Assessing the spatial impact of climate on wheat productivity and the potential value of climate forecasts at a regional level

Abstract

Quantification of the spatial impact of climate on crop productivity and the potential value of seasonal climate forecasts can effectively assist the strategic planning of crop layout and help to understand to what extent climate risk can be managed through responsive management strategies at a regional level. A simulation study was carried out to assess the climate impact on the performance of a dryland wheat-fallow system and the potential value of seasonal climate forecasts in nitrogen management in the Murray-Darling Basin (MDB) of Australia. Daily climate data (1889–2002) from 57 stations were used with the agricultural systems simulator (APSIM) to simulate wheat productivity and nitrogen requirement as affected by climate. On a good soil, simulated grain yield ranged from 7 t/ha in the east border regions. Optimal nitrogen rates ranged from 200 kgN/ha/yr. Simulated gross margin was in the range of –$20/ha to $700/ha, increasing eastwards. Wheat yield was closely related to rainfall in the growing season and the stored soil moisture at sowing time. The impact of stored soil moisture increased from southwest to northeast. Simulated annual deep drainage ranged from zero in western inland to >200 mm in the east. Nitrogen management, optimised based on ‘perfect’ knowledge of daily weather in the coming season, could add value of $26∼$79/ha compared to management optimised based on historical climate, with the maximum occurring in central to western part of MDB. It would also reduce the nitrogen application by 5∼25 kgN/ha in the main cropping areas. Comparison of simulation results with the current land use mapping in MDB revealed that the western boundary of the current cropping zone approximated the isolines of 160 mm of growing season rainfall, 2.5t/ha of wheat grain yield, and $150/ha of gross margin in QLD and NSW. In VIC and SA, the 160-mm isohyets corresponded relatively lower simulated yield due to less stored soil water. Impacts of other factors like soil types were also discussed.