AN: G32A-14
TI: Rapid Finite-Fault Determination from Geodetic Data
AU: M H Murray
AU: D S Dreger
AU: D S Neuhauser
AU: L S Gee
AU: P Segall
AU: B Romanowicz
EM: mmurray@seismo.berkeley.edu; Geophysics Dept., Stanford Univ., CA
AB:
We are developing methods to combine geodetic and seismic
information to rapidly estimate finite-fault geometry and
rupture propagation of earthquakes to aid in hazard
mitigation and emergency response activities.
Finite-fault parameters help to predict static surface
deformation changes to flood plains or navigational hazards
in coastal areas, stress changes on nearby faults that
advance or retard the probability of future earthquakes,
and strong ground motion maps for rapid damage assessment.
These methods will augment the Seismographic Station's
Rapid Earthquake Data Integration (REDI) project, which
currently relies solely on seismic data for earthquake
notification, by adding complementary information from
coseismic static displacements measured geodetically.
The displacements can be used to resolve nodal plane
ambiguity and infer the location and extent of the rupture
plane before it is well defined by aftershock locations,
and to infer distributions of fault slip that are unbiased
by rupture velocity and dislocation rise time assumptions.
To be useful in emergency situations, the geodetic
observations must be processed and combined with the
seismic observations quickly and automatically.
We use frame relay telemetry at many of the
Bay Area Regional Deformation (BARD) permanent
GPS stations for rapid, robust access to the observations,
and are developing near real-time procedures for detecting
coseismic offsets in the presence of high-frequency noise
sources, such as multipath and loss of phase lock.
We use non-linear optimization methods, including hybrid
Monte Carlo/gradient techniques, to rapidly estimate
finite-fault parameters from the coseismic offsets when
the fault geometry is not known a priori.
And we are investigating several approaches to combine the
geodetic and seismic observations, such as predicting
ground motion from the finite-fault parameters using
point-source summation forward modeling techniques, as well
as simultaneously inverting both geodetic and seismic
observations.
SC: G
DE: 7215
DE: 1209
MN: Fall Meeting 1996