This abstract presents the current state of a new wildfire simulation model designed for operational and educational use in Switzerland. Although Central Europe has historically been considered a low-risk zone [1], Switzerland, particularly the Locarnese region of Canton Ticino, is currently experiencing an upward trend in wildfire probability [2]. Driven by climate change, this heightened risk profile is expected to increase the frequency of extreme events. Due to the country's rugged topography and weather patterns, Swiss wildfires are predominantly slope-driven, spreading rapidly through forest and shrubland vegetation on steep terrain [3]. To mitigate potential damage, reliable simulation tools are essential for optimizing fire management operations and suppression strategies [4]. Because wildfire behaviour is heavily environment dependent, established software packages, such as FARSITE [5] in the United States and Prometheus [6] in Canada, are highly specialized and parametrized for their native regions, necessitating a similarly tailored approach for the Swiss context.
Among other things, this work presents a novel wildfire simulation model engineered to overcome the computational challenges of high resolution geographic data. While standard models rely on grids with 25 to 50 meters, our approach utilizes regional datasets, such as swissALTI3D [7], featuring cell resolutions as high as 0.5 meters. The resulting framework satisfies a crucial need in capturing intricate terrain features while maintaining the rapid processing speeds required for operational deployment. To achieve this, the new framework, based on a custom implementation of Cell2Fire [8] as growth model, introduces an innovative method for extracting vector-based fire fronts from local spread ellipses, paired with a new time-of-arrival calculation. This architecture preserves rapid simulation speeds and native compatibility with parallel processing. Ultimately, the model yields performance gains over traditional approaches.
The proof-of-concept simulation, as shown in Fig. 1, was run on real data to show the general performance of the model without optimization. The simulation landscape utilizes a 1×1 km geographic tile sourced from swissALTI3D [7] data. At a 0.5-meter cell resolution, this translates to a 2000×2000 cell computational grid. The corresponding forest mix rate [9], which defines fuel availability and combustible cells, is illustrated in Fig. 2.
Future research will expand this prototype into a comprehensive framework by integrating a customized fire behaviour model, a dynamic weather simulation, and a Geographic Information System (GIS) for spatial visualization and fire suppression strategy evaluation. Currently under development, the new fire behaviour model accounts for the unique fuel, climate, and topographical variations of the Swiss Alps. It is based on the semi-empirical mathematical model proposed by Rothermel [10], incorporating subsequent refinements and regional dataset adjustments. The framework leverages a diverse array of high-resolution spatial data, including key inputs such as the swissALTI3D digital terrain model [7], the National Forest Inventory (NFI) vegetation height models [11], and Sentinel-based Vegetation Health Indices (VHI) [12].
Against the backdrop of escalating wildfire risks in Switzerland, this study provides the foundational computational infrastructure necessary to empower Swiss authorities with proactive decision-making and risk-management tools.
References
[1] J. Miller, A. Böhnisch, R. Ludwig, and M. I. Brunner, “Climate change impacts on regional fire weather in heterogeneous landscapes of central Europe,” Natural Hazards and Earth System Sciences, vol. 24, no. 2, pp. 411–428, 2024. [Online]. Available: https://nhess.copernicus.org/articles/24/411/2024/
[2] A. M. Bardsley, D. K. Bardsley, M. Conedera, M. P¨utz, G. M. Robinson, and D. Weber, “Wildfire, Environmental Risk and Deliberative Planning in the Locarnese Region of Switzerland,” Environmental Management, vol. 68, no. 6, pp. 785–801, Dec. 2021. [Online]. Available: https://doi.org/10.1007/s00267-021-01523-5
[3] M. Conedera, J. Feusi, G. B. Pezzatti, and P. Krebs, “Linking the future likelihood of large fires to occur on mountain slopes with fuel connectivity and topography,” Natural Hazards, vol. 120, no. 5, pp. 4657–4673, Mar. 2024. [Online]. Available: https://doi.org/10.1007/s11069-023-06395-y
[4] Voltolina, D., Cappellini, G., Apuani, T., Sterlacchini, S.: Pyros: a raster–vector spatial simulation model for predicting wildland surface fire spread and growth. International Journal of Wildland Fire 33(3) (2024). https://doi.org/10.1071/WF22142
[5] M. A. Finney, “FARSITE: Fire Area Simulator-model development and evaluation,” U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Tech. Rep., 1998. [Online]. Available: http://dx.doi.org/10.2737/RMRS-RP-4
[6] Tymstra, C., Bryce, R., Wotton, B., Taylor, S., Armitage, O.: Development and structure of Prometheus: the Canadian wildland fire growth simulation model. Natural Resources Canada, Canadian Forest Service, Northern Forestry Centre, Information Report NOR-X-417.(Edmonton, AB) (2010)
[7] Federal Office of Topography swisstopo, “swissALTI3D,” Aug. 2024. [Online]. Available: https://www.swisstopo.admin.ch/de/hoehenmodell-swissalti3d
[8] C. Pais, J. Carrasco, D. L. Martell, A. Weintraub, and D. L. Woodruff, “Cell2Fire: A Cell-Based Forest Fire Growth Model to Support Strategic Landscape Management Planning,” Frontiers in Forests and Global Change, vol. 4, 2021. [Online]. Available: https://www.frontiersin.org/journals/forests-and-global-change/articles/10.3389/ffgc.2021.692706
[9] Federal Office for the Environment FOEN, “Forest mix rate NFI,” Oct. 2023. [Online]. Available: https://www.geocat.ch/geonetwork/srv/eng/catalog.search#/metadata/b07a4589-159b-4c3a-b1e7-dbdad11f7e62
[10] Rothermel, R.C. (1972) A Mathematical Model for Predicting Fire Spread in Wildland Fuels, Vol. 115. Intermountain Forest & Range Experiment Station, Forest Service, US Department of Agriculture, Ogden. https://www.fs.usda.gov/treesearch/pubs/32533
[11] Federal Office for the Environment [BAFU]. (2021). Vegetation height model NFI [Dataset]. Federal Directorate of Topography swisstopo; geo.admin.ch. https://data.geo.admin.ch/browser/index.html#/collections/ch.bafu.landesforstinventar-vegetationshoehenmodell
[12] Federal Office of Topography swisstopo. (2024a). Satellite images swissEO—VHI (Contains modified Copernicus Sentinel data) [Satellite imagery]. https://www.swisstopo.admin.ch/en/satelliteimage-swisseo-vhi