| As a basic parameter in the process of land surface water cycle,soil moisture has a significant impact on vegetation growth and regional ecosystem stability.The northern and southern mountains of Lanzhou are located in the western edge of the Loess Plateau.The regional climate is dry,the annual precipitation is small,the evaporation is strong,the spatial and temporal changes of soil moisture are complex,and the harsh natural conditions lead to the difficulty of vegetation growth in the northern and southern mountains.However,since the implementation of the north and south mountain environmental greening project,the large-scale vegetation restoration has had an important impact on the ecological environment of the area,and also changed the process and distribution characteristics of the regional surface water and soil.Therefore,it is of great practical significance to carry out remote sensing inversion and monitoring of soil moisture in the northern and southern mountains,obtain the spatial and temporal distribution results of regional soil moisture,and quantitatively estimate the ecological water demand of vegetation in the northern and southern mountains according to the water demand model,so as to promote the rational allocation and efficient use of water resources for afforestation and sprinkler irrigation in the northern and southern mountains.Based on this,this paper takes the northern and southern mountains of Lanzhou as the research area,takes Sentinel-2 L2A and Landsat8 OLI remote sensing images as data sources,combines the measured data of surface soil moisture in the northern and southern mountains,and constructs soil moisture inversion models with vertical drought index(PDI),improved vertical drought index(MPDI)and vegetation adjusted vertical drought index(VAPDI)respectively,Finally,the Sentinel-2 L2A remote sensing image with high inversion accuracy and MPDI drought index were selected to retrieve and analyze the surface soil moisture of the northern and southern mountains in Lanzhou city.Finally,the vegetation water demand of the northern and southern mountains in Lanzhou city was quantitatively estimated based on the vegetation water demand model,and the temporal and spatial variation characteristics were analyzed.The main conclusions are as follows:(1)PDI,MPDI,and VAPDI based on Sentinel-2 L2A and Landsat 8 OLI remote sensing images have varying degrees of linear negative correlation with the measured data of surface soil moisture in the north and south mountains,with the average R~2being 0.37,0.64 and 0.59 respectively,Which shows that the use of high-resolution optical remote sensing image red light band(Red)and near-infrared band(NIR)reflectivity calculation of remote sensing drought index to construct a soil moisture remote sensing inversion model has a certain feasibility.From the quantitative evaluation results of the four model verification accuracy evaluation indexes of the three soil moisture remote sensing inversion model verification results,the MPDI soil moisture inversion model of the two remote sensing images of Sentinel-2 L2A and Landsat 8 OLI has the highest accuracy,and is based on The verification accuracy of the inversion model constructed by Sentinel-2 L2A as the data source is better than that of Landsat 8 OLI images.Therefore,the MPDI inversion model based on Sentinel-2L2A remote sensing images has the best inversion effect.(2)Using the Sentinel-2 L2A remote sensing image MPDI soil moisture remote sensing inversion model to dynamically monitor and analyze the temporal and spatial changes of surface soil moisture at 9 time points in the growing season of the northern and southern mountains of Lanzhou City in 2019,the results show that:the study area The average soil moisture at different time points showed a fluctuating growth trend.The soil moisture content was low in late spring and early summer,and the soil moisture content was relatively high in summer,and then began to decrease.Therefore,we must pay attention to and prevent the drought in late spring,early summer and autumn.occur.From the perspective of spatial distribution,in the 9 time points of the growing season,except for the humid and drought-free margins in the southeast and southwest of Nanshan Mountains,other regions experienced different degrees of drought in different periods;among them,mild drought occurred.The area is mainly distributed in a flat and strip shape in most areas of the North Mountain and the northern part of the Nanshan Mountain;the area where drought occurs occupies a small area,mainly in the shape of a strip and scattered points in the valleys of the north and south mountains;Areas with heavy droughts occupies the smallest area,accounting for less than 1%in multiple periods of time.They are mainly distributed in the eastern part of Beishan.This area is greatly affected by human activities and has low vegetation coverage.Soil moisture is more affected by surface evapotranspiration.Therefore,the soil moisture content is low and the soil drought degree is large.(3)According to the estimation results of vegetation water demand in the northern and southern mountains of Lanzhou City,the average value of vegetation water demand in the southern and northern mountains of Lanzhou City from April to October showed a fluctuating downward trend.The maximum value of vegetation water demand was3985.66×10~4m~3,which appeared in July.The minimum value is 979.11×10~4m~3,which appears in October.From April to September,the vegetation water demand in most areas of the north and south mountains is more than 2000×10~4m~3,accounting for more than 50%of the total area of woodland,grassland and arable land,and they are mainly distributed in flat and strip shapes in the north and south mountains.The north and southeast areas of Nanshan.(4)The overall soil moisture content in the environmental greening projects in the north and south mountains is relatively low.Therefore,while preventing the occurrence of drought in the entire north and south mountains,attention and prevention of drought within the scope of the greening projects should be emphasized.In terms of vegetation water demand,attention should be paid to vegetation water demand from April to August within the scope of the greening project.Among them,the water demand in the green area reaches the highest in July and August in summer.Therefore,the irrigation of vegetation within the green area should be increased during this period.Water consumption ensures the normal growth of vegetation. |