Volcanic source inversion using a genetic algorithm and an elastic-gravitational layered earth model for magmatic intrusions

  • Authors:

  • K. F. Tiampo;J. Fernández;G. Jentzsch;M. Charco;J. B. Rundle

  • Affiliations:

  • Department of Earth Sciences, University of Western Ontario, London, ON, Canada;Instituto de Astronomía y Geodesia (CSIC-UCM). Fac. C. Matemáticas. Ciudad Universitaria, 28040-Madrid, Spain;Institute for Geosciences, FSU Jena, Burgweg 11, D-07749 Jena, Germany;Instituto de Astronomía y Geodesia (CSIC-UCM). Fac. C. Matemáticas. Ciudad Universitaria, 28040-Madrid, Spain;Center for Computational Science and Engineering, University of California, Davis, CA 95616, USA

  • Venue:
  • Computers & Geosciences
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

Here we present an inversion methodology using the combination of a genetic algorithm (GA) inversion program, and an elastic-gravitational earth model to determine the parameters of a volcanic intrusion. Results from the integration of the elastic-gravitational model, a suite of FORTRAN 77 programs developed to compute the displacements due to volcanic loading, with the GA inversion code, written in the C programming language, are presented. These codes allow for the calculation of displacements (horizontal and vertical), tilt, vertical strain and potential and gravity changes on the surface of an elastic-gravitational layered Earth model due to the magmatic intrusion. We detail the appropriate methodology for examining the sensitivity of the model to variation in the constituent parameters using the GA, and present, for the first time, a Monte Carlo technique for evaluating the propagation of error through the GA inversion process. One application example is given at Mayon volcano, Philippines, for the inversion program, the sensitivity analysis, and the error evaluation. The integration of the GA with the complex elastic-gravitational model is a blueprint for an efficient nonlinear inversion methodology and its implementation into an effective tool for the evaluation of parameter sensitivity. Finally, the extension of this inversion algorithm and the error assessment methodology has important implications to the modeling and data assimilation of a number of other nonlinear applications in the field of geosciences.