This paper describes (1) the development and validation of a native pasture model that accounts for species abundance, (2) the integration of the native pasture model into a catchment framework and (3) the likely biophysical trade-offs associated with increasing the proportion of native perennial pasture species within a pilot catchment. The rationale for developing the model was to build on the conceptualisation of contemporary pasture models and include new algorithms that better represent native grass species commonly found in temperate climate regions of Australia. The native pasture model is phenologically based and simulates both daily pasture growth and soil water dynamics. It was developed and validated against data from field experiments conducted at Wagga Wagga in NSW, Australia, for two native grass species, Austrodanthonia spp. (winter dominant C-3) and Bothriochloa macra (summer dominant C-4). The model predicted seasonal pattern and inter-annual variation of soil moisture contents and green leaf area index (GLAI) for various planting densities. There was a high correlation between the simulated and measured soil moisture contents (r(2) = 0.81, p < 0.0001 for Austrodanthonia spp; r(2) = 0.73, p < 0.0001 for Bothriochloa macra) and GLAI (r(2) = 0.66, p < 0.0001 for Austrodanthonia spp; r(2) = 0.83, p < 0.0001 for Bothriochloa macra) at actual measurement dates. The native pasture model was integrated into a physically-based catchment modelling framework (Catchment Analysis Tool, CAT) to provide a flexible platform for simulating the environmental impacts of various pasture management options and the likely consequences of land use changes on catchment dynamics. This catchment model comprises numerous vegetation modules (including cropping, forestry and alternative pasture models) of varying complexity. An example of how the native pasture model can be used within the CAT framework is described by the application to the Bet Bet catchment which has been identified as a high salt exporting catchment within the Murray Darling Basin of Australia. The Bet Bet catchment currently yields and exports annually approximately 36,056 ML and 20,000 tonnes of salt respectively. The grazing of unimproved pastures is the dominant land use within the catchment such that economically viable options for increasing perenniality in the catchment are limited to working with the native perennial grasses already present and targeted tree planting. The study evaluated the water yield, saturated area and salt load trade-offs associated with increasing the proportion of native perennial pasture species. Scenarios considered varying proportions of summer or winter dominant native pastures across the catchment in comparison to reafforestation. Results indicate that complete reafforestation of the grazing area would reduce both stream flow and salt exports, by 6,299 ML yr(-1) and 10,486 tonnes salt yr(-1) respectively. Complete catchment coverage of Bothriochloa macra was estimated to increase stream flow by 3,987 ML yr(-1) as well as to reduce groundwater inflows by 1,429 ML yr(-1)(2,745 tonnes salt yr(-1)). In contrast, Austrodanthonia spp. pasture was estimated to increase stream flow by 4,378 ML yr(-1) and increase salt export by 714 tonnes salt yr(-1). Targeted restoration of either summer or winter dominant native pastures within the steeper regions of the catchment with slopes greater than 5% (representing 11% of the catchment area) was considered. Results show that restoration of Bothriochloa macra in the upper catchment increased stream flow by 132 ML yr(-1) and reduced salt loads by 472 tonnes salt yr(-1), whereas restoration of Austrodanthonia spp. increased stream flow by 409 ML yr(-1) and reduced salt loads by 109 tonnes salt yr(-1). In contrast, reafforestation in the upper catchment reduced stream flow by 5,650 ML yr(-1) and salt loads by 8,689 tonnes salt yr(-1). Overall, results suggest that increasing the proportion of summer active native perennial grasses across large parts of the catchment may be an effective salinity management option for reducing recharge and salt loads to rivers. Integration of a validated native pasture model into the CAT framework is shown to provide the capability to simulate the long-term effects of land use changes in the landscape on catchment dynamics.
