ARCHIVED - Assessment of Canada’s Hydrokinetic Power Potential

Information Archived on the Web

Information identified as archived on the Web is for reference, research or recordkeeping purposes. It has not been altered or updated after the date of archiving. Web pages that are archived on the Web are not subject to the Government of Canada Web Standards. As per the Communications Policy of the Government of Canada, you can request alternate formats. Please "contact us" to request a format other than those available.

Download Full Document (PDF document, 2 207,9 Ko)

Canadian Hydraulics Centre
National Research Council of Canada

CETC Number Canmet_March_2010 / 2010-09-16

Phase I Report Methodology and Data Review

Executive Summary

Natural Resources Canada (NRCan) has identified a need to assess Canada’s hydrokinetic potential as a national renewable resource. To reach this objective a three-phase project was commissioned:

Phase I

Methodology Review and Data Review – a review of the methodologies, techniques and available data for conducting a regional hydrokinetic power assessment as well as a selection of proposed methodologies and validation datasets and locations.

Phase II

Methodology Validation – the implementation of a recommended set of methodologies against the validation datasets including sensitivity/uncertainty analysis.

Phase III

Assessment Determination – the application of the recommended methodologies to conduct a nation-wide assessment on the hydrokinetic potential for theoretical energy extraction.

The National Research Council – Canadian Hydraulics Centre (NRC-CHC) was commissioned to contribute to the first two phases of this project. This report constitutes the findings of Phase I.

Section 1 provides an introduction and outlines the goals and structure of this study of hydrokinetic potential. The project goals are to ultimately characterize and quantify the hydrokinetic power potential in Canada. Section 2 of this report identifies the need to estimate a number of hydrologic and physical characteristics at ungauged channel reaches to provide an estimate of hydrokinetic energy potential:
  • Flow characteristics as flow duration curves (FDCs)
  • Channel geometry
  • Channel slope; and
  • Channel roughness.

Section 3 presents the findings from a literature review investigating methodologies for regional estimation of flow, geometry slope and roughness characteristics in ungauged basins. The investigation identified many techniques for flow regionalization but primarily for extreme value estimations. The use of regionalization techniques for FDC estimation, as required for this study, was much less common but some studies existed and a few studies have been recently published with applications in Canada.

The investigation discovered few studies employing channel geometry estimation techniques at a regional scale. Most of the studies related geometry predictions to a channel-forming discharge. Other more promising approaches were identified which relate channel geometries to physiographic watershed characteristics or digital maps that include channel widths for larger rivers. The regional estimation of slope showed a predominance of digital elevation model (DEM) use in the studies investigated, although limitations and cautions in their use for channel slope estimation were identified by many. Some of the limitations of DEMs, including precision issues in low-relief basins, were shown to be surmountable through the use of fitted equations relating channel slope to watershed drainage area. No appropriate regionalization techniques for the estimation of river channel roughness were identified.

In addition to regional estimation techniques for flow and channel characteristics in ungauged basins, uncertainty and error propagation techniques were investigated.

Section 4 presents the results of a data source review and resource investigation. The section identifies a number of national databases that can be employed to act as inputs and validation datasets for the regionalization routines employed in this study. The data sets include measured or calculated properties at a national scale including regional climate data, hydrometric data, digital soil and land use maps, hydro network maps and digital elevation data. Of particular utility is the Water Survey of Canada measurement database, which includes at-site measurements of velocities and cross sections for thousands of water survey hydrometric stations across Canada. This particular database is seen as invaluable in the validation of the various flow and channel geometry regionalization techniques. No databases were discovered that included information that could be used for roughness or slope validation.

Section 5 presents the findings of an investigation of previous hydrokinetic and small hydropower resource assessment studies conducted at a regional or national scale. Many studies were discovered that assessed hydropower resources, usually requiring an average annual flow and an estimated penstock height, and not requiring geometry, slope or roughness estimates. Fewer studies have investigated regional hydrokinetic energy potential, and none at the scale or as inclusive as suggested in this study. Methodology validation of regionalization techniques was rarely performed in the studies examined.

Sections 6 and 7 outline the recommended approach for Phase II and the associated tasks, respectively. It is recommended that a number of flow regionalization techniques be employed and validated in this study including some conceptually simple methods (e.g. Area-Ratio), commonly employed and endorsed methods (e.g. RETScreen) and methods recently developed and employed by the academic community (e.g. CCA with graphical FDC). The recommended approaches for geometry estimation include the use of digital maps of river edges where available, and the use of physiographic and climactic data to drive regression analysis. Channel slope estimation is recommended to be estimated using available DEM data with the investigation of functional smoothing in low-gradient channels. Finally, lacking any regional data or regionalization techniques, roughness is to be estimated as a potential range of values based on published roughness estimates. Validation and uncertainty estimates are to be employed using jack-knife and bootstrap techniques where applicable.