Mechanisms and control of water logging and groundwater flow in the ‘Ucarro’ sub-catchment

Abstract

The occurrence of seasonal water logging was studied across a small (639 ha) catchment in the south-west of Western Australia. There are two groundwater systems in the catchment, a seasonal aquifer perched on the ‘B’ horizon of a duplex soil, and a deeper, perennial system which upslope is 6–8 m below the surface, but intersects the surface in the valley floor. The perched aquifer causes water logging across about 25% of the catchment. Water logging risk was predicted using a steady state hydrological model (Topog_Simul, [Water Resource Res. 22 (1986) 794]) and a hydrogeomorphic classification [Hydrogeological characterisation of catchments using Hydrogeomorphic Analysis of Regional Spatial Data (HARSD): characterisation of Axe Creek Catchment, Vic., Australia. In: Tanaguchi, M. (Ed.), Subsurface Hydrological Response to Land Cover and Land Use Change. Kluwer Academic Publishers, Dordrecht] and both models compared favourably with occurrence mapped from air photography. The patterns of water logging occurrence also compared well with simulations using a dynamic catchment model (Topog_Dynamic, [Modelling drainage and transient water logging in an agricultural catchment. In: Proceedings of the 25th Hydrology and Water Resources Symposium and 2nd International Conference on Water Resources and Environment Research, Brisbane, 6–8 July 1999. Water ’99. Publ. Inst. Eng. Aust., p. 999; Agric. Water Manage. 39 (1999b) 283]). The dynamic modelling showed that the impact on transient water logging of an existing drain and tree system in the catchment is only local to the area of the trees and drains, and does not contribute in a major way to the transient water logging further down the slope. However, this modelling, as well as long-term groundwater simulations (FlowTube, [Calibration and modelling of groundwater processes in the Liverpool Plains. CSIRO Land and Water Technical Report 5/00.]), showed that over time the trees would help lower groundwater relative to the “do nothing different” scenario. The long-term modelling also showed that a substantial portion of the catchment would have to be planted in order to have a major impact on groundwater discharge at the valley floor, and hence on salinity of both the soil and the stream.