New Exploratory Tools for Extremal Dependence: χ Networks and Annual Extremal Networks

Whitney K. Huang, Daniel S. Cooley, Imme Ebert-Uphoff, Chen Chen, Snigdhansu B Chatterjee

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Understanding dependence structure among extreme values plays an important role in risk assessment in environmental studies. In this work, we propose the χ network and the annual extremal network for exploring the extremal dependence structure of environmental processes. A χ network is constructed by connecting pairs whose estimated upper tail dependence coefficient, χ^ , exceeds a prescribed threshold. We develop an initial χ network estimator, and we use a spatial block bootstrap to assess both the bias and variance of our estimator. We then develop a method to correct the bias of the initial estimator by incorporating the spatial structure in χ. In addition to the χ network, which assesses spatial extremal dependence over an extended period of time, we further introduce an annual extremal network to explore the year-to-year temporal variation of extremal connections. We illustrate the χ and the annual extremal networks by analyzing the hurricane season maximum precipitation at the US Gulf Coast and surrounding area. Analysis suggests there exists long distance extremal dependence for precipitation extremes in the study region and the strength of the extremal dependence may depend on some regional scale meteorological conditions, for example, sea surface temperature.

Original languageEnglish (US)
Pages (from-to)484-501
Number of pages18
JournalJournal of Agricultural, Biological, and Environmental Statistics
Volume24
Issue number3
DOIs
StatePublished - Sep 15 2019

Fingerprint

Hurricanes
Precipitation (meteorology)
Risk assessment
hurricane
Annual
Coastal zones
risk assessment
temporal variation
sea surface temperature
Cyclonic Storms
coast
hurricanes
environmental assessment
Oceans and Seas
surface temperature
tail
coasts
Temperature
Dependence Structure
Estimator

Keywords

  • External dependence
  • Hurricanes
  • Networks
  • Precipitation
  • Spatial extremes

Cite this

New Exploratory Tools for Extremal Dependence : χ Networks and Annual Extremal Networks. / Huang, Whitney K.; Cooley, Daniel S.; Ebert-Uphoff, Imme; Chen, Chen; Chatterjee, Snigdhansu B.

In: Journal of Agricultural, Biological, and Environmental Statistics, Vol. 24, No. 3, 15.09.2019, p. 484-501.

Research output: Contribution to journalArticle

@article{ea1e67fa7da3496ab4715ec9a9611a53,
title = "New Exploratory Tools for Extremal Dependence: χ Networks and Annual Extremal Networks",
abstract = "Understanding dependence structure among extreme values plays an important role in risk assessment in environmental studies. In this work, we propose the χ network and the annual extremal network for exploring the extremal dependence structure of environmental processes. A χ network is constructed by connecting pairs whose estimated upper tail dependence coefficient, χ^ , exceeds a prescribed threshold. We develop an initial χ network estimator, and we use a spatial block bootstrap to assess both the bias and variance of our estimator. We then develop a method to correct the bias of the initial estimator by incorporating the spatial structure in χ. In addition to the χ network, which assesses spatial extremal dependence over an extended period of time, we further introduce an annual extremal network to explore the year-to-year temporal variation of extremal connections. We illustrate the χ and the annual extremal networks by analyzing the hurricane season maximum precipitation at the US Gulf Coast and surrounding area. Analysis suggests there exists long distance extremal dependence for precipitation extremes in the study region and the strength of the extremal dependence may depend on some regional scale meteorological conditions, for example, sea surface temperature.",
keywords = "External dependence, Hurricanes, Networks, Precipitation, Spatial extremes",
author = "Huang, {Whitney K.} and Cooley, {Daniel S.} and Imme Ebert-Uphoff and Chen Chen and Chatterjee, {Snigdhansu B}",
year = "2019",
month = "9",
day = "15",
doi = "10.1007/s13253-019-00356-4",
language = "English (US)",
volume = "24",
pages = "484--501",
journal = "Journal of Agricultural, Biological, and Environmental Statistics",
issn = "1085-7117",
publisher = "Springer New York",
number = "3",

}

TY - JOUR

T1 - New Exploratory Tools for Extremal Dependence

T2 - χ Networks and Annual Extremal Networks

AU - Huang, Whitney K.

AU - Cooley, Daniel S.

AU - Ebert-Uphoff, Imme

AU - Chen, Chen

AU - Chatterjee, Snigdhansu B

PY - 2019/9/15

Y1 - 2019/9/15

N2 - Understanding dependence structure among extreme values plays an important role in risk assessment in environmental studies. In this work, we propose the χ network and the annual extremal network for exploring the extremal dependence structure of environmental processes. A χ network is constructed by connecting pairs whose estimated upper tail dependence coefficient, χ^ , exceeds a prescribed threshold. We develop an initial χ network estimator, and we use a spatial block bootstrap to assess both the bias and variance of our estimator. We then develop a method to correct the bias of the initial estimator by incorporating the spatial structure in χ. In addition to the χ network, which assesses spatial extremal dependence over an extended period of time, we further introduce an annual extremal network to explore the year-to-year temporal variation of extremal connections. We illustrate the χ and the annual extremal networks by analyzing the hurricane season maximum precipitation at the US Gulf Coast and surrounding area. Analysis suggests there exists long distance extremal dependence for precipitation extremes in the study region and the strength of the extremal dependence may depend on some regional scale meteorological conditions, for example, sea surface temperature.

AB - Understanding dependence structure among extreme values plays an important role in risk assessment in environmental studies. In this work, we propose the χ network and the annual extremal network for exploring the extremal dependence structure of environmental processes. A χ network is constructed by connecting pairs whose estimated upper tail dependence coefficient, χ^ , exceeds a prescribed threshold. We develop an initial χ network estimator, and we use a spatial block bootstrap to assess both the bias and variance of our estimator. We then develop a method to correct the bias of the initial estimator by incorporating the spatial structure in χ. In addition to the χ network, which assesses spatial extremal dependence over an extended period of time, we further introduce an annual extremal network to explore the year-to-year temporal variation of extremal connections. We illustrate the χ and the annual extremal networks by analyzing the hurricane season maximum precipitation at the US Gulf Coast and surrounding area. Analysis suggests there exists long distance extremal dependence for precipitation extremes in the study region and the strength of the extremal dependence may depend on some regional scale meteorological conditions, for example, sea surface temperature.

KW - External dependence

KW - Hurricanes

KW - Networks

KW - Precipitation

KW - Spatial extremes

UR - http://www.scopus.com/inward/record.url?scp=85062030039&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062030039&partnerID=8YFLogxK

U2 - 10.1007/s13253-019-00356-4

DO - 10.1007/s13253-019-00356-4

M3 - Article

AN - SCOPUS:85062030039

VL - 24

SP - 484

EP - 501

JO - Journal of Agricultural, Biological, and Environmental Statistics

JF - Journal of Agricultural, Biological, and Environmental Statistics

SN - 1085-7117

IS - 3

ER -