| Abdullah Khan |
Abstract
Utilizing SAS ODA web application, this paper analyzes a panel dataset regarding corn yields, surface temperatures, and precipitation in the South Atlantic Census Region. The contiguous eight states that belong to the U.S. South Atlantic census region include Delaware, Florida, Georgia, Maryland, North Carolina, South Carolina, Virginia, and West Virginia. Corn (Maize) is a versatile crop utilized both as a grain and vegetable. However, according to a few recent agro-climatic simulation studies, corn and other agricultural products' yields may be adversely impacted due to climate change. Most of these studies rely on data representing larger spatial units. Recognizing the imperative of microclimate studies, this paper analyzes county-level data, state-level data, and census division-level data to examine trends of corn yields, surface temperature, and precipitation and explore direction, magnitude and statistical significance of correlation coefficients and regression coefficients of variables of interest. At the individual state-level, most of the states within the South Atlantic division displayed positive correlations between corn yields and temperatures and corn yields and precipitation with an exception for two states. West Virginia data showed a statistically significant negative correlation between temperature and corn yield. Florida showed a negative correlation between temperature and corn yield, but this result is not statistically significant. According to the regression results, for the South Atlantic division the estimated coefficients of temperature variable are negative and statistically significant across the baseline model, and the variant models with log-transformed variables and annual growth rate variables.
Key words: SAS, corn yields, surface temperature, precipitation, U.S. South Atlantic Census region, Data Visualization, Correlation, Trends, Regression |