Climatic Change (2007) 84:281–294

DOI 10.1007/s10584-007-9261-4

Impacts from decommissioning of hydroelectric

dams: a life cycle perspective

Sergio Pacca

Received: 25 May 2004 / Accepted: 14 March 2007 / Published online: 24 May 2007
# Springer Science + Business Media B.V. 2007

Abstract Greenhouse gas (GHG) emissions from hydroelectric dams are often portrayed as
nonexistent by the hydropower industry and have been largely ignored in global
comparisons of different sources of electricity. However, the life cycle assessment (LCA)
of any hydroelectric plant shows that GHG emissions occur at different phases of the power
plant’s life. This work examines the role of decommissioning hydroelectric dams in
greenhouse gas emissions. Accumulated sediments in reservoirs contain noticeable levels of
carbon, which may be released to the atmosphere upon decommissioning of the dam. The
rate of sediment accumulation and the sediment volume for six of the ten largest United
States hydroelectric power plants is surveyed. The amount of sediments and the respective
carbon content at the moment of dam decommissioning (100 years after construction) was
estimated. The released carbon is partitioned into CO2 and CH4 emissions and converted to
CO2 equivalent emissions using the global warming potential (GWP) method. The global
warming effect (GWE) due to dam decommissioning is normalized to the total electricity
produced over the lifetime of each power plant. The estimated GWE of the power plants
range from 128–380 g of CO2eq./kWh when 11% of the total available sediment organic
carbon (SOC) is mineralized and between 35 and 104 g of CO2eq./kWh when 3% of the
total SOC is mineralized. Though these values are below emission factors for coal power
plants (890 g of CO2eq./kWh), the amount of greenhouse gases emitted by the sediments
upon dam decommissioning is a notable amount that should not be ignored and must be
taken into account when considering construction and relicensing of hydroelectric dams.

1 Introduction

Greenhouse gas (GHG) emissions from reservoirs have become a topic of investigation for
researchers interested in the assessment of the impacts of hydroelectricity on the global
climate (Rudd et al. 1993; Gagnon and Chamberland 1993; Rosa and Schaeffer 1994;

S. Pacca (*)
Escola de Arte Ciências e Humanidades, Universidade de São Paulo, Sao Paulo, Brazil
e-mail: spacca@usp.br
282 Climatic Change (2007) 84:281–294

Fearnside 1995; Duchemin et al. 1995; Galy-Lacaux et al. 1999; Delmas et al. 2001;
Fearnside 2004; Rosa et al. 2004). Other studies have revealed the life-cycle carbon
emissions of hydroelectric plants normalized by their energy output (Chamberland and
Levesque 1996; Gagnon and van de Vate 1997; Tahara et al. 1997; Dones and Frischknecht
1998; Gagnon et al. 2002; Uchiyama 2002). Such studies use a life-cycle assessment (LCA)
approach to find out the carbon emissions of hydroelectric plants and compare those
emissions with other electricity generation technologies. Nevertheless, most of these studies
are incomplete because they do not assess the decommissioning of the dam.
A LCA is a method to evaluate the environmental impacts of a product or service from
“cradle to grave.” In the case of power plants, such impacts manifest themselves during
various phases of the plants’ life time. The importance of these impacts from each given
phase in the overall emissions of the power plant varies and is a function of the technology,
location, and design of the power plant. Most studies claiming to use a LCA method have
only considered the construction phase of hydroelectric plants; while, other non life-cycle
studies have looked at emissions from biomass decay in reservoirs (Rosa et al. 2003;
Fearnside 2004; Rosa et al. 2004; Abril et al. 2006; Santos et al. 2006). A study combining
dam construction, equipment manufacturing, emissions from flooded biomass, and
displacement of active ecosystems is also available (Pacca and Horvath 2002). While
Bioscience (2002) published a special issue about dam decommissioning, there still exists a
general lack of information about the impacts from decommissioning and its global
environmental implications (Hart and Poff 2002). To correctly calculate the greenhouse gas
emissions from hydroelectric plants, these impacts must be known so that they can be
factored into the LCA.
Other electricity generation technologies have been subject to investigations on their
end-of-life repercussions. Nuclear power, for instance, demands an impact assessment
beyond the operational life of the power plant because of long lasting radiation in the fuel
rods and contaminated materials produced during the demolition of old reactors (Wald
2003; OECD 2003). So, decision makers must foresee impacts, such as the disposal of
radioactive materials, when they decide to build and run a nuclear power plant. Similar to
the concern with nuclear power plants, impacts due to dam decommissioning should also be
assessed previous to construction. One remaining question is how to assess such impacts in
the future and how to allocate them to the electricity produced over the power plant’s life.
During their operation, reservoirs accumulate sediments and the environmental implications
of that material during the decommissioning and demolition of the dam is not yet clear.
Over 300 dams have been removed in the United States as they become old enough to
require Federal Energy Regulatory Agency relicensing and as costs of repairs and
maintenance outweigh the benefits of their operation (Hotchkiss et al. 2001). Usually, the
potential negative impacts from the release of trapped sediments are not addressed in
assessments. While these negative impacts have not been researched, they may include
downstream fish kills, filling-in of riffle-pool habitats, blockage of upstream navigational
channels, increased downstream deposition, and destabilization of stream banks. Several
case studies have been done to address sedimentation issues, and they observed that
sediments released after decommissioning harm the environment (Simons and Simons
1991; Stoker and Willians 1991; Pansic et al. 1995; Klumpp et al. 2003). After dam
demolishment, the options are to allow sediments to travel downstream, to dredge or
otherwise control sediments, or to leave sediments in place. However, any option affects the
biochemical properties of the sediments.
Economically, the costs of managing sediments may lead to the earlier retirement of
dams. Sediment management already accounts for 48% of the total dam decommissioning
Climatic Change (2007) 84:281–294 283

