Hydropower and climate change, insights from the integrated water-energy modelling of the Drin Basin

Abstract

The understanding of the transboundary impact of Climate Change on hydropower is not well-established in the literature, where few studies take a system perspective to understand the relative roles of different technological solutions for coordinated water and energy management. This study contributes to addressing this gap by introducing an open-source, long-term, technologically-detailed water and energy resources cost-minimisation model for the Drin River Basin, built in OSeMOSYS. The analysis shows that climate change results in a 15-52% annual decline in hydro generation from the basin by mid-century. Albania needs to triple its investments in solar and wind to mitigate the risk of climate change. Changing the operational rules of hydropower plants has a minor impact on the electricity supply. However, it can spare significant storage volume for flood control.

Full text

Energy Strategy Reviews 48 (2023) 101098
Available online 25 May 2023
2211-467X/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Hydropower and climate change, insights from the integrated water-energy
modelling of the Drin Basin
Youssef Almulla
a
,
*
, Klodian Zaimi
b
, Emir Fejzi´
c
a
, Vignesh Sridharan
c
, Lucia de Strasser
d
,
Francesco Gardumi
a
a
Department of Energy Technology, KTH, Royal Institute of Technology, Brinellv¨
agen 68, 10044, Stockholm, Sweden
b
Polytechnic University of Tirana (UPT), Bulevardi D¨
eshmor¨
et e Kombit Nr. 4, Tiran¨
e, Albania
c
Chemical Engineering Department, Imperial College London, SW7 2AZ, London, UK
d
The United Nations Economic Commission for Europe (UNECE), Bureau S411, Palais des Nations, 1211, Geneva, 10, Switzerland
ARTICLE INFO
Handling Editor: Mark Howells
Keywords:
Hydropower
Climate Change
Water-Energy-Nexus
Transboundary water
Modelling
ABSTRACT
The understanding of the transboundary impact of Climate Change on hydropower is not well-established in the
literature, where few studies take a system perspective to understand the relative roles of different technological
solutions for coordinated water and energy management. This study contributes to addressing this gap by
introducing an open-source, long-term, technologically-detailed water and energy resources cost-minimisation
model for the Drin River Basin, built in OSeMOSYS.
The analysis shows that climate change results in a 15–52% annual decline in hydro generation from the basin
by mid-century. Albania needs to triple its investments in solar and wind to mitigate the risk of climate change.
Changing the operational rules of hydropower plants has a minor impact on the electricity supply. However, it
can spare signicant storage volume for ood control.
1. Introduction
Hydropower is currently the largest source of renewable energy
generation worldwide. The total installed capacity reached 1330 GW in
2020 [1] representing 15.6% of the global electricity generation [2], or
about 60% of all renewable generation globally. The International En-
ergy Agency (IEA) underlines the importance of hydropower in its ‘Net
Zero by 2050’ report, suggesting that the world will need to double the
hydropower capacity by mid-century to keep the global temperature
increase below 1.5◦Celsius over pre-industrial times [3]. This means
that the same capacity that was installed in the last 100 years would
need to be built in the next 30 years [1].
Hydropower is to a great extent intertwined with climate change. On
the one hand, hydropower contributes signicantly to the mitigation of
greenhouse gas (GHG) emissions that cause global warming. On the
other hand, climate change alters river discharge and impacts
hydropower generation [4]. Therefore hydropower output is largely
uncertain and depends on rainfall and temperature patterns [5]. The
Sixth Assessment Report (AR6) from the Intergovernmental Panel on
Climate Change (IPCC) estimates that global hydropower production
declined by ~4–5% compared to long-term average production since the
1980s due to droughts. Changes can be far more severe according to the
climate change scenarios and according to the region. In parts of Europe
and the Mediterranean, hydropower generation could be reduced by up
to 40% under a temperature increase of 3 ◦C. Over 20 million people
have been internally displaced annually by climate extreme events (e.g.
wildres, droughts, and oods) since 2008, with storms and oods being
the most common drivers. Flood risks and societal damages are pro-
jected to increase with every increment of global warming, causing
signicant damage to infrastructure, food security, electricity genera-
tion and other societal domains [6].
Abbreviations: AR6, The Sixth Assessment Report; CC, Climatic Change; CIIs, Climate Impact Indicators; DRB, Drin River Basin; ECVs, Essential Climate Variables;
E-Hype, Europe-Hydrological Predictions for the Environment; GCMs, Global Circulation Models; GHG, Greenhouse gas; GWh, Gigawatt hour; HPP, Hydropower
Plant; IEA, International Energy Agency; IPCC, Intergovernmental Panel on Climate Change; MCM, Million Cubic Meter; MW, Mega Watt; OSeMOSYS, Open Source
Energy Modelling System; RCMs, Regional Climate Models; RCPs, Representative Concentration Pathways; RQs, Research Questions; SMHI, Swedish Meteorological
and Hydrological Institute; TWh, Terawatt hour; WEAP, Water Evaluation and Planning system.
* Corresponding author.
E-mail address: almulla@kth.se (Y. Almulla).
Contents lists available at ScienceDirect
Energy Strategy Reviews
journal homepage: www.elsevier.com/locate/esr
https://doi.org/10.1016/j.esr.2023.101098
Received 14 November 2022; Received in revised form 5 April 2023; Accepted 1 May 2023

Energy Strategy Reviews 48 (2023) 101098
2
1.1. Literature review and research gaps
Shared basins of rivers and lakes crossing international borders,
usually referred to as ‘transboundary basins’, make up about half the
Earth’s surface area and 40% of the world’s population lives in prox-
imity to such basins [7]. It is estimated that more than 70% of the new
hydropower projects have transboundary dimensions [8]. The man-
agement and use of these resources call for collaboration between
different sectors and different states, especially in cases of resource
scarcity or oods [9]. Despite that, there is a gap in the literature in
understanding the operation and cross-border usage of transboundary
dams [8,10], especially under extreme climate conditions. Llamosas
et al. conducted a systematic review of over 1200 peer-reviewed articles
on transboundary hydropower dams from 2009 to 2019. The study
concluded that most studies on transboundary hydropower dams focus
on water management and water allocation and focus less on hydro-
power benets and temporal and spatial variations [10]. In another
study, Llamosas and Sovacool explore the impacts of transboundary
dams in three major hydropower states (Laos, Paraguay, and Bhutan).
The study integrates elements from the energy security and justice
frameworks with the concept of technological capabilities to explore the
distribution of energy benets between riparian countries in trans-
boundary basins [8]. Almulla et al. explore solutions to motivate
transboundary hydropower cooperation in the Drina River Basin in
South-East Europe where the level of cooperation among the riparian
countries is low. A multi-country energy model with a simplied hy-
drological system was developed to represent the cascade of hydro-
power plants (HPPs) in the Drina basin and explore, among others,
cooperation vs. no cooperation scenarios [11]. Gonzalez et al. assess the
benets of cooperation in the management of new dams in trans-
boundary water resource systems that do not have formal sharing ar-
rangements. A multi-criteria comparison of uncooperative historical
reservoir operations vs. adopting new cooperative rules was used to
estimate the benets of cooperation in the Volta transboundary river
basin in West Africa [12]. Similarly, Basheer et al. use an integrated
hydrological and macroeconomic modelling framework to study the
Nile river basin [13]. All the aforementioned studies explore different
aspects of transboundary hydropower but lack the consideration of
climate change impact. In other words there is a clear gap in the liter-
ature in understanding the country-wide or region-wide impacts of
climate change in transboundary water basins. This gap was addressed
partially by Skoulikaris et al. [14] by exploring the impacts of climate
change on a transboundary river basin shared between Bulgaria and
Greece. The Water Evaluation and Planning (WEAP) is used to simulate
two large HPPs under various climate scenarios and for short-term
(2021–2050) and long-term (2071–2100) future periods. However,
this study limits the consideration of the energy system to two HPPs in
the basin. Spalding-Fecher et al. [15] explored a range of climate change
and socioeconomic scenarios for the Zambesi River Basin. More specif-
ically, it studies the impact of climate change on individual HPPs as well
as on the entire electricity generation systems of riparian countries. The
study soft-links two modelling frameworks namely: The Long-range
Energy Alternatives Planning (LEAP) and The Water Evaluation and
Planning (WEAP). Although this study addresses the gap at its core, it
can be argued that this is one early attempt and more investigation on a
wider spectrum of basins and climates is needed. Additionally, the
modelling frameworks (WEAP and LEAP) are not open-source, which
challenges replicability in any future work.
Hence, the aim of this study is to explore the impact of climate
change and oods on an electricity supply system highly reliant on hy-
dropower and vulnerable to the risk of oods. The Drin River Basin
(DRB) is chosen to conduct this study due to its characteristics as shown
in sub-section 1.2 and due to ongoing work on transboundary water-
energy cooperation under the framework of the UN Water Convention
[16] and the Memorandum of Understanding (Drin MoU) for the sus-
tainable management of the Drin Basin [17]. An open-source modelling
framework is used for this purpose to ensure transparency and motivate
reproducible research [18].
1.2. Case application: the Drin River basin (DRB)
The DRB in Southeast Europe is shared between Albania, North
Macedonia, Greece, Montenegro and Kosovo.
1
The basin covers an area
of 14,173 km
2
[19] with a length of 285 km [20]. The river originates
from Ohrid Lake and Prespa lakes in North Macedonia where it is called
the Black Drin. The other stream, the White Drin, originates in Kosovo
and converges with the Black Drin to form the Drin that then ows into
Albania and discharges to the Adriatic Sea (Fig. 1). The Drin Basin
comprises the sub-basins of the Black Drin, White Drin, Drin and
Buna/Bojana rivers, of the Prespa, Ohrid and Skadar/Shkod¨
er lakes, the
underlying aquifers, and the adjacent coastal and marine area [17].
The Drin Basin is naturally prone to a high risk of oods. The fre-
quency and intensity of these oods are increasing over time, likely due
to climatic changes and ow regulation practices [21]. The construction
of hydropower infrastructure along the mainstream of the Drin River
and the ow regulation practices aggravate the risk. For example, the
2010 oods were exacerbated by the operation of the three hydropower
plants in Albania which were forced to release water from the dams
inundating over 10,000 ha of land and affecting over 2000 houses
downstream in the district of Shkod¨
er in Albania [17].
The ow regime in the Drin River is altered by the operation of two
cascades of HPPs [19]. The rst cascade is on the Black Drin (in North
Macedonia) and consists of two large dams and HPPs (Globocica and
Spilje). The second cascade is along the Drin River and encompasses
(currently) three large dams and HPPs (Fierza, Koman, and Vau i Dej¨
es)
which serve the Albanian grid. Additionally, a new hydropower project
(Skavica HPP) is under development upstream of this cascade [21]. The
operational rules for the HPPs in the basin were dened in the 1980s and
have not undergone any revisions since then [22]. The ow in the Drin
River uctuates over the year and between the years. The water level
can be particularly high in the months of highest rainfall, from
December to February. The frequent oods in the lowlands
2
of the Drin
basin are a pressing problem for the region [17]. Among other factors,
the oods in the lowlands are determined by the ow regime of the Drin
River, especially since other rivers that are discharging into the lowlands
have no ood control [23].
The basin suffered from unsustainable management approaches and
conicting priorities between upstream and downstream countries for
several years [23]. Currently, there is relatively good cooperation at the
transboundary level for cascade dam operations, but it is restricted to
emergency situations in Albania and North Macedonia [17]. At the na-
tional level, hydrometeorological monitoring systems in Albania and
North Macedonia are not integrated with the dam operators. The oper-
ators do not feed the ood forecasting system with information. This
results in sub-optimal ood forecasting and possibly sub-optimal dam
operations [17]. Therefore, understanding the basin-wide impact of the
dam operation is crucial to mitigating the risk of oods in the basin.
From the electricity system perspective, the total installed capacity in
the countries sharing the DRB is 6461 MW. The power system is domi-
nated by hydropower with over 50% of the installed capacity. Then
comes thermal power with 43%, while the wind makes only 3% of the
capacity [24–27] as shown in Fig. 2 below, Figure C 1, Figure C 2 and
Figure C 3 in Appendix C. The Drin basin contributes 2015 MW of hy-
dropower which makes up 31% of the total installed capacity in the
riparians. This reects the importance of the basin for the energy sector
in the region.
The operation of the dams has mostly been driven by the objective of
1
UN administered territory under UN Security Council resolution 1244 [23].
2
The lowlands of the Drin Basin are the areas downstream of the Vau i Dej¨
es
dam and more specically surrounding Lake Skadar/Shkod¨
er [17].
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
3
Fig. 1. Extention of the Drin River Basin showing the main tributaries, main reservioirs and hydropower plants. (Elaboration by the author based on original map
from Drin Corda [17]).
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
4
maximising electricity production. Usually, hydropower reservoirs
require water levels to be kept at a maximum design level to store as
much energy as possible for daily hydropower generation [28]. This is a
typical prot-driven mindset that is noticed in other basins as well. For
instance in the Evros/Marita river basin shared between Bulgaria,
Greece and Turkey, the operators of the dams pefer to ll the reservoirs
with water to maximise electricity generation even if ood events are
projected in the area [29,30].
Operation criteria that take into account the need for containing
oods may signicantly reduce oods costs. However, there are con-
cerns by operators that this practice may reduce the gains from elec-
tricity generation to a non-negligible extent. To avoid ooding
downstream, dams and reservoirs can be effectively used to regulate
river levels by temporarily storing the ood volume and releasing it
later. Additionally, the implementation of the Skavica hydropower
project could represent an opportunity from the point of view of
increasing the electricity supply and at the same time adding extra
storage capacity to improve ood control in the basin as will be shown
later.
Therefore and in order to comprehensively and objectively assess the
concerns of the hydropower operators and to provide scientic ground
for dialogues among different stakeholders, we nd a strong need for
quantitative modelling of the hydropower cascade in the Drin Basin. In
addition, since the hydropower cascade is part of larger systems (the
hydrological system, the electricity generation systems of the riparian’s
and the climate systems that affect them), we suggest the use of inte-
grated systems modelling approaches. Furthermore, the hydrological
system being affected and constrained by the climatic system, we decide
to combine a simulation modelling approach (for the water ows) with
an economic optimisation modelling approach (for the power infra-
structure system). Methods will be detailed in section 2.
With the chosen approach, we inform stakeholders on the above is-
sues with a cost-benet analysis. It addresses specically the following
research questions (RQs):
1. What is the impact of climate change on hydropower generation in
the Drin Basin?
2. Which technologies should be invested in to mitigate the risk of
climate change?
3. What is the impact of keeping larger ood buffers in reservoirs on
electricity generation in each HPP and each country?
4. What benets would the new HPP (Skavica) bring to the energy
system?
This paper adds a number of contributions to the established
literature:
First, it shows the rst application of a long-term electricity supply
model for the Drin Basin with national detail, able to investigate the
impact of climate change at the basin and the national levels. Previous
studies focused on the water system [19,20,23,27,31] and some aspects
of energy were considered [32]. However, these were limited to a few
HPPs in the basin and isolated from the rest of the energy system.
Therefore in this model we include representation of entire electricity
system of the riparian countries with the focus on detailed representa-
tion of the hydrocascade.
Second, this paper introduces an enhanced application of a long-
term energy system optimisation tool (compare to the previous study
on the Drina basin [11]) to explore the impact of changing the opera-
tional rules of HPPs on electricity generation. In this case, the aim is to
have better ood control in place. Therefore we create a detailed rep-
resentation of the dam and its operational rules as will be shown in
section 2.1, but the method can serve other purposes as well like max-
imising electricity generation or studying environmental ow re-
quirements, to name a few.
Third, while this study showcases the Drin basin, the methodology
and insights are valuable and replicable for long-term energy planning
in other transboundary river basins. The use of an open-source model-
ling framework and its availability on GitHub [33] promotes trans-
parency and reproducibility which enables researchers to implement the
methodology in other case studies.
2. Methodology
The analysis is carried out by developing a water-energy model for
the Drin basin using the Open Source Energy Modelling System (OSe-
MOSYS) [34]. The open-source nature with the source code available on
a Github reporsitory [35], the exible structure and the holistic
consideration of the nation-wide and regional electricity system make
OSeMOSYS the right t for this study [34,36]. The model can be cat-
egorised as a bottom-up (technology-rich) and dynamic (representing
several years in a time domain of decades and intra-annual steps)
cost-minimisation model. It calculates the whole electricity system
conguration (installed capacity and operation of electricity supply
technologies in all the riparians
3
) that meets electricity demands in each
of the riparians at the least Net Present Cost while complying with
several constraints. The choice of constraints characterises the paper.
