PLEASE POST
MMM SEMINAR NCAR
Why
use spectral (bin) microphysics to simulate precipitating systems?
Barry Lynn, Alexander
Khain, and Jimy Dudhia
Department
of the Atmospheric Sciences, The Hebrew University of Jerusalem
And
National Center for Atmospheric Research
Abstract
Forecast of the rate, duration, amount and spatial distribution of
precipitation remains a difficult problem for current high-resolution
weather prediction mesoscale models. These models use, as a rule,
bulk-parameterization schemes for the description of cloud microphysics. In
these schemes, the size distributions of raindrops and ice particles are
preset (e.g., they might use the Marshall-Palmer distribution). This
simplification allows one to reduce a complex equation system for microphysical
processes to a few equations for mass contents (or mass contents and number
concentrations). This makes the schemes computationally efficient,
but also introduces errors in the description of microphysical mechanisms
that are responsible for formation of precipitation. We will present several
examples of such errors.
A more precise description of microphysical processes is referred to
as spectral (bin) microphysics (SBM). This takes into account factors
influencing drop spectrum formation (e.g., atmospheric aerosols) that are
required for improved simulation of precipitation formation. SBM is based
on solving an equation system for size distribution functions that includes
water drops, ice particles, and atmospheric aerosols. Each size distribution is
described by several tens of bins representing particles of certain mass, bulk
density and shape. This method is more time consuming as compared to
bulk-parameterization schemes, but the use of PC cluster technology makes
possible the use of the SBM in a three-dimensional simulation environment as a
potent tool for microphysical research.
We implemented SBM (developed by
the Hebrew University of Jerusalem) into the three-dimensional mesoscale model
known as MM5. The MM5-SBM was used to simulate a rain event over Florida on 27
July 1991. This event was accompanied be a squall-line formation. It is shown
that the MM5-SBM reproduces precipitation rate, rain amounts and location,
radar reflectivity, and cloud structure better than bulk parameterizations
currently implemented in MM5.
Possible applications of the first mesoscale model with spectral (bin)
microphysics are discussed.
Thursday, September 2, 2004, 3:30 PM
Refreshments 3:15 PM
NCAR-Foothills
Laboratory
3450 Mitchell Lane
Bldg 2, Rm 1022