costs (Annandale and Morris 1998), and if the costs of environmental impacts from
sediments are added, the significance of sediment management can be even higher.
The volume of all storage reservoirs in the world is 4900 km3, and the annual rate of
sedimentation in reservoirs worldwide is roughly 50 km3 (Sloff 1991). A recent study
concluded that 47% of the major world rivers have their sediment flux reduced due to
impoundments (Syvitski et al. 2005). The annual rate of sedimentation in U.S. reservoirs is
about 1.2 km3 (Glymph 1973). Accordingly, most reservoirs in the world could be filled
with sediments within the next 100 years. Actually, reservoirs are designed based on a
100 year sediment storage capacity (Poff and Hart 2002) and a hydroelectric plant’s lifetime
is in the range of 50–100 years or more (Oud 2002). Sedimentation problems in reservoirs
should be considered during the planning phase and its potential impacts should be
contemplated when comparing between hydroelectric plants and other electricity sources.
Because of the considerable volume of sediments trapped in reservoirs and their
noticeable carbon content, I hypothesize that dam decommissioning is a source of GHG
emissions. The rate of sediment accumulation and the volume of sediment for six of the ten
largest US hydroelectric power plants is assessed. The results express the global warming
effect (GWE) of the carbon in the sediments since it is most likely that still-water saturated
sediments will be exposed to the atmosphere and release the stored carbon. The addition of
this global warming potential to a hydroelectric plant’s total GHG emissions, when
compared to the lifetime energy output of the hydroelectric plant, may demonstrate that
hydroelectricity is a less effective way to combat global warming than previously assumed.

2 Source and fate of sediments

There are two basic sources for sediments in a reservoir:

1. Sediments produced outside the reservoir (allochthonous) load into the reservoir
through river flow.
2. Sediments produced out of dead plankton born in the reservoir (autochthonous).
Artificial impoundments affect the dynamics for both sediment sources. As soon as the
river is dammed, sediment accumulation begins in the reservoir. Water flow velocity
decreases within the reservoir, which allows increased particle deposition and lowers the
turbidity of the water, in turn, increasing light penetration (Klumpp et al. 2003). Thus, the
primary productivity in reservoirs tends to be high, which contributes to the fixation of
CO2. Both the increased particle deposition rate and production rate lead to sediment build-
up above the dam that persists throughout the dam’s life.
Carbon contained in a reservoir’s sediments may originate from carbon previously
available in the watershed and transported through runoff to the river bed and loaded in
reservoirs or may be captured from the atmosphere by autotrophs, such as blue algae and
macrophytes. If the latter prevails, carbon fixed from the atmosphere would be released
when sediments are exposed to the atmosphere, leading to a net zero effect on the global
atmospheric carbon concentration. However, because carbon is absorbed in the form of
CO2, and because of anoxic digestion, it may be released in the form of CH4, which would
potentially result in a negative net impact because of the higher radiative efficiency of CH4
compared to CO2. If no dam exists, the carbon contained in sediment is transported in
runoff flows through the environment. While environmental impacts of erosion have long
been recognized, lesser attention has been devoted to the consequences on the global
carbon cycle (Lal 2003).
284 Climatic Change (2007) 84:281–294

Models to calculate the theoretical volume of sediments deposited in reservoirs are

available (Morris and Fan 1997). However, in this paper, information on actual reservoir
sedimentation rates is used.
Each reservoir’s sediment volume after 100 years is calculated using coefficients from a
regression of the actual sedimentation rate time series. Figure 1 shows the results of the
regressions for the reservoirs managed by the US Army Corps of Engineers (USACE).
Because the forecast based on the regression expresses a tendency that combines the effect
of all data points, I use this method to project the future volume of sediments trapped in the
reservoirs.
Figure 2 shows a plot of the 100 year sediment volume versus the installed power
capacity of the reservoirs analyzed. Although sediment yield has been correlated to the size
of the drainage basin (US Bureau of Reclamation 1987), the installed power of
hydroelectric plants seems to have a direct relationship with the amount of sediments
trapped behind the dams. The installed power is a function of the flow volume and the head
of the power plant.

3 Characterization of sediments in lakes and reservoirs

Because runoff is the major source of sediments, the carbon content in reservoir sediments
is comparable to the carbon content in soils of the donor watershed. Sediments transported
by rivers contain 2–3% of particulate organic carbon (POC) (Lal 2003). After precipitation
to the bottom of reservoirs, soil’s POC may be the main source of organic carbon in
oligotrophic lakes (Kalff 2002). A sediment survey in Puget Sound in Washington State has
shown a range of organic carbon (OC) content between 0.4 and 3.7% based on dry weight
(Pamatmat and Jones 1973).

