Knowledge Network Node

Study on Soil Moisture at Transitional Region between Oasis and Desert Based on Remote Sensing and Hydrological Model in the Weigan-Kuqa River Basin,XinjiangCN

杨爱霞

新疆大学

Abstract:Soil moisture is an important foundation for the exchange of water and energy between the land and atmosphere system.It is also the key link between the surface and groundwater circulation and the carbon cycle in the land.At the same time,soil moisture is one of the most commonly used surface model parameters for estimating the yield of crops,land degradation and drought monitoring,especially has a very important significance on hydrology and meteorology,oasis agriculture,ecological environment and sustainable development in arid region.In present,from the perspective of soil moisture information acquisition method,it can be roughly divided into three categories: ground observation,model simulation and remote sensing monitoring,each one has its advantages and limitations.Therefore,how to integrate the advantages of each soil moisture monitoring method for achieving efficient and high-precision continuous spatial and temporal monitoring of soil moisture is a very hot research topic.Weigan-Kuqa oasis is an oasis of relatively early development in Xinjiang,as it is located in the inland,the climate is hot and dry.In this paper,the Weigan-Kuqa oasis desert transition zone is a typical research target area at arid and semi-arid area.This research combines geostatistics,geographic information system technology and numerical simulation method to study the regional soil moisture characteristics,which from measured to prediction,mechanism to mechanism,and point scale to regional scale.The aim is to obtain accurate estimates of point scale soil moisture as well as regional scales.The main research results are as follows:(1)Based on the spatial variability analysis and correlation analysis of soil moisture and physical and chemical properties and terrain factors in the study area.First,the spatial variation of soil water content show a trend of high in the west and low in the east.In the vertical direction,the soil water content is increased with the depth of the soil profile.From the seasonal changes,soil water content in April was lower than in July.Second,the correlation analysis between topographic factors and soil water content show that there is a significant correlation between terrain humidity index and soil water content.In large-scale soil moisture research,the terrain humidity index is a good substitute index.Third,the importance of variables by random forests show that soil bulk density and soil texture are more important factors on soil water content.Thus,in the future study of the spatial variation of soil water content,it is necessary to consider the effect of soil bulk density and texture on soil water content.(2)Based on TVDI model by optical remote sensing Landsat 8 image,the soil moisture data of the 7th time series from Landsat 8 image are obtained.The soil moisture in April,July and October obtained by remote sensing image inversion compared with the measured soil moisture,the fitting effect is relatively good.The correlation coefficient between the inversion of soil moisture and the measured soil moisture reached about 0.5.Through the construction of the semi empirical model by AIEM model,using the backscattering coefficient obtained by microwave remote sensing Sentinel-1A,can effectively reflect the soil water content in different months of the study area.According to fitting the measured soil moisture data and simulation of soil moisture data,the correlation coefficient R2 reached about 0.8 between them.The inversion accuracy can meet the requirement of soil moisture inversion at regional scale,and the result can correctly reflect the spatial distribution trend of soil moisture in the study area.(3)The measured values of the soil water characteristics of the soils with different depths(010cm,1020cm and 2040cm)obtained by vertical infiltration experiments in one-dimensional soil column are fitting well with the simulated values.The water suction of different soil profile depth decreased with the increase of soil moisture,and the depth of different soil profiles showed a rapid decline after the slow decline in the trend.The soil water capacity of each layer decreased with the increase of soil suction,different depths show to maintain a smooth trend of change after a rapid decline at first.Using empirical formula to calculate the different depth of soil water capacity and soil water suction show a good power function relationship,and the water capacity of different soil profiles(010cm,1020cm and 2040cm)is almost the same as the trend of soil suction.The soil diffusion rate and soil water content obtained at different depths obtained by the one-dimensional soil column infiltration experiments are fitting well,the coefficient of determination R2 is above 0.8.(4)Throughout the scatter plot between simulated and measured values of the layers of soil moisture,the results show that the simulated values of soil water content in different depths are basically the same as those of the measured values.The above four layers of soil profile(010cm,1020cm,2040 and 4060cm)simulated soil moisture changes in the larger fluctuations,two layers of soil profile(60 80 cm,80 100cm)soil water change is relatively stable,precipitation has a significant effect on the soil moisture in the shallow surface,and according to actual condition soil moisture is affected by precipitation.But in our research we compare the various evaluation indicators,finding out different layers of simulation results are different.we discover the accuracy of soil water movement parameters in the model needs to be further improved,but on the whole,the simulated values of soil water content were consistent with the measured values,which could reflect the changes of soil water content in different layers of bare or sparse vegetation.Therefore,to a certain extent,Hydrus model can be used to simulate the dynamic changes of soil moisture in arid area.The results of the study can provide an effective method for obtaining a wide range of soil moisture data for a long time,and it also can provide reference and basis for the two-dimensional or three-dimensional simulation of the application and promotion.(5)The results show that the Kalman filter can deal with the strong nonlinear problem well,and the soil water content of the assimilated surface soil is obviously improved compared with the simulated soil moisture content by Hydrus model alone.After assimilation of optical remote sensing observations,the mean square error and mean error of the simulated of the surface soil water content from 0.1195 and 0.1078 to 0.0377 and 0.1195,respectively,and the accuracy of the assimilation is improved by comparing the assimilated values with the simulated values.After assimilation of microwave remote sensing observations,for the surface soil water content,the simulated root mean square error and mean error are reduced from 0.0999 and 0.0829 to 0.0236 and 0.0332,respectively,the result of the assimilation is better.The study also show that deep soil water content do not change much,indicating that the assimilation of active microwave remote sensing data has great potential in the estimation of surface soil moisture content.
  • Series:

    (D) Agriculture

  • Subject:

    Fundamental Science of Agriculture; Agronomy

  • Classification Code:

    S152.7

Tutor:

丁建丽;

Download the mobile appuse the app to scan this coderead the article.

Download:875 Page:151 Size:4615K

Related Literature
  • Similar Article
  • Reader Recommendationr
  • Related Funding Articles
  • Citation Network
  • Study Results