Groundwater-dependent ecosystems (GDEs) rely on access to groundwater on a permanent or intermittent basis for some or all of their water requirements (Queensland Government, 2018). Remotely sensed data from Digital Earth Australia (DEA) (Geoscience Australia, 2018) were used to map potential aquatic and other GDEs and enhance understanding of surface water – groundwater interactions in the Upper Burdekin region. Two Landsat TM satellite products (Water Observations from Space (WOfS; Mueller et al. 2016) summary statistic and Tasselled Cap Index (TCI) wetness summary)) were used to investigate the persistence of surface water and soil moisture in the landscape to identify perennial streams, springs and other parts of the landscape that may rely on groundwater discharge. The WOfS summary statistic represents, for each pixel, the percentage of time that water is detected at the surface relative to the total number of clear observations. Due to the 25-m by 25-m pixel size of Landsat data, only features at least 25 m wide are detected and only features covering multiple pixels are consistently detected. The WOfS summary statistic was produced over the McBride and Nulla Basalt provinces for the entire period of available data (1987 to 2018). Pixels were polygonised and classified in order to visually enhance key data in the imagery, such as the identification of standing water for at least 80% of the time. The TCI is a method of reducing six surface reflectance bands of satellite data to three bands (Brightness, Greenness, Wetness) using a Principal Components Analysis (PCA) and Procrustes' Rotation (Roberts et al., 2018). The published coefficients of Crist (1985) are applied to DEA's Landsat data to generate a TCI composite. The resulting Tasselled Cap bands are a linear combination of the original surface reflectance bands that correlate with the Brightness (bare earth), Greenness and Wetness of the landscape. The TCI wetness summary (or Tasselled Cap Wetness (TCW) percentage exceedance composite), derived from the Wetness band, represents the behaviour of water in the landscape, as defined by the presence of water, moist soil or wet vegetation at each pixel through time. The summary shows the percentage of observed scenes where the Wetness layer of the Tasselled Cap transform is above the threshold, i.e. where each pixel has been observed as ‘wet’ according to the TCI. Areas that retain surface water or wetness in the landscape during the dry season are potential areas of groundwater discharge and associated GDEs. The TCW threshold is set at -600 to calculate the percentage exceedance. This threshold is based on scientific judgment and is currently in the research/testing phase. It is based on Australian conditions and conservative in nature. The dry season, when surface runoff to streams and rainfall are minimal, is particularly useful for identifying and mapping groundwater-fed streams, springs and other ecosystems that rely on access to groundwater during periods of limited rainfall. The Upper Burdekin region was especially dry between May and October 2013, with low rainfall totals in the months preceding this dry season and overall below-average rainfall conditions (i.e. decline in rainfall residual mass). The TCW exceedance composite was classified into percentage intervals to distinguish areas that were wet for different proportions of time during the 2013 dry season. Field validation of the remote sensing data products would be required to confirm the preliminary identification of parts of the landscape where groundwater discharges to the surface and potentially supports GDEs. This release includes the classified WOfS summary statistic and classified TCW percentage exceedance composite (May-October 2013) data products for the McBride and Nulla basalt provinces in the Upper Burdekin region, North Queensland. <b>References: </b> Crist EP (1985) A TM Tasseled Cap equivalent transformation for reflectance factor data. Remote Sensing of Environment 17(3), 301–306. Doi: 10.1016/0034-4257(85)90102-6. Geoscience Australia (2018) Digital Earth Australia. Geoscience Australia, http://www.ga.gov.au/dea. 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, ISSN 0034-4257. Queensland Government (2018) Groundwater dependent ecosystems, WetlandInfo 2014. Queensland Government, Brisbane, https://wetlandinfo.des.qld.gov.au/wetlands/ecology/aquatic-ecosystems-natural/groundwater-dependent/. Roberts D, Dunn B and Mueller N (2018) Open Data Cube Products Using High-Dimensional Statistics of Time Series. International Geoscience and Remote Sensing Symposium. Valencia, Spain: IEEE Geoscience and Remote Sensing Society.

