330 TECTONICS/Faults Menard HW (1974) Geology, Resources and Society San Francisco: WH Freeman and Co Musson R (1996) British earthquakes and the seismicity of the UK Geoscientist 16: 24 25 Musson R, Neislon G, and Burton PW (1990) Microseismic Reports on Historic British Earthquakes XIV: 22 April 1984 Colchester BGS Seismology Report W1/90/33 Edinburgh: British Geological Survey Scarth A (1997) Savage Earth London: HarperCollins Tuliani LI (1999) Seismicity and Earthquake Risk: On the Basis of Thermodynamic and Rheological Parameters of the Tectonosphere Moscow: Scientific World Van Andel TJ (1994) New Views on an Old Planet Cam bridge: Cambridge University Press Wong IG (2000) Earthquake mechanisms and plate tecton ics In: Hancock PL and Skinner BJ (eds.) The Oxford Companion to the Earth, pp 287 289 Oxford: Oxford University Press Faults S Stein, Northwestern University, Evanston, IL, USA ß 2005, Elsevier Ltd All Rights Reserved Introduction Faults are surfaces in the Earth along which one side moves or has moved with respect to the other They are identified either when an earthquake occurs or by geological mapping showing that motion across the fault has occurred in the past Many faults are inactive, in the sense that there has been no motion across them within some defined time interval, typically the past million years or less Other faults are active, in the sense that recent motion has occurred and hence motion might be expected in the future Faults, and the earthquakes on them, are studied to understand both the regional tectonics and the mechanics of faulting Typically, earthquakes occur on previously identified faults, demonstrate that the fault is active, and provide information on the fault’s geometry and the motion on it For example, in the famous 1906 San Francisco earthquake, one of the first earthquakes to be carefully studied, several metres of relative motion occurred along several hundred kilometres of the San Andreas Fault Hence, H Reid proposed the elasticrebound theory of earthquakes, in which materials on opposite sides of the fault move relative to each other, but friction ‘locks’ the fault and prevents it from slipping (Figure 1) Eventually more strain accumulates than the fault rocks can withstand, and the fault slips in an earthquake The motion is sometimes revealed after earthquakes by linear features, including roads and rows of trees (Figure 2) Those who study earthquakes seek to understand both the geological processes causing earthquakes and the physics of faulting These issues are important for society because knowing where and when earthquakes are likely and the expected ground motion during them can help to mitigate the risk that they pose The largest earthquakes occur at plate boundaries We view them as the most dramatic part of the seismic cycle, which takes place on segments of the plate boundary over hundreds or thousands of years During the interseismic stage, which makes up most of the cycle, steady motion occurs at a distance from the locked fault Immediately prior to rupture there is the preseismic stage, during which small earthquakes (foreshocks) and other possible precursory effects may occur The earthquake is the coseismic phase, during which rapid fault slip generates seismic waves During these few seconds, metres of slip on the fault ‘catch up’ with the few millimetres per year of motion that have occurred over hundreds of years at a distance from the fault Finally, a postseismic phase occurs after the earthquake, during which aftershocks and transient afterslip occur for a period of years before the fault resumes steady interseismic behaviour Because this cycle extends over hundreds of years, we not have observations of it in any one place Instead, our view of the seismic cycle is based on a combination of observations from different places It is far from clear how good this view is and how well our models represent the complexity of the seismic cycle As a result, earthquake and fault studies remain active research areas that integrate a variety of techniques Seismology is used to study the motion during earthquakes Historical records often provide data on the earthquake cycle for a given fault segment Field studies provide information about the location, geometry, and history of faults Geodetic measurements are used to study ground deformation before, during, and after earthquakes and thus provide information about the processes associated with fault locking and afterslip Results for individual earthquakes are combined with those from other analyses, including laboratory studies of rock deformation, to understand how the earthquakes in a region reflect the large-scale tectonic processes causing them and to study the physics of faulting