Methods | Safe Route Spearhead

Method

Objectives

The main objectives of this project are to:

1) Calculate a potential avalanche danger surface for the entire Spearhead Traverse

2) Propose the best route for completing the traverse that considers the calculated potential avalanche danger as well as other factors

Landcover

Again, because the Spearhead Range sits in both NTS 092G15 and 092J02 grids two separate landcover files were downloaded and then joined together using the 'Mosaic' tool.
To isolate just the Spearhead Range the new landcover data was then clipped to the "Study Area."
From personal experience on the Spearhead Range it was decided that many landcover classes should be combined to simplify the data for the MCE. Many of the landcover classifications would have similar effects on avalanches anyways. In reclassifying the landcover classes they were also organized in respect to their estimated effects on avalanche danger.

- Shadow and Developed Land categories became "No Effect," and were defined as having no influence on avalanche activity.

- Water, Snow/Ice, Rock/Rubble, and Exposed Land categories became "Open Land," and were defined as having a positive influence on avalanche activity
- Shrub Low, Wetland - Shrub, Wetland - Herb, Herb, and Grassland categories became "Grassland/Shrub," and were defined as having a positive influence on avalanche activity.

- Wetland - Treed, Coniferous Sparse, and Broadleaf Sparse categories became "Sparse Trees," and were defined as having a positive influence on avalanche activity.

- Coniferous Open and Broadleaf Open became "Open Trees," and were defined as having a neutral influence on avalanche activity

- Coniferous Dense became "Dense Trees," and was defined as having a negative influence on avalanche activity.

The new landcover classes were then converted to a raster surface, and normalized using the 'Fuzzy Membership' tool in preparation for the MCE with the calculated avalanche danger zones.

Spearhead DEM

Spearhead Slope

Spearhead Slope Classes

Raw landcover data. Note it is quite "busy" and cluttered.

Landcover data after being transferred to raster and normalized with the 'Fuzzy Membership' tool. White areas represent the landcover categories that are most likely to be associated with avalanche danger zones and dark gray areas represent landcover categories that are least likely to be associated with avalanche danger zones. Black areas are categories that were decided to have no affect on avalanche zones (mostly shadows, developed areas, or flat lakes).

Multi-Criteria Evaluation

Using the 'weighted sum' tool the normalized slope and landcover data was evaluated in an MCE to produce the potential avalanche danger surface (PADS). The MCE was done 3 times using different weights to see which weights rendered what were thought to be the most accurate results. In the end a 4:1 slope:landcover weighting system was used. This decision was informed by information from the Canadian Avalanche Association that says that slope is the most significant factor in avalanche activity and landcover is not. PADS values were classified using Natural Breaks because that was decided to be the least artificial classification scheme.

1. Calculating a Potential Avalanche Danger Surface (PADS)
Slope

The Spearhead Range sits in both NTS 092G15 and 092J02 grids and so two separate DEM's were downloaded for those regions and then merged together using the 'Mosaic' tool
To isolate just the Spearhead Range a new polygon feature class was created called "Study Area," which was drawn to cover the range. The DEM was then clipped to the "Study Area" using the 'clip' tool.
The slopes over the entire range were then calculated using the 'Slope' tool.
In order to identify the slopes in relation to their affect on avalanches, the calculated slope values were reclassified into three classes that reflected their "danger"; 0 - 24.9° = 1, 25 - 45° = 3, and +45° = 2 (in accordance to the Canadian Avalanche Association's slope classes).
These slope classes were then normalized using the 'Fuzzy Membership' tool to prepare them for a multi-criteria evaluation (MCE) with landcover.

The calculated Potential Avalanche Danger Surface based on a 4:1 weighted sum of slope and landcover, displayed in natural breaks. The data was chosen to be displayed in 4 classes to adhere to the Canadian Avalanche Association's Avalanche Terrain Exposure Scale's (ATES) 3 classes, with the addition of a low danger class meant to define areas of nearly total safety.

Methods Page 2

Data

The data used in this project was acquired primarily from a data clearinghouse, Geobase (http://www.geobase.ca/geobase/en/index.html).
All data was reprojected to BC Environment Albers projection to ensure spatial consistency, and because this projection preserves area which is useful in such a project.
DEM's downloaded: DEM092G15 2nd Ed. and DEM092J02 2nd Ed., both originated by The Government of Canada, Natural Resources Canada, Earth Sciences Sector, Mapping Information Branch, Centre for Topographic Information - Sherbrooke. Both published 2008-05-14
Landcover rasters downloaded: LCC2000-V 092G and LCC2000-V 092J, both originated by The Government of Canada, Natural Resources Canada, Earth Sciences Sector, Mapping Information Branch, Centre For Topographic Information - Sherbrooke. Both published 2009-05-06
ArcGIS 10.1 software was used to perform numerous operations on the data (tools/functions include Analysis Tools, Conversion Tools, Spatial Analyst Tools, and Data Management Tools), and ArcCatalog was used for all data management.
Supplemental information was acquired from the Canadian Avalanche Association and the Canadian Avalanche Centre, and was informed by my own backcountry experience in the region.