This report presents key results of groundwater level interpretations from the Upper Burdekin Groundwater Project in North Queensland, conducted as part of Exploring for the Future (EFTF)—an eight year, $225 million Australian Government funded geoscience data and information acquisition program focused on better understanding the potential mineral, energy and groundwater resources across Australia. The Upper Burdekin Groundwater Project is a collaborative study between Geoscience Australia and the Queensland Government. It focuses on basalt groundwater resources in two geographically separate areas: the Nulla Basalt Province (NBP) in the south and the McBride Basalt Province (MBP) in the north. This report interprets groundwater levels measured in both provinces by Geoscience Australia and the Queensland Government to provide recommendations for resource management. The NBP and MBP basalt aquifers are heterogeneous, fractured, vesicular systems. Several lava flows are mapped at surface in both provinces, and the degree of hydraulic connectivity between these flows is unclear. Although there was some uncertainty due to monitoring well construction issues, barometric efficiency analyses from supporting project documents suggest that the basalts of the NBP and MBP were unconfined where monitored during the EFTF project. That finding generally matches observations presented here. Longer term groundwater hydrographs suggest that groundwater levels have been declining in the NBP and MBP following major flooding in 2010-2011 related to one of the strongest La Niña events on record. Groundwater levels are yet to decline to pre-flood elevations in places. Importantly, these longer term hydrographs set the project in context: the EFTF monitoring period is only a small fraction of a much longer-functioning, dynamic groundwater system. Nulla Basalt Province The NBP is elongated east-west, and is situated entirely within the Burdekin River catchment. Volcanic vents in the west identify that area as the main extrusive centre. Regionally, groundwater migrates through the basalts of the NBP from the western high ground towards the Burdekin River in the east. Although lava flows of the NBP reach the Burdekin River, direct groundwater discharge in this area has not yet been proven. However, groundwater does discharge to various springs and surface watercourses in the NBP that are known tributaries of the Burdekin River. Despite the presence of many registered extraction bores, no clear signs of pumping were observed in groundwater hydrographs from the NBP during the EFTF monitoring period. Water levels in many bores responded to major rainfall events, ranging from a simple change in declining hydrograph slope to a water level increase of ~6.8 m in the central west. While some responses could have been induced by loading, electrical conductivity loggers and the extent of water level rise showed that many were clearly caused by recharge. At nested monitoring locations, groundwater levels remained commensurate with downward flow potentials throughout the EFTF monitoring period. McBride Basalt Province The MBP is approximately circular in plan, with volcanic vents present in a north-northeast trending band through the province centre. Lava flows extend away from the high ground of the province centre towards lower ground near the edges. In part due to its geometry, the MBP is situated within four river catchments; only surface water landing in the east flows into the Burdekin River. Regionally, groundwater migrates through the basalts of the MBP from the central high ground radially towards the edges. Direct groundwater discharge from the MBP basalts into the Burdekin River has been shown in this project. Similarly to the NBP, groundwater is also known to discharge to numerous springs and surface watercourses in the MBP. Water levels in many bores responded to major rainfall events. Responses ranged from a change in declining hydrograph slope to a water level increase of ~6.8 m in the southeast. While some responses could have been induced by loading, the extent of water level rise showed that others were clearly caused by recharge. No nested monitoring locations were installed for the EFTF project, so vertical head gradients are currently unknown. Although there are numerous registered extraction bores in the MBP, groundwater level response to pumping was only definitively identified in the east in bore RN12010016. However, several registered bores with high estimated yields have been installed in the northeast since EFTF fieldwork completion. It is possible that these higher yielding extraction bores may induce visible drawdown in monitoring bores in the future. Their high estimated yields may be associated with lava tubes; features not reported in the literature reviewed for this project for the NBP, but identified at surface and potentially in several Queensland Government bores drilled in the MBP. Conclusions and recommendations The Upper Burdekin Groundwater Project has provided abundant information on various aspects of the hydrogeology of the Nulla and McBride basalt provinces. General groundwater flow processes are understood at a regional scale for the EFTF monitoring period, but more detailed investigations and longer term monitoring are required to fully evaluate local conditions. One of the main observations of this study are the long term groundwater level declines in both the NBP and MBP following the 2010-2011 La Niña-associated floods. Groundwater levels are yet to reduce to pre-flood elevations in places, showing that the EFTF monitoring period represents only a small fraction of a much longer-functioning, dynamic groundwater system. It is unclear what, if any, contribution groundwater extraction has made to regional water level declines. Numerous correlations were assessed between groundwater hydrograph characteristics and potentially influencing factors, but the results were mostly inconclusive. There is uncertainty in hydraulic connectivity across lava flow boundaries and between intra-lava flow aquifers. Although interesting groundwater processes were identified at many bores, at the current bore spacing it is not generally possible to interpolate between locations with any certainty. Knowledge gaps and suggestions for further investigation are recorded in Section 5 of the report. The gaps identified should assist planning of future work to inform: - Further characterisation of groundwater resources. - Protection of groundwater dependent ecosystems. - Appropriate groundwater resource management.

