Susceptibility and vulnerability models were run separately before being combined into the final risk model. The application of susceptibility, vulnerability and the final risk models were conducted via the model builder application in ArcGIS. Both susceptibility and vulnerability consisted of specialised factors which were inputted into their respective models. All components of the susceptibility and vulnerability models were clipped to the B.C. Southern Interior and subsequently reclassified and standardised for use in the raster calculator. Once the individual layers were created, they were combined together in the raster calculator to create the final wildfire risk layer. 

Wildfire Susceptibility
For ease of viewing and describing the susceptibility model, specific components of susceptibility have been categorised and displayed in the proceeding figures. Susceptibility contained a constraint for two factors: water and land use (Figure 2). Landuse consisted of urban areas and barren land, which were assumed to contribute negligible susceptibility to wildfires. Three categories within the susceptibility model describe the fuel type, fuel conditions and ignition potential. Fuel conditions (Figure 3) consisted of maximum temperature, mean annual precipitation, and total biomass accumulation. The fuel conditions were categorized into fuel types, each layer describing the forest stand and species type (Figure 4). Species layers were weighted and converted from polygon to raster formats and combined into a final species raster layer. The ignition potential consisted of vector file highway systems and raster data for elevation (Figure 5). Highways were converted to raster format and calculated for Euclidean distance. All inputs were standardised, weighted and processed using the MCE equation (Eq. 2) in the raster calculator to create the final susceptibility layer (Figure 6).  

Equation 2. MCE equation for fire Susceptibility Layer:

Susceptibility Layer = ((Water Constraint)*(Bare Land Constraint))*((0.365)(Mean Annual Precipitation) + (0.158)( Max Annual Temperature) + (0.158) (Biomass) + (0.158)(Euclidean Distance) + (0.064)(Stand Age) + (0.064)(Elevation) + (0.031)(Stand Type))

Figure 2. Constraint component for susceptibility model with respective input (blue), tools (yellow) and output (green) layers. 

Figure 3. Fuel conditions component for susceptibility model with respective input (blue), tools (yellow) and output (green) layers. 

Figure 4. Fuel type component for susceptibility model with respective input (blue), tools (yellow) and output (green) layers. 

Figure 5. Ignition component for susceptibility model with respective input (blue), tools (yellow) and output (green) layers.

Figure 6. Overview of the susceptibility model with input layers (white) and their respective susceptibility categories (green). all inputs were calculated using the raster calculator (orange) to create the susceptibility map (blue). 

Wildfire Vulnerability
The vulnerability model was categorised into three categories: land value, ecological vulnerability and potential for loss of life. Weighted factors were entered into the model as inputs for reclassification and clipping. Land value consisted of timber harvest sites, protected areas and land use (Figure 7). Ecological vulnerability inputs consisted of species at risk and biodiversity indices (Figure 8). Potential for loss of life input was a shapefile for a measure of remoteness, which was clipped and subsequently converted to raster format (Figure 9). The vulnerability model contained the same constraints as susceptibility for both water and land use. The specific weights of each factor were reclassified (standardised) then combined into the raster calculator using the MCE equation (Eq.3) to create the vulnerability map (figure 10).

 Equation 3. MCE equation for Fire Vulnerability Layer:
Vulnerability Layer= ((Water Constraint)*(Bare Land Constraint))*((0.519) (Remoteness) + (0.201)(Biodiversity Index) + (0.201)(Species at Risk) + (0.079)(Index Ranked Land use))

Figure 7. Land value component of the vulnerability model. Input layers (blue) were clipped and reclassified with the clip and reclassify tools (orange) in ArcGIS to create the final layers to be used in the raster calculator of the vulnerability model. 

Figure 8. Ecological vulnerability component of the vulnerability model. Input layers of species at risk and diversity indices (blue) were subsequently clipped and reclassified (orange) into output layers (green) for raster calculations. 

Figure 9. Potential for loss of life component of the vulnerability model. The remoteness data layer (blue) was clipped, converted to raster and reclassified using the clip, polygon to raster and reclassify tolls (orange) in ArcGIS. Output layers from the tools are coloured green. 

Figure 10. Overview of the model to create the vulnerability map. All vulnerability layer inputs and their respective categories (green) are inputted into the raster calculator. Remoteness was a provided data set that was not created in this study   

Risk Model Overview
The final fire risk layer will be determined by combining the equal weights of the final fire vulnerability layer with the final fire susceptibility layer (Eq. 4) (Figure 11). 

Equation 4. Equation for the final risk model.
Risk = (0.5)(Vulnerability Layer) + (0.5)(Susceptibility) 

Figure 11. Risk model process showing both vulnerability and susceptibility inputs into the raster calculator resulting in the final output.