Australia’s National Science Agency Hydrogeological assessment of the Cambrian Limestone Aquifer and Dook Creek Aquifer in the Roper catchment, Northern Territory A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid Andrew R Taylor1, Russell S Crosbie1, Chris Turnadge1, Sebastien Lamontagne1, Alec Deslandes1, Phil J Davies1, Karen Barry1, Axel Suckow1, Anthony Knapton2, Sarah Marshall1, Geoff Hodgson1, Steven Tickell3, Clement Duvert4, Lindsay B Hutley4 and Katelyn Dooley1 1 CSIRO; 2CloudGMS; 3 Northern Territory Department of Environment, Parks and Water Security 4 Charles Darwin University A black background with purple text Description automatically generated A logo with black text Description automatically generated A blue and white cloud logo Description automatically generated Email CSIRO Enquiries ISBN 978-1-4863-1923-7 (print) ISBN 978-1-4863-1924-4 (online) Citation Taylor AR, Crosbie RS, Turnadge C, Lamontagne S, Deslandes A, Davies PJ, Barry K, Suckow A, Knapton A, Marshall S, Hodgson G, Tickell S, Duvert C, Hutley LB and Dooley K (2023) Hydrogeological assessment of the Cambrian Limestone Aquifer and the Dook Creek Aquifer in the Roper catchment, Northern Territory. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Copyright © Commonwealth Scientific and Industrial Research Organisation 2024. To the extent permitted by law, all rights are reserved and no part of this publication covered by copyright may be reproduced or copied in any form or by any means except with the written permission of CSIRO. Important disclaimer CSIRO advises that the information contained in this publication comprises general statements based on scientific research. The reader is advised and needs to be aware that such information may be incomplete or unable to be used in any specific situation. No reliance or actions must therefore be made on that information without seeking prior expert professional, scientific and technical advice. To the extent permitted by law, CSIRO (including its employees and consultants) excludes all liability to any person for any consequences, including but not limited to all losses, damages, costs, expenses and any other compensation, arising directly or indirectly from using this publication (in part or in whole) and any information or material contained in it. CSIRO is committed to providing web accessible content wherever possible. If you are having difficulties with accessing this document, please contact csiroenquiries@csiro.au. CSIRO Roper River Water Resource Assessment acknowledgements This report was funded through the National Water Grid’s Science Program, which sits within the Australian Government’s Department of Climate Change, Energy, the Environment and Water. Aspects of the Assessment have been undertaken in conjunction with the Northern Territory Government. The Assessment was guided by two committees: i. The Assessment’s Governance Committee: CRC for Northern Australia/James Cook University; CSIRO; National Water Grid (Department of Climate Change, Energy, the Environment and Water); NT Department of Environment, Parks and Water Security; NT Department of Industry, Tourism and Trade; Office of Northern Australia; Qld Department of Agriculture and Fisheries; Qld Department of Regional Development, Manufacturing and Water ii. The Assessment’s joint Roper and Victoria River catchments Steering Committee: Amateur Fishermen’s Association of the NT; Austrade; Centrefarm; CSIRO, National Water Grid (Department of Climate Change, Energy, the Environment and Water); Northern Land Council; NT Cattlemen’s Association; NT Department of Environment, Parks Australia; Parks and Water Security; NT Department of Industry, Tourism and Trade; Regional Development Australia; NT Farmers; NT Seafood Council; Office of Northern Australia; Roper Gulf Regional Council Shire Responsibility for the Assessment’s content lies with CSIRO. The Assessment’s committees did not have an opportunity to review the Assessment results or outputs prior to its release. This report was reviewed by Matthias Raiber and Rebecca Doble of CSIRO. Acknowledgement of Country CSIRO acknowledges the Traditional Owners of the lands, seas and waters, of the area that we live and work on across Australia. We acknowledge their continuing connection to their culture and pay our respects to their Elders past and present. CSIRO would also like to acknowledge a large number of different stakeholders that assisted the authors of this research: (i) Dion Bununjoa a Traditional Owner from Weemol that provided cultural heritage guidance during our field sampling, (ii) Peter Davidson assisted the field team while sampling at Momob Outstation, (iii) the Mangarrayi Rangers that provided permission to access Mangarrayi Aboriginal Land Trust for field sampling, (iv) Traditional Owner Tristan of Wubalawun Aboriginal Land Trust that provided permission to undertake field sampling, (v) Bridie Verik Lord, Sam Tapp and Jamalia Irwin who provided assistance and support from the Northen Land Council, (vi) Paul Burke of the NT Farmers Association and Will Evans of the Cattlemen's Association for useful discussions related to irrigated agriculture and the pastoral industry, (vii) Des Yin Foo, Simon Cruickshank, Peter Waugh, Claire Carter, Steve Tickell, Adrian Costar, Michelle Rodrigo, Liza Schenkel, Maddison Clonan, Michael Short, Trevelyan Edwards, and Tobiah Amery of NT Government for assistance with various aspects of the research program, (viii) the following pastoral station owners and managers who provided access to stations and guidance to groundwater bores for field sampling: Ian Hoare and Andrew of Elsey Station, Cricket of Warloch Station, Todd Trengrove of West Elsey Station, Lindsay of Vermelha Station, Jess and Joanne of Maryfield Station, Des and Emily Carey of Kalala Station, Andrew and Kerry of Mountain Valley Station, (ix) Rachael Waters for assistance from the Roper Gulf Council, and (x) Colton Perna of Charles Darwin University for assistance with sampling the Roper River by boat. Photo L: groundwater-fed waterhole in the lower reach of Elsey Creek. R upper photo: above-ground casing of a monitoring bore installed in the Dook Creek Formation. R lower photo: groundwater discharging from a submersible groundwater pump installed in the monitoring bore in the photo above. Source: CSIRO Director’s foreword Sustainable regional development is a priority for the Australian and Northern Territory governments. Across northern Australia, however, there is a scarcity of scientific information on land and water resources to complement local information held by Indigenous owners and landholders. Sustainable regional development requires knowledge of the scale, nature, location and distribution of the likely environmental, social and economic opportunities and the risks of any proposed development. Especially where resource use is contested, this knowledge informs the consultation and planning that underpins the resource security required to unlock investment. In 2019 the Australian Government commissioned CSIRO to complete the Roper River Water Resource Assessment. In response, CSIRO accessed expertise and collaborations from across Australia to provide data and insight to support consideration of the use of land and water resources for development in the Roper catchment. While the Assessment focuses mainly on the potential for agriculture, the detailed information provided on land and water resources, their potential uses and the impacts of those uses are relevant to a wider range of regional-scale planning considerations by Indigenous owners, landholders, citizens, investors, local government, the Northern Territory and federal governments. Importantly the Assessment will not recommend one development over another, nor assume any particular development pathway. It provides a range of possibilities and the information required to interpret them - including risks that may attend any opportunities - consistent with regional values and aspirations. All data and reports produced by the Assessment will be publicly available. Chris Chilcott Project Director C:\Users\bru119\AppData\Local\Microsoft\Windows\Temporary Internet Files\Content.Word\C_Chilcott_high.jpg The Roper River Water Resource Assessment Team Project Director Chris Chilcott Project Leaders Cuan Petheram, Ian Watson Project Support Caroline Bruce Communications Chanel Koeleman/Kate Cranney, Siobhan Duffy, Amy Edwards Activities Agriculture and socioeconomics Chris Stokes, Caroline Bruce, Shokhrukh Jalilov, Diane Jarvis1, Adam Liedloff, Yvette Oliver, Alex Peachey2, Allan Peake, Maxine Piggott, Perry Poulton, Di Prestwidge, Thomas Vanderbyl7, Tony Webster, Steve Yeates Climate David McJannet, Lynn Seo Ecology Groundwater hydrology Indigenous water values, rights, interests and development goals Danial Stratford, Laura Blamey, Rik Buckworth, Pascal Castellazzi, Bayley Costin, Roy Aijun Deng, Ruan Gannon, Sophie Gilbey, Rob Kenyon, Darran King, Keller Kopf3, Stacey Kopf3, Simon Linke, Heather McGinness, Linda Merrin, Colton Perna3, Eva Plaganyi, Rocio Ponce Reyes, Jodie Pritchard, Nathan Waltham9 Andrew R. Taylor, Karen Barry, Russell Crosbie, Phil Davies, Alec Deslandes, Katelyn Dooley, Clement Duvert8, Geoff Hodgson, Lindsay Hutley8, Anthony Knapton4, Sebastien Lamontagne, Steven Tickell5, Sarah Marshall, Axel Suckow, Chris Turnadge Pethie Lyons, Marcus Barber, Peta Braedon, Kristina Fisher, Petina Pert Land suitability Ian Watson, Jenet Austin, Elisabeth Bui, Bart Edmeades5, John Gallant, Linda Gregory, Jason Hill5, Seonaid Philip, Ross Searle, Uta Stockmann, Mark Thomas, Francis Wait5, Peter L. Wilson, Peter R. Wilson Surface water hydrology Justin Hughes, Shaun Kim, Steve Marvanek, Catherine Ticehurst, Biao Wang Surface water storage Cuan Petheram, Fred Baynes6, Kevin Devlin7, Arthur Read, Lee Rogers, Ang Yang, Note: Assessment team as at June 15, 2023. All contributors are affiliated with CSIRO unless indicated otherwise. Activity Leaders are underlined. 1James Cook University; 2NT Department of Industry, Tourism and Trade; 3 Research Institute for the Environment and Livelihoods. College of Engineering, IT & Environment. Charles Darwin University; 4CloudGMS; 5NT Department of Environment, Parks and Water Security; 6Baynes Geologic; 7independent consultant; 8Charles Darwin University; 9Centre for Tropical Water and Aquatic Ecosystem Research. James Cook University. ii | Hydrogeological assessment of the Roper catchment Shortened forms SHORT FORM FULL FORM 3H tritium AET actual evapotranspiration AMS accelerator mass spectrometry ANU Australian National University APV Antrim Plateau Volcanics Ar argon BCs boundary conditions BE barometric efficiency BMM binary mixing model BoM Bureau of Meteorology Ca calcium CBE charge balance error CFCs chlorofluorocarbons Cl chloride CLA Cambrian Limestone Aquifer CMB chloride mass balance DCA Dook Creek Aquifer DCF Dook Creek Formation DEA Digital Earth Australia DIWA Directory of Important Wetlands in Australia DOI Digital Object Identifier DRBWCD Daly Roper Beetaloo Water Control District EC electrical conductivity EM exponential model ET evapotranspiration GDE groundwater-dependent ecosystem GEMIS Geoscience Exploration and Mining Information Systems GPS global positioning system GWL depth to groundwater GWWAP Georgina Wiso Water Allocation Plan H1301 bromotrifluoromethane HCO3 bicarbonate He helium SHORT FORM FULL FORM HNO3 nitric acid ICPOES inductively coupled plasma optical emission spectrometry IRMS isotope ratio mass spectrometer K hydraulic connectivity Kr krypton LMWL local meteoric water line LPM lumped parameter model LWMZ Larrimah Water Management Zone M2 primary lunar constituent mAHD metres above Australian Height Datum mBGL metres below ground level mBTOC metres below top of casing Mg magnesium MRT mean residence time MTLAWAP Mataranka Tindall Limestone Aquifer Water Allocation Plan MWMZ Mataranka Water Management Zone Na sodium Ne neon NGIS National Groundwater Information System NR Maps Natural Resources Maps NT Northern Territory NTGS Northern Territory Geological Survey PET potential evapotranspiration PM piston flow model pmC percent modern carbon PWC Power and Water Corporation RoWRA Roper River Water Resource Assessment RSWL reduced standing water level S storage coefficient S1 solar diurnal S2 solar semi-diurnal NR Maps Natural Resources Maps NT Northern Territory NTGS Northern Territory Geological Survey PET potential evapotranspiration PM piston flow model K hydraulic conductivity pmC percent modern carbon PWC Power and Water Corporation RoWRA Roper River Water Resource Assessment RSWL reduced standing water level S storage coefficient S1 solar diurnal S2 solar semi-diurnal SF6 sulfur hexafluoride SI saturation indices SO4 sulfate SRTM Shuttle Radar Topography Mission STP standard temperature and pressure SWL standing water level Sy specific yield T transmissivity TDIC total dissolved inorganic carbon TDS total dissolved solids VPDB Vienna Pee Dee Belemnite VSMOW Vienna Standard Mean Ocean Water WOfS Water Observations from Space WSSR weighted sum of squared residuals Xe xenon Units UNIT DESCRIPTION μS microsiemens ‰ per mille Bq becquerel cc or cc(STP) cubic centimetre of gas at standard temperature and pressure cm centimetre d day fmol femtomole g gram GL gigalitre km kilometre L litre m metre meq milliequivalent mg milligrams ML megalitres mM millimole pMC percent modern carbon pMol picomole s second TU tritium unit y year μm micrometre Preface Sustainable regional development is a priority for the Australian and Northern Territory governments. For example, in 2023 the Northern Territory Government committed to the implementation of a new Territory Water Plan. One of the priority actions announced by the government was the acceleration of the existing water science program ‘to support best practice water resource management and sustainable development’. The efficient use of Australia’s natural resources by food producers and processors requires a good understanding of soil, water and energy resources so they can be managed sustainably. Finely tuned strategic planning will be required to ensure that investment and government expenditure on development are soundly targeted and designed. Northern Australia presents a globally unique opportunity (a greenfield development opportunity in a first-world country) to strategically consider and plan development. Northern Australia also contains ecological and cultural assets of high value and decisions about development will need to be made within that context. Good information is critical to these decisions. Most of northern Australia’s land and water resources, however, have not been mapped in sufficient detail to provide for reliable resource allocation, mitigate investment or environmental risks, or build policy settings that can support decisions. Better data are required to inform decisions on private investment and government expenditure, to account for intersections between existing and potential resource users, and to ensure that net development benefits are maximised. In consultation with the Northern Territory Government, the Australian Government prioritised the catchment of the Roper River for investigation (Preface Figure 1-1) and establishment of baseline information on soil, water and the environment. Northern Australia is defined as the part of Australia north of the Tropic of Capricorn. The Murray– Darling Basin and major irrigation areas and major dams (greater than 500 GL capacity) in Australia are shown for context. The Roper River Water Resource Assessment (the Assessment) provides a comprehensive and integrated evaluation of the feasibility, economic viability and sustainability of water and agricultural development. While agricultural developments are the primary focus of the Assessment, it also considers opportunities for and intersections between other types of water-dependent development. For example, the Assessment explores the nature, scale, location and impacts of developments relating to industrial and urban development and aquaculture, in relevant locations. The Assessment was designed to inform consideration of development, not to enable any particular development to occur. As such, the Assessment informs – but does not seek to replace – existing planning, regulatory or approval processes. Importantly, the Assessment does not assume a given policy or regulatory environment. As policy and regulations can change, this enables the results to be applied to the widest range of uses for the longest possible time frame. Preface Figure 1-1 Map of Australia showing Assessment area It was not the intention – and nor was it possible – for the Assessment to generate new information on all topics related to water and irrigation development in northern Australia. Topics not directly examined in the Assessment are discussed with reference to and in the context of the existing literature. Functionally, the Assessment adopted an activities-based approach (reflected in the content and structure of the outputs and products), comprising eight activity groups; each contributes its part to create a cohesive picture of regional development opportunities, costs and benefits. Preface Figure 1-2 illustrates the high-level links between the eight activities and the general flow of information in the Assessment. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au Preface Figure 1-2 Schematic diagram of the high-level linkages between the eight activities and the general flow of information in the Assessment. Assessment reporting structure Development opportunities and their impacts are frequently highly interdependent and consequently, so is the research undertaken through this Assessment. While each report may be read as a stand-alone document, the suite of reports most reliably informs discussion and decisions concerning regional development when read as a whole. The Assessment has produced a series of cascading reports and information products: • Technical reports; that present scientific work at a level of detail sufficient for technical and scientific experts to reproduce the work. Each of the eight activities has one or more corresponding technical report. • A Catchment report; that for the Roper catchment synthesises key material from the technical reports, providing well-informed (but not necessarily-scientifically trained) readers with the information required to make decisions about the opportunities, costs and benefits associated with irrigated agriculture and other development options. • A Summary report; that for the Roper catchment provides a summary and narrative for a general public audience in plain English. • A Summary factsheet; that for the Roper catchment provides key findings for a general public audience in the shortest possible format. The Assessment has also developed online information products to enable the reader to better access information that is not readily available in a static form. All of these reports, information tools and data products are available online at https://www.csiro.au/roperriver. The website provides readers with a communications suite including factsheets, multimedia content, FAQs, reports and links to other related sites, particularly about other research in northern Australia. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au Executive summary The catchment of the Roper River in the Northern Territory encompasses an area of approximately 77,400 km2 and lies in the wet–dry tropics of northern Australia. The catchment overlies five major geological provinces including: (i) the McArthur Basin, which underlies the centre, north and east of the catchment, (ii) the interconnected Daly, Wiso and Georgina basins, which overlie the McArthur Basin in the south to south-west of the catchment, and (iii) the geological Carpentaria Basin, which overlies most of the Daly, Wiso and Georgina basins in the south to south-west, as well as minor parts of the McArthur Basin in the north-west of the catchment. Within the Roper catchment, the availability and quality of groundwater resources are heavily influenced by the north–south climate gradient as well as the physical characteristics of rocks of the major geological provinces. Numerous aquifer types are hosted in each of the provinces across the catchment including: (i) highly productive regional-scale weathered, fractured or karstic limestones of the connected Daly, Wiso and Georgina basins in the south-west, (ii) highly productive intermediate-scale weathered, fractured or karstic dolostones and porous sandstones of the McArthur Basin in the north and east, (iii) variably productive local-scale fractured and weathered sandstone, siltstone and mudstones of the Roper River and Katherine River groups in the centre and north, (iv) poorly productive minor-scale siltstones of the Daly and Georgina basins and minor-scale basalt of the Kalkarindji Igneous Province, (v) variably productive Cretaceous sandstone of the geological Carpentaria Basin, and (vi) minor-scale unconsolidated Quaternary regolith and alluvium. This hydrogeological assessment involved several key components: (i) a literature and data review of all previous hydrogeological investigations, (ii) regional-scale desktop data collation and analyses, including digitising data contained in hand-written and typed drilling records, evaluating groundwater levels, groundwater salinity and bore yields, and reinterpreting pumping test data, (iii) a regional-scale recharge modelling assessment of all aquifers, (iv) identification of the most promising intermediate- to regional-scale aquifers for undertaking targeted field, desktop and modelling investigations, and (v) targeted investigations of the most promising aquifers to provide detailed information to underpin the future planning, investment and management of key groundwater resources. The literature review provided good insight into all aquifers and their current knowledge gaps. The regional-scale assessment proved useful as a screening tool for identifying spatial trends in important groundwater attributes, including the spatial extent of aquifers, ranges for groundwater levels, groundwater salinity, bore yields and aquifer hydraulic properties. It also helped identify the most promising aquifers for targeted investigations and provided important information to: (i) strategically select bores appropriate for environmental tracer sampling, (ii) identify strategic locations for surface water sampling for environmental tracers, and (iii) provide baseline datasets for use in detailed desktop analyses and modelling. A spatial evaluation of key groundwater attributes highlighted the geographic regions beneath the catchment that contained aquifers hosting either local-scale (1 to 10 km), intermediate-scale (10 to 50 km), or regional-scale (>50 to >100 km) groundwater flow systems. Local-scale aquifers generally had variable groundwater salinities (<500 to >7500 mg/L) and variable but low bore yields (mostly <2 L/second). Localised aquifers included the various sedimentary and igneous fractured and weathered rock aquifers hosted in the Proterozoic Roper River and Katherine River groups and sandstone aquifers hosted in the Cretaceous geological Carpentaria Basin, all of which are currently poorly characterised. However, these aquifers can provide an important localised source of water for stock and domestic use and in a few places across the catchment support community water supplies. Intermediate- to regional-scale aquifers generally had lower groundwater salinities (<500 to 3000 mg/L) and higher bore yields (ranging from 15 to 50 L/second where pumping tests had been conducted). These aquifers include: (i) the regional-scale Cambrian Limestone Aquifer (CLA) hosted in three equivalent hydrogeological units (Tindall Limestone, Montejinni Limestone and Gum Ridge Formation) of the interconnected Daly, Wiso and Georgina basins, (ii) the intermediate-scale Dook Creek Aquifer (DCA) hosted in the dolostones in the Dook Creek Formation of the Mount Rigg Group of the McArthur Basin, and (iii) intermediate-scale sandstone aquifers hosted in the Yalwarra Volcanics, Knuckey Formation and Mount Birch Sandstone in the Nathan Group of the McArthur Basin. The regional-scale CLA and intermediatescale DCA were the focus for targeted investigations in this assessment. Cambrian Limestone Aquifer Targeted investigations of the CLA in the catchment have helped validate and refine the existing hydrogeological conceptual model for the aquifer. Recharge was found to occur across almost the entire spatial extent of the aquifer where it is unconfined (northern aquifer outcrop, or beneath an extensive veneer of overlying Cretaceous strata) but was most prominent around aquifer margins. Localised recharge occurs: (i) via direct infiltration of rainfall and streamflow via sinkholes in and near the aquifer outcrop, or (ii) where surficial features (sinkholes, ephemeral streams or waterholes) are incised through thinner (i.e. <30 m thick) parts of overlying Cretaceous strata along aquifer margins. Broad diffuse recharge occurs via vertical leakage through overlying Cretaceous strata across much larger parts of the aquifer, though rates are lower where the overlying strata is thicker (i.e. >30 m). These geological controls (presence and permeability of karstic features and presence, thickness and permeability of overlying Cretaceous strata) have a large influence on recharge processes and rates to the underlying CLA (i.e. localised preferential leakage versus minor broad diffuse leakage). In addition, while localised recharge in the aquifer outcrop is an important process, particularly for supporting spring flow and baseflow to parts of the regional groundwater discharge zone, the outcrop is storage limited (i.e. the watertable is ≤5 m below ground level (mBGL)) and is a net discharge zone (i.e. phreatophytes transpire shallow groundwater and groundwater discharges from the aquifer to springs and streams). Recharge rates estimated by both upscaled chloride mass balance (CMB) and tritium (3H) concentrations in groundwater indicate contemporary recharge rates (i.e. over several recent decades) ranging between 3 and 70 mm/year. Numerical modelling of annual recharge indicated mean and median annual recharge fluxes across the CLA in the catchment were 995 and 469 GL/year, respectively. Mapping the hydraulic head across the CLA has further confirmed that regional groundwater flow (i.e. flow paths >100 km) beneath the catchment is generally from south to north following topography. Regional flow paths are influenced by the presence of a structural high of the Antrim Plateau Volcanics (APV) near Larrimah which guides flow paths to the north (Georgina Flow Path) and north-west (Flora Flow Path). Groundwater chemistry and the stable hydrogen and oxygen isotopes in groundwater and surface water have further confirmed that the upper Roper River, its tributaries and the Mataranka Springs Complex receive groundwater from flow paths in the CLA that occur at a variety of scales and from spatially variable directions (i.e. local-scale flow from the northern aquifer margin, intermediate-scale flow from the west and a combination of local and regional-scale flow from the south). These variable sources of discharge are an important consideration for adaptive management of the CLA into the future. In addition, results of new drilling in the regional groundwater discharge zone further highlight important vertical and horizontal geological controls on groundwater flow and discharge. Very low permeability siltstone and sandstone and the crystalline basement of the McArthur Basin were encountered at depths of only about 100 mBGL within 10 km of the Roper River. The structure of these basement units influences the upward dipping of the APV and overlying CLA which outcrops at the surface, and the upper Roper River and parts of its main tributaries are incised in the aquifer. Furthermore, these structural controls also influence the position of Slat and Elsey Creeks near their junction with the Roper River. Semi-quantitative estimates of mean residence times (MRTs) using concentrations of 3H and carbon-14 (14C) indicate that the timescales for groundwater flow also vary significantly. MRTs range from several years for localised flow paths of less than 5 km in and near the aquifer outcrop to many hundreds of years for regional flow paths longer than 150 km from the south. However, there is a degree of uncertainty to these estimates due to carbonate dissolution across the aquifer. Nevertheless, they are consistent with the timescales for flow estimates using numerical modelling. Numerical scenario-based groundwater flow modelling of the CLA was conducted to evaluate: (i) the current water balance, and (ii) the potential impacts of changes in climate and/or potential increased groundwater resource development on the streamflow of the upper Roper River and groundwater levels in the vicinity of existing licensed groundwater users and environmental receptors in the Mataranka region. Mean annual recharge modelling indicated that recent climate conditions have led to significantly higher recharge than the long-term mean. Using the historical climate to simulate the future climate to 2070 suggests that hypothetical increase in groundwater extraction of up to 105 GL/year will result in a further 2% reduction in discharge to the Roper River and additional drawdown in groundwater levels compared to current licensed groundwater extraction. Drawdown levels were estimated to decrease by 0.9 m in the regional groundwater discharge zone and up to 11.8 m at a distance of 110 km south of the river. Increased groundwater extraction does not linearly correspond to a proportional decrease in groundwater discharge to the Roper River. This is because a large proportion of groundwater extraction is partitioned into reductions in discharge to evapotranspiration (ET), and capture of groundwater throughflow. This is also an important finding for underpinning adaptive management of the CLA in the future. Simulating a future drier climate (10% reduction in rainfall) to 2070 with current licensed groundwater extraction results in a further 21% reduction in discharge to the Roper River and additional drawdown in groundwater levels compared to current licensed groundwater extraction under a historical climate. Drawdown levels would be expected to drop by 0.9 m in the regional groundwater discharge zone and up to 3.9 m at a distance of 110 km south of the river. Simulations of the hypothetical climate scenarios have highlighted that the potential impact on water resources in the Roper catchment of climate variability is more significant than that of current groundwater extraction due to its influence on spatial and temporal variability in groundwater recharge to and discharge from the karstic groundwater system. Furthermore, the extensive spatial coverage, thickness, and variable hydraulic properties of the karstic groundwater system significantly influence the time lags for hydrological impacts of both climate variability and groundwater extraction to propagate through the system. In some parts of the CLA, these processes take many hundreds of years to occur. Dook Creek Aquifer Targeted investigations of the DCA in the Roper catchment have also helped validate and refine the existing hydrogeological conceptual model for the aquifer. Recharge to the DCA in the Roper catchment occurs across the unconfined part of the aquifer west of the Central Arnhem Road where it outcrops and subcrops beneath a thin and patchy veneer of overlying Cretaceous strata. Localised recharge is the most dominant recharge process across the DCA. This recharge occurs by direct infiltration of rainfall and streamflow via sinkholes or via vertical leakage through thin (i.e. ≤ 20m) overlying Cretaceous strata. Recharge rates estimated by both upscaled CMB and 3H concentrations in groundwater indicate a range in contemporary recharge rates for the DCA from 3 to 100 mm/year. Numerical modelling of annual recharge indicated mean and median annual recharge fluxes across the DCA in the Roper catchment of 150 and 90 GL/year, respectively. Modelling of hydraulic head data for the DCA indicates that regional-scale (i.e. >100 km) to intermediate-scale (i.e. 20 to 50 km) groundwater flow is generally from south-west to north-east flowing a subdued form of the topographic gradient. While longer flow paths follow the topographic gradient to the north-east and discharge outside the catchment, shallower and shorter intermediate-scale flows are truncated by the position of the northern tributaries of the Roper River. That is where Flying Fox Creek and the Mainoru and Wilton rivers are incised in the DCA outcrop and receive lateral outflow from the aquifer, or where the lower reaches join the contact with the rocks of the Roper Group. More abundant spring complexes around the Wilton River associated with the Bulman fault indicate this is the main discharge for the DCA in the Roper catchment. Timescales for groundwater flow are highly variable in the karstic dolostone aquifer and also where the aquifer is hosted in siltstone lithologies of the DCA. Semi-quantitative estimates of MRTs for groundwater flow derived using 3H and 14C concentrations in groundwater also vary significantly across the aquifer. They range from several years for localised flow paths of less than 5 km in shallow unconfined parts of the DCA to a few thousand to many thousands of years for regional flow paths more than 100 km towards discharge areas or within lowpermeability siltstone parts of the aquifer. There is, however, a degree of uncertainty in these MRT estimates due to carbonate dissolution across the aquifer. Numerical scenario-based groundwater flow modelling of the DCA in the Roper catchment was conducted for the same reasons as described for the CLA. However, there is very little groundwater development currently across the aquifer, and the environmental receptors associated with the aquifer occur mostly between Flying Fox Creek and the Wilton River near Bulman. Using the historical climate to simulate the future climate to 2070 suggests that a hypothetical increase in groundwater extraction of up to 18 GL/year in the DCA will result in 11% reductions in discharge to both the Wilton River and Flying Fox Creek. Additional drawdown in groundwater levels is less than 2 m across the unconfined area of the aquifer. As in the CLA, increased groundwater extraction in the DCA does not linearly correspond to a proportional decrease in groundwater discharge to the Wilton River and Flying Fox Creek. A proportion of the groundwater extraction is partitioned into reductions in discharge to ET and capture of groundwater throughflow. Simulating a future drier climate (10% reduction in rainfall) to 2070 with current licensed groundwater extraction (only 0.1 GL/year) results in further 22 and 45% reductions in groundwater discharge to the Wilton River and Flying Fox Creek, respectively. Reductions in groundwater levels were less than 4 m across the unconfined area of the aquifer compared to current licensed groundwater extraction (0.1 GL/year) under a historical climate. As in the CLA, simulations of the hypothetical climate scenarios highlight the potential importance of considering the influence of climate variability on the water balance of the DCA in the future if the groundwater resource is to be further developed. Furthermore, and as in the CLA, the extensive spatial coverage, thickness and variable hydraulic properties of the karstic groundwater system significantly influence the time lags for hydrological impacts of both climate variability and groundwater extraction to propagate through the system. In some parts of the DCA, these processes take many hundreds of years to occur. This is a potentially useful finding for underpinning adaptive management of the DCA in the future if it is to be further developed. Groundwater development opportunities The unconfined parts of the regional-scale CLA and intermediate-scale DCA provide the greatest potential opportunities for future groundwater resource development across the catchment. Numerical modelling of the water balance for both aquifers have demonstrated that with appropriately sited borefields, it is estimated from numerical modelling that between 35 and 105 GL/year (~3 to 10% of recharge to the entire CLA) could potentially be extracted from the CLA in the vicinity of, and to the south of, Larrimah (i.e. groundwater extraction occurring between 60 and 160 km from existing groundwater users and ecologically and culturally important GDEs near Mataranka, depending upon community and government acceptance of potential impacts on groundwater-dependent ecosystems and existing groundwater users. This is the unconfined part of the CLA in the southern Daly Basin near Larrimah south to the northern Georgina Basin near Daly Waters where the aquifer has a sufficient saturated thickness (i.e. >20 m). The CLA in the southern Daly Basin west of Larrimah and south-west to the northern Wiso Basin has a thin (i.e. <20 m) saturated thickness and historical drilling has shown a low probability for installing productive (i.e. >30 L/second) high-yielding bores. This part of the CLA in the Roper catchment is therefore not likely to be suitable for groundwater-based irrigation. In addition, numerical modelling has also demonstrated between 6 and 18 GL/year could potentially be extracted from the western unconfined part of the DCA, west of the Central Arnhem Road between Flying Fox Creek and Bulman, depending upon community and government acceptance of potential impacts on groundwater-dependent ecosystems and existing groundwater users. These findings have been confirmed by the assessments targeted field, desktop and modelling investigations that demonstrate the unconfined parts of both aquifers: • have a large spatial extent that coincides with land suitable for agricultural intensification, making it accessible to many stakeholders away from existing water users and groundwaterdependent ecosystems (GDEs) • have depths to groundwater across large areas of less than 100 mBGL, thereby exhibiting economical pumping costs • can be intersected with drilling at economical depths in many places • have predominantly good-quality groundwater that is suitable for a wide range of purposes, noting some parts of the CLA host slightly brackish groundwater • have potential to support multiple dispersed groundwater developments – large (5 to 10 GL/year) across the CLA and small to moderate (1 to 3 GL/year) across the DCA. Sparse hydrogeological data for the intermediate-scale aquifers hosted in sandstone of the Yalwarra Volcanics, Knuckey Formation and Mount Birch Sandstone may also offer some small- to moderate-scale opportunities on the southern side of the Roper River south of Ngukurr. However, further drilling and pumping tests are required to confirm their presence, extent and prospectivity. Constraints on groundwater development opportunities Cambrian Limestone Aquifer Potential constraints on larger-scale (i.e. >5 GL/year) potential future groundwater resource development opportunities for groundwater-based irrigation from the CLA in the Roper catchment include: • being constrained to areas across the CLA where water is currently available for consumptive use such as those more than 50 km south of Mataranka • being constrained to areas where the saturated thickness of the aquifer is sufficient (i.e. >20 m) to support the installation of production bores (thus, mostly in the southern Daly Basin and northern Georgina Basin around and south of Larrimah) • installation of successful productive high-yielding (i.e. bore yields >30 L/second) bores may require the drilling of multiple investigations holes to identify productive parts of the aquifer which can create uncertainty in the cost of drilling programs • being constrained to areas where the depth to groundwater is less than or equal to 100 mBGL depending on the value of irrigated crops being grown, which may make bores prohibitively expensive in southern parts of the Roper catchment (where the depth is greater than 100 m BGL) when growing low-value crops • the need to consider crop selection and irrigation of fine textured soils across parts of the CLA in the southern Daly Basin and northern Georgina Basin as some areas exhibit slightly brackish groundwater (i.e. 1000 to 3000 mg/L TDS) which can degrade crops depending on their specific salt tolerance as well as cause soil structure decline such as dispersion • the need to maintain water quality across the aquifer given some parts of the CLA are brackish and increases in irrigated agriculture have the potential to increase salinity of the aquifer over time • the need to maintain water availability to all groundwater users and groundwater-dependent ecosystems as future demand for water increases and climate variability has a large influence on the aquifer water balance. The latter reflected by natural variations in recharge to the aquifer under the current and historical climate which is important consideration for adaptive management on the aquifer into the future. Dook Creek Aquifer Potential constraints on moderate-scale (i.e. 1 to 3 GL/year) potential future groundwater resource development opportunities for groundwater-based irrigation from the DCA in the Roper catchment include: • being constrained to areas across the DCA at sufficient distances away from Flying Fox Creek and the Wilton River as they only exhibit minor flows (i.e. <0.7 m3/s) • the need to better characterise other potential areas of groundwater–surface water interactions as identified, as these are likely to be a key constraint on future groundwater resource development in certain locations • being constrained to areas across the DCA away from Bulman, Weemol and numerous other culturally important outstations (Momob, Bodeidei Camp, Morbon/Blue Water) • the need to better characterise the water balance, saturated thickness, groundwater levels and hydraulic properties of the aquifer, as the DCA is currently data sparse across large areas • installation of successful productive high-yielding (i.e. bore yields >30 L/second) bores may require the drilling of multiple investigations holes to identify productive parts of the aquifer which can create uncertainty in the cost of drilling programs • the need to maintain water availability to all groundwater users and groundwater-dependent ecosystems as future demand for water increases and climate variability has a large influence on the aquifer water balance. The latter reflected by natural variations in recharge to the aquifer under the current and historical climate which is important consideration for adaptive management on the aquifer into the future. Contents Director’s foreword .......................................................................................................................... i The Roper River Water Resource Assessment Team ...................................................................... ii Shortened forms .............................................................................................................................iii Units ............................................................................................................................... vi Preface .............................................................................................................................. vii Executive summary .......................................................................................................................... x Part I Introduction and overview 1 1 Introduction ........................................................................................................................ 2 1.1 Overview ................................................................................................................ 2 1.2 Aims of the groundwater hydrology activity ......................................................... 3 1.3 Summary of previous hydrogeological investigations .......................................... 3 1.4 Current licensed water entitlements .................................................................... 5 1.5 Promising aquifers for targeted field, desktop and modelling investigations ...... 8 1.6 Report structure .................................................................................................... 8 2 Study area ......................................................................................................................... 10 2.1 Physiography and demography ........................................................................... 10 2.2 Climate ................................................................................................................. 13 2.3 Geology ................................................................................................................ 16 2.4 Hydrogeology....................................................................................................... 26 2.5 Surface water hydrology ..................................................................................... 37 2.6 Water-dependent ecosystems ............................................................................ 39 Part II Methods 41 3 Regional desktop assessment of the Roper catchment ................................................... 42 3.1 Geology, hydrogeology, and aquifer types ......................................................... 42 3.2 Groundwater levels ............................................................................................. 43 3.3 Groundwater salinity ........................................................................................... 44 3.4 Bore yields ........................................................................................................... 44 3.5 Aquifer hydraulic properties ............................................................................... 44 3.6 Recharge estimation ............................................................................................ 47 3.7 Identifying potential groundwater discharge areas using remote sensing ........ 59 4 Targeted field, desktop, and modelling investigations .................................................... 64 4.1 Hydrogeological framework ................................................................................ 64 4.2 Groundwater recharge and flow ......................................................................... 69 4.3 Numerical groundwater flow modelling ............................................................. 88 Part III Results 105 5 Regional assessment of the Roper catchment ............................................................... 106 5.1 Groundwater levels ........................................................................................... 106 5.2 Groundwater salinity ......................................................................................... 113 5.3 Bore yields ......................................................................................................... 114 5.4 Aquifer hydraulic properties ............................................................................. 116 5.5 Recharge estimation .......................................................................................... 129 5.6 Identifying potential groundwater discharge areas using remote sensing ...... 138 6 Targeted field, desktop and modelling investigations ................................................... 159 6.1 Hydrogeological framework .............................................................................. 159 6.2 Groundwater recharge and flow ....................................................................... 173 6.3 Numerical groundwater flow modelling ........................................................... 231 Part IV Discussion and conclusions 245 7 Discussion ....................................................................................................................... 246 7.1 Cambrian Limestone Aquifer ............................................................................. 248 7.2 Dook Creek Aquifer ........................................................................................... 255 7.3 Potential opportunities for future groundwater resource development ......... 259 7.4 Constraints on potential future groundwater resource development ............. 265 8 Summary and conclusions .............................................................................................. 267 8.1 Opportunities and constraints for future groundwater development ............. 267 8.2 Potential options for future work...................................................................... 268 References ........................................................................................................................... 270 Part IV Appendices 291 Figures Figure 1-1 Location, type and volume of annual licensed surface water and groundwater entitlements .................................................................................................................................... 7 Figure 2-1 The Roper catchment study area showing the Roper River and its tributaries and the spatial extent of the different physiographic regions .................................................................. 12 Figure 2-2 Historical (a) median annual rainfall and (b) median annual potential evaporation across the Roper catchment ......................................................................................................... 13 Figure 2-3 Historical monthly rainfall (left) and time series of annual rainfall (right) in the Roper catchment at Ngukurr, Mataranka, Larrimah and Bulman ........................................................... 14 Figure 2-4 Historical monthly potential evaporation (left) and time series of annual potential evaporation (right) in the Roper catchment at Ngukurr, Mataranka, Larrimah and Bulman ...... 15 Figure 2-5 Major geological provinces of the Roper catchment .................................................. 17 Figure 2-6 Surface geology of the Roper catchment .................................................................... 19 Figure 2-7 Simplified regional geology of the Roper catchment .................................................. 21 Figure 2-8 Simplified regional geology of the Roper catchment including the entire spatial extent of the Mount Rigg Group of the McArthur Basin and the Tindall Limestone and equivalents of the Daly, Wiso and Georgina basins ..................................................................... 23 Figure 2-9 Spatial changes in modelled depth to basement beneath the Roper catchment ...... 25 Figure 2-10 Aquifer types occurring within and beneath the Roper catchment .......................... 27 Figure 2-11 Simplified regional hydrogeology of the Roper catchment....................................... 31 Figure 2-12 Entire spatial extent of the Cambrian Limestone Aquifer and Dook Creek Aquifer (Proterozoic dolostone) beneath and beyond the Roper catchment .......................................... 35 Figure 2-13 Streamflow observation data availability in the Roper catchment and median annual streamflow (50% exceedance) under Scenario A ............................................................. 38 Figure 2-14 Location of ecological assets related to groundwater including Directory of Important Wetlands in Australia (DIWA) sites, groundwater-dependent ecosystems identified in the GDE Atlas and mapped spring locations across the Roper catchment .................................. 40 Figure 3-1. Methods used showing calculating the point recharge, upscaling using regression kriging and reporting of the results at the scale of the region and surface geology group ......... 48 Figure 3-2 Investigation area used for estimating recharge using the chloride mass balance method .......................................................................................................................................... 49 Figure 3-3 Chloride deposition across the study region: (a) mean, (b) standard deviation, and (c) skewness (Wilkins et al., 2022a) ................................................................................................... 51 Figure 3-4 Runoff coefficient across the investigation area ......................................................... 53 Figure 3-5 Covariates used in upscaling: (a) rainfall, (b) clay content of the soil, (c) Normalised Difference Vegetation Index (NDVI), and (d) simplified geology .................................................. 56 Figure 3-6 (a) Catchments used for baseflow analysis for rejection sampling, (b) additional internally draining catchments used with excess water for rejection sampling .......................... 59 Figure 3-7 Relationship used for determining the threshold October actual evapotranspiration for estimating potential discharge areas from aridity index ........................................................ 62 Figure 4-1 (a) Rotary drilling an investigation hole, (b) collected rock cuttings being logged by depth, (c) installing PVC casing, and (d) an complete final bore with steel standpipe set in concrete ........................................................................................................................................ 65 Figure 4-2 Spatial distribution of drilling and newly installed monitoring bores in Elsey National Park ............................................................................................................................................... 67 Figure 4-3 Target area for groundwater sampling across the unconfined part of the Cambrian Limestone Aquifer and adjacent aquifers between Daly Waters and Mataranka ....................... 72 Figure 4-4 Target area for groundwater sampling across the outcropping area of the Dook Creek Aquifer between Flying Fox Creek and the Goyder River ............................................................. 73 Figure 4-5 Inserting a battery-driven 12-volt Proactive plastic submersible Monsoon pump into a bore casing ................................................................................................................................. 76 Figure 4-6 (a) Collecting a groundwater sample in the field for radon-222 and (b) extracting the radon-222 from the water sample into mineral oil ...................................................................... 77 Figure 4-7 Collecting a groundwater sample for sulfur hexafluoride (SF6) .................................. 78 Figure 4-8 Preparing to collect a groundwater sample for dissolved noble gases using a copper tube ............................................................................................................................................... 79 Figure 4-9 Schematic cross-section representations of advective flow in the (a) piston flow model and (b) exponential model in idealised unconfined aquifers ............................................ 83 Figure 4-10 Target area for spring and surface water sampling in the groundwater discharge zone for the Cambrian Limestone Aquifer in the Roper catchment ............................................ 86 Figure 4-11 Target area for spring and surface water sampling across the outcropping and subcropping areas of the Dook Creek Aquifer .............................................................................. 87 Figure 4-12 Location of the Roper catchment and its relationship to the groundwater systems of the Dook Creek Aquifer and the Cambrian Limestone Aquifer in the Daly, Wiso and Georgina basins ............................................................................................................................................ 92 Figure 4-13 Spatial locations for hypothetical groundwater extraction sites across the Cambrian Limestone Aquifer relative to land that is potentially suitable for agricultural intensification, proposed water management zones and the groundwater discharge zone for the aquifer ..... 100 Figure 4-14 Spatial locations for hypothetical groundwater extraction sites across the Dook Creek Aquifer relative to land that is potentially suitable for agricultural intensification and groundwater discharge zone for the aquifer, including the major tributaries of the Roper River and discrete springs .................................................................................................................... 101 Figure 4-15 Spatial map showing the extent of the two proposed water management zones for reporting mean annual water balance results across the DR2 Cambrian Limestone Aquifer model, the extent of the DR2 Dook Creek Aquifer model within the Roper catchment, as well as groundwater-level reporting sites and groundwater discharge reporting sites ........................ 104 Figure 5-1 Static groundwater levels for (a) the major aquifers hosted in the Tindall Limestone and equivalents and the Mount Rigg and Nathan groups and (b) other minor aquifers of the Roper catchment ......................................................................................................................... 107 Figure 5-2 Groundwater hydrographs (green) measured using automated loggers installed in four groundwater wells in the Cambrian Limestone Aquifer in the vicinity of Mataranka and Elsey National Park ...................................................................................................................... 108 Figure 5-3 Groundwater hydrographs (green) measured using automated loggers installed in four groundwater wells in the Cambrian Limestone Aquifer (a,d) approximately 30 km south of Elsey National Park and (b,c) in the vicinity of Jilkminggan ........................................................ 110 Figure 5-4 Groundwater hydrographs (green) measured using automated loggers installed in four groundwater bores in aquifers of the (a,b) Dook Creek Formation, (c) Antrim Plateau Volcanics and (d) Cretaceous sandstone .................................................................................... 112 Figure 5-5 Groundwater salinity for (a) the major aquifers hosted in the Tindall Limestone and equivalents and the Mount Rigg and Nathan groups and (b) other minor aquifers of the Roper catchment ................................................................................................................................... 114 Figure 5-6 Groundwater bore yields for (a) the major aquifers hosted in the Tindall Limestone and equivalents and the Mount Rigg and Nathan groups and (b) other minor aquifers of the Roper catchment ......................................................................................................................... 115 Figure 5-7 Locations of groundwater bores at which two-bore constant-rate discharge pumping tests (coloured squares) or single-bore constant-rate discharge pumping tests (coloured circles) were undertaken ......................................................................................................................... 117 Figure 5-8 Locations of 31 groundwater bores at which time series of pressure were recorded using automated loggers (coloured circles) at a 4-hourly frequency or greater, for a continuous duration of 6 months or more .................................................................................................... 124 Figure 5-9 (a) Time series and (b) amplitude spectrum of hydraulic head measured at bore RN035926, located approximately 9 km south-east of Mataranka. (c) Time series and (d) amplitude spectrum of barometric pressure measured at same location over a shorter time period. (e) Time series and (f) amplitude spectrum of theoretical Earth tide vertical strain estimated over the same time period and at the same location ............................................... 126 Figure 5-10 (a) Chloride in groundwater observations within the study region and (b) the median of the point-scale estimates of recharge derived from them ....................................... 130 Figure 5-11 Point-scale relationships between recharge and (a) rainfall, (b) clay content of the soil, and (c) Normalised Difference Vegetation Index (NDVI) .................................................... 131 Figure 5-12 Point-scale relationships between rainfall and recharge by geology class ............. 131 Figure 5-13 (a-f) Coefficients used in the regression equations for upscaling the 1000 replicates, (g) the R2 for each of the 1000 replicates ................................................................................... 132 Figure 5-14 (a) The median value of the 1000 replicates of upscaled recharge using the regression equation and (b) the median of the residuals kriged to a regular grid and the points used ............................................................................................................................................. 133 Figure 5-15 The (a) 50th, (b) 5th and (c) 95th percentiles of upscaled recharge from the 1000 replicates using regression kriging .............................................................................................. 134 Figure 5-16 (a) Scatterplot of the 50th percentile of the recharge calculated at a point scale versus the 50th percentile of the regression kriging upscaled recharge estimate (red line is 1:1 for reference) and (b) the R2 of all 1000 replicates for the point estimates of recharge versus the upscaled estimates of recharge as a boxplot ............................................................................. 135 Figure 5-17 The (a) 50th, (b) 5th and (c) 95th percentiles of constrained recharge for the modelled area ............................................................................................................................. 136 Figure 5-18 The (a) 50th, (b) 5th and (c) 95th percentiles of constrained recharge for the Roper catchment ................................................................................................................................... 137 Figure 5-19 Water bodies in the Roper catchment identified (from Digital Earth Australia) showing the proportion of time that water bodies are inundated (from Water Observations from Space) ................................................................................................................................. 140 Figure 5-20 Excess water across the Roper catchment .............................................................. 141 Figure 5-21 Mataranka area showing (a) the location in relation to aquifers and streams with the Digital Earth Australia water bodies classified by proportion of time inundated according to Water Observations from Space, (b) mean annual excess water, (c) mean CMRSET actual evapotranspiration (AET) for February with the colour scale maximum at the mean FAO56 potential evapotranspiration, the same is also shown for (d) June, (e) August and (f) October 143 Figure 5-22 Maiwok and Flying Fox creeks over the Dook Creek Formation showing (a) the location in relation to aquifers and streams with the DEA water bodies classified by proportion of time inundated according to Water Observations from Space, (b) mean annual excess water, (c) mean CMRSET actual evapotranspiration for February with the colour scale maximum at the mean FAO56 potential evapotranspiration, and the same for (d) June, (e) August and (f) October ....................................................................................................................................... 145 Figure 5-23 Mainoru River over the Dook Creek Formation showing (a) the location in relation to aquifers and streams with the DEA water bodies classified by proportion of time inundated according to Water Observations from Space, (b) mean annual excess water, (c) mean CMRSET actual evapotranspiration (AET) for February with the colour scale maximum at the mean FAO56 potential evapotranspiration, and the same for (d) June, (e) August and (f) October ... 146 Figure 5-24 Wilton River over the Dook Creek Formation showing (a) the location in relation to aquifers and streams with the DEA water bodies classified by proportion of time inundated according to Water Observations from Space, (b) mean annual excess water, (c) mean CMRSET actual evapotranspiration (AET) for February with the colour scale maximum at the mean FAO56 potential evapotranspiration, and the same for (d) June, (e) August and (f) October ... 147 Figure 5-25 Guyuyu Creek and Goyder River over the Dook Creek Formation showing (a) the location in relation to aquifers and streams with the DEA water bodies classified by proportion of time inundated according to Water Observations from Space, (b) mean annual excess water, (c) mean CMRSET actual evapotranspiration (AET) for February with the colour scale maximum at the mean FAO56 potential evapotranspiration, and the same for (d) June, (e) August and (f) October ....................................................................................................................................... 149 Figure 5-26 Ngukurr over the Dook Creek Formation showing (a) the location in relation to aquifers and streams with the Digital Earth Australia water bodies classified by proportion of time inundated according to Water Observations from Space, (b) mean annual excess water, (c) mean CMRSET actual evapotranspiration (AET) for February with the colour scale maximum at the mean FAO56 potential evapotranspiration, and the same for (d) June, (e) August and (f) October ....................................................................................................................................... 150 Figure 5-27 Potential discharge zones to the streams overlying the outcropping area of the Dook Creek Formation Note: coloured areas are exaggerated in size so that they can be seen at this scale. ..................................................................................................................................... 151 Figure 5-28 Time series of actual evapotranspiration for potential discharge zones on Dook Creek Formation (DCF) ................................................................................................................ 152 Figure 5-29 (a) Mean October actual evapotranspiration (AET) for the known discharge area around Mataranka and (b) areas remaining with a high AET after excluding areas with a low October AET ................................................................................................................................ 153 Figure 5-30 (a) Coefficient of variation (CV) of October actual evapotranspiration (AET) in areas remaining after excluding areas of low October AET overlying the Dook Creek Formation (outside the Roper catchment) and (b) areas remaining after excluding areas with a high CV of October AET ................................................................................................................................ 154 Figure 5-31 Areas of potential groundwater discharge across the Roper catchment ............... 156 Figure 6-1 Locations of four regional-scale hydrogeological cross-sections traversing the Cambrian Limestone Aquifer in the Daly, Wiso and Georgina basins ........................................ 161 Figure 6-2 Hydrogeological cross-section A–B traversing from north-west of Mataranka in the Daly Basin to the south-east in the Georgina Basin due east of Daly Waters ............................ 162 Figure 6-3 Hydrogeological cross-section C–D traversing from south-west in the Wiso Basin to the north-east in the Daly Basin south-east of Mataranka ........................................................ 163 Figure 6-4 Hydrogeological cross-section E–F traversing from south-west in the northern Wiso Basin to the north-east in the southern Daly Basin east of Larrimah ........................................ 164 Figure 6-5 Hydrogeological cross-section G–H traversing from south-west in the northern Wiso Basin to the north-east in the northern Georgina Basin north east of Daly Waters .................. 164 Figure 6-6 Locations of three local-scale hydrogeological cross-sections traversing the regional groundwater discharge zone of the Cambrian Limestone Aquifer in the Daly Basin around Mataranka ................................................................................................................................... 165 Figure 6-7 Hydrogeological cross-section 1 traversing from west around the Birdum Creek fault to east in the Daly Basin across the regional groundwater discharge zone ............................... 166 Figure 6-8 Hydrogeological cross-section 2 traversing from west around the edge of the Jinduckin Formation to the east/south-east in the Daly Basin across the regional groundwater discharge zone, the Roper River and Elsey Creek ....................................................................... 167 Figure 6-9 Hydrogeological cross-section 3 traversing from north around the Roper River and Rainbow Spring to south in the Daly Basin across Elsey Creek .................................................. 167 Figure 6-10 Location of the north-west to south-east hydrogeological cross-sections traversing the region around Bulman, Weemol and surrounding outstations ........................................... 168 Figure 6-11 North-west to south-east cross-section traversing the Dook Creek Formation ..... 169 Figure 6-12 Interpolated spatial map of the depth to the top of the Cambrian Limestone Aquifer below the land surface ................................................................................................................ 171 Figure 6-13 Interpolated spatial map of the depth to the top of the Dook Creek Aquifer ........ 172 Figure 6-14 Potentiometric (reduced standing water level, RSWL) surface for the Cambrian Limestone Aquifer in the Roper catchment, including an extended buffer region of up to 100 km across the aquifer outside the southern catchment boundary .................................................. 174 Figure 6-15 Piezometric cross-sections through the potentiometric (reduced standing water level, RSWL) surface of the Cambrian Limestone Aquifer .......................................................... 175 Figure 6-16 Interpolated depth to groundwater (standing water level, SWL) surface for the Cambrian Limestone Aquifer in the Roper catchment, including an extended buffer region up to 100 km around the catchment ................................................................................................... 176 Figure 6-17 Modelled depth to groundwater (standing water level, SWL) in the Dook Creek Aquifer relative to the land surface ............................................................................................ 177 Figure 6-18 Spatial distribution of groundwater sampling sites across the Cambrian Limestone Aquifer and adjacent aquifers between Daly Waters and Mataranka ....................................... 179 Figure 6-19 Spatial distribution of groundwater sampling sites across the Dook Creek Formation and adjacent aquifers between Flying Fox Creek and the Goyder River .................................... 180 Figure 6-20 Spatial distribution of water quality types for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units .................................................................................................................. 181 Figure 6-21 Piper diagram showing major ion composition for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units .................................................................................................................. 182 Figure 6-22 Major ion ratio plots for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifer hosted in adjacent hydrogeological units ....... 184 Figure 6-23 Spatial distribution of water quality types for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units .................................................................................................................. 186 Figure 6-24 Piper diagram showing major ion composition for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units .................................................................................................................. 187 Figure 6-25 Major ion ratio plots for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and adjacent hydrogeological units ................................ 189 Figure 6-26 Stable hydrogen and oxygen isotope composition for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units compared to rainfall ................................................................. 191 Figure 6-27 Spatial distribution of deuterium (δ2H) isotopic composition for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units ............................................................................................... 192 Figure 6-28 Stable hydrogen and oxygen isotope composition for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and adjacent hydrogeological units compared to rainfall ................................................................................ 193 Figure 6-29 Spatial distribution of deuterium (δ2H) isotopic composition for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units ............................................................................................... 194 Figure 6-30 Strontium isotope composition relative to strontium concentration for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units ................................................................................... 195 Figure 6-31 Spatial distribution of strontium isotopic composition for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units ................................................................................................... 196 Figure 6-32 Strontium isotope composition relative to strontium concentration for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and adjacent hydrogeological units .................................................................................................................. 197 Figure 6-33 Spatial distribution of the strontium isotopic composition for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units ............................................................................................... 197 Figure 6-34 Tritium concentration for groundwater and spring samples from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units versus (a) distance from the edge of the Cambrian siltstone (Anthony Lagoon Formation – see Figure 6-18) and (b) depth below the watertable ........................................................................................... 198 Figure 6-35 Spatial distribution of the tritium concentration for groundwater and spring samples from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units .................................................................................................................. 199 Figure 6-36 Spatial distribution of the tritium concentration for groundwater and spring samples from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units .................................................................................................................. 201 Figure 6-37 Tritium concentrations for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units versus (a) distance from the southern edge of the Dook Creek Formation – see Figure 6-19) and (b) depth below the watertable .................................................................................................. 202 Figure 6-38 Measured concentrations of (a) xenon versus krypton and (b) xenon versus neon in groundwater and spring samples collected from the Cambrian Limestone Aquifer and aquifers hosted in adjacent hydrogeological units ................................................................................... 203 Figure 6-39 Measured concentrations of (a) xenon versus krypton and (b) xenon versus neon in groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in the adjacent Mountain Valley Limestone .............................................. 204 Figure 6-40 Concentrations of various anthropogenic gas tracers in groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units versus distance from the edge of the Cambrian siltstone (Anthony Lagoon Formation – see Figure 6-18): (a) SF6, (b) H1301, (c) CFC-11 and (d) CFC-12 206 Figure 6-41 Concentrations of various anthropogenic gas tracers in groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units versus depth below the watertable: (a) SF6, (b) H1301, (c) CFC- 11 and (d) CFC-12 ........................................................................................................................ 207 Figure 6-42 Concentrations of various anthropogenic gas tracers in groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in the adjacent Mountain Valley Limestone versus depth below the watertable: (a) SF6, (b) H1301, (c) CFC-11 and (d) CFC-12................................................................................................................. 209 Figure 6-43 Measured concentrations of carbon isotopes (a) 14C versus distance from the edge of the Cambrian siltstone (Anthony Lagoon Formation – see Figure 6-18), (b) 14C versus depth below the watertable, (c) TDIC versus distance from the Cambrian Silestone (see Figure 6-18) and (d) 13C versus 14C in groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units ........................... 211 Figure 6-44 Measured tritium and 14C concentrations for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units .................................................................................................................. 212 Figure 6-45 Measured concentrations of carbon isotopes (a) 14C versus tritium (b) 14C versus total dissolved inorganic carbon (TDIC), (c) ) 13C versus 14C, and (d) 14C versus TDIC in groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in the adjacent Mountain Valley Limestone .............................................. 213 Figure 6-46 Measured helium (He) concentrations (a) helium-3 to helium-4 ratio versus neon to helium ratio and (b) helium versus 14C in groundwater and spring samples collected from aquifers hosted the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units ..................................................................................................................................................... 215 Figure 6-47 Measured helium (He) concentrations (a) helium-3 to helium-4 ratio versus neon (Ne) to helium ratio and (b) helium versus 14C in groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units .................................................................................................................. 216 Figure 6-48 Measured 3H versus predicted mean residence time for groundwater flow for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units ..................................................................... 218 Figure 6-49 Locations of hydrometric monitoring including discharge and salinity for end-of-dryseason surveys undertaken between 2003 and 2015 ................................................................ 220 Figure 6-50 Locations of hydrometric monitoring including discharge and salinity for end-of-dryseason surveys undertaken between 2016 and 2022 ................................................................ 221 Figure 6-51 Spatial distribution of surface water and spring sampling sites in the groundwater discharge zone for the Cambrian Limestone Aquifer in the Roper catchment .......................... 222 Figure 6-52 Longitudinal trends in environmental tracers along the upper Roper River and its tributaries during October 2023 (a) electrical conductivity (EC) and discharge measurements, (b) 2H, (c) 18O, (d) 222Rn and (e) tritium ....................................................................................... 224 Figure 6-53 Stable hydrogen and oxygen isotopic composition of (a) Mataranka rainfall versus surface water in the upper Roper River from this assessment and (b) Roper River surface water samples and spring samples from this assessment versus 2019 to 2021 spring and regional groundwater ............................................................................................................................... 228 Figure 6-54 Location of the study sites where tree water source field investigations were conducted ................................................................................................................................... 229 Figure 6-55 Stable hydrogen and oxygen isotopic composition of rainfall versus soil, groundwater and stem water from sampling across six field sites during Phase I of field investigations .............................................................................................................................. 230 Figure 6-56 Stable hydrogen and oxygen isotopic composition of rainfall versus soil, groundwater and stem water from sampling across six field sites during Phase I of field investigations .............................................................................................................................. 231 Figure 6-57 Modelled mean annual recharge (a) across the entire DR2 FEFLOW model domain for the Cambrian Limestone Aquifer and (b) for the spatial extent of the CLA within the Roper catchment ................................................................................................................................... 232 Figure 6-58 Modelled mean annual recharge (a) across the entire DC2 FEFLOW model domain for the Dook Creek Aquifer and (b) for the spatial extent of the DCA within the Roper catchment ..................................................................................................................................................... 233 Figure 6-59 Modelled drawdown in groundwater level in the Cambrian Limestone Aquifer under (a) Scenario B70, historical climate and 70 GL/y of hypothetical future development for the year 2070 (i.e. 50 years) and (b) Scenario B′70, historical climate and 70 GL/y of hypothetical future development at quasi-equilibrium conditions for the year 2346 (i.e. 436 years) ...................... 235 Figure 6-60 Modelled drawdown in groundwater level in the Cambrian Limestone Aquifer under (a) Scenario Ddry70 at 2070 (i.e. after 50 years) (b) Scenario D′dry70 at 2346 (i.e. after 436 years), (c) Scenario Dmid70 at 2070 (i.e. after 50 years), (d) Scenario D′mid70 at 2346 (i.e. after 436 years), (e) Scenario Dwet70 at 2070 (i.e. after 50 years) and (f) Scenario D′wet70 at 2346 (i.e. after 436 years) .................................................................................................................... 236 Figure 6-61 Modelled drawdown in groundwater level in the Dook Creek Aquifer (CLA) under (a) Scenario B12, historical climate and 12 GL/y of hypothetical future development for the year 2070 (i.e. after 50 years) and (b) Scenario B′12, historical climate and 12 GL/y of hypothetical future development at quasi-equilibrium conditions (i.e. after 436 years) ............................... 237 Figure 6-62 Modelled drawdown in groundwater level in the Dook Creek Aquifer under (a) Scenario Ddry12 at 2070 (i.e. after 50 years), (b) Scenario D′dry12 at 2346 (i.e. after 436 years), (c) Scenario Dmid12 at 2070 (i.e. after 50 years), (d) Scenario D′mid12 at 2346 (i.e. after 436 years), (e) Scenario Dwet12 at 2070 (i.e. after 50 years) and (f) Scenario D′wet12 at 2346 (i.e. after 436 years) ........................................................................................................................... 238 Figure 7-1 Conceptual hydrogeological block model of the Cambrian Limestone Aquifer and aquifers hosted in adjacent hydrogeological units ..................................................................... 249 Figure 7-2 Two-dimensional conceptual schematic of the variability in the functionality of bores installed in the Cambrian Limestone Aquifer system and the system’s variability in karstic features and their interconnectivity ........................................................................................... 250 Figure 7-3 Conceptual schematic representing six different types of karstic sinkhole features (sometimes referred to as dolines) often found in different karst landscapes .......................... 252 Figure 7-4 Two-dimensional hydrogeological conceptual model of groundwater flow processes in the Dook Creek Aquifer ........................................................................................................... 257 Figure 7-5 Hydrogeological units with potential for future groundwater resource development ..................................................................................................................................................... 264 Tables Table 4-1 Location and target aquifers for new bores installed in Elsey National Park ............... 66 Table 4-2 Summary of modelling scenarios A, B, C and D for both aquifers using 4 × 109 years historical climate and combinations of current and hypothetical future groundwater development ................................................................................................................................. 95 Table 4-3 Fifty-year time periods used for climate sequences ..................................................... 97 Table 4-4 Summary of modelling scenarios A, B, C and D for both aquifers using the moving 11 × 50-year window of climate sequences and combinations of current and hypothetical future groundwater development ........................................................................................................... 97 Table 4-5 Gauging sites and the corresponding river branch name ........................................... 103 Table 5-1 Summary of groundwater-level data for bores in aquifers hosted in different hydrogeological units of the Roper catchment .......................................................................... 106 Table 5-2 Summary of groundwater salinity data as TDS for bores in aquifers hosted in different hydrogeological units of the Roper catchment .......................................................................... 113 Table 5-3 Summary of groundwater bore yield data for bores in aquifers hosted in different hydrogeological units of the Roper catchment .......................................................................... 115 Table 5-4 Hydraulic properties estimated from two-bore constant-rate discharge pumping tests performed in the Cambrian Limestone Aquifer across the Roper catchment ........................... 118 Table 5-5 Hydraulic properties estimated from two-bore constant-rate discharge pumping tests performed in the Dook Creek Aquifer across the Roper catchment .......................................... 119 Table 5-6 Hydraulic properties estimated from two-bore constant-rate discharge pumping tests performed on bores in aquifers hosted in the Nathan Group across the Roper catchment ..... 