Code management group minutes 24Oct2012

From UKCA
Revision as of 14:29, 1 November 2012 by Mdalvi (talk | contribs)

UKCA Code Management Meeting, 24th October 2012 (tele-conference).

Participants:
Met Office: Nick Savage (NS), Colin Johnson (CJ), Mohit Dalvi (MD)
Cambridge : Alex Archibald (AA)
Leeds : Graham Mann (GM)
Oxford : Zak Kipling (ZK)

Not present but on CMG; Sandip Dhomse (Leeds, SD), Luke Abraham (Cambridge, LA), Paul Telford (Cambridge, PT)

Summary of action points arising in the meeting

  • Carried forward:LA to check admin on UKCA email lists. NS to take this forward.
  • NS to circulate list of Somewhen tickets on UKCA
  • CJ to circulate provisional list of UKCA configurations at/ post UM8.4
  • ZK to request Philip to circulate information/ updates on JASMIN
  • MD to circulate link for external release of Iris postprocessing package (Offline discussion- will now be done by NS with more info on UKCA - Iris experience)
  • CJ/MD to port a UM8.2 TropIsop+MODE job to PUMA including latest improvements/ fixes.
  • NS to set-up doodle poll for next CMG

Status of previous Actions

  • GM to circulate minutes for approval and then send round full UKCA group. Done
  • GM to email out to gather info for science plan ahead of 11th Oct UKCA meeting. Most groups have responded. GM will collate into document by October end
  • LA to check admin on email lists & add extra person(s?)to cover whilst he is on leave Carried forward
  • NS to update on rationalisation of tracers at next meeting. Unused tracer removed in UM8.3. Will need to account for nitrate, seasalt, Na/Cl tracers in future
  • NS to circulate full ticket list for v8.4 code-freeze and put on PUMA UKCA Trac. Done. somewhen tickets to be circulated
  • NS to update on additional resource for Gmode nitrate implementation & testing. Request sent to Met Office/ NERC Exec, may be considered in relation to next ESM
  • MD to contact SD for other required bits of code/obs. Aim still to put together post-processing package by end Sept. Check again on GM IDL code and contact GM if still problem. Scripts from Alex and Fiona now being linked as package. Aerosol/ AEROCOM scripts resource-demanding and will need to be optimised, even converted to work from PP files. Targeted for November.
  • NS to circulate doodle poll for date of next CMG meeting around 25th Oct. Done
  • LA/PT, MD & GM to update standard release UKCA aerosol jobs in next few weeks. UM8.2 StratTrop job now avilable on MONSooN, TropIsop+MODE being transferred. UM7.3 Aerosol+RADAER+ACTIVATE begin tested. Actual science configuration of jobs will now be decided by the UKCA Board
  • NS to add “Tips & Tricks” page off front page of UKCA website. Done

Other Discussions

  • UKCA standard jobs: At UKCA Board meeting, three supported configurations agreed (Global StratTrop+MODE+RADAER+ACTIVATE; Global Offline oxidants+MODE; Regional AQUM). GM/AA inquired regarding N48 versions. Groups are welcome to set-up, evaluate and use any variants, but these may not be centrally supported or tested. For the medium-term, vn7.3 N48L60 and N96L63 likely to be used.
  • JASMIN: Philip has requested a joint proposal to exploit data storage and analysis capabilities. Colin to discuss with Philip and get more information on status of JASMIN, though this may have to wait till Luke re-joins.
  • CJ updated on progress with UM8.2 MODE configuration:
    • Aitken/Accum mode ratio still not ideal;
    • Primary emissions code now included as branch, diagnostics being checked;
    • Offline oxid code working but may need some re-organisation in view of retirement of ifdefs in future UM version;
    • RADAERv2 is tested and ready to lodge, possibly by Nicolas' replacement;
    • Plumbing of ACTIVATE to LS precip/ Radiation ready and targeted for UM8.4;
    • NoCLASSIC and Nitrate in MODE will have to be deferred to UM8.6.
  • Other priorities for UM8.4:
    • Updated reaction rates - may need to be evaluted for TropIsop separately.
    • Lightning NOx - will need discussion with Convection team to select the most appropriate input diagnostics.
    • Interactive BVOCs - actual computation should be released as part of JULES soon, will need plumbing into UKCA.
    • Use of NetCDF emission files to replace current system, will be lodged by Carlos. AA inquired regarding online remapping to model grid from a standard set. This will have to be evaluated vis-a-vis model timings.
    • Section 34 Tracer reassignment- can be possible if list is finalised soon and MD's other priorities, else deferred to UM8.6.
  • Iris: AQUM team are building a verification suite based on this. NS will demonstrate to UKCA CMG/ users once this is in a mature stage. MD (NS) to circulate information and external link

Date of Next meeting

Early December, NS to setup doodle poll