Change-point analysis as a tool to detect abrupt climate variations

Claudie Beaulieu, Jie Chen, Jorge L. Sarmiento

Research output: Contribution to journalArticle

71 Citations (Scopus)

Abstract

Recently, there have been an increasing number of studies using change-point methods to detect artificial or natural discontinuities and regime shifts in climate. However, a major drawback with most of the currently used change-point methods is the lack of flexibility (able to detect one specific type of shift under the assumption that the residuals are independent). As temporal variations in climate are complex, it may be difficult to identify change points with very simple models. Moreover, climate time series are known to exhibit autocorrelation, which corresponds to a model misspecification if not taken into account and can lead to the detection of non-existent shifts. In this study, we extend a method known as the informational approach for change-point detection to take into account the presence of autocorrelation in the model. The usefulness and flexibility of this approach are demonstrated through applications. Furthermore, it is highly desirable to develop techniques that can detect shifts soon after they occur for climate monitoring. To address this, we also carried out a simulation study in order to investigate the number of years after which an abrupt shift is detectable. We use two decision rules in order to decide whether a shift is detected or not, which represents a trade-off between increasing our chances of detecting a shift and reducing the risk of detecting a shift while in reality there is none. We show that, as of now, we have good chances to detect an abrupt shift with a magnitude that is larger than that of the standard deviation in the series of observations. For shifts with a very large magnitude (three times the standard deviation), our simulation study shows that after only 4 years the probabilities of shift detection reach nearly 100 per cent. This reveals that the approach has potential for climate monitoring.

Original languageEnglish (US)
Pages (from-to)1228-1249
Number of pages22
JournalPhilosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences
Volume370
Issue number1962
DOIs
StatePublished - Mar 13 2012

Fingerprint

Change-point Analysis
Climate
climate
Change Point
Autocorrelation
shift
Monitoring
Standard deviation
Flexibility
Simulation Study
Time series
Change-point Detection
Model Misspecification
Decision Rules
Discontinuity
autocorrelation
standard deviation
Trade-offs
flexibility
Series

Keywords

  • Abrupt climate change
  • Autocorrelation
  • Change-point detection
  • Regime shift

ASJC Scopus subject areas

  • Mathematics(all)
  • Engineering(all)
  • Physics and Astronomy(all)

Cite this

Change-point analysis as a tool to detect abrupt climate variations. / Beaulieu, Claudie; Chen, Jie; Sarmiento, Jorge L.

In: Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 370, No. 1962, 13.03.2012, p. 1228-1249.

Research output: Contribution to journalArticle

@article{1194af8c7cfd43cdb9dd68558e80db9c,
title = "Change-point analysis as a tool to detect abrupt climate variations",
abstract = "Recently, there have been an increasing number of studies using change-point methods to detect artificial or natural discontinuities and regime shifts in climate. However, a major drawback with most of the currently used change-point methods is the lack of flexibility (able to detect one specific type of shift under the assumption that the residuals are independent). As temporal variations in climate are complex, it may be difficult to identify change points with very simple models. Moreover, climate time series are known to exhibit autocorrelation, which corresponds to a model misspecification if not taken into account and can lead to the detection of non-existent shifts. In this study, we extend a method known as the informational approach for change-point detection to take into account the presence of autocorrelation in the model. The usefulness and flexibility of this approach are demonstrated through applications. Furthermore, it is highly desirable to develop techniques that can detect shifts soon after they occur for climate monitoring. To address this, we also carried out a simulation study in order to investigate the number of years after which an abrupt shift is detectable. We use two decision rules in order to decide whether a shift is detected or not, which represents a trade-off between increasing our chances of detecting a shift and reducing the risk of detecting a shift while in reality there is none. We show that, as of now, we have good chances to detect an abrupt shift with a magnitude that is larger than that of the standard deviation in the series of observations. For shifts with a very large magnitude (three times the standard deviation), our simulation study shows that after only 4 years the probabilities of shift detection reach nearly 100 per cent. This reveals that the approach has potential for climate monitoring.",
keywords = "Abrupt climate change, Autocorrelation, Change-point detection, Regime shift",
author = "Claudie Beaulieu and Jie Chen and Sarmiento, {Jorge L.}",
year = "2012",
month = "3",
day = "13",
doi = "10.1098/rsta.2011.0383",
language = "English (US)",
volume = "370",
pages = "1228--1249",
journal = "Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences",
issn = "0962-8428",
publisher = "Royal Society of London",
number = "1962",