Fig. 1 Projected sediment content after 100 years of operation

Climatic Change (2007) 84:281–294 285

Fig. 2 Sediment volume after 1.2E+10 m 3

100 years versus installed capacity
Hoover
1.0E+10 m 3

)
3
100 yr. Sediment Volume (m
3
8.0E+09 m

6.0E+09 m 3

4.0E+09 m 3 Garrison
Glen Canyon
Fort Peck Oahe
2.0E+09 m 3
Fort Randall

0.0E+00 m 3
0 500 1,000 1,500 2,000 2,500
Installed capacity (MW)

The specific weight of dried sediment samples collected in Lake Mead (Hoover dam) in
September 1948 and October 1964 were 1,041 and 961 kg/m3, respectively (Dendy and
Champion 1975), and the average of five sampling campaigns in the Fort Peck reservoir
was 929 kg/m3 (Dendy and Champion 1970). Therefore, a carbon content of 2% by dry
sediment weight and a sediment density of 1,000 kg/m3 seems plausible for sediments at
the bottom of reservoirs.
Based on the projected sediment volume, the sediment density, and the carbon content, it
is possible to calculate the total sediment organic carbon (SOC) in the reservoirs. Figure 3
shows the estimated stored SOC in reservoirs versus their respective surface area. Aberg et al.
(2004) concluded that the carbon balances in lakes and reservoirs are comparable. Therefore,
I compared the estimated SOC values with known values for lakes in Alberta, Canada
3.50E+08
Athabasca
3.00E+08

2.50E+08
Claire
Estimated SOC (Mg)

Hoover
2.00E+08 Lesser Slave

lakes
1.50E+08
reservoirs

1.00E+08
Bistcho Glen Canyon
Cold Garrison
5.00E+07 Utikuma Oahe
Fort Randall
La Biche
Fort Peck
0.00E+00
0.00E+00 5.00E+04 1.00E+05 1.50E+05 2.00E+05 2.50E+05

Surface area (ha)

Fig. 3 Estimated stored SOC versus lake or reservoir surface area

286 Climatic Change (2007) 84:281–294

(Campbell et al. 2000). Figure 3 shows that in most of the cases, SOC volume is proportional
to surface area; however, this is not the case for the Hoover and the Oahe dam.
The shape of sediment deposits on the bottom of reservoirs can be characterized by the
area increment method, which assumes that the elevation area curve due to sedimentation
follows the original depth profile of the reservoir (USACE 1995). The area increment
method applied to Fort Peck dam indicates that the surface area of the 100 year sediment
deposit is 2.2×108 m2 (USACE 1995). The same reasoning is applied to Lake Mead
(Hoover dam). By combining information from the capacity area ratio of the dam (US
Bureau of Reclamation 1987) with the 100 year projected sediment volume, the surface
area of the sediment deposit in Lake Mead equals 2.5×108 m2.
The drainage of water through the sediments can be characterized by Darcy’s law.
According to this law, the drainage velocity in a reservoir is proportional to the head of
water above the sediments. As the water table subsides, the drainage velocity decreases and
evaporation becomes the main driver of water disappearance. On average, it takes 1 year to
evaporate 1 m of water in the main stem reservoirs along the Missouri River (USACE
1979).1

4 Mineralization of carbon in sediments

Carbon mineralization rate is the rate of carbon release from sediments in the form of CO2
and CH4 (Liikanen et al. 2002). Sediments at the bottom of reservoirs are substrates for
colonies of decomposers, and decomposition processes in sediments produce CO2, CH4,
and N2O (Liikanen et al. 2002). Mineralization is caused by microorganisms, and CH4
production results from fermentation.
The presence of microorganisms is fundamental for mineralization. In freshwater
ecosystems, decomposers are two to three orders of magnitude more abundant in sediments
than in the water (den Heyer and Kalff 1998). Undisturbed sediments may be a notable
source of CH4 because steady conditions support the development of methanogenic cultures
in contrast to agitated environments that have been shown to produce less CH4
(Dannenberg et al. 1997). Mora-Naranjo et al. (2004) developed a model of methane
emissions in landfills that shows that at 85% of water content, the production of methane
reaches a maximum level in 12 days. Another model of methane emissions from rice
paddies shows that the emissions reach a peak after 26 days (Van Bodegom et al. 2001).
When a reservoir is drained, a slab of eutrophic water rests on top of the sediments. As a
eutrophic reservoir becomes anoxic, its sediments switch from being a nutrient sink to a
source (Harris 1999). Draining reservoirs leads to alternating exposure to oxic and anoxic
conditions that stimulate the overall mineralization of organic matter (OM). The
bioavailability of OM is always greatest under oxic conditions, and the maximum
mineralization of OM occurs under oxic conditions (Bastviken et al. 2004).
The rate of carbon mineralization is directly proportional to temperature increases
(Bartlett and Harriss 1993; Mora-Naranjo et al. 2004). So, as reservoirs are drained, and the
amount of solar radiation transferred and absorbed increases as a result of the shallower
water table, the substrate temperature also increases and more mineralization occurs.

