African Geographical Review
ISSN: 1937-6812 (Print) 2163-2642 (Online) Journal homepage: http://www.tandfonline.com/loi/rafg20
Impact of climate change and variability on
hydropower in Ghana
Sylvester Afram Boadi & Kwadwo Owusu
To cite this article: Sylvester Afram Boadi & Kwadwo Owusu (2017): Impact of climate
change and variability on hydropower in Ghana, African Geographical Review, DOI:
10.1080/19376812.2017.1284598
To link to this article:  https://doi.org/10.1080/19376812.2017.1284598
Published online: 13 Feb 2017.
Submit your article to this journal 
Article views: 163
View Crossmark data
Citing articles: 1 View citing articles 
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=rafg20
African Geographical Review, 2017
http://dx.doi.org/10.1080/19376812.2017.1284598
Impact of climate change and variability on hydropower in Ghana
Sylvester Afram Boadi* and Kwadwo Owusu
Climate Change & Sustainable Development Programme, University of Ghana, P.O. Box LG 59,
Legon, Accra, Ghana
(Received 21 December 2015; accepted 15 November 2016)
Ghana continues to rely heavily on hydropower for her electricity needs. This
hydropower reliance cannot ensure sustainable development since there is a strong
association between hydropower production and climate variability and change
including ENSO-related lake water levels reduction. Using regression analysis this
study found that rainfall variability accounted for 21% of the inter-annual fluctuations
in power generation from the Akosombo Hydroelectric power station between 1970
and 1990 while ENSO and lake water level accounted for 72.4% of the inter-annual
fluctuations between 1991 and 2010. There is therefore the need to diversify power
production to attain energy security in Ghana.
Keywords: climate change and variability; El Niño-Southern Oscillation;
hydropower; energy security; Ghana
Background
The impacts of climate variability and change are real and would continue to affect
sensitive sectors of the global economy. Productive sectors such as agriculture, water,
health, energy, and transport among others bear the brunt of these variability and change
in the world’s climate (IPCC, 2007). The reliance on climate sensitive sectors such as
hydropower has become a challenge to sustainable development as a result of climate
variability and change impacts on power generation (Okudzeto, Mariki, Paepe, &
Sedegah, 2014). There is evidence showing that African countries have not been able to
meet their development targets due to climate variability and change impacts on hydro-
power production (Beilfuss, 2012), resulting from both anthropogenic climate change
and natural variability such as the El Niño-Southern Oscillation (ENSO) which is the
largest known source of climate variability in the tropics (Camberlin, Janicot, &
Poccard, 2001; Collins et al., 2010). Ghana has the potential to grow its economy
through industrialization, increased productivity, job creation, and equitable wealth
distribution (Ministry of Energy [MoE], 2010a). However, in order to achieve this
economic development, there is the need for adequate generation and supply of energy.
In addition to an annually increasing power demand locally, the Volta River Author-
ity (VRA) exports power to the Communauté Electrique du Benin [CEB] – comprising
Togo and Benin, SONABEL (Burkina Faso), while also interchanging power with Com-
pagnie Ivoirienne d’Electricité [CIE] of La Cote d’Ivoire, as part of efforts to achieve a
West African Power Pool (VRA, 2015). According to Badu (2013), Ghana requires
6000 MW of power by 2015 to guarantee sustainable growth and development.
*Corresponding author. Email: safram_boadi@st.ug.edu.gh
© 2017 The African Specialty Group of the American Association of Geographers
2 S.A. Boadi and K. Owusu
However, the country’s current total installed capacity is just about 2846.5 MW (VRA,
[VRA], 2014). Ghana, thus, has an electricity generation deficit of over 3000 MW.
Power ration has been frequent in the past three decades and the country has seen an
intensification of power cuts that has resulted in the collapse of many manufacturing
industries in the last three years.
Ghana’s energy strategy to bridge this power deficit however weighs heavily on
hydropower and included the recently completed 400 MW Bui Hydroelectric power pro-
ject, the development of a planned 625 MW Western Rivers Hydropower project and
90 MW Juale Hydropower project, the implementation of the Aboadze TICO Power
plant steam turbine and the operationalization of the 125 MW Osagyefo Power Barge
(MoE 2010b). Also, the government has set a target of generating 10% of power from
other renewable sources which include mini hydro plants (Mathrani et al., 2013). This
reveals Ghana’s continuous dependence on hydroelectricity for the attainment of secure
and sustainable electricity supply. Investments in the major hydropower infrastructure in
the country were done in the early 1960s when the rainfall pattern was favorable
(Owusu, 2009). These investments were made with the assumption that mean rainfall
will continue at levels similar to those prior to independence (Owusu, Waylen, & Qiu,
2008). Historic records have however shown that rainfall pattern over West Africa has
undergone a period of decline, accompanied by a series of severe droughts and a shift
in the rainfall regime since the 1970s (Kasei, 2009; Owusu, 2009). Owusu et al. (2008)
for instance described the declining rainfall total in the Volta Basin since the early
1970s as having a serious negative impact on power production from the Akosombo
Hydropower Dam. Other studies have also demonstrated the strong effect of ENSO on
rainfall distribution and runoff in West Africa (Losada et al., 2012; Rodríguez-Fonseca
et al., 2011). Ghana’s power sector is already showing signs of susceptibility to variabil-
ity and change in rainfall (Bekoe & Logah, 2013), including variability due to ENSO
(Owusu et al., 2008; Waylen & Owusu, 2014).
