SUBSURFACE CHARACTERIZATION USING SYNCHROSQUEEZING TRANSFORM WITH HIGH-ORDER APPROXIMATIONS

摘要

Time-frequency analysis methods always play an important role in subsurface seismic characterization. Due to the advantages in characterizing non-stationary signals, time-frequency analysis often obtains higher resolution than the competing methods. In this paper, we present a novel technique for subsurface seismic characterization based on a synchrosqueezing transform with high-order approximation. The new synchrosqueezing transform can obtain more accurate instantaneous frequencies by using the higher order approximations for both amplitude and phase in order to achieve a highly energy-concentrated time-frequency representation. We use a synthetic example to demonstrate an excellent time-frequency representation using the proposed method, i.e., to extract the time-frequency variation relation. Applications of the proposed method on two field data sets demonstrate the potential of the new method in detecting low-frequency anomaly and detecting paleo-channels with a higher time-frequency resolution. These two geological features are crucial for subsurface characterization since they are usually related with oil & gas.