Dr. Joyce Penner

Dr. Joyce Penner

Ralph J. Cicerone Distinguished University Professor of Atmospheric Science

Department of Climate and Space Sciences and Engineering, College of Engineering, University of Michigan

Global Climate Modeling

Figure 6: This figure compares the response of 3 climate models to different treatments in aerosol/cloud interactions. There is wide variation due to basic differences in the choices made to represent aerosols and clouds in climate models. This figure compares the response of 3 climate models to different treatments in aerosol/cloud interactions. There is wide variation due to basic differences in the choices made to represent aerosols and clouds in climate models.

Aerosol/climate interactions

We are coupling our aerosol model to both the CAM5 NCAR climate model and the GFDL climate model. Our goal is to understand the variation of the response of clouds and cloud forcing by aerosols in these two major U.S. climate models. We hope to be able to use this intercomparison to determine the range of uncertainty in GCM’s and to examine the “best” method for parameterizing these interactions by comparing to observations.

Non-hydrostatic climate models

One problem with the treatment of aerosol/cloud interactions in GCM’s is that they do not fully resolve these interactions (their grids are too large). We are building an adaptive grid dynamical core for climate models that can span from hydrostatic treatments (appropriate at large grid resolutions) to non-hydrostatic treatments (for grid size < 10 km). This approach would allow us to fully resolve the response of clouds to changes in aerosols when conditions merit a high resolution treatment. The model is based on using a Lagrangian vertical coordinate that is able to seamlessly treat both hydrostatic and non-hydrostatic regimes.

Figure 7: The tracer distribution at different times. Clockwise: days 1, 10, 25, and 100. The wire shows the refinement levels. The tracer distribution at different times. Clockwise: days 1, 10, 25, and 100. The wire shows the refinement levels.
Figure 8: A rising bubble of air in our non-hydrostatic Lagrangian vertical coordinate model.A rising bubble of air in our non-hydrostatic Lagrangian vertical coordinate model.