Skip to content

Teleconference 6.4.2022

Marijana Crepulja edited this page Apr 7, 2022 · 4 revisions

Agenda

Discussion of the options, and determining best practices, for updating existing BUFR Tables D and B. This will help us to address issue #106

Discussion

Options

1. Update existing BUFR decriptor

  • Impact assessment
    • How many sequences are using descriptor?
    • Are there data already exchanged using that descriptor?
    • Is there impact on existing software?

2. Update existing BUFR template

  • Impact assessment
    • Are there data already exchanged using particular BUFR template?
    • Is there impact on existing software?

3. Generate new BUFR template

  • Impact assessment
    • There will be no impact on existing software.

Procedure of inclusion

  • Do we introduce testing (validation) period for member states to test their BUFR data handling software?
  • How do we inform, do we circulate email to focal points and ask for feedback? How long will we allow to receive feedback? Could it be the validation period we have within our FT procedure.
  • What if we get feedback that some countries are not able to use the data with the change?

Summary of discussion

  • Three solutions to issue #106 were discussed.
    1. Update of descriptor B 0-01-087 to increase the bit width
    2. Update of sequence 3-15-013 to include 2-02-129 change data width
    3. Create new sequence based on 3-15-013 and including 2-02-129
  • Sibylle and Marian explained some of the issues with updating the sequence (option ii), with the discussion focusing on the principle of allowing the modification of existing sequences. The main concern was that the impact on users was unclear, with existing software relying on the position of elements within a sequence. This reliance was partly due to optimization of the software for decoding BUFR but also due to outdated software.
  • It was noted (Jeff, Simon) that this issue was the perfect case for allowing changes to a Table D sequence given that the sequence was not yet in use but no consensus could be reached. The concern was that this would set a precedent and open the possibility of changing other Table D sequences in the future.
  • Option i was also discussed, however similar issues exist. The impact of modifying 0-01-087 is not known as the descriptor is already in use and there may be unintended consequences. However, it was noted there will also be similar issues with other marine platforms using 0-01-087 using identifiers beginning 99*****.
  • Option iii was decided as the simplest option. Jeff and David agreed to communicate the proposed solution to the impacted data providers / community.
  • In addition to option iii. the original solution remained a potential solution, with the element 0-01-087 set to missing and the station identification included in the WIGOS ID elements/section.
  • David questioned whether a new regulation was required making it clear that descriptors (Table B and Table D) could not be changed after their validation.
  • Following discussion David agreed to raise this topic with the Secretariat and to investigate a position paper proposing this.

Decisions

  • option iii. selected as the best option for issue #106

Actions

  • Jeff to communicate the decision with the affected parties. (post meeting note: completed 06/04/22)
  • Jeff and David to work with the parties to propose and validate the new sequence if required.
  • David to internally discuss the addition of a new regulation to the MoC making it clear that descriptors could not be changed after first introduction. (post meeting note: discussion within the Secretariat (06/04/22) - the regulations governing the changes are already in place. Additions to codes tables can be made via the fast track process but other changes, such as changing existing descriptors, need to be made via the complex procedure. Changing existing descriptors is not ruled out but the process is slower.)

Participants

No. Name Agency Apr. 06
1 Marijana CREPULJA ECMWF X
2 Abderrazak LEMKHENTER Météo Maroc X
3 Fang ZHAO CMA X
4 Sergio Henrique Soares FERREIRA National Institute for Space Research, Brazil X
5 Jeffrey ATOR NOAA X
6 Sibylle KREBBER DWD X
7 Sébastien VILLAUME ECMWF X
8 Simon Stuart ELLIOTT EUMETSAT X
9 Marian MAJAN HMEI X
10 David BERRY WMO Secretariat X
Clone this wiki locally