Welcome to I R N   International Rivers Network
  rivers, dams &
climate change
  mailing lists
 Dam Reservoirs and Carbon Cycling

Available in PDF format (2.5 MB)

November 2004

A Preliminary Review of the Impact of Dam Reservoirs on Carbon Cycling
By Payal Parekh 1

Image of Report Cover  


The International Hydropower Association (IHA) asserts that hydropower has a very low, or even positive impact on climate change because reservoirs (i.e., artificial lakes) sequester large amounts of carbon. The important question is whether reservoirs are important sinks for anthropogenic carbon. IHA uses an estimate indicating that reservoirs sequester 2.5% of global CO2 emissions. For several reasons this estimate gives a misleading indication of the climate impact of reservoirs. In some cases reservoirs will only be temporary sinks for carbon due to measures to mitigate reservoir sedimentation and dam decommissioning. Where reservoirs are permanent sinks their ability to sequester carbon will end once their storage capacity is filled with sediments. Furthermore, emissions of CO2 and CH4 from reservoirs may have a much higher negative climate impact than the positive impact due to carbon sequestration. In addition, reservoirs trap a significant amount of nutrients including silica that would otherwise have flowed to coastal waters. Silica is an important nutrient for diatom growth which plays an important role in the biological pump that transfers carbon to the deep ocean. A large amount of further research is needed to come up with any reliable estimates of the full climate impacts of reservoir construction.


The International Hydropower Association (IHA) claims in a brochure, "Greenhouse Gas (GHG) Emissions from Reservoirs," that "reservoirs2 in all climates sequester large amounts of carbon." This report is a preliminary attempt to review:

    1) whether reservoirs are capable of absorbing significant amounts of anthropogenic carbon, and

    2) the effect reservoirs have on CO2 uptake in coastal regions.

The global carbon cycle: long timescales

Carbon is exchanged between four spheres: the biosphere, lithosphere, hydrosphere and atmosphere. The amount of carbon stored in each sphere is listed in Table 1.

Table 1. Carbon storage sites

Sphere Size (Pg C (1015 g C))
Atmosphere 750
Surface Ocean 1,020
Deep Ocean 38,100
Biosphere 2,190
Lithosphere 5,000
Source: Climate Change 2001: The Scientific Basis, IPCC.

In the biosphere, plants uptake CO2 from the atmosphere by photosynthesis, which converts CO2 into organic matter. Approximately one–third of the carbon is stored as cellulose in the stems and branches of trees, while two–thirds is stored in the soil as humus, dead organic matter. The decay of organic matter (respiration/remineralization) returns CO2 to the atmosphere. In a non–perturbed state, photosynthesis and respiration are in balance.

Carbon dioxide is exchanged between the atmosphere and the oceans by both physical and biological processes. The physical exchange process, called the solubility pump, is driven by chemical gradients between the air and the oceans. Upon exchange, CO2 is dissolved in surface waters and transferred to the deep ocean by sinking water masses. Upwelled water masses return CO2 to the atmosphere.

The biological pump is the process by which marine phytoplankton uptake CO2 within the euphotic zone (portion of the water column where light penetrates) and convert it to organic matter. Upon death, a portion of the dissolved organic matter and particulate organic matter (i.e., calcareous and siliceous shells produced by marine organisms) are transported to deeper waters by sinking, effectively sequestering carbon from the atmosphere for on the order of 1000 years.

Approximately 1% of organic matter escapes remineralization and is buried on the seafloor. On geological timescales (104 – 106 yrs), these marine sediments are either released via volcanism or eventually uplifted by tectonics and exposed to atmospheric oxygen (weathering), resulting in the release of CO2 to the atmosphere, and thus closing the cycle. Figure 1 illustrates this cycle.

Figure 1: Main components of the natural carbon cycle; fluxes and storage capacity in Pg C yr–1

Figure 1

Source: Climate Change 2001: The Scientific Basis, IPCC.

The global carbon cycle: short timescales

Table 2: Human perturbation to the global carbon budget

CO2 sources Flux (Pg Cyr–1)
Fossil fuel combustion 6.3 +/– 0.4
CO2 sinks  
Atmosphere 3.2+/– 0.1
Ocean 1.7 +/– 0.5
Land 1.4 +/– 0.7
Source: Climate Change 2001: The Scientific Basis, IPCC.

The burning of fossil fuels and cement production is causing CO2 to be released to the atmosphere at much faster rates than due to natural processes. Half of the additional CO2 stays in the atmosphere, with the ocean, forests and soils absorbing the rest. Table 2 summarizes the perturbed global carbon budget.

Can dam reservoirs absorb significant quantities of anthropogenic CO2?

The IHA brochure, "Greenhouse Gas Emissions from Reservoirs," implies that hydropower is a climate–friendly energy source because "reservoirs sequester large amounts of carbon." This claim is largely based on research by Dean and Gorham (1998). Assuming an average sedimentation rate, bulk density and percentage organic carbon in sediments for all the world's reservoirs, Dean and Gorham (1998) estimate an accumulation rate of 0.16 x 1015 grams per year (0.16 Pg yr–1) of organic carbon for reservoirs.

