Estimation Of Component Temperatures Of Vegetative Canopy With Vis Nir And Tir Multiple Angular Data Through Inversion Of Vegetative Canopy Radiative Transfer Model

Estimation of Component Temperatures of Vegetative Canopy with Vis/NIR and TIR Multiple-angular Data Through Inversion of Vegetative Canopy Radiative Transfer Model

The separation of component temperature is the basic step for the application of two-source algorithm. Multi-angular thermal infrared measurements provide a chance for the estimation of component temperatures (namely, soil and vegetation temperatures) with remotely-sensed data. The objective of this study is to explore the factors that affect the estimation of component temperatures and propose new algorithm for inverting the canopy radiative transfer models to compute component temperatures.The objectives of this dissertation include: (1) finding an appropriate candidate leaf angle distribution functions for modeling and inversion, (2) evaluating the scaling behavior of Beer's law and its effect on the estimation of component temperatures, (3) proposing an analytical model for directional brightness temperature at top of canopy, (4) retrieving component temperatures with neural network and simplex algorithms. The effects of leaf angle distribution function on extinction coefficient, which is a key parameter for simulating the radiative transfer through vegetative canopy, is explored to improve the radiative transfer modeling. These contributions will enhance our understanding of the basic problems existing in thermal IR remote sensing and improve the simulation of land surface energy balance. Further work can be conducted to continue the enhancement and application of proposed algorithm to remote sensing images.

IEEE International Geoscience and Remote Sensing Symposium Proceedings

IGARSS.