The overdependence on hydropower has exposed the country to the impacts of
climate variability and change. One of the potential effects of variability and change in
climate, especially reduction in the amount and variation in the distribution of rainfall,
is the possibility of changes to river flow and runoff which will affect energy supply
from hydropower sources (Energy Commission & United Nations Ghana, 2012;
Harrison, Whittington, & Gundry, 1998). It appears that the country has ignored the vis-
ible climatic challenges faced by the existing hydropower infrastructure and continues
to pursue a hydro-dependent energy policy and strategy. This poses a developmental
challenge as even in a stable climate, the year to year variability mostly associated with
ENSO continues to affect power production from hydro sources. Meanwhile Ghana’s
climate has been projected to be warmer and drier due to climate change (Environmen-
tal Protection Agency & Ministry of Environment, Science, Technology and Innovation
[EPA & MESTI], 2015; McSweeney, New, & Lizcano, 2010; Minia, 2008). It is on this
premise that this study seeks to use empirical records from the Akosombo Hydroelectric
Power Station in the Volta Basin to demonstrate that climate variability including vari-
ability due to ENSO, will make a hydropower-dependent energy strategy unsustainable
for current and future power needs. The objective of this study is therefore to identify
the influence of variability and change in rainfall including ENSO effect on the
Akosombo Hydroelectric Power project’s (the largest dam in Ghana) capacity to provide
secure and sustainable electricity for Ghana’s development.
African Geographical Review 3
Climate variability and change impacts on hydropower generation
Energy systems are increasingly being affected by changing climatic trends. For exam-
ple, increasing variability, frequent extremes, and large inter-annual variation in climate
parameters are severely affecting hydropower production in some regions (Ebinger &
Vergara, 2011). Hydropower is especially vulnerable as it relies on water resources such
as runoff, river flow (Hamududu, 2012), and precipitation. Beilfuss (2012) evaluated the
hydrological risk of hydro-dependent power systems in Southern Africa in the face of
climate change using the Zambezi Basin as a case study. He found that droughts have
significant impacts on river flow and hydropower production in the basin. The IPCC
(2001, 2013) has categorized the Zambezi Basin as the river basin exhibiting the most
severe effects of climate change because of the impacts of temperature increases and
reductions in rainfall. There are two large hydropower dams operating on the main-
stream Zambezi River namely, Kariba Dam and Cahora Bassa Dam. Climate variability
and associated droughts have had significant impacts on the potential of these dams to
meet the firm power requirements and also the total power generation goals (Beilfuss,
2012). The World Bank (2010) quantified the potential climate change impacts in the
Zambezi River Basin to be a firm energy fall of 32% per year and an average annual
energy generation fall of 21% per year under moderate climate change scenarios. Under
the less optimistic scenarios, firm power reduces by 43% per year while average annual
energy falls by 25%. Yamba et al. (2011) also provided evidence of recent drought
occurrences in the South Africa Development Community region which resulted in a
deficiency of water supply and subsequently affected hydropower generation in most of
the drought affected countries.
Grijsen (2014) in a study presented a Climate Risk Assessment (CRA) for the five
main basins in Cameroon, focusing on climate change impacts on water resources avail-
ability for power generation mostly in the Sanaga, Benue, Nyong, and Ntem River
Basins. The study recorded a 16% reduction in runoff from the Sanaga basin due to
decrease in precipitation around 1970. Munang and Ayongue (2010) in their study
recorded an increase in temperature of 0.8 °C, a decline in rainfall by 112 mm/yr
(6.5%) and reduction in runoff by 142 m3/s (7.5%) for the Sanaga Basin. According to
Grijsen (2014), the cumulative effect of these variability in climate will either reduce
power production of the Lagdo dam in the Niger Basin by 35% or increase it by 15%
by year 2050. The study also found a possible 15% reduction or 5% increase in the
power generated by the Edea, Song Loulou, Lom Pangar, and Nachtigal power plants in
the Sanaga Basin by 2050 based on a different projection. Minor to moderate impacts
were also recorded for power generation from the other basins.
These impacts from climatic changes on hydropower is magnified by the
year-to-year variability in rainfall, such as that introduced by ENSO, which is known to
influence many extreme events such as hurricanes, floods, and droughts as well as the
hydrology of many regions of the world (Ward et al., 2014). Joly and Voldoire (2009)
report that a significant part of the West African Monsoon’s inter-annual variability can
be explained by ENSO teleconnections.
On a decadal scale, observational studies as recorded by Rodríguez-Fonseca et al.
(2011) indicate that variability of West African Monsoon is related to a global SST
inter-hemispheric pattern, which was partly responsible for the transition between the
wet 1950s and 1960s and the dry 1970s and 1980s. The dry periods of the 1980s were
caused by severe drought conditions over West Africa corresponding to a decrease in
precipitation in the whole monsoon system (Mohino et al., 2011). These findings mean
4 S.A. Boadi and K. Owusu
that, in El Niño years, the dry period separating the two rainy seasons gets longer
through a reduced occurrence of the rainy seasons (Camberlin et al., 2001). Funk et al.
(2005) attributed reductions in the levels of East African lakes to ENSO-events. Hydro-
power generation depends on lake water levels which led Seitz and Nyangena (2009) to
conclude that, these reductions in East African lake levels are likely to disrupt power
generation.
In Ghana, climate variability and change including ENSO has been recorded to
reduce the hydropower generation capacity of the Akosombo Generating Station
(Owusu et al., 2008; Bekoe & Logah, 2013; Waylen & Owusu, 2014). The growth rate
of manufacturing sector in the country has currently slowed down since the first half of
2013 as a result of power rationing which started in September 2012 (Okudzeto et al.,
2014). This clearly suggest that Ghana has been struggling to provide sufficient electric-
ity for its economic sectors (Clark, Davis, Eberhard, Gratwick, & Wamukonya, 2005)
under the current energy policy and strategy which is 52% hydropower driven (www.
vra.com). It is therefore very risky for Ghana to continue the current hydro-dependent
energy policy and strategy. Given the current trend of climate variability and change
impacts on hydropower and the projected increase in temperature and reduction in rain-
fall (EPA & MESTI, 2015), there is a strong possibility that the country is investing in
a power technology that may not be sustainable going into the future.
