Extreme weather events have nuanced implications for crop producers. While they can reduce local yields, widespread production losses often drive price increases. This study presents a panel approach that accounts for the price-yield correlation to assess the impact of such events on crop revenues, focusing on U.S. corn and soybeans. It conducts two key analyses: (1) quantifies the revenue impacts of the historic 1988 and 2012 U.S. heatwaves, and (2) examines the implications of climate change on crop revenue variability. The results show that compensatory price increases often substantially offset yield losses, especially when price responsiveness to supply shocks is strong. In particular, U.S. corn in 2012 and soybeans in 1988 saw crop revenues rise by more than 8% compared to normal weather conditions, while U.S. corn in 1988 and soybeans in 2012 experienced decreases of no more than 4%. The study highlights the importance of crop-specific and time-varying price responsiveness to supply shocks. Furthermore, it demonstrates that if growing-season weather during 1997–2019 had exhibited the volatility projected for 2036–2065 under a moderate emissions scenario, revenue variability for corn and soybeans in median U.S. counties would have increased by more than 60%, with more pronounced impacts in regions outside the major Corn Belt. These findings underscore the significant economic risks posed by climate change–induced variability in agricultural revenues.

Record-high prevented planting of staple crops such as corn and soybeans in the United States (US) Corn Belt due to heavy rainfall in recent years has spurred concern over crop production, as growing evidence suggests winter and spring precipitation extremes will occur more frequently in the coming decades. Using county-level data, we examine the effects of planting-season water surplus—precipitation minus evaporative demand—on prevented planting of corn and soybeans in the US Corn Belt. Using monthly water surplus data, we show significant impacts of excess moisture on preventing planting and suggest a 58%–177% increase in prevented planting during the months of April–June per standard deviation increase in water surplus. Downscaled climate change projections are used to estimate future changes in prevented planting during the mid-century (2036–2065) under the moderate emission scenario (RCP4.5). Our model predicts a decrease in prevented planting of approximately 111,000 acres (12%) for corn and 80,000 acres (16%) for soybeans in the US Corn Belt, relative to historical levels from 1950 to 2005. However, if we consider only precipitation and disregard evaporative demand, the alternative model indicates an increase of approximately 260,000 acres (30%) for corn and 86,000 acres (19%) for soybeans. Geographically, we find that prevented planting will slightly increase in some parts of Iowa, Minnesota, and Wisconsin and generally decrease in the other parts of the US Corn Belt. This work collectively highlights the value of incorporating water surplus data in assessing prevented-planting impacts and is the first known study to examine changing risk of prevented planting under future climate scenarios that may help inform adaptation efforts to avoid losses. 

As climate change increases the frequency and severity of extreme heat events, farmers are expected to face greater variability in crop yields. Using 10 million field-level observations, this study examines how farmers in the U.S. Corn Belt adjust their crop rotation decisions—focusing on corn and soybeans—in response to yield shocks that are largely driven by serially uncorrelated weather. The findings suggest that farmers tend to shift toward soybeans—a crop more tolerant of extreme heat—following a hotter-than-normal growing season, regardless of which crop experienced greater yield loss. However, I find little evidence of effective adaptation: years in which farmers plant more soybeans are not, on average, hotter than usual. Moreover, the effects of shocks occurring more than one year prior are substantially smaller than those of the most recent shock. This pattern is consistent with recency bias, whereby farmers overweight recent experiences when making decisions. Overall, the results suggest that short-term weather shocks can have lingering effects on land-use decisions in commercial agricultural systems, with potential implications for both agricultural production and environmental outcomes such as water quality. 

Using crime data from 21 major U.S. cities spanning 2007–20, this paper studies the impacts of wildfire smoke on crimes. Wildfire smoke simultaneously increases concentrations of multiple pollutants—PM2.5, NO2, and ozone—all of which are associated with heightened violent crime, and it introduces unique factors—such as odor and reduced visibility—underscoring the necessity of considering wildfire as a single treatment variable. Our results show that wildfire smoke significantly increases crime, with particularly notable impacts on violent and drug-related crimes. Our findings show that crime rates surge on the first and second consecutive days of smoke exposure but diminish from the third day onward, possibly due to adaptive behaviors to limit smoke exposure. On days with air quality alerts, crime falls, suggesting these alerts prompt individuals to adapt to smoke exposure. A back-of-the-envelope calculation estimates that eliminating wildfires could save approximately $619 million annually in reduced violent.

Technological advances in satellite image processing have made crop maps readily available over the last decade. Because of the diversity and complexity of crop production in California, however, reliable crop maps for the state are still scant. To fill this gap, we created field-level crop maps of California (hereinafter, Field-Level California Crop Map (FLCCM)) for 2007-2021. We leverage highly accurate ground-truth labels that exist in 2014, 2016, and 2018 to train our crop classifier using probability random forests. We then feed to our classifier the data for predictors that are available from 2007 to 2021. We release three types of crop predictions and their corresponding accuracy measures as well as class probability in three formats (.csv, .shp, .rds). Our training algorithm can be applied to other settings in which field-level ground-truth data are scarce but fine-resolution pixel-level data are relatively more abundant.