Annually varved lake sediments constitute excellent environmental archives, and have the potential to play an important role in understanding past seasonal climates. In this thesis we use functional data analysis methods to infer on past environmental and climate changes. In particular we analyze the varved sediments collected from the bottom of two lakes in southern Finland, lake Nautajärvi covering the past 9900 years, and lake Korttajärvi covering the past 8500 years. We propose two functional clustering methods, k-means alignment algorithm and double clustering Bagging Voronoi algorithm, used to group the seasonal patterns and also longer-term trends into different types that can be associated with different weather and climatic conditions. The resulting clusters are compared with the ones obtained performing similar analysis on lake Kassjön in northern Sweden. The time dynamics show great potential for seasonal climate interpretation.
Gli strati annuali di sedimenti (varve) provenienti dal fondo dei laghi, costituiscono eccellenti archivi ambientali e svolgono un ruolo importante nella comprensione del sistema climatico. In questa tesi vengono utilizzati metodi di analisi funzionale per dedurre i cambiamenti metereologici e climatici del passato. In particolare vengono analizzati il lago Nautajärvi e il lago Korttajärvi situati nel sul della Finlandia, i cui sedimenti coprono rispettivamente 9900 e 8500 anni. Vengono proposti due algoritmi di clustering funzionale, il k-means alignment e il Double Clustering Bagging Voronoi, allo scopo di suddividere i patterns stagionali e trends a lungo termine in gruppi associati a diverse condizioni metereologiche e climatiche. I cluster risultanti sono confrontati con quelli ottenuti da studi simili effettuati sul lago Kassjön in Svezia. La dinamica temporale mostra un grande potenziale per l’interpretazione dei climi stagionali passati.
Functional clustering of varved lake sediment to reconstruct past climate in Fennoscandia
POZZARI, ANNA
2016/2017
Abstract
Annually varved lake sediments constitute excellent environmental archives, and have the potential to play an important role in understanding past seasonal climates. In this thesis we use functional data analysis methods to infer on past environmental and climate changes. In particular we analyze the varved sediments collected from the bottom of two lakes in southern Finland, lake Nautajärvi covering the past 9900 years, and lake Korttajärvi covering the past 8500 years. We propose two functional clustering methods, k-means alignment algorithm and double clustering Bagging Voronoi algorithm, used to group the seasonal patterns and also longer-term trends into different types that can be associated with different weather and climatic conditions. The resulting clusters are compared with the ones obtained performing similar analysis on lake Kassjön in northern Sweden. The time dynamics show great potential for seasonal climate interpretation.| File | Dimensione | Formato | |
|---|---|---|---|
|
2017_12_Pozzari.pdf
accessibile in internet per tutti
Descrizione: Testo della tesi
Dimensione
5.53 MB
Formato
Adobe PDF
|
5.53 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/10589/137329