Research On Snow Parameters Retrieval Based On Multi-source Data In Black Soil Region Of Songnen Plain

Snow is a unique environmental factor in the seasonal snow area,and it is important to study the soil moisture forecast,hydrology and climate in specific areas by using some means to obtain the snow parameter information.Therefore,perfecting the snow monitoring system,realizing the quantitative study of snow parameters,improving the accuracy of snow cover drawing and snow deep inversion,it is very important to the agriculture and economic and social development of the seasonal snow area in our country.Because the meteorological station and other ground routine monitoring means is affected by the terrain and other natural factors,the obtained snow parameter information is difficult to reflect the real condition of the remote area,the data precision is poor,it is difficult to meet the application precision demand in the small area.Therefore,remote sensing monitoring means,which have the advantages of macroscopic,rapid and so on,came into being and were quickly popularized and applied.The black soil area of Songnen Plain is a typical black soil concentration area in the northeastern part of China,is an important commodity grain production base in our country,because the surface of the region is covered by snow for a long time,the phenomenon of frozen soil is serious,the amount of snow melt directly affects soil moisture in spring,and then it has influence on the ploughing,so it is very important to carry out the study of snow parameter inversion in this area.The inversion results of snow parameters will be directly used as input parameters for calculating snow equivalent and provide data support for the forecast of the spring tillage entropy in this area.The inversion of snow parameters mainly includes information extraction of snow cover and inversion of snow depth etc.This paper sums up the theory and main methods of snow cover information extraction and snow deep inversion.In order to solve the problem of large error from two value classification in the study of snow information extraction,and to solve the common problem of mixed pixels in order to improve the accuracy of the snow cover rate inversion in the research area,it is necessary to explore the snow cover rate inversion algorithm for different regions.The statistical regression method establishes a general model through the linear relationship between snow cover fraction from MODIS and NDSI,and uses NDSI from MODIS data to reverse the snow cover fraction,which improves the computational efficiency of the snow cover fraction in the large area,and is advantageous for popularization and application.In view of the differences in geographical environment in different regions,in order to effectively improve the accuracy of estimating the snow cover fraction in the local area,according to the unique geographical conditions of the black soil region in the Songnen Plain,the MOD09 GA image is regarded as the data source and OLI image as "truth" data,This study establishes a linear model between the snow cover fraction and NDSI in the black soil region of theSongnen Plain,obtains the snow cover fraction of the pixel in the study area,and completes the mapping of the snow cover fraction of the sub-pixel.The results show that the error analysis results of the inversion results are improved compared with the MOD10A1 FSC datas,which can better meet the requirement and reality of the current snow cover inversion research.In view of the fact that there is no mature snow depth inversion algorithm for reference in this research area,based on the improved Chang algorithm,the snow depth inversion is studied by using the bright temperature data of the SSMI/S Microwave Imager,the snow depth estimates obtained by meteorological stations in different years,and the inversion algorithm are validated and analyzed,The results show that the algorithm is feasible for snow depth inversion in this research area,and better reflects the real snow depth in the area.In addition,the paper uses the MOD10A2 data and the snow depth data set provided by the scientific data center of the arid area,the changes of snow cover area and snow depth in different snow seasons were analyzed,and the change trend was basically in accordance with the actual surface snow cover and snow depth distribution.