2.6.1

#640 Hot fix that solves MPI library compile error under certain preprocessor flags. Otherwise idential to 2.6

2.6

Main Changes:

• #550 Scalar geometry description for oasis and yac
• #498 Multio support for FESOM standalone
• #465 Support for 3rd order Adams Bashford solver
• #489 Couple Iceberg model
• #576 Add active age tracers
• #587 Add netcdf compression for default output
• #597 Remove parms library
• #354 Add ugrid compliance for default output
• #613 Add ability to use fractions of JRA55 forcing cycle
• #497 Add split-explicit ssh solver
• #583 Add RECOM OBGC coupling
• #637 Ensure functionality of Icepack 1.2 coupling

New diagnostics

• #479 Add diagnostic for first moment of horizontal tracer fluxes
• #578 Add Discret Variance Decay (DVD) diagnostic

Complete list of changes
Milestone 2.6

2.5.0.5_de

What's Changed

• Updates 2.5.0.4 de by @raguridan in #592

New Contributors

• @raguridan made their first contribution in #592

Full Changelog: 2.5.0.4_de...2.5.0.5_de

2.5.0.4_de

Full Changelog: 2.5.0.2_de...2.5.0.4_de

Add MareNostrum 5 support.

2.5.0.1_de

DestinE version 2.5.0.1_de

The _de indicates it's DestinE specific release.

What's Changed

• Removes Ollie Scripts by @pgierz in #482
• Fix a few typos on the Getting Started docs by @kinow in #484
• Refactoring razvan multio by @dsidoren in #487
• remove unused TODO block for IFS currents coupling by @sebastianbeyer in #490
• Add same information from GH readme to RTD docs by @kinow in #488
• fix that ensures we dont check for salinity <= 3 inside ice shelf by @JanStreffing in #494
• fixes #496 blas discovery by @suvarchal in #499
• updating documentation for fesom standalone and job_levante by @fernandadialzira in #501
• Refactoring runoff mapper by @dsidoren in #502
• some updates important for mio functionality by @suvarchal in #505
• add readthedocs.yaml by @koldunovn in #507
• change readthedocs to python 3.9 by @koldunovn in #508
• install theme for docs by @koldunovn in #509
• small fix for docs by @koldunovn in #511
• update docker image to refactoring by @koldunovn in #516
• update docker image to refactoring again, sorry suvi :) by @koldunovn in #517
• Add lumi CI/CD workflow by @tiggi in #519
• Lumi CICD sync fix by @tiggi in #520
• an OMP bux fix in tracer FCT by @dsidoren in #522
• Bugfixes to parms, widely compatable FindNETCDF, and new envs for atos and lumi by @suvarchal in #523
• Refactoring fct omp bugfix by @dsidoren in #525
• added OMP for mEVP (aEVP is still not covered). tested for CORE2 mesh… by @dsidoren in #527
• fix OpenMP in EVPdynamics by @sebastianbeyer in #534
• fix OpenMP in EVPdynamics by @sebastianbeyer in #535
• production_DE fixup by @wdeconinck in #531
• Multio: change some gridType metadataA by @sebastianbeyer in #540
• multio: set unstructuredGridType from namelist by @sebastianbeyer in #541
• Make async_threads_module and fesom_C an OBJECT library by @wdeconinck in #537
• production_DE with atosecmwf platform strategy by @trackow in #542
• dont write redi output when using multio by @sebastianbeyer in #547
• Always convert to double precision before multio by @sebastianbeyer in #548
• Remove old Cray exception for OpenMP - builds with tagged ifs-bundle by @trackow in #549
• Lumi gpu production de by @dsidoren in #555
• IFS-interface: read precomputed weights matching mpi ranks by @sebastianbeyer in #556
• fixed GPU bug when dynamics%use_wsplit=.TRUE. The problem was that by @dsidoren in #560
• production_DE fix in ice_fem_fct for occurring Overflows by @trackow in #562

New Contributors

• @fernandadialzira made their first contribution in #501
• @tiggi made their first contribution in #519

Full Changelog: AWI-CM3_v3.2...2.5.0.1_de

AWI-CM3_v3.2

Version of FESOM2 that was used for AWI-CM3.2

Includes all changes from FESOM 2.1, FESOM 2.1.1 and FESOM 2.5.

