Anatomy of a small intraplate rupture: definitely not a high stress drop event

Abstract

The magnitude mbLg = 5.0 Mont-Laurier earthquake of 19 October 1990, in Quebec, Canada, was one of the largest to have occurred in eastern North America during the past decade. High-frequency ground motions recorded on regional network instruments exceeded anticipated values for an event of its size by a factor of 3. A commonly favored explanation for the discrepancy is that the source was a rare "high stress drop" event. In this article, detailed fault slip models are derived to fit waveform and spectral characteristics of the network data. The results establish that the effective rupture stress was normal (less than 100 bars), that the fault rupture developed symmetrically, and that the average slip time for points inside the rupture area ( approx. 0.1s) was significantly less than that associated with the standard Brune source spectral model. The rupture area developed in at least four distinct episodes, each extending the previously ruptured area. Taken together with similar results for the mbLg = 6.5 Saguenay earthquake of November 1988, the results indicate that the widely-used assumption in hazard analyses that earthquake spectra are adequately represented by the standard Brune complete stress drop model is unreliable for the interpretation and prediction of strong ground motion.