Developing nationwide avalanche terrain maps for Norway

Larsen, H.T., Hendrikx, J., Slåtten, M.S. et al. Developing nationwide avalanche terrain maps for Norway. Nat Hazards 103, 2829–2847 (2020). https://doi.org/10.1007/s11069-020-04104-7

The Avalanche Terrain Exposure Scale (ATES) is a terrain classification system that was designed by Parks Canada to communicate the difficulties and risks of traveling in avalanche-prone areas. The avalanche hazard maps created with ATES are indispensable for backcountry travel and route planning. However, they are very time consuming to create for large swaths of backcountry terrain. This article describes the development of an automated algorithm capable of high spatial resolution ATES mapping for the entire mainland of Norway.

To accomplish this, the authors began with a 10-meter resolution raster digital terrain model (DTM) of Norway produced by the Norwegian Mapping Authority. They delineated a slope raster in Esri ArcMap 10.6, and assigned slope angles greater than 40° class 3 (complex), between 40° and 25° class 2 (challenging), and below 25° class 1 (simple). They also chose to use the optional class 0 for slopes with inclines of less than 15°, because as they noted, subsequent analyses might increase their terrain class if they are found to be in an avalanche path's runout. Next they calculated avalanche path start zone density using a modified potential release area (PRA) algorithm developed by Veitinger and Sovilla (2016). The criteria used by this algorithm are; slope, wind shelter index and roughness calculated from the DTM, average snow depth, and prevailing wind direction. To estimate the potential avalanche runouts, they used the hydrologic terrain analysis software TauDEM and TauDEM toolbox for Esri ArcMap (Tarboton 2005). Finally, they automated the previous steps in a script using Python 2.7, and the flowchart for this is shown in the graphic above.

When completed, the automated ATES algorithm took approximately 500 hours to process the entire mainland of Norway. This might seem like a lot of time, but when you think about the time it would probably take to manually produce a high resolution hazard map of 365,246 km², it seems like the blink of an eye! This article demonstrates that GI systems could generate ATES maps for the entirety of the American backcountry using Python scripted automation. This is welcome news, since I believe that ATES maps should be displayed at every major backcountry trailhead.