58 Projecting carbon dioxide exchange from a Canadian boreal hydroelectric reservoir for the possible engineering life span of the reservoir

Wednesday, 30 May 2012
Rooftop Ballroom (Omni Parker House)
Youngil Kim, McGill University, Montreal, QC, Canada; and N. T. Roulet, C. Li, S. Frolking, I. B. Strachan, J. Wu, and A. Tremblay

Quantification of greenhouse gas emission from boreal hydroelectric reservoirs is important in relation to the impact of energy development on climate change. We built a process-based reservoir model of FF-DNDC in order to project carbon dioxide (CO2) flux from a boreal hydroelectric reservoir over the life span of the reservoir (i.e., 100 years). The framework of this reservoir model is Forest-DNDC (a terrestrial biogeochemical model) that supports detailed soil C processes with considering the change in redox chemistry in response to flooding. To represent the reservoir C flux, this terrestrial model was modified by changing soil decomposition rates, adding sedimentation, and replicating temperature at the water-sediment interface. The study site is the Eastmain-1 reservoir in northern Québec, Canada, where main flooded lands comprised boreal forests and peatlands, and inundation commenced in 2006. FF-DNDC was used to estimate CO2 emission from the reservoir surface over the first four years of flooding, and then used to project the potential emission over 100 years based on the future climate change scenarios. From 2006 to 2009, modeled daily CO2 emissions averaged 1.42 g C m–2 d–1 (a range of 0.75 to 3.24 g C m–2 d–1) from flooded forest and 0.74 g C m–2 d–1 (a range of 0.51 to 1.09 g C m–2 d–1) from flooded peatland. Simulated daily CO2 emissions for both flooded forest and peatland decreased with time since flooding. When comparing between the modeled emissions and the measured emissions using an eddy covariance system at flooded forest, the modeled emissions were larger than the observed ones. However, no statistically significant difference was found between the modeled and measured emissions. The 100 year simulations showed a decrease of emissions over time from the highest amount in the first year. For the all climate change scenarios, annual CO2 emissions were fluctuated from 679 to 8 g C m–2 yr–1 with an average of 57 g C m–2 yr–1 at flooded forest, and they did from 291 to 101 g C m–2 yr–1 with an average of 173 g C m–2 yr–1 at flooded peatland. The model outputs were converted as the net emissions per energy production for the entire area of the reservoir. Compared to the net emissions (i.e., difference in emissions from flooded to natural ecosystems) from other methods of energy production, the simulated emission from the Eastmain-1 reservoir was significantly smaller than fossil fuel energy by burning coal, oil, or natural gas.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner