Premium
Evaluating the Contribution of Weather to Maize and Wheat Yield Trends in 12 U.S. Counties
Author(s) -
MaltaisLandry Gabriel,
Lobell David B.
Publication year - 2012
Publication title -
agronomy journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.752
H-Index - 131
eISSN - 1435-0645
pISSN - 0002-1962
DOI - 10.2134/agronj2011.0220
Subject(s) - dssat , yield (engineering) , crop simulation model , environmental science , crop yield , crop , agronomy , climate change , weather patterns , weather station , yield gap , zea mays , climatology , geography , meteorology , biology , ecology , materials science , metallurgy , geology
Evaluating the contribution of weather and its individual components to recent yield trends can be useful to predict the response of crop production to future climate change, but different modeling approaches can yield diverging results. We used two common approaches to evaluate the effect of weather trends on maize ( Zea mays L.) and wheat (Triticum aestivum L.) production in 12 U.S. counties, and investigate sources of disparities between the two methods. We first used the Decision Support System for Agrotechnology Transfer (DSSAT) model from 1984 to 2008 to evaluate the contribution of weather changes to simulated yield trends in six counties for each crop, each county being located in one of the top 10 U.S. producing states for that crop. A parallel analysis was conducted by multiplying inter‐annual weather sensitivity of county‐level yields with observed weather trends to estimate weather contributions to empirical yield trends. Weather had a low (maize) to high (wheat) contribution to simulated yield trends, with rain having the largest effect. In contrast, weather and rain had lower contributions to empirical yield trends. Along with evidence from previous studies, this suggests that DSSAT may be too sensitive to water thus inflating the importance of rain. Moreover, the time period used to compute yield trends also had a large effect on the importance of weather and its individual components. Our results highlight the importance of using multiple computation approaches and different time periods when estimating weather‐related yield trends.