Data and methodology
Study area
The Volta River Basin which is the area of study is located between Latitudes 5°N and
14°N and Longitude 2°E and 5°W (Figure 1). It has a total surface area of about
414,000 km2 shared between six riparian nations: Benin, Burkina Faso, Cote d’Ivoire,
Ghana, Mali and Togo (Owusu et al., 2008; Kasei, 2009). The Basin can be divided
into a north portion and a south portion. Ghana has about 40% of the total area of the
Volta Basin which covers about 70% of the country’s land area. The Basin in Ghana is
dotted by rivers such as the Oti and Daka, the White and Black Volta, and the Pru, Sene
and Afram rivers (Aquastat Survey, 2005).
The vegetation in the south Volta Basin in Ghana ranges from tropical humid forest
and dry forest in the interior to savannah in the northern part of the country. The climate
is mainly semi-arid to arid. The rainfall regime is divided into two: a dry season and a
rainy season. In the south Basin, rainfall amounts exceed potential evaporation in 6 to
9 months. Temperature in the Basin ranges between 27 °C in the south and 36 °C in the
north (Kasei, 2009). On the whole, the mean annual temperature for the Volta Basin
indicates an increasing trend. The increase in temperature in the south Volta Basin was
about 1 °C between 1991 and 2000.
The Akosombo hydroelectric power project is located in the south Volta Basin in
Ghana. The project is part of the VRA which was established in 1961. The VRA was
tasked to manage the development of the Volta River Basin, which included the con-
struction and supervision of the dam, the power station and the power transmission
work. The VRA is also responsible for the reservoir impounded by the lake. The dam
was built between 1961 and 1965. The primary purpose of the dam was to provide elec-
tricity for the Aluminum industry operated by the Volta Aluminum Company (VALCO).
The Akosombo hydroelectric power project was called the largest single investment in
the economic development plans of Ghana. Its original electrical output was 912 MW,
African Geographical Review 5
Figure 1. A Map of the Volta Basin showing the Synoptic stations used for the study. Source:
GIS Lab, Department of Geography and Resource Development, University of Ghana, Legon.
6 S.A. Boadi and K. Owusu
which was upgraded to 1020 MW in a retrofit project completed in 2006. The construc-
tion of the dam flooded parts of the Volta River Basin, and subsequently creating the
Volta Lake. The lake covers 8502 square kilometers which is about 3.6% of Ghana’s
total land area. The dam’s power plant contains six 170 MW Francis turbines. Each tur-
bine is supplied with water via a 112–116 m long and 7.2 m diameter penstock with a
maximum of 68.8 m of hydraulic head afforded. The minimum operating water level is
73.15 m (240 ft) and the maximum level is 84.73 m [278 ft] (www.vra.com).
Data
The research was carried out using a quantitative approach. Rainfall data for six Synop-
tic stations in the South Volta Basin were collected from the Ghana Meteorological
Agency (GMet). The Synoptic stations are: Bole, Kete-Krachi, Navrongo, Wa, Tamale
and Yendi (Figure 1). All these stations are located upstream of the Akosombo dam as
lake water levels are only influenced by rainfall recorded upstream (Khan & Short,
2001). These six stations were used by Bekoe and Logah (2013) and were also part of
the 28 stations employed by Kasei (2009) in his study of the water resources in the
Volta Basin. The rainfall data collected for the study consist of monthly rainfall totals
for these six Synoptic stations for the period 1970–2010.
Data on the Volta Lake water levels behind the Akosombo dam, monthly lake
inflows, and monthly power generation at the Akosombo Power Station for the years
1970–2010 were obtained from the VRA. Finally, ENSO indices (1970–2010) were
downloaded from the Center for Ocean-Atmosphere Prediction Studies (COAPS) hosted
by the Florida State University. The index used for the ENSO signal is the Japan Mete-
orological Agency (JMA) Index. This index uses a 5-month running mean of spatially
averaged sea surface temperature anomalies (SSTA) over the tropical Pacific (4°S–4°N,
150°W–90°W). Events are classified as either El Niño (Warm Phase), Neutral Phase, or
La Niña (Cold Phase). Years are defined as 6 consecutive month periods including
October–November–December (OND) at or above the +0.5 anomaly for warm (El Niño)
events and at or below the −0.5 anomaly for cold (La Niña) events. All other years
between +0.5 and −0.5 are classified as the Neutral Phase of ENSO. The JMA Index
ENSO year runs from October through to September (http://coaps.fsu.edu/jma).
The main limitation of this study was the limited number of Synoptic stations used in
the rainfall analysis. As can be seen from the map of the study area (Figure 1), the Basin
extends beyond the six stations used in the study. Due to the problem of data availability,
which is a major problem for most climatological studies in Africa, and also the fact that
only rainfall received in the upper catchment of rivers (in this case, upstream of the Ako-
sombo dam) influence the water available in storage reservoirs downstream (Khan &
Short, 2001), data for the study was limited to rainfall inputs from Bole, Kete-Krachi,
Navrongo, Wa, Tamale and Yendi. Even though the Basin extends beyond these six (6)
Synoptic stations, the rainfall drivers are the same for the entire basin and the data quality
was very high in the 6 stations used. Thus, the stations although few, still give a good rep-
resentation of the rainfall for the entire lower Volta Basin.
Methodology
The monthly power generation data from the Akosombo Power Station were aggregated
to derive the annual power generation for the period under study (1970–2010). These
annual power generation data were used to generate a line graph showing the power
African Geographical Review 7
generation pattern for the Akosombo Power Station from 1970 to 2010. Additionally,
the annual variations in power generation were determined by subtracting the total
power generation of a current year from that of the previous year (total power genera-
tion of present year – total power generation of past year).
