What can be learned from underdetermined geodetic slip inversions: The parkfield GPS network example

A. Ziv, M. P. Doin, R. Grandin

Research output: Contribution to journalArticlepeer-review

Abstract

Often geodetic data are inverted for fault slip using less independent constraints than model parameters, and the solution is non-unique. That underdetermined geodetic slip inversions cannot provide unique slip distributions does not mean that they cannot provide other unique information regarding the slip distribution. In order to see which of the slip distribution attributes are obtainable by underdetermined inversions, we considered a synthetic GPS data set and inverted it for slip.We set the fault and network geometries to be identical to those of the Parkfield segment and the 14 SCIGN GPS sites next to it. We show that while slip inversions of such data yield robust estimate of the geodetic potency and the moment centroid, neither the spreadness nor the skewness may be resolved given the SCIGN network configuration. Furthermore, we show that randomly constructed networks are better configured than the Parkfield network, in the sense that they better recover the macroscopic attributes of the slip distribution. Finally, we show that the moment magnitude may be recovered using individual GPS stations, provided that these stations are not located in close proximity to the fault zone.

Original languageEnglish
Pages (from-to)1900-1908
Number of pages9
JournalGeophysical Journal International
Volume194
Issue number3
DOIs
StatePublished - Sep 2013

Keywords

  • Earthquake source observations
  • Inverse theory
  • North America

Fingerprint

Dive into the research topics of 'What can be learned from underdetermined geodetic slip inversions: The parkfield GPS network example'. Together they form a unique fingerprint.

Cite this