Fire Risk Assessment on Wildland–Urban Interface and Adjoined Urban Areas: Estimation Vegetation Ignitability by Artificial Neural Network

Maria Mahamed; Lea Wittenberg; Haim Kutiel; Anna Brook

doi:10.3390/fire5060184

Fire Risk Assessment on Wildland–Urban Interface and Adjoined Urban Areas: Estimation Vegetation Ignitability by Artificial Neural Network

Maria Mahamed, Lea Wittenberg, Haim Kutiel, Anna Brook

Department of Geography and Environmental Studies

Research output: Contribution to journal › Article › peer-review

Abstract

Fire risk assessment on the wildland–urban interface (WUI) and adjoined urban areas is crucial to prevent human losses and structural damages. One of many interacting and dynamic factors influencing the structure and function of fire-prone ecosystems is vegetation ignitability, which plays a significant role in spreading fire. This study sought to identify areas with a high-level probability of ignition from time series multispectral images by designing a pattern recognition neural network (PRNN). The temporal behavior of six vegetation indices (VIs) before the considered wildfire event provided the input data for the PRNN. In total, we tested eight combinations of inputs for PRNN: the temporal behavior of each chosen VI, the temporal behavior of all indices together, and the values of VIs at specific dates selected based on factor analysis. The reference output data for training was a map of areas ignited in the wildfire. Among the considered inputs, the MSAVI dataset, which reflects changes in vegetation biomass and canopy cover, showed the best performance. The precision of the presented PRNN (RMSE = 0.85) in identification areas with a high potential of ignitability gives ground for the application of the proposed method in risk assessment and fuel treatment planning on WUI and adjoined urban areas.

Original language	English
Article number	184
Journal	Fire
Volume	5
Issue number	6
DOIs	https://doi.org/10.3390/fire5060184
State	Published - Dec 2022

Bibliographical note

Funding Information:
C. Bonanad declares that she has received funding for attendance at meetings and conferences, and speaker fees from BMS/Pfizer, Daiichi-Sankyo, Boehringer and Bayer. F. Formiga received funding for attendance at meetings and congresses, honoraria and funding from Pfizer for attending meetings and congresses; speaker fees from BMS/Pfizer, Daiichi-Sankyo, and Bayer; and funding for participating in research by BMS/Pfizer. M. Anguita declares that he has received speaker fees from BMS/Pfizer, Novartis, Bayer, Daiichi-Sankyo and Boehringer, and funding for participating in research projects from BMS/Pfizer, Abbot, Novartis, Bayer, and Daiichi-Sankyo. R. Petidier declares that he has received speaker fees from Bayer, BMS/Pfizer and Daiichy-Sankyo. A. Gullón declares that she has received speaker fees from BMS/Pfizer and Bayer and funding for participating in a research project from Daiichi-Sankyo.

Publisher Copyright:
© 2022 by the authors.

Keywords

artificial neural network
fire risk assessment
vegetation ignitability
wildland–urban interface

ASJC Scopus subject areas

Forestry
Building and Construction
Safety, Risk, Reliability and Quality
Environmental Science (miscellaneous)
Safety Research
Earth and Planetary Sciences (miscellaneous)

Access to Document

10.3390/fire5060184

Cite this

@article{865b22ad386044d4a2ab3c9350229f12,

title = "Fire Risk Assessment on Wildland–Urban Interface and Adjoined Urban Areas: Estimation Vegetation Ignitability by Artificial Neural Network",

abstract = "Fire risk assessment on the wildland–urban interface (WUI) and adjoined urban areas is crucial to prevent human losses and structural damages. One of many interacting and dynamic factors influencing the structure and function of fire-prone ecosystems is vegetation ignitability, which plays a significant role in spreading fire. This study sought to identify areas with a high-level probability of ignition from time series multispectral images by designing a pattern recognition neural network (PRNN). The temporal behavior of six vegetation indices (VIs) before the considered wildfire event provided the input data for the PRNN. In total, we tested eight combinations of inputs for PRNN: the temporal behavior of each chosen VI, the temporal behavior of all indices together, and the values of VIs at specific dates selected based on factor analysis. The reference output data for training was a map of areas ignited in the wildfire. Among the considered inputs, the MSAVI dataset, which reflects changes in vegetation biomass and canopy cover, showed the best performance. The precision of the presented PRNN (RMSE = 0.85) in identification areas with a high potential of ignitability gives ground for the application of the proposed method in risk assessment and fuel treatment planning on WUI and adjoined urban areas.",

keywords = "artificial neural network, fire risk assessment, vegetation ignitability, wildland–urban interface",

