Evaluation of the Relationship between Land Surface Temperature and Ammonia Emissions in the Midwest – Preliminary Results

The purpose of this study is to determine how land surface temperature impacts ammonia emissions in the Midwest region of the United States. The motivation behind this research project is to determine if land surface temperature (LST) has an impact on ammonia emissions. Ammonia is a precursor gas for PM_2.5and is a major component in the global nitrogen cycle. In order to accomplish this goal, Python programming and a high-performance computing system were used to synthesize satellite data. The source for the land surface temperature data was the MOD11A1 product on the MODIS satellite and the ammonia data source was an algorithm generated from the AIRS/AMSU satellite. The data for each product was first plotted and visually analyzed for relationships. Data was plotted for the months of April-September in 2014 and 2015. These months were chosen because of the wide range of temperatures and because the agriculture growing season falls in these months. After visually inspecting the data, it was found that there was some form of correlation between the land surface temperature and ammonia emissions. Months with higher land surface temperatures generally saw higher amounts of ammonia emissions. In order to confirm this visual analysis, the data sets had to be manipulated in order to perform a true statistical analysis. The LST data was regridded so the resolutions of the two data sets matched. The ammonia data set was subsetted in order to look at areas of interest within the plots. Correlation plots were constructed for each month to determine if there was a statistical relationship between the two data sets. It was found that there was a weak relationship, with August 2015 having the highest R value of 0.47. Moving forward, further analysis will be performed on smaller regions of the plot to see if there are better correlations. To do this, time series will be extracted from the data, and additional statistical analyses will be performed.