1. College of Earth Sciences，Jilin University，Changchun 130061，China;

 In order to explore the characteristics of sedimentary cycle identification under different scale wavelet transform methods based on seismic data，following the basis and idea of using continuous wavelet transform of logging data to divide stratigraphic cycles，synthetic record is used as a bridge to successfully use seismic data to divide 31 sand groups of the 3rd Member of Paleogene Shahejie Formation in the Qikou depression into cycles． Through the establishment of an ideal acoustic curve model，the corresponding characteristics of the cycle interface and the transition interface within the cycle，the energy spectrum and the wavelet coefficient curves of different scales are summarized，and the actual logging data are subjected to continuous wavelet transform to complete the cycle division． At the same time，the seismic trace beside the well was extracted，and the relationship between the cycle and the wavelet coefficient curve of the seismic trace at different scales was obtained，and the optimal method for dividing the cycle using the wavelet transform coefficient curve of the seismic data at different scales was summarized． Practical application results show that under the condition that the large-scale ( a100) wavelet coefficients of