Just like many others on the lab, I joggle between several projects which involve the impacts of water vapor, clouds and aerosols on climate change. I work closely with my colleague in the Microwave Atmospheric Science group, Dr. Jonathan Jiang, on most of these projects. One of the topics we have been working on is to analyze observed cloud profiles and use them to evaluate cloud simulations in climate models. Our unique approach is to examine cloud profiles binned by large-scale state variables, so as to directly link cloud structure with ambient environment conditions, the basis for cloud parameterization in models. This type of analysis results are often called “Bony Diagram”, a tribute to Dr. Sandrine Bony’s pioneering work (e.g. Bony et al, Climate Dynamics, 2004). We assembled a large amount of data (most of them from NASA satellites) to represent large-scale regimes and sort CloudSat observed liquid and ice water content by these “regime discriminators”. We found two dominant cloud modes, the “deep” and “shallow” clouds, associated with distinctly different large-scale regimes. There is some trace of middle-level clouds, but the signal is rather weak. We then compared the observed cloud structure with 3 leading climate models in the US, the GEOS-5, NCAR CAM and GFDL AM2. We found that the model simulated cloud profiles have large differences from the CloudSat observations: the cloud ice amounts in deep convective clouds are underestimated, the liquid water contents in the boundary layer clouds are also too little, and all models have some middle-level clouds in the upwelling regime which CloudSat lacks. The comparison probably indicates problems with both CloudSat retrieval and model physics. We are working on more diagnosis to understand the model-data discrepancies. This work is partly supported by the JPL's Global Change and Energy (GC&E) strategic fund. Dr. Joao Teixeira strongly supports our efforts. Vince Perun in the MLS group helps a great deal in processing the data. I gave a one-hour talk about this study at the Aerosol-Cloud Seminar organized by Dr. Dong Wu on May 27, 2009.