Abstract ATM-9321354 Shukla, Jagadish Schneider, Edwin K. Kinter, James L. Straus, David M. Institute of Global Environment and Society, Inc. Title: Predictability and Variability of the Present Climate This research encompasses a number of closely related projects which are aimed at understanding the low frequency (monthly to decadal) variations of the coupled climate system consisting of the atmosphere, hydrosphere, land biosystems, oceans and ice, primarily in the present climatic regime. Goals include both the ability to simulate these low frequency variations with complex numerical models of the individual and coupled systems, and to estimate the degree to which these variations are predictable. The methods to be used will involve extensive experimentation with existing numerical models and schemes to couple them, and extensive diagnosis and analysis of observational data from both existing records and from the retrospective assimilation of atmospheric data planned by the major operational centers of the world for the near future. In addition to the intrinsic importance of understanding and predicting the low frequency variability of the climate system, this work will help to define the accuracy and coverage requirements for current and future climate observing systems in order to adequately monitor the current behavior of the climate system. 1. Predictability of the coupled ocean-land-atmosphere system. The predictability of this system on both the seasonal and interannual time scales will be systematically explored, the former by a large number of seasonal atmospheric simulations, and the latter from integrations of the fully coupled system. 2. Land-surface-climate interactions. The importance of vegetation change, albedo feedback and the global boundary conditions in the Sahel drought will be explored. The potential impacts of man's activity on climate through the (hypothetical) doubling of the world's deserts, global deforestati on and the desiccation of the Aral Sea region will be addressed. 3. Climate diagnostics. Diagnostics of the planned reassimilation of atmospheric data includes a detailed examination of the hydrological cycle, and global energy cycle and the critical tropical-extratropical interactions. These efforts will be aided by the availability of consistent estimates of the subtle quantities of divergence, specific humidity and diabetic heating. 4. Simulation of the present climate. To test the ability of numerical models to simulate the observed low frequency anomalies, the PIs plan an integration of the atmospheric model for about 100 years using observed boundary conditions. Similarly, they will integrate the coupled climate system model for several hundred years in order to test the realism of its overall statistical behavior. 5. Climate observing system simulation. The PIs will estimate the sensitivity of the atmospheric circulation to many types of changes in the boundary conditions. They will "observe" numerically modelled climate system using methods similar to the current (and future) observing systems for the real climate to explore their sampling and accuracy characteristics systematically. This research is important because it seeks to enhance knowledge about climate processes and improvements in predicting climate variation and change. Part of this research is funded under the USGCRP CMAP project.