FP_ILM: Extremely fast volcanic SO2 plume height retrieval based on S5P/TROPOMI data using inverse learning machines

We present here a novel method for the precise and extremely fast retrieval of volcanic SO2 layer height (LH) based on S5P/TROPOMI data. We have developed the Full-Physics Inverse Learning Machine (FP_ILM) algorithm using a combined principal components analysis (PCA) and neural network approach (NN) to extract the information about the volcanic SO2 LH from high-resolution UV backscatter measurement of TROPOMI aboard Sentinel-5 Precursor.

The SO2 LH is essential for accurate determination of SO2 emitted by volcanic eruptions. So far UV based SO2 plume height retrieval algorithms are very time-consuming and therefore not suitable for near-real-time applications. The FP_ILM approach however enables for the first time to extract the SO2 LH information in a matter of seconds for an entire S5P orbit and thus applicable in NRT application.

The FP_ILM SO2 LH product is developed as part of ESA’s ‘Sentinel-5p+ Innovation - SO2 Layer Height project’ (S5P+I: SO2 LH) project, dedicated to the generation of an SO2 LH product and its extensive verification with collocated ground- and space-born measurements.