Simulation Of Runoff In The Source Region Of The Hanjiang River Based On Satellite Remote Sensing Precipitation Products

Precipitation change affects the water cycle and water balance of the watershed,and causes a series of water resources and water environment problems.The distributed hydrological model itself can simulate the hydrological process of the watershed well.Meanwhile,the application of the distributed hydrological model combined with the precipitation data retrieved by satellite remote sensing can realize the simulation of the hydrological process in areas without observed precipitation data,and provide a scientific basis for the management and regulation of regional water resources.The SWAT model driven by SM2RAIN-ASCAT precipitation product was retrieved based on the observation data of rain measuring stations and systematically evaluated the application ability of remote sensing data of ASCAT-SM2 satellite in hydrological model and the accuracy of it,which is expected to provide a more reliable data basis for hydrological research in areas without observable data.Its main conclusions are as follows:(1)Using linear trend analysis method,Mann-Kendall trend test method and Pettitt mutation test method to analyze the runoff evolution characteristics of the source region of the Han River in recent 60 years at the southern foot of the Qinling Mountains,it is found that the average annual runoff of the source region of the Han River(1960-2018)shows a decreasing trend(P>0.05),the annual runoff depth decreases at a rate of 1.40mm per 10 years.The abrupt change of annual runoff occurred in 1990,but did not pass the significant test.The reduction was 131.03mm before and after mutation.The cumulative anomaly curve and double cumulative anomaly curve(DMC)were used to analyze the relationship between runoff and precipitation in the source region of the Hanjiang River.According to the results of the mutation point analysis,the baseline period(1960-1990)and the variation period(1991-2018)of runoff change were divided by the mutation years.The drought index in the source region of the Hanjiang River was 0.79<1,which indicated that the climate in the basin was humid and the proportion of precipitation to runoff was high.Quantitative calculation of elastic coefficient method is used to precipitation change,the han river source area runoff changes and human activities affect the sensitivity of the potential evaporation,found that the sensitivity of the climate change is higher than the influence of human activity,and the most sensitive to precipitation change,including the sensitivity of the hanjiang river source area runoff to climate change in the trend of the increase was not significant,sensitivity to human activities were not significant weakening trend.The contribution of precipitation,potential evapotranspiration and underlying surface characteristics to runoff reduction were 71.71%,-8.04%and 36.33%,respectively.In conclusion,the contribution of climate change to runoff reduction in the source region of the Han River is greater than that of human activities,and the contribution of precipitation change is the largest.(2)The SWAT hydrological model of the source area of the Hanjiang River was constructed,and the model parameters were calibrated and verified using the annual and monthly runoff of the Wuhou Town Hydrological Station from 2007 to 2018.And through simulation,the range of sensitive parameters of the SWAT model suitable for watershed change was determined.The simulation results show that the SWAT model can meet the accuracy requirements of the evaluation and analysis of runoff changes in the headwater area of the Han River.The Nash coefficient NS and the determination coefficient both exceeded0.7,meeting the accuracy requirements of the hydrological model runoff simulation.(3)Evaluating the accuracy of the SM2RAIN-ASCAT precipitation products retrieved from satellite remote sensing precipitation data,it is found that the SM2RAIN-ASCAT remote sensing precipitation products are generally not accurate enough to capture single-point precipitation events,and the estimation of heavy precipitation is poor,and the correlation coefficient r with the measured daily precipitation is lower than 0.48,which is in good agreement with the measured monthly precipitation,and the correlation coefficient r with the measured monthly precipitation is higher than 0.65.The SM2RAIN-ASCAT precipitation product derived from the satellite remote sensing data ASCAT drives the SWAT model in the headwater area of the Hanjiang River.It simulates the annual and monthly runoff of the Wuhou Station from 2007 to 2018.The R~2 and Nash coefficients are 0.77?0.76 and 0.66?0.62,which meets the model evaluation standard.In summary,it is proved that the SM2RAIN-ASCAT remote sensing precipitation product in the source area of the Hanjiang River is reliable and effective in estimating actual precipitation,and can be used in the SWAT hydrological model for runoff simulation.