When clean energy is dirty: A geospatial analysis of Canadian hydroelectric dams constructed between 1981 and 2011 and demographic indicators of environmental marginalization

Abstract

The increasing urgency for climate action has pushed several governments towards investments in long-term, renewable energy sources. With its abundant supply of freshwater rivers, Canada’s shift away from oil and gas-generated power has been wholly successful, with almost 80% of all power generation in Canada attributed to hydroelectricity. However, trends in dam placement show a disproportionate number of Indigenous and other marginalized communities displaced or otherwise impacted by the environmental devastation caused by reservoir filling and interrupted water flow. A geospatial regression was conducted on sixteen hydropower dams constructed between the years 1981 and 2011 across Canada to determine if there is a significant spatial trend in constructing hydro projects disproportionately within disadvantaged minority communities. The demographic variables were chosen to align with the Canadian Marginalization Index (CMI), and describe ethnic minority and Indigenous population, dependency, education, and unstable housing. Each variable was applied to the dam location using the most recent census year prior to construction. A generalized linear regression (GLR) conducted in ArcGIS Pro determines if there is a significant relationship between demography in census subdivisions (CSDs) containing dams to those surrounding the dam site. Subdivision areas pre-dating the year 1991 returned inconsistent data due to the large variation in census subdivision size and population. Of the 16 dams analyzed, five returned statistically significant results demonstrating factors representing both higher and lower marginalization factors in dam site areas as compared to those surrounding, depending on the model. Indigenous population dynamics around dam sites return varied outputs, but results are consistent with hydro development political timelines that reveal patterns of historical ignorance to Indigenous land claims and treaty rights. Conversely, dam sites across Canada tend to be in rural and “unorganized” subdivisions, many of which have populations too small for collection or are designated “non-response” zones. These areas contribute largely to the unknowns surrounding community dynamics in the areas, and although the results give indication of the determining factors, finer study areas and more consistent rural data is required to obtain a more representative picture of marginalization in these areas. While lack of demographic data for rural areas of Canada and census years prior to 1981 is a hurdle to this research, greater efforts must occur to ensure the steps we are taking to address climate action do not come at the expense of our most vulnerable communities.

KEYWORDS: Geospatial analysis, Indigenous, Hydroelectricity, Census of Canada, Marginalization index