Berlin 2012 – wissenschaftliches Programm
DY 22.55: Poster
Mittwoch, 28. März 2012, 17:00–19:00, Poster A
Detecting long-range teleconnections in the climate network via ordinal pattern time-series analysis — •Juan Ignacio Deza1, Marcelo Barreiro2, and Cristina Masoller1 — 1Departament de Física i Enginyeria Nuclear, Universitat Politecnica de Catalunya, Colom 11. E-08222, Terrassa, Barcelona, Spain — 2Instituto de Física, Facultad de Ciencias, Universidad de la República, Iguá, 4225, Montevideo, Uruguay
We study the global climate network by means of nonlinear time series analysis of monthly averaged surface air temperature (SAT) anomalies. We employ various methods to infer climate interdependencies, like the linear cross correlation and the mutual information computed from SAT data and from its symbolic, ordinal-pattern-based representation. Special emphasis is made on detecting weak and long range connections (teleconnections). We find that the linear cross correlation measure tends to emphasize the connections between the tropics and the extratropics, while the nonlinear methods emphasize the regional internal atmospheric dynamics. We also report evidence of correlations between the poles and different regions, some of them very well limited geographically.