Energy 281 (2023) 128237

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

A baseline long term technoeconomic electricity supply model for highly

regulated fuel importing countries prone to shortages: Case of Jordan
Mahmood Shihadeh Saadeh a, *, Zakariya Dalala b, Osama Saadeh b, Taco Niet c
a
Department of Electrical Engineering, Faculty of Engineering, The Hashemite University, Zarqa, Jordan
b
Energy Engineering Department, German Jordanian University, Jordan
c
School of Sustainable Energy Engineering, Simon Fraser University, Canada

A R T I C L E I N F O A B S T R A C T

Handling editor: X Zhang Jordan, like other fuel-importing countries, is highly susceptible to fuel shortages and price fluctuations due to
global events. Political turmoil in the region has frequently disrupted Jordan’s energy supply, including the
Keywords: recent cessation of natural gas imports from Egypt, which is examined in this study. To address these in­
Energy system analysis terruptions, Jordan signed a 15-year contract to import natural gas from the Leviathan gas field. Jordan has long
Technoeconomic modeling
term power purchase agreements with independent power producers. thus, creating a safe investment envi­
Energy policy
ronment in the electricity production sector. In this research a long-term energy model was developed and
Optimization
OSeMOSYS modeling implemented for Jordan with the Open-Source long-term energy modeling tool OSeMOSYS. The model is used to
assess current policy and compare options moving forward. This study brings into focus the issues facing fuel
importing countries prone to shortages. Contractual aspects were incorporated in the model and investigated.
The need to diversify natural gas import contracts with relatively shorter durations was demonstrated. The
impact of different fuel pricing projections was highlighted and needs to be considered when drawing up such
contracts. It was also demonstrated that the current renewable energy outlook needs modification. potential of
renewable energy needs added consideration from stakeholders and policy makers.

1. Introduction and literature review actual model characteristics such as technology parameters and demand
information. Scenario data can be obtained from historical records,
Technoeconomic modeling is important when planning long term future forecasts, fuel contracts and power purchase agreements.
energy policy. Having a properly and thoroughly developed model for The authors in Ref. [2] describe the Tunisian electricity system and
the target region of interest is vital to ensure that long-term energy developed an OSeMOSYS model that is a long-term optimization model
planning and energy goals are coherent. Many researchers have devel­ (2010–2030). The purpose of the model is to assist policy makers by
oped long-term technoeconomic models to assist regional long term unveiling the potential benefits of increasing renewable energy pene­
energy planning [1–7]. There are few studies that actually describe tration. The model considers the characteristics and challenges of the
modeling efforts for systems with particular characteristics [8,9]. In Tunisian system and has been calibrated with real data (2010–2015).
Ref. [8] Technoeconomic modeling for the power generation sector was The authors investigated electricity demand and capacity, energy pro­
discussed for developing countries, Egypt was chosen as a case study. duction, reserve capacity evolution, CO2 emissions, cost and job crea­
Peculiarities of energy access for Sub-Saharan Africa were discussed in tion benefits of increasing renewable energy.
Ref. [9]. The Korean government’s energy policy has changed dramatically,
OSeMOSYS is an open-source optimization software for long term prioritizing social and environmental concerns. The authors in Ref. [7]
technoeconomic modeling [10–13]. It works as a bottom-up optimiza­ developed a multi-criteria decision making (MCDM) model which is
tion modeling framework. “Set data” in OSeMOSYS can be considered linked to an energy model to assess the economic impact. It was shown
the backbone of the model that defines its components. Sets that can be how new policies affect the total cost and environmental emissions.
defined for a model include daily time brackets, day type, fuel, Region, The authors in Ref. [14] proposed a methodology to represent the
season, technologies, time slices, and years. The “scenario data” are the impact of short-term operational constraints on long term planning by

* Corresponding author. Department of Electrical Engineering, Faculty of Engineering, The Hashemite University, Zarqa, Jordan.
E-mail addresses: [email protected] (M.S. Saadeh), [email protected] (Z. Dalala), [email protected] (O. Saadeh), [email protected] (T. Niet).