}

TY - JOUR

T1 - Change-point analysis as a tool to detect abrupt climate variations

AU - Beaulieu, Claudie

AU - Chen, Jie

AU - Sarmiento, Jorge L.

PY - 2012/3/13

Y1 - 2012/3/13

N2 - Recently, there have been an increasing number of studies using change-point methods to detect artificial or natural discontinuities and regime shifts in climate. However, a major drawback with most of the currently used change-point methods is the lack of flexibility (able to detect one specific type of shift under the assumption that the residuals are independent). As temporal variations in climate are complex, it may be difficult to identify change points with very simple models. Moreover, climate time series are known to exhibit autocorrelation, which corresponds to a model misspecification if not taken into account and can lead to the detection of non-existent shifts. In this study, we extend a method known as the informational approach for change-point detection to take into account the presence of autocorrelation in the model. The usefulness and flexibility of this approach are demonstrated through applications. Furthermore, it is highly desirable to develop techniques that can detect shifts soon after they occur for climate monitoring. To address this, we also carried out a simulation study in order to investigate the number of years after which an abrupt shift is detectable. We use two decision rules in order to decide whether a shift is detected or not, which represents a trade-off between increasing our chances of detecting a shift and reducing the risk of detecting a shift while in reality there is none. We show that, as of now, we have good chances to detect an abrupt shift with a magnitude that is larger than that of the standard deviation in the series of observations. For shifts with a very large magnitude (three times the standard deviation), our simulation study shows that after only 4 years the probabilities of shift detection reach nearly 100 per cent. This reveals that the approach has potential for climate monitoring.

AB - Recently, there have been an increasing number of studies using change-point methods to detect artificial or natural discontinuities and regime shifts in climate. However, a major drawback with most of the currently used change-point methods is the lack of flexibility (able to detect one specific type of shift under the assumption that the residuals are independent). As temporal variations in climate are complex, it may be difficult to identify change points with very simple models. Moreover, climate time series are known to exhibit autocorrelation, which corresponds to a model misspecification if not taken into account and can lead to the detection of non-existent shifts. In this study, we extend a method known as the informational approach for change-point detection to take into account the presence of autocorrelation in the model. The usefulness and flexibility of this approach are demonstrated through applications. Furthermore, it is highly desirable to develop techniques that can detect shifts soon after they occur for climate monitoring. To address this, we also carried out a simulation study in order to investigate the number of years after which an abrupt shift is detectable. We use two decision rules in order to decide whether a shift is detected or not, which represents a trade-off between increasing our chances of detecting a shift and reducing the risk of detecting a shift while in reality there is none. We show that, as of now, we have good chances to detect an abrupt shift with a magnitude that is larger than that of the standard deviation in the series of observations. For shifts with a very large magnitude (three times the standard deviation), our simulation study shows that after only 4 years the probabilities of shift detection reach nearly 100 per cent. This reveals that the approach has potential for climate monitoring.

KW - Abrupt climate change

KW - Autocorrelation

KW - Change-point detection

KW - Regime shift

UR - http://www.scopus.com/inward/record.url?scp=84863116010&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863116010&partnerID=8YFLogxK

U2 - 10.1098/rsta.2011.0383

DO - 10.1098/rsta.2011.0383

M3 - Article

C2 - 22291231

AN - SCOPUS:84863116010

VL - 370

SP - 1228

EP - 1249

JO - Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences

JF - Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences

SN - 0962-8428

IS - 1962

ER -