Drivers of spatial and temporal variation in soybean yield and irrigation requirements in the western US Corn Belt

• We assessed yield and irrigation amount in soybean fields using on-farm data.
• Producers are achieving yields relatively close to the attainable yield.
• Variation in rainfed yield was related to rainfall during the reproductive phase.
• Sowing date explained most of the inter-annual variation in irrigated yield.
• Efficiency in use of rainfall versus irrigation water to produce yield was similar.

Both rainfed and irrigated soybean production are important in Nebraska (western US Corn Belt), accounting for a respective 48 and 52% of the state's soybean production of 7 Mt on a respective 55 and 45% share of the state soybean area of 1.9 Mha. To date, no assessment of factors that may account for regional and inter-annual variation in yield and irrigation amount has been performed. To accomplish that objective, we evaluated a database containing on-farm field yields and total irrigation amount used in those fields. These data have been collected annually from ca. 1000 soybean fields in six regions of Nebraska during the past eight years. Distributions of farm yield and irrigation amount were analyzed and the impact of selected weather variables and key management factors on these two variables was assessed. For irrigated soybean, attainable yields were estimated from the 95th percentile of the yield distribution, and yield gaps were then calculated as the difference between the attainable yield and average farm yield. The interquartile range for yield and irrigation amount was used as a measure of the management gap between skillfully and sub-optimally managed fields. Distribution of irrigated yield and irrigation amount were skewed, indicating that many producers achieved yields relatively close to the attainable yield, but also that irrigation in excess of the amount needed may have occurred in an important fraction of the total fields. Variation in rainfed yield was strongly related to July–August total rainfall and seasonal water deficit, but, in contrast, no single meteorological factor could consistently explain variation in irrigated yield. In fact, sowing date explained most of the observed inter-annual variation in irrigated yield in all regions. Amount of irrigation applied in each year depended on both rainfall and reference evapotranspiration. Efficiency in use of irrigation water versus rainfall to produce seed yield was remarkably similar. Across regions, attainable yield of irrigated soybean in NE averaged 4.7 Mg ha−1, with the yield gap averaging about 16% of the attainable yield. Variation in yield and irrigation amount among fields was more important than variation among years and even regions. Further research is needed to identify the causes for this observed field-to-field variation in yield and irrigation amount within the same year and region.