← Back to Sessions

Inversion of Dispersion and H/V Curves for Shear-Wave Velocity Versus Depth Profiles in the Atlantic Coastal Plain of the Eastern U.S. Using Legacy Crustal-Scale Refraction Seismic Data

Session: Advances in Upper Crustal Geophysical Characterization II

Type: Oral

Date: 4/21/2021

Presentation Time: 02:30 PM Pacific

Description: 

Shallow shear-wave velocities for site response often are estimated using joint inversions of ambient-noise horizontal-to-vertical (H/V) spectral ratios and dispersion curves derived from ambient noise or small active sources. Here we evaluate carrying out these inversions using crustal-scale seismic refraction data, which have been acquired over many sedimentary basins of interest for hazard studies. We use data from the 2014­–2015 Eastern North American Margin (ENAM) experiment in Virginia and North Carolina. The ENAM project deployed a pair of ~215-km-long, NW–SE linear arrays with ~300-m receiver spacing to record 11 dynamite shots, and offshore airguns were recorded by 80 continuously recording seismometers with ~4 km spacing. The arrays crossed the onland portion of the Atlantic Coastal Plain sediments, which are a seaward-thickening wedge of Cretaceous and younger sediments up to 1.1 km thick beneath the ENAM arrays and deposited mostly on crystalline bedrock. Atlantic Coastal Plain strata are known to substantially amplify ground motions. We compute dispersion curves from 2- to 6-km-long portions of the dense arrays near each dynamite shot, and we compute ambient-noise H/V ratios from 24 half–hour recordings from the continuously recording seismometers. We use a genetic inversion algorithm in which 50 forward 1-D velocity models in each “generation” are evaluated for misfits compared to the observed data, with subsequent generations constructed from elements of the models with the smallest misfits. At each site we iterate for 150 to 300 generations from five initial sets of models, resulting in 37,000 to 70,000 forward models. Velocities to depths of about 400 m are defined well, with deeper velocities less defined. Best–fit models show large velocity increases near the estimated depth of basement rocks. Reliability of the results is supported by the consistency of the profiles from different starting models and nearby sites, small misfits to the observed data, and correspondence of large velocity increases with depths of basement rocks.

Presenting Author: Thomas L. Pratt

Student Presenter: No

Authors