The relationship of source parameters of oceanic transform earthquakes to plate velocity and transform length

The source parameters of large earthquakes on oceanic transform faults are closely related to the thermal and mechanical properties of oceanic lithosphere. Several characteristics of these earthquakes (including magnitude, moment, and apparent stress η are synthesized according to local plate velocity V , ridge‐ridge offset L , and average fault width W estimated by Brune's method. Several relationships result: (1) the maximum moment M 0 on a transform decreases with V , (2) maximum M 0 appears to increase with L for L < 400 km and may decrease for greater offsets, (3) η does not clearly depend on either V or L , (4) the maximum estimated W ( V ) decreases with V , (5) the minimum estimated W ( L ) increases with L , and (6) the largest earthquakes on long transforms occur near the transform center. Most of these relationships are consistent with the hypothesis that seismic failure occurs only at temperatures below a fixed value. An inversion of slip rate and magnitude data by transform supports this hypothesis and gives an estimate for the nominal temperature of the boundary separating stick slip and stable sliding. Though the actual thermal structure around oceanic transforms is not known, the idealized spreading plate models used in the inversion give a temperature range for the brittle to ductile boundary of 75°–150°C. If possible uncertainties in the thermal structure are allowed for, a range of 50°–300°C provides a conservative bound on the true limiting temperature. These temperature ranges are consistent with focal depths of transform earthquakes and with laboratory measurements of fault slip in rocks of compositions that are representative candidates for the material being faulted in oceanic transforms.