These include rst and foremost water availability along the Drin Basin
as predicted in different climate change scenarios, operational rules of
hydro power plants and dams, existing and planned infrastructure,
availability and cost of other resources (renewable and not) for elec-
tricity supply, and policy constraints. The time domain of the study is
between 2020 and 2050 and each year is split into 52-time steps rep-
resenting each week in a year. This resolution is used to ensure an
adequate representation of the water variability within a manageable
computational time. It is important to understand that this is a
techno-economic model, where both the water and energy systems are
represented in an engineering way, as commodity demand-supply
chains divided into series of processes/pieces of infrastructure and
commodities owing between them (respecting 1st Law balances of
mass and energy).
The model framework builds on a previous effort [11] that represents
a ‘hydrological system’ and links it to the ‘electricity system’ within
OSeMOSYS. In this study, we take this approach one step further. We
enhance the representation of the hydrological system and water storage
to model the changes in water availability imposed by either external
factors (i.e. climate change), or internal factors (i.e. change in HPPs’
operational rules). So the water-energy model of the Drin River Basin
Fig. 2. Total installed capacity by share (%) of different electricity generation
technologies in the Drin countries.
3
Excluding Greece since it covers only 1% of the basin area and has not
energy infrastructure in the basin.
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
5
encompasses two integrated systems: the hydrological system and the
electricity system, the boundaries of the represented systems are shown
in Fig. 3. Within the rst system, we detail the cascades of HPPs in the
DRB (Fig. 4) and introduce the reservoir operational rules. The second
system covers the rest of the electricity infrastructure of the four riparian
countries. The following subsections give an overview of the different
elements of the model. A detailed description of the whole linear pro-
gram can be found in the OSeMOSYS publication [34], the GitHub re-
pository [35] and OSeMOSYS documentation [37].
2.1. Representation of the hydrological system
The hydrological system in OSeMOSYS represents the Drin river with
its two main tributaries the White Drin and the Black Drin. The two
cascades of the ve large hydropower reservoirs on the Black Drin and
the Drin rivers are detailed. Additionally, a provision is made for the
planned Skavica HPP in Albania (between Spilje and Fierza) for one of
the analysed scenarios. The volume and the ow in each river segment
are determined based on the historical data on river discharge obtained
from the Europe-Hydrological Predictions for the Environment (E-
HYPE),
4
a hydrological model from the Swedish Meteorological and
Hydrological Institute (SMHI) [38]. Daily time series data is extracted
for the period 1981–2010 and processed to represent the average water
discharge in each river segment on weekly temporal resolution.
Fig. 5 illustrates the modelling concept, representing any of the HPPs
and related dams, water inputs and water outputs. As mentioned, since
OSeMOSYS is a techno-economic modelling tool working with mass and
energy balances, also the water system is represented in a techno-
economic fashion. The river segments upstream and downstream of a
power plant are represented in an aggregated way, as a generic ‘tech-
nology’ (water source) providing or receiving a certain water volume
ow. The river segment upstream feeds water to a dam. The water
available in the dam can be fed to the hydropower plant when this needs
to generate power (depending on user-dened electricity demands and
load proles), be stored when the dam is not full, or be released through
a spillway. The capacity of the dam and the spillway are user inputs
(Table 1), dened using data provided by the local stakeholders and
utilities or found in public sources. A minimum level of operation of the
spillway can be dened, to ensure compliance with regulations on
environmental ows. The dam can be discharged only down to the
minimum storage level dened by the operators and it can be lled up
only to the maximum level allowed by the buffer volume used for ood
containment.
The water balance is being respected in all parts of the system. This
implies that the water inows from different streams of the river and
catchments are aggregated in two steps, rst in the river segment rep-
resented before any of the reservoirs and second in the river segment
after the HPP.
2.2. Representation of the HPPs’ operational rules in OSeMOSYS
The operation of the hydropower plants is managed by a set of rules
that dictate the quantities of water to be stored and to be released in
various conditions in order to meet the basic (designed) power output
[39]. Hydropower generation is largely dependent on water inow,
which means that electricity generation (or at least its share in different
seasons) can be different in wet and dry years. This underlines the
importance of revising the reservoir operation rules to mitigate the risk
of climate change [39–41].
Different mathematical modelling methods and tools are used to
analyse reservoir operation policies. The plethora of methods and tools
include simulation models such as the standard operating policy (SOP)
also known as the S-shaped curve of operation; optimisation models
such as Stochastic dynamic programming (SDP) and the linear decision
rule (LDR) [33]; Multi-objective Optimisation model; Dynamic
Real-Time Reservoir Operation Rules, to name a few. The reservoir
operation rules can be based on long-term historical series of inow or
real-time water inow [40].
In this study, the existing operational rules of the reservoirs in the
Fig. 3. Schematic overview of the Drin water-energy model.
4
The HYdrological Predictions for the Environment (HYPE) distributed hy-
drological model when applied across Europe, is called (E-HYPE). The model is
forced by daily precipitation and temperature and then calculates ow paths in
the soil based on several parameters such as: snow melt, evapotranspiration,
surface runoff, inltration, to name a few [38].
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
6
Fig. 4. Structure of the hydropower cascade in the Drin River Basin, as represented in OSeMOSYS.
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
7
Drin basin, obtained from local sources [22], are simulated using OSe-
MOSYS. This includes dening, for each reservoir and each week, the
total storage capacity, the minimum storage level, the maximum storage
level and the water inow. The latter is based on historical time series
data from E-HYPE. Since this is not a standard application of OSe-
MOSYS, some changes are introduced to the storage parameters and
equations. We use the version of OSeMOSYS code with updated storage
equations developed by Neit [42] and by Kuling and Neit [43] and we
change the denition of the following parameters to be able to capture
the variation in each time slice. This required introducing a new
dimension representing time slices in the denition of each parameter:
•TechnologyToStorage{r in REGION, t in TECHNOLOGY, s in STOR-
AGE, l in TIMESLICE, m in MODE_OF_OPERATION};
•TechnologyFromStorage{r in REGION, t in TECHNOLOGY, s in
STORAGE, l in TIMESLICE, m in MODE_OF_OPERATION};
•StorageMaxCapacity{r in REGION, s in STORAGE, l in TIMESLICE, y
in YEAR};
•ResidualStorageCapacity{r in REGION, s in STORAGE, l in TIME-
SLICE, y in YEAR};
Also, the respective equations are adjusted to be consistent with this
change in the parameters (as illustrated in:Appendix A.
•s.t. S1_StorageLevelYearStart{r in REGION, s in STORAGE, y in
YEAR},
•s.t. S2_StorageLevelTSStart{r in REGION, s in STORAGE, l in TIME-
SLICE, y in YEAR}
•s.t. SC8_StorageRelling{s in STORAGE, r in REGION}.
•s.t. SC1_LowerLimit{r in REGION, s in STORAGE, l in TIMESLICE, y
in YEAR}:
•s.t. SC2_UpperLimit{r in REGION, s in STORAGE, l in TIMESLICE, y
in YEAR}:
•s.t. SC7_StorageMaxUpperLimit{s in STORAGE,l in TIMESLICE, y in
YEAR, r in REGION}:
Additionally, the variation in water levels is reected in the
parameter (TechnologyFromStorage). This denes which technology (in
this case HPP) receives the water from the reservoir and at which vol-
ume to generate one unit of electricity in each time slice. In other words,
it denes how much water passes through the penstock to the turbine
each week to produce one unit of electricity. This value (we call it Rate)
is calculated externally using the following correlation:
Rate =Q/E(1)
where: Rate =the ratio of the water volume required to produce one
unit of electricity in (m
3
/GWh), Q =the volume of water (m
3
) and E =
Fig. 5. Detail of the hydropower cascade model.
Table 1
Characteristics of the large dams and hydropower plants along the Drin River.
# Plant Reservoir Storage Volume
(MCM)
a
Power Capacity
(MW)
Started
Operation
Net Head
(m)
Water Inow to turbines
(m3/sec)
Avg. output
(GWh)
Spillway capacity
(m3/sec)
1 Globocica 55.3 42 1965 97.5 2 X 25 180 1100
2 Spilje 506 84 1969 91.3 3 X 36 288 2200
3 Skavica
b
2300 196 2025 about 140 2 X 87 NA 2800
4 Fierza 2350 500 1976 118 4 X 124 1568 2670
5 Koman 188 600 1985 96 4 X 150 1955 3400
6 Vau i Dej¨
es 310 250 1970 52 5 X 113 990 6700
Total Drin River
Basin
5709 1672 4570 18,870
a
MCM: Million cubic meters.
b
Skavica hydropower plant is introduced as a new capacity from 2025.
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
8
the electricity generated (GWh). Energy and water ows are calculated
using:
E=P t; (2)
P=
ρ
q g h (3)
where: P =power (GW), t =time (h),
ρ
=density (kg/m3) (~1000 kg/
m3 for water), q =water ow (m
3
/s), g =acceleration of gravity (9.81
m/s
2
) and h =head (m) or “the difference in elevation between the
headwater surface above and the tailwater surface below a hydroelectric
power plant under specied conditions” [44]. The operational rules
dictate the minimum and maximum ranges of water levels and the river
ow regulates how much water is available in each time slice. The
modelled electricity generation from the cascade of HPPs in the Drin
Basin is validated using historical data from the power utility in each
country as shown in Figure C 9.
The updated version of the code is one of the contributions of this
study. It enhances the representation of hydropower plants and their
operational rules which allows exploring the impact of new rules as
shown in section 2.4. This, to the best of our knowledge, is done for the
rst time in OSeMOSYS. The updated version of the code is open and
available for researchers to use in the GitHub repository [33].
2.3. Representation of the electricity system
The multi-regional OSeMOSYS model of the Drin basin encompasses
the electricity supply systems of Albania, North Macedonia, Montenegro
and Kosovo. The entire electricity system of each of the riparians is
modelled as its own entity and linked to the system of the other riparians
via transmission links. The reason is that the model aims to represent the
least-cost ways for the electricity supply system of riparian countries to
develop and meet future demands, as well as to assess the impacts of the
change in water availability on the electricity system.
The current and potential electricity supply technologies in each
country are represented. These include HPPs (inside and outside the
basin), thermal plants, non-hydro renewables (solar and wind) as well as
electricity trade interconnectors (between riparians and with countries
outside of the basin). The model calculates the least-cost electricity
supply mix (in terms of operation and new investments along the time
domain of the study) that meets given electricity demand projections
while facing constraints in natural resources. The above technologies
compete in gaining shares of electricity generation, based on their cost
and the availability of primary resources. Existing and new hydropower
is mainly constrained by the availability of water dictated by climatic
changes. New hydropower is also constrained by limited room for new
large infrastructure projects. Other renewables are mainly constrained
by variability in resource availability (expressed through maximum
capacity factors dened by time step). Fossil fuel power plants in-
vestments are mainly constrained by fuel prices and policies. Each group
of supply technologies is represented in an aggregated form, except for
the ve HPPs in the Drin basin which are modelled individually, since
one objective of this study is to explore the role of the HPPs in mitigating
ood risk in the Drin basin.
In summary, with this model setup, we are simulating the ow in
different parts of the river segments to be able to better represent water
availability for the HPPs in the Drin basin. We are using this as an input
to the model to optimise the intra-annual operation of the HPPs, the
generation from other technologies and the long-term investments to
minimize the overall system cost. This model structure distinguishes
itself from other hydrological modelling tools (i.e. WEAP [45]) that have
more detailed hydrological representation but take a simulation
approach and lacks the energy perspective.
2.4. Scenarios
We use the model described above to investigate four categories of
scenarios and sub-scenarios as shown in Fig. 6. All scenarios were co-
created with stakeholders through workshops and consultation meet-
ings. Each scenario aims at representing certain dynamics as follows.
2.4.1. Reference (REF)
This scenario represents the current situation of the electricity sys-
tem as well as the committed projects
5
in the Drin riparian countries,
with particular detail for the ve existing HPPs in the Drin Basin
(excluding the upcoming Skavica HPP). It assumes that the hydrological
conditions affecting water availability in the basin are similar to the
conditions that have been observed in the recent historical record
(1981–2010). This scenario establishes a baseline approximating pre-
sent conditions that are used to estimate the impact of changes expected
in the future.
2.4.2. Climate change (CC)
Climate change is expected to affect the region through changing
temperatures, precipitations and water availability. This scenario ad-
dresses RQs 1 and 2. It is based on hydrological data from SMHI which
provides inputs for about 40 climate impact indicators (CIIs) and 4 time-
series of essential climate variables (ECVs) [46]. The indicators are
based on the outputs of different modelling chains. The climate tem-
perature and precipitation indicators are based on the outputs of
Regional Climate Models (RCMs) driven by Global Circulation Models
(GCMs). The water quantity indicators are based on the outputs of three
hydrological models driven by the bias-corrected output of regional
climate models [38,46].
Multiple climates and multiple hydrological models are used to
address uncertainty in climate change impacts and to better dene
robust signals of future change. Emission scenarios were modelled using
different Representative Concentration Pathways (RCPs) [46]. In this
study, we use the outputs of the hydrological model (E-HYPE v3.1.2) and
generate an ensemble (from different GCMs and RCMs) for each RCP
(2.6, 4.5 and 8.5) as shown in Table C 3. Three sub-scenarios are
developed to represent the changes in the river discharge in the Drin
basin under each of the three RCPs for the period 2020–2050 and
explore the impact on the energy system. More specically, the key
variable here is the river discharge at various points of the cascade. This
is introduced to OSeMOSYS as a capacity factor in each time slice for the
river segments represented in Fig. 5. Thereby acting as a constraint to
the operation of HPPs and varying their generation according to changes
in water availability.
2.4.3. Flood protection (FP)
In this scenario, which aims at addressing RQ3, a new set of opera-
tional rules is suggested based on consultation with stakeholders to
improve ood management in the basin by increasing the buffer volume
in selected dams. The storage capacity in Spilje HPP and Fierza HPP is
the main inuencer on ood control [20] since they have the largest
storage capacity in their respective country (Table 1). Therefore, new
operational rules are studied for these two dams. The buffer volume in
each dam is increased by 5%, 10%, 15% and 20% in the wet season
(from October to May). This means increasing the free volume spared in
the dam for ood control and reducing the storage volume dedicated to
electricity generation. The minimum storage level that should be
maintained in the dams did not allow for going beyond 20%. Since the
changes in the water levels are not directly proportional to the changes
in the volume, the Volume-Elevation (VE) curves [47] of the reservoirs
of interest are used to calculate the changes in the water level or head
5
The Skavica HPP project is not considered in this scenario to give it a special
focus in a separate scenario.
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
9
(see Figure C 7 and Figure C 8). This is then translated into changes in
the volume of water passing through the turbine to generate electricity
(as explained in section 2.2). In this way, we aspire to quantify the
trade-offs between the security of electricity supply and ood
mitigation.
2.4.4. New dam (skavica) (ND)
The Albanian government announced the development of a new
reservoir and HPP on the Drin River in Skavica [48]. In this scenario, we
assume the power plant is installed and starts operation by 2025 and we
explore the impact that the power plant could have on power generation
and electricity imports in Albania (RQ4). Adding the Skavica dam to the
cascade means water ow along the Black Drin part of the cascade has to
pass through the Skavica HPP and/or dam before continuing down-
stream to merge with the White Drin (Fig. 4).
2.5. Key model inputs and assumptions
From the electricity system’s perspective, the main inputs are pro-
jections of electricity demand in each country and techno-economic
assumptions regarding supply technologies and their input fuels and
transmission and distribution networks. The demand for each country is
based on national projections [49–51] which in most cases are up to
2030–2035. For the remaining period until 2050, extrapolation is made
based on the average annual growth rate of electricity demand
(2015–2050) for each country from the South East Europe Electricity
Roadmap (SEERMAP) [52–55]. The electricity demand projections are
shown in Figure C 4. Additionally, the model requires the load prole in
each time slice. This is obtained for Albania for three seasons (winter,
summer and intermediate). The prole is then applied to other riparians
since they have almost similar daily load proles (Figure C 5) but are
normalized across different seasons of the year [49–51]. We carry out an
extensive review of the literature and local sources to compile the list of
the existing and planned electricity generation technologies in each
country. Full lists of power infrastructure capacities are summarised in
Appendix B. Special focus is given to the main HPPs in the Drin basin
with the characteristics shown in the following table [26,56].
