Dear WG4 Members,

I attended the joint EUROCS/GCSS workshop last week in Lisbon. EUROCS
is a European Union project similar in scope to GCSS (see their web
page at http://www.cnrm.meteo.fr/gcss/EUROCS/EUROCS.html if you need
more information). The WG1 of GCSS (boundary layer clouds) had their
meeting there as well; WG1 and boundary layer working group of EUROCOS
work very closely. Below I include a report from the discussions at the
workshop. It is relevant to future activities within WG4, so please
keep reading and give me you feedback.
 
1. THE NEXT WG4 CASE.

We agreed in Silver Springs meeting last fall that the next test case
will be associated with the diurnal cycle of deep convection over
land.  In Lisbon, Christian Jakob from ECMWF presented an appealing
case for the convection over the Amazon basin. An existence of the data
from the TRMM/LBA campaign in this region and an experience some WG4
members have with this case (NASA Goddard group for example) would be
very useful. I hope we can discuss this case before the October WG4
meeting in Boulder and use it as the next deep convection case. The
fact that large-scale models allow deep convection to develop much too
early during the day over the Amazon basin (one of the major points in
Christian's presentation) make this case highly relevant to the
convection parameterization issue within SCMs. Moreover, LBA
observations provide information not only on the boundary layer
development early during the day, but also on the modifications
associated with precipitation and downdrafts in late afternoon hours
(one of conclusions of SCMs running the ARM case was that the downdraft
mass flux was much smaller than predicted by CRMs; these certainly
affects the boundary layer characteristics). Running CRMs for this
case will be important for the improvements of the SCMs, which is one
of the goals of the GCSS. Setting up this case requires some
decision-making upfront, however, and your input is needed.

The deep convection over the LBA area falls into two categories (Tao,
please correct if I miss anything): i) monsoon type with surface
westerly flow and popcorn-type deep convection, and ii) break type with
surface easterly flow and organized squall line type convection.  The
first type should be our primary interest as in this case convection
develops gradually over the domain during the day and the influence of
organized systems propagating from outside the area (as typically
happens in the break type convection) is likely not an issue. According
to W.K. Tao, there is only a single day (February 23) if we insist on
having aircraft microphysical measurements, TRMM satellite overpasses,
and radar data to compare model results with the observations. However,
there are other cases I think where the radar data are available (e.g.,
Rutledge et al. paper 5B.5 from the 24th AMS Conference on Hurricanes
and Tropical Meteorology, 2000, Ft. Lauderdale, shows some radar data
from February 25th case). I feel that we should select a day where
there is a continues radar coverage from the morning hours (when
convection begins) till the deep convection phase in the afternoon, if
at all possible. Satellite data should be also useful with this
respect. These data should allow us to verify the model-predicted mean
rate at which convection deepens over a large area as the diurnal cycle
progresses.

Unfortunately, there are some issues as far as the forcing of the model
is concerned. LBA had a sounding network, but I am not sure if the
budgets have been derived from the data yet (we need Q1 and Q2 to drive
our models) and if we can trust these products. Another possibility is
to use large-scale advective tendencies from the ECMWF model, but since this
model features too rapid convective development in the late morning,
the large-scale vertical velocity may have nothing to do with the real
world. We may decide just to neglect the large-scale forcing
altogether and rely on the surface and radiative forcings only.
However, according to W.K. Tao, surface fluxes derived from LBA
observations seem too high. This statement is based on some simulations
they did using the February 23rd case. I think some experience from the
EUROCOS case (see below) might help us. I can run a few test cases
before we agree on the final TRMM-LBA setup.

An important aspect of this case is that we may benefit from
collaboration with the WG1 (boundary layer clouds) which recently
considered a scattered shallow convection case and growing boundary
layer (a report on that was presented in Lisbon). The agreement between
CRMs (LES models in fact) was impressive and it was emphasized that SCMs
(a few that participated) require a lot of work to match CRMs in the
scattered shallow convection case.

