A stochastic vision of the paleoclimate. Modeling and predictability - HAL-SHS - Sciences de l'Homme et de la Société Accéder directement au contenu
Article Dans Une Revue Georeview Année : 2014

A stochastic vision of the paleoclimate. Modeling and predictability

Résumé

The objective of the paper is related to the use of stochastic methods to appreciate if the recent climate trend is similar to the decennial trends of the last thousands years. If not, it means that the present climate change is unprecedented. If yes, it means that the extraordinary climatic events of the recent years are a natural expression of climate variability, but the memory of several generations and of instrumental Background There are uncertainties in the process of detecting a climate change. First, one may propose, like Klemes (1974), that no historical time can be stated for sure as stationary or not by the mathematicians. This is due to the short duration of instrumental series as compared to the (1989), is to know how long a sub period of a historical series should be in order to identify a local stationary state. At mean latitudes, the alternance of glacial and interglacial periods throughout the Quaternary clearly indicates that temperature averages have varied between two states, cold and warm, for geologic ages. This signifies a non-stationary state. At least at the average level of the series, the number of variations is equal to the number of stationary breaks. But at the level of peculiar sub periods, either glacial or interglacial stages, these have lasted enough to be considered as stationary from the point of view of the average temperature. If one changes the time scale to the annual level, and if one makes reference to the period of instrumental observations, of about 150 to 200 years, the bibliographic results are contradictory. The last decade is an exception, as many authors have pointed out the manifestation of global warming for the last 10 years. The annual hydro climatic series, studied by the water resources managers, are independent of time and can be considered as stationary ng stationary on average (Yevjevich, 1972; Salas et al., 1980; Kottegoda, 1980; Vandewiele, 1988). For 23 series of intermediate duration (96 to 1164 years) and analyzed at annual resolution, Hipel and McLeod (1994) have discovered the same type of result. On the basis of the data collected, the water resources managers assumed a hydrologic stationarity, as any trend cannot firmly established (Matalas, 1997). The trends encountered are suitable to be described as realizations of stochastic processes. In order to describe this type of process, it is reasonable to propose an ARIMA type of stationary model.

Domaines

Géographie
Fichier principal
Vignette du fichier
Haidu, 2014, A stochastic vision of the paleoclimate. Modelling and predictibility.pdf (932.26 Ko) Télécharger le fichier
Loading...

Dates et versions

hal-02488172 , version 1 (22-02-2020)

Licence

Paternité

Identifiants

Citer

Ionel Haidu. A stochastic vision of the paleoclimate. Modeling and predictability. Georeview, 2014, Special Issue: Late Pleistocene and Holocene climatic variability in the Carpathian-Balkan region- Abstracts Volume, 24 (2), ⟨10.4316/GEOREVIEW.2014.0.0.218⟩. ⟨hal-02488172⟩
37 Consultations
15 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More