Study On The Spatial Distirbution And Spectral Properties Of Soil Salinization Around Ebinur Lake

Soil salinization in arid and semi-arid land degradation has become a globalenvironmental problem. Salinization is the general term of the salinization oralkalization of soil. Soil salinization and secondary salinization has become a majorecological and environmental problems, a serious impact on industrial andagricultural production, thus more attention has been paid. Xinjiang is a majorregional distribution of saline soil in China, different levels of saline soil distributed inalmost the whole territory, which has a great impact on agricultural production.Remote sensing technology can provide information of saline nature, scope, andsalinization in real time. Therefore, the application of remote sensing technology tothe dynamic monitoring of large-scale saline soil, and timely access to the informationof the evolution of soil salinization, and to develop a reasonable program ofgovernance, control its further deterioration has become an urgent task.This paper takes the wide range of saline soil around the Ebinur District as atypical region. Through the field survey and targeted soil, the spectra of samplescollected, based on geography, soil science, environmental science, ecology andremote sensing science theory, and the3S technology as a means of using the methodof combining qualitative and quantitative theories, has a systematical discussion ofremote monitoring system of the salinization of the soil physical and chemicalproperties, spatial distribution characteristics, spectral characteristics andsoilsalinization around Ebinur Lake.The first chapter mainly introduces the purpose and significance of the study ofsaline soil, making a summary of the study of salinization of the soil at home andabroad, illustrating the study thoughts, study contents and the method applied. Thesecond chapter makes a detailed instruction of the procedure in the general situationof the research area, wild investigation, data collection and the sample processing. Chapter III offers a discusstion of the physical, chemical properties and spatialdistribution of soil salinization in the study area on the basis of measured analysis onsoil salinity. The fourth chapter discusses the soil spectral features around EbinurLake and establishes the prediction model. The fifth chapter focuses on therelationship between the different salt content measured spectral reflectance andimaging reflectivity, and establishes the soil salinization quantitative remote sensingmonitoring model. Chapter VI makes a summary of the major achievements and thewhole study progress.The main findings are as follows:①The soil profile in the mass fraction of salt ions was V-shaped distribution inthe research area, strong salt accumulation.②Establish spatial distribution model ofwater content, EC,SO₄^2-,Cl^-,Na⁺andMg²⁺ y=ax²+bx+c. The various elementsin the performance in the spatial distribution have some common characters, mainlyin the "u" distribution. However, the various elements of the content is larger near thelake than at the farmland edge.③Soluble salt ions, and EC was highly significantcorrelation between the establishment of model y=a×EC±b. Predictive ability ofEC onNa was the highest, R²=0.984, followed bySO₄^2-.Cl is the most active butthe accuracy is relative the least.④Salinized soil has good spectral responsecharacteristics. If the salt content increases, the spectral reflectance increases,showing a positive correlation. When the salt content is≥100g/kg, the spectralcharacteristics are obvious and easy to distinguish, but when the salt content is in the20g/kg¹00g/kg, the spectral curve characteristics are similar and difficult todistinguish.⑤Pretreatment of atmospheric correction and radiation correction of TMimage, extraction of image spectral reflectance, and synchronous spectral data werecompared with field measurements to establish the regression model y ax b, toobtain the correlation coefficients of the image spectrum value and the measuredspectral values. In general, the higher the salt content in the soil is, the better the correlation of the image with the measured spectral reflectance is.