1
For example, assuming that the hydraulic conductivity for sediments is 10−6 m/s (Nazaroff and Alvarez-
Cohen 2001), it takes 2.5 years to drain a 1 m deep slab of water over the sediments in Fort Peck reservoir. In
comparison the evaporation rate of the Missouri stem reservoirs averages 1 m per year (USACE 1979).
Climatic Change (2007) 84:281–294 287

The water table dynamics also affect carbon mineralization rates. In a landfill, water
transports methanogenic microorganisms and is essential for their survival. Because
compaction prevents water and gas circulation within the landfill, it is common to find
municipal solid waste in its original form years after it was buried (Meraz et al. 2004).
Additionally, the height of the water table beneath the landfill surface controls the oxidation
of methane, and in the absence of water, methane is oxidized into CO2.
Thus, the extent of carbon mineralization in a substrate is a function of moisture content.
Hackley et al. (1999) did however find that an earthen pier built on a shallow wetland
contained up to 70% of CH4 concentrations within the headspace of several piezometers
screened at depths ranging from 8 to 17 m. This demonstrates that if the appropriate
conditions are available, methane can be produced at considerable depths. Despite this,
different studies assessing methane production in landfills assume that mineralization
occurs only within a 1 m deep substrate slab (Jones and Nedwell 1992; Kightley et al.
1995; Perera et al. 2002).
Assuming that carbon within a 1 m deep slab of sediments in a reservoir being drained is
subject to mineralization and that the surface area of the 100 year sediment deposit in the
Fort Peck and the Hoover dam is 2.5×108 and 2.2×108 m2, 11 and 3% of the total 100 year
SOC volume is mineralized, respectively.2 This is a conservative assumption because the
mineralization could reach depths beyond the 1 m deep zone.
For each mole of CO2 produced, fermentation produces one mole of CH4 (Kalff 2002).
Thus, stoichiometry imposes a 50% ratio limit in the maximum amount of carbon
mineralized as CH4 by fermentation, which yields a unitary molar ratio between CO2 and
CH4. However, depending on the aeration of the substrate, part of the methane produced
is converted into CO2. When the water table of a wetland is between 20 and 10 cm above
the surface of the sediments, the CO2:CH4 ratio is 10 (Potter 1997). I use this molar ratio in
the assessment of the greenhouse gas emissions. The reason to use this value is because the
drainage of the last 1 m slab of water covering the reservoir’s sediments is one order of
magnitude higher than the average time required for bacteria to maximize methane
production in shallow aquatic environments. Thus, in comparison with time scales
characteristics of microbiological activity, the drainage of the reservoir takes a considerable
amount of time. Therefore, most of the carbon in the sediments is mineralized under anoxic
conditions. If a 1 m high slab of water above the sediments is considered, a significant
amount of organic carbon is mineralized before all water drains out and sediments are
exposed to air. Even after the water table reaches the sediments surface, anoxic
mineralization occurs because sediments below this level are still underwater.

5 Conversion of methane emissions in carbon dioxide equivalents

The global warming effect of greenhouse gas emissions from sediments during the
decommissioning of reservoirs is a function of the amount of sediments mineralized and
released as CO2 and CH4. To calculate the potential impact of GHG emissions from
sediments after decommissioning, I convert CH4 emissions to CO2 equivalent using the
Global Warming Potential (GWP) method. This allows for the comparison of potential
impacts from CH4 and CO2 emissions on the global climate system due to dam
decommissioning. The GWP is a ratio between the radiative efficiency of methane times
the integral of the function representing disappearance of methane in the atmosphere over a

2
The total volume of each reservoir, which was used to derive the percentages, is shown on Table 1.
288 Climatic Change (2007) 84:281–294

given time interval and the radiative efficiency of CO2 times the integral of the function
representing the evolution of the airborne fraction of CO2 in the atmosphere (Houghton
et al. 2001).
The temporal component in the calculation is fundamental for assessing the potential
effects of CH4 compared to the effects of CO2. For example, the IPCC publishes the GWP
for 20, 100, and 500 years, which correspond to 62, 23, and 7 g of CO2 per g of CH4,
respectively (Houghton et al. 2001). Thus, a shorter time period reflects a greater impact of
methane emissions on the global climate forcing.
In a previous analysis of GHG emissions from hydroelectric plants, Fearnside (1997)
considered that reservoirs produce a large pulse of emissions in the first years after filling
due to the decay of flooded biomass. In this case, the use of a 100-year GWP factor to
convert CH4 emissions into CO2 equivalent emissions coincides with an analysis of the
environmental repercussions of the hydroelectric plant 100 years after construction. In
reality, a GWP calculated based on a 100-year time period corresponds to discounting the
impact of each GHG relative to the impact of CO2 over 100 years after its emission
(Fearnside 2002).
Usually, analysts assume a 100 year life time for hydroelectric plants in their
assessments (Gagnon et al. 2002; Oud 2002). In this study, the period of analysis reflects
200 years, which corresponds to the 100 year electricity generation period of the power
plant plus 100 years to compare the potential of the methane released during
decommissioning to the potential global warming effect of CO2. That is, methane
emissions released during the end-of-life of the hydroelectric plant will be compared over
a 100 year period to CO2 equivalent emissions, which represent the potential impact on
global climate. Because the emissions of GHG from sediments occur after electricity
generation comes to an end, a GWP based on a shorter period could be applied. The
adoption of a GWP based on a shorter time period would incorporate the effects of the
emissions from the sediments closer to the 100 years of power generation. However, to be
consistent with the majority of analyses using GWPs, the assessment in this work uses a
100-year GWP, which equals 23.

