Project

General

Profile

[CCLM-Converction Permitting Scale] Korea Domain (3km) simulation configuration

Added by Donghyun Lee almost 4 years ago

Dear CLM-community members,

Unlikely numerous CPCS (Converction-Permitting Climate Simulation) study over Europe,
CPCS study over Asia, and my country Korea is very limited.

We are planning to simulate CCLM-CPCS, or CCLM-CPM (convection permitting model), in order to better understand future precipitation change.
By considering EASM (East Asian Summer Monsoon) which dominantly affects climate over Korea, we experimentally set simulation domain.
However, we are wondering about not only CPCS setting of DYNCTL and PHYCTL in CCLM, and but also CPCS domain.
We are expecting professional advices for CCLM CPCS configuration.

Here, we attach PPT file including our experiment details and our questions, and YUSPECIF of 25km setting.

Thank you for reading our challenging questions.
Best regards,
Donghyun Lee

[CPM-CCLM] KOREA v2.pptx [CPM-CCLM] KOREA v2.pptx 138 KB Korea CPCS explanations and questions
YUSPECIF [25km] YUSPECIF [25km] 47.9 KB YUSPECIF used in East Asia 25km simulation

Replies (6)

RE: [CCLM-Converction Permitting Scale] Korea Domain (3km) simulation configuration - Added by Donghyun Lee almost 4 years ago

I am afraid of pptx file, so I convert pptx to pdf. You can check this,
Sorry for any inconvenience.

Donghyun Lee

RE: [CCLM-Converction Permitting Scale] Korea Domain (3km) simulation configuration - Added by Andreas Dobler almost 4 years ago

Interesting and challenging task!

My first “professional advice” would be to have a look at a recent paper by Brisson et al.
“Modelling strategies for performing convection-permitting climate simulations”
https://www.schweizerbart.de/papers/metz/detail/25/84891/Modelling_strategies_for_performing_convection_permitting_climate_simulations
DOI: 10.1127/metz/2015/0598

One example: They found that the best downscaling strategy for their case is to use two nesting steps (25 km and 2.8 km) and an intermediate step (7 km) was found to be redundant. So maybe you can skip the 12.5 km nest.
Also, including graupel parametrization in the CPCS had a positive effect (itype_gscp = 4, I think). There is also information on the domain size, so the paper could give you some hints.

I recently did some 0.022° runs nested into 0.22° over Svalbard. The scripts defining the set-ups are attached.
The results look good so far, but I didn’t have time to do a deep analysis yet. Surely, the setup is not the final one.
For instance, I didn’t include the graupel parametrization in the CPCS. But generally, it looks close to your chooses.
dt=30 seems a little high to me and could yield some CFL violations. I started with dt=20 and had to go down to dt=10, but this could be due to the high latitudes in my case.

Best regards
Andreas

cosmo_NORWAY22_job.sh cosmo_NORWAY22_job.sh 5.68 KB Runscript (generating INPUT_* files) for 25km run
cosmo_NORWAY022_job.sh cosmo_NORWAY022_job.sh 5.48 KB Runscript (generating INPUT_* files) for 2.5km run

RE: [CCLM-Converction Permitting Scale] Korea Domain (3km) simulation configuration - Added by Donghyun Lee almost 4 years ago

Thank you very much for your thoughtful comments.

Now I read the paper you mentioned (Brisson et al., 2015), and it is really helpful for me to understand better configuration of both CPCS and CCLM.

(1) Although Brisson’s study shows redundant of intermediate step, we are willing to consider triple-nesting not only for CPCS, but for analyze CCLM 12.5km with other type of RCMs’ 12.5km results over Korea region.

(2) Since I do not know about itype_gscp options well, graupel parameterization is the most impressive one for me. However, contrary to Brisson’s dry-bias over Belgian region in CCLM 25km, we find little wet-bias over Korea region in our CCLM 25km run (BGF: ERA-INTERIM), so I want to do sensitivity test (short-period, less than 1 year) for itype_gscp (3 or 4) and their impacts over Korea region.

(3) As Brisson recommends, we will apply lateral boundary fields wider than our initial setting.

(4) I appreciate your kindness providing simulation information. We will reflect your set-ups carefully for our simulation. And also, as you pointed, we will carefully set the 'dt’ to escape CFL violations.

