Yield response and water productivity of soybean ( Glycine max L.) to deficit irrigation and sowing time in south-eastern Australia

The aim of this study was to investigate irrigation strategies and sowing dates that would maximise soybean yield and water productivity. It is based on field experiments conducted during two seasons and simulation modelling. Irrigation treatments were 33%, 66%, 66% plus 100% during pod development and pod -fill stages, and 100% of crop evapotranspiration (ET c ). In the first season experiment, cvs. Bidgee and Snowy were sown on 15 Nov. In the second season, cv. Bidgee was sown on 15 Nov and 15 Dec. Five sowing dates (1, 15, 30 Nov and 15, 31 Dec) and eight irrigation scenarios were analysed in-silico using the crop model APSIM. For the simulations, the first four irrigation scenarios were set by applying 50%ET c during one of the four growth stages: vegetative stage, flowering stage, pod development and pod -filling stage, and maturity stage. The other five irrigation treatments were 0%ET c , 25%ET c , 50%ET c , 75%ET c , and 100%ET c throughout the growing season. Soil water content and above -ground dry matter were measured at regular time intervals. Seed yield, 100 -seed weight, oil and protein contents were determined at harvest. Water deficit during pod development and pod -filling stage had significant effect on seed yield. The flowering stage was the next most sensitive stage for water deficit. During both cropping seasons, the 33% treatment yielded 51% of the fully irrigated reference. The latter had a significantly higher water productivity than all the deficit treatments. Early -sown soybean had higher yield than the late sown soybean. Sowing as late as early December was found to be still suitable for double cropping without significant yield loss. The result of this study is particularly useful for soybean farmers in water -scarce regions, such as south-eastern Australia, who practice double cropping with a tight cropping calendar.