First retrievals of volcanic SO2 heights from hyper-spectral satellite UV measurements
Kai Yang(1), Xiong Liu(1), Nickolay Krotkov(1), Arlin Krueger(2), Simon Carn(3) and Pawan K. Bhartia(4)
(1) GSFC/NASA and GEST/UMBC, 8800 Greenbelt Rd, Greenbelt, MD 20771, United States
(2) JCET/UMBC, 5523 Research Park Drive, Baltimore, MD 21228, United States
(3) Michigan Technological University, 1400 Townsend Drive, Houghton, MI 49931, United States
(4) GSFC/NASA, 8800 Greenbelt Rd, Greenbelt, MD 20771, United States
Sulfur dioxide (SO2) emissions, from both anthropogenic and volcanic sources, perturb atmospheric composition and chemistry. While in the planetary boundary layer (PBL), SO2 and sulfate aerosol (formed upon oxidation of SO2) cause detrimental effects on environment and human health. Once in the free troposphere and stratosphere, sulfate aerosols may have a long atmospheric residence time, from days to years depending on the injection altitude, and affect radiative forcing of climate directly by reflecting incoming solar radiation and indirectly by modifying the reflectivity, precipitation and lifetime of clouds. To quantify climatic impacts of sulfate aerosols, it is important to measure both the strength and altitude of SO2 emissions. In this presentation, we will describe results of our recent work aimed at understanding physical mechanisms through which the vertical distribution of an SO2 plume perturbs natural solar backscattered ultraviolet (BUV) radiances at the top of atmosphere. This new understanding results in a practical satellite retrieval technique, named Extended Iterative Spectral Fitting (EISF) algorithm for simultaneous retrieval of SO2 plume height and amount. The EISF algorithm reduces systematic errors of previous satellite BUV algorithms that rely on prescribed SO2 vertical distribution and an assumption of linear BUV response to SO2 loading. Using satellite BUV data for recent eruptions as examples, we demonstrate improvement of the retrieval accuracy of volcanic SO2 columns and first retrievals of the effective altitude of the SO2 plumes. Specifically, we will show results from applications of the EISF algorithm to observations from the Dutch-Finnish Ozone Monitoring Instrument (OMI). The OMI offers unprecedented high-sensitivity to SO2 by combining its hyper-spectral UV capability with high signal-to-noise ratio in SO2 sensitive UV spectral region and small footprint (13×24 km2 at nadir) measurements with daily global coverage. The examples show that a wide range of volcanic SO2 effective plume altitudes (from lower troposphere to lower stratosphere) can be estimated directly from hyper-spectral BUV measurements while improving the accuracy of the retrieved SO2 columns. These examples also show that hyper-spectral measurements reduce the impact of measurement noise, leading to higher precision. As a result, weaker SO2 loadings, such as those from anthropogenic emissions and volcanic degassing, can now be more reliably detected and quantified.