Accounting for Missing AOD Data with UVAI for Improved Aerosol Forecast in Wildfire/Dust Storm Cases

Authors

  • Jenny Chen Aspiring Scientists' Summer Internship Program Intern
  • Naphat Siripun Aspiring Scientists' Summer Internship Program Intern
  • Stephanie Song Aspiring Scientists' Summer Internship Program Intern
  • Charu Mehta Aspiring Scientists' Summer Internship Program Intern
  • Dr. Yunyao Li Aspiring Scientists' Summer Internship Program Mentor

DOI:

https://doi.org/10.13021/jssr2022.3469

Abstract

Aerosol Optical Depth (AOD) and UV Aerosol Index (UVAI) are aerosol extinction measures used in air quality forecasts for the amount of aerosol in the air. AOD measures both aerosol absorption and extinction at the visible spectrum but is unable to read areas with clouds/snow, while UVAI is only sensitive to aerosol absorption at the UV spectrum but is not limited to cloud-free areas. In this project I focused on filling missing AOD data with UVAI data using different correlation methods for 2 special cases: 1/15/2021, a thick dust storm around the Texas Panhandle, and 8/7/21 containing part of the 2021 California Dixie fire in NE California. The correlation between TropoMAER UVAI and VIIRS AOD was compared among different distance cutoffs to fire source and latitude/longitude boxes. The boxes worked best for the special cases, yielding correlation results of around r=0.82 for both the dust and fire cases. Using the linear regression equations from the resulting correlations, I replotted the data for both cases, which led to visibly wider coverages compared to the original AOD data. For the fire case specifically, most of the null values in the boxed region in which the correlation was calculated were filled in. In conclusion, the box method seemed to be effective in calculating correlations/regressions that produced plots with significant improvements in AOD coverage, especially for the fire aerosol case. This finding can be useful concerning aerosol forecasts as it addresses a disadvantage of AOD using existing data.

Published

2022-12-13

Issue

Section

College of Science: Department of Atmospheric, Oceanic & Earth Sciences

Categories