120 Table 5-7 Hydraulic properties estimated from two-bore constant-rate discharge pumping tests performed on bores in aquifers hosted in the Bessie Creek Sandstone across the Roper catchment ................................................................................................................................... 121 Table 5-8 Hydraulic properties estimated from two-bore constant-rate discharge pumping tests performed in different hydrogeological units hosting minor aquifers across the Roper catchment ................................................................................................................................... 122 Table 5-9 Estimated amplitudes (in mm) of M2 and S2 Earth tide constituents in 31 measured groundwater hydraulic head time series and resulting interpreted aquifer confinement status ..................................................................................................................................................... 127 Table 5-10 Mean recharge rates over each of the major aquifers. The 50th percentile is outside the brackets and the 5th and 95th percentiles are within the brackets, giving a range for the uncertainty .................................................................................................................................. 138 Table 5-11 Mean recharge rates over the simplified surface geology classes. The 50th percentile is outside the brackets and the 5th and 95th percentiles are within the brackets, giving a range for the uncertainty ...................................................................................................................... 138 Table 5-12 Calculation of volume of groundwater discharge from the Dook Creek Formation due to evapotranspiration ................................................................................................................. 153 Table 5-13 Summary of areas identified as potential groundwater discharge areas................. 157 Table 6-1 Summary of details for new monitoring bores installed in Elsey National Park ........ 160 Table 6-2 Carbonate speciation and mineral saturation indices from geochemical modelling for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units ..................................................................... 185 Table 6-3 Carbonate speciation and mineral saturation indices from geochemical modelling for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units .............................................................. 190 Table 6-4 Summary of measured dissolved noble gas concentrations in surface water and spring samples ....................................................................................................................................... 225 Table 6-5 Mean modelled groundwater levels at six reporting locations within the Cambrian Limestone Aquifer for each future hypothetical groundwater development and future climate scenario compared to Scenario A′N at 2070 (i.e. after 50 years) ............................................... 240 Table 6-6 Mean modelled groundwater levels at five reporting locations within the Dook Creek Aquifer (DCA) for each future hypothetical groundwater development and future climate scenario compared to Scenario A′N at 2070 (i.e. after 50 years) ............................................... 241 Table 6-7 Change in mean modelled discharge to the Roper River (G9030013), Wilton River (G9030003) and Flying Fox Creek (G9030108) under current and future climate and development scenarios ............................................................................................................... 242 Table 7-1 Opportunity-level estimates of the potential scale of groundwater resource development opportunities in the Roper catchment ................................................................. 261 Appendix figures Apx Figure A.1-1 Bore log diagram for RN043046 installed in the Tindall Limestone ............... 292 Apx Figure A.1-2 Bore log diagram for RN043049 installed in the Tindall Limestone ............... 293 Apx Figure A.1-3 Bore log diagram for RN043045 installed in the Antrim Plateau Volcanics.... 294 Apx Figure A.1-4 Bore log diagram for RN043047 installed in the Antrim Plateau Volcanics.... 295 Appendix tables Apx Table A.2-1 Summary of details for bores sampled during this assessment ....................... 296 Apx Table A.2-2 Summary of measured field parameters at each groundwater bore sampled in this assessment ........................................................................................................................... 297 Apx Table A.2-3 Summary of measured laboratory chemical analyses for groundwater samples collected at each bore site .......................................................................................................... 299 Apx Table A.2-4 Summary of measured environmental tracer analyses for groundwater samples collected at each bore site .......................................................................................................... 300 Apx Table A.2-5 Summary of measured noble gas analyses for groundwater samples collected at each bore site .......................................................................................................................... 302 Part I Introduction and overview 1 Introduction 1.1 Overview In northern Australia, future planning, investment and management of water resources relies on new or improved information across large areas which can lack data. Future industries, communities and the environment depend on groundwater as a reliable year-round water source where surface water resources are absent or ephemeral. Identifying opportunities and risks for future groundwater development requires understanding the coincidence of, and relationships between, suitable land, geology, productive and spatially extensive groundwater systems, existing water users and groundwater-dependent ecosystems (GDEs). In addition, evaluating potential hydrological impacts on existing users and ecologically and/or culturally important GDEs requires understanding the influences on hydrological processes of the distinctive seasonality and large inter-annual variability in rainfall and potential evaporation. In many places across the catchment of the Roper River, rivers and creeks are intermittent or ephemeral; groundwater is often the most attractive water resource depending on its availability at a particular location. Groundwater is the major water resource for towns and communities, the pastoral industry and irrigated agriculture but has only been characterised where current demand exists (Department of Environment Parks and Water Security, 2018). The regional aquifer hosted in the interconnected Tindall Limestone and its lithological and geochronological equivalents (Montejinni Limestone and Gum Ridge Formation) – referred to hereafter as the Cambrian Limestone Aquifer (CLA) – has been partially characterised around Mataranka where licensed entitlements are most prominent (Bruwer and Tickell, 2015). Other productive carbonate and sandstone aquifers, such as those hosted in the Dook Creek Formation (DCF) and Nathan Group, respectively, have been the subject of localised investigations for community water supplies at Beswick, Bulman and Ngukurr (Knapton, 2009c; Sumner, 2008; Yin Foo, 1991). As the nature of each aquifer varies across the catchment with spatial changes in geology and hydrogeology, it is important to collect new information to provide a more robust spatial understanding of the groundwater flow systems they host. Key characteristics include their: (i) extent and depth in the subsurface, (ii) thickness, (iii) storage and flow, and (iv) connectivity with other aquifers and aquitards, and with aquatic and terrestrial GDEs. GDEs include groundwater-fed rivers and creeks, riparian vegetation, wetlands, waterholes, springs, spring-fed vegetation, stygofauna and the marine environment. Consequently, identifying and evaluating promising aquifers with potential for future groundwater resource development and providing information to guide groundwater planning, investment and management involves: • identifying and characterising the depth and spatial extent of target aquifers • conceptualising the nature of their flow systems • estimating water balances • estimating potential extractable volumes from the aquifers and the associated drawdown in groundwater level or changes in flow over time and distance at various locations throughout each aquifer. This information will facilitate groundwater resource planning by describing current baseline conditions and predicted future changes to groundwater levels and flow in the aquifer at particular locations. Planning decisions require value judgments as to what is an acceptable impact (in terms of change to the natural flow regime) on receptors such as environmental assets or existing users at a given location. These decisions can be complex and require considerable input from a wide range of stakeholders, particularly government regulators and communities. In this assessment, information on the spatial changes to groundwater levels or flow for variations in specified extraction rates has been provided for target aquifers, but no expert judgment has been provided on what drawdown or flow is acceptable at a specified location. The information provided can be used by stakeholders to inform discussions on future groundwater resource planning, investment and management. 1.2 Aims of the groundwater hydrology activity The purpose of the groundwater hydrology activity in the Roper River Water Resource Assessment (RoWRA) is to provide an overview of groundwater resources in the Roper catchment within the context of identifying and characterising potential opportunities for, and risks associated with, groundwater resource development in the most promising aquifers at an intermediate to regional scale. As part of this regional-scale assessment, CSIRO partnered with the Northern Territory (NT) Department of Environment, Parks and Water Security (DEPWS), Charles Darwin University and CloudGMS. Key questions the assessment seeks to address include: • What types of aquifers exist and what is the nature of the flow systems they host? • What important attributes help identify aquifers in the catchment deemed promising for future groundwater resource development, and how do the attributes vary spatially? • Is it possible to estimate aquifer recharge (within a range) and are these ranges reasonable when considering rainfall, runoff, evapotranspiration, hydraulic properties and groundwater levels? • For the most promising aquifers identified, what component of their water balance can be estimated with some level of confidence and what components remain uncertain? • What are the ranges in potential extractable volumes for these most promising aquifers? • What are the main risks to future groundwater development across the catchment? 1.3 Summary of previous hydrogeological investigations Before the 1980s, groundwater resources of the Roper catchment were poorly understood in comparison to surface water resources and the regional geology. The earliest reported water resource investigations occurred in the late 1950s to early 1970s; these focused on surface water availability, particularly permanent waterholes as sources of water for agriculture, horticulture and community water supplies (Augustine, 1958; Brooks and Kneebone, 1963; Kneebone, 1960). From the late 1970s to the late 1990s, the search for reliable year-round water supplies for the Indigenous communities of Barunga (Bamyili), Beswick, Bulman, Ngukurr (formerly Roper Mission) and Minyerri involved numerous hydrogeological investigations that generated some of the first information on groundwater resources (Baker et al., 1992; Karp, 1984a; Tyson and Yin Foo, 1992; Verma and Rowston, 1992; Yin Foo, 1991). The key outputs from this work included new information on aquifer geometry and extents, groundwater quality and aquifer hydraulic properties. The findings highlighted the groundwater potential of multiple hydrogeological units hosting aquifers including the Cretaceous Sandstone; Dook Creek and Knuckey formations; Yalwarra Volcanics; and Bessie Creek, Hodgson and Mount Birch sandstones. Since the late 1990s, localised hydrogeological investigations have continued around the locations of Indigenous communities in efforts to ensure their water security. At the regional scale, other investigations have centred on the prospective groundwater resources of the CLA. Given the large spatial extent of the CLA, initial efforts in the early 2000s focused on water resource surveys which collated all available hydrological and hydrogeological information across the Sturt Plateau, Katherine and central and south-west Arnhem Land regions (George, 2001; Yin Foo, 2002; Zaar and Tien, 2003). Outputs from these investigations included datasets for: (i) surface water flows and quality, (ii) groundwater bores yields, (iii) aquifer types, (iv) aquifer hydraulic properties, (v) aquifer extents, and (vi) groundwater quality and chemical composition. These water resource characteristics were integrated to produce the first water resource maps providing information on potential water availability for the Katherine, Sturt Plateau and south-west Arnhem Land regions (George and Jolly, 2002; Yin Foo and Matthews, 2002). By the mid-2000s, demand increased for groundwater from the CLA for stock, domestic and agricultural purposes. Investigations shifted to further system characterisation and an initial quantification of the groundwater balance to better manage the prospective resource. This included better characterisation of the aquifer extent and geometry and an initial estimation of the recharge and discharge components of the groundwater balance (Jolly et al., 2004; Knapton, 2004). This work further emphasised the hydraulic connection between the CLA and the upper Roper River and its tributaries, resulting in further work to better map and characterise the sources of discharge from the aquifer at key springs and tributaries of the river (Karp, 2008). The work by Karp (2008) highlighted the need for future groundwater resource planning for the CLA, particularly in accounting for water availability for existing users across the aquifer and the importance of the aquifer for supporting GDEs. In 2009, the Gulf Water Study, partly funded by the Australian Government’s Water Smart Australia Program, undertook a regional hydrology study encompassing most of the Roper catchment. The key aims of the Gulf Water Study were to better understand the environmental and cultural importance of water resources in the Gulf region (Zaar, 2009a). Key outputs from this work included: • four water resource maps (Robinson–Calvert, Dunmarra–Hodgson, Limmen Bight–McArthur, and Roper River regions) integrating all available surface and groundwater data to provide information on potential water availability (Zaar et al., 2009a; 2009b; 2009c; 2009d) • four water resource reports for the corresponding regions (Fulton and Zaar, 2009; Zaar, 2009a; 2009b; 2009c) • an integrated surface–groundwater model (Knapton, 2009a; 2009b; 2009c) encompassing the groundwater resources of the entire CLA across the interconnected Daly, Georgina and Wiso basins, the Dook Creek Aquifer (DCA) in the McArthur Basin, and surface water resources of 12 sub-catchments within the Roper catchment. The model was initially developed to provide essential information to underpin water resource planning decisions for water management zones across different parts of the CLA. Since the late 2000s, ongoing water resource investigations have refined and improved characterisation, simulation and quantification of both surface water resources of the Roper River and groundwater resources of the CLA. Key investigations include hydrographic and water quality surveys of the Roper River spanning 2013 to 2018 (Kerle and Cruickshank, 2014; Schult, 2018; Schult and Novak, 2017; Wagenaar and Tickell, 2013; Waugh and Kerle, 2014). Hydrogeological assessments have further improved hydrogeological information for the CLA, including aquifer geometry, aquifer hydraulic properties and groundwater quality across the southern part of the Daly Basin (Bruwer and Tickell, 2015), the northern part of the Georgina Basin (Tickell and Bruwer, 2018) and the Beetaloo Sub-basin of the McArthur Basin (Evans et al., 2020; Fulton and Knapton, 2015). Collectively, these hydrological and hydrogeological investigations have contributed to the upgrade and refinement of the integrated surface–groundwater model used to underpin water resource planning decisions in the Roper catchment (Knapton, 2020), in particular, in the proposed Mataranka Tindall Limestone Aquifer Water Allocation Plan (MTLAWAP) area in the Roper catchment (Department of Environment and Natural Resources, 2017). 1.4 Current licensed water entitlements 1.4.1 Surface water entitlements Licensed surface water entitlements are sparse across the Roper catchment, which occupies the eastern part of the Daly Roper Beetaloo Water Control District (DRBWCD). The Roper catchment (and the DRBWCD) also contain the proposed MTLAWAP area and the current Georgina Wiso Water Allocation Plan (GWWAP) area (Figure 1-1). Four surface water licences have been granted for a combination of public water supplies and cultural and industrial uses; all fall outside proposed water allocation planning areas (Department of Environment Parks and Water Security, 2018). The largest entitlement (80 ML/year) is for public water supply at Barunga (Figure 1-1). The water is sourced from Beswick Creek which is fed by water discharging from Bamyili Spring. The next largest entitlement (26 ML/year) is from the Roper River for industrial purposes. Minor entitlements of less than 8 ML/year from the Roper River have been granted for cultural and industrial purposes. 1.4.2 Groundwater entitlements Licensed groundwater entitlements have been granted across the central and south-eastern parts of the Roper catchment, most prominently around Mataranka. Most of the groundwater entitlements (32 GL/year) are for water sourced from the CLA in the south-west of the catchment (Department of Environment Parks and Water Security, 2018). These licensed entitlements all occur within the proposed MTLAWAP with the exception of one licence for public water supply at Daly Waters (Figure 1-1). Only very minor entitlements (about 1 GL/year) are sourced elsewhere, mostly from localised fractured and weathered rock aquifers hosted in the Roper Group. The purpose of the majority of these entitlements is irrigated agriculture (31.1 GL/year). Licensed entitlements totalling 354 ML/year have also been granted for public water supplies at Mataranka, Jilkminggan, Larrimah and Daly Waters. The remaining entitlements (309 ML/year) have been granted for industrial purposes, including tourist accommodation, and council and cement operations. Other licensed entitlements come from aquifers hosting intermediate- to local-scale groundwater systems inside the DRBWCD but outside the proposed MTLAWAP and current GWWAP areas (Figure 1-1). Licences have been granted for public water supply at Beswick (190 ML/year), Barunga (280 ML/year) and Minyerri (150 ML/year). Groundwater for Barunga is sourced from localised aquifers hosted in Cretaceous sandstone. Groundwater for Beswick is sourced from the intermediate-scale DCA hosted in the Dook Creek Formation. Groundwater for Minyerri is sourced from a localised aquifer hosted in the Bessie Creek Sandstone. Figure 1-1 Location, type and volume of annual licensed surface water and groundwater entitlements Data sources: Licensed entitlement data sourced from Department of Environment Parks and Water Security (2018); Daly Roper Beetaloo Water Control District (DRBWCD) sourced from Department of Environment, Parks and Water Security (2019b); Water allocation plan areas sourced from Department of Environment, Parks and Water Security (2019a) A map of a large area with green and brown lines Description automatically generated 1.5 Promising aquifers for targeted field, desktop and modelling investigations Previous hydrogeological investigations in the Roper catchment and current licensed water entitlements overwhelmingly demonstrate that the CLA offers the most promising regional-scale opportunities for future groundwater resource development. In addition, previous investigations also highlight the potential of the DCA to offer promising opportunities at an intermediate scale, but further investigation is required to provide more information on the resource. These aquifers are spatially extensive, host good-quality (i.e. low-salinity) water and existing drilling and pumping test data indicate their potential to yield sufficient water for irrigation. They are also promising because: • they occur across large areas at economically viable depths for drilling to intersect the resource • the depth to groundwater across these areas also makes it viable to pump the resource to the surface • their large spatial extents provide opportunities for development where soils are suitable for agricultural intensification away from existing users and GDEs. In terms of potential economic constraints on future groundwater resource development, a maximum drilling depth of 300 m below ground level (mBGL) and a maximum depth to groundwater of 100 mBGL was adopted for this assessment. These thresholds were determined on the basis that the drilling, construction and installation of bores and the associated pumping costs would likely be prohibitively expensive for all but the highest value irrigated crops. Accordingly, the CLA and DCA were the primary focus for targeted field, desktop and modelling investigations in this assessment. Some components of the work also focused on further characterising the nature of hydraulic connectivity with adjacent aquifers, and groundwater– surface water interactions between the aquifers, surface water features and GDEs. Note, however, that other local-scale aquifers hosting groundwater resources may present equally attractive future water supply options in localised parts of the Roper catchment, such as aquifers hosted in multiple hydrogeological units of the Nathan Group (Yalwarra Volcanics, Knuckey Formation and Mount Birch Sandstone) near Ngukurr in the east of the catchment. These aquifers are high yielding and host good-quality (low-salinity) water, but they have a limited extent and require further localised investigations to provide confidence in the resource. Nevertheless, some of the data for these aquifers has been collated, interpreted and summarised in parts of this report. 1.6 Report structure This report describes the methods and tools used to document the regional assessment of hydrogeological systems of the Roper catchment and the targeted field, desktop and modelling investigations of the CLA and DCA. The report is structured as follows: • Chapter 2 outlines the geography, demography, climate, geology and hydrogeology of the study area. • Chapters 3 and 4 detail the methods and tools used to conduct the regional hydrogeological assessment (water levels, bore yields, water quality, recharge estimation and discharge mapping) and targeted field, desktop and modelling investigations, respectively. • Chapters 5 and 6 present and discuss the results of the regional assessment and targeted field, desktop and modelling investigations, respectively. • Chapter 7 discusses the opportunities and constraints associated with further development of groundwater resources from the CLA and DCA, and from other aquifers in the catchment. • Chapter 8 states the conclusions of this assessment and potential options for future work. 2 Study area 2.1 Physiography and demography The Roper catchment has an area of approximately 77,400 km2 and it flows into the western Gulf of Carpentaria. It exhibits four different physiographic regions described in Plumb and Roberts (1992) including: (i) the Cretaceous Tableland (referred to as the Sturt Plateau) in the south-west, (ii) the Gulf Fall in the centre, (iii) Wilton River Plateau in the north, and (iv) the Coastal Plain in the east (Figure 2-1). The Sturt Plateau is a relatively flat to undulating plain with low rounded crests and isolated ridges comprised mostly of deeply weathered Cretaceous claystone and siltstone sediments and lateritic surfaces (Yin Foo, 2002). There is a very subtle topographic gradient which is highest in the south around Daly Waters to slightly lower in the north around the lower reaches of Elsey Creek and Mataranka. Most rainfall across the plateau is evapotranspired though some rainfall also recharges groundwater which eventually discharges to a combination of the Mataranka Springs Complex and the upper Roper River and its tributaries. The remaining rainfall is evapotranspired, and only minor ephemeral flows occur in Western, Birdum and Elsey creeks and the Strangways River and its tributaries in the late wet season (Hughes et al., 2023; Knapton, 2020; Yin Foo, 2002). The dissected Gulf Fall physiographic region occupies most of the central part of the catchment. It extends from the north-eastern edge of the Sturt Plateau around Mataranka to the estuarine coastal plain in the east, east of Ngukurr. The region exhibits a complex landscape of residual rises and hills, strike ridges, mesas and plateaux and intervening fluvial valleys. The rocks and sediments of the Gulf Fall region are mostly comprised of sandstones from the Proterozoic eon, mudstones and siltstones intruded by igneous dolerite dykes and patches of surficial colluvium and alluvium (Abbott et al., 2001). Rainfall across the Gulf Fall region is mostly partitioned between runoff and evapotranspiration, and generally there is little groundwater recharge to the weathered and fractured rock. Run off accumulates as streamflow in the mid to lower reaches of the Roper River and its tributaries (Wilton, Mainoru and Hodgson rivers, Flying Fox and Maiwok creeks) (Hughes et al., 2023; Knapton, 2009a; 2009c). The Wilton River Plateau in the northern part of the catchment is a level to gently undulating sandstone plateau comprised mostly of Proterozoic and Cretaceous sandstone and patches of surficial colluvium and alluvium (Sweet et al., 1999). Most rainfall across the plateau is evapotranspired or becomes groundwater recharge which eventually discharges to the upper reaches of Flying Fox Creek and the Mainoru and Wilton rivers (Knapton, 2009a; 2009c). The Coastal Plain extends east of the Gulf Fall region as an extensive area of salt flats, tidal flats and mangroves comprised mostly of surficial alluvium and colluvium underlain by Proterozoic sandstone (Abbott et al., 2001). Land tenure across the Roper catchment is generally either pastoral lease (37,749 km2) or freehold Aboriginal Land Trusts (35,261 km2). Other significant tenures include Crown leases (3999 km2) and vacant Crown land (229 km2) (Department of Environment Parks and Water Security, 2021). The Roper catchment is sparsely populated but includes the communities of Mataranka, Larrimah, Daly Waters, Beswick, Barunga (Bamyili), Minyerri and Ngukurr as well as several smaller Indigenous communities, outstations and roadhouses. Most community localities have a population of fewer than 500 people; exceptions include Beswick, Minyerri and Ngukurr. Ngukurr, with a population of approximately 1000 people, is the largest community (Australian Bureau of Statistics, 2021). Land use mapping (Burgess et al., 2015; Department of Environment Parks and Water Security, 2017) shows the predominant land uses in the catchment are grazing of native vegetation (35,329 km2) and managed resource protection and other minimal uses (34,677 km2). Other substantial land uses include marshes and wetlands (3273 km2), nature conservation (3159 km2), estuaries (547 km2), transport and communications (228 km2), and grazing of modified pastures (76 km2). Irrigated plantation forests (8.5 km2; sandalwood) and irrigated seasonal and perennial horticulture and cropping (11.9 km2; watermelon, mango production in the Mataranka area) and mining (0.8 km2) account for a small proportion of the total area. Figure 2-1 The Roper catchment study area showing the Roper River and its tributaries and the spatial extent of the different physiographic regions Data source: Physiographic regions sourced from Plumb and Roberts (1992) Location map \\FS1-CBR.nexus.csiro.au\{lw-rowra}\work\3_Land_suitability\3_Roper\1_GIS\1_Map_docs\1_Exports\LL-R-507_location_v4-10_10-8.png For more information on this figure please contact CSIRO on enquiries@csiro.au 2.2 Climate The following climate summary, unless stated otherwise, comes from a companion technical report on climate data characterisation of the Roper catchment by McJannet et al. (2023). The climate of the Roper catchment is semi-arid monsoonal with distinctive wet and dry seasons. The catchment is characterised by these distinctive seasons due to its location in the Australian summer monsoon belt. The wet season (September to April) is characterised by higher temperatures and higher humidity and is when the catchment experiences most of the annual rainfall. The dry season (May to August), in comparison, is cooler, has lower humidity and little to no rainfall. The median annual rainfall, averaged over the Roper catchment for the 109-year historical period (1 September 1910 to 31 August 2019) is 770 mm. There is a clear north to south rainfall gradient with annual rainfall highest (~1000 mm) in the northern part of the catchment and lowest (~600 mm) in the most southerly part the catchment (Figure 2-2a). The median annual areal potential evaporation (as calculated using Morton’s areal wet potential) averaged over the Roper catchment for the 109-year historical period is 1900 mm. Similar to the median annual rainfall, there is a clear gradient in areal potential evaporation but in the reverse direction: the highest (~1900 mm) annual areal potential evaporation occurs in the south and the lowest (~1820 mm) occurs in the far north. Figure 2-2 Historical (a) median annual rainfall and (b) median annual potential evaporation across the Roper catchment Approximately 96% of rain falls in the Roper catchment during the wet-season months (1 November to 30 April) see Figure 2-3. The highest monthly rainfall totals typically occur during the mid-wet season (January, February and March). Tropical cyclones and tropical lows contribute a considerable proportion of total annual rainfall, but the actual amount is highly variable from one year to the next. A map of different colors Description automatically generated Areal potential evaporation in the Roper catchment exceeds 1800 mm in most years (Figure 2-4). Evaporation is high all year round but exhibits a strong seasonal pattern, ranging from about 200 mm/month during the October ‘build-up’ to about 120 mm/month during the middle of the dry season (June) (Figure 2-4). Figure 2-3 Historical monthly rainfall (left) and time series of annual rainfall (right) in the Roper catchment at Ngukurr, Mataranka, Larrimah and Bulman ‘A range’ is the 10th to 90th percentile monthly rainfall. The solid blue line in the right column is the 10-year running mean. "\\fs1-cbr\{lw-rowra}\work\1_Climate\3_Roper\2_Reporting\plot\climate_report\annual_monthly_rainfall_range_4_station.png" For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au Figure 2-4 Historical monthly potential evaporation (left) and time series of annual potential evaporation (right) in the Roper catchment at Ngukurr, Mataranka, Larrimah and Bulman ‘A range’ is the 10th to 90th percentile monthly rainfall. Note: The ‘A mean’ line is directly under the ‘A median’ line in these figures. The solid blue line in the right column is the 10-year running mean. "\\fs1-cbr\{lw-rowra}\work\1_Climate\3_Roper\2_Reporting\plot\climate_report\annual_monthly_pet_range_4_station.png" For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au 2.3 Geology 2.3.1 Geological provinces Five major geological provinces occur in the Roper catchment (Figure 2-5). From oldest to youngest these are: (i) the McArthur Basin, which underlies the centre, north and east of the catchment, (ii) the three interconnected Daly, Wiso and Georgina basins, which overlie the McArthur Basin in the south to south-west of the catchment, and (iii) the geological Carpentaria Basin, which overlies most of the Daly, Wiso and Georgina basins in the south to south-west of the catchment and also part of the McArthur Basin in the west (Figure 2-5). The Paleo-to Mesoproterozoic McArthur Basin is an intracratonic sedimentary basin comprised mostly of sandstones and siltstones that in some places have been intruded by minor igneous rocks (Ahmad et al., 2013). The McArthur Basin extends well beyond the Roper catchment; it is exposed over an area of about 180,000 km2 across the NT. It hosts stacked sedimentary sequences of rocks that collectively have a maximum thickness of about 10 km and is bound to the north and east by the Arafura Basin and Gulf of Carpentaria, respectively (Ahmad et al., 2013). To the south it is bound by the Tomkinson Province and to the west by the Pine Creek Orogen. In the Roper catchment, the McArthur Basin is undulating with isolated ranges of quartzite and igneous rocks dissected by river valleys. Topographic features include the Shadforth and McKay hills in the north; the Strangman and Bold ranges and Collara Mountains in the centre; and the Hartz, Downers and High Black ranges south of the Roper River. The Neoproterozoic to Palaeozoic Daly, Wiso and Georgina basins are interconnected intracratonic sedimentary basins comprised of sandstone, siltstone, limestone and dolostone (Kruse et al., 2013; Kruse and Munson, 2013a; 2013b). In the Roper catchment, the basins mostly unconformably overlie the early Cambrian Antrim Plateau Volcanics (APV) of the Kalkarindji Igneous Province (Glass et al., 2013). The Roper catchment coincides with only small parts of the interconnected basins (southern Daly Basin and northern Wiso and Georgina basins) which collectively have a total area of about 460,000 km2. The basins all vary in thickness, generally between 80 and 300 m in the Roper catchment, though the Georgina Basin south of Daly Waters can be up to about 400 m thick. The basins rarely outcrop in the catchment as they are obscured by the overlying rocks and sediments of the Mesozoic geological Carpentaria Basin (Munson et al., 2013) The Mesozoic geological Carpentaria Basin is an intracratonic sedimentary basin comprised mostly of claystone, siltstone and sandstone. It also extends well beyond the Roper catchment with a total area of about 200,000 km2 over the central to northern part of the NT (Munson et al., 2013). Its thickness is highly variable, with a maximum thickness in the catchment of about 100 m in the northern Georgina Basin, but it is much thinner (<20 m) in the Wilton River Plateau region of the McArthur Basin (Munson et al., 2013). Figure 2-5 Major geological provinces of the Roper catchment Data source: geological provinces adapted from Raymond (2018) 2.3.2 Surface geology The surface geology of the Roper catchment is presented in Figure 2-6. The oldest rocks in the catchment are the Proterozoic rocks of the McArthur Basin which partially outcrop in the central to northern part of the catchment. These rocks are comprised of stacked and/or repeated thick A map of a large area with rivers and roads Description automatically generated sequences of sandstone and siltstone beds occasionally intruded by dolerite (Abbott et al., 2001; Sweet et al., 1999). They were deposited in a series of basins extending across the area and then folded, faulted and intruded by igneous rocks to form mountain ranges. Towards the end of the Proterozoic, the ranges had been eroded down to a level not far above that of the current topography. During the Cambrian period, there was widespread extrusion of basalt lava, which was followed by deposition of limestones and dolostones (Glass et al., 2013; Kruse and Munson, 2013a). The Cambrian strata only occur in the south-western portion of the Roper catchment where the limestones, dolostones and basalt are mostly obscured beneath a veneer of surficial Cretaceous and Cenozoic rocks and sediments. Only minor outcrops of the limestone and basalt occur around Mataranka and south of Minyerri (Figure 2-6). Erosion recommenced after the Cambrian period and continued to the mid-Cretaceous when subsidence and high global sea levels resulted in deposition of a thin succession of Cretaceous shallow marine sandstone, siltstone and claystone across the Roper catchment (Munson et al., 2013). The most prominent outcrops of the Cretaceous rocks occur in the south-west and northwest of the catchment (Figure 2-6). The present landscape has been produced by warping and dissection of a series of erosion surfaces formed during several cycles of erosion that started in the late Cretaceous and ended in the mid-Cenozoic. During this time, stable crustal conditions and subaerial exposure led to patchy erosion of the Cretaceous rocks and prolonged subaerial weathering of the remaining Cretaceous and Proterozoic rocks, resulting in the formation of weathered profiles and associated laterite capping. Between the mid-Cenozoic and the present day, there has been gentle uplift and warping of the various surfaces and their weathered capping. Continued erosion has led to the emergence of the present-day landscape; this involved the removal of Cretaceous strata from most of the region and the etching out of structures in the underlying Proterozoic rocks of the McArthur Basin. Floodplain and coastal sediments composed of sand, silt and clay were deposited on the margins of modern drainage systems and the coastline, respectively. Figure 2-6 Surface geology of the Roper catchment Data source: surface geology sourced from Raymond et al. (2012) 2.3.3 Regional geology A simplified regional geology of the Roper catchment is presented in Figure 2-7. The surficial veneer of Cretaceous to Cenozoic cover presented in Figure 2-6 in Section 2.3.2 has been removed to show the main subsurface geological units. This has been done to highlight the location and extent of units beneath the catchment important for either hosting or controlling the occurrence of groundwater resources. The regional geology is entirely related to the rocks within the major A map of a large area Description automatically generated geological provinces of the catchment (Figure 2-5). The subsurface of the central to northern part of the catchment is dominated by the Proterozoic Roper, Katherine River and Mount Rigg geological groups. The subsurface of the south-west of the catchment is dominated by the Cambrian Tindall Limestone and its lithological and geochronological equivalents (Montejinni Limestone and Gum Ridge Formation). Small portions of Proterozoic Bukalara Sandstone and Nathan Group and Cambrian APV make up the remainder of the catchment (Figure 2-7). The Roper Group occupies the largest portion of the subsurface beneath the catchment. It exhibits a series of stacked sandstone and siltstone units that are folded and faulted and dip in different directions in the subsurface as well as being intruded in places by the Derim Derim Dolerite. Regional faults across this part of the catchment include the Bulman and Showell Creek faults and those in the Flying Fox fault zone (Abbott et al., 2001). The most prominent geological units of the Roper Group across this part of the catchment occurring beneath Cretaceous to Cenozoic cover include the (i) Limmen Sandstone, (ii) Mainoru Formation, (iii) Crawford Formation, (iv) Jalboi Formation, (v) Hodgson Sandstone, and (vi) Bessie Creek Sandstone. These units are comprised of fine to coarse quartzose sandstone with the exception of the Mainoru and Jalboi formations which are mostly comprised of siltstone and mudstone. Some of these geological units, such as the Limmen, Hodgson and Bessie Creek sandstones, host localised and isolated low-yielding aquifers which supply an important source of groundwater, as does the Derim Derim Dolerite which in places intrudes through these units (see Section 2.4). The Tindall Limestone and its lithological and geochronological equivalents (Montejinni Limestone and Gum Ridge Formation) cover a large portion of the subsurface in the south-western part of the catchment (Figure 2-7). These equivalent units are mostly flat lying to gently dipping and are mostly comprised of partially dolomitised limestone with some mudstone and siltstone (Geoscience Australia and Australian Stratigraphy Commission, 2017). Collectively, the units extend for tens to a few hundred kilometres to the west, south and east of the Roper catchment (Figure 2-8). Large parts of the limestone rocks have been eroded by dissolution, forming karsts including sinkholes, caves, caverns and springs which host the CLA, the largest and most important groundwater system beneath the catchment. Very small portions of the limestone unit to the west of Mataranka and east of Daly Waters are overlain by the siltstone and mudstone of the Jinduckin Formation and its equivalent Anthony Lagoon Formation, respectively. Almost all of the limestone and siltstone units are obscured by a veneer of overlying Cretaceous sandstone, siltstone and claystone; the exception is a small area near Mataranka. Near Larrimah, the APV, which is mostly comprised of tholeiitic basalt, is most prominent west of the Birdum Creek fault with minor occurrences also south of Barunga and Minyerri (Geoscience Australia and Australian Stratigraphy Commission, 2017) (see Figure 2-7). Figure 2-7 Simplified regional geology of the Roper catchment To show the spatial extent of key regional geological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The extent of the surficial Cretaceous to Quaternary rocks and sediments is shown on the lower right inset. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008). Geological faults data source: Department of Industry, Tourism and Trade (2010) The Proterozoic Mount Rigg Group is most prominent around Bulman where it outcrops forming the southern part of the Arnhem Shelf and Wilton River Plateau in the north-west of the catchment (Figure 2-7). It also outcrops in small patches near Barunga and in the form of its geological equivalent, the Nathan Group, around Ngukurr. The reminder of the Mount Rigg Group is obscured by a veneer of overlying Cretaceous rocks and sediments. Key geological units hosted in the Mount Rigg Group include the Dook Creek Formation (DCF) and Bone Creek Sandstone. The DCF is mostly comprised of dolostone, chert, sandstone and siltstone. West of the Central Arnhem Road, (Figure 2-7) the DCF gently dips in the subsurface, but east of the Central Arnhem Road the formation dips steeply beneath the Roper Group (Abbott et al., 2001; Rawlings, 2015). Similar to the Cambrian limestone units, the DCF dolostone have been eroded by dissolution, forming karsts including sinkholes, caves, caverns and springs. The dolostone hosts the second-largest groundwater system beneath the catchment, the Dook Creek Aquifer (DCA), and is a major water source for the communities of Beswick and Bulman. The DCA hosted in the Dook Creek Formation of the Mount Rigg Group, extends for tens of kilometres to the north-east of the Roper catchment (Figure 2-8). The Nathan Group hosts the Walmudga and Knuckey formations, Yalwarra Volcanics and Mount Birch Sandstone (Abbott et al., 2001). The Nathan Group only outcrops around Ngukurr where it is heavily faulted and folded and dips in the subsurface in various directions beneath the Roper Group (Abbott et al., 2001; Sumner, 2008). The Knuckey Formation, Yalwarra Volcanics and Mount Birch Sandstone units host moderately yielding aquifers of good-quality water which is partly used for Ngukurr’s water supply (see Section 2.4). The Proterozoic Bukalara Sandstone and equivalents exhibit small outcrops around Barunga and south of Minyerri. The sandstone is generally flat lying to gently dipping beneath the APV (Ahmad et al., 2013). The Bukalara Sandstone is mostly comprised of fine to coarse sandstone with minor shale and conglomerate (Geoscience Australia and Australian Stratigraphy Commission, 2017; Karp, 1984b). Fractured and weathered components of the sandstone host localised and isolated low-yielding aquifers (see Section 2.4). Figure 2-8 Simplified regional geology of the Roper catchment including the entire spatial extent of the Mount Rigg Group of the McArthur Basin and the Tindall Limestone and equivalents of the Daly, Wiso and Georgina basins To show the spatial extent of key regional geological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The extent of the surficial Cretaceous to Quaternary rocks and sediments is shown on the lower right inset. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008) Geological faults data source: Department of Industry, Tourism and Trade (2010) 2.3.4 Depth to basement ‐‐ The NT SEEBASE is a depth-to-basement structural model that provides a modelled depth to basement based on the shape and depth of major basement structures and the base of sedimentary sections. Depth to basement is modelled using a variety of data sources including magnetic and gravity data, digital elevation model data and existing geological maps and publications (Northern Territory Geological Survey and Geognostics Australia Pty Ltd, 2021). Spatial changes in depth to basement have been shown to highlight the major structural features beneath the Roper catchment; these include major deposition features, basement highs and major shallow to surficial faults (Figure 2-9). The major deposition features and structural highs beneath the catchment are associated with the greater McArthur Basin and its various subbasins. Beneath the entire catchment there is significant variability in depth influenced by major structural controls. In the south-west of the catchment to the north-west of Larrimah is one of the depocenters of the Beetaloo Sub-basin of the greater McArthur Basin known as the Gorrie depocenter (Northern Territory Geological Survey and Geognostics Australia Pty Ltd, 2021). The maximum depth to basement at this location is approximately 9.2 km where the depocenter hosts a significantly thick sequence of Paleo-to Mesoproterozoic rocks overlain by a thinner sequence of Cambrian to Cenozoic rocks and sediments. A major structural high exists between Mataranka, Larrimah and Daly Waters associated with two major fault zones to the east of the Birdum Creek fault. There is significant variability in depth to basement in a north–south direction with a minimum depth of approximately 0.2 km between Mataranka and Larrimah, and depths varying between approximately 4 to 5 km around Daly Waters in the south. Due north of Mataranka, a shallow depocenter known as the Maiwok Sub-basin exists just south of the Flying Fox fault zone. The maximum depth to basement at this location is approximately 6.2 km. Around Minyerri, depth to basement varies between approximately 2 and 3 km except for the depocenter of the Saint Vidgeon Sub-basin to the north-east of Minyerri, where the maximum depth is approximately 6 km (Northern Territory Geological Survey and Geognostics Australia Pty Ltd, 2021). Across the northern part of the McArthur Basin, depths to basement vary from as little as approximately 0.3 km on the Arnhem Shelf north of Bulman to approximately 3.5 km depth south-west of Bulman. The only other depocenter beneath the catchment occurs in the northeast around the Phelp River fault zone which exhibits significant structure and variability. Depths to basement around the fault zone range from approximately 3.5 km just north-east of Ngukurr to approximately 9.3 km due east of Ngukurr toward the catchment boundary on the Gulf of Carpentaria. 24 | Hydrogeological assessment of the Roper catchment Figure 2-9 Spatial changes in modelled depth to basement beneath the Roper catchment Data source: NT SEEBASE and GIS Digital Information Package – Northern Territory Geological Survey and Geognostics Australia Pty Ltd (2021) 2.4 Hydrogeology 2.4.1 Aquifer types Within the Roper catchment the availability and quality of groundwater resources are heavily influenced by the physical characteristics of rocks of the major geological units across the region (Figure 2-10). Aquifer types include: (i) intermediate- to regional-scale weathered fractured or karstic limestones, dolostones and sandstones of the Daly, Wiso and Georgina basins, and Mount Rigg and Nathan groups, (ii) localised fractured and weathered sandstone, siltstone and mudstone rocks of the Roper River and Katherine River groups, siltstones of the Daly, Wiso and Georgina basins, and the igneous rocks of the Kalkarindji Igneous Province, and (iii) surficial rocks and sediments including Cretaceous sandstone of the geological Carpentaria Basin and unconsolidated Quaternary regolith and alluvium. As shown in Figure 2-10, the major rock types (fractured, weathered, sedimentary, karstic) regardless of their spatial extent host aquifer systems of different scale ranging from minor and localised up to intermediate and regional scale. In this assessment, major aquifer systems are those that contain regional- and intermediate-scale groundwater systems, which have adequate storage volumes (i.e. gigalitres) that could potentially yield water at a sufficient rate (>10 L/second) and have water of sufficient quality (<1000 mg/L total dissolved solids (TDS)) for a range of irrigated cropping. Minor aquifers are those that contain local-scale groundwater systems with lower storage (i.e. megalitres). The yields from minor aquifers are variable but are often low (<5 L/second). Minor aquifers also often have variable water quality ranging from fresh (~500 mg/L TDS) to brackish (~3000 mg/L TDS) and are mostly suitable for stock and domestic use only. Aquifer systems can be interconnected across multiple hydrogeological units which are summarised in the sections below. Figure 2-10 Aquifer types occurring within and beneath the Roper catchment Note: this map does not show localised surficial aquifers hosted in sediments and rocks including unconsolidated Quaternary regolith and alluvium and consolidated Cretaceous sandstone. Aquifer type data source: Department of Environment Parks and Water Security (2008) 2.4.2 Hydrogeological units The regional distribution of hydrogeological units within and beneath the Roper catchment is shown in Figure 2-11. The rocks and sediments of these geological units host a diverse range of aquifers that vary in extent, storage and productivity. Major aquifers in the Roper catchment are found in: (i) the Cambrian limestone of the interconnected Daly, Wiso and Georgina basins beneath the south-west of the catchment, and (ii) the Proterozoic dolostones and sandstones of the McArthur Basin in the north and east of the catchment. Minor localised aquifers are found in: (i) the Proterozoic sedimentary and igneous rocks of the McArthur Basin beneath large parts of the centre of the catchment, (ii) the Cambrian basalt of the Kalkarindji Igneous Province that mostly occurs beneath the Cambrian limestone in the south-west of the catchment, (iii) the Cambrian siltstones of the Daly, Wiso and Georgina basins, and (iv) the surficial Cretaceous sandstones of the geological Carpentaria Basin and Cenozoic alluvium. Limestone aquifers The Cambrian limestone aquifers of the interconnected Daly, Wiso and Georgina basins outcrop around Mataranka and extend beneath the south to south-west of the Roper catchment, occupying about 36% of the total catchment area (Figure 2-11). They consist of three equivalent hydrogeological units (Tindall Limestone, Montejinni Limestone and Gum Ridge Formation) which extend to the west, and far south and east of the Roper catchment, hosting the interconnected CLA (Figure 2-12). The combined total area of the interconnected regional-scale CLA beyond the Roper catchment is approximately 460,000 km2. Within the south-west of the Roper catchment, the CLA underlies an area of approximately 27,500 km2. The CLA is a highly productive but complex regional-scale aquifer; it is the largest groundwater resource beneath and extending beyond the Roper catchment. The complexity of the system arises from: (i) the variability of and interconnectivity between fractures, fissures and karsts across the spatial extent of the three equivalent hydrogeological units, (ii) the temporal and spatial variability in recharge processes and fluxes to the aquifer, and (iii) the structural geological controls on groundwater flow. Groundwater resources from the aquifer have mostly been developed for groundwater-based irrigated agriculture but also for community water supplies at Mataranka, Larrimah and Daly Waters. For more information on current groundwater entitlements, see Section 1.4.1. The heterogeneity in the amount of weathering by dissolution of the limestone and the nature and interconnectivity of karstic features across large areas affects permeability, groundwater flow and bore yields across the aquifer (Bruwer and Tickell, 2015; Jolly et al., 2004). Recharge across the aquifer is spatially variable and occurs via a variety of processes either directly in the aquifer outcrop or where it is unconfined beneath overlying Cretaceous strata. Within the Roper catchment, the aquifer outcrops around Mataranka (see lower right inset of Figure 2-11), but the remainder is mostly unconfined beneath a veneer of overlying Cretaceous strata that varies in both thickness and permeability. However, the aquifer is confined in places by the Cambrian siltstone of the Jinduckin and Anthony Lagoon formations (mostly outside the Roper catchment), which influences spatial variability in recharge to the aquifer (Bruwer and Tickell, 2015; Knapton, 2020). Recharge processes include a combination of: (i) localised preferential infiltration of rainfall and streamflow via sinkholes directly in the aquifer outcrop or where surficial features such as ephemeral streams or waterholes are incised through overlying Cretaceous strata, or (ii) via broad diffuse vertical infiltration of rainfall through the overlying sandstone, siltstone and claystone where the aquifer is unconfined (Knapton, 2020). Recharge to the CLA beneath the Roper catchment has been estimated in many previous investigations using a variety of methods (chloride mass balance (CMB), hydrograph analyses, water balance modelling, interpretation of environmental tracer concentrations in groundwater) that have different assumptions and account for varied spatial and temporal scales. Estimates from these previous investigations relevant to the CLA within the Roper catchment ranged between 1 and 130 mm/year (Bruwer and Tickell, 2015; Crosbie and Rachakonda, 2021; Deslandes et al., 2019; Jolly et al., 2004; Knapton, 2020; Tickell and Bruwer, 2018; Yin Foo and Matthews, 2001). Groundwater flow in the aquifer system is complex due to variability in the amount and connectivity of karstic features across the aquifer and spatial and temporal variability in recharge and discharge across large areas. At a local scale, groundwater flow can occur via preferential flow in secondary porosity features (sinkholes, caves and caverns), but across the aquifer extent, regional flow occurs via the interconnection of these karstic features similar to porous media (Knapton, 2020). Regional groundwater flow in the aquifer is generally from south to north. In the Georgina Basin, regional flow sometimes referred to as the ‘Georgina Flow Path’ is from south to north into the Daly Basin and toward the Roper River (Bruwer and Tickell, 2015; Tickell and Bruwer, 2018). In the Wiso Basin, regional flow sometimes referred to as the ‘Flora Flow Path’ is from the south to the north-west, towards the Flora River just west of the catchment boundary (Knapton, 2020). Some intermediate- to local-scale flow in the Daly Basin comes from in the west around the Murrawal Plateau near the western catchment boundary east of Katherine and flows east toward the Roper River but flow also comes from the north from the aquifers northern margin flowing south towards the Roper River (Knapton, 2020). The aquifer is partly absent where the underlying igneous basalt of the APV forms a high just west of Larrimah. This partly interrupts the aquifer’s continuous spatial extent and influences the directions of regional groundwater flow in both the Wiso and Georgina basins (Yin Foo and Matthews, 2001). The aquifer in the Roper catchment discharges via a combination of: (i) diffuse seepage to streams (Roper Creek, upper Roper River, Waterhouse River and Elsey Creek), (ii) localised spring discharge (Bitter, Rainbow, Botanic Walk and Fig Tree springs) including a few instream springs in the bed of the upper Roper River and its tributaries, (iii) transpiration from riparian and spring-fed vegetation in and around Elsey National Park, and (iv) extraction of groundwater. The sources of groundwater discharge to the upper Roper River and its tributaries and springs comes from a combination of regional discharge from the Georgina and Daly basins in the south, intermediate to local discharge from the Daly Basin to the west and north, and localised discharge from the aquifer outcrop around Mataranka (Department of Environment and Natural Resources, 2017; Karp, 2008; Lamontagne et al., 2021). Mean dry-season discharge from the CLA to the upper Roper River and its tributaries varies from year to year but on average ranges from 1 to 4 m3/second based on historical streamflow gauge data (Knapton, 2020). Previous spot measurements of the end-of-dryseason spring flow during the past two decades at Bitter, Rainbow and Fig Tree springs indicate they had a mean flow of about 0.8, 0.4 and 0.05 m3/second, respectively (Karp, 2008; Wagenaar and Tickell, 2013; Waugh and Kerle, 2014). Outside the Roper catchment, discharge from the aquifer occurs to the Katherine, Flora, Daly and Douglas rivers via diffuse seepage and localised spring discharge, transpiration from riparian and spring-fed vegetation and groundwater extraction. Bore yields are highly variable given the complex nature of the karstic aquifer but often range from 15 to 45 L/second from appropriately constructed production bores. The productive (highyielding) part of the limestone aquifer occurs in the weathered, fractured and karstic parts of the aquifer. Pumping tests have been conducted on many bores across the CLA within the Roper catchment, and aquifer hydraulic properties have been calculated from these tests. Previous estimates of transmissivity (T) and hydraulic conductivity (K) been summarised in Knapton (2020) with transmissivity ranging between 66 and 10,000 m2/day and hydraulic conductivity ranging between 1 and 71 m/day (Bruwer and Tickell, 2015; Tickell and Bruwer, 2018). Specific yield (Sy) for the aquifers has been estimated to range from 0.01 to 0.06 and a storage coefficient (S) of approximately 0.000001 (Knapton, 2020). Groundwater is generally fresh with an electrical conductivity (EC) less than 750 μS/cm and has a calcium (Ca)–bicarbonate (HCO3) composition north of Mataranka in the Daly Basin (Bruwer and Tickell, 2015). South of Mataranka in the southern Daly Basin and northern Georgina Basin, the EC is on average higher, ranging between 1000 and 3000 μS/cm, and has a sodium (Na)–chloride (Cl) composition (Bruwer and Tickell, 2015; Jolly et al., 2004). In addition to stock and domestic use of groundwater, there is approximately 32 GL/year of licensed groundwater entitlements for the CLA, most of which is exclusively for irrigated agriculture (see Section 1.4.1). Figure 2-11 Simplified regional hydrogeology of the Roper catchment To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The extent of the surficial Cretaceous to Quaternary rocks and sediments is shown on the lower right inset. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008) Geological faults data source: Department of Industry, Tourism and Trade (2010) Spring data source: Department of Environment Parks and Water Security (2014c) Sinkhole data source: Department of Environment Parks and Water Security (2014b) Dolostone aquifers The Mesoproterozoic dolostones aquifers are hosted in hydrogeological units of the Mount Rigg and Nathan groups of the McArthur Basin. They predominantly occur beneath the northern and eastern parts of the Roper catchment (Figure 2-11). These dolostones host productive intermediate- to local-scale aquifers. The most significant dolostone aquifer is hosted in the Dook Creek Formation of the Mount Rigg Group in the north-east of the catchment. The aquifer outcrops and subcrops between Barunga and Bulman and to the north-east of the catchment boundary (Figure 2-11 and Figure 2-12). This aquifer system, referred to as the Dook Creek Aquifer (DCA), occupies an area of approximately 14,100 km2 within and beneath the Roper catchment (about 18% of the Roper catchment). Where the DCA extends to the north-east outside the Roper catchment it occupies an area of approximately 21,800 km2. Like the CLA, the DCA is a complex aquifer due to the variability and interconnectivity between secondary porosity features (fractures, fissures and karsts) across the spatial extent of the formation. The variability in the dissolution of the dolostone, and the size, type and interconnectivity between karstic features, affect permeabilities, groundwater flow and bore yields across the aquifer (Knapton, 2009c; Zaar and Tien, 2003). Where the aquifer is unconfined to the west of the Central Arnhem Road (see Figure 2-7), recharge is spatially variable and occurs via: (i) broad diffuse vertical infiltration of rainfall, (ii) in some places, localised preferential infiltration via sinkholes directly in the outcrop, or (iii) through the overlying Cretaceous strata (sandstone, siltstone and claystone). East of the Central Arnhem Road (see Figure 2-7) the aquifer is confined by the sandstone of the Roper Group, which influences the spatial variability in recharge and discharge to and from the aquifer (Knapton, 2009c; Zaar and Tien, 2003). Data on the DCA are sparse and there are limited recharge estimates for the aquifer. Water balance modelling by Knapton (2009c) and Williams et al. (2003) estimated mean annual recharge rates ranging between 20 and 90 mm/year, respectively. The heterogeneity in the amount and connectivity of karstic features across the aquifer and spatial and temporal variability in annual recharge and discharge have a significant influence on groundwater flow in the aquifer. Groundwater flow is generally in a north-easterly direction, though groundwater-level data for the aquifer are sparse. Flow occurs at a local scale (i.e. a few hundred metres) through sinkholes and karstic features as well as at an intermediate scale (i.e. a few kilometres to tens of kilometres) where karstic features are interconnected across large areas (Knapton, 2009c; Zaar and Tien, 2003). In the Roper catchment, the shallowest flow paths discharge via diffuse seepage to streams (Flying Fox Creek, Mainoru and Wilton rivers), transpiration from riparian and spring-fed vegetation, and extraction of groundwater. Discharge also occurs preferentially via springs including at Bodeidei, Lindsay, Top and Weemol springs (Knapton, 2009c; Zaar and Tien, 2003). North-east of the Roper catchment, discharge from the aquifer occurs via diffuse seepage to streams (Guyuyu Creek and the Goyder River), transpiration from riparian and spring-fed vegetation, extraction of groundwater, and localised discharge at springs such as Emu and Wurrkal springs (Knapton, 2009c). Bore yields are variable given the complex nature of the karstic aquifer but can range from 15 to 45 L/second from appropriately constructed production bores. The productive (high-yielding) part of the dolostone aquifer occurs in the weathered, fractured and karstic zone. Pumping tests have been conducted on localised parts of the unconfined aquifer where community water supplies have been developed at Beswick and Bulman. Previous estimates of transmissivity range from 700 to 6000 m2/day and a specific yield of approximately 0.02 (Verma and Rowston, 1992; Yin Foo, 1991). Groundwater quality data for the aquifer is sparse except for near Beswick and Bulman. Groundwater is generally fresh (EC <700 μS/cm) and has a Ca–HCO3 composition (Mann and Yin Foo, 1993; Verma and Rowston, 1992). There is only one licensed groundwater entitlement (of 190 ML/year) from the aquifer, which is for Beswick’s water supply (see Section 1.4.1). Sandstone aquifers The Proterozoic sandstones of the McArthur Basin are mostly located across large parts of the centre and north of the Roper catchment. They include rocks of the Nathan Group, Bukalara Sandstone and Roper Group (Figure 2-11). These aquifers host local-scale flow systems, which provide an important source of groundwater in places for community water supplies and stock and domestic use. The most productive of these aquifers within and beneath the catchment are hosted in the Nathan Group around Ngukurr. Nathan Group The Mesoproterozoic Nathan Group exhibits three hydrogeological units that host local-scale aquifers around Ngukurr (Figure 2-11) including the: (i) quartz sandstone of the basaltic Yalwarra Volcanics, (ii) dolomitic siltstone and sandstone of the Knuckey Formation, and (iii) quartz sandstone of the Mount Birch Sandstone (Abbott et al., 2001; Sumner, 2008). Recharge to the aquifers occurs via diffuse seepage of intense wet-season rainfall where the aquifers either outcrop or subcrop beneath surficial Cenozoic cover north of Ngukurr and south of the Collara Mountains (Abbott et al., 2001; Sumner, 2008). Some recharge also occurs via leakage from the Roper River where the river channel is incised into and traverses the outcrop of both the Mount Birch Sandstone and Yalwarra Volcanics (Sumner, 2008). Recharge estimates are sparse for the aquifers though Sumner (2008) estimated recharge to be approximately 20 mm/year. Discharge from the aquifers is inferred from a comparison of temporal logged groundwater-level data and streamflow data. Investigations by both Baker et al. (1992) and Sumner (2008) indicate a connection with the Roper River which recharges the aquifer in the wet season before the aquifers subsequently discharge back to the river in the dry season. No terrestrial springs have been identified associated with the aquifers, though it is likely that preferential discharge may occur to the river via fracture flow where the aquifer outcrops in the river bed (Sumner, 2008). A previous investigation of groundwater–surface water connectivity using environmental tracers by Cook (2003) also confirms a source of groundwater discharge to the Roper River just upstream of Ngukurr. Production bores associated with Ngukurr’s water supply exhibit individual bore yields often around 15 L/second with maximum yields of up to 30 L/second where production bores have been constructed and pump tested (Baker et al., 1992; Sumner, 2008). Interpretation of pumping test data on production bores installed for Ngukurr’s water supply has yielded previous estimates of aquifer hydraulic properties. Sumner (2008) summarised previous estimates of transmissivity for all three aquifers as ranging between 63 and 878 m2/day and storativity ranging between 0.001 and 0.0038. Groundwater quality data collated by Sumner (Sumner, 2008) indicates groundwater in the aquifers is fresh to brackish with an EC ranging between about 500 and 1000 μS/cm. Chemical analyses of the borefield water indicates the ionic composition of the groundwater is magnesium (Mg) to HCO3. There are currently no licensed water entitlements for Ngukurr’s water supply. Water is sourced mostly from the Nathan Group aquifers and is supplemented by a small amount of surface water from the Roper River. Water demand for the community was estimated in the early 1990s and late 2000s as ranging between 500 and 600 ML/year (Baker et al., 1992; Sumner, 2008). Bukalara Sandstone The Neoproterozoic Bukalara Sandstone hosts local-scale aquifers where it outcrops to the south of Minyerri (Figure 2-11). Information for the coarse sandstone aquifers is sparse. Recharge to the aquifer is inferred to occur via infiltration of intense wet-season rainfall directly in the aquifer outcrop where it is unconfined. Elsewhere the aquifer is confined by the overlying APV (Karp, 1984b). Discharge occurs via groundwater extraction and localised discharge from springs occurring at the contact between the sandstone and the basalt of the underlying APV (Fulton and Zaar, 2009). Indicative bore yields for the aquifers typically range from 0.3 to 5 L/second though few pumping tests have been undertaken (Fulton and Zaar, 2009). Two pumping tests indicate that aquifer transmissivity ranges from 18 to 79 m2/day, but there are no estimates of the range in storage coefficients (Evans et al., 2020). Water quality data for the aquifers are sparse but indicate that the salinity expressed as TDS ranges from fresh to brackish (73 to 1510 mg/L) (Fulton and Zaar, 2009). The ionic composition of groundwater is fairly mixed and ranges from Ca–Mg–HCO3 to Na to sulfate (SO4) to Cl (Evans et al., 2020). Limmen, Hodgson and Bessie Creek sandstones The Mesoproterozoic Limmen, Hodgson and Bessie Creek sandstones of the Roper Group host local-scale aquifers in places across the central part of the Roper catchment (Figure 2-11). Aquifers hosted in the Bessie Creek Sandstone around Minyerri have the most information as they have been the subject of multiple investigations for community water supplies at Minyerri (Jamieson and Pidsley, 1992; Karp, 1984a; 1984b; Verma, 1990). Limited bore yield and water quality information from either investigation or stock and domestic bores for the Hodgson and Limmen sandstones indicate their potential as aquifers, but information is sparse and has not been summarised in detail here (Verma, 1990). Recharge to the Bessie Creek Sandstone occurs where the aquifer subcrops beneath surficial Cenozoic cover south of Minyerri. The sandstone is folded and faulted and groundwater movement is controlled by secondary porosity features (fractures and joints) at contact zones or fault zones with the adjacent Velkerri and Corcoran formations (Jamieson and Pidsley, 1992). Groundwater flow is inferred to be highly heterogenous due to the fractured nature of the aquifers but is inferred to follow a subdued form of topography with discharge occurring via contact springs and groundwater extraction (Fulton and Zaar, 2009; Jamieson and Pidsley, 1992). Hydraulic properties estimated from pumping tests on bores installed in the aquifers indicate transmissivities of between 3 and 10 m2/day and a specific yield of about 0.02 to 0.03 (Jamieson and Pidsley, 1992). Bore yields from pumping tests indicate yields are variable and low, ranging between 0.1 and 4.0 L/second (Jamieson and Pidsley, 1992; Verma, 1990). Bore yields of between 2 and 4 L/second are often not sustainable for several hours as most of the groundwater storage occurs in the bore casing and surrounding fractures, which eventually desaturate (Jamieson and Pidsley, 1992; Verma, 1990). Salinity in the aquifers is generally fresh, with EC ranging between 45 and 530 μS/cm. However, the water can often be slightly acidic with the pH commonly observed in the 5.5 to 6.5 range (Jamieson and Pidsley, 1992; Verma, 1990). Figure 2-12 Entire spatial extent of the Cambrian Limestone Aquifer and Dook Creek Aquifer (Proterozoic dolostone) beneath and beyond the Roper catchment The lower right map inset shows the spatial extents of surficial Cretaceous to Quaternary regolith sediments in the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008) Geological faults data source: Department of Industry, Tourism and Trade (2010) Basalt aquifers The Cambrian APV of the Kalkarindji Igneous Province underlies most of the CLA except for near Larrimah in the south-west of the catchment; there it comes up close to the surface and is overlain by either unsaturated limestone or Cretaceous sediments (Figure 2-11) (Yin Foo, 2002). The APV is comprised mostly of basalt and basalt breccia, and the upper part of the basaltic unit hosts localised partial aquifers where it is fractured and weathered (Department of Environment Parks and Water Security, 2014a). Elsewhere the remainder of the basalt is a consolidated aquitard confining underlying aquifers hosted in Proterozoic sandstones (ELA, 2022; Evans et al., 2020; Yin Foo, 2002). Information for the basalt aquifers is limited to areas where it either subcrops beneath Cambrian limestone and Cretaceous cover or where it outcrops south of Barunga and south of Minyerri. Recharge to the APV occurs where the aquifer either outcrops or subcrops beneath thin sequences of Cretaceous sediments and Cambrian limestone. The APV also receives recharge from localised vertical leakage from the CLA where its upper parts are fractured or weathered and connected to the overlying CLA (ELA, 2022; Evans et al., 2020). Localised groundwater flow and storage is influenced by the amount of weathering and secondary porosity features such as fractures and joints and the degree to which these features are interconnected. Bore yields are highly variable but generally low (<2 L/second). Water quality is also variable, ranging in water type from Ca–HCO3 to Na–HCO3, and salinity ranges from fresh to saline (ELA, 2022; Evans et al., 2020). There is little information about discharge from the APV, but it is inferred to occur via: (i) groundwater extraction for stock and domestic use, (ii) evapotranspiration where groundwater levels are shallow (<5 mBGL), and (iii) localised spring discharge west of the Roper catchment around Top Springs in the catchment of the Victoria River (Randal, 1973). Siltstone aquifers The Cambrian siltstones of the connected Daly, Wiso and Georgina basins host three lithologically equivalent hydrogeological units: (i) the Jinduckin Formation, (ii) the Hooker Creek Formation, and (iii) the Anthony Lagoon Formation. These units are mostly comprised of siltstone, which generally confines the CLA, but where they consist of interbedded lenses of sandstone and dolostone they also host local-scale aquifers of limited extent, low permeability and low bore yield (Knapton, 2020). These aquifers are highly variable in composition and have a limited extent in the catchment and are therefore only briefly discussed (Figure 2-11). Surficial aquifers Surficial sediments and rocks that host minor local-scale aquifers include unconsolidated Quaternary alluvium, and consolidated Cretaceous sandstone of the geological Carpentaria Basin. Alluvium mostly occurs near the mouth of the Roper River and in association with minor parts of the rivers, creeks and their floodplains and channels throughout the catchment. However, these aquifers have limited extent, are poorly characterised, and has very sparse information on bore yields and water quality. Aquifers hosted in Cretaceous sandstone predominantly occur across the south and north-east of the catchment where they overlie the CLA and DCA, respectively. Aquifers hosted in the Cretaceous rocks are mostly comprised of sandstone. Individual bore yields can often be a few litres per second but are generally less than 4 L/second (Evans et al., 2020). Water quality for these aquifers is variable and can range from fresh to brackish (<1000 mg/L TDS). Recharge to these aquifers occurs via diffuse rainfall infiltration through overlying regolith. The main discharge mechanisms are bores extracting groundwater for stock and domestic use, evaporation (through the soil or plants) from shallow watertables and discharge to rivers and creeks. These aquifers offer little potential for future groundwater resource development beyond stock and domestic purposes. Unconsolidated alluvial sediments (i.e. sand, silt or clay transported and deposited at some stage by flowing surface water) are sparse across most of the Roper catchment. Occasionally, they are present in conjunction with the Roper River and its tributaries, their channels and floodplains. However, the largest occurrence of unconsolidated alluvial sediments occurs at the mouth of the Roper River in association with the Limmen Bight Tidal Wetlands (see Section 2.6). There are very few groundwater bores and little information associated with these sediments, but given their limited extent and thickness, and proximity to the coast, they appear to host little groundwater suitable for potential use. 2.5 Surface water hydrology The Roper catchment is 77,400 km2 in area and features flat, tidally affected coastal plains that extend 20 to 60 km inland. These plains typically lie at less than 10 metres above Australian Height Datum (mAHD) and are prone to seasonal flooding. The Roper River extends approximately 300 km inland from the river mouth with major tributaries (the Wilton and Hodgson rivers) entering the river mid-catchment from the north and south, respectively. In headwater areas situated in the north-western part of the Roper catchment, altitudes reach up to 420 mAHD. Tidal influence on streamflow is detectable as far upstream as Roper Bar (around 10 km downstream of gauge 9030250), around 130 km from the Roper mouth (Figure 2-13). Due to the difficulty of streamflow measurement in tidally affected rivers, the lowermost reliable stream gauge on the Roper River is at Red Rock, a further 10 km upstream of Roper Bar. Only five current gauges in the Roper catchment are suitable for analysis of the surface water flow (and a further five gauges provide historical data). Based on modelled runoff, the median and mean annual discharge from the Roper catchment are 4341 and 5560 GL, respectively. The flows in the streams of the Roper catchment are highly seasonal with only the Roper River itself being largely perennial. The northern tributaries generally flow for more than 50% of the time while the more arid southern tributaries flow less than 50% of the time (with a minimum of 13% at gauge 9030124 on Daly Waters Creek). Dry-season flows in the Roper River are sustained through spring flow from the Mataranka Thermal Pools. It has been estimated that a flow rate of 2.5 m3/second downstream from the springs at gauge 9030013 is required for the river to continue flowing into the estuary (Jolly et al., 2004). Figure 2-13 Streamflow observation data availability in the Roper catchment and median annual streamflow (50% exceedance) under Scenario A Figure source: Figure 2-4 in Hughes et al. (2023) A map of a river Description automatically generated 2.6 Water-dependent ecosystems The largely intact habitats and landscapes of the Roper catchment provide near-natural ecosystem services that support high biodiversity, recreational activities, tourism, traditional and commercial fisheries, and areas of agricultural production. Within the freshwater sections of the Roper catchment are extensive areas with high habitat values, including ephemeral and persistent rivers, wetlands, floodplains and groundwater-dependent ecosystems (GDEs) as captured in the Groundwater Dependent Ecosystems Atlas (Doody et al., 2017). GDEs include the Mataranka Thermal Pools, which are listed in the Directory of Important Wetlands in Australia (DIWA) and support baseflow to the upper Roper River and its tributaries as well as a range of spring-fed vegetation. For the marine and estuarine environments, the Roper River provides some of the largest flows into the western Gulf of Carpentaria, supporting extensive intertidal, estuarine and marine communities including those in the Limmen Bight Marine Park. The dynamic between wet and dry seasons provides both challenges and opportunities for biota (Warfe et al., 2011). During the dry season, river flows are reduced and streams in the catchment recede, many to isolated pools. However, in parts of the Roper catchment, persistence of surface water during the dry season is supported by spring discharge from aquifers including the CLA and DCA (Faulks, 2001). The Roper catchment includes two DIWA sites (Figure 2-14): the groundwater-fed Mataranka Thermal Pools and the coastal Limmen Bight (Port Roper) Tidal Wetlands System (Environment Australia, 2001). The Mataranka Thermal Pools DIWA site, which is located in 4 ha of Elsey National Park, is maintained by permanent springs (Department of Agriculture, 2019). Groundwater-dependent vegetation fringes the pools that are supported by persistent discharge from the CLA. The Limmen Bight (Port Roper) Tidal Wetlands System DIWA site is the secondlargest area of saline coastal flats in the NT (1848 km2 excluding subtidal seagrass areas) (Department of Agriculture, 2019). Limmen Bight forms a highly important habitat system of tidal wetlands (intertidal mud flats, saline coastal flats and estuaries), and while the whole site is tidal, it receives large volumes of freshwater inflows from the contributing catchments (Department of Agriculture, 2019). Stygofauna sometimes referred to as subterranean aquatic fauna or aquatic invertebrates have also been found at widely separated sites in the saturated zone of karstic parts of the CLA in the south-west of the Roper catchment (see Figure 2-11) (Rees et al., 2020). Little is currently known about the biodiversity and ecological integrity of Stygofauna in the CLA or their level of water dependency (Rees et al., 2020). Figure 2-14 Location of ecological assets related to groundwater including Directory of Important Wetlands in Australia (DIWA) sites, groundwater-dependent ecosystems identified in the GDE Atlas and mapped spring locations across the Roper catchment Data sources: nationally important wetlands – Environment Australia (2001); known and potential GDEs – Doody et al. (2017) Spring data source: Department of Environment Parks and Water Security (2014c) Part II Methods 3 Regional desktop assessment of the Roper catchment Groundwater investigations were conducted using several approaches and at two different scales. A regional assessment of the Roper catchment was performed as desktop analyses or modelling using historical hydrogeological information related to all aquifers in the catchment. The regional assessment was conducted to essentially compile spatial datasets of geology, aquifer properties, bore locations and construction, and key groundwater attributes (water levels, water quality and bore yield) for aquifers in all hydrogeological units across the catchment. This baseline information was used to provide a regional overview of the available hydrogeological data across the catchment, specifically to: • generate geology, hydrogeology and aquifer type maps across the catchment • evaluate the range in key groundwater attribute data (depth to groundwater, salinity, indicative bore yield) for all aquifers across the catchment and identify what it is currently being used for (stock, domestic, community water supply, industrial or irrigation uses) • derive spatial maps of the locations of pumping tests and time series groundwater-level data that can be used to assess the nature of existing aquifer hydraulic property data across the catchment and identify where further analyse are required • derive spatial estimates of groundwater recharge for all aquifers across the catchment • derive spatial maps of groundwater–surface water connectivity and groundwater discharge across the catchment • generate spatial datasets to underpin further targeted field (groundwater sampling and drilling) desktop (piezometric cross-sections, potentiometric surfaces, spatial maps of depth to aquifers and groundwater) and modelling investigations of the CLA and DCA described in detail in Section 4. 3.1 Geology, hydrogeology, and aquifer types Analysis was undertaken to assess publicly available databases for digital spatial datasets on surface and basement geology and point locations with lithological and/or stratigraphic logs to generate maps of geology, hydrogeology and aquifer type and to attribute aquifers to individual groundwater bores. Multiple datasets were collated from sources including: • the Geoscience Exploration and Mining Information Systems (GEMIS) online database (https://geoscience.nt.gov.au/gemis/ntgsjspui/community-list) for accessing geoscientific reports and data kept by the Northern Territory Geological Survey (NTGS) • Natural Resources Maps (NR Maps) online (https://nrmaps.nt.gov.au/nrmaps.html) web mapping tool for accessing and mapping natural resources data • the Northern Territory Government (NTG) Open Data Portal that contains datasets made available from NT Government agencies; and geoscientific data available from Geoscience Australia (https://www.ga.gov.au/data-pubs). Data and information from publicly available literature were also used where applicable. Spatial data were then used to generate the maps of surface geology, regional geology, regional hydrogeology and aquifer type presented in the figures in sections 2.3 and 2.4. Lithological and/or stratigraphic logs for all available groundwater bores, exploration and stratigraphic holes and petroleum wells drilled in the Roper catchment were also sourced from the NTG Open Data Portal (Department of Environment Parks and Water Security, 2014a; Department of Industry Tourism and Trade, 2000; 2010) and used to overlay these point locations on the digital geology datasets. Available lithological and/or stratigraphic logs were then used in conjunction with bore construction (i.e. screened interval or open hole section) to attribute an aquifer to each site with bore construction information available. Aquifers were assigned either using existing interpretation from the local drillers, geologists or hydrogeologists who generated the logs during drilling investigations or using interpretation from hydrogeologists in the project team where logs were available. Where no log was available from drilling records, the aquifer was assigned as unknown. For the CLA and DCA, the surface water catchment is not the groundwater flow divide, so groundwater bores were also attributed to these two aquifers outside the Roper catchment. The aquifer attribution spatial dataset was then used to: (i) map key groundwater data (water levels, water quality, and bore yield) by aquifer representing the major and minor aquifers across the catchment, (ii) attribute aquifers to time series water level and aquifer hydraulic property spatial datasets before further data analyses, (iii) generate aquifer-specific spatial datasets for undertaking further targeted field, desktop and modelling investigations of the CLA and DCA, as described in detail in Section 4. 