Monitoring land surface processes with thermal infrared data: Calibration of SVAT parameters based on the optimisation of diurnal surface temperature cycling features

Land Surface Models (LSM) have been designed to describe water and energy transfers at the soil–vegetation–atmosphere interface, and are therefore essential in many environmental disciplines. These numerical models, driven by the boundary conditions in the atmosphere and in the soil, require adequate knowledge of those vegetation and soil characteristics which are determinant in the characterisation of mass and energy transfers. In view of the fact that, firstly this information is often only partially known, and secondly the transfers are sometimes incorrectly represented, these models can rapidly drift and need to be regularly corrected. To this aim, remote sensing is a promising tool and many studies are currently devoted to the development of assimilation techniques to control their inputs or internal variables. The research presented in this paper contributes to this effort. Its ambition is to explore new methodologies, designed to make use of remote sensing thermal infrared data recorded from space. This study is based on the analysis of links between the characteristics of the diurnal cycle of the surface brightness temperature and the soil–atmosphere interface parameters and variables. The proposed methodology takes advantage of these temperatures cycling features, instead of absolute temperature values, to calibrate the LSM. The results show that the model parameters have a significant impact on the diurnal temperature dynamics, sometimes to a greater extent than on the temperature itself, and that these relationships have diurnal and seasonal variations. As a consequence, the use of TIR data for LSM calibration can be optimised by considering only those parts of the information which are really relevant to parameter calibration.