INFRA (Evan)

Pandora and our two capabilities for the obs in ... is merged.  Also, you can now build Jedi during a Skylab experiment by adding a line to your experiment YAML that just says, Build Jedi. Documentation is provided in Skylab experiments, readme with an example also in Skylab experiments, Jedi C test, yaml, which also hint there's in addition to building Jedi, you can run Jedi C tests with Skylab. on R2d2 we have this new authentication system that was necessary to make the web API work. we're still targeting spec stack 1.9 for the end of the year.

Algo (Christian)

Clementine has created localizations with bump for the ensemble B matrix for her geos experiment.
She also has first results with an EDA of 32 members, and is now working on cycling with Ricardo's help.
Christian is also starting to test added abilities to shift the end of the DA window and append observations within the new time window while finalizing some design choices for this particular continuous DA mode as well as making some improvements in how observations are appended between outer loops to ensure consistency.

Steve H.  has added the ability to add a "receipt time"  to observations with the IODA converters which we will use in continuous DA testing.Nate has PRs in Ewok and Skylab for an experiment that does multi-scale/spectral localization with a low-resolution MPAS ensemble.

Interfaces (Francois)

- For the continuous DA application, we have developed a standalone filtering application for ioda obs files. This currently includes a time window filter and an optional receipt time filter.
- NOAA-EMC contributed a bug fix for a file name collision issue when running the BUFR reader with multiple MPI tasks.
- The MetOffice contributed  ObsSpace utilities that list out all of the groups or all of the variables in the ObsSpace data container.
- Successfully handed off the ObsSpace Data Frame development from Phil U to Steve V. One result of this effort is that we have received an excellent set of documentation from Phil.
- Have begun implementing incremental steps in IODA toward the goal of being able to test the new container alongside the existing ObsGroups.
- Did some cleanup in VADER
- Continued work on enabling GPU support for interpolation in JEDI/Atlas. Started investigations into ecTrans to identify work needed to enable GPU support (beyond what  ECMWF has already done.

SOCA (Travis) 

Kriti continues Working on hybrid, TLM. Hamideh is working on coupled oasim hofx. Travis is working on a secrete side project he will discuss it later. it is about automation of tutorials testing.

Compo (Jerome)

Here are the updates for COMPO:

• 1 week of the 4DEnVar experiment with 8, 16, 24 members is finished. We will run 32 members experiment next. 

• All the observations for the experiment period are now on R2D2 (Thanks, Eric and Ashley)

• Adding/updating compo documentation:

• Column retrieval 

• PM 

• MODIS AOD converter is up-to-date and thanks to Ashley the r2d2 ingest suite is implemented. PRs are waiting to merge.

Obs (Ben R, Hui, Francois)

ufo:

• - PR#3541: superobbing PR from UKMO  interesting update allowing superobbing for those observations that do not require H(x)

• - PR#3440: generalization of a surface data operator - UKMO introduced a surface-p corrected, NOAA introduced an extension for surface-t these are being combined thanks to Hui Liu and Mike Cooke

• - PR#3547: Jianjun from NOAA has been introducing some elements for all-sky treatment particularly with CRTM which will allow reproduction of GSI and hopefully more flexibility to experiment and expand on the surfaces over which all-sky is implemented.

• ozone assimilation at NRL can now successfully produce H(x) from NEPTUNE beginning implementation of methodology

converters:

• - PR#1573: compo team added update to the pandora network converter, being used in the TEMPO validation, adding a site classification

• - PR#1576: ground-based TEC converter for the old TENET format, new format is eminent in the new year

• - update to TROPICS converter regarding quality flags

• - PR#1587: working with NOAA and USAF partners, have converter update for tomorrow.io sounder both L1-TB and L1-TBR and Muon Space OSW L2 and L3 products

space weather:

• IODA observation conventions have some updates to accomodate the new observation types to date, ionosonde, space-based TEC and ground-based TEC

• GeoVaLs conventions for space weather models is underway with naming conventions being introduced there as well

CRTM (Ben J)

v3.2 status: expected Dec. 23.  Enable default netCDF for all files inputs/outputs.

Working with commercial vendors, Tomorrow.io, on coefficient generation / testing (BenR mentioned)

Continuing collaboration with international NWP centers on accelerating the use of visible radiance / reflectance data use in data assimilation, with a particular focus toward operational NWP capabilities.  Workshop in early March, Frascati Italy on this topic.
 
Visibile Radiance assimilation progress has been made on the CRTM and UFO side, but still need scientific updates on surface / cloud reflectances.

Cheng continues her collaboration with the PACE project.

EMC (Dan)

• Workflow changes to run nightly experiments against develop versions of JEDI. Added regression to the some JEDI Var tasks.
• Changes to History IO in fv3-jedi.
• Preparing B matrix PRs. Adding regional GSI-bec and MGBF to Saber. With testing in maps-jedi and fv3-jedi.
• Adding filters to UFO to handle screening of clouds in all-sky assimilation over land.
• Changes to SOCA to allow Arctic and Antarctic sea ice to be handled in one task.

NASA (Ricardo)

Ricardo was making changes in SWLL so it is up to date with the variable name change code sprint. Ricardo found out than GNSS-RO is not Handled correctly. which is the fact that the 3d pressure is not being handled quite correctly in the linear variable transform.

Navy (Sara)

ongoing work with ozone, GNSS-Ro  assimilation. There's a group of us working on new LETKF type solvers, as well as other LETKF applications. They have a new hire up at NRL DC, working on the LETKF.

MMM (Jake)

1. Andy Stokely is updating Fortran-based obs2ioda conversion code to
    write out ioda-hdf5 format data, which can be directly ingested into JEDI.
    Current obs2ioda writes out the older version of ioda file, which needs to
    go through additional ioda-upgrader steps to be assimilated in JEDI.

2. Tao Sun is implementing the linear version of LETKF for MPAS-JEDI.

MetOffice (Tom Hill)

• UFO PR #3519 adds the ability to save the MPI rank for each observation into the ObsSpace.  Which has been added to aid debugging. 

• Several PRs added into UFO to deal with issues where one of the MPI ranks has zero observations.  Ongoing work to find and fix other functions. 

ROMS (Hernan)

• 4dvar is now working with roms-jedi. overall is working well

OAR (Ming)

• gsl, regional hybrid-3denvar