A space-time conditional intensity model for evaluating a wildfire hazard index

Roger Peng, Frederic Paik Schoenberg, James A. Woods

Research output: Contribution to journalArticle

Abstract

Numerical indices are commonly used as tools to aid wildfire management and hazard assessment. Although the use of such indices is widespread, assessment of these indices in their respective regions of application is rare. We evaluate the effectiveness of the burning index (BI) for predicting wildfire occurrences in Los Angeles County, California using space-time point-process models. These models are based on an additive decomposition of the conditional intensity, with separate terms used to describe spatial and seasonal variability as well as contributions from the BI. We fit the models to wildfire and BI data from the years 1976-2000 using a combination of nonparametric kernel-smoothing methods and parametric maximum likelihood. In addition to using the Akaike information criterion (AIC) to compare competing models, we use new multidimensional residual methods based on approximate random thinning and rescaling to detect departures from the models and to ascertain the precise contribution of the BI to predicting wildfire occurrence. We find that although the BI appears to have a positive impact on wildfire prediction, the contribution is relatively small after taking into account natural seasonal and spatial variation. In particular, the BI does not appear to take into account increased activity during the years 1979-1981 and can overpredict during the early months of the year.

Original languageEnglish (US)
Pages (from-to)26-35
Number of pages10
JournalJournal of the American Statistical Association
Volume100
Issue number469
DOIs
StatePublished - Mar 2005

Fingerprint

Hazard
Space-time
Model
Nonparametric Smoothing
Wildfire
Kernel Smoothing
Akaike Information Criterion
Smoothing Methods
Thinning
Kernel Methods
Rescaling
Point Process
Process Model
Maximum Likelihood
Decompose
Evaluate
Prediction
Term

Keywords

  • Conditional intensity model
  • Model evaluation
  • Point process residual analysis
  • Random rescaling
  • Random thinning
  • Wildfire risk

ASJC Scopus subject areas

  • Mathematics(all)
  • Statistics and Probability

Cite this

A space-time conditional intensity model for evaluating a wildfire hazard index. / Peng, Roger; Schoenberg, Frederic Paik; Woods, James A.

In: Journal of the American Statistical Association, Vol. 100, No. 469, 03.2005, p. 26-35.

Research output: Contribution to journalArticle

Peng, Roger ; Schoenberg, Frederic Paik ; Woods, James A. / A space-time conditional intensity model for evaluating a wildfire hazard index. In: Journal of the American Statistical Association. 2005 ; Vol. 100, No. 469. pp. 26-35.
@article{394af8ba46d64f5a88c63f0529e432fb,
title = "A space-time conditional intensity model for evaluating a wildfire hazard index",
abstract = "Numerical indices are commonly used as tools to aid wildfire management and hazard assessment. Although the use of such indices is widespread, assessment of these indices in their respective regions of application is rare. We evaluate the effectiveness of the burning index (BI) for predicting wildfire occurrences in Los Angeles County, California using space-time point-process models. These models are based on an additive decomposition of the conditional intensity, with separate terms used to describe spatial and seasonal variability as well as contributions from the BI. We fit the models to wildfire and BI data from the years 1976-2000 using a combination of nonparametric kernel-smoothing methods and parametric maximum likelihood. In addition to using the Akaike information criterion (AIC) to compare competing models, we use new multidimensional residual methods based on approximate random thinning and rescaling to detect departures from the models and to ascertain the precise contribution of the BI to predicting wildfire occurrence. We find that although the BI appears to have a positive impact on wildfire prediction, the contribution is relatively small after taking into account natural seasonal and spatial variation. In particular, the BI does not appear to take into account increased activity during the years 1979-1981 and can overpredict during the early months of the year.",
keywords = "Conditional intensity model, Model evaluation, Point process residual analysis, Random rescaling, Random thinning, Wildfire risk",
author = "Roger Peng and Schoenberg, {Frederic Paik} and Woods, {James A.}",
year = "2005",
month = "3",
doi = "10.1198/016214504000001763",
language = "English (US)",
volume = "100",
pages = "26--35",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor and Francis Ltd.",
number = "469",

}

TY - JOUR

T1 - A space-time conditional intensity model for evaluating a wildfire hazard index

AU - Peng, Roger

AU - Schoenberg, Frederic Paik

AU - Woods, James A.

PY - 2005/3

Y1 - 2005/3

N2 - Numerical indices are commonly used as tools to aid wildfire management and hazard assessment. Although the use of such indices is widespread, assessment of these indices in their respective regions of application is rare. We evaluate the effectiveness of the burning index (BI) for predicting wildfire occurrences in Los Angeles County, California using space-time point-process models. These models are based on an additive decomposition of the conditional intensity, with separate terms used to describe spatial and seasonal variability as well as contributions from the BI. We fit the models to wildfire and BI data from the years 1976-2000 using a combination of nonparametric kernel-smoothing methods and parametric maximum likelihood. In addition to using the Akaike information criterion (AIC) to compare competing models, we use new multidimensional residual methods based on approximate random thinning and rescaling to detect departures from the models and to ascertain the precise contribution of the BI to predicting wildfire occurrence. We find that although the BI appears to have a positive impact on wildfire prediction, the contribution is relatively small after taking into account natural seasonal and spatial variation. In particular, the BI does not appear to take into account increased activity during the years 1979-1981 and can overpredict during the early months of the year.

AB - Numerical indices are commonly used as tools to aid wildfire management and hazard assessment. Although the use of such indices is widespread, assessment of these indices in their respective regions of application is rare. We evaluate the effectiveness of the burning index (BI) for predicting wildfire occurrences in Los Angeles County, California using space-time point-process models. These models are based on an additive decomposition of the conditional intensity, with separate terms used to describe spatial and seasonal variability as well as contributions from the BI. We fit the models to wildfire and BI data from the years 1976-2000 using a combination of nonparametric kernel-smoothing methods and parametric maximum likelihood. In addition to using the Akaike information criterion (AIC) to compare competing models, we use new multidimensional residual methods based on approximate random thinning and rescaling to detect departures from the models and to ascertain the precise contribution of the BI to predicting wildfire occurrence. We find that although the BI appears to have a positive impact on wildfire prediction, the contribution is relatively small after taking into account natural seasonal and spatial variation. In particular, the BI does not appear to take into account increased activity during the years 1979-1981 and can overpredict during the early months of the year.

KW - Conditional intensity model

KW - Model evaluation

KW - Point process residual analysis

KW - Random rescaling

KW - Random thinning

KW - Wildfire risk

UR - http://www.scopus.com/inward/record.url?scp=14944365234&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=14944365234&partnerID=8YFLogxK

U2 - 10.1198/016214504000001763

DO - 10.1198/016214504000001763

M3 - Article

AN - SCOPUS:14944365234

VL - 100

SP - 26

EP - 35

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 469

ER -