Identifying the Most Effective Catchment Management Options

Catchment management strategies have been developed throughout Australia to protect, preserve and reclaim the anthropocentric value of natural resources. These catchment management strategies commonly encourage landowners to adopt a range of interventions to achieve numerous outcomes. However, it is difficult for landowners, and supporting catchment management staff, to identify which management options will be most effective in achieving the desired management outcomes for the landowner's specific conditions. Bayesian networks have been used to model catchment management systems to assess the impact of management intervention and can also be used to identify which management options are critical to the achievement of desired catchment outcomes. This paper describes an approach used to identify critical management interventions in the Shepparton Irrigation Region Land and Water Salinity Management Plan using a Bayesian network model and mutual information analysis. A Bayesian network model was used to predict the condition of management options for site specific conditions if management objectives were to be achieved. Mutual information was used to assess the information content of and conditional dependencies between predicted probability distributions describing the condition of management options. The information content of the predicted probability distributions was then used to rank the management options in order of their effectiveness and identify the strength of interactions between management options. The results indicate that the most effective management options in the Shepparton Irrigation Region are the installation of a farm reuse system, the improvement of irrigation management and the conversion of surface irrigation to pressurised irrigation. These management options were shown to the most effective across all of the case-study area suggesting that the management options promoted by the Shepparton Irrigation Region Land and Water Management Salinity Plan are robust.