Non-hydro renewable energy technologies, mainly solar and wind,
are also considered in the model. All the existing and planned projects
are forced into the model as shown in Table B 7 - Table B 10. This
provides an outlook of investments until 2030. For the remaining period,
the model is allowed to gradually increase investments in solar and wind
if deemed necessary and cost-competitive (see Table 3). The capacity
factors for solar and wind technologies are extracted from the ‘Renew-
ables.Ninja’ dataset [57,58] for each country, with a spatial differenti-
ation between the inside and outside of the DRB and processed to weekly
time resolution (Table C 1 and Table C 2).
Electricity trade is modelled in a simplied manner due to the lack of
detailed data and lack of spatial detail in the modelling tool. Each
country is connected to a generic export and a generic import inter-
connection. An upper cap equivalent to the maximum historical level of
trade between 2011 and 2018 [27,51,66] is assumed for the rst
modelling year and an annual increase of 10% is allowed in the rst
decade. It is then lowered to 5% for the rest of the modelling period to
have a more representative outlook by avoiding massive imports.
Different limits to trade can be further detailed and explored in any
future work. The prices of exports and imports are based on average
historical records (2011–2018) for Albania [66]. An annual price in-
crease of 5% is assumed for the 2025–2034 period, followed by 2% for
2035–2044 and 1% for 2045–2050.
3. Results
This section highlights key insights from the scenario analysis. The
outcomes from each scenario (and its sub-scenarios) are discussed from
the energy system perspective. The focus is on the Drin basin, mainly the
cascade of the ve HPPs (Globocica, Spilje, Fierza, Koman and Vau i
Dej¨
es) but it also extends to some aspects of the national electricity
system.
3.1. Reference scenario
The DRB has signicant importance in terms of the security of the
electricity supply. Under the conditions of this scenario and before any
expansion in generation capacity by 2025 onwards, the HPPs in the
basin generate about 6405 GWh distributed between the countries as
shown in Fig. 7. The cascade of the ve large HPPs shared between
North Macedonia and Albania makes up about 82% of the basin’s hydro
generation.
The DRB supplies the Albanian grid with about 5.1 TWh of the total
7.5 TWh of electricity generated in Albania (i.e. ca 70%), as shown in
Fig. 8. While in Montenegro, the generation from the Drin basin is 0.9
TWh out of 3.4 TWh or about 27% of the total generation. Due to the
small capacities in the North Macedonian HPPs, the Drin Basin
Fig. 6. Scenarios tree of the Drin water-energy nexus assessment.
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
10
contributes by 0.44 TWh out of the total generation of 6.3 TWh or about
7% only of the national generation. Again, these values represent the
average generation for the years 2021–2024 before the expansion pro-
jects. In the following years (2025–2050), other technologies like solar
and wind add to the mix in each country. This results in a signicant
decrease in the share of Drin basin generation in Albania (from 70% to
56%). Changes are less important in other countries.
3.2. Climate change (CC) scenario
The impact of changing climate varies under different projections.
Looking at the change in electricity generation in the last decade of the
modelling period (2041–2050), it is noticed that the average annual
decline in generation from Fierza HPP due to climate change ranges
between 15%, 36% and 40% for RCPs 2.6, 4.5 and 8.5 respectively if
compared with the reference scenario. Koman HPP has a similar drop
(13%, 31% and 37%). However, Vau i Dej¨
es experiences a higher decline
ranging between (19%, 38% and 44%) under the same climate
conditions.
If we compare the absolute values, for example under the RCP8.5
scenario Koman HPP loses on average 785 GWh per year, Fierza HPP
726 GWh and Vau i Dej¨
es 518 GWh as shown in Fig. 9. This is due to the
differences in installed capacity in each HPP (see Table 1). In other
words, the Albanian national grid receives less hydropower from the
Drin basin ranging from 775 GWh to 1754 GWh and 2028 GWh under
RCPs 2.6, 4.5 and 8.5 respectively. This means a reduced annual hydro
generation of 15%, 34% and 40% respectively as shown in Table 4.
In North Macedonia, the impact of climate change has a similar
trend. The north Macedonian grid loses about 100–230 GWh of elec-
tricity each year on average in the last decade from the HPP along the
Drin river as shown in Fig. 10. This makes up about 23–52% of the
generation if compared to the reference scenario.
The drop in Globocica HPP ranges between 65, 71 and 74 GWh,
while the drop in generation from Spilje HPP ranges between 38, 83 and
157 GWh under RCP2.6, RCP4.5 and RCP8.5 respectively. The change in
the North Macedonian part should be taken with caution. The hydro-
logical data used in this analysis projects a signicant decline in the
water ow from the river segments in the north Macedonian part even at
the beginning of the modelling period. It also shows differences in the
historical period (1981–2010) between the reference scenario and the
climate change scenarios. This highlights the importance of understating
climate projections when extracting modelling insights and designing
long-term energy strategies.
In conclusion, the results of the CC scenario show that Albania’s
electricity system suffers higher consequences due to climate change
compared to North Macedonia. As shown in Table 4, Albania loses about
Table 2
Techno-Economic characteristics of the power supply technologies.
Type Capital Cost
a
(million USD ‘mUSD’) Variable Cost Fixed O&M Operational
Life
Capacity Factor
AL MK ME XK (mUSD/
TWh)
(mUSD/
GW)
(Years) (%)
Large Hydro - Dam (New) 1169 (Kalivac HPP) - 3092 (Skavica
HPP)
2552 3453 2240 NA 3.4 50 36–39
Medium Hydro - Run off
river
NA 2355 2355 NA NA 3.4 50 26–29
Small Hydro NA NA NA NA NA 3.4 50 11
Solar PV 905 975 900 2128 NA 35.5 20 see table Table C
1
Wind 1288 1700 1866–2191 1802 NA 29 25 see Table C 2
Coal Power plant - Existing NA NA NA NA 4.18 29 30 65
Coal Power plant - New NA 1555 1490 3000 5.18 29 30 70
CHP - NEW NA 733 NA NA 1.58 9.2 30 65
Combined Cycle - New 1501 1232 NA NA 1.58 9.2 30 65
a
Capital costs are based on announced projects in each country. For (Medium hydro – Run-off-river), capital costs are based on the average of 3 projects in North
Macedonia.
Sources [26,27,48,59–65]
:
Table 3
Key assumptions under each scenario.
Scenarios
Reference Climate Change Flood Protection New Dam
(Skavica)
Water Flow The water ow in the cascade is based on the historical
mean ow from (1981–2009).
The water ow in the cascade is based on the climate
projections for three RCPs.
Same as in the
reference scenario.
New operational rules
are introduced for
Spilje and Fierza.
same as in the
reference
scenario.
Skavica dam Skavica new dam is not introduced in this scenario. The model is allowed to invest in Skavica HPP but the dam is not added to the
cascade.
Skavica new dam
is detailed in the
cascade.
Trade limits Electricity trade limits are based on historical values from (2011–2018) and the model is allowed to increase by 10% annually (2021–2029) and then 5% annually
(2030–2050).
Renewable
capacity
Renewable (solar and wind) energy installations are
limited to the conrmed projects up until 2030. The
constraints are relaxed for the following years and for
each technology an annual increase of 10% is allowed.
Renewable capacity limits are relaxed for Albania and
North Macedonia to allow an annual increase of
30–45% between (2026–2030) and 10% for the rest of
the modelling period.
Same as in the
reference scenario.
Same as in the
reference
scenario.
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
11
775–2000 GWh of its generation from the Drin basin due to climate
change. This can be attributed to the large capacities and the great de-
pendency of the Albanian electricity system on hydropower in the basin.
In North Macedonia, the basin contributes by 7% only of the total supply
and the size of its dams is much smaller. This makes the country’s elec-
tricity system less vulnerable to changes in climate and river ow along
the Drin and the decline in electricity supply is between 100 and 230 GWh.
Another result can be observed, which is partially linked to the ones
shown so far. Non-hydro RES namely solar and wind have great po-
tential (given the resources) and are cost-competitive. When the con-
straints on the annual installed capacity for solar and wind are relaxed,
they become cost-competitive and gain signicant shares in the elec-
tricity supply, making up for the lost hydro generation. They also offset
imports, taking a dominating role in the supply. As shown in Fig. 11,
from 2041 to 2050 Albania triples its generation from solar and wind
under climate change conditions to overcome the decline in hydro
generation from the Drin Basin. This requires increasing the solar ca-
pacity from 660 MW in the Reference scenario to about 2250 MW under
the climate change scenario and sensitivities. Similarly, wind capacity
increases from 330 MW to 970 MW. In North Macedonia, since the
decline in hydro is relatively smaller, it is enough to increase solar ca-
pacity from 1600 MW to 2280 MW under the climate change scenario
and sensitivities, while wind has almost the same capacity (640 MW)
and generation (740 GWh) in all scenarios. This is due to the relatively
lower investment cost required for solar projects compared to wind as
shown in Table 2. The other reason is the relatively low capacity factor
for wind technology compared to solar technology as shown in Figure C
6, Table C 1 and Table C 2.
It is noteworthy that the investments in solar and wind not only help
mitigate climate impact but help also enhance energy independence and
reduce electricity imports in both countries, as shown in Fig. 11 _a.
Imports decrease by 65% in Albania and by 25% in North Macedonia.
This is driven in the model by the gradual increase in import prices in
each decade. This insight highlights the importance and cost-
competitiveness of non-hydro renewables not only from a climate
change perspective but also from an energy security standpoint. Such
insight would not be possible to achieve without developing a national-
scale energy model and integrating the hydrological aspects (e.g. water
ow) to reect the dynamics of climate change and hydropower.
The uncertainty in climate change projections, although not the
focus of this study, cannot be ignored [67]. The uncertainty in climate
projections is usually attributed to three sources: a) the future
Fig. 7. Average annual electricity generation (in GWh) from the hydropower plants in the Drin Cascades under the reference scenario, between 2021 and 2024.
Fig. 8. Electricity generation (GWh) from hydropower in the Drin basin
compared to the national electricity generation in Albania, North Macedonia
and Montenegro under the reference scenario. (Note: Kosovo does not have any
hydro capacity in the basin.)
Fig. 9. Average electricity generation in each decade from the hydropower
plants in the Albanian cascade under different scenarios (REF, RCP2.6, RCP4.5
and RCP8.5).
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
12
Greenhouse Gas (GHG) emissions concentration, b) the climate sensi-
tivity which refers to the response in climate (e.g. increase in tempera-
ture) due to the increase in emissions, and c) the climate models
limitations [68]. We explored three RCPs to minimize the uncertainty in
future emission levels. We used an ensemble of different models to
minimize the uncertainty in climate models and climate sensitivity.
More specically, the E-HYPE model provides 11 projections for river
ow based on inputs from different GCM and RCM models (Table C 3).
The projections are distributed unequally among the RCPs. A mean -
value ensemble for each RCP (2.6, 4.5 and 8.5) from the given pro-
jections is generated to dene robust signals of future changes within a
concise number of scenarios. Although this is a well-known approach to
reducing the uncertainty in climate models [69] and is used widely in
the scientic community [70–72], this approach has its limitations. For
example, this approach may hide the effect of extreme events. An
alternative approach can be exploring a larger number of climate change
scenarios from projections representing the full range of GCM and RCM
results as done by Ref. [73].
Table 4
Summary of the average electricity generation in each of the HPPs in (GWh) under the Climate Change scenario – Average generation in (2041–2050).
country N.Macedonia Albania
Scenario/HPP Globocica Spilje Total Fierza Koman Vau i Dej¨
es Total
Generation (GWh) - REF scenario 127 319 446 1848 2094 1174 5117
RCP26 Generation (GWh) 62 281 343 1571 1816 955 4342
Diff. compare to REF (GWh) −65 −38 −103 −277 −278 −220 −775
% Difference −51 −12 −23 −15 −13 −19 −15
RCP45 Generation (GWh) 56 236 292 1189 1445 729 3363
Diff. compare to REF (GWh) −71 −83 −154 −658 −650 −446 −1754
% Difference −56 −26 −35 −36 −31 −38 −34
RCP85 Generation (GWh) 53 162 215 1122 1309 657 3088
Diff. compare to REF (GWh) −74 −157 −231 −726 −785 −518 −2028
% Difference −58 −49 −52 −39 −37 −44 −40
Fig. 10. Total change in electricity generation from Globocica and Spilje hy-
dropower plants (combined) in North Macedonia.
Fig. 11. Solar and wind in Albania and North Macedonia, a) Average generation (GWh) in the period 2041–2050, b) Total installed capacity (MW) in the
same period.
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
13
3.3. Flood protection scenario
As previously mentioned, the changes under this scenario are
implemented on Spilje and Fierza dams due to the prominence of their
storage capacities and their impact on ood protection in the basin.
Results show that increasing the buffer volume in Spilje by 5% means
gaining an additional buffer volume of 7–9 million cubic meters (MCM),
while the 20% increase results in an additional buffer of about 26–34
MCM as shown in Fig. 12. This comes at the cost of lowering the water
level in the Spilje dam, based on the proposed operational rules, from 2
to 4.3 m as shown in Table C 4.
Due to its large volume, the changes in the Fierza dam are much
larger. Adding a 5% buffer translates into 36–68 MCM of additional
storage capacity while adding a 20% buffer means 144–270 MCM of
additional storage capacity (Fig. 13). Achieving this large volume
requires changing the water level in the Fierza dam from 0.5 to 7.8 m as
shown in Table C 5.
With these changes, the model results show that increasing the buffer
volume results in a very small change in electricity generation. The
average losses in generation range from 7 to 10 GWh per year in Spilje
and about 5–28 GWh per year in Fierza as shown in Table 5. This rep-
resents 2.2–3.2% drop in Spilje and 0.3–1.5% drop in Fierza, for the 5%
and 20% scenarios, respectively.
In monetary terms, the annual losses from electricity sales range
between EUR 560 k – 800 k in North Macedonia and between EUR 400 k
– 2,240 k in Albania for the 5% and 20% scenarios respectively
(Table 5). These losses are very low if compared to the savings in terms
of avoided ood damages. For example, in 2010 the ood event resulted
in unprecedented ooded areas and damage in the region. The total
countrywide damage and losses in Montenegro only exceeded EUR 40
million and 1.5% of the population was evacuated. It is believed that the
oods were exacerbated by the release of 3000 m
3
/s of water into the
Buna/Bojana River from the reservoirs in Albania (Fierza, Koman and
Vau i Dej¨
es) [17].
Additionally, looking at the overall generation from all ve HPPs in
the Drin River Basin (Fig. 14) the change is almost negligible as other
HPPs tend to produce more electricity. This indicates that sparing
additional volumes to have better ood control may not jeopardize the
security of the electricity supply. This indicative insight should motivate
the decision-makers in the energy sector to rethink the existing opera-
tional rules of the HPPs, which were set three decades ago, to achieve
ood-smart management by increasing the buffer volumes in the dams.
3.4. New dam scenario
This scenario explores the impact of the new Skavica dam on the
electricity system in Albania. The results show that once it starts oper-
ation in 2025, Skavica (196 MW capacity) can add 550 GWh of elec-
tricity to the Albanian grid, if it operates on a system-wide cost-
minimisation perspective. This improves energy independence and re-
duces electricity imports cumulatively by more than 18,000 GWh be-
tween 2025 and 2050 as shown in Fig. 15.
Besides the gains specic to the ND scenario, shown in Fig. 15, the
level of imports reduces annually in both the REF and ND scenarios also
because the generation from solar and wind increases over the years
(Fig. 16) and the price of imports increases in each decade. This makes it
increasingly economically protable to rely on domestic supply.
Furthermore, and from the ood management perspective, Skavica
adds 2300 MCM of storage capacity. This increases the total storage
capacity in the basin from 3409 MCM today to 5709 MCM, which is
needed to mitigate ood risk. This has not been investigated in the ood
protection scenario as the exact impacts of the dam on the water ows
and the operational rules of the dam are not known yet. However, it is
arguable that the presence of this dam would also allow the increases in
buffer volumes to be shared between more plants, with reduced effects
on the electricity generation of each plant and increased ood risk
mitigation.
Fig. 13. Additional buffer volume gained under 5% and 20% increase in
Fierza dam.
Fig. 12. Additional buffer volume gained by 5% and 20% increase in Spilje
dam (Million Cubic Metres).
Table 5
Summary of the changes in terms of electricity generation in Spilje and Fierza hydropower plants.
