Robust Satellite Techniques for short-term seismic hazard forecast over California using a multi-year continuous GOES-17/18 TIR record

Robust Satellite Techniques (RST) applied to long, homogeneous time series of thermal infrared (TIR) satellite radiances have been used for almost three decades to detect space–time anomalies potentially related to the preparation phase of strong earthquakes. Previous multi-year investigations—often based on background datasets longer than a decade and carried out across different continents and tectonic settings—showed that over two thirds of the identified, space–time persistent anomalies fall within a predefined spatial and temporal window around earthquakes with magnitude M ≥ 4, with a false-positive rate below one third. In addition, Molchan error diagram analyses provided evidence that the observed association departs from random-guess behavior.After comprehensive tests performed over Greece, Italy, Turkey, and Japan, we focus here on California and critically discuss the most recent advances achieved using GOES observations. Preliminary experiments based on GOES-17 TIR radiances and a four-year background (2019–2022) yielded encouraging results for M6+ earthquakes, with a gain probability of approximately 1.6 and 67% of events successfully alerted. However, the four-year limit reflects data-availability constraints linked to the operational discontinuity and subsequent decommissioning of GOES-17, while RST is known to require background datasets that are both homogeneous and statistically robust.To increase the statistical significance of the assessment, we extend the analysis to a seven-year dataset (2019–2025) by integrating GOES-17 and GOES-18. The feasibility of this integration was preliminarily verified using the temporal overlap between the two sensors, checking for the absence of significant differences and ensuring consistency with the homogeneity requirements of RST. We finally discuss the results obtained from the extended GOES-17/18 record and their implications for short-term seismic hazard experiments.