Frictional coefficient on faults in a seismogenic region inferred from earthquake mechanism solutions

The coefficient of friction associated with faults in a seismogenic region is inferred from numerous focal mechanism solutions. Earthquake slip can occur on a favorably oriented fault plane when preexisting fault planes are oriented in various directions. The fault plane is thought to be selected when the Coulomb failure stress given by sigma(f) = tau - mu(sigma(n) - p) - S-0 has a maximum value along a slip direction on the plane. Here, tau is the shear stress, mu is the coefficient of friction, sigma(n) is the normal stress, p is the pore pressure, and So is the cohesion. Consequently, the geometry of the plane and the slip direction include information concerning the coefficient of friction for the faults. In the present paper the coefficient of friction is inferred from the distribution of P axis orientations, assuming hydrostatic pore pressure. The method is very simple and has the advantage that the results obtained do not depend on pore pressure, when pore pressure is not proportional to the normal stress. Observed distributions of P axis azimuths and plunges display a very narrow peak having a width of about 20 degrees. The coefficient of friction on the faults can then be estimated from the width to be less than or equal to mu 0.2. When the stress is balanced by friction, the shear stress can be calculated. Here, the shear stress (in megapascals) has a value of 4.0z, where z is the depth (in kilometers), under the assumption of hydrostatic pore pressure and a coefficient of friction of 0.2. The value obtained for the coeffcient of friction is very small compared with those derived from laboratory frictional experiments. It is thought that this small value of mu is attributed to the friction itself, when pore pressure is not proportional to the normal stress.