Present included: Stephan Siemen, John Schattel, Joseph Matula, Adrian Custer, Marie-Françoise Voidrot

During the 2nd workshop on the use of GIS/OGC standards in Meteorology in Toulouse, the working group on the "Roadmap" stated that the MDWG should define an Interoperability Experiment in January 2010 and begin it in March 2010. (Report and planning available at : http://external.opengeospatial.org/twiki_public/bin/view/MetOceanDWG/MetocWMS_IE_Project )

The aim of this telco was to define the objectives of this IE.

The Hydro DWG has define an IE which can be a reference of work (link also available ont the same twiki page in the Resources part) Some use cases have been made available on the twiki (Airport and Landfall) or through the email list which are very different.

There are two types of use cases to be defined and this has to be clarified on the twiki and via the email list :

* There are use cases for the modelling thread that can be complex and very developped as the process will be long (in term of years and not weeks or monthes). and they will be structuring for the metadata schemas at WMO level.

* The use cases for the WMS best practices definition have to be more simple and to focus on short term tests over the proposals under validation on time handling. We have worked on the time issues, let's define small uses cases mostly dealing with time. (no problem of licensing of data or security, or whichever).

These use cases could cover each of the user profiles defined in the performance twiki page : http://external.opengeospatial.org/twiki_public/bin/view/MetOceanDWG/MetOceanPerformances

It could be interesting to have some more detailed information on the way the different implementations already available through the IE volunteerings have dealed with the time issues. (http://external.opengeospatial.org/twiki_public/bin/view/MetOceanDWG/MetocWMS_IE_Volunteerings). John will put some links to some US services available. Joseph says some are already available for IBL client.

For Numerical model output, one map is defined by th efollowing parameters :

• Type of the model (ECMWF, UKMet, NWS, ARPEGE French Model...)
• Resolution of the GRID (The same model can be provided with deifferent resolution depending on the area of the world, the field, the level...)
• The meteorological field (temperature, humidity, pressure ... but also accumulation of rain over a period...)
• the type of level (hPa, meters, Flight Level, Potential Vorticity, aeronautic level -Jet, tropopause...- ...)
• The value of the level
• the domain (in lat, lon, projection...)
• the Run and the time offset or the validity date

The IE should help us check if we can define a single best practices doc with a good selection of default vaues for the mass market or if we have to define two of them (one for mass market and one for professional use).

Everybody agrees on the fact that the first IE should be simple to be possible on a short time and open to all. Not too simple though because the simple test of combining a field from two different MWS servers into a single WMS client has been done already. (A space for Retex of this ype of tests should be feeded at : http://external.opengeospatial.org/twiki_public/bin/view/MetOceanDWG/MetocWMS_IE_Implementations

Marie-Francoise will announce the agenda for the next teleconf (probably IE or SLD as this issue will have to improve soon too)

-- MarieFrancoiseVoidrotMartinez - 27 Jan 2010