Coordinated merges:
- YAML refactoring for `obsdatain/generate/obsdataout` sections in the `obs space` section: https://github.com/orgs/JCSDA-internal/teams/jedi-models/discussions/81 (merge on Friday 9/16)
- New rmsByLevel () function will need to be implemented on model side of Increment<MODEL>.h: https://github.com/orgs/JCSDA-internal/teams/jedi-models/discussions/80 (merge on Wednesday 9/21)
OBS:
Fabio added observations to the skylab experiment.
Lindsay working on extending plotting capabilities.
Francois V is working on CRTM2.4 for the release, updating the known good output in UFO to be ObsValue (not GSI H(x)) and adjusting the tolerance.
CRTM:
Patrick:
- Added the iasi_metop-c coefficients to the ufo testing tarball for Fabio
- Ongoing work on converting the CRTM coefficient generation package (CRTM_coef) to a bundle
- Continuing work on new JPSS-2 ATMS coefficients.Patrick: iasii metopc coeffs to the ufo testing data
Cheng: working on CRTM2.4.1 code structure and aerosol related tests for AOD changes
COMPO:
Maryam: working on ewok with COMPO, successfully running with geos-cf and tropomi and o2 data.
LAND:
Andy: since getting back have dealt interface changes for the WRF-Hydro folks due to OOPS PRs:
feature/flip_geovals #1846
feature/add_rms_by_level #1901
Have also tested and begun using spack-stack (GNU) on Cheyenne
SOCA:
Working on marine Skylab v2 in ewok. Rewriting operator for vertical interpolation to work with composite operator, converters for swot, and oasim.
ALGO:
New capability of using masks in the interpolation (contact Francois H for details if interested).
SABER code sprint: work on SABER blocks refactoring (removing templates, passing correct functionspaces and variables around); interpolation block; spectral covariance blocks split up.
Particle flow filter PR in oops in review.
EDA ens variance in the ewok suite.
Work in progress on ensemble and hybrid covariances in Var ewok.
Question on time interpolation for 4DEnVar: https://github.com/JCSDA-internal/oops/issues/1894
INFRA:
Finalized change in ioda that allows to write a single file.
New spack stack release: new pio, eccodes, pyeccodes - finalized next week and rolled out on all systems.
Containers (3 new containers: intel, gnu, clang) to replace CI and EDU containers. (Intel container is pretty big due to intel one api)
Parallel cluster runs on AWS: ATMS obs seem to be a problem (not reproducible on other machines).
R2D2 ingest: full fv3 C386 ensemble on s3, orion
EWOK: adding computation of ens variance in EDA.
R2D2: scripts for re-converting GFS backgrounds