Skip to content

Conversation

@xylar
Copy link
Collaborator

@xylar xylar commented Dec 10, 2017

The name of the MOC mask file is constructed automatically from the
mesh name.

This approach is needed in anticipation of having other region masks (notably masks for individual ice shelves, see #237)

Example config files have been updated accordingly.

The name of the mask file is constructed automatically from the
mesh name.
@xylar
Copy link
Collaborator Author

xylar commented Dec 10, 2017

This needs to be merged before #237 and #246

@xylar
Copy link
Collaborator Author

xylar commented Dec 10, 2017

I need to re-test all of the example config files to make sure they still work. Here's the checklist:

  • anvil/config.20170926.FCT2.A_WCYCL1850S.ne30_oECv3.anvil
  • configs/edison/config.20170807.beta1.G_oQU240.edison
  • configs/edison/config.20170915.beta2.A_WCYCL1850S.ne30_oECv3_ICG.edison
  • configs/edison/config.20171102.beta3rc02_1850.ne30_oECv3_ICG.edison
  • configs/edison/config.B_low_res_ice_shelves_1696_JWolfe_layout_Edison
  • configs/lanl/config.MatchBoth_orig
  • configs/olcf/config.20170313.beta1.A_WCYCL1850S.ne30_oECv3_ICG.edison
  • configs/olcf/config.20170915.beta2.A_WCYCL1850S.ne30_oECv3_ICG.edison
  • configs/olcf/config.GMPAS-IAF_oRRS18to6v3.titan
  • configs/theta/config.20171031.tenYearTest.GMPAS-IAF.T62_oEC60to30v3wLI.60layer.theta

@xylar xylar force-pushed the switch_to_mask_directory branch 3 times, most recently from 8172aad to 3d79120 Compare December 10, 2017 20:50
@xylar
Copy link
Collaborator Author

xylar commented Dec 10, 2017

@milenaveneziani, could you check the 3 configs on OLCF by making sure the MOC gets computed correctly when you run the analysis? I need to get myself set up on titan and rhea but I'm not there yet. I'd like to make sure the region mask files there have the right naming convention. I had to fix the names on LANL IC and theta.

@milenaveneziani
Copy link
Collaborator

@xylar: I am testing with the config files you listed on titan. First test (18to6 run) failed because I don't have permission to read the model data (I have asked @vanroekel to change permissions when he has a chance), but I have changed the region mask filename in a way that is compatible with how this PR expects it. Second test (beta1 run) passed. Third test (beta2 run) is going, but I found a tiny typo in the corresponding config file: I will point it to you in the review section.

[timeSeriesSeaIceAreaVol]
## options related to plotting time series of sea ice area and volume

# plot on polar plot
Copy link
Collaborator

@milenaveneziani milenaveneziani Dec 11, 2017

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could you add these lines in this file, right before line 46:

# provide an absolute path to put HTML in an alternative location (e.g. a web
# portal)
# htmlSubdirectory = /global/project/projectdirs/acme/www/USERNAME/RUNNAME
htmlSubdirectory = html

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

done

# ./run_mpas_analysis --list
# to list all task names and their tags
# an equivalent syntax can be used on the command line to override this
# option:
Copy link
Collaborator

@milenaveneziani milenaveneziani Dec 11, 2017

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could you add these, right before line 35:

# provide an absolute path to put HTML in an alternative location (e.g. a web
# portal)
# htmlSubdirectory = /ccs/proj/cli115/www/USERNAME/RUNNAME
htmlSubdirectory = html

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

done

## circulation (MOC)

# Region names for basin MOC calculation.
# Supported options are Atlantic and IndoPacific
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could you change mpasMeshName to oEC60to30v3?

And then add, before line 54:

# provide an absolute path to put HTML in an alternative location (e.g. a web
# portal)
# htmlSubdirectory = /ccs/proj/cli115/www/USERNAME/RUNNAME
htmlSubdirectory = html

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

done

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think you forgot to adjust the mpasMeshName.

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, I missed that. I'll fix it now.

# ./run_mpas_analysis --list
# to list all task names and their tags
# an equivalent syntax can be used on the command line to override this
# option:
Copy link
Collaborator

@milenaveneziani milenaveneziani Dec 11, 2017

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This one is missing a few things (mind you, I haven't had a chance to test this yet, because of the nco problems with large files, which I mentioned in the email):

  • as noted by @vanroekel, the input baseDirectory should be changed to /lustre/atlas1/cli115/proj-shared/vanroek/run
  • mpasMeshName
  • there is the piece with Directory with mapping files, but not this newer part:
# Directory for region mask files
regionMaskDirectory = 
  • these lines before line 37:
# provide an absolute path to put HTML in an alternative location (e.g. a web
# portal)
# htmlSubdirectory = /ccs/proj/cli115/www/USERNAME/RUNNAME
htmlSubdirectory = html

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I do see the regionMaskDirectory part in my version at the bottom. I have changed the rest as you suggested.

regionMaskDirectory = /projects/OceanClimate/mpas-analysis_data/mpas_analysis/region_masks

[climatologyMapSoseTemperature]
# Times for comparison times (Jan, Feb, Mar, Apr, May, Jun, Jul, Aug, Sep, Oct,
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

do we have a htmlSubdirectory on theta?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No, as far as I'm aware, there's no web portal on theta.

@milenaveneziani
Copy link
Collaborator

Test with beta2 run also passed. I am down to testing the high-res case.

@vanroekel
Copy link
Collaborator

@milenaveneziani, FYI that the high res data on titan is now at

/lustre/atlas1/cli115/proj-shared/vanroek/run

@xylar xylar force-pushed the switch_to_mask_directory branch from 3d79120 to 01d9f83 Compare December 12, 2017 08:53
@milenaveneziani
Copy link
Collaborator

@xylar: after changing the mpasMeshName in one of the config files, you can merge this.

Although the olcf config file for the high-res cannot be tested for the MOC right now, due to the need for nco 4.7.0, I have changed the MOC transect filename for the RRS18to6 to reflect changes here, so we shouldn't have problems when the time comes.

Remove --purge option from job scripts
@xylar xylar force-pushed the switch_to_mask_directory branch from 01d9f83 to b7adb36 Compare December 12, 2017 16:14
@xylar
Copy link
Collaborator Author

xylar commented Dec 12, 2017

@milenaveneziani, I really appreciate your efforts on the RRS18to6. We'll get it working but I agree that it shouldn't hold up this PR as long as we're pretty sure the config file will be correct once we get it working.

@xylar xylar merged commit ced7dda into MPAS-Dev:develop Dec 12, 2017
@xylar xylar deleted the switch_to_mask_directory branch December 12, 2017 16:19
@xylar xylar mentioned this pull request Jan 26, 2018
xylar added a commit that referenced this pull request Jan 26, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants