Land Surface Remote Sensing and Their Application in Climate Models 

Our group has been actively engaged with scientists who collectively have considerable experience in climate modeling and in the development of various MODIS algorithms to make better use of the latest NASA satellite data to improve the representation of land in climate models. Using the NASA remote sensing land products, such as the land surface albedo, leaf area index (LAI), fraction of photosynthetically active radiation (FPAR), fractional vegetation cover, soil emissivity and plant phenology, we compared these data with climate models to identify and attribute the model deficiencies in land surface parameters and parameterization, generated more realistic and consistent land surface datasets, developed simple parameterization schemes based on satellite observation for climate models, and evaluated the impacts of improvements on the simulated climatology of land surface variables in climate models. For example, our studies found that the largest albedo biases over semiarid and snow covered vegetated surfaces and the largest emissivity biases over the arid and semi-arid regions in climate models, which are mainly attributed to lack of land surface heterogeneity characterization and inaccurate specification of land surface parameters in the models. See details here

  Our current research efforts have also been focused on the evaluations of soil moisture results from climate models and the generation of a vegetation water content dataset using passive microwave observations. Because of its sensitivity to both soil moisture and vegetation properties, and the ability to penetrate the atmosphere, passive microwave radiation observed on TOA has demonstrated unique strength for monitoring land surface characteristics. For more details, click here