Adds penetrating shortwave radiation by vanroekel · Pull Request #570 · E3SM-Project/Omega

vanroekel · 2026-09-22T16:52:30Z

This allows shortwave radiation to enter the subsurface ocean instead of simply being deposited in the top layer. The penetration with depth follows the implementation of Manizza et al (2008), their equation 4. It splits the incident solar into three components, near IR, 'blue' band, and 'red' band. The solar flux at the bottom of the ocean is set to zero to ensure there are no energy leaks, i.e. the column integrated heating is identical for penetrating shortwave and when deposited in the top layer.

A unit test has been added with a very shallow column and it tests equivalence of heating in the penetrating case and top layer only case.

Checklist

Copilot

Copilot review overview

🟡 Changes recommended

Unresolved moderate issues remain in halo exchange and regression-test coverage.

Get a fresh assessment by requesting another Copilot review.

Review effort: Lite
Findings: 2 · 1

Open (3)

Assert Foxx_swnet in coupling tests · New
Strengthen assertions for shortwave attenuation profile · New
Remove redundant shadowed metadata declaration · New

What changed in this PR

Adds configurable, depth-dependent three-band penetrating shortwave heating to the ocean.

Changes:

Adds shortwave tendency and extinction-coefficient auxiliary fields.
Imports coupled Foxx_swnet forcing.
Updates configuration, tests, and documentation.

File	Reviewed change / findings
`components/omega/test/ocn/TendencyTermsTest.cpp`	Adds conservation coverage. Moderate (3 votes): strengthen assertions to validate per-layer attenuation.
`components/omega/test/ocn/SfcCouplingTest.cpp`	Updates coupling fixtures. Moderate (2 votes): verify `Foxx_swnet` in both coupling paths.
`components/omega/src/ocn/TendencyTerms.h`	Defines the shortwave penetration kernel.
`components/omega/src/ocn/TendencyTerms.cpp`	Constructs the new kernel.
`components/omega/src/ocn/Tendencies.h`	Registers the new tendency.
`components/omega/src/ocn/Tendencies.cpp`	Applies the tendency. Moderate (1 vote): exchange extinction-field halos or restrict processing to owned cells.
`components/omega/src/ocn/SfcCoupling.h`	Adds coupled shortwave storage.
`components/omega/src/ocn/SfcCoupling.cpp`	Imports coupled shortwave flux. Moderate (1 vote): add `Foxx_swnet` coupling assertions.
`components/omega/src/ocn/OceanInit.cpp`	Reads extinction inputs. Nit (2 votes): remove redundant shadowed declarations at lines 153 and 242.
`components/omega/src/ocn/Forcing.cpp`	Enables and exchanges shortwave forcing.
`components/omega/src/ocn/auxiliaryVars/ShortwavePenAuxVars.h`	Declares extinction auxiliary fields.
`components/omega/src/ocn/auxiliaryVars/ShortwavePenAuxVars.cpp`	Registers extinction auxiliary fields.
`components/omega/src/ocn/AuxiliaryState.h`	Adds shortwave auxiliary state. Moderate (1 vote): exchange both extinction arrays in `exchangeHalo()`.
`components/omega/src/ocn/AuxiliaryState.cpp`	Initializes auxiliary fields. Moderate (1 vote): exchange extinction halos; Nit (1 vote): reorder the `ShortwavePenAux` initializer.
`components/omega/src/drivers/coupled/omega_cpl_indices.F90`	Adds the coupled shortwave import index.
`components/omega/doc/userGuide/TendencyTerms.md`	Documents the tendency.
`components/omega/doc/userGuide/Forcing.md`	Documents forcing configuration and behavior.
`components/omega/doc/userGuide/AuxiliaryVariables.md`	Documents extinction fields.
`components/omega/doc/devGuide/Forcing.md`	Documents implementation details.
`components/omega/doc/devGuide/AuxiliaryVariables.md`	Documents auxiliary-field handling.
`components/omega/configs/Default.yml`	Adds feature defaults and input configuration.

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

vanroekel · 2026-09-24T05:06:17Z

Here is a picture showing heating over 3 hours from different configurations. Jerlov use 2 exponentials and Manizza use 3. The default for omega will be somewhat like Manizza type 1 but with spatial variability

katsmith133

Just a quick first round of reviewing. Nice job on this @vanroekel!

