CRTM Technical Meeting Protocol
Core Topic of the Meeting: General Update
Date: 2020-10-19 Time: 15:04 h
Location: Virtual (Google Hangouts)
Invited Speakers: N/A
Meeting Chair: Patrick Stegmann (JCSDA)
Keeper of the Minutes: Cheng Dang (JCSDA)
Attendees: Benjamin Johnson (absent due to emergency JEDI meeting), Patrick Stegmann, Cheng Dang, Hongli Wang, Daniel Abdi, Yingtao Ma, Nick Nalli
Introduction by Patrick:
Ben cannot join due to an emergency JEDI meeting.
Agenda Item 1: | Cheng’s Update (Aerosols) |
Discussion: |
Update by Cheng: - CMAQ modifications are merged into the CRTM developing branch. - Both netCDF and binary IO are activated. - Preliminary sensitivity tests on CMAQ/GOCART tables.
|
Result: |
|
Tasks: | - Cheng: meet with Hongli on “how to select CMAQ aerosol look up table”. |
Responsible People: | Hongli: we proposed to finish the GSI by next year (4 months from now), so we are interested in if there is any plan to evaluate the CRTM-CMAQ AOD.
Yingtao: Do you have any AOD measurements on for model evaluation?
Hongli: VIIRUS, only retrievals. We have the error/uncertainty/bias of observation, but we hope the CMAQ simulation with CRTM does not have the bias.
Hongli: I also need help on developing the GSI-interface.
Cheng: we could set a meeting together to go through the process, and Yingtao is welcomed to join.
|
Deadline: |
|
Agenda Item 2: | Patrick: Update on CRTM core team development |
| - CRTM TL/AD CTESTS - Side note: Convergence tests may pass or fail depending on the compiler (pass for ifort and fail for gcc) - Transmittance coefficient package: VIIRUS comparison (data from the default, Yingtao, and Patrick) - Improvements for UMBC profiles. (In the troposphere, the difference is quite significant) - Possible explanations for the differences: mismatching training profiles (ECMWF83 vs. UMBC48)? SRF influence? No ODAS merge with ODPS for water line absorptions. Unknown unknows. Mismatching set of training absorbers. Different absorbers?
|
Result: |
|
Tasks: | - Check the influence of different absorbers - Include ODAS
|
Responsible People: | Yingtao: I used the UMBC48 profiles too. I cannot recall the details right now.
Yingtao: another feature you might want to check is the coefficient profiles (print in Fortran). I did that, and I cannot get always reproduce Yong Chen’s results.
Patrick: yes, I have done that, and usually the results do not match each other, and there are different responses for different channels.
Yingtao (shared screen): Figures on TauCoeff for Ch12 from g17_lbl_v11.7 @60K and Yong Chen’s original results. Do you see a similar result?
Patrick: no. I have not check the vertical profiles yet.
Yingtao: what’s your plan on this work after the quarterly review?
Patrick: Continue the transmittance package and ensure that it is in reasonable agreement with the previous coefficient sets from the trunk.
Yingtao: I contacted Zhenlong for his 4-day long data simulation.
Patrick: If you want we can schedule a separate meeting on the NLTE work.
Yingtao: Ah yes, the NLTE work, I haven’t continued working on that yet.
|
Deadline: |
|
Agenda Item 3: | Updates and questions by in-kind contributors |
Discussion: |
Nick: I have most research done, but still working with Jim on the assimilation side. What I need to focus on now is writing the paper. I’m waiting for more ship data to do another valuation on the model. For warm water campaign, the results look pretty well. I will probably give a 1-hour GPS-science seminar on results about that.
Daniel: I’m trying to get JEDI complied, but there are some difficulties with the shell scripts and packages. Do you have any suggestions?
Patrick: We are doing some modifications to the JCSDA repo, so some links might not be functional.
Daniel: I have JEDI-internal repo, but I still need more modules on EC-build and etc. Is this something I can get?
Patrick: Maybe you can get them on the git repo as well.
Daniel: I’ll look into that.
|
Result: |
|
Tasks: | - |
Responsible People: | - |
Deadline: |
|
Conclusion:
15:40h Final end of meeting.