Estimation of Earthquake Forecasting Probability Based on Satellite Thermal Anomalies

An approach is proposed for estimating the probability of earthquake forecasting based on pre-seismic thermal anomalies detected through satellite remote sensing products. A spatially self-adaptive scheme identifies optimal anomaly recognition criteria for refining calculated anomalies. Bayesian probability theory is then applied to calculate the conditional probability of earthquake occurrence given the observed anomaly. This probabilistic approach accounts for uncertainties associated with anomaly-based forecasting. Analysis of global seismic regions reveals that certain parameters, such as water vapor anomaly, exhibit superior performance in terms of higher probability gains. The adoption of the Bayesian method to incorporate satellite observations of thermal anomalies into operational earthquake forecasting systems could significantly enhance forecast capacity. The probabilistic framework addresses the limitations of anomaly correlation methods and provides more quantitative estimates of earthquake likelihood.