Parameter Spilje HPP Fierza HPP
+5% Buffer (FP_05) +20% Buffer (FP_20) +5% Buffer (FP_05) +20% Buffer (FP_20)
% change in generation - 2.2% - 3.2% - 0.3% - 1.5%
Mean annual change in generation (GWh) - 7 - 10 - 5 - 28
Losses in monetary values (kEuros)
a
−560 −800 −400 −2240
a
Based on average household electricity prices in each country (2013–2019) [74].
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
14
4. Discussions
In our scenario analysis, the variation in water availability due to
changing climate results in a signicant decline (15–52% annually) in
hydro generation in the basin by the mid-century. The losses (in absolute
terms) in the Albanian cascade are larger than in other riparians and
reach on average up to 775, 1750 and 2000 GWh on annual basis under
RCPs 2.6, 4.5 and 8.5 respectively. Taking a least-cost perspective on
electricity supply planning, non-hydro renewables play an important
role in mitigating the impact of climate change on the security of the
electricity supply, especially in the long term.
Climate change has a long-term impact by nature, which highlights
the need for long-term planning and investment outlook. The analysis
shows that solar and wind have the potential to play an important role in
the electricity supply mix of the Drin riparian countries and compensate
for the declines potentially caused by climate change. However, the
ofcially announced investments are not enough to reect this high
potential. Revisiting the investment plans may be needed. For example,
in the model results solar and wind technologies in Albania require triple
the capacity (compared to the reference case) to compensate for the
decline of hydro in the basin. Additionally, solar and wind energy also
enhance energy independence and reduce electricity imports in Albania
by 65% and North Macedonia by 25% compared to the reference case.
The overall system cost
6
increases from EUR 34.7 billion in the reference
scenario to EUR 58.7, 114 and 155 billion under RCPs 2.6, 4.5 and 8.5
respectively. This change in the system cost is mainly driven by the in-
crease in solar and wind installations to compensate for the decline in
hydro generation from the Drin basin and the change in electricity
imporst. Which can be seen as another indicator of the impact of climate
change on the electricity systems of the Drin River Basin countries.
That being said, it is important to note that as the share of Variable
Renewable Energy (VRE) sources, such as solar and wind, increases in a
power system, there may be a need for greater system exibility (e.g.
energy storage) to maintain a balance between supply and demand.
However, this aspect is not addressed in the Drin model structure or the
analysis, as the study’s focus is on the long-term impact of climate
change. Therefore, weekly averages are used for both nal electricity
demand and VRE capacity factor, which is considered appropriate for
the study’s long-term scope also the storage modelling was limited to
hydro storage. Moreover, the modelling complexity and temporal res-
olution are kept at this level to facilitate capacity-building efforts. If the
short-term analysis or system exibility question needs to be addressed,
alternative modelling tools such as electricity market and dispatch
models (e.g., PLEXOS [75]) or modications to the OSeMOSYS code can
be explored, as studied by Welsch et al. [76].
Changing the operational rules of the dams to accommodate oods
Fig. 14. Change in electricity generation under the ood control scenario. A) Impact on Fierza HPP in Albania, b) Impact on Spilje HPP in North Macedonia and C)
Total generation from the ve HPPs in the Drin Basin from both Albania and North Macedonia.
6
The system cost: is the total electricity generation cost in the four countries
modelled (Albania, North Macedonia, Montenegro and Kosovo).
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
15
has a minor impact on the annual electricity generation in the basin,
according to our scenarios. The losses in generation from the studied
HPPs (Spilje and Fierza) are in the range of 0.5–3% annually. On the
other hand, such changes have the potential to spare an additional 7–34
MCM of volume to be used for ood control. This insight could motivate
the stakeholders from the energy sector to cooperate with other sectors
in minimising ood risks by rethinking the existing operational rules of
the HPPs, which were set three decades ago.
The investment in the new Skavica hydropower plant has a positive
impact on energy independence. The additional 196 MW capacity adds
about 550 GWh of hydropower to the Albanian grid and reduces the
imports by 18 TWh between 2025 and 2050, from a cost-minimisation
perspective. Furthermore, Skavica adds 2300 MCM of storage capac-
ity, which provides room for further mitigating ood risk.
Future work can focus on addressing a number of limitations in this
study. For example, sensitivity analyses could be carried out to explore
the impact of different operational rules and different prices for elec-
tricity imports. More importantly, the impact of Skavica on the opera-
tion of the other HPPs in the Albanian cascade can be further detailed
once actual operational data from Skavica is obtained. Increasing the
time resolution, expanding storage modelling and exploring the short-
term dynamic of VRE investments and grid stability might be another
area of improvement. In terms of climate change projections, future
work could explore a broader range of climate projections and scenarios
using different hydrological models to quantify the uncertainty of
climate change impact.
Fig. 16. Electricity generation in Albania from solar and wind technologies under the reference (REF) and new dam, Skavica (ND) scenarios.
Fig. 15. Change in annual electricity imports in Albania between the REF scenario (Without Skavaica) and the ND scenarios (with Skavica).
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
16
5. Conclusion
The Drin River Basin (DRB) plays an important role in the energy
systems of Albania, Montenegro and North Macedonia. It supplies about
70% of the total electricity supply in Albania, 27% in Montenegro and
7% in North Macedonia. The cascade of the ve large HPPs shared be-
tween North Macedonia (upstream) and Albania (downstream) makes
up about 82% of the basin’s generation.
The basin is facing several challenges of which climate change and
oods come at the top of the list. This study explored the impact of
climate change and oods on the energy system by conducting a techno-
economic optimisation analysis of the Drin River Basin. A multi-country
water-electricity system model for the four riparian countries was
developed using the Open Source energy Modelling System (OSe-
MOSYS). The model includes a representation of the hydrological sys-
tem in the basin and integrates it with the electricity system to reect the
links between climate change, water ows, hydropower operation and
national electricity supply security.
This study shows that the vulnerability of the Albanian power system
to climate change is higher than the North Macedonian one. VRE has the
potential to mitigate the risk of CC and increase the security of electicity
supply. Furthermore, the energy sector can play an important role in
mitigating the impact of oods and increasing the nations preparedness
for extreme climate events.
The insights drawn from this analysis informed the dialogue on the
sustainable management of resources in the Drin basin between stake-
holders from the water and energy sectors in the riparian countries [77,
78]. An integrated water-energy management plan could be a tool for
each dam operator to plan, not only in the shorter term but also in the
long-term, for both the production and ood control services. The model
developed in this analysis was used as a showcase for capacity building
on integrated water-energy modelling [79].
Credit author statement
Youssef Almulla: Conceptualization, Methodology, Software,
Formal analysis, Visualization and Writing - Original Draft; Klodian
Zaimi: Methodology, Formal analysis, Writing - Review & Editing; Emir
Fejzi´
c: Methodology, Writing - Review & Editing; Vignesh Sridharan:
Software, Visualization, Writing- original draft preparation.; Lucia de
Strasser: Conceptualization, Project administration, Funding acquisi-
tion and supervision.; Francesco Gardumi: Conceptualization, Writing
- Review & Editing and supervision, Project administration and
supervision.
Disclaimer
The views expressed in this article are those of the authors and do not
necessarily represent the views of the United Nations or its Member
States.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Data availability
on a Github repository: https://github.com/KTH-dESA/Drin_Water_
Energy_Nexus_Model
Acknowledgement
This work was funded by the United Nations Economic Commission
for Europe (UNECE) through grant number: UNOG2020-470000. We
would like also to acknowledge Global Water Partnership – Mediterra-
nean (GWP-Med) and the Drin Core Group (DCG) for their valuable
contribution to this work by providing some of the input data and
validating modelling assumptions.
Appendix A
Changes in OSeMOSYS Storage Equations:
The following storage equations and constraints are updated to introduce the weekly changes in the operational rules of hydropower plants:
Storage Equations:
s.t.S1 StorageLevelYearStart{r in REGION,s in STORAGE,y in YEAR}:if y =min{yy in YEAR}min(yy)then StorageLevelStart[r,s]
else StorageLevelYearStart[r,s,y−1] + sum{l in TIMESLICE}(sum{t in TECHNOLOGY,m in MODE OF OPERATION
:TechnologyToStorage[r,t,s,l,m]>0}(RateOfActivity[r,l,t,m,y−1]
∗TechnologyToStorage[r,t,s,l,m]) − (sum{t in TECHNOLOGY,m in MODE OF OPERATION
:TechnologyFromStorage[r,t,s,l,m]>0}RateOfActivity[r,l,t,m,y−1] ∗ TechnologyFromStorage[r,t,s,l,m]))
∗YearSplit[l,y−1] = StorageLevelYearStart[r,s,y]Equation A1
s.t.S2 StorageLevelTSStart{r in REGION,s in STORAGE,l in TIMESLICE,y in YEAR}
:if l =min{ll in TIMESLICE}min(ll)then StorageLevelYearStart[r,s,y]
else StorageLevelTSStart[r,s,l−1,y] + (((sum{t in TECHNOLOGY,m in MODE OF OPERATION
:TechnologyToStorage[r,t,s,l,m]>0}RateOfActivity[r,l−1,t,m,y]
∗TechnologyToStorage[r,t,s,l,m]) − (sum{t in TECHNOLOGY,m in MODE OF OPERATION
:TechnologyFromStorage[r,t,s,l,m]>0}RateOfActivity[r,l−1,t,m,y] ∗ TechnologyFromStorage[r,t,s,l,m]))
∗YearSplit[l−1,y]) = StorageLevelTSStart[r,s,l,y]Equation A2
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
17
s.t.SC8 StorageRefilling{s in STORAGE,r in REGION}:sum{y in YEAR,l in TIMESLICE} (sum{t in TECHNOLOGY,m in MODE OF OPERATION
:TechnologyToStorage[r,t,s,l,m]>0}(RateOfActivity[r,l,t,m,y]
∗TechnologyToStorage[r,t,s,l,m])− (sum{t in TECHNOLOGY,m in MODE OF OPERATION
:TechnologyFromStorage[r,t,s,l,m]>0}RateOfActivity[r,l,t,m,y]
∗TechnologyFromStorage[r,t,s,l,m])) ∗ YearSplit[l,y]=0
Equation A3
Storage constraints:
s.t.SC1 LowerLimit{r in REGION,s in STORAGE,l in TIMESLICE,y in YEAR}:MinStorageCharge[r,s,y] ∗ (sum{yy in YEAR
:y−yy <OperationalLifeStorage[r,s]&& y−yy >=0}NewStorageCapacity[r,s,yy]+ResidualStorageCapacity[r,s,l,y]) <=StorageLevelTSStart[r,s,l,y]
Equation A4
s.t.SC2 UpperLimit{r in REGION,s in STORAGE,l in TIMESLICE,y in YEAR}:StorageLevelTSStart[r,s,l,y]<= (sum{yy in YEAR
:y−yy <OperationalLifeStorage[r,s]&& y−yy >=0}NewStorageCapacity[r,s,yy] + ResidualStorageCapacity[r,s,l,y]) Equation A5
s.t.SC7 StorageMaxUpperLimit{s in STORAGE,l in TIMESLICE,y in YEAR,r in REGION}:(sum{yy in YEAR
:y−yy <OperationalLifeStorage[r,s]&& y−yy >=0}NewStorageCapacity[r,s,yy]+ResidualStorageCapacity[r,s,l,y]) <=StorageMaxCapacity[r,s,l,y]
Equation A6
Appendix B
Table B1
List of existing and planned Thermal power plants in the four countries.
Country Power Plant Capacity
(MW)
Assumptions:
Albania Vlore (CC gas) 98 Assumed to continue for the entire modelling period. With a low capacity factor
of 30% only.
Albania GPP Korça shpk 500 The project was not granted environmental permission and not ofcially
cancelled but it is difcult to get through without this permission. (Removed from
the model)
North
Macedonia
Bitola 699 Operational Coal PP
North
Macedonia
REK Oslomej 125 Not operating since 2015, not considered in the model
North
Macedonia
Negotino (heavy fuel oil PP) 198 Assumed operational until 2025.
North
Macedonia
TE-TO 230 Operational CHP, phased out by 2040
North
Macedonia
Kogel 30 Operational CHP, phased out by 2040
North
Macedonia
Energitica 30 Operational CHP, phased out by 2041
North
Macedonia
Bitola (revitalization) 650 New: allowed after 2025
North
Macedonia
Oslomej (revitalization) 109 New: 2023
North
Macedonia
New Lignite PP 300 New: 2035
North
Macedonia
New CHP 450 New: 2025
North
Macedonia
Exist CHP (revitalization) 260 New: 2021
North
Macedonia
New Cas CHP 40 New: 2023
North
Macedonia
New Cas CHP 30 New: 2023
North
Macedonia
New Cas CHP 30 New: 2023
North
Macedonia
New NGCC by 2033 230 New: 2033
North
Macedonia
A new NG-red power plant in the southwestern city of Bitola 800 Not added yet to the model as there is no clear date. The project cost is estimated
at 400 million euros.
North
Macedonia
shut down the coal-red REK Bitola, with an installed capacity
of 675 MW, and convert it to gas from a planned pipeline
675 (699) Not added yet to the model as there is no clear date.
North
Macedonia
Negotino project is related to the plan to switch the 210 MW
thermal power plant TEC Negotino from fuel oil to natural gas.
210 (198) Not added yet to the model as there is no clear date.
Montenegro Pljevlja TPP 225 According to EPCG: Pljevlja TPP will continue at least for 5 years.
It is considered in the model for the entire modelling period.
(continued on next page)
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
18
Table B1 (continued )
Country Power Plant Capacity
(MW)
Assumptions:
Kosovo Kosova A: 610 MW 610 The new capacity is 915 MW for both units.
Kosovo A will phase out by 2030 and Kosovo B by 2040. Kosovo Kosova B:678 MW 678
Kosovo New (Kosova e Re) 500 under construction, expected by 2023.
Table B2
List of existing and planned hydropower plants in the four countries
Country Type Plants Location Capacity
(MW)
Note
Albania Hydro dam Okshtun +Temove +Lubalesh 1 Inside the basin 15
Albania Hydro dam NEW Skavica Hpp Inside the basin 196 Assumed after 2025.
All other scenarios except
reference.
Albania Hydro dam Fierza HPP Inside the basin 500
Albania Hydro dam Koman HPP Inside the basin 600
Albania Hydro dam Vau I dejes HPP Inside the basin 250
Albania Hydro dam Ulza: 25.2 MW, Shkopeti: 25 MW, Banja:73 MW Outside the
basin
123.2
Albania Hydro dam Moglice: 192 MW,2019
Kalivac: 120 MW, 2020
Shala: 83.5 MW, 2021
Outside the
basin
395.5 New projects added in respective
years
Albania Runoff River Ashta 1: 22.2 MW
Ashta 2: 34.2 MW
Liapaj: 13.62 MW
Bele2: 11 MW
Lubalesh 2+Gjorice: 10.9 MW
Inside the basin 91.92
Albania Runoff River Bistrica 1 & 2 cascade: 27.5 MW
Sllabinje (Fterre Sarande): 13.8 MW
Tervol: 12 MW
Private Generator: 191.4 MW
Outside the
basin
240
Albania Small Hydro (<10
MW)
See Table B 3 Inside the basin 110.18
Albania Small Hydro (<10
MW)
See Table B 4 Outside the
basin
190.23
North
Macedonia
Hydro dam Globicica HPP Inside the basin 42
North
Macedonia
Hydro dam Shpilje HPP Inside the basin 84
North
Macedonia
Hydro dam Kalimanci: 13.6 MW
Tikvesh: 112 MW
Vrutok: 165.6 MW
Kozjak: 82.5 MW
Sveta Petka: 36 MW
Outside the
basin
409.7
North
Macedonia
Hydro dam Globocica II: 20 MW, 2035
Veles:96 MW, 2030.
