Calculating salt loads and salinity risk in floodplain landscapes: a systems-based multi-scale approach
Lawrie, K, Tan, KP, Pain, C, Clarke, J, Apps, H, Halas, L, Wong, VNL, Gibson, D & Cullen, K 2008, 'Calculating salt loads and salinity risk in floodplain landscapes: a systems-based multi-scale approach', paper presented to the 33rd International Geological Congress, Oslo, Norway, 6-14 August.
A new approach to the calculation of salt loads and salinity risk in floodplain landscapes of the Murray Basin in south-eastern Australia involves the integration of satellite imagery (SPOT5 and LANDSAT), airborne electromagnetics, high resolution LiDAR digital elevation data, geomorphic, vegetation and geological mapping, and petrophysical attributes from laboratory analysis of field and borehole samples. Satellite imagery, validated by a field mapping program, was used to map surface salt and halophytic vegetation. AEM data, validated by drilling data, were used to produce maps of the key elements of the hydrogeological system including the thickness and extent of the main aquitards and aquifers and internal variations in lithology that could be related to significant variations in hydraulic conductivity. Calibration of hydrogeological parameters was accomplished using samples from a drilling program on the floodplains of the middle reaches of the Murray River. Pore fluids were collected from uncontaminated drill cores and analysed for total dissolved solutes. Other petrophysical attributes measured include lithology texture, porosity and permeability. These data are used to convert the electrical conductivity data to useable attributes such as water quality or tonnes per hectare salt. Previous studies have demonstrated that the causal relationship between apparent conductivity (ECa) and salt can be expressed as a linear function. The average conductivity in the material between surface and groundwater is used to produce a near-surface salt store map, and maps of water quality beneath the watertable. The average ECa is translated into total salt as tonnes per hectare. The spatial distribution of fine textured materials is then overlain onto the total salt image.
On average, the bulk density of sediments is approximately 1.6 g/cm3 (or t/m3). Because ArcGIS calculates the total-salt as voxels (i.e. 40 x 40 x 1 m3), the conversion of mass of sediments into voxel and finally to tonnes/hectare is carried out as follows: ECa = 0.1073 x Total-Salt mS/m; Total-Salt = ECa / 0.1073 g/tonne; Since 1 voxel = 40 x 40 x 1 = 1600 m3, taking the bulk density of dry sediments to be 1.6 t/m3, 1 voxel will contain 2560 tonnes (i.e. 1600 x 1.6) of dry sediments. Thus, to convert g/tonne to tonnes/voxel Total-Salt (where ECa is in S/m instead of mS/m); Total-Salt = ((ECa x 1000) / 0.1073)/1000000 x 2560 t / voxel. Since 1 hectare consists 6.25 voxels, Total-Salt = ((ECa x 1000) /0.1073)/1000000 x 2560 x 6.25 t / hectare; Total-Salt = ECa x 149.11 t / hectare. We conclude that salt stored in fine-textured material is unlikely to be mobilised whereas salt contained in permeable sands is considered to have a higher risk of mobility if connected to appropriate flowpaths. The derived maps and data are then input into hydrogeological and salinity impact models to establish the links between irrigation and salt discharge to the river and floodplain.