https://doi.org/10.1016/j.energy.2023.128237
Received 16 August 2022; Received in revised form 16 May 2023; Accepted 21 June 2023
Available online 30 June 2023
0360-5442/© 2023 Elsevier Ltd. All rights reserved.
M.S. Saadeh et al. Energy 281 (2023) 128237

direct integration of technical constraints in the planning model. The and then into Jordan Electric Power Company (JEPCO) in 1962, and is
authors concluded that incorporating short-term operational constraints currently responsible for distribution operation in the central part of the
in long term models lead to better utilization of existing capacities and country. Irbid district electric company (IDECO) was established in 1957
reduce the requirement of new generation capacity. Thus, over­ and is responsible for all distribution operation in the northern part of
estimation of generation capacity and associated investment can be the country.
avoided. In 1967, the Jordan Electricity Authority (JEA) was founded by the
There are very few existing technoeconomic energy supply modeling Jordanian government. JEA was responsible for all generation and
efforts for the Jordanian energy system. Such modeling efforts do not transmission of electricity, in addition to distribution operation in areas
consider existing long term contractually binding agreements. Jordan not covered by IDECO or JEPCO. In 1997 JEA was restructured into
has long term contracts to import fuel; supply fuel to independent power three separate entities, the National Electric Power Company (NEPCO),
producers (IPPs); adhere to long-term power purchase agreements a public share holding company solely owned by the government,
(PPAs); while subsidizing the distribution sector energy costs, regardless Central Electricity generation company (CEGCO), and the electricity
of volatile market prices. All of the above is important when specifying distribution company (EDCO). NEPCO is responsible for electric trans­
scenarios that are simulated and studied. mission, CEGCO is responsible for electricity generation, while EDCO is
The authors in Ref. [15] investigated the water-energy nexus in responsible for electricity distribution in the southern part of the
Jordan. An OSeMOSYS model of the energy system was created to country. The Electricity Regulatory Commission (ERC) was founded in
investigate the effect of the energy system on water consumption. En­ 2001, which later became the Energy and Minerals Regulation Com­
ergy independence and water desalination feasibility was assessed for a mission (EMRC).
sustainable choice for the country’s future. The work was completed The government has restructured the energy sector for the purpose of
within the context of the FAO project “Implementing the 2030 Agenda reducing its role as well as enhancing the role of the private sector. There
for water Efficiency/Productivity and water Sustainability in MENA has been rising energy demand resulting in a significant increase in per-
Countries”. The authors in Ref. [15] introduce long term tech­ capita consumption. Long-term fixed generation cost, minimum guar­
noeconomic modeling but do not take into account the existing long anteed generation limits along with fixed fuel price enhanced private
term contractually binding agreements of the Jordanian system. sector participation and attracted local and foreign private investors for
The authors in Ref. [16] modeled Jordan’s energy system using the the generation sector through power purchase agreements (PPAs) with
Long-Range Energy Alternative Planning system (LEAP) software. The several Independent Power producers (IPPs). The privatization in the
authors sought to help in planning for optimal electricity generation generation sector has led to investment in the electricity generation
mix. The study found that more electricity should be generated from infrastructure and an increase in generation capacity with sufficient
conventional power plants (53%), followed by renewable energy tech­ spinning reserve.
nologies (28.3%), nuclear power generation (13.5%), oil shale (3.18%), Jordan currently has six IPPs with a total installed generation ca­
and finally from coal fired generation at (1.9%). pacity of 5845 MW. The peak power demand is currently 3146 MW with
In this research characteristics of highly regulated fuel importing a growth rate of 2.9%. Demand is supplied to customers by three private
countries prone to shortages are considered. This is of utmost impor­ distribution companies. The total generated electrical energy in 2020
tance when it comes to technoeconomic modeling, as they significantly was 20,744 GWh. The generation capacity is much larger than the peak
affect the model. Jordan is chosen as a case study representing such a power demand [18].
market. The problem being addressed for the Jordanian case study in The Jordanian electricity sector is highly regulated, as the govern­
this research is the interruption of fuel supply from Egypt in 2013, which ment is the sole importer of primary fuel used for electricity production.
was replaced with a long-term contract to import gas from the leviathan NEPCO is also the sole wholesale buyer and seller of electricity. Resi­
gas field. This is a repeat of the past, by maintaining a single primary dential consumption is approximately 45% of total electrical energy
supplier for fuel. consumption; share per sector is shown in Fig. 1. The electricity cost is
On contrary of the work done in Refs. [15,16], the methodological subsidized for residential consumers. The Jordanian government has
approach adopted in this research is focused on the contractual aspects made reform efforts to achieve energy sector financial sustainability and
of the Jordanian energy system. current policy is modeled, investigated independence.
and assessed. The energy economic model is created to support the Reform has been achieved through many steps of restructuring and
decision-making process related to future energy investments and privatization [18]. The goals of Jordan’s energy strategy per the
planning, both contractual and technical. The model provides insights 2020–2030 published energy sector strategy are to diversify energy
into the status quo of the national energy strategy achievements thus far, sources, increase share of local energy sources, increase energy effi­
while providing means to investigate different scenarios moving for­ ciency, reduce total energy budget and make Jordan a regional energy
ward. The goal is to keep decision makers informed from an hub [17].
Energy-economic perspective. This model is a base line model where Historically, Jordan relied on one primary source from neighboring
different scenarios represent changing circumstances and can be countries for the import of hydrocarbons. Hydrocarbons were typically
implemented. Scenarios can be used to assess different aspects of na­ obtained at a subsidized price below the market rate [19,20]. In the past,
tional energy policy and strategies. The model is dynamic and can be
used to simulate various future scenarios and can be easily updated to
represent the most recent situation [17].