Best regards,
Donghyun

RE: [CCLM-Converction Permitting Scale] Korea Domain (3km) simulation configuration - Added by Erwan Brisson almost 4 years ago

Donghyun and Andreas, I am really happy to see that people are using the results of Brisson et al. (2015) for preparing their model setups. If you have any further question on this research, do not hesitate to contact me.

For your information in the CRCS working group, we have two projects going on, which aim at providing appropriate configurations for the COSMO-CLM at convection permitting scales. Hopefully, we will be able to provide the community with this information in the coming year. For one of this project, we set up a list of parameters which may be relevant to test.

Concerning the graupel parameterization. I personally believe that it is a mistake to turn the graupel parameterization off in convectively active areas with moderate or high temperature. The parameterization of graupel should not be seen as a switch for adjusting the climatology of precipitation. As explained in Brisson et al. (2015) and many other studies, graupel plays a major role in the production of intense precipitation through riming. It also has an impact on the vertical distribution of latent heat release and other processes which are mainly occurring in deep convection. In my opinion, the graupel parameterization is critical for representing realistically deep convection for the present-day and other climates.
Andreas, correct me if I am wrong, but at the latitude of Svalbard, the occurrence of convection is probably low (the ratio of convective precipitation/stratiform precipitation is <<1), and the atmosphere is cold enough which limits the relevance of riming processes (which require both the presence of solid and liquid water).
Of course, it does not hurt to test the sensitivity of the graupel parameterization. However, I would not turn it off to compensate an overestimation of precipitation in the parent simulation.

In the CRCS working group, there exist a namelist market, where CRCS group members are invited to share their namelists. http://redc.clm-community.eu/projects/cclm/wiki/CRCS_namelist_market . I guess this is a useful page for you.

Kind regards,
Erwan

RE: [CCLM-Converction Permitting Scale] Korea Domain (3km) simulation configuration - Added by Erwan Brisson almost 4 years ago

Sorry, part of the previous message was removed when posting. Here is another try.

Donghyun and Andreas, I am really happy to see that people are using the results of Brisson et al. (2015) for preparing their model setups. If you have any further question on this research, do not hesitate to contact me.

For your information in the CRCS working group, we have two projects going on, which aim at providing appropriate configurations for the COSMO-CLM at convection permitting scales. Hopefully, we will be able to provide the community with this information in the coming year. For one of this project, we set up a list of parameters which may be relevant to test.

Concerning the graupel parameterization. I personally believe that it is a mistake to turn the graupel parameterization off in convectively active areas with moderate or high temperature. The parameterization of graupel should not be seen as a switch for adjusting the climatology of precipitation. As explained in Brisson et al. (2015) and many other studies, graupel plays a major role in the production of intense precipitation through riming. It also has an impact on the vertical distribution of latent heat release and other processes which are mainly occurring in deep convection. In my opinion, the graupel parameterization is critical for representing realistically deep convection for the present-day and other climates.
Andreas, correct me if I am wrong, but at the latitude of Svalbard, the occurrence of convection is probably low (the ratio of convective precipitation/stratiform precipitation is much lower than 1), and the atmosphere is cold enough which limits the relevance of riming processes (which require both the presence of solid and liquid water).
Of course, it does not hurt to test the sensitivity of the graupel parameterization. However, I would not turn it off to compensate an overestimation of precipitation in the parent simulation.

In the CRCS working group, there exist a namelist market, where CRCS group members are invited to share their namelists. http://redc.clm-community.eu/projects/cclm/wiki/CRCS_namelist_market . I guess this is a useful page for you.

Kind regards,
Erwan

RE: [CCLM-Converction Permitting Scale] Korea Domain (3km) simulation configuration - Added by Andreas Dobler almost 4 years ago

Hi Erwan,

thanks for the input.

Convection is more common than I did expect. Especially in the Barents sea in winter when cold air outbreaks hit the relatively warm sea surface (see attached image).
Svalbard is still out of the region where convection occurs regularly, but climate projections show a different picture
I guess riming and deep convection is of less importance than in other areas (since the liquid water in the air during winter is probably negligible).
I may have time to run a short sensitivity test on the graupel parameterization for the area. We’ll see.

Best regards
Andy

    (1-6/6)