The ENSO phases were used to group the mean total rainfall for the lower Volta
Basin which had been computed from the monthly rainfall data collected from GMet.
The average rainfall for the long-term period, El Niño years and La Niña years were
calculated. These averages were used to generate a bar plot to compare the mean rainfall
received for the different phases of ENSO.
A regression analysis was conducted in analyzing the power outputs (Saadia, 2008),
to establish the influence of rainfall, ENSO, lake level elevation, and net lake inflow on
annual fluctuations in power generation. Preliminary test was performed which revealed
that most of the assumptions for a regression test; normality of distribution, no signifi-
cant outliers, heteroscedasticity and colinearity were not violated. The rainfall, ENSO,
lake level elevation, net lake inflows and power generation data were divided into two
periods: P1 (1970–1990) and P2 (1991–2010) to correspond to the widely reported fail-
ure of the West African Monsoon in the 1970s and 1980s, and its recovery afterwards
(Giannini, Saravanan, & Chang, 2003; Rodríguez-Fonseca et al., 2011). Significance
levels of 0.05 were used throughout.
Annual fluctuations in power generation from Akosombo Power Station for the two
periods were regressed against rainfall, ENSO, lake level elevation and lake net inflow
data using a stepwise multiple regression to determine if a change in these variables has
significant effect on annual variation in power generation. A multiple regression equa-
tion is described by four (4) parameters: Y, the predicted variable (dependent variable);
a, is the model constant; B, is the slope; and X, represents the different predictive
variables (independent variables).
Y ¼ aþ B1X1 þ B2X2 þ B3X3 þ B4X4 þ e
In this application, Y is power generation, a is the constant (y-intercept), B is the slope,
X1 is the rainfall inputs, X2 is ENSO, X3 is the lake level elevation, X4 is net lake
inflow, and ε is the error.
Results and discussion
The annual power generation derived from the monthly power generation totals at the
Akosombo Hydroelectric Power Station is used for the analysis presented in this sec-
tion. Figure 2 shows the trend of the annual power generation at the Akosombo Power
Station. Annual power generation was generally on the increase from 1970 until it
reached an annual total of 5277 GWh in 1980. After 1980, a decreasing trend was expe-
rienced, with output plummeting to the lowest annual total of 1468 GWh in 1984. Other
low power output years include: 1998, 2003 and 2007. It can be seen from Figure 4
which shows the lake level elevations behind the Akosombo dam that all these low
power output years correspond to very low lake water levels. Figure 2 however shows
that total annual power output has been on the increase over the study period. This
could be attributed to a series of retrofitting exercises which were commissioned in
1989 and ended in 2006 (www.vra.com). The years in which retrofits were completed
are shown in Figure 2.
In order to test for how much of the annual fluctuations in power production is
explained by rainfall, ENSO, lake level elevation and net lake inflow, a stepwise
8 S.A. Boadi and K. Owusu
Figure 2. Annual power generation at Akosombo Power Station from 1970 to 2010. Source:
Based on data from the VRA (2015).
Table 1. Summary results of a stepwise multiple regression analysis for power generation
(Period 1).
Regression Standard error of Standardized
Variables coefficient regression coefficient coefficient (β) t Sig
Mean Basin 2.964 1.312 .460 2.260 .036
Rainfall
Note: R2 = .212, Adjusted R2 = .170; F(1, 19) = 5.106, p = .036.
multiple regression was ran using mean total basin rainfall, ENSO indices, lake level
elevation and net lake inflow as predictor variables. The regression was ran using the
two year categories, P1 and P2. The results of the stepwise multiple regression analysis
for P1 and P2 are shown in Tables 1 and 2.
The overall model for P1 was statistically significant, p = .036, R2 = .212, Adjusted
R2 = .170. This means that the model has explanatory power. As can be seen from
Table 1, the correlation associated with the t distribution of mean basin rainfall is statis-
tically significant (t = 2.260, p = .036). This means that basin rainfall helps predict
annual variation in power generation. The regression coefficients associated with
monthly total basin rainfall is 2.964 which implies that if mean basin rainfall increases
by a unit (1 mm), one will expect power generation to increase by 2.964 GWh. The
other variables (ENSO, lake level elevation and net lake inflow) were omitted from the
P1 model because their contribution to the annual fluctuations in power generation for
the period between 1970 and 1990 were not statistically significant.
The overall model for P2 was statistically significant, p = .000, R2 = .753, Adjusted
R2 = .724. Based on interpretations of beta weights, ENSO (β = −.702) was by far the
best predictor of annual fluctuations in power generation compared to Lake level
elevation (β = .499). As can be seen from Table 2, the correlation associated with the
African Geographical Review 9
Table 2. Summary results of a stepwise multiple regression analysis for power generation
(Period 2).
Regression Standard error of Standardized
Variables coefficient regression coefficient coefficient (β) t Sig
ENSO –1169.642 200.733 –.702 – .000
5.827
Lake level 69.930 16.902 .499 4.137 .001
elevation
Note: R2 = .753, Adjusted R2 = .742; F(2, 17) = 25.947, p = .000.
t distribution of ENSO is statistically significant (t = −5.827, p = .000). This means that
ENSO helps predict annual variation in power generation in P2. Table 2 also shows that
the correlation associated with the t distribution of Lake level elevation is statistically
significant (t = 4.137, p = .001), which also means that Lake level elevation helps pre-
dict annual variation in power generation. The regression coefficients associated with
ENSO, and Lake level elevation are −1169.642 and 69.930, respectively. This means
that if the ENSO indices increase by a unit, holding Lake level elevation constant, one
will expect power generation to decrease by 1169.642 GWh. Also, if Lake level eleva-
tion increases by one unit (1 ft) holding ENSO constant, one will expect power genera-
tion to increase by 69.930 GWh. It can be seen that mean basin rainfall and net lake
inflow were omitted from the regression model because their contribution to the annual
fluctuations in power generation for the period between 1991 and 2010 were not
statistically significant.
