Microseism Directivity from Noise Cross-correlation

Abstract:

The direction of microseisms (0.05-0.35 Hz) is determined from noise cross-correlation with conventional pre-processing, where only earthquakes and local cultural transients (e.g. trawling, fish hitting) are removed. Noise cross-correlation has been successfully used to obtain accurate travel times, generally by retaining the phase (but not the amplitude) using one-bit or running-absolute-mean pre-processing. These methods give cross-correlation functions with modified amplitudes, and tend to accentuate long duration signals, thereby biasing the dominant source direction. Here, to avoid introducing bias, the microseism source direction in the Cascadia region was determined from cross-correlation functions for OBS data using conventional pre-processing. No significantly dominant (in the sense of total energy) microseism source directions are detected. Moreover, pelagic-generated signals tend to be weaker but have longer duration, in contrast to coastally-generated signals that tend to be stronger but have shorter duration.