Response And Numerical Simulation Of Soil Water Infiltration To Precipitation Change In The Converted Grassland In The Loess Hilly Region

Precipitation is the main source of soil water in the loess hilly region,and the amount of precipitation is very important to soil water supply and vegetation growth.In this study,under the condition of natural rainfall in the field,7 precipitation treatments were set as 60%,40%,20%,natural rainfall,20%,40% and 60%,respectively,and TDR was used to carry out in-situ fixed point monitoring and numerical simulation of soil moisture content.The relationship between different precipitation conditions and soil water infiltration and its infiltration law were explored,and the influence of different precipitation treatments on soil water infiltration was clarified,so as to better utilize precipitation resources and provide powerful theoretical guidance for vegetation restoration on the Loess Plateau.The following results are obtained:1.In the growing season,the average moisture content of 0-180 cm soil under different precipitation treatments showed the same trend over time,reaching the peak value in June(9.02%～14.54%)and October(16.59%～24.58%).Increasing rainfall by 40% increased soil water content the most(32.43%),and decreasing rainfall by 60% decreased soil water content the most(15.71%).2.The soil moisture content in 0-40 cm and 50-90 cm soil layer is easily affected by meteorological environment,while the soil moisture content in 100-180 cm soil layer is relatively stable.The average moisture content of 0-40 cm soil layer increased with the increase of rainfall.In the 50-90 cm soil layer,the 20% soil moisture content increased by42.8%,and the 60% soil moisture content decreased by 12.95%.In the 100-180 cm soil layer,the precipitation increase of 40% was the highest and increased by 44.04%,while the precipitation increase of 20% was the lowest and decreased by 9.61%.3.The variation trend of soil moisture content profile was different under different precipitation treatments.The overall trend of soil moisture content increased by 20% showed an "M" trend with the soil depth,while the overall trend of the remaining treatments showed an "S" trend with the soil depth increased by-decreasing-increasing.The minimum soil moisture content was concentrated at 10 cm,and the coefficient of variation of soil moisture content at 10 cm was the largest,ranging from 0.47(40% rain increase)to 0.83(60% rain decrease).The maximum values were 20-40 cm and 180 cm,and the coefficients of variation were the smallest at 180 cm,ranging from 0.09(40% rain reduction)to 0.29(control).4.The infiltration amount and infiltration rate of different precipitation treatments decreased with the passage of time,and decreased with the increase of soil depth.Under heavy rain conditions,the infiltration amount and infiltration rate increased with the increase of precipitation treatment rainfall,and the maximum infiltration depth was 30 cm(60% rain reduction)-80 cm(40% rain increase).Under the condition of rainstorm,there is a threshold value at 20% rainfall increase.When the rainfall increase is higher than 20%,the infiltration quantity and infiltration rate will be promoted;otherwise,the infiltration rate will be decreased.The maximum infiltration depth is 70 cm(60% rainfall decrease)-140 cm(40%rainfall increase).Under the condition of heavy rain(25.2 mm),the infiltration volume of each precipitation treatment was 1.76～6.36 mm,and the average infiltration rate was0.013～0.047 mm/h.Under the condition of rainstorm(73.6 mm),the infiltration of each precipitation treatment ranges from 5.06 mm to 12.09 mm,and the average infiltration rate ranges from 0.046 mm to 0.110 mm/h,with the highest precipitation increase of 60% and the lowest precipitation decrease of 60%.5.The rainfall infiltration models with different precipitation treatments show better fitting results(except 0-20 cm on the surface and 100-150 cm on the bottom),which can be used for soil moisture prediction.The fitting value of soil moisture content at 0-150 cm depth showed the same trend with the measured value over time.The simulated results of other soil layers could better reflect the rainfall infiltration process,except that NSE<0.5 for the fitting rainfall infiltration process of shallow layer and bottom layer showed relatively poor results.RMSE for each precipitation treatment were 0.005-0.042(60% rain increase),0.006-0.036(40% rain increase),0.006-0.038(20% rain increase),0.003-0.023(control),0.005-0.033(20% rain decrease)and 0.006-0.028(40% rain decrease),respectively %),0.001～0.027(60% rain reduction),the RMSE decreased with the increase of soil layer range,and the accuracy was improved.The simulation accuracy was the highest in control treatment.