Main menu


offline CICE model

11 posts / 0 new
Last post
offline CICE model

I  ran  the offline CICE model .

I did a test : change the Great lakes to ocean and modify related files. 

But after I  ran  the CICE offline model , the ice thickness is all euqal to 0 in the Great lakes.

I checked the sea surface temperature I modified in POP file and there exsits lower temperature than 0 in some months.

But why the results have no ice ?

Waiting for somebody helping me ! Thanks !


Are you pointing CICE to the same modified mask as you are using for POP? You need to make sure that grid_file points to the modified horiz_grid file from POP and kmt_file points to the modified topography file.



The grid_file I used is horiz_grid_20010402.ieeer8. This file includes hte htn , angle and so on.

Variables in this files have no missing value. I didn't change this file. The model also can run.

Should I also change this file?


Not the grid_file, what are you doing for the kmt_file? This uses the ocean column depths as the land/ocean mask. If this is not the same as POP you will have problems.



I changed the kmt file and added values in my added area.

The changes I ran  CICE offfline model are : 

POP file:  inputdata/ocn/docn7/SOM/ ---- change sss sst mask and set other variables to 0  

grid file and kmt file : topography_20090204.ieeei4  horiz_grid_20010402.ieeer8  region_mask_20090205.ieeei4 --- add values in my own area incluing kmt andregion_mask 

domain file : inputdata/share/domains/  ----------- change frac and mask 

The model can run successfully , just without ice values in my added area in winter .

And another thing is that when I output the sst in output flies , there is no value in my added area. BUT I did change the pop file. 


Thank you for your patience very much! Look forward to your guidance~


I really don't understand your issue. The CICE still thinks those are land points. So something is not consistent between POP and CICE.




Sorry for my not clear expression!

I'm embarrassed to go on asking you.

I want to change the great lakes into ocean and let the model deal with the great lakes same as ocean, so can this be achieved ?

Now I am trying it .

Then I found the result of ice in winter are all 0 in the great lakes area.

 So yesterday I added the value in great lakes area in ice initial file :

But the model crashed with "ERROR: ice: Vertical thermo error"

I am really really sorry for my confused questions.


Sounds like you were using the same mask, but then initializing the sea ice in the great lakes to 0. One has to be very careful with modifying the initial files. The internal energy and everything need to be consistent. I think the best approach is to still initialize the sea ice to 0 in the great lakes and let it grow on it's own. Are you getting even a tiny bit of ice after a few days?



Thank you very much ~

I have adopted your way to initialize the sea ice to 0 .

But the results are very strange.

The sst output have no value(missing) in the great lakes.

And the air temperature output is also strange, in the great lakes with Kelvin-format( always -273.15) and other area with normal centigrade-format.

So I think in the model it have not identify the great lakes as ocean.

Could you give some hints?


In the D compset (active CICE only), the DOCN runs in SOM mode. That means you have a prognostic SST. I assume you are using some sort of qflux and mixed-layer depth forcing file. What are these values in Great Lakes? This file also has an initial SST value to start the SOM off with. Now I thought you were also running a G compset (active CICE and POP). In this case you need to make sure you have initial values for temperature and salinity in the Great Lakes. Try writing out some coupler history files using HIST_OPTION and HIST_N. Look at the fields coming from the ocean and make sure these are reasonable for the Great Lakes. I'm afraid this is all I can suggest.



Sorry for disturbing you again and again!

I want to know if I want to give value of q-flux and mixed-layer depth in the Great lakes I added to ocean , how could I give?

is there nc-format of kmt file with resolution of gx1v6 ? I tried with gx1v5-res nc-format kmt file, but it can't work.

Can I give them equal to one point in the other ocean area ? 

Log in or register to post comments

Who's new

  • huzhy2010@...
  • praveenkumar.jrf@...
  • garimam@...
  • tarique
  • m.valmartin@...