Submitted to: Book Chapter
Publication Type: Book / chapter
Publication Acceptance Date: 4/16/2007
Publication Date: 5/20/2008
Citation: Jacob, F., Schmugge, T., Olioso, A., French, A.N., Ogawa, K., Petitcolin, F., Chehbouni, G., Pinheiro, A., Privette, J. 2008. Potential of thermal infrared remote sensing for the monitoring of land surfaces. Book Chapter. In: Advances in Land Remote Sensing: System, Modeling, Inversion and Application (S. Liang, Ed.). Chapter 10:243-269. Interpretive Summary: The use of thermal infrared remote sensing over vegetated land surfaces can be very useful for understanding hydrological, meteorological, and related agricultural growth processes over short and long term studies. With thermal infrared, the physical temperature of plants can be measured from satellites, which in turn can help reveal how much water is available for plant growth and whether or not plants are water-stressed. Thermal infrared data can also reveal regional and temporal changes that reflect climatological and landcover changes not readily seen from surveys on the ground, including subtle shifts in vegetation patterns. How these practical short and long term applications might be realized is the subject of this review paper, which considers technological and algorithmic developments of thermal infrared remote sensing from two perspectives. In one, the theoretical basis for measuring vegetation temperatures with thermal infrared data is discussed. In the other, practical ways to accurately temperatures are accurately reviewed. Future improvements with thermal infrared remote sensing—which will allow vegetation temperatures to be measured as easily as satellites can currently observe ocean temperatures—will be achieved by finding ways to merge theory with experience and by greatly increasing the number of thermal infrared-capable satellites.
Technical Abstract: Thermal infrared remote sensing over vegetated land surfaces is expected to provide valuable information for documenting soil-vegetation-atmosphere exchanges about heat, water and mass. On the one hand, the community benefits of various TIR remote sensing observations according to four dimensions: spatial, temporal, angular and directional. On the other hand, several land surface products are targeted for documenting various land surface models dealing with different processes. For linking the remotely sensed observations to the targeted products, the community has developed modeling tools and inversion strategies. However, using the latter faces several difficulties to be overcome. Due to the numerous environmental factors which drive the radiative regime within natural media, complex modeling tools required significant documentation, making their inversion extremely complicated. Inverting simple modeling tools faces the ill posed problem, which can be fortunately solved using a priori information from various sources. Overall, the community made significant progress for the recovery of some TIR derived land surface products, with performances rather close to the required accuracies. Brightness and radiometric temperatures can be derived with accuracies close to 1 K, though refinements still are necessary. The use of aerodynamic temperature, or its replacement by radiometric temperature, still is a serious challenge, regarding the difficulties faced for adequately performing the inherent parameterizations. Therefore, the next challenge for the community is the derivation of more elaborated products: soil and vegetation temperatures with or without their sunlit and shaded components, and canopy temperature profile. This requires the use of multiangular systems, currently under development at the ground level. Such systems will be of interest for developing inversion strategies from methods widely used over the solar domain. Finally, it is worth relocating these investigations within their context. On the one hand, some of the difficulties can be overcome via the development of land surface process models. On the other hand, serious challenges have to be handled with temporal and spatial issues, in the context of monitoring land surface processes with adequate spatial scale and temporal samplings.