Study On The Application Of Thermal Infrared Remote Sensing Land Surface Temperature Retrieval Methods And Comparative Analysis

Land surface temperature is playing a very important role in ground -atmosphere interaction. It is a key parameter in the global change studies, playing very important meanings in researches such as hydrology, ecology, environment and biological geochemistry, and there is important application value in calculating agricultural weather, hot inertia etc, Achievement of quantitative and accurate land surface promotes the process of prediction of drought disaster,study of crops , estimation on crop output, numerical weather forecast,global climatic change and global carbon balance. So it becomes one of important tasks in quantitative remote sensing research to make use of satellite remote sensing to retrieve land surface temperature at present. This text has explained the principle and different method of land surface temperature remote sensing retrieval and gave some comments on some key problems and advantages and disadvantages of different land surface temperature. Finally, this text expected the development trend of remote sensing land surface temperature retrieval.This text introduced basic inference procedure of present existing thermal infrared remote sensing land surface temperature methods such as split window algorithm, single window algorithm and multi-channel algorithm so on. And explored its actual application and compared the results of land surface temperature retrieval of different type of data. By comparison, it can be found out that single window algorithm is high dimension and less sensitive to land surface emissivity compared with multi channel window algorithm aiming at NOAA, MODIS etc. In terms of retrieval technology and accuracy, single window algorithm has big advantages. But it needs a lot of funds if it is used to retrieve large regional land surface temperature. At present split window algorithm has broadest application and the most mature algorithm. Accuracy is comparatively high. Relatively it doesn't need atmospheric outline value. But at present it also wasn't perfect. On the foundation of thermal infrared remote sensing land temperature retrieval method, I compared and analyzed three kinds of land surface temperature retrieval methods. I made use of MODIS split window algorithm and temperature and emissivity separation method respectively and get the result of land surface temperature retrieval result of MODIS and ASTER data in 19 November 2003. The land surface temperature retrieval result of MODIS shows that the lowest temperature is 293.76 K and the highest land surface temperature is 307.17K. The result of ASTER data land surface temperature retrieval shows that the lowest land surface temperature is 292.86K and the highest land surface temperature is 308.37 K. and I also made use of MODIS and ETM+ data to retrieve the land surface temperature with the split window algorithm and single window algorithms on January 10, 2003. The result of MODIS land surface temperature shows that the lowest land surface temperature was 284.97 K and the highest land surface temperature was 294.10K. The result of ETM+ land surface temperature retrieval on January 10, 2003 shows that the lowest land surface temperature was 283.68K and the highest land surface temperature is 292.88K.By comparison, it can be known that temperature difference of different land features is prominent. Among them the urban surface temperature is the highest. The temperature of river and lake is the lowest, secondly the forest land. It can be concluded that expansion of city green area can city water body can prevent or weaken the urban heat island effect. In a word, study of land surface temperature retrieval mainly based on split window algorithm method, in order to improve the accuracy of land surface temperature retrieval, different authors have carried on research from multi-respects. But because of the complexity of the land surface temperature retrieval, until now, research of land surface temperature retrieval still takes feasibility research as the core mainly. Correction of atmospheric effect, unknown of land surface emissivity and skin effect of land temperature retrieval still restrains the actual application of land surface temperature remote sensed retrieval method.