Results
Slope Path Including Alberta
The least cost path obtained using only slope across the two provinces was remarkably similar throughout BC. In Alberta; however, the path generated in GIS differed from the proposed path by Enbridge in that the generated path avoided an area west of Edmonton with a slight elevation gain rather than just going right through as in the proposed pipeline path.
Map showing least-cost path using only slope
as a cost factor from Bruderheim to Kitimat
Results Table
Summarized here are the total lengths of each path, the length of each parameter that the pipeline crossed, and the total number of times each pipeline crossed each parameter.
Note: These results are not completely accurate and thus are only estimates due to the coarse scale nature of this project. The analysis spanned from the British Columbia-Alberta border to Kitimat, BC using a raster resolution of 100x100 therefore the accuracy of these lengths is limited.
Discussion
Least Cost Paths
The Least Cost Paths map above shows the locations of all 3 least cost path routes produced in the analysis, as well as the route that is currently being proposed by Enbridge (Northern Gateway Pipeline, 2009). Least Cost Paths 1 and 3 both follow a very similar route that travels north much more than the proposed route and least cost path 2. Least Cost Path 2 and the route proposed by Enbridge both follow a similar route, although path 2 is much more contorted.Least Cost Path 1
Even though lakes were assigned the highest cost (100) Least Cost Path 1 still passed through 6 lakes accounting for 54 km of pipe that would have crossed lakes. The path is approximately 797km (as seen in the Results Table) therefore approximately 7.1% of this pipe crossed through a lake. Upon further analysis of the various layers used to create the cost surface, it was found that all lakes have a slope of 0⁰ and are all located in relatively low elevation areas. In contrast, the areas surrounding the lakes generally have slopes that are greater than zero and are found in higher elevations. As was shown in the cost table for this path, higher elevations and higher slopes were given fairly high costs and low elevations and low slopes were given low costs. This means that although the lakes have a high cost individually, the cumulative cost of all of the superimposed layers in the vicinity of a lake are quite low. On the other hand, the areas surrounding the lakes would contain cumulative costs because of the slope values. The path is therefore more likely to travel through the lakes which is what seemed to occur in this case.Least Cost Path 2
In the production of the cost surface for least cost path 2, cost values for several layers were changed as can be seen in the cost assignment table for this path. The most drastic change was made to the lakes layer where cost was increased from 100 to 1000 in order to ensure that the pipe did not cross any lakes like it did in path 1. The costs of high elevation values were also reduced because there were doubts as to whether elevation played much of a role in determining a route for the pipeline. As can be seen in The Effects of Elevation on the Three Least Cost Paths map below, all 3 routes tend to stay within the 500-1000m elevation range where it is present, even when the costs of higher elevations were reduced for least cost path 2. Even near the end where it looks like path 2 is crossing through several high elevation areas, zooming in reveals that the path travels through lower lying valleys. The fact that there are fewer lakes present along the route near path 2 depicts how the cost of 1000 assigned to it deters the path from going near these areas as compared to least cost paths 1 and 3. In comparison to the proposed route, path 2 is much longer and contorted. Path 2 crosses less of the parameters in the analysis (provincial parks, old growth forest, ect.) as can be seen in the Least Cost Paths map above. It also stays within lower elevation areas while the proposed route crosses various high elevation areas as can be seen in the map immediately below. This avoidance of parameters and affinity for lower elevation areas due to the costs we assigned in our model seem to account for why path 2 is so much more contorted and longer than the proposed route.
Least Cost Path 3
The question of how much emphasis the high cost values of elevation and slope had on Least Cost Path 1 lead to the creation of Least Cost Path 3. The same costs for elevation and slope that were used in the production of Least Cost Path 1 were used for path 3 but all of the other parameter costs were reduced to 10, except for the lakes layer which was given a cost of 1000. The idea was to produce a path in which only elevation, slope, and lakes would have an effect on the proposed route. As can be seen in both the Least Cost Paths and The Effect of Elevation on the Three Least Cost Paths maps, path 3 followed a very similar route to path 1. Differences arose in the centre of BC where path 3 travelled a little further North, which seems to be an attempt to avoid the many lakes that path 1 crossed just further south. A map of the effects of slope on the paths can be seen below where, as predicted, the paths are constrained to low slope areas due to the high cost associated with steep slopes in all 3 models.
