BACK TO KNMI EUROCS PAGE


EUROCS Intercomparison for Single Column Models

Diurnal Cycle of Cumulus Clouds over land

Geert Lenderink and Pier Siebesma

Case | Motivation | First Experiment | Outlook | Results | Timetable 



Case

The present SCM intercomparison is based on the 6th GCSS WG 1 intercomparison case. This case is based on observations at the ARM  site on the Southern Great Plains (USA) on 21st june 1997. During that day Cumulus clouds developed at the top of an initially clear convective boundary layer. The intercomparison case has been developed and coordinated by Andy Brown (UKMO), responsible for the LES models, and Adrian Lock (UKMO), responsible for the SCM models. The original case description by Andy Brown can be obtained here (compessed postscript).

Motivation

Results of several LES models have been presented at a meeting in Boulder, January 2000. On average, the LES results agreed rather well with each other. All models produced Cumulus clouds in reasonable correspondece with the measurements. Some results of the LES models can be obtained by anonymous ftp to email.meto.gov.uk (cd pub/apr/arm).

However, results of SCMs differed more strongly, and did not match the LES results as closely. Although most SCMs were able to simulate a Cumulus cloud, the SCMs typically suffered from the following deficiencies:

Finally, it was felt by most SCM modellers that SCMs are much more sensitive to changes in the forcing or initial profiles than LES models seem to be. The present intercomparison will be focussed on these problems. Some results from a version of the KNMI SCM - illustrating these problems - can be obtained here.

First Experiment

As a first step to tackle these problems we request you to run (or repeat) the GCSS case.
So what to do ?

  1. Read the summary of the original case description.  Mainly, look at the large scale forcing, the surface fluxes, and surface roughness length.
  2. Read the following specific further instructions, including the initial profiles  (already interpolated to the used grid) and the output to be generated.
If you have no experience at all with SCM modeling of Cu clouds you might want to run the GCSS BOMEX case too. In this case, a quasi-stationary state of Cu clouds above sea is simulated. If you would like us to look at these results, you might as well sent them in.

If you have not sent in a model description, please do so now.


Outlook

In a latter stage, we will define a set of sensitivity runs. We will also perform LES runs to answer specific questions resulting from the SCM results. These will probably include several closure assumptions, like cloud base massflux.


Results
 



Time table
 
Case release 1 September 2000
Sent in results 15 May 2001 ***
Workshop Lisbon 28-31 May 2001

*I** THIS IS A HARD DEADLINE: RESULTS SENT IN AFTER THIS DEADLINE WILL NOT BE PROCESSED !!!!
 


Geert Lenderink