You are on page 1of 2

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/372066793

Engineering Applications of Artificial Intelligence Development of novel


optimized deep learning algorithms for wildfire modeling: A case study of
Maui, Hawai'i

Article · January 2023


DOI: 10.1016/j.engappai.2023.106699

CITATIONS READS

0 10

7 authors, including:

Fatemeh Rezaie Mahdi Panahi


University of Science and Technology, Korea Yonsei University
59 PUBLICATIONS   1,995 CITATIONS    88 PUBLICATIONS   5,423 CITATIONS   

SEE PROFILE SEE PROFILE

Saro Lee
Korea Institute of Geoscience and Mineral Resources
284 PUBLICATIONS   22,364 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Determining of suitability of existing schools and site selection of new school buildings using GIS and metaheuristic algorithm. View project

gis and rs View project

All content following this page was uploaded by Mahdi Panahi on 04 July 2023.

The user has requested enhancement of the downloaded file.


Engineering Applications of Artificial Intelligence 125 (2023) 106699

Contents lists available at ScienceDirect

Engineering Applications of Artificial Intelligence


journal homepage: www.elsevier.com/locate/engappai

Development of novel optimized deep learning algorithms for wildfire


modeling: A case study of Maui, Hawai‘i
Fatemeh Rezaie a,b,c ,∗, Mahdi Panahi c,d , Sayed M. Bateni c , Saro Lee a,b , Changhyun Jun e ,∗∗,
Clay Trauernicht f , Christopher M.U. Neale g
a
Geoscience Data Center, Korea Institute of Geoscience and Mineral Resources (KIGAM), 124, Gwahak-ro, Yuseong-gu, Daejeon 34132, Republic of Korea
b
Department of Geophysical Exploration, Korea University of Science and Technology, 217 Gajeong-ro, Yuseong-gu, Daejeon 34113, Republic of Korea
c Department of Civil and Environmental Engineering and Water Resources Research Center, University of Hawai‘i at Manoa, Honolulu, HI, 96822, USA
d Department of Computer Engineering, Chosun University, Gwangju, 61452, Republic of Korea
e Department of Civil and Environmental Engineering, College of Engineering, Chung-Ang University, Seoul 06974, Republic of Korea
f
Department of Natural Resources and Environmental Management, University of Hawai‘i at Manoa, Honolulu, HI, 96822, USA
g
Daugherty Water for Food Global Institute, University of Nebraska, Lincoln, NE, USA

ARTICLE INFO ABSTRACT


Keywords: To address the growing global concern regarding increased wildfire occurrences and their widespread socio-
Wildfire susceptibility map ecological impacts, cost-effective and practical approaches must be urgently identified to accurately predict the
Convolutional neural network probability of wildfire incidents. The objective of this study was to develop deep learning models to estimate
Group method of data handling
the likelihood of wildfire incidents and compare the predictive capability of the methods. To this end, the
Ant colony optimization
group method of data handling (GMDH) and convolutional neural network (CNN) algorithms were coupled
Hawai‘i
with the biogeography-based optimization (BBO) and ant colony optimization (ACO) algorithms to enhance the
predictive performance of the models. Overall, 1745 historical wildfires were identified in the island of Maui,
Hawai‘i. Among them, 1221 events (70%) were randomly selected to generate the models, and the remaining
524 (30%) were used for validation. The frequency ratio, information gain ratio (IGR), and variance inflation
factor methods were used to select the optimal number of predictor variables for model development. 13
influencing factors: elevation, aspect, slope, plan curvature, slope length, valley depth, topographic wetness
index, mean annual wind speed, mean annual air temperature, mean monthly rainfall, distance to the road,
distance to the river, and normalized difference vegetation index were selected to generate the proposed
models and detect fire-prone areas. The IGR values indicated that the key parameters for predicting wildfire
susceptibility were the distance to the road, mean annual air temperature, elevation, and slope. Finally, the
area under the receiver operating characteristic curve (AUROC) was computed to verify the reliability and
accuracy of the wildfire susceptibility maps. Both the optimization algorithms enhanced the performances
of the GMDH and CNN models. The ACO most notably enhanced the CNN performance compared with the
models (AUROCTraining = 0.889 and AUROCTesting = 0.885). The findings demonstrated the potential of coupled
models in overcoming the limitations of individual models for mapping fire-susceptible areas and analyzing
the multifactorial aspects that lead to wildfire occurrence. Overall, the proposed frameworks can be used to
predict the spatio-temporal patterns of wildfire occurrence, which are of significance for land management,
wildfire prevention, and mitigation of wildfire consequences.

1. Introduction fire-promoting) invasive species instead of native plants. Furthermore,


wildfires affect the water quality, greenhouse gas emissions, surface
In the last decade, the wildfire frequency and intensity has increased erosion and runoff, soil properties, biodiversity, biogeochemical cycle,
exponentially, resulting in significant changes in large surfaces of nat- vegetation succession, and wildlife habitats at the local and regional
ural and human-made landscapes. Moreover, the ecosystem has been scales (Cai et al., 2021; Xie et al., 2022; Xu et al., 2020; Xue et al.,
disrupted owing to the enhanced spreading of fire-adapted (and often 2022). All these changes negatively alter the soil composition, threaten

∗ Corresponding author at: Geoscience Data Center, Korea Institute of Geoscience and Mineral Resources (KIGAM), 124, Gwahak-ro, Yuseong-gu, Daejeon
34132, Republic of Korea.
∗∗ Corresponding author.
E-mail addresses: rezaie@kigam.re.kr (F. Rezaie), mpanahi@hawaii.edu (M. Panahi), smbateni@hawaii.edu (S.M. Bateni), leesaro@kigam.re.kr (S. Lee),
cjun@cau.ac.kr (C. Jun), trauerni@hawaii.edu (C. Trauernicht), cneale2@unl.edu (C.M.U. Neale).

https://doi.org/10.1016/j.engappai.2023.106699
Received 6 February 2023; Received in revised form 8 May 2023; Accepted 19 June 2023
Available online xxxx
0952-1976/© 2023 Elsevier Ltd. All rights reserved.

View publication stats

You might also like