Developing stress-monitoring sites using cross-hole seismology to stress-forecast the times and magnitudes of future earthquakes

A new understanding of rockmass deformation suggests that changing stress i n the crust modifies the geometry of the distributions of fluid-saturated g rain-boundary cracks and pores pervading almost all rocks in the crust. The se stress-aligned micro cracks cause the widely observed splitting of seism ic shear-waves, which are sensitive to the details of the microcrack geomet ry. This means that analysing shear-wave splitting along appropriate ray pa ths above small earthquakes can monitor the build up of stress before large earthquakes. This allowed the time and magnitude of an M = 5 earthquake in Iceland to be successfully 'stress-forecast'. Such forecasting, using smal l earthquakes as the source of shear-waves, is possible only on those rare occasions when the comparatively severe restrictions on the geometry of sou rce, receiver, and large earthquake, allow shear-wave splitting to be analy sed along appropriate ray paths. This paper suggests that in the absence of such pronounced local seismicity and optimum recording geometry, the time and magnitude of future earthquakes can be estimated by analysing shear-wav e splitting in controlled-source cross-well seismology between three boreho les. We suggest that stress-forecasting at such stress-monitoring sites (SM Ss) is the best deterministic option for reliable forewarning of large eart hquakes near any earthquake-vulnerable township, or vibration-sensitive ins tallation. A preliminary stress-monitoring site is currently being set up i n Northern Iceland. (C) 2001 Elsevier Science B.V. All rights reserved.