Understanding cirrus clouds using explainable machine learning

• DOI: 10.48550/arxiv.2305.02090
• Publication Date: 2023-01-01
• Authors: Kai Jeggle; David Neubauer; Gustau Camps-Valls; Ulrike Lohmann
• Keywords: FOS: Computer and information sciences; Computer Science - Machine Learning; Shapley values; FOS: Physical sciences; QA75.5-76.95; eXplainable AI; Machine Learning (cs.LG); cirrus clouds; Environmental sciences; Physics - Atmospheric and Oceanic Physics; machine learning; 13. Climate action; Electronic computers. Computer science; SHAP; Atmospheric and Oceanic Physics (physics.ao-ph); GE1-350; explainable machine learning

Abstract Cirrus clouds are key modulators of Earth’s climate. Their dependencies on meteorological and aerosol conditions are among the largest uncertainties in global climate models. This work uses 3 years of satellite and reanalysis data to study the link between cirrus drivers and cloud properties. We use a gradient-boosted machine learning model and a long short-term memory network with an attention layer to predict the ice water content and ice crystal number concentration. The models show that meteorological and aerosol conditions can predict cirrus properties with R2 = 0.49. Feature attributions are calculated with SHapley Additive exPlanations to quantify the link between meteorological and aerosol conditions and cirrus properties. For instance, the minimum concentration of supermicron-sized dust particles required to cause a decrease in ice crystal number concentration predictions is 2 × 10−4 mg/m3. The last 15 hr before the observation predict all cirrus properties.