UERRA MST meeting no 10, 18 November 2015




Per Undén, Manola Brunet, Andrea Kaiser-Weiss, Richard Renshaw, Albert Klein Tank




Progress reports updates and next steps in the plan

Deliverables' status

Copernicus Climate Change services and UERRA approach, white paper (15 min)

GA 2016 and project review meeting(s), participation

UERRA User Workshop

Other meetings and communication

Next MST meeting


Progress reports and next steps


The Data digitization and QC has continued and a large number of additional data will be ready at the end of year (from 6M before to 9M). In addition URV is working on filling data gaps for the RA in WP2 over Sweden and Norway in the 60's. These are already digitized data that are extracted with a view to make synoptic observations from the station records and then to encode them in BUFR. There is significant effort to combine the station records into observations of all variables into something that looks like a SYNOP or CLIMAT and then to do the BUFR coding. MF has done the BUFR encoding for French data so they can be used in the RAs and this has now been proven to work. We will ask and expect that the actual BUFR coding can be done in collaboration between URV and MF.

It is important that the rescued and quality controlled data are passed on to the international data bases as well and it is confirmed that this will be done when all the data are processed and ready.

Action point: URV will send the rescued and developed data when they are all ready to the international data bases (ISPD and ISD, maybe more?)

Linden Ashcroft is extracting the Swedish observations and has investigated and prepared for the Norwegian data and scripts for access will be made.

For the uncertainty work and gridding enhancements, Richard Cornes has moved to KNMI and is working on these issues, partly with the previous tasks from CRU in combination with the KNMI work.



SMHI has started the long RA s from 1960 but there are sometimes problems with some areas of zero humidity which were found in the 5 year runs with ALADIN only. It is being investigated together with attempts to bring in additional French observations in various BUFR formats.

The cloud fraction reanalysis system has been developed and tested with structure functions and for 22 km before doing the analysis at 5 km. The SEVIRI period is fine but before that (MVERI and NOAA) data has no cloud mask (only albedo) but the CM-SAF will soon provide one that can be used. It is proposed to start with the second half of the period and do the early one when the cloud mask is there. The deliverable will thus be delayed but a few months is not a problem for the evaluation in WP3. The deliverable report will also be put back to when the work is done and not to do extra work on a preliminary version (Per will communicate with the PO and provided that this is agreed)

The Met Office RA system is technically nearly complete at ECMWF now and the TOVS data are now possible to assimilate. The rejection list system is also in place. The first ensemble test (as in the Deliverable 2.1 report on the web site) for one month has been done but shows problems in the ensemble spread. The lateral boundaries have to be provided form a global system and there may be problems also with the lateral boundary conditions being used. It is ongoing work and under investigation.

The first tests with the hybrid combination of the COSMO LETKF with ensemble nudging at UBO showed problems in loss of humidity. Apart from the need of more ensemble boundary conditions, there was a technical problem in this combination of systems obliterating part of the observations. In a recent teleconference Richard and Per agreed with Jan Keller that the work with the LETKF would continue for the rest of this year, but after that when Jan is not available the production of the 5 years may be done with the ensemble nudging if the LETKF hybrid is not ready. Then that work will be resumed after Jan's leave.

At MF Rachid Abida has started work in the Project and has addressed the problem of the data gaps over France. The French climate stations have been encoded into BUFR and various versions tried in order to assimilate them in the SMHI and MF systems.



The WP3 work is going ahead as planned. The evaluation work and system has been applied for winds and temperatures with station and tower measurements. The skill scores based on contingency tables have been developed and everything is in the R scripts and on the git hub. Gerard van der Schrier has worked on Climate Information from the RA. Cristian has worked on PDF based verification and will implement this in a coming visit to DWD 14/15 January.

The git hub is shared with everybody working in this area but the software and information can be shared more generally in UERRA. This will be done in the form of a table summarizing the procedures and experiences (D3.4).

Most of the evaluation has so far been done with the readily available COSMO RA but the EURO4M 2008-2009 data in MARS will soon be accessed. The COSMO RA will be transferred for archiving very soon. Per has had a telecon with Manuel and Richard at ECMWF and the first test data from UERRA should be available early in January.

The evaluation with satellite data has not started yet but is a next step to do in the spring.

In 2016 it is planned to include and demonstrate some UERRA RA data. So far COSMO and global RA have been used and soon EURO4M. A demo will be made for the User WS.



The Data plan in collaboration with CLIPC has now been written (D4.2) and will be ready before end of December.

The ESGF systems are still down due to the security issue, so there has been no progress possible.

For archiving in GRIB2 in MARS at ECMWF, most of the parameters have been agreed on and defined. A few need WMO definitions and ECMWF will submit them through a fast track system expected to be done within a month. Meanwhile the rest of the parameters can start to be archived, once the WP2 providers can extract exactly the agreed parameters from their test files.



There has not been any clarification or advise yet how to do the EU-briefs in the DoW.

Action: Per will contact our PO again.


See also under the User WS. The invitation and programme have been prepared and discussed this week and will soon be sent out.


The contacts through tele conferences between the 5 projects will resume. The common web page is probably satisfactory, TBD.


Deliverables' status

D 1.3 DARE data gaps pre 1950. M24. Will be done in December.

D 1.9 – E-OBS impact of data density (M15). Delayed and KNMI wishes to coordinate this with other work and a paper and is expected for December.

D 1.10 and 12 (Gridding improvements and resolution enhancement) (M24). These are expected to be a little delayed as explained before.

D 2.1 – Ensemble variational DA development (M21). Done and delivered.

D 2.2 Ensemble variational observations. M24. This will be done on time.

D 2.5 – HARMONIE physics ensemble (M12). Substantially delayed but the runs are soon completed. The data extraction and the analysis of results and a report is worked on and should be ready before end of the year.

D 2.10 MESAN cloud analysis. M24. Delayed some months into 2016 (for the SEVIRI period).

D 3.4 Evaluation experiences. M24. Is being done and in time.

D 4.2 – Being done and ready for the end of December (M12)

D 7.3 – M18. EU brief. Albert is prepared to do this but we are lacking a good choice of venue and event and Per is consulting our PO for advise.

D 8.2 – User Workshop I (M24). The planning is ongoing and a draft programme exists. It will soon be announced and Per will include this at the EMS/ECAM meeting.

D 9.2 – Web portal : Common web page. The common web page under clipc.eu will be submitted as a deliverable by CLIPC (although it appears in all 5 projects). (M6).

D 9.3 Lessons learned. M24.

Copernicus climate change services, plans and white paper

Richard has written a paper outlining the ideas and with many questions to be answered. It may be difficult to ask ECMWF but there is a Q/A page on the Copernicus web site. We hope that a coming Workshop will be held to further define things. A side meeting will be organised at the GA to discuss further.

There were severe concerns at DWD about the data policy of Corpernicus and this was discussed. Per and Richard will check with their representatives. (At SMHI there does not seem to be any such concerns and the data policy should be open).

GA 2016 and Project review meeting(s)

There will be a session for the review or a review meeting with our PO like at the end of the Tuesday's proceedings.

N.B. there is no Period Report now, it is due for 1 July 2016.

UERRA User Workshop

See above under WP8.


Other meetings, publications, dissemination and communications


Next MST meeting

In connection with the GA.

Action points

1. URV will send the processed data to the international data bases when they are complete.

2. Per will ask Monika again for advice on the EU brief(s).