9306754 Pielke The poorly understood impact of a variety of land/surface factors on the atmosphere is one of the major uncertainties limiting atmospheric predictability. The Principal Investigator will continue an ongoing research program studying land/surface processes with the following four specific goals: 1. Validation of a vegetation/soil parameterization and exploring coupling of a mesoscale atmospheric numerical model to an ecosystem dynamic model. 2. Investigation of the effect of landscape patchiness on: boundary layer structure; development of mesoscale thermally-forced flows and deep cumulus convection. An emphasis will be placed on effects of Amazon deforestation. 3. Evaluation of the predictability of atmospheric flows as a function of spatial scales of land/surface patchiness. This includes further investigation of the hypotheses that inhomogeneous land/surface features can improve predictability as compared to a homogenous surface. 4. Development of a parameterization of landscape effects for large-scale atmospheric models. If successful, this research will produce better land/surface parameterizations which will lead to improvements in atmospheric numerical modeling including both short term forecasting (mesoscale) and climate modeling. ***