東華大學圖書館 |

The Use of Wavelet Transform-Based Methods to Analyze Variability in Hydrological Data, Multiscale Linkages to Large-Scale Climate Oscillations, and for Hydrological Record Extension.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	The Use of Wavelet Transform-Based Methods to Analyze Variability in Hydrological Data, Multiscale Linkages to Large-Scale Climate Oscillations, and for Hydrological Record Extension./
作者:	Nalley, Deasy Praity.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2020,
面頁冊數:	308 p.
附註:	Source: Dissertations Abstracts International, Volume: 82-10, Section: B.
Contained By:	Dissertations Abstracts International82-10B.
標題:	Decomposition. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28383663
ISBN:	9798708718594

The Use of Wavelet Transform-Based Methods to Analyze Variability in Hydrological Data, Multiscale Linkages to Large-Scale Climate Oscillations, and for Hydrological Record Extension.
Nalley, Deasy Praity.

The Use of Wavelet Transform-Based Methods to Analyze Variability in Hydrological Data, Multiscale Linkages to Large-Scale Climate Oscillations, and for Hydrological Record Extension. - Ann Arbor : ProQuest Dissertations & Theses, 2020 - 308 p.

Source: Dissertations Abstracts International, Volume: 82-10, Section: B.

Thesis (Ph.D.)--McGill University (Canada), 2020.

This item must not be sold to any third party vendors.

Time series analyses performed on hydro-meteorological data such as precipitation and streamflow provide useful information regarding the evolution of climatic changes and variability. Quantifying these changes is challenging due to factors such as the existence of multiple periodicities in the data that may be associated with activities in large-scale climate oscillations. The unavailability of sufficient data further presents challenges to time series analyses and to the identification of linkages among hydro-meteorological and climatic variables.The goals of the current research are twofold: 1) to assess multiscale linkages between a hydrometeorological variable and one or more large-scale climate oscillation; and 2) to address the issue of water-resources data scarcity by extending streamflow records. Precipitation and streamflow data used are obtained from stations across Canada. The climate oscillations included in the analysis are El Nino-Southern Oscillation (ENSO), the North Atlantic Oscillation (NAO) and the Pacific Decadal Oscillation (PDO).The research goals are accomplished through the development of innovative wavelet-based approaches, which consist of four parts (i.e., the four main components of the thesis). The applications of WT in this research are summarized in two main ways: 1) WT methods are used to assess the individual and combined influences of large-scale climate oscillations on streamflow and precipitation data; and 2) the discrete wavelet transform (DWT) is used as a pre-processing technique to improve the performance of eight existing record extension methods (namely, ordinary least squares, maintenance of variance extension (MOVE) types 1-4, Kendall-Theil robust line (KTRL), KTRL2 and robust line of organic correlation (RLOC)).The first main part serves as the foundation of the research - it demonstrates the utility of continuous wavelet transform (CWT) and wavelet coherence (WTC) in assessing linkages between streamflow and ENSO, NAO and PDO. The procedures are applied to data from eight stations in Quebec and Ontario. Strong seasonality in streamflow data is depicted via the CWT spectra, and significant periodicities within 4-8 years and greater than 8 years are related to ENSO and NAO, and PDO, respectively. Extending the use of CWT, the second section proposes the inclusion of two procedures for quantifying the linkages between streamflow (data from 73 stations are used) and a climate oscillation. First, the main change point in streamflow is identified; second, significant lowfrequency components (e.g., trends) in streamflow that are hidden by the dominant higherfrequency variations are revealed using the inverse CWT. Most of the main change points in streamflow occur during the late 1960s-early 1970s, where intense activities or phase shifts occur, especially in ENSO and NAO. The intra-annual to annual variability in streamflow conceal the effect of lower-scale components, as their significance is only noticeable after data reconstructions.The third part of this thesis extends the application of WTC, in which multiple wavelet coherence (MWC) is used, as a multiscale and multivariate analysis method, to analyze the combined influences of ENSO, NAO and PDO. The combined influences imply three two-factor combinations (ENSO-NAO, ENSO-PDO and NAO-PDO) and one three-factor combination (ENSO-NAO-PDO). The procedures are applied to streamflow and precipitation data from three watersheds with distinct climatic conditions - in Alberta, Ontario and Newfoundland. There are consistent increases in the values of the average wavelet coherence (AWC) and the percentage of significant periodicities (PoSP) for all watersheds, from the one factor (i.e., a single oscillation), to the two-factor combinations, with the highest AWC and PoSP values reflected in the three-factor combinations.Finally, building on the utility of WT demonstrated through parts 1-3, the last part of the thesis incorporates DWT as a pre-processing method in extending streamflow records, using data from 67 stations spread across Canada. The performances of the WT-based methods and the conventional methods alone (without data pre-processing via DWT) are compared. Improved accuracy and precision for the WT-based methods are observed, especially for MOVE-WT (types 1 and 2), KTRLWT, KTRL2-WT and RLOC-WT.Through these four subjects of exploration, this thesis demonstrates the effectiveness of WT approaches - both CWT (and its derivatives, WTC and MWC) and DWT - in analyzing hydrometeorological data. Collectively, the results of the current research enhance the understanding of variability in streamflow and precipitation over many areas in Canada. This is useful for improving forecasting in operational hydrology and for sustainable management of water resources as a whole.

