Geog 168 Lab 6 - Suitability Analysis II

Above, clockwise: Elevation, Fuel Risk, Final Fuel/Elevation Model, and Fuel Type. Work from the exercise for Lab 6:

I created my fire map with the intent to focus on two major aspects of fires: fuel and elevation. The location of the Station Fire was a typical mixed shrub-and-forest environment that was not only incredibly dry, but was also at a higher elevation overall than a fire might be along the coast or in a suburban area. In the exercise fire model, we worked with raster images for vegetation type, fuel, and slope. However, I decided to incorporate elevation as part of my fire model because my experience backpacking taught me that in dry years, campfires (and sometimes hiking altogether) are prohibited in areas of high elevation due to the risk of forest fires. Fires are more difficult to extinguish at higher elevations because its usually rugged terrain and alpine environments take much longer to regenerate than others because the growing season is shorter. These environments might also be relatively isolated--not many areas of Los Angeles have native conifer trees, and plants and animals adapted to these environments are much more vulnerable if an area is destroyed. I also worked with elevation directly instead of calculating the slopes of the area because--while slope is a very important factor in fighting fires--I had more trouble exporting a slope raster from my DEM than in the lab exercise. I don't know exactly why, but I concluded that I had more than enough information on land cover, fuel hazard, and elevation profile to make a meaningful map.

First I downloaded land cover and elevation data. I retrieved the elevation data from the USGS Seamless server, and the land cover data from the California Department of Forestry and Fire Protection. While I thought that the land cover data was important (coloring on the Fuel Type map indicates which types are worse for fires), however, I was more interested in the risk associated with that fuel. For example, tree stands that are ablaze are much more hazardous than a fire in a meadow. Grass fires light quickly, but burn "cool" because there is little fuel on the ground. Live hardwoods in relatively thin stands can survive these fires. Conifers engulfed in flames, however, burn long and hot. Burrowing animals might not survive, and the area is completely devastated. To get a good idea of locations where high elevation (and therefore more delicate environments) meet high-risk fuels, I decided to use the raster calculator to make a model.

I first reclassified the elevation data into about 7 classifications, in increments of 500 feet. Next I reclassified the land cover data as well--twice. The first time I consolidated similar land cover classifications so that it would be easier to read. For example, Class 1 "Grass" and Class 2 "Sparse pine/meadow" were grouped together. The result is the "Fuel Types" map in the lower left corner of the map sheet. I should note that these land cover classes (retrieved from the California Dept of Forestry/Fire) were organized by degree of hazard, so I changed the colors for these land cover types appropriately (e.g., Grass is a lighter color than Brush). Next I reclassified the raster further. I gave urban, desert, barren and other non-forest environments a low value because they are less likely to burn as fast or furious as a forest fire. I separated the rest of the classes by their degree of hazard, giving more hazardous fuel types (conifer stands) higher values. Next I used the raster calculator to bring the rasters together. My calculation was (Elevation_Reclassification) + (2*(Land_Cover_Reclassification)) + (2*(Fuel_Risk)). The fuel risk raster (in the upper right corner) was obtained from the same website as the land cover raster and accounted for slope and other factors (such as proximity to water sources) in assessing the risk of fuels in this area. I thought that fuel type and fuel risk were more important to determining overall fire risk than elevation, so I gave each twice as much weight in the calculation as the elevation raster.

The result is the Elevation/Fuel Risk model in the lower right corner of the map sheet. As a final touch, I added the roads to the area because I thought it was worthwhile for the viewer to know that the most hazardous areas for a fire in La Cañada Flintridge were actually some of the closest to the local roads. I also added a general elevation map to the map sheet, and hillshade to two maps so they looked more polished. Finally, I added a small map of LA County in case the viewer didn't know the context of the fire. I didn't have much trouble with this lab but I think if I could have obtained the slope it might have altered my fire model. I also think that knowing which areas burned most recently might have added to the map, but I didn't want to make it too complex.