An examination of the Stafford Smith-Morton ecological model: a case study in the Tanami Desert, Australia

Abstract

The pattern of substrate, climatic, vegetation and fire features in the Tanami Desert were considered in relation to the ecological model for arid Australia proposed by Stafford Smith and Morton. The nature and accuracy of spatial data used to describe and quantify the pattern of the landscape features were also examined. Components of the ecological model were accurately reflected in the study area. For example, substrates identified as the most productive amounted to less than 8% of the region, and there was substantial spatial and interannual variation in rainfall. However, a strong climatic gradient was also evident in the study area, a feature not accommodated for in the model proposed by Stafford Smith and Morton. Vegetative ground and shrub cover decreased from north to south and was strongly associated with increasing aridity and lower maximum and minimum temperatures. Spinifex (Triodia spp.) cover showed a curvilinear response. The spatial data for both substrate and fire history were reasonably accurate (around 90%) when compared with ground-truthed data, and is considered suitable to reflect ecological pattern and process in the Tanami Desert. Both the adequacy of the ecological model and accuracy of spatial data are important issues to consider before the development of statistical modelling for prediction of species distribution.