A P-S converted-wave survey with severe statics problems on a Quaternary plain

Abstract

Shear wave (S-wave) velocity structure down to basement is one of the key elements for the estimation of strong ground motion generated by earthquakes. A seismic reflection survey is one of the best technologies to image underground structure and to determine seismic velocity. However, an S-wave reflection survey is not as effective or as efficient as a P-S converted-wave survey. Therefore, a P-S converted-wave survey is preferable to an S-wave reflection survey as a technology to get S-wave information for deeper strata. In Quaternary plains, especially in alluvial plains, the S-wave velocity in the surface strata is very low and varies quickly from place to place, especially near rivers or ponds where unconsolidated sediments exist. The thickness of surface strata also varies. These factors can make the receiver-side statics (S-wave statics) in P-S converted-wave processing very complicated. Without proper statics correction, the reflection images are distorted. We conducted an experimental three-component survey on a Quaternary plain. In this survey, S-wave statics are ten times larger than P-wave statics, and the maximum difference is more than 350 ms. The survey shows a good example of large and rapid variation of S-wave statics. P-wave statics and S-wave statics have some similarity in short distance variation, but they differ in spatial trend. Therefore, it was not possible just to modify the P-wave statics and use them as S-wave statics. It was difficult to pick the S-wave first break because of the rapid variations in the S-wave statics. To solve this problem, we developed an iterative procedure, re-picking the first break after applying previously determined statics. By this procedure, we could calculate more correct refraction statics, resulting in a much better image for the P-S converted-wave section. This example shows the importance of the receiver-side statics corrections.