Environmental: Measuring Greenhouse Gas Emissions from a Canadian Reservoir

Hydro-Quebec undertook a research project at its Eastmain-1 project to measure gross greenhouse gas emissions from a newly created reservoir. The utility determined that gross emissions returned to levels associated with natural aquatic ecosystems by the third year after flooding.

Hydropower is an important electricity generating resource in Canada, representing about 66 percent of the country’s total installed capacity. Hydroelectric plants are attractive because they emit significantly fewer greenhouse gases (GHG) per terawatt-hour than do thermal plants, ranging from 35 to 70 times less.1 However, there is growing concern about determining how, and if, freshwater reservoirs contribute to the increase in GHGs in the atmosphere.2,3 The determination of GHG emissions from reservoirs is becoming more and more relevant both to ensure that methods of energy production are compared adequately and to evaluate facilities for carbon dioxide (CO2) credits.4 Three GHGs are related to the creation of reservoirs and the use of hydropower plants with reservoirs: CO2, methane (CH4), and nitrous oxide (N2O).5,6

Most of the data available on GHGs in the literature are gross emissions measured at the surface of water bodies and from established reservoirs (greater than 10 years old).4 However, the energy sector should work toward a new approach based on net GHG emissions, which are the emissions related to the creation of a reservoir minus what would have been emitted or absorbed by the natural systems over a 100-year period.

The goal of the Easmain-1 net GHG emissions project, which began in 2003, is to contribute to the existing body of knowledge. The 480-MW Eastmain-1 powerhouse was commissioned in 2006. The main dam and 33 dikes form a reservoir with a surface area of 603 square kilometers. The reservoir is in the boreal ecoregion of Quebec, about 1,000 kilometers north of Montreal.

Within this project, we have used a variety of methods for measuring GHG from reservoir, including gas partial pressure, eddy covariance towers, floating chambers, and automated systems. In this article, we present the results of gross GHG emissions of a young boreal reservoir, measured using both floating chambers and automated systems, in the context of the first study evaluating net GHG emissions of a reservoir.

Automated systems

For the Eastmain-1 GHG project, we installed five automated systems in 2006. One system is installed on two units at the Eastmain-1 generating station and measures GHG year-round. This system is connected to distribution pipes that collect water from the scroll case. In addition, during the ice-free period (mid-June to end of October), four automated systems are installed on rafts collecting surface water (30 centimeters below the surface). In the Eastmain-1 reservoir, one system is at the mouth of the Eastmain River (entrance to the reservoir), one is over flooded peat, and a third is over flooded forest. The fourth automated system is installed on a reference lake, Lake Mitsumis (see Figure 1).

Figure 2 shows the continuous GHG monitor installed in the powerhouse. Hydro-Quebec personnel built this automated system, based on the GHG partial pressure technique, using commercially-available components. CO2, CH4, and O2 partial pressures are measured in water and in air using three different types of sensors (LI-820 CO2 Analyzer from LI-COR, Panterra CH4 Analyzer from Neodym Technologies, and S101 O2 Analyzer from Qubit Systems, respectively). A similar automated system is commercially available from AXYS Technologies Inc.


The sensors are housed in a “dry box,” while the pump, valves, and tubing are installed in a “wet box.” While in monitoring mode, the device activates once every three hours. The monitoring operating cycle requires 22 minutes: one cycle of 20 minutes in water with two measurements (at 10 and 20 minutes), then one cycle of 2 minutes in air with one measurement. Water and air temperature are also recorded using a Thermistor temperature analyzer from Campbell Scientific. A programmable electronic data logger manufactured by Campbell Scientific controls all the electrical devices and collects and stores data. Details are available regarding quality assurance, quality control methods, and equations for flux calculations from partial pressure.7,8 Solar panels or small wind turbines installed on rafts or on the shore provide the energy for the automated systems.

Floating chambers

Figure 3 shows the floating systems connected to gas analyzers and computers. The floating chambers technique is widely used to measure GHG fluxes over water bodies.4 Air is sampled through an opening at the top of the chamber, which has a surface area of 0.2 square meter, and is returned at the opposite end of the chamber. This configuration allows the trapped air to be continuously mixed and enables a more representative measurement of gas concentrations.9 It takes about 10 minutes to obtain a reliable flux measurement. Three fluxes are measured at each station, and the floating chamber is moved to another station of the water body. Depending on the size of the water body or reservoir, a minimum of about ten stations is required to obtain a representative mean value of GHG fluxes.

CO2 is measured with a non-dispersive infrared instrument (Ciras-SC from PP Systems), and CH4 and N2O are measured with a Fourier transform infrared instrument (Gasmet DX-4010 from Temet Instruments). The gas measurement accuracy is 0.1 percent and 1 percent for these instruments, respectively. The floating system takes continuous readings, and the data logger stores a value every 20 seconds over a period of 5 to 10 minutes. All samples are plotted on a graph to obtain a slope and to calculate the flux of CO2, CH4, or N2O per square meter. Details on how the floating chamber is used and on equations and calculations are available.9,10,11


Figure 4 shows the variation of CO2 concentration in the water in the Eastmain-1 reservoir over 16 months, measured using the automated system installed in the Eastmain-1 powerhouse. CO2 concentration increases in the late fall as the ice cover starts to form and decreases in the spring after the ice breaks up.

Comparison of mean fluxes of CO2, CH4, and N2O measured using the floating chambers in the La Grande complex and the Eastmain-1 reservoir provide valuable insights. We can observe a rapid increase in mean CO2 and CH4 emissions from the newly flooded Eastmain-1 reservoir that were about five times higher the first year after flooding than those of natural aquatic ecosystems before impoundment. From 2003 to 2005, mean CO2 was 1,352 ± 1,431 milligrams per square meter per day and mean CH4 was 1.7 ± 1.8 milligrams per square meter per day; in 2006, these values were 6,580 ± 3,567 milligrams CO2 per square meter per day and 7.8 ± 9.5 milligrams CH4 per square meter per day. These emissions then decreased rapidly the second year after flooding and returned to values not significantly different than natural aquatic ecosystems in the second year after impoundment for CH4 and the third year after impoundment for CO2. In 2007, mean CH4 was 3.2 ± 3.1 milligrams per square meter per day; in 2008, mean CO2 was 1,942 ± 1,175 milligrams per square meter per day.

We have not observed any difference before or after flooding for N2O fluxes. CO2, CH4, and N2O fluxes measured in aquatic ecosystems of the Eastmain-1 area are comparable to those reported in the literature for boreal natural lakes and reservoirs on the Cote-Nord and James Bay region of Quebec and Labrador.8,10,11,12

Many parameters can influence GHG fluxes, such as water residence time, the type of flooded vegetation (peat, forest soils, agricultural lands, etc.), and the ratio of surface area flooded to water volume. These parameters influence both the intensity (maximum fluxes reached) and duration of GHG emissions. Generally, in boreal ecosystems, the CO2 maximum flux is reached within the first year, and the time to return to natural values is generally within the first ten years after impoundment. The increase of GHG emissions in a newly formed reservoir is always coupled to an overall increased biological production (fish communities, invertebrates, planktonic communities, etc.).4 Because trees are very slow to decompose, the CO2 emitted is related to the natural decomposition of a fraction of the labile organic matter contained in flooded soils. After this transition period (less than 10 years), CO2 emissions are comparable to natural water bodies and are related to the carbon entering the reservoir as runoff from the watershed. The reservoir effect is over after that period.

We have calculated an annual integrated GHG flux, called a mass balance analysis, using both the automated systems and floating chambers, and they show a difference of only 12 percent between the two techniques. The automated system is a very interesting technique to measure GHG from reservoirs because it reduces the overall cost of measuring in comparison to a traditional field campaign using floating chambers, it requires only one person for maintenance, the systems can be accessed remotely, and it reduces the costs related to safety measures needed to sample water bodies using boats. In fact, automated systems generate about ten times more data for a cost about ten to 20 times cheaper.


Monitoring of natural aquatic ecosystems and the newly flooded Eastmain-1 reservoir demonstrates the following:

The automated systems (at left) used to measure greenhouse gas emissions from the Eastmain-1 reservoir were powered using solar panels.

— CO2 and CH4 gross fluxes increase rapidly within the first year after flooding;

— Eastmain-1 reservoir gross CH4 fluxes returned to values similar to natural aquatic ecosystems the second year after flooding;

— Eastmain-1 reservoir CO2 gross fluxes returned to values similar to natural aquatic ecosystems the third year after flooding; which is well within the ten years generally observed for boreal reservoirs; and

— Automated systems and floating chambers techniques give similar results on annual mass balance.

With a relatively small surface area (603 square kilometres) and a short water residence time (3 to 6 months), Eastmain-1 is a good example of a reservoir that emits a small amount of GHGs. The most efficient thermal power plant, natural gas combined cycle, emits about 380,000 tons of CO2 equivalent per terawatt-hour, which is 23 times more than Eastmain-1.4,13 This good performance of the Eastmain-1 reservoir will be improved, as an additional 780-MW power plant is being installed with the construction of the Eastmain-1A-Rupert diversion. Boreal hydropower plants should be considered as part of the solution to reduce the effect on climate change.


