Available online at www.sciencedirect.

com

ScienceDirect
Energy Procedia 59 (2014) 330 335

European Geosciences Union General Assembly 2014, EGU 2014

Trade-offs of water use for hydropower generation and biofuel

production in the Zambezi basin in Mozambique
Philipp Stanzela, *, Harald Klinga, Kit Nicholsonb
a
Poyry, Hydropower and Renewable Energy,Water Resources Division, Laaer-Berg-Strae 43, 1100 Vienna, Austria
b
Hydroc Consult, Siegum 4, 24960 Siegum, Germany

Abstract

In the Zambezi basin in Mozambique, hydropower production is an important economic resource, with substantial development
envisaged for the next decades. Irrigated agriculture currently plays a minor role, but irrigation development has a large potential
and is an important government policy goal, especially aiming at the cultivation of biofuel crops. This contribution assesses
interrelations and trade-offs between these two water-dependent development options. Scenario simulations under different
climate and development assumptions show that adverse impacts of irrigation withdrawal on hydropower are low. Consequently,
the use of water for irrigated agriculture can generate higher economic benefits than the use for hydropower production.

2014
2014TheTheAuthors. Published
Authors. by by
Published Elsevier Ltd.Ltd.
Elsevier This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of the Austrian Academy of Sciences.
Peer-review under responsibility of the Austrian Academy of Sciences
Keywords: Zambezi River basin; hydropower; irrigation development; biofuel production.

1. Introduction

Hydropower is the most important energy source in Mozambique, as in several other southern African countries
[1]. In the Zambezi basin, it is one of the major economic resources, and substantial hydropower development is
envisaged for the next decades [2]. In Mozambique, the extension of the large Cahora Bassa hydropower plant
(HPP) and the construction of several new facilities downstream are planned. Irrigated agriculture currently plays a
minor role, but has a large potential due to available land and water resources. Irrigation development, especially for

* Corresponding author. Tel.: +43-664-828-72-33 ; fax: +43-1-53-605-165.

E-mail address: philipp.stanzel@poyry.com

1876-6102 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of the Austrian Academy of Sciences
doi:10.1016/j.egypro.2014.10.385
 Philipp Stanzel et al. / Energy Procedia 59 (2014) 330 335 331

the production of biofuels, is an important government policy goal in Mozambique [3]. This contribution assesses
interrelations and trade-offs between these two development options with high dependence on water availability for
a future period between 2021 and 2050.

2. Study Area

Focus area is the Zambezi River between Cahora Bassa reservoir and the planned HPP Lupata (Fig. 1). The
extended Cahora Bassa HPP and the planned HPPs of Mphanda Nkuwa, Boroma and Lupata are considered in the
assessment of hydropower generation. Cahora Bassa reservoir, the planned future reservoirs and the catchment areas
contributing to this stretch of the Zambezi are subject of the analyses of irrigation development.

Lake
Chiwa
Luia
Cahora Bassa
Lake Cahora Malawi
Bassa
Mphanda Nkuwa
Revubue
Boroma

10S
Zambezi
Mozambique
Lupata
Luenha

Zimbabwe
20S Hydropower plant
River Shire
200 km N 20 km Country border
20E 30E 40E

Fig. 1. Study area: Lower Zambezi in Mozambique (adapted Zambezi DSS screenshot)

3. Water availability under future climate conditions

River discharge, irrigation water availability and hydropower production in the Lower Zambezi and its tributaries
under two projected future climates are simulated with a hydrological model and a reservoir operation and
hydropower model. The climate scenarios represent a future wetting and a future drying climate and are based on
bias corrected climate model simulations available from the EU WATCH project (by models CNRM-CM3 and
ECHAM5/MPIOM). The projected warming is similar in both scenarios, with +1.7C between the reference period
1961-1990 and the future period 2021-2050 in the CNRM model data and +1.4C in the ECHAM model data.
Fig. 2 shows the changes in mean monthly inflow to Cahora Bassa reservoir between the reference period and the
future period, as simulated by the hydrological model under the assumed climatic changes.

