Diffuse Flank Degassing as the Dominant Source of the Large-Scale Thermal Anomaly Preceding the 2006 Augustine Eruption

Pre-eruptive, long-term, large-scale thermal anomalies detectable in 1 km resolution MODIS Thermal Infrared (TIR) radiance data have been consistently observed at long-dormant volcanoes years before eruptions. However, the physical mechanisms driving these signals remain unresolved. This study addresses a critical question: is the large-scale thermal anomaly primarily governed by localized high-temperature conduit heating or by spatially distributed, low-intensity heat release from diffuse magmatic degassing along volcanic flanks? Resolving this mechanism is vital for interpreting TIR data and for understanding heat and volatile transport during volcanic unrest.We investigate this question at Augustine Volcano during its 2006 eruption, where summit conduit warming preceded the large-scale thermal anomaly by approximately three months. To explain this temporal offset, we adopt a conceptual model following Zhan et al. (2022), based on magma ascent followed by conduit sealing. We simulate surface thermal evolution under two scenarios: (1) an area-integrated signal including both the conduit and flanks, and (2) a conduit-excluded signal (near-vent area, ~150 m radius removed) dominated by flank degassing. The simulations show that including the conduit produces rapid warming synchronous with summit heating, whereas conduit-excluded simulations yield a delayed warming that reproduces both the timing and magnitude of the observed large-scale anomalies.The strong agreement between conduit-excluded simulations and satellite observations provides robust evidence that the pre-eruptive thermal anomaly at Augustine was predominantly controlled by diffuse flank degassing rather than conduit heating. More broadly, our study establishes a physically-based framework for interpreting satellite thermal anomalies as indicators of evolving degassing pathways and subsurface permeability changes during prolonged volcanic unrest. This significantly enhances the utility of TIR monitoring for understanding volcanic heat transport processes and the state of unrest. Furthermore, we plan to apply this framework to a wide range of volcanoes to evaluate the generality of these findings.