Use of Short Time Series for Early Global Warming Trend Detection for Ocean Acoustic Travel Times.
1997
McCartor, Gary D. | Wortman, William R. | Bottone, Steven
Previous work indicated that for expected variability of ocean acoustic propagation times and the level of anticipated warming trends, it may take at least two decades of observation to establish, with an adequate level of confidence, that warming does or does not exist. In the current report, we have attempted to establish the extent to which the use of data from a shorter period may allow at least better definition of the period needed to establish these warming trends. Truncated simulated data sets, generally 5 or 10 years, from the MASIG model of acoustic propagation time variability over decades have been used to establish the extent to which the statistical trend extraction techniques give stable results using these shorter series to define the noise background. It is found that the use of these shorter series provides an apparent reduction of time required to detect trends; This unfortunately indicates that meaningful estimates of warming trends cannot be made with use of a lengthy data set approaching the full 20 year interval. Generally it is found that once the length of the background data is less than half of the full set, the estimates of times needed to detect trends are, falsely, reduced. This indicates that the variability which tends to mask trends is long period and that shorter observation intervals do not allow characterization of this variability. In another portion of this study, we fmd that the substantial short term variability seen in the first six months of the ATOC experiment is comparable with that seen by Spiesberger but that this variability plays almost no role for the long term trend detection problem, which is dominated by multiyear periods.
Afficher plus [+] Moins [-]Mots clés AGROVOC
Informations bibliographiques
Cette notice bibliographique a été fournie par AVANO
Découvrez la collection de ce fournisseur de données dans AGRIS
