You are viewing an old version of this page. View the current version.

Compare with Current View Page History

Version 1 Current »

INFRA - Evan

  • Preparing spack-stack-2.1.1 release, stay tuned
  • Preparing for a jedi-bundle release by the end of May, which is targeting to be based on spack-stack-2.1
  • JEDI CI caching is working
    • Planning on rolling this out over the next couple weeks.
    • We are seeing a ~2x speedup.

INTERFACES - SteveV

  • IODA - Lewis (Met Office), SteveV and SteveH
    • Refining features, implementation of the new OSDF container
    • Continuing exercising OSDF based ObsSpaces in workflows and unit testing
      • New OSDF writer is producing feedback files that are under evaluation
        • Eg., benchmark comparison with current ObsGroup based ObsSpaces feedback files
  • Vader - SteveV
    • Refining JEDI standards and design regarding variable naming and metadata
  • UFO - Raji, SteveV
    • General work to resolve issues, improve code, etc.
    • No announcements for the wider audience

ALGO - Christian

  • No major updates from Algo, just a request to review saber Prs related to GSI BEC.
  • Travis mentioned that a new implicit vertical diffusion is available in saber
    • Areas where explicit vertical diffusion can be replaced with the new implicit vertical diffusion which should noticeably speed things up
    • Hernan mentioned that he has tested the implicit vertical diffusion and it looks good

OBS - FrancoisV

  • IODA converter:
    • GNSS TEC bugfix #1782 
      • Metadata values were not being added when TENET obs values were missing
    • Add a ctest for the gnss spacebased TEC converter #1779
  • UFO:
    • Eigen fixes #418: Patches for Eigen 5.0.0 compatibility.
    • Fix cmake usage of gsw gsw_LIBRARIES #4135:
      • When gsw is installed from gitHub, it has a different CMakeLists.txt file than the one in JCSDA gitHub. Then the library target is not quite set up the same way, and the subsequent build would fail to find gsw's Fortran module files.
    • Fix issues with empty ObsSpace #4122
    • Fix several pieces of code that currently fail if the ObsSpace is empty
    • Add zenith angle and hofx Reflectivity for active sensor #4022
      • The zenith angle becomes missing after applying the filter Average Observations To GeoVals Model Levels. It is then reassigned the value 80 in ufo/src/ufo/operators/crtm/ufo_crtm_utils_mod.F90, which is unreasonable for DPR.
      • The issue was resolved through YAML configuration and not a code change.
  • PyIRI
    • Increment operators implemented #177
    • This PR implements methods in IncrementPyiri, which were needed by Skylab. It also adds an interface test for IncrementPyiri and implements the methods required for the interface test to pass. 

CRTM - BenJ

  • Transmittance coefficient generation package is nearly completed, working on cleanup and preparation for distribution to key partners and stakeholders
  • CRTM ocean emissivity estimates are poor over ocean above 200 GHz (e.g., ICI, PolSIR, AWS, etc.), so I've created a new LUT that is trained against PARMIO (a reference emissivity model).  The new LUT will support up to ~800 GHz with accurate emissivity and polarization states. 
  • Cheng is working on MPAS-JEDI evaluation of PolSIR forward operator and AWS MWR O-B.

EMC - Dan

  • We need the syncing of UFO public and private to be fixed with utmost urgency.
    • Evan manually sync'd UFO during the meeting
    • Evan will set up cron job to do manual sync's for short term
    • Work on refining the github actions/runners/etc will fix this for the long term
  • We have started work on i-jedi and have established basic functionality.
  • Working towards moving variable changes to Vader (soca first and then fv3-jedi et al)
  • MGBF into Saber PRs.
  • Other PRs into Saber in support of regional background error model. Thanks to the JEDI team for reviewing and merging PRs.

GMAO - no update

NRL - Dom

  • Good progress on 4DVar flow - satellite obs are working
  • Build flows are being upgraded 
    • Updated JEDI components commit/tags
    • Targeting spack-stack-2.2
  • Updating to a newer version of NEPTUNE which in general is going well, but have seen some cases where runtime slowed down

UKMO - MikeC

  • PR in IODA (#1733) which will change ODB file output
    • Just want people to be aware
    • Believe that this impact is limited to the Met Office workflows

MMM - BJ

  • Junmei Ban continues working on assimilating CRTM active sensors. Ben Ruston and Chris Thomas helped her configure complex obs filters, although there remains further work to be done.
  • Jim Wittig is working on preparing the spack-stack v2.2 on NCAR HPC, Derecho.
  • We still plan the next community release of mpas-jedi at the end of this month. We are working on several small PRs to clean up (or improve) some ctest yamls. As an example, ctests using the regional MPAS meshes will be added (open PR).

ROMS - no update

  • No labels