HR: 14:55h
AN: T13H-06 INVITED    [Abstracts]
TI: Do GPS Vectors "Know" About Crustal Heterogeneities?
AU: * Allmendinger, R W
EM: rwa1@cornell.edu
AF: Cornell University, Dept. of Earth & Atmospheric Sciences, Ithaca, NY 14853-1504, United States
AB: Unambiguous elastic rebound recorded by GPS data collected before and after major earthquakes has led to the conclusion that the majority of the tectonic signal captured by GPS is elastic and transitory. Elastic models used to understand plate kinematics commonly assume that the only heterogeneities in the earth are the discontinuities (i.e., faults) between blocks of uniform material properties. While a convenient first approximation, this simplification does not accord with geological observations that different crustal blocks have significantly different properties, inherited from hundreds of millions, if not billions, of years of tectonic evolution. A simpler approach, with no a priori mechanical assumptions, is to calculate strain and rotation rate directly from GPS velocity fields. The results of such analysis show a clear spatial correlation between stable blocks (commonly referred to by geologists as "rigid"), embedded within and at the margins of late Cenozoic orogens around the world, and regions of anomalously low strain. The best examples of these are the Tarim and Sichuan blocks in the Tibet- Himalayan orogen. In the southern Central Andes, significant strain rate gradients and EW-trending belts of constrictional strain are best explained in terms of rheological variation. In contrast, the Altiplano, geologically quiescent for the last 9 Ma, but generally thought to be weak, has a relatively high strain rate due to the geometry and locking depth of the subduction thrust. In the western United States, the overall strain rate is 4-5 times greater to the west of the Paleozoic margin of the continent than to the east, even though the magnitudes of crustal thinning measured by 2D dilation rate, are similar in the eastern and western Basin and Range. In the Andean and western US cases, other factors besides the presence of "rigid" or mobile crust probably help to determine the strain rate patterns. Additionally, in none of the cases studied are the GPS data sufficiently dense to distinguish between changes in block size and changes in block rheology. Thus a region of higher distributed strain could be modeled by either by an aggregate of smaller blocks, or a single larger but weaker block. The difference between these two alternatives may be more semantic than real.
DE: 1209 Tectonic deformation (6924)
DE: 1240 Satellite geodesy: results (6929, 7215, 7230, 7240)
DE: 8002 Continental neotectonics (8107)
DE: 8038 Regional crustal structure
DE: 8102 Continental contractional orogenic belts and inversion tectonics
SC: Tectonophysics [T]
MN: 2007 Fall Meeting