2016-07-07 Per Undén


UERRA MST meeting no 13, 6 July 2016




Per Undén, Manola Brunet, Andrea Kaiser-Weiss, Richard Renshaw, Else van den Besselaar


Progress reports updates from the WPs

Periodic report status

Deliverables' status

Copernicus Climate Change services and UERRA approach

Conferences: Abstracts to EMS/ECAM in September, other meetings.

UERRA Meetings, GA and showcase WS in November

AOB, Other meetings and communication, replacement of an EU DG CLIMA ESAB member

Next MST meeting

Progress reports in the WPs


The focus is on the quality control of the 8.2 M rescued station values. It is really resource consuming but progressing well and about 50% is done. It will be finished at the end of July or early August. Deliverable D1.5 is consequently delayed to September as notified before. There are large data sets available from other digitised sources like Catalonia, Norway and Sweden, some 114 M values. They are quality controlled and will be used in the spatial homogeneity checks.

The gridding resolution assessments at KNMI will be submitted (D 1.11) in July.


The ensemble assimilation at the Met Office is running now for 2008 and the year will soon be available. It had to be rerun due to an error in the SST. The SMHI HARMONIE runs are progressing well but a lot of computer resources are used (SBUs) but it is estimated to be enough. A special project is used as well, as a small addition of resources. There is work on the preparation of MARS archiving as it is urgently needed.

The MF downscaling reanalysis is now running and an ensemble is created. A paper about the precipitation reanalysis has been published in Tellus.

The cloud fraction analysis at SMHI is ready to start running. Questions about which final resolution to project it at has been addressed and 5 km does not add information and is prohibitively expensive so 11 km is the most likely choice. The overlap of cloud data with model background was another issue.

At University of Bonn Lilo is running the 5 years but is still with 2008 as there have been many technical problems with the handling of boundaries. Then there are changes due to the new Cray processors.

Previous action: The priority of which year (2008 or 2009) to run first is 2008.


WP3 is working with the precipitation evaluation (representation of extreme events, Gerard) and other aspects of the precipitation validation. DWD is working with the radiation against satellite and about instantaneous vs integrated values. The diurnal and annual cycles are studied. We will come back to the winds once we have access to the WP2 data (MARS).

Due to the availability of the reanalyses in MARS, Andrea gave advance notification of a possible delay in D3.5 due in October.


The MARS work at ECMWF is now very critical for the UERRA RA as SMHI is in production. The tools and grib_api are now ready for conversion and archiving in GRIB2 and it is waiting for the data samples to be in correct format. Scripts have been provided but there is some work to get everything out correctly as defined.

The ESGF system is now working at KNMI and it is ready for UERRA data to be put there. On should contact Wim Som de Cerff (and at the CLIPC This email address is being protected from spambots. You need JavaScript enabled to view it. mail address ) when you want to put data in there.


KNMI will now participate in and disseminate material at an African Workshop in Angola 28 November to 9 December, arranged in the context of GFCS end ETCCDI.



No new activities.


The contacts through tele conferences between the 5 projects took place 16 June. There was information about the projects that were present (2 missing). The Lessons learned is still not finished due to some not submitted their reports.

Periodic Report status

The inputs have been provided by the WP leaders and will be edited and compiled by the Coordinator and sent out for comments before August (to be completed in August).

Deliverables' status

D 1.5 Dare quality control. It will be delayed by 2-3 months (from M30).

D 1.10 (UEA) and 12 (KNMI) delivered (

D 1.11 Gridding resolution assessments. Due M30 and will be done in July.

D 2.3 Ensemble variational DA diagnostics delayed until September.

D 2.6 HARMONIE initial production. M30. Diagnostics will be carried out for some of the data and a report written but somewhat delayed.

D 2.9 MESCAN ensemble. M30. It will be ready in September.

D 2.10 MESAN cloud analysis. M24. Delayed some months into or the end of summer.

D 7.9. African Workshop. Delayed but will be done in connection with other activities in the late autumn.

D 9.3 Lessons learned. M24. In process but not all contributions yet.

