GEOSCAN, résultats de la recherche


TitreReading the bass line: how well do moisture-sensitive tree rings track decadal variability?
AuteurSt. George, S; Ault, T R
SourceAmerican Geophysical Union Fall Meeting 2010, abstracts; PP52A-02, 2010 p. 1 (Accès ouvert)
LiensOnline - En ligne
Séries alt.Ressources naturelles Canada, Contribution externe 20200317
ÉditeurAmerican Geophysical Union
RéunionAmerican Geophysical Union Fall Meeting 2010; San Francisco, CA; US; décembre 13-17, 2010
Documentsite Web
Mediaen ligne; numérique
Formatshtml; pdf
ProgrammeGéosciences de changements climatiques, Impacts des changements climatiques et adaptation dans le secteur des ressources naturelles et d'autres secteurs clés de l'économie
Diffusé2010 12 01
Résumé(disponible en anglais seulement)
Most dendroclimatic studies assess past changes in decadal variability by first reconstructing an annually-resolved target variable, and then applying some form of filter that emphasizes variability within a specific frequency band. We evaluate the ability of a network of tree-ring records along the central Pacific Coast of the United States (hereafter, the CPC) to estimate the behavior of an exceptionally vigorous decadal pattern in winter precipitation. The CPC is one of the few regions in North America where precipitation records exhibited strong variability at decadal timescales during the last century. Fewer than one-quarter of all tree-ring chronologies from this region are good proxies for the decadal pattern, but Monte Carlo analysis demonstrates that the level of similarity observed between the ring-width network and winter precipitation was not likely to occur due to chance. By screening the network to retain those tree-ring chronologies that are optimal predictors of our decadal target, we produce an estimate of that component that is better than those obtained from either projecting the signal over all records or over some function (either the network's mean or its leading principal component) that describes tree growth across the entire network. Projecting the pattern over the entire length of the tree-ring chronologies indicated that decadal variability in regional precipitation was most vigorous during the mid and late-20th century. Between 1650 and 1930, the amplitude of the decadal pattern was relatively weak and the proxy estimates show a limited number of decadal events separated by longer intervals of lower variance. Our results indicate that strong decadal variability is a relatively new feature of the winter climate of the CPC region, and that this type of behavior has been uncommon for most of the last three and a half centuries. They also provide another example of the benefits of reconstruction approaches that evaluate the ability of proxy records to track climate variability at specific timescales.