3.2 Groundwater levels Depth to groundwater is an important attribute for assessing local and regional trends in water levels, identifying recharge and discharge areas, and understanding the spatial changes in depths from which water would have to be pumped to the surface for use. The primary source of static groundwater-level data for groundwater bores across the Roper catchment was the NTG Open Data Portal, which provides a direct link to the digital spatial groundwater database for the entire NT (Department of Environment Parks and Water Security, 2014a). In addition to digital data from the groundwater database, drilling records (hand-written, typed or digital) were also accessed via NR Maps (Department of Environment Parks and Water Security, 2021) to undertake the aquifer attributions described in Section 3.1. These drilling records also include observations of groundwater level, groundwater salinity and bore yield from testing undertaken when the bores were first constructed and installed. Some data from old drilling records have not been digitised and added to the territory-wide groundwater database. Where the database lacked groundwater- level data, observations were obtained from the drilling records. Static groundwater-level observations were then linked to the aquifer attribution dataset described in Section 3.1 and then mapped in classes in ArcGIS by aquifer where aquifer information was available. Chapter 3 Regional desktop assessment of the Roper catchment | 43 3.3 Groundwater salinity Groundwater salinity is an important attribute for assessing the suitability of groundwater for different uses (i.e. domestic use, stock water, irrigation water or industrial applications) and for identifying local and regional trends in spatial changes in salinity. Salinity data in the form of total dissolved solids (TDS) was collated from the territory-wide groundwater database accessed via the NTG Open Data Portal (Department of Environment Parks and Water Security, 2014a). Before mapping and evaluating the salinity data, the validity and quality of the analytical data were reviewed by calculating the charge balance error (CBE). Only water samples returning a CBE ±5% were used. The salinity data were then linked to the aquifer attribution dataset described in Section 3.1 and then symbolised and mapped in ArcGIS, by aquifer where aquifer information was available, in classes suitable for different water uses including stock water and irrigation. 3.4 Bore yields Bore yield data provide an indication of the influence the physical properties of the rocks and sediments of different aquifers have on controlling the storage and flow of groundwater at different locations. Note that long-term (12 to 48 hour) pumping tests on production bores provide the best indication of bore yield at a certain location in an aquifer. However, the majority of bore yield data in most locations is obtained by short-term (i.e. a few hours) discharge testing by air lifting or submersible pumps on small diameter (<150 mm) stock and investigation bores. In most cases, the bore construction and size of the air compressor or submersible pump limit the yield from these bores. Nevertheless, the limited testing does provide a useful indication of aquifer potential. Bore yield data were collated from the territory-wide groundwater database accessed via the NTG Open Data Portal (Department of Environment Parks and Water Security, 2014a). In addition to digital data from the groundwater database, drilling records (hand-written, typed or digital) were also accessed via NR Maps (Department of Environment Parks and Water Security, 2021) to undertake the aquifer attributions described in Section 3.1. Drilling records were reviewed for bore yield data, and where the territory-wide database did not have a bore yield observation for a specific bore site, data from the drilling records was used where available. Bore yield data were then linked to the aquifer attribution dataset described in Section 3.1 and then symbolised and mapped in different classes in ArcGIS by aquifer where aquifer information was available. 3.5 Aquifer hydraulic properties 3.5.1 Pumping test reanalyses Appropriate estimates of subsurface hydraulic conditions and properties are required to reliably predict both current states (e.g. water balances) and future states (e.g. predicted drawdown or streamflow depletion) of groundwater flow systems. Hydraulic conditions include aquifer confinement status (e.g. confined, leaky, unconfined), dimension of flow (ranging from linear to radial to spherical) and flow domain type (e.g. single or dual domain). Hydraulic properties include transmission and storage capabilities (described by transmissivity and storage coefficient parameters, respectively), primarily of aquifers and, less commonly, of aquitards. Constant-rate discharge pumping tests provide the most direct and scale-appropriate means of estimating subsurface hydraulic conditions and properties. Hydraulic conditions can be identified from qualitative assessment of measured time–drawdown responses and their temporal derivatives, which are compared to established diagnostic reference plots (Gringarten, 2008; Renard, 2009). Hydraulic properties are estimated by matching modelled solutions to measured time–drawdown responses. Confounding processes, including borehole storage and well skin effects, can also be identified from measured time–drawdown responses. In the NT, locations of historical pumping tests were first collated by Nguyen and Tickell (2014). This dataset has been periodically updated since first publication. Historically, the estimation of subsurface hydraulic properties from these tests was limited to aquifer transmissivity and, to a much lesser extent, storativity. This was largely due to ubiquitous use of the Cooper and Jacob (1946) approximation of the Theis (1935) solution. A limited number of publications have collated, digitised and reinterpreted selected tests identified by Nguyen and Tickell (2014) (e.g. Turnadge et al., 2018; Ball et al., 2019; Amery and Tickell, 2022). In the present assessment, all historical pumping tests undertaken in the Roper catchment were collated, digitised and reanalysed to reestimate hydraulic conditions and properties. Data sources A total of 144 groundwater bores at which historical pumping tests had been performed were identified within the Roper catchment from the Tickell and Nguyen (2014) dataset. A subset of 23 groundwater bores were identified at which two-well constant-rate discharge tests had been performed (two bores: one pump tested, the other monitored) and where a meaningful drawdown response was achieved (i.e. where the maximum drawdown was great than 1 m). Other reasons for excluding historical tests from further analysis included: absence of time–drawdown data; failure to maintain a constant rate of extraction; step changes in drawdown not being attributable to variations in extraction rate; and the inability to estimate physically reasonable parameters during inverse modelling. For each of the 23 pumping tests, hydraulic responses recorded during drawdown, and recovery phases where available, were digitised. Reanalysis methods Time–drawdown data were assessed qualitatively and interpreted quantitatively using the industry standard software package AQTESOLV v4.5 (Duffield, 2007). For responses identified as either unconfined or dual-domain confined, the Moench (1997) solution was used to estimate aquifer and aquitard parameters. For responses identified as leaky confined, the Hantush (1960) solution was used. A symmetric 95% confidence interval (CI; approximated via the two-sigma rule) associated with each estimated hydraulic property value was calculated through the linear propagation of variance. While this parametric approach to estimating parameter uncertainty is admittedly not ideal, it does provide an indication of the relative degree of confidence in estimated hydraulic properties. 3.5.2 Passive time series analyses Active methods of hydraulic testing, which includes both pump and slug testing, are the industry standard for estimating subsurface hydraulic conditions and properties. However, the labour and operational costs of these tests limit their widespread application. Alternatively, subsurface hydraulic conditions and properties can be estimated via the analysis of groundwater-level data collected at relatively high temporal resolution (e.g. hourly) using automated loggers (Turnadge et al., 2019). These analyses infer hydraulic conditions and properties from groundwater pressure responses to passive forcings, including periodic solid body Earth and atmospheric tides (Rau et al., 2020) and aperiodic barometric loading (Rasmussen and Crawford, 1997). Hydraulic conditions, and specifically the degree of aquifer confinement, can be estimated from groundwater responses to Earth tides (Turnadge et al., 2019). For groundwater bores identified as accessing confined aquifers, specific storage can be estimated as a function of barometric efficiency, which represents the proportion of barometric loading that is accommodated by the compressibility of the aquifer (Rau et al., 2020). In this assessment, all available groundwater-level data recorded at high temporal resolution using automated loggers in the Roper catchment were reanalysed to reestimate hydraulic conditions and properties. Data sources Time series of water levels from 31 groundwater bores located in the catchment were obtained from the NT Department of Environment, Parks and Water Security (DEPWS). Three of the 31 DEPWS datasets were unsuitable for analysis, due to large and/or repeated gaps in the records. Time series of water levels from an additional 14 groundwater bores located in the Roper catchment were obtained from the NT Power and Water Corporation (PWC). Eleven of the 14 PWC datasets were unsuitable for analysis due to having limited measurement duration (e.g. <1 year) or coarse sampling resolution (e.g. 6-hourly) or being affected by nearby groundwater extraction. A total of 31 datasets were therefore analysed for groundwater responses to passive forcings. Outlying data values (typically resulting from temporary removal of loggers) were omitted. Short periods of missing data (i.e. in the order of hours) were filled using linear interpolation. Where data were missing over longer periods, datasets were subdivided into shorter, continuous time series. Time series of solid body Earth tides were calculated using the Kudryavtsev (2004) catalogue using the Fortran code ETERNA PREDICT (Wenzel, 1996) via the Python package PyGTide (Rau et al., 2022). Time series of barometric pressure corresponding to the duration of groundwater pressure monitoring and recorded at a sufficient sampling interval (e.g. hourly) were only available at a single location: bore RN035926 (approximately 9 km south-east of Mataranka). Reanalysis methods Both Earth and atmospheric tides are composed of many harmonic constituents, each with a unique frequency (Melchior, 1983). Atmospheric tides are dominated by the solar diurnal (S1) and solar semi-diurnal (S2) constituents, which occur at 1 and 2 cycles/day, respectively. Frequency domain responses of groundwater levels in unconfined aquifers, which are directly hydraulically connected to the atmospheric via the unsaturated zone, are dominated by the same two constituents. In comparison, solid body Earth tides are dominated by up to a dozen harmonic constituents. These include the 1 and 2 cycles/day constituents but also uniquely include the primary lunar constituent (M2), which occurs at approximately 1.9 cycles/day. Frequency domain responses in confined aquifers typically feature attenuated responses to Earth tide constituents. For this reason, the magnitude of the M2 constituent in groundwater-level time series can be used as an indicator of the degree of aquifer confinement. The concept of barometric efficiency (BE) describes the proportion of mechanical loading imposed upon a confined aquifer by the atmosphere. Small (i.e. near-zero) BE values indicate low aquifer compressibility; therefore, a high proportion of the load is accommodated by the aquifer matrix. Conversely, large (i.e. near-unity) BE values indicate high aquifer compressibility. Traditionally, BE values were calculated by directly relating time series (or their temporal derivatives) of groundwater pressure to those of atmospheric pressure. These methods were confounded by the effects of Earth and atmospheric tides on confined aquifer pressures. Acworth et al. (2016) developed a method of BE estimation that accounted for these confounding effects, which was improved further by Rau et al. (2020). In addition to measured groundwater and barometric pressure time series, the latter solution also incorporates a synthetically generated time series of subsurface strain resulting from solid body Earth tides. This can be calculated using the software package ETERNA PREDICT (Wenzel, 1996) based on established tidal catalogues (e.g. Hartmann and Wenzel (1995); Kudryavtsev (2004)). The specific storage (𝑆𝑆𝑠𝑠) of confined aquifers is conventionally defined as 𝑆𝑆𝑠𝑠 = 𝜌𝜌𝑤𝑤 𝑔𝑔 (𝛽𝛽𝑚𝑚 + 𝛽𝛽𝑤𝑤 𝜃𝜃𝑒𝑒), where 𝜌𝜌𝑤𝑤 = freshwater density = 1000 kg.m–3, 𝑔𝑔 = gravitational acceleration = 9.80665 m.s–2, 𝛽𝛽𝑚𝑚 = aquifer matrix compressibility ∈ (10–14, 10–6) Pa–1, 𝛽𝛽𝑤𝑤 = water compressibility = 4.58 × 10–10 Pa–1, and 𝜃𝜃𝑒𝑒 = aquifer matrix effective porosity ∈ (0.01, 0.33) (Domenico and Schwartz, 1998). Since barometric efficiency is a proxy for compressibility, specific storage values can alternatively be estimated as a function of BE as 𝑆𝑆𝑠𝑠 = 𝜌𝜌𝑤𝑤 𝑔𝑔 𝛽𝛽𝑤𝑤 𝜃𝜃𝑒𝑒/𝐵𝐵𝐵𝐵 (Jacob, 1940). Using these input values, aquifer-specific storage values estimable via Jacob’s solution range over approximately four orders of magnitude, from 10–8 m–1 to 10–4 m–1. All components of Jacob’s solution can be estimated within a single order of magnitude, with the exception of aquifer effective porosity. For this reason, aquifer-specific storage values estimated using Jacob’s solution are most sensitive to the latter input parameter, which can in practice be one or more orders of magnitude less than 1%. In such cases, the lower limit for estimated specific storage values will be less than 4 × 10–8 m–1. In this assessment, the frequency domain response (i.e. amplitude spectrum) of water levels measured in groundwater bores was calculated using the discrete Fourier transform (Bendat and Piersol, 1993). For bores at which the M2 constituent was found to be dominant, barometric efficiency values were calculated using the Rau et al. (2020) solution. Specific storage values were then estimated from BE values using the Jacob (1940) solution. 3.6 Recharge estimation The method used in this study for estimating recharge at the catchment scale using the chloride mass balance (CMB) method is an evolution of that used in previous assessments (Taylor et al., 2018a; Taylor et al., 2018b; Turnadge et al., 2018), the Bioregional Assessment Program (Crosbie et al., 2018) and the Geological and Bioregional Assessment Program (Crosbie and Rachakonda, 2021), and most recently in the Sydney region (Wilkins et al., 2022b) and Great Artesian Basin (Crosbie et al., 2022). It includes four steps (Figure 3-1): • estimating recharge using the CMB method at a point scale, including estimating the chloride in rainfall, runoff and groundwater • regression kriging the point estimates of recharge: upscaling recharge using regression equations, kriging the residuals of the point-scale recharge and aggregating the regression rasters and residual rasters to form the regression kriging upscaled recharge rasters • evaluating the uncertainty in the spatial recharge estimates using 1000 replicates of the two processes above • constraining the uncertainty using observations of baseflow at the low end and excess water at the high end. These four steps are described in detail in the following subsections. Results are found in Section 5.5. Figure 3-1. Methods used showing calculating the point recharge, upscaling using regression kriging and reporting of the results at the scale of the region and surface geology group The area under investigation for recharge using the CMB method is substantially larger than the Roper catchment because the surface water catchment is not the groundwater flow divide. It includes both the Roper and multiple Southern Gulf catchments (Settlement Creek, Nicholson, Leichhardt and Morning Inlet catchments) as they are linked by the Georgina Basin, the gulf catchments in between and the northern-flowing catchments that receive groundwater flow from the Dook Creek Formation (Figure 3-2). Stochastically generate chloride deposition surface (single realisation for all points) Stochastically generate chloride in runoff (each point individually generated) Stochastically generate chloride in groundwater (each point individually generated) Calculate point recharge Develop regression equation using covariates Calculate recharge residuals Krige residuals to regular grid Upscale recharge to regular grid using covariates Add recharge and residuals grid Raster Stack of 1000 replicates of upscaled recharge estimates Constrain uncertainty using baseflow and excess water Summary of replicates as 5th, 50th and 95th percentiles Repeat 1000 times Results aggregated by aquifer and surface geology Figure 3-2 Investigation area used for estimating recharge using the chloride mass balance method 3.6.1 Point-scale chloride mass balance The CMB method (Anderson, 1945) is the most widely used approach for estimating recharge, both globally (Scanlon et al., 2006) and in Australia (Crosbie et al., 2010). It is popular because it is robust over many climate zones and is cost effective, requiring only analyses of chloride in groundwater and rainfall. It is relatively insensitive to the mechanism of recharge, whether that be A map of the united states Description automatically generated diffuse recharge through the soil matrix, bypass flow through macropores or even captured surface water in sinkholes (Alcalá et al., 2011; Bazuhair and Wood, 1996). At its simplest, the CMB method can be used to estimate recharge (R, measured in mm/year) knowing only the chloride deposition from rainfall (D, measured in kg per ha per year) and the chloride concentration of the groundwater (Clgw, measured in mg/L-1 ): 𝑅𝑅 = 100 𝐷𝐷 𝐶𝐶𝑙𝑙𝑔𝑔𝑔𝑔 . (1) This works because the chloride deposited on the land surface from the atmosphere is excluded from evaporation and transpiration, so the chloride becomes concentrated as it travels through the soil profile to the groundwater. This evapoconcentration allows estimation of the recharge rate. Wood (1999) listed four important assumptions behind estimating recharge using the chloride mass balance method: • Chloride in groundwater originates from rainfall on the aquifer and not from flow from underlying or overlying aquifers. • Chloride is conservative in the system. • Steady-state conditions are assumed in that the fluxes of chloride and water have not changed over time. • There is no recycling of chloride within the aquifer. These assumptions are discussed in the following paragraphs. For most of the study area, the chloride in groundwater is due to rainfall. In some alluvial areas additional chloride is added to the groundwater system through losing streams and recharge due to overbank flooding. Observations of chloride in groundwater from these areas must be identified and removed from the analysis. In confined aquifers there can be some mixing of groundwater from layers above and below, so to avoid these complications, the analysis only includes observations of groundwater chloride observations from outcropping areas. Chloride is conservative in the system when the system has no sources or sinks. Chloride is exported from the system through runoff, which can be accounted for by incorporating runoff into Equation (1), as in Equation (4) below. The steady-state assumption is difficult to meet in areas that have undergone recent land use change (Cartwright et al., 2007), such as native vegetation clearing for agriculture. There has been limited land clearing in the areas under investigation, so the assumption is made that groundwater observations are from the current land use. To avoid ‘old’ water and only sample ‘young’ water, chloride in groundwater observations from outcrop areas and in bores that are less than 100 m deep are used. Recycling of chloride occurs when groundwater is evaporated or transpired and then returned to the groundwater system. Recycling can occur in groundwater discharge areas (Bazuhair and Wood, 1996), flow through lakes (Howcroft et al., 2017) and through irrigation with groundwater (Wood and Sanford, 1995). These areas can be easily identified as having actual evapotranspiration (AET) greater than rainfall and so any observations of chloride in groundwater from these areas can be excluded. Chloride in rainfall The chloride deposition rate (kg per ha per year) has been obtained from a national gridded product at a 5 km resolution (Wilkins et al., 2022a). Because of the sparsity of observational data concerning chloride in rainfall, uncertainty in the result is high in certain regions. Figure 3-3 shows the chloride deposition rate in the study area. The mean (μ), standard deviation (σ) and skewness (g) from 1000 replicate models are used to provide a Pearson Type III distribution of chloride deposition at each point location in the study area (Figure 3-3). The deposition according to the Pearson Type III distribution is given as (Pilgrim, 1987): 𝐷𝐷 = 𝜇𝜇 + 𝐾𝐾𝑌𝑌. 𝜎𝜎 (2) where KY is a frequency factor calculated from g and a standard normal deviate (z): 𝐾𝐾𝑌𝑌 = 2 𝑔𝑔 􀵤􁉄􁉀𝑧𝑧 − 𝑔𝑔 6􁉁 𝑔𝑔 6 + 1􁉅 3 − 1􀵨. (3) The standard normal deviate is generated stochastically for each replicate and used to generate the chloride deposition rate for input into the recharge estimation at each point location. The same standard normal deviate is used for each bore location in a replicate to maintain spatial consistency in the recharge estimates. The same process is repeated 1000 times to generate the inputs for the 1000 replicates. Figure 3-3 Chloride deposition across the study region: (a) mean, (b) standard deviation, and (c) skewness (Wilkins et al., 2022a) Blue squares indicate points where chloride deposition has been measured. Chloride in runoff The chloride exported in runoff can be accounted for by modifying Equation (1) by assuming that the chloride exported in surface runoff is proportional to the runoff coefficient. The runoff coefficient (RC), which is the proportion of rainfall that becomes streamflow, can be obtained from the output of the AWRA model (Vaze et al., 2013) on a gridded basis and therefore can then be used to estimate RC at each point within the study area (Figure 3-4). To account for runoff, Equation (1) can be modified to: 𝑅𝑅 = 100 𝐷𝐷(1−𝛼𝛼.𝑅𝑅𝑅𝑅 ) 𝐶𝐶𝐶𝐶 𝑔𝑔𝑔𝑔 . (4) Here α is a scale factor, which is included because large rainfall events that result in large runoff volumes are often below average in chloride concentration and therefore do not reduce chloride deposition proportionally. For this assessment, α is sampled stochastically from a uniform distribution with a value between 0.33 and 0.66 (Crosbie et al., 2018) during the uncertainty estimation (the second step of Figure 3-1). The value of α is equal for each bore in a replicate to maintain spatial consistency but is sampled individually for each replicate. Figure 3-4 Runoff coefficient across the investigation area Chloride in groundwater Measurements of chloride in groundwater were sourced from two datasets: • the NT Government’s territory-wide groundwater database (Department of Environment Parks and Water Security, 2014a) • the Bureau of Meteorology’s (BoM’s) National Groundwater Information System (NGIS) (BoM, 2020) There were too many individual observations of chloride in groundwater to enable quality assurance and quality control to be undertaken on each observation, so an automated procedure was employed to select the observations used in estimating recharge. This procedure was based on previous work (Crosbie et al., 2018; Crosbie and Rachakonda, 2021) and involved the following steps: 1. Duplicate records were removed. 2. The mean value of chloride in groundwater was calculated for each location where multiple analyses had been undertaken. 3. Any location where the excess water (rainfall minus actual evapotranspiration) was less than −25 mm/year was excluded. Negative excess water means the actual evapotranspiration is greater than the rainfall, which indicates an additional source of water and chloride, such as groundwater discharge, irrigation or flooding, which violates the assumptions of the CMB method previously described. 4. Any bore greater than 100 m deep was excluded. The ideal sampling point for estimating recharge would be immediately below the watertable where water has recharged the groundwater store recently. Monitoring bores are not constructed this way, so limiting the depth is a compromise to ensure that water is sampled from outcrop areas and is as recent as possible. 5. Chloride in groundwater concentrations of less than 2 mg/L were excluded, as these measurements are probably not representative of groundwater. These data are closer to that of rainfall and probably indicate a mis-transcribed value entered in the database or a sample contaminated with rainfall through short circuiting of the bore casing. 6. Chloride in groundwater concentrations of greater than 2000 mg/L were excluded as these data are probably not representative of groundwater in recharge areas. They are more likely to be discharge areas or downstream of discharge areas. 7. Bores that located on alluvium as mapped in Geoscience Australia’s surface geology maps (Raymond et al., 2012) were excluded, as the alluvium is likely to have additional chloride present that is not directly from rainfall. This includes areas subject to flooding, irrigation or groundwater discharge from deeper or adjoining layers. 3.6.2 Upscaling point estimates of recharge The upscaling of the point estimates of recharge using regression kriging (Hengl et al., 2004) includes three steps (detailed in subsequent sections): 1. developing a regression equation to predict recharge across the study area using covariates that can be mapped on a regular grid 2. kriging the residuals between the point estimates of recharge and the regression equation estimates of recharge to provide a surface of residuals 3. adding the residual surface to the regression surface, which provides a spatial estimate of recharge that is informed locally by point data and away from point data is dependent upon a global relationship between the covariates and recharge. This process is repeated for 1000 stochastically generated replicates to quantify the uncertainty in the recharge estimates, as outlined in Figure 3-1. Regression equations Previous studies have shown that recharge is better approximated by a log-normal distribution than a normal distribution (Cook et al., 1989; Eriksson, 1985), and this extends to its relationship with rainfall (Petheram et al., 2002). Reviews of recharge in Australia and globally have shown that recharge estimates depend upon rainfall, soil type and vegetation (Crosbie et al., 2010; Kim and Jackson, 2012; Scanlon et al., 2006), and these have been used successfully as covariates to upscale modelled point estimates of recharge (Crosbie et al., 2013). Surface geology has also been used as a covariate in upscaling recharge estimates (Crosbie et al., 2015; Crosbie et al., 2018). Based on these previous studies, the covariates used are: • annual mean rainfall over the 30-year period from 1991 to 2020 (Jones et al., 2009) (Figure 3-5a) • mean clay content of the top 2 m of the soil profile (Grundy et al., 2015) (Figure 3-5b) • Normalised Difference Vegetation Index (NDVI) as a measure of vegetation density (BoM, 2019) (Figure 3-5c) • a simplification of the surface geology (Raymond et al., 2012) (Figure 3-5d). Figure 3-5 Covariates used in upscaling: (a) rainfall, (b) clay content of the soil, (c) Normalised Difference Vegetation Index (NDVI), and (d) simplified geology The regression equation used in the upscaling uses the covariates of log-transformed rainfall (log10 𝑃𝑃), clay content (C) and NDVI as a measure of vegetation density (V), and is: log10 𝑅𝑅 = 𝛽𝛽𝑔𝑔𝑔𝑔 𝑔𝑔 + 𝛽𝛽1. log10 𝑃𝑃 + 𝛽𝛽2. 𝐶𝐶 + 𝛽𝛽3. 𝑉𝑉. (5) where βgeo, β1 , β2 and β3 are fitting parameters fitted through least squares regression (which will be different for each of the 1000 replicates). The βgeo parameter is a set of three values representing the intercept value for each of the three geology classes (high, medium and low recharge). Now consider a single replicate (i.e. a single choice for the chloride deposition and runoff) and the most successful covariates. To form the regression equation for this replicate, bootstrapping is used to reduce the effect of outliers and estimate the uncertainty in the regression equation (Efron and Tibshirani, 1994). That is, the same number of observations are randomly selected from the population of all observations, with replacement. This ensures that the same number of observations is used in the regression equation for each replicate but not every point will be included in each replicate, and some points will be included more than once. This minimises the bias in the regression equation that can occur due to outliers in the dataset. Kriging residuals The difference in the recharge estimated through the regression equations and through the pointscale CMB method are calculated for each replicate. The residual at each observation point is defined by: Residual = log10(fitted) − log10(predicted) (6) Here ‘fitted’ is the recharge (in mm/year) according to the regression equation, while ‘predicted’ is the recharge (in mm/year) according to the CMB method. These residuals are fitted to a spherical semi-variogram using gstat (Pebesma, 2004) in R (R Core Team, 2016) and then kriged to a 0.01° (~1000 m) grid to create a residuals surface. The nugget, sill and the range parameters are fitted to the variogram separately for each replicate. Aggregating recharge rasters For each replicate, the residual raster was added to the recharge raster upscaled using the regression equations to create a regression kriging recharge raster, at a 0.01° (~1000 m) resolution. Summary statistics were calculated from the 1000 replicates at the 5th, 50th and 95th percentiles, with the 50th percentile being assumed to be the ‘most likely’ estimate of recharge and the difference between the 5th and 95th percentiles a measure of the uncertainty. Constraining the uncertainty The CMB estimates of net recharge are constrained by the baseflow and the excess water estimates using a rejection sampling approach (Tarantola, 2005; Von Neumann, 1951). This is the same process used to constrain the CMB estimates of recharge in the CLA (Crosbie and Rachakonda, 2021). It relies on the baseflow being less than the recharge and the excess water being greater than the recharge. Three catchments were selected that are known to have high levels of baseflow and to drain large parts of the study area (Drysdale et al., 2002; Yin Foo and Matthews, 2001). The selected gauges were the first gauge downstream of the known groundwater discharge zones (Figure 3-6a): Roper River at Elsey Homestead (G9030013), Gregory River at Riversleigh No.2 (912105A), and Lawn Hill Creek at Lawn Hill No.2 (912103A). The baseflow separation method used here is the digital recursive filter suggested by Lyne and Hollick (1979). Although it has no physical basis it is the most commonly applied method of baseflow separation in Australia (Grayson et al., 1996). The baseflow separation was undertaken using the BaseJumper program (Murphy et al., 2008) with a filter parameter of 0.925. It was assumed that the error in the baseflow estimates could be up to ±30% based on studies in other regions (Coxon et al., 2015; Petersen-Øverleir et al., 2009). The CMRSET algorithm for AET (Guerschman et al., 2011) was used with MODIS data to create an 8-day time series at 250 m resolution over the study area. These data were aggregated to a longterm mean for the period 2001 to 2018. The long-term mean AET was subtracted from the longterm mean rainfall over the same period (Jones et al., 2009) to create a long-term mean excess water data layer. The long-term mean excess water was extracted for the three catchments used for the baseflow separation (Figure 3-6a) and for three selected internally draining catchments (BoM, 2012) (Figure 3-6b). Based on an evaluation of CMRSET water balances against runoff from stream gauges (King et al., 2011) and reviews of remotely sensed evapotranspiration (ET) (Glenn et al., 2011; Kalma et al., 2008), an error of up to ±30% has been assumed. Both the baseflow (BF) and excess water have an estimated uncertainty of ±30%. Assuming that the uncertainty is normally distributed and using the ‘six‐sigma rule’ gives the standard deviation of the baseflow prediction as 10% of the estimated value. This can then be used to estimate a probability distribution which can be randomly sampled from: 𝐵𝐵𝐵𝐵𝑖𝑖 = 𝜇𝜇𝑖𝑖 + 𝑧𝑧. 𝜎𝜎𝑖𝑖 (7) where z is the randomly selected standard normal deviate and μI and σI are the mean and standard deviation of the baseflow for the ith catchment. In this way the same standard normal deviate is used to estimate the baseflow for each of the three catchments for each repetition of the rejection sampling algorithm. The uncertainty in the excess water is defined similarly: 𝐸𝐸 𝐸𝐸𝑖𝑖 = 𝑃𝑃 − (𝜇𝜇𝑖𝑖 + 𝑧𝑧. 𝜎𝜎𝑖𝑖). (8) The mean recharge across the six catchments has been extracted from the 1000 upscaled replicates of the CMB method. The rejection sampling algorithm was run 10,000 times with a randomly selected standard normal deviate for the baseflow distribution, a second randomly selected standard normal deviate for the excess water distribution and a randomly selected run number from the 1000 replicates of upscaled CMB recharge. A selection was retained in the posterior distribution if the following constraints were met: • The upscaled CMB estimates of recharge were greater than the baseflow estimates for each of the three catchments. • The upscaled CMB estimates of recharge were less than the excess water estimates in each of the three additional internally draining catchments. Figure 3-6 (a) Catchments used for baseflow analysis for rejection sampling, (b) additional internally draining catchments used with excess water for rejection sampling 3.6.3 Extracting recharge values for zones of interest The 1000 m resolution rasters produced may be used to estimate recharge aggregated over any area. Mean recharge rates have been extracted for the major aquifers (Figure 2-11) using the simplified geology (Figure 3-5d). 3.7 Identifying potential groundwater discharge areas using remote sensing Identifying areas of groundwater discharge has conventionally involved extensive observations in the field. This includes identifying the presence of ‘phreatophytes’ (deep-rooted plants that depend on groundwater), springs and seeps on the ground or increases in baseflow along river reaches. These approaches are labour intensive and impractical at the regional scale. As an alternative, remote sensing products offer an opportunity to identify and delineate the areas of potential groundwater discharge for further investigation in the field. 3.7.1 Digital Earth Australia The Digital Earth Australia (DEA) suite of products is derived from the Landsat Open Data Cube produced by Geoscience Australia (Lewis et al., 2017). This dataset includes all Landsat imagery at a 25 m resolution dating back to 1987. The water bodies dataset (Krause et al., 2021) and Water Observations from Space (WOfS) dataset (Mueller et al., 2016) are both potentially useful. The WOfS dataset classifies each pixel in each Landsat image as being inundated with water or not. The dataset contains a summary layer that is the proportion of images from the Data Cube in which a pixel is inundated; this can discriminate between areas that are permanently inundated and areas that are infrequently inundated (and dryland areas). The DEA water bodies dataset is a polygon representation of contiguous areas at least 5 pixels (~0.3 ha) in size that are inundated in more than 10% of images. Identifying permanent water bodies can potentially narrow the search for groundwater discharge areas. The maximum from each pixel within a water body of the proportion of inundated images has been assigned to each polygon. This method can establish the permanence of a water body, but it has an issue with scale: a large water body will be assigned a degree of permanence based on a single 25 × 25 m pixel. Any water bodies that hold water throughout the dry season can be further investigated for the source of water. Permanent water could result from surface water that replenishes a deep waterhole during the wet season where the depth of water is greater than the evaporation through the dry season. Alternatively, it could result from a continuous discharge of groundwater throughout the year. Note that this method also identifies anthropogenic water storages such as instream or farm dams. 3.7.2 Excess water The CMRSET v2.2 dataset (Guerschman et al., 2022; McVicar, 2022) uses high-resolution/lowfrequency satellite data (e.g. Landsat and Sentinel-2) blended with low-resolution/high-frequency satellite data (e.g. MODIS and VIIRS) to produce a monthly actual evapotranspiration product at a resolution of 30 m from 2002 onwards. Excess water is calculated by subtracting the annual mean AET from the annual mean rainfall (P). Areas with positive values of excess water (i.e. P > AET) are exporting water elsewhere, either through runoff or through groundwater recharge. Areas with negative values of excess water (i.e. P < AET) are importing water from either a surface source such as floodplains and wetlands or a groundwater source such as springs and seeps. Areas with high negative excess water have been used previously to identify groundwater discharge areas (Crosbie and Rachakonda, 2021). These areas are indicative of having a water source throughout the year rather than only in the wet season for the majority of the landscape that is reliant upon rainfall. 3.7.3 Potential discharge areas associated with carbonate rocks The Digital Earth Australia products are very useful for showing where open water sits in the landscape when it can be seen by satellites. This has problems where there is a tree canopy that can hide the small water bodies from the satellites’ view, thus some perennial water bodies are not identified. It also cannot identify areas of groundwater discharge where there is no associated surface water body, e.g. discharge by terrestrial vegetation. The excess water mapping derived from the CMRSET data has the opposite problem in that too many areas identified as having evapotranspiration greater than rainfall are not groundwater discharge areas. These areas are primarily streamlines and floodplains that are subject to wet-season run-on that stores water for a limited time into the dry season. Both of these products supply useful information on groundwater discharge areas, but neither is sufficient for mapping potential groundwater discharge areas alone. The use of CMRSET AET throughout the year was investigated further as a method of mapping potential discharge areas associated with the CLA around Mataranka, and the DCA in the north of the catchment and smaller productive aquifers hosted in the Knuckey Formation around Ngukurr. Spatial patterns of AET were investigated for the monthly averages in February when the catchment was at its wettest and for June, August and October as the dry season progressed. As the discharge areas around Mataranka are well known and have been mapped previously, the focus of further work was on the Dook Creek Formation where only discrete springs and waterholes have previously been mapped (the Knuckey Creek Formation did not show much evidence of groundwater discharge). The areas of groundwater discharge showed consistently higher AET through the dry season than the surrounding areas, particularly in October when stored soil water in run-on areas had become depleted. To delineate areas of potential groundwater discharge, a threshold of 2.5 mm/day for the mean AET in October was adopted based on the mapped areas of known groundwater discharge around Mataranka. This threshold was then applied to the areas overlying the Dook Creek Formation. 3.7.4 Potential discharge areas over whole catchment When the methods applied in the Dook Creek Formation for identifying groundwater discharge areas were trialled across the rest of the Roper catchment and areas outside the Roper catchment, it became apparent that the 2.5 mm/day threshold for October actual evapotranspiration was not high enough. For the Dook Creek Formation, and Mataranka, the annual mean rainfall is approximately 1000 mm/year and the 2.5 mm/day threshold worked well. When applied to areas of higher rainfall, too many potential discharge areas were identified, and too few areas were identified in low rainfall areas. In October the potential evapotranspiration (PET) over the Dook Creek Formation is 6.9 mm/d, which gives a threshold as previously applied of mean October AET being at least 36% of mean October PET. If this threshold is defined as a ratio of AET to PET, and is scaled based on aridity index (P/PET), then the threshold is calculated in the form of the Budyko equation (Budyko, 1974): 𝐴𝐴𝐴𝐴 𝐴𝐴 𝑃𝑃𝑃𝑃𝑃𝑃 = 1 + 𝑃𝑃 𝑃𝑃𝑃𝑃𝑃𝑃 − 􁉈1 + 􀵬 𝑃𝑃 𝑃𝑃𝑃𝑃𝑃𝑃 􀵰 𝑤𝑤 􁉉 1􀵗𝑤𝑤 . (9) In this application the exponent w is not the same as used in Budyko equation for estimating catchment AET and subsequent runoff. It is an empirical scaler and was determined through trial and error to be 1.65 (Figure 3-7). Using this equation the Dook Creek Formation has a mean aridity index of 0.6 and so would have a threshold of 36% of PET or 2.5 mm/d, the same as used previously. Figure 3-7 Relationship used for determining the threshold October actual evapotranspiration for estimating potential discharge areas from aridity index The areas identified by this threshold approach include all areas that have a high October evapotranspiration, irrespective of source. This includes floodplains and wetlands that can retain water through the dry season and actively managed areas such as dams and irrigation areas. Groundwater discharge areas tend to be a stable water source and do not exhibit high interannual variability. Wetlands and floodplains tend to have higher inter-annual variability due to their dependence upon retaining water from the previous wet season. This distinction is used to further aid in identifying groundwater discharge areas using the coefficient of variation of the October evapotranspiration. A threshold of 0.5 has been used with areas identified with high October evapotranspiration and high variability are excluded as probably not being groundwater discharge areas. The CMRSET algorithm over estimates evapotranspiration from bare soils, particularly dark or red soils. If it is assumed that areas of groundwater discharge are more likely to be vegetated, then false positives due to bare soils can be minimised by removing areas with high fractions of bare soils. This was done using the DEA Fractional Cover dataset (GA, 2021). This dataset identifies areas of green growing vegetation; brown, dead or senescing vegetation; and bare soil. Areas with a mean October bare soil fraction of greater than 15% have been excluded. To minimise isolated pixels that are more likely to be false positives, identified areas were collated as contiguous polygons and any less than 0.5 ha in size were excluded. There were still some false positives in the dataset so each polygon was inspected in relation to geological contacts, its topographic position and watertable level where this data exists. Each polygon was assigned to one of the following classes: • perennial groundwater discharge • seasonally varying • coastal • recharge feature • mis-identified? The ‘perennial groundwater discharge’ class is most commonly springs and seeps. These can be on streamlines but are frequently not. They are usually associated with structural features such as fractures, joints and faults, a change in geological unit (one unit onlaps another), or a break of slope, and they are generally not in the alluvial areas low in the catchment. The ‘seasonally varying’ class is associated with the thin and storage-limited alluvium of the major rivers. These areas are conceptualised as discharging groundwater during the dry season from an alluvial store that is recharged via surface water in the wet season. The ‘coastal class’ is defined as polygons on estuarine or delta deposits or on the shoreline itself. These areas may have a groundwater discharge component as well as the tidal sea water component. ‘Recharge features’ hold water through the dry season but are above the watertable. These are frequently associated with sinkholes. Areas in the ‘mis-identified?’ class meet the criteria associated with the remote sensing products but do not have the geological attributes of the other classes and are not inferred to be discharge features. 4 Targeted field, desktop, and modelling investigations Targeted field, desktop, and modelling investigations primary focused on the CLA and DCA; however, components of the work also focused on their hydraulic connection to adjacent aquifers and GDEs including springs, surface water and groundwater-dependent vegetation. The regional assessment of the Roper catchment (see Section 3) provided detailed information and datasets to inform the targeted field, desktop and modelling investigations, which are described below in detail. The field and desktop investigations were conducted by CSIRO but included collaboration with DEPWS, Charles Darwin University and CloudGMS on key components as part of contractual agreements, including: • drilling the CLA and underlying hydrostratigraphic units in Elsey National Park in the vicinity of the upper Roper River and the Mataranka Springs Complex • sampling groundwater and monitoring water levels across parts of the CLA and DCA • sampling surface water along key reaches of the Upper Roper and its tributaries • scenario-based groundwater flow modelling of the CLA and DCA. 4.1 Hydrogeological framework 4.1.1 Drilling investigations Cambrian Limestone Aquifer Inter-aquifer connectivity between the CLA and aquifers hosted in underlying hydrogeological units and the locations and sources of groundwater–surface water interactions in the vicinity of the upper Roper River and its tributaries, and the Mataranka Springs Complex was a potential source of uncertainty identified in recent investigations by Huddlestone-Holmes et al. (2021) and Lamontagne et al. (2021). To address this potential uncertainty, drilling was carried out in Elsey National Park in collaboration with DEPWS to: (i) to map and characterise the lithology and stratigraphy of hydrogeological units under the CLA, and (ii) provide purpose-built infrastructure discretely screened in these underlying units to obtain hydraulic, geochemical and environmental tracer data to characterise their connectivity with the overlying CLA, the Mataranka Springs Complex and the upper Roper River. Two drill sites were chosen based on the digital geology mapping, the existence of access tracks to existing groundwater bores and lithological information from bores in units underlying the CLA in the near vicinity of the upper Roper River and Mataranka Springs Complex. Cultural heritage clearance for drilling to be undertaken at two different sites was provided by the Aboriginal Areas Protection Authority. A moderate-depth (>100 mBGL) investigation hole was drilled at two different sites in Elsey National Park to confirm: (i) the thickness and composition of the CLA at each site, and (ii) the presence, nature and composition of different hydrogeological units underlying the CLA in the vicinity of the upper Roper River and the Mataranka Springs Complex. Information from each bore hole was then used to inform the installation of two new purpose-built monitoring bores (one in the CLA and one in the most immediate underlying hydrogeological unit) at each of the two sites. In total, five bore holes were drilled and four monitoring bores were installed while one site was backfilled (Figure 4-1). The monitoring bores (i) characterise the water levels, water quality and bore yields for the CLA and aquifers hosted in underlying hydrogeological units, (ii) generate updated hydrogeological data for use in new cross-sections, and (iii) provide new infrastructure to assess inter-aquifer connectivity and groundwater–surface water connectivity in the vicinity of the upper Roper River and the Mataranka Springs Complex (Figure 4-2). Details of the four new monitoring bores are presented in Table 4-1 and their locations shown in Figure 4-2. Drilling services were provided by the Water Resources Division of DEPWS. Drilling was conducted in November 2022 using a custom-built Atlas Copco T3WDH drill rig. Access to each site was via existing access roads. Drill pads of approximately 15 × 15 m were bulldozed to provide suitable sites to undertake the drilling. All holes were drilling using the rotary air method with occasional use of foam to stabilise the borehole and assist in returning drill cutting to the surface. Drill cuttings were collected and logged in 3 m intervals and used to select bore completion specifications including construction material, bore diameter and completion depth (including bore depth, screen depth and screen length). Figure 4-1 (a) Rotary drilling an investigation hole, (b) collected rock cuttings being logged by depth, (c) installing PVC casing, and (d) an complete final bore with steel standpipe set in concrete Bores were completed in discrete aquifers with short vertical screens and close spatial vicinity adjacent to each other to represent ‘nested’ vertical sites. Bore construction (i.e. bore diameter, screen type and screen length) was based on an assessment of aquifer composition and thickness and initial bore yields by airlifting. Bores in minor, low-yielding aquifers were constructed with 100 mm diameter PVC casing and machine-slotted short, screened intervals (i.e. 6 m) that were cemented from overlying aquifers. Bores in larger, high-yielding aquifers such as the CLA were constructed with larger diameter (i.e. approximately 150 to 200 mm) steel casing with stainless steel wire-wound short, screened intervals that were cemented from overlying aquifers. Bore design was necessary to enable meaningful collection and analysis of groundwater chemistry and A screenshot of a computer Description automatically generated environmental tracers to characterise and quantify groundwater flow, vertical inter-aquifer connectivity and groundwater–surface water connectivity at both sites. All bores were constructed in accordance with the minimum construction requirements (National Uniform Drillers Licensing Committee, 2020) and developed by airlifting until field parameters (including pH, electrical conductivity and temperature) had stabilised and discharge water was clear of suspended sediment. Bore development was typically achieved over a period of 1 to 2 hours. All bores were completed with a steel standpipe with lockable lid set in a 1 m2 concrete slab. Table 4-1 Location and target aquifers for new bores installed in Elsey National Park BORE NO. COMPLETED DATE DATUM LATITUDE LONGITUDE UTM ZONE EASTING NORTHING AQUIFER RN043045 1/11/2022 GDA94 -14.9036704 133.0928172 53 294839 8351450 Antrim Plateau Volcanics RN043046 4/11/2022 GDA94 -14.9036543 133.0930496 53 294864 8351452 Tindall Limestone RN043047 17/11/2022 GDA94 -15.0160943 133.1975983 53 306215 8339104 Antrim Plateau Volcanics *RN043048 1/12/2022 GDA94 -15.0160608 133.1973846 53 306193 8339106 Antrim Plateau Volcanics RN043049 30/11/2022 GDA94 -15.0160772 133.1977193 53 306228 8339106 Tindall Limestone *RN043048 was backfilled. Figure 4-2 Spatial distribution of drilling and newly installed monitoring bores in Elsey National Park To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The red box in the top right inset indicates the location and geographical extent of the map figure within the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008) Geological faults data source: Department of Industry, Tourism and Trade (2010) Spring data source: Department of Environment Parks and Water Security (2014c) A map of the sea Description automatically generated 4.1.2 Hydrogeological cross-sections The lithological and/or stratigraphic logs collated and reviewed in Section 3.1 and the new logs generated from the drilling program described in Section 4.1.1 were used to conduct a spatial analysis to generate hydrogeological cross-sections across key areas of both the CLA and DCA. Cambrian Limestone Aquifer Existing cross-sections of the CLA from water resource maps (Tickell, 2016) and previous hydrogeological investigations (Bruwer and Tickell, 2015; Knapton, 2020) and their locations across the aquifer were reviewed. In addition, the recently developed Leapfrog hydrostratigraphic model that forms the geometry of the layers in the DR2 FEFLOW groundwater model (Knapton, 2020) was interrogated by CloudGMS to strategically extract three hydrogeological cross-sections traversing west to east across the CLA in the Daly, Wiso and Georgina basins beneath the Roper catchment. Cross-sections were generated to highlight two-dimensional spatial changes in the thickness and geometry of the CLA and adjacent overlying and underlying hydrogeological units traversing recharge and throughflow areas. New cross-sections were also generated traversing the regional groundwater discharge zone in the vicinity of the upper Roper River and some of its major tributaries (Waterhouse River and Elsey Creek) and the Mataranka Springs Complex. Dook Creek Aquifer While there have been some useful historical water resource and hydrogeological investigations of the DCA (Knapton, 2009c; Williams et al., 2003; Zaar and Tien, 2003), hydrogeological data are sparse relative to the CLA. The only existing cross-section found in the literature is a section traversing the DCA in the vicinity of the Goyder River north-east and outside the Roper catchment (Prowse et al., 1999; Zaar and Tien, 2003). Most of the drilling across the DCA has focused on community water supplies at Bulman and Weemol and numerous surrounding outstations (Momob, Bodeidei Camp, Morbon/Blue Water). This part of the DCA is also an important groundwater discharge zone with numerous discrete springs (Weemol, Wiamuna, Mount Catt) and groundwater seepage providing baseflow to the Wilton River and its tributaries (Knapton, 2009c; Zaar and Tien, 2003). A spatial analysis was undertaken, and stratigraphic picks from lithological and/or stratigraphic logs were used to generate a new cross-section across the region around Bulman extending from Mount Jean in the north-west to Mount Catt in the south-east. 4.1.3 Depth to top of major aquifers in key hydrogeological units Understanding spatial changes in the depth to the top of the CLA and DCA is important for assessing the costs of drilling to intersect the aquifers and assessing construction and installation costs of different types of groundwater infrastructure (i.e. monitoring bores or production bores for community water supply or irrigation). In addition, spatial changes in depth can be important for understanding groundwater flow processes including recharge, discharge, inter-aquifer or aquifer–aquitard connectivity, and changes in the confinement status of aquifers in different locations. Cambrian Limestone Aquifer CloudGMS used the recently developed Leapfrog hydrostratigraphic model that forms the geometry of the layers in the DR2 version of the FEFLOW groundwater model (Knapton, 2020) to extract the gridded model layer for the top of the CLA within and beneath the Roper catchment. The gridded surface of the top of the CLA from Leapfrog was then mapped and classified into depth classes using ArcGIS. Dook Creek Aquifer Hydrostratigraphic data used to generate the contoured surface of the DCA in the initial DCA FEFLOW model (Knapton, 2009c) was imported by CloudGMS into Leapfrog, and a refined gridded surface was generated by CloudGMS to form the key layer in the DR2 version of the DCA FEFLOW model (Knapton et al., 2023). CloudGMS then extracted the gridded surface of the top of the DCA from Leapfrog which was then mapped and classified into depth classes in ArcGIS. 4.2 Groundwater recharge and flow 4.2.1 Regional groundwater levels Cambrian Limestone Aquifer Groundwater-level observation data from the NT groundwater database (Department of Environment Parks and Water Security, 2014a) were used to derive both a potentiometric surface for the CLA and a depth to standing water level (SWL) surface. The data selected as input to the surface included the entire extent of the CLA within the Roper catchment as well as an area of the CLA within a 100 km buffer around the coincidence between the aquifer and catchment boundary. This resulted in the selection of 3825 bores. These bores were combined with the aquifer attribution dataset described in Section 3.1 to ensure bores with water level data were only screened to the CLA. This resulted in 2011 bores with water level data attributed to the CLA. It is important to note that in some places groundwater associated with the CLA is hosted in parts of the overlying Cambrian siltstone (Jinduckin Formation and Anthony Lagoon Formation) and Cretaceous sediments, but groundwater levels in aquifers hosted in these overlying hydrogeological units were not included at the scale of the potentiometric surface. Groundwater-level measurements from the NT groundwater database were available as a standing water level in metres below ground level (mBGL). However, producing a potentiometric surface requires a height of the groundwater level above a reference datum. To convert SWL measurements to a reduced standing water level (RSWL) in metres above Australian Height Datum (mAHD), they were subtracted from the elevation of the ground at the location of the bore. The SWL measurements were also used independently to produce a depth to groundwater (SWL) surface. The ground elevation data used was derived from the hydrologically enforced Shuttle Radar Topography Mission (SRTM)–derived 1 Second Digital Elevation Model (Gallant et al., 2011), with the elevation at the pixel closest to each bore used to represent its reference height. This does not account for any height of the bore standpipe/collar height above the ground. Groundwater-level measurements were taken at varying times, usually at the time of bore construction. The paucity of data in both time and space means it is not possible to construct maps of the potentiometric surface for a single time period. Due to accessibility requirements, all groundwater-level measurement were collected during the dry season, so the data represent a dry-season surface. Analysis of spatial attributes of groundwater-level data It is important to understand the spatial structure of the water level data to understand the interpolation best suited to this data type. To understand the spatial structure of the data, an experimental semi-variogram was plotted for both the groundwater-level data and the elevation data at bore locations. A semi-variogram (or simply a variogram) is a plot displaying the distance between points (lag) and their similarity. It reflects the covariance between water level measurements. The experimental variogram is used to understand the form of water level data required to model a theoretical variogram, which can be used for data interpolation using a geostatistical method known as kriging (Cressie, 1990). A theoretical variogram has several key components including a nugget, sill and a range. A regression analysis was also conducted comparing groundwater-level data to elevation data at bore locations. This is important to understand the relationship between groundwater and topography across different aquifers. It can help explain the degree of confinement of an aquifer system and to determine whether groundwater flow is recharge or topography driven (Haitjema and Mitchell-Bruker, 2005). The regression analysis assisted with decisions regarding the best form of interpolation to use to define the potentiometric surface across unconfined, semi-confined and confined parts of the aquifer. Watertable interpolation There are many different options for watertable interpolation. Several approaches were compared, including kriging, inverse distance weighting and co-kriging (using land elevation as a covariate). GIS software was used for the analysis and the regression statistics of each interpolation method to compare each interpolation approach. Ultimately, simple kriging was chosen as the best approach for the interpolation. Kriging is superior to inverse distance weighting for several reasons, including that it accounts for whether data are clustered, so points that are close together in space may be weighted lower than data that are stand-alone. Kriging is also a probabilistic form of interpolation, and each mean estimate of the potentiometric surface is accompanied by an estimate of data variance. This helps to interpret the degree of uncertainty in the final surface. Co-kriging was not chosen because of the lack of correlation between the digital elevation model and the watertable in the southern parts of the aquifer where land elevation is greater. The CLA is overlain in places, mostly outside the Roper catchment, by a sequence of siltstone and mudstone known as the Jinduckin Formation (and its equivalents such as the Anthony Lagoon Formation), resulting in partial confinement of the aquifer, particularly towards the south and west (see Figure 2-11). Dook Creek Aquifer Groundwater-level observation data from the NT groundwater database (Department of Environment Parks and Water Security, 2014a) for the DCA are sparse compared to data for the CLA. However, the DC2 DCA FEFLOW groundwater model contains an output of calibrated depth to groundwater based on calibration to both spatial and temporal groundwater levels and groundwater discharge to streams (see Section 4.3). The calibrated depth to groundwater grid was exported and used to create a spatial map of modelled depth to groundwater across the DCA. 4.2.2 Groundwater sampling campaigns To better characterise groundwater flow processes across key parts of both the CLA and DCA, the aquifer attribution spatial dataset described in Section 3.1 was used for a spatial analysis to identify bores discretely screened in both aquifers with suitable bore construction (i.e. shortscreened intervals) for groundwater sampling for chemistry and environmental tracers. Cambrian Limestone Aquifer Previous hydrogeological investigations of the CLA have highlighted the complex nature of groundwater flow processes (recharge, throughflow and discharge) occurring due to its karstic nature, large spatial extent and spatial changes in confinement status (Bruwer and Tickell, 2015; Jolly et al., 2004; Karp, 2008; Yin Foo, 2002). In addition, a recent investigation of the sources of discharge at the Mataranka Springs Complex by Lamontagne et al. (2021) identified the potential for inter-aquifer connectivity between the CLA and APV or other deeper underlying hydrogeological units around Mataranka. Furthermore, recent work on the mapping of the hydrostratigraphy of the CLA by Bruwer and Tickell (2015) and Knapton (2020) indicate the potentially most prospective part of the CLA (larger saturated thickness, higher transmissivities) occurs along the eastern margin of the CLA in the southern Daly Basin and northern Georgina Basin beneath the Roper catchment. This area coincides with numerous water management zones for the proposed Mataranka Tindall Limestone Aquifer Water Allocation Plan (MTLAWAP) and the current Georgina Wiso Water Allocation Plan (GWWAP). Groundwater sampling The spatial aquifer attribution dataset generated from analyses described in Section 3.1 was used to collate lithology and stratigraphy data for groundwater bores in the CLA and adjacent hydrogeological units underlying the CLA between Daly Waters and Mataranka. These analyses were used to: (i) evaluate available groundwater-level data for the aquifers, (ii) identify a subset of bores that followed the regional hydraulic gradient in the unconfined part of the aquifer, and (iii) identify a subset of bores discretely screened to the CLA and underlying hydrogeological units that were suitable for groundwater sampling (particularly for environmental tracers) by evaluating their construction details. A groundwater sampling campaign for chemistry and environmental tracers in groundwater was conducted to better characterise and quantify groundwater flow processes and inter-aquifer connectivity between Daly Waters and Mataranka (Figure 4-3). Permissions to access bores were sought from Traditional Owners and pastoral station owners and managers. Two sampling programs occurred over two dry seasons in August 2022 and April 2023 and included the newly drilled bores described in Section 4.1.1. Specific details of the methods of groundwater sampling and field observations are described in sections 4.2.3 to 4.2.6. Spring and surface water sampling were also conducted to confirm the nature and sources of localised and regional groundwater discharge zones around Mataranka (see Section 4.2.11). Figure 4-3 Target area for groundwater sampling across the unconfined part of the Cambrian Limestone Aquifer and adjacent aquifers between Daly Waters and Mataranka To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left map inset with red polygon indicates the location and geographic extent of the map figure within the Roper catchment. The pink polygon on the map indicates the target area for groundwater sampling. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008). Geological faults data source: Department of Industry, Tourism and Trade (2010). Water management zones: Department of Environment Parks and Water Security (2019c). Dook Creek Aquifer Groundwater sampling The few previous hydrogeological investigations of the DCA have mostly been at a local scale associated with either community water supplies (Verma and Rowston, 1992; Yin Foo, 1983) or the Arafura Swamp (Williams et al., 2003). However, two regional-scale studies, the water resources of West Arnhem Land by Zaar and Tien (2003) and the Gulf Water Study by Zaar (2009d) and Knapton (2009c), have covered larger parts of the DCA. Drilling investigations, groundwaterlevel monitoring and groundwater, spring and surface water sampling and water balance modelling have indicated the potential prospectivity of the aquifer based on the water quality, saturated thickness, bore yields and magnitude in components of the groundwater balance. The spatial aquifer attribution dataset generated from analyses described in Section 3.1 was used to collate lithology and stratigraphy data for groundwater bores in the DCA and some adjacent hydrogeological units across the unconfined part of the aquifer between Flying Fox Creek and the Goyder River, north-east of the catchment boundary. These analyses were used to: (i) evaluate available groundwater-level data for the aquifers, (ii) identify a subset of bores in the aquifer outcrop where it is shallow and unconfined, and (iii) identify a subset of bores discretely screened to the DCA and underlying hydrogeological units that were suitable for groundwater sampling (particularly for environmental tracers) by evaluating their construction details. A groundwater sampling campaign for chemistry and environmental tracers in groundwater was conducted to better characterise and quantify groundwater flow processes in the unconfined part of the aquifer between Flying Fox Creek and the Goyder River (Figure 4-4). The sampling program occurred during the dry season in August 2022. Specific details of the methods of groundwater sampling and field observations are described in sections 4.2.3 to 4.2.6. Springs were also sampled to confirm previous investigations of the water sources for key discrete springs associated with the DCA (see Section 4.2.11). Figure 4-4 Target area for groundwater sampling across the outcropping area of the Dook Creek Aquifer between Flying Fox Creek and the Goyder River To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left map inset with red polygon indicates the location and geographic extent of the map figure within the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008) Geological faults data source: Department of Industry, Tourism and Trade (2010) A map of a mountain Description automatically generated 4.2.3 General chemistry and environmental tracers General chemistry Characterising the chemical composition of groundwater is important for evaluating its suitability for different uses (i.e. stock water or irrigation). In addition, it provides a basis for understanding the recharge sources to aquifers and the nature and scale of groundwater flow processes including, inter-aquifer, aquifer–aquitard and groundwater–surface water connectivity. The chemical composition, salinity and acidity or alkalinity in water samples result from processes occurring in different components of the hydrological cycle. For example, dissolved constituents including salts and metals (ions) are present in water as a result of the cycling of salts with a contemporary marine origin, connate salts with a historical marine origin, and salts and metals resulting from either terrestrial hydrochemical weathering or subsurface water-rock interactions such as carbonate dissolution. Additional chemical reactions which alter the chemical composition of water, such as redox reactions and ion exchange, also occur during infiltration of rainfall and along flow paths with increasing aquifer depth. In this assessment, two water sources (groundwater and surface water) were sampled at multiple locations across different parts of both the CLA and DCA. Their ionic compositions were analysed to: (i) characterise the composition and sources of ions for each water type, (ii) characterise the suitability of different water sources for different uses, and (iii) assist in understanding the nature and scale of groundwater flow processes in each aquifer. Environmental tracers Environmental tracers (tracers) are widely used when coupled with traditional hydrogeological data (i.e. groundwater level, groundwater chemistry, aquifer type, aquifer geometry and aquifer properties) for developing a hydrogeological conceptual model (Cook and Böhlke, 2000), particularly in data-sparse regions. Tracers have proved highly invaluable in previous hydrogeological investigations in northern Australia when seeking to develop or refine hydrogeological conceptual models through characterising and where possible quantifying groundwater flow processes (i.e. recharge, throughflow and discharge) (Deslandes et al., 2019; Harrington et al., 2014; Lamontagne et al., 2021; Smerdon et al., 2012; Taylor et al., 2018b; Taylor et al., 2017; Turnadge et al., 2018). A wide range of tracers exist to characterise: • infiltration conditions (i.e. presence or absence of excess air, temperature of water during recharge, evaporation before or during recharge) • water–rock interactions (i.e. the dominant aquifer(s) which water has travelled through) • groundwater–surface water interactions (i.e. groundwater recharge from, or discharge to, creeks) • groundwater discharge (i.e. via plants as transpiration, inflow to creeks, seepage through exposed sediments and rocks, or as submarine groundwater discharge to the marine environment) • inter-aquifer and aquifer–aquitard connectivity (i.e. mixing of different groundwater sources between multiple aquifers or aquifers and aquitards). In addition to characterising and conceptualising groundwater systems, the sampling and application of multiple tracers under appropriate circumstances (i.e. using specialised sampling techniques and at groundwater infrastructure with adequate bore construction) can be used to quantify groundwater flow processes. A common application of tracers is to characterise and quantify mean residence times (MRTs) for different groundwater flow processes. Residence times for groundwater can vary significantly (i.e. a few years to tens of thousands of years) depending on the scale and physical properties of different hydrogeological units as well as changes in hydraulic gradients. In this assessment, MRTs derived from interpreting tracers have been adopted as the preferred measure of the cumulative time groundwater has spent in aquifers since the time of recharge. Briefly, stable hydrogen and oxygen isotopes of water (2H and 18O) composition of groundwater are useful for identifying water sources and recharge mechanisms for an aquifer (Dogramaci et al., 2012; Harrington et al., 2002). The strontium isotope ratio (87Sr/86Sr) of groundwater can be used to identify the aquifer material (host rocks) which groundwater has spent most of its time flowing through (Harrington and Herczeg, 2003; Raiber et al., 2009). Concentrations of anthropogenic gas tracers – including chlorofluorocarbons (CFCs), sulfur hexafluoride (SF6) and bromotrifluoromethane (Halon 1301 or H1301) – can be used to characterise residence times for groundwater ranging from years to decades. The concentrations of these anthropogenic gases are known from temporal atmospheric monitoring. They are soluble in water and represent the air– water equilibrium at the time of recharge. Tritium (3H) and carbon-14 (14C) are radioactive isotopes with half-lives of 12.32 and 5730 years, respectively (Godwin, 1962; Lucas and Unterweger, 2000). They are present in the atmosphere both naturally from the interaction of nitrogen with cosmic rays and by release from nuclear weapons testing in the mid-1900s (Kalin, 2000). Given the respective half-lives of 3H and 14C and known concentrations in the atmosphere, they can be used to characterise MRTs in groundwater of up to about 70 years and 40,000 years, respectively. The noble gases of helium (He), neon (Ne), argon (Ar), krypton (Kr) and xenon (Xe) are chemically inert, so they accumulate over time in groundwater. In particular, helium-4 (4He) is a stable isotope of He and is produced by the decay of uranium and thorium radioactive isotopes naturally occurring in geological units, so it provides a good indicator for very long groundwater residence times (Gardner et al., 2011; Mahara et al., 2009; Solomon, 2000). Neon, argon, krypton and xenon are also useful tracers for inferring conditions during recharge, including recharge temperatures and indications of entrapped air (excess air) during localised recharge events through the unsaturated zone. They can also be useful for indicating entrapped excess air from fluctuations in the groundwater level due to groundwater recharge or extraction from groundwater pumping. 4.2.4 Groundwater sampling The geographical coordinates of each bore site were recorded using a handheld global positioning system (GPS). Where possible, the water level was manually observed using a portable electronic submersible Solinst 101 Water Level Meter. On some occasions, downhole equipment such as water supply pumps prevented a water level being obtained, so the most recent historical measurement was obtained from the NT groundwater database (Department of Environment Parks and Water Security, 2014a). The water level and bore construction data (i.e. bore diameter, casing length and screened interval) were used to calculate the total purge volume at each site (usually being three standing water columns) or until field parameters (see Section 4.2.6) stabilised. Groundwater was carefully purged using three different types of submersible pumps including a stainless steel 230V Grundfos MP 1 / Redi-Flo2, a 12 V battery-driven Proactive stainless steel Monsoon and a 12 V battery-driven Proactive plastic Monsoon (Figure 4-5) or using an existing pump installed for water supply. Purging was carried out to ensure water samples obtained from the aquifer were representative and not just samples of stagnant water sitting in the bore casing. On some occasions, some bores needed less purging due to recent pumping to turkey nests, tanks or troughs. Sampling was then undertaken under low-flow rates to ensure samples were carefully collected according to a variety of sampling protocols. Figure 4-5 Inserting a battery-driven 12-volt Proactive plastic submersible Monsoon pump into a bore casing 4.2.5 Sample collection Groundwater sampling was conducted by initially establishing a gas-tight connection to either the discharge hose on existing headworks of equipped sub-artesian water supply bores or the discharge hose on submersible pumps deployed into the casing of sub-artesian monitoring bores. Where submersible pumps were used on sub-artesian monitoring bores, the pump was lowered into the casing and installed at a minimum of 10 m below the measured standing water level. In the case of headworks of equipped sub-artesian water supply bores, a gas-tight connection was established by securing copper or nylon tubing using either Swagelok fittings or Philmac connections directly to the headworks. For sub-artesian monitoring bores, nylon tubing was used A person holding a rope Description automatically generated at the discharge hose on the submersible pump. These gas-tight connections are crucial for preventing the contamination of environmental tracer samples with the atmosphere, as well as any soft plastic components from hoses or existing pumps that may contain CFCs. Groundwater samples were collected for general chemistry (pH, EC and alkalinity) as well as major and minor ions and environmental tracers including 2H,18O, 87Sr/86Sr, radon-222 (222Rn), 3H, CFCs (CFC-11, CFC-12 and CFC-113), SF6, H1301, 13C, 14C and noble gases (He, Ne, Ar, Kr, Xe). Samples for major and minor ions were collected by first filling a bulk 1 L container. Duplicate samples were filtered using a syringe and 1.2/0.45 μm Acrodisc syringe filters into 125 mL polyethylene terephthalate plastic bottles. One bottle was acidified with a few drops of nitric acid (HNO3) for cations. Samples of water for 2H and 18O were collected in duplicate and placed in a 28 mL gastight glass bottle (McCartney bottle) to prevent evaporation. Strontium isotope samples were collected and filtered with 1.2/0.45 μm Acrodisc syringe filters and placed into 250 mL polyethylene terephthalate plastic bottles. Radon samples were collected with the nylon hose inserted to the base of the 1.25 L polyethylene terephthalate plastic bottle and bottom-filling the bottle until overflowing (Figure 4-6a). A subsample of 222Rn was later extracted into mineral oil (Figure 4-6b) following the method outlined in Leaney and Herczeg (2006). Tritium samples were collected in 1 L polyethylene terephthalate plastic bottles under gently flowing conditions and capped without a head space to avoid contact with the atmosphere. (a) (b) Figure 4-6 (a) Collecting a groundwater sample in the field for radon-222 and (b) extracting the radon-222 from the water sample into mineral oil SF6 samples were collected following the protocol outlined by GNS Science (https://www.gns.cri.nz/assets/Laboratory-files/Tritium-and-Water-Dating-Laboratory/Nov-22- SF6-Sampling-Instructions.pdf). Briefly this involved inserting the nylon discharge hose into the bottom of a 1 L amber glass bottle and then bottom-filling the bottle placed in a 10 L stainless steel bucket (Figure 4-7). Samples were collected using a gentle pumping rate, and bottles were reverse- or bottom-filled until the 10 L bucket overflowed to eliminate gas exchange with the atmosphere. Figure 4-7 Collecting a groundwater sample for sulfur hexafluoride (SF6) Samples for CFCs were collected following the protocol outlined by GNS Science (https://www.gns.cri.nz/assets/Laboratory-files/Tritium-and-Water-Dating-Laboratory/Nov-22- CFC-Sampling-Instructions.pdf). Briefly this involved placing the nylon discharge hose into the bottom of a 125 mL glass bottle placed inside a 10 L steel bucket. Samples were collected using a gentle pumping rate, and bottles were reverse- or bottom-filled until the 10 L bucket overflowed. Samples were collected in triplicate and capped under water to prevent exposure to the atmosphere. Samples for 14C/13C were collected in 500 mL polyethylene terephthalate plastic bottles. Duplicate samples for dissolved noble gases were collected in copper tubes following Weiss (1968). Briefly, this involves creating a gas-tight connection between the copper tube and the discharge hose connected to either the existing headworks on an equipped water supply bore or the submersible pump on a monitoring bore. The copper tube is gently flushed to remove any bubbles and then back pressure is gently applied using a flow regulator before the copper tube is clamped at each end without trapping any air (Figure 4-8). Figure 4-8 Preparing to collect a groundwater sample for dissolved noble gases using a copper tube 4.2.6 Field parameters Field parameters were monitored and recorded during purging and sampling under gently flowing conditions. Parameters were measured using a calibrated YSI Pro Plus multiparameter meter which simultaneously measures EC, pH, temperature and dissolved oxygen. Alkalinity was measured in the field using a HACH digital titration kit. The presence of H2S was noted when detected by odour during sampling as was the number of gas bubbles. 4.2.7 Analytical methods Major and minor ion chemistry and EC, pH and total alkalinity were measured by the CSIRO Analytical Services Unit (ASU; Waite Campus, Adelaide). Major and minor cations were analysed by inductively coupled plasma optical emission spectrometry (ICPOES), and major anions were analysed by ion chromatography. Measurement precision of most analytes by ICPOES is ±0.05 mg/L. The major anions analysed included Cl, HCO3, SO42, bromide (Br) and fluoride (F). The major and minor soluble cations measured included Ca, potassium (K), Mg, Na, aluminium (Al), arsenic (As), boron (B), cadmium (Cd), cobalt (Co), chromium (Cr), copper (Cu), iron (Fe), manganese (Mn), molybdenum (Mo), nickel (Ni), phosphorus (P), lead (Pb), antimony (Sb), selenium (Se), silicon (Si), Sr and zinc (Zn). Deuterium and 18O were analysed by the Stable Isotope Geochemistry Laboratory in Earth Sciences at the University of Queensland. Measurements by the University of Queensland were carried out using a dual inlet isotope ratio mass spectrometer (IRMS) and reported with a precision of ±2‰ and ±0.1‰ for 2H and 18O, respectively. Radon-222 activity was measured by liquid scintillation counting at CSIRO’s Noble Gas Facility, Waite Campus, Adelaide, using an LKB Wallac Quantulus liquid scintillation counter with a detection limit approximately 0.003 Bq/L and a precision of 3 to 5%. Tritium was measured by electrolytic enrichment and liquid scintillation counting at GNS Science New Zealand and reported with a detection limit of 0.025 TU (the most accurate tritium dating lab globally) (Morgenstern and Taylor, 2009). CFC samples were also measured by GNS Science New Zealand. CFC samples (CFC-11, CFC-12, CFC-113) were measured by gas chromatography with an electron capture detector after quantitative stripping from water aliquots of a defined volume under a stream of high-purity nitrogen following Plummer and Busenberg (1992). Sulfur hexafluoride and H1301 were also measured by gas chromatography with electron capture detection similar to that outlined in Busenberg and Plummer (2000) with a measurement precision of ±0.05 and ±0.2 fmol/kg–1 for SF6 and H1301, respectively. Carbon-14 samples were measured by single stage accelerator mass spectrometry (AMS) at the Australian National University (ANU), Canberra. Samples were prepared by first precipitating the dissolved inorganic carbon from 1 L of groundwater as strontium carbonate (SrCO3) under alkaline conditions (pH > 11). The SrCO3 precipitate was then acidified and purified cryogenically into aliquots of carbon dioxide (CO2) for measurement by AMS (Fallon et al., 2010). Values were reported with a measurement precision of ±0.2% pMC. Samples for 13C were analysed using precipitated SrCO3 and freeze-drying samples under vacuum and immediately weighing approximately 1.6 mg into a 5 mL exetainer vial. The vial was flushed with helium and acidified with 0.2 mL of >101% phosphoric acid (H3PO4). The CO2 was measured on a Sercon 20-20 IRMS in continuous flow mode. The precision was 0.1‰ and was based on duplicate/triplicate measurements of unknown groundwater samples. Samples for dissolved noble gases were analysed at CSIRO’s Noble Gas Facility, Waite Campus, Adelaide. Dissolved gases were first separated from water using an offline extraction system. Gas subsamples were then analysed on a fully automated facility which conducts: (i) drying of all gases, (ii) raw gas analysis to determine the dinitrogen gas (N₂) to Ar ratio on a quadrupole mass spectrometer, (iii) separation of noble gases from reactive gases using a variety of reactive getter systems, (iv) separation of the noble gases by cryogenic techniques, and (v) measurement of gas volumes and their isotopic composition using a spinning rotor gauge, quadrupole mass spectrometers, and a high-resolution Helix MC noble gas mass spectrometer (Suckow et al., 2019). All noble gas concentrations are given in cubic centimetre of gas per gram (ccSTP/g where STP stands for standard temperature and pressure and the gram is related to the weight of the sample). Post processing of all noble gas measurements was done using the LabData laboratory information management and database system (Suckow and Dumke, 2001). 4.2.8 Interpretation of general chemistry Multiple approaches can be used for interpretating the ionic composition of a water sample, from simple ion plots and piper diagrams (Peeters, 2014; Piper, 1944) to more complex geochemical speciation modelling (Parkhurst and Appelo, 2013). In this assessment, the charge balance error (CBE) was initially calculated for each water sample to evaluate the viability of each sample with respect to analytical errors or preservation of the sample from field collection. Where the CBE was acceptable (±5%), ion plots were derived and the ionic composition of groundwater and surface water compared to the seawater dilution line to evaluate if the sources of ions in water were from rainfall of marine origin or from other sources (e.g. water–rock interactions). Piper diagrams were used to plot groundwater and surface water samples to evaluate the hydrochemical evolution of groundwater to characterise the nature (i.e. inter-aquifer, aquifer–aquitard and groundwater– surface water connectivity) and scale of groundwater systems (i.e. local, intermediate or regional). Spatial maps of water quality were also derived to look at the spatial evolution of groundwater in the CLA, DCA and adjacent aquifers. RockWare Inc’s AqQA program was used to generate piper diagrams and to assess the salinity hazard for different water samples. PHREEQC (Parkhurst and Appelo, 2013) was used to: (i) determine the saturation indices (SI) for groundwater with respect to calcite, gypsum, and dolomite and (ii) ascertain the chemical speciation of aqueous CO2 and HCO3 in relation to the total dissolved inorganic carbon (TDIC) in groundwater. The latter was necessary for interpreting 14C values in groundwater for the potential addition of dead carbon from either mixing with atmospheric CO2 during recharge or isotopic exchange during flow through the carbonate aquifers, and to derive an MRT. 4.2.9 Interpretation of environmental tracers A range of methods exist to interpret different environmental tracers observed in groundwater including radioactive isotopes such as 14C, accumulating tracers such as 4He, or those with known temporal concentrations in the atmosphere such as CFCs and SF6.This includes the use of multiple one- and two-dimensional analytical solutions or lumped parameter models (LPMs) that can be parametrised using aquifer thickness and geometry (Appelo and Postma, 1996; Małoszewski and Zuber, 1982; Vogel, 1967; Zuber, 1986). In some cases, more complex numerical groundwater flow and solute transport models have been used where detailed information on aquifers is known (Reilly et al., 1994; Salamon et al., 2006; Sanford, 2011; Voss and Wood, 1994). In this assessment, existing hydrogeological conceptual models for the CLA (Bruwer and Tickell, 2015; Jolly et al., 2004; Karp, 2008; Knapton, 2004; 2009c; 2020; Tickell and Bruwer, 2018) and DCA (Knapton, 2009c; Williams et al., 2003; Zaar and Tien, 2003) were validated or refined by interpreting measured tracer concentrations in groundwater from shallow unconfined parts of both aquifers (along topographic gradients and previously reported hydraulic gradients) and their adjacent aquifers, and comparing them to concentrations in mapped discharge areas (springs and seepage zones along streamlines). In addition, measured tracer concentrations in groundwater were used for estimating MRTs and areal or spatially averaged recharge rates for a range of aquifer thicknesses. Measured tracer concentrations in springs and surface water were used to characterise groundwater–surface water connectivity. 