- Title
- Analysis of climate dynamics as a complex system: new insights from the development of advanced methodologies in information theory and complex networks.
- Creator
- Carpi, Laura
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2013
- Description
- Research Doctorate - Doctor of Philosophy (PhD)
- Description
- Growing interest in climate prediction highlights the importance of achieving a better understanding of the underlying mechanisms behind climate variability. One of the dominant modes of interannual climate variability with worldwide weather and socioeconomic impacts is the El Niño Southern Oscillation (ENSO). Different methodologies have been used to improve the understanding of the ENSO dynamics, however research is still needed. During the last two decades the development and use of complex networks theory has led to important advances in the analysis of climate dynamics. Specfically, the analysis of climate networks, that is networks constructed using climate data, has provided valuable insights into different aspects of the climate dynamics that could not be captured using the classic methods frequently used in climatology. The goal of this thesis is to implement and develop new methodologies based on Information Theory quantifers and Complex Networks for the analysis of climate variability at a variety of temporal scales. Initial work focussed on the use of well-known Information Theory quantifers to study several dynamical systems including the logistic map and stochastic noises with different degrees of correlation. The Bandt & Pompe methodology, a symbolisation technique for time series analysis is used for the estimation of the probability distribution functions and missing ordinal patterns. Temporal changes in the dynamics of the Holocene ENSO proxy record of the Laguna Pallcacocha sedimentary data were also studied using entropy quantifiers, missing ordinal patterns, the Fisher Information measure, and the Shannon-Jensen Statistical Complexity. The analysis using these Information Theory Quantifers enabled the detection of climatic cycles with a period close to 2000 years during the mid-to-late Holocene. This work also enabled the establishment of connections between changes in entropy and epochs of rapid climate change (RCC), which indicated that the dynamics of the ENSO proxy record during the RCC interval 9000-8000 BP displays very low entropy (high predictability) that is remarkably different from that of the other RCCs of the Holocene. In a second stage, a new methodology based on Information Theory quantifers was developed to analyse dynamical changes in complex networks. It was shown that the metric properties of this quantifer make it a powerful tool to compare different states evolving networks topologies, and also to measure the distance between two different network structures. This novel integrative approach was applied to study the temporal evolution of the surface air temperature climate network for the Tropical Pacific region. The evolving characteristics of the network were analysed for temporal windows of one year duration over the 1948 - 2009 period. Results successfully captured the cyclic behaviour of El Niño Southern Oscillation, which was shown to involve alternating states of lower/higher efficiency in information transfer during El Niño/La Niña years respectively, reflecting a higher climatic stability for La Niña years that is in agreement with observations. The study also detected a change in the dynamics of the network structure that coincides with the 1976/77 climate shift, after which conditions of less efficient information transfer are more frequent and intense.
- Subject
- climate networks; information theory; complex networks; ENSO; thesis by publication
- Identifier
- http://hdl.handle.net/1959.13/940859
- Identifier
- uon:13115
- Rights
- Copyright 2013 Laura Carpi
- Language
- eng
- Full Text
- Hits: 1655
- Visitors: 1779
- Downloads: 381
Thumbnail | File | Description | Size | Format | |||
---|---|---|---|---|---|---|---|
View Details Download | ATTACHMENT01 | Abstract | 636 KB | Adobe Acrobat PDF | View Details Download | ||
View Details Download | ATTACHMENT02 | Thesis | 6 MB | Adobe Acrobat PDF | View Details Download |