Assessing Wildfire Drivers in the Marshall Fire Using Remote Sensing and Random Forest
1. College of Earthquake Engineering and Architectural Safety, University of Emergency Management, Langfang 065201, China
2. School of Computer Science and Engineering, Chongqing University of Science and Technology, Chongqing 401331, China
| Abstract: | This study focuses on the 2021 Marshall Fire occurring in the Wildland-Urban Interface (WUI) of Colorado, USA, aiming to quantify fire driving factors and devise targeted mitigation strategies. Based on multi-source datasets including ERA5 meteorological reanalysis, MODIS NDVI, SRTM terrain and NLCD land cover products, nine environmental driving factors covering topography, meteorology and vegetation were selected. A random forest model was adopted to quantify the contribution of each factor to the spatial distribution of wildfire and divide fire risk grades for differentiated risk reduction management. The model attained an overall classification accuracy of 93.7%, revealing favorable simulation performance. Topographic elevation served as the dominant driving factor with an importance value of 0.452, followed by relative humidity (0.140) and wind speed (0.116). Burned regions were characterized by higher elevation, lower ambient relative humidity and steeper terrain gradients. Although the regional averaged wind speed was marginally higher in unburned zones, valley-induced localized extreme gusts made wind speed a vital fire-controlling factor. According to factor spatial differentiation, the study partitions the study area into high, medium and low fire risk zones and proposes corresponding risk reduction measures. The research results can support wildfire risk mapping, risk zoning and differentiated prevention & risk reduction management for North American WUI regions. |
| Keywords: | Wildland-Urban Interface Wildfire; Random Forest; Driving Factor; Marshall Fire; Risk Zoning; Disaster Reduction |
| DOI: | 10.57237/j.se.2026.03.001 |
View:
Downloads:
Received: 8 June 2026
Accepted: 22 June 2026
Published Online: 20 July 2026
References
| [1] | Holden, Z. A.; Morgan, P.; Crimmins, M. A.; et al. Wildfire susceptibility and topographic controls. For. Ecol. Manage. 2009, 258, 1275–1283. https://doi.org/10.1016/j.foreco.2009.06.021 |
| [2] | Povak, N. A.; Hessburg, P. F.; Salter, R. B. Evidence for scale-dependent topographic controls on wildfire spread. Ecosphere, 2018, 9(10): e02443. https://doi.org/10.1002/ecs2.2443 |
| [3] | Jolly, W. M.; Cochrane, M. A.; Freeborn, P. H.; et al. Climate-induced variations in global wildfire danger. Nat. Commun. 2015, 6, 7537. https://doi.org/10.1038/ncomms8537 |
| [4] | Abatzoglou, J. T.; Williams, A. P. Impact of anthropogenic climate change on wildfire across western US forests. Proc. Natl. Acad. Sci. USA 2016, 113, 11770–11775. https://doi.org/10.1073/pnas.1607171113 |
| [5] | Chuvieco, E.; Aguado, I.; Yebra, M.; et al. Development of a Framework for Fire Risk Assessment Using Remote Sensing and Geographic Information System Technologies. Ecological Modelling, 2010, 221(1): 46–58. https://doi.org/10.1016/j.ecolmodel.2008.11.017 |
| [6] | Breiman, L. Random Forests. Mach. Learn. 2001, 45, 5–32. https://doi.org/10.1023/A:1010933404324 |
| [7] | Oliveira, S.; Oehler, F.; San-Miguel-Ayanz, J.; Camia, A.; Pereira, J. M. C. Modeling spatial patterns of fire occurrence in Mediterranean Europe using Multiple Regression and Random Forest. Forest Ecology and Management, 2012, 275: 117–129. https://doi.org/10.1016/j.foreco.2012.03.003 |
| [8] | Giglio, L.; Schroeder, W.; Justice, C. O. The collection 6 MODIS active fire detection algorithm and fire products. Remote Sensing of Environment, 2016, 178: 31–41. https://doi.org/10.1016/j.rse.2016.02.054 |
| [9] | Van Wagtendonk, J. W.; Root, R. R.; Key, C. H. Comparison of AVIRIS and Landsat ETM+ detection capabilities for burn severity. Remote Sensing of Environment, 2004, 92(3): 397–408. https://doi.org/10.1016/j.rse.2003.12.015 |
| [10] | Eidenshink, J.; Schwind, B.; Brewer, K.; et al. A Project for Monitoring Trends in Burn Severity. Fire Ecology, 2007, 3(1): 3–21. https://doi.org/10.4996/fireecology.0301003 |
| [11] | Breiman, L. Statistical Modeling: The Two Cultures. Statistical Science, 2001, 16(3): 199–231. https://doi.org/10.1214/ss/1009213726 |
| [12] | Syphard, A. D.; Brennan, T. J.; Keeley, J. E. The role of defensible space for residential structure protection during wildfires. International Journal of Wildland Fire, 2014, 23(8): 1165–1175. https://doi.org/10.1071/WF13158 |
| [13] | Finney, M. A. The Challenge of Quantitative Risk Analysis for Wildland Fire. Forest Ecology and Management, 2005, 211(1–2): 97–108. https://doi.org/10.1016/j.foreco.2005.02.010 |
| [14] | Calkin, D. E.; Cohen, J. D.; Finney, M. A.; Thompson, M. P. How risk management can prevent future wildfire disasters in the wildland–urban interface. Proceedings of the National Academy of Sciences, 2014, 111(2): 746–751. https://doi.org/10.1073/pnas.1315088111 |
| [15] | Stephens, S. L.; Collins, B. M.; Biber, E.; Fulé, P. Z. U.S. federal fire and forest policy. https://doi.org/10.1890/120116 |
Join the Editorial Board
We invite active, qualified and high profile scientists and researchers to join as Editorial Board Members.
Join UsBecome a Reviewer
Scholars with a strong interest in reviewing are invited to join the reviewer panel to ensure the quality of the research to be published.
Join Us