Study On Temporal-Spatial Distribution And Impact Factors Of Soil Salinization Based On 3S Technology

Soil salinization is the result of natural factors such as geology,hydrology,and climate,as well as human irrational farming and irrigation.The soil salinization in Hetao Irrigation District,Inner Mongolia,which is an important food production base in China,had restricted the development of local agricultural economy.It is of great theoretical and practical significance to use scientific and effective means to obtain soil salinization information and analyze its temporal and spatial dynamics,and to comprehensively explore its influencing factors,which can provide scientific support for the improvement of saline-alkali soilification and prevention.In this paper,the Shenwu Irrigation Area in Dengkou,Inner Mongolia,was selected as the research area.The Landsat ETM+/OLI image was used as the data source.We used the 3S technical means and constructed the MSDI(Modified Salinization Detection Index)model based on the 2D feature space theory.After that,the salinization information was quantitatively extracted and classified,and the spatial and temporal distribution of salinization since 2000 was analyzed combined with the salt data of field soil samples.Finally,the factors affecting soil salinization were analyzed initially from the aspects of natural and human factors.The main research results are as follows:1)This study selected the salinity index(SI)and modified soil-adjusted vegetation index(MSAVI)as the characteristic variables to construct the MSDI model,and used this model to analyze the correlation between SI,MSAVI,NDVI and the salt soil sample,the results showed that: the correlation between MSAVI and soil salinity is 0.8365,which higher than NDVI and SI;the correlation coefficient between MSDI and soil salinity is 0.8568.The overall accuracy of the model for extracting salinization information is 87.5%,and the Kappa coefficient is 0.726,which can effectively quantitatively analyze and monitor soil salinization.2)We used the MSDI model to extract and classify the salinization information of the 2001,2010 and 2017 images in the study area,and analyzed the spatial distribution and dynamic changes of soil salinization according to the soil sample salt data: the soil salt type is mainly sodium salt type,sulfate-chloride and chloride-sulfate type,heavy saline-alkali distribution in the northwest,sand and water body margins,medium and light saline-alkali land is fragmented and scattered.Since 2000,the area of non-salt alkali increased from 18.5% to 30.47%,and the desert area decreased by 52.18%.The area of saline-alkali land increased from 2000 to 2010,and then gradually decreased.The overall salinization situation was effectively improved.3)Based on comprehensive geological,climatic,hydrological and socioeconomic data,the influencing factors of soil salinization in the study area were preliminarily analyzed: long-term subsidence in geological structure,sedimentation of rivers and lakes,and arid and hot paleoclimate are intrinsic factors of the formation of high salinity parent material layer;sandy soil provides favorable conditions for the distribution and accumulation of salt;dry and hot rain is an important cause of soil salinization;groundwater movement mode makes salt difficult to excrete;population,economy,policy and the density of the canal network,etc.,indirectly or directly affect the extent and distribution of soil salinization by changing the land use pattern.