AGF-01_Debreu_domcfg

[smasson]

Workplan action

AGRIF / Configuration domain generation tool refactoring & Agrif friendly

Wikipage: wiki:2021WP/AGF-01_Debreu_domcfg

[smasson]

In 14610:

AGFdomcfg: create dev branch, #2638

[smasson]

In 14612:

AGFdomcfg: create AGRIF dev branch, #2638

[smasson]

In 14613:

AGFdomcfg: change svn:externals to use AGRIF/dev_r14608_AGRIF_domcfg, #2638

[ldebreu]

In 14619:

AGFdomcfg:uncomment the call to agrif_set_external_mapping, call agrif_nemo_init even on the root grid #2638

[ldebreu]

In 14620:

AGFdomcfg: allow the conv to understand the IOMSG keyworkd #2638

[ldebreu]

In 14621:

AGFdomcfg: allow the conv to understand the IOMSG keyworkd #2638

[ldebreu]

In 14622:

AGFdomcfg: 1) define some AGRIF variables in real8 2) remove the wrong indices translation in modarrays #2638

[ldebreu]

In 14623:

AGFdomcfg: 1) Update DOMAINcfg to be compliant with the removal of halo cells 2) Update most of the LBC … subroutines to a recent NEMO 4 version #2638

[ldebreu]

In 14624:

AGFdomcfg: Update namelist files (jpiglo, jpjglo are replaced by NiGlo0, NjGlo0 version #2638

[ldebreu]

In 14625:

AGFdomcfg: Update AGRIF_FixedGrids.in, put an updated version for the AGRIF_DEMO test #2638

[ldebreu]

In 14626:

AGFdomcfg: Update make_namelist.py (Use of NiGlo0, NjGlo0 instead of jpiglo, jpjglo #2638

[jchanut]

In 14628:

AGFdomcfg: Fixes a couple of tiny compilation problems #2638

[jchanut]

In 14629:

AGFdomcfg: add missing block in reference namelists, #2638

[jchanut]

In 14630:

AGFdomcfg: 1) restore use of analytical grids (i.e. case where jphgr_msh>0). That may be useful to set up test cases with AGRIF. 2) Add random topography over a flat bottom if nn_bathy = -1 (in place of a Gaussian bump). This illustrates well where the interface matching and update are done. #2638

[jchanut]

In 14633:

AGFdomcfg: fixes internal flags in the East-West periodic case and in cmems demo, #2638

[jchanut]

In 14634:

AGFdomcfg: temporary removal of lbclnk in procnames: this prevents using the tool in mpp mode, #2638

[jchanut]

In 14640:

AGFdomcfg: slight changes in boundary matching location to comply with what is expected in NEMO, #2638

[jchanut]

In 14641:

1) Revise boundary checking with AGRIF (unify vertical remaping case or not) 2) Disable parent volume check without vertical remaping until we sort out what to do in the damned overlapping zone. At this stage DOMAINcfg produces meshes in agreement with what NEMO expects, except for cyclic East-West child grids for which a mismatch persists at boundaries. Child grids over North Pole Fold or East-West boundaries are however correct, #2638

[jchanut]

In 14655:

AGRIF: Set ghost cell number to 0 over root grid to properly define analytical grids in zooms, #2638

[jchanut]

In 14656:

Revise VORTEX child grid position by taking into account Parent grid ghosts. No need to change AGRIF_FixedGrids.in file anymore if changing the number of ghostcells, #2638

[jchanut]

In 14673:

Update ICE_AGRIF child domain as for VORTEX (position dependent on ghost cell number), #2638

[jchanut]

In 14684:

Allow chosing type of input bathymetry in child grids again (wrongly introduced @14630), #2638

[jchanut]

In 14687:

#Changes to accomodate AGRIF new grid positionning implemented at r14674 and r14675, #2638

[jchanut]

In 14688:

Last round of changes following r14674, r14675 and r14687 (AGRIF new grid positionning), #2638

[jchanut]

In 14698:

#2638: initialize agrif ghost cells number (to zero) on root grid (DOMAINcfg)

[jchanut]

In 14702:

#2638, push restrictions in agrif mapping in case of East-West periodic or North Fold lbcs

[jchanut]

In 14703:

#2638, push restrictions in agrif mapping in case of East-West periodic or North Fold lbcs + Disable interp if lk_north=.false.

[jchanut]

In 14720:

#2638, use i refinement factor in place of j to define e1x over child grids

[jchanut]

In 14721:

#2638, (partially) correct bathymetry transitioning in corners + add lk_north=.false. case