A visual inspection of the line graph showing the trend of power generation from
the Akosombo Power Station (Figure 2) indicates lower power generations in 1984,
1998, 2003 and 2007 which is consistent with the findings of Amekor (2007) and
Bekoe and Logah (2013). It is clearly seen from Figure 2 that the total annual power
output of 1468 GWh recorded in 1984 is the lowest generated over the 41-year study
period. This happened immediately after the strong El Niño conditions in 1982/83 and
its resultant severe droughts in 1983 which lowered the Volta Lake water levels in 1984
(Figure 4). This supports the findings of Leemhuis, Jung, Kasei, and Liebe (2009) who
stated that the 1984 power shortages in the country were as a result of low lake level in
the previous years.
The regression analysis examining the effect of mean annual rainfall, ENSO, lake
level elevation and net lake inflow on fluctuations in power generations for P1 suggest
that, rainfall contributed significantly in creating differences in power generation. It can
be observed from Table 1 that the regression model explains 21.2% of the variability
and change in inter-annual fluctuations in power generation. The model shows that, for
every 1 mm increase in rainfall there is a 2.964 GWh increase in power generation
implies that if rainfall reduces, power output will also reduce. This finding confirms the
findings of Gyau-Boakye (2001), and Lautze, Barry, and Youkhana (2006) that mean
annual rainfall influences power generation from the Akosombo Power Station. Figure 3
which illustrates the distribution of mean rainfall for the different phases of ENSO from
1970–2010 was used to show the relationship between rainfall and ENSO. ENSO as
pointed out earlier is the largest determinant of rainfall variability in the tropics as a
whole (Mude et al., 2007). During El Niño years, the average rainfall received in the
basin is about 1040 mm compared to 1209 mm of rainfall received in La Niña years.
The long-term rainfall mean for the lower Volta Basin section in Ghana is 1124 mm.
10 S.A. Boadi and K. Owusu
Figure 3. Distribution of mean rainfall for the different ENSO phases from 1970 to 2010.
Source: Based on data from GMet.
Figure 4. Monthly Volta Lake elevations from 1970 to 2010. Source: Based on data from the
VRA (2015).
This means that the occurrence of El Niño and La Niña in the Eastern Pacific are asso-
ciated with below-normal (low) and above-normal (high) rainfall inputs, respectively,
for the lower Volta Basin which is consistent with the findings of Owusu et al. (2008)
and Joly and Voldoire (2009). The IPCC (2001) found the ENSO phenomenon to be
occurring more frequently since the mid-1970s with the El Niño event becoming more
common, persistent and intense. Low rainfall episodes in the lower Volta Basin with its
attendant influence on hydropower generation in Ghana could therefore be projected to
continue.
African Geographical Review 11
The total power generation from the Akosombo Hydroelectric Power Station has
generally been on the increase after 1991 mainly due to the host of retrofitting exercises
carried out during that period (VRA, 2004). However, Figure 2 shows that inter-annual
variation in power output has become a more common feature during this period. The
regression analysis for the second period, P2, indicate that ENSO and lake level eleva-
tion are the significant predictors of the inter-annual fluctuations in power generated
from the Akosombo Power Station. The model as shown in Table 2, explains 72.4% of
the variability in the inter-annual fluctuations in power generation across the second per-
iod (1991 – 2010). A study of the beta weights shows that ENSO (β = −.702) makes
the highest contribution to fluctuations in annual power generation compared to lake
level elevation (β = .449) for the P2 period. The influence of ENSO on the fluctuation
of power generation was not surprising as the study had earlier established that the El
Niño event is associated with low rainfall inputs in the lower Volta Basin.
Again, it can be seen from Table 2 that, an increase of 1 ft in lake level elevation
across the P2 period, corresponds to 69.93 GWh increase in annual power generation,
holding ENSO constant. This was also expected because the amount of power generated
at a hydropower station is dependent on the flow rate, water level and the overall energy
conversion efficiency of the generating plant (Kaunda, Kimambo, & Nielsen, 2012).
This shows how significant ENSO and lake water levels are to hydroelectric power
generation from the Akosombo Hydroelectric Power Station. It also suggests a strong
relationship between ENSO, lake level elevation and power generation over the 20-year
period (P2).
The regression analysis for the two periods suggests that power generation was sig-
nificantly influenced by rainfall when the rainfall inputs for the whole Volta Basin were
in a low phase. However, when the rainfall pattern shifted during the second period
(P2), fluctuations in power generation were significantly dictated by ENSO and lake
water levels. It is also possible that the low frequency of El Niño and La Niña episodes
in P1 as opposed to the high frequency and magnitude of El Niño events in P2 accounts
for the significant influence. The results from the regression analysis for both periods,
thus, reveal the influence of climate variability and change on fluctuations in power gen-
eration through its influence on mean basin rainfall, and lake level elevation, which sup-
ports the findings of Saadia (2008), Amekor (2007) and Kasei (2009). The IPCC (2013)
found El Niño events to be occurring more frequently and with increased magnitude
since the mid-1970s. Since El Niño is associated with a reduction of 1169.64 GWh in
annual power generation, the study argues that fluctuations in power generation at the
Akosombo Hydroelectric power station and other hydroelectric station in the country
will persist especially in a future where El Niño is predicted to be more frequent and
severe as a result of anthropogenic climate change and resultant warming of the oceans.
