Unified model for intraplate earthquakes - Intraplate Earthquakes - page 276

of deaths and destruction ( www.earthquake.usgs.gov/earthquake/world/most destructive.
php ) . The earliest attempt to explain IPEs on a global scale was by Sykes ( 1978 ) , who
noted that due to a sparsity of seismic stations and short historical records the nature of
IPEs was not well understood. Over the past four decades, since the development of the
theory of plate tectonics, which explains nearly 95% of the Earth's seismic energy release
near or along plate boundaries, there have been continuing improvements in the quality
and quantity of data associated with IPEs. These include the development of improved
techniques in locating IPEs with better instrumentation and denser seismic networks; pale-
oseismology to document prehistoric earthquakes; global positioning systems to detect
minute earth movements; in situ stress measurements; improved geophysical techniques,
especially seismic tomography; mathematical models to explore various hypotheses; and
improved analytical techniques to process various types of data. Improvements in obser-
vational data have been accompanied by the formulation of explanatory models which fall
into two groups. The first were aimed at explaining the seismicity at a particular location,
the New Madrid seismic zone (NMSZ) in the United States in particular. Most of these
were conceptual in nature and are generally not testable. The second were based on global
observations of geological and mechanical similarities, and spatial association between
geological features and seismicity. However, a unified model for the genesis of IPEs has
been missing.
Among the global observations are the apparent affinity of IPEs to rifted regions (John-
ston and Canter, 1990), and the association of IPEs with identifiable geological features,
e.g., with buried plutons (Long, 1976 ) , fault bends (King, 1986 ) and rift pillows (Zoback and
Richardson, 1996 ) . These earthquakes occur due to reactivation caused by a regional, uni-
form compressional stress field (Zoback, 1992a ) . Simultaneously, sedimentologists devel-
oped basin inversion models to show how reactivation of elongate rift structures produces
the sedimentary basins in central Europe (Ziegler, 1987; Nielsen et al ., this volume). Finally,
in situ stress measurements and analysis of seismicity data revealed that in the vicinity of
some of these earthquakes the local maximum horizontal stress field was rotated (Zoback
and Richardson, 1996 ) . These different observations and explanations were akin to the
proverbial description of an elephant by blind men, each examining one part of the ele-
phant's anatomy. I show that these observations, hypotheses, and ideas are not mutually
exclusive, rather they are all different parts of a synoptic view. In this chapter I consider
these earlier observations and integrate them with newer observations, data analyses, and
mathematical models to develop a unified model for IPEs. In this testable model, the main
idea is that IPEs are associated with stress build-up at local stress concentrators (LSC) due
to a uniform, far-field regional stress field associated with plate tectonic forces. These LSCs
were formed in and are preferentially located in former rift zones, and their reactivation
occurs by the present-day compressive stress fields in the mid-continental regions. I will
refer to this phenomenon as “reactivation by stress inversion.” This stress accumulation
causes detectable changes in the nature and direction of the regional stress field in their
vicinity over wavelengths of tens to hundreds of kilometers.