Migrate NREL references to NLR (National Laboratory of the Rockies) by jason-curtis · Pull Request #2705 · pvlib/pvlib-python

jason-curtis · 2026-03-03T01:28:26Z

Summary

Update all nrel.gov URLs to nlr.gov equivalents following the NREL → NLR rename (December 2025). The nrel.gov domain stops resolving May 29, 2026.
Update critical API base URLs in psm4.py and midc.py (runtime paths)
Add NLR_API_KEY environment variable support with NREL_API_KEY deprecation fallback
Update all documentation, docstrings, comments, and test references

Verified Domain Mapping

Old domain	New domain	Status
`developer.nrel.gov`	`developer.nlr.gov`	301 redirect, API works
`www.nrel.gov`	`www.nlr.gov`	301 redirect, pages + PDFs work
`sam.nrel.gov`	`sam.nlr.gov`	301 redirect
`midcdmz.nrel.gov`	`midcdmz.nlr.gov`	Both serve (no redirect yet)

What was NOT changed (intentionally)

github.com/NREL/* URLs (org not renamed)
raw.githubusercontent.com/NREL/* URLs
web.archive.org URLs (archived snapshots)
Bibliographic report IDs (NREL/TP-5K00-76626 etc.)
Historical whatsnew entries (v0.14.x and earlier)
Publisher names in bibliographic citations (cited as published)
NREL-PySAM package references (PyPI package not renamed)
The nrel_api_key fixture name (kept for backwards compat; nlr_api_key added as alias)

CI Note for Maintainers

The CI workflow now sets both NLR_API_KEY and NREL_API_KEY from the existing NRELAPIKEY secret. No action needed unless you want to rename the secret.

PR template checklist

Closes Transition timing from nrel.gov to nlr.gov #2701
I am familiar with the contributing guidelines
I attest that all AI-generated material has been vetted for accuracy and is in compliance with the pvlib license
(N/A) Tests added
Updates entries in docs/sphinx/source/reference for API changes. (see note later about adding a docs page about NREL vs NLR)
Adds description and name entries in the appropriate "what's new" file in docs/sphinx/source/whatsnew for all changes. Includes link to the GitHub Issue with :issue:`num` or this Pull Request with :pull:`num`. Includes contributor name and/or GitHub username (link with :ghuser:`user`).
(N/A) New code is fully documented. Includes numpydoc compliant docstrings, examples, and comments where necessary.
Pull request is nearly complete and ready for detailed review.
Maintainer: Appropriate GitHub Labels (including remote-data) and Milestone are assigned to the Pull Request and linked Issue.

Followups checklist

integrate PR template checklist
https://github.com/NREL -> https://github.com/NatLabRockies (verify subrepo names etc)
consider adding a docs page about NREL vs NLR explaining the name change and when we use which names
revert changes in scholarly-style citations to docs that were published under NREL name (@cwhanse this is your intent, correct? ). I'd suggest that we can refer to NREL in actual citations, but then elsewhere ("# data from NREL 'Slope-Aware Backtracking for Single-Axis Trackers'" we use either NLR or "NLR (NREL at time of publishing)".
fix spacing typos identified by @cwhanse

Test plan

grep -rn 'nrel\.gov' pvlib/ docs/ tests/ confirms only excluded items remain
pytest tests/ — 1345 passed (8 pre-existing failures unrelated to this PR)
Spot-checked new URLs with curl -sI — confirmed 200/301 responses

AI Attestation

This PR was generated with assistance from Claude (Anthropic). All changes were reviewed for correctness.

🤖 Generated with Claude Code

Update all nrel.gov URLs to nlr.gov equivalents following the NREL→NLR rename (Dec 2025). The nrel.gov domain stops resolving May 29, 2026. - Update API base URLs in psm4.py and midc.py (critical runtime paths) - Add NLR_API_KEY env var support with NREL_API_KEY deprecation fallback - Update all documentation URLs, prose references, and test comments - Preserve report IDs (NREL/TP-xxx), GitHub URLs, and citation publishers Closes pvlib#2701 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

github-actions · 2026-03-03T01:28:45Z

Hey @jason-curtis! 🎉

Thanks for opening your first pull request! We appreciate your
contribution. Please ensure you have reviewed and understood the
contributing guidelines.