ISBN: 9798708718594Subjects--Topical Terms:

3561186
Decomposition.
Subjects--Index Terms:

Wavelet transform

The Use of Wavelet Transform-Based Methods to Analyze Variability in Hydrological Data, Multiscale Linkages to Large-Scale Climate Oscillations, and for Hydrological Record Extension.
LDR:10134nmm a2200373 4500 001 2343463
005 20220512065813.5
008 241004s2020 ||||||||||||||||| ||eng d
020 $a 9798708718594
035 $a (MiAaPQ)AAI28383663
035 $a (MiAaPQ)McGill_tx31qp414
035 $a AAI28383663
040 $a MiAaPQ $c MiAaPQ
100 1 $a Nalley, Deasy Praity. $3 3682030
245 1 4 $a The Use of Wavelet Transform-Based Methods to Analyze Variability in Hydrological Data, Multiscale Linkages to Large-Scale Climate Oscillations, and for Hydrological Record Extension.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2020
300 $a 308 p.
500 $a Source: Dissertations Abstracts International, Volume: 82-10, Section: B.
500 $a Advisor: Biswas, Asim;Adamowski, Jan.
502 $a Thesis (Ph.D.)--McGill University (Canada), 2020.
506 $a This item must not be sold to any third party vendors.
520 $a Time series analyses performed on hydro-meteorological data such as precipitation and streamflow provide useful information regarding the evolution of climatic changes and variability. Quantifying these changes is challenging due to factors such as the existence of multiple periodicities in the data that may be associated with activities in large-scale climate oscillations. The unavailability of sufficient data further presents challenges to time series analyses and to the identification of linkages among hydro-meteorological and climatic variables.The goals of the current research are twofold: 1) to assess multiscale linkages between a hydrometeorological variable and one or more large-scale climate oscillation; and 2) to address the issue of water-resources data scarcity by extending streamflow records. Precipitation and streamflow data used are obtained from stations across Canada. The climate oscillations included in the analysis are El Nino-Southern Oscillation (ENSO), the North Atlantic Oscillation (NAO) and the Pacific Decadal Oscillation (PDO).The research goals are accomplished through the development of innovative wavelet-based approaches, which consist of four parts (i.e., the four main components of the thesis). The applications of WT in this research are summarized in two main ways: 1) WT methods are used to assess the individual and combined influences of large-scale climate oscillations on streamflow and precipitation data; and 2) the discrete wavelet transform (DWT) is used as a pre-processing technique to improve the performance of eight existing record extension methods (namely, ordinary least squares, maintenance of variance extension (MOVE) types 1-4, Kendall-Theil robust line (KTRL), KTRL2 and robust line of organic correlation (RLOC)).The first main part serves as the foundation of the research - it demonstrates the utility of continuous wavelet transform (CWT) and wavelet coherence (WTC) in assessing linkages between streamflow and ENSO, NAO and PDO. The procedures are applied to data from eight stations in Quebec and Ontario. Strong seasonality in streamflow data is depicted via the CWT spectra, and significant periodicities within 4-8 years and greater than 8 years are related to ENSO and NAO, and PDO, respectively. Extending the use of CWT, the second section proposes the inclusion of two procedures for quantifying the linkages between streamflow (data from 73 stations are used) and a climate