Moses Olanrewaju Falowo, Olisa-emeka Etinosa Omigie, Richard Ayobami Alakuko

Seismic data plays a crucial role in investigating subsurface geological structures, but it is often contaminated by noise that reduces interpretability. This study investigates the application of wavelet deconvolution for enhancing seismic data quality by attenuating noise. The methodology involves the preprocessing of field data from Wyoming's Teapot Dome, implementation of trace-by-trace wavelet deconvolution, and evaluation of improvements. Key preprocessing steps include geometry correction, brute stack, refraction statics, and airblast and ground roll noise removal. Optimal deconvolution parameters of 85s operator length and 16s gap length are determined through testing. Results demonstrate wavelet deconvolution effectively improves resolution and signal-to-noise ratio by reducing noise levels across the frequency spectrum, with maximum attenuation at lower frequencies below 10 Hz. The technique successfully preserves reflection signals and geological information while removing noise components. This study underscores the value of wavelet deconvolution for practical applications in enhancing seismic data quality and enabling accurate subsurface characterization. Further optimizations of the algorithm through adaptive wavelet selection and tuning could extend its performance across diverse datasets.