Attention-Based Wildland Fire Spread Modeling Using Fire-Tracking Satellite Observations
Modeling the spread of wildland fires is essential for assessing and managing fire risks. However, this task remains challenging due to the partially stochastic nature of fire behavior and the limited availability of observational data with high spatial and temporal resolutions. Herein, we propose a...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-07-01
|
| Series: | Fire |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2571-6255/6/8/289 |
| _version_ | 1797584796710338560 |
|---|---|
| author | Yufei Zou Mojtaba Sadeghi Yaling Liu Alexandra Puchko Son Le Yang Chen Niels Andela Pierre Gentine |
| author_facet | Yufei Zou Mojtaba Sadeghi Yaling Liu Alexandra Puchko Son Le Yang Chen Niels Andela Pierre Gentine |
| author_sort | Yufei Zou |
| collection | DOAJ |
| description | Modeling the spread of wildland fires is essential for assessing and managing fire risks. However, this task remains challenging due to the partially stochastic nature of fire behavior and the limited availability of observational data with high spatial and temporal resolutions. Herein, we propose an attention-based deep learning modeling approach that can be used to learn the complex behaviors of wildfires across different fire-prone regions. We integrate optimized spatial and channel attention modules with a convolutional neural network (CNN) modeling architecture and train the attention-based fire spread models using a recently derived fire-tracking satellite observational dataset in conjunction with corresponding fuel, terrain, and weather conditions. The evaluation results and their comparison with benchmark models, such as a deeper and more complex autoencoder model and the semi-empirical FARSITE fire behavior model, demonstrate the effectiveness of the attention-based models. These new data-driven fire spread models exhibit promising modeling performances in both the next-step prediction (i.e., predicting fire progression from one timestep earlier) and recursive prediction (i.e., recursively predicting final fire perimeters from initial ignition points) of observed large wildfires in California, and they provide a foundation for further practical applications including short-term active fire spread prediction and long-term fire risk assessment. |
| first_indexed | 2024-03-10T23:57:47Z |
| format | Article |
| id | doaj.art-e856112838494e8bb0b34543cff9f9c0 |
| institution | Directory Open Access Journal |
| issn | 2571-6255 |
| language | English |
| last_indexed | 2024-03-10T23:57:47Z |
| publishDate | 2023-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Fire |
| spelling | doaj.art-e856112838494e8bb0b34543cff9f9c02023-11-19T01:03:06ZengMDPI AGFire2571-62552023-07-016828910.3390/fire6080289Attention-Based Wildland Fire Spread Modeling Using Fire-Tracking Satellite ObservationsYufei Zou0Mojtaba Sadeghi1Yaling Liu2Alexandra Puchko3Son Le4Yang Chen5Niels Andela6Pierre Gentine7Our Kettle Inc., Kensington, CA 94707, USAOur Kettle Inc., Kensington, CA 94707, USAOur Kettle Inc., Kensington, CA 94707, USAOur Kettle Inc., Kensington, CA 94707, USAOur Kettle Inc., Kensington, CA 94707, USADepartment of Earth System Science, University of California, Irvine, CA 92697, USASchool of Earth and Environmental Sciences, Cardiff University, Cardiff CF10 3AT, UKDepartment of Earth and Environmental Engineering, Columbia University, New York, NY 10027, USAModeling the spread of wildland fires is essential for assessing and managing fire risks. However, this task remains challenging due to the partially stochastic nature of fire behavior and the limited availability of observational data with high spatial and temporal resolutions. Herein, we propose an attention-based deep learning modeling approach that can be used to learn the complex behaviors of wildfires across different fire-prone regions. We integrate optimized spatial and channel attention modules with a convolutional neural network (CNN) modeling architecture and train the attention-based fire spread models using a recently derived fire-tracking satellite observational dataset in conjunction with corresponding fuel, terrain, and weather conditions. The evaluation results and their comparison with benchmark models, such as a deeper and more complex autoencoder model and the semi-empirical FARSITE fire behavior model, demonstrate the effectiveness of the attention-based models. These new data-driven fire spread models exhibit promising modeling performances in both the next-step prediction (i.e., predicting fire progression from one timestep earlier) and recursive prediction (i.e., recursively predicting final fire perimeters from initial ignition points) of observed large wildfires in California, and they provide a foundation for further practical applications including short-term active fire spread prediction and long-term fire risk assessment.https://www.mdpi.com/2571-6255/6/8/289wildfire modelingdeep learningremote sensingfire risk assessment |
| spellingShingle | Yufei Zou Mojtaba Sadeghi Yaling Liu Alexandra Puchko Son Le Yang Chen Niels Andela Pierre Gentine Attention-Based Wildland Fire Spread Modeling Using Fire-Tracking Satellite Observations Fire wildfire modeling deep learning remote sensing fire risk assessment |
| title | Attention-Based Wildland Fire Spread Modeling Using Fire-Tracking Satellite Observations |
| title_full | Attention-Based Wildland Fire Spread Modeling Using Fire-Tracking Satellite Observations |
| title_fullStr | Attention-Based Wildland Fire Spread Modeling Using Fire-Tracking Satellite Observations |
| title_full_unstemmed | Attention-Based Wildland Fire Spread Modeling Using Fire-Tracking Satellite Observations |
| title_short | Attention-Based Wildland Fire Spread Modeling Using Fire-Tracking Satellite Observations |
| title_sort | attention based wildland fire spread modeling using fire tracking satellite observations |
| topic | wildfire modeling deep learning remote sensing fire risk assessment |
| url | https://www.mdpi.com/2571-6255/6/8/289 |
| work_keys_str_mv | AT yufeizou attentionbasedwildlandfirespreadmodelingusingfiretrackingsatelliteobservations AT mojtabasadeghi attentionbasedwildlandfirespreadmodelingusingfiretrackingsatelliteobservations AT yalingliu attentionbasedwildlandfirespreadmodelingusingfiretrackingsatelliteobservations AT alexandrapuchko attentionbasedwildlandfirespreadmodelingusingfiretrackingsatelliteobservations AT sonle attentionbasedwildlandfirespreadmodelingusingfiretrackingsatelliteobservations AT yangchen attentionbasedwildlandfirespreadmodelingusingfiretrackingsatelliteobservations AT nielsandela attentionbasedwildlandfirespreadmodelingusingfiretrackingsatelliteobservations AT pierregentine attentionbasedwildlandfirespreadmodelingusingfiretrackingsatelliteobservations |