BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Understanding variability and temporal trends in biosphere-atmosph
 ere CO2 exchange through integrating models with data - Dr. Trevor Keenan\
 , Harvard University
DTSTART:20120319T100000Z
DTEND:20120319T110000Z
UID:TALK36953@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:Process-based models of atmosphere-biosphere interactions\, al
 ong with empirical data mining techniques\, are the primary tools used for
  scaling disparate observations through space and time. In the past few de
 cades they have been developed in tandem with our understanding of ecologi
 cal theory\, resulting in models of various levels of complexity and detai
 l. Model intercomparisons\, however\, show a large range in model performa
 nce\, with no clear consensus as to whether model structural error (proces
 s mis-representation) or mis-parameterization is to blame. One potential r
 eason for this lies in difficulties in using data sources at different sca
 les to adequately test model performance and the common reliance on uni-va
 riate model validations. Another is the lack of quantification of the unce
 rtainty associated with model projections. In order to advance our ability
  to make policy-actionable projections of the future evolution of the eart
 h system\, we need to address these issues.\n \nIn this talk I will assess
  our ability to model land-atmosphere CO2 exchange at different spatial an
 d temporal scales\, both in the present climate and under future climate c
 hange. The analysis makes heavy use of model-data fusion techniques\, whic
 h constitute a powerful framework by which to combine models with various 
 data streams. Model benchmarking tools\, such as empirical data mining tec
 hniques\, also provide a strong alternative model evaluation. To illustrat
 e the potential benefits of such an approach\, we assess the performance o
 f 17 process-based models of atmosphere-biosphere interactions\, and two d
 ata mining tools\, across 11 long-term eddy covariance forest sites. The r
 esults highlight details of model performance often overlooked by conventi
 onal model-data comparisons\, and quantify the degree of coupling of terre
 strial carbon sequestration to climate anomalies at multiple sites and tim
 e scales.\n
LOCATION:Large lecture theatre\, Microsoft Research Ltd\, 7 J J Thomson Av
 enue (Off Madingley Road)\, Cambridge
END:VEVENT
END:VCALENDAR