no general updates
obs:
IODA converter:
- Add sequence number to Madrigal sTEC and vTEC converters #1740; general bug fixes #1742
- JCSDA has met with UKMO, NOAA EMC and GSL, NASA regarding preparation for MTG-IRS and reviewed next steps and IODA conventions.
UFO:
- Geographical polygon filter #3977. Thanks NOAA (Sam Trahan). The documentation for this filter PR974 in jedi-docs needs review
- The duplicate check filter was using an ncDiag variable now fixed with PR3972 . Documentation needs review PR980
- Add check for solar zenith and azimuth angle #3973:
- Add a check for solar zenith and azimuth angles in high frequency GMI channels.
- Change variable in Duplicate Check ObsFunction #3972
- Change PreUseFlag to use parameter variable name and add new parameters.
Tech Debt reduction:
- compare strings fortran removed as intrinsic: #3986
- check on `nGood` variable to avoid nonsense entries: #3987
- fix FORTRAN line continuation: #3990
Plan to discuss (1) the request about the phasing out AtmVertInterpLay (2) debt reduction in next OBSPROC meeting Feb 24.
space weather:
- IODA converters (mentioned above)
- vTEC refactor PR merged UFO #1740
- Pathsum H(x) test in pyIRI pyiri-jedi ##145
- Update ObsErrorPyiri to allow localization #141
- Eliminate fill values from heights returned from horizontal interpolator #148
- Request for creating waccm-x-jedi repo
- Also met with UKMO, NRL, SWARD(NASA/CU), HAO regarding the JEDI-space weather development and coordination
crtm:
- Working on CRTM version 3.2: created a version 3.1.3 release spack-stack to be able to have IFX support, this is not intended to be a commonly used release, but really more for spack-stack
- Working on visible transmittance coefficient for ABI GOES 19, obs team is testing currently.
- We have received updated MTG-IRS coefficient from NOAA STAR.
- We have also received updated AMSR3 coefficient as well.
- I'm working on restoring or adding missing Jacobians from the land surface, ice surface and snow surface models. This is all in support of v3.2.0.
infra:
- CI image revision for spack-stack 2.0
- NCEP spectral transform “sp” library removed - has been deprecated and replaced with NCEP “ip” general interpolation library..
- GSIBec removed from the image CI configures jedi bundle wiht BUILD_GSIBEC=ON
- For unit tests needing gsibec, do this https://github.com/JCSDA-internal/saber/pull/1191
- Note: gsibec 1.2.1 still requires sp (although can build without it but will not work fully) we need to update to 1.2.3 to get a version that works with ip.
- See spack stack issue spack-stack/issues/1648
- spack-stack 2.1
- Target feature freeze tomorrow.
interfaces:
- prepping OSDF container for testing by partners
- adding new ctests to improve coverage
- implementing a writer to enable end-to-end workflows
- thanks UKMO for contributing OSDF append/reduce capability
- UKMO working on ODB reader for OSDF, will reach out to EMC soon for bufr reader
algo:
- PR (from Anna Shlyaeva) introduced a hybrid block chain which refactors the hybrid covariance in SABER into a more elegant, less hacky format. New yaml key will be needed when using a SABER hybrid covariance. Ricardo and I discovered this the hard way.
- Ensemble management in block chains was also refactored, possibly with some improvement to memory usage.
- Big extensions on the way for adding scale-dependent background error covariances on the way