| Soil salinization is a global environmental problem,and its identification is mostly based on the soluble salt content in soil.Although this method is direct and accurate,its relatively high testing cost limits the development of large-scale work,and the sporadic test data of soluble salt content is insufficient to support the research of key scientific issues such as the spatial distribution characteristics of salinization and the cause of salinization.Satellite remote sensing interpretation is one of the most commonly used indirect identification methods for soil salinization in recent years.Although this method has the advantages of wide monitoring range,strong continuous monitoring ability,and fast information updating speed,the method is not only greatly affected by environmental factors such as vegetation,soil moisture content,but also affected by the method model and personnel experience,which directly affects the reliability of the retrieval results.It is not only of scientific significance,but also of practical significance to study efficient,accurate and low-cost identification and evaluation methods of soil salinization.Taking the soil salinization area in the west of Jilin Province as the research object,based on the soil geochemical composition,this paper studied the relationship between the total soluble salt and the soil chemical composition,proposed the geochemical identification and evaluation method and process of soil salinization,and provided a new way for the identification and evaluation of soil salinization.This paper systematically studies the chemical composition characteristics of salinized soil in the west of Jilin Province.The Ca O,Na2O,Cl,S and p H of salinized soil are significantly higher than that of non-salinized soil in this area,and also significantly higher than that of eastern China plain soil and Songliao plain soil.Through XRD analysis of surface salt crystal samples,the main minerals of crystal salt are sodium carbonate,sodium bicarbonate,sodium chloride,sodium sulfide and a small amount of calcium carbonate.From the perspective of mineralogy,it is confirmed that the saline soil in this area is mainly soda type.The main cations of soluble salt in saline soil are Na+,Ca2+and Mg2+,and the main anions are HCO3-,SO42-,CO32-and Cl-.The study found that there was a certain relationship between the total salt content of saline soil and soil chemical components Na2O,Mg O,Ca O,Cl,S,indicating that the total salt content of soil can be retrieved by soil chemical components to identify and evaluate soil salinization.Because soil chemical composition(such as Na2O)is a comprehensive reflection of various minerals and soluble salts in soil,soil chemical composition cannot be simply used for salinization evaluation,and the relationship between soil chemical composition and total salt content needs to be deeply explored.In fact,salinized samples are abnormal samples with some components(such as Na2O)significantly higher than the soil background value.Affected by the superposition of multiple parent bodies,the probability distribution pattern of soil element content is significantly deviated from the normal distribution,affecting the identification of geochemical anomaly samples,which is actually caused by the superposition of multiple normal distributions.In this paper,k-means classification method is used to decompose multiple parent bodies,and a method for extracting geochemical indicators of soil salinization based on the probability distribution of soil element content is proposed.Through partial least squares,machine learning and other methods,the calculation method of background value(theoretical value)of soil chemical composition was established.The difference between the measured value and theoretical value of soil samples was used as the geochemical index of soil salinization,and the extraction method of soil salinization geochemical index based on soil chemical composition was established.According to the mass balance between soil chemical composition and minerals,a calculation method of soluble salt content based on optimization theory is proposed.Based on the geochemical indicators of soil salinization extracted by the above methods,this paper uses 1:250000 and 1:50000 soil geochemical data to identify and evaluate the type and degree of soil salinization in the west of Jilin Province.The test results show that the results of the geochemical identification method of soil salinization proposed in this paper are consistent with the results of the total salt content method by more than 80%.Based on the geochemical indicators of soil salinization,this paper analyzes the natural influencing factors of soil salinization in the west of Jilin Province.The results show that the spatial distribution of soil salinization in the study area is significantly controlled by the regional structure,and the salinized soil is concentrated in the lake marsh depression limited by two NE and two NW faults and the western alluvial-lacustrine low-lying flat,and the Zhenlai area at the intersection of NE and NW faults is also the concentrated distribution area of salinized soil.The main parent materials of salinized soil in the study area are river and lake sediments and marsh sediments.In space,salinization is distributed in the distribution area of chlorite-rich soil.The statistical results show that the soil samples rich in chlorite,montmorillonite,plagioclase,amphibole and biotite have a high proportion of salinization.Based on the geochemical indicators of salinization,the spatial variation characteristics of salinization in the study area were analyzed by using semi-variance function and spatial derivation.On this basis,the causes of salinization in the west of Jilin Province are analyzed.Natural factors such as structure,landform,hydrology,climate and wind direction are the reasons why the degree of salinization in the eastern part of the study area is higher than that in the western part.In addition,human production and living activities are also important factors affecting salinization. |