However, before we agree on the details of the LBA-based case, we can
benefit from the EUROCOS deep convection case.  EUROCOS deep convection
participants agreed to look into an idealized case of diurnal cycle of
convection over land based on the first day of ARM subcase A. As you
may remember from WG4 previous meetings (and CRMs/SCMs papers now
submitted to QJ), the subcase A features two small events in the first
part of the period, and a major event towards the end. The first of the
two weaker events is dominated by diurnal effects, with a weak
large-scale forcing. You may also remember that original 2-km
horizontal resolution CRM simulations of this subcase featured a delay
in precipitation development. This delay was removed once a higher
horizontal resolution was used (both Jon Petch and Marat Khairoutdinov
presented some supporting results in Silver Springs).

Francoise Guichard and Jon Petch (leaders of the EUROCOS deep
convection group) agreed to prepare details of this case by the end of
June. The proposal is to use a smaller domain in 2D (say, 250 km), a
horizontal resolution of 250 m, and run the case for, say, 3 days,
using the perpetual large-scale forcing and surface fluxes. I think I
will be able to run this case before the October meeting. I will keep
all of you posted and hope that at least some of you will find time to
run this case and present results in Boulder. Participation of SCMs in
this case is important and the experience gained will benefit the
design of the LBA-based case.

2. UPCOMING MEETINGS.

WG2 (cirrus clouds, chaired by Dave Starr) is considering to have a
joint workshop with our group in Boulder in October. One possibility is
that they meet on Wed-Fri, October 24-26, which will give a one day
overlap with our meeting (which is Mon-Wed). They are still thinking
about this. I will let you know as soon as they decide. Some of us
might be interested to try our microphysical schemes on test cases WG2
is using.

Below is information about two upcoming workshops that some of you
might be interested in. Please contact Wei-Kuo Tao
(tao@agnes.gsfc.nasa.gov) for further details.
 

Second TRMM Latent Heating Algorithm Workshop

TRMM Heating Products: Requirements and Applications

M. Moncrieff, A. Hou and W.-K. Tao

October 10, 11 and 12 (half day) 2001 at NCAR

Objectives:

* Report on conclusions/action items from the 1st workshop and on
TRMM heating products in general.

* Discuss the use of cloud-resolving models in TRMM heating algorithm
development/refinement; in the analysis of TRMM-related field
experiments; in collaborative activities between TRMM and GCSS WG4;
and in understanding latent heating processes (e.g., organization,
convective/stratiform regions, role of cumulus with tops below the
melting level).

* Discuss the application of TRMM latent heating products to the
parameterization of tropical convection.

* Discuss the application of TRMM latent heating products to data assimilation.

* Identify further TRMM data products and analyses that may be
required in future.
 

Cumulus Parameterization Mini-Workshop

W.-K. Tao, D. Starr and Y. Sud
(Nov 13-15 2001, NASA/Goddard Space Flight Center)

Objectives: To review the state-of-the-art of cumulus
parameterization schemes used in GCMs, climate models and NWPs.  Also
to discuss the problems as well as the directions of new research
associated with cumulus parameterization.

A. Arakawa/UCLA: To discuss the theory of cumulus parameterization
(with review/overview)

D. Randall/CSU: To discuss the use of CRMs for cumulus
parameterization problem (GCSS)

A. Del Genio/GISS: To discuss the cumulus parameterization schemes
used in Climate models

S. Lord/NCEP: To discuss the cumulus parameterization used in NWP
 

Participants:

Invited Talks on the state-of-art of cumulus parameterization schemes
at DAO, NCAP, ECMWF, GISS, NCAR and Universities
Invited Talks - observational (need to include momentum transport, cirrus)
Invited talks - CRMs/SCMs
Radiation

Regards,

Wojtek.

=====================================================================
Dr. Wojciech W. Grabowski

Mesoscale and Microscale Meteorology Division (MMM)
National Center for Atmospheric Research (NCAR)
PO Box 3000, Boulder, Colorado 80307-3000

overnight mail: 3450 Mitchell Ln, Boulder, CO 80301

e-mail: grabow@ncar.ucar.edu
phone:  (303)-497-8974
fax:    (303)-497-8181
=====================================================================