Opened 4 months ago

Closed 3 months ago

#3231 closed help (fixed)

LGM input/output are something wrong

Reported by: yb19052 Owned by: um_support
Component: UM Model Keywords:
Cc: Platform: NEXCS
UM Version: 10.7


I have a suite (u-bp881) for the Last Glacial Maximum (LGM). It runs, but the variables “roughness length after timestep” in the start dump (bp881.astart) and in the restart file after one year (bp881a.da18510101_00) are based on the pre-industrial (PI) land-ocean mask. It seems that the other variables look okay.

Here, I have questions. Should I set up “roughness length after timestep (stash_rq: 26)” in the Configure ancils? Or, it is automatically created but is my input something wrong?

There are the files at the following locations,

I appreciate your kind cooperation.


Change History (7)

comment:1 Changed 4 months ago by grenville


I am not sure what you are asking - the roughness looks OK in relation to the land-sea mask. Other start files (for example /home/d03/gmslis/cylc-run/u-bg743/share/data/bg743.astart) have similar-looking fields.


comment:2 Changed 4 months ago by yb19052

Hi Grenville,

Thank you for your response.

I mean that "Roughness length on tiles" is based on the LGM land-ocean mask (which is different from the PI mask), but "Roughness length after timestep" is based on the PI land-ocean mask. Because I did not use the PI mask, the latter variable is something wrong I think. Please check the "land mask (No halo) (LAND=TRUE)" in bp881.astart.

I checked bg743.astart, and it is a suit based on the PI land-ocean mask.

Do you mean that we use the same "Roughness length after timestep" in the different past climates (e.g., the Last Glacial Maximum and Eocene)?

I appreciate your kind cooperation.


comment:3 Changed 4 months ago by grenville

Hi Kenji

We are still working on this.


comment:4 Changed 4 months ago by simon

Hi Kenji,

You need to go into the Configure ancils an set up a new section:
stash_req: 26
domain: whole grid
update_ancil: false
source: set to specified constant value
user_prog_rconst: 0.5

The roughness in the start dump will be filled with 0.5, but the model will apply the correct mask when it runs. What I think happened is that the reconfiguration somehow wasn't interpolating the roughness length to the correct mask, and then the model wasn't calculating the roughness length on those points with missing data.


comment:5 Changed 4 months ago by yb19052

Hi Simon,

I added the new section about "surface roughness" and am running the suite again.
I will check the output after one year run.


comment:6 Changed 3 months ago by yb19052

Hi Simon,
My suite works well. Thank you for your help.

comment:7 Changed 3 months ago by simon

  • Resolution set to fixed
  • Status changed from new to closed

OK, I'll close the ticket.

Note: See TracTickets for help on using tickets.