Jr. Unruh et al., KINEMATICS OF POSTSEISMIC RELAXATION FROM AFTERSHOCK FOCAL MECHANISMSOF THE 1994 NORTHRIDGE, CALIFORNIA, EARTHQUAKE, J GEO R-SOL, 102(B11), 1997, pp. 24589-24603
Geodetic observations of surface deformation associated with the 1994
Northridge, southern California, earthquake generally are reproduced b
y simple models of a large-scale elastic dislocation on a blind or bur
ied thrust fault. The smaller-scale aftershocks of the Northridge eart
hquake are distributed throughout much of the volume of crust that app
ears to have deformed elastically during the mainshock. These aftersho
cks, averaged over volumes that are large relative to their rupture ra
dii, reflect a distributed, permanent deformation that is accommodated
by local brittle fracture. We use a micropolar continuum model to inv
ert the aftershocks in such volumes for the average incremental strain
, and we compare that deformation both with the elastic strain from th
e dislocation model of the mainshock and with geodetically measured st
rain. Aftershock deformation that occurred at depths below about 6 km,
and which is associated with the primary rupture zone, is consistent
with slow continuation of the southwest-side-up reverse slip on the bl
ind Northridge thrust fault. In contrast, aftershock deformation from
the upper 5-7 km of the hanging wall block directly above the thrust f
ault can be characterized by horizontal NE-SW shortening and horizonta
l NW-SE (i.e., fault-parallel) extension. This pattern of deformation
is similar to that associated with the mainshock, as observed geodetic
ally and as calculated from the elastic dislocation model. We interpre
t that the aftershock activity in the hanging wall represents the quas
i-ductile accommodation by brittle deformation mechanisms of a permane
nt strain distributed through the hanging wall block. The aftershocks
along the mainshock rupture zone are interpreted as resulting from eit
her (1) the time-dependent release along a weakened fault zone of part
of the remaining accumulated elastic strain in the upper crust or (2)
the continued slip in the weakened fault zone driven by the deformati
on of a ductile-elastic lower crustal layer that relaxes under the str
ess transferred by the coseismic loss of cohesion in the upper crust.
In either case, the aftershock activity suggests that the crust underg
oes quasi-ductile flow as a brittle-elastic material, and is not a str
ictly elastic material.