GEWEX Cloud System Study - Working Group 4

Procedures to submit Case 4 results

***************************************************************
INITIAL DEADLINE TO SUBMIT YOUR RESULTS: APRIL 30

My intention is to have data available in May to work on the analysis. As you may remember, I submitted a paper to IUGG meeting in Japan in July, and I intend to present our results there (I will add names of those who submit data to the list of authors). I will also start writing a journal paper discussing this case. I understand that end of April is about a month away, but I hope that some of you will be able to run this case (some of you did it already), and submit the results. My plan is to submit the journal paper in fall, so even if you will not be able to run the simulation and prepare data files before summer, please consider doing this during summer.
***************************************************************

DATA SUBMISSION PROCEDURES

Please use case 3 documentation as a guide. It is located at http://www.met.utah.edu/skrueger/gcss/case3-docs.htm; CSS Working Group 4 Model Intercomparison Procedures for Case 3: Summer 1997 ARM SCM IOP, S. K. Krueger et al., Technical report, July 1999; hereinafter KCXY99) Similar procedures were used for EUROCS diurnal cycle case.

The results need to be submitted in a set of ASCII files. A single file is used for the evolution of each profile (group A below). A single file is used for all time series. File names will follow Case 3, with some changes. Each file should be named:

c4_MODEL_TYPE.pITEM.MODEL_ACRONYM - for profiles c4_MODEL_TYPE.tGROUP.MODEL_ACRONYM - for time series

where c4 stands for case 4 and, as in case 3, MODEL_TYPE should be 1 for 1D SCM, 2 for 2D CRM, and 3 for 3D CRM. ITEM in profile files identifies the item from the list below (as in KCXY99), and GROUP in time series should be 1 as only a single group is considered (as it was in KCXY99).

For instance, results for my model (EULAG), run in 2D configuration, will consist of set of profile files:

c4_2.p0.EULAG_2D
c4_2.p1.EULAG_2D
c4_2.p2.EULAG_2D
...
c4_2.p30.EULAG_2D

and a single file with time series:

c4_2.t1.EULAG_2D.

Results are supposed to be submitted at t=10, 20, ... , 720 min, e.g., for 36 time levels.

After creating the files, please tar all of them into a single file named c4_MODEL_TYPE.MODEL_ACRONYM.tar (e.g., c4_2.EULAG_2D.tar in my case), gzip it to create a smaller file (which will have name c4_2.EULAG_2D.tar.gz in my case), and e-mail it to me as an attachment. I will unzip/untar the file and use the set of your data files in the analysis.

If your model does not provide information about a specific quantity, please omit this file from the profile set, and use a negative value with" 8"s (e.g., -888.88) in the time series file.

As for the actual data format used while writing data to the files, you may follow formats prescribed in KCXY99 (the additional data have formats prescribed below). However, I will try to read the data using unformatted reads and as long as there is a space between subsequent entries, I should be able to read the data.

DATA TO BE SUBMITTED.

A. time evolution of mean (i.e., domain-averaged for CRMs) profiles

*********************************************************************
Results are to be submitted at t=10, 20, ... , 720 min, e.g., for 36 time levels. *********************************************************************

Please use the same list and file format as in KCXY99, except for the item 0 (i.e., height, not pressure). The list is included below. If you model uses pressure as vertical coordinate, convert pressure levels to height levels using initial profiles (i.e., input sounding). Another difference from KCXY99 is that in the definition of cloud fraction (item 9), the threshold cloud condensate should be taken as amin1(1.e-5,0.01*qsat) rather than 0.01*qsat as in KCXY99. The reason is that 0.01*qsat gives a relatively high threshold value (e.g., more than 0.1 g/kg) in the lower troposphere.

0. Height (m) FORMAT CHANGE - use F8.1

1. Temperature (K)

2. Water vapor mixing ratio (g/kg)

3. Relative humidity (unitless)

4. cloud water mixing ratio (g/kg)

5. cloud ice mixing ratio (g/kg)

6. Rain mixing ratio (g/kg)

7. Snow (slow-falling ice) mixing ratio (g/kg)

8. Graupel (fast-falling ice) mixing ratio (g/kg)

9. Cloud fraction (unitless) NOTE: threshold cloud condensate mixing ratio should be taken as amin1(1.e-5,0.01*qsat)

23. Updraft cloud mass flux (kg/m**2/s)

24. Downdraft cloud mass flux (kg/m**2/s)

30. Hydrometeor fraction (unitless).

B. time evolution of selected quantities

*********************************************************************
Results are to be submitted at t=10, 20, ... , 720 min, e.g., for 36 time levels.
*********************************************************************

The list below is an extended version from KCXY99.

1. Time (minutes) FORMAT CHANGE - use F7.1

2. Total cloud amount (as 3 on p. 12 in KCXY99; unitless)

3. Cold cloud amount (as 4 on p. 12 in KCXY99; unitless)

4. Height of the center of mass of cloud condensate (see below; km; F8.3)

5. Maximum cloud top height (cloud top height defined as on p. 12 and 13 of KCXY99; km; F8.3)

6. Cloud liquid water path (as 6 on p. 12 in KCXY99; kg/m**2)

7. Cloud ice path (as 7 on p. 12 in KCXY99; kg/m**2)

8. Surface rainfall rate (as 8 on p. 12 in KCXY99; mm/day)

9. Surface convective rainfall rate (as 9 on p. 12 in KCXY99; mm/day)

10. Surface stratiform rainfall rate (as 10 on p. 12 in KCXY99; mm/day)

11. Precipitable water (as 5 on p. 12 in KCXY99; kg/m**2)

12. Tropospheric density weighted temperature (K; F8.2)

additional definitions:

height of the center of mass of cloud condensate: sum(qz) / sum(q) where q is the total condensate (sum of cloud water, cloud ice, rain, snow, graupel, etc), z is height, and sum stands for summation over entire computational domain (1D for SCMs, 2D or 3D for CRMs).

tropospheric density weighted temperature: use limits of z=0 and z=12 km when deriving the density-weighted temperature. Note that temperature, not potential temperature, should be used here.