Estimating Spatial Heterogeneity in Hay Yield Responses to Weather Variations in Oklahoma

Hay is an important field crop in the U.S., with over 54 million harvested acres in 2015. In many southern states, hay is an important input for cattle production, and reducing forage costs is crucial for improving the profitability of livestock operations. It is well known that crop yields and quality are significantly influenced by weather variations, which can have different impacts across geographical regions and over years. This study quantifies possible heterogeneous impacts in hay yield responses to weather variations in Oklahoma hay yields. The paper uses panel data on hay yields for Oklahoma’s 77 counties from 1977 to 2007. The weather variables include temperature and precipitation. A geographically weighted regression (GWR) approach is used to estimate the local effects of weather variations on hay yields in geographic regions. The GWR allows the relationships between hay yields and weather variations to vary across geographic regions. Results suggest that geographic variation does exist in hay’s response to weather. Accordingly, it is important to model hay production within a framework that allows weather response parameters to vary. Hay producers can reduce their production risk by incorporating models that permit geographical variation in how the local climate impacts yields.