Perched water tables are often associated with irrigation areas overlying a low-conductivity layer. Perching leads to waterlogging and salinity, necessitating sub-surface drainage. It also affects water percolating below the impeding layer to the regional water table, through changes in magnitude of fluxes and timing of impacts at the water table. In this way, perching affects the growth of groundwater mounds, and associated increases in salinity in nearby land and streams. The degree of perching, and risk of salinity, are reduced by water use efficiency improvements. The residual risk from irrigation development and efficiency improvements is determined by both the magnitude and timing of impacts.
Salinity strategies for the Murray-Darling Basin in south-eastern Australia have reduced salinity in the River Murray by developing salt interception schemes and the encouragement of water use efficiency programs. Groundwater modelling along the lower stretches of the River Murray is used to guide design and management of groundwater pumping, attribution of costs and rehabilitation of floodplains, but does not currently simulate perching and its influence on the magnitude and timing of salinity impacts to the river, creating a gap in the modelling process between cause and effect.
This talk describes the trialing of integrated agronomic- unsaturated zone- groundwater modelling at Loxton, which represents perched water tables explicitly and links the relevant processes. The agronomic water balance model uses observed diversion volumes and empirical parameters to estimate a time series for irrigation accession flux. The unsaturated zone model uses this time series to estimate the time series for recharge and losses to the surface (mainly by drainage). The groundwater model uses the recharge output as part of estimating fluxes of groundwater to the River Murray and adjacent floodplains and groundwater pumping schemes. The magnitude and timing of the groundwater flux is directly related to the magnitude, timing and location of irrigation recharge, which, in turn, is directly related to the empirical factors and the thickness and properties of soil layers. Some of these parameters, along with hydrogeological parameters are difficult to measure at scale and need to be calibrated in a way that constrains non-uniqueness problems. The purpose of the trial is to develop approaches for the implementation and calibration of the integrated model.
The development of the unsaturated zone model that can adequately represent perching has been critical to the capacity to conduct such a trial. While numerical models, can simulate unsaturated zone processes, including perching, numerical stability and the temporal and spatial complexities of irrigation areas mean that such models are impractical. A quasi-analytical model, benchmarked against FEFLOW, has been developed to represent these processes and could be used in modelling applications elsewhere.