Parallelization of a Subgrid Orographic Precipitation
Scheme in an MM5-based Regional Climate Model

L. Ruby Leung (1), John G. Michalakes (2), and Xindi Bian (1)

(1) Atmospheric Science and Global Change Resource,
Pacific Northwest National Laboratory,
Richland, WA 99352, USA
{ruby.leung, xindi.bian}@pnnl.gov

(2) Mathematics and Computer Science Division,
Argonne National Laboratory, 
Argonne, IL 60439, USA
michalak@mcs.anl.gov


Regional Climate Models (RCMs) are practical downscaling tools to yield
regional climate information for assessing the impacts of climate
variability and change.  The Pacific Northwest National Laboratory
(PNNL) RCM, based on the Penn State/NCAR Mesoscale Model (MM5),
features a novel subgrid treatment of orographic precipitation for
coupling climate, hydrologic, and ecologic processes at the watershed
scale.  The parameterization aggregates subgrid variations of surface
topography into a finite number of surface elevation bands.  An airflow
model and a thermodynamic model are used to parameterize the orographic
uplift/descent as air parcels cross over mountain barriers or valleys.
The parameterization has significant performance advantages over
nesting to achieve comparable resolution of climate information;
however, previous implementations of the subgrid scheme required
significant modification to the host MM5 model, prohibiting its
incorporation within the NCAR-supported community version of MM5. With
this effort, software engineering challenges have been addressed to
incorporate, parallelize, and load-balance the PNNL subgrid scheme with
minimum changes to MM5. The result is an efficient, maintainable tool
for regional climate simulation and a step forward in the development
of an MM5-based community regional climate model.