desole mais l'article dont parles Hughues (voir ci dessus) n'est pas celui auquel tu fais reference!!!
les conclusions de l'article cite par Hughes sont les suivantes:
This study has identified several persistent cyclic variations in climatic and meteorological records from Svalbard and Greenland. Some of the cycles appear to correspond to known cyclic variations in the Moons' orbit around Earth, while others may correspond to solar variations. Notwithstanding the physical explanation for such cyclic variations, which is not the main focus of the present study, wavelet analysis of climatic and meteorological records represents a potentially useful means for climate analysis, as a supplement to Fourier analysis. In contrast to Fourier analysis, the wavelet analysis provides information on the time-dependant dynamics of observed recurrent climate variations, which is especially important to understand the physical explanation for observed variations and to evaluate the future development.
The present warm period following the Little Ice Age since about 1800 AD can be reproduced by a simple three input period only approach, based on the Greenland GISP2 temperature record. Apparently the present period of warming since the LIA to a high degree may be the result of natural climatic variations, known to characterise at least the previous 4000 years.
Both investigated records show high natural variability and exhibits long-term persistence, although on different time scales. The strength and persistence of several of the identified natural cyclic variations suggests that a natural cycle based forecasting of future climate may be potentially feasible, at least for limited time ranges. Our empirical experience suggests a realistic forecasting time range of about 10–25% of the total record length. In the case of Greenland, such forecasting suggests that the present post LIA warm period is likely to continue for most of the 21st century, before the overall Late Holocene cooling may again dominate, but this being dependant on the magnitude of the anthropogenic greenhouse enhancement.
Fourier and wavelet analyses deconstruct data series into their fundamental components. Natural cycles that have remained strong over several decades or centuries are likely to continue without major changes into at least the near future, and will therefore be essential for forecasting any future climatic development. Forecasts based on insights obtained by Fourier and wavelet analyses should therefore not be considered purely statistical, as they are based on observed dynamics characterising past climate change. Used together, Fourier and wavelet analyses represent potential important techniques for climate analysis.
The natural cycle climate model forecast described in the present paper is supplementary to forecasts (scenarios) derived from analytic climate models, and thereby represents a complimentary approach to climate forecasting based on such analytic models.