Study On Localization For Snow Depth Inversion Under Forest Using Passive Microwave Remote Sensing Based On HUT Model In Northeast China

Snow cover is one of the most active natural elements of the earth surface, and natural snow transfers and changes the interaction between microorganisms, plant, animal, nutrients, atmosphere and soil. High albedo of snow surface, latent heat of internal ice/water phase change and adiabatic effect of snow significantly affects the global energy and water cycle process, thereby affecting the climate change. Meanwhile climate change also leads to spatial and temporal redistribution of snow on global, regional and catchment scale, thereby affecting the snowmelt water resource circulation, distribution and management. Due to high latitudes of the northern hemisphere, the Northeast China is one of the main snow cover distribution region in China. On one hand, snow cover can cause the snow disaster; on the other hand, snowmelt water is an important supplement to water system and the soil moisture of agricultural farming in Northeast China. Snow cover is an important source of water resources, and water is a necessary condition for human. Snow cover is closely related to precipitation, surface temperature, surface albedo and radiation budget.Passive microwave remote sensing uses electromagnetic wave scattering and radiative transfer difference with different microwave frequency of snow particles to realize the quantitative inversion of snow parameters. Snow grain size distribution inside snow cover is an important factor to affect microwave radiation and scattering characteristics of snow cover. Inconsistency of snow grain size distribution is also an important factor to affect inversion accuracy of snow depth and snow water equivalent of passive microwave remote sensing.Forest is the main body of terrestrial ecosystems, and forest ecosystems have an important "buffer" and "valve" function in the process of biogeochemical cycle in the geosphere and biosphere. Forest ecosystems are the world’s largest carbon sink. Forest is the biggest and most perfect terrestrial ecosystem on earth, and it plays an irreplaceable role in the water cycle, carbon cycle, energy cycle and the ecological balance in the earth system. Forest is also the most important maintainer in the earth’s ecological system. Because the forest system has a multilayer structure and each layer of the microwave radiation transfer has different frequency and polarization characteristics, this complexity reduces the accuracy of snow parameters inversion under forest using remote sensing technology which is the problem in the research field.This dissertation is supported by the National Natural Science Foundation of China and dependent on Jingyuetan Remote Sensing Experiment Station for research of snow parameters inversion under forest using remote sensing technology. Main work is as follows:(1) Northeast China is a relatively independent geographical unit, and the characteristics of the snow cover are different from other regions. Therefore the characteristics of the snow cover are researched deeply in Northeast China to provide necessary snow cover parameters for radiative transfer model. The main research contents include: 1) snow characteristics including snow grain size, snow temperature and snow wetness in forest area and farmland area are studied, especially to study the snow grain size, because the snow grain size is the most sensitive parameters in radiative transfer model; 2) all layers of snow grain size, snow temperature and snow wetness are studied; 3) the snow grain size evolution model that is Jordan91 model is validated.(2) Chang algorithm, NASA 96 algorithm and FY3 B operational inversion algorithm are validated and analyzed in typical forest regions of Northeast China to obtain the accuracy of the above algorithms. This can provide guidance for the development of snow depth inversion algorithm which is applicable to the characteristics of snow cover in Northeast China.(3) With the experimental data, frequency, polarization and angle characteristic of microwave radiative transfer parameters in typical forest of Northeast China are explored, and the relationship between microwave radiative transfer parameters and forest geometric and physical parameters is studied.(4) Using the microwave radiative transfer parameters(effective transmissivity and effective scattering albedo) which are observed and parameterized in typical forest of Northeast China, semi empirical microwave radiative transfer model of snow-forest is constructed under the framework of the τ-ω model. This paper combines satellite remote sensing data with ground-based remote sensing data. Based on parameterization research of the forest system radiative transfer model, corresponding research and observation of microwave radiative transfer properties of different underlying surface medium are carried out, and radiative transfer equation of stratified medium is establish.(5) Considering the vertical distribution of snow grain size in improved HUT model, HUT model of multilayer is established. Using Jordan91 model for snow grain size evolution, combining with snow depth and other data of meteorological stations, a localized HUT inversion model is established.The main innovations of this dissertation are as follows:(1) Based on the observation and analysis of the characteristics of snow cover in Northeast China, combining with the meteorological data, the localized HUT snow radiative transfer model based on the Jordan91 snow grain size evolution model is proposed.(2) Based on τ-ω model, combining with satellite remote sensing data, the extinction coefficient of different forest types is studied. We use real and reliable experimental data for validation and evaluation of typical microwave radiative transfer model simulation in complex situations, which lay a theoretical basis for revealing the effect mechanism of the forest on the microwave radiative characteristics of snow and snow depth retrieval accuracy.