Research On The Single-channel Algorithm For Land Surface Temperature Retrieval From Landsat8 Data

Land surface temperature( LST), as an important parameter of the surface energy balance and water balance, plays a very important role in the energy exchange between land surface and the atmosphere, and the retrieval of LST by remote sensing has important reference value in the related research fields for climate, geological exploration, agricultural production, ecological protection and so on. Especially in recent years, with the development and progress of remote sensing technology, satellite data of high spatial resolution and high temporal resolution provides great possibility for the accurate retrieval of LST, the retrieval of LST is getting more and more attention and application.The Landsat8 satellite was launched in February 2013, compared with the same series of satellites launched before, it not only inherits the characteristics of high spatial resolution in thermal infrared channel, but introduces the quantum infrared photon detection technology in the TIRS for the first time, and further enhances its sensitivity of thermal infrared band, which is more conductive to separate the LST from atmospheric temperature. So, compared with the traditional satellite data, the Landsat8 satellite data is more suitable for accurate analysis of LST and the distribution of heat in space.So far, as the related research of thermal infrared field using Landsat8 data is few, this paper firstly recalculates the empirical formula in the Jim?nez-Mu?oz&Sobrino single channel algorithm combining the TIR1 response function, and gets the new coefficient matrix for Landsat8 satellite; then according to the theory of Jim?nez-Mu?oz&Sobrino single channel algorithm, revises the parameters acquisition method in the traditional radiative transfer method to solve the problem of high dependence on the real time sounding data in the practical application. Afterthat, the accuracy evaluation and sensitivity analysis for the two kinds of algorithms are carried out respectively using the Landsat8 simulation data. Finally, this paper selects several Landsat8 image data in different seasons of the same year, and uses these two single channel algorithms to retrieval the LST of Hailar study area, then compares the LST derived from the two algorithms to the simultaneously acquired LST in the study area. The results show that the errors between them are very small, which proves that the two single channel algorithms in this paper are feasible.In addition, although this paper has completely implemented the two kinds of single channel algorithms through a lot of work, and achieved ideal effect, due to the limitations of the remote sensing data and ground truth data, the unity of the study area and other problems, the research content of this paper remains to be further expanded.