Research And Application On Key Technology Of Long Time Snow Cover Dataset Generated Over Qinghai-Tibetan Plateau

Snow is an important member of the cryosphere and basis for study of the Earth’s radiation balance, global climate change, ecological environment. The Qinghai-Tibet Plateau (QTP), with an average elevation above 4000m, is located in the middle and low latitudes in the Northern Hemisphere. It is not only one of mainly China’s three snow areas, but also an important snow area of the Eurasian continent. The melting snow can supply many rivers and influence the global climate changes. However, large-scale and excessive snow can also bring inconvenience to local residents, and restrict the development of animal husbandry severely. Therefore, it is important to carry out monitoring the snow cover over the QTP.In this paper, the QTP is considered as the study area and the satellite remote sensing data are used to establish the snow discrimination model supplied by ground observation stations’data. The long time series of the snow cover dataset is generated by the multi-temporal and multi-sensor cloud removal technology. Then analyzing the temporal and spatial variation characteristics of snow cover, carrying out researching on snow cover automatic extraction system. It can provide a scientific basis for the study of snow warning and atmosphere couping system.Firstly, we complete data preprocessing of the FY-3/VIRR and FY-3/MERSI data and propose a general method of identifying snow pixel with multi-channel by dynamic threshold. Through the accuracy verification, it proves that the algorithm is feasible and easy to achieve good results, the snow extraction results can be used as a reliable data source of monitoring snow cover over the QTP.Secondly,7-day rolling combination algorithm and multi-sensor combination algorithm are presented to reduce cloud and through combining the two algorithm to generate long sequence of the QTP snow cover dataset. It can effectively reduce the interference of cloud and improve the snow monitoring accuracy. Therefore, it is more suitable for snow distribution research over the QTP.Thirdly, in order to meet the massive remote sensing data processing, we design and develop an automatic system for snow cover extraction based on IDL and C# languages. The system has a high degree of automation, fast processing speed, good stability and friendly interface, which promote domestic and foreigh satellite remote sensing data to be fully applied and spread.Finally, based on the long sequence of the snow cover dataset and snow depth dataset, we analyze the temporal and spatial variation characteristics of snow cover, snow days and snow depth over the QTP. It has great importance to QTP snow resource utilization, climate research, environmental protection and disaster prevention.