Minutes of Telecon 22 August 2011

Attendees

* Jon Blower (University of Reading) * Frederic Guillaud & Marie-Francoise Voidrot (Meteo-France)

Discussions :

1) Met OCean WMS Best Practices draft on Time handling for climatological and climate change data

Based on the statement that we should not use things defined in WMS for things they have not been designed for, Jon has convinced the other participants that we should not put into the TIME parameter semantics and tricky interpretations that the clients will not be able to understand. Thinking that if we avoid specific DIMENSIONS we will be more interoperable with general purpose clients is an illusion. The TIME dimension is fine for observations and forecasting data but cannot bring all the complexity of the climatology and climate change data. If we really want to avoid Dimensions, Jon suggests that if there isn't too many climatologies , we should use one layer for each one. Forecasting outputs really requires Dimensions as there are thousands of combinations parameters and time steps.

Frederic has submitted a proposal by email today based on the definition of dimensions and referring to a lot af existing works. This paper could be discussed in the teleconf of the 23rd next week.

Action : Meteo-France will look for infos around the number of climatologies.

2) Layers naming

Jon informs that a mapping between Cf and GCMD has been done already.

3) Next telco

The next teleconf will be on August 23rd.

-- MarieFrancoiseVoidrotMartinez - 30 Aug 2011