Enhancing prediction of wildfire occurrence and behavior in Alaska using spatio-temporal clustering and ensemble machine learning
Wildfires are an integral part of Alaska’s ecological landscape, shaping its boreal forests and tundra. However, recent shifts in wildfire frequency, intensity, and seasonality pose unprecedented challenges for fire management in Alaska’s remote and ecologically vulnerable regions. This study addres...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-03-01
|
| Series: | Ecological Informatics |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1574954124005053 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832595374020231168 |
|---|---|
| author | A. Ahajjam M. Allgaier R. Chance E. Chukwuemeka J. Putkonen T. Pasch |
| author_facet | A. Ahajjam M. Allgaier R. Chance E. Chukwuemeka J. Putkonen T. Pasch |
| author_sort | A. Ahajjam |
| collection | DOAJ |
| description | Wildfires are an integral part of Alaska’s ecological landscape, shaping its boreal forests and tundra. However, recent shifts in wildfire frequency, intensity, and seasonality pose unprecedented challenges for fire management in Alaska’s remote and ecologically vulnerable regions. This study addresses the challenge of wildfire occurrence and behavior prediction in Alaska by developing a comprehensive framework that leverages satellite-based data, geospatial features, advanced optimization, and machine learning (ML). First, NASA’s Fire Information for Resource Management System (FIRMS) dataset spanning +20 years is processed using a spatio-temporal clustering algorithm to create refined wildfire datasets. A sequential Genetic Algorithm (GA) is employed for cost-effective feature selection from 49 geospatial features, including remote sensing and reanalysis data. Histogram Gradient Boosting (HistGB) is then used for predictive modeling of wildfire occurrence, burnt area, and wildfire duration. This ensemble model’s performance is benchmarked across four prediction horizons (same-day, +7 days, +30 days, +90 days) and against various conventional ML and deep learning techniques. Results highlight key factors influencing wildfire dynamics in Alaska and demonstrate substantial improvements in prediction accuracy (e.g., an average improvement of 72.62% in wildfire occurrence accuracy regardless of prediction horizon), offering valuable insights for risk assessment and resource allocation in wildfire management in Alaska. |
| format | Article |
| id | doaj-art-0c7e2cda728840f89fd0c7474833f434 |
| institution | Kabale University |
| issn | 1574-9541 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Ecological Informatics |
| spelling | doaj-art-0c7e2cda728840f89fd0c7474833f4342025-01-19T06:24:40ZengElsevierEcological Informatics1574-95412025-03-0185102963Enhancing prediction of wildfire occurrence and behavior in Alaska using spatio-temporal clustering and ensemble machine learningA. Ahajjam0M. Allgaier1R. Chance2E. Chukwuemeka3J. Putkonen4T. Pasch5School of Electrical Engineering and Computer Science, University of North Dakota, Upson Hall I, Grand Forks, 58202-7165, ND, USA; Corresponding author.Department of Physics & Astrophysics, University of North Dakota, Witmer Hall, Grand Forks, 58202-7129, ND, USAHarold Hamm School of Geology and Geologic Engineering, University of North Dakota, Leonard Hall, Grand Forks, 58202-8358, ND, USAResearch Institute for Autonomous System, University of North Dakota, 4201 James Ray Dr, Grand Forks, 58202, ND, USAHarold Hamm School of Geology and Geologic Engineering, University of North Dakota, Leonard Hall, Grand Forks, 58202-8358, ND, USADepartment of Communication, University of North Dakota, O’Kelly Hall, Grand Forks, 58202, ND, USAWildfires are an integral part of Alaska’s ecological landscape, shaping its boreal forests and tundra. However, recent shifts in wildfire frequency, intensity, and seasonality pose unprecedented challenges for fire management in Alaska’s remote and ecologically vulnerable regions. This study addresses the challenge of wildfire occurrence and behavior prediction in Alaska by developing a comprehensive framework that leverages satellite-based data, geospatial features, advanced optimization, and machine learning (ML). First, NASA’s Fire Information for Resource Management System (FIRMS) dataset spanning +20 years is processed using a spatio-temporal clustering algorithm to create refined wildfire datasets. A sequential Genetic Algorithm (GA) is employed for cost-effective feature selection from 49 geospatial features, including remote sensing and reanalysis data. Histogram Gradient Boosting (HistGB) is then used for predictive modeling of wildfire occurrence, burnt area, and wildfire duration. This ensemble model’s performance is benchmarked across four prediction horizons (same-day, +7 days, +30 days, +90 days) and against various conventional ML and deep learning techniques. Results highlight key factors influencing wildfire dynamics in Alaska and demonstrate substantial improvements in prediction accuracy (e.g., an average improvement of 72.62% in wildfire occurrence accuracy regardless of prediction horizon), offering valuable insights for risk assessment and resource allocation in wildfire management in Alaska.http://www.sciencedirect.com/science/article/pii/S1574954124005053WildfiresAlaskaPermafrost lanscapeMachine learningDeep learningFeature selection |
| spellingShingle | A. Ahajjam M. Allgaier R. Chance E. Chukwuemeka J. Putkonen T. Pasch Enhancing prediction of wildfire occurrence and behavior in Alaska using spatio-temporal clustering and ensemble machine learning Ecological Informatics Wildfires Alaska Permafrost lanscape Machine learning Deep learning Feature selection |
| title | Enhancing prediction of wildfire occurrence and behavior in Alaska using spatio-temporal clustering and ensemble machine learning |
| title_full | Enhancing prediction of wildfire occurrence and behavior in Alaska using spatio-temporal clustering and ensemble machine learning |
| title_fullStr | Enhancing prediction of wildfire occurrence and behavior in Alaska using spatio-temporal clustering and ensemble machine learning |
| title_full_unstemmed | Enhancing prediction of wildfire occurrence and behavior in Alaska using spatio-temporal clustering and ensemble machine learning |
| title_short | Enhancing prediction of wildfire occurrence and behavior in Alaska using spatio-temporal clustering and ensemble machine learning |
| title_sort | enhancing prediction of wildfire occurrence and behavior in alaska using spatio temporal clustering and ensemble machine learning |
| topic | Wildfires Alaska Permafrost lanscape Machine learning Deep learning Feature selection |
| url | http://www.sciencedirect.com/science/article/pii/S1574954124005053 |
| work_keys_str_mv | AT aahajjam enhancingpredictionofwildfireoccurrenceandbehaviorinalaskausingspatiotemporalclusteringandensemblemachinelearning AT mallgaier enhancingpredictionofwildfireoccurrenceandbehaviorinalaskausingspatiotemporalclusteringandensemblemachinelearning AT rchance enhancingpredictionofwildfireoccurrenceandbehaviorinalaskausingspatiotemporalclusteringandensemblemachinelearning AT echukwuemeka enhancingpredictionofwildfireoccurrenceandbehaviorinalaskausingspatiotemporalclusteringandensemblemachinelearning AT jputkonen enhancingpredictionofwildfireoccurrenceandbehaviorinalaskausingspatiotemporalclusteringandensemblemachinelearning AT tpasch enhancingpredictionofwildfireoccurrenceandbehaviorinalaskausingspatiotemporalclusteringandensemblemachinelearning |