Chebren:458 MW, 2029
Gradec:75.34 MW, 2030
Galiste:77.9 MW, 2035
Inside the basin New projects
North
Macedonia
Runoff River Vrben: 12.8 MW
Raven:21 MW
Outside the
basin
North
Macedonia
Runoff River Vardar Valley 1: 45 MW, 2025
Vardar Valley 2: 152.51 MW, 2030
Small hydro: 160 MW, 2019
Outside the
basin
New projects
North
Macedonia
Small Hydro (<10
MW)
See Table B 5 Outside the
basin
97
Montenegro Hydro dam Peru´
cica HPP: 307 MW, Unit8 with 58.5 MW is planned
for 2025
Moraˇ
ca 238.4 MW, assumed by 2035
Inside the basin 307
Montenegro Hydro dam Piva HPP Outside the
basin
342
Montenegro Hydro dam Komarnica: 172 MW, 2029
Kruˇ
sevo V1: 82 MW, assumed by 2035
Kruˇ
sevo V2: 100 MW, assumed by 2038
Outside the
basin
354 New projects
Montenegro Runoff River See Table B 6 Outside the
basin
50 New projects
Montenegro Runoff River See Table B 6 Inside the basin 20 New projects
Kosovo Hydro dam Ujmani +Lumbardhi +Dikance +Burimi +Radavci Outside the
basin
61
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
19
Table B3
List of small hydropower plants in Albania - Inside the Drin Basin
Country Type Plant Location Capacity (MW)
Albania Small Hydro Pobreg Inside the basin 12.7
Albania Small Hydro Dardhe Inside the basin 5.80
Albania Small Hydro Selishte Inside the basin 2.00
Albania Small Hydro Bele 1 Inside the basin 5.00
Albania Small Hydro Topojan 2 Inside the basin 5.80
Albania Small Hydro Shemri Inside the basin 1.00
Albania Small Hydro Mgulle Inside the basin 0.80
Albania Small Hydro Tucep Inside the basin 0.40
Albania Small Hydro Ostreni i Vogel Inside the basin 0.32
Albania Small Hydro Murdhar 1 Inside the basin 2.68
Albania Small Hydro Murdhar 2 Inside the basin 1.00
Albania Small Hydro Trebisht Inside the basin 1.78
Albania Small Hydro Topojan 1 Inside the basin 2.90
Albania Small Hydro Orgjost I Ri Inside the basin 4.80
Albania Small Hydro Truen Inside the basin 2.50
Albania Small Hydro Kacni Inside the basin 3.87
Albania Small Hydro Borove Inside the basin 1.92
Albania Small Hydro Zabzun Inside the basin 0.30
Albania Small Hydro Sebishte Inside the basin 2.84
Albania Small Hydro Prodan 1 Inside the basin 0.38
Albania Small Hydro Prodan 2 Inside the basin 0.80
Albania Small Hydro Okshtun Ekologjik Inside the basin 0.45
Albania Small Hydro Ternove Inside the basin 0.92
Albania Small Hydro Lubalesh 1 Inside the basin 4.60
Albania Small Hydro Lubalesh 2 Inside the basin 5.10
Albania Small Hydro Gjorice Inside the basin 4.18
Albania Small Hydro Bulqize Inside the basin 0.60
Albania Small Hydro Orgjost Inside the basin 1.20
Albania Small Hydro Lekbibaj Inside the basin 1.40
Albania Small Hydro Zerqan Inside the basin 0.63
Albania Small Hydro Shoshan (Shoshaj) Inside the basin 0.00
Albania Small Hydro Homesh Inside the basin 0.33
Albania Small Hydro Muhur Inside the basin 0.25
Albania Small Hydro Dukagjin Inside the basin 0.64
Albania Small Hydro Lure (also Lura) Inside the basin 0.75
Albania Small Hydro Borje Inside the basin 1.50
Albania Small Hydro Oreshke Inside the basin 5.60
Albania Small Hydro Carnaleva Inside the basin 2.95
Albania Small Hydro Carnaleva 1 Inside the basin 3.27
Albania Small Hydro Lura 1 Inside the basin 6.54
Albania Small Hydro Lura 2 Inside the basin 4.02
Albania Small Hydro Lura 3 Inside the basin 5.66
Table B4
List of small hydropower plants in Albania - Outside the Drin Basin
Country Type Plant Location Capacity (MW)
Albania Small Hydro Vlushe Outside the basin 14.2
Albania Small Hydro Martanesh (Bulqize) Outside the basin 10.5
Albania Small Hydro Labinot-Mal (Elbasan) Outside the basin 0.25
Albania Small Hydro Gjanc Outside the basin 2.96
Albania Small Hydro Smokthine Outside the basin 9.20
Albania Small Hydro Bene Outside the basin 1.00
Albania Small Hydro Selce Outside the basin 1.60
Albania Small Hydro Bogove (Skrapar) Outside the basin 2.50
Albania Small Hydro Xhyre (Librazhd) Outside the basin 0.25
Albania Small Hydro Vithkuq (Korce) Outside the basin 2.72
Albania Small Hydro Orenje (Librazhd) Outside the basin 0.88
Albania Small Hydro Bishnica 2 Outside the basin 2.50
Albania Small Hydro Dishnica Outside the basin 0.20
Albania Small Hydro Lubonje Outside the basin 0.30
Albania Small Hydro Gizavesh Outside the basin 0.50
Albania Small Hydro Carshove Outside the basin 1.50
Albania Small Hydro Sasaj (Sarande) Outside the basin 7.00
Albania Small Hydro Klos (Mirdite) Outside the basin 1.95
Albania Small Hydro Peshku (Burrel) Outside the basin 3.43
Albania Small Hydro Kumbull- Merkurth (Mirdite) Outside the basin 0.83
Albania Small Hydro Picar 1 (Gjirokaster) Outside the basin 0.20
Albania Small Hydro Qafzeze Outside the basin 0.40
Albania Small Hydro Mollaj Outside the basin 0.60
Albania Small Hydro Kryezi 1 Outside the basin 0.60
(continued on next page)
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
20
Table B4 (continued )
Country Type Plant Location Capacity (MW)
Albania Small Hydro Kryezi i Eperm Outside the basin 0.20
Albania Small Hydro Shkalle Outside the basin 1.60
Albania Small Hydro Cerunje 1 Outside the basin 2.30
Albania Small Hydro Cerunje 2 Outside the basin 2.80
Albania Small Hydro Klos Outside the basin 2.30
Albania Small Hydro Rrype Outside the basin 3.60
Albania Small Hydro Koka1 Outside the basin 3.20
Albania Small Hydro Rapuni 1 Outside the basin 4.10
Albania Small Hydro Rapuni 2 Outside the basin 4.00
Albania Small Hydro Qarr Outside the basin 1.00
Albania Small Hydro Kaltanj Outside the basin 0.50
Albania Small Hydro Langarica 3 Outside the basin 2.20
Albania Small Hydro Gostivisht Outside the basin 1.30
Albania Small Hydro Ura e Dashit Outside the basin 1.20
Albania Small Hydro Sotira 1&2 Outside the basin 2.20
Albania Small Hydro Kozel Outside the basin 0.50
Albania Small Hydro Helmes 1 Outside the basin 0.80
Albania Small Hydro Helmes 2 Outside the basin 0.50
Albania Small Hydro Cekrez 1 Outside the basin 0.43
Albania Small Hydro Cekrez 2 Outside the basin 0.23
Albania Small Hydro Radove Outside the basin 2.50
Albania Small Hydro Perrollaj Outside the basin 0.50
Albania Small Hydro Stravaj Outside the basin 3.60
Albania Small Hydro Shutine Outside the basin 2.40
Albania Small Hydro Gurshpate 1 Outside the basin 0.84
Albania Small Hydro Gurshpate 2 Outside the basin 0.83
Albania Small Hydro Hurdhas 1 Outside the basin 1.71
Albania Small Hydro Hurdhas 2 Outside the basin 1.30
Albania Small Hydro Hurdhas 3 Outside the basin 1.20
Albania Small Hydro Treska 2 Outside the basin 0.62
Albania Small Hydro Treska 3 Outside the basin 0.40
Albania Small Hydro Treska 4 Outside the basin 3.60
Albania Small Hydro Lengarica Outside the basin 8.94
Albania Small Hydro Driza Outside the basin 3.41
Albania Small Hydro Bistrica 3 Outside the basin 1.53
Albania Small Hydro Strelce Outside the basin 1.5
Albania Small Hydro Treska 1 Outside the basin 0.13
Albania Small Hydro Lanabregas 1 +2 Outside the basin 5.00
Albania Small Hydro Borshi (also Borsh) Outside the basin 0.25
Albania Small Hydro Funares Outside the basin 1.92
Albania Small Hydro Lunik Outside the basin 0.20
Albania Small Hydro Nikolica (also Nikolice) Outside the basin 0.70
Albania Small Hydro Vithkuq Outside the basin 0.00
Albania Small Hydro Velcan Outside the basin 1.20
Albania Small Hydro Leshnice (also Leshnica) Outside the basin 0.38
Albania Small Hydro Kerpice Outside the basin 0.42
Albania Small Hydro Barmash Outside the basin 0.63
Albania Small Hydro Marjan Outside the basin 0.20
Albania Small Hydro Arras (also Arres or Arrez) Outside the basin 4.80
Albania Small Hydro Ujanik Outside the basin 0.63
Albania Small Hydro Voskopoje Outside the basin n/a
Albania Small Hydro Piqeras (also Piqerras) Outside the basin n/a
Albania Small Hydro Rajan Outside the basin 1.02
Albania Small Hydro Lozhan Outside the basin 0.10
Albania Small Hydro Faqekuq 1 Outside the basin 3.00
Albania Small Hydro Faqekuq 2 Outside the basin 3.40
Albania Small Hydro Stranik Outside the basin 1.60
Albania Small Hydro Zall Tore Outside the basin 2.60
Albania Small Hydro Belesova 1 (Lumas Berat) Outside the basin 0.15
Albania Small Hydro Belesova 2 Outside the basin 0.28
Albania Small Hydro Fterra 1 Outside the basin 1.08
Albania Small Hydro Fterra 2 Outside the basin 2.00
Albania Small Hydro Verba 1 Outside the basin 2.00
Albania Small Hydro Verba 2 Outside the basin 3.00
Albania Small Hydro Çarshove Outside the basin 0.07
Albania Small Hydro Rehove Outside the basin 0.10
Albania Small Hydro Nice Outside the basin 0.60
Albania Small Hydro Nishove 1 Outside the basin 1.11
Albania Small Hydro Meshanik Outside the basin 1.65
Albania Small Hydro Guve Outside the basin n/a
Albania Small Hydro Bence Outside the basin 5.4
Albania Small Hydro Tepelene Outside the basin n/a
Albania Small Hydro Ujanik 2 Outside the basin 1.9
Albania Small Hydro Kaskada e Luses 1-7 Outside the basin 6.8
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
21
Table B5
List of small hydropower plants in North Macedonia
Country Type Plant Location Capacity (MW)
North Macedonia Small Hydro Matka (New) Outside the basin 9.6
North Macedonia Small Hydro Pena Outside the basin 3.3
North Macedonia Small Hydro Zrnovci Outside the basin 1.6
North Macedonia Small Hydro Pesoˇ
cani Outside the basin 2.88
North Macedonia Small Hydro Sapunˇ
cica Outside the basin 2.9
North Macedonia Small Hydro Dosnica Outside the basin 5.1
North Macedonia Small Hydro Popova ˇ
Sapka cascade Outside the basin 4.8
North Macedonia Small Hydro Turija Outside the basin 2
North Macedonia Small Hydro Babuna Outside the basin 0.64
North Macedonia Small Hydro Belica 1 cascade Outside the basin 0.25
North Macedonia Small Hydro Belica 2 cascade Outside the basin 1
North Macedonia Small Hydro Lukar Kavadarci Outside the basin 1.3
North Macedonia Small Hydro Filternica Outside the basin 0.38
North Macedonia Small Hydro Streˇ
zevo 1 Outside the basin 2.4
North Macedonia Small Hydro Bioloˇ
ski Outside the basin 0.13
North Macedonia Small Hydro Dovlednjik Outside the basin 0.46
North Macedonia Small Hydro Filternica Outside the basin 0.38
North Macedonia Small Hydro MHE (Ohrid 1) Outside the basin 0.12
North Macedonia Small Hydro MHE (Ohrid 2) Outside the basin 0.32
North Macedonia Small Hydro MHE (Ohrid 3) Outside the basin 0.23
North Macedonia Small Hydro MHE (Belica 1) Outside the basin 1
North Macedonia Small Hydro MHE (Belica 2) Outside the basin 1
North Macedonia Small Hydro DIKOM Outside the basin 0.03
North Macedonia Small Hydro Hidroenergo Projekt Vodovod Bitola Outside the basin 0.4
North Macedonia Small Hydro Studencica Outside the basin 0.6
North Macedonia Small Hydro Krkljanska reka Outside the basin 0.38
North Macedonia Small Hydro Slatino Outside the basin 0.56
North Macedonia Small Hydro Brbushnica Outside the basin 0.58
North Macedonia Small Hydro Kranska reka Outside the basin 0.58
North Macedonia Small Hydro Kriva reka 2 Outside the basin 0.58
North Macedonia Small Hydro Brajcino 1 Outside the basin 0.7
North Macedonia Small Hydro Kamenicka reka Outside the basin 1.2
North Macedonia Small Hydro Ljubanska Outside the basin 0.23
North Macedonia Small Hydro Pesocan 393 Outside the basin 0.99
North Macedonia Small Hydro Selecka reka, s. Burinec Outside the basin 1.72
North Macedonia Small Hydro Zelengrad Outside the basin 0.13
North Macedonia Small Hydro Brestjanska reka Outside the basin 0.67
North Macedonia Small Hydro Ratevo Outside the basin 0.4
North Macedonia Small Hydro Mini Turija Outside the basin 0.16
North Macedonia Small Hydro Gradecka Outside the basin 0.92
North Macedonia Small Hydro Tresonce Outside the basin 1.98
North Macedonia Small Hydro Pesocan 392 Outside the basin 1.13
North Macedonia Small Hydro Golemaca 259 Outside the basin 0.42
North Macedonia Small Hydro Mala reka Outside the basin 0.27
North Macedonia Small Hydro Dobrenoec Outside the basin 0.48
North Macedonia Small Hydro Bistrica 97 Outside the basin 2.64
North Macedonia Small Hydro Bistrica 98 Outside the basin 3.2
North Macedonia Small Hydro Brajcino 2 Outside the basin 1.47
North Macedonia Small Hydro Galicka reka 3 Outside the basin 1.28
North Macedonia Small Hydro Esterec 372 Outside the basin 0.38
North Macedonia Small Hydro Bistrica 99 Outside the basin 3.28
North Macedonia Small Hydro Eksploatacionen minimum Outside the basin 0.32
North Macedonia Small Hydro Brza voda 3 95 Outside the basin 0.72
North Macedonia Small Hydro Toplec Outside the basin 0.33
North Macedonia Small Hydro Brza voda 2 94 Outside the basin 0.96
North Macedonia Small Hydro Brza voda 1 96 Outside the basin 0.96
North Macedonia Small Hydro Patiska reka 146 Outside the basin 0.71
North Macedonia Small Hydro Golemo Ilino 257 Outside the basin 0.46
North Macedonia Small Hydro Baciska reka 2 28 Outside the basin 1.17
North Macedonia Small Hydro Kusnica 256 Outside the basin 0.25
North Macedonia Small Hydro Kamena reka 125 Outside the basin 2.4
North Macedonia Small Hydro Konjarka 236 Outside the basin 1
North Macedonia Small Hydro Kriva reka 1179 -1 Outside the basin 0.54
North Macedonia Small Hydro Kriva reka 2179 -2 Outside the basin 0.99
North Macedonia Small Hydro Kalin Kamen 1 Outside the basin 0.25
North Macedonia Small Hydro Kalin Kamen 2 Outside the basin 0.32
North Macedonia Small Hydro Bosava 1 Outside the basin 2.88
North Macedonia Small Hydro Bosava 2 Outside the basin 2.88
(continued on next page)
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
22
Table B5 (continued )
Country Type Plant Location Capacity (MW)
North Macedonia Small Hydro Bosava 3 Outside the basin 1.92
North Macedonia Small Hydro Bosava 4 Outside the basin 1.92
North Macedonia Small Hydro Bosava 5 Outside the basin 1.44
North Macedonia Small Hydro Stanecka reka Outside the basin 0.14
North Macedonia Small Hydro Kazani 208 Outside the basin 1.06
North Macedonia Small Hydro Vejacka reka 93 Outside the basin 1.31
North Macedonia Small Hydro Jablanica 399 Outside the basin 3.28
Table B6
List of small hydropower plants in Montenegro
Country Type Plant Location Capacity (MW)
Montenegro Small Hydro sHPP Rijeka Crnojevi´
ca Outside the basin 0.5
Montenegro Small Hydro sHPP Lijeva rijeka Outside the basin 0.05
Montenegro Small Hydro sHPP Podgor Outside the basin 0.4
Montenegro Small Hydro sHPP ˇ
Savnik Outside the basin 0.2
Montenegro Small Hydro sHPP Glava Zete Outside the basin 5.36
Montenegro Small Hydro sHPP Slap Zete Outside the basin 1.2
Montenegro Small Hydro sHPP Rijeka Muˇ
sovi´
ca Inside the basin 1.3
Montenegro Small Hydro sHPP Jezerˇ
stica Inside the basin 0.844
Montenegro Small Hydro sHPP Bistrica Inside the basin 5.6
Montenegro Small Hydro sHPP Rmuˇ
s Inside the basin 0.474
Montenegro Small Hydro sHPP Spaljevi´
ci 1 Inside the basin 0.65
Montenegro Small Hydro sHPP Orah Inside the basin 0.954
Montenegro Small Hydro sHPP ˇ
Sekular Inside the basin 1.665
Montenegro Small Hydro sHPP Vrelo Inside the basin 0.615
Montenegro Small Hydro sHPP Bradavec Inside the basin 0.954
Montenegro Small Hydro sHPP Piˇ
sevska rijeka Inside the basin 1.08
Montenegro Small Hydro sHPP Jara Inside the basin 4.568
Montenegro Small Hydro sHPP Babino polje Inside the basin 2.214
Montenegro Small Hydro sHPP Bistrica Majstorovina Inside the basin 3.6
Montenegro Small Hydro sHPP ˇ
Seremet Potok Inside the basin 0.792
Table B7
List of renewable energy projects in Albania
Country Technology Plant Operator Capacity (MW) Started Operation Location
Albania Hydro Dam Skavica KESH 196 2025 Inside Drin
River Basin
Albania Hydro Dam
(Francis)
Moglice Statkraft 184 (2 ×92 MW) 2019 (shifted to
2021)
Outside Drin
River Basin
Albania Hydro Dam
(Francis)
Kalivac Ayen Enerji & Fusha 111 2020 (shifted to
2022)
Outside Drin
River Basin
Albania Hydro Dam Shala 83.5 2021 (shifted to
2023)
Inside Drin
River Basin
Albania Hydro Pocem (stopped)
Albania Solar PV Karavasta Voltalia 140 assumed by 2025 Outside Drin
River Basin
Albania Solar PV oating PV Statkraft 2 assumed by 2025 Outside Drin
River Basin
Albania Solar PV Fier Solar 2.5 assumed by 2025 Outside Drin
River Basin
Albania Solar PV solar power plant at Vau i Dej¨
es KESH 5.1 assumed by 2025 Inside Drin
River Basin
Albania Solar PV oating solar PV at Vau i Dej¨
es KESH 12.9 assumed by 2028 Inside Drin
River Basin
Albania Solar PV Durres (Spitalle solar park) French company Voltalia 100 assumed by 2028 Outside Drin
River Basin
Albania Solar PV Blue 1 and Blue 2 by Blessed Investment and Matrix
Konstruksion, registered in Albania
100 assumed by 2030 Outside Drin
River Basin
(continued on next page)
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
23
Table B7 (continued )
Country Technology Plant Operator Capacity (MW) Started Operation Location
Albania Solar PV Additional capacity an annual increase of 10%
allowed from 2026
onwards.
