Understanding the natural and anthropogenic controls on aquifer salinization is of great societal relevance because of its impact on irrigated agriculture and drinking water supplies within semi-arid regions of the world where fresh water is in short supply. Because the dissolved solids concentration of groundwater recharge represents a measure of precipitation minus evapotranspiration, the amount and spatial distribution of salinity within shallow aquifer systems may represent a useful indicator of past climate. A research plan is outlined in this proposal to study salinity evolution within sedimentary basins on time scales of 104-105 years. A new mathematical model (MWT3D SALT) is in the process of being developed to solve governing equations representing groundwater, salt transport, groundwater/surface water interactions and evapotranspiration within the root zone. This three-dimensional model will be capable of representing four aquifer salinization mechanisms: 1) evaporative concentration and reflux of salts from saline lakes; 2) concentration of salts within the root zone and near the water table by deeply rooted phreatophytes; 3) diffusion of connate groundwater from marine confining units; 4) direct evaporation of groundwater across the capillary fringe where the water table encroaches upon the land surface. The model will be applied both in a generic sensitivity study and to the hydrogeologically and paleochimatologically well studied Murray Basin, Australia. The generic sensitivity study will help to determine what specific land surface, subsurface, and vegetation conditions play an important role in aquifer salinization. For the Murray Basin application, the model will first be calibrated/validated using modern hydrologic stresses, published aquifer parameters, and records of long-term (~50 years) groundwater level fluctuations. Once we are satisfied that the Murray Basin model can reproduce historical hydrologic changes at the regional scale, we will attempt to reconstruct Quaternary records of paleo-lake level fluctuations, the spatial occurrence of shallow evaporite deposits, and the present-day distribution of groundwater salinity across the basin. The paleohydrologic model will be driven either by estimates of groundwater recharge taken from composite lake core records for Australia and Tasmania and by paleo global circulation model (Paleo-GCM) output. If we can use lake core records and/or paleo-GCM output to reconstruct the present-day salinity distribution across sedimentary basins, then we will have discovered a powerful new approach for regional Quaternary paleoclimate studies in semi-arid regions of the world. PIs Person and Hanor have proven strengths in basin-scale mathematical modeling and groundwater brine geochemistry and are well suited to carry out the proposed study.