I haven't tested it, but I can do that on Frontier and post the results.

katsmith133 · 2026-09-30T16:00:29Z

      BottomDragCoeff: 1.0e-3
    SfcThicknessForcingTendencyEnable: false
    SfcTracerForcingTendencyEnable: false
+    PenetratingShortwaveTendencyEnable: false


This is bigger than just this PR, but I feel like we can organize this section a bit better. Perhaps more like the BottomDragTendency section above, where the parameters that go with that tendency are a subsection of that tendency flag. That way its clear what parameters go with what tendencies and it keeps it a bit cleaner? To start, I would suggest making NearIrFraction, NearIrCoeff, BlueFraction, and RedFraction sub parameters under a header of PenetratingShortwaveTendency along with an Enable parameter. And then one of us can do a clean up PR that tidies up the other ones too?

This is fixed in the new version and reflected in the docs as well

katsmith133 · 2026-09-30T16:06:32Z

+Omega:
+  Tendencies:
+    SfcTracerForcingTendencyEnable: true
+    PenetratingShortwaveTendencyEnable: true


Is there a warning or something for if the user sets PenetratingShortwaveTendencyEnable: true, but SfcTracerForcingTendencyEnable: false?

Also, need to add BlueFraction: 0.19 and RedFraction: 0.23 here too?

Added the Blue/Red fractions and added an error if PenetratingShortwaveTendencyEnable: true and SfcTracerForcingTendencyEnable: false. See L144 of OceanInit.cpp

katsmith133 · 2026-09-30T16:10:44Z

+
+The near-infrared fraction and extinction coefficient can be changed under
+`Omega.Tendencies` with `NearIrFraction` and `NearIrCoeff`. The red and
+blue/green band fractions remain fixed at 0.23 and 0.19, respectively.


This can lead to totals less than or greater than 1, yes? Is that a problem we anticipate and/or should mitigate somehow?

Also, where did these fractions/values come from? I don't seem them in the Manizza 2008 paper.

I see the 0.58, but not the 0.23 and 0.19.

The values were a mistake they should have both been 0.21 as per Manizza et al, that is fixed. There is also now a check that the NearIRFraction and BlueFraction and RedFraction sum to 1, see L365 of Tendencies.cpp

katsmith133 · 2026-09-30T17:19:35Z

   VorticityAux.registerFields(GroupName, AuxMeshName);
   VelocityDel2Aux.registerFields(GroupName, AuxMeshName);
   SurfTracerRestAux.registerFields(GroupName, AuxMeshName);
+   // Only the Default state's fields belong to the shared IO group created


I am not sure I understand why ShortwavePenAux is treated differently here. Can you elaborate?

This is removed, not sure why it was different

katsmith133 · 2026-09-30T17:21:04Z

   }

+   OMEGA_SCOPE(LocPenetratingShortwave, PenetratingShortwave);
+   if (LocPenetratingShortwave.Enabled) {


Does not SfcTracerForcingTendencyEnable also have to be enabled for this to go through? Or is that checked elsewhere

This is fixed.

katsmith133 · 2026-09-30T17:22:47Z


-   TracerForcingFieldsEnabled =
-       SfcThicknessForcingEnabled || SfcTracerForcingEnabled;
+   bool PenetratingShortwaveEnabled = false;


I guess perhaps this is where we can double up and check if SfcTracerForcingTendencyEnable is also on and set PenetratingShortwaveEnabled based on that too?

Or is it L186 below that is what is really needed? Really we need TracerForcingFieldsEnabled enabled, right? Or do we also need SfcTracerForcingTendencyEnable enabled as well?

Actually looking here I think we need to strip all of the penetrating shortwave. Copilot loaded all this to check if tracer forcing is on at L189, but we don't want tracer forcing enabled if only Penetrating Shortwave is on. I'm going to remove it. I had co pilot add an error out in a different place if the two aren't on.

katsmith133 · 2026-09-30T17:31:33Z

+      const Real Kb          = ExtinctionCoeffBlue(ICell);
+      const Real ZSurface    = GeomZInterface(ICell, KTop);
+
+      Real FluxAtTop = SurfaceFlux;


Can we amend these to be FluxAtLayerTop and FluxAtLayerBottom? I think that helps make this read clearer.

Done, fixed it here and in TendenciesTest as well

katsmith133 · 2026-09-30T17:36:25Z

+   const auto Mesh   = HorzMesh::getDefault();
+   const auto VCoord = VertCoord::getDefault();
+
+   // Arbitrary test values for the incident surface shortwave flux and the


Does this test the no penetrative flux beyond the bottom most layer part? I feel like with 60 x 10m layers, the test probably doesn't have any extra leftover flux by the time it gets to the bottom.

