America's Farming Future: The Impact of Climate Change on Crop Yields

Watch a video Describing my Project


Crop yields are strongly dependent on average summertime temperatures and extreme heat waves, both of which are projected to increase in the coming century. Here I create a statistical model to predict US yields to 2100 for three crops: maize, soybeans, and rice, for both a low and high-emissions future scenarios (RCP 4.5 and 8.5) in order to place a monetary cost on climate change. The model is based on linear regressions between historical crop yields and daily weather observations since 1970 for every county in the US. After computing correlations between the crop yields and numerous weather statistics, yields were found to be most strongly dependent on heat waves, summer average temperature, and killing degree days. Geographically, counties further south are more sensitive to heat extremes, implying that growing regions will shift northwards in the future. The model shows that climate change will have a strong influence on maize and soybean yields, and less on rice. Crop yields are predicted to decrease by 3.8%, 2.4%, and 0.83% per decade for maize, soy, and rice respectively for a high emissions scenario, and about half as much for a low emissions scenario, compared to a historical increase of 24%, 18%, and 17% per decade since 1970 due to improvements in plant breeds and farming practices. Climate change results in a loss of $23 billion for maize and $11.5 billion for soybeans per year in 2100 for the high emissions scenario, in today’s prices. This study demonstrates the importance of accounting for future costs of climate change when choosing today’s energy policies, and motivates continued improvements in agricultural technology to compensate for warming temperatures.