2. Characteristics of the Jordanian energy system

Located in the middle east, Jordan is strategically located at the

intersection of Asia, Africa, and Europe. This strategic location makes
the country a potential energy hub for the region. Even though all
neighboring countries have large oil and gas reserves, Jordan lacks any
hydrocarbons.
The first electrical company in Jordan was established in 1937 with a
single rented engine under the name “Amman Company”. As demand
increased, it expanded into “Jordan Electricity Partnership” in 1945, Fig. 1. Electrical consumption share per sector.

2
M.S. Saadeh et al. Energy 281 (2023) 128237

Jordan has addressed financial crisis by seeking an alternative supplier By assuring a steady supply of fuel through long term fuel purchase
to replace the lost primary supplier. The continuous reliance on a single agreements and assuring profits for IPPs through long-term power
primary source at subsidized prices has compromised the country’s purchase agreements, a stable energy market was created. Overall
energy security. This heavy dependence on one primary source for fuel installed technologies and capacities of IPPs operating in Jordan are
has resulted in economic crises for Jordan whenever disruptions summarized in Table 2. Profit was guaranteed for distribution com­
occurred [21]. panies equivalent to 10% of assets. Profit guarantee along with Tariff
The purpose of electricity price subsidization is to shield the poor setting created a great infrastructure investment incentive for distribu­
from fuel price fluctuation. With time, the prices of electricity subsidies tion companies [30].
increased, and the beneficiaries also increased. Most of subsidy cost was Work is underway to repair damage to the electrical grid tie to Syria
no longer benefiting the poor [22]. The government instituted a major and Lebanon to allow Jordan to export electrical energy to Lebanon
subsidy reform program in 2008 to address the issue [23]. The subsidy [31]. Work is also underway to repair and expand the Arab gas pipeline
reform program was halted in 2011 due to the Arab Spring unrest [24]. to allow Egypt to export natural gas to Lebanon through Jordan via the
The latest of the series of disruptions to Jordanian fuel supply was the pipe line and to allow Iraq to export natural gas and fuel from the port of
disruption to natural gas imports from Egypt. The disruption to natural Aqaba through Jordan also via the Arab pipeline [32,33]. All of which
gas imports from Egypt was caused by a series of bombings targeting the must be considered as future development paths for scenarios.
pipeline supplying natural gas to Jordan. The disruption of natural gas
supply from Egypt left Jordan without its single supplier. No replace­ 3. Methodology
ment for subsidized source of fuel was available.
Due to the interruption of Egyptian natural gas imports in 2012, fuel This section presents the methodology used to develop the tech­
oil had to be used for electricity generation which was threefold more noeconomic model for the Jordanian energy system. The modeling and
expensive at the time [25]. Some of the IPPs are tri-fuel plants that can optimization tool, the data required, the assumptions, and the modeled
use natural gas, heavy fuel, and diesel. The use of oil and diesel for scenarios are described.
electrical generation increased the financial burden of electrical gener­
ation. The use of heavy fuel for electricity generation in 2012 caused a 3.1. OSeMOSYS modeling tool
20% increase in total cost of energy imports. The increase in oil prices
also caused an eightfold increase in the cost of subsidies between 2010 In OSeMOSYS the demand is defined in the model scenario. Future
and 2012. Subsidies on fossil fuel amounted to 2.8% of GDP making up supply by each technology is computed to satisfy the projected demand.
to 8.8% of government expenditures [24,26]. Therefore, Jordan sought Transfer functions that connect fuels, technologies and final demands
to Increase energy security by diversifying the energy mix and increase are computed within the software based on technology, fuel data and
renewable energy share [27]. Also, another major subsidy reform pro­ parameters. The activity by each technology is computed in a manner
gram was initiated in 2012 [25]. that minimizes the overall cost of the technology mix that satisfies the
To ensure a steady supply of fuel at a stable price, Jordan signed long projected demand. GNU linear programing kit (GLPK) solver is used to
term fuel contracts. Advantages of long-term fuel contracts are discussed solve the OSeMOSYS developed model minimization problem. GLPK is
in Ref. [28]. For modeling purposes, as will be discussed in the results known for its efficiency and flexibility. While GLPK may not always be
section, both contract and anticipated market prices are inserted into the the fastest solver for particular problems, it is extremely robust. GLPK
model. Table 1 presents Jordanian long term leviathan natural gas handles a wide variety of problem types and is open source, making it
contract along with spot market price as in Refs. [18,29]. the go to solver for academic and research purposes where transparency
Jordanian gas transmission pipeline network is a low-pressure gas and replicability is crucial.
network. There is dockside floating regasification facility at Aqaba, the In OSeMOSYS, time slices are used to represent different periods
only seaport in Jordan. The purpose of the dockside floating facility is to within a year, enabling the model to account for variations in energy
receive and process high pressure liquefied natural gas to help with supply and demand over time. These time slices can be configured to
seasonal short term natural gas shortages. represent specific hours of the day, days of the week, seasons, or even
To summarize the Jordanian government has contracts to import fuel specific events. This flexibility allows the model to reflect the variability
at a predefined long-term price, sells imported fuel at a fixed price and intermittency inherent in many energy systems, particularly those
despite volatile market prices to IPPs, purchase generated electricity with a high penetration of renewable resources. By using time slices,
from IPPs at a fixed tariff, sell electricity at a fixed subsidized tariff to the OSeMOSYS is able to more accurately represent the operation of energy
distribution sector. These contracts were necessary to attract much systems under different conditions, leading to more reliable and accu­
needed international investment in an aging infrastructure. rate modeling results. This feature is instrumental in comprehensive
There are devastating financial consequences of deviation from energy planning, allowing for informed decision-making regarding
scenarios considered in making such binding contractual decisions. The
national energy policy is implemented through contractual decisions
Table 2
that are based on the current energy situation. failing to consider growth
Installed technologies in the Jordanian electrical network.
projections and planning, and not fulfilling commitments, can have se­
Technology 2016 2017 2018 2019
vere consequences due to the binding and long-term nature of these
contracts. Although intended to alleviate the strain on the economy, CCGT Installed Capacity (MW) 3800 4200 4200 4200
such a policy can instead intensify the burden and impose restrictions on PV Capacity (MW) 209.8 312.8 648.8 1028.8
Wind Capacity (MW) 197 197 372.2 616.2
future energy policy.