4.2.10 Groundwater residence times and recharge rates Groundwater residence times Residence times for groundwater flow and mixing were estimated by interpreting measured tracer concentrations with LPMs. The LPMs tested and evaluated in this Assessment were: • the piston flow model, which assumes flow from the recharge to the discharge area with no mixing (see Figure 4-9a, which is adapted from Jurgens et al. (2012)) • the exponential model, in which residence times for groundwater are stratified and increase logarithmically from the watertable to the base of the aquifer (Appelo and Postma, 1996; Vogel, 1967) (see Figure 4-9b, which is adapted from Jurgens et al. (2012)) • a binary mixing model, which accounts for possible mixtures between different water sources with both short (i.e. a few years) and long (i.e. hundreds to thousands of years) residence times. The LPMs were used to predict both the MRT for a given tracer sample and the distribution of residence times through the entire aquifer. They used a few key assumptions and parameters: • Aquifer geometry was simplified based on the saturated thickness of aquifers estimated from measurements of groundwater levels at the time of field sampling and the base of the aquifer determined from lithological logs that were incorporated on to hydrogeological cross-sections (see Section 6.1.2). • Elevations of groundwater bores were taken from either surveyed elevation data or use of the 1- second SRTM-derived hydrologically enforced digital elevation model. • Mean annual recharge temperature was based on observations from the Bureau of Meteorology weather stations at Mataranka for the CLA and Bulman for the DCA (see Section 0). • Groundwater salinity was based on the mean total dissolved solids (TDS) estimated from the analyses of major and minor ions in groundwater (see Section 5.2). • Transient atmospheric measurements of CFCs, SF6, H1301 and tritium in precipitations were obtained from existing published databases (Bullister, 2017; Newland et al., 2013). In this assessment, the predicated residence time distribution outputs from LPMs served as a basis for determining residence times from the tracers. Where multiple tracers are sampled at a given site, at least three different approaches exist to determine MRTs: • Based on some initial analysis, the most reliable tracer is identified and a simple model with only one free parameter (such as the piston flow model or the exponential model) is used to derive a MRT from that single tracer. • A MRT is derived individually from an LPM for all tracers that are considered unaffected by any secondary processes, and either the range of MRTs is reported or a mean calculated. • Tracers are corrected for secondary effects such as excess air where possible and the complexity of the model is increased (e.g. by including a time lag for the unsaturated zone) until all tracers (or a selected more reliable subset of all tracers) can be reasonably modelled by one single LPM to obtain a single MRT from multiple tracers. This assessment used a combination of these approaches based on an evaluation of the measured concentrations provided from the various analytical laboratories. Figure 4-9 Schematic cross-section representations of advective flow in the (a) piston flow model and (b) exponential model in idealised unconfined aquifers Grey dots represent a sandstone aquifer. Nabla (inverted delta) represents the depth of the watertable below the land surface. Grey lines represent the evolution of groundwater flow lines from their origin at the watertable to the base of the aquifer and side of the cross-section. Figure source: Figure (a) is adapted from Figure 1 in Jurgens (2012) and Figure (b) is adapted from Figure 2 in Jurgens (2012) Groundwater recharge Recharge rates were derived using multiple approaches: (i) the CMB method was employed where appropriate (i.e. where certain assumptions can be met) as described in Section 3.6, and (ii) environmental tracers were applied where appropriate (i.e. adequate bore construction, valid analytical results, available groundwater-level and aquifer property information). Unconfined aquifer – piston flow model Some groundwater samples were collected from shallow unconfined portions of the aquifers. Long-term mean annual recharge rates were estimated based on the assumption of mostly vertical advective flow. A key assumption here was that the depth of the sample is small relative to the total aquifer thickness. Accordingly, the recharge rate R (mm/year) can be given by the following equation from Cook and Böhlke (2000): 𝑅𝑅 = 𝑧𝑧𝑧𝑧 𝑡𝑡 (9) where z is the depth of the sample below the watertable, taken as the mid-point of the screen (m), ε is aquifer porosity and t is the MRT (years) based on the concentration of 3H, for example, at depth. Mean residence times derived from tracers were used as input into Equation (9). Unconfined aquifer – exponential model Some groundwater samples were collected from depths greater than 20 m below the watertable; therefore, the assumption of only vertical advective flow may not be valid, particularly with increasing depth below the watertable. In this case, the exponential model was considered to estimate long-term mean recharge rates where groundwater flow is initially vertical, but as the depth below the watertable increases, it eventually becomes horizontal. That is, the conceptual model is that the MRT exponentially increases with depth below the watertable (Figure 4-9b Vogel, 1967). Accordingly, MRTs for advective flow using 3H, for example, were derived using the LPMs based on the use of the convolution of a time-varying input signal with a response function (Małoszewski and Zuber, 1982). That is: 𝐶𝐶𝑜𝑜 𝑜𝑜𝑜𝑜(𝑡𝑡) = 􀶱 𝐶𝐶𝑖𝑖𝑖𝑖(𝑡𝑡ʹ) ⋅ 𝑔𝑔(𝑡𝑡 − 𝑡𝑡ʹ) ⋅ 𝑒𝑒−𝜆𝜆(𝑡𝑡 – 𝑡𝑡 ʹ)𝑑𝑑𝑑𝑑 ʹ 𝑡𝑡 −∞ (10) where Cout(t) is the time-dependent 3H concentration at a given sampling site, Cin(t’) is the 3H time series in precipitation, t – t’ is the time difference between the concentration in precipitation and groundwater sampling at the bore, and λ is the radioactive decay constant of 3H. For the exponential model, precipitation has a weight function describing the exponential decreasing contributions with time: 𝑔𝑔(𝑡𝑡 − 𝑡𝑡ʹ) = 1 𝜏𝜏 ⋅ 𝑒𝑒𝑒𝑒𝑒𝑒 􁉆 −(𝑡𝑡 − 𝑡𝑡ʹ) 𝜏𝜏 􁉇. (11) 4.2.11 Groundwater–surface water interactions Cambrian Limestone Aquifer Spring and surface water sampling was conducted in conjunction with the groundwater sampling (see Section 4.2.2) to better characterise the sources of discharge to the upper Roper River, its major tributaries and the Mataranka Springs Complex. The target area for surface water sampling was the headwaters of the Roper River, the major regional groundwater discharge zone for the CLA in the Roper catchment (Bruwer and Tickell, 2015; Jolly et al., 2004; Karp, 2008; Knapton, 2020; Yin Foo, 2002). The source of discharge is from two regional flow paths that congregate near Mataranka: the ‘Daly flow path’ from the north and the ‘Georgina flow path’ from the south, that congregate near Mataranka and then mix with both local and intermediate-scale flow within and near the aquifer outcrop (Bruwer and Tickell, 2015; Department of Environment and Natural Resources, 2017; Karp, 2008; Knapton, 2020; Tickell, 2016). While major springs in Elsey National Park (Rainbow Spring and Bitter Spring) are obvious groundwater discharge features in the landscape, groundwater discharge also occurs via numerous smaller seeps and directly through the river bed of the Roper River and its tributaries. In addition, remote sensing estimates of evapotranspiration suggest the whole land surface within and surrounding the park is a groundwater discharge zone via tree transpiration and shallow watertable evaporation (Crosbie and Rachakonda, 2021). Review of historical hydrometric data Historical data for the source and magnitude of baseflow to the Roper River was assessed by reviewing end-of-dry-season discharge surveys made by the NT Government along the river since 2003. Hydrographic data, including stream stage, streamflow and stream water quality, were collated for various locations of regular monitoring around the upper Roper River and are available via the NT Water Data Portal (Northern Territory Government, 2023). Several historical hydrographic surveys of the upper Roper River were also reviewed and evaluated (Kerle and Cruickshank, 2014; Schult, 2018; Schult and Novak, 2017; Wagenaar and Tickell, 2013; Waugh and Kerle, 2014). Longitudinal synoptic surface water and spring sampling campaign A longitudinal synoptic surface water and spring sampling campaign was conducted in October 2022 in an area spanning from Roper Creek in the west to just downstream of the Roper River and Elsey Creek junction in the east (Figure 4-10). Surface water samples were collected by boat at most locations using a submersible plastic pump deployed about 30 cm below the water surface in areas of flowing water. Due to access constraints, springs and some tributaries were sampled from land using a 3 m extension pole to ensure the pump was well submersed in flowing water. At each sampling location, chemical parameters including pH, EC, temperature and dissolved oxygen were simultaneously measured under flowing conditions using a using a calibrated YSI Pro Plus multiparameter meter. Water samples were collected for major well as major and minor ions and environmental tracers including: 2H,18O, 222Rn, 3H, and noble gases (He, Ne, Ar, Kr, Xe). The survey was conducted to: (i) further characterise the spatial occurrences of groundwater discharge, (ii) further determine whether any aquifers other than the CLA contribute to groundwater discharge to the Mataranka Springs Complex and upper Roper River, and (iii) identify key tree species that are phreatophytes in the regional groundwater discharge zone (see Section 4.2.12). Figure 4-10 Target area for spring and surface water sampling in the groundwater discharge zone for the Cambrian Limestone Aquifer in the Roper catchment To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left map inset with red polygon indicates the location and geographic extent of the map figure within the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008) Geological faults data source: Department of Industry, Tourism and Trade (2010) Groundwater discharge zone data source: Department of Environment and Natural Resources (2017) Spring data source: Department of Environment Parks and Water Security (2014c) Dook Creek Aquifer Spring water sampling was conducted in conjunction with the groundwater sampling (see Section 4.2.2) to further characterise and validate the sources of discharge via discrete springs associated with the DCA. Previous investigations by Zaar and Tien (2003) indicate that spring-fed reaches of the Mainoru, Wilton and Goyder rivers sustain dry-season baseflows of between 10 and 100 L/second at the end of the dry season. The target area for spring water sampling was along the Wilton and Goyder rivers where spring complexes are more abundant (Figure 4-11). Sampling was undertaken in August 2022 at the same time as the groundwater sampling campaign described in Section 4.2.2. Figure 4-11 Target area for spring and surface water sampling across the outcropping and subcropping areas of the Dook Creek Aquifer To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left map inset with red polygon indicates the location and geographic extent of the map figure within the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008) Geological faults data source: Department of Industry, Tourism and Trade (2010) Groundwater discharge zone data source: Department of Environment and Natural Resources (2017) Spring data source: Department of Environment Parks and Water Security (2014c) 4.2.12 Tree water use Unless stated otherwise the following contextual and methods summary comes from a companion technical report by Duvert et al. (2023) on characterising tree water sourcing at the Mataranka Springs Complex. Elsey National Park, and small areas adjacent the park to the east, south and west, is an area of significant groundwater discharge from the CLA to the Roper River (Jolly et al., 2004; Karp, 2008; Lamontagne et al., 2021; Watson Resource Consulting, 1999; Yin Foo, 2002) (Figure 4-10). According to the Australian Groundwater Dependent Ecosystems Atlas (Doody et al., 2017) and recent remote sensing studies that show where evapotranspiration rates exceed rainfall (Crosbie and Rachakonda, 2021), there is a high probability that the extensive vegetation in Elsey National Park and the riparian vegetation along adjacent creek lines rely on groundwater from the CLA. However, the exact proportion of this area of groundwater dependency, the sources of water for trees or other vegetation types, and the species of vegetation most dependent on groundwater are unclear. The potential occurrence of a groundwater depth threshold beyond which vegetation rely on groundwater is also of particular interest given it could be a key control on groundwater transpiration. Both the area of groundwater dependency and potential depth threshold are important factors in further characterising and quantifying groundwater evapotranspiration from the CLA. Targeted investigations of tree water uptake were undertaken in Elsey National Park (Figure 4-10). The work involved using the stable isotopes of water (oxygen-18, 18O and deuterium, 2H) to trace sources of tree water and measuring soil matric potential (a measure of the availability of soil water for plants) and pre-dawn leaf water potential (a measure of plant water stress). The main objective was to evaluate the contribution of different water sources (soil water at different depths, groundwater from the CLA) to tree water uptake along a groundwater depth gradient. This involved two sampling campaigns during two consecutive dry seasons. The first sampling campaign (Phase I) was conducted, from 29 September 2021 to 15 October 2021, at the end of the dry season when trees are most likely to rely on groundwater. The aim of Phase I was to assess tree water sources along a groundwater depth gradient, from a site within riparian vegetation and a very shallow depth to groundwater (<1 m) by Elsey Creek, to a site of upland savanna woodland with a much greater depth to groundwater (>20 m) south-west of Elsey National Park. Sites were selected based on their proximity to existing bores, so that the watertable depth and isotopic signature of the groundwater end-member (source) could be determined at each site. The second sampling campaign (Phase II) was conducted the following dry season from 19 September 2022 to 22 September 2022. It focused on site 4 from Phase I where the tree species identified included both groundwater and soil water users. The aim of Phase II was to characterise the deep soil profile at this site and to conduct a more in-depth investigation at the transition between savanna woodlands and the seasonal swamp forest. The watertable is likely to be shallower in the swamp than in the savanna (the elevation difference between the savanna and swamp locations is approximately 5 m). 4.3 Numerical groundwater flow modelling Unless stated otherwise the following contextual and methods summary comes from a companion technical report on groundwater flow modelling by Knapton et al. (2023). The CLA and DCA are the two largest, most productive and potentially most promising aquifers within and beneath the Roper catchment for future groundwater-based development (Bruwer and Tickell, 2015; Tickell and Bruwer, 2018; Zaar and Tien, 2003). Parts of these aquifers coincide with land recently identified as potentially suitable for agricultural intensification (Thomas et al., 2022). To better understand the potential opportunities and risks associated with future groundwater resource development, water resource modelling was undertaken to provide new information that could potentially help inform future water resource planning, investment, and management across parts of the CLA and DCA considered potentially suitable for groundwater-based irrigation. CSIRO engaged CloudGMS Pty Ltd to run, process and evaluate the results of multiple future hypothetical groundwater development and climate scenarios using the two different existing finite element groundwater models of the CLA and DCA. The specific objectives of this modelling investigation were to assess the possible effects of changes in rainfall and potential evaporation and future hypothetical development of groundwater resources on the availability of groundwater for environmental requirements of surface water resources (e.g. environmental receptors such as the upper Roper River, its major tributaries and the Mataranka Springs Complex in the Mataranka region, and the Wilton and Mainoru rivers, Flying Fox Creek and various discrete springs in the Bulman region) and existing licensed water users in key parts of the Roper catchment. To do this, eight scenarios were simulated using both the CLA and DCA groundwater models. The climate data used as input to the groundwater models was sourced from climate analyses undertaken in a companion technical report on the climate of the Roper catchment (McJannet et al., 2023). The locations for future hypothetical groundwater development are outlined in Section 4.3.6. This section provides a succinct summary of the methods used for the numerical modelling, and Section 6.3 provides a summary of key results. Further details on model construction, calibration, sensitivity analysis and predictions are provided in Knapton et al. (2023). 4.3.1 Model descriptions Cambrian Limestone Aquifer Groundwater flow in the CLA is modelled using a calibrated three-dimensional finite element groundwater model referred to as DR2 as detailed in Knapton (2020). The DR2 groundwater model covers an area of 159,000 km2 and represents the unconfined and confined areas of the CLA in the Daly Basin, the Wiso Basin to the south and the Georgina Basin to the south-east (Figure 4-12). The groundwater model was developed using finite element methods in the FEFLOW simulation code (Diersch, 2008) and consists of three layers. The CLA groundwater system is conceptually characterised as an equivalent porous medium. This simplification allows for the development of a more manageable and computationally efficient model while still capturing the essential characteristics of the groundwater system using calibrated regional aquifer parameters to reproduce the observed groundwater levels and discharge to the rivers. Recharge is applied to the top slice of the groundwater model according to recharge zones based primarily on the mapped surface geology (Knapton, 2005; Knapton, 2006). The time series recharge flux estimates of recharge have been determined using MIKE SHE (DHI, 2008) and more recently the LUMPREM utility (Doherty, 2020). These are process-based models and include an estimate of preferential or bypass flow. The groundwater model includes boundary conditions (BCs) that define the interaction between the rivers and the groundwater system. Discharge from the rivers is implemented using Cauchy or transfer BCs (like river cells in ModFlow). The groundwater model assumes that recharge and discharge to the rivers where they are in connection with the aquifer is relatively uniform between adjacent nodes. The transfer in/out rates vary spatially across the model domain, and areas of preferential recharge and discharge along the rivers are simulated by adjusting the transfer in/out parameters. Springs are not included in the model as discrete pathways because they are too poorly understood and at a scale too small to be adequately represented. Extraction for stock and domestic and horticultural use is simulated from the model domain via well BCs at model nodes. Pumping rates were applied as a steady-state value equal to the annual pumped volume for the bore converted to m3/day. The extent of the CLA forms the boundary of the model domain where the APV outcrop or occurs above the groundwater level is implemented as no-flow BC (see APV outcrop near Larrimah in Figure 2-11). The method used to calibrate the DR2 groundwater model is documented in detail by Knapton (2020). The calibration process used a combination of pilot points (Doherty, 2003) and PEST, an automated nonlinear parameter estimation code (Doherty, 2004). Pilot points were strategically placed throughout the model domain to allow for flexible spatial parameterisation, capturing the spatial variability of hydraulic properties (i.e. hydraulic conductivity, storage coefficient and transfer in/out). PEST was then used to iteratively calibrate the model by adjusting the pilot point parameters and recharge parameters to minimise the objective function (i.e. discrepancies between observed and simulated data). The observed data used to define the objective function included available historical groundwater levels (9151 head observations) in the CLA and discharge measurements for the Roper, Flora, Katherine and Douglas rivers. The results of the calibration process are detailed in Knapton (2020). Dook Creek Aquifer The groundwater flow model of the DCA in the Roper catchment is based on a calibrated threedimensional finite element groundwater model referred to as DC2 and is described in Knapton (2009c). The DC2 groundwater flow model encompasses an area of approximately 22,220 km2. It incorporates the entire extent of the unconfined areas of the Dook Creek Formation and includes the entire catchments of the Flying Fox Creek, Mainoru River, Wilton River, Guyuyu Creek and Goyder River (Figure 4-12). The groundwater model was developed using finite element methods in the FEFLOW simulation code (Diersch, 2008) and consists of three layers. Like the CLA, the DCA groundwater system is conceptually characterised as an equivalent porous medium. This simplification allows for a more manageable and efficient model while still capturing the essential characteristics of the groundwater system using calibrated regional aquifer parameters to reproduce the observed groundwater levels and discharge to the rivers. Based on the conceptual model, the DCA groundwater system is unconfined to the north-west of the Central Arnhem Highway and confined beneath the Limmen Sandstone to the south-east. The extent of the confined region of the DCA model was arbitrarily defined using the subcatchments of the rivers which source flows from the Dook Creek Formation. To represent recharge and evapotranspirational losses, the DC2 groundwater model uses areal flux BCs applied to the top slice of the model using a combination of element distributions and functions. 90 | Hydrogeological assessment of the Roper catchment The DC2 groundwater model includes BCs that represent the discharge to the northern tributaries of the Roper River, implemented using the Cauchy or transfer BCs (Diersch, 2008). The transfer BC describes a reference hydraulic head which has an imperfect hydraulic contact with the groundwater body caused by a colmation layer (related to the river bed conductance). Pumping bores for stock and domestic and horticultural use were implemented using multilayer well BC. Pumping rates were applied as a steady-state value equal to the annual pumped volume for the bore converted to cubic metres per day. Given that there has been very little development of the DCA groundwater resources in the Roper catchment, they are only relevant to scenario modelling to examine the effects of future development. Discrete springs discharging from the DCA through the confining Limmen Sandstone (e.g. Lindsay Spring and Top Spring) were simulated using constant head BC. The extent of the DCA forms the boundary of the model domain and is implemented as no-flow BC. The calibration of the DC2 groundwater flow model is documented by Knapton (2009). The calibration was undertaken using the automatic calibration code PEST (Doherty, 2015). PEST uses the weighted sum of squared residuals to determine the objective function or ‘goodness of fit’ between the simulated response and the observed response (i.e. simulated and measured groundwater levels and groundwater discharge). PEST generates new model parameters as pilot point values (i.e. hydraulic conductivity, specific yield and transfer out distributions). The distribution of hydraulic conductivity, storage coefficient and transfer out pilot point values were interpolated to the model mesh. Figure 4-12 Location of the Roper catchment and its relationship to the groundwater systems of the Dook Creek Aquifer and the Cambrian Limestone Aquifer in the Daly, Wiso and Georgina basins Figure source: Figure 1 in Knapton et al. (2023) 4.3.2 Modelling time periods The CLA and DCA are both regional-scale to intermediate-scale groundwater flow systems, respectively, so it may take a few hundred to several hundreds of years, respectively, before the system re-establishes a quasi-equilibrium state, where the groundwater flow patterns stabilise following a change in model state, such as climate or development. Consequently, the impacts of development or future climate will in part depend on the timescale over which the modelling results are reported. For this reason, the results are reported over two different time periods. Location of the Roper catchment and its relationship to the groundwater systems of the Dook Creek Aquifer and the Cambrian Limestone Aquifer in the Daly, Wiso and Georgina basins. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au Time periods for reporting The first involved running both models (CLA and DCA) for 436 years to examine the impacts of changes in climate and increases in groundwater extraction at quasi-equilibrium conditions. This would represent a worst-case scenario noting there would be opportunities for management interventions and changes in government policy. The second time period over which the results were reported involved running both models to 2070 (~50 years). This was considered a pragmatic time period over which to consider the impacts of changes in climate and groundwater extraction because: (i) it is equivalent to more than twice the length the investment period of a typical agricultural enterprise, (ii) it is about the service life of an appropriately constructed groundwater production borefields, (iii) it is about five times the length of the current period over which NT water licences are assigned, and (iv) it is consistent with the time period over which future climate projections have been evaluated. For the purposes of this report the modelling results are reported over two timescales: • Quasi-equilibrium, designated by a prime symbol (′), where the models were run once for 436 years using 4 × 109-year sequences of the historical climate with only the results of the last 109-year sequence evaluated. • Projected 2070 model state, where the models were run for 159 years comprising a warm-up period of 109-years using the historical climate data (1910 to 2019) and thereafter using the 50- year historical and future climate and groundwater development inputs to evaluate the 2059 to 2069 representative conditions (nominally representing 2070 conditions). The 50-year historical and future climate and groundwater development inputs are comprised of an ensemble of 11 × 50-year climate sequences (based on a 50-year moving window over the 1910 to 2019 historical climate) run through the models, with the last 10 years evaluated (2070 conditions). The quasi-equilibrium timescale represents likely worst-case conditions if the climate and hypothetical development remained unchanged over the entire timescale (which is unlikely). The projected 2070 model states are more representative of a result that is consistent with the time frames over which projected future climate scenarios were evaluated in the companion technical report on climate (McJannet et al., 2023) and time frames over which water planning and investment decisions are made. The eleven 50-year climate sequence ensemble used for the 2070 reporting time period was chosen to encapsulate the effects of inter-decadal variability. 4.3.3 Historical and future climate datasets Establishing future climate scenarios The climate sequences used for the two different time periods for reporting (436 years and 159 years) were used to assess five different scenarios in both models: • Scenario AN – historical climate (1910 to 2019) and no groundwater development • Scenario A – historical climate (1910 to 2019) and current levels of development • Scenario B – historical climate (1910 to 2019) and hypothetical development • Scenario C – future climate (dry, mid and wet) and current levels of development • Scenario D – future climate (dry, mid and wet) and hypothetical levels of development. Future climate sequences were generated using methods outlined in the companion technical report by McJannet et al. (2023). Briefly, the potential impacts of future climate change were evaluated within a sensitivity analysis framework. Future climate sequences were evaluated using seasonal scaling factors from selected global climate models to scale the historical climate data. The entire climate sequence was then re-scaled using an annual scaling factor representing a percentage change from the long-term annual rainfall and potential evaporation (PET). This provided a method to generate future climate datasets based on the sensitivity to climate change (±10% rainfall and PET). Percentage change in long-term annual rainfall and PET were based on the 10, 50 and 90% exceedance values shown in McJannet et al. (2023). These datasets were applied to both groundwater models, resulting in changes to mean annual recharge. The following variations are used in the future climate scenarios C and D: • Scenario Cdry represents a reduction in the long-term mean annual rainfall by 10% and an increase in the long-term mean annual PET by 10%. • Scenario Cmid represents a reduction in the long-term mean annual rainfall by 2% and an increase in the long-term mean annual PET by 7.5%. • Scenario Cwet represents an increase in the long-term mean annual rainfall by 10% and an increase in the long-term mean annual PET by 5%. • Scenario D includes all of the variants of Scenario C of the climate change scenarios (i.e. Cdry, Cmid, Cwet) with the addition of proposed variations in future hypothetical groundwater development, nominally called scenarios Ddry, Dmid, Dwet. In considering changes to the hydrological regime (groundwater-level drawdown and groundwater flux) under different scenarios, all results were assessed relative to current conditions (i.e. Scenario AN, which is based on historical climate and current groundwater development). 4.3.4 Adopted model scenarios and naming conventions Quasi-equilibrium conditions (436 years) The first set of scenarios examine the impacts of hypothetical future development assuming the groundwater systems in both aquifers achieves quasi-equilibrium. The results from this set of scenarios are designated with a prime (′) to distinguish them from the results for the scenarios representing 2059 to 2069 (nominally 2070) conditions (see Section 4.3.5). The first two scenarios (Scenario A′N and Scenario A′) are historical climate scenarios based on the 109-year historical climate sequence repeated four times to achieve quasi-equilibrium. The historical climate is taken as the observed climate (rainfall and PET) for water years from 1910 to 2019. Scenario A′N assumes no groundwater development and Scenario A′ assumes the current levels of groundwater development (32 GL/year for the CLA and 0.1 GL/year for the DCA) described in Section 1.4.2 and shown in Table 4-2. Scenario A′N is used as the baseline against which the A′ and B′ assessments of relative change are made. The second scenario (Scenario B′) is also a historical climate scenario. Under this scenario, the current level and future level of groundwater development (35, 70 and 105 GL/year for the CLA and 6, 12 and 18 GL/year for the DCA) described in Table 4-2 are used. Scenario B′ is therefore used to assess potential water availability assuming future development under historical climate used in Scenario A′N and Scenario A′. The third scenario (Scenario C′) is a future climate and current development scenario. It was based on a 109-year climate series derived from scaling rainfall and PET described in Section 4.3.3. Scenario C′ assumes the current level of groundwater development described in Table 4-2. The fourth scenario (Scenario D′) is a future climate and future development scenario. It uses the same climate sequences as Scenario C′, and the current level plus future level of groundwater development as described in Table 4-2. To examine quasi-equilibrium conditions, modelled data from the last 109-year climate replica (based on the baseline period of 1910 to 2019) were evaluated for both aquifers for each scenario described below (A,B,C and D) and changes in the hydrological regime reported relative to Scenario A′N. Table 4-2 Summary of modelling scenarios A, B, C and D for both aquifers using 4 × 109 years historical climate and combinations of current and hypothetical future groundwater development SCENARIO CLA MODEL SCENARIO DCA MODEL A′N Historical climate and no development A′N Historical climate and no development A′ Historical climate and 32 GL/y current development A′ Historical climate and 0.1 GL/y current development B′35 Historical climate and current development + 7 × 5 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 35 GL/y) B′6 Historical climate and current development + 6 × 1 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 6 GL/y) B′70 Historical climate and current development + 7 × 10 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 70 GL/y) B′12 Historical climate and current development + 6 × 2 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 12 GL/y) B′105 Historical climate and current development + 7 × 15 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 105 GL/y) B′18 Historical climate and current development + 6 × 3 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 18 GL/y) C′dry C′dry corresponding to a 10% reduction in mean annual rainfall and 10% increase in potential evaporation relative to the historical climate (1910–2019) and 32 GL/y current development C′dry C′dry corresponding to a 10% reduction in mean annual rainfall and 10% increase in potential evaporation relative to the historical climate (1910–2019) and 0.1 GL/y current development C′mid C′mid corresponding to a 2% reduction in mean annual rainfall and 7.5% increase in potential evaporation relative to the historical climate (1910–2019) and 32 GL/y current development C′mid C′mid corresponding to a 2% reduction in mean annual rainfall and 7.5% increase in potential evaporation relative to the historical climate (1910–2019) and 0.1 GL/y current development C′wet C′wet corresponding to a 10% increase in mean annual rainfall and a 5% increase in potential evaporation relative to the historical climate (1910–2019) and 32 GL/y current development C′wet C′wet corresponding to a 10% increase in mean annual rainfall and a 5% increase in potential evaporation relative to the historical climate (1910–2019) and 0.1 GL/y current development D′dry35 C′dry climate and current development + 7 × 5 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to additional 35 GL/y) D′dry6 C′dry climate and current development + 6 × 1 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to additional 6 GL/y) D′dry70 C′dry climate and current development + 7 × 10 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums an additional 70 GL/y) D′dry12 C′dry climate and current development + 6 × 2 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 12 GL/y) SCENARIO CLA MODEL SCENARIO DCA MODEL D′dry105 C′dry climate and current development + 7 × 15 GL/y hypothetical future enterprises (i.e. total future extraction sums to an additional 105 GL/y) D′dry18 C′dry climate and current development + 6 × 3 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 18 GL/y) D′mid35 C′mid climate and current development + 7 × 5 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 35 GL/y) D′mid6 C′mid climate and current development + 6 × 1 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 6 GL/y) D′mid70 C′mid climate and current development + 7 × 10 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 70 GL/y) D′mid12 C′mid climate and current development + 6 × 2 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 12 GL/y) D′mid105 C′mid climate and current development + 7 × 15 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 105 GL/y) D′mid18 C′mid climate and current development + 6 × 3 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 18 GL/y) D′wet35 C′wet climate and current development + 7 × 5 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 35 GL/y) D′wet6 C′wet climate and current development + 6 × 1 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 6 GL/y) D′wet70 C′wet climate and current development + 7 × 10 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 70 GL/y) D′wet12 C′wet climate and current development + 6 × 2 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 12 GL/y) D′wet105 C′wet climate and current development + 7 × 15 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 105 GL/y) D′wet18 C′wet climate and current development + 6 × 3 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 18 GL/y) 4.3.5 Projected 2059 to 2069 conditions Projected 50-year conditions (2070) The second set of scenarios examines how different groundwater development scenarios might affect water resources in the CLA over a specific period in the future (2059 to 2069), based on historical and future climate data. The historical and future climate inputs comprise 11 × 50-year historical climate sequences taken from the observed climate (rainfall and PET) using the water years from 1910 to 2019. The scenarios comprise a warm-up period of 109 years of historical climate data used to prime the CLA groundwater model to 2019 and thereafter using the 50-year historical and future climate and groundwater development inputs to evaluate the 2059 to 2069 representative conditions (nominally representing 2070 conditions, 50-years after hypothetical future groundwater development started). The time periods for each of the 50-year climate sequences used in the 2019 to 2069 scenarios are summarised in Table 4-3. Note, that to account for 11 × 50-year historical climate sequences in a the 109-year period (1910 to 2019), climate sequences 8, 9, 10, 11 (i.e. the last four decades of the historical climate) used decadal blocks from the first four decades of the sequence from 1910 to 1939 to provide climate data. Table 4-3 Fifty-year time periods used for climate sequences CLIMATE SEQUENCE DATE FROM DATE TO 1 01/09/1910 31/08/1959 2 01/09/1920 31/08/1969 3 01/09/1930 31/08/1979 4 01/09/1940 31/08/1989 5 01/09/1950 31/08/1999 6 01/09/1960 31/08/2009 7 01/09/1970 31/08/2019 8 01/09/1980 to 31/08/2019 01/01/1900 to 31/08/1909 9 01/09/1990 to 31/08/2019 01/01/1900 to 31/08/1919 10 01/09/2000 to 31/08/2019 01/01/1900 to 31/08/1929 11 01/09/2010 to 31/08/2019 01/01/1900 to 31/08/1939 The first two scenarios (Scenario A′N and Scenario A′) represent ‘recent climate’ scenarios and are based on the 11 × 50-year historical climate sequences without development and with current levels of groundwater development, respectively. Scenario A′N is used as the baseline against which A, B, C and D assessments of relative change in hydrological regime are made. The second scenario (Scenario B) is also a ‘recent climate’ scenario. It is based on the 11 × 50-year climate sequences used in scenarios A′N and A′. Under this scenario, the current level and future level of groundwater development described in Section 4.3.4 are used. That is current development of 32 GL/year for the CLA and 0.1 GL/year for the DCA and hypothetical future groundwater development of 35, 70 and 105 GL/year for the CLA and 6, 12 and 18 GL/year for the DCA (see Table 4-4). Scenario B is, therefore, used to assess potential water availability assuming hypothetical future groundwater development. The third scenario (Scenario C) is a future climate and current development scenario. It is based on the 11 × 50-year climate sequences derived from scaling rainfall and PET described in Section 4.3.3. Scenario C employs the current level of groundwater development. The fourth scenario (Scenario D) is a future climate and future development scenario. It uses the same 11 climate sequences as Scenario C but considered the same hypothetical future groundwater development described in Section 4.3.4 (see Table 4-4). Table 4-4 Summary of modelling scenarios A, B, C and D for both aquifers using the moving 11 × 50-year window of climate sequences and combinations of current and hypothetical future groundwater development SCENARIO CLA MODEL SCENARIO DCA MODEL A′N Historical climate and no development A′N’ Historical climate and no development A′ Historical climate and 32 GL/y current development A′ Historical climate and 0.1 GL/y current development B′35 Historical climate and current development + 7 × 5 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 35 GL/y) B′6 Historical climate and current development + 6 × 1 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 6 GL/y) B′70 Historical climate and current development + 7 × 10 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 70 GL/y) B′12 Historical climate and current development + 6 × 2 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 12 GL/y) B′105 Historical climate and current development + 7 × 15 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 105 GL/y) B′18 Historical climate and current development + 6 × 3 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 18 GL/y) C′dry C′dry corresponding to a 10% reduction in mean annual rainfall and 10% increase in potential evaporation relative to the historical climate (1910–2019) and 32GL/y current development C′dry C′dry corresponding to a 10% reduction in mean annual rainfall and 10% increase in potential evaporation relative to the historical climate (1910– 2019) and 0.1 GL/y current development C′mid C′mid corresponding to a 2% reduction in mean annual rainfall and 7.5% increase in potential evaporation relative to the historical climate (1910–2019) and 32 GL/y current development C′mid C′mid corresponding to a 2% reduction in mean annual rainfall and 7.5% increase in potential evaporation relative to the historical climate (1910– 2019) and 0.1 GL/y current development C′wet C′wet corresponding to a 10% increase in mean annual rainfall and a 5% increase in potential evaporation relative to the historical climate (1910–2019) and 32 GL/y current development C′wet C′wet corresponding to a 10% increase in mean annual rainfall and a 5% increase in potential evaporation relative to the historical climate (1910– 2019) and 0.1 GL/y current development D′dry35 C′dry climate and current development + 7 × 5 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to additional 35 GL/y) D′dry6 C′dry climate and current development + 6 × 1 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to additional 6 GL/y) D′dry70 C′dry climate and current development + 7 × 10 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums an additional 70 GL/y) D′dry12 C′dry climate and current development + 6 × 2 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 12 GL/y) D′dry105 C′dry climate and current development + 7 × 15 GL/y hypothetical future enterprises (i.e. total future extraction sums to an additional 105 GL/y) D′dry18 C′dry climate and current development + 6 × 3 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 18 GL/y) D′mid35 C′mid climate and current development + 7 × 5 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 35 GL/y) D′mid6 C′mid climate and current development + 6 × 1 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 6 GL/y) D′mid70 C′mid climate and current development + 7 × 10 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 70 GL/y) D′mid12 C′mid climate and current development + 6 × 2 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 12 GL/y) D′mid105 C′mid climate and current development + 7 × 15 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 105 GL/y) D′mid18 C′mid climate and current development + 6 × 3 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 18 GL/y) D′wet35 C′wet climate and current development + 7 × 5 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 35 GL/y) D′wet6 C′wet climate and current development + 6 × 1 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 6 GL/y) D′wet70 C′wet climate and current development + 7 × 10 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 70 GL/y) D′wet12 C′wet climate and current development + 6 × 2 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 12 GL/y) D′wet105 C′wet climate and current development + 7 × 15 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 105 GL/y) D′wet18 C′wet climate and current development + 6 × 3 GL/y hypothetical future enterprises (i.e. total hypothetical future extraction sums to an additional 18 GL/y) 4.3.6 Locations for hypothetical future groundwater development Cambrian Limestone Aquifer Locations for hypothetical future groundwater development across the CLA (Figure 4-13) were selected based on spatial analyses of land suitability grids derived by Thomas et al. (2022) overlain on: (i) the spatial hydrogeology data (Department of Environment Parks and Water Security, 2008; Department of Industry Tourism and Trade, 2014) (see Figure 4-3), (ii) the spatial water quality and bore yield data (Department of Environment Parks and Water Security, 2014a) (see Figure 5-5 and Figure 5-6), (iii) the spatial water management zone data (Department of Environment Parks and Water Security, 2019c) (see Figure 4-13), and (iii) the spatial licensed entitlement data (Department of Environment Parks and Water Security, 2018) (see Figure 1-1). In addition, the DR2 (CLA) Leapfrog geological model was used to: (i) extract gridded depth to the top of the CLA, and (ii) extract gridded depth to groundwater and (iii) extract multiple hydrogeological cross-sections to evaluate aquifer saturated thickness (see Section 6). From this, based on potential water availability in the proposed MTLAWAP (Department of Environment and Natural Resources, 2017) and the current Georgina Wiso Water Allocation Plan (Northern Territory Government, 2022), seven hypothetical locations were chosen where there is: (i) potentially suitable land for agricultural intensification, (ii) a suitable aquifer exhibiting a saturated thickness greater than 20 m, (iii) suitable groundwater quality for irrigation (≤1000 mg/L TDS), (iv) indicative bore yields to indicate sufficient water could be extracted for groundwater-based irrigation (>15 L/second), (v) a threshold drilling depth to intersect the aquifer of less than 200 mBGL, and (vi) a depth to groundwater of less than 100 mBGL. Each location was placed at a distance of greater than 10 km from existing licensed water users and other hypothetical development locations. Figure 4-13 Spatial locations for hypothetical groundwater extraction sites across the Cambrian Limestone Aquifer relative to land that is potentially suitable for agricultural intensification, proposed water management zones and the groundwater discharge zone for the aquifer The extent of the different proposed and current water management zones is symbolised by both colour and pattern. Proposed water management zone data source: Department of Environment Parks and Water Security (2019c) Land versatility data source: Thomas et al. (2022) Water management zone data source: Department of Environment Parks and Water Security (2019c) Dook Creek Aquifer Locations for hypothetical future groundwater development across the DCA (Figure 4-14) were selected based on spatial analyses similar to those specified above but with slight differences. The same land suitability grids derived by Thomas et al. (2022) were overlain on: (i) the spatial hydrogeology data (Department of Environment Parks and Water Security, 2008; Department of Industry Tourism and Trade, 2014) (see Figure 4-4), and (ii) the spatial water quality and bore yield data (see Figure 5-5 and Figure 5-6). In addition, the DR2 (DCA) Leapfrog geological model was used to: (i) extract gridded depth to the top of the DCA, and (ii) gridded depth to groundwater (see Section 6). No licensed groundwater use across the DCA coincides with the DC2 (DCA) model domain and there are no water management zones as there is no water allocation plan for the A map of a large land Description automatically generated groundwater resource. However, based on the spatial analyses and the scale (i.e. size and thickness) of the aquifer, six hypothetical locations were chosen where there is: (i) potentially suitable land for agricultural intensification, (ii) a suitable aquifer exhibiting a saturated thickness greater than 20 m, (iii) suitable groundwater quality for irrigation (≤1000 mg/L TDS), (iv) indicative bore yields to indicate sufficient water could be extracted for groundwater-based irrigation (>15 L/second), (v) a threshold drilling depth to intersect the aquifer of less than 200 mBGL, and (vi) a depth to groundwater of less than 100 mBGL. Each location was placed at a distance of greater than 10 km from the Bulman community (the only existing water user, other than those using small quantities of water for stock and domestic use only) and the other hypothetical development locations. Figure 4-14 Spatial locations for hypothetical groundwater extraction sites across the Dook Creek Aquifer relative to land that is potentially suitable for agricultural intensification and groundwater discharge zone for the aquifer, including the major tributaries of the Roper River and discrete springs Spring data source: Department of Environment Parks and Water Security (2014c) Land versatility data source: Thomas et al. (2022) 4.3.7 Mean annual water balances Mean annual water balances are reported for two proposed management zones within the DR2 CLA model domain: the Mataranka Water Management Zone (MWMZ), which combines the north and south Mataranka zones on Figure 4-13, and the Larrimah Water Management Zone (LWMZ) (Figure 4-15). For the DCA, the mean annual water balance is reported for the extent DC2 DCA model domain within the Roper catchment (see Figure 4-15). Mean annual water balances are reported for both the quasi-equilibrium and projected 2059 to 2069 timescales and water balance components are reported in gigalitres per year. The quasi-equilibrium mean annual water balances are calculated for the last 109-year climate sequence replicate, whereas the projected 2059 to 2069 annual water balances are calculated for the final ten-year period from 2059 to 2069 (nominally representing 2070 conditions). 4.3.8 Groundwater levels CLA groundwater levels are reported for six sites (see Figure 4-15). RN035796 and RN019012 are in the proposed MWMZ. RN035796 is in the groundwater discharge zone along the upper Roper River and represents areas where the Tindall Limestone outcrops, and RN019012 represents areas where Cretaceous cover exists over the Tindall Limestone. RN028082 and RN029013 are in the proposed LWMZ. RN028082 is in the Tindall Limestone, and RN029013 is in the Gum Ridge Formation. RN024536 is located near the southern extent of the Roper catchment about 30 km east of Daly Waters and is also located in the Gum Ridge Formation. RN005621 is located about 100 km south of the Roper catchment. DCA groundwater levels are documented for five sites (see Figure 4-15). RN006546 represents part of the aquifer near Mountain Valley homestead, RN027811 represents part of the aquifer near the Bulman groundwater supply borefield, RN028226 represents part of the aquifer nearby where groundwater discharges to Wilton River, RN031983 represents part of the aquifer nearby where groundwater discharges to Flying Fox Creek, and RN036302 is located on Mainoru Station near the Mainoru River. The mean groundwater level for each site was calculated to provide a simple measure of the effects of each scenario on the groundwater systems of the CLA and DCA. The mean groundwater level for the quasi-equilibrium conditions was calculated for the fourth (i.e. last) 109-year climate sequence replicate. The mean groundwater level for each of the projected 2059 to 2069 scenarios was calculated from the last 10 years of all eleven 50-year sequences (nominally representing 2070 conditions). 4.3.9 Groundwater drawdown To demonstrate the spatial extent of the effects associated with each scenario, the groundwater drawdown has been calculated for the final time step of each scenario and presented as contours. Drawdowns calculated for scenarios A′, B′, C′ and D′ all use Scenario A′N as the reference head. A single set of drawdown contours is presented for each of the quasi-equilibrium scenarios at the final time step. The drawdowns from the 11 sets of projected 2059 to 2069 results at the final time step are summarised as percentiles with contours representing p5, p50, and p95 conditions. 4.3.10 Groundwater discharge metrics Groundwater discharge to rivers are presented as hydrographs in cubic metres per second for selected gauge sites along the upper Roper River (G9030013, DR2 CLA model), Flying Fox Creek (G9030108, DC2 DCA model) and Wilton River (G9030003, DC2 DCA model) (Figure 4-15). These sites are considered to represent the dry-season flow dynamics of the upper Roper River (DR2 CLA model) and its northern tributaries (DC2 DCA model), especially with respect to the impacts of groundwater extraction on dry-season low flows. The gauging sites and the corresponding river branch are presented in Table 4-5. Discharge at each site was calculated by summing fluxes at the Cauchy BC nodes assigned upstream of each site. Mean groundwater discharge at the gauge sites along the Roper River, Flying Fox Creek and Wilton River have also been calculated to provide simple measures that reflect changes to the discharge regime under each scenario. The mean groundwater discharge for the quasi-equilibrium scenarios was calculated from the fourth (i.e. last) 109-year climate sequence replicate. The mean groundwater discharge for the projected 2059 to 2069 scenarios was calculated from the last 10 years of all eleven 50-year sequences (nominally representing 2070 conditions). Table 4-5 Gauging sites and the corresponding river branch name GAUGE SITE BRANCH G9030013 Roper River at Elsey Homestead G9030003 Wilton River at Bulman Waterhole G9030108 Flying Fox Creek at Central Arnhem Road Figure 4-15 Spatial map showing the extent of the two proposed water management zones for reporting mean annual water balance results across the DR2 Cambrian Limestone Aquifer model, the extent of the DR2 Dook Creek Aquifer model within the Roper catchment, as well as groundwater-level reporting sites and groundwater discharge reporting sites Water management zone data source: Department of Environment Parks and Water Security (2019c) Part III Results 5 Regional assessment of the Roper catchment 5.1 Groundwater levels 5.1.1 Static groundwater levels Using the aquifer attribution dataset described in Section 3.1, Table 5-1 summarises the available groundwater-level data for bores in aquifers hosted in different hydrogeological units across the Roper catchment (Department of Environment Parks and Water Security, 2014a). Static standing water level data are available for 873 bores across the catchment, of which 56 could not be attributed to an aquifer (Figure 5-1). Most bores with information come from the CLA hosted in the Tindall Limestone and equivalents (n = 437), followed by bores in localised fractured rock aquifers hosted in the Roper Group (n = 128) and bores in aquifers hosted in the Mount Rigg and Nathan groups (n = 71; most data are from bores in the DCA). There is sparse information for localised fractured rock aquifers hosted in the Katherine River Group and aquifers hosted in unconsolidated sediments which contain little usable groundwater. The only clear spatial trend in groundwater-level data comes from the regional-scale CLA where water levels range from 1 mBGL where the aquifer outcrops around Mataranka to as low as 136 mBGL just south of Daly Waters along the southern catchment boundary (Figure 5-1). Other aquifers are data sparse, making it difficult to infer spatial trends, or their variability in water level simply reflects the localised nature of their flow systems, such as the fractured rock aquifers of the Roper Group, APV and Derim Derim Dolerite. Other than parts of the CLA and the APV which underlies the CLA, the maximum static groundwater levels are mostly less than 65 mBGL. There is only one bore with water level information for the Jinduckin Formation; it had a groundwater level of 63 mBGL. Table 5-1 Summary of groundwater-level data for bores in aquifers hosted in different hydrogeological units of the Roper catchment GROUNDWATER LEVEL STATISTIC HYDROGEOLOGICAL UNIT TINDALL LIMESTONE AND EQUIVALENTS MOUNT RIGG AND NATHAN GROUPS UNCONSOLIDATED SEDIMENTS CRETACEOUS ANTRIM PLATEAU VOLCANICS BUKALARA SANDSTONE DERIM DERIM DOLERITE ROPER GROUP KATHERINE RIVER GROUP Range min (mBGL) 1 1 1 4 3 2 -4 −10 18.7 Range max (mBGL) 136 43 15 46 131 63 61 37 29 Mean (mBGL) 44 12 6 21 50 22 13 10 24.9 Median (mBGL) 45 9 4 18 55 22 10 8 27 Count 437 71 19 45 57 27 29 128 3 Figure 5-1 Static groundwater levels for (a) the major aquifers hosted in the Tindall Limestone and equivalents and the Mount Rigg and Nathan groups and (b) other minor aquifers of the Roper catchment Symbol shape indicates different aquifers that bores are installed in; colour indicates the groundwater-level classes. Groundwater-level data source: Department of Environment Parks and Water Security (2014a) 5.1.2 Temporal groundwater levels Groundwater hydrographs from 12 selected bores with an appropriate time series (i.e. >5 years of data) were interpreted in terms of wet-season rainfall responses during the observation periods. Effects of surface water–groundwater connectivity were excluded as all bores were located distant from rivers and creeks (>100 m). Effects resulting from groundwater extraction were also excluded, as all bores either form part of a regular groundwater monitoring network maintained by DEPWS or were not located adjacent to production bores used as part of existing licensed extraction. Cambrian Limestone Aquifer Cambrian Limestone Aquifer hydrographs measured at four groundwater bores in the vicinity of Mataranka and Elsey National Park during the period 2007 to 2021 featured wet-season responses that ranged in magnitude from 0.3 to 7.8 m (Figure 5-2). Relatively small wet-season responses were observed at RN008299 (0.6 to 2.6 m) and RN035926 (0.3 to 2.0 m). Moderate magnitude responses were observed at RN035796 (1.0 to 5.2 m). Relatively large responses were observed at RN034032 (4.9 to 7.8 m). Atypically large responses to the 2020–21 wet season were recorded at RN008299 and RN035796 (2.6 and 5.2 m, respectively). A persistent maximum hydraulic head of approximately 121 mAHD (equivalent to a depth of approximately 2.5 m below ground surface) was observed during each wet season at RN034032, with the exception of 2018–19 and 2019–20 (which recorded 119 mAHD and 120 mAHD, respectively). This apparent upper limit was attributed A map of different countries/regions Description automatically generated to evapotranspiration flux losses from shallow groundwater. The monotonic recession of groundwater levels following the cessation of wet-season rainfall was consistently observed at all four bores. Recessions persisted until the onset of each subsequent wet season. The inter-annual variability of wet-season responses at each bore was high, typically in the order of metres. An inter-annual declining trend in minimum (i.e. end-of-dry-season) groundwater levels was observed at all four bores. Figure 5-2 Groundwater hydrographs (green) measured using automated loggers installed in four groundwater wells in the Cambrian Limestone Aquifer in the vicinity of Mataranka and Elsey National Park Blue lines show daily rainfall totals recorded at BoM weather station 14610 (Mataranka) between 2007 and 2022. Cambrian Limestone Aquifer hydrographs measured at four groundwater bores distant from Mataranka and Elsey National Park during the period 2007 to 2021 featured wet-season responses that ranged from 0.1 to 8.0 m (Figure 5-3). Relatively small wet-season responses were observed at RN029091 (0.2 to 0.9 m) and RN038812 (approximately 0.2 m). Moderate magnitude responses were observed at RN031985 (1.2 to 4.0 m). Response magnitudes at RN035793 were highly variable, ranging from less than 1 m in 2018–19 and 2019–20, to metre-scale responses in five of the 7 years recorded. An atypically large response to the 2020–21 wet season was recorded at RN035793 (8.0 m). The monotonic recession of groundwater levels following the cessation of wetseason rainfall was observed at three of the four bores (i.e. excluding RN038812). Recessions persisted until the onset of each subsequent wet season. The inter-annual variability of wetseason responses at each bore was limited, typically in the order of decimetres. An inter-annual declining trend in minimum (i.e. end-of-dry-season) groundwater levels was observed at two of the four bores (RN029091 and RN035793). In contrast, hydraulic head values at RN038812 slightly increased from 152.0 to 152.5 mAHD over the period 2014 to 2019. Figure 5-3 Groundwater hydrographs (green) measured using automated loggers installed in four groundwater wells in the Cambrian Limestone Aquifer (a,d) approximately 30 km south of Elsey National Park and (b,c) in the vicinity of Jilkminggan Blue lines show daily rainfall totals recorded at BoM weather station 14623 (Elsey) between 2007 and 2022. Other aquifers Hydrographs recorded in other aquifers in the Roper catchment were generally of limited duration (e.g. 12 months), representing only a single wet-season response. This included two bores completed in the Dook Creek Formation (RN031981 and RN031983), one bore completed in the Antrim Plateau Volcanics (RN034030) and one bore completed in Cretaceous sandstone (RN030825) (Figure 5-4). Responses of 3.2 m and 4.8 m to the 2007–08 wet season were recorded in the two Dook Creek Formation bores (RN031981 and RN031983, respectively). A response of 10.3 m to the 2010–11 wet season was recorded in RN030825. Six wet-season responses were recorded at RN034030 over the period 2008 to 2016. These ranged in magnitude from 1.3 m in 2011–12 to 2.8 m in 2008–09. Monotonic recession of groundwater levels following the cessation of wet-season rainfall was consistently observed at all four bores. Recessions persisted until the onset of each subsequent wet season. The inter-annual variability of wet-season responses at RN034030 was limited, typically in the order of decimetres. Minimum (i.e. end-of-dry-season) groundwater levels at RN034030 were consistent over the period of monitoring. Figure 5-4 Groundwater hydrographs (green) measured using automated loggers installed in four groundwater bores in aquifers of the (a,b) Dook Creek Formation, (c) Antrim Plateau Volcanics and (d) Cretaceous sandstone Blue lines show daily rainfall totals recorded at BoM weather stations (a,b) 14627 (Bulman), (c) 14623 (Elsey) and (d) 14919 (Maranboy). 5.2 Groundwater salinity Using the aquifer attribution dataset described in Section 3.1, Table 5-2 summarises the available groundwater salinity data for bores in aquifers hosted in different hydrogeological units across the Roper catchment. Salinity data of an acceptable quality (CBE within ±5%) are available for 303 bores across the catchment, of which 36 could not be attributed to an aquifer (Figure 5-5). Most bores with information come from the CLA hosted in the Tindall Limestone and equivalents (n = 133) followed by bores in localised fractured rock aquifers hosted in the Roper Group (n = 43) and bores in aquifers hosted the Mount Rigg and Nathan groups (n = 32 – most from the DCA). There is sparse information for localised aquifers, including the Cretaceous rocks, Bukalara Sandstone, unconsolidated sediments and various fractured rock aquifers hosted in the APV and Derim Derim Dolerite, which are only used for stock and domestic water supplies. Groundwater in the major aquifers such as the CLA hosted in the Tindall Limestone and equivalents and the DCA hosted in the Mount Rigg Group have a mostly fresh salinity. The mean and median salinities for the Tindall Limestone and equivalents are 844 and 850 mg/L, respectively (Table 5-2). The slightly higher mean and median salinity along with a maximum salinity of 2080 mg/L, reflects that some parts of the aquifers host brackish groundwater. Salinity varies between fresh and slightly brackish in parts of the northern Georgina and southern Daly basins in the eastern and central parts of the aquifer compared to the western part of the aquifer in the Wiso Basin (Figure 5-5). Groundwater in the Mount Rigg and Nathan Group aquifers, particularly the DCA, is fresh with a mean and median salinity of 390 and 327 mg/L respectively. The mean and median salinity for the minor aquifers across the Roper catchment suggests most groundwater is fresh (<800 mg/L) with the exception of localised aquifers hosted in the fractured rocks of the Derim Derim Dolerite and the Roper Group (mean salinity >900 mg/L). There is only one bore with groundwater salinity information for the Jinduckin Formation which had a salinity of 319 mg/L. Table 5-2 Summary of groundwater salinity data as TDS for bores in aquifers hosted in different hydrogeological units of the Roper catchment GROUNDWATER SALINITY STATISTIC HYDROGEOLOGICAL UNIT TINDALL LIMESTONE AND EQUIVALENTS MOUNT RIGG AND NATHAN GROUP UNCONSOLIDATED SEDIMENTS CRETACEOUS ANTRIM PLATEAU VOLCANICS BUKALARA SANDSTONE DERIM DERIM DOLERITE ROPER GROUP Range min (mg/L) 172 39 260 19 229 130 380 45 Range max (mg/L) 2080 1041 800 1570 2616 1510 3060 5475 Mean (mg/L) 844 390 429 425 615 622 1193 951 Median (mg/L) 850 327 343 330 479 490 665 380 Count 133 32 10 11 21 12 4 43 Groundwater salinity data as total dissolved solids (TDS) Figure 5-5 Groundwater salinity for (a) the major aquifers hosted in the Tindall Limestone and equivalents and the Mount Rigg and Nathan groups and (b) other minor aquifers of the Roper catchment Symbol shape indicates different aquifer that bores are installed in; colour indicates the groundwater salinity classes as total dissolved solids (TDS). Salinity data source: Department of Environment, Parks and Water Security (2014) 5.3 Bore yields Using the aquifer attribution dataset described in Section 3.1, Table 5-3 summarises the available groundwater bore yield data for bores in aquifers hosted in different hydrogeological units across the Roper catchment. Bore yield data are available for 928 bores across the catchment, of which 63 could not be attributed to an aquifer (Figure 5-6). Most bores with information come from the CLA hosted in the Tindall Limestone and equivalents (n = 444) followed by bores in localised fractured rock aquifers hosted in the Roper Group (n = 144) and bores in aquifers hosted the Mount Rigg and Nathan groups (n = 85 – most from the DCA). There is sparse information for localised aquifers hosted in both the fractured and weathered rocks of the Roper Group and the unconsolidated sediments. Bore yields in the major aquifers such as the CLA hosted in the Tindall Limestone and equivalents and the DCA hosted in the Mount Rigg Group exhibit relatively high yields with mean values greater than 5 L/second and maximum yields greater than 40 L/second (Figure 5-6). Bore yields from aquifers hosted in the Cretaceous rocks, Bukalara Sandstone and the APV and Derim Derim Dolerite indicate that where either porous parts of the sandstones exist (Cretaceous rocks and Bukalara Sandstone) or large and or interconnected fractures occur (across parts of the APV and dolerite), yields can be locally high (>10 L/second) but aquifers are storage limited. The low (i.e. mean and median <2 L/second) and variable bore yields from aquifers hosted in the fractured and weathered rock aquifers of the Roper River and Katherine River groups and the aquifers hosted in unconsolidated sediments indicate they store and yield little water. As 114 | Hydrogeological assessment of the Roper catchment A map of land with green and blue dots Description automatically generated described in Section 3.4, the yield value is indicative as most bore testing is by short-term (i.e. a few hours) discharge testing by air lifting or use of a submersible pump on small diameter (<150 mm) stock and investigation bores. Nevertheless, these values provide a good indication of the potential for the aquifer at specific locations to yield water at a sufficient rate for different purposes. Table 5-3 Summary of groundwater bore yield data for bores in aquifers hosted in different hydrogeological units of the Roper catchment BORE YIELD STATISTIC HYDROGEOLOGICAL UNIT TINDALL LIMESTONE AND EQUIVALENTS MOUNT RIGG AND NATHAN GROUP UNCONSOLIDATED SEDIMENTS CRETACEOUS ANTRIM PLATEAU VOLCANICS BUKALARA SANDSTONE DERIM DERIM DOLERITE ROPER GROUP KATHERINE RIVER GROUP Range min (L/s) 0.1 0.1 0.2 0.2 0.1 0.8 0.1 0.04 0 Range max (L/s) 100 45 4 15 25 12 10 33.5 2.5 Mean (L/s) 7.6 6.6 1.4 5.2 2.4 3.4 1.8 1.7 1.1 Median (L/s) 4 4 1 3 1.6 3 1.2 0.8 1.1 Count 444 85 18 46 60 31 31 144 5 Figure 5-6 Groundwater bore yields for (a) the major aquifers hosted in the Tindall Limestone and equivalents and the Mount Rigg and Nathan groups and (b) other minor aquifers of the Roper catchment Symbol shape indicates different aquifers within which bores are sited; colour indicates bore yield classes. Bore yield data source: Department of Environment, Parks and Water Security (2014) 5.4 Aquifer hydraulic properties 5.4.1 Pumping test reanalyses Time–drawdown data were collated from 23 two-bore (extraction bore and monitoring bore) constant-rate discharge pumping tests performed across the Assessment region (Figure 5-7). This represented only 2% of the registered bores in the Roper catchment (i.e. those known to not be abandoned, backfilled or decommissioned) and only 16% of the bores known to have been hydraulically tested in the catchment. Tests performed on the CLA were limited to two bores located approximately 26 km due east of Mataranka. Tests performed in the DCA were limited to five bores located approximately 4 km east of Beswick and one bore located 500 m west of Bulman. Tests performed on aquifers hosted in the Nathan Group were limited to four bores located within 6 km of Ngukurr. Tests performed on aquifers hosted in the Bessie Creek Sandstone were limited to five bores located within 4 km of Minyerri. Tests performed on bores in other aquifers included the Moroak Sandstone (RN035879, RN036300), Hodgson Sandstone (RN035798), and APV (RN035876) near Minyerri, the Mount Rigg Group near Barunga (RN008420), and unconsolidated sediments near Beswick (RN009587). Figure 5-7 Locations of groundwater bores at which two-bore constant-rate discharge pumping tests (coloured squares) or single-bore constant-rate discharge pumping tests (coloured circles) were undertaken Also shown are groundwater bores at which pumping tests had not been undertaken (small white circles), and the spatial extents of the Cambrian Limestone Aquifer (blue shading) and Dook Creek Aquifer (green shading). Cambrian Limestone Aquifer Two constant-rate discharge pumping tests were undertaken in the CLA, at bores RN008361 and RN008362 Table 5-4. Unconfined conditions were identified from qualitative inspection of time– drawdown responses; they were consistent with bore screen depth, overlying lithology, and comparisons between water cut depths and standing water levels. Hydraulic properties had not previously been estimated from these two tests. Aquifer transmissivity values of 1048 and 1064 m2/day were estimated with a high degree of confidence (i.e. a maximum 95% with a confidence interval of ±3%). The order of magnitude estimated was consistent with transmissivity values estimated at other bores completed in the CLA. Aquifer-specific yield values of 0.06% and 0.13% were estimated with a moderate degree of confidence (i.e. 95% with confidence intervals of ±27 and ±18%, respectively). Such low specific yield values (i.e. <0.1%) are consistent with a karstic limestone aquifer. Table 5-4 Hydraulic properties estimated from two-bore constant-rate discharge pumping tests performed in the Cambrian Limestone Aquifer across the Roper catchment To maximise readability, the en dash (–) indicates fields for which values were not able to be collated or estimated. PUMPING TEST DETAILS AND HYDRAULIC PROPERTIES REGISTERED BORE RN008361 RN008362 Test duration (h) 24 25 Aquifer type Unconfined Unconfined Historical transmissivity (T) (m2/d) – – Estimated T (m2/d) 1064 1048 T ±95% CI (%) 3 2 Historical storage coefficient (S) or specific yield (Sy) – – Estimated S or Sy 6 × 10−4 1 × 10−3 S or Sy ±95% CI (%) 27 18 CI = confidence interval Dook Creek Aquifer Six constant-rate discharge pumping tests on the DCA were identified, at bores RN009629, RN021686, RN021710, RN021711, RN028477 and RN028478. All tests were of 24 hours duration except for the last two bores where the tests were 8 hours. Time–drawdown responses indicated a variety of confinement types, including unconfined (RN009629), dual-domain confined (RN021686, RN021710, RN021711) and leaky confined (RN028477, RN028478) (Table 5-5). These interpretations were consistent with bore screen depth, overlying lithology, estimated storage coefficient magnitudes, and comparisons between water cut depths and standing water levels. Tests performed at RN028477 and RN028478 were excluded from quantitative analysis because 95% confidence intervals associated with estimated aquifer storativity values exceeded 100%. Historical estimates of aquifer transmissivity from all tests were highly variable, ranging from less than 10 m2/day at RN021686 to more than 500 m2/day at RN021710 and RN021711. Aquifer transmissivity values estimated from non-leaky test responses were also highly variable, ranging from 7 m2/day to 341 m2/day. A high degree of confidence (i.e. a maximum 95% with a confidence interval of ±6%) was associated with these estimates. Aquifer storage properties had not previously been estimated from these six tests. Aquifer storativity values ranging from 5 × 10–4 to 8 × 10–3 were estimated from non-leaky test responses with a moderate to low degree of confidence (i.e. 95% with confidence intervals ranging from ±14% to ±48%). Table 5-5 Hydraulic properties estimated from two-bore constant-rate discharge pumping tests performed in the Dook Creek Aquifer across the Roper catchment The en dash (–) indicates fields for which values were not able to be collated or estimated. PUMPING TEST DETAILS AND HYDRAULIC PROPERTIES REGISTERED BORE RN009629 RN021686 RN021710 RN021711 RN028477 RN028478 Duration (h) 24 24 24 24 8 8 Aquifer type Uncon fined Dual domain Dual domain Dual domain Leaky Leaky Historical transmissivity (T) (m2/d) 84 8 582 509 6 – Estimated T (m2/d) 84 7 235 341 – – T ±95% CI (%) 6 5 3 5 – – Historical storage coefficient (S) or specific yield (Sy) – – – – – – Estimated S or Sy 8 × 10−3 1 × 10−3 5 × 10−4 3 × 10−3 – – S or Sy ±95% CI (%) 20 48 14 24 – – CI = confidence interval Nathan Group aquifers Four constant-rate discharge pumping tests in aquifers of the Nathan Group were identified, at bores RN035721, RN035317, RN030405 and RN035316 (Table 5-6). Time–drawdown responses indicated dual-domain confined conditions at all four bores. These interpretations were consistent with bore screen depth, overlying lithology, estimated storage coefficient magnitudes, and comparisons between water cut depths and standing water levels. Tests performed at RN035317 and RN030405 were excluded from quantitative analysis because 95% confidence intervals associated with estimated aquifer storativity values exceeded 100%. Historical estimates of aquifer transmissivity from all tests were variable, ranging from 116 m2/day at RN035316 to 1422 m2/day at RN035721. Estimated aquifer transmissivity values of 62 m2/day and 705 m2/day were lower than prior historical estimates. A high degree of confidence (i.e. a maximum 95% with a confidence interval of ±13%) was associated with these estimated transmissivity values. Aquifer storage properties had not previously been estimated from these four tests. Aquifer storativity values of 4 × 10–4 and 4 × 10–3 were estimated with a low degree of confidence (i.e. 95% with confidence interval values of ±33% and ±83%, respectively). Table 5-6 Hydraulic properties estimated from two-bore constant-rate discharge pumping tests performed on bores in aquifers hosted in the Nathan Group across the Roper catchment The en dash (–) indicates fields for which values were not able to be collated or estimated. PUMPING TEST DETAILS AND HYDRAULIC PROPERTIES REGISTERED BORE RN035721 RN035317 RN030405 RN035316 Duration (h) 8 24 21 24 Aquifer type Dual domain Dual domain Dual domain Dual domain Historical transmissivity (T) (m2/d) 1422 132 407 116 Estimated T (m2/d) 705 – – 62 T ±95% CI (%) 13 – – 5 Historical storage coefficient (S) or specific yield (Sy) – – – – Estimated S or Sy 4×10−3 – – 4×10−4 S or Sy ±95% CI (%) 83 – – 33 CI = confidence interval Bessie Creek Sandstone aquifers Five constant-rate discharge pumping tests on aquifers hosted in the Bessie Creek Sandstone were identified, at bores RN025383, RN027907, RN027909, RN027960 and RN027961 (Table 5-7). Time– drawdown responses indicated leaky or dual-domain confined conditions. These interpretations were consistent with bore screen depth, overlying lithology, estimated storage coefficient magnitudes, and comparisons between water cut depths and standing water levels. Historical estimates of aquifer transmissivity from these five tests were relatively small (≤10 m2/day) and had not previously been interpreted at RN027907. Estimated aquifer transmissivity values were order-of-magnitude consistent with prior historical estimates, ranging from less than 1 to 5 m2/day. High-to-moderate degrees of confidence (i.e. a maximum 95% with a confidence interval of ±18%) were variously associated with these estimates. Aquifer storage properties had not previously been estimated from these five tests, with the exception of RN025383 (where S = 3 × 10–4). Estimated aquifer storativity values were highly variable, ranging over 2.5 orders of magnitude, from 1 × 10–6 to 3 × 10–4. The 95% confidence intervals associated with these estimates were also variable, ranging from ±7% (high confidence) to ±88% (low confidence). Table 5-7 Hydraulic properties estimated from two-bore constant-rate discharge pumping tests performed on bores in aquifers hosted in the Bessie Creek Sandstone across the Roper catchment The en dash (–) indicates fields for which values were not able to be collated or estimated. PUMPING TEST DETAILS AND HYDRAULIC PROPERTIES REGISTERED BORE RN025383 RN027907 RN027909 RN027960 RN027961 Duration (h) 16 8 8 8 24 Aquifer type Leaky Leaky Dual domain Dual domain Dual domain Historical transmissivity (T) (m2/d) 3 – 10 5 3 Estimated T (m2/d) 1 3 5 1 <1 T ±95% CI (%) 18 8 15 7 9 Historical storage coefficient (S) or specific yield (Sy) 3 × 10−4 – – – – Estimated S or Sy 3 × 10−4 5 × 10−6 1 × 10−6 2 × 10−4 1 × 10−6 S or Sy ±95% CI (%) 7 10 88 22 23 CI = confidence interval Other minor aquifers Six constant-rate discharge pumping tests in minor aquifers of the Roper catchment were identified, at bores RN035876, RN035798, RN035879, RN036300 and RN008420 (Table 5-8). Time– drawdown responses indicated dual-domain confined conditions at all bores with the exception of leaky confined conditions in a Cretaceous sediments/rocks aquifer at RN009587. These interpretations were consistent with bore screen depth, overlying lithology, estimated storage coefficient magnitudes, and comparisons between water cut depths and standing water levels. The test performed at RN035798 was excluded from quantitative analysis because the 95% confidence interval associated with the estimated aquifer storativity value exceeded 100%. Historical estimates of aquifer transmissivity ranged from 2 and 4 m2/day in Moroak Sandstone aquifers to 234 m2/day in the Hodgson Sandstone. Estimated aquifer transmissivity values were order-of-magnitude consistent with prior historical estimates, with the exception of RN008420 (Mount Rigg Group). High-to-moderate degrees of confidence (i.e. 95% with confidence interval values ranging from ±2% to ±51%) were variously associated with these estimates. Aquifer storage properties had not previously been estimated from any of the six tests. Estimated aquifer storativity values were highly variable, ranging from 4 × 10–9 (Moroak Sandstone, RN035879) to 5 × 10–4 (Cretaceous sediments/rocks, RN009587). The 95% confidence intervals associated with these estimates were also highly variable, ranging from ±15% (RN036300) to ±43% (RN035879) (both Moroak Sandstone). Table 5-8 Hydraulic properties estimated from two-bore constant-rate discharge pumping tests performed in different hydrogeological units hosting minor aquifers across the Roper catchment The en dash (–) indicates fields for which values were not able to be collated or estimated. PUMPING TEST DETAILS AND HYDRAULIC PROPERTIES HYDROGEOLOGICAL UNIT ANTRIM PLATEAU VOLCANICS HODGSON SANDSTONE MOROAK SANDSTONE MOROAK SANDSTONE MOUNT RIGG GROUP CRETACEOUS SEDIMENTS AND ROCKS Registered bore RN035876 RN035798 RN035879 RN036300 RN008420 RN009587 Duration (h) 8 8 8 24 24 24 Aquifer type Dual domain Dual domain Dual domain Dual domain Dual domain Leaky Historical transmissivity (T) (m2/d) 40 234 4 2 12 40 Estimated T (m2/d) 16 168 2 2 1 32 T ±95% CI (%) 7 – 10 2 10 51 Historical storage coefficient (S) or specific yield (Sy) – – – – – – Estimated S or Sy 6×10−6 – 2×10−6 4×10−9 6×10−6 5×10−4 S or Sy ±95% CI (%) 20 – 43 15 19 24 CI = confidence interval Summary A total of 144 groundwater bores were identified from an existing territory-wide dataset (Nguyen and Tickell, 2014) at which historical pumping tests had been performed within the Roper catchment. Of these, only 23 two-bore constant-rate discharge tests achieved meaningful drawdown responses. These tests were performed on bores in aquifers of the Dook Creek Formation (6), Bessie Creek Sandstone (5), Nathan Group (4) and Cambrian limestone (2), as well as other relatively minor aquifers (6). Reinterpretation of these tests provided a range of benefits, including estimates of: aquifer confinement type; aquifer storage properties (for the first time); and 95% confidence intervals associated with both aquifer transmissivity and storativity parameters. Reinterpreted aquifer transmissivity values were generally lower than historical estimates, due to accounting for either aquifer depressurisation or aquitard leakage during model inversion. More generally, the reanalysis of these data has highlighted the paucity of hydraulic testing data across aquifers in the Roper catchment. 