ALOS PALSAR Reveals Volcanoes Sink After Large Earthquakes
by Matt Pritchard, Cornell University
Large earthquakes are known to trigger volcanic activity at volcanoes hundreds of km from their epicenters, but interferometric synthetic aperture radar (InSAR) data from Japan’s Advanced Land Observing Satellite (ALOS) Phased Array L-Band SAR (PALSAR) have revealed a new phenomenon — enhanced subsidence at some, but not all, volcanoes near two large earthquakes in Chile and Japan. The findings come from two unrelated research teams at Cornell University and Kyoto University.
Five volcanic areas were observed to subside up to 15 cm after the 2010 Maule, Chile, and 2011 Tohoku-Oki earthquakes, magnitudes 8.8 and 9.0 respectively. In both South America and Japan, the size and shape of the deformation zones are similar: All are oriented approximately north–south and measure 15-30 km long and 10-15 km wide (Figures 1-3). The South American volcanoes are Caldera del Atuel in Argentina and Tinguiririca, Calabozos Caldera, Cerro Azul, and Nevados de Chillán, all in Chile. The Japanese volcanoes are Mt. Akitakoma, Mt. Kurikoma, Mt. Zao, Mt. Azuma, and Mt. Nasu.
ALOS PALSAR data were critical for observing the volcano subsidence in both Japan and South America. Sufficient SAR data from other satellites were not acquired at all of the volcanoes, and the available data have low coherence because of the vegetated terrain in both areas. While Japan’s dense GPS network confirmed the enhanced subsidence at a few of the volcanoes, the network was still not robust enough to determine the extent of the subsidence and its relation to the volcanoes. However, the GPS data are critical in showing that deformation took less than 24 hours to reach a maximum, indicating a predominantly elastic response.
Several possible mechanisms could cause the deformation, and future work will focus on understanding which are involved. Some deformation mechanisms can be ruled out because the dense seismic sensing network in Japan recorded intensified seismic activity at only one of the deforming volcanoes (Mt. Akitakoma) and no change in gas emissions. Because of the similarities in the patterns and amount of deformation after the two earthquakes, the mechanism of deformation is probably similar. For the Japanese volcanoes, enhanced subsidence of the volcanic regions because of a mechanically weak magma chamber is suggested by the authors of that study.
In South America, the authors of that study propose that co-seismic shaking and the opening of fractures could have released hydrothermal fluids that then caused enhanced subsidence, partially explaining observed increases in stream flow.
It is not understood why some volcanoes in the rupture zone subsided but others did not. For example, the geothermally active Laguna del Maule, which is currently one of the fastest deforming volcanoes (>20 cm/yr) without a recent eruption had no coseismic deformation; nor did the deformation rate change between March 2010 and January 2011 (the date of the last acquired ALOS PALSAR data in this region). One possible explanation for the lack of deformation at Laguna del Maule isthat the earthquake slip was at a minimum near the Laguna del Maule system (Figure 1).
Additional discoveries using ALOS-1 data from 2006-2011 will be enabled by the more than 1,000,000 scenes that will be part of the PALSAR datapool in early 2014, including all data for the Americas and selected locations around the world. These data are available to U.S.-based members of the Western North America InSAR Consortium (WInSAR) once they sign a data use agreement. More data from around the world can be added to the ASF DAAC datapool through a simple proposal process. Please contact firstname.lastname@example.org for details.