Table 1
Projected fuel prices from [18,29].
Fuel type Quantity 2020 2025 2030 2035 2040 2045 2050

Oil USD/barrel 77.1 97.2 123.5 148.4 173.9 199.8 225.7

USD/MMBtu 13.3 16.8 21.3 25.6 30 34.4 38.9
Natural Gas Contract Price USD/MMBtu 6.3 6.3 6.3 6.3 6.3 6.3 –
Natural Gas at Henry Hub USD/MMBtu 2.1 3.3 4.3 5 5.7 6.2 6.9

3
M.S. Saadeh et al. Energy 281 (2023) 128237

investment and operation of energy technologies. • Fixed Cost [M$/GW or $/kW]: defines the fixed Operation & Main­
The main equations of the OSeMOSYS model include the objective tenance cost
function and the system constraints. The objective function is the model • Variable Cost [M$/GWh]: defines the variable costs of the plant
period cost by region. The model period cost by region is the sum of total • Availability Factor: capacity available on average over one year
discounted cost of all modeled years. The total discounted cost is the sum expressed as a fraction of the total installed capacity, with value
of total discounted cost by technology over all technologies. The total ranging from 0 to 1. It gives the possibility to account for planned
discounted cost by technology is the sum of discounted operating cost, outages
capital cost, emission penalties, minus the salvage cost. The discounted • Capacity Factor: represents the variability in generation of the power
operating cost is the annual operating cost of technology discounted plant at each point in time, depending on the availability of resources
through the parameter discount rate. Discount rate is a region-specific • Operational Life [yr]: defines the lifetime of the technology
value for the discount rate. Annual operating cost of technology t is • Residual Capacity [GW]: defines the existing total capacity available
the undiscounted sum of the annual variable operating cost (depends on in each year for a specific type of power plant
activity) and the annual fixed operating costs (depends on capacity). • Total Annual Max Capacity and Total Annual Min Capacity [GW]:
OSeMOSYS balances generation and demand for each time slice by a total maximum/minimum existing (residual plus cumulatively
combination of system constraints and optimization algorithms. The installed) capacity allowed for a technology in a specified year
energy system is modeled as a network of interconnected technologies • Total Annual Max Capacity Investment and Total Annual Min Ca­
converting fuels to final demand. Each of which is represented by a se­ pacity Investment [GW]: maximum/minimum planned new capacity
ries of mathematical equations [10]. At its core, OSeMOSYS seeks to allowed to be installed for a technology in a specified year
minimize the total system cost while satisfying a set of energy balance • Capacity to Activity Unit: defines how much activity can occur by
constraints for each time slice. This essentially means that the total one unit of capacity in one year.
generation of energy from all sources in the system must meet the total
demand in each time slice. The model solves for the optimal generation Cost variables were used to capture the financial aspects of the long
mix and transmission flows that meet these requirements at the lowest term contracts in the model.
cost. In OSeMOSYS, the final demand is characterized by these
For each time slice, the model compares the available capacity of parameters.
different energy sources (considering factors like the capacity factor for
intermittent sources like wind and solar) against the demand in that time • YearSplit: defines the duration of a modeled time slice expressed as a
slice. The model then dispatches the various sources in order to increase fraction of the year; the sum of each entry over one modeled year
marginal cost (from the cheapest to the most expensive) until the de­ should be equal to 1.
mand is met. This is known as economic dispatch. In this way, OSe­ • Specified Demand Profile: defines the share of energy demand
MOSYS ensures that demand is met in a cost-effective manner while required in each time slice.
adhering to the physical and economic constraints of the system. If the • Specified Annual Demand [PJ]: defines the total specified demand
available capacity from all sources is insufficient to meet demand in a for each modeled year, linked to a specific ‘time of use’ during the
time slice, the model may also choose to invest in new capacity, year.
considering the investment and operating costs of different technologies. • Accumulated Annual Demand [PJ]: defines the total accumulated
The equations used by OSeMOSYS to perform said optimization can be demand of a commodity that is constant throughout the year, for
found in Ref. [10]. each modeled year.
System constraints include.
A model was developed for the energy sector in Jordan using OSe­
• Energy Balance: Ensuring that energy supply meets demand for each MOSYS. The data used for the model is described in Ref. [18]. To be able
energy carrier in each time slice. to properly model the Jordanian network, the energy sector was thor­
• Capacity Transfer: Describing the available capacity of each tech­ oughly analyzed. The modeled years were 2020–2070.
nology in each period, accounting for the capacity from the previous The model was calibrated using historic data (2015–2020). The
period, new capacity additions, and retirements. model was deployed and run to determine the cheapest generation
• Activity Constraints: Defining the level of operation of each tech­ technology and modeled period cost. Overall cost, year-by-year energy
nology in each period, which is limited by the available capacity. share by technology, total installed capacities, and imported fuel results
• Resource Availability: Limiting the use of each resource by its are presented and discussed in the results section.
availability.
• Technology Constraints: Describing the operation and performance 3.2. OSeMOSYS model implementation and data
of each technology.
The model was implemented while taking into consideration.
The power of OSeMOSYS lies in its simplicity and flexibility. It allows
users to create a model that is as simple or as complex as needed, 1) Fuel import sources and contracts, minimum and maximum import
depending on the data availability and the level of detail required for the limits, contract duration and price.
analysis by the objectives of the study. Activity constraints, total tech­ 2) Projected fuel prices for the model duration.
nology annual activity lower limit, total technology annual activity 3) All electrical generation sources, generation technologies, initial
upper limit, total technology model period activity lower limit, and total cost, variable and fixed maintenance cost, commission dates and
technology model period upper limit were used to model the contractual expected lifetime.
limitations in the model. 4) Power purchase agreements, minimum and maximum generated
In OSeMOSYS, power plants are represented by technologies. Tech­ power, spinning reserve and capacity.
nologies are characterized by the following parameters. 5) Current and projected load profile.
6) Signed and committed new generation plans.
• Output Activity Ratio: defines the rate of fuel provided (i.e.
Electricity) The Jordanian model was created as a single region model with a
• Capital Cost [M$/GW or $/kW]: defines the overnight investment single mode of operation for the years 2020–2070 with no storage fa­
cost of the plant cilities. Since no storage facilities are modeled, daily time brackets and

