Secondary salinisation of soil and water resources is an acute natural resources management issue over large parts of Australia. This paper focuses on the situation of the Liverpool Plains catchment in Northern New South Wales which covers 1.2 million hectares. More than 1,000 farms operate in the catchment. 200,000 hectares of prime floodplain land are at risk of becoming salt-affected due to rising saline groundwater tables.

The Liverpool Plains are famous for their vast alluvial floodplains where self-mulching black clays provide the production basis for an extensive dryland cropping industry. Irrigation is practised on a small proportion of the plains where high-yielding aquifers allow groundwater pumping. The plains are semi-surrounded by ranges and interspersed with hills. These areas are farmed to some extent but predominantly grow cattle.

The hydrogeology of the catchment constitutes a predisposition for rising groundwater tables due to horizontal and vertical constrictions to lateral groundwater flow. However, for widespread dryland salinisation to occur, the sufficient condition is an increase in recharge to the groundwater system. This can arise from increased rainfall or a change in the vegetation cover to the effect that a smaller proportion of the rain is evapotransporated. Significant vegetation change has taken place in the time since European settlement which saw phases of extensive clearing of native vegetation and pastures have been increasingly replaced with cropping systems.

In economic terms, floodplain salinity constitutes a complex externality problem because of spatial effects and time lags associated with the biophysical processes involved. A large part of the water that is causing salinity comes from upstream locations and has been recharged years earlier. The external costs to affected farmers are large and the associated loss of soil productivity may jeopardise the viability of their businesses in less than a decade after emergence of the problem.

This paper presents the concept of a dynamic spatial optimisation model for analysing catchment management problems. The model is called SMAC. Applying biophysical criteria, the catchment is spatially discretised into four distinct areas. In addition to its land use and hydrogeological characteristics, each area has a model farm attributed to it which represents its typical farming structures. Price and cost functions apply. Rising groundwater tables cause salinisation which causes yield decline, reduction in land use options, and loss of land value. The objective function maximises present net income of the catchment from agriculture and residual land value over a 30-year period.

The paper also presents the results of a series of model runs which explore the trade-offs involved in catchment management for soil salinity management and the extent to which external costs are optimally internalised. Three sets of trade-offs govern decision making for salinity management. The first set trades off costs arising from soil salinisation against those arising from salinity prevention. The second set compares the efficiency of recharge-reduction measures in upstream areas with those in the plains. The third set compares the efficiency of the range of water table management measures at each site within the systems setting.

The results are based on best available biophysical data. Despite the resulting preliminary character of the analysis some major conclusions can be drawn with respect to the economically optimal level of salinity and salinity management strategies. Sensitivity testing of results to assumed costs and climatic variability is shown.