Probabilistic retrieval of volcanic SO 2 layer height and partial column density using the Cross-track Infrared Sounder (CrIS)

Abstract. During most volcanic eruptions and many periods of volcanic unrest, detectable quantities of sulfur dioxide ( SO2 ) are injected into the atmosphere at a wide range of altitudes, from ground level to the lower stratosphere. Because the fine ash fraction of a volcanic plume is, at times, colocated with SO2 emissions, global tracking of volcanic SO2 is useful in tracking the hazard long after ash detection becomes dominated by noise. Typically, retrievals of SO2 vertical column density (VCD) have relied heavily on hyperspectral ultraviolet measurements. More recently, infrared sounders have provided additional VCD measurements and estimates of the SO2 layer altitude, adding significant value to real-time monitoring of volcanic emissions and climatological analyses. These methods can provide fast and accurate physics-based retrievals of VCD and altitude without regard to solar irradiance, meaning that they are effective day and night and can observe high-latitude SO2 even in the winter. In this study, we detail a probabilistic enhancement of an infrared SO2 retrieval method, based on a modified trace gas retrieval, to estimate SO2 VCD and altitude probabilistically using the Cross-track Infrared Sounder (CrIS) on the Joint Polar Satellite System (JPSS) series of satellites. The methodology requires the characterization of real SO2 -free spectra aggregated seasonally and spatially. The probabilistic approach replaces altitude and VCD estimates with probability density functions for the layer height and the partial VCD at multiple heights, fully quantifying the retrieval uncertainty and allowing the estimation of SO2 partitioning by layer. This framework adds significant value over basic VCD and altitude retrieval because it can be used to assign probabilities of SO2 occurrence to different atmospheric intervals. We highlight analyses of several recent significant eruptions, including the 22 June 2019 eruption of Raikoke volcano, in the Kuril Islands; the mid-December 2016 eruption of Bogoslof volcano, in the Aleutian Islands; and the 26 June 2018 eruption of Sierra Negra volcano, in the Galapagos Islands. This retrieval method is currently being implemented in the VOLcanic Cloud Analysis Toolkit (VOLCAT), where it will be used to generate additional cloud object properties for real-time detection, probabilistic characterization, and tracking of volcanic clouds in support of aviation safety.