Geological Society of London
Download file
Download file
2 files

Volcanic hotspots of the central and southern Andes as seen from space by ASTER and MODVOLC between the years 2000 and 2010

Download all (3.17 MB)
posted on 2016-06-21, 11:14 authored by J. A. Jay, M. Welch, M. E. Pritchard, P. J. Mares, M. E. Mnich, A. K. Melkonian, F. Aguilera, J. A. Naranjo, M. Sunagua, J. Clavero

We examine 150 volcanoes and geothermal areas in the central, southern and austral Andes for thermal anomalies between the years 2000 and 2010 from two different spaceborne sensors: (1) those automatically detected by the MODVOLC algorithm from MODIS; and (2) manually identified hotspots in night-time images from ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer). Based on previous work, we expected to find eight volcanoes displaying thermal anomalies (Ubinas, Villarrica, Copahue, Láscar, Llaima, Chaitén, Lonquimay and Chiliques). We document 35 volcanic areas with pixel-integrated temperatures of 4 up to more than 100 K above background in at least two images, and another 16 areas that have questionable hotspots with either smaller anomalies or a hotspot identified in only one image. Most of the thermal anomalies are related to known activity (i.e. lava and pyroclastic flows, growing lava domes, fumaroles, and lakes) while others are of unknown origin or reflect activity at volcanoes that were not thought to have surface activity. A handful of volcanoes exhibit temporal variations in the magnitude and location of their temperature anomalies that can be related to both documented and undocumented pulses of activity. Our survey reveals that low-amplitude volcanic hotspots detectable from space are more common than expected, based on lower spatial resolution data, and that these features could be more widely used to monitor changes in the activity of remote volcanoes. We find no evidence from ASTER or MODVOLC that the thermal anomalies were affected by six earthquakes with Mw above 7 in our study area from 2000 to 2010, although the observations may not have been optimal to detect such anomalies.