Construction Costs
Out of the 3 least cost paths produced, Least Cost Path 2 (LCP 2) was the shortest at 725.9km. LCP 2 also did not cross any lakes, while LCP 1 crossed 6, and LCP 3 crossed 1. From these parameters it can therefore be assumed that LCP 2 would have the lowest total construction costs out of the 3 routes produced. In comparison to the produced routes, the route proposed by Enbridge is significantly shorter (608.9km) because it is much straighter. This route also did not cross any lakes therefore we can assume that total construction costs for the route proposed by Enbridge would be much less than any of the routes produced in this analysis. It is possible that elevation did not play a big role when the proposed route was assigned cost values because the proposed route travels strait through several mountain ranges while the least cost paths produced in this analysis contort around them.
Comparing Enbridge’s Proposed Route to Least Cost Path 2: Environmental Risk
As can be seen in the Results Table, the currently proposed pipeline route crossed approximately 9.6km of provincial parks, 18.1km of agricultural land reserve, and 11.2km of old growth forest. In comparison, Least Cost Path 2 did not cross any provincial parks, 1.4km of agricultural land reserve (approximately 7.7% of the amount crossed by the currently proposed route), and 1.1 km of old growth forest (approximately 9.9% of the amount crossed by the currently proposed route). These results exemplify the fact that LCP 2 has much lower environmental risks than the currently proposed route even though it is 117km shorter.
Cost Surface and Least Cost Analysis
This is a map that spans from Kitimat, BC to the location at the Alberta-British Columbia border where the proposed pipeline is supposed to cross. The 3 cost paths produced as well as the digitized proposed pipeline (Northern Gateway Pipelines, 2009) are shown above. The parameters that were used to produce the cost surface are also shown.
Environmental Justice Analysis
One of our analyses was to see if there were any environmental justice issues with the proposed Enbridge pipeline path or with our proposed paths. Environmental justice can be defined as “the fair treatment and meaningful involvement of all people regardless of race, color, national origin, or income with respect to the development, implementation, and enforcement of environmental laws, regulations, and policies”
(http://www.epa.gov/environmentaljustice/).The main reason for this sort of analysis is because there have been issues in the past where hazardous and polluting industries were predominantly taking place through areas with minorities (especially aborigines) and lower income communities. We used average income, unemployment rates, and aboriginal population statistics to determine whether there were any potential environmental justice issues along both the generated and proposed pipeline routes.
Average Income Analysis
Average income of dissemination areas were plotted near the pipeline route for this analysis. There does not appear to be any correlation between the proposed Enbridge path and dissemination areas with lower incomes.
An interesting note is that there is a noticeable segment of the Enbridge pipeline that passes through a higher-than-average income dissemination area ($57,153).
For our cost paths 1 and 3, there were more areas with lower average incomes than the Enbridge path (lowest being $15,383). Also of note was that this path did not pass through the higher-than-average income dissemination area that the Enbridge path passed through.
Cost path #2 follows a path that is more similar to the proposed Enbridge path, and as a result there does not seem to be any correlation between this path and dissemination areas with lower incomes.
Map showing three generated paths overlain with average income of dissemination areas.
-
Unemployment Analysis
Unemployment rate of census subdivisions were plotted near the pipeline route for this analysis. Analysis using dissemination areas for unemployment rate was not possible due to privacy protection by Statistics Canada.
There did not seem to be significant correlation between the proposed Enbridge path and census subdivisions with high unemployment. Although the pipeline does pass through two census subdivisions with higher than average unemployment rate --above 11.4% whereas the Canadian average was 6.3% in 2006—the majority of the census subdivisions that the pipeline goes through are close to or below average. Similar results were concluded for cost path #2.
Our least cost path #1 does not pass through areas with significant unemployment until about the last third of its path. In this last section it passes through several small census subdivisions with unemployment rates that were quite high. The highest being that of Gitwangak 1 area which had an unemployment rate of 43.6%.
Our least cost path #3 did not deviate too far from #1 and as a result also passed through much of the same areas with high unemployment rate.
Map showing three generated paths overlain with unemployment rates of dissemination areas.
-
Proportion of Aboriginal Peoples near Pipeline Analysis
In this analysis, aboriginal population was standardized with total population to see whether or not the Enbridge pipeline passed through areas with high proportions of aborigines Census subdivisions were used for this analysis once again due to privacy protection by Statistics Canada.
Although there did not seem to be significant correlation between the Enbridge path passing though census subdivisions with high proportions of aborigines, there does seem to be some subdivisions with high aboriginal proportions that were close to the pipeline. Cost path #2 had similar results.
Our cost paths 1 and 3 do pass through more areas with higher percentage aboriginal populations, however.
Map showing three generated paths overlain with aboriginal population proportion of dissemination areas.