Parameterization of Riming Intensity and Its Impact on Ice Fall Speed Using ARM Data

Abstract

Riming within mixed-phase clouds can have a large impact on the prediction of clouds and precipitation within weather and climate models. The increase of ice particle fall speed due to riming has not been considered in most general circulation models (GCMs), and many weather models only consider ice particles that are either unrimed or heavily rimed (not a continuum of riming amount). Using the Atmospheric Radiation Measurement (ARM) Program dataset at the Southern Great Plains (SGP) site of the United States, a new parameterization for riming is derived, which includes a diagnosed rimed mass fraction and its impact on the ice particle fall speed. When evaluated against a vertical-pointing Doppler radar for stratiform mixed-phase clouds, the new parameterization produces better ice fall speeds than a conventional parameterization. The new parameterization is tested in the recently developed Geophysical Fluid Dynamics Laboratory (GFDL) atmospheric model (AM3) using prescribed sea surface temperature (SST) simulations. Compared with the standard (CTL) simulation, the new parameterization increases ice amount aloft by ?20%?30% globally, which reduces the global mean outgoing longwave radiation (OLR) by ?2.8 W m?2 and the top-of-atmosphere (TOA) shortwave absorption by ?1.5 W m?2. Global mean precipitation is also slightly reduced, especially over the tropics. Overall, the new parameterization produces a comparable climatology with the CTL simulation and it improves the physical basis for using a fall velocity larger than a conventional parameterization in the current AM3.