The SXM Formation in the Sichuan Basin is a favorable area for tight sandstone gas reservoirs. The channel sands are vertically superposed in multiple stages and are very different in exploration and development effects, so is urgent to implement the sand distribution. Log analysis shows that the Vp of sandstone is larger or smaller than that of the surrounding rock (mudstone), that is, Vp is difficult to effectively distinguish lithology, while Vs of sandstone is always higher than that of mudstone (sensitive to lithology), so multi-wave seismic data is beneficial to fine description of channel sand. According to the log interpretation results (Vp, Vs, density and porosity) of reservoirs, a porosity change model was established (porosity 5% to 14%), and multi-wave forward modeling was carried out. Forward modeling and actual multi-wave seismic data analysis show that for high impedance sand (porosity 5% to7%), PP and PS peaks are at the top of the sand, with strong reflections, and PS amplitude is stronger than PP. It shows that both PP and PS can identify this type of reservoir, and the effect of PS is better. For medium impedance sand (good reservoir, porosity 7% to 10%), PP peaks are significantly weakened and even polarity reversal occurs, and PS peak is the top of the sand, and strong reflection, that is, PS is more conducive to identifying this type of reservoir, and PP is likely to miss this type of reservoir due to weaker amplitude. For low-impedance sand (high quality reservoir, porosity greater than 10%), PP peaks describe the bottom of sand (top is trough and bottom is peak), PS peaks still indicate the top of sand, and PP amplitude is significantly stronger than PS amplitude, indicating that PP is more conducive to identifying this type reservoir. In summary, the amplitude of PP changes with the porosity, and the peak of the high impedance (low porosity) is the top of the sand, while the polarity reversal (PP peak is sand’s bottom) occurs at the low impedance (high porosity), and PS can indicate the top of the sand more stably with sand porosity changing (the phase remains unchanged). Therefore, the difference in the amplitude and phase of PP and PS can be used to characterize the porosity change and accurately predict the sand distribution. Finally, the maximum peak amplitude of PP and PS are extracted and combined with phase to describe channel sand distribution. The combination of PP- and PS-wave clearly identifies more sands than PP-wave.