5.4.2 Passive time series analyses Groundwater pressure data measured using automated loggers were obtained from 45 bores located within the Roper catchment. Of these, time series data from 31 bores were found to be suitable for passive time series analysis; that is, they were sampled at 4-hourly frequency or greater for a continuous duration of 6 months or more (Figure 5-8). These represented only 3% of the total number of registered functional bores in the Roper catchment (i.e. those known to not be abandoned, backfilled or decommissioned). Datasets from the majority (21 bores) were recorded in the CLA. Of these, 16 bores were located within 30 km of Mataranka and 5 bores were located up to 130 km farther south (RN028087, RN029091, RN035929, RN038812, RN038813). Due to the use of relatively long screens during bore construction, data from an additional two bores (RN035790 and RN035927, both located on the border of Elsey National Park) represented the combined response of aquifers hosted in both the CLA and APV. The eight remaining datasets were obtained from: four bores in Cretaceous sediments/rocks, located within 23 km of Barunga (RN022743, RN030825, RN031497, RN035863); two bores in the Nathan Group, located 5 km northwest of Ngukurr (RN035312, RN035315); one bore in the Dook Creek Formation, located 25 km northwest of Bulman (RN031981); and one bore in the Antrim Plateau Volcanics, located 20 km south-east of Mataranka (RN034030). Figure 5-8 Locations of 31 groundwater bores at which time series of pressure were recorded using automated loggers (coloured circles) at a 4-hourly frequency or greater, for a continuous duration of 6 months or more To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. Results Amplitude spectra for the frequency band 0.5 to 2.5 cycles/day were calculated for visual inspection using the fast Fourier transform from 31 measured groundwater hydraulic head time series. Amplitudes of the primary lunar constituent (M2) and S2 tidal constituents (occurring at 1.9 and 2.0 cycles/day respectively) were calculated separately using a harmonic least squares method. For example, from the time series of hydraulic head recorded at bore RN035926 (Figure 5-9a), the relevant amplitude spectrum was calculated (Figure 5-9b). Similarly, from time series of measured barometric pressure (Figure 5-9c) and theoretical Earth tide vertical strain (Figure 5-9d) at the same location, equivalent amplitude spectra were also calculated (Figure 5-9e and Figure 5-9f, respectively). The example presented in Figure 5-9b is indicative of unconfined conditions, since the groundwater time series is dominated by the S1 and S2 components (occurring at 1 and 2 cycles/day, respectively), which are present in both the barometric pressure and Earth tide time series. In comparison, a groundwater time series dominated by the M2 constituent (at 1.9 cycles/day), which is present only in the Earth tide time series, would be indicative of confined conditions. Alternatively, semi-confined conditions would be indicated by the presence of an M2 constituent amplitude that exceeded local spectral noise but was smaller than that of the S2 constituent. Figure 5-9 (a) Time series and (b) amplitude spectrum of hydraulic head measured at bore RN035926, located approximately 9 km south-east of Mataranka. (c) Time series and (d) amplitude spectrum of barometric pressure measured at same location over a shorter time period. (e) Time series and (f) amplitude spectrum of theoretical Earth tide vertical strain estimated over the same time period and at the same location Amplitude spectra (coloured lines) were calculated using the fast Fourier transform. Amplitudes of the M2 and S2 tidal constituents (labelled open black circles) were calculated using harmonic least squares. Aquifer confinement status The amplitude of the M2 constituent did not exceed local spectral noise, and therefore indicated unconfined conditions, at 19 of 31 bores (Table 5-9). The amplitude of the M2 constituent was not dominant but did exceed local spectral noise (and was equal to at least 15% of the S2 constituent amplitude) at ten bores, indicating semi-confined conditions at these locations. Of the 23 groundwater bores fully or partially screened in the CLA, unconfined and semi-confined conditions were identified at 14 and 9 locations, respectively. The absence of confined conditions was consistent with both (i) the relatively shallow occurrence of the lower CLA, and (ii) the absence of the confining upper CLA across the majority of the Roper catchment. Similarly, unconfined conditions were consistently identified in four bores completed in Cretaceous sediments and rocks. Table 5-9 Estimated amplitudes (in mm) of M2 and S2 Earth tide constituents in 31 measured groundwater hydraulic head time series and resulting interpreted aquifer confinement status DETAILS REGISTERED BORE BORE RN RN008299 RN022743 RN028087 RNO29091 RN030825 RN031497 RN031981 RN031984 RN031985 Aquifer CLA CS/R CLA CLA CS/R CS/R DCF CLA CLA 𝑨𝑨𝑴𝑴𝟐𝟐 𝑮𝑮𝑮𝑮 0.61 1.12 0.66 2.43 0.20 0.88 2.07 0.50 1.16 𝑨𝑨𝑺𝑺𝟐𝟐 𝑮𝑮𝑮𝑮 3.40 11.78 13.66 9.93 2.32 11.19 6.35 1.82 5.53 Ratio 0.18 0.10 0.05 0.24 0.09 0.08 0.33 0.27 0.21 Status S U U S U U S S S BORE RN RN034030 RN034031 RN034032 RN034038 RN03403 9 RN034230 RN034231 RN035315 RN035512 Aquifer APV CLA CLA CLA CLA CLA CLA YV NG 𝑨𝑨𝑴𝑴𝟐𝟐 𝑮𝑮𝑮𝑮 13.42 0.30 0.07 1.03 0.11 0.60 1.26 0.44 7.12 𝑨𝑨𝑺𝑺𝟐𝟐 𝑮𝑮𝑮𝑮 5.33 2.77 0.89 6.81 1.03 1.82 2.78 0.09 54.05 Ratio 2.52 0.11 0.08 0.15 0.11 0.33 0.45 4.89 0.13 Status C U U S U S S C U BORE RN RN035519 RN035790 RN035792 RN035793 RN03579 6 RN035863 RN035926 RN035927 RN035929 Aquifer CLA CLA/APV CLA CLA CLA CS/R CLA CLA/APV CLA 𝑨𝑨𝑴𝑴𝟐𝟐 𝑮𝑮𝑮𝑮 0.92 0.41 0.58 0.09 1.23 0.67 0.47 0.03 0.42 𝑨𝑨𝑺𝑺𝟐𝟐 𝑮𝑮𝑮𝑮 8.12 1.72 5.69 0.78 2.31 11.34 4.05 1.47 8.78 Ratio 0.11 0.24 0.10 0.12 0.53 0.06 0.12 0.02 0.05 Status U S U U S U U U U BORE RN RN036305 RN038812 RN038813 RN041443 Aquifer CLA CLA CLA CLA 𝑨𝑨𝑴𝑴𝟐𝟐 𝑮𝑮𝑮𝑮 0.58 0.65 0.68 0.11 𝑨𝑨𝑺𝑺𝟐𝟐 𝑮𝑮𝑮𝑮 4.38 13.41 13.30 1.03 Ratio 0.13 0.05 0.05 0.11 Status U U U U RN = registered number; Ratio = 𝑨𝑨𝑴𝑴𝑴𝑴 𝑮𝑮𝑮𝑮 / 𝑨𝑨𝑺𝑺𝑺𝑺 𝑮𝑮𝑮𝑮; Aquifer – APV = Antrim Plateau Volcanics, CLA = Cambrian Limestone Aquifer, CS/R = Cretaceous sediments/rocks, DCF = Dook Creek Formation, NG = Nathan Group, YV = Yalwarra Volcanics; Status – U = unconfined, S = semi-confined, and C = confined. The M2 constituent was dominant only at RN034030 and RN035315, indicating confined conditions at these two locations. Bore RN034030 is located approximately 1.2 km north of the Roper Highway and within the eastern boundary of Elsey National Park. The bore is screened at 27 to 29 mBGL in localised aquifers hosted in the APV but is hydraulically disconnected (via annulus cement) from only the uppermost 4 m of overburden. The 27 m of overburden thickness includes layers of limestone, siltstone and clays of variable thickness, which could plausibly of confine the APV below the CLA. Water cuts of 0.3 and 1.0 L/second were observed during construction of RN034030 at 14.0 to 17.3 and 20.0 to 28.4 mBGL, respectively. A standing water level of 3.4 mBGL was recorded after bore completion. These observations indicate a confining pressure of approximately 24 mH2O (235 kPa). These various independent lines of evidence support the estimation of confined conditions at RN034030 from the interpreted Earth tide response. Bore RN035315 is located approximately 6 km northwest of Ngukurr. The bore features three discrete screened intervals between 47 and 66 mBGL, all of which intersect the Yalwarra Volcanics aquifer (Groves, 2021). Water cuts of 0.5 to 3.3 L/second were encountered during drilling at four discrete depths between 37 and 65.3 mBGL. A standing water level of 39 mBGL was recorded after bore completion. These observations indicate a dry overburden thickness of approximately 37 m overlying the Yalwarra Volcanics aquifer at RN035315 and a minimum confining pressure of approximately 8 mH2O (78 kPa). These various independent lines of evidence support the estimation of confined conditions at RN035315 from the interpreted Earth tide response. Estimation of aquifer barometric efficiency and specific storage The estimation of aquifer barometric efficiency and specific storage values is generally only appropriate to confined conditions. Confined conditions were identified at two bores (RN034030 and RN035315) from analyses of responses to Earth tides. Barometric pressure observations recorded over a corresponding time period and at an hourly sampling interval were available at a single location, at bore RN035926. This barometric logger was located sufficiently nearby (within 12 km) to permit the estimation of barometric efficiency at RN034030. Conversely, the barometric logger was located too far (approximately 170 km) from RN035315 to permit analysis. Using the Rau et al. (2020) solution, aquifer barometric efficiency at RN034030 was estimated at 32%. Corresponding aquifer-specific storage values were estimated using the Jacob (1940) solution to range from 1 × 10–7 m–1 to 5 × 10–6 m–1 for assumed aquifer effective porosity end member values of 1% and 33%, respectively. Summary Data recorded by automated loggers that were suitable for time series analyses were obtained from a total of 31 groundwater bores in the Roper catchment. Responses to Earth tides at two bores indicated confined conditions, with the remainder of responses indicating unconfined or semi-confined conditions. Of the 23 groundwater bores fully or partially screened in the CLA, unconfined and semi-confined conditions were identified at 14 and 9 locations, respectively. The absence of confined conditions was consistent with both (i) the relatively shallow occurrence of the lower CLA, and (ii) absence of the confining upper CLA across the majority of the Roper catchment. The estimation of aquifer barometric efficiency and specific storage was constrained (partly by barometric pressure data availability) to a single bore, RN034030. Aquifer barometric efficiency at RN034030 was estimated at 32%. Corresponding aquifer-specific storage values were estimated to range from 1 × 10–7 m–1 to 5 × 10–6 m–1 for assumed aquifer effective porosity end member values of 1% and 33% respectively. The paucity of continuous time series of groundwater and barometric pressure in the Roper catchment limited the application of analyses based on responses to Earth and atmospheric tides. This result highlights the need for ongoing monitoring of groundwater and barometric pressures at sufficient sampling periods (i.e. hourly or less) over sufficient and corresponding durations (i.e. 9 months or more). Estimates of aquifer transmission and storage properties derived from tidal analyses would supplement existing estimates from relatively higher-cost methods of active hydraulic testing, such as pump and slug tests. 5.5 Recharge estimation This section contains the results of estimating recharge using the chloride mass balance method. It follows the same format as outlined in the methods section (Section 3.6). The results are presented at the point scale, then upscaled to the entire study area, constrained using observations of baseflow and excess water, and finally recharge is extracted for various areas of interest. 5.5.1 Point-scale chloride mass balance The database of chloride in groundwater measurements contains 2319 locations with at least one chloride in groundwater observation (Figure 5-10). The chloride in groundwater ranged from 2 to 25,000 mg/L with a median of 122 mg/L. At these points the mean chloride deposition in rainfall ranged from 54 kg per ha year along the coast to 2.0 kg per ha per year further inland. Of these 2319 points, 1604 were retained after being assessed against the criteria in Section 3.6.1. The median of the 1000 replicates of point recharge at these 849 points is shown in Figure 5-10. It ranges from 0.1 to 679 mm/year, with a mean of 45 mm/year and a median of 4 mm/year, demonstrating the skewed distribution of recharge. Figure 5-10 (a) Chloride in groundwater observations within the study region and (b) the median of the point-scale estimates of recharge derived from them 5.5.2 Upscaling point estimates of recharge Regression equations The median of the 1000 replicates of point recharge is used in an exploration of the covariates chosen for upscaling (Figure 5-11and Figure 5-12). There is a positive correlation between rainfall and recharge (Figure 5-11a): as rainfall increases so does recharge. Clay content of the top 2 m of the soil profile shows a negative correlation (Figure 5-11b): as clay content increases recharge decreases. The relationship with the Normalised Difference Vegetation Index (NDVI) is not as expected: a higher vegetation density should result in lower recharge. The positive correlation seen (Figure 5-11c) is due to the relationship between rainfall and NDVI. As higher vegetation density is expected with higher rainfall, the relationship with rainfall needs to be accounted for before the negative correlation with recharge can be seen. The three geology classes show a similar slope of regression line (Figure 5-12) but differ in their intercepts. The high-recharge class has the largest intercept, and the low-recharge class has the smallest intercept. Figure 5-11 Point-scale relationships between recharge and (a) rainfall, (b) clay content of the soil, and (c) Normalised Difference Vegetation Index (NDVI) Figure 5-12 Point-scale relationships between rainfall and recharge by geology class The combined results of the covariates in predicting recharge is analysed through multiple linear regression. The parameter results are shown as boxplots in Figure 5-13. The value of βgeo for the high-recharge class is greater (less negative) than the values for the medium- and low-recharge geology classes, indicating that the high-recharge class does indeed have higher recharge than the other classes, all other variables being equal. The coefficients for the clay content and NDVI are negative, as they should be: increases in clay content or NDVI should lead to decreased recharge. The results of the regression equation can be measured by the R2 as a measure of the goodness of fit of the models. The R2 had a mean of 0.59 and a range of 0.45 to 0.72 for the 1000 replicates. This means that the regression equation is only able to explain (on average) 59% of the variance in the recharge estimates. This is probably a reflection of a noisy dataset rather than a poor model and is similar to other predictions of recharge using regression analysis (Crosbie et al., 2018; Crosbie and Rachakonda, 2021). Figure 5-13 (a-f) Coefficients used in the regression equations for upscaling the 1000 replicates, (g) the R2 for each of the 1000 replicates The line in the centre of the box is the median, the box represents the interquartile range (the 25th to 75th percentiles), the whiskers represent the 10th and 90th percentiles and the dots are any data points outside the 10th and 90th percentiles. NDVI = Normalised Difference Vegetation Index, Geo_H = high-recharge class, Geo_M = moderate-recharge class, Geo_L = low-recharge class. Kriging residuals The point estimates of recharge can be upscaled using the fitted regression equations and the spatially mapped covariates. The mean of the 1000 replicates is shown in Figure 5-14a. The spatial trends in the recharge follow the rainfall trends with the highest recharge in the areas with the highest rainfall and vice versa. The residuals of the fitted recharge from the regression equation were kriged to create a surface, the mean of the 1000 replicates is shown in Figure 5-14b. This surface has a mean of zero and a mean absolute residual of the upscaled surface of 0.21. In areas without data points, the residual surface tends to zero. This is by design, as in these areas the recharge estimates will depend upon the regression equations. In areas with data points the regression estimates will be ‘corrected’ toward the point values of recharge. The variogram used has a nugget fixed at 0.1 which prevented over-fitting; this can be seen in the lack of ‘bullseyes’ around each data point. The range of the variograms is approximately 50 km, which is the distance that each data point can influence. If some structure can be found in the patterns of the residual surface, then this could be brought into the regression equations to reduce the residuals and increase the fit of the regression equations, but the cause of the patterns could not be identified in this study. Figure 5-14 (a) The median value of the 1000 replicates of upscaled recharge using the regression equation and (b) the median of the residuals kriged to a regular grid and the points used Aggregating recharge rasters When the residuals are added back to the regression estimates of recharge, the final set of regression kriging upscaled recharge estimates is created. These are displayed as the 5th, 50th and 95th percentiles of the 1000 replicates in Figure 5-15. Over the entire region being investigated, the 50th percentile surface of recharge has a mean of 25 mm/year, the 5th percentile has a mean of 13 mm/year and the 95th percentile has a mean of 47 mm/year. Figure 5-15 The (a) 50th, (b) 5th and (c) 95th percentiles of upscaled recharge from the 1000 replicates using regression kriging The 50th percentile upscaled recharge estimates on a 0.05° (~1000 m) grid are compared to the median point estimates of recharge in Figure 5-16. This shows that the mean R2 for the 1000 replicates has increased from 0.59 for the regression equation to 0.87 for the regression kriging, demonstrating a much better fit to the point estimates of recharge from the CMB method. Figure 5-16 (a) Scatterplot of the 50th percentile of the recharge calculated at a point scale versus the 50th percentile of the regression kriging upscaled recharge estimate (red line is 1:1 for reference) and (b) the R2 of all 1000 replicates for the point estimates of recharge versus the upscaled estimates of recharge as a boxplot Constraining the uncertainty The 1000 replicates were constrained using the baseflow at the low end and the excess water at the high end in an attempt at reducing the uncertainty in the recharge estimates. Of the 10,000 random selections from the 1,000 replicates, 5,282 passed the rejection sampling criteria. This process only saw marginal improvement in the uncertainty with more samples being rejected at the low end by the baseflow observations than by the excess water observations at the high end. The constrained estimates of recharge are displayed as the 5th, 50th and 95th percentiles in Figure 5-17. Over the entire region being investigated, the 50th percentile surface of recharge has a mean of 27 mm/year, the 5th percentile has a mean of 15 mm/year and the 95th percentile has a mean of 49 mm/year. Figure 5-17 The (a) 50th, (b) 5th and (c) 95th percentiles of constrained recharge for the modelled area The constrained recharge when clipped to the Roper catchment is shown in Figure 5-18. The 50th percentile surface of recharge ranges from 0.7 to 317 mm/year with a mean of 23 mm/year. The 5th percentile surface of recharge ranges from 0.5 to 197 mm/year with a mean of 14 mm/year, and the 95th percentile surface of recharge ranges from 1.1 to 541 mm/year with a mean of 39 mm/year. Figure 5-18 The (a) 50th, (b) 5th and (c) 95th percentiles of constrained recharge for the Roper catchment 5.5.3 Extracting recharge values for zones of interest Recharge rates were extracted for the major aquifers as their full extents, not just the outcropping areas (Table 5-10). The recharge to the CLA in the Daly and Georgina basins is comparatively low within the Roper catchment as they are buried beneath a veneer of overlying Cretaceous strata that limit recharge, in both cases the majority of these aquifers are outside the Roper catchment. Similarly, the CLA in the Wiso Basin only has a small part of its extent within the Roper catchment but has higher recharge than the CLA in the Daly and Georgina basins within the catchment. The Dook Creek Formation has about half of its area within the Roper catchment and has the highest recharge rate of the major aquifers due to the higher rainfall in the area, its mean rate is similar for the whole extent and the portion within the Roper catchment. Table 5-10 Mean recharge rates over each of the major aquifers. The 50th percentile is outside the brackets and the 5th and 95th percentiles are within the brackets, giving a range for the uncertainty MAJOR AQUIFERS RECHARGE (mm/y) CLA – Daly Basin 8 (5–13) CLA – Wiso Basin 23 (12–37) CLA – Georgina Basin 5 (3–9) DCA – Dook Creek Formation (whole) 56 (35–88) DCA – Dook Creek Formation (Roper) 58 (36–93) Knuckey Formation 11 (6–22) The mean recharge rates aggregated to the simplified surface geology classes are shown in Table 5-11. The highest recharge was seen in the JK (Jurassic–Cretaceous) partial aquifer of the Great Artesian Basin (outside the Roper catchment) as this is in the highest rainfall part of the catchment. Table 5-11 Mean recharge rates over the simplified surface geology classes. The 50th percentile is outside the brackets and the 5th and 95th percentiles are within the brackets, giving a range for the uncertainty SIMPLIFIED SURFACE GEOLOGY RECHARGE (mm/y) Alluvium colluvium 16 (10–27) Aquitard 28 (16–52) Cenozoic aquifer 2 (1–4) Cenozoic low-permeability sediments 18 (10–30) JK partial aquifer 131 (78–216) Mesozoic aquitard 17 (10–29) Palaeozoic or older aquifer 40 (26–63) Palaeozoic or older partial aquifer 30 (18–52) Quaternary sediments undifferentiated 15 (7–31) Volcanics 52 (31–93) 5.6 Identifying potential groundwater discharge areas using remote sensing 5.6.1 Digital Earth Australia The Digital Earth Australia water bodies dataset (Krause et al., 2021) identifies 2447 water bodies that are greater than 0.3 ha in size and contain water more than 10% of the time (Figure 5-19). Only 141 of these water bodies contain water more than 75% of the time and can be considered semi-permanent or permanent. Most of these occur along the Roper and Arnold rivers, the latter is the south-eastern tributary of the Hodgson River. The water bodies along the Arnold River are associated with underlying sandstones of the Roper Group (Bessie Creek and Hodgson sandstones) and are not focused upon here due to the low potential for using the underlying aquifers as a source of groundwater for irrigation. In the region around Mataranka, where there is known groundwater discharge (Karp, 2008), this dataset is identifying some permanent water bodies but not others. Nothing is identified from Bitter or Rainbow springs where vegetation is quite dense. Permanent water bodies are identified downstream of Fig Tree Spring, where the tufa aquifer outcrops and in-river springs have been previously mapped by Karp (2008), and further downstream in the Roper River from the confluence of Elsey Creek down to Red Lily Lagoon. Off the streamlines within Elsey National Park, many temporary (<75% of the time inundated) water bodies are identified within the known groundwater discharge zone (Karp, 2008; Tickell, 2016). The lower reaches of Salt and Elsey creeks are known to be groundwater discharge zones but are not identified here as permanent water bodies due to dense vegetation, only Warloch Ponds on Elsey Creek has been identified as a permanent water body. Some permanent water bodies are identified in the reach between Red Lily Lagoon and Ngukurr where water bodies like Boomerah Lagoon, Duck Ponds and Yawurrwarda have been previously mapped (Kneebone, 1960; Zaar et al., 2009d). Some are also identified along the lower reach of the Wilton River where Lake Allen is (Brooks and Kneebone, 1963). However, a much larger amount and extent of permanent water is identified further downstream of Ngukurr. These are associated with the Limmen Bight (Port Roper) Tidal Wetland System, which is recognised nationally through its listing on the Directory of Important Wetlands (Department of the Environment and Energy, 2010). This wetland system consists of intertidal mud flats, saline coastal flats and estuaries. Its water sources are listed as inundation by surface water or sea water during spring tides or storm surges – no mention is made of it being a groundwater discharge area. The streams crossing the Dook Creek Formation have been previously identified as receiving groundwater discharge (Zaar and Tien, 2003). At the western end, no permanent water bodies are identified along Maiwok or Flying Fox creeks, which is consistent with findings by Augustine (1962). There are several permanent water bodies on the Mainoru River tributaries, but these are upstream of the Dook Creek Formation overlying Roper Group sediments. Several permanent water bodies occur near the Wilton River at the northern extremity of the Dook Creek Formation where there are springs associated with the Roper Group and spring-fed vegetation (Zaar and Tickell, 2003). There is also a large permanent (>90%) water body where the Wilton River leaves the Dook Creek Formation where groundwater discharge occurs at Weemol Spring (Zaar and Tien, 2003). A few more permanent water bodies occur on the Wilton River further downstream from the Dook Creek Formation, approaching the confluence with the Roper River overlying Roper Group sediments. Lake Allen is the largest of these (Brooks and Kneebone, 1963). Outside and to the north-east of the Roper catchment are permanent water bodies identified on Guyuyu Creek and the Goyder River where they traverse the Dook Creek Formation (not shown on Figure 5-19), and multiple springs have been previously mapped (Department of Environment Parks and Water Security, 2014c). No permanent water bodies are identified south of the Roper River in the Nathan Group carbonates near Ngukurr, but a few waterholes were previously mapped in this area (Brooks and Kneebone, 1963; Kneebone, 1960; Zaar et al., 2009d). On the western side of the Roper catchment, overlying the Wiso Basin, numerous water bodies are identified that are associated with sinkholes. Some of these are identified as permanent (including Chowyung Waterhole, Brolga Waterhole, Rocky Hole, Lake Duggan) but are considered recharge features that are capturing surface water (Yin Foo, 2002), as are Stuart Swamp near Daly Waters. Permanent water bodies were also mapped along the Arnold River, an eastern tributary of the Hodgson River south of Minyerri, where numerous waterholes are known to exist (Paine, 1963). Figure 5-19 Water bodies in the Roper catchment identified (from Digital Earth Australia) showing the proportion of time that water bodies are inundated (from Water Observations from Space) Note: the size of the water bodies has been exaggerated so that they can be seen at the catchment scale. 5.6.2 Excess water The map of excess water, calculated as precipitation (P) minus AET from the CMRSET v2.2 dataset, across the Roper catchment (Figure 5-20) shows which areas are exporting water (green) and which are importing water (purple). Areas that are importing water are potentially groundwater discharge areas (or floodplains, irrigation areas etc.). At the scale of the Roper catchment, only the very large areas importing water can be seen. These include the known groundwater discharge zone for the CLA in and near Elsey National Park around Mataranka and the Limmen Bight (Port Roper) Tidal Wetland System on the coast. More detail is shown the next section. Figure 5-20 Excess water across the Roper catchment 5.6.3 Potential discharge areas on carbonates The area around Mataranka within Elsey National Park is a known groundwater discharge area and therefore a good place to test potential identification methods. Applying the maximum proportion of time inundated (from the Water Observations from Space (WOfS) dataset) to the water bodies dataset (from Digital Earth Australia (DEA)) shows that there are many temporary water bodies within this area, and some permanent water bodies are associated with perennial groundwater discharge from the CLA along reaches of the upper Roper River and lower reach of Elsey Creek (Figure 5-21a). These same areas can also be seen as areas of negative excess water, indicating that evapotranspiration is much greater than rainfall (Figure 5-21b). This is not necessarily identifying groundwater discharge areas but only identifying potential discharge areas because other sources of water could be contributing (e.g. runoff from further up the catchment filling waterholes and contributing to bed and bank storage). A further exploration is undertaken of the seasonality of the evapotranspiration. During the wet season, water availability is high due to the rainfall and so evapotranspiration is at a maximum (relative to potential evapotranspiration). Figure 5-21c for February evapotranspiration shows that most of the area has an actual evapotranspiration between 50 and 100% of the potential evapotranspiration. During the dry season, most of the area has low evapotranspiration as there is no rainfall input Into the system so the evapotranspiration needs to come from storage (Figure 5-21d,e,f). Groundwater discharging from regional groundwater systems would be expected to maintain high evapotranspiration throughout the dry season as the flow paths are long and the dynamics damped. This can be seen in the persistently high evapotranspiration through the dry season along Elsey Creek and the upper Roper River and in the area in Elsey National Park with temporary water bodies. The areas known to be groundwater discharge areas can be seen as having high evapotranspiration in October at the end of the dry season. This observation is enough to identify potential groundwater discharge areas but cannot be used definitively as the source of the high evapotranspiration has not been identified (i.e. soil water versus groundwater). Analysis around Mataranka has shown that high October evapotranspiration has the potential to be used as a screening tool to identify areas of potential groundwater discharge for further targeted investigations. An investigation of tree water sources in Elsey National Park was carried out as part of this assessment as described in Section 4.2.12. Results of this investigation are presented in Section 6.2.15. Figure 5-21 Mataranka area showing (a) the location in relation to aquifers and streams with the Digital Earth Australia water bodies classified by proportion of time inundated according to Water Observations from Space, (b) mean annual excess water, (c) mean CMRSET actual evapotranspiration (AET) for February with the colour scale maximum at the mean FAO56 potential evapotranspiration, the same is also shown for (d) June, (e) August and (f) October The same approach (i.e. identifying areas with high evapotranspiration in October)can be used in other less-studied regions of the Roper catchment to investigate the potential for groundwater discharge locations. The areas underlain by the carbonate rocks of the Dook Creek Formation and the sandstones of the Roper Group are examined here in more detail. A small temporary water body is identified in the DEA dataset near where Maiwok Creek completes its crossing of the outcropping area of the Dook Creek Formation. This area has a high evapotranspiration rate in February (Figure 5-22c) but the high rate does not continue through the dry season (Figure 5-22d,e,f), so this is not a prospective area for groundwater discharge. No permanent water bodies are identified along Flying Fox Creek within the reach that crosses the Dook Creek Formation using the DEA dataset (Figure 5-22a). However, (Figure 5-22b) shows that the lower part of the reach has a highly negative excess water between Jamberline Waterhole and Lunaarracomp waterhole (Zaar and Tickell, 2003) that continues downstream of the outcrop of the Dook Creek Formation. This area continues to have high actual evapotranspiration relative to potential evapotranspiration throughout the dry season (Figure 5-22d,e,f) suggesting it may be a groundwater discharge area. A group of temporary water bodies to the north of the Mainoru River are identified in the DEA dataset (Figure 5-23a). Although these have a high negative excess water (Figure 5-23b) and high evapotranspiration throughout the dry season (Figure 5-23d,e,f), they are probably internally draining wetlands that act as recharge features in the landscape. Downstream of the outcropping area of the Dook Creek Formation are areas of high excess negative water (Figure 5-23b) that coincide with the braided section of the Mainoru River. Near the downstream edge of the Dook Creek Formation are several temporary water bodies identified from the DEA dataset and also Top Springs (Figure 5-23a). Although small at the scale of the maps, these areas have high negative excess water (Figure 5-23b) and high evapotranspiration throughout the dry season (Figure 5-23d,e,f), indicative of groundwater discharge. Two more springs (White Rocks Spring and Lindsay Spring) are mapped to the north of Mainoru River at the edge of the outcrop of the Dook Creek Formation (Figure 5-23a). Although mapped as discrete discharge features, neither seems significant at this scale as they are not associated with a permanent water body or high negative excess water. Many temporary water bodies identified are around the Wilton River over the outcrop of the Dook Creek Formation, and two could be considered permanent (Figure 5-24a). There are also a number of springs. Although small at this scale, some areas downstream of Weemol and Wiamuna Hot Springs have high negative excess water (Figure 5-24b) and continue to have high evapotranspiration throughout the dry season (Figure 5-24d,e,f), indicative of groundwater discharge. Two springs further north (Little Spring and Bodeidei Spring) also have high negative excess water and high evapotranspiration throughout the dry season, although to a lesser degree. Just downstream of the outcrop of the Dook Creek Formation is a permanent water body near Bulman Waterhole on the Wilton River (Figure 5-24a) that has high negative excess water (Figure 5-24b) and maintains high evapotranspiration through the dry season (Figure 5-24d,e,f). This appears to be fed by a 5 km stretch of the Wilton River over the Dook Creek Formation that also has high negative excess water and maintains high evapotranspiration through the dry season. Figure 5-22 Maiwok and Flying Fox creeks over the Dook Creek Formation showing (a) the location in relation to aquifers and streams with the DEA water bodies classified by proportion of time inundated according to Water Observations from Space, (b) mean annual excess water, (c) mean CMRSET actual evapotranspiration for February with the colour scale maximum at the mean FAO56 potential evapotranspiration, and the same for (d) June, (e) August and (f) October Figure 5-23 Mainoru River over the Dook Creek Formation showing (a) the location in relation to aquifers and streams with the DEA water bodies classified by proportion of time inundated according to Water Observations from Space, (b) mean annual excess water, (c) mean CMRSET actual evapotranspiration (AET) for February with the colour scale maximum at the mean FAO56 potential evapotranspiration, and the same for (d) June, (e) August and (f) October Figure 5-24 Wilton River over the Dook Creek Formation showing (a) the location in relation to aquifers and streams with the DEA water bodies classified by proportion of time inundated according to Water Observations from Space, (b) mean annual excess water, (c) mean CMRSET actual evapotranspiration (AET) for February with the colour scale maximum at the mean FAO56 potential evapotranspiration, and the same for (d) June, (e) August and (f) October The outcropping area of the Dook Creek Formation continues to the north-east outside the Roper catchment where there are numerous discrete discharge locations on the Goyder River and Guyuyu Creek, which both flow north to the Arafura Sea. There are many temporary water bodies identified by the DEA dataset, and a few that can be considered semi-permanent (Figure 5-25a). Many of these water bodies are off the stream network and associated with sinkholes so are probably recharge features. These water bodies have high negative excess water (Figure 5-25b). Four mapped springs, which discharge into Annie Creek (Emu Springs), Goyder River (Wurrkal Spring and Emu Spring No2) and Guyuyu Creek (Gakorrorodjh) (Figure 5-25a), all have high negative excess water (Figure 5-25b). This region is the last point of discharge for aquifers hosted in the Dook Creek Formation. Many of the water bodies maintain high evapotranspiration through the early dry season (Figure 5-25d,e) with less evapotranspiration in the late dry season (Figure 5-25f). This is consistent with a waterhole or wetland sourced from dry season run-on and then drying out through the dry season. High dry-season evapotranspiration is maintained from Wurrkal Spring and downstream on the Goyder River, indicating year-round groundwater discharge (Figure 5-25d,e,f). This is also seen in the last 10 km of Guyuyu Creek flowing across the Dook Creek Formation. The Knuckey Formation is another carbonate formation that surrounds Ngukurr. In this area, permanent water bodies are identified along the Roper and Wilton rivers and some temporary water bodies to the south-east (in the catchment of the Towns River) (Figure 5-26a). The major river channels of the Roper, Wilton and Hodgson rivers show high negative excess water indicative of water inflows from upstream (Figure 5-26b). There is high evapotranspiration during the wet season (Figure 5-26c) but this does not persist through the dry season except for the major river channels (Figure 5-26d,e,f). Although springs are mapped in the area (Figure 5-26a), there is no evidence that they are a major source of groundwater discharge. The Roper River increases in width in this reach which could be indicative of some groundwater discharge from the Knuckey Formation, which is consistent with findings by Sumner (2008). Figure 5-25 Guyuyu Creek and Goyder River over the Dook Creek Formation showing (a) the location in relation to aquifers and streams with the DEA water bodies classified by proportion of time inundated according to Water Observations from Space, (b) mean annual excess water, (c) mean CMRSET actual evapotranspiration (AET) for February with the colour scale maximum at the mean FAO56 potential evapotranspiration, and the same for (d) June, (e) August and (f) October Figure 5-26 Ngukurr over the Dook Creek Formation showing (a) the location in relation to aquifers and streams with the Digital Earth Australia water bodies classified by proportion of time inundated according to Water Observations from Space, (b) mean annual excess water, (c) mean CMRSET actual evapotranspiration (AET) for February with the colour scale maximum at the mean FAO56 potential evapotranspiration, and the same for (d) June, (e) August and (f) October The previous figures showed that the areas of probable groundwater discharge are characterised by actual evapotranspiration that continues through the dry season at a relatively high rate. A mean rate of 2.5 mm/day in October was selected as a threshold for areas across the outcropping area of the Dook Creek Formation to delineate potential discharge areas (Figure 5-27). This shows that around 1% of the outcropping area of the Dook Creek Formation is a potential discharge area. Figure 5-27 Potential discharge zones to the streams overlying the outcropping area of the Dook Creek Formation Note: coloured areas are exaggerated in size so that they can be seen at this scale. The time series of actual evapotranspiration for these areas shows a general trend through all months (Figure 5-28). All potential discharge areas across the outcropping area of the Dook Creek Formation have actual evapotranspiration that is below potential evapotranspiration for all months, with February being the closest. The gap between actual evapotranspiration and potential evapotranspiration widens as the dry season continues from June through to October. Similarly, the gap between the actual evapotranspiration of areas not considered potential discharge areas falls further behind the potential discharge areas as the dry season progresses, indicating the depletion of soil water storage. The north-flowing rivers have consistently higher actual evapotranspiration than the tributaries of the Roper River through the dry season, which is potentially a sign of greater volumes of discharge to the north than the south. Figure 5-28 Time series of actual evapotranspiration for potential discharge zones on Dook Creek Formation (DCF) If it is assumed that the difference in ET between the rest of the Dook Creek Formation and the identified potential discharge areas is due to groundwater discharge, then the volume of groundwater being discharged can be calculated (Table 5-12). This works out to be 17.9 GL/year, which is only 5% of the recharge over the outcropping area of the Dook Creek Formation estimated using the chloride mass balance method (Table 5-10). Most of the groundwater recharge is being discharged elsewhere, including baseflow to streams, smaller areas of discharge via ET not identified here, groundwater extraction and groundwater flow to the confined parts of the aquifer hosted in the Dook Creek Formation, and groundwater flow to other adjacent hydrogeological units. Table 5-12 Calculation of volume of groundwater discharge from the Dook Creek Formation due to evapotranspiration POTENTIAL DISCHARGE AREA AREA (km2) AET (mm/y) EXCESS AET (mm/y) EXCESS AET (ML/y) Guyuyu Creek 8 1111 296 2,452 Goyder River 5 1208 394 1,980 Wilton River 45 1062 248 11,044 Mainoru River 4 1042 228 819 Flying Fox Creek 6 1066 252 1,613 Rest of DCF 6527 814 0 0 Total 6595 17,908 AET = actual evapotranspiration. 5.6.4 Potential discharge areas over whole catchment Applying Equation (9) over the Roper catchment resulted in a threshold value of mean October AET that had to be exceeded for a 30 × 30 m pixel to be considered a potential groundwater discharge area. This threshold resulted in most of the catchment being excluded as potential groundwater discharge areas. Figure 5-29 shows the known groundwater discharge area around Elsey National Park, and this threshold of October AET has correctly identified the spring features, the areas that have been previously mapped, and Elsey Creek, Roper Creek and Waterhouse River. Figure 5-29 (a) Mean October actual evapotranspiration (AET) for the known discharge area around Mataranka and (b) areas remaining with a high AET after excluding areas with a low October AET Some areas have a high October AET but the source of water is likely run-on of surface water in the wet season that has been stored in wetlands, floodplains and sinkholes before being evapotranspired in the dry season. These areas can often be identified as having a higher variability in inter-annual October AET than groundwater discharge sites, which are less variable. Figure 5-30a shows some example wetlands and sinkholes between Guyuyu Creek and the Goyder River overlying the outcropping area of the Dook Creek Formation that are excluded after filtering with the coefficient of variation of October evapotranspiration (Figure 5-30b). Figure 5-30 (a) Coefficient of variation (CV) of October actual evapotranspiration (AET) in areas remaining after excluding areas of low October AET overlying the Dook Creek Formation (outside the Roper catchment) and (b) areas remaining after excluding areas with a high CV of October AET In more arid areas with a high proportion of bare soils, dark coloured or red soils can lead to evapotranspiration being overestimated by the CMRSET algorithm. These areas are not filtered out by the criteria based on high October evapotranspiration with low variability. However, it is often clear that their topographic position and lack of vegetation would preclude them from being a discharge area and that they need to be excluded. The DEA Fractional Cover dataset (Lymburner, 2021) has been applied to successfully exclude these areas. There were 15,029 polygons identified in the Roper catchment as potential groundwater discharge features. Of these, 3052 were larger than 0.5 ha and were considered further. It is these larger potential discharge areas that are important at this regional scale when considering the water resource assessment. Smaller discharge features can potentially be identified down to a 900 m2 pixel size, but these would be very localised in scale and difficult to consider at the regional scale. Not all these 3052 polygons are groundwater discharge features: any permanent water bodies that are above the watertable and act as recharge features will also have been identified. Without adequate depth to groundwater data across the entire catchment, these features cannot be algorithmically excluded and must be manually excluded. Some other features identified as potential groundwater discharge areas make little sense when investigated in detail because of topographic position, geologic conditions or simply an unknown reason for overestimated October evapotranspiration. These areas all need to be inspected individually and classified manually. So each polygon was inspected individually and assigned to one of five classes (Figure 5-31, Table 5-13): • perennial groundwater discharge • seasonally varying • coastal • recharge feature • mis-identified? The ‘perennial groundwater discharge’ class included 1342 polygons for a total area of 3717 ha. These features are predominantly springs in the mid to upper reaches of catchments and are related to geological contacts (Figure 5-31). The ‘seasonally varying’ class included 1111 polygons for a total area of 5238 ha (Figure 5-31). These features are mostly associated with the alluvium in the lower reaches of the Roper River and its tributaries (although not exclusively so). They are conceptualised as being recharged by surface water during the wet season and then discharging this water during the dry season through evapotranspiration. There may also be a component of groundwater discharge from localscale groundwater systems hosted in the Roper Group where large parts of the middle and lower reaches of the river traverse this hydrogeological unit. The ‘coastal’ class included 565 polygons for a total area of 9416 ha (Figure 5-31). These features are within the estuary of the Roper River and are associated with the Limmen Bight (Port Roper) Tidal Wetland System. These areas may have a component of groundwater discharge along with the evapotranspiration of sea water. The ‘recharge’ class included 29 polygons for a total of 32 ha. These features were identified as they hold water semi-permanently or permanently but are tens of metres above the watertable, such as across parts of the Sturt Plateau. They are mainly sinkholes and waterholes that capture run-on during the wet season and retain this water through the dry season. These identified recharge features are a by-product of identifying discharge areas; there are many more recharge features that have not been identified here. The ‘mis-identified’ class included five polygons for a total of 4 ha. These features seem to have anomalously high October evapotranspiration without a geological explanation. A map of a large green and red area Description automatically generated Figure 5-31 Areas of potential groundwater discharge across the Roper catchment Note: the size of the polygons has been greatly exaggerated so they can be seen at this scale. Table 5-13 Summary of areas identified as potential groundwater discharge areas CLASS COUNT AREA (ha) Perennial groundwater discharge 1342 3417 Seasonally varying 1111 5,238 Coastal 565 9,416 Recharge 29 32 Mis-identified? 5 4 Total 3052 18,187 The territory-wide springs database (Department of Environment Parks and Water Security, 2014c) is the best comparison dataset to assess the performance of this method. This database has been compiled over decades of field work identifying springs in the field, but it does not purport to be complete or to capture all the major springs. Of the 132 springs identified, 3 are classified as ‘major’: Rainbow Springs and Bitter Springs at Mataranka, and Mount Catt Spring on the Wilton River. Rainbow and Bitter springs were correctly identified here, but Mount Catt Spring was excluded as having an area of less than 0.5 ha of high evapotranspiration – it covered three pixels rather than the threshold of six pixels used here. If the 0.5 ha criterion is ignored, then the median distance between the springs in the NT springs database and those identified here is 280 m. Considering the low-level of accuracy of the mapping in the database (the location of some springs is noted as approximate and others as being hundreds of metres upstream it is estimated that this method correctly identified half of the springs in the NT springs database in the study area. Some non-identification of springs could be scale dependent: a pixel in the remote sensing data is 900 m2, so small seeps and soaks will not be identified. A better test of the accuracy of the springs identified here would be field verification of the more than 1000 springs identified that are not currently in the territory-wide springs database. 5.6.5 Summary of potential discharge areas from remote sensing Mapping water bodies from DEA and assessing their permanence using WOfS is useful for identifying water bodies larger than 0.3 ha. The pixel size is 25 m from Landsat, so minor pools and waterholes within streams are often too small to be identified using this platform. (Along much of its length the Roper River is not identified as a water body.) Excess water calculated using CMRSET data identified many areas that had evapotranspiration that was much greater than rainfall but some of these (mostly streamlines and wetlands receiving run-on) were not groundwater discharge areas. October evapotranspiration from CMRSET provided a much more useful guide to the location of groundwater discharge areas. Areas that are still using a lot of water at the end of the dry season have another source other than rainfall; in a lot of cases this will be groundwater. A threshold value of 2.5 mm/day was useful across the outcropping area of the Dook Creek Formation where there was not a strong climate gradient but did not scale well to the whole catchment. To enable this method to be used across the whole of the Roper catchment, this threshold value became dependent upon the aridity index which changes spatially due to the climate gradient at the scale of the catchment (Barrow, 1992). The method developed here using the threshold October evapotranspiration identified 1348 potential springs across the Roper catchment compared to the 132 springs that have been documented from field studies. This method successfully identified about half of the springs in the database; some differences will be due to the scale of investigation. The next step is field verification of the potential springs that have not yet been mapped in the territory-wide springs database including: (i) field site visit during the late dry-season (i.e. September to October) and recording observations of the presence of absence of water at the land surface and the geographic coordinates of the site, (ii) in-situ observations of physical (i.e. temperature) and chemical parameters (i.e. EC, pH) of the water, and (iii) collection of water samples for major ion chemistry to compare with groundwater chemistry data. 6 Targeted field, desktop and modelling investigations 6.1 Hydrogeological framework 6.1.1 Drilling investigations Cambrian Limestone Aquifer The results of drilling investigations are summarised in Table 6-1, including drilled depth, screen interval, water level and water quality parameters taken during bore development. Drilling and installing four monitoring bores required five boreholes with a cumulative total of 505 m of drilling (one borehole, RN043048, was backfilled and decommissioned). All drilling was conducted within Elsey National Park where the Tindall Limestone either outcrops or occurs at shallow depths beneath a thin veneer (>15 mBGL) of overlying Cenozoic (sand, loam and clay) and or Cretaceous (sandy clay) sediments. Boreholes RN043045 and RN043047 were used as moderate-depth (>100 mBGL) investigation holes to characterise the stratigraphy at the two locations and guide the targeted installation of the four monitoring bores. Boreholes RN043045 and RN043047 were completed at depths of 186 and 174 mBGL, respectively. Both boreholes intersected the middle Cambrian Tindall Limestone, which ranged in thickness from about 28 m at RN043045 to 35 m at RN043047. The limestone was mostly comprised of hard to weathered and karstic limestone with minor siltstone, mudstone and sandstone interbeds. Underlying the Tindall Limestone, the early Cambrian APV, composed of hard, dark grey, fine-grained basalt, was encountered in both boreholes. It ranged in thickness from 63 m in RN043047 and 76 m in RN043045. Underlying the APV at the most northern site (RN043045) closest to the upper Roper River and Bitter Springs, was the Mesoproterozoic Walmudga Formation of the Nathan Group in the McArthur Basin. The Walmudga Formation, composed mostly of purple fine-grained siltstone, had a partial thickness of 68 m which is where the borehole ceased at a total depth of 186 mBGL. Underlying the APV at the most southern site (RN043047) closest to the southern boundary of the national park and just west of Salt Creek, was the Paleoproterozoic Urapunga Granite, which forms the crystalline basement of the McArthur Basin. For more detailed information on the lithology and stratigraphy logs and bore construction diagrams see Appendix A.1 containing Apx Figure A.1-1 to Apx Figure A.1-4. At both drilling sites in the national park (southern and northern), two monitoring bores were constructed and installed on the same drill pad adjacent each other, one in the Tindall Limestone and one in the underlying APV. Both underlying APV monitoring bores (RN043045 and RN043047) were constructed with 100 mm inside diameter Class 12 PVC pipe and discretely screened in the APV by cementing off the top 10 m of the weathered and fractured basalt, followed by another metre of bentonite seal before the screened interval composed of 8 m lengths of 2 mm slotted casing. Both bores had completion standing water levels (SWLs) ranging from 16.5 to 19.5 mBGL, very low bore yields (<0.4 L/second), slightly brackish salinities and a slightly alkaline pH (see Table 6-1). The two monitoring bores in the Tindall Limestone (RN043046 and RN043049) were constructed with steel casing ranging in diameter from 158 to 206 mm inside diameter discretely screened in the Tindall Limestone by cementing off the top 6 m of overlying Cenozoic and or Cretaceous cover. Both bores were completed with stainless steel wire-wound screens ranging in length from 2 to 4.3 m with a 3 mm aperture. Note that RN043049 had two screened intervals Table 6-1. In addition, cementing and bentonite sealing of more of the overlying rocks on both bores could not be conducted due to the cavernous and karstic nature of the limestone. For more detail on the lithology and stratigraphy logs and bore construction diagrams see Apx Figure A.1-1 to Apx Figure A.1-4 in Appendix in A.1. Bore reports for all four sites can also be found online: • RN043045 – https://ntlis.nt.gov.au/hpa-services/borereport?bore=RN043045 • RN043046 – https://ntlis.nt.gov.au/hpa-services/borereport?bore=RN043046 • RN043047 – https://ntlis.nt.gov.au/hpa-services/borereport?bore=RN043047 • RN043048 – https://ntlis.nt.gov.au/hpa-services/borereport?bore=RN043048 • RN043049 – https://ntlis.nt.gov.au/hpa-services/borereport?bore=RN043049. Table 6-1 Summary of details for new monitoring bores installed in Elsey National Park BORE NO. COMPLETED DATE PURPOSE AQUIFER COMPLETED DEPTH (mBGL) SCREEN INTERVAL (mBGL) GROUNDWATER LEVEL (mBGL) BORE YIELD (L/s) SALINITY AS EC (μS/cm) pH RN043045 1/11/2022 Monitoring Antrim Plateau Volcanics 62.7 56.7 – 62.2 19.5 0.3 982 8.0 RN043046 4/11/2022 Monitoring Tindall Limestone 47 37.6 – 41.9 5.4 30+ 727 7.3 RN043047 17/11/2022 Monitoring Antrim Plateau Volcanics 74 68.0 – 73.5 16.5 0.1 2152 6.5 *RN043048 1/12/2022 Backfilled Antrim Plateau Volcanics 47.7 – 5.7 4 1564 7.4 #RN043049 30/11/2022 Monitoring Tindall Limestone 49.7 (1) 31.0 – 33.0 (2) 44.7 – 48.7 5.9 50+ 1566 7.5 #RN043049 has two separate screened intervals. EC = electrical conductivity. *RN043048 was backfilled. 6.1.2 Hydrogeological cross-sections Cambrian Limestone Aquifer Regional-scale sections The locations of four regional-scale hydrogeological cross-sections are shown in Figure 6-1. One was adapted from Tickell (2016) and traverses from north to south. The other three were extracted from the DR2 Leapfrog model (Knapton, 2020) and traverse west to east across the CLA in the Wiso, Daly and Georgina basins beneath the Roper catchment. The four sections are presented in Figure 6-2, Figure 6-3, Figure 6-4, and Figure 6-5. These sections summarise the twodimensional spatial context of the three key hydrogeological units at a regional scale. They provide the vertical and horizontal framework that underpins part of the hydrogeological conceptual model for regional groundwater flow in parts of the CLA. Figure 6-1 Locations of four regional-scale hydrogeological cross-sections traversing the Cambrian Limestone Aquifer in the Daly, Wiso and Georgina basins To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left inset with red polygon indicates the location of the map extent in the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008). Geological faults data source: Department of Industry, Tourism and Trade (2010). Section A–B The north–south section (A–B) presented in Figure 6-2 shows that the CLA hosted in the Cambrian limestone is thickest (100 to 150 m) either side of the regional groundwater discharge zone around Mataranka. Here a structural high in the underlying basalt which hosts the APV reduces the CLA to a thickness of about 50 to 75 m at the Roper River (RN035796) and Elsey Creek (RN036305) sites, respectively. This is consistent with the thin depth to basement (<200 mBGL) shown in Figure 2-9, noting that the APV hosted in the basalt is likely to be thinner than shown on this section as most groundwater bores do not penetrate the entire APV in this area. This has been further highlighted by the results of the new drilling (not incorporated in this existing section, see Figure 6-7) where crystalline basement was intersected at about 100 mBGL in RN043045 at about 7 km from Fig Tree Spring and the Roper River. This is also consistent with drilling at RN038810 which intersected crystalline basement at a depth of about 160 mBGL and about 50 km south of the Roper River. In addition, the section clearly highlights the absence or thinness of the overlying veneer of Cretaceous rocks and sediments where the CLA outcrops or subcrops beneath thin cover in the regional groundwater discharge zone around Mataranka. Furthermore, it highlights the spatial variability in the thickness of cover heading north or south of the Roper River, which is a key control on vertical leakage through the Cretaceous unit to recharge the underlying CLA. Figure 6-2 Hydrogeological cross-section A–B traversing from north-west of Mataranka in the Daly Basin to the south-east in the Georgina Basin due east of Daly Waters Cross-section location is shown in Figure 6-1. Section C–D The first of the west-to-east-traversing sections (C–D) is presented in Figure 6-3. This section starts on the western edge of the Cambrian Limestone in the Wiso Basin (outside the Roper catchment) and traverses the CLA over the structural high of the basalt and Birdum Creek fault around Larrimah, finishing on the eastern margin of the Cambrian limestone (east of Larrimah). A key feature of this section is the variable but thin saturated thickness of the CLA in the southern Daly Basin, west of the structural high of the basalt adjacent the Birdum Creek fault (CLA unsaturated at RN031902). In some parts of the southern Daly Basin and northern Wiso Basin west of the Birdum creek fault, the CLA has a saturated thickness of less than or equal to 30 m, and it is unsaturated around the structural high of the basalt near Larrimah. East of the Birdum Creek fault in the southern Daly Basin (RN038811), the saturated thickness of the CLA increases substantially (>100 m) for most of the remaining cross section before pinching out at RN031168. Figure 6-3 Hydrogeological cross-section C–D traversing from south-west in the Wiso Basin to the north-east in the Daly Basin south-east of Mataranka Cross-section location is shown in Figure 6-1. Section E–F Section E–F presented in Figure 6-4 exhibits similar spatial features to that shown in Section C–D (Figure 6-3). The CLA in the Wiso Basin has a thin and variable saturated thickness. It is partially unsaturated where the APV exhibits a high just south of Larrimah, and the CLA has a much larger saturated thickness in the southern Daly Basin before pinching out to the east at stratigraphic hole LRC008. The depth to groundwater in the CLA also increases in part of this section, particularly in the western part of the Roper catchment in the Wiso Basin where the Cretaceous cover is thicker (~100 m). Figure 6-4 Hydrogeological cross-section E–F traversing from south-west in the northern Wiso Basin to the northeast in the southern Daly Basin east of Larrimah Cross-section location is shown in Figure 6-1. Section G – H Section G–H, presented in Figure 6-5, exhibits a change to a more uniform saturated thickness for the CLA hosted in the Cambrian limestone across most of the cross section between the northern Wiso Basin in the west, and the northern Georgina Basin in the east. The CLA mostly exhibits a saturated thickness of approximately 100 m, though it becomes partially saturated at the edges of the aquifer where it pinches out. Also, the Cambrian siltstone of the Anthony Lagoon Formation (equivalent of the Jinduckin Formation) can be seen overlying the western part of the CLA between RN038150 and RN026477 (Figure 6-1, shows the northern edge of the Cambrian siltstone adjacent the cross section at RN038150). Figure 6-5 Hydrogeological cross-section G–H traversing from south-west in the northern Wiso Basin to the northeast in the northern Georgina Basin north east of Daly Waters Cross-section location is shown in Figure 6-1. Local-scale sections – Mataranka Three new hydrogeological cross-sections were developed from spatial analyses of existing hydrogeological data and incorporating information from the new drilling in Elsey National Park (see Section 6.1.1) (Figure 6-6). These sections provide intimate hydrogeological data at a local scale around the regional groundwater discharge zone. The three new sections summarise the twodimensional spatial context for multiple hydrogeological units in the area. They provide details on the vertical and horizontal framework that underpins part of the hydrogeological conceptual model for groundwater discharge around the upper Roper River and Mataranka Springs Complex. A map of a city Description automatically generated Figure 6-6 Locations of three local-scale hydrogeological cross-sections traversing the regional groundwater discharge zone of the Cambrian Limestone Aquifer in the Daly Basin around Mataranka To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower right inset indicates the map extent in the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008). Geological faults data source: Department of Industry, Tourism and Trade (2010) Groundwater discharge zone data source: Department of Environment and Natural Resources (2017) Spring data source: Department of Environment Parks and Water Security (2014c) Section 1 Section 1 presented in Figure 6-7 and traverses the Cambrian limestone (Cmt) from west of the Birdum Creek fault to just east of Elsey Creek. The section shows the offset of the Cambrian limestone (Cmt – Limestone and siltstone) on the western end of the section at the Birdum Creek fault (near RN030950). Here the Jinduckin Formation (Coij–Siltstone and limestone) is deposited over the Cambrian limestone. The Cretaceous cover (Kl) in this western part of the section is also reasonably thick (>100 m) between the fault and the Stuart Highway (RN031483). A pronounced feature is the basement high exhibited by the Proterozoic Urapunga Granite (Pg) and Roper Group sandstone and siltstone (Pr) of the McArthur Basin and its structural influence on the outcropping and subcropping of the Cambrian limestone in the discharge zone for the CLA around the lower reach of Elsey Creek (a key finding from the new drilling). The outcropping of the Roper Group sandstone and siltstone (Pr) on the eastern part of the cross section influences the position of Elsey Creek to the west of RN036898. The groundwater level in the CLA in this part of the regional groundwater discharge zone is very shallow (<5 mBGL). Figure 6-7 Hydrogeological cross-section 1 traversing from west around the Birdum Creek fault to east in the Daly Basin across the regional groundwater discharge zone Cross-section location is shown in Figure 6-6. Section 2 Section 2 is presented in Figure 6-8. It traverses the Cambrian limestone (Cmt) from the eastern edge of the Jinduckin Formation (Coij – Siltstone and limestone), through the Mataranka Springs Complex and terminates at the eastern margin of the Cambrian limestone. Like Section 1, Section 2 shows the structural influence of the Roper Group Sandstone and siltstone (Pr) which outcrops mid-section, as well as at the eastern margin of the CLA. These structural highs influence the location of the lower reach of Elsey Creek, which joins the Roper River between the two outcropping parts of the Roper Group units. At a local scale, the outcrop of the Roper Group also influences groundwater flow and discharge by subtly guiding part of the flow towards the Mataranka Springs Complex and the other part towards Elsey Creek and its junction with the Roper River. As for the groundwater levels in Section 1, groundwater levels in Section 2 indicate a downward gradient from the CLA to the localised fractured and weathered rock aquifers hosted in the underlying APV and Roper Group aquifers. A diagram of a section of a variety of layers Description automatically generated with medium confidence Figure 6-8 Hydrogeological cross-section 2 traversing from west around the edge of the Jinduckin Formation to the east/south-east in the Daly Basin across the regional groundwater discharge zone, the Roper River and Elsey Creek Cross-section location is shown in Figure 6-6. Section 3 Section 3 is presented in Figure 6-9. It traverses the Cambrian limestone (Cmt) from just north of Rainbow Spring at RN035792, across the Roper River and near the edge of the Roper Group outcrop/subcrop to the southern side of Elsey Creek. The main feature of this section is the prominence of the outcropping Cambrian limestone and the influence of the topography of the underlying basalt (Cla) of the APV to the thinning and outcrop of the CLA around the Roper River. The CLA is only tens of metres thick in the vicinity of the Roper River from a north to south orientation. A diagram of layers of the earth Description automatically generated Figure 6-9 Hydrogeological cross-section 3 traversing from north around the Roper River and Rainbow Spring to south in the Daly Basin across Elsey Creek Cross-sections location is shown in Figure 6-6. Dook Creek Aquifer Hydrogeological data for the DCA is sparse, though several drilling investigations have been carried out for community water supplies at Bulman and Weemol as well as numerous surrounding outstations (Momob, Bodeidei Camp, Morbon/Blue Water). Using stratigraphic and groundwaterlevel data, a north-west to south-east cross-section traversing these communities was generated with the location shown in Figure 6-10. A map of a large area Description automatically generated Figure 6-10 Location of the north-west to south-east hydrogeological cross-sections traversing the region around Bulman, Weemol and surrounding outstations Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008). Geological faults data source: Department of Industry, Tourism and Trade (2010) Spring data source: Department of Environment Parks and Water Security (2014c) Section A–B The north-west to south-east (A–B) cross-section traversing the communities and outstations around Bulman is presented in Figure 6-11. The cross-section highlights part of the horizontal and vertical framework that underpins the hydrogeological conceptual model for groundwater flow in the DCA. The section highlights the extensive exposure and unconfined nature of the south-eastdipping DCA outcrop at this location, extending about 42 km from the eastern edge of the Bone Creek Sandstone to the western edge of the Limmen Sandstone. The DCA outcrop is intruded in places by igneous dykes of the Derim Derim Dolerite, which can sometimes truncate shallow flow paths. Depending on the nature of the dyke (i.e. spatial extent and degree of weathering), this can lead to the formation of contact springs, resulting in discrete points of localised discharge from shallow parts of the DCA Figure 6-11. In other places, discrete karstic springs occur in topographic depressions where the watertable intersects the land surface. In addition, the DCA can also be interconnected with parts of the overlying fractured and weathered rock aquifers of the Limmen Sandstone of the Roper Group where it is thin (<50 m) and extensively fractured or weathered. The south-east-dipping nature of the DCA and existing drilling and groundwater-level observations suggest that the DCA has a reasonably thick saturated thickness in places (>150 m). However, further hydrogeological information from drilling investigations is required to characterise spatial changes in the total thickness of the Dook Creek Formation as well as the saturated thickness of the DCA. Figure 6-11 North-west to south-east cross-section traversing the Dook Creek Formation See Figure 6-10 for the spatial location of the cross-section. 6.1.3 Depth to top of key hydrogeological units Cambrian Limestone Aquifer Figure 6-12 shows the spatial interpolated depth to the top of CLA below the natural surface. Over 2600 groundwater bores, mineral exploration holes or stratigraphic wells were used in the Leapfrog hydrostratigraphic model that forms the geometry of the layers in the DR2 FEFLOW groundwater model. Only a partial extent of the interpolated surface for the CLA has been shown beyond the Roper catchment. This output represents the general depth required to drill and intersect the CLA, noting that at a local scale the aquifer can be interconnected with the overlying Cambrian siltstone (Jinduckin Formation and Anthony Lagoon Formation) where it is semiconfined or confined (mostly outside the Roper catchment) or the overlying surficial veneer of Cretaceous rocks and sediments where it is unconfined. The CLA beneath the Roper catchment is generally flat and only dips subtly towards the southern boundary of the surface water catchment. Depth to the top of the CLA, while spatially variable, occurs at less than 200 mBGL across its extent in the south to south-west of the catchment (Figure 6-12). The top of the aquifer is generally shallow (<50 mBGL) along the north and north-eastern margins of the aquifer in the Daly Basin (see Section 2.3.1 and Figure 2-5 for information on the spatial occurrence and extent of the geological basins). The depth to the top of the CLA then generally increases subtly in a southerly and south-easterly direction away from the northern aquifer boundary beneath the catchment. Depths increase initially to about 50 to 100 mBGL in the subsurface around Larrimah before increasing to between 100 and 200 mBGL toward the southeast, south and south-western catchment boundaries. The most abrupt increase in depth to the top of the aquifer away from the northern aquifer boundary is immediately west of Mataranka where the aquifer is confined by a small portion of the overlying Cambrian siltstone (see Section 2.4 and Figure 2-11 for the spatial occurrence of hydrogeological units). Here the depth to the top of the aquifer is between 200 and 250 mBGL (Figure 6-12). The greatest depth to the top of the aquifer occurs in the Daly Basin west of the Roper catchment boundary where the aquifer is overlain sequentially by the Jinduckin Formation, Oolloo Dolostone and Florina Formation. Figure 6-12 Interpolated spatial map of the depth to the top of the Cambrian Limestone Aquifer below the land surface Only a partial spatial extent of the Cambrian Limestone Aquifer is shown beyond the Roper catchment boundary. Depths are in metres below the ground level (mBGL). Stratigraphic data points represents bores with stratigraphic data providing information about changes in geology with depth. Aquifer extent data source: Knapton (Knapton, 2020) Dook Creek Aquifer Figure 6-13 shows the spatial interpolated depth to the top of DCA below the natural surface. Over 50 groundwater bores, mineral exploration holes or stratigraphic wells were used in the Leapfrog hydrostratigraphic model that forms the geometry of the layers in the DC2 FEFLOW groundwater model. The dip in the DCA is derived from a combination of the digital outcrop geology data (Department of Industry Tourism and Trade, 2014) and data on the intersection of the DCA in a deep stratigraphic well to the south-east. The extent of the interpolated surface for the DCA is shown beyond the Roper catchment because the surface water catchment is not the groundwater flow divide. The extent of the aquifer is that used for the DC2 FEFLOW groundwater model and represents almost the entire unconfined portion of the DCA with the exception of a small outcrop to the south-west around Barunga and Beswick. This output represents the general depth required to drill and intersect the DCA, noting that at a local scale the aquifer can be interconnected in places with the overlying Derim Derim Dolerite, Limmen Sandstone, and Cretaceous rocks and sediments where it is unconfined. Depth to the top of the DCA, while spatially variable, occurs generally at depths of less than 100 mBGL across the western unconfined portion of the aquifer, west of the Central Arnhem Road (Figure 6-13). However, information is sparse. Depth to the top of the DCA east of the Central Arnhem Road increases from about 100 mBGL to 500 mBGL far beneath the confluence of the Mainoru and Wilton rivers. Beneath the lower reaches of the Wilton River, the depth to the top of the DCA is greater than 1000 mBGL (Figure 6-13). Figure 6-13 Interpolated spatial map of the depth to the top of the Dook Creek Aquifer Depths are in metres below the ground surface. Stratigraphic data points represents bores with stratigraphic data providing information about changes in geology with depth. Aquifer extent data source : Knapton (2009c) 6.2 Groundwater recharge and flow 6.2.1 Potentiometric surface Cambrian Limestone Aquifer Mapping the potentiometric surface of the CLA (Figure 6-14) shows groundwater levels are highest in the south, extending above 170 mAHD in some parts to the south-west and south-east of the catchment boundary. It is important to note that this southern boundary does not represent the boundary of the entire CLA, which continues to extend in a southerly direction across parts of the Wiso and Georgina basins. Groundwater levels decline gradually from where they are highest in the south and south-west towards the north and north-west where they are lowest. This gradient in groundwater levels, represents the major directions of regional groundwater flow (i.e. the Georgina and Flora flow paths) which are influenced by the presence of the Cambrian basalt of the APV, a structural high, where the CLA is unsaturated (Figure 6-14). The areas of lowest reduced groundwater levels (~75 mAHD) are in the north-west around the lower reaches of the Flora River and upper reaches of the Katherine River outside of the Roper catchment. In the north around Mataranka, reduced groundwater levels are approximately 120 mAHD near the upper Roper River and Mataranka Springs Complex. It is important to note that the surface in places can represent more subtle hydraulic gradients than those previously presented in other investigations (Bruwer and Tickell, 2015; ELA, 2022; Knapton, 2020), because groundwater levels from the overlying and hydraulically connected Cambrian siltstone (i.e. Jinduckin Formation) and Cretaceous strata were excluded. For example, there is the subtle decline in groundwater levels from approximately 140 mAHD on the western catchment boundary to less than 120 mAHD east at the upper Roper River and Mataranka Spring Complex. This further emphasises that the regional groundwater discharge zone in the catchment receives flow originating from the west, south-west and south-east of the CLA. A map of a mountain Description automatically generated Figure 6-14 Potentiometric (reduced standing water level, RSWL) surface for the Cambrian Limestone Aquifer in the Roper catchment, including an extended buffer region of up to 100 km across the aquifer outside the southern catchment boundary Bore locations with groundwater-level data used to generate the surface are shown as red dots. Yellow lines represent the location of two regional piezometric cross-sections: A–A′ through the northern Wiso Basin to southern Daly Basin and B–B′ through the northern Georgina Basin to southern Daly Basin. Purple polygon of the Cambrian basalt is where the Cambrian Limestone Aquifer is unsaturated. Black arrows are the inferred regional and intermediate-scale groundwater flow directions. 6.2.2 Piezometric cross-section Cambrian Limestone Aquifer The decline in reduced groundwater levels from south to north is also evident in two piezometric cross-sections from the potentiometric surface (Figure 6-15). The potentiometric surface is relatively flat in the southern parts of the region, reflecting the confined or partially confined aquifer conditions resulting from thick overlying strata in the Georgina Basin (Anthony Lagoon Formation and surficial Cretaceous cover). Groundwater levels on the cross-sections decline as they approach the discharge areas in the northern part of the aquifer where there appears to be some correlation with the digital elevation model, unlike in the southern parts of the aquifer. Figure 6-15 Piezometric cross-sections through the potentiometric (reduced standing water level, RSWL) surface of the Cambrian Limestone Aquifer Transect locations are shown in Figure 6-14. DEM = digital elevation model. 6.2.3 Depth to groundwater Cambrian Limestone Aquifer A depth to groundwater surface was created using the same groundwater-level input dataset (i.e. standing water level) as for the potentiometric surface (see Figure 6-14) but without use of the digital elevation model (Figure 6-16). Groundwater levels are shallower (closer to the surface) in the north in the vicinity of the known groundwater discharge areas around Katherine and Mataranka. In these areas, groundwater levels are less than 5 m deep in some places around Mataranka and less than 10 m deep in some places around Katherine. The shallow groundwater levels in these areas are important for supporting a range of groundwater-dependent ecosystems including, but not limited to, groundwater-fed streams, springs and phreatophytic vegetation. Much deeper groundwater levels, up to and greater than 130 m below the land surface, are observed in southern parts of the aquifer coincident with the southern catchment boundary. Deep groundwater levels (>120 m) below the land surface also occur south of Katherine where the CLA is confined by the overlying Cambrian siltstone of the Jinduckin Formation. Figure 6-16 Interpolated depth to groundwater (standing water level, SWL) surface for the Cambrian Limestone Aquifer in the Roper catchment, including an extended buffer region up to 100 km around the catchment Bore locations with groundwater-level data in mBGL used to generate the surface are shown as red dots. Purple polygon of the Cambrian basalt is where the Cambrian Limestone Aquifer is unsaturated. Dook Creek Aquifer Depth to groundwater data are sparse across the DCA compared to across the CLA. However, an export of the calibrated depth to groundwater surface from the DC2 DCA groundwater model was used in ArcGIS to map spatial changes in depth to groundwater across the DCA (Figure 6-17). The modelled depth to groundwater indicates that groundwater depths in the western part of the aquifer (west of Central Arnhem Road) where the aquifer is unconfined, range from 10 to 110 m below ground surface. Groundwater is shallowest (i.e. <10 mBGL) in the vicinity of groundwater discharge zones around the lower reaches of Flying Fox Creek and the Mainoru and Wilton rivers. In contrast, in the eastern part of the aquifer (east of the Central Arnhem Road) where the aquifer transitions from unconfined to confined, groundwater depths range from 80 m below ground surface to 150 m above ground surface (Figure 6-17). This eastern part of the aquifer is deep (i.e. >500 mBGL, see Figure 6-13) and confined by the overlying Roper Group, and groundwater is modelled to be under natural pressure and in parts is highly artesian (Figure 6-17). Figure 6-17 Modelled depth to groundwater (standing water level, SWL) in the Dook Creek Aquifer relative to the land surface Mapped spatial extent of most of the DCA both within and beyond the Roper catchment boundary. Depth is in metres below ground level (mBGL). Positive values are depth below the land surface, representing sub-artesian conditions; negative values are depths above the land surface, representing artesian conditions. Aquifer extent data source: Knapton (2009c) 6.2.4 Field water sampling campaigns Cambrian Limestone Aquifer Groundwater samples were collected at 24 groundwater bores in two field trips during 2022 and 2023. These bores were constructed in aquifers hosted in five different hydrogeological units: (i) Tindall Limestone (17 bores), with one site being sampled twice for comparison, (ii) Gum Ridge Formation (3), (iii) Antrim Plateau Volcanics (2), (iv) Cretaceous Sandstone (1), and (v) Abner Sandstone (1). The locations of the 24 bores are shown in Figure 6-18. Please see Appendix A.2 for bore coordinates, groundwater levels and some construction details, which are summarised in Apx Table A.2-1. In addition, field parameters collected during sampling are summarised in Apx Table A.2-2 and analytical data from the various chemistry and environmental tracer laboratories are summarised in Apx Table A.2-3 to Apx Table A.2-5. Surface water samples were also collected and interpreted in conjunction with groundwater samples from the CLA and adjacent aquifers. Surface water samples were also collected from 15 locations: 2 springs and 13 stream sites. Surface water results are briefly discussed in the following sections of the report, but for more detail on groundwater–surface water interactions see Section 6.2.14. A map of water with black and blue lines Description automatically generated with medium confidence Figure 6-18 Spatial distribution of groundwater sampling sites across the Cambrian Limestone Aquifer and adjacent aquifers between Daly Waters and Mataranka To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left map inset with red polygon indicates the location and geographic extent of the map figure within the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008). Geological faults data source: Department of Industry, Tourism and Trade (2010) Dook Creek Aquifer Groundwater samples were collected at nine sites during August 2022, comprising bores constructed in aquifers hosted in two different hydrogeological units: (i) the Dook Creek Formation (6), and (ii) the Mountain Valley Limestone (1). Samples were also collected from two key springs. The locations of all water sampling sites are shown in Figure 6-19. Please see Appendix for more details. Apx Table A.2-1 summarises bore coordinates, groundwater levels and construction details. In addition, Apx Table A.2-2 summarises the field parameters collected during sampling and monitoring. A map of a river Description automatically generated Figure 6-19 Spatial distribution of groundwater sampling sites across the Dook Creek Formation and adjacent aquifers between Flying Fox Creek and the Goyder River To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left map inset with red polygon indicates the location and geographic extent of the map figure within the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008) 6.2.5 Chemistry Cambrian Limestone Aquifer Groundwater samples from the CLA (Tindall Limestone and Gum Ridge Formation) were either of a Ca–HCO3 or Na–HCO3 ionic composition, reflecting the chemical composition of the limestone and dolomite of both hydrogeological units (see Figure 6-20, Figure 6-21 and Table 6-2). Groundwater from the CLA with a Ca–HCO3 composition mainly occurred along or nearby the eastern and northern margins of the CLA where the aquifer is recharged either directly in outcrop areas or in subcrop areas beneath a thin (<30 m) veneer of Cretaceous rocks and sediments (Figure 6-20). Bitter Spring and Rainbow Spring also had a Ca–HCO3 composition reflective of the CLA as their 180 | Hydrogeological assessment of the Roper catchment primary water source. Groundwater from the CLA with a Na–HCO3 composition mainly occurred in central parts of the aquifer where recharge is lower and throughflow occurs beneath a thicker (>30 m) veneer of Cretaceous rocks and sediments, or the central to eastern portions of the aquifer outcrop in the northern regional groundwater discharge zone nearby Salt Creek (Figure 6-21). Groundwater from localised aquifers hosted in the overlying Cretaceous rocks and sediments were either of a Mg–HCO3 or Na–HCO3 composition, based mostly on historical data as only one bore was sampled (Figure 6-21 and Table 6-2). There is evidence of vertical leakage from aquifers hosted in the Cretaceous rocks and sediments, particularly around the eastern and northern margins of the CLA where the veneer of overlying Cretaceous cover is thin. Figure 6-20 Spatial distribution of water quality types for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left map inset with red polygon indicates the location and geographic extent of the map figure within the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008). Geological faults data source: Department of Industry, Tourism and Trade (2010) Groundwater from localised aquifers hosted in the APV had a variable composition ranging between Ca–HCO3, Na–HCO3, Ca–Cl, and Na–Cl reflecting the heterogeneous nature of the fractured and weathered rock aquifers Figure 6-21 and Table 6-2. Some samples from the APV exhibited a Ca–HCO3 or Ca–Cl composition and had a lower salinity (i.e. a TDS of <700 mg/L) than other APV samples, which indicated they had been recharged via vertical leakage from the overlying CLA. Only one sample was collected from a localised aquifer hosted in the Abner Sandstone of the Roper Group. It had a Na–Cl composition, which is different from the composition of aquifers of the CLA. Historical chemistry samples from aquifers hosted in the Bukalara Sandstone and APV were evaluated in the vicinity of G9055086 spring in Hot Springs Valley to the east of the CLA (Figure 6-18). Aquifers hosted in the Bukalara Sandstone had a varied composition reflecting their heterogenous nature. Compositions ranged from Ca–HCO3 and Na–HCO3 to Ca–SO4 and Mg–SO4 in unconfined parts of the aquifer. In confined part of the sandstone aquifers, the composition was Ca–Cl (Figure 6-21 and Table 6-2). G9055086 spring had a Ca–Cl composition, which is similar to some of the localised aquifers in the APV. However, G9055086 spring had a much lower salinity (<500 mg/L) and a much higher temperature (~65 °C) than that of aquifers hosted in the APV and Bukalara Sandstone (with salinity >1000 mg/L TDS and temperature <40 °C). The Ca–Cl composition of G9055086 spring may reflect its long residence time travelling through the APV aquitard, but other physical and chemical parameters suggest a deeper groundwater source from the Proterozoic hydrogeological units of the McArthur Basin. A diagram of a diagram of a triangle Description automatically generated with medium confidence Figure 6-21 Piper diagram showing major ion composition for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units Samples are symbolised by aquifers or springs sampled. Filled shapes are samples collected in this assessment. Hollow shapes are historical data available from Department of Environment Parks and Water Security (2014a). Plots of the major cation and anion concentrations relative to chloride concentration (Figure 6-22) have been used to investigate hydrochemical trends in groundwater. All groundwater samples from the CLA (Tindall Limestone and Gum Ridge Formation) are enriched in Ca, Mg and HCO3 relative to rainfall of marine origin, and most samples are also slightly enriched in Na (Figure 6-22). This trend reflects the carbonate-rich minerals of the limestones and dolomites that comprise the transmissive parts of the Tindall Limestone and Gum Ridge Formation. The hydrochemical trend from recharge areas to throughflow and discharge areas is evident from the exponential shift in Ca, Mg and HCO3 concentrations relative to chloride. These observations can be made because Cl/Br ratios provide no evidence of halite dissolution (Figure 6-22e), so Cl can be considered to be only sourced from rainfall. Although not shown, all groundwater samples were below saturation with respect to halite which was determined using PHREEQC. What is notable from the Cl/Br plot is the different sources of recharge and hydrochemical evolution and mixing with flow in different parts of the CLA shown by the two linear lines approaching the seawater dilution line along different slopes (Figure 6-22e). One line evolves from a Ca–HCO3 low-Cl water type along the eastern margins to a Ca–HCO3 with increased Cl water type with throughflow and discharge to the north. The other line evolves from a Na–HCO3 low-Cl water type to the west, where recharge occurs through thicker Cretaceous cover or leakage from the Jinduckin Formation, which evolves to a Na–HCO3 with increased Cl water type with throughflow and discharge to the east/north-east. 0.0 0.5 1.0 1.5 2.0 0.0 5.0 10.0 15.0 Na/Cl (meq/L) Cl (meq/L) Tindall Limestone Antrim Plateau Volcanics Gumridge Formation Cretaceous Sandstone Springs Abner Sandstone (a) 0.0 2.0 4.0 6.0 8.0 10.0 0.0 5.0 10.0 15.0 Ca/Cl (meq/L) Cl (meq/L) Tindall Limestone Antrim Plateau Volcanics Gumridge Formation Cretaceous Sandstone Springs Abner Sandstone (b) 0.0 2.5 5.0 7.5 10.0 12.5 0.0 5.0 10.0 15.0 HCO3/Cl (meq/L) Cl (meq/L) Tindall Limestone Antrim Plateau Volcanics Gumridge Formation Cretaceous Sandstone Springs Abner Sandstone (c) 0.0 1.0 2.0 3.0 4.0 5.0 6.0 0.0 5.0 10.0 15.0 Mg/Cl (meq/L) Cl (meq/L) Tindall Limestone Antrim Plateau Volcanics Gumridge Formation Cretaceous Sandstone Springs Abner Sandstone (d) 0 150 300 450 600 750 0.0 5.0 10.0 15.0 Cl/Br (meq/L) Cl (meq/L) Tindall Limestone Antrim Plateau Volcanics Gumridge Formation Cretaceous Sandstone Springs Abner Sandstone (e) 0 50 100 150 200 250 0.0 5.0 10.0 15.0 Na/HCO3 (meq/L) Cl (meq/L) Tindall Limestone Antrim Plateau Volcanics Gumridge Formation Cretaceous Sandstone Springs Abner Sandstone (f) 0 10 20 30 40 50 60 0.0 5.0 10.0 15.0 (Ca+Mg)/HCO3 (meq/L) Cl (meq/L) Tindall Limestone Antrim Plateau Volcanics Gumridge Formation Cretaceous Sandstone Springs Abner Sandstone Seawater dilution Calcite dissolution (g) 0 10 20 30 40 50 0.0 5.0 10.0 15.0 Ca/SO4 (meq/L) Cl (meq/L) Tindall Limestone Antrim Plateau Volcanics Gumridge Formation Cretaceous Sandstone Springs Abner Sandstone Seawater dilution Gypsum dissolution (h) Figure 6-22 Major ion ratio plots for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifer hosted in adjacent hydrogeological units Samples are symbolised by aquifers or springs sampled. Na/HCO3 ratios for all groundwater samples are significantly lower than expected for rainfall of marine origin. Therefore, even though Na has been added to groundwater through processes such as cation exchange, the addition of Ca and HCO3 is far more pronounced. These trends point towards water–rock interactions and the primary source of these ions being from carbonate dissolution. PHREEQC modelling indicates that all groundwater samples from the CLA have a significant portion of their dissolved inorganic carbon in the form of aqueous HCO3 compared to aqueous CO2 (Table 6-2). All groundwater samples from the CLA are either close to saturation or supersaturated with respect to calcite and dolomite. Gypsum dissolution is unlikely to be a source of Ca, as most groundwater samples exhibit Ca/SO4 ratios much higher than unity (Figure 6-22). Table 6-2 Carbonate speciation and mineral saturation indices from geochemical modelling for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units BORE REGISTERED NUMBER HYDROGEOLOGICAL UNIT WATER TYPE AQUEOUS CO2 (mM) AQUEOUS HCO3 (mM) SI CALCITE SI DOLOMITE SI GYPSUM RN031167 Tindall Limestone Ca–HCO3 1.26 8.85 0.59 1.16 −1.38 RN033095 Tindall Limestone Na–HCO3 1.13 8.02 0.50 1.01 −1.40 RN034324 Tindall Limestone Ca–HCO3 2.05 9.09 0.36 0.73 −1.38 RN036503 Tindall Limestone Ca–HCO3 1.35 8.18 0.47 0.93 −1.47 RN029537 Tindall Limestone Ca–HCO3 1.58 9.80 0.49 1.05 −1.52 RN040211 Tindall Limestone Na–HCO3 1.61 8.88 0.44 0.92 −1.43 RN040219 Tindall Limestone Na–HCO3 1.45 8.65 0.45 0.88 −1.40 RN029706 Tindall Limestone Ca–HCO3 1.42 8.74 0.50 0.91 −1.47 RN039137 Tindall Limestone Ca–HCO3 1.22 8.53 0.53 1.01 −1.43 RN033297 Tindall Limestone Ca–HCO3 0.82 6.53 0.52 0.73 −1.98 RN039136 Tindall Limestone Ca–HCO3 1.05 8.22 0.61 1.11 −1.44 RN043049 Tindall Limestone Na–HCO3 3.42 9.49 0.06 0.22 −1.40 RN034031 Tindall Limestone Na–HCO3 3.54 9.76 0.02 0.24 −1.38 RN034230 Tindall Limestone Ca–HCO3 1.99 6.79 −0.01 −0.15 −2.24 RN043046 Tindall Limestone Ca–HCO3 1.97 6.70 −0.03 −0.17 −2.31 RN032164 Tindall Limestone Na–HCO3 4.28 9.10 −0.02 −0.08 −1.46 RN035464 Tindall Limestone Ca–HCO3 3.79 8.14 −0.04 −0.26 −1.77 RN041491 Gum Ridge Formation Na–HCO3 1.81 8.92 0.41 0.82 −1.41 RN041203 Gum Ridge Formation Na–HCO3 1.01 8.55 0.56 1.14 −1.45 RN038185 Gum Ridge Formation Ca–HCO3 0.69 7.13 0.67 1.28 −1.52 RN043045 Antrim Plateau Volcanics Ca–Cl 0.28 2.59 0.10 −0.20 −1.33 RN043047 Antrim Plateau Volcanics Ca–Cl 0.01 1.59 0.53 0.70 −1.28 RN036507 Cretaceous Sandstone Mg–HCO3 3.08 5.87 −0.59 −0.95 −3.33 RN024602 Abner Sandstone Na–Cl 60.53 0.21 −6.50 −13.13 −5.67 Rainbow Spring Tindall Limestone Ca–HCO3 0.67 5.72 0.39 0.70 −2.11 Bitter Spring Tindall Limestone Ca–HCO3 1.03 6.49 0.34 0.63 −1.75 G9055086 spring Unknown Ca–Cl 0.66 1.45 −0.58 −1.46 −3.27 RN = registered number for a groundwater bore; SI = saturation index Dook Creek Aquifer Groundwater samples from the DCA were generally of a Ca–HCO3 ionic composition, reflecting the chemical composition of the dolomite aquifer (Figure 6-23 and Figure 6-24). However, two samples from the DCA were of a Mg–HCO3 and Na–Cl ionic composition. The former reflecting the siltstone composition of the DCA at that location, and the latter reflecting its shallow location in the discharge zone near the Goyder River. Samples, including historical data from the DCA, tend to form a cluster and show little hydrochemical evolution from recharge to discharge areas, reflecting the more localised to intermediate-scale shallow flow in the DCA compared to that of the CLA. Both Weemol and Wurrkal Spring had a Ca–HCO3 composition reflecting the DCA as their water source. Figure 6-23 Spatial distribution of water quality types for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left map inset with red polygon indicates the location and geographic extent of the map figure within the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008). Geological faults data source: Department of Industry, Tourism and Trade (2010) Historical chemistry samples were collated and evaluated from aquifers hosted in the Derim Derim Dolerite, which intrudes the DCA in places, and the Limmen Sandstone, which overlies the DCA east of the Central Arnhem Road (Figure 6-24). Historical samples from the Derim Derim Dolerite had a Mg–HCO3 ionic composition as did some groundwater samples from the DCA, indicating their interconnectivity where the DCA has been discretely intruded (see Figure 6-11 for dolerite intrusions). Historical samples from aquifers hosted in the Limmen Sandstone generally had a Mg– HCO3 ionic composition similar to parts of the DCA, indicating their interconnectivity where both aquifer outcrops are adjacent or the overlying Limmen Sandstone is thin (<30 m) and fractured. One historical sample from Mount Catt Spring (hollow hourglass) had a Mg–HCO3 ionic composition which is likely sourced from aquifer hosted in the Limmen Sandstone given its position in the aquifer outcrop (see Figure 6-11). One groundwater sample was collected from an aquifer hosted in the Mountain Valley Limestone which in minor places overlies the DCA. It had a Mg–HCO3 ionic composition. Figure 6-24 Piper diagram showing major ion composition for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units Samples are symbolised by aquifers or springs sampled. Filled shapes are samples collected in this assessment. Hollow shapes are historical data available from Department of Environment Parks and Water Security (2014a). As for the plots generated for the CLA, the major cation and anion concentrations for the DCA have been plotted relative to chloride to investigate hydrochemical trends in groundwater (Figure 6-25). All groundwater samples from the DCA are enriched in Ca, Mg and HCO3 relative to rainfall of marine origin, and most samples are slightly enriched in Na (Figure 6-25). This trend again reflects the carbonate-rich minerals of the dolomites that comprise the transmissive parts of the DCA. The addition of Ca, Mg and HCO3 is far more pronounced than for other major ions as evident from the vertical shift above the seawater dilution line relative to chloride. Observations of the Cl/Br ratios do not provide any evidence of halite dissolution. Although not shown, all groundwater samples were below saturation with respect to halite, which was determined using PHREEQC. Na/HCO3 for groundwater samples are significantly lower than expected for rainfall of marine origin. Therefore, even though Na has been added to groundwater through processes such as cation exchange, the addition of Ca and HCO3 is far more pronounced. These trends point towards water–rock interactions and the primary source of these ions being from carbonate dissolution. PHREEQC modelling indicates that all groundwater samples from the DCA have a significant portion of their dissolved inorganic carbon in the form of aqueous HCO3 compared to aqueous CO2 (Table 6-3). Most groundwater samples from the DCA are either close to saturation or are supersaturated with respect to calcite and dolomite except where bores intersect siltstone parts of the aquifer. Gypsum dissolution is unlikely to be a source of Ca as most groundwater samples exhibit Ca/SO4 ratios much higher than unity (Figure 6-25). 