University of Twente Student Theses
Fuel type aggregation for wildfire simulation optimization
Meer, Gosse Luurt van der (2025) Fuel type aggregation for wildfire simulation optimization.
|
PDF
19MB |
| Abstract: | Wildfire behaviour modelling plays an important role in the field of wildfire management and in conducting fire risk assessments. This research is focused on the influence fuel type aggregation on the accuracy of wildfire simulations outcomes, by using the Flammap fire behaviour modelling software. The FBFM40 fuel type classification (Scott & Burgan, 2005) has been systematically aggregated based on the following fire behaviour characteristics: fuel load, rate of spread, and flame lengths or no specific characteristic. The River Road East Fire, 2023, in the Lolo National Forest, Montana (USA), has been selected as the study area. This due to its representative boreal forest ecosystem and recent fire occurrence. This study uses the Minimum Travel Time (MTT) fire spread model to analyse changes in simulation accuracy across multiple levels of fuel type aggregation. The Sørensen Similarity Index (SSI) has been used to quantify the under- and over-simulation, while entropy levels have been calculated to evaluate the fuel type diversity left within the input data. The results indicate that fuel type aggregation mainly impacts the over-simulation, especially after the 6 aggregation steps and or when the entropy levels reached < 0.61. The rate of spread showed the greatest influence on the simulation accuracy. The findings suggest that maintaining an optimal level between the amount of unique fuel types left in the simulation and the fuel type diversity (entropy) is essential to balance the computational efficiency and simulation accuracy. This study shows potential risks of oversimplification for the input fuel type classification to use in fire behaviour modelling Future research should explore dynamic aggregation methods with more focus on small-scale differences as well as fire suppression efforts. Recommended would be further validation of entropy-based thresholds across diverse wildfire-prone regions or complete ecosystems. |
| Item Type: | Essay (Master) |
| Faculty: | ITC: Faculty of Geo-information Science and Earth Observation |
| Subject: | 38 earth sciences |
| Programme: | Spatial Engineering MSc (60962) |
| Link to this item: | https://purl.utwente.nl/essays/106110 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page