Improved Event Location Uncertainty Estimates

摘要: Abstract : The objective of this project was to develop methodologies that improve location uncertainties in the presence correlated, systematic model errors and non-Gaussian measurement errors. We have developed a methodology based on copula theory obtain robust estimates variogram models for travel-time error. Using methodology, we produced generic, transportable regional Pn teleseismic P phases. These are employed estimate network covariance matrix describes spatial correlation structure predictions validated map phase picking delay variance as function SNR, well epicentral distance body wave magnitude considered surrogate SNR. pick delays implemented corrections, while error variances add diagonal full, non-diagonal data matrix. Our representation assumes bulk 3D velocity heterogeneities Earth accounted by calibrated predictions. linearized iterative algorithm makes use full matrix, hence accounts correlated structure. By relocating large number GT events (GT0-2 nuclear explosions, GT5 earthquakes Reciprocal Cluster Analysis show ignoring leads rapidly deteriorating ellipse coverage bias with increasingly networks. Calibrated travel-times responsible improvements, but they do little coverage. While effect event locations is small order 1-2 km), tend tighten clusters reducing separation between events.