Cloud Feedbacks



Other MIPs



Google Calendar

Lab Calendar

Site Map


Privacy & Legal Notice

Thanks to Our Sponsors:

PCMDI > WCRP CMIP3 Model Output > Diagnostic Subprojects Printer Friendly Version
<< Back to WCRP CMIP3 Subprojects

Attribution for 20th Century Regional Climate Trends

PI: Arun Kumar
Institution: National Oceanic and Atmospheric Administration
Additional Investigators: Martin Hoerling, Peitao Peng, S. D. Schubert
Abstract: Climate varies on multi-decadal time scales, in many instances due solely to the strong nonlinear dynamics of coupled ocean-atmosphere interactions. The suggestion for a climate trend can be easily rendered when examining only a portion of such low frequency swings, when in fact no drift is occurring. A new attribution challenge exists in todayís climate system however, where atmospheric chemistry is known to be changing due to human activity, and true climate trends are anticipated. On regional scales, differentiating the intrinsic low frequency variations from ones that are forced is a key to knowing whether climate is indeed veering toward a new direction. Beyond the policy-relevance of such knowledge, and its societal value, there also exist immediate consequences for operational long-lead climate predictions. The practice of seasonal climate prediction at the Climate Prediction Center includes utilization of US trends (generally the past 30-years) in surface temperature and precipitation through a method referred to as Optimal Climate Normal. The skill of seasonal predictions of US surface temperature, based on trend information alone, have proven comparable to predictors that have characteristic interannual variability, such as ENSO. Despite the success of observed climate trend information in seasonal predictions, understanding the causes for such variability from the observed data alone remains poor. Are such trends merely part of atmospheric internal variability, or can the trends be attributed to an external forcing? To the extent that a forcing has been important, is it manifest as the direct and local response to increasing greenhouse gases, or are the non-local feedbacks associated with the oceanís response to increasing greenhouse gases also important? A more complete understanding can be gained from the analysis of model simulations for the 20th century where the climate, including its low frequency statistical fluctuations is compared with similar runs where the observed evolution of external forcing is specified. Analysis for simulations from more than one model is also essential to gain confidence in validity of conclusions. The focus of this proposal, therefore, will be analysis of differences in atmospheric trends between control simulations and 20th century simulations to year 2000, and to analyze their relationships with the observed trends.

Data Sets Required: Analysis will rely on monthly mean fields from three sets of simulations: (1) 100 year (minimum) control run, and (2) 20th century AMIP simulations to year 2000 forced with observed SSTs and observed climate forcing, and (3) 20th century coupled simulations with observed climate forcing. Our main focus will be on the analysis of monthly mean upper level heights, land surface temperature, and precipitation.

    Add Publication

    << Back to WCRP CMIP3 Subprojects
    For questions or comments regarding this website, please contact the Webmaster.
    Lawrence Livermore National Laboratory  |  Physical & Life Sciences Directorate