This was a mistake it should have been much shorter, column should be 16m then.

vanroekel · 2026-10-04T05:05:54Z

Polaris `omega_pr` suite

PR build: /lcrc/group/e3sm/ac.vanroekel/solar_test_build
PR workdir: /lcrc/group/e3sm/ac.vanroekel/pr_suite_solar
Machine: chrysalis
Partition: debug
Compiler: intel
Build type: Release
Log: /lcrc/group/e3sm/ac.vanroekel/pr_suite_solar/polaris_omega_pr.o1301380
Result: All tests passed

vanroekel · 2026-10-05T03:07:04Z

Polaris `omega_pr` suite

PR build: /pscratch/sd/l/lvroekel/solar_pr_suite_cpu_build
PR workdir: /pscratch/sd/l/lvroekel/solar_pr_suite_cpu
Machine: pm-cpu
Compiler: gnu
Build type: Release
Log: /pscratch/sd/l/lvroekel/solar_pr_suite_cpu/polaris_omega_pr.o59302032
Result: All tests passed

vanroekel · 2026-10-05T03:07:30Z

Polaris `omega_pr` suite

PR build: /pscratch/sd/l/lvroekel/solar_pr_suite_gpu_build
PR workdir: /pscratch/sd/l/lvroekel/solar_pr_suite_gpu
Machine: pm-gpu
Compiler: gnugpu
Build type: Release
Log: /pscratch/sd/l/lvroekel/solar_pr_suite_gpu/polaris_omega_pr.o59302521
Result: All tests passed

vanroekel · 2026-10-05T04:57:59Z

Polaris `omega_pr` suite

PR build: /lustre/orion/cli115/scratch/vanroek/solar_omega_build_craygnu-mphipcc
PR workdir: /lustre/orion/cli115/scratch/vanroek/solar_frontier_craygnu-mphipcc_suite
Machine: frontier
Partition: batch
Compiler: craygnu-mphipcc
Build type: Release
Log: /lustre/orion/cli115/scratch/vanroek/solar_frontier_craygnu-mphipcc_suite/polaris_omega_pr.o5602787
Result: All tests passed

alicebarthel · 2026-10-07T18:55:38Z

                                             OnCell);
   }

+   if (TracerForcingFieldsEnabled) {


Do we need a Halo exchange here? I'd think running on CellOwned would be enough since this is cell to cell?

I think we may? The tracer tends iterate over nCellsAll so we'd have to make sure missing shortwave isn't needed. Can we punt on this until a later PR? Or @brian-oneill do you have thoughts on this one?

actually thinking further on this I think it can be removed. I'm doing the surface tracer forcing over NCellsOwned and removing this exchange

Yes, I agree. It doesn't look like this exchange was necessary

alicebarthel · 2026-10-07T18:56:41Z

       Err,
       "Tendencies: SfcTracerForcingTendencyEnable not found in TendConfig");

+   Config PenSWConfig("PenetratingShortwaveTendency");


Very minor formatting but using SW capitalized as a shorthand for Shortwave seems strange.

Suggested change

Config PenSWConfig("PenetratingShortwaveTendency");

Config PenSwConfig("PenetratingShortwaveTendency");

alicebarthel · 2026-10-07T18:58:24Z

+   Real NearIrFraction = 0.58_Real;
+   Real NearIrCoeff    = 2.86_Real;
+   Real RedFraction    = 0.21_Real;
+   Real BlueFraction   = 0.21_Real;


I'd prefer if we don't carry default values within the code. Default values should be in the Default.yml file. I'd rather expose that the value was not passed properly than run with a hard-coded value.

I don't agree with the proposed change. We do this approach in other places of the code (e.g. Vmix, the submeso PR) it's also what MPAS did too. I don't understand the need for another check within classes.

I am not saying to add a check. I am saying that the fail should be triggered by the check when initializing/creating the class. So a default value here is not necessary except for ctests, which in this case already set the values.

I still don't understand. What check do you mean? Also how is the approach here different than what I see in other classes?

I'm confused because I thought the yaml checks we have are if the config is in the yaml and not a value check. Or maybe this is what you mean? You want to have a check that it's in the yaml? I would guess from the code, but maybe I misunderstand that even with those values it will fail. So I think these values may be redundant.

