 |
| Titel |
Causal Modes in the Low-frequency variability of Mediterranean and Middle-Eastern climates |
| VerfasserIn |
Erik Casagrande, Matteo Zampieri, Vincenzo Artale, Silvio Gualdi, Annalisa Molini |
| Konferenz |
EGU General Assembly 2015
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 17 (2015) |
| Datensatznummer |
250112528
|
| Publikation (Nr.) |
 EGU/EGU2015-12688.pdf |
|
|
|
|
|
| Zusammenfassung |
| In the last three decades, the Mediterranean and the Middle East experienced a phase of
warming larger than the one that could be expected from global warming, and largely
ascribable to natural (e.g. internal) climate variability. To better understand this process
we explore here the presence of causal relationships among the diverse modes of
variability of the climate system, focusing in particular on inter-annual and decadal
scales of variability, influencing the climate of Mediterranean and Middle-Eastern
regions. Causality measures used in this study include time and frequency-domain
Granger causality (GC) and the phase slope index (Ψ), a directional coupling statistic
developed by Nolte et. al. in 2007. GC metrics are applied to signals before and after the
filtering of high frequency (inter-annual) components, while Ψ is designed to discern
between low-frequency causal flow and higher frequency components. To assure the
necessary sample size, the analysis is based on the preindustrial runs of the Fifth
Coupled Model Intercomparison Project (CMIP5), which are free from external
perturbation and last some hundred years. We selected the runs based on ENSO
stationarity – to ensure that the simulations reached the equilibrium – and the consistent
representation of the Atlantic Multidecadal Oscillation (AMO), which is considered
one of the main drivers for the low-frequency (decadal) climate variability of the
Mediterranean and the Middle East in summer. Finally, we discuss the potential of
causality metrics for the predictability of future decadal variability in these regions. |
|
|
|
|
|
|