Contributions of historical changes in sowing date and climate to U.S. maize yield trend: An evaluation using large-area crop modeling and data assimilation

Abstract

The consequences of observed changes in climate and management on yield trends in major crop-producing regions have implications for future food availability and access. We presents an assessment of the impacts of historical changes in sowing date and climate to the maize yield trend in the United States (U.S.) Corn Belt from 1980 to 2006, using large-area crop modeling and data assimilation technique (the model optimization based on the Markov Chain Monte Carlo method). The model calibrated at a regional scale successfully captured the major characteristics of the reported changes in yield and the timing and length of maize growth periods over the Corn Belt. The simulation results using the calibrated model indicate that while the climate change observed for that period likely contributed to decrease the yield trend, the positive contribution from the reported earlier shift of sowing date offset the negative impacts. With given spread in the assessment results across previous studies and this study, the credence of the conclusion that the negative impacts of the climate change on the U.S. maize yield trend are more likely attributed to the decreasing growing-season precipitation trend than to the temperature trend increased. This study addressed an emerging use of large-area crop modeling and data assimilation to attribute observed change in crop yield trend to climate and management.