2026 onwards Outside Drin
River Basin
Albania Wind Onshore three projects have been
authorised for construction with a total capacity
of 9 MW which qualies for FiT support (MIE, 2019).
9 assumed by 2025 Outside Drin
River Basin
Albania Wind Onshore At the end of 2020, a 150 (130)
MW wind tender was launched
(MIE 2019)
150 assumed by 2027 Outside Drin
River Basin
Albania Wind Onshore WPP in Tepelena region Alb-Building 12 Building Permit
issued (assumed by
2023)
Outside Drin
River Basin
Albania Wind Onshore Additional capacity An annual increase of 10%
allowed from 2030
onwards.
2030 onwards Outside Drin
River Basin
Table B8
List of renewable energy projects in North Macedonia
Country Technology Power plant option Operator Installed capacity (MW) Start year (potential) Inside/outside Drin
River Basin
North
Macedonia
Large hydro Tenovo-Kozjak
project
Project increasing supply of existing Kozjak,
Malka & Sv. Petka HPP
2030 Outside Drin River
Basin
North
Macedonia
Large hydro Globocica II 20 2035 Inside Drin River
Basin
North
Macedonia
Large hydro Veles 96 2030 Outside Drin River
Basin
North
Macedonia
Large hydro Cebren (or
Chebren)
458 2029 Outside Drin River
Basin
North
Macedonia
Large hydro Gradec 75.34 2030 Outside Drin River
Basin
North
Macedonia
Large hydro Galiste 77.9 2035 Outside Drin River
Basin
North
Macedonia
Small hydro Vardar Valley
SHPPs 1
45 2025 Outside Drin River
Basin
North
Macedonia
Small hydro Vardar Valley
SHPPs 2
153 2030 Outside Drin River
Basin
North
Macedonia
Small hydro Small hydro Max. 135-160 2019 Outside Drin River
Basin
North
Macedonia
Biogas Biogas with FiT 18 2020 Outside Drin River
Basin
North
Macedonia
Biogas Biogas without FiT 10 2025 Outside Drin River
Basin
North
Macedonia
Biomass PP or CHP on
biomass
12.5–15 2020 Outside Drin River
Basin
North
Macedonia
Wind
Onshore
Bogdanci Phase I ELEM 36.8 2014 Outside Drin River
Basin
North
Macedonia
Wind
Onshore
Bogdanci Phase II ELEM 13.8 Proposed Outside Drin River
Basin
North
Macedonia
Wind
Onshore
Miravci Phase I ELEM 14 Preliminary Desing Outside Drin River
Basin
North
Macedonia
Wind
Onshore
Miravci Phase II ELEM 36 Preliminary Desing Outside Drin River
Basin
North
Macedonia
Wind
Onshore
Bogoslovec Thor Impex D.O.
O.E.L
33 Building Permit issued Outside Drin River
Basin
North
Macedonia
Wind Wind with FiT 64 2021 Outside Drin River
Basin
North
Macedonia
Wind Wind with FiP 50 2022 Outside Drin River
Basin
North
Macedonia
Wind Wind without FiP
or FiT
100–500 2025 Outside Drin River
Basin
North
Macedonia
PV Oslomej PV 100 assume 50 MW in 2025//50
MW in 2027
Outside Drin River
Basin
North
Macedonia
PV PV with FiP 200 2020 Outside Drin River
Basin
North
Macedonia
PV PV without FiP 400–800 2020 Outside Drin River
Basin
North
Macedonia
PV PV rooftop 250–400 2019 Outside Drin River
Basin
North
Macedonia
PV Voishanci PV 1.48 2020 (opr) Outside Drin River
Basin
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
24
Table B9
List of renewable energy projects in Montenegro
Country Technology Plant River Capacity (MW) Started Operation Location
Montenegro Dam (Francis) HPP Andrijevo Moraˇ
ca 127.4 (2 ×63.7 MW) no date. Assumed by
2035
Inside Drin
River Basin
Montenegro Dam
(gravitational)
Francis
HPP Raslovi´
ci Moraˇ
ca 37 (2 ×18.5 MW) no date. Assumed by
2035
Inside Drin
River Basin
Montenegro Dam
(gravitational)
Francis
HPP Milunovi´
ci Moraˇ
ca 37 (2 ×18.5 MW) no date. Assumed by
2035
Inside Drin
River Basin
Montenegro Dam
(gravitational)
Francis
HPP Zlatica Moraˇ
ca 37 (2 ×18.5 MW) no date. Assumed by
2035
Inside Drin
River Basin
Montenegro Arch Dam HPP Komarnica Komarnica 168 2029 (EPCG) Outside Drin
River Basin
Montenegro Dam HPP Kruˇ
sevo V1 82 MW no date. Assumed by
2035
Outside Drin
River Basin
Montenegro Dam HPP Kruˇ
sevo V2 90–100 MW no date. Assumed by
2038
Outside Drin
River Basin
Montenegro Dam HPP PERU´
CICA -
UNIT 8
58.5 MW 2025 Outside Drin
River Basin
Montenegro Small Hydro HPP Ibriˇ
stica Moraˇ
ca 3.1 not considered Outside Drin
Basin
Montenegro Small Hydro ˇ
Stitariˇ
cka Tara 0.9 not considered Outside Drin
Basin
Montenegro Wind WPP Gvozd N/A 54.6 2023 Inside Drin
River Basin
Montenegro Solar PV Solar PV plant -
Briska Gora
N/A phase 1: 50 MW by 2022, phase 2 increases to 250 by 2024.
(some plans mention it will increase to 262 MW)
2022 (50 MW)/2024
(200 MW)
Outside Drin
River Basin
Table B10
List of renewable energy projects in Kosovo
Country Technology Plant Units Capacity
(MW)
Expected year of
commissioning
Inside/Outside Drin River
Basin
Kosovo Wind WPP Zatric I, II – 64.8 – Inside Drin River Basin
Kosovo Wind WPP Bajgora
Consists of 3 wind farms (Selac I, II,
III)
27 turbines x 3,83
MW
105 2021 Outside Drin River Basin
Kosovo Wind WPP Koznice – 34.5 2022 Outside Drin River Basin
Kosovo Wind Budakova 46 2026 Outside Drin River Basin
Kosovo Wind WPP Cicavica 17 turbines x 3 MW 51 – Outside Drin River Basin
Kosovo Wind wind farms PE Kamenica-1 and 2 2 * 34.8 69.6 2024 Outside Drin River Basin
Kosovo Solar PS Kamenica-3 30 30 2024 Outside Drin River Basin
Kosovo Hydro dam HPP Lepenc I 10 2020 Outside Drin River Basin
Appendix C
Fig. C1. Total installed capacity in the Drin basin countries by technology.
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
25
Fig. C2. Total installed electricity generation capacity in Albania.
Fig. C3. Total installed capacity in North Macedonia by technology.
Fig. C4. Final electricity demand (TWh) projections for the Drin countries from 2020 to 2050.
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
26
Fig. C5. Daily load prole for Albania, Kosovo and North Macedonia.
Fig. C6. Average capacity factor for solar and wind technologies in each country [57,58].
Table C1
Average weekly capacity factor for solar technology in each country [57,58].
Week Albania North Macedonia Montenegro Kosovo
Inside DRB Outside DRB Inside DRB Outside DRB Inside DRB Outside DRB Inside DRB Outside DRB
1 4.6% 15.2% 4.6% 11.1% 16.7% 6.3% 4.8% 6.6%
2 7.1% 15.7% 5.8% 10.9% 12.4% 6.5% 5.9% 5.9%
3 11.1% 15.1% 7.2% 15.8% 15.1% 11.4% 8.6% 13.0%
4 8.5% 6.5% 7.2% 6.3% 6.8% 7.8% 8.6% 9.2%
5 6.3% 7.0% 8.6% 6.6% 11.1% 5.3% 6.2% 6.6%
6 5.9% 9.4% 8.2% 7.6% 6.1% 6.7% 5.2% 6.8%
7 16.5% 19.2% 20.8% 19.4% 23.3% 11.7% 10.1% 12.0%
8 21.0% 18.4% 21.0% 19.5% 21.6% 18.3% 14.6% 19.9%
9 11.7% 9.1% 13.0% 12.7% 10.2% 8.8% 13.7% 11.6%
10 8.6% 14.3% 11.5% 10.0% 16.2% 6.7% 7.6% 10.4%
11 13.1% 18.4% 14.1% 8.3% 23.8% 7.7% 6.5% 8.6%
12 17.2% 17.4% 18.6% 18.0% 20.7% 14.5% 12.2% 19.3%
13 12.1% 11.0% 7.7% 5.5% 12.5% 11.4% 9.5% 10.4%
14 15.8% 18.1% 17.0% 18.1% 15.9% 10.5% 11.6% 15.6%
15 15.3% 21.2% 20.4% 17.2% 22.1% 8.7% 8.9% 15.2%
16 21.9% 22.7% 22.8% 22.1% 22.7% 14.8% 18.5% 20.1%
17 24.7% 23.6% 25.1% 23.8% 23.7% 24.3% 23.3% 26.9%
18 16.3% 18.1% 18.8% 19.8% 18.9% 17.8% 16.9% 20.1%
19 22.9% 23.2% 22.5% 21.4% 24.3% 19.5% 23.7% 23.6%
20 23.5% 25.0% 24.9% 21.7% 26.5% 23.5% 19.7% 21.2%
21 19.4% 18.3% 20.0% 20.2% 20.6% 14.9% 19.9% 19.6%
22 21.2% 22.0% 21.2% 21.2% 25.5% 16.2% 19.3% 20.0%
23 24.4% 22.4% 24.6% 25.3% 27.0% 26.2% 24.8% 26.9%
24 23.4% 22.8% 24.2% 24.1% 26.8% 24.6% 24.1% 26.9%
25 19.9% 21.9% 21.5% 17.9% 24.2% 12.0% 16.6% 16.0%
26 24.4% 23.8% 23.0% 23.6% 23.3% 17.1% 19.5% 18.6%
27 23.3% 23.8% 21.9% 20.4% 26.9% 23.2% 20.8% 21.9%
28 24.9% 24.5% 26.0% 25.9% 26.6% 24.5% 24.9% 25.8%
29 26.3% 24.3% 26.9% 26.2% 26.8% 23.6% 26.4% 25.7%
30 22.5% 22.6% 24.4% 24.5% 25.3% 21.0% 24.2% 25.9%
(continued on next page)
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
27
Table C1 (continued )
Week Albania North Macedonia Montenegro Kosovo
Inside DRB Outside DRB Inside DRB Outside DRB Inside DRB Outside DRB Inside DRB Outside DRB
31 25.6% 23.8% 25.6% 23.8% 26.6% 21.6% 25.5% 25.3%
32 21.2% 20.3% 23.0% 19.9% 23.8% 22.5% 21.6% 23.0%
33 20.2% 20.1% 23.6% 21.7% 22.8% 16.9% 20.6% 22.6%
34 20.1% 22.7% 22.1% 18.8% 24.8% 12.8% 19.5% 16.6%
35 21.9% 21.2% 23.5% 22.3% 23.5% 21.3% 21.8% 22.8%
36 23.4% 22.3% 25.5% 24.2% 23.0% 21.1% 23.6% 21.6%
37 17.8% 17.6% 19.4% 18.0% 21.2% 16.9% 16.4% 18.3%
38 24.0% 21.4% 24.5% 24.0% 21.9% 21.5% 23.2% 22.6%
39 11.2% 14.5% 11.9% 9.5% 17.4% 12.9% 9.4% 9.8%
40 16.2% 17.6% 17.1% 16.1% 21.5% 10.8% 11.2% 14.2%
41 9.7% 10.4% 10.6% 9.4% 8.7% 6.7% 8.8% 8.6%
42 12.8% 11.9% 13.0% 13.9% 13.1% 8.6% 12.8% 13.2%
43 11.2% 13.2% 12.8% 12.3% 13.2% 8.0% 8.9% 9.2%
44 18.9% 15.2% 19.0% 20.1% 20.1% 19.3% 19.2% 18.4%
45 21.2% 19.2% 22.1% 21.0% 21.1% 20.7% 20.3% 19.7%
46 19.3% 16.3% 19.8% 17.6% 16.7% 17.7% 18.2% 17.5%
47 12.4% 11.3% 13.9% 13.9% 11.0% 12.6% 13.8% 12.7%
48 5.8% 6.8% 8.4% 7.2% 5.9% 3.3% 4.1% 3.4%
49 17.5% 16.7% 17.3% 14.8% 17.0% 8.4% 14.3% 14.8%
50 12.4% 14.9% 14.0% 12.9% 15.9% 10.4% 13.0% 10.0%
51 15.8% 13.7% 15.9% 16.4% 17.0% 11.4% 11.0% 8.7%
52 17.3% 16.0% 18.9% 18.5% 18.2% 17.4% 18.6% 16.6%
Table C2
Average weekly capacity factor for Wind technology in each country [57,58].
