Skip to content

fix: correct Wigner 3j argument order in reduced_dipole#1134

Merged
riclarsson merged 2 commits into
atmtools:mainfrom
olemke:port-lm-bugfix
Jun 23, 2026
Merged

fix: correct Wigner 3j argument order in reduced_dipole#1134
riclarsson merged 2 commits into
atmtools:mainfrom
olemke:port-lm-bugfix

Conversation

@olemke

@olemke olemke commented Jun 22, 2026

Copy link
Copy Markdown
Member

Fixes an incorrect argument ordering in the Wigner 3j symbol computed inside reduced_dipole within the Hartmann ECS line-shape model. The previous call swapped the (li, lf - li, -lf) triplet with (lf, li - lf, -li), producing wrong reduced dipole values and thus erroneous scaling of relaxation-matrix off-diagonal elements. The corrected arguments restore the physically intended coupling for the rotational quantum numbers.

Same fix as in ARTS 2.6 PR #1133.

Changes

  • src/core/lbl/lbl_lineshape_voigt_ecs_hartmann.cpp: Updated both branches of the iseven(Jf + lf + 1) conditional in reduced_dipole to call wigner3j(Jf, k, Ji, lf, li - lf, -li) instead of wigner3j(Jf, k, Ji, li, lf - li, -lf), ensuring the bra-side and ket-side angular momentum couplings are passed in the correct order.

Breaking Changes

None — this is a bug fix. Results previously computed with the Hartmann ECS line-shape model for affected transitions will change numerically to the correct values.

olemke added 2 commits June 22, 2026 11:17
Swap the m-arguments passed to wigner3j in reduced_dipole so the
second triplet matches the intended (lf, li-lf, -li) ordering rather
than (li, lf-li, -lf).

Same fix as for ARTS 2.6 in PR atmtools#1133.
Fix will be done in separate PR.
@riclarsson riclarsson merged commit c2a1b3d into atmtools:main Jun 23, 2026
9 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants