Unified model for intraplate earthquakes - Intraplate Earthquakes - page 279

latitudes. Stein et al .( 1979 ) , Quinlan ( 1984 ) , Hasagawa and Basham ( 1989 ) , and Wu
( 1998 ) , among others, suggested that as a result of glacial isostatic adjustment (GIA),
stresses accumulated during loading can be released by a reduction of the normal stress
due to the upward flexure of the lithosphere due to the removal of the surface load. Wu
( 1998 ) estimated that the contribution of GIA towards fault instability in Laurentia and
Fennoscandia is a few MPa, in agreement with Zoback's (1992b) estimate of 10 MPa.

Wu and Johnston ( 2000 ) estimated that, because the flexural stresses decay rapidly from
the ice margin, deglaciation has no effect hundreds of kilometers away from the edge of the
ice sheet. They estimated that its influence on the seismogenic regions in Eastern Canada
was < 0.5 MPa, and concluded that the magnitude of post-glacial stress is too small to
trigger the M8 earthquake near New Madrid, contrary to the conclusion of Grollimund and
Zoback ( 2001 ) .
In a related study, Calais et al .( 2010 ) suggested that the seismicity in the NMSZ results
from stress changes caused by the upward flexure of the lithosphere associated with river
incision in the northern Mississippi Embayment. In their model the removal of

6m
of sediments over a 60 km wide area between 16 and 12 kyr BP, and another 6 m of
sediments over a 30 km wide area between 12 and 10 kyr BP imposed additional tension
stresses ( < 0.5 MPa) at depths between 5 and 15 km. Calais et al .( 2010 ) suggest that these
stress changes were adequate to trigger the 1811-1812 sequence of M 7 earthquakes on

critically stressed faults in the NMSZ.
Pollitz et al .( 2001 ) proposed that the seismicity results from the downward pull of the
brittle upper crust by a suddenly sinking mafic body. According to this hypothesis, sinking
began several thousand years ago due to perturbations in the lower crust related to the last
North American glaciation.

11.3.4 Unverifiable models

Most of the above models are generally not testable since they are based on unverifiable
assumptions. For example, as Calais et al .( 2010 ) point out, there is no evidence of a weak
lower crust or a sinking high-density body, as was suggested by Pollitz et al .( 2001 ) . Also,
the magnitudes of the perturbing stresses associated with these models are about an order
of magnitude lower than those of the regional stress field. However, I will show that they do
contain some plausible elements, and can play a role in altering the stress regime at higher
latitudes. Those will be incorporated in the unified model.

11.4 Local perturbation of the regional stress field: local stress
concentrator models

The observation of a spatial association of seismicity locations with structures delineated
by geophysical and geological data led to the suggestion of a causal relation between the
two. The seismicity locations were inferred to be associated with local stress build-up, S L ,
on structures due to S T. From elasticity theory (e.g., Jaeger and Cook, 1979 ) a far-field stress