Week Albania North Macedonia Montenegro Kosovo
Inside DRB Outside DRB Inside DRB Outside DRB Inside DRB Outside DRB Inside DRB Outside DRB
1 0.03% 26.6% 26.0% 31.2% 39.2% 36.2% 24.2% 28.8%
2 0.01% 24.3% 17.6% 21.6% 29.3% 22.4% 9.1% 16.5%
3 0.01% 21.0% 8.4% 19.6% 11.7% 20.8% 10.6% 23.2%
4 0.02% 26.4% 14.5% 24.4% 31.1% 25.6% 13.2% 23.8%
5 0.02% 50.8% 27.0% 24.5% 34.9% 43.1% 18.6% 34.3%
6 0.02% 16.4% 22.6% 26.9% 35.6% 22.4% 16.9% 19.1%
7 0.04% 42.0% 21.9% 24.4% 49.4% 47.8% 33.8% 37.8%
8 0.03% 21.4% 18.6% 21.3% 31.7% 31.3% 23.5% 26.1%
9 0.02% 17.4% 9.4% 12.0% 26.0% 28.7% 19.6% 27.8%
10 0.01% 22.1% 34.7% 34.8% 9.0% 28.9% 18.4% 26.4%
11 0.02% 22.5% 7.1% 7.2% 20.6% 24.1% 19.4% 24.0%
12 0.01% 13.1% 10.3% 10.4% 30.6% 20.4% 12.5% 21.2%
13 0.03% 13.7% 18.7% 15.0% 40.9% 33.7% 24.0% 28.8%
14 0.01% 24.2% 13.8% 21.5% 13.7% 20.4% 7.3% 20.1%
15 0.01% 13.0% 14.6% 24.8% 6.7% 6.8% 5.4% 10.3%
16 0.01% 10.1% 11.3% 15.2% 13.8% 17.1% 9.2% 17.9%
17 0.01% 28.3% 8.1% 17.7% 10.6% 22.5% 11.0% 23.7%
18 0.02% 20.2% 12.5% 14.0% 12.3% 16.2% 15.4% 23.6%
19 0.01% 25.0% 7.8% 8.5% 11.0% 17.0% 9.7% 19.2%
20 0.01% 13.4% 16.6% 12.9% 8.4% 12.8% 9.9% 14.9%
21 0.01% 10.4% 11.9% 8.9% 7.5% 13.0% 10.5% 15.1%
22 0.01% 22.0% 7.4% 12.9% 14.0% 17.0% 9.5% 17.4%
23 0.01% 11.1% 9.1% 14.7% 5.3% 11.4% 7.5% 10.9%
24 0.00% 7.9% 5.2% 8.7% 6.7% 6.7% 4.5% 8.1%
25 0.01% 9.3% 9.8% 13.6% 6.5% 6.6% 3.9% 5.9%
26 0.02% 16.2% 8.9% 15.6% 29.9% 24.7% 18.9% 24.4%
27 0.01% 9.2% 6.3% 12.5% 7.9% 7.0% 5.8% 10.0%
28 0.01% 14.7% 4.3% 13.7% 10.8% 10.1% 7.8% 11.3%
29 0.01% 10.1% 7.5% 11.7% 10.5% 8.8% 5.6% 9.6%
30 0.01% 20.1% 5.2% 10.2% 14.7% 17.8% 10.4% 15.4%
31 0.01% 15.6% 8.3% 7.9% 8.0% 9.4% 8.8% 11.8%
32 0.00% 10.7% 8.3% 14.4% 5.7% 5.7% 4.1% 7.5%
33 0.01% 15.0% 5.2% 7.7% 10.4% 9.0% 6.6% 12.2%
34 0.00% 6.8% 9.2% 8.0% 10.9% 10.5% 6.0% 11.5%
(continued on next page)
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
28
Table C2 (continued )
Week Albania North Macedonia Montenegro Kosovo
Inside DRB Outside DRB Inside DRB Outside DRB Inside DRB Outside DRB Inside DRB Outside DRB
35 0.01% 6.1% 4.1% 6.0% 9.1% 6.9% 5.2% 10.5%
36 0.00% 7.6% 13.5% 9.8% 10.7% 9.4% 5.3% 12.2%
37 0.01% 16.3% 10.0% 17.0% 17.4% 19.3% 13.2% 21.0%
38 0.01% 9.3% 2.9% 6.1% 17.3% 13.9% 9.4% 14.0%
39 0.01% 14.2% 10.8% 10.8% 6.9% 9.7% 7.1% 8.8%
40 0.01% 12.9% 8.8% 9.8% 10.5% 15.0% 8.2% 13.6%
41 0.01% 9.6% 12.4% 16.3% 18.1% 18.1% 9.8% 15.6%
42 0.00% 3.7% 7.8% 13.4% 2.7% 5.9% 1.9% 5.3%
43 0.00% 5.6% 23.0% 22.3% 6.4% 4.7% 2.5% 5.6%
44 0.01% 22.4% 25.4% 9.8% 22.5% 23.0% 11.4% 20.6%
45 0.03% 45.7% 5.0% 8.2% 36.2% 43.0% 25.9% 38.2%
46 0.02% 50.4% 4.0% 15.8% 31.0% 41.6% 15.5% 28.5%
47 0.01% 25.7% 23.8% 23.7% 13.4% 17.7% 8.0% 14.7%
48 0.02% 20.3% 11.7% 16.2% 22.1% 21.5% 18.2% 24.1%
49 0.01% 16.5% 5.1% 9.4% 18.9% 15.3% 8.1% 12.8%
50 0.01% 21.1% 6.3% 10.0% 22.5% 23.3% 11.9% 17.4%
51 0.02% 47.7% 7.4% 6.9% 31.6% 43.9% 24.4% 34.7%
52 0.02% 42.0% 4.9% 7.7% 29.5% 44.6% 25.0% 44.1%
Table C3
Climate model data used in the hydrological model E-HYPE [46].
Hydrological Model RCP GCM (4) RCM (4) Period (input dataset) Period (adjusted in the model) Institute
E-HYPE v3.1.2 2.6 EC-EARTH RCA4 1970–2100 2020–2050 SMHI
MPI-ESM-LR REMO2009 1951–2100 2020–2050 CSC
4.5 EC-EARTH RCA4 1970–2100 2020–2050 SMHI
EC-EARTH RACMo22 E 1951–2100 2020–2050 KNMI
HadGEM2-ES RCA4 1970–2098 2020–2050 SMHI
MPI-ESM-LR REMO2009 1951–2100 2020–2050 CSC
CM5A WRF33 1971–2100 2020–2050 IPSL
8.5 EC-EARTH RCA4 1970–2100 2020–2050 SMHI
EC-EARTH RACMO22E 1951–2100 2020–2050 KNMI
HadGEM2-ES RCA4 1970–2098 2020–2050 SMHI
MPI-ESM-LR REMO2009 1951–2100 2020–2050 CSC
Fig. C7. The Elevation-Volume curve of Spilje Reservoir.
Table C4
Changes in the water level (masl) in Spilje reservoir
month hist_level (m) level (m) +5% buffer level (m) +20% buffer Diff in m (+5%) Diff in m (+20%)
1 569 566.3 564.7 2.7 4.3
2 566 564 562.2 2 3.8
3 567 564.7 563 2.3 4
4 570 567.1 565.7 2.9 4.3
(continued on next page)
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
29
Table C4 (continued )
month hist_level (m) level (m) +5% buffer level (m) +20% buffer Diff in m (+5%) Diff in m (+20%)
5 576 576 576 0 0
6 578 578 578 0 0
7 576 576 576 0 0
8 575 575 575 0 0
9 572 572 572 0 0
10 570 567.1 565.7 2.9 4.3
11 568 565.5 563.9 2.5 4.1
12 569 566.3 564.7 2.7 4.3
Fig. C8. The Elevation-Volume curve of Fierza Reservoir.
Table C5
Changes in the water level (masl) in Fierza reservoir.
month hist_level (m) 5%_level (m) 20%_level (m) Diff in m (5% level) Diff in m (20% level)
1 279 278.5 275 0.5 4.2
2 276 274.9 270 1.1 5.8
3 280 279.7 276 0.3 3.8
4 285 285.3 283 −0.3 1.7
5 290 290 290 0 0
6 296 296 296 0 0
7 293 293 293 0 0
8 286 286 286 0 0
9 275 275 275 0 0
10 272 270.1 264 1.9 7.8
11 276 274.9 270 1.1 5.8
12 279 278.5 275 0.5 4.2
Comparison of betweem modelled and historical generation for the cascade of hydropower plants in the Drin River Basin
The validation of the OSeMOSYS model used in this study was conducted based on historical electricity generation data (in GWh) obtained from
the utilitites in North Macedonia and Albania. For the upstream plants, Globocia and Spilje, data from the North Macedonian utility (ELEM) were
obtained for the period 1999–2015 [80]. For the downstream plants, Fierza, Koman and Vau i Dej¨
es, data from the Albanian utility (KESH) were
obtained for the period 2004–2014 [81]. In both cases, minimum, maximum and average generation data were used to compare the modelled
generation for each plant under the reference scenarios (REF). As shown in Figure C 9, the modelled electricity generation correlate to the average
historical generation in each of the hydropower plants.
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
30
Fig. C9. Comparison between modelled output (reference scenario) and the historical output (min, max and avg) for the hydropower plants along the cascade in the
Drin River Basin.
References
[1] International Hydropower Association. IHA, Hydropower Status Report - Sector
Trends and Insights’, ‘2021. London, UK, 2021. Accessed: Mar. 10, 2022. [Online].
Available: https://assets-global.website-les.com/5f749e4b9399c80b5e421384/6
0c37321987070812596e26a_IHA20212405-status-report-02_LR.pdf.
[2] International Energy Agency. IEA, Renewables 2021 - Analysis and Forecast to
2026, Paris, France, 2021 [Online]. Available: https://iea.blob.core.windows.net
/assets/5ae32253-7409-4f9a-a91d-1493ffb9777a/Renewables2021-Analysisan
dforecastto2026.pdf.
[3] International Energy Agency, IEA, ‘Net Zero by 2050 - A Roadmap for the Global
Energy Sector’, Paris, France, 2021 [Online]. Available: https://iea.blob.core.win
dows.net/assets/deebef5d-0c34-4539-9d0c-10b13d840027/NetZeroby2050-AR
oadmapfortheGlobalEnergySector_CORR.pdf.
[4] L. Berga, The role of hydropower in climate change mitigation and adaptation: a
review, Engineering 2 (3) (2016) 313–318, https://doi.org/10.1016/J.
ENG.2016.03.004.
[5] International Energy Agency, IEA, ‘Global Energy Review 2020’, Paris, France,
2020 [Online]. Available: https://iea.blob.core.windows.net/assets/7e802f6
a-0b30-4714-abb1-46f21a7a9530/Global_Energy_Review_2020.pdf.
[6] The Intergovernmental Panel on Climate Change. IPCC, ‘IPCC WGII Sixth
Assessment Report - Technical Summary’, The Intergovernmental Panel on Climate
Change (2021). Accessed: Mar. 10, 2022. [Online]. Available: https://report.ipcc.
ch/ar6wg2/pdf/IPCC_AR6_WGII_FinalDraft_TechnicalSummary.pdf.
[7] Z. Yihdego, J. Gibson, Implementing International Watercourses Law through the
WEF Nexus and SDGs: an Integrated Approach Illustrated in the Zambezi River
Basin, Koninklijke Brill NV, The Netherlands, 2020 [Online]. Available: https
://click.endnote.com/viewer?doi=10.1163%2F9789004444713&token=WzMzN
DM5MTMsIjEwLjExNjMvOTc4OTAwNDQ0NDcxMyJd.XHRJXQjkOcr
ubSYyEqQfjR4EK0U.
[8] C. Llamosas, B.K. Sovacool, Transboundary hydropower in contested contexts:
energy security, capabilities, and justice in comparative perspective, Energy
Strategy Rev. 37 (Sep. 2021) 100698, https://doi.org/10.1016/j.esr.2021.100698.
[9] UNECE, Reconciling Resource Uses in Transboundary Basins: Assessment of the
Water-Food-Energy-Ecosystems Nexus, United Nations Economic Comision for
Europe, 2015 [Online]. Available: http://www.unece.org/index.php?id=41427&L
=0.
[10] C. Llamosas, B.K. Sovacool, The future of hydropower? A systematic review of the
drivers, benets and governance dynamics of transboundary dams, Renew.
Sustain. Energy Rev. 137 (Mar. 2021) 110495, https://doi.org/10.1016/j.
rser.2020.110495.
[11] Y. Almulla, E. Ramos, F. Gardumi, C. Taliotis, A. Lipponen, M. Howells, The Role of
Energy-Water Nexus to Motivate Transboundary Cooperation: an Indicative
Analysis of the Drina River Basin, IJSEPM, 2018, https://doi.org/10.5278/
ijsepm.2018.18.2.
[12] J.M. Gonzalez, et al., Quantifying cooperation benets for new dams in
transboundary water systems without formal operating rules, Front. Environ. Sci. 9
(May 2021), 596612, https://doi.org/10.3389/fenvs.2021.596612.
[13] M. Basheer, et al., Collaborative management of the Grand Ethiopian Renaissance
Dam increases economic benets and resilience, Nat. Commun. 12 (1) (Sep. 2021),
https://doi.org/10.1038/s41467-021-25877-w. Art. no. 1.
[14] C. Skoulikaris, C. Makris, M. Katirtzidou, V. Baltikas, Y. Krestenitis, Assessing the
vulnerability of a deltaic environment due to climate change impact on surface and
coastal waters: the case of nestos river (Greece), Environ. Model. Assess. 26 (4)
(Aug. 2021) 459–486, https://doi.org/10.1007/s10666-020-09746-2.
[15] R. Spalding-Fecher, B. Joyce, H. Winkler, Climate change and hydropower in the
southern african power pool and zambezi River Basin: system-wide impacts and
policy implications, Energy Pol. 103 (Apr. 2017) 84–97, https://doi.org/10.1016/
j.enpol.2016.12.009.
[16] Convention on the Protection and Use of Transboundary Watercourses and
International Lakes, Routledge, 1992, https://doi.org/10.4324/9781315066547-
50.
[17] P. Whalley, D. Faloutsos, ‘Drin Basin - Transboundary Diagnostic Analysis (TDA)’,
Global Water Partnership Mediterranean (GWP-Med), Greece, 2020. Accessed: Jun.
07, 2021. [Online]. Available: https://www.gwp.org/globalassets/global/gwp-
med-les/list-of-programmes/gef-drin-project/drin-docs/tda_nal.pdf.
[18] S. Feger, ‘Interactive Tools for Reproducible Science’, Text.PhDThesis, Ludwig-
Maximilians-Universit¨
at München, 2020. Accessed: Oct. 12, 2022. [Online].
Available: https://edoc.ub.uni-muenchen.de/26940/.
[19] C. Papadaki, E. Dimitriou, river ow alterations caused by intense anthropogenic
uses and future climate variability implications in the balkans, Hydrology 8 (1)
(2021) 7, https://doi.org/10.3390/hydrology8010007.
[20] S. Kovaçi, M. Ifti, Analysis of distributions and dynamics for water side inows in
the Drin River basin, Albania 3 (3) (2015) 6.
[21] Global Water Partnership Mediterranean, GWP-med, ‘Transboundary Diagnostic
Analysis Thematic Report on the Resource Nexus (Phase I of the Water-Food-
Energy-Ecosystems Nexus Assessment of the Drin Basin)’, Athens, 2020. Accessed:
Jul. 15, 2021. [Online]. Available: https://unece.org/DAM/env/water/nexus/Dri
n_TDA_-_Nexus_Thematic_Report_Final.pdf.
[22] Albanian Power Corporation. KESH, ‘Technical regulation of the maximum
discharges for Drin cascade’, Albanian Power Corporation (KESH), Tirana.
[Online]. Available: http://www.kesh.al/en/csr/society/emergency-prepar
edness/.
[23] M. Scoullos, Transboundary IWRM attempts in the mediterranean emphasis on the
Drin River case and the involvement of stakeholders, in: Integrated Water
Resources Management in the Mediterranean Region, Springer, Dordrecht, 2012,
https://doi.org/10.1007/978-94-007-4756-2_1. Accessed: Mar. 14, 2022.
[Online]. Available:.
[24] Energy Regulatory Authority, The Situation of the Power Sector and ERE Activity
during 2019 - Annual Report, Tirana, 2020 [Online]. Available: https://ere.gov.al/
doc/ERE_annual_report_2019_26102020.pdf.
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
31
[25] Energy Regulatory Authority, The Situation of the Power Sector and ERE Activity
during 2018 - Annual Report, Tirana, 2019 [Online]. Available: https://www.ere.
gov.al/doc/Annual_Report_2018.pdf.
[26] I. Thermos, Albania and North Macedonia: The evolution of the electricity system
under the scope of climate change (2019). Accessed: Jul. 13, 2021. [Online].
Available: http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-264262.
