Assessing exceptional drought with a cropping systems simulator: A case study for grain production in northeast Australia

The potential for cropping systems simulation to assist in the identification and quantification of 'exceptional circumstances' of drought impact on agriculture was assessed. The work focused on the grain production region of northeast Australia as a case study and used the crop-soil management simulator, APSIM (Agricultural Production Systems Simulator). Wheat and sorghum production was simulated for a range of cropping systems and locations and the results averaged as a notional per unit area 'farm production'. Moving averages over time bases from 1 to 4 years were used to examine variability and drought occurrence. Droughts over different time periods were identified in absolute terms (e.g. simulated production relative to a median or some lower percentile) or in probabilistic terms (e.g. the estimated recurrence interval or odds of experiencing a worse drought). The identification of droughts using agricultural systems simulation was contrasted with rainfall estimates of drought for the period of intel est and some differences in ranking were observed. As the simulated agricultural drought reflects the impact of factors such as rainfall timing, intensity, and amount and effects of soil water storage on cropping system performance, it is considered a superior measure of drought severity on farm performance. Sensitivity analyses revealed little to no effects of the choice of crop model or cropping system configuration (e,g, planting criteria) on the identification of exceptional drought circumstances. (C) 1998 Elsevier Science Ltd All rights reserved.