0.0 1.0 2.0 3.0 4.0 5.0 0.0 0.5 1.0 1.5 2.0 Na/Cl (meq/L) Cl (meq/L) Dook Creek Formation Mountain Valley Limestone Springs Seawater dilution (a) 0.0 10.0 20.0 30.0 40.0 50.0 0.0 0.5 1.0 1.5 2.0 Ca/Cl (meq/L) Cl (meq/L) Dook Creek Formation Mountain Valley Limestone Springs Seawater dilution (b) 0.0 10.0 20.0 30.0 40.0 50.0 60.0 0.0 0.5 1.0 1.5 2.0 HCO3/Cl (meq/L) Cl (meq/L) Dook Creek Formation Mountain Valley Limestone Springs Seawater dilution (c) 0.0 10.0 20.0 30.0 40.0 0.0 0.5 1.0 1.5 2.0 Mg/Cl (meq/L) Cl (meq/L) Dook Creek Formation Mountain Valley Limestone Springs Seawater dilution (d) 0 150 300 450 600 750 0.0 0.5 1.0 1.5 2.0 Cl/Br (meq/L) Cl (meq/L) Dook Creek Formation Mountain Valley Limestone Springs Seawater dilution (e) 0 50 100 150 200 250 0.0 0.5 1.0 1.5 2.0 Na/HCO3 (meq/L) Cl (meq/L) Dook Creek Formation Mountain Valley Limestone Springs Seawater dilution (f) 0 10 20 30 40 50 60 0.0 0.5 1.0 1.5 2.0 (Ca+Mg)/HCO3 (meq/L) Cl (meq/L) Dook Creek Formation Mountain Valley Limestone Springs Seawater dilution Calcite dissolution (g) 0 100 200 300 400 0.0 0.5 1.0 1.5 2.0 Ca/SO4 (meq/L) Cl (meq/L) Dook Creek Formation Mountain Valley Limestone Springs Seawater dilution Gypsum dissolution (h) Figure 6-25 Major ion ratio plots for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and adjacent hydrogeological units Samples are symbolised by aquifers or springs sampled. Table 6-3 Carbonate speciation and mineral saturation indices from geochemical modelling for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units BORE REGISTERED NUMBER HYDROGEOLOGICAL UNIT WATER TYPE AQUEOUS CO2 (mM) AQUEOUS HCO3 (mM) SI CALCITE SI DOLOMITE SI GYPSUM RN036303 Mountain Valley Limestone Mg–HCO3 1.41 7.77 −0.16 0.08 −3.39 RN031981 Dook Creek Formation Ca–HCO3 0.87 6.95 0.22 0.53 −2.89 RN028226 Dook Creek Formation Ca–HCO3 1.75 2.94 −1.18 −2.35 −4.24 RN028228 Dook Creek Formation Mg–HCO3 0.90 4.29 −0.29 −0.42 −4.06 RN031970 Dook Creek Formation Na–Cl 0.33 0.20 −4.00 −7.38 −5.38 RN028224 Dook Creek Formation Ca–HCO3 1.50 2.18 −1.49 −2.95 −4.35 RN031983 Dook Creek Formation Mg–HCO3 1.46 10.5 0.49 1.23 −3.13 Weemol Spring Dook Creek Formation Ca–HCO3 1.24 6.72 0.18 0.42 −3.74 Wurrkal Spring Dook Creek Formation Ca–HCO3 0.59 6.26 0.4 0.86 −3.92 RN = registered number for a groundwater bore; SI = saturation index 6.2.6 Stable hydrogen and oxygen isotopes Cambrian Limestone Aquifer The stable hydrogen and isotopic compositions of groundwater samples from the CLA (Tindall Limestone and Gum Ridge Formation) and aquifers hosted in adjacent hydrogeological units are shown in (Figure 6-26 and Figure 6-27). For comparison, the isotopic composition of rainfall near Darwin (the closest rainfall isotopic data available) spanning 1963 to 2010 (IAEA/WMO, 2023) and local meteoric water line (LMWL) for Darwin is also shown. The range in isotopic composition for all groundwater samples is from −8.9 to −7.6‰ for δ18O, and from −63 to −50‰ for δ2H. For the Tindall Limestone, δ18O ranges from −8.6 to −7.6‰ and δ2H ranges from −63 to −50‰. And for the Gum Ridge Formation, δ18O ranges from −8.6 to −7.7‰ and δ2H ranges from −61 to −53‰. The similarity of these ranges reflects: (i) their correlation with the most depleted wet-season rainfall being the source of recharge to the aquifer and (ii) their interconnectivity and mixing between aquifers in both hydrogeological units. LMWL – Darwin: δ2H = 7.82δ18O + 10.3 -70 -65 -60 -55 -50 -45 -40 -10.0 -9.5 -9.0 -8.5 -8.0 -7.5 -7.0 -6.5 -6.0 -5.5 -5.0 δ2H (‰VSMOW) δ18O (‰VSMOW) Antrim Plateau Volcanics Cretaceous Sandstone Abner Sandstone Tindall Limestone Gum Ridge Formation Springs LMWL – Darwin Darwin monthly rainfall Figure 6-26 Stable hydrogen and oxygen isotope composition for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units compared to rainfall Samples are symbolised by aquifers or springs sampled. LMWL = local meteoric water line. VSMOW = Vienna Standard Mean Ocean Water. Rainfall data source: IAEA/WMO (2023) Similar to the hydrochemical trends discussed in Section 6.2.5, there is an isotopic trend spatially across the CLA with evaporative enrichment in groundwater indicating different recharge mechanisms. The most enriched samples, which plot furthest to the right of the Darwin LMWL (Figure 6-26), are either from the central part of the aquifer where recharge has occurred through thick (>30 m) Cretaceous cover, or directly in parts of the northern aquifer outcrop where evapotranspiration of groundwater occurs from the shallow watertable (~5 mBGL) (see Figure 6-27). Most groundwater samples from along or near the eastern margin of the aquifer, where recharge occurs beneath a thin (<30 m) veneer of Cretaceous cover, plot closer to the Darwin LMWL. Rainbow and Bitter springs had a similar isotopic composition (~−8.3‰ for δ18O and −56.0‰ for δ2H) and plot close to the LMWL (see Figure 6-26 and Figure 6-27), though Bitter Spring is slightly more enriched for δ2H. The groundwater sample from the Abner Sandstone is isotopically depleted compared to samples from the CLA. Samples from the Antrim Plateau Volcanics (APV) and Cretaceous Sandstone fall within the same range of isotopic compositions as the CLA, reflecting their interconnectivity via vertical leakage from aquifers hosted in the Cretaceous cover through to the APV underlying the CLA. The isotopic composition for G9055086 spring in Hot Spring Valley is similar to the other springs. Figure 6-27 Spatial distribution of deuterium (δ2H) isotopic composition for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left map inset with red polygon indicates the location and geographic extent of the map figure within the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008). Geological faults data source: Department of Industry, Tourism and Trade (2010) Dook Creek Aquifer The stable hydrogen and isotopic compositions of groundwater samples from the DCA and aquifers hosted in the adjacent Mountain Valley Limestone are shown in Figure 6-28 and Figure 6-29. For comparison, the isotopic composition of rainfall near Darwin and the local meteoric water line for Darwin is also shown. The range in isotopic composition for all groundwater samples is from −8.4 to −3.6‰ for δ18O and from −53 to −30‰ for δ2H. For the DCA, δ18O ranges from −8.0 to −3.6‰ and δ2H ranges from −49 to −30‰ (Figure 6-28). For the aquifer hosted in the Mountain Valley Limestone, the isotopic composition is −8.4‰ for δ18O and −53‰ for δ2H. Samples from all aquifers plot around the LMWL and exhibit little enrichment from evaporation, suggesting groundwater recharge is rapid and highly localised in the aquifer outcrop. This is not surprising given the large area of aquifer outcrop and shallow watertable for the DCA and the fact the Mountain Valley Limestone overlies the DCA. LMWL – Darwin: δ2H = 7.82δ18O + 10.3 -70 -65 -60 -55 -50 -45 -40 -35 -30 -10.0 -9.0 -8.0 -7.0 -6.0 -5.0 δ2H (‰ VSMOW) δ18O (‰ VSMOW) Mountain Valley Limestone Dook Creek Formation Springs LMWL – Darwin Darwin rainfall Figure 6-28 Stable hydrogen and oxygen isotope composition for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and adjacent hydrogeological units compared to rainfall Samples are symbolised by aquifers or springs sampled. LMWL = local meteoric water line. VSMOW = Vienna Standard Mean Ocean Water. Rainfall data source: IAEA/WMO (2023) In comparison to the CLA, DCA groundwater exhibits a less depleted isotopic composition for rainfall, suggesting that while the DCA is in a higher-rainfall zone, smaller wet-season rainfall events generate net recharge to the shallow aquifer. The isotopic composition of spring waters correlates with localised discharge from different shallow parts of the aquifer (Figure 6-29). For example, Weemol Spring has a more depleted isotopic composition (−8.3‰ for δ18O and −54‰ for δ2H) similar to that of the Mountain Valley Limestone. However, no samples were collected from this aquifer in the near vicinity of the spring. The composition maybe also be reflected by the fact that groundwater from the DCA is connected with aquifers hosted in the Derim Derim Dolerite, which intrudes the DCA in close vicinity to the spring. This is consistent with the hydrogeological cross-section in Section 6.1.2 (Figure 6-11) and the hydrochemistry presented in Section 6.2.5 (Figure 6-24), suggesting the water source for the spring is the DCA. The isotopic composition of Wurrkal Spring is more enriched (−7.3‰ for δ18O and −47‰ for δ2H) than that of Weemol Spring but correlates to groundwater in the DCA near the Goyder River. Figure 6-29 Spatial distribution of deuterium (δ2H) isotopic composition for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left map inset with red polygon indicates the location and geographic extent of the map figure within the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008). Geological faults data source: Department of Industry, Tourism and Trade (2010) 6.2.7 Strontium isotopes Cambrian Limestone Aquifer The strontium (87Sr/86Sr) isotopic compositions of all groundwater samples collected from the CLA and aquifers hosted in adjacent hydrogeological units are presented in Figure 6-30. The range in isotopic composition for all groundwater samples is quite narrow, ranging from 0.712009 to 0.718965. This narrow range reflects groundwater that has acquired strontium either from rainfall of marine origin or weathering of carbonate minerals. Groundwater samples form two clusters in isotopic composition: one group ranging from 0.712009 to 0.713704 and the other ranging from 0.716017 to 0.718501 (Figure 6-30). These clusters reflect the different water–rock interactions for groundwater recharge and flow mechanisms across different spatial parts of the CLA. 0.710000 0.715000 0.720000 0.725000 0.730000 0.735000 0.740000 0.745000 1 10 100 1000 87Sr/86Sr 1/Sr (mg/L) Tindall Limestone Gum Ridge Formation Antrim Plateau Volcanics Cretaceous Sandstone Abner Sandstone Springs G9055086 Rainbow Spring Bitter Spring Figure 6-30 Strontium isotope composition relative to strontium concentration for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units Samples are symbolised by aquifers or springs sampled. Strontium concentration plotted on a logarithmic scale. The samples in the cluster with a lower composition (0.712009 to 0.713704) are either from the western part of the northern aquifer outcrop or along or near the eastern margin of the CLA where recharge occurs beneath a thin veneer (<30 m) of Cretaceous cover (Figure 6-31). The samples in the cluster with a slightly higher composition (0.716017 to 0.718501) are from either: (i) the central part of the CLA where recharge and flow occur beneath a thick (>30 m) veneer of Cretaceous cover and, in some cases, Cambrian siltstone (Jinduckin Formation), or (ii) the central to eastern part of the northern regional groundwater discharge zone near Salt Creek (Figure 6-31). Rainbow and Bitter springs have a strontium isotope composition similar to groundwater from the central part of the CLA. G9055086 spring has a vastly different strontium isotope composition more reflective of silicate weathering, indicting its water source is not from the CLA. Figure 6-31 Spatial distribution of strontium isotopic composition for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left map inset with red polygon indicates the location and geographic extent of the map figure within the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008). Geological faults data source: Department of Industry, Tourism and Trade (2010) Dook Creek Aquifer The strontium (87Sr/86Sr) isotopic composition of all groundwater samples collected from the DCA and aquifers hosted in adjacent hydrogeological units are presented in Figure 6-32 and Figure 6-33. The range in isotopic composition for all groundwater samples is broad, ranging from 0.720925 to 0.765986. The lower end of this range is more typical of strontium derived from either rainfall of marine origin or carbonate mineral weathering. The upper end of this range is more typical of strontium derived from silicate weathering. Groundwater samples with a lower strontium isotopic composition (0.720925 to 0.738322) correspond to samples from the more transmissive dolomite parts of the DCA or the aquifer hosted in the Mountain Valley Limestone. These samples also have a similar isotopic composition to both Weemol and Wurrkal Spring (Figure 6-32). 196 | Hydrogeological assessment of the Roper catchment Groundwater samples with a higher strontium isotopic composition (0.745805 to 0.765986) correspond to samples from siltstone parts of the DCA. This composition is more reflective of the weathering of silicates from the micas, quartz, feldspars and clay mineral composition of the siltstone. 0.710000 0.720000 0.730000 0.740000 0.750000 0.760000 0.770000 1 10 100 1000 87Sr/86Sr 1/Sr (mg/L) Mountain Valley Limestone Dook Creek Formation Springs Weemol Spring Wurrkal Spring Figure 6-32 Strontium isotope composition relative to strontium concentration for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and adjacent hydrogeological units Samples are symbolised by aquifers or springs sampled. Figure 6-33 Spatial distribution of the strontium isotopic composition for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left map inset with red polygon indicates the location and geographic extent of the map figure within the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008). Geological faults data source: Department of Industry, Tourism and Trade (2010) 6.2.8 Tritium Cambrian Limestone Aquifer Tritium concentrations in groundwater and spring samples collected from the CLA and aquifers hosted in adjacent hydrogeological units are summarised in Figure 6-34. Figure 6-34a, summarises 3H concentrations versus distance along the regional hydraulic gradient from the northern edge of the Anthony Lagoon Formation where the CLA becomes unconfined; recharge generally increases. Figure 6-34b, summarises 3H concentrations versus depth below the watertable. Tritium concentrations in groundwater ranged from 0.025 TU (the detection limit) to 0.63 TU, though more than 50% of the samples contained tritium below the detection limit, similar to that of historical samples. A diagram of a person's body Description automatically generated with medium confidence Figure 6-34 Tritium concentration for groundwater and spring samples from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units versus (a) distance from the edge of the Cambrian siltstone (Anthony Lagoon Formation – see Figure 6-18) and (b) depth below the watertable SS = Sandstone. Plot (b) includes predicted curves for 3H compositions assuming different groundwater flow conditions. PM = piston flow model; Vogel = exponential model. Samples are symbolised by aquifers or springs sampled. Filled shapes represent samples collected in this assessment. Hollow shapes represent historical sample results (Deslandes et al., 2019; Frery et al., 2022; Lamontagne et al., 2021; Tickell and Bruwer, 2018). Samples with concentrations of tritium greater than 0.1 TU were found either along or near the eastern margin of the aquifer, where recharge occurs either directly in the aquifer outcrop or beneath a thin veneer of Cretaceous cover, or where recharge occurs directly in the eastern part of the northern aquifer outcrop near Mataranka (Figure 6-35). However, there was significant spatial variability in the concentrations across the aquifer, reflecting both spatial and temporal variability in recharge processes across different parts of the aquifer sampled. Furthermore, this variability also reflects the mixing in the aquifer between groundwater that has recently been recharged (i.e. on a decadal scale) versus groundwater in throughflow and discharge areas with longer residence times that may be tritium-free (i.e. >100 years in the saturated zone). This is evident by the concentration in discharge water from the two key springs (Rainbow and Bitter springs) where 3H concentrations are 0.06 TU and below the detection limit, respectively (Figure 6-35). Tritium concentrations in aquifers hosted in both the Abner and Cretaceous sandstones had low but measurable concentrations of 3H (0.1 and 0.18 TU, respectively) reflecting decadal scale recharge directly in their aquifer outcrops (Figure 6-34a). Figure 6-34b summarises 3H concentrations in groundwater from all aquifers versus depth below the watertable. The samples with moderate to high concentrations of 3H generally sit amongst residence time distribution model curves with a vertical flow velocity ranging between 400 and 1400 mm/year. Samples with low concentrations of 3H and occur at shallow depths in the aquifer (<15 mBGL) or the springs, plot to the left of these model curves and occur in the regional groundwater discharge zone where there is significant mixing of recharge and discharge waters in the CLA. There is also one sample from the Gum Ridge Formation where recharge is very low. Figure 6-35 Spatial distribution of the tritium concentration for groundwater and spring samples from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left map inset with red polygon indicates the location and geographic extent of the map figure within the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008). Geological faults data source: Department of Industry, Tourism and Trade (2010) Dook Creek Aquifer Tritium concentrations in groundwater and spring samples collected from the DCA and aquifers hosted in the adjacent Mountain Valley Limestone are summarised in Figure 6-36 and Figure 6-37. Figure 6-37a, summarises 3H concentrations versus distance along the intermediate- to regionalscale hydraulic gradient from the south-western edge of the Dook Creek Formation outcrop where the DCA is unconfined. Figure 6-37b, summarises 3H concentrations versus depth below the watertable. Tritium concentrations in groundwater ranged from 0.026 TU (close to the detection limit) to 0.79 TU. Samples from the DCA with the highest concentrations of tritium (0.34, 0.48 and 0.79 TU) occur directly in the aquifer outcrop in areas with a relatively shallow watertable (<15 mBGL) or occur in a valley where both runoff and overbank flooding can accumulate (Figure 6-36). However, other parts of the DCA exhibit significant spatial variability in the concentrations across the aquifer, reflecting both spatial and temporal variability in recharge and flow across the parts of the aquifer sampled. This indicates that at least the shallow flow paths in the unconfined parts of the aquifer are of a local to intermediate scale. For example, along the upper reaches of the Wilton River where the Roper catchment is slightly more elevated, and rainfall has a greater tendency to run off, as reflected by the low 3H concentration (close to the detection limit) in groundwater (Figure 6-36). In addition, some parts of the aquifer are predominantly throughflow and discharge areas, and groundwater has a longer residence time (>100 years) and therefore can be tritium-free. This is evident by the concentration in discharge water from the two key springs (Weemol and Wurrkal springs), where 3H concentrations are 0.14 and below detection limit, respectively. Tritium concentrations in aquifers hosted in the Mountain Valley Limestone were low (0.028 TU) where the one sample was collected, and the watertable was 25 mBGL. A map of a river Description automatically generated Figure 6-36 Spatial distribution of the tritium concentration for groundwater and spring samples from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left map inset with red polygon indicates the location and geographic extent of the map figure within the Roper catchment. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008). Geological faults data source: Department of Industry, Tourism and Trade (2010) Figure 6-37b presents 3H concentrations in groundwater from all aquifers versus depth below the watertable. The samples with moderate to high concentrations of 3H generally sit among residence time distribution model curves with a vertical flow velocity ranging between 600 and 850 mm/year. Samples with low concentrations of 3H occur at shallow depths in the aquifer (<20 mBGL) and springs; they plot to the left of these model curves and occur in groundwater discharge zones where there is significant mixing of recharge and discharge waters in the CLA. Figure 6-37 Tritium concentrations for groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units versus (a) distance from the southern edge of the Dook Creek Formation – see Figure 6-19) and (b) depth below the watertable Plot (b) includes predicted curves for 3H compositions assuming different groundwater flow conditions. PM = piston flow model; Vogel = exponential model. Samples are symbolised by aquifers or springs sampled. Filled shapes represent samples collected in this assessment. Hollow shapes represent historical sample results (Deslandes et al., 2019; Frery et al., 2022; Lamontagne et al., 2021; Tickell and Bruwer, 2018) 6.2.9 Noble gases Cambrian Limestone Aquifer Concentrations of stable noble gases (Ne, Kr and Xe) from groundwater samples and spring samples collected from the CLA and aquifers hosted in adjacent hydrogeological units are summarised in Figure 6-38. Noble gases are presented as plots of one element against another so the blue line represents the solubility equilibrium at different temperatures and green lines starting at different solubility equilibrium temperatures demonstrate the addition of excess air. Noble gas concentrations indicate infiltration temperatures between 29 and 41 °C, which is consistent with mean wet-season temperatures across large areas of the CLA from the semi-arid to arid zone (Figure 6-38). Samples from the Gum Ridge Formation tend to have the highest temperatures, while samples from the Tindall Limestone exhibit a wide range, and samples from the sandstone aquifers exhibit lower temperatures (i.e. higher Xe concentrations). This is consistent with the climate gradient at the land surface across the area of the aquifers sampled. The heavier stable noble gas concentrations (Xe and Ne) indicate higher excess air values – where the points in the Xe–Ne plot from the blue solubility equilibrium line extend only to about 5 cc/kg along the green line of excess air, they extend in the Xe–Kr plot to values higher than 10 cc/kg. This means excess air is enriched in the heavier elements, which will be of importance for the interpretation of the anthropogenic gas tracers (see Section 6.2.10). Helium will be discussed in Section 6.2.12, since it is a good indicator of groundwater residence times as opposed to infiltration conditions. A diagram of different types of data Description automatically generated with medium confidence Figure 6-38 Measured concentrations of (a) xenon versus krypton and (b) xenon versus neon in groundwater and spring samples collected from the Cambrian Limestone Aquifer and aquifers hosted in adjacent hydrogeological units SS = Sandstone; Sol. Eq. = solubility equilibrium; Ex. Air = excess air. Samples are symbolised by aquifers or springs sampled. Filled shapes represent samples collected in this assessment. Hollow shapes represent historical sample results (Deslandes et al., 2019; Frery et al., 2022; Lamontagne et al., 2021; Tickell and Bruwer, 2018) Dook Creek Aquifer Concentrations of stable noble gases (He, Ne, Ar and Kr) from groundwater samples and spring samples collected from the DCA and aquifers hosted in the adjacent Mountain Valley Limestone are summarised in Figure 6-39. Noble gas concentrations indicate infiltration temperatures between 25 and 32°C (Figure 6-39). This temperature range is narrower than that across parts of the CLA, which is expected because the DCA occurs in the northern part of the catchment which has a monsoonal climate with higher rainfall and lower temperatures. Helium will be discussed in Section 6.2.12 since it is a good indicator of groundwater residence times as opposed to infiltration conditions. A diagram of different types of data Description automatically generated with medium confidence Figure 6-39 Measured concentrations of (a) xenon versus krypton and (b) xenon versus neon in groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in the adjacent Mountain Valley Limestone Sol. Eq. = solubility equilibrium; Ex. Air = excess air. Samples are symbolised by aquifers or springs sampled. 6.2.10 Anthropogenic gases – chlorofluorocarbons, sulfur hexafluoride, bromotrifluoromethane Cambrian Limestone Aquifer Concentrations of SF6, H1301, CFC-11 and CFC-12 in groundwater and spring samples collected from the CLA and aquifers hosted in adjacent hydrogeological units versus distance along the regional hydraulic gradient are summarised in Figure 6-40. Concentrations of SF6 in groundwater samples collected in this assessment range from 0.2 to 12.1 fMol/kg, which is similar to the range in samples collected from historical investigations for most aquifers. The expected measurable SF6 concentration for groundwater in equilibrium with atmospheric concentrations (i.e. no excess air) is 0.15 to 1.7 fMol/kg. Four samples collected in this assessment are greater than 1.7 fMol/kg as are many historical samples, indicating they are affected by excess air entrapped in the saturated zone (Figure 6-40a). Air entrapment can be caused by watertable fluctuations, either from natural recharge and discharge processes or by groundwater extraction inducing rapid drawdown and recovery at a given bore location. The range in concentrations exhibited reflect significant variation spatially across the aquifer, correlating with spatial variability in excess air from either different recharge processes (i.e. localised versus diffuse recharge) or variations in the magnitude and duration of groundwater extraction. The samples with the highest concentrations (>2 fMol/kg) are either from along or near the eastern margin of the aquifer, where recharge occurs beneath a thin (<30 m) veneer of Cretaceous rocks and sediments, or from the eastern part of the northern aquifer outcrop around Mataranka. Samples from the APV that underlies the CLA have moderate concentrations of SF6 (0.87 and 1.51 fMol/kg). Samples from aquifers hosted in the Abner and Cretaceous sandstones have relatively low to moderate concentrations of SF6 0.21 and 0.59, respectively (Figure 6-40a). Concentrations of H1301 in groundwater range from 0.3 to 2.26 fMol/kg, which falls within the expected measurable range (0.3 to 4.0 fMol/kg) assuming groundwater is in equilibrium with atmospheric concentrations (Figure 6-40b). As for SF6 concentrations in groundwater, H1301 concentrations exhibit significant spatial variability across different parts of the aquifer. Samples from the CLA with the highest concentrations of H1301 (>1.3 fMol/kg) are either along or near the eastern margin of the aquifer or from the eastern part of the northern aquifer outcrop around Mataranka. Samples from the APV that underlies the CLA have moderate concentrations of SF6 (0.74 and 1.66 fMol/kg). These results reflect differences in either: (i) localised recharge via sinkholes, (ii) licensed groundwater extraction between Mataranka and Larrimah for the Tindall Limestone, (iii) or localised episodic recharge across broad areas in the southern more arid parts of the aquifer as indicted for historical samples from the Gum Ridge Formation (Figure 6-40b). Concentrations of CFC-11 and CFC-12 in groundwater range from the detection limit of 0.05 to 0.73 pMol/kg for CFC-11 (Figure 6-40c) and from the detection limit of 0.05 to 0.74 pMol/kg for CFC-12 (Figure 6-40d). Concentrations of CFC-11 appear to be slightly degraded compared to the expected measurable composition for groundwater in equilibrium with atmospheric concentrations. Similar to the other gas tracers (SF6 and H1301), there is significant spatial variability in concentrations of CFC-11 and CFC-12 across different parts of the aquifer. Concentrations of CFC-12, while varying spatially, exhibit some similar spatial trends to the other gas tracers, though not always. Some of this is a result of degradation of CFC-11 or the recent plateau (over the past two decades) and subtle decline in atmospheric concentrations of CFC-11 and CFC-12, rendering it a partially less reliable tracer. Concentrations of CFC-12 from aquifers hosted in other adjacent (overlying or underlying) aquifers are generally low (<0.035 pMol/kg), reflecting of their localised nature. However, the Cretaceous Sandstone has a CFC-12 concentration of 0.56 pMol/kg, reflecting some recharge on a decadal scale (Figure 6-40d). A group of graphs showing different colored squares Description automatically generated with medium confidence Figure 6-40 Concentrations of various anthropogenic gas tracers in groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units versus distance from the edge of the Cambrian siltstone (Anthony Lagoon Formation – see Figure 6-18): (a) SF6, (b) H1301, (c) CFC- 11 and (d) CFC-12 SS = Sandstone. Samples are symbolised by aquifers or springs sampled. Filled shapes represent samples collected in this assessment. Hollow shapes represent historical sample results (Deslandes et al., 2019; Frery et al., 2022; Lamontagne et al., 2021; Tickell and Bruwer, 2018) Figure 6-41 summarises concentrations of SF6, H1301, CFC-11 and CFC-12 in groundwater and spring samples collected from the CLA and aquifers hosted in adjacent hydrogeological units versus depth below the watertable. Plots for SF6 (Figure 6-41a) and H1301 (Figure 6-41b) have been corrected for excess air, and residence time distribution model curves account for initial recharge conditions (i.e. recharge temperature, ground surface elevation and salinity). Nearly all groundwater samples, regardless of aquifer or gas tracer, collected in this assessment and historical samples sit between residence times distribution model curves with a vertical flow velocity ranging between 400 and 1400 mm/year. The exceptions are samples from the regional groundwater discharge zone where significant mixing in the aquifer occurs and one sample from the Gum Ridge Formation where recharge is very low. A group of graphs showing different types of data Description automatically generated with medium confidence Figure 6-41 Concentrations of various anthropogenic gas tracers in groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units versus depth below the watertable: (a) SF6, (b) H1301, (c) CFC-11 and (d) CFC-12 SS = Sandstone. Plots include predicted curves for anthropogenic gas compositions assuming different groundwater flow conditions. PM = piston flow model; Vogel = exponential model. Assumed groundwater flow conditions for predicted curves include: (i) groundwater salinity = 1 g/L, (ii) mean annual recharge temperature = 30 °C, (iii) mean ground surface elevation = 130 (mAHD), and (iv) corrections for excess air. Samples are symbolised by aquifers or springs sampled. Filled shapes represent samples collected in this assessment. Hollow shapes represent historical sample results (Deslandes et al., 2019; Frery et al., 2022; Lamontagne et al., 2021; Tickell and Bruwer, 2018) Dook Creek Aquifer Figure 6-42 summarises concentrations of SF6, H1301, CFC-11 and CFC-12 in groundwater and spring samples collected from the CLA and aquifers hosted in adjacent hydrogeological units versus depth below the watertable. Concentrations versus distance from the south-western edge of the outcrop were not plotted due to the localised to intermediate scale of shallow flow in the aquifer (see Section 6.2.8). Concentrations of SF6 in groundwater ranged between 0.15 fMol/kg (the detection limit) and 13.57 fMol/kg (Figure 6-42a). As in the CLA, samples from the DCA exhibited significant spatial variability across different parts of the aquifer and in two locations are affected by excess air as a result of different recharge and discharge processes. For example, one site with an SF6 concentration of 13.57 fMol/kg is in a recharge zone where flooding accumulates and localised recharge occurs. The other site, with an SF6 concentration of 1.95 fMol/kg, is in a discharge zone with a shallow watertable that fluctuates up to the land surface during the wet season. The sample from the aquifer hosted in the Mountain Valley Limestone contains very little SF6 (0.18 fMol/kg). Concentrations of H1301 in groundwater ranged between 0.3 fMol/kg (the detection limit) and 1.56 fMol/kg (Figure 6-42b). Samples with the highest concentrations occur in both recharge and discharge areas of the DCA, and the sample from the Mountain Valley Limestone contained little H1303, which is similar to the SF6 results. Concentrations of CFC-11 and CFC-12 in groundwater ranged between 0.05 pMol/kg (the detection limit) and 0.2 for CFC-11 (Figure 6-42c) and between 0.05 pMol/kg (the detection limit) and 0.49 pMol/kg for CFC-12 (Figure 6-42d). The sample from the Mountain Valley Limestone contained little CFC-11 and CFC-12, which is consistent with the low concentrations of the other gas tracers (SF6 and H1301). When concentrations of SF6, H1301 and, CFC-11 and CFC-12 in groundwater are plotted with depth below the watertable and residence time distribution model curves, nearly all samples sit among modelled vertical velocities of 600 to 850 mm/year (Figure 6-42a to d). Figure 6-42 Concentrations of various anthropogenic gas tracers in groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in the adjacent Mountain Valley Limestone versus depth below the watertable: (a) SF6, (b) H1301, (c) CFC-11 and (d) CFC-12 Plots include predicted curves for anthropogenic gas compositions assuming different groundwater flow conditions. PM = piston flow model; Vogel = exponential model. Assumed groundwater flow conditions for predicted curves include: (i) groundwater salinity = 1 g/L, (ii) mean annual recharge temperature = 30 °C, (iii) mean ground surface elevation = 130 (mAHD), and (iv) corrections for excess air. Samples are symbolised by aquifers or springs sampled. 6.2.11 Carbon isotopes Cambrian Limestone Aquifer Figure 6-43 is a summary of the carbon isotopic composition of groundwater and spring samples collected from the CLA and aquifers hosted in adjacent hydrogeological units. Carbon-14 concentrations are expressed as a percentage equivalent to percent modern carbon (pmC). Carbon-13 concentrations are expressed as a percentage 13C equivalent to the standard, Vienna Pee Dee Belemnite (VPDB). Carbon-14 (14C) concentrations for all groundwater samples collected in this assessment range from 48 to 82%. While for historical samples, samples from the Tindall Limestone exhibit a similar range in 14C composition except for some samples from the confined parts of the Gum Ridge Formation which in some cases are less than 10% (Figure 6-43a). Carbon- 13 concentrations for groundwater samples collected in this assessment exhibit a narrow range from −11 to −8% VPDB (Figure 6-43d). Similar to the 14C composition in groundwater, concentrations of 13C in historical samples from the Tindall Limestone also exhibit a narrow range in composition except that samples are slightly more depleted (i.e. −12 to −10% VPDB). In addition, some of the historical samples from the confined parts of the Gum Ridge Formation have a more enriched 13C composition of about −7% VPDB (Figure 6-43d). The only clear trend in 14C concentrations with distance from the northern edge of the Cambrian siltstone is that the samples from deep confined parts of the Gum Ridge Formation have significantly lower concentrations than those from the shallower unconfined parts of the Tindall Limestone. There is also no clear trend in 13C compositions. For example, samples either near or along the eastern and northern margins where the aquifer outcrops do not always have the most depleted 13C composition, which reflects the spatial variability in groundwater recharge and flow as well as water–rock interactions. Figure 6-43b summarises concentrations of 14C versus depth below the watertable and Figure 6-43c summarises total dissolved inorganic carbon (TDIC) versus distance from the northern edge of the Cambrian siltstone (see Figure 6-18). No decrease in concentrations of 14C in groundwater with depth below the watertable is exhibited (Figure 6-43). However, the shallowest samples from the eastern and northern margins of the Tindall Limestone have the highest concentrations of 14C (>70%), an indication of areas with higher recharge. There is no major trend for increasing TDIC with distance from the northern edge of the Cambrian siltstone for samples collected in this assessment (Figure 6-43). This is also reflected by the fact that all these samples are either saturated or close to saturation with respect to calcite and dolomite, and all have a significant portion of their TDIC in the form of aqueous HCO3 compared to aqueous CO2 (see Section 6.2.5). However, the historical samples exhibit a larger range in TDIC with depth (a factor of three in some cases) and distance from the Cambrian siltstone (Figure 6-43). This is further highlighted by the trend lines shown in Figure 6-43d. The trend lines likely reflects different degrees of mineral weathering depending on residence time in the aquifer and mixing of different spatial flow paths, but no geochemical modelling was conducted on historical samples to determine saturation indices or carbonate speciation. Overall, the variability in carbon isotope composition reflects the spatial variability in groundwater recharge and flow, as well as the heterogeneity of carbonate dissolution, particularly with depth in the aquifer. This may reflect either longer residence times for groundwater or mixing in the aquifer. However, the fact that most groundwater samples have a portion of their TDIC which is 14C-free due to weathering of carbonate minerals or oxidation of organic matter, indicates that use of 14C for estimating mean residence times (MRTs) for groundwater flow requires careful interpretation. Figure 6-43 Measured concentrations of carbon isotopes (a) 14C versus distance from the edge of the Cambrian siltstone (Anthony Lagoon Formation – see Figure 6-18), (b) 14C versus depth below the watertable, (c) TDIC versus distance from the Cambrian Silestone (see Figure 6-18) and (d) 13C versus 14C in groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units SS = Sandstone. Samples are symbolised by aquifers or springs sampled. Purple lines in plot (d) are trend lines for mixing across different parts of the CLA. Characterising groundwater flow using the radioactive decay of 14C requires an estimate of the initial concentration of 14C at the time of recharge. Given the lack of clear trends in either 13C composition or TDIC versus 14C concentrations, common chemical or isotope mass balance corrections (Fontes and Garnier, 1979; Ingerson and Pearson, 1964; Mook, 1976; Tamers, 1967) have not been used in this assessment to estimate the initial concentration of 14C. Alternatively, 3H concentrations in groundwater, which are highest in recharge areas along aquifer margins, have been used to estimate an upper and lower range in the initial 14C composition during recharge and its subsequent temporal decay (Figure 6-44). Tritium and 14C compositions in groundwater indicate that a correction factor of between 0.65 and 0.85 applied to the initial atmospheric 14C concentration encompasses some of the samples collected in this assessment as well as historical samples. Figure 6-44 Measured tritium and 14C concentrations for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units SS = Sandstone. Plot includes predicted curves for an upper and lower range in 14C and 3H composition assuming different groundwater flow conditions: (i) piston flow model (PM), exponential model (EM) and binary mixing model (BMM). Samples are symbolised by aquifers or springs sampled. Filled shapes represent samples collected in this assessment. Hollow shapes represent historical sample results (Deslandes et al., 2019; Frery et al., 2022; Lamontagne et al., 2021; Tickell and Bruwer, 2018) Dook Creek Aquifer Figure 6-45 is a summary of the carbon isotopic composition of groundwater and spring samples collected from the DCA and aquifers hosted in the adjacent Mountain Valley Limestone. Carbon-14 concentrations for all groundwater samples range from 3 to 83% (Figure 6-45b). Samples from areas with higher recharge exhibit the highest range in 14C concentrations (68 to 83%), while samples from discharge areas exhibit lower concentrations (3 to 48%) (Figure 6-45b). Carbon-13 concentrations in groundwater, however, exhibit considerable variability regardless of spatial location. Some samples have a more depleted composition ranging between −24 and −21% VPDB, while others exhibit a more enriched composition ranging between −17 and −13% VPDB (Figure 6-45c). Similar to variability in 13C composition and TDIC for the CLA, samples from the DCA exhibit a wide range in TDIC, varying by up to a factor of 50 from 0.5 to 25 mMol/L (Figure 6-45b). This indicates variations in the degree of carbonate dissolution within the dolomite aquifer. This is consistent with the fact that all samples have a larger portion of their dissolved inorganic carbon in the form of aqueous HCO3 than as aqueous CO2 (See Section 6.2.5, Table 6-3). In addition, most groundwater samples from the DCA are either close to saturation or are supersaturated with respect to calcite and dolomite except where bores intersect siltstone parts of the aquifer. Only three samples have a reasonably narrow 13C composition and a consistent TDIC concentration where radioactive decay of 14C is considered unaffected by carbonate dissolution. All three samples are located in discharge areas (see Figure 6-19). As for the interpretation of carbon isotopes for the CLA, 3H concentrations in groundwater, which are highest in recharge areas, have been used to estimate an upper and lower range in 14C composition during recharge and its subsequent temporal decay (Figure 6-45a). Tritium and 14C compositions in groundwater indicate that a correction factor of between 0.5 and 0.8 applied to the initial atmospheric 14C concentration encompasses most samples collected across the DCA. Figure 6-45 Measured concentrations of carbon isotopes (a) 14C versus tritium (b) 14C versus total dissolved inorganic carbon (TDIC), (c) ) 13C versus 14C, and (d) 14C versus TDIC in groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in the adjacent Mountain Valley Limestone Plot (a) includes predicted curves for an upper and lower range in 14C composition assuming different groundwater flow conditions: (i) piston flow model (PM), exponential model (EM) and binary mixing model (BMM). Samples are symbolised by aquifers or springs sampled. Pink lines in plots (b), (c) and (d) are lines representing alteration of 14C due to the addition of 14C-free TDIC. 6.2.12 Helium Cambrian Limestone Aquifer Figure 6-46 summarises measured dissolved helium concentrations in groundwater collected from the CLA and aquifers hosted in adjacent hydrogeological units. Dissolved helium concentrations in groundwater varied from atmospheric equilibrium (4.5E–08 cc/g) up to 9.8E–05 cc/g. Dissolved helium concentrations in the Tindall Limestone exhibited a reasonably narrow range for samples collected in this assessment, with concentrations varying between 4.4E–08 and 1.9E–07 cc/g. There was considerable spatial variability in dissolved helium in groundwater, which is indicative of mixing in the aquifer, though in some cases the lowest concentrations occurred along the eastern and northern margins of the aquifer near the aquifer outcrop. Samples with slightly higher concentrations came from either: (i) deeper parts of the Tindall Limestone in the central part of the aquifer beneath a thick veneer of overlying Cretaceous rocks and sediments, (ii) bores adjacent to outcropping Proterozoic sedimentary and igneous rocks of the McArthur Basin and APV respectively, or (iii) the two major springs (Rainbow and Bitter springs). The slightly higher concentrations near the northern aquifer outcrop reflects the mixing in the regional groundwater discharge zone of flow paths that range in both scale (i.e. local, intermediate and regional) as well as mean residence times (i.e. short, intermediate or long). These samples fall along a mixing line between solubility equilibrium and the accumulation of some terrigenic helium from radioactive decay of uranium and thorium in the limestone and siltstone rocks, which is indicative of longer mean residence times for groundwater flow (Figure 6-46). Samples from the Tindall Limestone that fell below the mixing line are indicative of the addition of excess air (see Section 6.2.9). All samples with excess air occur in shallow unconfined parts of the aquifer either directly in the aquifer outcrop or in subcrop beneath a thin veneer of Cretaceous rocks and sediments. Samples from aquifers hosted in other hydrogeological units and major springs generally fall within the same range as aquifers hosted in the Tindall Limestone. The exceptions are some samples from the APV, deep confined parts of the Gum Ridge Formation and the one sample from G9055086 spring in Hot Springs Valley. Samples from the APV exhibited much higher concentrations of dissolved helium (2.2E–06 and 9.8E–05 cc/g) than those in the Tindall Limestone (Figure 6-46). A few historical samples from the deep confined parts of the Gum Ridge Formation exhibit similar concentrations to those from the APV. These higher concentrations reflect an accumulation of terrigenic helium associated with longer mean residence times for groundwater flow. In the case of samples from the APV, this is indicative of limited flow in the localised lowyielding and storage-limited fractured and weathered rock aquifers. In the case of the historical samples from the Gum Ridge Formation, it is indicative of deeper confined groundwater flow. These processes are further highlighted by the samples plotting closer to the predicted compositions for a binary mixing model (BMM), reflecting vertical leakage of some recently recharged groundwater mixing with groundwater in the underlying aquifers. For example, vertical leakage from the CLA to the underlying APV and vertical leakage from the Anthony Lagoon Formation to the Gum Ridge Formation. The higher concentrations in the sample collected from G9055086 spring in Hot Springs Valley are indicative of a water source from a deep confined aquifer in the Proterozoic rocks with much longer residence times for groundwater flow. Groundwater from the deep confined aquifer with a high helium concentration would likely mix with shallower groundwater with lower helium concentrations once it passes through the APV aquitard and shallower overlying unconfined aquifer before discharging at the spring. Figure 6-46 Measured helium (He) concentrations (a) helium-3 to helium-4 ratio versus neon to helium ratio and (b) helium versus 14C in groundwater and spring samples collected from aquifers hosted the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units SS = Sandstone. Plot (a) includes trend lines for helium and neon compositions assuming accumulation of either helium-3 or helium-4. Plot (b) includes predicted helium compositions assuming different groundwater flow conditions: (i) piston flow model (PM), exponential model (EM) and binary mixing model (BMM). Samples are symbolised by aquifers or springs sampled. Dook Creek Aquifer Figure 6-47 summarises measured dissolved helium concentrations in groundwater collected from the DCA and aquifers hosted in the adjacent Mountain Valley Limestone. Dissolved helium concentrations in groundwater varied from atmospheric equilibrium (4.5E–08 cc/g) up to 1.4E– 06 cc/g. Dissolved helium concentrations in the DCA exhibited a reasonably narrow range for samples (5.8E–08 to 8.2E–07 cc/g). The one exception is a sample collected in a discharge area (see Figure 6-19) which exhibits a much higher concentration (1.4E–06 cc/g) (Figure 6-47). All samples fall along a mixing line between solubility equilibrium and the accumulation of some terrigenic helium from radioactive decay of uranium and thorium in the dolostone and siltstone rocks. Only the samples from Weemol Spring, the aquifer hosted in the Mountain Valley Limestone and one sample from the DCA collected in a discharge area (see Figure 6-19) are indicative of accumulation of some terrigenic helium associated with longer mean residence times for groundwater flow (Figure 6-47). A graph of a function Description automatically generated with medium confidence Figure 6-47 Measured helium (He) concentrations (a) helium-3 to helium-4 ratio versus neon (Ne) to helium ratio and (b) helium versus 14C in groundwater and spring samples collected from aquifers hosted in the Dook Creek Formation and aquifers hosted in adjacent hydrogeological units Plot (a) includes trend lines for helium and neon compositions assuming accumulation of either helium-3 or helium-4. Plot (b) includes predicted helium compositions assuming different groundwater flow conditions: (i) piston flow model (PM), exponential model (EM) and binary mixing model (BMM). Samples are symbolised by aquifers or springs sampled. 6.2.13 Groundwater residence times and recharge rates Cambrian Limestone Aquifer Because the anthropogenic gas tracers (SF6, H1301, CFC-11 and CFC-12) are affected by excess air entrapped in the saturated zone, 3H was deemed the most suitable tracer for estimating recharge to shallow unconfined parts of the CLA. The 3H concentrations in groundwater versus depth below the watertable (see Section 6.2.8 and Figure 6-34) showed that all samples collected in this assessment at a depth of greater than 30 m below the watertable contained measurable 3H while only about 50% of historical samples did. Furthermore, the variability in presence or absence of measurable 3H in groundwater clearly reflects the variability in MRTs across the unconfined parts of the CLA at the scale of the aquifer within the Roper catchment. The highest measured concentration of 3H occurred at a depth of about 20 m below the watertable. This implies that the vertical component of groundwater flow has taken about 60 years to reach this depth in the aquifer, allowing for semi-quantitative estimates of vertical groundwater flow. Using (i) MRTs from precited 3H compositions that fall within an upper and lower bound for vertical velocities of between 400 and 1400 mm/year (described in Section 6.2.8 and presented in Figure 6-34) and (ii) an aquifer porosity of 5% based on storage parameters derived from interpretation of pumping tests across the unconfined parts of the CLA by Amery and Tickell (2022), contemporary mean annual recharge to the shallow (<30 m below the watertable) northern and eastern parts of the CLA in the Roper catchment varies between 20 and 70 mm/year. These estimated recharge rates are consistent with previous estimates for the same parts of the CLA by Bruwer and Tickell (2015) and Jolly et al. (2004). Furthermore, they are within an order of magnitude of the estimates derived for the CLA using the upscaled CMB (see Section 5.5.2). Differences in recharge estimates arise from differences in the: (i) timescales the tracers integrate over compared to CMB, (ii) data inputs used and their uncertainty and (iii) assumptions used in each method. When predicting MRTs for deeper unconfined parts of the CLA or other adjacent low-yielding overlying or underlying aquifers with MRTs longer than several decades, the predicted curves for different 3H compositions described in the carbon isotopes section (section 6.2.11 and Figure 6-44) are required. Predicted curves of 3H compositions shown in Figure 6-48a assume groundwater flow paths with different mixtures of short and long mean residence times for groundwater flow (i.e. different proportions of 3H in groundwater). When predicting 3H compositions in groundwater that decrease exponentially with both depth and residence time (Figure 6-48b), most samples collected at a depth of greater than 30 m below the watertable, and some greater than 50 m below the watertable, for the Tindall Limestone fit close to the predicted curve for the exponential model. Furthermore, MRTs range from about 100 to 500 years. Though there is a high degree of uncertainty in the MRTs due to carbonate dissolution in the aquifer. Samples also collected from the low-yielding aquifers hosted in the overlying Cretaceous Sandstone and underlying APV also have MRTs of several hundred years. These semi-quantitative estimates of the timescale for flow using environmental tracers are consistent with the timescales for modelled groundwater flow in the CLA within the Roper catchment (see Section 6.3). Assuming (i) an aquifer thickness of 125 m (±25 m) derived from the hydrogeological cross-sections (see Section 6.1.2) and (ii) an aquifer porosity of 5% (±4.9%, i.e. from 0.1 to 9.9%), estimates of mean annual recharge rates for deeper parts of the CLA or other adjacent low-yielding aquifers with longer MRTs range from 3 to 70 mm/year. The lower rates of recharge are most applicable to the deepest samples in the CLA (samples collected at a depth of >50m below the watertable) or samples from the low-yielding Cretaceous and Proterozoic sandstones. Figure 6-48 Measured 3H versus predicted mean residence time for groundwater flow for groundwater and spring samples collected from aquifers hosted in the Cambrian Limestone and aquifers hosted in adjacent hydrogeological units SS = Sandstone. Plot (a) includes predicted curves for 3H composition in groundwater for groundwater flow paths with different mixtures of short and long mean residence times for groundwater flow (i.e. different proportions of 3H). Plot (b) includes predicted curves for 3H isotope compositions assuming different groundwater flow conditions. Piston flow model (PM) and exponential model (EM). Samples are symbolised by aquifers or springs sampled. Filled shapes represent samples collected in this assessment. Hollow shapes represent historical sample results (Deslandes et al., 2019; Frery et al., 2022; Lamontagne et al., 2021; Tickell and Bruwer, 2018). Dook Creek Aquifer Tritium was also deemed to be the most suitable tracer for estimating recharge to shallow unconfined parts of the DCA because most of the anthropogenic gas tracers (SF6, H1301, CFC-11 and CFC-12) are affected by excess air entrapped in the saturated zone. Based on the 3H concentrations in groundwater versus depth below the watertable (see Section 6.2.8 and Figure 6-37), most samples collected in this assessment at a depth of greater than 40 m below the watertable contained measurable 3H. The highest measured concentration of 3H occurred at a depth of about 20 m below the watertable, the concentration was higher than that for the CLA, which could be expected because the DCA is in a higher-rainfall zone. Using (i) MRTs from precited 3H compositions that fall within an upper and lower bound for vertical velocities of between 600 and 800 mm/year (described in Section 6.2.8 and presented in Figure 6-37) and (ii) an aquifer porosity of 5% based on storage parameters derived from interpretation of pumping tests across the unconfined parts of the DCA (Mann and Yin Foo, 1993; Verma and Rowston, 1992), contemporary mean annual recharge to the shallow (<40 m below the watertable) parts of the DCA in the Roper catchment vary between 30 and 40 mm/year. These estimated recharge rates are consistent with previous estimates of recharge to the DCA using water balance modelling (Knapton, 2009c; Williams et al., 2003). Furthermore, they are reasonably consistent with the range in recharge estimates derived for the DCA using the upscaled CMB (see Section 5.5.2). When predicting MRTs for deeper unconfined parts of the DCA or aquifers hosted in the adjacent Mountain Valley Limestone where groundwater flow may not always be vertical and MRTs may be longer than several decades, the exponential model is required. Assuming (i) an aquifer thickness of 100 m (±25 m) derived from the hydrogeological cross-sections (see Section 6.1.2) and (ii) an aquifer porosity of 5% (±4.9%), estimates of mean annual recharge rates for deeper parts of the DCA or aquifers hosted in the adjacent Mountain Valley Limestone range from 3 to 100 mm/year. For samples found in discharge areas of the DCA and having MRTs longer than several decades, the predicted curves for different 3H compositions described in the carbon isotopes section (section 6.2.11 and Figure 6-45) are required. Using the exponential model and the corrected 14C composition based on 3H in groundwater, MRTs range from a few thousand years to many thousands of years, though there is a high degree of uncertainty in the MRTs due to carbonate dissolution in the aquifer. Nevertheless, these MRTs correspond to recharge rates of a few millimetres per year, which is not surprising given all three samples were collected in net discharge zones for the DCA. 6.2.14 Groundwater–surface water interactions Data from over 23 hydrometric monitoring sites have been collated along the Roper River since 2003, but many sites have been monitored infrequently. Hydrometric surveys were undertaken in 2003 and 2004 and then every year since 2013. The number of monitoring sites has varied from year to year, with more detailed surveys (10 sites or more) made in 2013–14 and 2021–22 (Figure 6-49 and Figure 6-50). Water quality monitoring for salinity measured as EC in μS/cm was also available at most sites. Review of historical hydrometric data While a lack of consistency from year to year in the sites that were gauged complicated the review, a few trends are apparent. In most years, end-of-dry-season discharge gradually increased from the junction with the Roper Creek to upstream of Red Lily Lagoon (~32 km downstream). At the permanent gauging site G9030176 (the closest from the junction), end-of-dry-season discharge fluctuated between 2.14 and 3.67 m3/s, with a tendency towards lower discharge in more recent years. In contrast, discharge at stations G9030022 and G9030023 (~26 and ~31 km downstream, respectively) have ranged from 3.23 to 5.64 m3/s during the same period. Thus, when accounting for inputs from tributaries, 1 to 2 m3/s is typically gained within the river channel over the ‘gaining’ section of the river. Discharge tends to decrease from Red Lily Lagoon to near the estuarine section of the river (see G9035122 and G9030250; Figure 6-49 and Figure 6-50). Figure 6-49 Locations of hydrometric monitoring including discharge and salinity for end-of-dry-season surveys undertaken between 2003 and 2015 The thick black line represents the Roper River and the thin black lines represent its tributaries. Green sites are the main springs of the Mataranka Springs Complex. Discharge measurements are the top value at specified locations with values expressed in m3/second. Salinity measurements are the bottom value at specified locations with values expressed in μS/cm. Figure 6-50 Locations of hydrometric monitoring including discharge and salinity for end-of-dry-season surveys undertaken between 2016 and 2022 The thick black line represents the Roper River and the thin black lines represent its tributaries. Green sites are the main springs of the Mataranka Springs Complex. Discharge measurements are the top value at specified locations with values expressed in m3/second. Salinity measurements are the bottom value at specified locations with values expressed in μS/cm. Longitudinal surface water and spring sampling campaign Water samples were collected at 13 surface water locations and two key springs (Bitter and Rainbow springs). Locations of the sampling sites are shown in Figure 6-51. Sampling sites extend from Roper Creek in the west to just downstream of the Roper River and Elsey Creek junction in the east. A map of a river Description automatically generated Figure 6-51 Spatial distribution of surface water and spring sampling sites in the groundwater discharge zone for the Cambrian Limestone Aquifer in the Roper catchment To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The lower left map inset with red polygon indicates the location and geographic extent of the map figure within the Roper catchment. Distance in kilometres for sampling sites, represents distance downstream of the junctions between the Waterhouse and Roper rivers. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008) Geological faults data source: Department of Industry, Tourism and Trade (2010) Groundwater discharge zone data source: Department of Environment and Natural Resources (2017) Spring data source: Department of Environment Parks and Water Security (2014c) Streamflow at the time of the synoptic survey At the time of the synoptic surface water survey and spring sampling campaign, NT Government hydrographers measured discharge at six locations along the study reach targeted by the synoptic survey. These were Roper Creek (G9035085), Waterhouse River (G9035407), Elsey Creek (G9035193), Fig Tree Spring (G9030157) and Roper River (G9030176 and G9030013; Figure 6-49 and Figure 6-50). In October 2022, Roper Creek contributed about half the baseflow (1.65 m3/s), followed by Waterhouse River (0.42 m3/s), Elsey Creek (0.2 m3/s), Salt Creek (<0.2 m3/s, estimated) and Fig Tree Spring (0.01 m3/s). Salinity measurements as EC in springs and surface water clearly showed that groundwater from the eastern end of Elsey National Park is more saline than those samples taken upstream (Figure 6-52a), with Salt Creek EC approximately 3100 μS/cm. Consequently, Roper River EC (~1220 mS/cm) is similar to Bitter Spring at the confluence of the Roper Creek and Waterhouse River but increased to 1440 mS/cm as it flowed through Elsey National Park (Figure 6-52a). This pattern was also found in previous hydrographic surveys (Figure 6-50 and Figure 6-52). Stable hydrogen and oxygen isotopes There were marked variations in δ18O and especially δ2H along the Roper River (Figure 6-52b,c). For example, at the Roper Creek – Waterhouse River junction, δ2H was −56‰, but 10 km downstream the signature had increased to δ2H = −50‰ and then decreased again to −52‰ downstream from Elsey Creek. The variations in δ2H in the river are consistent with the signature of nearby springs and tributaries. In the mid-reach section, Fig Tree Spring (δ2H = −47‰) and Salt Creek (δ2H = −42‰) have a more enriched signature while Elsey Creek is relatively depleted (δ2H = −58‰). Variations in 18O were less pronounced, exhibiting a narrow range from −8.3 to −7.6‰. The two exceptions were Fig Tree Spring (δ18O = −6.6‰) and Salt Creek (δ18O = −5.7‰). Radon-222 Radon-222 activity in Rainbow and Bitter springs was very high (10 to 12 Bq/L) and similar to values measured at these springs since 2019 (Geological and Bioregional Assessment Program, 2021). However, while not sampled in 2022, radon in Fig Tree Spring was previously found to be more elevated and also much more variable over time (21.5 ± 10.0 Bq/L; mean ± SD) than Rainbow and Bitter springs (Geological and Bioregional Assessment Program, 2021). Radon-222 was also very high in Elsey Creek (22 Bq/L) but low in Salt Creek (0.11 Bq/L). Radon-222 activities in the Roper River were relatively high for surface water throughout the study reach (0.26 to 6.0 Bq/L) clearly indicating the study reach was fed by groundwater discharge but tended to be lower away from major springs and tributaries (Figure 6-52d). Tritium The trends for tritium are similar to the ones for the stable isotopes of water, with a tendency for tritium to increase mid-reach and decrease downstream from Elsey Creek (Figure 6-52e). Here again, the variability in the river matches the one observed in the springs and tributaries. Tritium is very low upstream in the Roper River (0.05 to 0.09 TU), consistent with the low values (<0.09 TU) found in Rainbow Spring, Bitter Spring, Roper Creek and Waterhouse River. This indicates most of the baseflow is from regional groundwater sources in the Daly Basin at the western end of Elsey National Park (which has very little or no post-1950 rainfall). Roper River tritium gradually increases to approximately 0.14 TU as it flows through Elsey National Park, indicating a stronger component of discharge from recent (post-1950) localised rainfall recharge. This is consistent with the elevated tritium activities previously found in Fig Tree Spring (0.34 TU (Lamontagne et al., 2021) and in October 2022 in Salt Creek (0.50 TU). In contrast, Elsey Creek had the lowest tritium activity (0.064 TU), and tritium decreased slightly in the Roper River below its junction (Figure 6-52e). As tritium is probably 1 to 2 TU in current rainfall at Mataranka (Tadros et al., 2014), a component of the groundwater at Fig Tree Spring and Salt Creek was very likely recharged locally in the aquifer outcrop during the past decade. A screenshot of a graph Description automatically generated Figure 6-52 Longitudinal trends in environmental tracers along the upper Roper River and its tributaries during October 2023 (a) electrical conductivity (EC) and discharge measurements, (b) 2H, (c) 18O, (d) 222Rn and (e) tritium Bitter Spring_up = most upstream Bitter Spring sampling site. Negative values of distance are upstream of the confluence of the Roper and Waterhouse rivers; positive values are downstream of the confluence. Blue values on plot (a) are historical discharge measurements in m3/second. Samples symbols indicate location, and those labelled ‘monitoring’ are historical data from an investigation by Lamontagne et al. (2021). Table 6-4 summarises measured dissolved noble gas concentrations in surface water and the two key springs. Concentrations of dissolved helium in surface water exhibited little variation, as found previously by Lamontagne et al. (2021). The only locations where helium was above atmospheric equilibrium concentrations were Rainbow and Bitter springs. This is consistent with slightly higher helium concentrations in groundwater in some of the southern and central parts of the CLA or in bores adjacent to outcropping Proterozoic sedimentary rocks of the McArthur Basin and igneous rocks of the APV. Helium concentrations were equivalent to atmospheric equilibrium in the Roper River, suggesting that in-river groundwater inputs are only from the CLA. However, the mass balance for helium-4 in rivers is strongly influenced by gas evasion losses, which could be large for dissolved helium in the context of the Roper River. Table 6-4 Summary of measured dissolved noble gas concentrations in surface water and spring samples Sample concentrations are expressed as cubic centimetres at standard temperature and pressure per gram of water. SAMPLING SITE He cc/g Ne cc/g Ar cc/g Kr cc/g Xe cc/g Bitter Spring – upstream 1.2E-07 1.9E-07 2.7E-04 5.4E-08 6.7E-09 Rainbow Spring 2.1E-07 2.1E-07 2.8E-04 5.8E-08 7.4E-09 Waterhouse River 4.9E-08 1.8E-07 2.5E-04 5.5E-08 6.8E-09 Roper Creek 4.2E-08 1.6E-07 2.6E-04 5.5E-08 6.8E-09 Roper River – 0 km 5.1E-08 1.8E-07 2.6E-04 5.7E-08 6.9E-09 Roper River – 1.5 km 5.3E-08 1.8E-07 2.5E-04 5.5E-08 6.8E-09 Roper River – 7.2 km 4.0E-08 1.6E-07 2.6E-04 5.6E-08 7.0E-09 *Fig Tree Spring 4.6E-08 1.9E-07 3.0E-04 6.0E-08 7.5E-09 Roper River – 10.1 km 3.8E-08 1.5E-07 2.5E-04 5.7E-08 7.1E-09 Roper River – 10.3 km 3.7E-08 1.5E-07 2.5E-04 5.4E-08 6.8E-09 Salt Creek 4.2E-08 1.7E-07 2.5E-04 5.6E-08 6.8E-09 Roper River – 12 km 3.9E-08 1.6E-07 2.5E-04 5.5E-08 6.8E-09 Roper River – 16 km 4.2E-08 1.7E-07 2.6E-04 5.7E-08 7.1E-09 Elsey Creek 4.5E-08 1.7E-07 2.5E-04 5.4E-08 6.6E-09 Roper River – 17 km 4.7E-08 1.8E-07 2.6E-04 5.6E-08 6.9E-09 Roper River – 20 km 4.5E-08 1.7E-07 2.6E-04 5.6E-08 6.9E-09 *Fig Tree Spring values are from a historical sampling campaign (Lamontagne et al., 2021). Stable isotopic composition in springs Owing to the north–south climate gradient in the NT, the stable hydrogen and oxygen isotopes of water are one of the more powerful environmental tracers to determine the origin of groundwater at a spring. The processes at play include the continental effect (the tendency for rainfall in a particular travelling air mass to become isotopically depleted over distance (Clark and Fritz, 1997)), and the amount effect (the tendency in semi-arid climates for larger rain events to be isotopically depleted and to disproportionally contribute to annual recharge (Dogramaci et al., 2012; Harrington et al., 2002)). Thus, along a north–south axis across the CLA, groundwater tends to become more isotopically depleted. In addition, while the exact processes are unclear at present, groundwater recharged farther south in the semi-arid Georgina Basin tends to have an evaporation signal (i.e. it tends to deviate to the right of the meteoric water line and be isotopically enriched relative to the original recharge (Tickell and Bruwer, 2018)). Thus, for the springs in the Mataranka area, the stable isotopes of water are a particularly useful marker for groundwater originating in the Daly Basin (i.e. coming from the north and west) or from the Georgina Basin of the CLA (i.e. coming from the south). To further evaluate the variations in isotopic composition for the major springs in the area measured since 2019, samples collected in the assessment have been compared to: (i) results from a one-off end-of-dry-season survey in October 2019 (Lamontagne et al., 2021), and (ii) samples from a historical bi-weekly monitoring program from late 2019 to early 2021. During this period, Mataranka had a very large event-to-event variability in isotopic composition with a tendency for more isotopically enriched rain during a relative dry wet season in 2019–20 (Figure 6-53a). However, from 2019 to 2022, Rainbow Spring had a nearly constant isotopic composition (δ18O ~−8.4‰ and δ2H ~−56‰) on or very close to either the Darwin or Alice Springs meteoric water lines (Figure 6-53b). This was previously interpreted as evidence that Rainbow Spring is primarily fed by the Daly flow path (i.e. groundwater flowing from the north of the Roper River). Unlike Rainbow Spring, which has three to four main vents close together in one initial waterhole approximately 30 m2 in size, Bitter Spring consists of an unknown number of vents (probably >10) spread in a concentric fashion over approximately 1 ha. These join to form a swiftly flowing stream (the main channel for Roper Creek). As for Rainbow Spring, isotopic values measured at Bitter Spring_up (most upstream Bitter Spring sampling site) and Bitter Spring_down (most downstream Bitter Spring sampling site) have been relatively constant over time, but δ2H is always approximately 1‰ more depleted than at Rainbow Spring (Figure 6-53b). These two sampling sites are on the main channel approximately 50 to 200 m downstream from the initial vents. Lamontagne et al. (2021) also sampled two of the vents (Bitter Vent 1 and 2; Figure 6-53b) upstream from Bitter Spring_up and showed that these were isotopically depleted (δ2H ~−61‰) relative to Bitter Spring_up and Bitter Spring_down. Both Bitter Vent 1 and 2 also fell on the Georgina water line (the anticipated isotopic composition for groundwater flowing northward from the Georgina Basin). The signature at the two vents led Lamontagne et al. (2021) to conclude that Bitter Spring was fed through the Georgina flow path of the CLA. Warlock Pond Spring, at the headwaters of Elsey Creek south of Mataranka, also had a signature similar to Bitter Vent 1 and 2 and was attributed a similar origin. However, there is clearly at least one other source of groundwater providing additional baseflow to Roper Creek. Unlike Rainbow and Bitter springs, Fig Tree Spring had a large temporal variability in isotopic composition between 2019 and 2021 (Figure 6-53b). In general, its isotopic composition was intermediate between expectations from the local meteoric water lines and the Georgina water line, albeit with a stronger evaporation signal than the other springs. However, Fig Tree Spring tended to be closer to the Georgina water line during a dry year and closer to the meteoric water line during a wet year (Geological and Bioregional Assessment Program, 2021). There was also evidence of a component of quick flow at Fig Tree Spring, with recent rainfall with an unusual isotopic composition occasionally detected at the scale of weeks at the outlet (Geological and Bioregional Assessment Program, 2021). Thus, unlike Rainbow and Bitter springs, there is a component of local recharge to Fig Tree Spring, but this component appears variable from year to year. Stable isotopic composition in surface water There were noticeable differences in isotopic composition between the various tributaries of the Roper River in October 2022 (Figure 6-53b). At its junction with the Roper River, Waterhouse River was isotopically enriched (δ18O = −8.14‰; δ2H = −51.8‰) relative to the other tributaries and plotted close to the local meteoric water lines. It was also enriched relative to Rainbow Spring (δ18O = −8.32‰; δ2H = −56.0‰) which discharges into it approximately 2 km upstream. Similarly, at its junction with the Roper River, Roper Creek (δ18O = −7.98; δ2H = −55.4‰) was more isotopically enriched than Bitter Spring_up (δ18O = −8.27; δ2H = −56.8‰) approximately 8 km upstream. However, Roper Creek had an intermediate composition relative to the local meteoric and Georgina water lines. At its confluence with the Roper River, Elsey Creek had the most depleted isotopic composition of all the tributaries (δ18O = −8.04; δ2H = −58.1‰); it plotted close to the Georgina water line and very close to the average composition for Georgina Basin groundwater from the CLA. Its isotopic composition was also similar to one of its headwater springs (Warlock Pond Spring) sampled in 2019. In contrast, Salt Creek had the most enriched isotopic composition (δ18O = −5.71; δ2H = −41.8‰) and plotted close the Georgina water line. Its isotopic composition was most similar to Fig Tree Spring. The contrast in isotopic composition between Elsey Creek and Salt Creek is interesting considering these tributaries join the Roper River within 4 km of one another. Below the confluence of Roper Creek and the Waterhouse River, the Roper River was most similar in isotopic composition to the Roper Creek (δ18O = −8.12; δ2H = −55.7‰), consistent with the latter’s four-fold-higher discharge relative to the Waterhouse River in October 2022 (Figure 6-50). As described in the previous section (Figure 6-52b and c), farther downstream the isotopic composition of the Roper River first increased mid-reach (near Fig Tree Spring and Salt Creek) and then decreased for 2H past Salt Creek. However, when viewed on a δ18O–δ2H plot (Figure 6-53b), these variations in the Roper River cannot solely be attributed to the tributaries because the most enriched Roper River samples plot on the local meteoric water lines and are also very similar to the average for Daly Basin regional groundwater (Figure 6-53b). In other words, the lack of an evaporation signal in the Roper River water at mid-reach stations argues against a partially evaporated in-river source similar to Salt Creek and Fig Tree Spring. Downstream from Salt Creek and Elsey Creek, the isotopic composition of the Roper River moves off the meteoric water lines again, consistent with inputs containing an evaporation signal. Figure 6-53 Stable hydrogen and oxygen isotopic composition of (a) Mataranka rainfall versus surface water in the upper Roper River from this assessment and (b) Roper River surface water samples and spring samples from this assessment versus 2019 to 2021 spring and regional groundwater Symbol represent wet or dry years (in a) and sampling locations (in b). Samples labelled ‘monitoring’ or with a year are historical data from an investigation by Lamontagne et al. (2021). 