[27] Ministry of Economy of N.Macedonia, Strategy for Energy Development of the
Republic of North Macedonia up to 2040 - Final Draft for Public Consultations,
Skopje, Oct. 2019 [Online]. Available: https://economy.gov.
mk/Upload/Documents/Energy%20Development%20Strategy_FINAL%20DRAFT
%20-%20For%20public%20consultations_ENG_29.10.2019(3).pdf.
[28] Global Water Partnership Mediterranean, GWP-med, ‘Thematic Report on the
Resource NEXUS - Phase I of the Water-Food-Energy-Ecosystems Nexus Assessment
of the Drin Basin’, Athens, Greece, 2019. Accessed: May 13, 2022. [Online].
Available: https://www.gwp.org/globalassets/global/gwp-med-files/list-of-prog
rammes/see-nexus/drin-phase-i-nexus-assessment/report-nexus-phase-i-assessme
nt.pdf.
[29] P. Angelidis, M. Kotsikas, N. Kotsovinos, Management of upstream dams and ood
protection of the transboundary river evros/maritza, Water Resour. Manag. 24 (11)
(Sep. 2010) 2467–2484, https://doi.org/10.1007/s11269-009-9563-6.
[30] C. Skoulikaris, Transboundary cooperation through water related EU directives’
implementation process. The case of shared waters between Bulgaria and Greece,
Water Resour. Manag. 35 (14) (Nov. 2021) 4977–4993, https://doi.org/10.1007/
s11269-021-02983-4.
[31] F. Zhushi Etemi, et al., Correlation between physical and chemical parameters of
water and biotic indices: the case study the White Drin River basin, Kosovo,
J. Water Land Dev. 46 (2020) 229–241, 2020, Accessed: Mar. 14, 2022. [Online].
Available: https://journals.pan.pl/dlibra/publication/134585/edition/117620.
[32] The Deutsche Gesellschaft für Internationale Zusammenarbeit GmbH, GIZ,
‘Preliminary Flood Risk Assessment for the Drin Drim – Buna Bojana River Basin’,
2018 [Online]. Available: https://www.giz.de/en/downloads/Report%20%E2%
80%93%20Preliminary%20Flood%20Risk%20Assessment%20for%20the%
20DrinDrim%20%E2%80%93%20BunaBojana%20River%20Basin%20(2018).pdf.
[33] Y. Almulla, V. Sridharan, E. Fejzic, F. Gardumi, Repository of the Drin Nexus
Assessment Model, KTH division of Energy Systems, Jun. 07, 2022. Accessed: Jun.
07, 2022. [Online]. Available: https://github.com/KTH-dESA/Drin_Water_Ener
gy_Nexus_Model.
[34] M. Howells, et al., OSeMOSYS: the open source energy modeling system, Energy
Pol. 39 (10) (Oct. 2011) 5850–5870, https://doi.org/10.1016/j.
enpol.2011.06.033.
[35] OSeMOSYS repository, OSeMOSYS (2021). Accessed: Mar. 16, 2023. [Online].
Available: https://github.com/OSeMOSYS/OSeMOSYS.
[36] F. Gardumi, et al., From the development of an open-source energy modelling tool
to its application and the creation of communities of practice: the example of
OSeMOSYS, Energy Strategy Rev. 20 (Apr. 2018) 209–228, https://doi.org/
10.1016/j.esr.2018.03.005.
[37] KTH, division of Energy Systems Analysis, KTH-dESA, ‘OSeMOSYS
Documentation’, 2018. https://osemosys.readthedocs.io/en/latest/. (Accessed 29
March 2023). accessed.
[38] C. Donnelly, J.C.M. Andersson, B. Arheimer, Using ow signatures and catchment
similarities to evaluate the E-HYPE multi-basin model across Europe, Hydrol. Sci.
J. 61 (2) (Jan. 2016) 255–273, https://doi.org/10.1080/02626667.2015.1027710.
[39] J. Chang, X. Wang, Y. Li, Y. Wang, H. Zhang, Hydropower plant operation rules
optimization response to climate change, Energy 160 (Oct. 2018) 886–897,
https://doi.org/10.1016/j.energy.2018.07.066.
[40] M. Ahmadi, O.B. Haddad, H.A. Lo´
aiciga, Adaptive reservoir operation rules under
climatic change, Water Resour. Manag. 29 (4) (2015) 1247–1266, Mar, https://
doi.org/10.1007/s11269-014-0871-0.
[41] N. Ehsani, C.J. V¨
or¨
osmarty, B.M. Fekete, E.Z. Stakhiv, Reservoir operations under
climate change: storage capacity options to mitigate risk, J. Hydrol. 555 (Dec.
2017) 435–446, https://doi.org/10.1016/j.jhydrol.2017.09.008.
[42] T. Niet, Alternate Storage Code for OSeMOSYS, Feb. 2020. Accessed: Oct. 14, 2022.
[Online]. Available: https://github.com/OSeMOSYS/OSeMOSYS_GNU_MathProg/
releases/tag/AlternateStorageCode_v0.1.
[43] K. Kuling, T. Niet, ‘A Comparison of Two Different Methods for Modelling Storage
with OSeMOSYS.’, Presented at the the 43rd International Association for Energy
Economics Conference, France, Paris, France, Paris, Jun. 2021.
[44] U. department of I.-B. of R. USBR, Reclamation, Managing Water in the West -
Hydroelectric Power, Jul. 2005. Accessed: Feb. 05, 2023. [Online]. Available: https
://www.usbr.gov/power/edu/pamphlet.pdf.
[45] Stockholm Environment Institute. SEI, ‘WEAP (Water Evaluation And Planning)’.
https://www.weap21.org (accessed May 20, 2022).
[46] The Swedish Meteorological and Hydrological Institute, SMHI, ‘Denition of New
Climate Impact Indicators for Water Management, Including Data Requirements -
SWICCA Deliverable 4.1 Version 1’, 2016. Accessed: Mar. 04, 2022. [Online].
Available: http://136.156.132.3/joan/c3s_cds_forms/sis-water-quality-swicca
/denition-new-climate-indicators-for-water-management.pdf.
[47] Y. Li, H. Gao, G.H. Allen, Z. Zhang, Constructing reservoir area–volume–elevation
curve from TanDEM-X DEM data, IEEE J. Sel. Top. Appl. Earth Obs. Rem. Sens. 14
(2021) 2249–2257, https://doi.org/10.1109/jstars.2021.3051103.
[48] I. Todorovic, ‘Albania’s Hydropower Plant Skavica to Cost up to EUR 510 Million’,
Balkan Green Energy News, Jun. 05, 2020 accessed Mar. 08, 2022), https://balkan
greenenergynews.com/albanias-hydropower-plant-skavica-to-cost-up-to-eur-510-
million/.
[49] Energy Regulatory Authority - Albania, State of the Energy Sector and ERE Activity
during the Year 2018 - Annual Report, 2019 [Online]. Available: https://www.ere.
gov.al/doc/Raporti_vjetor_ERE_2018_perfundimtar.pdf.
[50] Government of Montenegro, Draft Decision on the Energy Balance of Montenegro
for 2020, Government of Montenegro, 2019 accessed Mar. 04, 2022), https://www
.gov.me/en/documents/e82c924c-7591-4178-9733-08854e05cfd3.
[51] Energy Regulatory Ofce Kosovo, Statement of Security of Supply for Kosovo
(Electricity, Natural Gas and Oil), Pristina, 2019. Accessed: Jul. 13, 2021. [Online].
Available: http://ero-ks.org/201
9/Publikimet/Deklarate_mbi_Sigurine_e_Furnizimit_ne_Kosove
(energji_elektrike_gaz_natyror_nafte)ZRRE_31_07_2019_eng.pdf.
[52] Regional Centre for Energy Policy Research, REKK, ‘South East Europe Electricity
Roadmap (SEERMAP) - Country Report: Albania 2017’, Regional Centre for Energy
Policy Research (REKK), 2017. Accessed: Mar. 04, 2022. [Online]. Available:
https://rekk.hu/downloads/projects/SEERMAP_CR_ALBANIA_A4_ONLINE.pdf.
[53] Regional Centre for Energy Policy Research, REKK, ‘South East Europe Electricity
Roadmap (SEERMAP) - Former Yugoslavian Republic of Macedonia 2017’,
Regional Centre for Energy Policy Research (REKK), 2017. Accessed: Mar. 04,
2022. [Online]. Available: https://rekk.hu/downloads/projects/SEERMAP_CR_M
ACEDONIA_A4_ONLINE.pdf.
[54] Regional Centre for Energy Policy Research, REKK, ‘South East Europe Electricity
Roadmap (SEERMAP) - Country Report: Montenegro 2017’, Regional Centre for
Energy Policy Research (REKK), 2017. Accessed: Mar. 04, 2022. [Online].
Available: https://rekk.hu/downloads/projects/SEERMAP_CR_MONTENEGRO
_A4_ONLINE.pdf.
[55] Regional Centre for Energy Policy Research, REKK, ‘South East Europe Electricity
Roadmap (SEERMAP) - Country Report: Kosovo 2017’, Regional Centre for Energy
Policy Research (REKK), 2017. Accessed: Mar. 04, 2022. [Online]. Available: http
s://rekk.hu/downloads/projects/SEERMAP_CR_KOSOVO_A4_ONLINE.pdf.
[56] Albanian Power Corporation, KESH, ‘KESH Activity Report 2013-2016’, Tirana,
2016. Accessed: Jul. 12, 2021. [Online]. Available: http://www.kesh.al/wp-cont
ent/uploads/2020/05/KESH-Activity-Report-2016.pdf.
[57] S. Pfenninger, I. Staffell, Long-term patterns of European PV output using 30 years
of validated hourly reanalysis and satellite data, Energy 114 (Nov. 2016)
1251–1265, https://doi.org/10.1016/j.energy.2016.08.060.
[58] I. Staffell, S. Pfenninger, Using bias-corrected reanalysis to simulate current and
future wind power output, Energy 114 (Nov. 2016) 1224–1239, https://doi.org/
10.1016/j.energy.2016.08.068.
[59] V. Spasic, EUR 300 Million Investments in Renewable Energy Projects in Albania,
Balkan Green Energy News, 2017 accessed Mar. 18, 2022, https://balkangreenener
gynews.com/eur-300-million-investments-in-renewable-energy-projects-in-albania
/.
[60] IEA-ETSAP, Coal-Fired Power, IEA-ETSAP, Apr. 2010 [Online]. Available:
https://www.iea-etsap.org/E-TechDS/PDF/E01-coal-red-power-GS-AD-gct_
FINAL.pdf.
[61] IEA-ETSAP, Gas-red Power, IEA-ETSAP, Apr. 2010 [Online]. Available: https
://iea-etsap.org/E-TechDS/PDF/E02-gas_red_power-GS-AD-gct_FINAL.pdf.
[62] Montenegrin Electric Enterprise, EPCG, ‘Electricity Production and Electro
Energetic Balance’, Elektroprivreda Crne Gore AD Nikˇ
si´
c, 2020 accessed Mar. 18,
2022), https://www.epcg.com/o-nama/proizvodnja-i-elektroenergetski-bilans.
[63] CEE Bankwatch, Kosova e Re lignite power plant, Kosovo, Bankwatch, 2020
accessed Mar. 18, 2021), https://bankwatch.org/project/kosova-e-re-lignite-powe
r-plant-kosovo.
[64] Minister of Economic Development - Kosovo, Energy Strategy of the Republic of
Kosovo, Pristina, 2017. Accessed: Mar. 18, 2021. [Online]. Available: https://rise.
esmap.org/data/les/library/kosovo/Renewable%20Energy/Kosovo_Energy_Str
ategy.pdf.
[65] Kosovo Energy Corporation J.S.C, Termoelektrana “Kosova A” – Korporata
Energjetike e Kosov¨
es, 2020. http://kek-energy.com/kek/sr/termoelektrana-koso
va-a/. (Accessed 16 April 2020). accessed.
[66] Institute of Statistics, INSTAT, ‘Balance of Electric Power, 2012-2019, Tirana, 2020
[Online]. Available: http://www.instat.gov.al/en/themes/environment-and-energ
y/energy/.
[67] V. Sridharan, et al., The impact of climate change on crop production in
Uganda—an integrated systems assessment with water and energy implications,
Water 11 (9) (Sep. 2019), https://doi.org/10.3390/w11091805. Art. no. 9.
[68] Intergovernmental Panel on Climate Change, IPCC, ‘chapter 4: future global
climate: scenario-based projections and near-term information, in: Climate Change
2021: the Physical Science Basis. Contribution of Working Group I to the Sixth
Assessment Report of the Intergovernmental Panel on Climate Change, 2021,
p. 120.
[69] J. Cronin, G. Anandarajah, O. Dessens, Climate change impacts on the energy
system: a review of trends and gaps, Climatic Change 151 (2) (Nov. 2018) 79–93,
https://doi.org/10.1007/s10584-018-2265-4.
[70] P. Seljom, et al., Modelling the effects of climate change on the energy system—a
case study of Norway, Energy Pol. 39 (11) (Nov. 2011) 7310–7321, https://doi.
org/10.1016/j.enpol.2011.08.054.
[71] S.C. Parkinson, N. Djilali, Robust response to hydro-climatic change in electricity
generation planning, Climatic Change 130 (4) (Jun. 2015) 475–489, https://doi.
org/10.1007/s10584-015-1359-5.
[72] K.B. Karnauskas, J.K. Lundquist, L. Zhang, Southward shift of the global wind
energy resource under high carbon dioxide emissions, Nat. Geosci. 11 (1) (Jan.
2018), https://doi.org/10.1038/s41561-017-0029-9. Art. no. 1.
[73] P.E. Carvajal, G. Anandarajah, Y. Mulugetta, O. Dessens, Assessing uncertainty of
climate change impacts on long-term hydropower generation using the CMIP5
Y. Almulla et al.

Energy Strategy Reviews 48 (2023) 101098
32
ensemble—the case of Ecuador, Climatic Change 144 (4) (Oct. 2017) 611–624,
https://doi.org/10.1007/s10584-017-2055-4.
[74] European Union Agency for the Cooperation of Energy Regulators, ACER and
Council of European energy regulators, in: CEER, ‘Annual Report on the Results of
Monitoring the Internal Electricity and Natural Gas Markets in 2019’, Slovenia and
Belgium, Oct. 2020. Accessed: Nov. 03, 2022. [Online]. Available: https://www.
acer.europa.eu/Ofcial_documents/Acts_of_the_Agency/Publication/ACER%
20Market%20Monitoring%20Report%202019%20-%20Energy%20Retail%20and
%20Consumer%20Protection%20Volume.pdf.
[75] A.M. Foley, B.P. ´
O Gallach´
oir, J. Hur, R. Baldick, E.J. McKeogh, A strategic review
of electricity systems models, Energy 35 (12) (2010) 4522–4530.
[76] M. Welsch, et al., Incorporating exibility requirements into long-term energy
system models – a case study on high levels of renewable electricity penetration in
Ireland, Appl. Energy 135 (Dec. 2014) 600–615, https://doi.org/10.1016/j.
apenergy.2014.08.072.
[77] Global Water Partnership Mediterranean, GWP-med, ‘Phase II Nexus Assessment in
the Drin Basin’, Global Water Partnership, Dec. 2020 accessed Mar. 11, 2022), https
://www.gwp.org/en/GWP-Mediterranean/WE-ACT/Programmes-per-theme/Wate
r-Food-Energy-Nexus/seenexus/drin-II/.
[78] Global Water Partnership Mediterranean, GWP-med, ‘Phase II of the Nexus
Assessment in the Drin Basin Moving Ahead’, Global Water Partnership, Sep. 2020
accessed Mar. 11, 2022), https://www.gwp.org/en/GWP-Mediterranean
/WE-ACT/News-List-Page/phase-2-of-nexus-assessment/.
[79] Global Water Partnership Mediterranean, GWP-med, ‘Introduction to the Climate-
Land-Energy-Water (CLEWs) Modelling Framework and its Use in the Nexus
Assessment of the Drin River Basin’, Global Water Partnership, Sep. 2021 accessed
Mar. 11, 2022), https://www.gwp.org/en/GWP-Mediterranean/WE-ACT/News-
List-Page/2021/drin-clews-workshop/.
[80] P. plants of N. M. ELEM, Hydroenergy, Production - Hydroenergy (2015).
https://www.elem.com.mk/wp-content/uploads/2015/hidroenergija_en.html.
[81] Albanian Power Corporation. KESH, ‘Drini Cascade’, Korporata Elektroenergjitike
Shqiptare. https://www.kesh.al/en/asset/drini-cascade/(accessed Mar. 13, 2023).
Y. Almulla et al.