I am working with a global fully distributed (gridded) hydrological model and I am having some issues regarding two sets of data extremely important for the correct representation of the soil evaporation and evapotranspiration processes: Leaf Area Index (LAI) and Soil Depth (rooting and total).
As I'm not an expert in neither of these two thematic areas, my perplexities regards both the quality of the source data and how the data should be unmixed basing on land use classes: as for instance forest, irrigated crops, and (rainfed) shrubland.
I was wondering whether you could point to some good reference datasets and some unmixing methodology that you found particularly useful in your experience.
EDIT- just to clarify why the unmixing is needed: the observed (remotely sensed) LAI is usually made available at a certain spatial resolution (good products are offered in the order of magnitude of few hundred meters or 1 km), and these territorial units are characterized by different land use classes. Having a correct representation of the LAI and soil depth for the specific cell and the specific land use class is crucial to analyze the change in the hydrological process associated to a dynamic land use change throughout the simulations.