6.2.15 Tree water use This section provides a high-level summary of the results from a companion technical report by Duvert et al. (2023) on characterising tree water sources in and around Elsey National Park near the Mataranka Springs Complex. Figure 6-54 shows the locations where Phase I and II of field investigations were conducted across two different dry seasons in 2021 and 2022. For more details on the sampling locations including the details of groundwater bores samples please see the companion technical report by Duvert et al. (2023). In addition, while the integration of results for pre-dawn leaf water potential of target vegetation and gravimetric water content of the soil investigated is important, the focus of this high-level summary will centre on the application of the isotopic composition of groundwater, stem water and soil water. For a more detailed description of all the results for the tree water sourcing investigation, see Duvert et al. (2023). Figure 6-54 Location of the study sites where tree water source field investigations were conducted (a) The six sites investigated during Phase I and their approximate depth to groundwater (GWL) and (b) highlights of the field sampling locations on the transect examined at site 4 during Phase II. Figure 6-55 shows the stable hydrogen and oxygen isotopic composition of rainfall versus soil, groundwater and stem water from sampling across six field sites during Phase I of field investigations (field sampling locations are shown in Figure 6-54a). Groundwater samples plotted very close to the local meteoric water line, with the exception of site 1 where groundwater might have undergone some degree of evaporation due to the very shallow watertable (<1 mBGL). Some soil water samples plotted close to groundwater (sites 1, 4), while at other sites, soil water followed a clear evaporative pattern (sites 2, 3). Water extracted from stems had a broad range of isotopic compositions, but all plotted much lower than the source water samples (Figure 6-55). A negative δ2H offset of stem water relative to soil and groundwater was discovered and is not uncommon in stem water investigations (e.g. Barbeta et al., 2019; de la Casa et al., 2022; Duvert et al., 2022; Tetzlaff et al., 2021). However, this will not be discussed in detail here. Please see Duvert et al. (2023) for more detail. Given the observed δ2H offset in stem water, the interpretation of sources of water used by vegetation are based solely on δ18O data. Water extracted from stems had a broad range of δ18O compositions, with low (−9 to −6‰; sites 1, 3, 4), intermediate (−6 to −4‰; sites 2, 3, 5) to much higher values (−3 to 1‰; site 6) (Figure 6-55). Generally, it is reasonable to assume that stem water samples with low δ18O are more likely to originate from groundwater or from the capillary fringe, while stem water samples with higher δ18O are more likely to originate from soil water that has undergone evaporative enrichment. Based on this reasoning, all trees at site 1 and a subset of trees at sites 2, 3, 4 and 5 (i.e. sites where the watertable is less than 10 m below ground level) may be groundwater users, while all trees at site 6 and a subset of trees at sites 2, 3, 4 and 5 may preferentially use soil water. Exclusive soil water use is expected at site 6 as the watertable stands at greater than 20 m below the ground level at this site. In terms of patterns of tree water use across species, Melaleuca dealbata was an exclusive groundwater user (site 1), consistent with the fact that this species is commonly observed in swamp and seasonally inundated areas. Hakea arborescens also appeared to be a groundwater user (sites 3, 4), despite being a typical savanna species. Other species seemed to use either soil or groundwater or a combination of the two sources, such as Corymbia confertiflora (sites 4, 5), Erythrophleum chlorostachys (sites 4, 6), and Corymbia bella (sites 3, 4). A screenshot of a video game Description automatically generated Figure 6-55 Stable hydrogen and oxygen isotopic composition of rainfall versus soil, groundwater and stem water from sampling across six field sites during Phase I of field investigations Note that the deeper soil horizons (>2.5 mBGL) were not sampled during Phase I due to access constraints. The black line represents the Mataranka local meteoric water line (Geological and Bioregional Assessment Program, 2021). The isotopic data collected during Phase II of field investigations further highlight the large variations found within specific sampling locations (Figure 6-56; see Figure 6-54b for field sampling locations). This is particularly true for the ‘savanna’ and ‘transition’ locations, where stem water samples spanned δ18O values between −8.3 and −5.0‰ and between −6.7 and −3.7‰, respectively. At the ‘swamp’ location, the isotopic composition of stem water was more consistently low for both sampled species, with δ18O values ranging from −7.7 to −6.0‰ (Figure 6-56), suggesting a more prevalent groundwater use at this location. This is not unexpected as the depth to groundwater at this location might be approximately 1 to 2 m, given that the elevation difference between the ‘savanna’ and ‘swamp’ locations is approximately 5 m. The observed variability also holds for individual species. For instance, H. arborescens had trees spanning δ18O values between −7.4 and −3.7‰ across the transect. Again, this may reflect the importance of small-scale heterogeneities in the subsurface and variability in water availability across space, as well as changes in topography, elevation and depth to groundwater. A group of graphs with different colored dots Description automatically generated Figure 6-56 Stable hydrogen and oxygen isotopic composition of rainfall versus soil, groundwater and stem water from sampling across six field sites during Phase I of field investigations The larger green squares correspond to soil samples with Ψsoil >−6 mPa, that is, those for which water was likely available to trees. The black line represents the Mataranka local meteoric water line (Geological and Bioregional Assessment Program, 2021). Species are Bauhinia cunninghamii, Corymbia bella, Corymbia confertiflora, Erythrophleum chlorostachys, Hakea arborescens, Melaleuca alsophila and Melaleuca argentea. 6.3 Numerical groundwater flow modelling This section provides a high-level summary of the results of modelling future climate and future hypothetical groundwater developments in the CLA and DCA using the DR2 and DC2 FEFLOW models, respectively. The complete details of the results are documented in the companion technical report on groundwater modelling (Knapton et al., 2023). In reporting the results of the hypothetical groundwater development scenarios, no judgment is made as to whether the impacts of the modelled groundwater-level drawdown to receptors such as groundwater-dependent environmental assets or existing users are acceptable. With the CLA and DCA being regional-scale to intermediate-scale groundwater flow systems, changes in climate and increases in groundwater extraction can take many hundreds of years to fully propagate through the system. Consequently, the time period over which results are reported becomes relevant. The results reported in this section involve running the model to 2070 (~50 years) and also out to quasi-equilibrium (436 years). The 50-year time period is considered a pragmatic time period over which to consider the impacts of changes in climate and groundwater extraction because it is: (i) equivalent to more than twice the length of the investment period of a typical agricultural enterprise, (ii) roughly equivalent to the service life of a commissioned groundwater borefield, and (iii) consistent with the time period over which future climate projections have been evaluated. Also note that this time period is about five times the length of the current period over which NT water licences are assigned. Running the model over 436 years to quasi-equilibrium is beyond the current ability to project a future climate but is useful to demonstrate the long-term impacts of additional extraction in regional groundwater systems with significant time lags between changes in stresses and the changes in discharge. 6.3.1 Modelled mean annual recharge Cambrian Limestone Aquifer Figure 6-57 summarises the modelled mean annual recharge estimated across the entire DR2 FEFLOW model domain for the CLA. Modelled mean annual recharge for the entire model domain ranges from 2 to 23,743 GL/year with a mean and median of 995 and 469 GL/year, respectively (Figure 6-57a). Modelled mean annual recharge for the spatial extent of the CLA within the southwest of the Roper catchment ranges from 1 to 2,177 GL/year with a mean and median of 243 and 119 GL/year, respectively (Figure 6-57b). 1 10 100 1000 10000 100000 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 Modelled mean annual recharge (GL) Date Modelled mean annual recharge 10-year moving average (a) 1 10 100 1000 10000 100000 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 Modelled mean annual recharge (GL) Date Modelled mean annual recharge 10-year moving average (b) Figure 6-57 Modelled mean annual recharge (a) across the entire DR2 FEFLOW model domain for the Cambrian Limestone Aquifer and (b) for the spatial extent of the CLA within the Roper catchment The dashed black line is the 10-year moving mean. Modelled estimates are presented on a logarithmic scale. Dook Creek Aquifer Figure 6-58 summarises the modelled mean annual recharge estimated across the entire DC2 FEFLOW model domain for the DCA. Modelled mean annual recharge ranges from 1 to 1328 GL/year with a mean and median modelled mean annual recharge of 231 and 138 GL/year, respectively (Figure 6-58a). Modelled mean annual recharge for the spatial extent of the DCA within the north-west of the Roper catchment ranges from 1 to 892 GL/year with a mean and median of 150 and 90 GL/year, respectively (Figure 6-58b). 0 200 400 600 800 1000 1200 1400 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 Modelled mean annual recharge (GL) Date Modelled mean annual recharge 10-year moving average (a) 0 200 400 600 800 1000 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 Modelled mean annual recharge (GL) Date Modelled mean annual recharge 10-year moving average (b) Figure 6-58 Modelled mean annual recharge (a) across the entire DC2 FEFLOW model domain for the Dook Creek Aquifer and (b) for the spatial extent of the DCA within the Roper catchment Dashed black line is the 10-year moving mean. 6.3.2 Changes in water balance under future climate and hypothetical groundwater development Cambrian Limestone Aquifer The water balance results for the proposed Mataranka Water Management Zone (MWMZ) and proposed Larrimah Water Management Zone (LWMZ) in the CLA show that increased groundwater extraction will alter flow paths. Under the quasi-equilibrium no-development scenario (Scenario A′N) for the MWMZ, the inputs from recharge and flow from the north, west and south are balanced by discharge to the atmosphere as evapotranspiration and discharge to the Roper River. With current levels of development (Scenario A′), discharge due to extraction is balanced by a reduction in discharge via evapotranspiration and discharge to the river. Under the highest simulated level of hypothetical future extraction (Scenario B′105) the increased extraction is outside the MWMZ but the evapotranspiration and discharge to the river are further reduced due to the MWMZ becoming a net exporting zone compared to a net importing zone without extraction. The LWMZ does not have any discharge directly to evapotranspiration or streams as the watertable is too deep. Under natural conditions (Scenario A′N), inputs from recharge are exported out of the zone as lateral flow. At the timescale of quasi-equilibrium (436 years), under all levels of simulated extraction (scenarios A′, B′35, B′70 & B′105), the direction of flow is reversed and the LWMZ becomes a net importing zone. Under a projected future dry or medium (mid) climate, the decrease in rainfall and increase in potential evapotranspiration led to a reduction in recharge, which is balanced by a reduction in all forms of discharge at quasi-equilibrium. Under a wet future climate, the increased rainfall and potential evapotranspiration (PET) lead to an increase in recharge, which is balanced by an increase in all forms of discharge under the no development or natural conditions (Scenario A′N). Even with the maximum simulated hypothetical developments in the CLA, there is an increase in discharge from the CLA under the wet future climate scenario. Dook Creek Aquifer In the DCA under the quasi-equilibrium (436 years) no-development scenario (Scenario A′N), the inflows through recharge are balanced by discharge to evapotranspiration, rivers, springs and lateral flow out to the north-west of the Roper catchment. Under the hypothetical future development scenarios (B′6, B′12 & B′18), the increased extraction is balanced by a reduction in all forms of discharge. Under a future dry or medium (mid) climate, the decrease in rainfall and increase in PET led to a reduction in recharge, which is balanced by a reduction in all forms of discharge at quasiequilibrium. Under a wet future climate, the increased rainfall and PET lead to an increase in recharge. This increased recharge is balanced by an increase in all forms of discharge under the no development or natural conditions (Scenario A′N), which is the current state for the DCA which has no current licensed groundwater extraction. At the maximum simulated hypothetical developments in the DCA, there is an increase in discharge under the wet future climate scenario. 6.3.3 Cumulative spatial drawdown under future climate and hypothetical groundwater development Cambrian Limestone Aquifer Drawdown in groundwater levels due to additional hypothetical developments in the CLA is concentric around the seven hypothetical development sites between Larrimah and Daly Waters. Under Scenario B′70 after 50 years (Figure 6-59a), the maximum drawdown (>10 m) is centred roughly on the centroid of the future hypothetical developments, and a drawdown of 1 m extends 230 to 240 km south-east of the Roper River. Under Scenario B′70 after 436 years (Figure 6-59b), the maximum drawdown is greater (<25 m) and the 1 m drawdown contour extends further (360 to 370 km south of the Larrimah Water Management Zone and 460 to 470 km south-east of the Roper River), demonstrating that the full impacts of additional development are not realised within 50 years. The results for the scenarios with 35 GL/year of additional hypothetical development follow the same pattern of drawdown with a smaller magnitude than those for the 70 GL/year scenarios, while the scenarios with 105 GL/year of additional hypothetical development show greater drawdown. Figure 6-59 Modelled drawdown in groundwater level in the Cambrian Limestone Aquifer under (a) Scenario B70, historical climate and 70 GL/y of hypothetical future development for the year 2070 (i.e. 50 years) and (b) Scenario B′70, historical climate and 70 GL/y of hypothetical future development at quasi-equilibrium conditions for the year 2346 (i.e. 436 years) Drawdown contours under Scenario B′70 shown as the 5th, 50th and 95th percentiles of 11 replicates representing the inter-decadal variability in climate. See companion technical report on groundwater modelling (Knapton et al., 2023) for more information. The future climate projections range from a reduction in recharge under the dry and medium (mid) scenarios to an increase in recharge under the wet scenario. When combined with the hypothetical future developments, the results are more variable than those seen under the historical climate and future hypothetical developments scenario. After 50 years, the drawdown in the CLA under scenarios Ddry70 (Figure 6-60a) and Dmid70 (Figure 6-60c) are greater than the drawdown under Scenario B70 (Figure 6-59a). In contrast, the drawdown under Scenario D′wet70 (Figure 6-60e) is less and shows an increase in areas away from the hypothetical future developments. These trends continue out to 436 years: Scenario D′dry70 shows greater than 40 m drawdown in the south-east of the model domain (Figure 6-60b), whereas Scenario D’wet70 shows greater than 50 m increase in groundwater level (Figure 6-60f). The results for the scenarios with 35 GL/year of additional hypothetical development follow the same pattern of drawdown though with a smaller magnitude than those for the 70 GL/year, while the scenarios with 105 GL/year of additional hypothetical development show greater drawdown. Figure 6-60 Modelled drawdown in groundwater level in the Cambrian Limestone Aquifer under (a) Scenario Ddry70 at 2070 (i.e. after 50 years) (b) Scenario D′dry70 at 2346 (i.e. after 436 years), (c) Scenario Dmid70 at 2070 (i.e. after 50 years), (d) Scenario D′mid70 at 2346 (i.e. after 436 years), (e) Scenario Dwet70 at 2070 (i.e. after 50 years) and (f) Scenario D′wet70 at 2346 (i.e. after 436 years) Dashed lines represent negative drawdown, i.e. an increase in groundwater levels relative to Scenario AN. See the companion technical report on groundwater modelling (Knapton et al., 2023) for more information. Dook Creek Aquifer Drawdown in groundwater levels due to additional hypothetical development in the DCA is concentric around the six hypothetical extraction sites. Under Scenario B12 after 50 years (Figure 6-61a), drawdown exceeds 2 m around the hypothetical developments and extends as far as the unconfined portion of the aquifer. Under Scenario B’12 after 436 years (Figure 6-61b), the drawdown has a similar magnitude but extends further into the confined portion of the aquifer. This demonstrates that the full impacts of additional development are not realised within 50 years. The results for the scenarios with 6 GL/year of additional hypothetical development follow the same pattern of drawdown with a smaller magnitude than those for the 12 GL/year, while the scenarios with 18 GL/year of additional hypothetical development show greater drawdown. Figure 6-61 Modelled drawdown in groundwater level in the Dook Creek Aquifer (CLA) under (a) Scenario B12, historical climate and 12 GL/y of hypothetical future development for the year 2070 (i.e. after 50 years) and (b) Scenario B′12, historical climate and 12 GL/y of hypothetical future development at quasi-equilibrium conditions (i.e. after 436 years) Drawdown contours under Scenario B12 shown as the 5th, 50th and 95th percentiles of 11 replicates representing the inter-decadal variability in climate. See the companion technical report on groundwater modelling (Knapton et al., 2023) for more information. As for the CLA, the future climate projections for the DCA result in a decrease in recharge for the dry and medium (mid) scenarios and an increase in recharge for the wet future climate, leading to results that are more variable than those seen under the historical climate and future hypothetical developments scenario. After 50 years, the drawdown in the DCA under scenarios Ddry12 (Figure 6-62a) and Dmid12 (Figure 6-62c) are greater than the drawdown seen under Scenario B12 (Figure 6-61a). In contrast, the drawdown under Scenario Dwet12 (Figure 6-62e) is less and shows an increase in areas away from the hypothetical future developments. These trends continue out to 436 years: Scenario D′dry12 shows greater than 5 m drawdown in the north-west of the model domain (Figure 6-62b) and Scenario D′wet12 shows greater than 5 m increase in groundwater level (Figure 6-62f). The results for the scenarios with 6 GL/year of additional hypothetical development follow the same pattern of drawdown with a smaller magnitude than those for the 12 GL/year, while the scenarios with 18 GL/year of additional hypothetical development show greater drawdown. Figure 6-62 Modelled drawdown in groundwater level in the Dook Creek Aquifer under (a) Scenario Ddry12 at 2070 (i.e. after 50 years), (b) Scenario D′dry12 at 2346 (i.e. after 436 years), (c) Scenario Dmid12 at 2070 (i.e. after 50 years), (d) Scenario D′mid12 at 2346 (i.e. after 436 years), (e) Scenario Dwet12 at 2070 (i.e. after 50 years) and (f) Scenario D′wet12 at 2346 (i.e. after 436 years) Dashed lines represent negative drawdown, i.e. an increase in groundwater levels relative to Scenario AN. See the companion technical report on groundwater modelling (Knapton et al., 2023) for more information. 6.3.4 Cumulative drawdown at specified reporting locations Cambrian Limestone Aquifer The potential impacts, in terms of groundwater drawdown, of the three different hypothetical groundwater extraction rates (5, 10 and 15 GL/year) at seven hypothetical locations within the CLA at 2070 (i.e. after 50 years) are reported at six bores installed in a range of different hydrogeological settings and proximities to existing licensed water users and environmental assets. At the largest cumulative extraction rate (105 GL/year, Scenario B105), the maximum mean modelled drawdown in groundwater level after the 50-year period (~2070) is about 12 m, occurring in the centre of the hypothetical extraction sites (RN029013 – see Table 6-5). Drawdown of about 1 m in groundwater level – a value that can be considered measurable – is modelled to extend more than 100 km north of the centre of the hypothetical development sites to the groundwater discharge zone (RN035796 in Table 6-5), as well as south and outside the catchment south of Daly Waters (RN005621 in Table 6-5). The widespread propagation of drawdown arising from groundwater extraction is due to the low storage properties of the limestone aquifer. At the centre of the hypothetical extraction sites (RN029013), the modelled groundwater drawdown under scenarios B35, B70 and B105 exceeds the groundwater drawdown under Scenario Cdry. However, at Mataranka the modelled groundwater drawdown under Scenario Cdry (117.6 m) exceeds the drawdown under scenarios B35, B70 and B105 (118.4 m). A dry future climate and hypothetical groundwater development (i.e. Scenario D) exacerbates the groundwater drawdown modelled under Scenario B. Table 6-5 Mean modelled groundwater levels at six reporting locations within the Cambrian Limestone Aquifer for each future hypothetical groundwater development and future climate scenario compared to Scenario A′N at 2070 (i.e. after 50 years) Locations of groundwater-level reporting sites are shown in Figure 4-15. Additional maps of groundwater drawdown are provided in the companion technical report on groundwater modelling, Knapton et al. (2023). SCENARIO RN005621 –NEAR NEWCASTLE WATERS (mAHD) DIFF TO A′N (m) RN024536 – NEAR DALY WATERS (mAHD) DIFF TO A′N (m) RN028082 – NEAR LARRIMAH (mAHD) DIFF TO A′N (m) RN029012 –SOUTH OF MATARANKA (mAHD) DIFF TO A′N (m) RN029013 – SOUTH OF LARRIMAH (mAHD) DIFF TO A′N (m) RN035796 – AT MATARANKA (mAHD) DIFF TO A′N (m) AN 160.5 – 154.2 – 142.7 – 134.1 – 148.6 – 119.0 – A 160.3 −0.2 153.5 −0.7 140.4 −2.3 132.5 −1.6 147.0 −1.6 118.5 −0.5 B35 160.1 −0.4 151.1 −3.1 137.6 −5.1 131.6 −2.5 143.6 −5.0 118.4 −0.6 B70 159.9 −0.6 148.7 −5.5 134.8 −7.9 130.7 −3.4 140.2 −8.4 118.4 −0.6 B105 159.7 −0.8 146.4 −7.8 132.0 −10.7 129.7 −4.4 136.8 −11.8 118.4 −0.6 Cdry 158.8 −1.7 151.9 −2.3 137.5 −5.2 127.9 −6.2 144.7 −3.9 117.6 −1.4 Cmid 159.6 −0.9 152.7 −1.5 138.3 −4.4 128.9 −5.2 145.6 −3 118.1 −0.9 Cwet 162.8 2.3 155.4 1.2 140.5 −2.2 131.2 −2.9 148.0 −0.6 119.3 0.3 Ddry35 158.6 −1.9 149.5 −4.7 134.7 −8 126.9 −7.2 141.4 −7.2 117.5 −1.5 Ddry70 158.4 −2.1 147.1 −7.1 131.8 −10.9 125.9 −8.2 138.0 −10.6 117.5 −1.5 Ddry105 158.2 −2.3 144.7 −9.5 128.9 −13.8 124.9 −9.2 134.6 −14 117.4 −1.6 Dmid35 159.4 −1.1 150.3 −3.9 135.5 −7.2 127.9 −6.2 142.2 −6.4 118.1 −0.9 Dmid70 159.3 −1.2 147.9 −6.3 132.7 −10 126.9 −7.2 138.8 −9.8 118.0 −1.0 Dmid105 159.1 −1.4 145.5 −8.7 129.8 −12.9 126.0 −8.1 135.4 −13.2 118.0 −1.0 Dwet35 162.6 2.1 153.1 −1.1 137.8 −4.9 130.3 −3.8 144.6 −4 119.3 0.3 Dwet70 162.4 1.9 150.7 −3.5 134.9 −7.8 129.4 −4.7 141.3 −7.3 119.3 0.3 Dwet105 162.2 1.7 148.3 −5.9 132.1 −10.6 128.4 −5.7 137.9 −10.7 119.2 0.2 Dook Creek Aquifer The potential impacts of the three different hypothetical groundwater extraction rates (1, 2 and 3 GL/year) at each of the six locations within the DCA have been reported at specified locations. This includes six bores installed in a range of different settings across the shallow unconfined parts of the aquifer, such as in close vicinity to groundwater discharge zones at Flying Fox Creek and the Wilton River and to existing groundwater users such as the communities at and near Bulman (Table 6-6). At the largest cumulative extraction rate (18 GL/year, Scenario B′18), the maximum mean modelled drawdown in groundwater level after the 50-year period (~2070) is <2 m also occurring in the centre of the hypothetical extraction sites. Drawdown of about 1 m in groundwater level – a value that can be considered measurable – is modelled to extend >50 km west of the centre of the hypothetical groundwater extraction sites to Flying Fox Creek (RN031983 in Table 6-6), as well as > 50 km east and outside the catchment east of the Wilton River (RN028226 in Table 6-6). Larger drawdowns are projected for this level of hypothetical future development under a dry or median future climate (Scenarios Ddry18, Dmid18). Similar to the CLA, the widespread propagation of small drawdown impacts is due to the low storage properties of the dolostone aquifer. Table 6-6 Mean modelled groundwater levels at five reporting locations within the Dook Creek Aquifer (DCA) for each future hypothetical groundwater development and future climate scenario compared to Scenario A′N at 2070 (i.e. after 50 years) Locations shown on Figure 4-15. Maps of groundwater drawdown are provided in the companion technical report on groundwater modelling, Knapton et al. (2023). SCENARIO RN006546 – EAST OF FLYING FOX CK (mAHD) DIFF TO A′N (m) RN027811 – WEST OF WILTON R (mAHD) DIFF TO A′N (m) RN028226 – EAST OF WILTON R (mAHD) DIFF TO A′N (m) RN031983 – WEST OF FLYING FOX CK (mAHD) DIFF TO A′N (m) RN036302 – WEST OF MAINORU R (mAHD) DIFF TO A′N (m) AN 139.8 – 97.3 – 93.3 – 146.1 – 134.0 – A 139.8 0.0 97.3 0.0 93.3 0.0 146.1 0.0 134.0 0.0 B6 139.4 –0.4 97.0 –0.3 93.1 –0.2 145.8 –0.3 133.5 –0.5 B12 138.9 −0.9 96.7 −0.6 92.9 −0.4 145.4 −0.7 132.9 −1.1 B18 138.4 −1.4 96.4 −0.9 92.7 −0.6 145.1 −1.0 132.3 −1.7 Cdry 135.9 −3.9 95.6 −1.7 92.1 −1.2 142.1 −4.0 130.5 −3.5 Cmid 138.9 −0.9 96.9 −0.4 93.0 −0.3 145.1 −1.0 133.3 −0.7 Cwet 143.5 3.7 98.7 1.4 94.1 0.8 149.9 3.8 136.8 2.8 Ddry6 135.3 −4.5 95.2 −2.1 91.8 −1.5 141.6 −4.5 129.7 −4.3 Ddry12 134.6 −5.2 94.8 −2.5 91.5 −1.8 141.1 −5.0 128.9 −5.1 Ddry18 133.9 −5.9 94.4 −2.9 91.2 −2.1 140.6 −5.5 128 −6 Dmid6 138.4 −1.4 96.6 −0.7 92.8 −0.5 144.8 −1.3 132.7 −1.3 Dmid12 137.9 −1.9 96.3 −1 92.6 −0.7 144.4 −1.7 132 −2 Dmid18 137.3 −2.5 95.9 −1.4 92.4 −0.9 144 −2.1 131.3 −2.7 ′wet6 143.2 3.4 98.5 1.2 94.0 0.7 149.7 3.6 136.4 2.4 Dwet12 142.8 3.0 98.2 0.9 93.9 0.6 149.4 3.3 136.1 2.1 SCENARIO RN006546 – EAST OF FLYING FOX CK (mAHD) DIFF TO A′N (m) RN027811 – WEST OF WILTON R (mAHD) DIFF TO A′N (m) RN028226 – EAST OF WILTON R (mAHD) DIFF TO A′N (m) RN031983 – WEST OF FLYING FOX CK (mAHD) DIFF TO A′N (m) RN036302 – WEST OF MAINORU R (mAHD) DIFF TO A′N (m) Dwet18 142.5 2.7 98.0 0.7 93.8 0.5 149.2 3.1 135.7 1.7 6.3.5 Changes in groundwater discharge under future climate and hypothetical groundwater development Cambrian Limestone Aquifer Changes in discharge to streams are where changes due to future climate or development are most likely to affect assets of value to the community but are unlikely to be noticed immediately due to the time lags for groundwater flow in the regional groundwater systems. The current development under the current climate (Scenario A) is projected to reduce discharge to the Roper River by 9% at 2070; this increases to 10% under the 70 GL/year hypothetical development scenario (Scenario B70) (see Table 6-7). Fifty years is not long enough for the system to come to quasi-equilibrium: the reduction in discharge is projected to be 15% after 436 years at quasiequilibrium (Scenario B′70). The largest reductions in discharge are seen under the dry future climate with the maximum simulated hypothetical future groundwater development. Scenario Ddry105 has a projected reduction in discharge to the Roper River of 33% for 2070, increasing to 66% after 436 years at quasi-equilibrium (Scenario D′dry105) (see Table 6-7). The less-extreme median (mid) future climate and medium future development sees a projected reduction in discharge of 21% for the Roper River at 2070 (Scenario Dmid70). Dook Creek Aquifer Results similar to those for the CLA are seen in the Wilton River and Flying Fox Creek. A reduction in discharge of 7% is projected in 2070 for the medium simulated hypothetical development (Scenario B12), which becomes 11% after 436 years at quasi-equilibrium (see Table 6-7). As seen in the CLA, the largest reductions in discharge are seen under the dry future climate with the maximum simulated hypothetical future groundwater development. Results under Scenario Ddry18 see a reduction in discharge to the Wilton River of 34% and to Flying Fox Creek of 54% for 2070. These reductions become 59% and 69% respectively after 436 years at quasi-equilibrium (Scenario D′dry18) (see Table 6-7). The less-extreme median (mid) future climate and medium future development sees projected reductions in discharge of 13% and 19% for the Wilton River and Flying Fox Creek, respectively, at 2070 (Scenario Dmid12). Table 6-7 Change in mean modelled discharge to the Roper River (G9030013), Wilton River (G9030003) and Flying Fox Creek (G9030108) under current and future climate and development scenarios Results after 50 years, representing 2070, are shown with results after 436 years, at quasi-equilibrium, given in the brackets. All results are reported relative to the historical climate with no groundwater development – Scenario A′N. SCENARIO ROPER RIVER SCENARIO WILTON RIVER FLYING FOX CREEK A′N 3.3m3/s A′N 0.11 m3/s 0.63 m3/s A (A′) −9% (−9%) A (A′) 0% (0%) 0% (0%) B′35 (B′35) −10% (−12%) B′6 (B′6) −3% (−5%) −3% (−5%) B′70 (B′70) −10% (−15%) B′12 (B′12) −7% (−11%) −7% (−11%) SCENARIO ROPER RIVER SCENARIO WILTON RIVER FLYING FOX CREEK B′105 (B′105) −11% (−18%) B′18 (B′18) −11% (−16%) −11% (−16%) C′dry (C′dry) −30% (−45%) C′dry (C′dry) −22% (−35%) −45% (−57%) C′mid (C′mid) −19% (−26%) C′mid (C′mid) −6% (−8%) −12% (−15%) C′wet (C′wet) +4% (+10%) C′wet (C′wet) +22% (+33%) +54% (+66%) D′dry35 (D′dry35) −31% (−52%) D′dry6 (D′dry6) −26% (−42%) −48% (−62%) D′dry70 (D′dry70) −32% (−59%) D′dry12 (D′dry12) −30% (−50%) −51% (−66%) D′dry105 (D′dry105) −33% (−66%) D′dry18 (D′dry18) −34% (−59%) −54% (−69%) D′mid35 (D′mid35) −20% (−31%) D′mid6 (D′mid6) −9% (−14%) −16% (−20%) D′mid70 (D′mid70) −21% (−37%) D′mid12 (D′mid12) −13% (−19%) −19% (−26%) D′mid105 (D′mid105) −22% (−44%) D′mid18 (D′mid18) −17% (−26%) −23% (−31%) D′wet35 (D′wet35) +3% (+7%) D′wet6 (D′wet6) +19% (+29%) +51% (+62%) D′wet70 (D′wet70) +3% (+4%) D′wet12 (D′wet12) +16% (+24%) +48% (+58%) D′wet105 (D′wet105) +2% (+2%) D′wet18 (D′wet18) +13% (+19%) +45% (+54%) Part IV Discussion and conclusions 7 Discussion This hydrogeological assessment involved several key components: (i) a literature and data review of all previous hydrogeological investigations in the catchment, (ii) a regional-scale desktop data collation and analyses, including digitising data contained in hand-written and typed drilling records, evaluating groundwater levels, groundwater salinity and bore yields, and reinterpreting pumping test data (iii) a regional-scale recharge modelling assessment of all aquifers in the catchment, and (iv) targeted field, desktop and modelling investigations of the Cambrian Limestone Aquifer (CLA) and Dook Creek Aquifer (DCA). The literature review provided good insight into all aquifers in the catchment and their current knowledge gaps. The regional-scale assessment proved useful as a screening tool for identifying spatial trends in important groundwater attributes, including the spatial extent of aquifers, ranges for groundwater levels, groundwater salinity, bore yields and aquifer hydraulic properties. It also helped identify the most promising aquifers for targeted investigations (CLA and DCA) and provided important information to: (i) strategically select appropriate bores for environmental tracer sampling, (ii) identify strategic locations for surface water sampling for environmental tracers, and (iii) provide baseline datasets for use in detailed desktop analyses and modelling. Regional-scale desktop analyses, particularly the digitising of hand-written and typed lithology and bore construction data published in drilling records, resulted in a fundamental aquifer attribution dataset. It provided the aquifer-specific data for input to evaluate: (i) groundwater levels for characterising the scale and direction of groundwater flow from recharge to discharge areas, (ii) groundwater quality and salinity for characterising the suitability of groundwater for irrigated agriculture, (iii) bore yields and historical pumping tests and refine estimates of aquifer hydraulic properties for characterising the suitability of aquifers that potentially yield water at sufficient rate for irrigation, and (iv) current groundwater demand across different parts of the catchment. The groundwater quality data were also used to provide the chloride concentrations in groundwater for input into the regional-scale recharge modelling using the chloride mass balance (CMB) method. Regional-scale recharge modelling produced in this assessment was useful for providing groundwater recharge estimates that could be compared with: (i) previous recharge estimates across the DCA and CLA derived using CMB, (ii) arithmetic water balances, and (iii) numerical modelling. The recharge estimation techniques highlighted two key things: (i) the upscaled CMB is the only technique that estimates net recharge at a catchment or regional scale and is therefore considered the most useful (noting it cannot estimate recharge from overbank flooding, or to deep semi-confined and confined parts of aquifers), and (ii) while the upscaled CMB developed in this assessment is an improvement on past estimates of recharge in northern Australia, better accounting for spatial changes in both the physical properties of the underlying geology and runoff in areas of high rainfall and steep terrain which constrain recharge, remains challenging. Furthermore, it must be stressed that the upscaled CMB and other recharge estimation techniques do not account for aquifer storage, so it is not always clear whether the aquifers can accept these rates of recharge in any given year. The methods also do not account for preferential recharge from flooding or through localised features in aquifer outcrops. Therefore, the key features of an aquifer must be carefully conceptualised before simply deriving a recharge flux based on the surface area of an aquifer outcrop and an estimated recharge rate. Nevertheless, the estimates are a good starting point for deriving a water balance for target aquifers, either arithmetically or through use in a groundwater model. The literature and data reviews and the regional desktop assessment identified the regional- to intermediate-scale CLA and DCA as the most promising aquifers with potential for future groundwater resource development. The CLA and DCA were, therefore, the focus of targeted investigations. However, less extensive but productive aquifers of the Yalwarra Volcanics, Knuckey Formation and Mount Birch Sandstone that host fresh groundwater resources on the southern side of the Roper River may also offer potential opportunities for future groundwater resource development but require further investigation. However, these aquifers were not the subject of targeted investigations and exhibit only sparse hydrogeological information (lithology, water quality and bore yield) which highlights their potential for future development. Further drilling and pumping test investigations would be required to confirm those aquifers’ spatial extent, saturated thickness, water quality and potential to yield sufficient water to support irrigation without depleting the aquifer or inducing leakage from the Roper River, which also hosts saline water in in its lower reaches from large tidal influences. Other aquifers in the catchment – including those hosted in the Proterozoic sedimentary and igneous rocks of the Roper River and Katherine river groups, the Cambrian basalt of the APV, and the Cretaceous sandstone – host localised groundwater systems of variable water quality and bores yield and therefore only offer potential as localised or conjunctive water resources. The targeted investigations of the CLA and DCA have provided an up-to-date evaluation of both karstic aquifers and further validated and refined previous hydrogeological conceptualisations of both systems. For the CLA, this includes but is not limited too historical investigations by Jolly et al. (2004), Karp (2008), Knapton (2009c), Tickell (2005) and Yin Foo (2002) and more recent investigations by Bruwer and Tickell (2015), Deslandes et al. (2019), ELA (2022), Evans et al. (2020), Fulton and Knapton (2015), Knapton (2020) and Lamontagne et al. (2021). For the DCA, this includes but is not limited too historical investigations by Knapton (2009c), Williams et al. (2003) and Zaar and Tien (2003). Targeted field, desktop and modelling investigations of the CLA were particularly useful for: • characterising the spatial variability in groundwater recharge rates and fluxes, particularly the provision of independent estimates of recharge using environmental tracers • characterising the spatial variability in the direction, scale and mean residence times (MRTs) for groundwater flow processes across different parts of the aquifer • characterising the spatial changes in both depth to the top of the CLA and depth to groundwater • better characterising the spatial occurrence, geological controls and sources of groundwater– surface water connectivity along the upper Roper River, its major tributaries and the Mataranka Springs Complex • characterising the species of phreatophytes and sources of their water use in the regional groundwater discharge zone of the CLA near Mataranka • characterising the nature and scale of potential opportunities for future groundwater resource development as well as the key constraints, including the potential influences of climate variability and hypothetical groundwater extraction on the CLA’s water balance. Targeted field, desktop and modelling investigations of the DCA were particularly useful for: • characterising the spatial variability in groundwater recharge rates and fluxes, particularly the provision of independent estimates of recharge using environmental tracers • characterising the spatial variability in the direction, scale and mean residence times (MRTs) for groundwater flow processes across different parts of the aquifer • characterising the spatial changes in both depth to the top of the DCA and depth to groundwater • further characterising potential groundwater discharge zones using remote sensing estimates of actual evapotranspiration • characterising the nature and scale of potential opportunities for future groundwater resource development as well as the key constraints, including the potential influences of climate variability and hypothetical groundwater extraction on the DCA’s water balance. 7.1 Cambrian Limestone Aquifer 7.1.1 Regional hydrogeological context Within and beneath the south-west of the Roper catchment, the CLA underlies an area of approximately 27,500 km2. The aquifer only outcrops across a small area in the north near Mataranka which hosts the Mataranka Springs Complex, the only regional groundwater discharge zone for the aquifer in the catchment. The margins of the aquifer in the catchment occur along the north, north of Mataranka and the Roper River, and along the east, running in a north-west to south-east direction tens of kilometres east of Mataranka, Larrimah, and Daly Waters, respectively (see Figure 2-11). The CLA extends well beyond the southern and western catchment boundaries, occupying a total area (including its extent within and beneath the Roper catchment) of approximately 460,000 km2 (see Figure 2-12). However, the hydraulic properties and therefore scale of groundwater flow vary significantly across the extent of the aquifer. They are influenced by both topographic and geological structures and highs as well as the degree of interconnectivity between karstic features (i.e. sinkholes, caves, caverns and springs) across large areas. Within the Roper catchment, the CLA is mostly flat lying to gently dipping. It mostly overlies the Cambrian basalt of the Antrim Plateau Volcanics (APV) of the Kalkarindji Igneous Province, but in minor places near the upper Roper River, it also overlies Proterozoic sandstone and siltstone of the McArthur Basin. These low-permeability hydrogeological units exhibit key geological controls on groundwater flow and discharge from the CLA. Overlying the CLA is a veneer of Cretaceous sandstone, siltstone and claystone of variable thickness which has a large influence on the spatial occurrence and magnitude of vertical recharge to the underlying CLA, except where the CLA outcrops around Mataranka (Figure 7-1). Regional groundwater flow is generally from south to north, from the northern Georgina and Wiso basins into the Daly basin where the aquifer discharges: (i) to the north via localised discharge through the Mataranka Springs Complex and more broadly via diffuse or lateral outflow along the upper Roper River and tributaries where they are incised into the aquifer, and (ii) to the northwest via localised spring discharge and lateral outflow along the Flora, Daly and Katherine rivers where they are incised in the aquifer outside of the Roper catchment. Flow paths hosted in the CLA within the Roper catchment range from a few kilometres for localised flow paths in the northern aquifer outcrop up to about 150 km for regional flow originating from the northern Georgina and Wiso basins at the southern catchment boundary. A diagram of a golf course Description automatically generated Figure 7-1 Conceptual hydrogeological block model of the Cambrian Limestone Aquifer and aquifers hosted in adjacent hydrogeological units Textured and coloured geological units highlight the structural controls on vertical and horizontal groundwater flow in the Cambrian Limestone Aquifer in the regional groundwater discharge zone around the Mataranka Springs Complex and the upper Roper River and its tributaries. Blue arrows highlight the spatial variability in groundwater flow directions that converge in the discharge zone at different springs and different parts of the upper Roper River and its tributaries. Figure adapted and updated from Department of Environment and Natural Resources (2017) Groundwater resources hosted in the CLA occur across a reasonably large area (~27,500 km2) beneath the south-west of the catchment, though the aquifers exhibit a large spatial variability in their saturated thickness. The western parts of the CLA in the north-eastern part of the Wiso Basin and south-western part of the Daly Basin have a thin (i.e. <20 m) and variable saturated thickness, while some parts of the CLA in the northern Georgina Basin and south-eastern Daly Basin have a much larger saturated thickness (i.e. >100 m) (see Section 6.1.2). The depth to the top of the CLA is less than 250 mBGL across the entire spatial extent of the aquifers in the south-west of the catchment, and the depth to groundwater is less than 100 mBGL except for along the southern catchment boundary (see Sections 6.1.3 and 6.2.3). Groundwater salinity in the CLA ranges from fresh (i.e. <500 mg/L) in the western part of the CLA in the south-western Daly Basin and northeastern Wiso Basin to slightly brackish (i.e. 500 to <3000 mg/L) in the south-eastern Daly Basin and northern Georgina Basin (see Figure 5-5). Bore yields exhibit significant variability due to the high heterogeneity in the karstic nature of the aquifers, but where appropriately constructed production bores have been installed, bore yields often exceed 15 L/second. Where the aquifer is highly karstic (i.e. karsts are connected across a large area near the bore screen) and the saturated thickness of the aquifer is greater than 20 m, bore yields often exceed 50 L/second (Figure 7-2). Larger bore yields and areas of the CLA with a large, saturated thickness (i.e. >20 m) occur in the north-eastern Daly Basin and northern Georgina Basin coincide with areas where the demand for groundwater from the aquifer is currently highest in the Roper catchment. Hence, this area of the CLA was considered the most likely prospect for future groundwater resource development for groundwater-based irrigation and was the focus of targeted investigations. A diagram of a soil layer Description automatically generated Figure 7-2 Two-dimensional conceptual schematic of the variability in the functionality of bores installed in the Cambrian Limestone Aquifer system and the system’s variability in karstic features and their interconnectivity Figure adapted from Department of Environment and Natural Resources (2016) 7.1.2 Conceptual model Figure 7-1 is a three-dimensional representation of the hydrogeological conceptual model for the CLA within the Roper catchment. It highlights the key geological controls on groundwater flow processes, the sources and directions of groundwater flow, and the spatial occurrence of key discharge processes in the regional groundwater discharge zone around Mataranka, which were further characterised from targeted field investigations in this assessment. Recharge Recharge to the CLA in the Roper catchment occurs across almost the entire spatial extent of the aquifer in the catchment as it is unconfined both at the northern aquifer outcrop and also beneath the extensive veneer of overlying Cretaceous strata. However, recharge processes and rates, are highly spatially and temporally variable. Higher rates of recharge occur as localised recharge following intense wet-season rainfall at aquifer margins, as reflected in places across the northern and eastern parts of the aquifer by a Ca–HCO3 ionic composition of groundwater (see Section 6.2.5), a depleted isotopic composition in groundwater (see Section 6.2.6) and higher concentrations of tritium (3H) in groundwater (see Section 6.2.8). Localised recharge occurs via: (i) direct infiltration of rainfall and streamflow via sinkholes in and near the northern aquifer outcrop, or (ii) where surficial features including sinkholes, ephemeral streams or waterholes are incised through thinner (i.e. <30 m thick) parts of the overlying Cretaceous strata around aquifer margins (northern and eastern margin of the aquifer in the Roper catchment). It is important to note, however, that despite the northern aquifer outcrop being an area of highly localised recharge, the outcrop is predominantly a net regional groundwater discharge zone exhibiting a shallow watertable (i.e. <5 mBGL). While high rates of localised recharge can occur in this zone, recharge fluxes are constrained by the lack of available aquifer storage (i.e. the aquifer completely fills prior to groundwater discharging laterally when adjacent stream levels decline and via evapotranspiration from phreatophytes). Localised recharge to the aquifer outcrop is also important for providing the source of discharge to maintain spring flows at Fig Tree Spring and contributing additional baseflow (i.e. in addition to regional groundwater flow and discharge) from lateral outflow to the eastern parts of the upper Roper River downstream of the spring. Evidence for this includes: (i) changes in dry season baseflows, (ii) a more enriched isotopic composition, and (iii) higher concentrations of 3H in surface water compared to springs and reaches of the upper Roper River and its tributaries upstream to the west of Fig Tree Spring (see Section 6.2.14). Broader recharge to the CLA occurs as diffuse vertical leakage the overlying Cretaceous strata, though recharge rates are highly spatially variable and are often much lower through thicker parts (i.e. >30 m thick) of the overlying strata. Evidence of this leakage is reflected by: (i) a Na–HCO3 composition in groundwater (see Section 6.2.5), (ii) a more enriched stable isotopic composition in groundwater (see Section 6.2.6), and (iii) measurable but lower concentrations of 3H (i.e. <0.1 TU) in groundwater (see Section 6.2.8). Differences between localised recharge events and delayed diffuse recharge events is also reflected in the reasonably quick (days to weeks) rainfall-recharge responses in groundwater hydrographs in and near the northern aquifer outcrop, as opposed to the delayed (weeks to a month or two) and subdued rainfall-recharge responses much further south of the aquifer outcrop beneath a thicker (i.e. >30 m) veneer of overlying Cretaceous strata (see Section 5.1.2). Recharge rates estimated by both upscaled CMB and 3H concentrations in groundwater indicate a range in contemporary recharge rates (i.e. over several recent decades) for the CLA from 3 to 70 mm/year. These rates are consistent with a range of previous estimates provided from historical investigations across the same parts of the CLA by Bruwer and Tickell (2015), Crosbie and Rachakonda (2021), Jolly et al. (2004) and Knapton (2009c, 2020). In addition, numerical modelling of mean annual recharge further highlights the temporal variability in recharge across the aquifer. Estimates of mean annual recharge fluxes to the CLA for the entire model domain ranged from 2 to 23,743 GL/year with a mean and median of 995 and 469 GL/year, respectively. While for the CLA within the Roper catchment, estimates ranged from 1 to 2,177 GL/year with a mean and median of 243 and 119 GL/year, respectively. Numerical modelling of mean annual recharge has also further highlighted the sensitivity of the CLA’s water balance to climate variability. Despite sinkholes being a direct conduit for localised recharge, they are also mapped in places at the surface across thick (i.e. >30 m) Cretaceous strata overlying large areas of the central parts of the aquifer away from the northern and eastern margins. In these areas, low but measurable concentrations of 3H (i.e. <0.1 TU) in groundwater that is brackish (i.e. salinity >1,000 mg/L) and with a Na–HCO3 ionic composition occur in the CLA (see Section 6.2.8). This highlights that not all sinkholes are permeable and connected to the underlying CLA particularly where Cretaceous strata is thick (i.e. >30 m). Figure 7-3 summarises the conceptual representations of six different types of karstic sinkhole features (sometimes referred to as dolines) often found in different karst landscapes (Geological Society of London, 2012). Highly permeable solution sinkholes created from dissolution weathering often occur across outcropping parts of the CLA including the northern aquifer outcrop around Mataranka where localised recharge rates can be high (Bruwer and Tickell, 2015; Deslandes, 2019; Karp, 2002). In contrast, dropout sinkholes which are also highly permeable as reflected by higher concentrations of 3H (i.e. >0.1 TU) in groundwater with a fresher salinity (i.e. <1,000 mg/L) and a Ca–HCO3 ionic composition occur (see Section 6.2.8), occur near the northern aquifer outcrop or along the eastern margin of the CLA where the overlying Cretaceous strata are thin (i.e. <30 m) (Figure 7-3). Where Cretaceous strata are thicker (i.e. >30 m), buried sinkholes can be common but are often filled with compacted sediments and are unlikely to enhance groundwater recharge, particularly if filled with low-permeability claystone. Collapse sinkholes that generate subterranean caves or caverns are quite common across parts of the CLA and when intersected with drilling often result in a loss of circulation. Production bores installed in these locations often exhibit some of the highest bore yields across the aquifer (Bruwer and Tickell, 2015). Figure 7-3 Conceptual schematic representing six different types of karstic sinkhole features (sometimes referred to as dolines) often found in different karst landscapes Figure sourced from Geological Society of London (2012) Groundwater flow and residence times Regional groundwater flow (i.e. flow paths >100 km) in the CLA beneath the Roper catchment is generally from south to north following a subdued form of the topographic gradient. The direction of regional flow from the northern Georgina Basin (the Georgina flow path) is influenced by the structural high of the APV adjacent to the Birdum Creek fault (see Section 6.1.2) which guides flow north toward Mataranka. In the northern Wiso Basin (the Flora Flow Path), regional flow is also influenced by the presence of the structural high of the APV which guides flow to the north-west in the southern Daly Basin toward Katherine (see 6.2.1). In the northern Daly Basin (the Daly flow path), an intermediate-scale flow path (i.e. flow path of ~50 km) occurs from the west (east of Katherine) on the western catchment boundary and flows east toward Mataranka. Smaller localised flow paths also occur in the northern aquifer outcrop of the Daly Basin. These include flow from the northern aquifer margin south toward the Roper River and flow from localised lateral outflow from the aquifer outcrop north toward the river (Figure 7-1). Semi-quantitative estimates of mean residence times for groundwater flow derived using 3H and carbon-14 (14C) concentrations in groundwater (see Section 6.2.13) also vary significantly across the aquifer. These range from several years for localised flow paths of less than 5 km in and near the aquifer outcrop to many hundreds of years for regional flow paths greater than 150 km from the southern basins. There is a high degree of uncertainty in the MRTs due to carbonate dissolution in the aquifer, but these estimates agree with the ranges in timescales for flow based on numerical modelling (ELA, 2022; Knapton et al., 2023). Discharge Discharge from the CLA in the Roper catchment occurs via a combination of: (i) lateral outflow to streams where they are incised in the aquifer outcrop (Roper Creek, upper Roper River, Waterhouse River and Elsey Creek), (ii) localised spring discharge (Bitter, Rainbow, and Fig Tree springs and other minor springs) including instream springs associated with larger karstic features in the bed of the upper Roper River and its tributaries, (iii) transpiration from riparian, spring-fed and other phreatophytic vegetation where the watertable for the CLA is shallow (i.e. ≤ 5 mBGL) in and around Elsey National Park, and (iv) extraction of groundwater for stock and domestic use, irrigated agriculture and community water supply. The results of drilling investigations conducted in Elsey National Park (see Section 6.1.1) and the development of new hydrogeological cross-sections highlight some of the key geological controls on vertical and horizontal groundwater flow and discharge in the regional groundwater discharge zone around Mataranka. A basement high associated with the Urapunga Granite (crystalline basement of the McArthur basin) was encountered at a depth of about 100 mBGL less than 10 km south of the Roper River in bore RN043047 (see Appendix A.1.2). In addition, Proterozoic sandstone and siltstone of the Roper Group of the McArthur Basin were also encountered at a depth of about 100 mBGL 1 km north of the Roper River and Bitter Spring in bore RN043045 (see Appendix A.1.2). These basement highs associated with very low-permeability hydrogeological units influence the upward dipping of the overlying strata (see Section 6.1.2, Figure 6-7 and Figure 6-8), including the Cambrian basalt of the APV of the Kalkarindji Igneous Province which dips upward close to the land surface (i.e. <50 mBGL). The Cambrian Limestone overlying the APV, in this case the Tindall Limestone of the Daly Basin, also dips upward to outcrop at the land surface, resulting in the regional groundwater discharge zone around the upper Roper River near Mataranka. This basement high is consistent with data in the regional depth to basement shown by the NT SEEBASE for the area near Mataranka (see Section 2.3.4). In addition, these structural geological controls influence the position of the CLA’s northern aquifer outcrop, and the positions of Salt and Elsey creeks, which terminate at junctions with the Roper River located between two subcropping areas of the Proterozoic sandstone and siltstone of the Roper Group of the McArthur Basin (see Figure 4-10 in Section 4.2.11, also Figure 7-1). The review of historical hydrographic surveys and the synoptic longitudinal survey of the upper Roper River, its tributaries and major springs of the Mataranka Springs Complex, and the results from new drilling, have demonstrated the complexity and spatial variability in a combination of localised, intermediate- and regional-scale groundwater discharge from the CLA at different locations within the regional groundwater discharge zone (see Sections 6.1.2 and 6.2.14). The slightly brackish salinities, depleted isotopic composition and low 3H concentrations in water sampled at Roper Creek and Bitter Springs suggest their source of discharge is primarily from eastward-flowing groundwater in the CLA originating from a groundwater flow divide near the western margin of the Roper catchment approximately 50 km west of Mataranka (Figure 7-1). The slightly fresher salinities than for Roper Creek and Bitter Springs and depleted isotopic composition reflected in water sampled from Rainbow Spring and the Waterhouse River are more consistent with flow in the CLA originating from the northern aquifer margin in the Daly Basin flowing south to the Roper River (Bruwer and Tickell, 2015). More-enriched isotopic compositions and the highest salinity waters sampled in Fig Tree Spring and Salt Creek suggest their sources of discharge are more reflective of much shorter localised flow within the aquifer outcrop. The isotopic compositions of water sampled in Elsey Creek and the Roper River downstream of its junction with Elsey Creek suggest their source of discharge originated from regional groundwater flow from the south in the Georgina Basin flowing north to the north-eastern parts of the upper Roper River. The high temperature (~65 °C), Ca–Cl composition, vastly different strontium isotopic composition compared to samples from the CLA or APV, and higher concentrations of dissolved helium of water sampled from G9055086 spring slightly east of the CLA’s eastern boundary clearly indicate its water source is from much deeper Proterozoic hydrogeological units of the McArthur Basin (see Section 6.2.4 to 6.2.12). Tree water use investigations have demonstrated that patterns of tree water use are highly variable in and near Esley National Park, as well as being variable among individual tree species. The data showed that phreatophytic vegetation such as Melaleuca dealbata and Hakea arborescens exclusively source their water from the CLA, but coexist with vegetation that are soil water users across areas in and near Elsey National Park where the watertable is <7 mBGL. The identification of potential groundwater discharge areas using remote sensed data has also validated previous mapped and modelled extents of the regional groundwater discharge zone around Mataranka (Department of Environment and Natural Resources, 2017; Knapton, 2020; Tickell, 2016). In addition, it has identified potential groundwater discharge areas along the upper and middle reaches of Maryfield Creek and the middle reaches of the Arnold River which are adjacent but just outside the northern margins of the CLA (see Section 5.6). 7.1.3 Scenario-based numerical groundwater flow modelling of hypothetical future changes in climate and groundwater extraction Scenario-based numerical groundwater flow modelling of the CLA within the Roper catchment in this assessment was conducted to evaluate: (i) the current water balance, and (ii) the potential impacts of changes in rainfall and potential evaporation and/or potential increased groundwater resource development on the streamflow of the upper Roper River and groundwater levels in the vicinity of existing licensed groundwater users and environmental receptors in the Mataranka region. Mean annual recharge modelling using the historical climate record for the DR2 FEFLOW model domain (much larger than the Roper catchment) indicates that recent climate conditions have led to significantly higher recharge than the long-term mean. Using the historical climate to simulate the future climate to 2070 suggests that a hypothetical increase in groundwater extraction of up to 105 GL/year will result in a further 2% reduction in discharge to the Roper River and additional drawdown in groundwater levels compared to current licensed groundwater extraction. Drawdown levels would be expected to drop by 0.9 m in the regional groundwater discharge zone and up to 11.8 m at a distance of 110 km south of the river. Increased groundwater extraction does not linearly correspond to a proportional decrease in groundwater discharge to the Roper River, especially in the case of the CLA because a large proportion of groundwater extraction is partitioned between reductions in discharge to ET and capture of groundwater throughflow that would otherwise flow downgradient towards the Roper River. Simulating a future drier climate (10% reduction in rainfall) to 2070 with current licensed groundwater extraction compared to current licensed groundwater extraction under an historical climate, results in a further 21% reduction in discharge to the Roper River and additional drawdown in groundwater levels. Drawdown levels would be expected to drop by 0.9 m in the regional groundwater discharge zone and up to 3.9 m at a distance of 110 km south of the river. Simulations of the hypothetical climate scenarios highlight the potential importance of considering the influence of climate variability on the water balance of the CLA. The simulations also highlight the importance of some key factors associated with the hydrogeological conceptual model of the CLA including: (i) the spatial variability in localised recharge and discharge in the northern aquifer outcrop and the spatially extensive diffuse recharge across large parts of the CLA via vertical leakage through the overlying Cretaceous strata in the Roper catchment, and (ii) the influence of the extensive spatial coverage, thickness and variable hydraulic properties of the karstic groundwater system significantly influences the time lags for hydrological impacts of both climate variability and groundwater extraction to propagate through the system. With regard to the time lags, the full impacts of changes in stresses on the CLA due to groundwater extraction or climate, are not evident in the 50-year future scenarios. They will require hundreds of years to come to a state of quasi-equilibrium as demonstrated by the quasi-equilibrium (436-year) modelling scenarios (Knapton et al., 2023). These key findings are potentially important for underpinning adaptive management of the CLA into the future (i.e. conducting ongoing periodic reviews of the spatial location and magnitude of individual licensed entitlements, as well as overall allocation from the CLA across the different water allocation planning areas and their water management zones). 7.2 Dook Creek Aquifer 7.2.1 Regional hydrogeological context Within the northern part of the Roper catchment, the DCA occupies an area of approximately 14,100 km2 within and beneath the catchment (about 18% of the Roper catchment). The DCA is hosted in the Proterozoic dolostones of the Mount Rigg Group of the McArthur Basin. The aquifer outcrops and subcrops between Barunga and Bulman and to the north-east of the catchment boundary. Where the DCA extends to the north-east outside the Roper catchment it occupies a total area (including its extent within and beneath the Roper Catchment) of approximately 21,800 km2. The DCA dips subtly in the subsurface west of the Central Arnhem Road where it is entirely unconfined across outcropping areas and subcropping areas beneath a thin (i.e. 20 m thick) and patchy veneer of overlying Cretaceous sandstone, siltstone and claystone. East of the Central Arnhem Road, the DCA dips steeply in the subsurface and becomes confined and is modelled to be artesian beneath the Proterozoic sandstones and siltstones of the Roper Group of the McArthur Basin (Figure 7-5). As with the CLA, the hydraulic properties of the DCA vary significantly across the extent of the aquifer influencing the scale of groundwater flow. Key influences include both topographic and geological structures and contacts as well as the degree of interconnectivity between karstic features (i.e. sinkholes, caves, caverns and springs) across large areas. Regional- (i.e. >100 km) to intermediate-scale (i.e. 20 to 50 km) groundwater flow in the DCA is modelled to be generally from south-west to north-east and follows a subdued form of topography (Figure 7-4). However, groundwater-level data for the aquifer are sparse except near Barunga and Bulman. Recharge occurs as a combination of localised recharge in the aquifer outcrop and broad diffuse recharge through the overlying Cretaceous strata. Groundwater discharge occurs via: (i) a combination of diffuse seepage or outflow to streams (Flying Fox Creek, Mainoru and Wilton rivers) where they are incised in the aquifer outcrop, (ii) localised spring discharge (e.g. Top, Lindsay and Weemol springs) at contacts with occasional dolerite intrusions or more broadly the Proterozoic rocks of the Roper Group, (iii) transpiration from riparian and springfed vegetation, and (iv) minor extraction of groundwater. Groundwater resources hosted in the DCA occur across a reasonable sized area (~14,100 km2) within and beneath the northern part of the catchment, though estimates of spatial changes in the total saturated thickness of the aquifer are sparse. The unconfined parts of the DCA, west of the Central Arnhem Road where the aquifer outcrops and subcrops beneath a thin veneer of Cretaceous strata, occur mostly at depths of less than 50 mBGL but in places occur at depths up to 100 mBGL (Figure 7-5). The depths to groundwater in the eastern unconfined parts of the aquifer range from 10 to 110 m below ground surface. Groundwater is shallowest (i.e. <10 mBGL) in the vicinity of groundwater discharge zones around the lower reaches of Flying Fox Creek and the Mainoru and Wilton rivers. Where the aquifer transitions from unconfined to confined, east of the Central Arnhem Road, groundwater depths range from 80 m below ground surface to 150 m above ground surface. Groundwater salinity is fresh (i.e. <500 mg/L) across the unconfined parts of the aquifer. As in the CLA, bore yields exhibit significant variability due to the high heterogeneity in the karstic nature of the dolostone aquifer. Where appropriately constructed production bores have been installed, bore yields often exceed 15 L/second, and in the few locations where pumping tests have been undertaken, yields can exceed 30 L/second. Despite sparse hydrogeological data across the DCA, key attributes (i.e. fresh and shallow groundwater, high bore-yields and economical depths to access the aquifer) highlight that this portion of the aquifer is the most likely prospect for future groundwater resource development and was the focus of targeted investigations. A graph of a graph showing the different types of data Description automatically generated with medium confidence Figure 7-4 Two-dimensional hydrogeological conceptual model of groundwater flow processes in the Dook Creek Aquifer 7.2.2 Conceptual model Recharge Recharge to the DCA in the Roper catchment occurs across the unconfined part of the aquifer (west of the Central Arnhem Road) where it outcrops and subcrops beneath a thin and patchy veneer of overlying Cretaceous strata (Figure 7-5). However, recharge rates and processes are highly spatially and temporally variable due to the heterogeneity in the karstic features across the aquifer and the presence or absence of the patchy veneer of overlying Cretaceous strata. Localised recharge occurs: (i) via direct infiltration of rainfall and streamflow via sinkholes in and near the aquifer margins of the outcropping area, or (ii) where surficial features including sinkholes, ephemeral streams or waterholes are incised through the thin overlying Cretaceous strata across the unconfined areas of the aquifer. The less depleted isotopic composition of groundwater which falls on the Darwin local meteoric water line compared to parts of groundwater in the CLA suggests that smaller wet-season rainfall events in this higher-rainfall zone of the catchment generate localised net recharge to the shallow aquifer (see Section 6.2.6). These localised recharge events are also reflected in reasonably quick (days to weeks) rainfall-recharge responses in groundwater hydrographs (see Section 5.1.2) and fresh groundwater salinities across parts of the unconfined extent of the aquifer (see Section 5.2). Recharge rates estimated by both upscaled CMB and 3H concentrations in groundwater indicate a range in contemporary recharge rates (i.e. over several recent decades) for the DCA ranging from 3 to 100 mm/year. These rates are consistent with a range of previous estimates provided from investigations across the same parts of the CLA by Knapton (2009c) and Williams (2003). In addition, numerical modelling of mean annual recharge also further highlights the temporal variability in recharge across the aquifer. Modelled mean annual recharge for the entire DC2 model domain ranges from 1 to 1328 GL/year with a mean and median modelled mean annual recharge of 231 and 138 GL/year, respectively. Modelled mean annual recharge for the spatial extent of the DCA within the north-west of the Roper catchment ranges from 1 to 892GL/year with a mean and median of 150 and 90 GL/year, respectively. Groundwater flow and residence times Regional- (i.e. >100 km) to intermediate-scale (i.e. 20 to 50 km) groundwater flow in the DCA is generally from south-west to north-east following a subdued form of the topographic gradient. While longer flow paths follow the topographic gradient to the north-east, shallower and shorter intermediate-scale flows are truncated by the position of the northern tributaries of the Roper River. These tributaries, Flying Fox Creek and the Mainoru and Wilton rivers, are incised in the DCA outcrop and receive lateral outflow from the aquifer where reaches of streams are incised into the watertable or where the lower reaches join the contact with the rocks of the Roper Group, roughly around the Central Arnhem Road where the aquifer dips steeply to the east and becomes confined (see Section 6.1.3, Figure 6-13). In addition, occasional dolerite dykes associated with the Derim Derim dolerite intrude the DCA and truncate flow at localised points where contact springs sometimes occur. Timescales for groundwater flow can be highly variable in the karstic dolostone aquifer and also where the aquifer is hosted in siltstone lithologies of the DCA. Semi-quantitative estimates of mean residence times for groundwater flow derived using 3H and 14C concentrations in groundwater (see Section 6.2.13) also vary significantly across the aquifer. They range from several years for localised flow paths of less than 5 km in shallow unconfined parts of the DCA to a few thousand to many thousands of years for regional flow paths more than 100 km towards discharge areas or within low-permeability siltstone parts of the aquifer. There is, however, a high degree of uncertainty in the MRT estimates due to carbonate dissolution in the aquifer. Discharge Discharge from the DCA occurs via: (i) a combination of diffuse seepage to streams (Flying Fox Creek, Mainoru and Wilton rivers) where they are incised in the aquifer outcrop, (ii) localised spring discharge (e.g. Top, Lindsay and Weemol springs) at contacts with occasional dolerite intrusions or more broadly the Proterozoic rocks of the Roper Group, (iii) transpiration from riparian and spring-fed vegetation of which Pandanus is prominent around springs, and (iv) groundwater extraction for community water supply at Bulman and Weemol. Spring complexes and groundwater-fed reaches of streams are most prominent around the Wilton River in the Roper catchment in the vicinity of thew Bulman fault (see Figure 2-9). Identifying potential groundwater discharge areas using remote sensed data has also validated previous mapped locations of groundwater discharge zone around reaches of Flying Fox Creek and the Mainoru and Wilton rivers (see Section 5.6). In addition, preliminary estimates of discharge fluxes have been estimated for polygons representing the spatial extent of these potential discharge areas and provide promising locations to further investigate groundwater–surface water interactions across parts of the DCA. 7.2.3 Scenario-based numerical groundwater flow modelling of hypothetical future changes in climate and groundwater extraction Scenario-based numerical groundwater flow modelling of the DCA within the Roper catchment in this assessment was conducted to evaluate: (i) the current water balance, and (ii) the potential impacts of changes in rainfall and potential evaporation and/or potential increased groundwater resource development on the streamflow of the northern tributaries of the Roper River and groundwater levels in the vicinity of existing groundwater users and environmental receptors in the Bulman region near the Wilton River and across other parts of the DCA where groundwater discharges to springs and streams (e.g. Flying Fox Creek). Mean annual recharge modelling using the historical climate record for the DC2 FEFLOW model domain (extending slightly north-east of the Roper catchment boundary) indicates that recent climate conditions have led to slightly higher recharge than the long-term mean. It is important to note that very little groundwater extraction occurs across the DCA except for community water supplies at Bulman and Weemol (~0.1 GL/year), which results in no changes to natural groundwater discharge. Using the historical climate to simulate the future climate to 2070 suggests that a hypothetical increase in groundwater extraction of up to 18 GL/year in the DCA will result in 11% reductions in discharge to both the Wilton River and Flying Fox Creek. Additional drawdown in groundwater levels is greater than 2 m across the unconfined area of the aquifer. As in the CLA, increased groundwater extraction does not linearly correspond to a proportional decrease in groundwater discharge to the Wilton River and Flying Fox Creek. A proportion of the groundwater extraction is partitioned into reductions in discharge to ET and capture of groundwater throughflow that would otherwise flow downgradient and outside the catchment toward the Goyder River. Simulating a future drier climate (10% reduction in rainfall) to 2070 with current licensed groundwater extraction (only 0.1 GL/year) results in further 22 and 45% reductions in groundwater discharge to the Wilton River and Flying Fox Creek, respectively. Reductions in groundwater levels were less than 4 m across the unconfined area of the aquifer compared to current licensed groundwater extraction (0.1 GL/year) under a historical climate. As in the CLA, simulations of the hypothetical climate scenarios highlight the potential importance of considering the influence of climate variability on the water balance of the DCA. The simulations also highlight the importance of some key factors associated with the hydrogeological conceptual model of the DCA including: (i) the spatial variability in localised recharge and discharge across the aquifer, and (ii) the significant influence of the spatial coverage, thickness and variable hydraulic properties of the karstic groundwater system on the time lags for hydrological impacts of both climate variability and groundwater extraction to propagate through the system to key springs and groundwater-fed streams. For the time lags, the full impacts of changes in stresses on the aquifer system due to extraction or climate, are not evident in the 50-year future scenarios for the DCA. They will require hundreds of years to come to a state of quasi-equilibrium as demonstrated by the quasi-equilibrium (436-year) modelling scenarios (Knapton et al., 2023). These key findings are potentially useful to underpin adaptive management of the DCA should it’s groundwater resource be further developed in the future. 7.3 Potential opportunities for future groundwater resource development Planning future groundwater resource developments and authorising licensed groundwater entitlements require value judgments as to what is an acceptable impact to receptors such as environmental assets or existing users at a given location. These decisions can be complex and typically require considerable input from a wide range of stakeholders, particularly government regulators and communities, to agree upon what is an acceptable change to a water resource’s condition (i.e. water balance). This hydrological assessment provides scientific information to help inform these potential future decisions, including: (i) identifying aquifers potentially suitable for future groundwater resource development, (ii) characterising their depth, spatial extent, saturated thickness, hydraulic properties and water quality, (iii) conceptualising the nature of their flow systems, (iv) estimating aquifer water balances, and (v) providing initial estimates of potential extractable volumes of groundwater and the associated drawdown in groundwater level over time and distance relative to existing water users and environmental receptors such as groundwater-dependent ecosystems (GDEs), as well as changes in the aquifer water balance. The changes in drawdown and components of the aquifer water balance over time at different locations provide information on the potential risks of changes in aquifer storage, and therefore water availability, to existing users or the environment. It is important to note that the results of the hypothetical future groundwater development modelling scenarios do not provide an assessment of or expert judgment or opinion as to whether the potential impacts of groundwater-level drawdown are acceptable to receptors such as environmental assets or existing users. Instead, they provide defensible information to inform future groundwater resource planning and investment and management. The hydrogeological units of the Roper catchment (Figure 7-5) contain a variety of local-, intermediate- and regional-scale aquifers that host localised to regional-scale groundwater flow systems. The intermediate- to regional-scale limestone and dolostone aquifers are present in the subsurface across large areas, collectively occurring beneath about 50% of the catchment. Given their large spatial extent, they also underlie and coincide frequently with larger areas of soil suitable for irrigated agriculture (Thomas et al., 2022). They contain mostly low-salinity water (<1000 mg/L total dissolved solids (TDS)) and can yield water at a sufficient rate to support irrigation development (>10 L/second). These aquifers also exhibit large magnitudes in components of their water balance (recharge and discharge) and contain larger volumes of groundwater in storage (gigalitres to teralitres) than do local-scale aquifers. Furthermore, their storage and discharge characteristics are often less affected by short-term (yearly) variations in recharge rates caused by inter-annual variability in rainfall. The larger spatial extent of these intermediate- to regional-scale aquifers provides greater opportunities for groundwater resource development away from existing water users and GDEs at the land surface, such as springs, springfed vegetation and surface water, which can be ecologically and culturally significant. In contrast, local-scale aquifers in the Roper catchment, such as fractured and weathered rock and alluvial aquifers, host local-scale groundwater systems that are highly variable in composition, salinity and yield. They also have a small and variable spatial extent and less storage compared to the larger aquifers, limiting groundwater resource development to localised opportunities such as stock and domestic use, or as a conjunctive water resource (i.e. combined use of surface water, groundwater or rainwater). This assessment identified five hydrogeological units hosting aquifers that may have potential for future groundwater resource development of various scales in the Roper catchment (see Table 7-1 for details and Figure 7-5 for the locations of different hydrogeological units): • Cambrian limestone • Proterozoic dolostone and sandstone • Cretaceous sandstone, siltstone and claystone • Proterozoic sedimentary and igneous rocks • Cambrian basalt. Table 7-1 Opportunity-level estimates of the potential scale of groundwater resource development opportunities in the Roper catchment For locations of the hydrogeological units see Figure 7-5. The indicative scale of the groundwater resource is based on the magnitude of the inputs and outputs of the groundwater balance. The actual scale will depend upon government and community acceptance of potential impacts of groundwater-dependent ecosystems and existing groundwater users. HYDROGEOLOGICAL UNIT LOCATION LEVEL OF KNOWLEDGE INDICATIVE SCALE OF RESOURCE (GL/y) † COMMENT Cambrian limestone South to southwestern part of the catchment Medium to high 35–105 Most promising regional-scale aquifer, the CLA, is typically tens of metres thick with high bore yields (5–50 L/s) and good water quality (<1000 mg/L TDS). Has potential to support multiple large-scale (5–15 GL/y) developments. Greatest opportunities exist in the Georgina Basin Water Management Zone of the Georgina Wiso water allocation plan area as well as the Larrimah Water Management Zone in the proposed Mataranka water allocation plan area, respectively. Opportunities are limited where water management zones have reached or are close to full allocation, or if the nature and cumulative scale of development will potentially affect water availability to existing licensed water users and the Mataranka Springs Complex and upper Roper River and its major tributaries. Proterozoic dolostone and sandstone Northeastern part of the catchment Low to medium <20 Promising intermediate-scale aquifer, the DCA, hosted in the Proterozoic dolostone (Dook Creek Formation). The aquifer is typically tens of metres thick with high bore yields (5–50 L/s) and good water quality (<500 mg/L TDS). Has potential to support multiple small to intermediate-scale (1–3 GL/y) developments. Greatest opportunities exist where the aquifer is unconfined west of the Central Arnhem Road between Flying Fox Creek and the Wilton River. Opportunities are limited near community water supplies for Beswick, Bulman, and Weemol, where the aquifer is connected to Flying Fox Creek and the Mainoru and Wilton rivers, and where springs occur (e.g. Top, Lindsay and Weemol springs). The Proterozoic dolostone and sandstone aquifers (Yalwarra Volcanics, Knuckey Formation and Mount Birch Sandstone) on the southern side of the Roper River away from Ngukurr may offer some potential opportunities for future groundwater development but this requires further investigation. Cretaceous sandstone, siltstone and claystone Southern part of the catchment Low <5 Local-scale sandstone aquifers occurring as localised basal quartzose sandstones. Variable bore yields (often <5 L/s) and variable water quality. Only likely to offer potential for smallscale (<0.5 GL/y) localised developments (i.e. mostly suited to stock and domestic water supplies) or as a conjunctive water resource where basal sandstone units are present. Proterozoic sedimentary and igneous rocks Central part of the catchment Low <5 Local-scale fractured and weathered rock aquifers composed mostly of sandstone and siltstone with some dolerite. Variable bore yields (often <2 L/s) and variable water quality. Only likely to offer potential for small-scale (<0.25 GL/y) localised developments (i.e. mostly suited to stock and domestic water supplies) or as a conjunctive water resource in the outcropping area where fracturing and weathering is high. HYDROGEOLOGICAL UNIT LOCATION LEVEL OF KNOWLEDGE INDICATIVE SCALE OF RESOURCE (GL/y) † COMMENT Cambrian basalt Small patches in the south of the catchment Low <5 Local-scale fractured and weathered rock aquifers composed mostly of basalt and breccia. Variable bore yields (often <2 L/s) and variable water quality. Only likely to offer potential for small-scale (<0.25 GL/y) localised developments (i.e. mostly suited to stock and domestic water supplies) or as a conjunctive water resource in the outcropping/subcropping areas where fracturing and weathering are high. †Actual scale will depend upon government and community acceptance of impacts to GDEs and existing water users. 7.3.1 Cambrian Limestone Aquifer and Dook Creek Aquifer Groundwater is already widely used in parts of the Roper catchment for a variety of purposes. It offers year-round niche opportunities that are geographically distinct from surface water development opportunities. The two most extensive and productive groundwater systems in the Roper catchment, the regional-scale CLA hosted in the Cambrian limestone and the intermediatescale DCA hosted in the Proterozoic dolostone of the Dook Creek Formation, offer the most promising opportunities for potential future groundwater resource development and were the focus of targeted investigations in this assessment. Existing licensed groundwater extraction in the CLA totals 32 GL/year, with 24 GL/year allocated in the vicinity of Mataranka. However, actual groundwater use is less. There is currently very little development of groundwater from the DCA other than stock and domestic bores and there is no water allocation plan. Assuming full use of existing groundwater licences in the CLA, groundwater discharge from the CLA to the Roper River near Mataranka was modelled to reduce by 9% by about the year 2070. With appropriately sited borefields, it is estimated from numerical modelling that between 35 and 105 GL/year (~3 to 10% of recharge to the entire CLA) could potentially be extracted from the CLA in the vicinity of, and to the south of, Larrimah (i.e. groundwater extraction occurring between 60 and 160 km from existing groundwater users and ecologically and culturally important GDEs near Mataranka, depending upon community and government acceptance of potential impacts on groundwater-dependent ecosystems and existing groundwater users. This is the unconfined part of the CLA in the southern Daly Basin near Larrimah south to the northern Georgina Basin near Daly Waters where the aquifer has a sufficient saturated thickness (i.e. >20 m). The CLA in the southern Daly Basin west of Larrimah and south-west to the northern Wiso Basin has a thin (i.e. <20 m) saturated thickness and historical drilling has shown a low probability for installing productive (i.e. >30 L/second) high-yielding bores. This part of the CLA in the Roper catchment is therefore not likely to be suitable for groundwater-based irrigation. Between 6 and 18 GL/year could potentially be extracted from the western unconfined part of the DCA, west of the Central Arnhem Road between Flying Fox Creek and Bulman, depending upon community and government acceptance of potential impacts on groundwater-dependent ecosystems and existing groundwater users. Due to the long-time lags associated with groundwater flow over long distances, the additional hypothetical extractions in the CLA result in only a further 2% reduction in modelled groundwater discharge to the Roper River near Mataranka by about the year 2070. However, the modelled reduction in groundwater levels ranges from about 12 m at the centre of the hypothetical developments to 0.5 m up to 110 km away by about 2070. Furthermore, numerical modelling incorporating climate variability, suggests that variations in climate are likely to have a bigger impact on the water balance of both the CLA and DCA which is an important consideration for managing these water resources into the future. Ultimately, the location and scale of actual future groundwater development will depend upon government and community acceptance of hydraulic impacts (i.e. changes in condition) to GDEs and water availability to existing groundwater users. Furthermore, any proponent seeking a groundwater license will most likely be required to undertake a hydrogeological assessment and ensure their proposed extraction meets licensing conditions in relation to groundwater drawdown. 