Can we identify the fault which triggered a British earthquake?
In areas of high seismicity and dense monitoring, for example along the San Andreas fault complex, major faults can be mapped at the surface and often correlated with specific earthquakes. Occasionally major earthquakes can occur on previously unknown 'blind' faults with no surface representation, as with the 1994 Northridge earthquake. It is more difficult, if not impossible, to identify the causative faults in areas of low seismicity. The length of the fault involved in generating small magnitude events need only be of the order of a few hundred metres and the faults generally show no related surface features. Location errors for the calculated hypocentre also need to be considered. These vary according to the magnitude of the event, and the station density. If the earthquake occurred offshore or near the coast, there is an asymmetrical distribution of the monitoring stations and correspondingly much greater location errors.
Onshore, surface geological maps are highly detailed for the UK showing an abundance of mapped faults. For a given epicentre, if surface fault density is high and location errors are large, the error 'circle' can encompass many possible causative faults. The causative fault may be listric in nature, shallowing with depth, and extrapolation between the focus at depth and any surface feature vertically above would not be relevant. Deeper earthquakes in the mid-lower crust may occur on faults that have no connection to the surface and, therefore, no related surface feature.
Two of the main tools for obtaining further hypocentral parameters are focal mechanism studies and spectral analysis. The former, involves mapping the pattern of dilatations and compressional P-wave first arrivals which plot in 4 quadrants, separated by a pair of focal planes, one of which represents the fault plane. The focal mechanism provides information on the type of fault movement and the local stress regime operational. Spectral analysis of the recorded ground motion involves plotting the spectral level against the frequency for the seismic wave spectrum and provides an indication of the size of the radius of the circular fault plane, the seismic moment and moment magnitude (Mw).
