The loess tableland area, located in the south of the Ordos Basin, shows complicated landforms and geological conditions. They lead to a mass of noise on seismic profiles, especially surface waves. Conventional methods are difficult to suppress 3D surface waves since they are hyperbolic in the far-array. This paper proposed a new model-based, data-driven noise attenuation method, which uses dispersion curve analysis of surface wave and joint inversion based on genetic algorithm (GA) and conjugate gradient algorithm, to build an accurate surface wave model, and then to subtract the model directly from seismic data to improve the signal to noise ratio (SNR). Numerical tests show that joint inversion of dispersion waves was stable, accurate, and efficient in multi-type models, including models with a low-velocity layer and with a high-velocity layer. In practical applications in the south of Ordos Basin, the 3-D surface wave attenuation method is effective in surface wave suppression. The low-frequency information and the reflect signal are efficiently preserved. Seismic faces are more apparent for reservoir characterization. This method provided credible data for the exploration of the loess tableland area. Keywords Surface Wave, Model-Based, Genetic Algorithm, Dispersion Curves

In order to obtain accuracy interval velocity and improve the quality of imaging of velocity volume, so as to better describe the geological structure and reservoir of gas hydrate in specific area, a new interval velocity inversion combination method is proposed in this paper. By introducing wave ray tracing algorithm and velocity field structuring and smoothing algorithm, this new inversion solution can deduce more accuracy and reasonable interval velocity than other normal inversion methods. Empirical results shows that velocity profile has higher resolution and can finely reflect the occurrence status and spatial distribution of gas hydrate.