... personal wiki, blog and notes
Bryan's Blog 2009/07/08
UM Crack Cocaine
One of the items of discussion at yesterday's NCAS advisory group was whether or not NCAS should be working with more than one mesocale model (currently we do, with significant efforts invested in both the Met Office Unified Model and the U.S. WRF model).
The issues on the table included:
Can we support the model users for two models? Answer: yes, because we (NCAS CMS) support the academic Unified Model users, and NCAR support the WRF users, so WRF requires relatively little NCAS support.
Does the use of WRF contribute towards numerical weather prediction (NWP) (as opposed to understanding of weather processes)? Answer: not directly, in the UK! But we're told that because WRF is currently "easier" to use than the UM, it's easier to build a wider community of modellers, who one might then move onto the more complex environment of the Unified Model, and then contribute directly to UK NWP improvements. (And, in any case, the NCAS role in this area is really about "basic underpinning science" not about direct measured improvemetns in NWP, which in the UK is primarily a Met Office issue).
I just couldn't help feeling that WRF was being seen as the entry level drug for the academic UK NWP community to the UM crack-cocaine. But maybe I was just looking for anything to enliven two days of committee meetings ...
by Bryan Lawrence : 2009/07/08 : 1 trackback : 4 comments (permalink)
cf standard names growth
Adequate descriptions of scientific data depend on precise descriptions of what the data actually are. At the heart of that are what we call "phenomenon descriptions", which at BADC and in much of the climate community, we handle using "CF standard names".
Alison Pamment here at BADC is the international CF standard names manager. Today she posted an update to the CF standard names list, and buried in her email to the CF list was this:
The current version is now version 12, dated 6 July 2009. This has been a very large modification to the table - 804 new names, 67 aliases, 19 clarifications to definitions and 3 modifications to units - almost doubling the length of the table to just over 1900 entries.
This specific large update is part of the preparation for CMIP5, and represents a huge amount of work. We don't do nearly enough to recognise this sort of essential underpinning work, without which we couldn't build automatic tools to find data ... and without which, scientific users couldn't have confidence in knowing exactly what a particular variable measured (or simulated). Well done Alison, and all the folk who worked hard proposing and defining the names!