If AI is used for any portion of this PR, you must vet the content
for technical accuracy.

williamhobbs · 2026-03-03T01:44:08Z

Looks like https://github.com/NREL has moved to https://github.com/NatLabRockies.

For reference descriptions, like “the NREL SPA”, would it be helpful to add some sort of “formerly known as NREL…”? I could see this causing some confusion, especially to people newer to solar modeling.

cwhanse · 2026-03-03T14:13:40Z

@jason-curtis does Claude strip out pvlib's PR template?

cwhanse

When referring to the publisher, I think it should stay NREL.

I'm on the fence about NLR SPA vs. NREL SPA. NLR continues to distribute and support code, so that's NLR SPA. The report is NREL SPA. Even NLR's webpage has both: "...the following NLR technical report:"..."Reda, I., ..., NREL Report No...."

So either NLR SPA or NREL SPA is OK with me.

Thanks so much @jason-curtis for taking the initiative here.

docs/examples/spectrum/plot_spectrl2_fig51A.py

pvlib/spectrum/spectrl2.py

pvlib/clearsky.py

pvlib/snow.py

pvlib/solarposition.py

pvlib/spa.py

tests/test_shading.py

jason-curtis · 2026-03-03T15:57:06Z

does Claude strip out pvlib's PR template? - @cwhanse

Oops, sorry. Claude likes to create things directly with gh pr create --body [...] so it bypasses the template. I'll think to check for templates in the future.

williamhobbs · 2026-03-03T16:15:33Z

Does this look like the right checklist

I think it's best to edit the original comment and put the checklist there.

cwhanse · 2026-03-03T16:36:00Z

Claude likes to create things directly with gh pr create --body [...] so it bypasses the template.

good to know, that helps as we maintainers learn to work with AI in this process.

revert changes in scholarly-style citations to docs that were published under NREL name

My intent is to preserve NREL when identifying the publisher of a document. Use NLR when identifying where one can find the document now. Use NLR when describing "ownership" of data or a model, e.g., NLR PVWatts, NLR PSM4, etc.

docs/examples/irradiance-decomposition/plot_diffuse_fraction.py

docs/examples/spectrum/average_photon_energy.py

docs/sphinx/source/whatsnew/v0.15.1.rst

docs/tutorials/tmy_to_power.ipynb

pvlib/spectrum/irradiance.py

pvlib/pvsystem.py

AdamRJensen · 2026-03-03T16:44:08Z

The SPA should still be called NREL as that's who published it. We do not use the source code distributed by NLR.

jason-curtis · 2026-03-03T16:55:30Z

it's now distributed by NLR though, right? <https://optionzero.co>

…

On Tue, Mar 3, 2026 at 8:47 AM Adam R. Jensen ***@***.***> wrote: *AdamRJensen* left a comment (pvlib/pvlib-python#2705) <#2705 (comment)> The SPA should still be called NREL as that's who published it. We do not use the source code distributed by NLR. — Reply to this email directly, view it on GitHub <#2705 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAZPVU7P35IO6E4KIJ5XGF34O4EA3AVCNFSM6AAAAACWE32HH2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTSOJSGI3DGMRTGA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

cwhanse · 2026-03-03T17:13:36Z

NREL if the reference is to an NREL publication. It's NLR's SPA model now when we are referring to where one can find more information about the algorithm or NLR's implementation.