Copernicus climate change services, plans

The UERRA pre-operational service is employing multi-model ensemble in addition to individual ensembles. It was not a clear conclusion about the multi-model approach at the Copernicus RRA WS last May and ECMWF did not seem so convinced and there is a limited money stream of 6.5 M Euro (possibly a bit more). It should still be possible to run a similar service in C3S bearing in mind that the cost of UERRA is 7M and a bit over half is for the RA/WP2, but in C3S one needs to cover the computing costs as well. We believe it is important for the uncertainties to have several models in the service. It remains to see from the evaluation of ensembles from each system vs the uncertainties comparing each system and only partial results will be seen in the autumn.

There is also a benefit of having the main modelling groups in Europe involved in the reanalyses and benefitting from developments everywhere. One can compare and should look at other efforts in multi-model products lige TIGGE and CMIP5, CORDEX, to see the benefits and to look at how products can be created. It is unclear what sort of products will be required in C3S, if a combination or uncertainty ranges can be made from ensembles and multi-model ensembles.

It will be further discussed at the EMS meeting, informally, and we need to prepare some scenarios before the call and probably meet to discuss it (the GA in November is convenient (Per's comment) but maybe too late?). MF is very interested to continue the downscaling analysis ensembles coupled with the regional upper air RA from SMHI as in UERRA but it is dependent on good observations. There are a number of countries interested and planning or doing high resolution sub-regional RA, like the Alps or Scandinavian mountains and it was discussed at the WS. It is not feasible to produce a pan-European RA at 2.5 km whereas there are many groups who would like to do it over a small area. It was concluded at the WS that one should stick to about 9 km for the whole domain. Also, it is not feasible with the money stream and from aspect of creating pan-European products to try to involve many more partners doing sub-regions.

Meetings and communication

ACRE climate meeting in Ireland, July (KNMI and URV participation) (see also above).

This was to a great deal oriented towards answering to the Copernicus Observation rescue call. Manola informed about the issues and the meeting pointed out that there are serious data gaps in the 60s and early 70s in MARS for global reanalyses but there is very little money for actual Data rescue. More about coordinating and surveying other efforts. The call is moreover focussed on global data and not on the regional issues. The main players in the field were present. A big issue is still access to data and to have more open data.

EMS/ECAM 12-16 September, Trieste. Many of us will attend and from the MST. We will have a small side meeting for the MST and for the C3S. Per will have two presentations.

NMM Nordic Meteorological Meeting, 22-24 August, Stockholm. Per will present UERRA (and J-N Thépaut is on the speaker list)

REA will organise a collocation workshop with the Projects 28-29 September


Soci, C., Bazile, E., Besson, F., & Landelius, T. (2016). High-resolution precipitation re-analysis system for climatological purposes. Tellus A, 68. doi:http://dx.doi.org/10.3402/tellusa.v68.29879

GA eo 2016 and Showcase event and Project review meeting(s)

There will be a session for the review or a review meeting with our PO will be on September 7. for the Coordinator and WP leaders 2 and 3)

The next GA is announced 21-23 and the showcase WS 23-24 November.

There is some concern how much will be available for November, but ECMWF sounded quite hopeful for the MARS developments and not it is mainly the question to populate the archives from the UERRA runs. It is a showcase event, i.e. we will demonstrate new products and not the final ones. The advantage with the location at ECMWF is that there is a good potential to reach many users. It is still felt that we shall pursue with the WS. To postpone is is still possible although it is tight with meeting facilities if at ECMWF and moreover, the co-location with the GA is an advantage. Next year we will anyway have two WSs , one second WP8 and one final WP3 and both oriented towards products and users.

We will still have a final GA next year, although not strictly necessary since the budget in the DoW did not have more than 4 GAs. It is useful to sum up the results at the end.


We have now got a positive response from our contacts in DG CLIMA to replace Alfonso Gutirtez. It is George Paunescu who has accepted to become our new ESAB member.

Next MST meeting

Probably in connection with the EMS meeting

Action points

No particular ones.