Evidences of volcanic unrest on high-temperature fumaroles by satellite thermal monitoring: The case of Santa Ana volcano, El Salvador

Abstract

On October 1st, 2005, Santa Ana volcano (El Salvador) underwent a VEI 3 phreatomagmatic eruption after approximately one century of rest. Casualties and damages to some of the local infrastructures and surrounding plantations were followed by the evacuation of the nearby communities. The analysis of MODIS (Moderate Resolution Imaging Spectroradiometer) infrared data reveals that the main explosion was preceded by a one-year-long thermal unrest, associated to the development of a fumaroles field, located at the western rim of the summit crater lake. By combining space-based thermal flux and ground-based measurements (seismicity, sulfur emissions and lake temperatures), we suggest that the activity observed at Santa Ana between 2004 and 2005 was driven by the gradual intrusion of an undegassed magma body at a very shallow depth. Magma injection induced thermal anomalies associated with sustained degassing from the fumaroles field and promoted the interaction between the magmatic-hydrothermal system and the overlying water table. This process culminated into the VEI 3 phreatomagmatic eruption of October 2005 that strongly modified the shallow structure of the crater area. The subsequent three-years-long activity resulted from self-sealing of the fracture system and by the opening of a new fracture network directly connecting the deeper hydrothermal system with the crater lake. Our results show that satellite-based thermal data allow us to detect the expansion of the high-temperature fumarolic field. This may precede an explosive eruption and/or a lava dome extrusion. In particular, we show that thermal records can be analyzed with other geochemical (i.e. SO2 emissions) and geophysical (seismicity) data to track a shallow magmatic intrusion interacting with the surrounding hydrothermal system. This provides a remarkable support for volcano monitoring and eruption forecasting, particularly in remote areas where permanent ground data acquisition is hazardous, expensive and difficult.