International Rivers' Comments on Greenhouse Gas Emissions to UNESCO/IHA

Friday, March 13, 2009

International Rivers' Comments on the UNESCO/IHA Measurement Specification Guidance

The document provides a lot of detail on measurements needed to assess greenhouse gas emissions from reservoirs. But it does not present the relative importance/weighting of each of the components of the equation and therefore how each of the measurements fits into the overall calculation of the climate impact of building a dam and reservoir. Explaining the relative importance of the different processes is important as this would clarify which processes are the main contributors to GHG emissions from reservoirs. Then attention can be focused on the measurements of processes that account for the bulk of the emissions. Additionally, the document does not discuss the accuracy and precision (i.e. what are the error bars) for any of the measurement techniques discussed.

The document narrowly defines gross emissions as "those measured at the water-air surface" (p. 16, lines 510-511). Emissions from the construction of the dam are not included. This includes the use of fossil fuels by machinery and the production of building materials, in particular cement, and deforestation for the construction of roads, power lines, resettlement sites etc. Greenhouse gas emissions from the above-water decay of semi-submerged trees would also not be considered under this definition. A recent publication from Fearnside (2008) discusses this in further detail.

While the document discusses the importance of making measurements downstream of the reservoir site (p. 22, lines 780-788), it does not specifically call for assessing the change in nutrient delivery and primary productivity in the coastal zone. Primary productivity in the ocean is an important global sink of carbon that is weakened by the damming of rivers. This is particularly important in light of the fact that the Amazon River plume provides the fuel for the carbon drawdown of 27 million metric tons annually (Subramaniam et al., 2008). It is likely that river plumes in other parts of the world play a similar role.

On a related note, the document assumes that carbon sedimented in the reservoir is a net sink (p. 13-14, lines 418-426). The sequestration that occurs in a reservoir has a limited lifespan (because of e.g. dam decommissioning, dam failure or efforts to scour/dredge the reservoir to increase its lifespan). Under unperturbed conditions, this carbon may have been deposited in the ocean, which is likely to be more effective at carbon sequestration.

While the document considers the possibility of methane production by methanogenesis (p. 13, line 420-421), the procedures for measuring carbon mass flow and carbon storage in sediments (p. 39, lines 1342-1359) fails to recommend the use of flux chambers to quantify the outgassing of methane. In fact, this section is rather sparse on details concerning measurements of carbon flows in sediments.

On page 26, Figure 4, the relationship between reservoir age and CO2 flux is displayed only for boreal dams. Does this relationship also hold for tropical dams, the primary contributor of greenhouse gas emissions from hydropower? The gray bar, which represents the "natural variation range" is misleading. Prior to the construction of the dam, many of these reservoirs were actually wetlands or forest, i.e. carbon sinks, not sources. Even if the CO2 flux out of the reservoir is comparable to emissions from natural lakes, this is of little relevance given that reservoirs normally flood mainly terrestrial rather than lacustrine ecosystems.


Fearnside, P.M. (2008)
Hidrelétricas como Fábricas de Metano e o Papel dos Reservatórios em Áreas de Floresta Topical na Emissão de Gases de Efeito Estufa, Oecologia Brasiliensis, 12:100-115.

Subramaniam, A. et al. (2008) Amazon River Enhances Diazotrophy and Carbon Sequestration in the Tropical North Atlantic Ocean, PNAS, 105:10460-10465.