Source Parameter Estimation Using Waveform Correlation and Grid Search Techniques

Abstract:

Moment tensor catalogs from earthquake sequences are routinely used for studying the tectonics of seismically active regions, constraining regional scaled magnitudes, and developing amplitude attenuation models. In this study, we extended the low magnitude cut off of these catalogs by computing moment tensor solutions for small magnitude events using historic solutions form highly correlated reference events. Forward models for correlated events (i.e. correlation coefficient ≥ 0.5) within 200 km are used for estimating magnitude source parameters. This is achieved via a grid search technique for scaling observed long period surface waves. Preliminary results obtained from the 2008-02-21 Storfjorden aftershock sequence show our technique can estimate moment magnitudes below 3.25 at a variance reduction level approaching 70%.