Crustal structure in the Changbaishan volcanic area, China, determined by modeling receiver functions

Abstract

During 1998–1999, we installed a temporary broadband seismic network in the Changbaishan volcanic region, NE China. We estimated crustal structure using teleseismic seismograms collected at the network. We detected a near surface region of strong anisotropy directly under the main volcanic edifice of the volcanic area. We modeled 109 receiver functions from 19 broadband stations using three techniques. First we used a “slant-stacking” method to model the principal crustal P reverberation phases to estimate crustal thickness and the average crustal P to S speed ratio ( v p/ v s), assuming an average P-wave velocity in the crust. We then estimated crustal S-wave velocity ( v s) and v p/ v s profiles by modeling stacked receiver functions using a direct search. Finally, we inverted several receiver functions recorded at stations closest to the main volcanic edifice using least squares to estimate v s velocity profiles, assuming a v p/ v s value. The results from the three estimation techniques were consistent, and generally we found that the receiver functions constrained estimates of changes in wave speeds better than absolute values. We resolved that the crust is 30–39 km thick under the volcanic region and 28–32 km thick away from the volcanic region, with a midcrust velocity transition at about 10–15 km depth. We estimated that the average crust P-wave velocity is about 6.0–6.2 km/s surrounding the main volcanic region, while it is slightly lower in the vicinity of the main volcanic edifice. The estimates of v p/ v s were more ambiguous, but we inferred that the bulk crustal Poisson's ratio (which is related to v p/ v s) ranges between 0.20 and 0.30, with a suggestion that the Poisson's ratio is lower under the central volcanic region compared to the surrounding areas. We resolved low S-wave velocities (down to about 3 km/s) in the middle crust in the region of the main volcanic edifice. The low velocity anomaly extends from about 5–10 to 15–25 km below the surface, probably indicating a region of elevated temperatures. We were unable to determine if partial melt is present with the data we considered in this paper.