So maybe I'm understanding your reply. but let me know

Agreed, this is broader than just this class. I tend to code things up with default values too. I just wonder if once the development is done, it is better to remove the default value assignment to ensure that the value is actually passed from the yaml file rather than inadvertently relying on the hard-coded default value, that's all. I don't expect it to be an issue, just redundant.

Are you basically advocating this section should be

bool Enabled;
Real NearIrFraction;
Real NearIrCoeff;
Real RedFraction;
Real BlueFraction;

?

If so I think I'd still prefer to leave the code as is. The model will still fail if the parameters aren't in the yaml file, I feel having the defaults is nice for clarity.

Basically if the parameter is not in the yaml file, the code should fail. Which I expect it to do given the checks at the class creation. So there should always be a value passed, so a local default value is unnecessary (and potentially confusing if we change default values down the line).

Leaving the code as is is fine by me, it is a preference rather than a requirement.

Fair point. I'd prefer though this not be changed here alone. Can you create an issue for discussion? I like having the default, but I definitely see your point about changing of defaults. it would be good to get input from the broader team.

alicebarthel · 2026-10-07T19:09:01Z

+
+      Real FluxAtLayerTop = SurfaceFlux;
+      for (I4 K = KTop; K <= KBot; ++K) {
+         Real FluxAtLayerBottom;


Declaring inside the loop seemed strange to me but it seems like for a double/float type it makes no difference to modern compilers.

Yes that is odd. I'll fix it momentarily

brian-oneill · 2026-10-07T19:27:57Z

@vanroekel The PenetratingShortwaveOnCell functor is written to use within a flat parallelFor wrapper, but contains a serial for loop which has only a superficial dependence between iterations. This is a perfect candidate for hierarchical parallelism. I see two possibilities: the serial for could be changed to a parallelForInner and redundantly calculate the top and bottom fluxes each iteration, or using two parallelForInner loops and scratch memory to calculate the interface fluxes first, then the tendencies.

I'm happy to do the refactor, do I have your permission to push a change to this branch?

alicebarthel · 2026-10-07T19:29:28Z

+   const Real TotalFraction = this->PenetratingShortwave.NearIrFraction +
+                              this->PenetratingShortwave.RedFraction +
+                              this->PenetratingShortwave.BlueFraction;
+   if (std::abs(TotalFraction - 1.0_Real) > 1.0e-5_Real) {


Suggested change

if (std::abs(TotalFraction - 1.0_Real) > 1.0e-5_Real) {

if (Kokkos::abs(TotalFraction - 1.0_Real) > 1.0e-5_Real) {

alicebarthel · 2026-10-07T19:34:41Z

 #include "VertCoord.h"
 #include "auxiliaryVars/KineticAuxVars.h"
 #include "auxiliaryVars/PseudoThicknessAuxVars.h"
+#include "auxiliaryVars/ShortwavePenAuxVars.h"


Does it make more sense to use forcingVars rather than auxiliaryVars? I know surface restoring is still in Aux but that was left over from the early design. Not a deal breaker, just a scope question.

alicebarthel

This looks great @vanroekel !
Approving based on visual inspection, my comments are all minor. Is this in need of additional testing in polaris?
The ctests coverage looks good but let me know if you need anything else on this.

vanroekel · 2026-10-07T20:11:36Z

Thanks for the review @alicebarthel I'll address your comments now. I think the test coverage is good. I just need to test aurora. I'm turning on the polaris solar test I'm adding (E3SM-Project/polaris#796) in the pr suite so it's getting tested on every machine.

vanroekel · 2026-10-07T20:42:46Z

@brian-oneill I'd be happy to have you do that. Let me address and push fixes to @alicebarthel comments then you can take on the refactor. I'll let you know when ready

brian-oneill · 2026-10-08T19:22:15Z

The move from aux state to forcing was causing a failure in the FILL_VALUE_TEST, so I just pushed a small change to resolve that. I'm running some tests using the shortwave_pen test from E3SM-Project/polaris#796. Once that all looks good, I'll push the refactor.

brian-oneill · 2026-10-09T02:18:20Z

Testing

Ran the ctests at cbc3a06 with all tests passing using the following configurations:

Chrysalis with intel
Frontier with craygnu
Frontier with craygnu-mphipcc
pm-cpu with gnu
pm-gpu with gnugpu

brian-oneill · 2026-10-09T02:39:54Z

Testing

The following results are from tests using the ocean/column/shortwave_pen test from E3SM-Project/polaris#796 with 3d189cd as the baseline build and cbc3a06 as the PR build

