BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Unnatural Complexity: Towards Achievable Aims for Complicated Simu
 lation Models in Applications with Data (Weather-like) and those without (
 Climate-like) - Leonard A. Smith\, London School of Economics\, London\, U
 .K.
DTSTART:20070813T130000Z
DTEND:20070813T134500Z
UID:TALK7793@talks.cam.ac.uk
CONTACT:Nick Watkins
DESCRIPTION:Complexity in Nature is seen in the behaviour of systems often
  viewed as having many interacting parts\, while the success of scientific
  models is most apparent in systems which have been isolated as much as po
 ssible. Natural complexity flows from the seamless interaction of these pa
 rts as a single whole.  While state-of-the-art scientific simulation model
 s are\, arguably\, unnaturally complex.  Rather than aiming to simplify ei
 ther the system studied or the mathematical framework employed\, various i
 mportant bits of a natural system are modelled explicitly. These component
 s are then bolted together\, along with an assortment of parameterisations
  to handle those other known aspects of the system which are not "resolved
  explicitly".\n\nHow can these simulation models be gainfully employed to 
 address questions of decision-support in high-priority areas of Earth syst
 em science? And how might the dialogue between model and observation be im
 proved to speed the advancement of both? Can we find a more coherent appro
 ach interpreting data in a field often characterised by what it is not (no
 n-linear\, non-Brownian\, non- Gaussian tails\, ...)? In particular\, how 
 might we develop faith that these unnaturally complex models can extrapola
 te beyond observed conditions to provide a "comprehensive\, physically con
 sistent\, prudent projection" of future likely conditions? Does Bayesian e
 mulation provide a decision-relevant alternative to naive realism? Do our 
 simulation models have a well-defined mathematical target?  Do we have gro
 unds to believe that unnaturally complex models will be structurally robus
 t?\n\nThese questions will be addressed within the framework of nonlinear 
 dynamical systems. While modern computing power allows us much more compli
 cated models\, similar issues concerned Maxwell and Poincare. We will ask 
 what achievable aims we might target given the mathematical nature of our 
 models\; weather-like applications where we have a historical archive of f
 orecast-observation pairs will be contrasted with climate-like application
 s where by definition no such archive exists. Resource allocation (in term
 s of experimental design) for model improvement will be contrasted with th
 at for decision support.  Our ability to forecast\, and in particular to c
 onstruct decision-relevant probability forecasts from ensembles of simulat
 ions\, will be examined.  Probabilistic Similarity and Shadowing are sugge
 sted as primary tools to evaluate the decision-support relevance of ensemb
 le simulations with imperfect models.  Applications in weather forecasting
  and climate modelling are contrasted with those involving Near-Earth Obje
 cts to suggest necessary conditions for deploying probability forecasts (a
 s such) from imperfect models in practice.  This appears to be somewhat mo
 re difficult than using state-of-the-art simulation models to improve our 
 understanding of natural complex systems.\n
LOCATION:Law Faculty\, Cambridge
END:VEVENT
END:VCALENDAR