CO2 emissions from fossil fuel burning and cement production are currently estimated at 6.3 Pg C yr–1. Thus, using the numbers from Dean and Gorham (1988) reservoirs are absorbing 2.5% of anthropogenic emissions.

However, this calculation omits several important factors which could turn reservoirs from apparent global sinks to significant global sources of carbon. An assessment of the net impact of reservoirs upon global carbon fluxes should account for factors including in particular:

  • Conversion of pre–dam sinks and sources of carbon.
  • Estimates of the area flooded globally by dam reservoirs range from 500,000 (McAllister et al., 2000) to 1,500,000 km2 (St. Louis et al., 2000). Ecosystems now converted to reservoirs would have been a complex mosaic of sources and sinks of greenhouse gases. Changes in carbon fluxes from riparian ecosystems downstream of dams due to reservoir operation (e.g., forest loss and wetland desiccation) should also be accounted for in estimates of reservoir climate impact.

  • The fraction of carbon entering reservoirs that is converted into CO2 and CH4 and then released to the atmosphere.
  • St. Louis et al. (2000) estimate global emissions of carbon (in the form of both CO2 and CH4) from reservoirs as 0.27 Pg C yr–1. This is 60% higher than the amount of carbon Dean and Gorham (1998) estimate to be absorbed by reservoirs. These emissions are due to the decomposition of both terrestrial biomass flooded by the reservoir and aquatic biomass that grows in the reservoir, as well as carbon entering the reservoir from upstream. The climate impact of carbon emitted from reservoirs is likely to be higher than if this carbon were emitted to the atmosphere under pre–dam conditions due to the significant proportion of methane in reservoir emissions (CH4 has a global warming potential 23 times higher than CO2).

  • The eventual release of sequestered carbon due to measures to conserve reservoir storage capacity, and dam decommissioning.
  • A proportion of carbon sequestrated in reservoirs will be released due to measures such as sluicing or venting which seek to mobilize reservoir sediments and reverse loss of water storage capacity. These techniques are not widespread but work for specific types of reservoirs.

    Dam decommissioning is now increasingly seen as an option, especially for older dams and where reservoirs can no longer serve a useful purpose due to loss of water storage capacity to sediments. The impetus for decommissioning is usually safety or a desire for ecosystem restoration. A certain portion of the accumulated sediments will be released after decommissioning, while another portion may be stabilized in place.

  • The capacity of reservoirs to sequester carbon has a limited lifespan.
  • The ability of a reservoir to sequester carbon is limited by its storage capacity. Measures such as sluicing and venting are ineffective for the majority of reservoirs. These will typically become filled with sediments after 50–200 years (Morris and Fan, 1998). Reservoirs with low sedimentation rates (such as those in high mountain areas with thin soils) will necessarily sequester small amounts of carbon.

  • The fraction of carbon sequestered in reservoirs that would under pre–dam conditions have been sequestered downstream and off–shore.
  • Dams have significantly reduced the transport of sediments from the continents to the world’s oceans. Vörösmarty et al. (2003) estimate that 25% of sediment flux destined for coastal areas is impounded behind reservoirs. Whether the carbon associated with this sediment flux and river discharge is more effectively sequestered in reservoirs or in coastal/deltaic environments depends on both the characteristics of the reservoir and the type of marine environment into which the sediments are discharged.

    Hedges and Keil (1996) estimate that approximately 0.4–0.6 Pg C yr–1 is carried to the oceans by rivers, divided equally between particulate and dissolved matter. Hedges (1992) estimates that globally just under 20% of this terrigenous input is preserved in marine sediments. The remaining 80% is degraded by microbial processes.

    The IHA states in its brochure that "dating of carbon in sediments of continental margins show that the carbon is typically very old, confirming a very low rate of carbon burial" without providing any references. Yet, Bauer et al. (2001) report varying contributions of young organic matter of terrestrial and/or riverine origin in the Mid–Atlantic Bight continental shelf region.

  • The impact upon oceanic carbon uptake of trapping sediments and nutrients behind dams and altering timing and amount of freshwater flows into the oceans.
  • Riverine input is the major external source of nutrients to the ocean. While discharge of nitrate and phosphate downstream of dams by human activities compensates for what is trapped behind dams, in the case of silicate, there is no such compensation. Silicate is necessary for the growth of unicellular algae known as diatoms, which have silica shells. Diatoms are more efficient at transporting carbon than coccolithophorids (carbonate–secreting organisms) and represent 40% of oceanic primary productivity. Silica limitation in coastal regions can thus have an effect on the efficiency of the biological pump, the process by which CO2 is incorporated into organic matter.

    Dams have caused the silicate load into the Black Sea to drop by two–thirds (Humborg et al., 1997). A decrease in silicate input to the Bay of Bengal restricted blooms of diatoms to the monsoon months (Ittekkot et al., 2000).