7.3.2 Opportunities from other aquifers The less extensive but productive aquifers of the Yalwarra Volcanics, Knuckey Formation and Mount Birch Sandstone hosted in the Proterozoic dolostone and sandstone on the southern side of the Roper River may also offer potential for intermediate-scale opportunities for future groundwater resource development. However, these aquifers were not the subject of targeted investigations and exhibit only sparse hydrogeological information (lithology, water quality and bore yield) which highlights their potential for future development. Further drilling and pumping test investigations would be required to confirm their key attributes. Other aquifers in the catchment – including those hosted in the Proterozoic sedimentary and igneous rocks of the Roper River and Katherine River groups, the Cambrian basalt of the APV, and the Cretaceous sandstone – host localised groundwater systems of variable water quality and bores yield and therefore only offer potential as localised or conjunctive water resources. Figure 7-5 Hydrogeological units with potential for future groundwater resource development To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The extent of the surficial Cretaceous to Quaternary rocks and sediments is shown in the lower right inset. The right inset also shows the entire spatial extent of the Cambrian limestone and Proterozoic dolostone outside the Roper catchment. 7.4 Constraints on potential future groundwater resource development 7.4.1 Cambrian Limestone Aquifer Potential constraints on larger-scale (i.e. >5 GL/year) potential future groundwater resource development opportunities for groundwater-based irrigation from the CLA in the Roper catchment include: • being constrained to areas across the CLA where water is currently available for consumptive use such as those more than 50 km south of Mataranka • being constrained to areas where the saturated thickness of the aquifer is sufficient (i.e. >20 m) to support the installation of production bores (thus, mostly in the southern Daly Basin and northern Georgina Basin around and south of Larrimah) • installation of successful productive high-yielding (i.e. bore yields >30 L/second) bores may require the drilling of multiple investigations holes to identify productive parts of the aquifer which can create uncertainty in the cost of drilling programs • being constrained to areas where the depth to groundwater is less than or equal to 100 mBGL depending on the value of irrigated crops being grown, which may make bores prohibitively expensive in southern parts of the Roper catchment (where the depth is greater than 100 m BGL) when growing low-value crops • the need to consider crop selection and irrigation of fine textured soils across parts of the CLA in the southern Daly Basin and northern Georgina Basin as some areas exhibit slightly brackish groundwater (i.e. 1000 to 3000 mg/L TDS) which can degrade crops depending on their specific salt tolerance as well as cause soil structure decline such as dispersion • the need to maintain water quality across the aquifer given some parts of the CLA are brackish and increases in irrigated agriculture have the potential to increase salinity of the aquifer over time • the need to maintain water availability to all groundwater users and groundwater-dependent ecosystems as future demand for water increases and climate variability has a large influence on the aquifer water balance. The latter reflected by natural variations in recharge to the aquifer under the current and historical climate which is important consideration for adaptive management on the aquifer into the future. 7.4.2 Dook Creek Aquifer Potential constraints on moderate-scale (i.e. 1 to 3 GL/year) potential future groundwater resource development opportunities for groundwater-based irrigation from the DCA in the Roper catchment include: • being constrained to areas across the DCA at sufficient distances away from Flying Fox Creek and the Wilton River as they only exhibit minor flows (i.e. <0.7 m3/s) • the need to better characterise other potential areas of groundwater–surface water interactions as identified, as these are likely to be a key constraint on future groundwater resource development in certain locations • being constrained to areas across the DCA away from Bulman, Weemol and numerous other culturally important outstations (Momob, Bodeidei Camp, Morbon/Blue Water) • the need to better characterise the water balance, saturated thickness, groundwater levels and hydraulic properties of the aquifer, as the DCA is currently data sparse across large areas • installation of successful productive high-yielding (i.e. bore yields >30 L/second) bores may require the drilling of multiple investigations holes to identify productive parts of the aquifer which can create uncertainty in the cost of drilling programs • the need to maintain water availability to all groundwater users and groundwater-dependent ecosystems as future demand for water increases and climate variability has a large influence on the aquifer water balance. The latter reflected by natural variations in recharge to the aquifer under the current and historical climate which is important consideration for adaptive management on the aquifer into the future. 7.4.3 Other aquifers Key constraints on further potential future groundwater resource development opportunities for groundwater-based irrigation from the less extensive but productive sandstone aquifers of the Yalwarra Volcanics, Knuckey Formation and Mount Birch Sandstone include the need to identify and map productive aquifers hosted in these hydrogeological units on the southern side of the Roper River south of Ngukurr. While existing geological mapping and sparse hydrogeological data indicate their potential as a promising groundwater resource (i.e. host fresh groundwater, indicative bore yields are >10L/second) further drilling and pumping test investigations would be required. These investigations would confirm the aquifers’ spatial extent, saturated thickness, water quality and potential to yield sufficient water to support irrigation without depleting the aquifer or inducing leakage from the Roper River, which also hosts saline water in its lower reaches from large tidal influences. 8 Summary and conclusions 8.1 Opportunities and constraints for future groundwater development The regional desktop assessment of all available hydrogeological data proved useful as a screening tool for identifying potential groundwater resource development opportunities across the catchment. The unconfined parts of the CLA (the portion of the aquifer in the southern Daly Basin near Larrimah south to the northern Georgina Basin near Daly Waters) and DCA (the western portion of the aquifer west of the Central Arnhem Road between Flying Fox Creek and the Wilton River) have been identified as providing the greatest potential opportunities for future groundwater resource development across the Roper catchment. However, the nature and scale of these opportunities will ultimately depend upon community and government acceptance of impacts to GDEs and existing water users. These findings have been confirmed by targeted investigations that demonstrate the specified areas of the unconfined parts of both aquifers: • have a large spatial extent that coincides with portions of land suitable for agricultural intensification, making it accessible to many stakeholders in areas away from existing groundwater users and GDEs • have depths to groundwater of less than 100 mBGL across large areas, thereby exhibiting economical pumping costs • can be intersected with drilling at economical depths in many places • have predominantly good-quality groundwater that is suitable for a wide range of purposes, noting some parts of the CLA host slightly brackish groundwater • have potential to support multiple dispersed groundwater developments – large (5 to 10 GL/year) across the CLA and small to moderate (1 to 3 GL/year) across the DCA Furthermore, the assessment has further validated and refined the hydrogeological conceptual models for both aquifers, which include: • better characterising the key vertical and horizontal geological controls on groundwater discharge from the CLA and their influence on the position of Elsey and Salt creeks due to structural highs of the crystalline basement of the McArthur Basin beneath the regional groundwater discharge zone near Mataranka • better characterising the different spatial sources of discharge from the CLA in the regional groundwater discharge zone near Mataranka, an important consideration when evaluating potential changes in the water balance for the CLA due to climate variability or groundwater extraction • identifying other potential new areas of groundwater discharge that could inform further future investigations, such as of G9055086 spring in Hot Springs Valley to confirm its deeper source of groundwater • further confirming that the extensive spatial coverage, thickness and variable hydraulic properties of the two karstic groundwater systems (CLA and DCA) significantly influence the time lags for hydrological impacts of both climate variability and groundwater extraction to propagate through each system • identifying that the potential impact of climate variability on water resources in the Roper catchment is more significant than that of current groundwater extraction due to its influence on the spatial and temporal variability in groundwater recharge and discharge of both karstic groundwater systems. The latter two dot points are important considerations for managing the groundwater resources of both aquifers into the future. 8.2 Potential options for future work 8.2.1 Cambrian Limestone Aquifer The CLA has been the subject of multiple hydrogeological investigations over the past two decades, particularly the last decade. To fill key remaining knowledge gaps requires: (i) improved mapping and quantification of evapotranspiration disaggregated into evapotranspiration sourced from groundwater, rainfall and surface water from the regional groundwater discharge zone around Mataranka, (ii) further identification and mapping of the occurrence of phreatophytes and the application of suitable remotely sensed or other techniques for monitoring the temporal changes in their condition (i.e. to be used to establish environmental water requirements and thresholds for maintaining specified ecological conditions), (iii) more spatial and temporal hydraulic head data for areas greater than 50 km south of Mataranka which is likely to occur should further groundwater licences be granted across this southern region of the CLA in the southern Daly Basin and northern Georgina Basin, and, (iv) continuation of the hydrographic monitoring of spring flows and stream flows in the regional groundwater discharge zone near Mataranka. Improved mapping and quantification of evapotranspiration from the regional groundwater discharge zone would help better quantify this important component of the water balance for the CLA and assist in further constraining its overall water balance. This is important when considering numerical modelling has identified that future increases in groundwater extraction across the aquifer result in changes in reductions in the evapotranspiration discharge component of the waster balance. Identifying suitable remotely sensed or other techniques for monitoring the temporal changes in the condition of phreatophytes will provide a monitoring tool for this type of GDE’s condition and may assist in better characterising and quantifying their water requirements. Currently little spatial and temporal hydraulic head data for the CLA exists in the northern Wiso and northern Georgina basins, additional data would assist in getting a further baseline understanding of seasonal head responses to high inter-annual variability in climate and potential future groundwater extraction. This may come from proponents seeking groundwater licences as they may be required to undertake a hydrogeological assessment requiring the installation of monitoring bores to ensure their extraction is meeting licensing conditions (i.e. monitoring groundwater drawdown). In addition, better spatial and temporal head data would further validate or refine existing potentiometric surfaces for the CLA and the directions of regional groundwater flow, as well as further containing the calibration of the DR2 FEFLOW model. The hydrographic surveys of spring and stream flows have provided an important dataset to map the spatial variation in discharge from the CLA over time. This will be important in the future for monitoring the spatial and temporal cumulative hydraulic impacts from future climate variations and groundwater extraction, as well as providing another input datasets for constraining the calibration of the DR2 model. 8.2.2 Dook Creek Aquifer The DCA is yet to be developed and therefore is sparse in terms of key hydrogeological data (i.e. spatial and temporal groundwater-level data, water quality data, aquifer hydraulic properties and aquifer extent and geometry). To fill key remaining knowledge gaps requires: (i) improved hydrogeological mapping from drilling, pumping tests, water quality testing and groundwater-level monitoring, and (ii) further identification and characterisation of groundwater discharge areas along different reaches of Flying Fox Creek and the Mainoru and Wilton rivers, guided by areas identified in this assessment, with remotely sensed data. Improved hydrogeological mapping of the DCA will help: (i) confirm the potential prospectivity of the groundwater resource, (ii) help validate or refine the existing hydrogeological conceptual model, and (iii) potentially further constrain the water balance for the aquifer. Better characterising groundwater discharge areas across the DCA will also help: (i) further validate or refine the current conceptual model, (ii) potentially further constrain the water balance, and (iii) better understand potential constraints on future groundwater resource development in specific locations. 8.2.3 Nathan Group aquifers Further drilling and pumping test investigations would be required to determine whether any productive sandstone aquifers are hosted in the Yalwarra Volcanics, Knuckey Formation and Mount Birch Sandstone of the Nathan Group on the southern side of the Roper River south of Ngukurr, and to map them. These investigations would help confirm the aquifers’ spatial extent, saturated thickness, water quality and potential to yield sufficient water to support irrigation without depleting the aquifer or inducing leakage of saline water from the Roper River. References Abbott ST, Sweet IP, Plumb KA, Young DN, Cutovinos A, Ferenczi PA, Brakel AT and Pietsch BA (2001) Roper Region: Urapunga and Roper River Special, Northern Territory (2nd edn). 1:250,000 geological map series explanatory notes, SD 53–10, 11. Northern Territory Geological Survey and Geoscience Australia (National Geoscience Mapping Accord). Hyperlink to: Roper Region: Urapunga and Roper River Special, Northern Territory (2nd edn). 1:250,000 geological map series explanatory notes, SD 53–10, 11 . Acworth R, Halloran L, Rau G, Cuthbert M and Bernardi T (2016) An objective frequency domain method for quantifying confined aquifer compressible storage using Earth and atmospheric tides. Geophysical Research Letters 43(22), 8. Ahmad M, Dunster JN and Munson TJ (2013) McArthur Basin. In: Ahmad M and Munson TJ (compilers) Geology and mineral resources of the Northern Territory. Northern Territory Geological Survey, Special Publication 5. Hyperlink to: McArthur Basin. In: Ahmad M and Munson TJ (compilers) Geology and mineral resources of the Northern Territory . Alcalá FJ, Cantón Y, Contreras S, Were A, Serrano-Ortiz P, Puigdefábregas J, Solé-Benet A, Custodio E and Domingo F (2011) Diffuse and concentrated recharge evaluation using physical and tracer techniques: results from a semiarid carbonate massif aquifer in southeastern Spain. Environmental Earth Sciences 62(3), 541–557. DOI: 10.1007/s12665-010-0546-y. Amery T and Tickell S (2022) Beetaloo Sub-basin SREBA Water Studies: aquifer test data review for overlying primary aquifers. Northern Territory Government, Department of Environment, Parks and Water Security, Water Resources Division, Palmerston. Hyperlink to: Beetaloo Sub-basin SREBA Water Studies: aquifer test data review for overlying primary aquifers . Anderson VG (1945) Some effects of atmospheric evaporation and transpiration on the composition of natural water in Australia (continued). 4. Underground waters in riverless areas. J. Aust. Chem. Inst 12, 83–98. Appelo CAJ and Postma D (1996) Geochemistry, groundwater and pollution, 2nd edn. AA Balkema Publishers, Leiden, London, New York, Philadelphia, Singapore. Augustine P (1958) Beswick Station – report on water resources DLH-581. Northern Territory Administration, Darwin. Hyperlink to: Beswick Station – report on water resources DLH-581 . Augustine P (1962) Moroak station – WSDO – application for advance. Water Resources Branch. Northern Territory Administration, Darwin. Hyperlink to: Moroak station – WSDO – application for advance . Australian Bureau of Statistics (2021) Australian Bureau of Statistics, 2021 Census data. Australia’s national statistical agency, Australian Government. Baker A, Moretti A, Jolly P, Tyson P and Warren, T and Associates (1992) Ngukurr water supply evaluation. Power and Water Authority and Water Resources Division, Northern Territory Government, Darwin. Hyperlink to: Ngukurr water supply evaluation . Ball N, Booth E, McLeod B and Vornicu DE (2019) Historical pumping test data collation and reinterpretation for the Northern Territory. Department of Environment and Natural Resources, Northern Territory, Palmerston, NT. Hyperlink to: Historical pumping test data collation and reinterpretation for the Northern Territory . Barbeta A, Jones SP, Clavé L, Wingate L, Gimeno TE, Fréjaville B, Wohl S and Ogée J (2019) Unexplained hydrogen isotope offsets complicate the identification and quantification of tree water sources in a riparian forest. Hydrology and Earth System Sciences 23(4), 2129– 2146. DOI: 10.5194/hess-23-2129-2019. Barrow CJ (1992) World atlas of desertification (United Nations Environment Programme), edited by N. Middleton and D. S. G. Thomas. Edward Arnold, London, 1992. isbn 0 340 55512 2, £89.50 (hardback), ix + 69 pp. Land Degradation and Development 3(4), 249–249. DOI: 10.1002/ldr.3400030407. Bazuhair AS and Wood WW (1996) Chloride mass-balance method for estimating ground water recharge in arid areas – examples from Western Saudi Arabia. Journal of Hydrology 186(1– 4), 153–159. Bendat J and Piersol A (1993) Engineering applications of correlation and spectral analysis. John Wiley and Sons, New York, USA. BoM (2012) Australian hydrological geospatial fabric (Geofabric): Product guide Version 2.1. Bureau of Meteorology, Australian Government. BoM (2019) Normalized Difference Vegetation Index (NDVI): AVHRR, without atmospheric correction, Australia coverage. Bureau of Meteorology. Viewed April 2019. ( Hyperlink to: Normalized Difference Vegetation Index (NDVI): AVHRR, without atmospheric correction, Australia coverage ). BoM (2020) National Groundwater Information System: a spatial database maintained by the Bureau of Meteorology for GIS specialists that contains a range of groundwater information submitted by states and territories. Data licensed under a Creative Commons Australia Attribution Licence. ©Commonwealth of Australia (Bureau of Meteorology) License. ( Hyperlink to: National Groundwater Information System ). Brooks B and Kneebone D (1963) Surface water reconnaissance of Roper and Wilton river area. Water Resources Branch. Northern Territory Administration, Darwin. Hyperlink to: Surface water reconnaissance of Roper and Wilton river area . Bruwer Q and Tickell S (2015) Daly Basin groundwater resource assessment – North Mataranka to Daly Waters in the Northern Territory of Australia, December 2015. Hyperlink to: Daly Basin groundwater resource assessment – North Mataranka to Daly Waters in the Northern Territory of Australia, December 2015 . Budyko MI (1974) Climate and life. Academic Press, New York and London, Hyperlink to: Climate and life . Bullister JL (2017) Atmospheric histories (1765–2015) for CFC-11, CFC-12, CFC-113, CCl4, SF6 and N2O. Oak Ridge, Tennessee. Hyperlink to: Atmospheric histories (1765–2015) for CFC-11, CFC-12, CFC-113, CCl4, SF6 and N2O . Burgess J, McGrath N, Andrews K, Wright A and Hill JV (2015) Soil and land suitability assessment for irrigated agriculture in the Larrimah Area, Sturt Plateau. Department of Land Resource Management, Land Assessment Branch, Northern Territory Government, Darwin. Hyperlink to: Soil and land suitability assessment for irrigated agriculture in the Larrimah Area, Sturt Plateau . Busenberg E and Plummer LN (2000) Dating young groundwater with sulfur hexafluoride: natural and anthropogenic sources of sulfur hexafluoride. Water Resources Research 36(10), 3011– 3030. DOI: 10.1029/2000WR900151. Busenberg E and Plummer N (1992) Use of chlorofluorocarbons (CCl3F and CCl2F2) as hydrologic tracers and age‐dating tools: the alluvium and terrace system of central Oklahoma. Water Resources Research 28(9), 2257–2283. DOI: 10.1029/92WR01263. Cartwright I, Weaver TR, Stone D and Reid M (2007) Constraining modern and historical recharge from bore hydrographs, H-3, C-14 and chloride concentrations: applications to dual-porosity aquifers in dryland salinity areas, Murray Basin, Australia. Journal of Hydrology 332(1–2), 69– 92. Clark ID and Fritz P (1997) Environmental isotopes in hydrogeology (1st edn). CRC Press, Boca Raton. DOI: 10.1201/9781482242911. Cook PG (2003) Inferring groundwater inflow to the Roper River (NT) from environmental tracers. CSIRO Land and Water, Darwin. Cook PG and Böhlke J-K (2000) Determining timescales for groundwater flow and solute transport. In: Cook PG and Herczeg AL (eds), Environmental tracers in subsurface hydrology. Springer US, Boston, MA. Cook PG, Walker GR and Jolly ID (1989) Spatial variability of groundwater recharge in a semiarid region. Journal of Hydrology 111, 195–212. Cooper H and Jacob C (1946) A generalized graphical method for evaluating formation constants and summarizing well‐field history. Eos, Transactions, American Geophysical Union 27(4), 9. Coxon G, Freer J, Westerberg IK, Wagener T, Woods R and Smith PJ (2015) A novel framework for discharge uncertainty quantification applied to 500 UK gauging stations. Water Resources Research 51(7), 5531–5546. DOI: 10.1002/2014WR016532. Cressie N (1990) The origins of kriging. Mathematical Geology 22(3), 239–252. DOI: 10.1007/BF00889887. Crosbie R, Pickett T, Mpelasoka F, Hodgson G, Charles S and Barron O (2013) An assessment of the climate change impacts on groundwater recharge at a continental scale using a probabilistic approach with an ensemble of GCMs. Climatic Change 117, 41–53. DOI: 10.1007/s10584- 012-0558-6. Crosbie R, Raiber M, Cui T and Viney N (2015) Blending field observations and AWRA outputs for groundwater recharge estimation in a coal basin in eastern Australia. In: MODSIM2015, 21st International Congress on Modelling and Simulation. Modelling and Simulation Society of Australia and New Zealand, December 2015, 2033–2039. http://www.mssanz.org.au/modsim2015/L2032/crosbie.pdf. Crosbie RS, Jolly ID, Leaney FW and Petheram C (2010) Can the dataset of field based recharge estimates in Australia be used to predict recharge in data-poor areas? Hydrology and Earth System Sciences 14(10), 2023–2038. DOI: 10.5194/hess-14-2023-2010. Crosbie RS, Peeters LJM, Herron N, McVicar TR and Herr A (2018) Estimating groundwater recharge and its associated uncertainty: use of regression kriging and the chloride mass balance method. Journal of Hydrology 561, 1063–1080. DOI: 10.1016/j.jhydrol.2017.08.003. Crosbie RS and Rachakonda PK (2021) Constraining probabilistic chloride mass-balance recharge estimates using baseflow and remotely sensed evapotranspiration: the Cambrian Limestone Aquifer in northern Australia. Hydrogeology Journal 29(4), 1399–1419. DOI: 10.1007/s10040- 021-02323-1. Crosbie RS, Raiber M, Wilkins A, Dawes W, Louth-Robins T and Gao L (2022) Quantifying diffuse recharge to groundwater systems of the Great Artesian Basin, the NSW coalfields and surrounds. CSIRO Report EP2022–2355. de la Casa J, Barbeta A, Rodríguez-Uña A, Wingate L, Ogée J and Gimeno TE (2022) Isotopic offsets between bulk plant water and its sources are larger in cool and wet environments. Hydrology and Earth System Sciences 26, 4125–4146. DOI: 10.5194/hess-26-4125-2022. Department of Agriculture, Water and the Environment (2019) Directory of Important Wetlands in Australia – Information sheet. Limmen Bight (Port Roper) Tidal Wetlands System – NT007. Viewed 30 March 2022. https://www.environment.gov.au/cgi-bin/wetlands/report.pl. Department of Environment and Natural Resources (2016) Bore construction Darwin rural area most common in dolomite country. Department of Natural Resources and Environment. Data downloaded in September 2022. Department of Environment and Natural Resources (2017) Background Brief, Water Allocation Plan for the Tindall Limestone Aquifer, Mataranka – Daly Waters. Department of Environment and Natural Resources, Northern Territory Government, Darwin. https://depws.nt.gov.au/__data/assets/pdf_file/0005/438008/Background-Brief-Tindall- Limestone-Aquifer-Mataranka-to-Daly-Waters.pdf. Department of Environment Parks and Water Security (2008) Northern Territory Groundwater – Aquifers 1:2,000,000. ( Hyperlink to: Northern Territory Groundwater – Aquifers 1:2,000,000 ). Department of Environment Parks and Water Security (2014a) Digital groundwater database for the Northern Territory, supplied by the Department of Environment, Parks and Water Security, Copyright – Northern Territory of Australia. Data downloaded from the Northern Territory Government (NTG) Open Data Portal in September 2021. ( Hyperlink to: Digital groundwater database for the Northern Territory ). Department of Environment Parks and Water Security (2014b) Sinkholes of the Northern Territory. Northern Territory Department of Environment Parks and Water Security. ( Hyperlink to: Sinkholes of the Northern Territory ). Department of Environment Parks and Water Security (2014c) Springs of the Northern Territory. Northern Territory Department of Environment Parks and Water Security. ( Hyperlink to: Springs of the Northern Territory ). Department of Environment Parks and Water Security (2017) Land use mapping project of the Northern Territory, 2016–2022 (LUMP). Northern Territory Department of Environment Parks and Water Security ( Hyperlink to: Land use mapping project of the Northern Territory, 2016–2022 (LUMP) ). Department of Environment Parks and Water Security (2018) Water licensing – groundwater extraction points and licensed entitlements granted by the Controller of Water Resources. Northern Territory Department of Environment Parks and Water Security. Hyperlink to: Water licensing – groundwater extraction points and licensed entitlements granted by the Controller of Water Resources . Department of Environment Parks and Water Security (2019a) Water allocation plans in the Northern Territory. Northern Territory Government, Darwin. ( Hyperlink to: Water allocation plans in the Northern Territory ). Department of Environment Parks and Water Security (2019b) Water control districts in the Northern Territory. ( Hyperlink to: Water control districts in the Northern Territory , download). Department of Environment Parks and Water Security (2019c) Water management zones in the Northern Territory. Department of Environment, Parks and Water Security. ( Hyperlink to: Water management zones in the Northern Territory ). Department of Environment Parks and Water Security (2021) NR maps, the Department of Environment and Natural Resources web mapping tool for natural and cultural research data for the Northern Territory. (Data accessed January 2021, Hyperlink to: NR maps, the Department of Environment and Natural Resources web mapping tool for natural and cultural research data for the Northern Territory ). Department of Industry Tourism and Trade (2000) Northern Territory geological survey – drillhole collars. (The database is a capture, storage and maintenance of drillhole collars which forms part of the NTGS Geochemical database, Hyperlink to: Northern Territory geological survey – drillhole collars , download). Department of Industry Tourism and Trade (2010) Northern Territory petroleum wells. (Contains details of industry petroleum exploration and production wells drilled onshore and in Northern Territory administered waters since 1959, Hyperlink to: Northern Territory petroleum wells , download). Department of Industry Tourism and Trade (2014) Northern Territory geological map (Interp) 2500K. Northern Territory Department of Industry, Tourism and Trade. Northern Territory Department of Industry Trade and Tourism. Hyperlink to: Northern Territory geological map (Interp) 2500K . Department of the Environment and Energy (2010) A Directory of Important Wetlands in Australia (DIWA) spatial database. CSIRO, Canberra. Viewed 08 December 2018, Hyperlink to: A Directory of Important Wetlands in Australia (DIWA) spatial database . Department of Tourism Industry and Trade (2010) Northern Territory geological faults 2500K. ( Hyperlink to: Northern Territory geological faults 2500K , download). Deslandes A, Gerber C, Lamontagne S, Wilske C and Suckow A (2019) Environmental tracers in the Beetaloo Basin: aquifer and groundwater characterization. Land and Water, CSIRO, Australia. Hyperlink to: Environmental tracers in the Beetaloo Basin: aquifer and groundwater characterization . DHI (2008) Mike SHE User Manual. Danish Hydraulic Institute, Denmark. Diersch H-JG (2008) FEFLOW® 5.4 Users Manual. WASY – Institute for Water Resources Planning and System Research, Berlin, Germany. Dogramaci S, Skrzypek G, Dodson W and Grierson PF (2012) Stable isotope and hydrochemical evolution of groundwater in the semi-arid Hamersley Basin of subtropical northwest Australia. Journal of Hydrology 475, 281–293. DOI: 10.1016/j.jhydrol.2012.10.004. Doherty J (2003) Ground water model calibration using pilot points and regularization. Ground Water 41(2), 170–177. Doherty J (2004) PEST Model-Independent Parameter Estimation User Manual: 5th edn. Watermark Numerical Computing, Brisbane, Australia. Doherty J (2015) Calibration and uncertainty analysis for complex environmental models. Watermark Numerical Computing, Brisbane, Australia. Hyperlink to: Calibration and uncertainty analysis for complex environmental models . Doherty J (2020) A simple lumped parameter model for unsaturated zone processes. Watermark Numerical Computing, Brisbane, Australia. Domenico P and Schwartz F (1998) Physical and chemical hydrogeology. John Wiley and Sons, New York, USA. Doody TM, Barron OV, Dowsley K, Emelyanova I, Fawcett J, Overton IC, Pritchard JL, Van Dijk AIJM and Warren G (2017) Continental mapping of groundwater dependent ecosystems: a methodological framework to integrate diverse data and expert opinion. Journal of Hydrology: Regional Studies 10, 61–81. DOI: 10.1016/j.ejrh.2017.01.003. Drysdale RN, Taylor MP and Ihlenfeld C (2002) Factors controlling the chemical evolution of travertine-depositing rivers of the Barkly karst, northern Australia. Hydrological Processes 16(15), 2941–2962. DOI: 10.1002/hyp.1078. Duffield G (2007) AQTESOLV for Windows Version 4.5 User's Guide. Reston, VA, USA. Duvert C, Canham CA, Barbeta A, Alvarez Cortes D, Chandler L, Harford AJ, Leggett A, Setterfield SA, Humphrey CL and Hutley LB (2022) Deuterium depletion in xylem water and soil isotopic effects complicate the assessment of riparian tree water sources in the seasonal tropics. Ecohydrology 15(6), e2383. DOI: 10.1002/eco.2383. Duvert C, Hutley LB, Lamontagne S, Bourke A, Alvarez D, Irvine D and Taylor A (2023) Tree water sourcing at the Mataranka Springs Complex. A report from the CSIRO Roper River Water Resource Assessment to the Government of Australia. CSIRO, Australia, csiro:EP2023-4375. Efron B and Tibshirani RJ (1994) An introduction to the bootstrap. CRC press. ELA (Eco Logical Australia) (2022) Beetaloo Sub-basin SREBA: water quality and quantity baseline summary report. Northern Territory Government, Department of Environment, Parks and Water Security, Water Resources Division, Palmerston. Hyperlink to: Beetaloo Sub-basin SREBA: water quality and quantity baseline summary report . Environment Australia (2001) A directory of important wetlands in Australia, 3rd edn, Environment Australia, Canberra. Eriksson E (1985) Principles and applications of hydrochemistry. Chapman and Hall Ltd, New York. Evans TJ, Radke BM, Martinez J, Buchanan S, Cook SB, Raiber M, Ransley TR, Lai ÉCS, Skeers N, Woods M, Evenden C, Cassel R and Dunn B (2020) Hydrogeology of the Beetaloo GBA region. Technical appendix for Geological and Bioregional Assessment Program: Stage 2. Department of the Environment and Energy, Bureau of Meteorology, CSIRO and Geoscience Australia. Hyperlink to: Hydrogeology of the Beetaloo GBA region . Fallon SJ, Fifield LK and Chappell JM (2010) The next chapter in radiocarbon dating at the Australian National University: status report on the single stage AMS. Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms 268(7), 898–901. DOI: 10.1016/j.nimb.2009.10.059. Faulks JJ (2001) Roper River Catchment: An assessment of the physical and ecological condition of the Roper River and its major tributaries. Technical Report No. 36/2001. Natural Resources Division. Department of Lands, Planning and Environment. Katherine, NT. Fontes J-C and Garnier J-M (1979) Determination of the initial 14C activity of the total dissolved carbon: A review of the existing models and a new approach. Water Resources Research 15(2), 399–413. DOI: 10.1029/WR015i002p00399. Frery E, Byrne C, Crosbie R, Deslandes A, Evans T, Gerber C, Huddlestone-Holmes C, Markov J, Martinez J, Raiber M, Turnadge C, Suckow A and Wilske C (2022) Fault-related fluid flow implications for unconventional hydrocarbon development, Beetaloo Sub-Basin (Northern Territory, Australia). Geosciences 12(1), 37. Fulton S and Knapton A (2015) Beetaloo Basin hydrogeological assessment. CloudGMS. Hyperlink to: Beetaloo Basin hydrogeological assessment . Fulton S and Zaar U (2009) Gulf water study: Dunmarra and Hodgson river region. Deparment of Natural Resources, Environment, the Arts and Sport and Greening the Territory, Darwin. Hyperlink to: Gulf water study: Dunmarra and Hodgson river region . GA (2021) DEA fractional cover percentiles (Landsat). Geoscience Australia. ( Hyperlink to: DEA fractional cover percentiles (Landsat) ). Gallant J, Wilson N, Dowling T, Read A and Inskeep CS-dSDEMVRGA, Canberra (2011) SRTMderived 1 Second digital elevation models version 1.0. Record 1. Geoscience Australia, Canberra. https://pid.geoscience.gov.au/dataset/ga/72759. Gardner WP, Harrington GA, Solomon DK and Cook PG (2011) Using terrigenic 4He to identify and quantify regional groundwater discharge to streams. Water Resources Research 47(6). DOI: 10.1029/2010WR010276. Geological and Bioregional Assessment Program (2021) Fact sheet 17: Hydrochemistry, 18O, 2H and radon time series of the Mataranka Springs. Geological and Bioregional Assessment Program, Department of Agriculture, Water and the Environment, Bureau of Meteorology, CSIRO and Geoscience Australia. Hyperlink to: Fact sheet 17: Hydrochemistry, 18O, 2H and radon time series of the Mataranka Springs . Geological Society of London (2012) A hole lot of trouble. Geological Society of London. George D (2001) Water resources of the Katherine Region and South West Arnhem Land. Natural Resources Division, Department of Lands, Planning and Environment, Darwin. Hyperlink to: Water resources of the Katherine Region and South West Arnhem Land . George D and Jolly P (2002) Water resources map of the Katherine Region and South West Arnhem Land. Department of Lands, Planning and Environment. Natural Resources Division, Northern Territory Government, Darwin. ( Hyperlink to: Water resources map of the Katherine Region and South West Arnhem Land ). Geoscience Australia and Australian Stratigraphy Commission (2017) Australian Stratigraphic Units Database. ( Hyperlink to: Australian Stratigraphic Units Database ). Glass LM, Ahmad M and Dunster JN (2013) Kalkarindji Province. In: Ahmad M and Munson TJ (compilers) Geology and mineral resources of the Northern Territory. Northern Territory Geological Survey, Special Publication 5. Hyperlink to: Kalkarindji Province. In: Ahmad M and Munson TJ (compilers) Geology and mineral resources of the Northern Territory . Glenn EP, Doody TM, Guerschman JP, Huete AR, King EA, McVicar TR, Van Dijk AIJM, Van Niel TG, Yebra M and Zhang Y (2011) Actual evapotranspiration estimation by ground and remote sensing methods: the Australian experience. Hydrological Processes 25(26), 4103–4116. DOI: 10.1002/hyp.8391. Godwin H (1962) Half-life of radiocarbon. Nature 195(4845), 984–984. DOI: 10.1038/195984a0. Grayson R, Argent R, Nathan R, McMahon T and Mein R (1996) Hydrological recipes: estimation techniques in Australian hydrology. CRC Catchment Hydrology, Melbourne. Gringarten A (2008) From straight lines to deconvolution: the evolution of the state of the art in well test analysis. SPE Reservoir Evaluation and Engineering 11(1), 22. Groves H (2021) Groundwater extraction licence resource assessment: NGU1 and NGU2 (Ngukurr Community). Department of Environment, Parks and Water Security (Water Resources Division), Palmerston, Northern Territory. Grundy MJ, Rossel RAV, Searle RD, Wilson PL, Chen C and Gregory LJ (2015) Soil and landscape grid of Australia. Soil Research 53(8), 835–844. DOI: 10.1071/SR15191. Guerschman JP, McVicar TR, Vleeshower J, Van Niel TG, Peña-Arancibia JL and Chen Y (2022) Estimating actual evapotranspiration at field-to-continent scales by calibrating the CMRSET algorithm with MODIS, VIIRS, Landsat and Sentinel-2 data. Journal of Hydrology 605, 127318. DOI: 10.1016/j.jhydrol.2021.127318. Guerschman JP, Warren G, Byrne G, Lymburner L, Mueller N and Van Dijk AI (2011) MODIS-based standing water detection for flood and large reservoir mapping: algorithm development and applications for the Australian continent. CSIRO. Haitjema HM and Mitchell-Bruker S (2005) Are water tables a subdued replica of the topography? Groundwater 43(6), 781–786. DOI: 10.1111/j.1745-6584.2005.00090.x. Hantush M (1960) Modification of the theory of leaky aquifers. Journal of Geophysical Research 65(11), 13. Harrington GA, Cook PG and Herczeg AL (2002) Spatial and temporal variability of ground water recharge in Central Australia: a tracer approach. Groundwater 40(5), 518–527. DOI: 10.1111/j.1745-6584.2002.tb02536.x. Harrington GA, Gardner WP and Munday TJ (2014) Tracking groundwater discharge to a large river using tracers and geophysics. Groundwater 52(6), 837–852. DOI: 10.1111/gwat.12124. Harrington GA and Herczeg AL (2003) The importance of silicate weathering of a sedimentary aquifer in arid Central Australia indicated by very high 87Sr/86Sr ratios. Chemical Geology 199(3), 281–292. DOI: 10.1016/S0009-2541(03)00128-1. Hartmann T and Wenzel H (1995) The HW95 tidal potential catalogue. Geophysical Research Letters 22(24), 4. Hengl T, Heuvelink GBM and Stein A (2004) A generic framework for spatial prediction of soil variables based on regression-kriging. Geoderma 120(1–2), 75–93. DOI: 10.1016/j.geoderma.2003.08.018. Howcroft W, Cartwright I, Fifield LK and Cendón D (2017) Differences in groundwater and chloride residence times in saline groundwater: the Barwon River Catchment of Southeast Australia. Chemical Geology 451, 154–168. Huddlestone-Holmes CR, Brandon C, Crosbie RS, Davies PJ, Doble R, Frery E, Kear J, Holland KL, O’Grady A, Pavey CR, Peeters LJM, Tetreault-Campbell S, Aghbelagh YB, Arjomand E, Bailey AHE, Barry K, Donohue R, Evans TJ, Evenden C, Flett D, Frery E, Gonzalez D, Herr A, Kasperczyk D, King D, Lawrence E, Lawson C, Li LT, MacFarlane C, Mokany K, Mallants D, Markov J, Martinez C, Martinez J, Marvanek SP, McVicar TR, Merrin LE, Murray J, Northover S, Raiber M, Reese B, Stewart SB, Turnadge C, Watson G, Woods M and Zhejun P (2021) Impact assessment for the Beetaloo GBA region. Geological and Bioregional Assessment Program: Stage 3 synthesis. Department of Agriculture, Water and the Environment, Bureau of Meteorology, CSIRO and Geoscience Australia, Australia. Hyperlink to: Impact assessment for the Beetaloo GBA region (download). Hughes J, Yang A, Marvanek S, Wang B, Petheram C and S P (2023) River model calibration and scenario analysis for the Roper catchment. A report from the CSIRO Roper River Water Resource Assessment to the Government of Australia. CSIRO, Australia. IAEA/WMO (2023) Global network of isotopes in precipitation. The GNIP Database. IAEA. Ingerson R and Pearson FJ (1964) Estimation of age and rate of motion of groundwater by the 14C method. In Miyake Y and Koyama T (eds) Recent research in the field of hydrosphere, atmosphere and nuclear geochemistry. Marusen, Tokyo, 263–283. Jacob C (1940) On the flow of water in an elastic artesian aquifer. Eos, Transactions, American Geophysical Union 21(2), 13. Jamieson M and Pidsley D (1992) Miniyeri groundwater resources investigation DLH-624 1991– 1992. Power and Water Authority, Darwin. Hyperlink to: Miniyeri groundwater resources investigation DLH-624 1991–1992 . Jolly P, Knapton A and Tickell S (2004) Water availability from the aquifer in the Tindall Limestone south of the Roper River. Natural Systems Division Conservation and Natural Resources Group, Department of Infrastructure, Planning and Environment. Hyperlink to: Water availability from the aquifer in the Tindall Limestone south of the Roper River . Jones DA, Wang W and Fawcett R (2009) High-quality spatial climate data-sets for Australia. Australian Meteorological and Oceanographic Journal 58(4), 233–248. Jurgens BC, Böhlke JK and Eberts SM (2012) TracerLPM (Version 1): An Excel® workbook for interpreting groundwater age distributions from environmental tracer data. Techniques and Methods 4–F3. US Geological Survey, US Geological Survey. Hyperlink to: TracerLPM (Version 1): An Excel® workbook for interpreting groundwater age distributions from environmental tracer data . Kalin RM (2000) Radiocarbon dating of groundwater systems. In: Cook PG and Herczeg AL (eds) Environmental tracers in subsurface hydrology. Springer US, Boston, MA. Kalma JD, McVicar TR and McCabe MF (2008) Estimating land surface evaporation: a review of methods using remotely sensed surface temperature data. Surveys in Geophysics 29(4–5), 421–469. Karp D (1984a) Bore completion – Hodgson Downs outstation DLH-624. Water Resource Division, Investigation Branch, Groundwater Section. Hyperlink to: Bore completion – Hodgson Downs outstation DLH-624 . Karp D (1984b) Bore completion – Hodgson River outstation DLH-600. Northern Territory Government. Hyperlink to: Bore completion – Hodgson River outstation DLH-600 . Karp D (2002) Land degradation associated with sinkhole development in the Katherine Region. Department of Infrastructure, Planning and Environment, Darwin. https://territorystories.nt.gov.au/10070/229127/0. Karp D (2008) Surface and groundwater interaction the Mataranka Area. Department of Natural Resources, Environment, the Arts and Sport. Report No. 17/2008D, Darwin, NT. Hyperlink to: Surface and groundwater interaction the Mataranka Area . Kerle E and Cruickshank S (2014) End of dry season stream flow measurements, Roper River, October 2014. Department of Land Resource Management, Northern Territory Government, Palmerston. Hyperlink to: End of dry season stream flow measurements, Roper River, October 2014 . Kim JH and Jackson RB (2012) A global analysis of groundwater recharge for vegetation, climate, and soils. Vadose Zone Journal 11(1). DOI: 10.2136/vzj2011.0021RA. King E, van Niel T, van Dijk A, Wang Z, Paget M, Raupach T, Guerschman J, Haverd V, McVicar T, Miltenberg I, Raupach M, Renzullo L and Zhang Y (2011) Actual evapotranspiration estimates for Australia intercomparison and evaluation. Water for a Healthy Country National Research Flagship, CSIRO, Canberra. Knapton A (2004) Modelling of water extraction at the Shenandoah Station, Georgina Basin and effects on base flows in the Roper River. Department of Infrastructure, Planning and Environment. Natural Resources Division, Northern Territory Government, Darwin. Hyperlink to: Modelling of water extraction at the Shenandoah Station, Georgina Basin and effects on base flows in the Roper River . Knapton A (2005) Preliminary groundwater modelling of the Oolloo Dolostone. Department Natural Resources, Environment and the Arts, Alice Springs. Knapton A (2006) Regional groundwater modelling of the Cambrian Limestone Aquifer system of the Wiso Basin, Georgina Basin and Daly Basin. Department of Natural Resources, Environment, the Arts and Sport, Water Resources Branch. Northern Territory Government, Alice Springs. Knapton A (2009a) Gulf water study: an integrated surface – groundwater model of the Roper River catchment, Northern Territory. Part B: MIKE11 surface water model. Department of Natural Resources, Environment, the Arts and Sport, Water Resources Branch. Northern Territory Government, Darwin. Hyperlink to: Gulf water study: an integrated surface – groundwater model of the Roper River catchment, Northern Territory. Part B: MIKE11 surface water model . Knapton A (2009b) Gulf water study: an integrated surface – groundwater model of the Roper River catchment, Northern Territory. Part A: Coupled surface – groundwater model. Department of Natural Resources, Environment, the Arts and Sport, Water Resources Branch, Northern Territory Government, Darwin. Hyperlink to: Gulf water study: an integrated surface – groundwater model of the Roper River catchment, Northern Territory. Part A: Coupled surface – groundwater model . Knapton A (2009c) Gulf water study: an integrated surface – groundwater model of the Roper River catchment, Northern Territory. Part C – FEFLOW groundwater model. Department of Natural Resources, Environment, the Arts and Sport Water Resources Branch, Technical Report No. 32/2009D. Hyperlink to: Gulf water study: an integrated surface – groundwater model of the Roper River catchment, Northern Territory. Part C – FEFLOW groundwater model . Knapton A (2020) Upgrade of the coupled model of the Cambrian Limestone Aquifer and Roper River systems. CloudGMS and Department of Environment, Parks and Water Security, Northern Territory Government, Palmerston. Hyperlink to: Upgrade of the coupled model of the Cambrian Limestone Aquifer and Roper River systems . Knapton A, Taylor AR, Petheram C and Crosbie RS (2023) An investigation into the effects of climate change and groundwater development scenarios on the water resources of the Roper catchment using two finite element groundwater flow models. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Kneebone D (1960) Roper River mission water resources DLH-633 investigation field trip. Water Resources Branch, Northern Territory Administration, Darwin. Hyperlink to: Roper River mission water resources DLH-633 investigation field trip . Krause CE, Newey V, Alger MJ and Lymburner L (2021) Mapping and monitoring the multi-decadal dynamics of Australia’s open waterbodies using Landsat. Remote Sensing 13(8), 1437. Kruse PD, Dunster JN and Munson TJ (2013) Georgina Basin. In: Ahmad M and Munson TJ (compilers) Geology and mineral resources of the Northern Territory. Northern Territory Geological Survey, Special Publication 5, Hyperlink to: Georgina Basin. In: Ahmad M and Munson TJ (compilers) Geology and mineral resources of the Northern Territory . Kruse PD and Munson TJ (2013a) Daly Basin. In: Ahmad and Munson TJ (compilers) Geology and mineral resources of the Northern Territory. Northern Territory Geological Survey, Special Publication 5. Hyperlink to: Daly Basin. In: Ahmad and Munson TJ (compilers) Geology and mineral resources of the Northern Territory . Kruse PD and Munson TJ (2013b) Wiso Basin. In: Ahmad M and Munson TJ (compilers) Geology and mineral resources of the Northern Territory. Northern Territory Geological Survey, Special Publication 5. Hyperlink to: Wiso Basin. In: Ahmad M and Munson TJ (compilers) Geology and mineral resources of the Northern Territory . Kudryavtsev S (2004) Improved harmonic development of the Earth tide-generating potential. Journal of Geodesy 77, 10. Lamontagne S, Suckow A, Gerber C, Deslandes A, Wilske C and Tickell S (2021) Groundwater sources for the Mataranka Springs (Northern Territory, Australia). Scientific Reports 11(1), 24288. DOI: 10.1038/s41598-021-03701-1. Leaney FW and Herczeg AL (2006) A rapid field extraction method for determination of radon-222 in natural waters by liquid scintillation counting. Limnology and Oceanography: Methods 4(7), 254–259. DOI: 10.4319/lom.2006.4.254. Lewis A, Oliver S, Lymburner L, Evans B, Wyborn L, Mueller N, Raevksi G, Hooke J, Woodcock R, Sixsmith J, Wu W, Tan P, Li F, Killough B, Minchin S, Roberts D, Ayers D, Bala B, Dwyer J, Dekker A, Dhu T, Hicks A, Ip A, Purss M, Richards C, Sagar S, Trenham C, Wang P and Wang LW (2017) The Australian Geoscience Data Cube — foundations and lessons learned. Remote Sensing of Environment 202, 276–292. DOI: 10.1016/j.rse.2017.03.015. Lucas LL and Unterweger MP (2000) Comprehensive review and critical evaluation of the half-life of tritium. Journal of Research of the National Institute of Standards and Technology 105(4), 541–549. DOI: 10.6028/jres.105.043. Lymburner L (2021) Geoscience Australia Landsat fractional cover collection 3. Geoscience Australia, Canberra. ( Hyperlink to: Geoscience Australia Landsat fractional cover collection 3 ). Lyne V and Hollick M (1979) Stochastic time variable rainfall runoff modelling. In: Hydrology and Water Resources Symposium, 89–92. Mahara Y, Habermehl MA, Hasegawa T, Nakata K, Ransley TR, Hatano T, Mizuochi Y, Kobayashi H, Ninomiya A, Senior BR, Yasuda H and Ohta T (2009) Groundwater dating by estimation of groundwater flow velocity and dissolved 4He accumulation rate calibrated by 36Cl in the Great Artesian Basin, Australia. Earth and Planetary Science Letters 287(1), 43–56. DOI: https://doi.org/10.1016/j.epsl.2009.07.034. Małoszewski P and Zuber A (1982) Determining the turnover time of groundwater systems with the aid of environmental tracers: 1. Models and their applicability. Journal of Hydrology 57(3), 207–231. DOI: 10.1016/0022-1694(82)90147-0. Mann J and Yin Foo D (1993) Beswick water supply: bore completion report, RN 28477 and RN 28478. Power and Water Authority, Darwin. Hyperlink to: Beswick water supply: bore completion report, RN 28477 and RN 28478 . McJannet D, Yang A and Seo L (2023) Climate data characterisation for hydrological and agricultural scenario modelling across the Victoria, Roper and Southern Gulf catchments. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. McVicar T, Vleeshouwer J, Van Niel T and Guerschman J (2022) Actual evapotranspiration for Australia using CMRSET algorithm. Version 1.0.0. Terrestrial Ecosystem Research Network (TERN). DOI: 10.25901/gg27-ck96. Melchior P (1983) The tides of the planet Earth. Pergamon Press, Oxford, UK. Moench A (1997) Flow to a well of finite diameter in a homogeneous, anisotropic water table aquifer. Water Resources Research 33(6), 11. Mook WG (1976) The dissolution-exchange model for dating groundwater with 14C. International Atomic Energy Agency. Morgenstern U and Taylor CB (2009) Ultra low-level tritium measurement using electrolytic enrichment and LSC. Isotopes in Environmental and Health Studies 45(2), 96–117. DOI: 10.1080/10256010902931194. Mueller N, Lewis A, Roberts D, Ring S, Melrose R, Sixsmith J, Lymburner L, McIntyre A, Tan P, Curnow S and Ip A (2016) Water observations from space: mapping surface water from 25 years of Landsat imagery across Australia. Remote Sensing of Environment 174, 341–352. DOI: 10.1016/j.rse.2015.11.003. Munson TJ, Ahmad M and Dunster JN (2013) Carpentaria Basin. In: Ahmad M and Munson TJ (compilers) Geology and mineral resources of the Northern Territory. Northern Territory Geological Survey, Special Publication 5. Hyperlink to: Carpentaria Basin. In: Ahmad M and Munson TJ (compilers) Geology and mineral resources of the Northern Territory . Murphy R, Neal B, Morden R, Nathan RN and Evans R (2008) BaseJumper – tool for analysing timetrends in baseflow. In: Water Down Under, Adelaide, Engineers Australia. National Uniform Drillers Licensing Committee (2020) Minimum construction requirements for water bores in Australia (4th edn). National Uniform Drillers Licensing Committee 2020. Hyperlink to: Minimum construction requirements for water bores in Australia (4th edn) . Newland MJ, Reeves CE, Oram DE, Laube JC, Sturges WT, Hogan C, Begley P and Fraser PJ (2013) Southern hemispheric halon trends and global halon emissions, 1978–2011. Atmospheric Chemistry and Physics 13(11), 5551–5565. DOI: 10.5194/acp-13-5551-2013. Nguyen D and Tickell S (2014) Pumping tests in the Northern Territory: a new spatial dataset. Palmerston, NT. ( Hyperlink to: Pumping tests in the Northern Territory: a new spatial dataset ). Northern Territory Geological Survey and Geognostics Australia Pty Ltd (2021) Northern Territory SEEBASE® and GIS. Northern Territory Geological Survey, Digital Information Package DIP 030. ( Hyperlink to: Northern Territory SEEBASE® and GIS ). Northern Territory Government (2022) Draft Georgina Wiso Water Allocation Plan 2022–2030 Department of Environment, Parks and Water Security: Northern Territory, Australia. Hyperlink to: Draft Georgina Wiso Water Allocation Plan 2022–2030 (download). Northern Territory Government (2023) The NT Water Data WebPortal v2023.3.123. A browserbased information and data presentation system integrating data collection, data storage, reporting, data computation, data management, and real-time information. Hyperlink to: The NT Water Data WebPortal v2023.3.123 . Paine AGL (1963) Tanumbirini, Northern Territory: 1:250,000 Geological Series - Explanatory Notes. Record SE5302. Geoscience Australia, Canberra., https://ecat.ga.gov.au/geonetwork/srv/eng/catalog.search#/metadata/79532. Parkhurst DL and Appelo CAJ (2013) Description of input and examples for PHREEQC version 3 — A computer program for speciation, batch-reaction, one-dimensional transport, and inverse geochemical calculations. U.S. Geological Survey techniques and methods, book 6, chap. A43. Hyperlink to: Description of input and examples for PHREEQC version 3 — A computer program for speciation, batch-reaction, one-dimensional transport, and inverse geochemical calculations . Pebesma EJ (2004) Multivariable geostatistics in S: the gstat package. Computers and Geosciences 30(7), 683–691. DOI: 10.1016/j.cageo.2004.03.012. Peeters L (2014) A background color scheme for piper plots to spatially visualize hydrochemical patterns. Groundwater 52(1), 2–6. DOI: 10.1111/gwat.12118. Petersen-Øverleir A, Soot A and Reitan T (2009) Bayesian rating curve inference as a streamflow data quality assessment tool. Water Resources Management 23(9), 1835–1842. DOI: 10.1007/s11269-008-9354-5. Petheram C, Walker G, Grayson R, Thierfelder T and Zhang L (2002) Towards a framework for predicting impacts of land-use on recharge: 1. A review of recharge studies in Australia. Australian Journal of Soil Research 40(3), 397–417. Pilgrim DH (1987) Australian rainfall and runoff: a guide to flood estimation. The Institution of Engineers, Australia, Barton, ACT. Piper AM (1944) A graphic procedure in the geochemical interpretation of water‐analyses. Eos, Transactions, American Geophysical Union 25(6), 914–923. Plumb KA and Roberts HG (1992) The geology of Arnhem Land, Northern Territory. Department of Primary Industries and Energy, Bureau of Mineral Resources, Geology and Geophysics. Record 1992/55. Hyperlink to: The geology of Arnhem Land, Northern Territory (download). Prowse G, Zaar U and Matthews I (1999) Water resources map of East Central Arnhem Land. Natural Resources Division, Department of Land, Planning and Environment, Northern Territory. R Core Team (2016) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Hyperlink to: R: a language and environment for statistical computing . Raiber M, Webb JA and Bennetts DA (2009) Strontium isotopes as tracers to delineate aquifer interactions and the influence of rainfall in the basalt plains of southeastern Australia. Journal of Hydrology 367(3), 188–199. DOI: 10.1016/j.jhydrol.2008.12.020. Randal MA (1973) Groundwater in the northern Wiso Basin and environs, Northern Territory. Bulletin 123. Department of Minerals and Energy, Bureau of Mineral Resources, Geology and Geophysics, Australian Government Publishing Service. Hyperlink to: Groundwater in the northern Wiso Basin and environs, Northern Territory . Rasmussen T and Crawford L (1997) Identifying and removing barometric pressure effects in confined and unconfined aquifers. Groundwater 35(3), 10. Rau G, Cuthbert M, Acworth R and Blum P (2020) Disentangling the groundwater response to Earth and atmospheric tides to improve subsurface characterisation. Hydrology and Earth System Sciences 24(12), 14. Rau GC, Eulenfeld T, Howe D, Rietbroek R, Gosselin J, Staniewicz S and Taylor D (2022) PyGTide: a Python module and wrapper for ETERNA PREDICT to compute synthetic model tides on Earth. Institute of Applied Geosciences, Karlsruhe Institute of Technology. Karlsruhe, Germany. ( Hyperlink to: PyGTide: a Python module and wrapper for ETERNA PREDICT to compute synthetic model tides on Earth ). Rawlings DJ (2015) Broughton-1 stratigraphic log, Pacific Oil and Gas. Northern Territory Geological Survey, Technical note 2015-004 (poster). Hyperlink to: Broughton-1 stratigraphic log, Pacific Oil and Gas (download). Raymond OL (2018) Australian geological provinces 2018.01 edn. Geoscience Australia, Canberra. Hyperlink to: Australian geological provinces 2018.01 edn . Raymond OL, Liu S, Gallagher R, Highet LM and Zhang W (2012) Surface geology of Australia, 1:1,000,000 scale, 2012 edn [Digital Dataset]. Geoscience Australia, Commonwealth of Australia, Canberra. https://ecat.ga.gov.au/geonetwork/srv/eng/catalog.search Rees GN, Oberprieler S, Nielsen D, Watson G, Shackleton M and Davis JA (2020) Characterisation of the stygofauna and microbial assemblages of the Beetaloo Sub-basin, Northern Territory. CSIRO, Australia., https://gisera.csiro.au/wp-content/uploads/2021/03/GISERA-Project18- Stygofauna_final-report-20201208.pdf. Reilly TE, Plummer LN, Phillips PJ and Busenberg E (1994) The use of simulation and multiple environmental tracers to quantify groundwater flow in a shallow aquifer. Water Resources Research 30(2), 421–433. DOI: 10.1029/93WR02655. Renard P, Glenz D and Mejias M (2009) Understanding diagnostic plots for well-test interpretation. Hydrogeology Journal 17(3), 589-600. Doi: 10.1007/s10040-008-0392-0. Salamon P, Fernàndez-Garcia D and Gómez-Hernández JJ (2006) A review and numerical assessment of the random walk particle tracking method. Journal of Contaminant Hydrology 87(3–4), 277–305. DOI: 10.1016/j.jconhyd.2006.05.005. Sanford W (2011) Calibration of models using groundwater age. Hydrogeology Journal 19, 13–16. Scanlon BR, Keese KE, Flint AL, Flint LE, Gaye CB, Edmunds WM and Simmers I (2006) Global synthesis of groundwater recharge in semiarid and arid regions. Hydrological Processes 20(15), 3335–3370. Schult J (2018) Dry season water quality at Elsey Station in the upper reaches of the Roper River. Department of Environment and Natural Resources, Northern Territory Government, Palmerston. Hyperlink to: Dry season water quality at Elsey Station in the upper reaches of the Roper River . Schult J and Novak P (2017) Water quality of the Roper River 2012–2016. Department of Environment and Natural Resources, Darwin. Hyperlink to: Water quality of the Roper River 2012–2016 . Smerdon BD, Payton Gardner W, Harrington GA and Tickell SJ (2012) Identifying the contribution of regional groundwater to the baseflow of a tropical river (Daly River, Australia). Journal of Hydrology 464–465, 107–115. DOI: 10.1016/j.jhydrol.2012.06.058. Solomon DK (2000) 4He in groundwater. In: Cook PG and Herczeg AL (eds), Environmental tracers in subsurface hydrology. Springer US, Boston, MA. Suckow A, Deslandes A, Gerber C, Mallants D and Smith S (2019) Commissioning of the CSIRO Noble Gas Facilities at Waite. CSIRO. Hyperlink to: Commissioning of the CSIRO Noble Gas Facilities at Waite . Suckow A and Dumke I (2001) A database system for geochemical, isotope hydrological, and geochronological laboratories. Radiocarbon 43(2A), 325–337. DOI: 10.1017/S0033822200038182. Sumner J (2008) Ngukurr groundwater investigation 2006–07. Department of Natural Resources, Environment, the Arts and Sport, Northern Territory Government, Alice Springs. Hyperlink to: Ngukurr groundwater investigation 2006–07 . Sweet IP, Brakel AT, Rawlings DJ, Haines PW, Plumb KA and Wygralak AS (1999) Mount Marumba, Northern Territory: explanatory notes – 1:250,000 geological map series. Australian Geological Survey Organisation and Northern Territory Geological Survey (National Geoscience Mapping Accord), Explanatory Notes, SD 53–6 (2nd edn). Hyperlink to: Mount Marumba, Northern Territory: explanatory notes – 1:250,000 geological map series . Tadros CV, Hughes CE, Crawford J, Hollins SE and Chisari R (2014) Tritium in Australian precipitation: a 50 year record. Journal of Hydrology 513, 262–273. DOI: 10.1016/j.jhydrol.2014.03.031. Tamers MA (1967) Radiocarbon ages of groundwater in an arid zone unconfined aquifer. In: Stout GE (ed.) Isotope techniques in the hydrologic cycle. American Geophysical Union. DOI: 10.1029/GM011p0143 Tarantola A (2005) Inverse problem theory and methods for model parameter estimation. Society for Industrial and Applied Mathematics, Philadelphia, PA, USA. Taylor A, Doble R, Crosbie R, Barry K, Harrington G, Davies P and Thomas M (2018a) Hydrogeological assessment of the Bulimba Formation – Mitchell catchment, Queensland. A technical report to the Australian Government from the CSIRO Northern Australia Water Resource Assessment, part of the National Water Infrastructure Development Fund: Water Resource Assessments. CSIRO. Hyperlink to: Hydrogeological assessment of the Bulimba Formation – Mitchell catchment, Queensland . Taylor A, Harrington G, Clohessy S, Dawes W, Crosbie R, Doble R, Wohling D, Batlle-Aguilar J, Davies P, Thomas M and Suckow A (2018b) Hydrogeological assessment of the Grant Group and Poole Sandstone – Fitzroy catchment, Western Australia. A technical report to the Australian Government from the CSIRO Northern Australia Water Resource Assessment, part of the National Water Infrastructure Development Fund: Water Resource Assessment. CSIRO. DOI: https://doi.org/10.25919/5b86ed7e26b48. Taylor AR, Smith SD, Lamontagne S and Suckow A (2017) Characterising alluvial aquifers in a remote ephemeral catchment (Flinders River, Queensland) using a direct push tracer approach. Journal of Hydrology 556, 600–610. DOI: 10.1016/j.jhydrol.2017.10.030. Tetzlaff D, Buttle J, Carey SK, Kohn MJ, Laudon H, McNamara JP, Smith A, Sprenger M and Soulsby C (2021) Stable isotopes of water reveal differences in plant – soil water relationships across northern environments. Hydrological Processes 35(1), e14023. DOI: 10.1002/hyp.14023. Theis C (1935) The relation between the lowering of the piezometric surface and the rate and duration of discharge of a well using ground‐water storage. Eos, Transactions, American Geophysical Union 16(2), 6. Thomas M, Philip S, Stockman U, Wilson PR, Searle R, Hill J, Bui E, Gregory L, Watson I, Wilson PL and Gallant G (2022) Soils and land suitability for the Roper catchment, Northern Territory. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Tickell S (2005) Groundwater resources of the Tindall Limestone. Northern Territory Government, Darwin. Retrieved 8 November 2023, https://hdl.handle.net/10070/613966. Tickell S (2016) Tindall Aquifer King River to Daly Waters map. Water Resources, Department of Land Resource Management, Northern Territory of Australia. Hyperlink to: Tindall Aquifer King River to Daly Waters map (download). Tickell SJ and Bruwer Q (2018) Georgina Basin groundwater assessment: Daly Waters to Tennant Creek. Department of Environment and Natural Resources, Northern Territory Government, Palmerston. Hyperlink to: Georgina Basin groundwater assessment: Daly Waters to Tennant Creek . Turnadge C, Crosbie R, Barron O and Rau G (2019) Comparing methods of barometric efficiency characterization for specific storage estimation. Groundwater 57(6), 16. Turnadge C, Crosbie RS, Tickell SJ, Zaar U, Smith SD, Dawes WR, Davies PJ, Harrington GA and Taylor AR (2018) Hydrogeological characterisation of the Mary–Wildman rivers area, Northern Territory. CSIRO, Australia. Hyperlink to: Hydrogeological characterisation of the Mary–Wildman rivers area, Northern Territory . Tyson P and Yin Foo D (1992) Barunga (Bamyili) water resource evaluation DLH-580. Power and Water Authority, Darwin. Hyperlink to: Barunga (Bamyili) water resource evaluation DLH-580 . Vaze J, Viney N, Stenson M, Renzullo L, Van Dijk A, Dutta D, Crosbie R, Lerat J, Penton D and Vleeshouwer J (2013) The Australian Water Resource Assessment Modelling System (AWRA). Proceedings of the 20th International Congress on Modelling and Simulation, Adelaide, Australia. https://www.mssanz.org.au/modsim2013/L17/vaze.pdf Verma MN (1990) Water supply investigation Hodgson Downs (Miniyeri) community DLH-642 supply investigation. Power and Water Authority, Darwin. Hyperlink to: Water supply investigation Hodgson Downs (Miniyeri) community DLH-642 supply investigation . Verma MN and Rowston P (1992) Bulman groundwater resource evaluation DLH-587. Power and Water Authority, Darwin. Hyperlink to: Bulman groundwater resource evaluation DLH-587 . Vogel JC (1967) Investigation of groundwater flow with radiocarbon. IAEA, International Atomic Energy Agency (IAEA). Von Neumann J (1951) Various techniques used in connection with random digits. Applied Mathematics Series 12(36–38), 5. Voss CI and Wood WW (1994) Synthesis of geochemical, isotopic and groundwater modelling analysis to explain regional flow in a coastal aquifer of southern Oahu, Hawaii. In: Proceedings of a Final Research Coordination Meeting on “Mathematical Models and Their Applications to Isotope Studies in Groundwater Hydrology” held by the International Atomic Energy Agency in Vienna, Austria, June 1–4, 1993. Pp. 147–178. Wagenaar D and Tickell S (2013) Late dry season stream flows and groundwater levels, upper Roper River, October 2013. Department of Land Resource Management, Northern Territory Government, Palmerston. Hyperlink to: Late dry season stream flows and groundwater levels, upper Roper River, October 2013 . Warfe DM, Pettit NE, Davies PM, Pusey BJ, Hamilton SK, Kennard MJ, Townsend SA, Bayliss P, Ward DP, Douglas MM, Burford MA, Finn M, Bunn SE and Halliday IA (2011) The 'wet–dry' in the wet–dry tropics drives river ecosystem structure and processes in northern Australia. Freshwater Biology 56(11), 2169–2195. DOI: 10.1111/j.1365-2427.2011.02660.x. Watson Resource Consulting (1999) Water resource management Tindal Limestone aquifer Katherine to Mataranka. Northern Territory Government, Lands Planning Environment, Darwin. Viewed 9 November 2023, Hyperlink to: Water resource management Tindal Limestone aquifer Katherine to Mataranka . Waugh P and Kerle E (2014) End of wet season stream flow measurements, Roper River, May 2014. Department of Land Resource Management, Northern Territory Government, Palmerston, Hyperlink to: End of wet season stream flow measurements, Roper River, May 2014 . Weiss RF (1968) Piggyback sampler for dissolved gas studies on sealed water samples. Deep Sea Research and Oceanographic Abstracts 15(6), 695–699. DOI: 10.1016/0011-7471(68)90082- X. Wenzel H (1996) The nanogal software: Earth tide data processing package ETERNA 3.30. Bulletin d'Information des Marées Terrestres 124, 15. Wilkins A, Crosbie R, Louth-Robins T, Davies P, Raiber M, Dawes W and Gao L (2022a) Australian gridded chloride deposition-rate dataset. Data in Brief 42, 108189. DOI: 10.1016/j.dib.2022.108189. Wilkins A, Crosbie RS, Raiber M, Dawes W, Louth-Robins T and Gao L (2022b) Quantifying recharge to groundwater systems in the NSW coalfields, ACARP Report C33037. CSIRO Report EP2022- 1381. Williams D, Chudleigh I and Jolly P (2003) Arafura swamp water resources study 2003, Technical report. Northern Territory, Department of infrastructure, Planning and Environment, Natural Systems Division, Darwin. Hyperlink to: Arafura swamp water resources study 2003 . Wood WW (1999) Use and misuse of the chloride-mass balance method in estimating ground water recharge. Ground Water 37(1), 2–3. Wood WW and Sanford WE (1995) Chemical and isotopic methods for quantifying ground-water recharge in a regional, semiarid environment. Ground Water 33(3), 458–468. Yin Foo D (1983) Beswick water supply – investigation of groundwater resources, 1982 DLH-581. Department of Transport and Works, Darwin. Hyperlink to: Beswick water supply – investigation of groundwater resources, 1982 DLH-581 . Yin Foo D (1991) Beswick water supply: an assessment of available data 1991 DLH-581. Power and Water Authority, Darwin. Hyperlink to: Beswick water supply: an assessment of available data 1991 DLH-581 . Yin Foo D (2002) Water resources of the Sturt Plateau region, overview report. Department of Infrastructure, Planning and the Environment, Northern Territory Government, Darwin. https://territorystories.nt.gov.au/10070/672757/0/0. Yin Foo D and Matthews I (2001) Hydrogeology of the Sturt Plateau. 1:250,000 scale map explanatory notes. Department of Infrastructure, Planning and the Environment, Darwin. Hyperlink to: Hydrogeology of the Sturt Plateau. 1:250,000 scale map explanatory notes . Yin Foo D and Matthews I (2002) Sturt Plateau 1:250,000 scale hydrogeology map. Conservation and Natural Resources Group, Department of Infrastructure, Planning and the Environment, Darwin ( Hyperlink to: Sturt Plateau 1:250,000 scale hydrogeology map ). Zaar U (2009a) Gulf water study: final report. Department of Natural Resources, Environment, the Arts and Sport, Northern Territory Government, Palmerston. Hyperlink to: Gulf water study: final report . Zaar U (2009b) Gulf water study: Limmen Bight and the Macarthur rivers region. Department of Natural Resources, Environment the Arts and Sport, Darwin. Hyperlink to: Gulf water study: Limmen Bight and the Macarthur rivers region . Zaar U (2009c) Gulf water study: Robinson and Calvert rivers region. Department of Natural Resources, Environment the Arts and Sport, Northern Territory Government, Palmerston. Hyperlink to: Gulf water study: Robinson and Calvert rivers region . Zaar U (2009d) Gulf water study: Roper River region. Department of Natural Resources, Environment, the Arts and Sport, Northern Territory Government, Darwin. Hyperlink to: Gulf water study: Roper River region . Zaar U, Jolly P and Yin Foo D (2009a) Water resources map of the Limmen Bight River region. Department of Natural Resources, Environment, the Arts and Sport, Northern Territory Government, Darwin. ( Hyperlink to: Water resources map of the Limmen Bight River region ). Zaar U, Jolly P and Yin Foo D (2009b) Water resources map of the McArthur River region. Department of Natural Resources, Environment, the Arts and Sport. Northern Territory Government, Darwin. (https://data.nt.gov.au/dataset/gulf-water-study). Zaar U, Jolly P and Yin Foo D (2009c) Water Resources map of the Robinson and Calvert rivers Region. Department of Natural Resources, Environment, the Arts and Sport, Northern Territory Government, Darwin. (https://data.nt.gov.au/dataset/gulf-water-study). Zaar U, Jolly P and Yin Foo D (2009d) Water Resources map of the Roper River Region. Gulf Water Study. Department of Natural Resources, Environment, the Arts and Sport, Northern Territory Government, Darwin. (https://data.nt.gov.au/dataset/gulf-water-study). Zaar U and Tickell S (2003) Water resources of west central Arnhem Land. Department of Lands, Planning and Environment, Natural Resources Division, Northern Territory Government, Darwin Hyperlink to: Water resources of west central Arnhem Land ). Zaar U and Tien T (2003) Water resources of West Arnhem Land. Department of Lands, Planning and Environment, Natural Resources Division, Northern Territory Government, Darwin. Hyperlink to: Water resources of West Arnhem Land . Zuber A (1986) Mathematical models for the interpretation of environmental radioisotopes in groundwater systems. In: Fritz P and Fontes J-C (eds) Handbook of environmental isotope geochemistry, Vol. 2. Elsevier, Amsterdam, 1–59. Part IV Appendices A.1 Borehole log and bore construction diagrams A.1.1 Tindall Limestone Apx Figure A.1-1 Bore log diagram for RN043046 installed in the Tindall Limestone Apx Figure A.1-2 Bore log diagram for RN043049 installed in the Tindall Limestone A.1.2 Antrim Plateau Volcanics Apx Figure A.1-3 Bore log diagram for RN043045 installed in the Antrim Plateau Volcanics Apx Figure A.1-4 Bore log diagram for RN043047 installed in the Antrim Plateau Volcanics A.2 Field sampling data A.2.1 Groundwater bore data Apx Table A.2-1 Summary of details for bores sampled during this assessment BORE REGISTERED NUMBER DATE SAMPLED EASTING NORTHING HYDROGEOLOGICAL UNIT GROUNDWATER LEVEL (mBTOC) SCREEN (mBGL) BORE DEPTH (mBGL) RN031167 21/08/2022 320089 8320304 Tindall Limestone NA 33–40 42 RN033095 20/08/2022 295827 8314966 Tindall Limestone 40 61–65 67 RN034324 21/08/2022 320284 8307844 Tindall Limestone 39 46.5–47 49 RN036503 21/08/2022 330720 8306979 Tindall Limestone 23 32.2–34 40 RN029537 17/08/2022 336686 8295795 Tindall Limestone 34 42–48 48 RN040211 19/08/2022 304624 8293273 Tindall Limestone 45 102–108 108 RN040219 17/08/2022 316770 8288638 Tindall Limestone 49 63–70 70 RN029706 17/08/2022 337950 8278100 Tindall Limestone 45.4 71–77 83 RN039137 18/08/2022 329901 8264706 Tindall Limestone 41 65–77 77 RN033297 18/08/2022 357385 8249225 Tindall Limestone 49.5 65.5–72 72 RN033297–R 18/08/2022 357385 8249225 Tindall Limestone 49.5 65.5–72 72 RN039136 18/08/2022 344933 8248567 Tindall Limestone 46 65–77 77 RN041491 22/08/2022 321008 8225778 Gum Ridge Formation 59 64–70 94 RN041203 22/08/2022 334999 8222610 Gum Ridge Formation 54 80–86 86 RN038185 22/08/2022 352067 8220410 Gum Ridge Formation 55 67.5–75 76 RN036303 16/08/2022 393408 8460346 Mountain Valley Limestone 25.0 55–61.5 73.5 RN031981 17/08/2022 410843 8507485 Dook Creek Formation 4.7 17.7–29.7 189.6 RN028226 18/08/2022 430438 8498818 Dook Creek Formation 14.5 36.0–54.0 54.0 RN028228 19/08/2022 418278 8498259 Dook Creek Formation 24.6 38.0–47.0 NA RN031970 20/08/2022 481541 8570273 Dook Creek Formation 8.7 NA 119.4 BORE REGISTERED NUMBER DATE SAMPLED EASTING NORTHING HYDROGEOLOGICAL UNIT GROUNDWATER LEVEL (mBTOC) SCREEN (mBGL) BORE DEPTH (mBGL) RN028224 21/08/2022 420726 8523364 Dook Creek Formation 34.2 45.5–58.4 64.7 RN031983 22/08/2022 357941 8442661 Dook Creek Formation 11.5 53.7–65.7 244.3 RN043047 18/04/2023 306215 8339104 Antrim Plateau Volcanics 16.5 68–73.5 174.2 RN043049 18/04/2023 306228 8339106 Tindall Limestone 5.85 45–49 49.7 RN034031 18/04/2023 306203 8339110 Tindall Limestone 5.2 35.4–41.4 91.9 RN034230 19/04/2023 294850 8351457 Tindall Limestone NA 12.–18.4 67.9 RN043046 19/04/2023 294864 8351452 Tindall Limestone 5.4 37.6–41.9 47.1 RN032164 19/04/2023 298122 8307383 Tindall Limestone NA NA NA RN043045 19/04/2023 294839 8351450 Antrim Plateau Volcanics 19.5 56.7–62.2 186 RN035464 20/04/2023 322174 8338623 Tindall Limestone 5 14.5–17.5 18 RN036507 20/04/2023 335974 8342380 Cretaceous Sandstone 18 52.1–55 79 RN024602 20/04/2023 324084 8346852 Abner Sandstone 21 38–50 50 mBTOC = metres below top of casing; mBGL = metres below ground level; NA = not available. A.2.2 Measured field parameter data Apx Table A.2-2 Summary of measured field parameters at each groundwater bore sampled in this assessment BORE REGISTERED NUMBER DATE SAMPLED EASTING NORTHING HYDROGEOLOGICAL UNIT TEMP (OC) pH EC (μS/cm) DO (mg/L) TOTAL ALKALINITY AS CaCO3 (mg/L) RN031167 21/08/2022 320089 8320304 Tindall Limestone 32.6 7.13 1510 0.4 468 RN033095 20/08/2022 295827 8314966 Tindall Limestone 33.1 7.08 1850 1.3 423 RN034324 21/08/2022 320284 8307844 Tindall Limestone 32.6 6.94 1710 0.75 478 RN036503 21/08/2022 330720 8306979 Tindall Limestone 32.1 7.04 1340 0.4 431 RN029537 17/08/2022 336686 8295795 Tindall Limestone 31.9 7.05 1420 1 515 RN040211 19/08/2022 304624 8293273 Tindall Limestone 33.6 6.99 1860 1 468 RN040219 17/08/2022 316770 8288638 Tindall Limestone 32 7.03 1780 0.8 455 RN029706 17/08/2022 337950 8278100 Tindall Limestone 31.8 7.04 1360 0.6 459 BORE REGISTERED NUMBER DATE SAMPLED EASTING NORTHING HYDROGEOLOGICAL UNIT TEMP (OC) pH EC (μS/cm) DO (mg/L) TOTAL ALKALINITY AS CaCO3 (mg/L) RN039137 18/08/2022 329901 8264706 Tindall Limestone 31.6 7.11 1650 0.7 449 RN033297 18/08/2022 357385 8249225 Tindall Limestone 32 7.17 830 1.15 342 RN033297–R 18/08/2022 357385 8249225 Tindall Limestone 32.1 6.72 850 1.1 332 RN039136 18/08/2022 344933 8248567 Tindall Limestone 32.3 7.15 1360 0.7 434 RN041491 22/08/2022 321008 8225778 Gum Ridge Formation 33.8 6.94 1870 1.4 470 RN041203 22/08/2022 334999 8222610 Gum Ridge Formation 32.4 7.18 1800 0.8 450 RN038185 22/08/2022 352067 8220410 Gum Ridge Formation 32.6 7.265 1310 1.3 378 RN036303 16/08/2022 393408 8460346 Mountain Valley Limestone 31 6.84 770 0.28 795 RN031981 17/08/2022 410843 8507485 Dook Creek Formation 29.3 7.05 590 0.57 716 RN028226 18/08/2022 430438 8498818 Dook Creek Formation 31.4 6.42 270 0.16 296 RN028228 19/08/2022 418278 8498259 Dook Creek Formation 31.7 6.79 470 0.77 440 RN031970 20/08/2022 481541 8570273 Dook Creek Formation 30.7 4.75 120 0.21 Below range RN028224 21/08/2022 420726 8523364 Dook Creek Formation 31.6 6.12 180 0.32 220 RN031983 22/08/2022 357941 8442661 Dook Creek Formation 32.1 6.75 1130 1.15 1096 RN043047 18/04/2023 306215 8339104 Antrim Plateau Volcanics 33.5 7.82 1870 0.2 70 RN043049 18/04/2023 306228 8339106 Tindall Limestone 34 6.69 2170 0.4 516 RN034031 18/04/2023 306203 8339110 Tindall Limestone 31.6 6.685 2700 0.2 508 RN034230 19/04/2023 294850 8351457 Tindall Limestone 33.6 6.8 870 0.65 336 RN043046 19/04/2023 294864 8351452 Tindall Limestone 33.6 6.83 850 0.5 332 RN032164 19/04/2023 298122 8307383 Tindall Limestone 33.1 6.58 1890 0.9 454 RN043045 19/04/2023 294839 8351450 Antrim Plateau Volcanics 34.6 7.22 1420 0.1 125 RN035464 20/04/2023 322174 8338623 Tindall Limestone 31.6 6.63 1130 0.1 404 RN036507 20/04/2023 335974 8342380 Cretaceous Sandstone 31.8 6.56 690 1.4 281 RN024602 20/04/2023 324084 8346852 Abner Sandstone 32.2 3.85 40 0.4 5 –R = duplicate observations undertaken A.2.3 Laboratory chemical analyses data Apx Table A.2-3 Summary of measured laboratory chemical analyses for groundwater samples collected at each bore site BORE REGISTERED NUMBER DATE SAMPLED HYDROGEOLOGICAL UNIT EC (μS/CM) Ca (mg/L) Mg (mg/L) Na (mg/L) K (mg/L) SO4(mg/L) Cl(mg/L) TOTAL ALKALINITY AS CACO3 (mg/L) Br(mg/L) NO3(mg/L) RN031167 21/08/2022 Tindall Limestone 1460 173 70.2 102 9.93 127.2 118.8 1177 0.5 0.5 RN033095 20/08/2022 Tindall Limestone 1790 156 68.1 200 25.7 134.9 204.4 1220 0.8 2.3 RN034324 21/08/2022 Tindall Limestone 1660 157 70.7 140 14.2 139.1 148.1 1234 0.5 0.5 RN036503 21/08/2022 Tindall Limestone 1320 157 68.1 78.9 6.9 105.4 91.1 1150 0.5 0.5 RN029537 17/08/2022 Tindall Limestone 1380 139 71.6 87.8 7.91 106.6 91.8 1230 0.5 6.0 RN040211 19/08/2022 Tindall Limestone 1800 157 72.6 198 19.8 129.8 213.7 1242 0.8 2.0 RN040219 17/08/2022 Tindall Limestone 1730 156 64.4 184 26.1 134.4 194.0 1189 0.7 2.6 RN029706 17/08/2022 Tindall Limestone 1320 157 55.4 80.4 10.8 102.1 84.9 1186 0.5 0.5 RN039137 18/08/2022 Tindall Limestone 1610 160 63.2 159 22.5 120.5 163.9 1159 0.6 2.1 RN033297 18/08/2022 Tindall Limestone 800 144 30.1 22.4 3.69 27.2 25.7 891 0.5 2.9 RN039136 18/08/2022 Tindall Limestone 1320 167 56.6 89.7 12.1 105.9 97.8 1129 0.5 0.6 RN041491 22/08/2022 Gum Ridge Formation 1810 162 68.6 199 19.6 130.2 217.0 1195 0.8 3.6 RN041203 22/08/2022 Gum Ridge Formation 1750 141 65.9 202 30 132.9 214.6 1129 0.8 3.7 RN038185 22/08/2022 Gum Ridge Formation 1260 163 62.1 74.2 7.35 89.5 117.6 994 0.5 1.3 RN036303 16/08/2022 Mountain Valley Limestone 630 34.5 38.3 51.2 16.9 3.7 45.6 641 0.5 0.5 RN031981 17/08/2022 Dook Creek Formation 560 66.5 35.6 16.8 7.6 6.0 5.2 732 0.5 0.5 RN028226 18/08/2022 Dook Creek Formation 180 21.6 9.45 4.67 0.613 0.5 5.8 214 0.5 0.5 RN028228 19/08/2022 Dook Creek Formation 400 46.5 28.7 9.04 1.28 0.5 6.1 527 0.5 0.6 RN031970 20/08/2022 Dook Creek Formation 110 1.11 2.05 10.1 0.355 0.5 22.6 25 0.5 0.5 RN028224 21/08/2022 Dook Creek Formation 150 15.6 7.13 4.02 1.15 0.5 3.6 170 0.5 0.8 RN031983 22/08/2022 Dook Creek Formation 920 99.8 76.1 19 10.1 3.2 7.3 1284 0.5 0.5 BORE REGISTERED NUMBER DATE SAMPLED HYDROGEOLOGICAL UNIT EC (μS/CM) Ca (mg/L) Mg (mg/L) Na (mg/L) K (mg/L) SO4(mg/L) Cl(mg/L) TOTAL ALKALINITY AS CACO3 (mg/L) Br(mg/L) NO3(mg/L) RN043047 18/04/2023 Antrim Plateau Volcanics 1770 146 27.9 167 10.3 158 383 86 <2 <2 RN043049 18/04/2023 Tindall Limestone 2050 122 66.7 225 29.7 173 243 495 <2 <2 RN034031 18/04/2023 Tindall Limestone 2620 123 86.8 348 36.9 203 280 511 <2 <2 RN034230 19/04/2023 Tindall Limestone 820 86.1 27.6 29.7 6 22.1 33.9 350 <0.5 <0.5 RN043046 19/04/2023 Tindall Limestone 810 82.7 27.2 26.7 5.99 19.2 29.3 345 <0.5 <0.5 RN032164 19/04/2023 Tindall Limestone 1810 133 52.8 169 24.6 128 195 474 <2 2.1 RN043045 19/04/2023 Antrim Plateau Volcanics 1350 120 21.2 109 6.11 152 229 135 <1 <1 RN035464 20/04/2023 Tindall Limestone 1080 124 36.1 42.8 4.09 53.8 54.2 422 <0.5 <0.5 RN036507 20/04/2023 Cretaceous Sandstone 650 44.7 32.9 20 3.03 3.0 17.0 300 <0.5 9.3 RN024602 20/04/2023 Abner Sandstone 40 0.521 0.166 1.18 <0.2 <0.5 3.7 3.2 <0.5 <0.5 A.2.4 Environmental tracer data Apx Table A.2-4 Summary of measured environmental tracer analyses for groundwater samples collected at each bore site BORE REGISTERED NUMBER DATE SAMPLED HYDROGEOLOGICAL UNIT δ 2H (‰ VSMOW) δ 18O (‰ VSMOW) 87Sr/86Sr 3H (TU) CFC-11 (pMol/kg) CFC-12 (pMol/kg) ) 14C (pmC) δ13C (‰ PDB) 222 Rn (Bq/L) SF6 (fmol.kg-1) RN031167 21/08/2022 Tindall Limestone −8.4 −56 0.713 0.033 0.32 0.58 75.30 −9.94 1.09 0.79 RN033095 20/08/2022 Tindall Limestone −8.0 −55 0.717 0 – – 63.77 −9.45 – – RN034324 21/08/2022 Tindall Limestone −8.4 −55 0.713 0.046 0.43 0.74 74.12 −9.54 0.75 0.83 RN036503 21/08/2022 Tindall Limestone −8.4 −57 0.713 0.138 0.14 0.42 81.62 −9.54 6.25 0.78 RN029537 17/08/2022 Tindall Limestone −8.45 −59 0.712 0.061 0.47 0.72 78.03 −9.29 3.62 1.05 RN040211 19/08/2022 Tindall Limestone −8.4 −59 0.718 0 0.59 0.57 77.95 −9.44 9.54 0.54 RN040219 17/08/2022 Tindall Limestone −7.9 −55 0.717 −0.023 0.24 0.24 58.27 −8.65 3.57 1.54 RN029706 17/08/2022 Tindall Limestone −8.4 −54 0.713 0.024 0.38 0.51 58.08 −8.48 3.03 1.13 RN039137 18/08/2022 Tindall Limestone −8.0 −58 0.716 0.024 0.44 0.43 55.37 −8.62 5.22 0.87 BORE REGISTERED NUMBER DATE SAMPLED HYDROGEOLOGICAL UNIT δ 2H (‰ VSMOW) δ 18O (‰ VSMOW) 87Sr/86Sr 3H (TU) CFC-11 (pMol/kg) CFC-12 (pMol/kg) ) 14C (pmC) δ13C (‰ PDB) 222 Rn (Bq/L) SF6 (fmol.kg-1) RN033297 18/08/2022 Tindall Limestone – – – – – – – – 3.97 – RN033297–R 18/08/2022 Tindall Limestone −7.6 −50 0.712 0.627 0.73 0.67 56.02 −8.81 4.10 2.66 RN039136 18/08/2022 Tindall Limestone −8.2 −58 0.713 0.004 0.50 0.45 50.51 −8.49 1.72 0.70 RN041491 22/08/2022 Gum Ridge Formation −8.6 −61 0.718 0.005 0.26 0.21 69.52 −9.48 6.05 0.82 RN041203 22/08/2022 Gum Ridge Formation −7.7 −53 0.718 0.044 0.45 0.35 54.68 −9.02 2.28 na RN038185 22/08/2022 Gum Ridge Formation −8.3 −55 0.712 0.119 0.18 0.14 48.00 −8.82 18.1 12.13 RN036303 16/08/2022 Mountain Valley Limestone −8.4 −53 0.735 0.018 0.15 0.12 37.03 −17.38 5.95 0.21 RN031981 17/08/2022 Dook Creek Formation −7.4 −49 0.721 −0.005 0.07 0.02 2.73 −14.87 6.31 1.45 RN028226 18/08/2022 Dook Creek Formation −7.4 −48 0.751 0.335 0.02 0.03 68.39 −14.88 5.66 1.95 RN028228 19/08/2022 Dook Creek Formation −8.0 −47 0.733 0.791 0.21 0.49 73.31 −14.20 4.79 13.57 RN031970 20/08/2022 Dook Creek Formation −7.0 −42 0.738 0.072 0.05 0.04 68.30 −23.74 56.69 0.18 RN028224 21/08/2022 Dook Creek Formation −3.6 −30 0.746 0.03 0.08 0.04 72.77 −20.75 16.91 0.15 RN031983 22/08/2022 Dook Creek Formation −6.8 −46 0.766 0.481 0.11 0.14 83.38 −13.34 10.36 0.45 RN043047 18/04/2023 Antrim Plateau Volcanics −8.3 −59 0.714 −0.003 0.03 0.23 stc stc 11.6 1.51 RN043049 18/04/2023 Tindall Limestone −8.3 −60 0.716 0.129 0.05 0.40 stc stc 10.7 1.42 RN034031 18/04/2023 Tindall Limestone −8.0 −60 0.717 na 0.07 0.39 stc stc 14.6 1.86 RN034230 19/04/2023 Tindall Limestone −8.1 −60 0.718 0.063 0.15 0.29 stc stc 17.7 2.96 RN043046 19/04/2023 Tindall Limestone −8.4 −57 0.718 0.009 0.19 0.31 stc stc 11.3 2.92 RN032164 19/04/2023 Tindall Limestone −8.2 −58 0.717 0.006 0.29 0.33 stc stc 6.81 0.88 RN043045 19/04/2023 Antrim Plateau Volcanics −8.0 −62 0.712 −0.029 0.01 0.16 stc stc 20.5 0.87 RN035464 20/04/2023 Tindall Limestone −8.1 −58 0.714 0.292 0.03 0.21 stc stc 41.3 0.20 RN036507 20/04/2023 Cretaceous Sandstone −8.6 −63 0.716 0.181 0.52 0.58 stc stc 14.7 0.59 RN024602 20/04/2023 Abner Sandstone −8.0 −56 0.719 0.101 0.03 0.03 stc stc 49.9 0.21 –R = duplicate sample taken; – = no data A.2.5 Noble gas data Apx Table A.2-5 Summary of measured noble gas analyses for groundwater samples collected at each bore site BORE REGISTERERED NUMBER DATE SAMPLED HYDROGEOLOGICAL UNIT He (ccSTP/g) Ne (ccSTP/g) Ar (ccSTP/g) Kr (ccSTP/g) Xe (ccSTP/g) RN031167 21/08/2022 Tindall Limestone 7.9E-08 1.9E-07 2.6E-04 5.3E-08 6.4E-09 RN033095 20/08/2022 Tindall Limestone – – – – – RN034324 21/08/2022 Tindall Limestone 4.9E-08 1.8E-07 2.5E-04 5.3E-08 6.3E-09 RN036503 21/08/2022 Tindall Limestone 1.0E-07 2.2E-07 2.8E-04 5.7E-08 6.9E-09 RN029537 17/08/2022 Tindall Limestone 4.7E-08 1.7E-07 2.5E-04 5.1E-08 6.2E-09 RN040211 19/08/2022 Tindall Limestone 5.6E-08 1.8E-07 2.5E-04 5.2E-08 6.6E-09 RN040219 17/08/2022 Tindall Limestone 4.7E-08 1.6E-07 2.6E-04 5.5E-08 6.8E-09 RN029706 17/08/2022 Tindall Limestone 5.1E-08 1.8E-07 2.5E-04 5.1E-08 6.2E-09 RN039137 18/08/2022 Tindall Limestone 4.4E-08 1.5E-07 2.6E-04 5.4E-08 6.8E-09 RN033297 18/08/2022 Tindall Limestone 3.9E-08 1.4E-07 2.6E-04 5.3E-08 6.4E-09 RN033297–R 18/08/2022 Tindall Limestone 4.5E-08 1.9E-07 2.8E-04 5.8E-08 7.2E-09 RN039136 18/08/2022 Tindall Limestone 4.9E-08 1.7E-07 2.6E-04 5.4E-08 6.8E-09 RN041491 22/08/2022 Gum Ridge Formation 7.6E-08 1.8E-07 2.5E-04 5.2E-08 6.3E-09 RN041203 22/08/2022 Gum Ridge Formation 4.1E-08 1.6E-07 2.6E-04 5.4E-08 6.8E-09 RN038185 22/08/2022 Gum Ridge Formation 3.8E-08 1.5E-07 2.5E-04 5.3E-08 6.5E-09 RN036303 16/08/2022 Mountain Valley Limestone 3.4E-07 2.6E-07 3.2E-04 6.1E-08 7.7E-09 RN031981 17/08/2022 Dook Creek Formation 1.4E-06 2.0E-07 3.0E-04 6.1E-08 8.2E-09 RN028226 18/08/2022 Dook Creek Formation 8.2E-08 2.5E-07 3.1E-04 5.9E-08 7.6E-09 RN028228 19/08/2022 Dook Creek Formation – – – – – RN031970 20/08/2022 Dook Creek Formation 6.8E-08 2.7E-07 3.7E-04 7.3E-08 9.3E-09 RN028224 21/08/2022 Dook Creek Formation 7.1E-08 2.9E-07 3.7E-04 6.7E-08 8.5E-09 RN031983 22/08/2022 Dook Creek Formation 7.5E-08 2.1E-07 3.0E-04 6.0E-08 7.7E-09 –R = duplicate sample taken; – = no data