APPLICATIONS OF THE OZONE MONITORING INSTRUMENT IN OBSERVING VOLCANIC SULFUR DIOXIDE PLUMES AND SULFATE DEPOSITION
APPLICATIONS OF THE OZONE MONITORING INSTRUMENT IN OBSERVING VOLCANIC SULFUR DIOXIDE PLUMES AND SULFATE DEPOSITION
Files
Publication or External Link
Date
2021
Authors
Fedkin, Niko Markovich
Advisor
Dickerson, Russell R
Citation
DRUM DOI
Abstract
Sulfur dioxide (SO2), a gas emitted by both volcanoes and anthropogenic activity, is a major pollutant and a precursor to sulfate aerosols. Sulfates can be deposited back to the ground where they have adverse impact on the environment or reside in the stratosphere as aerosols and affect radiative forcing. I investigated two components that stem from SO2: the deposition of sulfate, and the remote sensing of the SO2 layer height, important for aviation safety and chemical modeling. In the first study, I used column SO2 data from the Ozone Monitoring Instrument (OMI), and sulfate wet deposition data from the National Atmospheric Deposition Program to investigate the temporal and spatial relationship between trends in SO2 emissions and the downward sulfate wet deposition over the northeastern U.S. from 2005 to 2015. The results showed that emission reductions are reflected in deposition reductions within this same region. Emission reductions along the Ohio River Valley led to decreases in sulfate deposition not only in eastern OH and western PA, but also further downwind at sites in Delaware and Maryland. The findings suggested that emissions and wet deposition are linked through not only the location of sources relative to the observing sites, but also photochemistry and weather patterns characteristic to the region in winter and summer. The second part of this dissertation focuses on SO2 layer height retrievals and their applications. To this end I applied the Full Physics Inverse Learning Machine (FP-ILM) algorithm to OMI radiances in the spectral range of 310-330 nm. This approach utilized radiative transfer calculations to generate a large dataset of synthetic radiance spectra for a wide range of geophysical parameters. The spectral information was then used to train a neural network to predict the SO2 height. The main advantage of the algorithm is its speed, retrieving plume height in less than 10 min for an entire OMI orbit. I also compared the SO2 height retrievals to other data sources and explored some potential applications, in particular their use in volcanic SO2 plume forecasts and estimating the total mass emitted from volcanic eruptions.