Detection of Thermal Features from Space at Indonesian Volcanoes Using the Medium Spatial Resolution ASTER Sensor (CE 2000-2020)

Thermal infrared (TIR) satellite observations enable volcanic thermal features (VTFs) to be monitored routinely and rapidly on a global scale. Routine analysis and detection of VTFs are available globally for low spatial resolution (>1km/pixel) TIR sensors, however medium resolution (90m/pixel) ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) TIR data from have yet to be systematically examined for VTFs at all active Indonesian volcanoes. Documenting VTFs over multiple decades allows for quantification of background volcanic activity and its relation to unrest or eruption. We expect the higher spatial resolution data will reveal more than the currently reported 30 volcanoes with VTFs or 22% of potentially active (Holocene and Pleistocene with recent thermal activity, according to the Smithsonian Institution) volcanoes in the country. We extended the ASTER Volcanic Thermal Output Database (AVTOD, Reath et al., 2019) to Indonesia by manually examining cloud-free night-time images over 135 potentially active volcanoes from CE 2000-2020. Each volcano is classified to: 1. those with VTFs, 2. those with potential VTFs that are inconclusive due to a lack of validation sources, and 3. those without measurable VTFs. We found at least 54 volcanoes with VTFs, despite the low proportion of cloud-free images (28%). Of these, 24 volcanoes have VTF detections from space documented for the first time (to our knowledge) and 29 volcanoes have VTFs not related to eruptive episodes (e.g., fumarolic VTFs). A further 10 volcanoes have inconclusive VTFs and 14 have volcanic lakes. With the inclusion of hotspots detected by ASTER, the percentage of Indonesian volcanoes with recorded satellite-detectable thermal features increases from 22% to 40%, similar to that in Latin America (26%, Reath et al., 2019) and the USA (33%, Reath et al., 2021). This multi-decadal country-wide survey highlights the potential of TIR monitoring and detection of low-temperature volcanic features that may be eruption precursors, and presents the possibility of using this database to improve the sensitivity of machine learning algorithms in detecting more subtle VTFs....