Using Depth Phases and Array Processing to Characterize Intermediate-depth and Deep Earthquake Clusters

Abstract:

The mechanism for intermediate depth and deep earthquakes is still under debate. The temperatures and pressures are above the point where ordinary fracture ought to occur. Key to constraining this mechanism using seismological data is the precise determination of hypocentral depth and the robust estimation of source parameters. It is well known that depth phases can provide unique information, which not only allows for significant improvement in event depth determination, but also holds the key to better constraining source properties. Nevertheless, routinely and systematically picking such phases at teleseismic or regional distances is problematic due to poor signal-to-noise ratios around the pP and sP phases. To overcome this limitation we have started taking advantage of the availability of dense seismic arrays. We recently proposed a relative earthquake relocation algorithm based on the precise picking of the P and pP phase arrivals using array processing techniques. We further improve our algorithm by implementing source deconvolution to enhance the signal to noise-ratio of the pP phase, which significantly reduces the uncertainty in our estimate of pP-P arrival times. We are thus able to precisely map intermediate-depth and deep seismicity in regions where it is tightly clustered. As a byproduct of our relocation scheme we get array-based estimates of the source time function for every event-pair in the cluster. We explore the Bucaramanga nest in northern South America, arguably the densest and most active intermediate-depth earthquake nest in the world, and provide evidence for at least two clusters within the nest separated by a depth of no more than 5 km. We also map subduction zone earthquake clusters in Chile and Argentina (depth>100Km) and perform relative measurements of source parameters in an attempt to model the diversity of their source processes.