This paper presents the results from the development of a drought forecasting model at seasonal to interannual time scales for the Murray-Darling Basin, Australia. The Murray-Darling river system has a catchment area of 1.073 million km2, or nearly 14% of Australia’s land area and is the fourth longest river system in the world. About half of Australia’s gross primary production comes from the land and water resources of the basin. However, Australia has a highly variable climate consistent with its location largely within the semi-arid subtropical latitudes of the southern hemisphere. Consequently, droughts are an accepted feature of the climate in this part of the world and the ability to reliably predict such events would provide considerable improvement in the management of the basin’s water resources. The forecasting methodology involves the spatial analysis of historical records of precipitation, normalized difference vegetation index (NDVI), El Nino/Southern Oscillation (ENSO) indicators and the standardized precipitation index (SPI). To characterize the spatial variability in the basin, an approach involving functional texture analysis was adopted. The results of these analyses were used to formulate an adaptive prediction model that can provide reliable operational forecasts at adequate lead times for effective water resource management.