Geologic Adaptation for Seismic Network Tracking

We demonstrate a practical method for automatically adapting a network of seismic unattended UGSs to their specific geologic setting. The demonstration relies on data generated from high-fidelity 3D seismic simulations of a moving vehicle traversing a complex terrain having heterogeneous geology, and significant topographic relief. The simulated data, allows an arbitrary deployment of sensor nodes to form a network. Importantly, the simulated data shows realistic variations in character with as much as 40 dB of signal power variation between hardrock materials and thick soils. The presence of large lateral geologic contrasts also produces significant deflections in surface wave raypaths and extensive regions with very low surface wave spatial coherence. These conditions present notably difficult problems for UGS tracking methods. The network demonstration has 14 UGS nodes with each node having 6 vertical component seismic sensors arranged in circular array pattern. In principle, the simulations approach allows an arbitrary number of UGS nodes and array geometries. The network is adapted to the local geology using a sequence of calibrated vertical impulses applied to the ground at several locations along the expected path of the target. Inversion of the seismic signals from the calibration pulses allows us to define passive target range estimation and Line of Bearing (LOB) correction functions. The LOB corrections compensate for raypath curvature. The range correction functions compensate for the large signal amplitude changes produced by the complex geology. The target-track LOB and Range information from each network node are fused with a least squares range-tracking methods. We further show how statistical rejection of out-laying track information also improves the track result. Report Documentation Page Report Date 00 Oct 2000 Report Type N/A Dates Covered (from... to) Title and Subtitle Geologic Adaptation for Seismic Network Tracking Contract Number