Salinity contours for alluvial resource units in the Murray-Darling Basin

Salinity contours based on interpolated 95th percentile bore observations for the main alluvial resource units used to calculate a groundwater stress metric.\nLineage: We used groundwater salinity data from 12,513 observation bores from the BOM NGIS (version 1.7.0) as inputs to calculate groundwater stress (iGF) in the 22 groundwater SDL management units analysed. The upper and lower two percent of salinity data were removed for data quality assurance to\nfilter out extreme values. From the remaining data, the 95th percentile salinity value was calculated for each bore for use in spatial\ninterpolation. A range of kriging interpolation models and parameter combinations\nwere tested to create prediction surfaces of salinity across SDL areas from bore data. Model fit was assessed using crossvalidation.\nThe semi-variogram model and parameter combination that minimised root-mean-square error (RMSE) was selected. A\nseparate kriging model was applied in the Shepparton Irrigation Region (SIR) (GS8a) as it overlaps and has a variable hydraulic\nconnection with the underlying Sedimentary Plain (GS8c). Continuous predicted\nvalues were contoured into salinity classes TDS (mg/L) following the Recharge Risk Assessment Method (RRAM) described in MDBA,\n2020a: 1) fresh < 1500 mg/L, 2) brackish 1500–3000 mg/L, 3) saline 3000–14,000 mg/L, and 4) highly saline > 14,000 mg/L. The\ntotal areas of these salinity classes within each groundwater SDL were used as inputs to calculate iGF.\nFull details are given in the research paper: https://doi.org/10.1016/j.ejrh.2023.101419