Abstract:In order to clarify the effect of grass cover in water level fluctuating zone (WLFZ) on agricultural non-point source pollution control in the Three Gorges Reservoir (TGR), the field scouring experiment was conducted to analyze the interception rate of nitrogen and phosphorus in surface runoff and the spatial change of nitrogen and phosphorus content in soil before and after scouring, and the effects of the typical grass cover on the transport of nitrogen and phosphorus pollutants in WLFZ were studied. The results showed that:(1) Grass buffer zone intercepted pollutants with a good overall efficiency, and interception rate for nitrate nitrogen was the highest, ranging from 93.5%~99.3%. The interception rate was directly proportional to the width of grass buffer zone and inversely proportional to the scouring intensity. (2) The maximum change rates of TN, TP, NO3--N and NH4+-N before and after scouring were 2.30, 1.96, 5.01 and 4.75, respectively, which all appeared in the lower slope position of the plot. With the increasing of grass buffer zone width and the decreasing of scouring intensity, the change rate of pollutant content increased. (3) The variation rates of nitrogen and phosphorus pollutants in 0-10 cm soil layer were larger, and the variation range was 0.91~5.01. The change rates of TN, TP and NH4+-N in 10-30 cm soil layer were generally close to 1, while the change rate of NO3--N in this soil layer was still large. (4) RDA analysis results showed that the most important factor affecting the change rate of pollutants content was the soil depth, followed by the slope position, the width of the grass buffer zone, and the scouring intensity. In conclusion, the grass buffer of the WLFZ could effectively intercept pollutants and promote their migration to the soil interior. In the process of restoration of herbaceous plants of the WLFZ and the prevention and control of non-point source pollution in the TGR, attention should be paid to increasing the coverage width of grass buffer, reducing the influence of high-intensity rainfall and runoff, promoting the pollutants to be intercepted on the up-slope and migrating to the soil, thus reducing the risk of pollutants being released into the overlying water. It could provide a theoretical basis for the restoration of herbaceous vegetation in the WLFZ and the prevention and control of agricultural non-point source pollution in the TGR.