6000 6000
a Mean monthly b Mean monthly
inflow to inflow to
5000 5000
Cahora Bassa Cahora Bassa
4000 4000
Inflow (m/s)
Inflow (m/s)

CNRM 2021-2050 ECHAM 2021-2050

3000 CNRM 1961-1990 3000 ECHAM 1961-1990

2000 2000

1000 1000

0 0
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
Month Month

Fig. 2. Simulated Zambezi inflow to Mozambique (Cahora Basssa) under (a) a wetting climate change scenario (CNRM) and (b) a drying climate
change scenario (ECHAM)
332 Philipp Stanzel et al. / Energy Procedia 59 (2014) 330 335

The hydrological and hydropower models and all input data sets are publicly accessible via the online water
management decision support system Zambezi DSS [4], accessible via www.zdss.ingc.gov.mz or
hydro.poyry.at/zambezi/.

4. Irrigation water demand for biofuel production

4.1. Scenarios of irrigation water demand

Potential water demand for large-scale irrigated agriculture is estimated for a mix of possible biofuel crops in
three scenarios with different irrigated area sizes. The irrigated areas in the Low Biofuel Scenario, the Medium
Biofuel Scenario and the High Biofuel Scenario, as listed in Table 1, are based on projections of the World Bank [2]
for future development in irrigated agriculture in Mozambique. The World Banks estimates for areas currently
under irrigation are considerably lower (around 6,000 ha).
Sweet sorghum and maize (for production of ethanol) and cotton seed and soybean (for biodiesel) are assumed as
suitable biofuel crops [2,3,5]. As these crops can also be used for food or fibre production, and areas with irrigated
vegetables are included, the scenarios can also be regarded as general agriculture development scenarios.
Irrigated areas and crop type are allocated to five points in the study area (see Fig. 3a for the High Biofuel
Scenario), according to information given by the World Bank (2010). The largest areas are allocated to the
surroundings of Cahora Bassa reservoir and Lupata reservoir, and to the Luenha basin, and comparatively small
irrigated areas are assumed for Mphanda Nkuwa reservoir and Boroma reservoir. Fig. 3b presents the resulting
overall mean annual irrigation demand for each crop in the High Biofuel Scenario and shows the high demand of
crops cultivated in winter, coinciding with dry season, and the comparatively low irrigation demand of crops
cultivated in summer, coinciding with wet season. Total annual irrigation demand in the Zambezi basin in
Mozambique rises from below 250 million cubic meters (MCM) per year in the Low Biofuel Scenario to over 2.500
MCM in the High Biofuel Scenario (Table 1).
For the areas around future reservoirs, water is assumed to be withdrawn from the reservoirs; for the Luenha
basin, water abstractions are assumed to taken directly from Luenha River.

Table 1. Total irrigated area and resulting mean annual irrigation water demand in the three biofuel scenarios
Scenario Irrigated area Irrigation demand
(ha) (MCM/y) (m/s)
Low Biofuel Scenario 55,000 238 7.5
Medium Biofuel Scenario 220,000 1,408 44.7
High Biofuel Scenario 385,000 2,579 81.8

a 40 000 b 1 000
Vegetables

Summer / wet season Summer maize

35 000
Annual irrigation demand (MCM)

Winter / dry season

Winter maize
800
Winter cotton

Winter sorghum

30 000
Vegetables (tomatoes)
Irrigated area (ha)

25 000
Winter maize

Soybeans 600
20 000 Beans

15 000 Summer cotton 400

Beans

Summer cotton
Soybeans
Summer maize

10 000 Winter cotton

Sorghum 200
Sorghum

5 000
Winter sorghum
0
0
Cahora Bassa Mphanda Nk. Boroma Luenha Lupata

Fig. 3. (a) Irrigated area for each abstraction point and crop type in the High Biofuel Scenario; (b) resulting mean annual irrigation demand for
each crop (sum of all abstraction points).
 Philipp Stanzel et al. / Energy Procedia 59 (2014) 330 335 333

4.2. Biofuel production

Yield estimates of biomass, biofuel (ethanol, biodiesel) and net energy are based on extensive review of literature
about existing biofuel production under similar conditions as in Mozambique. Resulting annual biomass yields rise
from 140.000 tons in the Low Biofuel Scenario to 1.7 million tons in the High Biofuel Scenario (Fig. 4a). This
increase in biomass yields corresponds to an increase in biofuel yields from 15.000 m to almost 180.000 m (Fig.
4b) and an increase in annual net energy yield from 0.2 million GJ to 1.8 million GJ (Fig. 4 c).
The different climate scenarios have almost no effect on irrigation water availability and resulting biofuel
production. Irrigation water demands can always be met in the new reservoirs on the Zambezi River, and frequently
cannot be met in the Luenha basin, independent of a drying or wetting climate trend. Because of limited water
availability in the Luenha, the theoretically possible yields cannot be achieved, with actual biomass and biofuel
yields between 8% and 11% lower than the potential in all scenarios.
a Ethanol Biodiesel b Ethanol Biodiesel c Ethanol Biodiesel
1 600 150 000 1 400 000
Annual biomass yield (1000 t)

Annual net energy yield (GJ)

Annual biofuel yield (m)

1 400 125 000 1 200 000

1 200 1 000 000
100 000
1 000
800 000
800 75 000
600 000
600 50 000
400 400 000
25 000 200 000
200
0 0 0
High Medium Low High Medium Low High Medium Low

Fig. 4. Mean annual yield of (a) biomass, (b) biofuel, and (c) net energy for ethanol and biodiesel in the three biofuel scenarios

5. Trade-offs

5.1. Water resources and hydropower production

Hydropower simulation results show that increasing irrigation withdrawals lead to decreasing energy production
in the four considered hydropower plants (Fig. 5a). However, due to the generally high water availability in the
Zambezi River, the impact is rather low. In the High Biofuel Scenario, hydropower generation is around 540 GWh/y
lower than in the Low Biofuel Scenario, which is less than 2% of the overall production (which varies considerably
according to the underlying climate assumption).

5.2. Cost and Benefits

The decrease in hydropower generation results in a decrease of gross revenues by 13 to 16 million USD per year
for the Medium Biofuel Scenario, compared to the Low Biofuel Scenario, and by 27 million USD per year for the
High Biofuel Scenario (Fig. 5b). Increased biofuel production leads to an increase in gross revenues of 64 million
USD for the Medium Biofuel Scenario and 128 million USD per year for the High Biofuel Scenario, compared to
the Low Biofuel Scenario (Fig. 5c).
The consideration of related cost emphasizes that the use of water for biofuel production can generate higher
economic benefits than its use for hydropower production. Fig. 6 presents the estimates of cost and value of biofuels
which are based on extensive literature review. There are wide variations in both, the cost depending on crops,
cultivation practice, the locations of production, processing and final sales, and the world oil price and the value of
production depending on the world oil price and on whether the biofuel is being used as a substitute for imported
fuel or as an export to global markets. With world oil prices staying over 80 USD/barrel, the production of biofuels
for oil import substitution will yield strong benefits, with more favourable conditions for ethanol than for biodiesel.
The production of biofuels for export is more challenging and requires highly efficient production.
334 Philipp Stanzel et al. / Energy Procedia 59 (2014) 330 335

a Hydropower generation b Reduction hydropower revenue c Increase biofuel revenue

(compared to Low Biofuel Scenario) (compared to Low Biofuel Scenario)
50000 -140 140
Low Biofuel Scenario 128 128

Revenue (million USD / y)

Revenue (million USD / y)
-120 120
Medium Biofuel Scenario Medium Biofuel Scenario Medium Biofuel Scenario
40000 -100 100
Energy (GWh / y)

High Biofuel Scenario High Biofuel Scenario High Biofuel Scenario

-80 80
30000 64 64
-60 60

20000 -40 40
-27 -27
-16 -13
-20 20
10000 0 0
ECHAM CNRM ECHAM CNRM ECHAM CNRM

Fig. 5.(a) Mean annual hydropower generation in three biofuel scenarios and two climate scenarios; (b) related reduction in hydropower revenue
and (c) increase in biofuel revenue in the Medium and High Biofuel scenarios, as compared to the Low Biofuel Scenario

a b

(USD / l)
(USD / l)

Fig. 6. Costs and value of biofuel production for (a) ethanol and (b) biodiesel in Mozambique

Table 2 summarizes the implications of an increased use of water for the irrigation of biofuel crops, moving from
the Low Biofuel Scenario to the High Biofuel Scenario, and using biofuels for import substitution of petroleum for
transport. For a conservative assumption of net benefits of biofuel production of 0.2 USD/l, the increase in produced
biofuels of 164 million litres would yield benefits of about 33 million USD, which is almost twice as much as the
related cost of 17 million USD of substituting hydropower with power from coal.