6 Summary of methods and results

Data on sediment accumulation was collected by contacting personnel from the Army
Corps of Engineers. The “Summary of Reservoir Surveys and Storage Changes – Missouri
River Main Stem” reports the total storage loss in 1,000 acre feet for the Fort Peck,
Garrison, Oahe, Big Bend, Fort Randall, and Gavins Point dams (USACE 2004). Changes
in storage capacity for Hoover dam were copied from the report on “Sediment Deposition
in U.S. Reservoirs” (Dendy and Champion 1975). The measured sediment volume in Lake
Powell (Glen Canyon Dam) was copied from the “1986 Lake Powell Survey” (Ferrari
1988).
Based on the reservoirs studied, the amount of carbon seems to be exponentially
proportional to the installed capacity of the power plant (Fig. 2). The total sediment volume
at the end of the plant’s life, which is set at 100 years, is estimated using a regression
applied to the time series with the actual sedimentation rate for each reservoir. Table 1
shows the 100 year sediment volume, the storage capacity of the reservoirs, the percentage
of the volume of each reservoir that will be filled with sediments after 100 years of
operation, the average annual gross electricity generation of each power plant, the 100 year
sediment mass, and the total mass of carbon available in the sediments. The SOC is
Table 1 Summary of results
Climatic Change (2007) 84:281–294

Reservoir Sediment Annual max . Percent filled after Average annual 100 year 100 year 100 year Potential CO2 Potential CH4
volume storage 100 years gross energy energy sediment mass carbon mass emissions emissions
(m3) (m3) (%) (MWh) (MWh) (Mg) (Mg) (Mg) (Mg)

Hoover 1.04E+10 3.52.E+10 30 4.1E+06 4.1E+08 1.0E+10 2.2E+08 7.3E+08 3.0E+07

Glen Canyon 3.49E+09 3.23.E+10 11 3.5E+06 3.5E+08 3.5E+09 7.4E+07 2.5E+08 9.9E+06
Garrison 3.11E+09 3.00.E+10 10 2.4E+06 2.4E+08 3.1E+09 6.6E+07 2.2E+08 8.8E+06
Oahe 2.42E+09 2.78.E+10 9 2.8E+06 2.8E+08 2.4E+09 5.2E+07 1.7E+08 6.9E+06
Fort Peck 1.98E+09 2.34.E+10 8 1.1E+06 1.1E+08 2.0E+09 4.2E+07 1.4E+08 5.6E+06
Fort Randall 1.92E+09 6.E+09 31 1.8E+06 1.8E+08 1.9E+09 4.1E+07 1.4E+08 5.5E+06
289
290 Climatic Change (2007) 84:281–294

estimated by multiplying the total volume of sediments by the sediment density (1,000 kg/m3)
and by the carbon density in the sediments (2%).
Carbon mineralization produces both CO2 and CH4; however, not all SOC is assumed to
be remineralized. The percentage of carbon mineralized is based on the estimations made
for the Fort Peck and Hoover dams, which are then applied to the other reservoirs.
According to the two case studies, 3 and 11% of the total available SOC is mineralized.
The CO2:CH4 molar ratio used in the analysis is 10, and a GWP of 23 is used to convert
CH4 into CO2 equivalents. Results in metric tons of CO2 equivalent are normalized by the
predicted electricity output of the power plant for 100 years. The average annual gross
energy output of the hydroelectric plants managed by the U.S. Army Corps of Engineers
was obtained from the Monthly Project Statistics (USACE 2004) and the net energy output
for the 2003 fiscal year is used for the Hoover and Glen Canyon power plant (US Bureau of
Reclamation 2004a, b). The final global warming effect (GWE) is expressed in grams of
CO2 equivalent per kWh.

7 Discussion

The present assessment allows for the comparison of the GWE due to the decommissioning
of dams with other life-cycle phases or with other studies of hydroelectric plants. The
estimated emissions due to the construction of the Glen Canyon Dam are 800,000 metric
tons of CO2 equivalent (Pacca and Horvath 2002). Emissions during the operation of the
dam, which are associated with the decay of the biomass in the reservoir, account for
3,500,000 metric tons of CO2 equivalent. Finally, emissions from decommissioning of the
reservoir dam account for 33,000,000 metric tons of CO2 equivalent (Fig. 4). Therefore,
emissions from the decommissioning account for more than nine times the emissions from
biomass decay in the reservoir, and render a tremendous contribution to the overall life
cycle emissions of hydroelectric plants. In the case of the Glen Canyon hydroelectric plant,
emissions from decommissioning are one order of magnitude higher than the emissions
from the biomass decay in the reservoir, which are one order of magnitude higher than the
emissions from the construction of the dam.
This analysis looks at carbon emissions at the time of dam decommissioning, which
assumes that all SOC will be emitted when the sediments are exposed to air. The emissions
should occur over a very short period of time since I am only looking at the immediate
effects during the hydroelectric power plant’s end-of-life.
Figure 5 shows a comparison of average emission factors from fossil fueled power plants
based on a literature review on LCA (Pacca 2003) and the emission factors found in this
study observing the maximum and the minimum molar ratios between CO2 and CH4.
Emission factors due to dam decommissioning are on the same order of magnitude as
emissions from fossil fueled power plants, if 11% of SOC is mineralized.
Various factors contribute to the mineralization of SOC during decommissioning. In this
study, based on calculations from known parameters of two existing reservoirs, I include a 3
or 11% SOC mineralization. The literature on methane emissions shows that methane may
be produced deeper in soil/landfill profiles, which would imply that more SOC could be
mineralized. If sediments are dredged from the reservoir, a greater fraction of the total SOC
might be mineralized. Additionally, it is important to remember that dredging can create
different conditions in a reservoir and how they affect the share of CH4 produced and
released. If oxidizing conditions are present, part of the CH4 is converted into CO2, and the
1:10 ratio that was applied in this analysis should be reassessed.
Climatic Change (2007) 84:281–294 291

1.E+09
5.2E+07
1.4E+07
5.0E+06
2.0E+06 max.
8.0E+05
min.
1.E+06
Mg of CO2 eq.

1.E+03

1.E+00
construction operation decommissioning
life cycle phases

Fig. 4 Life cycle emissions of Glen Canyon Dam

The difference in carbon found in the reservoir or flowing through the river is the timing
of mineralization because the carbon that would be mineralized in a river bed is otherwise
held in the reservoir to be mineralized later on. If the SOC in the bottom of the reservoir is
being fixed from the atmosphere instead of being transported through run-off, there is a
positive effect on global warming. This is because carbon is taken up from the atmosphere
in the form of CO2 and later released as CH4, which has a higher GWP than CO2. Water
always runs through the thalweg and the ultimate destination of carbon in the soil is the
ocean. A reservoir prevents carbon from reaching the ocean and eventually sends it back to
the atmosphere, which contributes to enhancing the earth’s radiative forcing.
The total filling of reservoirs by sediments may take more than 1,000 years. Therefore
sediment accumulation per se may not justify the removal of dams. Other reasons, such as
structural aging and environmental problems, may lead to the decommissioning of dams
and force water agencies to take some action to prevent negative environmental impacts.
Dams are not eternal structures. Thus, the end-of-life for hydroelectric plants needs to be

1000
890

750
g of CO2 eq./kWh

Max. value
500 454 Min. value
380 Avr. thermo.

263
250 194
157 148 128
104
72 53
43 40 35
0
Fort Peck Fort Garrison Glen Hoover Oahe Natural Coal
Randall Canyon Gas

Fig. 5 Emission factors for decommissioning versus GWE of fossil fueled power plants
292 Climatic Change (2007) 84:281–294

factored into the assessment of this energy source and its global climate change
implications should be accounted for.
This study calculates the global climate impacts of hydroelectric plants at the end of their
life cycle. The hydropower industry should recognize the impact of a reservoir’s sediments
and incorporate these results into their assessment of hydroelectric power plants. Decision
makers also need to consider the potential impacts from decommissioning dams when
reviewing the relicensing of hydroelectric power plants or the construction of new ones.
If the accumulation of sediments in reservoirs is a function of the installed capacity of
hydroelectric plants, it makes sense to support the installation of a chain of smaller power
plants on the same river instead of one larger hydroelectric plant so that harmful GHG
emissions are minimized. In addition, because small hydroelectric plants trap less sediment,
they favor management actions that can effectively limit sediment accumulation by flushing
smaller sediment amounts downstream. Thus, instead of becoming a carbon source,
sediments return to the stream flow that reaches the oceans, which is the ultimate global
carbon sink.

Acknowledgements I would like to thank Dr. Ian Campbell for his generous advice, Tiffany Bock for her
help with data collection, and Andrea Kraljevic for editing. I am also grateful to the UC Energy Institute
California Energy Studies Program.

References

Aberg J, Bergstrom A, Algesten G, Soderback K, Jansson M (2004) A comparison of the carbon balances of
a natural lake and a hydroelectric reservoir in Northern Sweden. Water Res 38(3):531–538
Abril G, Richard S, Guerin F (2006) In situ measurements of dissolved gases (CO2 and CH4) in a wide range
of concentrations in a tropical reservoir using an equilibrator. Sci Total Environ 354:246–251
Annandale GW, Morris GL (1998) Decommissioning of dams. In: Morris GL, Fan J (eds) Reservoir
sedimentation handbook. McGraw-Hill, New York
Bartlett KB, Harriss RC (1993) Review and assessment of methane emissions from wetlands. Chemosphere
26:261–320
Bastviken D, Persson L, Odham G, Tranvik L (2004) Degradation of dissolved organic matter in oxic and
anoxic lake water. Limnol Oceanogr 49(1):109–116
Campbell ID, Campbell C, Vitt DH, Laird LD, Trew D, Kotak B, LeClair D, Bayley S (2000) A first estimate
of organic carbon strorage in Holocene Lake sediments in Alberta, Canada. J Paleolimnol 24:395–400
Chamberland A, Levesque S (1996) Hydroelectricity, an options to reduce greenhouse gas emissions from
thermal power plants. Energy Convers Manag 37(6–8):885–890
Dannenberg S, Wudler J, Conrad R (1997) Agitation of anoxic paddy soils slurries affects the performance of
the methanogenic microbial community. Microbiol Ecol 22:257–263
Delmas R, Galy-Lacaux C, Richard S (2001) Emissions of greenhouse gases from the tropical hydroelectric
reservoir of Petit Saut (French Guiana) compared with emissions from thermal alternatives. Glob
Biogeochem Cycles 15(4):993–1003
Dendy FE, Champion WA (1970) Summary of reservoir sediment deposition surveys made in the United
States through 1970. Miscellaneous Publication No. 1266, Agricultural Research Service, USDA.
Dendy FE, Champion WA (1975) Sediment deposition in U.S. reservoirs: summary of data reported through
1975. Miscellaneous Publication Number 1362, United States Department of Agriculture.
den Heyer C, Kalff J (1998) Organic matter mineralization rates in sediments: a within- and among lake
study. Limnol Oceanogr 43(4):695–705
Dones R, Frischknecht R (1998) Life-cycle assessment of photovoltaic systems: results of Swiss studies on
energy chains. Prog Photovolt: Res Appl 6:117–125
Duchemin E, Lucotte M, Canuel R, Chamberland A (1995) Production of the greenhouse gases CH4 and
CO2 by hydroelectric reservoirs of the boreal region. Glob Biogeochem Cycles 9(4):529–540
Fearnside PM (1995) Hydroelectric dams in the Brazilian Amazonia as sources of greenhouse gases. Environ
Conserv 22:7–19
Climatic Change (2007) 84:281–294 293

Fearnside PM (1997) Greenhouse-gas emissions from Amazonian hydroelectric reservoirs: the example of
Brazil’s Tucurui dam as compared to fossil fuel alternatives. Environ Conserv 24:64–65
Fearnside PM (2002) The preference in global warming calculations: a proposal for a unified index. Ecol
Econ 41:21–31
Fearnside PM (2004) Greenhouse gas emissions from hydroelectric dams: controversies provide a
springboard for rethinking a supposedly ‘clean’ energy source. Clim Change 66(1–2):1–8
Ferrari RL (1988) Lake Powell survey. USBR Technical Report REC-ERC-88-6
Gagnon L, Chamberland A (1993) Emissions from hydroelectric reservoirs and comparison of
hydroelectricity, natural-gas and oil. Ambio 22(8):568–569
Gagnon L, van de Vate J (1997) Greenhouse gas emissions from hydropower: the state of research in 1996.
Energy Policy 25(1):7–13
Gagnon L, Bélanger C, Uchiyama Y (2002) Life-cycle assessment of electricity generation options: the status
of research in year 2001. Energy Policy 30(14):1267–1278
Galy-Lacaux C, Delmas R, Kouadio RS, Gosse P (1999) Long-term greenhouse gas emissions from
hydroelectric reservoirs in tropical forest regions. Glob Biogeochem Cycles 13:503–517
Glymph LM (1973) Summary: sedimentation of reservoirs. In: Ackermann WC, White GF, Worthington EB
(eds) Man-made lakes: their problems and environmental effects. Geophysical monograph series.
American Geophysical Union, Washington, DC
Hackley KC, Liu CL, Trainor D (1999) Isotopic identification of the source of methane in subsurface
sediments of an area surrounded by waste disposal facilities. Appl Geochem 14:119–131
Harris G (1999) This is not the end of limnology (or of science): the world may well be a lot simpler than we
think. Freshw Biol 42:689–706
Hart DD, Poff LN (2002) A special section on dam removal and river restoration. BioScience 52:653–655
Hotchkiss M, Barber M, Wohl E (2001) Dam decommissioning: decisions and unresolved sediment transport
issues. In: Proceedings of the seventh federal interagency sedimentation conference, Reno, NV
Houghton JT, Ding Y, Griggs DJ, Noguer M, van der Linden PJ, Dai X, Maskell K, Johnson CA (eds.)
(2001) Climate change 2001: the scientific basis, contribution of working group I to the third assessment
report of the Intergovernmental Panel on Climate Change. Intergovernamental Panel on Climate Change
(IPCC), Cambridge University Press, New York
Jones HA, Nedwell DB (1992) Methane emission and methane oxidation in land-fill cover soil. FEMS
Microbiol Ecol 102:185–195
Kalff J (2002) Limnology: inland water ecosystems. Prentice-Hall, Upper Saddle River, NJ
Kightley D, Nedwell DB, Cooper M (1995) Capacity for methane oxidation in landfill cover soils measured
in laboratory scale soil microorganisms. Appl Environ Microbiol 61:592–601
Klumpp C, Bountry J, Greimann B (2003) Case studies in dam decommissioning at the Bureau of
Reclamation. In: Proceedings world water and environmental resources congress, EWRI, Philadelphia,
PA
Lal R (2003) Soil erosion and the global carbon budget. Environ Int 29:437–450
Liikanen A, Flojt L, Martikainen P (2002) Gas dynamics in eutrophic lake sediments affected by oxygen,
nitrate, and sulfate. J Environ Qual 31:338–349
Meraz RL, Vidales A, Dominguez A (2004) A fractal-like kinetics equation to calculate landfill methane
production. Fuel 83:73–80
Mora-Naranjo N, Meima JA, Haarstrick A, Hempel DC (2004) Modeling and experimental investigation
of environmental influences on the acetate and methane formation in solid waste. Waste Manage 24:
763–773
Morris GL, Fan J (1997) Reservoir sedimentation handbook: design and management of dams, reservoirs,
and watersheds for sustainable use. McGraw-Hill, New York
Nazaroff WW, Alvarez-Cohen L (2001) Environmental engineering science. Wiley, New York
OECD (2003) Decommissioning nuclear power plants: policies, strategies, and costs. Nuclear Energy
Agency, Paris, France
Oud E (2002) The evolving context for hydropower development. Energy Policy 30(14):1215–1223
Pacca S (2003) Global warming effect applied to electricity generation technologies. PhD Dissertation,
University of California, Berkeley, CA
Pacca S, Horvath A (2002) Greenhouse gas emissions from building and operating electric power plants in
the Upper Colorado River Basin. Environ Sci Technol 36:3194–3200
Pamatmat MM, Jones RS (1973) Oxidation of organic matter in sediments. Ecological Research Series, EPA
660/3-73-005, Office of Research and Development, US EPA.
Pansic N, Austin RJ, Finis M (1995) Sediment management for dam decommissioning, sediment
management and erosion control on water resources projects. Fifteenth Annual USCOLD Lecture
Series, San Francisco California, May 15–19, United States Committee on Large Dams, Denver, CO
294 Climatic Change (2007) 84:281–294