* update links for redirects * roll back some NLR -> NREL for publisher reference * fix various typos and EOF removal Co-authored-by: Rajiv Daxini <143435106+RDaxini@users.noreply.github.com> Co-authored-by: Cliff Hansen <cwhanse@sandia.gov>

- Replace broken rredc.nlr.gov TMY3 link with nsrdb.nlr.gov archive - Use canonical https://docs.nlr.gov URL for PVUSA reference PDF - Add trailing newline Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

- Remove non-standard double URL in AM1.5 spectra reference - Remove "Pre-print:" prefix from Marion 2008 reference URL Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

NREL GitHub org has moved to NatLabRockies. Update all references in code and docs (excluding historical whatsnew entries). Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

jason-curtis · 2026-03-03T23:36:59Z

thanks for the feedback, folks. I believe I got it all. Also added a docs page about the migration.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

wholmgren

thanks @jason-curtis!

wholmgren · 2026-03-04T16:43:22Z

tests/conftest.py



+@pytest.fixture(scope="module")
+def nlr_api_key(nrel_api_key):


what's the point of the alias? I say just rename the existing function.

wholmgren · 2026-03-04T16:47:18Z

tests/conftest.py

+        demo_key = os.environ["NLR_API_KEY"]
    except KeyError:
-        warnings.warn(
-            "WARNING: NREL API KEY environment variable not set! "


I'd keep the existing pattern but change the text to "WARNING: NLR_API_KEY (formerly NREL_API_KEY) environment variable....". Don't bother with the deprecation that we have to go back and clean up in a few months.

wholmgren · 2026-03-04T16:58:06Z

docs/sphinx/source/whatsnew/v0.15.1.rst

+* Update all NREL references to NLR (National Laboratory of the Rockies)
+  following the laboratory rename and domain migration from ``nrel.gov``
+  to ``nlr.gov``. Add ``NLR_API_KEY`` environment variable support with
+  ``NREL_API_KEY`` fallback. (:issue:`2701`, :pull:`2705`)


update this pending resolution of need for fallback (see comment on the test fixture).

wholmgren · 2026-03-04T16:58:49Z

docs/sphinx/source/user_guide/index.rst

   :caption: Extras

   extras/nomenclature
+   extras/nrel_nlr


if we're adding a whole new doc page to explain our stance then we should probably link to it in the whats new

I see no reason for adding a page on this. That just seems like it'll be outdated soon and is a maintainance burden.

As evidenced by the 4 rounds of feedback on this PR, I think it's worth having something that explains why we're using NREL in some places and NLR in others but happy to defer to y'all.

wholmgren · 2026-03-04T17:05:41Z

tests/conftest.py



+@pytest.fixture(scope="module")
+def nlr_api_key(nrel_api_key):


test_psm4.py should point to this fixture

wholmgren · 2026-03-04T17:07:07Z

.github/workflows/pytest-remote-data.yml

        env:
          # copy GitHub Secrets into environment variables for the tests to access
+          NLR_API_KEY: ${{ secrets.NRELAPIKEY }}
          NREL_API_KEY: ${{ secrets.NRELAPIKEY }}


suggest we go ahead and remove this line. no need to update the secret name though

AdamRJensen

Reports that were published by NREL, should not be changed to NRL.

Osti.gov has been getting a lot better at giving DOIs, so most links can be updated. This avoids us having to change them again in the future.

docs/examples/irradiance-decomposition/plot_diffuse_fraction.py

docs/examples/spectrum/average_photon_energy.py

AdamRJensen · 2026-03-04T17:19:25Z

docs/sphinx/source/user_guide/index.rst

   :caption: Extras

   extras/nomenclature
+   extras/nrel_nlr


I see no reason for adding a page on this. That just seems like it'll be outdated soon and is a maintainance burden.

docs/tutorials/tmy_to_power.ipynb

pvlib/bifacial/infinite_sheds.py

AdamRJensen · 2026-03-04T17:36:34Z

tests/iotools/test_tmy.py

 def test_solaranywhere_tmy3(solaranywhere_index):
    # The SolarAnywhere TMY3 format specifies midnight as 00:00 whereas the
-    # NREL TMY3 format utilizes 24:00. The SolarAnywhere file is therefore
+    # NLR TMY3 format utilizes 24:00. The SolarAnywhere file is therefore


Change depending on decision of naming of TMY3 format.

pvlib/tracking.py

tests/test_tracking.py

Co-authored-by: Adam R. Jensen <39184289+AdamRJensen@users.noreply.github.com>

AdamRJensen · 2026-03-04T17:42:42Z

NREL if the reference is to an NREL publication. It's NLR's SPA model now, when we are referring to where one can find more information about the algorithm or NLR's implementation.

