Add missing omp private variables for data subglacial runoff

[jonbob]

Adds some variables were introduced in PR #6508 and not icluded in the appropriate omp private statement. These variables are only used for data subglacial runoff, so this does not impact current tests.

[BFB]

[github-actions]

PR Preview Action v1.4.8
🚀 Deployed preview to https://E3SM-Project.github.io/E3SM/pr-preview/pr-6656/
on branch gh-pages at 2024-10-03 19:41 UTC

[jonbob]

@vanroekel -- did you see any other loops where the new variables didn't get included?

[jonbob]

I ran the following test with and without these changes:

• PET_P640.TL319_IcoswISC30E3r5.GMPAS-JRA1p5-DIB-PISMF-DSGR.chrysalis_intel.mpaso-jra_1958
  I didn't see a FAIL in any testing, but it is still worth making these fixes

[rljacob]

You may want to add that test to e3sm_gpuomp

[vanroekel]

Approving by visual inspection. This is the only place I saw some omp variables missing. Thanks @jonbob

[irenavankova]

Approving based on visual inspection. Thank you for fixing this!

[irenavankova]

During the check I spotted another typo, on lines 3576-3577, there should be config_flux_attenuation_coefficient_subglacial_runoff instead of config_flux_attenuation_coefficient_runoff as follows:

transmissionCoeffTop = exp( max(zTop / config_flux_attenuation_coefficient_subglacial_runoff, -100.0_RKIND) )
transmissionCoeffBot = exp( max(zBot / config_flux_attenuation_coefficient_subglacial_runoff, -100.0_RKIND) )

@jonbob, could that be included in this pull request or do I need to open a new one?

[jonbob]

@irenavankova -- that should probably be a new PR. Do you want to make it?

[irenavankova]

Yes, I ll make it.

[jonbob]

@rljacob -- we could also change the test from the PR that brought this in. Right now we include

SMS_P480_Ld5.TL319_IcoswISC30E3r5.GMPAS-JRA1p5-DIB-PISMF-DSGR.mpaso-jra_1958

in the e3sm_ocnice_extra_coverage suite. We could change that to

PET_P480_Ld2.TL319_IcoswISC30E3r5.GMPAS-JRA1p5-DIB-PISMF-DSGR.mpaso-jra_1958

[rljacob]

Sure. And I realized this isn't omp-offload you're talking about so you don't need to move it to gpuomp

[jonbob]

Passes:

• PET_P480.TL319_IcoswISC30E3r5.GMPAS-JRA1p5-DIB-PISMF-DSGR.chrysalis_intel.mpaso-jra_1958 (new test)
• ERP_Ld3.ne30pg2_r05_IcoswISC30E3r5.WCYCL1850.chrysalis_intel.allactive-pioroot1

merged to next

[jonbob]

merged to master