Table 2. Implications of moving from Low to High Biofuel Scenario (increase in irrigated area of 330,000 ha)
Benefits from using biofuels Loss from switching from
for transport in Mozambique hydropower to coal
Impact on energy 1.64 GJ/y produced 1.94 GJ/y switched to coal
Units produced 164 million litres biofuels 540 GWh/y switched
Cost of production 0.2 - 1.0 USD/l 0.015 USD/kWh (hydro)
Value of production 0.045 USD/kWh (coal)
in Beira 0.85 - 1.0 USD/l
in Maputo 0.65 - 0.8 USD/l
Economic gain / loss per l or kWh about 0.2 USD/l 0.030 USD/kWh
Total economic gain/loss about 33 million USD gain about 17 million USD loss
per ha (330,000 ha) 100 USD/ha 51 USD/ha
per MCM water (2341 MCM) 1.4 USc/m 0.7 USc/m
 Philipp Stanzel et al. / Energy Procedia 59 (2014) 330 335 335

Generally, investment in irrigated agriculture is expected to have more impact on local economy and therefore
poverty reduction than investment in hydropower. The production of food crops instead of biofuel crops might have
even higher benefits. However, the intensification of agricultural production, whether for biofuel or food production,
requires careful management techniques to avoid the risks of increased erosion, contamination of ground water and
decreased biodiversity. For a comprehensive analysis of benefits and risks of large-scale biofuel development in
Mozambique, these issues need further investigation.

6. Conclusions

This contribution presents an assessment of interrelations and trade-offs of using the water resources in the Lower
Zambezi River in Mozambique for hydropower production and for large-scale irrigated agriculture for the
production of biofuel. The scenario simulations in this study yield the following main findings:

x Water demand for agricultural production with substantial irrigated areas can be met from future reservoirs on the
Zambezi River even under unfavorable conditions of climate change in the assumed drying scenario.
x The impact of irrigation withdrawal on water availability for hydropower is rather low, with a decrease of
hydropower generation of -2% for an increase of over 300.000 ha of irrigated area between the Low and the High
Biofuel scenarios.
x The use of water for cultivation of biofuels can generate higher economic benefits than the use for hydropower
production, especially when producing for import substitution, as long as biofuel production is efficient and
world oil prices stay high.
x More research on benefits of irrigated agriculture for food production and on environmental impacts of
large-scale irrigation development is needed.

Acknowledgements

This research was carried out for the National Institute of Disaster Management (INGC), Mozambique. Input data
for hydropower modelling is based on data by the Global Precipitation Climatology Centre in Germany, the Climate
Research Unit in the UK and on data sets from the EU WATCH project, generated by the Max-Planck-Institut fr
Meteorologie (MPI-M) and the Centre National de Recherche Mtorologique (CNRM).

References

[1] Southern Africa Power Pool. Annual report 2012. Harare: Southern Africa Power Pool; 2012.
[2] The World Bank. The Zambezi River Basin: A multi-sector investment opportunities analysis. Washington: The World Bank; 2010.
[3] Schut M, Slingerland M, Locke A. Biofuel developments in Mozambique. Update and analysis of policy, potential and reality. Energy Policy
2010;38:5151-5165.
[4] Kling H, Stanzel P, Preishuber M. Impact modelling of water resources development and climate scenarios on Zambezi River discharge.
J Hydrol: Reg Stud 2014;1:17-43.
[5] Yamba FD. Challenges and opportunities for biofuels production, marketing, economics and policy implications in Southern Africa. In:
UNIDO. Industrial biotechnology and biomass utilisation. Prospects and challenges for the developing world. Vienna, Austria: UNIDO;
2007. p. 81-96.