BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Rewarding strength\, discounting weakness: combining information f
 rom multiple climate simulators - Chandler\, R (UCL)
DTSTART:20101207T163000Z
DTEND:20101207T170000Z
UID:TALK28308@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:Although modern climate simulators represent our best availabl
 e understanding of the climate system\, projections can vary appreciably b
 etween them. Increasingly therefore\, users of climate projections are adv
 ised to consider information from an "ensemble" of different simulators or
  "multimodel ensemble" (MME).\n\nWhen analysing a MME the simplest approac
 h is to average each quantity of interest over all simulators\, possibly w
 eighting each simulator according to some measure of "quality". This appro
 ach has two drawbacks. Firstly\, it is heuristic: results can differ betwe
 en weighting schemes\, leaving users little better off than before. Second
 ly\, no simulator is uniformly better than all others: if projections of s
 everal different quantities are required the rankings of the simulators (a
 nd hence the implied weights) may differ considerably between quantities o
 f interest.\n\nA more sophisticated approach is to use modern statistical 
 techniques to derive probability density functions (pdfs) for the quantiti
 es of interest. However\, no systematic attempt has yet been made to sampl
 e the range of possible modelling decisions in building a MME: therefore i
 t is not clear to what extent the resulting "probabilities" are in any way
  relevant to the downstream user.\n\nThis talk presents a statistical fram
 ework that addresses all of these issues\, building on Leith and Chandler 
 (2010). The emphasis is on conceptual aspects\, although the framework has
  been applied in practice elsewhere. A mathematical analysis of the framew
 ork shows that:\n\n(a) Information from individual simulators is automatic
 ally weighted\, alongside that from historical observations and from prior
  knowledge. (b) The weights reflect the relative value of different inform
 ation sources for each quantity of interest. Thus each simulator is reward
 ed for its strengths\, whereas its weaknesses are discounted. (c) The weig
 hts for an individual simulator depend on its internal variability\, its e
 xpected consensus with other simulators\, the internal variability of the 
 real climate\, and the propensity of simulators collectively to deviate fr
 om the real climate. (d) Some subjective judgements are inevitable.\n\nRef
 erence: Leith\, N.A. and Chandler\, R.E. (2010). A framework for interpret
 ing climate model outputs. J. R. Statist. Soc. C 59(2): 279-296. 
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR