Yield gap analysis of US rice production systems shows opportunities for improvement

Many assessments of crop yield gaps based on comparisons to actual yields suggest grain yields in highly intensified agricultural systems are at or near the maximum yield attainable. However, these estimates can be biased in situations where yields are below full yield potential. Rice yields in the US continue to increase annually, suggesting that rice yields are not near the potential. In the interest of directing future efforts towards areas where improvement is most easily achieved, we estimated yield potential and yield gaps in US rice production systems, which are amongst the highest yielding rice systems globally. Zones around fourteen reference weather stations were created, and represented 87% of total US rice harvested area. Rice yield potential was estimated over a period of 13–15 years within each zone using the ORYZA(v3) crop model. Yield potential ranged from 11.5 to 14.5 Mg ha−1, while actual yields varied from 7.4 to 9.6 Mg ha−1, or 58–76% of yield potential. Assuming farmers could exploit up to 85% of yield potential, yield gaps ranged from 1.1 to 3.5 Mg ha−1. Yield gaps were smallest in northern California and the western rice area of Texas, and largest in the southern rice area of California, southern Louisiana, and northern Arkansas/southern Missouri. Areas with larger yield gaps exhibited greater annual yield increases over the study period (35.7 kg ha−1 year −1 per Mg yield gap). Adoption of optimum management and hybrid rice varieties over the study period may explain annual yield increases, and may provide a means to further increase production via expanded adoption of current technologies.