Biological and Environmental Research - Earth and Environmental System Sciences
Earth and Environmental System Modeling
07 June 2014

Assessing the CAM5 Physics Suite in the WRF-Chem Model


Coarse resolution global climate models do not accurately portray variability of the climate system at regional and local scales, such as extreme precipitation and pollution events. The complexity of climate models also makes it difficult to quickly develop and test new parameterizations for processes such as aerosol and cloud formation. To address these challenges, Department of Energy scientists at Pacific Northwest National Laboratory, in collaboration with a researcher from University of Wyoming, ported the CAM5 physics parameterization suite to the WRF-Chem model to minimize inconsistencies between the global and regional models, enabling more efficient parameterization development and to provide a consistent treatment of physical processes when configuring the CAM5 and WRF models for climate downscaling. Evaluated using DOE’s Atmospheric Radiation Measurement (ARM) Climate Research Facility observations, field campaign measurements, and satellite retrievals, simulations using high model resolution in WRF with the CAM5 physics suite exhibited decreased model biases versus typical coarse simulations with CAM5, boding well for anticipated higher resolution simulations in future versions of CAM. For example, improvements were seen in low aerosol concentrations near the surface in the Arctic and high cloud susceptibility to aerosols at increasing horizontal resolution without any modification to the model physics. The team also determined that the collocation of aerosols and clouds is very different at different model resolutions, resulting in differences in long-range aerosol transport and cloud susceptibility to aerosols. This framework, released as part of WRF (and WRF-Chem) version 3.5 in 2013, can be used to guide future parameterization development that improves the representation of aerosols, clouds, and aerosol-cloud interaction processes and their sub-grid variability and to perform more consistent regional downscaling modeling than was previously possible.