4
M.S. Saadeh et al. Energy 281 (2023) 128237

day type sets are not defined. The Sets that were defined for the model potential from various sources across different time slices. It is a measure
are fuel, region, technology, timeslice and year sets. of a power plant’s actual output relative to its maximum potential
The main fuel used in Jordan and considered in the model is natural output within a given time slice. A power plant’s capacity factor is the
gas. Fuels are defined in OSeMOSYS as energy carriers, i.e., electricity is ratio of the actual energy generated by a facility to the hypothetical
considered a fuel. Fuels defined in this model are natural gas, trans­ maximum possible if it were operating at full capacity continuously over
mission level electricity, and distribution level electricity. In OSe­ the same time slice. It can be calculated by dividing the total energy
MOSYS, Fuels are produced by technologies. Technologies defined in produced by a plant during a specific time slice by the energy it would
this model are Natural gas imports, power generation from each IPP, have produced at full capacity.
solar generation, wind generation, and electrical transmission. The This becomes particularly important when we consider the role of
various IPPs discussed previously have been implemented in the model renewable energy sources, such as wind and solar power, in the energy
as technologies. production scenarios modeled in OSeMOSYS. These sources often have
Fuels eventually feed a final demand. The final energy demand lower capacity factors due to variability in their energy production
considered in this model is the overall electrical power consumed in depending on the time slice (e.g., wind speeds fluctuate, and sunlight is
Jordan. The architecture of the interconnection of Fuels, Technologies, not available at night), but their environmental benefits and potential
and final demands are presented in a Reference Energy System (RES). cost savings mean they are still valuable additions to the energy mix.
The modeled RES for the Jordanian energy system is presented in Fig. 2. Model Objectives can be considered the most important factor when
Time slices represent different parts of the year with similar demand choosing the model time slices. The selection of the number and dura­
and renewable energy capacity factor behavior. Time slices enable the tion of time slices in energy system modeling, such as OSeMOSYS, can
modeling of complex systems over longer periods of time. The number of significantly influence the results, and is therefore a critical consider­
time slices and their durations needs to be determined in the model ation. This aspect of model design is largely dictated by the specific
based on the level of detail required by the model. Total time in a year is context and goals of the study, and different authors have proposed
split between time slices via the parameter YearSplit. The value of the various approaches, each with its own merits and drawbacks.
parameter YearSplit for all time slices must add up to one. For instance, the authors in Ref. [2] considered the societal demand
The selection of time slices depends primarily on model objectives behavior and took into account the effect of social behavior during the
and capacity factor considerations. Other factors include data avail­ month of Ramadhan on load demand. Their investigation concluded that
ability and quality and sufficient resolution. during the month Ramadhan there is a decrease in the difference be­
In OSeMOSYS modeling, the capacity factor is a crucial parameter, as tween peaks and troughs. They also concluded that there is an increase
it allows for a more realistic representation of the energy production in the necessity of supply during the night from dispatchable sources and

Fig. 2. Jordan’s reference energy system.

