Linking Camera Images, Geodetic Data, & Numerical Models: a Test-case in the March 2011 Collapse of Pu'u O'o Pit-crater, Hawai'i (Invited)
Abstract
Deformation is often obscured at active volcanic vents. It is hence commonly difficult to constrain the onset and evolution of high-strain processes in such very-near-field settings. Time-lapse cameras recently deployed at the highly-active Pu'u O'o vent, Hawaii, imaged a major pit-crater collapse on March 5th 2011 in unprecedented spatio-temporal detail. We analyzed these images with Digital Image Correlation (DIC) techniques to yield a semi-quantitative (pixel-unit) description of the collapse's structural development and associated surface displacements. We then ran numerical pit-crater collapse simulations based on the 2D Distinct Element Method (DEM) to examine how geometric and mechanical factors may influence such observations. The time-lapse images reveal initial stability during lava drain-back, followed by sudden broad sagging of the crater floor, and then collapse controlled by two ring faults. The associated surface displacement patterns derived from DIC analysis fit well with tilt measurements from a station located several hundred meters from the vent. One can hence clearly decipher how the very near-field structural development influences the far-field tilt signal. The DIC data are best reproduced in DEM models with a shallow, vertically-elongated magma reservoir, and/or with a host-rock that is reasonably strong at depth. This study highlights how new tools like DIC analysis and DEM models can help unravel physical processes at active volcanoes.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2013
- Bibcode:
- 2013AGUFM.V53F..05H
- Keywords:
-
- 4315 NATURAL HAZARDS Monitoring;
- forecasting;
- prediction;
- 8494 VOLCANOLOGY Instruments and techniques;
- 8419 VOLCANOLOGY Volcano monitoring;
- 8425 VOLCANOLOGY Effusive volcanism