One of the most challenging expressions of atmospheric circulation are compound events. They are characterized by multiple dependent variables that merge and cause extreme impacts. Some authors explain that comprehension of their causal chains is crucial to understand these phenomena. The state of art of causal models does not fulfill the requirements of the analysis of these phenomena, so a new method is developed. It is based on Reichenbach’s common cause principle and the goal is to determine the dependence between two stationary first order, two state Markov chains: six coefficients analyze the synchronism or asynchronism of two temporal series and their values are compared with the confidence interval of corresponding independent events. This method is applied to meteorological network in Europe and strong and soft connections between coupled stations are identified. The difference between strong and soft is based on the mean frequency of El Niño. Various behaviors of synchronism and asynchronism between the sites are identified and the distances are studied. So, the minimum distance between two teleconnected stations is established. The analysis results are compared with the synoptic circulation in Northern hemisphere and teleconnections and different patterns are verified. With the method correctness tested, it could be useful to study seasonal connections and to improve the knowledge of the challenging global atmospheric circulation.
Gli eventi composti sono una delle espressioni più complesse della circolazione atmosferica, in cui molteplici variabili dipendenti tra loro si uniscono per dar luogo ad eventi con impatti estremi. Alcuni autori sottolineano l’importanza della comprensione dei rapporti causali di questi fenomeni per una loro completa analisi. Lo stato delle conoscenze dei modelli causali trovati in letteratura non sembrava applicabile in questo campo. Per questo motivo si propone un nuovo metodo che, basato sul principio di causa comune di Reichenbach, individua una dipendenza tra due processi Markoviani stazionari a due stati di primo ordine: sei differenti coefficienti che individuano la sincronia e asincronia di variabili sono confrontati con l’intervallo di confidenza dei rispettivi eventi indipendenti congiunti. Il metodo è stato successivamente applicato alla rete pluviometrica europea e sono state individuate le connessioni forti e deboli a scala annuale, differenziate secondo la frequenza de El Niño. Per ognuna sono stati studiati i diversi comportamenti di sincronia e asincronia caratterizzanti le coppie di stazioni e analizzate le distanze. In questo modo è stata trovata la distanza per definire due siti teleconnessi tra loro. I risultati sono stati confrontati con la circolazione sinottica dell’emisfero boreale e le connessioni che ne influenzano il clima, trovando dei buoni riscontri. Verificata la correttezza del metodo, esso potrà essere raffinato per studiare le connessioni stagionali e potrà essere utile per una migliore comprensione della complicata circolazione atmosferica globale.
Studio dell'associazione in variabili binarie e connessioni in una rete meteorologica
MERONI, VIOLA
2018/2019
Abstract
One of the most challenging expressions of atmospheric circulation are compound events. They are characterized by multiple dependent variables that merge and cause extreme impacts. Some authors explain that comprehension of their causal chains is crucial to understand these phenomena. The state of art of causal models does not fulfill the requirements of the analysis of these phenomena, so a new method is developed. It is based on Reichenbach’s common cause principle and the goal is to determine the dependence between two stationary first order, two state Markov chains: six coefficients analyze the synchronism or asynchronism of two temporal series and their values are compared with the confidence interval of corresponding independent events. This method is applied to meteorological network in Europe and strong and soft connections between coupled stations are identified. The difference between strong and soft is based on the mean frequency of El Niño. Various behaviors of synchronism and asynchronism between the sites are identified and the distances are studied. So, the minimum distance between two teleconnected stations is established. The analysis results are compared with the synoptic circulation in Northern hemisphere and teleconnections and different patterns are verified. With the method correctness tested, it could be useful to study seasonal connections and to improve the knowledge of the challenging global atmospheric circulation.File | Dimensione | Formato | |
---|---|---|---|
Viola Meroni.pdf
non accessibile
Descrizione: Thesis text
Dimensione
16.45 MB
Formato
Adobe PDF
|
16.45 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/148635