5
M.S. Saadeh et al. Energy 281 (2023) 128237

from wind energy plants. They also considered the consumption Table 3
behavior during weekdays and weekend, both day and night resulting in Year split for each time slice.
a total of 16 time slices. Name Description Year Split
Incorporating societal demand behavior during Ramadhan into en­
ID Intermediate Day 0.3339
ergy system modeling adds significant complexity due to the lunar cal­ IN Intermediate Night 0.3264
endar basis of this month. The difference between lunar and Gregorian PD Peak Day 0.0495
calendars causing yearly shift for the month of Ramadhan is of consid­ PN Peak Night 0.0354
erable effect when it comes to long term modeling. This shifting nature SD Summer Day 0.046
SN Summer Night 0.0389
leads to yearly changes in weather conditions, daylight hours, and cor­ WD Winter Day 0.0779
responding energy demand patterns. While this inclusion can enhance WN Winter Night 0.092
the model’s realism and accuracy, it requires sophisticated handling of
temporal data and may increase computational demands.
The authors in Ref. [34] considered dividing one day per month into broader time slices were utilized. High temporal resolution time slices
four 6 h periods resulting in 48 time slices. The authors in Ref. [35] would be crucial if the study focused on short-term system operation or
chose a similar approach using a day split into early morning, morning, the precise matching of supply and demand on an hourly basis, which
afternoon and night to represent each month of the year. Capacity factor typically includes the consideration of energy storage. This approach
variations were not considered, and production followed an average allows us to efficiently capture the most relevant variations in supply
monthly pattern. and demand over the course of the year, while maintaining a level of
The authors in Ref. [8] as per the objectives of their study, took into simplicity that facilitates the robustness and transparency of the
account the demand profile for different consumption sectors. They modeling process.
considered residential, commercial, governmental, agricultural, indus­ Choice of time slices is based on thorough investigation of hourly
trial, and public lighting sectors. 15 time slices were used to capture the demand, solar energy, and wind energy profiles throughout the year.
variations in the demand profile. Numerical patterns are identified for each profile, and the year is split
The authors in Ref. [36] used high temporal resolution over one year. accordingly to reflect these patterns. Data sets of similar values from a
They used hourly time slices to capture the effect of intermittent numerical perspective were identified, and aggregated. This approach
renewable energy sources on the system. As per the objectives of their offers a balance between capturing the variability in demand and supply
study, system costs were not included in the analysis and a technical capacity while maintaining high computational efficiency.
regulation strategy was followed. In this research high temporal resolution data for demand, solar PV
The objective of the Brazilian model created by the authors in generation, and wind generation profiles were investigated. Hourly data
Ref. [5] was determining an optimal mix of solar PV, wind and hydro with 8760 data sets were numerically analyzed. Numerical patterns
power for a low carbon electricity supply in Brazil. The model used a were detected to represent the entire year. Each entire year was split
daily time series of power production. As necessitated by the objective of between those patterns. First the year was split between day and night to
the model, the authors assumed that sub daily variation on the demand account for demand patterns and solar energy availability. The year was
and supply sides were balanced by storage availability of up to 24 h. This also split into intermediate, peak, summer, and winter to represent du­
