 |
|
|
|
|
Stephen J. Ghan Pacific Northwest National Laboratory P.O. Box 999/K9-24 Richland, WA 99352 email: Steve.Ghan@pnl.gov Although we understand the basic physics of direct and indirect effects quite well now, and have applied much of that understanding to global aerosol models, there remains a major source of uncertainty: the poor representation of the influence of clouds on subgrid pollutant (primary aerosols, secondary aerosols, and precursor gases for secondary aerosols) transport, transformation, and removal in global models. This source of uncertainty has remained elusive and will continue to be elusive unless a bold step is taken. We propose to use grid cell mean statistics from cloud resolving models (CRMs) embedded within each global model grid cell to drive a physically-based treatment of pollutant processing by clouds. For example, the grid cell mean cloud mass flux can be used to treat vertical transport of pollutants, the mean updraft velocity can be used to determine the aerosol activation, the mean cloud fraction and in-cloud water content can be used to treat aqueous chemistry, and the mean precipitation fraction and precipitation rate can be used to treat precipitation scavenging. We call this the Explicit Cloud-Parameterized Pollutants (ECPP) approach. The method will be tested first using a single column model driven by statistics derived from cloud-resolving simulations of pollutant transport, transformation, and removal by clouds for a limited domain. It will be further tested using CRMs embedded within each grid cell of a regional circulation model, with pollutant processes either embedded within the CRMs or parameterized using the ECPP method. When the ECPP is mature it will be applied to a global climate model already running with embedded CRMs. [back to ASP ST Membership] |