Chrysalis, `intel`

Baseline workdir: /lcrc/group/e3sm/ac.boneill/polaris/baselines/intel_swpen/
Baseline build: /lcrc/group/e3sm/ac.boneill/lvrFork/build_intel
PR build: /lcrc/group/e3sm/ac.boneill/myFork/sw-refactor_intel
PR workdir: /lcrc/group/e3sm/ac.boneill/polaris/10-08-26/sw-refactor_intel
Machine: chrysalis
Partition: batch
Compiler: intel
Build type: Release
Log: /lcrc/group/e3sm/ac.boneill/polaris/10-08-26/sw-refactor_intel/1304646o.txt
Result: All tests passed

Frontier, `craygnu`

Baseline workdir: /lustre/orion/proj-shared/cli115/boneill/polaris/baselines/craygnu_swpen/
Baseline build: /lustre/orion/proj-shared/cli115/boneill/lvrFork/build_craygnu
PR build: /lustre/orion/proj-shared/cli115/boneill/myFork/sw-refactor_craygnu
PR workdir: /lustre/orion/proj-shared/cli115/boneill/polaris/10-08-26/sw-refactor_craygnu
Machine: frontier
Partition: batch
Compiler: craygnu
Build type: Release
Log: /lustre/orion/proj-shared/cli115/boneill/polaris/10-08-26/sw-refactor_craygnu/5638945o.txt
Result: All tests passed

Frontier, `craygnu-mphipcc`

Baseline workdir: /lustre/orion/proj-shared/cli115/boneill/polaris/baselines/craygnu-mphipcc_swpen
Baseline build: /lustre/orion/proj-shared/cli115/boneill/lvrFork/build_craygnu-mphipcc
PR build: /lustre/orion/proj-shared/cli115/boneill/myFork/sw-refactor_craygnu-mphipcc
PR workdir: /lustre/orion/proj-shared/cli115/boneill/polaris/10-08-26/sw-refactor_craygnu-mphipcc
Machine: frontier
Partition: batch
Compiler: craygnu-mphipcc
Build type: Release
Log: /lustre/orion/proj-shared/cli115/boneill/polaris/10-08-26/sw-refactor_craygnu-mphipcc/5638945o.txt
Result: All tests passed

pm-cpu, `gnu`

Baseline workdir: /global/cfs/cdirs/e3sm/boneill/polaris/baseline/gnu_swpen/
Baseline build: /global/cfs/cdirs/e3sm/boneill/lvrFork/build_gnu
PR build: /global/cfs/cdirs/e3sm/boneill/myFork/sw-refactor_gnu
PR workdir: /global/cfs/cdirs/e3sm/boneill/polaris/10-08-26/sw-refactor_gnu
Machine: pm-cpu
Compiler: gnu
Build type: Release
Log: /global/cfs/cdirs/e3sm/boneill/polaris/10-08-26/sw-refactor_gnu/59572864o.txt
Result: All tests passed

pm-gpu, `gnugpu`

Baseline workdir: /global/cfs/cdirs/e3sm/boneill/polaris/baseline/gnugpu_swpen
Baseline build: /global/cfs/cdirs/e3sm/boneill/lvrFork/build_gnugpu
PR build: /global/cfs/cdirs/e3sm/boneill/myFork/sw-refactor_gnugpu
PR workdir: /global/cfs/cdirs/e3sm/boneill/polaris/10-08-26/sw-refactor_gnugpu_1
Machine: pm-gpu
Compiler: gnugpu
Build type: Release
Log: /global/cfs/cdirs/e3sm/boneill/polaris/10-08-26/sw-refactor_gnugpu_1/59572867o.txt
Result: All tests passed

Aurora, `oneapi-ifxgpu`

Baseline workdir: /lus/flare/projects/E3SM_Dec/bjoneill/polaris/baselines/oneapi-ifxgpu_swpen
Baseline build: /lus/flare/projects/E3SM_Dec/bjoneill/lvrFork/build_oneapi-gpu
PR build: /lus/flare/projects/E3SM_Dec/bjoneill/myFork/sw-refactor_oneapi-ifxgpu
PR workdir: /lus/flare/projects/E3SM_Dec/bjoneill/polaris/10-08-26/sw-refactor_oneapi-ifxgpu_1
Machine: aurora
Compiler: oneapi-ifxgpu
Build type: Release
Log: /lus/flare/projects/E3SM_Dec/bjoneill/polaris/10-08-26/sw-refactor_oneapi-ifxgpu_1/8915070o.txt
Result: All tests passed

vanroekel · 2026-10-09T15:35:43Z

@brian-oneill and @katsmith133 I've rebased and retested on frontier. let me know if you have any further changes you'd like. Otherwise I think this is ready to go.