The downloadable C code should be referenced as NRL's. However, our implementations do not match the C-code and should reference the paper; and therefore should still be called NREL SPA.

- Rename nrel_api_key fixture to nlr_api_key directly (no alias) - Simplify fixture: check NLR_API_KEY, warn with DEMO_KEY fallback (no NREL_API_KEY deprecation logic) - Update test_psm4.py to use nlr_api_key fixture - Remove NREL_API_KEY from CI workflow - Update whatsnew: remove fallback mention, link nrel_nlr doc page

cwhanse · 2026-03-04T17:46:57Z

NREL if the reference is to an NREL publication. It's NLR's SPA model now, when we are referring to where one can find more information about the algorithm or NLR's implementation.

The downloadable C code should be referenced as NRL's. However, our implementations do not match the C-code and should reference the paper; and therefore should still be called NREL SPA.

pvlib's implementation doesn't match the NREL SPA code either, so I'm not understanding the concern, can you elaborate?

cwhanse · 2026-03-04T18:04:08Z

I think we (maintainers) need to agree on some rules to avoid giving @jason-curtis conflicting guidance. @pvlib/pvlib-maintainer do you agree with the following rules?

Citations to reports that were published by NREL, should not change the publisher to NRL.
URLs change to NLR, since NREL.gov will go away.
Replace URLs with DOIs where we can, to reduce future maintenance.

Open questions:
4. Is it NREL SPA / NREL TMY3, or is it NLR SPA / NLR TMY3? I think we'll use both, depending on the context. NLR is the entity that now has copyright of SPA C code and is the source for information to learn more. NREL is the entity that published (past tense) the technical report on which pvlib code is based. I think similar distinctions are important for SPA.

RDaxini · 2026-03-04T18:12:29Z

Would it help if I enquired internally at NREL to find out if there is any official guidance on this from NREL's perspective...? Or, is this wholly a pvlib decision?

disclaimer: there might not even be any official guidance, most of us are also still trying to figure these things out

cwhanse · 2026-03-04T18:14:37Z

Would it help if I enquired internally at NREL to find out if there is any official guidance on this from NREL's perspective...? Or, is this wholly a pvlib decision?

disclaimer: there might not even be any official guidance, most of us are also still trying to figure these things out

Yes, please do.

adriesse · 2026-03-04T19:02:37Z

I think we (maintainers) need to agree on some rules to avoid giving @jason-curtis conflicting guidance. @pvlib/pvlib-maintainer do you agree with the following rules?
1. Citations to reports that were published by NREL, should not change the publisher to NRL.

2. URLs change to NLR, since NREL.gov will go away.

3. Replace URLs with DOIs where we can, to reduce future maintenance.
Open questions: 4. Is it NREL SPA / NREL TMY3, or is it NLR SPA / NLR TMY3? I think we'll use both, depending on the context. NLR is the entity that now has copyright of SPA C code and is the source for information to learn more. NREL is the entity that published (past tense) the technical report on which pvlib code is based. I think similar distinctions are important for SPA.

I'm in favor of keeping the memory and legacy of NREL alive where possible and reasonable. I would also err on the side of caution by not changing something now, because we can always do it later. It doesn't have to be done with a single stroke of a felt-tipped pen.

There was an earlier name change from SERI to NREL, and I think folks in Golden have happily continued to improve the QC software named after it without changing the name. So for the open question, I vote for keeping NREL, at least for now.

RDaxini · 2026-03-04T22:35:22Z

Not sure what's up with the 👎, I was asking whether I should reach out to the NREL comms team or whether it's a pvlib decision for us to make alone—that's two options, so... 👎= ?

Anyway, I enquired already following @cwhanse's earlier response and was advised that anything published under the NREL brand should be cited and referenced to using the NREL brand. Only more recent work is branded as NLR and should be referenced as such. Recent works includes old software being updated by the now-NLR, if we are using the new version.
I think this is in line with the trend of our discussion anyway (#2705 (comment)) and I am on board with that. For 4. I think also think the answer is both as described in #2705 (comment).

I see no harm in having sought the lab comms team's input here.