These climate variability challenges when added to traditionally inherent hydropower
challenges such as risk from siltation and flooding (Harrison et al., 1998), upstream
water abstraction and transboundary water problems (Lautze et al., 2006), and technical
challenges (Sackey, 2007), will seriously affect the ability of Ghana’s hydropower
infrastructure to supply adequate and sufficient electricity for present and future genera-
tions. This means Ghana’s power output will continue to fluctuate and more likely wor-
sen unless the country changes its power policy direction. The continuous reliance on
hydropower for the provision of adequate and secure power for Ghana is therefore prob-
lematic and probably unsustainable considering the fact that average demand of electric-
ity keeps rising between 10 and 15% per year (Acheampong & Ankrah, 2014).
According to a report by the Ghana National Commission for UNESCO (2010) such a
12 S.A. Boadi and K. Owusu
situation of reducing power supply amidst significant annual increases in electricity
demand introduces a tight demand-supply balance with no reserve margin which leads
to the persistence of periodic load shedding and blackouts.
Previous research (Bukari, 2013; Okudzeto et al., 2014) had shown that the supply
of sufficient power generally boosts productivity and growth. Alternatively, inadequate
power supply stifles sustainable development. For example, a report by Power Systems
Energy Consulting (2010) showed that Ghana’s GDP growth reduces between 2 and 6%
annually as a result of erratic power supply. This means that persistent power shortages
will hinder the ability of the country to achieve sustainable development. There is there-
fore an urgent need for an effective and efficient energy policy with a clear direction on
diversifying away from hydropower if the government wants to attract and sustain
industries as a way of achieving development for the country.
Conclusion and recommendation
This study sought to demonstrate empirically that variability and change in rainfall as
well as the effects of ENSO will create huge problems for Ghana’s power sector unless
the country makes an effort to change the hydro-dependent energy policy it is currently
implementing. The findings of the study support the hypothesis that changes in rainfall
inputs in the lower Volta Basin influence power generation from the Akosombo Hydro-
electric power station. The generation of power from the Akosombo Power Station is
quite complex considering the array of factors which influence power generation in
Ghana (Stanturf et al., 2011). Aside climate variability and change, there are complex
challenges such as transboundary water issues, dam siltation, threats of structural dam-
age to the hydro-infrastructure as a result of flooding, among others challenges, dictating
the amount of power generated within any specific year.
The study found that mean basin rainfall, lake level elevations, and the ENSO phe-
nomena accounted for a greater part of the inter-annual variability in power output from
the Akosombo Hydroelectric power station. It was also revealed that climate variability
and change influenced hydropower generation through its effect on rainfall between
1970 and 1990. However, between 1991 and 2010, climate variability and change influ-
enced power outputs through its effects on ENSO and also through lake water levels
behind the Akosombo dam. The implication of these findings is that Ghana needs to
diversify away from hydropower in order to achieve energy security and sustainable
development. With rainfall predicted to become more variable and El Niño increasing in
both frequency and magnitude, the country must act swiftly to alter the current policy
pathway which weighs heavily on hydropower to other sources if the frequently recur-
ring power shortages and subsequent blackouts are to be resolved.
Based on the above findings, the study makes the following recommendations. The
Ministry of Energy and Petroleum, Ministry of Power and the VRA should consider the
possible diversification of Ghana’s electricity generation away from hydropower in order
to guarantee the supply of adequate and sufficient power for its citizens and meet the
desired target to export power to the West African sub-region. It is clear from the find-
ings that climate variability and change will continue to affect power output from Gha-
na’s hydropower sector and as such a continued reliance on this power source is not
encouraged. The expansion of the thermal sector looks more feasible especially with the
commencement of gas production from the country’s oil fields following the discovery
of large gas reserves.
African Geographical Review 13
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes on contributors
Sylvester Afram Boadi has just completed a Master of Philosophy in Climate Change and Sustain-
able Development at the University of Ghana, Legon. His area of research concentration is climate
variability and change impact on agriculture and energy systems.
Kwadwo Owusu is a senior lecturer and Coordinator of the Climate Change and Sustainable
Development Programme, University of Ghana. His areas of research interest are climatology, cli-
mate variability, and change impacts on agriculture and smallholder adaptation to climate change.
References
Acheampong, T., & Ankrah, F. (2014). Pricing and deregulation of the energy sector in Ghana:
Challenges and prospects (Ghana Energy Situation Report, Q1, 1024). Accra: IMANI Center
for Policy and Education.
Amekor, E. M. K. (2007). Managing the challenges of climate change and impacts of hydrological
change on hydro power production from the water resources of the Volta lake in Ghana – A
case study. In M. Heinonen (Ed.), The Proceedings of the Third International Conference on
Climate and Water (pp. 25–30). Helsinki: Finnish Environment Institute.
Badu, J. (2013). Options for addressing Ghana’s electricity shortfall and case against nuclear
power. Retrieved from http://www.nyakpo.com/wp-content/upload/2013/07/Addressing-
Energy-Shortage-in-Ghana-J-Badu.pdf
Beilfuss, R. (2012). A risky climate for southern African hydro: Addressing hydrological risks and
consequences for Zambezi River Basin Dams. Berkeley, CA: International Rivers.
Bekoe, E. O., & Logah, F. Y. (2013). The impact of droughts and climate change on electricity
generation in Ghana. Environmental Sciences, 1, 13–24.
Bukari, D. (2013). Energy access for the poor in Ghana: Where lays the answer? CEPMLP Annual
Review – CAR, 16, 21.
Camberlin, P., Janicot, S., & Poccard, I. (2001). Seasonality and atmospheric dynamics of the tele-
connection between African rainfall and tropical sea-surface temperature: Atlantic vs. ENSO.
International Journal of Climatology, 21, 973–1005. doi:10.1002/joc.673
Clark, A., Davis, M., Eberhard, A., Gratwick, K., & Wamukonya, N. (2005). Power sector reform
in Africa: assessing the impact on poor people. Washington, DC: Energy Sector Management
Assistance Programme, and World Bank.
Collins, M., An, S.-I., Cai, W., Ganachaud, A., Guilyardi, E., Jin, F.-F., … Wittenberg, A. (2010).
The impact of global warming on the tropical Pacific and El Niño. Nature Geoscience, 3,
391–397. doi:10.1038/ngeo868.
Ebinger, J. O., & Vergara, W. (2011). Climate impacts on energy systems: key issues for energy
sector adaptation. A World Bank Study. Washington, DC: World Bank.
Energy Commission & United Nations Ghana. (2012). Ghana – Sustainable energy for all action
plan. Retrieved from http://energycom.gov.gh/files/SE4ALL-GHANA%20ACTION%20PLAN.
pdf
EPA & MESTI. (2015). Ghana’s third national communication report to the UNFCCCC. Accra:
Environmental Protection Agency and Ministry of Environment, Science, Technology and
Innovation. Retrieved from unfccc.int/resource/docs/natc/ghanc3.pdf
Funk, C., Senay, G., Asfaw, A., Verdin, J., Rowland, J., Korecha, D., … Choularton, R. (2005).
Recent drought tendencies in Ethiopia and Equatorial-Subtropical Eastern Africa. FEWS-NET.
Retrieved from http://pdf.usaid.gov/pdf_docs/Pnadh997.pdf
Ghana National Commission for UNESCO. (2010). Energy, oil and gas: Powering the nation. In
Ghana and UNESCO Magazine (2nd ed., pp. 209–214). Retreived from www.natcomre
port.com/ghana/
Giannini, A., Saravanan, R., & Chang, P. (2003). Oceanic forcing of Sahel Rainfall on interannual
to interdecadal time scales. Science, 302, 1027–1030.
14 S.A. Boadi and K. Owusu
Grijsen, J. (2014). Understanding the impact of climate change on hydropower: The case of
Cameroon. Washington, DC: World Bank. Retrieved from https://openknowledge.worldbank.
org/handle/10986/18243 License: CC BY 3.0 IGO.
Gyau-Boakye, P. (2001). Environmental impacts of the Akosombo Dam and effects of climate
change on the lake levels. Environment, Development and Sustainability, 3, 17–29.
Hamududu, B. H. (2012). Impacts of climate change on water resources and hydropower systems
in central and southern Africa. (Unpublished doctoral dissertation). Norwegian University of
Science and Technology, Trondheim.
Harrison, G. P., Whittington, H. W., & Gundry, S. W. (1998, September 8–10). Climate change
impacts on hydroelectric power, Proceeding 33rd Universities Power Engineering Conference
(UPEC ‘98), Edinburgh, 391–394.
IPCC. (2001). Climate change 2001: Impacts, adaptation, and vulnerability. Contribution of
Working Group II to the Third Assessment Report of the Intergovernmental Panel on Climate
Change. In J. J. McCarthy, O. F. Canziani, N. A. Leary, D. J. Dokken, & K. S. White (Eds.).
Cambridge, MA: Cambridge University Press. 1032 pp.
IPCC. (2007). Climate change 2007: Impacts, adaptation and vulnerability. Contribution of Work-
ing Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change. In M. L. Parry, O. F. Canziani, J. P. Palutikof, P. J. van der Linden, & C. E. Hanson
(Eds.). Cambridge, MA: Cambridge University Press. 976 pp.
IPCC. (2013). Climate change 2013: The physical science basis. Contribution of Working Group I
to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. In T. F.
Stocker, D. Qin, G.-K. Plattner, M. Tignor, S. K. Allen, J. Boschung, A. Nauels, Y. Xia, V.
Bex, & P. M. Midgley (Eds.). Cambridge: Cambridge University Press. 1535 pp. doi:10.1017/
CBO9781107415324
Joly, M., & Voldoire, A. (2009). Influence of ENSO on the West African Monsoon: Temporal
aspects and atmospheric processes. Journal of Climate, 22, 3193–3210. doi:10.1175/
2008JCLI2450.1
Kasei, R. A. (2009). Modelling impacts of climate change on water resources in the Volta Basin,
West Africa. (Unpublished doctoral dissertation). Rheinischen Friedrich-Wilhelms-Universität
Bonn, Bonn.
Kaunda, C. S., Kimambo, C. Z., & Nielsen, T. K. (2012). Potential of small-scale hydropower for elec-
tricity generation in Sub-Saharan Africa. ISRN Renewable Energy, 2012, 15 pp. doi:10.5402/2012/
132606
Khan, S., & Short, L. (2001). Assessing the impact of rainfall variability on watertables in
irrigation areas (CSIRO Technical Report 10/01). Canberra: CSIRO.
Lautze, J., Barry, B., & Youkhana, E. (2006). Changing interfaces in Volta Basin water manage-
ment: Customary, national and transboundary (ZEF Working Paper Series 16). Bonn: Center
for Development Research.
Leemhuis, C., Jung, G., Kasei, R., & Liebe, J. (2009). The Volta Basin water allocation system:
Assessing the impact of small-scale reservoir development on the water resources of the Volta
Basin, West Africa. Advances in Geosciences, 21, 57–62.
Losada, T., Rodriguez-Fonseca, B., Mohino, E., Bader, J., Janicot, S., & Mechoso, C. R. (2012).
Tropical SST and Sahel rainfall: A non-stationary relationship. Geophysical Research Letters,
39(12), 1–7. doi:10.1029/2012GL052423
Mathrani, S., Santley, D., Hosier, R., Bertholet, F., Braud, A., Dawson-Amoah, G., … Reinoso,
G. (2013). Energizing economic growth in Ghana: Making the power and petroleum sectors
rise to the challenge. Washington, DC: World Bank.
McSweeney, C., New, M., & Lizcano, G. (2010). UNDP climate change country profiles: Ghana.
Retrieved from http://www.geog.ox.ac.uk/research/climate/projects/undp-cp/UNDP_reports/
Ghana/Ghana.lowres.report.pdf
Minia, Z. (2008). Climate change scenario development. In W. K. Agyemang-Bonsu (Ed.), Ghana
Climate change impacts, vulnerability and adaptation assessments (pp. 2–13). Accra: Envi-
ronmental Protection Agency.
Ministry of Energy. (2010a). National energy policy. Retreived from http://ghanaoilwatch.org/im
ages/laws/national_energy_policy.pdf
Ministry of Energy. (2010b). Republic of Ghana energy sector strategy and development plan.
Retrieved from http://ghanaoilwatch.org/images/laws/energy_strategy.pdf
African Geographical Review 15
Mohino, E., Rodríguez-Fonseca, B., Mechoso, C. R., Gervois, S., Ruti, P., & Chauvin, F. (2011).
Impacts of the tropical Pacific/Indian Oceans on the seasonal cycle of the West African
Monsoon. Journal of Climate, 24, 3878–3891.
Mude, A., Ouma, R., van de Steeg, J., Kaiuki, J., Opiyo, D., & Tipilda, A. (2007). Kenya adapta-
tion to climate change in the arid lands: Anticipating, adapting to and coping with climate
risks in Kenya – Operational recommendations for KACCAL. Nairobi: ILRI.
Munang, M., & Ayongue, S. N. (2010). Climate change impacts on the hydroelectricity potentials
of the Sanaga Basin: Past and future trends (Unpublished master’s thesis). Buea: University
of Buea.
Okudzeto, E., Mariki, W. A., Paepe, G. D., & Sedegah, K. (2014). Ghana 2014. Paris: AfDB,
OECD, UNDP. Retrieved from http://www.africaneconomicoutlook.org/fileadmin/uploads/aeo/
2014/PDF/CN_Long_EN/Ghana_ENG.pdf
Owusu, K. (2009). Changing rainfall climatology of west Africa: Implications for rainfed agricul-
ture in Ghana and water sharing in the Volta Basin (Unpublished doctoral dissertation).
Florida: University of Florida.
Owusu, K., Waylen, P., & Qiu, Y. (2008). Changing rainfall inputs in the Volta Basin: Implica-
tions for water sharing in Ghana. GeoJournal, 71, 201–210. doi:10.1007/s10708-008-9156-6
Power Systems Energy Consulting (PSEC). (2010). GRIDCo – Ghana wholesale power reliability
assessment 2010. Retrieved from http://www.gridcogh.com/media/photos/forms/relassessment/
2010_GRIDCo_Reliability_Assessment_Report.pdf
Rodríguez-Fonseca, B., Janicot, S., Mohino, E., Losada, T., Bader, J., Caminade, C., … Voldoire,
A. (2011). Interannual and decadal SST-forced responses of the West African monsoon. Atmo-
spheric Science Letters, 12, 67–74. doi:10.1002/asl.308
Saadia, B. (2008). Energy crises in Ghana; A consequence of climate change? (Unpublished
master’s thesis). Nottingham: University of Nottingham.
Sackey, T. (2007). Why the power shortage? In A. Brew-Hammond, & F. Kemausuor (Eds.),
Energy crisis in Ghana: Technology drought or policy? (pp. 6–15). Kumasi: University Press.
Seitz, J., & Nyangena, W. (2009). Economic impact of climate change in the East African com-
munity (EAC) (Final Report). Retrieved from http://www.global21.eu/download/Economic_
Impact_Climate_Change_EAC.pdf
Stanturf, J. A., Warren, M. L., Charnley, S., Jr., Polasky, S. C., Goodrick, S. L., Armah, F., &
Nyako, Y. A. (2011). Ghana climate change vulnerability and adaptation assessment.
Washington, DC: United States Agency for International Development.
Survey, A. (2005). Irrigation in Africa in Figures – Ghana. FAO Water Report, 29, 249–262.
VRA. (2004). Volta River Authority 43rd Annual Report and Accounts 2004. Retrieved from
http://vraghana.com/our_mandate/solar_energy.php
VRA. (2014). Volta River Authority NEWS. Newsletter, 5, October–November–December 2014.
Retrieved from http://www.vraghana.com/resources/newsletters/vra_news_october_decem
ber_2014.pdf
VRA. (2015). Volta River Authority NEWS. Newsletter, 6: April–June 2014. Retrieved from
http://www.vra.com/resources/newsletters/2015/VRA_Apr-Jun_2015.pdf
Ward, P. J., Jongman, B., Kummu, M., Dettinger, M. D., Weiland, F. C. S., & Winsemius, H. C.
(2014). Strong influence of El Niño Southern oscillation on flood risk around the world.
Proceedings of the National Academy of Sciences, 111, 15659–15664.
Waylen, P., & Owusu, K. (2014). Changes in expectations and extremes in the rainfall climatology
of Accra, Ghana, 1895–2005. Applied Geography, 52, 99–109. doi:10.1016/j.apgeog.2014.
04.011
World Bank. (2010). The Zambezi River Basin. A multi-sector investment opportunities analysis.
Washington, DC: The World Bank.
Yamba, F. D., Walimwipi, H., Jain, S., Zhou, P., Cuamba, B., & Mzezewa, C. (2011). Climate
change/variability implications on hydroelectricity generation in the Zambezi River Basin.
Mitigation and Adaptation Strategies for Global Change, 16, 617–628. doi:10.1007/s11027-
011-9283-0