brian-oneill

I was very thorough in testing the changes I pushed (which helped me catch a couple bugs along the way). So other than one minor cleanup below, I approve.

brian-oneill · 2026-10-09T16:48:16Z

+#include <cmath>
+#include <iostream>


These includes don't look necessary, probably a vestige of developing/debugging?

vanroekel requested review from alicebarthel and a lite review from Copilot September 22, 2026 16:52

vanroekel assigned sbrus89 Sep 22, 2026

vanroekel added enhancement New feature or request Omega labels Sep 22, 2026

Copilot started reviewing on behalf of vanroekel September 22, 2026 16:53 View session

Copilot AI reviewed Sep 22, 2026

View reviewed changes

Comment thread components/omega/test/ocn/SfcCouplingTest.cpp Outdated

Comment thread components/omega/test/ocn/TendencyTermsTest.cpp Outdated

Comment thread components/omega/src/ocn/OceanInit.cpp Outdated

vanroekel force-pushed the vanroekel/omega/penetrating_solar branch from ca8d98a to 828ae13 Compare September 24, 2026 05:04

katsmith133 reviewed Sep 30, 2026

View reviewed changes

vanroekel force-pushed the vanroekel/omega/penetrating_solar branch from 828ae13 to 585728b Compare October 2, 2026 16:17

vanroekel mentioned this pull request Oct 7, 2026

Updates mpas to omega conversion script for shortwave extinction coefficients E3SM-Project/polaris#838

Open

2 of 4 tasks

alicebarthel reviewed Oct 7, 2026

View reviewed changes

alicebarthel approved these changes Oct 7, 2026

View reviewed changes

alicebarthel mentioned this pull request Oct 7, 2026

Should local default values be removed or left in? #595

Open

vanroekel and others added 12 commits October 8, 2026 22:07

First draft of penetrating shortwave radiation

6dacc75

Fixes a SW bug and adds a SW unit test

342b135

fixes penetrating SW unit tests

037030e

Updates docs for penetrating shortwave

3daa76f

updates hard coded values to parameters

be1a4a9

addresses review comments

0ec0fac

Rebase fixes

a37f3d0

fixes documentation

75a45fc

moves shortwave to forcing and removes unneeded halo exchange

95c9568

Add forcing::init to FillValueTest

f893a6c

Refactor PenetratingShortwaveOnCell to use hierarchical parallelism

d88fd5e

Rebase fixes

9cc4be9

vanroekel force-pushed the vanroekel/omega/penetrating_solar branch from cbc3a06 to 9cc4be9 Compare October 9, 2026 14:10

vanroekel added 2 commits October 9, 2026 10:53

linting fixes

5cf190a

Fixes another rebase error

f8555e1

brian-oneill approved these changes Oct 9, 2026

View reviewed changes

removes unneeded imports

41c0616

	Config PenSWConfig("PenetratingShortwaveTendency");
	Config PenSwConfig("PenetratingShortwaveTendency");

	if (std::abs(TotalFraction - 1.0_Real) > 1.0e-5_Real) {
	if (Kokkos::abs(TotalFraction - 1.0_Real) > 1.0e-5_Real) {

Conversation

vanroekel commented Sep 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Copilot review overview

🟡 Changes recommended

Uh oh!

Uh oh!

Uh oh!

Uh oh!

vanroekel commented Sep 24, 2026

Uh oh!

katsmith133 left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

vanroekel commented Oct 4, 2026

Polaris omega_pr suite

Uh oh!

vanroekel commented Oct 5, 2026

Polaris omega_pr suite

Uh oh!

vanroekel commented Oct 5, 2026

Polaris omega_pr suite

Uh oh!

vanroekel commented Oct 5, 2026

Polaris omega_pr suite

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

vanroekel commented Sep 22, 2026 •

edited

Loading

Polaris `omega_pr` suite

Polaris `omega_pr` suite

Polaris `omega_pr` suite

Polaris `omega_pr` suite