Perera LAK, Achari G, Hettiaratchi JPA (2002) Determination of source strength of landfill gas: a numerical
modeling approach. J Environ Eng 128(5):461–471 (May)
Poff NL, Hart DD (2002) How dams vary and why it matters for the emerging science of dam removal.
BioScience 52(8):659–668
Potter CS (1997) An ecosystem simulation model for methane production and emission from wetlands. Glob
Biogeochem Cycles 11(4):495–506
Rosa LP, Schaeffer R (1994) Greenhouse-gas emissions from hydroelectric reservoirs. Ambio 23(2):164–165
Rosa LP, Santos MA, Matvienko B, Sikar E, Lourenço RSM, Menezes CF (2003) Biogenic gas production
from major Amazon reservoirs, Brazil. Hydrol Process 17:1443–1450
Rosa LP, Santos MA, Matvienko B, Sikar E (2004) Greenhouse gas emissions from hydroelectric reservoirs
in tropical regions. Clim Change 66(1–2):9–21
Rudd JWM, Harris R, Kelly CA, Hecky RE (1993) Are hydroelectric reservoirs significant sources of
greenhouse gases? Ambio 22(4):246–248
Santos MA, Rosa LP, Sikar B, Sikar E, Santos EO (2006) Gross greenhouse gas fluxes from hydropower
reservoir compared to thermo power plants. Energy Policy 34:481–488
Simons RK, Simons DB (1991) Sediment problems associated with dam removal, Muskegon River,
Michigan, ASCE Hydraulic Engineering. Proceedings of the 1991 National Conference, Nashville, TN
Sloff CJ (1991) Reservoir sedimentation: a literature review, communications on hydraulic and geotechnical
engineering. Report no. 91-2, ISSN 0169-6548, Delft University of Technology
Stoker BA, Williams DT (1991) Sediment modeling of dam removal alternatives, Elwha River, Washington,
Hydraulic Engineering, ASCE Hydraulic Engineering. Proceedings of the 1991 National Conference,
Nashville, TN
Syvitski JPM, Vörösmarty CJ, Ketter AJ, Green P (2005) Impacts of humans on the flux of terrestrial
sediment to the global coastal ocean. Science 308:376–380
Tahara K, Kojima T, Inaba A (1997) Evaluation of CO2 payback time of power plants by LCA. Energy
Convers Manag 38:S615–S620
Uchiyama, Y (2002) Present efforts of saving energy and future energy demand/supply in Japan. Energy
Convers Manag 43:1123–1131
USACE – US (1979) Engineering and Design – Missouri River Main Stem Reservoir, System Reservoir
Regulation Manual – Master Manual, Army Corps of Engineers, US Army Engineer Division, Omaha,
NE. http://www.nwd-mr.usace.army.mil/rcc/reports/pdfs/Master_Manual_1979.pdf
USACE – US (1995) Engineering and design – sedimentation investigations of rivers and reservoirs.
Publication Number: EM 1110-2-4000. Army Corps of Engineers, Omaha, NE
USACE – US (2004) Missouri River Region Water Management Information. In: Monthly Project Statistics,
Army Corps of Engineers, Northwestern Division Missouri River Basin. http://www.nwd-mr.usace.army.
mil/rcc/, Accessed May 2004.
US Bureau of Reclamation (1987) Design of small dams. United States Department of Interior, Bureau of
Reclamation, Denver, CO
US Bureau of Reclamation (2004a) Glen Canyon dam facts. US Bureau of Reclamation. http://www.uc.usbr.
gov/information/gcdfacts.html, Accessed March 2004.
US Bureau of Reclamation (2004b) Hoover power plant Boulder Canyon Project. US Bureau of Reclamation.
http://www.usbr.gov/power/data/sites/hoover/hoover.pdf, Accessed May 2004.
Van Bodegom PM, Wassman R, Metra-Corton TM (2001) A process based model for methane emission from
flooded rice paddies. Glob Biogeochem Cycles 15(1):247–263
Wald ML (2003) Dismantling nuclear reactors. Scientific American 288(3):60–69 (March)