assumption is not cost effective as it is not yet economically viable to rations with different levels of photovoltaic and wind power availability
store large quantities of electricity. They also assumed that wind power as well as demand levels. Year split for time slices has been fixed over
was distributed evenly over the day by dividing the production by 24 h, time for the entire modeled period as shown in Table 3.
which they state is a very rough estimate. Scenario data can by generally categorized as demand data and
The most detailed and most accurate timeslice representation was technology data. Demand data are presented in Table 4. Technology
carried out by the authors in Ref. [37]. The authors considered the data can be generally categorized as cost parameters, capacities, and
impact of time resolution on the storage capacity requirement for the activity limits. Technology data are obtained from IPP installations and
system. As per the objective of the study the authors considered models their prospective PPA contracts.
with 1248, 2184, 4392, and 8760 time slices. The authors concluded The main cost parameters are capital cost, fixed cost, and variable
that at least 4392 time slices are to be used for optimizing dispatch. cost. Capital cost in USD/kW represents the investment cost of the plant.
The objective of this study is to create a base line long term tech­ Fixed cost in USD/kW defines the fixed operation and maintenance cost.
noeconomic electricity supply model for highly regulated fuel importing Variable cost in USD/MWhr defines the variable cost of the plant.
countries taking Jordan as a case study. Jordan primarily depends on Generation cost related data are shown in Table 5 and fuel cost data are
natural gas for electricity production with low level of renewable energy shown in Table 1.
penetration. The model was used to provide insight into current natural The activity of each technology is subject to annual and model period
gas import policy and provide recommendations moving forward. In the constraints. Those constraints are total technology annual activity lower
projected 30-year renewable energy outlook for Jordan, the target for limit, total technology annual activity upper limit, total technology
renewable energy penetration is established at a modest 30%. This level model period activity lower limit, and total technology model period
of penetration does not necessitate the incorporation of large-scale en­ upper limit. Those limits are obtained for each technology from the
ergy storage facilities, as dispatchable power plants can adequately respective contract.
provide the difference between the energy demand and the renewable Total annual capacity for each technology is converted to capacity
energy profiles. Thus, energy storage was not considered in this study.
Furthermore, the economic feasibility of large-scale energy storage is yet
to be established. Table 4
Projected Peak Power Demand and generated energy [18].
A large number of time slices impacts problem size, data handling
and computational efficiency. The use of broader time slices allows the Year Peak Power Demand Generated Electrical Energy
utilization of available data more effectively, while still capturing the MW % Growth GWh % Growth
main trends and patterns in demand and supply capacity. As Per the 2020 3146 2.9 20,744 3.0
status que of the Jordanian energy system primarily depending on im­ 2022 3341 3.1 22,063 3.2
ported natural gas, the projected 30-year renewable energy outlook, and 2025 3645 2.9 24,250 3.2
the objectives of the study, there is no need for high resolution time 2030 4186 2.8 28,230 3.1
2040 5528 2.8 38,261 3.1
slices. A more computationally efficient approach was adopted, and

6
M.S. Saadeh et al. Energy 281 (2023) 128237

Table 5
Cost related data for each technology type.
Technology Type Input Efficiency Variable Operation and Fixed operation and 2019 investment Capacity Construction Plant
Fuel Maintenance cost maintenance cost cost factor time Life
(USD/MWhr) (USD/kW) (USD/kW) (years) (years)

