SSTAR: A user-friendly application to track subtle thermal anomalies at volcanoes

Over the past decades, satellite-based thermal monitoring of volcanoes has undergone a major transformation driven by new Earth-observing missions and advances in spectroscopic surface analysis, revealing diverse thermal responses of volcanic surfaces to subsurface processes and showing that many eruptions are preceded by subtle thermal anomalies. This growing body of evidence underscores the need for robust methods capable of detecting and tracking diffuse thermal unrest using existing satellite archives while fully exploiting the capabilities of current sensors in orbit. To meet this need, we present SSTAR (Subtle Surface Thermal Anomalies Recognizer), a user-friendly application designed to detect and analyze diffuse thermal anomalies, i.e., subtle surface warming on the order of ~1 K over large areas (several km²), using MODIS satellite data. Building on the statistical thermal anomaly detection framework of Girona et al. (2021) [https://doi.org/10.1038/s41561-021-00705-4], SSTAR operates at the pixel level to track the temporal evolution of thermal anomalies at specific sites and to map their spatiotemporal distribution across broad regions, incorporating dedicated filtering strategies to identify both long-term (years) and short-term (weeks) signals with uncertainty quantified through bootstrapping. We demonstrate the capabilities of SSTAR through its application to Shishaldin Volcano (Alaska), where the four eruptions that occurred over the past two decades are shown to have been systematically preceded by diffuse, low-amplitude thermal anomalies, highlighting the potential value of such signals as eruption precursors. SSTAR is distributed as a standalone application with an interactive interface accessible to non-specialists, while also providing full script access for MATLAB users who wish to adapt or extend the methodology for specialized applications. Beyond volcanology, it is expected to be useful for geothermal exploration, where the detection of faint and spatially coherent thermal anomalies may help identify subsurface fluid pathways and guide early-stage site characterization. An upcoming version will enable near-real-time tracking of diffuse thermal unrest, positioning SSTAR as a forward-looking tool for advancing satellite-based thermal monitoring of volcanic activity in the coming decades.