oscillation. First, the main change point in streamflow is identified; second, significant lowfrequency components (e.g., trends) in streamflow that are hidden by the dominant higherfrequency variations are revealed using the inverse CWT. Most of the main change points in streamflow occur during the late 1960s-early 1970s, where intense activities or phase shifts occur, especially in ENSO and NAO. The intra-annual to annual variability in streamflow conceal the effect of lower-scale components, as their significance is only noticeable after data reconstructions.The third part of this thesis extends the application of WTC, in which multiple wavelet coherence (MWC) is used, as a multiscale and multivariate analysis method, to analyze the combined influences of ENSO, NAO and PDO. The combined influences imply three two-factor combinations (ENSO-NAO, ENSO-PDO and NAO-PDO) and one three-factor combination (ENSO-NAO-PDO). The procedures are applied to streamflow and precipitation data from three watersheds with distinct climatic conditions - in Alberta, Ontario and Newfoundland. There are consistent increases in the values of the average wavelet coherence (AWC) and the percentage of significant periodicities (PoSP) for all watersheds, from the one factor (i.e., a single oscillation), to the two-factor combinations, with the highest AWC and PoSP values reflected in the three-factor combinations.Finally, building on the utility of WT demonstrated through parts 1-3, the last part of the thesis incorporates DWT as a pre-processing method in extending streamflow records, using data from 67 stations spread across Canada. The performances of the WT-based methods and the conventional methods alone (without data pre-processing via DWT) are compared. Improved accuracy and precision for the WT-based methods are observed, especially for MOVE-WT (types 1 and 2), KTRLWT, KTRL2-WT and RLOC-WT.Through these four subjects of exploration, this thesis demonstrates the effectiveness of WT approaches - both CWT (and its derivatives, WTC and MWC) and DWT - in analyzing hydrometeorological data. Collectively, the results of the current research enhance the understanding of variability in streamflow and precipitation over many areas in Canada. This is useful for improving forecasting in operational hydrology and for sustainable management of water resources as a whole.
520 $a Les analyses de series chronologiques de precipitation et de debit aident a comprendre l'evolution des changements climatiques et leur variabilite. L'indisponibilite des donnees et l'existence de multiples periodicites dans celles-ci, rend difficile la quantification de ces changements. Ces periodicites peuvent representer l'influence d'oscillations climatiques a grande echelle (intra-annuelles a multi-decennales). Les methodes de transformee en ondelettes (WT) permet d'evaluer les influences individuelles et simultanees des oscillations climatiques a petite et grande echelle sur les debits et de precipitations. La transformee en ondelettes discretes (DWT) permet un pretraitement de donnees visant a ameliorer les performances de huit methodes d'extrapolation de donnees existantes : les moindres carres ordinaires, extension du maintien de la variance (MOVE) types 1-4, ligne Kendall-Theil robuste (KTRL), KTRL2 et ligne robuste de correlation organique (RLOC)). Les donnees de precipitations et de debit proviennent de stations situees a travers le Canada. Les oscillations climatiques incluses dans l'analyse sont : El Nino - Oscillation australe (ENSO), l'oscillation nord-atlantique (ONA) et l'oscillation decennale du Pacifique (ODP).Developpant des approches novatrices basees sur les ondelettes, cette recherche comprend 4 volets. Le 1er, le fondement de cette recherche, demontre l'utilite de la transformee en ondelettes continue (CWT) et de la coherence en ondelettes (WTC) pour etablir des liens entre le flux des cours d'eau et ENSO, ONA et ODP. Ces techniques furent appliquees a des donnees provenant du Quebec et de l'Ontario. Des periodicites significatives sur 4 a 8 ans furent liees a ENSO et NAO, tandis que celles de plus de 8 ans furent liees au PDO.Le 2e volet propose l'inclusion de deux procedes pour quantifier les liens entre le debit (donnees de 73 stations) et les oscillations climatiques. Tout d'abord, le principal point de rupture dans le flux fut identifie, puis, grace a une CWT inverse, les composantes a basse frequence (c.a.d., les tendances) du debit, cachees par les variations dominantes de frequence superieure, furent revelees. La plupart des points de rupture eurent lieu entre 1965-1975, lorsque des activites intenses ou des changements de phase dans l'ENSO et le NAO eurent lieu. La variabilite intra-annuelle a annuelle du debit masqua les effets des composantes a plus petite echelle, leur importance n'etant perceptible qu'apres la reconstruction des donnees.Le 3e volet de cette these etend l'application du WTC : la coherence a ondelettes multiples (MWC) servit comme methode d'analyse multi-echelle et multivariee pour etudier les influences individuelles et combinees de ENSO, NAO et PDO. Ces influences impliquerent trois combinaisons a deux facteurs (ENSO-NAO, ENSO-PDO et NAO-PDO) et une combinaison a trois facteurs (ENSO-NAO-PDO). Les procedures furent appliquees aux donnees de trois bassins versants (Alberta, Ontario et Terre-Neuve). Une augmentation de la coherence moyenne des ondelettes (AWC) et du pourcentage de periodicites significatives (PoSP) furent notes pour tous les bassins versants, allant d'un seul facteur (une seule oscillation), a une combinaison de 2 facteurs, jusqu'a une combinaison a 3 facteurs, montrant l'AWC et les valeurs PoSP les plus eleves.Le 4e volet integre la DWT comme pretraitement dans l'extension des enregistrements de flux provenant de 67 stations. Les methodes basees sur la WT augmenterent l'exactitude et la precision, en particulier celles de MOVE-WT (types 1 et 2), KTRL-WT, KTRL2-WT et RLOC-WT.Cette etude demontre l'efficacite des approches WT dans l'analyse des donnees hydrometeorologiques, permettant d'ameliorer nos connaissances quant a la variabilite du debit et des precipitations dans de nombreuses regions du Canada et s'averant donc utile pour ameliorer les previsions hydrologiques operationnelles et la gestion durable des ressources en eau.
590 $a School code: 0781.
650 4 $a Decomposition. $3 3561186
650 4 $a Multivariate analysis. $3 517467
650 4 $a Hydrology. $3 545716
650 4 $a Accuracy. $3 3559958
650 4 $a Precipitation. $3 3680799
650 4 $a Wavelet transforms. $3 3681479
650 4 $a Influence. $3 1973172
650 4 $a Stream flow. $3 3560049
650 4 $a Hydrologic sciences. $3 3168407
650 4 $a Bioengineering. $3 657580
653 $a Wavelet transform
653 $a Hydrological data
653 $a Large-scale climate Oscillations,
653 $a Hydrological record extension
690 $a 0202
690 $a 0388
710 2 $a McGill University (Canada). $3 1018122
773 0 $t Dissertations Abstracts International $g 82-10B.
790 $a 0781
791 $a Ph.D.
792 $a 2020
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28383663