  1. “Assessment of Greenhouse Gas Emissions from the Full Energy Chain for Hydropower, Nuclear Power and Other Energy Sources,” presented at International Atomic Energy Agency advisory group meeting, Vienna, Austria, 1996.
  2. St.Louis, V.L., et al, “Reservoirs Surfaces as Sources of Greenhouse Gases to the Atmosphere: A Global Estimate,” BioScience, Volume 50, 2000, pages 766-775.
  3. Tremblay, Alain, M. Lambert, and L. Gagnon, “CO2 Fluxes from Natural Lakes and Hydroelectric Reservoirs in Canada,” Environmental Management, Volume 33, Supplement 1, July 2004, pages S509-S517.
  4. Tremblay, A., L. Varfalvy, C. Roehm, and M. Garneau, editors, Greenhouse Gas Emissions — Fluxes and Processes: Hydroelectric Reservoirs and Natural Environments, Springer Berlin, Heidelberg, N.Y., 2005.
  5. Climate Change: The Scientific Basis, Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, N.Y., 2001.
  6. Duchemin, E., et al, “Appendix 2 – Possible Approach for Estimating CO2 Emissions from Lands Converted to Permanently Flooded Lands. Basis for Future Methodological Development,” National Greenhouse Gas Inventories Guidelines, Vol. 4 — Agriculture, Forestry and Other Land Use, Institute for Global Environmental Strategies, Kanagawa, Japan, 2006.
  7. Bastien, J., J.-L. Frechette, and R.H. Hesslein, “Continuous Greenhouse Gas Monitoring System — Operating Manual,” Report prepared by Environnement Illimité Inc. for Manitoba Hydro and Hydro-Quebec, 2007.
  8. Tremblay, A., and J. Bastien, “Greenhouse Gases Fluxes from a New Reservoir and Natural Water Bodies in Québec, Canada,” Verhandlungen der Internationalen Vereinigung fur Theoretische und Angewandte Limnologie, Volume 30, No. 6, 2009, pages 866-869
  9. Lambert, M., and J.-L. Frechette, “Analytical Techniques for Measuring Fluxes of CO2 and CH4 from Hydroelectric Reservoirs and Natural Water Bodies,” In Greenhouse Gas Emissions: Fluxes and Processes, Hydroelectric Reservoirs and Natural Environments, Springer, Heidelberg, Germany, 2005.
  10. Demarty, M., J. Bastien, and A. Tremblay, 2008a. “Amenagement Hydroelectrique de l’Eastmain-1 — Etude des flux de gaz a effet de serre — Resultats ete-automne 2008,” Report prepared by Environnement Illimite Inc. for Hydro-Quebec Production, 2008.
  11. Demarty, M., J. Bastien, and A. Tremblay, “Etude des Flux de Gaz a Effet de Serre des Milieux Aquatiques de la Mauricie, de la Cote-Nord et du lac Peribonka — Suivi 2008,” Report prepared by Environnement Illimite Inc. for Hydro-Quebec Production, 2008.
  12. Bastien, J., A. Tremblay, and L. LeDrew, 2009. “Greenhouse Gases Fluxes from Smallwood Reservoir and Natural Water Bodies in Labrador, Newfoundland, Canada,” Verhandlungen der Internationalen Vereinigung fur Theoretische und Angewandte Limnologie, Volume 30, No. 6, 2009, pages 854-857.
  13. Tremblay, A., L. Varfalvy, and M. Lambert, “Greenhouse Gases from Boreal Hydroelectric Reservoirs: 15 Years of Data?” Proceedings of the 15th International Seminar on Hydropower Plants, Vienna University of Technology, Vienna, Austria, 2008.

Alain Tremblay, PhD, is a senior environment advisor with Hydro-Quebec Production (Generation) and manager of the greenhouse gas research program. Julie Bastien is a biologist and project manager and Maud Demarty, PhD, is a biologist with Environnement Illimite Inc. Bastien and Demarty performed the field measurements and statistical analyses and interpreted the results. Claude Demers is a senior science communicator with Hydro-Quebec Public Affairs.


More Hydro Review Current Issue Articles
More Hydro Review Archives Issue Articles

Previous articleDesign, construction sought for 2.6-MW Machai in Pakistan
Next articleAlstom to supply turbines for Angostura hydro project in Chile

No posts to display