    While there are no estimates available of the impact of decreased riverine discharge on the biological pump, it is possible to make a back of the envelope, worst–case calculation. Let us assume that primary productivity is stopped in coastal regions due to the complete blockage of riverine sediment and nutrient input by dams. Coastal areas account for 20–30% of global primary productivity (Hedges et al., 1997). A 20% shut–down of the biological pump could result in a 40 ppm increase in atmospheric CO2; a 30% shut–down in a 60 ppm increase (assuming that global turn–off of the biological pump results in approximately 200 ppm increase in atmospheric CO2 (Sarmiento and Toggweiler, 1984; Maier–Reimer et al., 1996)). By comparison the increase in atmospheric CO2 due to fossil fuel burning over the past 150 years is approximately 100 ppm.

    In coastal regions where upwelling rather than riverine input is the primary source of nutrients, decreased fresh water input could effect local circulation. For example, decrease in fresh water input in the East China Sea due to the Three Gorges Dam is expected to reduce cross–shelf upwelling and thus the supply of nutrients to surface waters, reducing the efficiency of the biological pump (Chen, 2002).


Reservoirs have been estimated to absorb 2.5% of anthropogenic carbon emissions globally. Global emissions of carbon from reservoirs, however, have been estimated to be 60% higher than this estimate of global reservoir carbon sequestration. The sequestration capacity of reservoirs has a limited lifespan. Trapping of riverine input of sediments and nutrients by dams can affect the long term silica budget of the oceans, decreasing the effectiveness of the biological pump that sequesters CO2 in deep waters. More studies are needed to quantify the climate impact of reservoirs and how dams affect the supply of nutrients to the ocean and biogeochemical cycling in coastal areas.


1) Payal Parekh, Dept. of Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139–4307 USA; parekh@ocean.mit.edu.

2) In this report, the term reservoir refers to artificial lakes.


Bauer, J.E., E.R.M. Druffel, D.M. Wolgast and S. Griffin (2001), Cycling of dissolved and particulate organic radiocarbon in the northwest Atlantic continental margin, Global Biogeochemical Cycles, 15, 615–636.

Calvert, S.E. and T.F. Pedersen, Organic Carbon Accumulation and Preservation in Marine Sediments: How Important is Anoxia?, Organic Matter: Productivity, Accumulation, and Preservation in Recent and Ancient Sediments, edited by J.K. Whelan and J.W. Farrington, 232–263, Columbia University Press, New York.

Chen, C. (2002), The impact of dams on fisheries: Case of the Three Gorges Dam, Challenges of a Changing Earth, edited by W. Steffen, J. Jager, D.J. Carson and C. Bradshaw, Chapter 16, Springer Verlag, Berlin.

Dean, W. E. and E. Gorham (1998), Magnitude and significance of carbon burial in lakes, reservoirs, and peatlands, Geology, 26, 535–538.

Duchemin, É., M. Lucotte and V. St. Louis (2002), Hydroelectric Reservoirs as an Anthropogenic Source of Greenhouse Gases, World Resource Review, 14, 334–353.

Hedges,J., R. Keil, and R. Benner (1997), What happens to terrestrial organic matter in the ocean?, Organic Geochemistry, 27, 195–212.

Humborg, C., V. Ittekkot, A. Cociasu and B. von Bodungen (1997), Effect of Danube river dam on Black Sea biogeochemistry and ecosystem structure, Nature, 386, 385–388.

Intergovernmental Panel on Climate Change (2001), Climate Change 2001: The Scientific Basis, World Meteorological Organization and United Nations Environment Programme.

International Hydropower Association, Greenhouse Gas Emissions from Reservoirs, brochure.

Ittekkot, V., C. Humborg and P. Schäfer (2000), Hydrological alterations and marine biogeochemistry: A silicate issue?, BioScience, 50, 776–782.

Maier–Reimer, E., U. Mikolajewicz and A. Winguth (1996), Future ocean uptake of CO2 – Interaction between ocean circulation and biology, Climate Dynamics, 12, 711–721.

Martin, J.H., G.A. Knauer, D.M. Karl and W.W. Broenkow (1987), VERTEX: carbon cycling in the northeast Pacific, Deep–Sea Research, 34, 267–285.

McAllister, D., J. Craig, N. Davidson, D. Murray and M. Sneddon (2000), Biodiversity Impacts of Large Dams. Contributing Paper for World Commission on Dams Thematic Review II.1.

Morris, G.L. and J. Fan (1998), Reservoir Sedimentation Handbook, McGraw Hill, NY.

Sarmiento, J.L. and J.R. Toggweiler (1984), A new model for the role of the oceans in determining atmospheric CO2, Nature, 308, 621–624.

St. Louis, V.L., C.A. Kelly, E. Duchemin, J.W.M. Rudd and D.M. Rosenberg (2000) "Reservoir surfaces as sources of greenhouse gases to the atmosphere: A global estimate," BioScience, 50, 766–775.

Vörösmarty, C.J., M. Meybeck, B. Fekete, K. Sharma, P. Green and J.P.M. Syvitski (2003), Anthropogenic sediment retention: major global impact from registered river impoundments, Global and Planetary Change, 39, 169–190.

Walsh, J.J. (1988), On the Nature of Continental Shelves, Academic Press, NY, 520 pp.

 Additional Information

For further information, please contact:

    Patrick McCully, International Rivers Network
    E–mail: patrick@irn.org
    Phone: +1 510–848–1155