This report provides an initial summary of the hydrogeochemistry of the McBride Basalt Province (MBP) and Nulla Basalt Province (NBP) of the Upper Burdekin Region of North Queensland, completed as part of Exploring for the Future (EFTF)—an eight year, $225 million Australian Government funded geoscience data and information acquisition program focused on better understanding the potential mineral, energy and groundwater resources across Australia. Groundwater hydrogeochemistry studies can improve system understanding by reflecting host formation compositions and groundwater processes. These studies also provide regional baseline groundwater datasets that can inform environmental monitoring, resource use and decision making. During 2017 and 2018 Geoscience Australia collected 38 groundwater samples and 80 surface water samples (including quality control samples) to evaluate groundwater system processes including potential flow paths, recharge and groundwater-surface water-interactions. These surveys were conducted across three months of fieldwork, sampling water for a comprehensive suite of hydrogeochemical parameters. The present report includes surface water and groundwater data and information on: 1) sampling sites; 2) field physicochemical parameters (EC, pH, Eh, DO and T); 3) field measurements of total alkalinity (HCO3-); 4) laboratory results of major anion and cation results; 5) laboratory results for isotopes of water (δ18O and δD), DIC (δ13C), and dissolved strontium (87Sr/86Sr); and 6) hydrogeochemical maps representing the spatial distribution of these parameters. Pending analyses include: CFCs, SF6 and radiogenic isotopes δ14C and δ36Cl. Analysis that were largely below detection limit include: trace element concentrations, dissolved sulfide (S2-), ferrous iron (Fe2+), and dissolved sulfate (affecting sampling of δ34S and δ18O). This study demonstrates that hydrogeochemistry surveys, with full suites of chemical parameters including isotopes, can reveal fundamental groundwater system processes such as groundwater flow paths, groundwater recharge and groundwater-surface water interactions. The chemical ‘fingerprints’ identified here indicate groundwater flow paths are largely restricted to within the MBP and NBP aquifers, which have little interaction with adjacent and underlying non-basaltic rocks. The results also indicate groundwater is largely recharged from rainfall in higher elevations of the basalt provinces, with variable rainfall inputs to groundwater from lower elevation and rivers along flow paths. Groundwater-surface water interactions show several chemical signatures linking groundwater to springs, tributary rivers and the Burdekin River. Results from the Upper Burdekin Hydrogeochemistry Survey for the MBP and NBP have been plotted and mapped with initial interpretations presented below. Further detailed interpretation of this hydrogeochemistry data will be the focus of future publications. This data release is part in a series of staged outputs from the EFTF program. Relevant data, information and images are available through the Geoscience Australia website.

This dataset includes point estimates of groundwater recharge in mm/year. Recharge rates have been estimated at monitoring bore locations in the basaltic aquifers of the Nulla and McBride basalt provinces. Recharge estimates have been calculated using the “chloride mass balance” method. The chloride mass balance process assumes that the chloride ion is a conservative tracer in precipitation, evapotranspiration, recharge and runoff; and that all the chloride is from rainfall, instead of for example halite saturation or dissolution processes. So the volumetric water balance and the flux of chloride balance must both be true. Assuming that runoff and evapotranspiration are negligible (so approximated by zero), the equation is simplified: Water balance P=ET+R+Q Water balance multiplied by chloride concentrations (chloridefluxbalance) P∙Cl_ppt=ET∙Cl_ET+R∙Cl_gw+Q∙Cl_riv | ΔCl_reac≈0 Assumptions to simplify equation P∙Cl_ppt=R∙Cl_gw | Q≈0 & ET≈0 Rearranging for recharge rate (unknown) R=P∙(Cl_ppt)/(Cl_gw ) | Q≈0 & ET≈0 Where P = precipitation rate; ET = evapotranspiration rate; R = recharge rate; Q = runoff to streams; Clppt = concentration of Cl in precipitation; ClET = concentration of chloride in evapotranspiration; Clgw = concentration of Cl in groundwater; Clriv = concentration of chloride in river runoff; ΔClreac = change in chloride concentrations from reactions.

The Upper Burdekin Chloride Mass Balance Recharge web service depicts the recharge rates have been estimated at borehole locations in the Nulla and McBride basalt provinces. Using rainfall rates, rainfall chemistry and groundwater chemistry, the recharge rates have been estimated through the Chloride Mass Balance approach.

<p>The outcrop extent of the Nulla Basalt Province, selected from the Queensland Detailed Surface Geology vector polygon mapping, March 2017. <p>© State of Queensland (Department of Natural Resources and Mines) 2017 Creative Commons Attribution

<p>The outcrop extent of the McBride Basalt Province, selected from the Queensland Detailed Surface Geology vector polygon mapping, March 2017. <p>© State of Queensland (Department of Natural Resources and Mines) 2017 Creative Commons Attribution

This grid dataset is an estimation of the relative surface potential for recharge within the Nulla Basalt Province. This process combined numerous factors together as to highlight the areas likely to have higher potential for recharge to occur. Soil permeability and surface geology are the primary inputs. Vegetation and slope were excluded from consideration, as these were considered to add too much complexity. Furthermore, this model does not include rainfall intensity – although this is known to vary spatially through average rainfall grids, this model is a depiction of the ground ability for recharge to occur should a significant rainfall event occur in each location. The relative surface potential recharge presented is estimated through a combination of soil and geological factors, weighting regions that are considered likely to have greater potential for recharge (e.g. younger basalts, vent-proximal facies, and highly permeable soils). Near-surface permeability of soil layers has been considered as a quantified input to the ability for water to infiltrate soil strata. It was hypothesised that locations proximal to volcanic vents would be preferential recharge sites, due to deeply penetrative columnar jointing. This suggestion is based on observations in South Iceland, where fully-penetrating columnar joint sets are more prevalent in proximal facies compared to distal facies in South Iceland (Bergh & Sigvaldson 1991). To incorporate this concept, preferential recharge sites are assumed to be within the polygons of vent-proximal facies as derived from detailed geological mapping datasets. Remaining geology has been categorised to provide higher potential recharge through younger lava flows. As such, a ranking between geological units has been used to provide the variation in potential recharge estimates. <b>Reference</b> Bergh, S. G., & Sigvaldason, G. E. (1991). Pleistocene mass-flow deposits of basaltic hyaloclastite on a shallow submarine shelf, South Iceland. Bulletin of Volcanology, 53(8), 597-611. doi:10.1007/bf00493688

The Upper Burdekin Basalt extents web service delivers province extents, detailed geology, spring locations and inferred regional groundwater contours for the formations of the Nulla and McBride Basalts. This work has been carried out as part of Geoscience Australia's Exploring for the Future program.

<p>Summary <p>Spring point locations compiled for the Nulla Basalt Province <p>A compilation of spring locations as identified through various methods, including existing Queensland Springs Database, topographic mapping, fieldwork visits, landholder citizen scientist mapping, and inspection for neighbouring similar features in Google Earth. This compilation has had locations adjusted through inspecting visible imagery and elevation data to identify the likely positions of springs at higher resolution.