AdamRJensen · 2026-03-04T22:56:36Z

Not sure what's up with the 👎, I was asking whether I should reach out to the NREL comms team or whether it's a pvlib decision for us to make alone—that's two options, so... 👎= ?

Certainly no harm. The thumbs down was just my way of saying "not worth it, don't waste your time". I admit that the thumbs down can be seen as a bit aggressive. But thanks for doing it 👍

adriesse · 2026-03-05T21:03:02Z

Not sure what's up with the 👎, I was asking whether I should reach out to the NREL comms team or whether it's a pvlib decision for us to make alone—that's two options, so... 👎= ?

My thumbs down was to the first question. If they told us something we didn't want to do, it could make it more awkward to do something different.

cwhanse reviewed Mar 3, 2026

View reviewed changes

RDaxini added this to the v0.15.1 milestone Mar 3, 2026

RDaxini added the documentation label Mar 3, 2026

cwhanse added documentation and removed documentation labels Mar 3, 2026

RDaxini reviewed Mar 3, 2026

View reviewed changes

jason-curtis and others added 5 commits March 3, 2026 15:04

Apply suggestions from code review

08f2c27

* update links for redirects * roll back some NLR -> NREL for publisher reference * fix various typos and EOF removal Co-authored-by: Rajiv Daxini <143435106+RDaxini@users.noreply.github.com> Co-authored-by: Cliff Hansen <cwhanse@sandia.gov>

Fix broken links in tmy_to_power tutorial

1ebb8d6

- Replace broken rredc.nlr.gov TMY3 link with nsrdb.nlr.gov archive - Use canonical https://docs.nlr.gov URL for PVUSA reference PDF - Add trailing newline Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Fix reference formatting in irradiance.py and pvsystem.py

0ae3558

- Remove non-standard double URL in AM1.5 spectra reference - Remove "Pre-print:" prefix from Marion 2008 reference URL Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Migrate github.com/NREL URLs to github.com/NatLabRockies

aec4682

NREL GitHub org has moved to NatLabRockies. Update all references in code and docs (excluding historical whatsnew entries). Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Add docs page explaining NREL to NLR rename

6563103

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

This was referenced Mar 3, 2026

Contributing guide should note that PRs must include the template checklist #2706

Open

Note in contributing guide that PRs must include the template checklist #2707

Open

Fix line too long in conftest.py docstring

fd771e6

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

wholmgren reviewed Mar 4, 2026

View reviewed changes

AdamRJensen reviewed Mar 4, 2026

View reviewed changes

Apply suggestions from @AdamRJensen CR

1820fdd

Co-authored-by: Adam R. Jensen <39184289+AdamRJensen@users.noreply.github.com>

chore: add runtime dirs to .gitignore

01ad3d8



		@pytest.fixture(scope="module")
		def nlr_api_key(nrel_api_key):

Conversation

jason-curtis commented Mar 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Verified Domain Mapping

What was NOT changed (intentionally)

CI Note for Maintainers

PR template checklist

Followups checklist

Test plan

AI Attestation

Uh oh!

github-actions bot commented Mar 3, 2026

Hey @jason-curtis! 🎉

Uh oh!

williamhobbs commented Mar 3, 2026

Uh oh!

cwhanse commented Mar 3, 2026

Uh oh!

cwhanse left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

jason-curtis commented Mar 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

williamhobbs commented Mar 3, 2026

Uh oh!

cwhanse commented Mar 3, 2026

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

AdamRJensen commented Mar 3, 2026

Uh oh!

jason-curtis commented Mar 3, 2026 via email

Uh oh!

cwhanse commented Mar 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jason-curtis commented Mar 3, 2026

Uh oh!

wholmgren 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!

AdamRJensen left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

jason-curtis commented Mar 3, 2026 •

edited

Loading

jason-curtis commented Mar 3, 2026 •

edited

Loading

cwhanse commented Mar 3, 2026 •

edited

Loading

RDaxini commented Mar 4, 2026 •

edited

Loading

AdamRJensen commented Mar 4, 2026 •

edited

Loading

adriesse commented Mar 5, 2026 •

edited

Loading