author = "Maria Mahamed and Lea Wittenberg and Haim Kutiel and Anna Brook",

note = "Funding Information: C. Bonanad declares that she has received funding for attendance at meetings and conferences, and speaker fees from BMS/Pfizer, Daiichi-Sankyo, Boehringer and Bayer. F. Formiga received funding for attendance at meetings and congresses, honoraria and funding from Pfizer for attending meetings and congresses; speaker fees from BMS/Pfizer, Daiichi-Sankyo, and Bayer; and funding for participating in research by BMS/Pfizer. M. Anguita declares that he has received speaker fees from BMS/Pfizer, Novartis, Bayer, Daiichi-Sankyo and Boehringer, and funding for participating in research projects from BMS/Pfizer, Abbot, Novartis, Bayer, and Daiichi-Sankyo. R. Petidier declares that he has received speaker fees from Bayer, BMS/Pfizer and Daiichy-Sankyo. A. Gull{\'o}n declares that she has received speaker fees from BMS/Pfizer and Bayer and funding for participating in a research project from Daiichi-Sankyo. Publisher Copyright: {\textcopyright} 2022 by the authors.",

year = "2022",

month = dec,

doi = "10.3390/fire5060184",

language = "English",

volume = "5",

journal = "Fire",

issn = "2571-6255",

publisher = "MDPI AG",

number = "6",

}

TY - JOUR

T1 - Fire Risk Assessment on Wildland–Urban Interface and Adjoined Urban Areas

T2 - Estimation Vegetation Ignitability by Artificial Neural Network

AU - Mahamed, Maria

AU - Wittenberg, Lea

AU - Kutiel, Haim

AU - Brook, Anna

N1 - Funding Information: C. Bonanad declares that she has received funding for attendance at meetings and conferences, and speaker fees from BMS/Pfizer, Daiichi-Sankyo, Boehringer and Bayer. F. Formiga received funding for attendance at meetings and congresses, honoraria and funding from Pfizer for attending meetings and congresses; speaker fees from BMS/Pfizer, Daiichi-Sankyo, and Bayer; and funding for participating in research by BMS/Pfizer. M. Anguita declares that he has received speaker fees from BMS/Pfizer, Novartis, Bayer, Daiichi-Sankyo and Boehringer, and funding for participating in research projects from BMS/Pfizer, Abbot, Novartis, Bayer, and Daiichi-Sankyo. R. Petidier declares that he has received speaker fees from Bayer, BMS/Pfizer and Daiichy-Sankyo. A. Gullón declares that she has received speaker fees from BMS/Pfizer and Bayer and funding for participating in a research project from Daiichi-Sankyo. Publisher Copyright: © 2022 by the authors.

PY - 2022/12

Y1 - 2022/12

N2 - Fire risk assessment on the wildland–urban interface (WUI) and adjoined urban areas is crucial to prevent human losses and structural damages. One of many interacting and dynamic factors influencing the structure and function of fire-prone ecosystems is vegetation ignitability, which plays a significant role in spreading fire. This study sought to identify areas with a high-level probability of ignition from time series multispectral images by designing a pattern recognition neural network (PRNN). The temporal behavior of six vegetation indices (VIs) before the considered wildfire event provided the input data for the PRNN. In total, we tested eight combinations of inputs for PRNN: the temporal behavior of each chosen VI, the temporal behavior of all indices together, and the values of VIs at specific dates selected based on factor analysis. The reference output data for training was a map of areas ignited in the wildfire. Among the considered inputs, the MSAVI dataset, which reflects changes in vegetation biomass and canopy cover, showed the best performance. The precision of the presented PRNN (RMSE = 0.85) in identification areas with a high potential of ignitability gives ground for the application of the proposed method in risk assessment and fuel treatment planning on WUI and adjoined urban areas.

AB - Fire risk assessment on the wildland–urban interface (WUI) and adjoined urban areas is crucial to prevent human losses and structural damages. One of many interacting and dynamic factors influencing the structure and function of fire-prone ecosystems is vegetation ignitability, which plays a significant role in spreading fire. This study sought to identify areas with a high-level probability of ignition from time series multispectral images by designing a pattern recognition neural network (PRNN). The temporal behavior of six vegetation indices (VIs) before the considered wildfire event provided the input data for the PRNN. In total, we tested eight combinations of inputs for PRNN: the temporal behavior of each chosen VI, the temporal behavior of all indices together, and the values of VIs at specific dates selected based on factor analysis. The reference output data for training was a map of areas ignited in the wildfire. Among the considered inputs, the MSAVI dataset, which reflects changes in vegetation biomass and canopy cover, showed the best performance. The precision of the presented PRNN (RMSE = 0.85) in identification areas with a high potential of ignitability gives ground for the application of the proposed method in risk assessment and fuel treatment planning on WUI and adjoined urban areas.

KW - artificial neural network

KW - fire risk assessment

KW - vegetation ignitability

KW - wildland–urban interface

UR - http://www.scopus.com/inward/record.url?scp=85144674523&partnerID=8YFLogxK

U2 - 10.3390/fire5060184

DO - 10.3390/fire5060184

M3 - Article

AN - SCOPUS:85144674523

SN - 2571-6255

VL - 5

JO - Fire

JF - Fire

IS - 6

M1 - 184

ER -

Fire Risk Assessment on Wildland–Urban Interface and Adjoined Urban Areas: Estimation Vegetation Ignitability by Artificial Neural Network

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this