oldrot2_mod.f

OLDROT2_MOD holds the terms of the time derivative of $\theta$ in the Maeder–Zahn (1998) advection–diffusion treatment of angular momentum ("MHP 06/02 added term for the time evolution of the angular velocity distribution"), at the start of the step and during the substeps, shared by 5 routines. It replaces COMMON/OLDROT2/ (MHP via Claude, 10/14/26); JSON comes from DIMENSIONS_MOD. All members are arrays of JSON.

What its variables hold, who sets them, who reads them. At each interface I (between shells I-1 and I):

• THM: $r^4/(\sqrt{m}\,G)$ (in logs, 2(log r) − ½ log m − log G), set by SETUPV ("added for $d\theta/dt$ term from Zahn & Maeder 1998").
• THO, QWRST: $\theta$ = THM $\bar\Omega\,d\Omega/dr$ and $d\Omega/dr$ of the old model, set by HPOINT; GETW copies them into THP, QWRMST on the first iteration.
• THP, QWRMST, WMST: $\theta$, $d\Omega/dr$ and $\Omega$ at the start of the current substep, set by GETW (from the old model on the first iteration, recomputed from the current $\Omega$ after it). THP is read by VCIRC; WMST by SECULR (the starting surface $\Omega$ for the wind); QWRMST nowhere (getw.f, line 275↗, getw.f, line 287↗): unused?
• THN: the $d\theta/dt$ term, FESTIME (THM $\bar\Omega\,d\Omega/dr$ − THP)/$\Delta t$, set and read by VCIRC, which adds it to the circulation velocities.
• FESTIME: $P/(g\,\Delta\nabla\,\rho)$ at the interface, set by SETUPV (0 where unused) and rescaled by VCIRC with the μ-gradient factor.
• DDELM: $\nabla_\mathrm{ad} - \nabla$, set by HPOINT from the old model and by SETUPV (with the μ term, not below a floor); read by SETUPV and VCIRC.

A blanket USE OLDROT2_MOD. DDELM has two meanings, set by HPOINT and then overwritten by SETUPV.