Geoelectric signals in China and the earthquake generation process

We present original apparent resistivity records for stations located in the Chinese provinces of Hebei and Sichuan, along with seismicity patterns, fault maps, and limited hydrologic data. These records are part of a 30-year dalabase of geoelectric signals t_ha__t have been recorded in China by a nationwide network of more than 100 stations et up by the State Seismological Bureau of the People's Republic of China. The da_t_8base includes the time period during which the M 7.8 1976 Tangsban and M 7.2 1976 Songpan earthquakes occurred. Resistivity data recorded in and around the epicentral region of the Tangshah event show coherent changes, both temptral and in areal distribution, years and months before and after the earthquakes. A regional change in water table occurred more than 3 years prior to the Tangshah earlhqnalre. The resistivity and water table signals appear to correlate in the 2to 3-year period prior to the earthquake. Rainfall records from the region around Tang•an, extending to the dries of Beijing and Tianjin, do not appear to account for the resistivity changes. Regional seismicity and active fault patterns in north China indicate that deformation occurs mainly along the borders of large tectonic blocks. The source region for the geoelectric signals appem• to lie within a large shear zone, at the edge of a tectonic block undergoing rotation. The Tangshah epicentral area is thus only part of a larger zone of deformation. We suggest that dilafional strain associated with the opening of fractures may lower resistivity and water table. For shallow crustal depths, such as those monitored by the Chinese geoelectric network, dilation and resistivity changes are expecte• to begin at stress levels well below the peak values of the seismic cycle. Our study suggests that in some cases the areal extent of geoelectric signals may be significantly larger than that of the eventual earthquake source dimension. The Chinese geoelectric database offers an oplxamnity to improve our understanding of possible geoelectric precursors to earthqu•s.