Improving Seismic First Arrival Picking in Noisy Data: A Wavelet-Based Denoising Technique

Abstract

Accurate seismic first arrival picking is fundamental for geophysical interpretation and subsurface imaging. This study evaluates the performance of wavelet-based denoising techniques combined with the Translation-Invariant Shrinkage (TIS) algorithm to enhance first arrival detection. The Higher Density Discrete Wavelet Transform (HDDWT) and Double Density Wavelet Transform (DDWT) are applied to synthetic and real seismic datasets with varying noise levels. Results indicate that HDDWT outperforms DDWT in preserving critical low-frequency components and maintaining signal fidelity, particularly under high noise conditions. The P-phase Picker algorithm, when integrated with HDDWT, achieves superior accuracy and reliability in first arrival detection. These findings underscore the potential of HDDWT and TIS as robust tools for improving seismic data quality and enhancing interpretation workflows.