---

Attenuation of Ground Roll in Seismic Data Using Model-Based Inversion and Genetic Algorithm.

Ahmed SN, Routa AK, Kumar P, Vedanti N.

JOURNAL OF THE GEOLOGICAL SOCIETY OF INDIA

https://doi.org/10.17491/jgsi/2025/174276 

This study addresses the issue of ground roll, a significant surface noise that can hide important seismic reflections. Since traditional filters often fail when the noise overlaps with real signals, a model-based method is used to predict and subtract ground roll from the data. By refining the model with a genetic algorithm and applying it repeatedly, the technique successfully removes noise and reveals clearer subsurface reflections in both 2D and 3D datasets, improving seismic imaging quality.

Fig:A 3D shot gather: (a) Pre-processed data, where coal layers are obscured; (b) f-k spectrum of the raw shot gather; (c) f-k filtered data, reflection events have not been completely restored;(d) Processed data, with restored amplitudes revealing primary events and effective attenuation of ground roll.

---