Cumulative Effects Assessment Practitioners' Guide

Eagle Terrace Sub-Division: Case Study Highlights

VECs: Elk, wolf, Swainson's Thrush

Issues: Loss of critical winter habitat, obstruction of regional wildlife movements

Approaches: Calculation of incremental land lost (direct and indirect) using a GIS due to changes between successive development scenarios

Lessons learned: Spatial boundaries based on available digital base-map; limitations of interpreting long-term ecological implications of changes

Background

In 1996 a proposal was submitted to the Town of Canmore in Alberta for the development of the 67 ha Eagle Terrace residential subdivision. The project would adjoin existing sub-divisions near the town, located in the Bow (River) Valley east of Banff National Park. Parts of this valley have experienced extensive human development since the beginning of this century. Continued growth in tourism has increased demand for resident and visitor facilities, resulting in development pressures for housing. This urban growth, combined with a major national transportation corridor (i.e., four-lane highway and a railway) has disturbed the important montane ecosystem that supports a wide variety of wildlife species and plants. Concerns were raised about the cumulative or nibbling loss of wildlife habitat in the mountain valley, and the obstruction of wildlife movement corridors as developments continue to advance up the lower slopes of the valley.

Assessment Approach

An EIA of the project was prepared which included a CEA chapter (Eagle Terrace 1996). The assessment method was based on the calculation of available habitat in the valley between successive development scenarios. This allowed the comparison of incremental losses of habitat. Three wildlife species were chosen as indicators: elk, wolf and Swainson's Thrush. Elk was used to assess use by ungulates and to serve as an ecological indicator of use of early seral habitats. Wolf was used to assess use by large carnivores and to serve as an ecological indicator of regional wildlife movements between Banff National Park and areas east of the Park. Swainson's Thrush was used to assess use by songbirds and to serve as an ecological indicator of localized fragmentation of forest habitat.

Ecosites (classifications of vegetation-landscape association that categorize an area based on its soils, drainage, and vegetation characteristics) were used to create a habitat base map that provided input into a GIS. The map defined a regional study area of approximately 17,000 ha, which extended west from Canmore to the boundary of Banff National Park, and east to the eastern edge of the Rocky Mountains. Twenty-four ecosites were interpreted, based on their suitability for providing basic wildlife habitat requirements (e.g., cover, forage), into three habitat suitability ratings (low, moderate and high) for both summer and winter.

The Eagle Terrace site was mapped along with other existing developments, foreseeable projects (i.e., those under application for approval or approved), and various infrastructure such as roads and railways. To represent loss of habitat due to alienation effects (i.e., due to sensory disturbances such as noise and light), an "alienation buffer" was defined for each indicator species. These buffers, which surrounded all disturbances, were 500 m for elk, 1000 m for wolf and 600 m for Swainson's Thrush.

Four development scenarios described changes in the valley. Each scenario was defined by a combination of habitat suitability and level of development, and represented a "snapshot" in time of the condition of the human and natural environment. The first scenario, Pristine, was represented by current valley conditions with all developments removed. The second scenario, Current, represented the existing Bow Valley with its current settlements, roads and other developments. The third scenario, Reasonably Foreseeable, included all developments in the Current scenario and projects which were already under construction or for which there was considerable likelihood that they would occur. The final scenario, Full Build, added the Eagle Terrace development to the last scenario.

Direct habitat loss (due to the overlap of various disturbances on the habitat suitability map), indirect habitat loss (due to the alienation buffers), and total or effective loss (i.e., direct and indirect) were determined for summer and winter habitat conditions. Losses were determined between each of the scenarios. The final calculation therefore provided an indication of the relative contribution of Eagle Terrace to changes in the valley. This contribution could also be compared to changes that had already occurred due to other projects.

It was determined that the Eagle Terrace project would incrementally contribute a small loss of montane and wildlife habitat relative to the losses that already had occurred. In general, existing developments contributed to a loss of 21% or 2789 ha of the important montane ecosite, while the Eagle Terrace development represented 2% of that loss. A large proportion of the most important habitat (i.e., high suitability in winter) was found to already have been lost: 59% for elk, 81% for wolf and 87% for Swainson's Thrush (see table below for an example of how the results were presented). The Eagle Terrace project contributed to less than 1% of this change. The implications of these changes on regional wildlife populations is avoidance of the area by ungulates which make use of the mountain benchlands for winter refuge, avoidance by carnivores which, make use of the valley corridor for regional movements, and avoidance by nesting songbirds.

Summer Habitat Losses for Swainson's Thrush
ScenarioHigh SuitabilityModerate SuitabilityLow SuitabilityTotal 4
Area 1%SA 2%RA 3Area%SA%RAArea%SA%RA
Current15899.486.5647038.246.91420.810.88201
Reasonably Foreseeable1751.070.912167.216.600.00.01391
Full Build 500.00.0100.10.200.00.010
Total176410.4 769645.4 1420.84 9602

1 area in hectares.
2 habitat loss as percentage of total land in study area (16,959 hectares).
3 habitat loss as percentage of habitat remaining after last scenario for that habitat suitability class (i.e., High,. Moderate,. Low) (Note: The total % is not additive and therefore has not been calculated).
4 total habitat loss.
5 represents incremental change due to addition of Eagle Terrace to Reasonably Foreseeable scenario.

Lessons Learned

  • The assessment approach provided a quantitative determination of regional level habitat changes due to successive developments and the determination of the relative contribution of the project under review in the future. However (as is still often true in the assessment of effects on wildlife), the ultimate ecological implications of these changes could only be qualitatively determined. It could, for example, be reasonably assumed that losses from existing developments had already considerably contributed to a steady decline in local and regional wildlife populations (corroborated to some extent by field data). This suggested that eventually much of the valley may become permanently alienated to wildlife, or at minimum, result in conditions unlikely to support wildlife populations at historical levels in the valley. That the latter conclusion could not be quantitatively determined indicates the difficulty practitioners experience in translating results from a GIS into the answer ultimately sought; namely, will the wildlife disappear? [It is notable that no other methodological tool yet exists that may provide a fully confident answer, although the Cumulative Effects Model (USFS 1990) currently used for the assessment of effects on grizzly bear is making considerable advances, especially when combined with the interpretation of genetic data.]
  • The availability of a digital ecosite map for use in a GIS was a major factor during the determination of spatial bounds. The map included many developments and extended for a considerable distance along the mountain valley, which ensured an adequate representation of natural conditions and human-caused disturbance conducive to a regional level analysis.