Firstly, we identified the factors that contribute to an avalanche. Terrain features and topography can have a greater overall impact on the formation of avalanche conditions (Choubin et al., 2019). This analysis will look at slope angle, elevation, convexity, and aspect as the factors that contribute to avalanche hazards. While wind and other meteorological factors also influence avalanche hazards, any consideration of meteorological factors in hazard identification shifts the scope of the project towards forecasting instead of hazard mapping.
The primary data source for this project was the BC TRIM. From the TRIM elevation points a TIN was made. We were able to create a DEM with a 25-meter resolution from the TINs. The DEM was an intermediate step in calculating the factors that contribute to an avalanche. Each of the factors that contribute to avalanche risk were inputted into a multi-criteria evaluation (MCE) to determine three regions of lowest hazard in each of our study areas. Within each of the three regions, we then utilized a least-cost path to determine the lowest risk route from the nearest road to the ridgeline.
Slope Angle
The most common slope angle for avalanches is 30-45 degrees, and slab avalanches can be initiated in a start zone up to 60 degrees (Floyer & Robine, 2018). Any slope angle above 60 degrees is too steep to hold enough snow to produce a large slide. Slopes above 60 degrees have frequent small slides as they are above the angle of repose for the snow (Maggioni & Gruber, 2003). Slopes between 25-30 degrees have a rare chance of sliding, and slopes under 25 degrees have a component of gravity force that is too high to allow for them to slide (Floyer & Robine, 2018; Maggioni & Gruber, 2003). Based on this research, we reclassified the slope angle raster to reflect the avalanche hazard.
| Start | End | New |
| 0 | 30 | 3 |
| 30 | 45 | 1 |
| 45 | 60 | 2 |
| 60 | 90 | 1 |
| NODATA | NODATA | NODATA |
Elevation
The start zone of an avalanche is strongly correlated with its elevation (Gleason, 1994). In this model, elevation zones were broken up into three bands: below treeline, treeline and alpine. These are the same regions that Avalanche Canada uses (Floyer & Robine, 2018). Each study area received its own elevation band classification as the treeline changes for each of the zones. We used treeline because trees act as anchors for the snowpack (Maggioni & Gruber, 2003; Floyer & Robine, 2018). Increased spacing between trees reduces their effectiveness as anchors.
Coast
| Band | Start (Meters) | End (Meters) | New |
| Alpine | 1700 | 99999 | 1 |
| Treeline | 1400 | 1700 | 2 |
| Below Treeline | 0 | 1500 | 3 |
| NODATA | NODATA | NODATA | NODATA |
Mix
| Band | Start (Meters) | End (Meters) | New |
| Alpine | 1800 | 99999 | 1 |
| Treeline | 1500 | 1800 | 2 |
| Below Treeline | 0 | 1500 | 3 |
| NODATA | NODATA | NODATA | NODATA |
Rockies
| Band | Start (Meters) | End (Meters) | New |
| Alpine | 1900 | 99999 | 1 |
| Treeline | 1600 | 1900 | 2 |
| Below Treeline | 0 | 1600 | 3 |
| NODATA | NODATA | NODATA | NODATA |
Convexity
The shape of the slope can determine the amount of tension in the snowpack. A convex slope that contains a rollover will typically have an avalanche trigger point just above the inflection point in the slope (Floyer & Robine 2018). Conversely, a concave slope is under compression as the weight of the snow is compressing the snow below it (Choubin et al., 2019). The compression of the snow on a concave slope can help reduce smaller slab avalanches (Floyer & Robine 2018). Based on this information, weights were applied to the convexity of the slopes.
| Start | End | New |
| -100 | 0 | 2 |
| 0 | 100 | 3 |
| NODATA | NODATA | NODATA |
Aspect
The aspect of the slopes has an impact on the metrological conditions that are experienced on the slope along with the snow stability. Northern slopes are cooler and typically more unstable as they can develop weak, faceted layers (Gleason, 1994). Southern slopes can also become unstable due to warming of the snowpack (Conway & Raymond, 1993). Within the model, weights were assigned for angles based on sun exposure.
| Start (Degrees) | End (Degrees) | New |
| -1 | 0 | 2 |
| 0 | 22.5 | 2 |
| 22.5 | 67.5 | 3 |
| 67.5 | 112.5 | 3 |
| 112.5 | 157.5 | 2 |
| 157.5 | 202.5 | 1 |
| 202.5 | 247.5 | 2 |
| 247.5 | 292.5 | 3 |
| 292.5 | 337.5 | 2 |
| 337.5 | 360 | 2 |
| NODATA | NODATA | NODATA |
Analytical Hierarchy Process
An analytical hierarchy process was used to weigh the different components of topographical avalanche hazards. Our selection of the weights was based on current literature and the way that each factor contributes to overall risk (Smith & McClung, 1997; Chobin et al., 2019; Floyer & Robine, 2018).
| Category | Weight |
| Aspect | 3.7 % |
| Curvature | 24.8 % |
| Elevation | 7.3 % |
| Slope Angle | 64.1 % |
Best Suitable Location
After coming up with an AHP and MCE, the locate tool was utilized within ArcGIS Pro to identify the best suitable location from the reweighted and combined raster’s. The locate tool was looking for a 25 km^2 region within the study area.
Best Descent Path
After a suitable location was identified within each study area, a point was added to the ridgeline within the region. Additionally, a point was added on the nearest road to the identified region. Between the two points, a least-cost path was performed using the weighted raster as the cost surface. The result yielded a line that was the least hazardous descent path from the top of the identified location to the nearest road (we preferred to pick a spot where parking was feasible).