In addition using this release FESOM2 is capable of writing grid corner points for OASIS, provided the mesh has no bad coastal nodes. #432 #450

FESOM2.5

Main changes

• Complete refactoring of the code in order to implement modular structure. Now variables are wrapped in derived types. For tracers [#167, #185 ], dynamics [ #214, #219, #228, ], ice [#251, ]

• Add OpenMP support [#196, #217 , #220, #222, #226, #227, #232, #233, #235, #236, #241, #242, #244, #246, #247, #256, #269, #293, #346, #406], with binary reproducibility #357

• Implementation of binary restarts and parallel reading of restarts #260, #265, #296

• Add capability to easilly extract dwarfs (computational kernels) from the code, so that they can be optimised separatelly (#297, #302, #303, #304, #367)

• Energy diagnostic #416, #417

Improvements for coupling with atmospheric models

• Multiple fixes for AWI-CM3 #352, #374
• FESOM main split in three parts (as required for IFS coupling) #195
• Add a switch to build FESOM as library (for IFS coupling) #207 #229, #231
• Implement IFS interface #224, #238, #239, #327
• IFS coupling related changes #321, #324, #331, #332, #429, #433
• Send ocean variables to IFS to output them on atmospheric grid #435
• Big IFS related update #388, that includes:
  • coupling of surface ocean currents (took nodal velocities instead of velocities on elements)
  • optimization of coupling interface (packing of fields, p2p instead of alltoallv) following what was done for NEMO4
  • extra diagnostics are available on IFS grid (MLD, ave T 300m, ave S 300m, 20C isotherm, etc) and thus in GRIB output
  • reduced standard I/O output in single-executable runs

Compiler and HPC support

• nvfortran compiler support #266, #290, #291, #317, #328, #329, #376
• Use consistent target attribute (for Cray compiler) #198
• Compiles with OpenMP with cray compiler #200
• Add Levante support #338, #358, #365, #394
• Add LUMI support #387, #412, #406
• Add albedo support #391, #392, #400, #419

New features

• Automatic rotation for vector fields, no need or rotation in postprocessing #301, #335
• Allow topography to be read on elements and nodes #359
• Support for driving FESOM with the windstress (instead of the wind speed -> bulk formulas) from reanalysis #366
• Add MLD3 with cmor compliant density threshold of 0.03 kg/m³ #383
• Chlorophyll climatology can be used instead of constant value. Shall be activated through namelist #339
• Add Langmuir Parameterisation for TKE, update IDEMIX parameter #340
• Transposed 3D output files via preprocessor definition #193
• add the output of density class thickness to the DMOC diagnostic for the proper remapping to zcoordinates #372
• Additional IO output for upper levels for nextGEMS #436

Improvements for High resolution simulations

• Brunt-Vaisala frequency is smoothed spatially before it is given to the mixing schemes. This damps some mode which is expressed as a grid noise in the modelled fields (mainly at the equator) #287
• Smoothing Bv in vertical is necessary in order to avoid vertival noise in Kv/Av #385
• Fix bug, when visc_gamma0, visc_gamma1 & visc_gamma2 were used instead gamma0_tra, gamma1_tra & gamma2_tra in diff_part_bh #276
• Multiple bugs fixed in biharmonic tracer diffusion, that allow better results for high resolution simulations #277, #282
• Bug fix in pressure_bv (depth computatiopn for pressure contribution) #320

Bug fixes

• Multiple TKE+IDEMIX fixes #351
• fix bug with reading runoff and SSS #177
• Remove last occurence of execute_command_line function, was failing too often #318
• Fix KPP bug related to sw_pene #237
• Bug fix in update_atm_forcing. press_air array was accessed even if not allocated. #248
• CFLMAX was not initialized in vert_vel_ale #253
• Fixed treatment of sw_pene in cvmix_KPP #330
• Fixed depth sign in the parameterisation of Langmuir circulations (LC in TKE) #353
• zisotherm computation improved #401
• diag_densMOC assumed that GM is used. An IF condition was introduced #410
• Multiple bugfixes for cavity #390
• Make sure that the restart works from zstar towrads linfs #375

Documentation

• Includes the formerly separate documentation repository directly along with the source code #344

AWI-CM3 version 3.1

Version of FESOM2 that was used for AWI-CM3.1

AWI-CM3 version 3.0 (Streffing et al. 2022)

Version of FESOM2.0 that was used for https://doi.org/10.5194/gmd-15-6399-2022
AWI-CM3 coupled climate model: Description and evaluation experiments for a prototype post-CMIP6 model

FESOM2.1.1

This release is the last one, where the global results with default values will be bit reproducible with release 2.1.0 (4979513).

Main changes

• Be able to call the FESOM executable with options for testing it without running a simulation (i.e. a minimal dry-run) #121
• Print information about the currently used MPI library at FESOM startup #122
• Fix mesh.diag files, so xraay can understand them #126 and add lon lat to the output #152
• Multiple fixes for KPP #128
• Fix bugs in the zstar volume conservation of the soufflet channel experiment (channel results changed) #131
• Several fixes for cavity #132 #135 #149
• Fix problem with FESOM2.0 reproducability bias on ollie #146
• Fix severe with uninitialized copies of variables #147
• Several fixes for AWI-CM3 #154 , #156
• Add NCEP reanalysis as one of the possible forcings #155
• Ignore the error message from reading the restart if the variable is not in the file #159