Analyzing the Relationship Between Ultraviolet Aerosol Index and Aerosol Optical Depth to Fill in Missing Data from a 2021 Wildfire

Abstract

Aerosol Optical Depth (AOD) is the measure of aerosols in the air during a certain period of time and indicates relatively hazy parts of the atmosphere. Yet, in many areas, when plotting the AOD on a map, it measures no data. As an alternative, the Ultraviolet Aerosol Index (UVAI), also determines the presence of dust and smoke, two ultraviolet-absorbing aerosols. I hypothesize that there is a way to fill in the “no data” parts of the AOD map using the data already given, including UVAI. To do this, I observed a specific time period (7/10/2021 - 7/22/2021) in which wildfires on the west coast and in Canada were prevalent. Then, I utilized Python libraries to create scatterplots calculating the correlation, as well as a linear regression equation, to analyze the relationship between UVAI and AOD for these dates. Using the linear regression equation, I could plug in UVAI values to get new, potentially correct, AOD values. I replotted the data with these new values and the results were drastically more coverage than the original AOD maps. Therefore, since the correlation between UVAI and AOD was high, the relationship between them was statistically significant enough to fill in the missing data. This new data gives a better understanding of where these wildfires originated and help find points that may have been missed by the photometer.