Combined Cycle Gas Natural 47.5 37.2 930 86 3 30

turbines Gas
Wind 8.4 1920 32 1 20
Solar PV utility 8.4 900 28 1 25
Solar PV rooftop 6 1000 28 1 25

available for each time slice through the parameter Capacity Factor. from the leviathan gas field are not enforced as per the 15-year agree­
Capacity factor represents the variability in generation of the power ment. Neither is the lower limit for production enforced for any of the
plant at different points in time, which depends on the availability of independent power producers (unlimited). The total model period cost is
resources. Capacity factor for each technology is specified for all defined 20785.8 million USD.
time slices as. This factor is defined for natural gas power plants,
photovoltaic power plants, and wind generation power plants. The ca­ 4. Results and discussion
pacity factor for natural gas plants is fixed at 95%, zero for utility
photovoltaic power plants during nights and an overall average of 13%. The simulator matches the absolute total energy production to the
For wind generation plants it has an average of 23%. total projected demand as in Table 4, while minimizing overall tech­
Only data relevant for discussing the various scenarios implemented nology mix cost. The main results are obtained for each scenario are 1)
and their relevant results has been presented. More data has been pre­ total production by technology, 2) model period cost by region, 3) total
sented in Ref. [18]. annual capacity, and 4) annual fuel imports.
The total production by technology represents the total yearly energy
3.3. Scenarios production for each technology over the modeled period. In the simu­
lator it is called “Total Technology Annual Activity”. The model period
Simulated scenarios reflected the enforcement and/or disregard for cost by region is calculated as a function of cost data entered to the
the limits on electrical generation and natural gas imports as per the model along with the total production by technology.
power purchase agreement and the leviathan gas field contract. Sce­ The total annual capacity is the total capacity installed of each
narios 1–4 are as described in Table 6 below. Enforcement and disregard technology annually over the modeled period. The total annual capacity
of limits was achieved through the “Total Technology Annual Activity of each technology represents the power plant installations and their
Lower Limit” and the “Total Technology Annual Activity Upper Limit”. capacities. The total annual capacity includes both the additional
The limits used were obtained from the relevant contracts. The power installed capacities and the pre-existing capacities. It is important to
purchase agreements from the IPPs only have lower limits, there are no investigate the total annual capacity result as it gives insight into the
contractual upper limits. The leviathan gas field contract has both lower sufficiency of the signed Power purchase agreements with the inde­
import limits and upper import limits. It is of importance to continuously pendent power producers. It is important here to note that for a natural
develop scenarios and analyze the results as contemporary issues arise to gas power plants it is not necessary for the activity to match the capacity,
help guide policy makers. Continuous evaluation of policy to provide usually activity is within the capacity. For renewable energy resources,
continuous feedback is also of importance. typically activity is matched to the capacity.
The annual fuel import resulting from minimizing overall technology
3.3.1. Scenario one (BAU) mix cost is important as it gives insight into sufficiency and cost effec­
In this scenario the upper and lower limit for importing natural gas tiveness of long-term fuel import agreements. This insight is only
from the leviathan gas field are enforced as per the 15-year agreement. evident when comparing annual fuel imports between different
This is the business as usual (BAU) scenario. The lower limit for pro­ scenarios.
duction is enforced for each independent power producer (generation For all four scenarios discussed below, the total production by
technology) from the relevant “take-or-pay” schemed power purchase technology, model period cost, total annual capacity, and annual fuel
agreement. The total model period cost is 25,112 million USD. imports were obtained.
Total production by technology (Optimal energy mix).
3.3.2. Scenario two (PPA limited) Fig. 3 shows the optimal energy mix for the (BAU) scenario. Natural
In this scenario the upper and lower limits for importing natural gas gas IPPs share energy production with no significant changes
from the leviathan gas field are not enforced. The lower limit for pro­ throughout the modeled period. Wind and solar energy provide the
duction is enforced for each independent power producer (generation largest share of energy in the later modeled years.
technology) from the relevant “take-or-pay” schemed power purchase Fig. 4 below shows the energy mix for the “PPA limited” scenario.
agreement (PPA limited). The total model period cost is 21035.9 million The energy mix in this scenario is similar to the energy mix in the “BAU”
USD. scenario, with dominance in renewable energy share beyond 2040. The
total model period cost is 21,035.9 million USD.
3.3.3. Scenario three (NGS limited)
In this scenario the upper and lower limit for importing natural gas
Table 6
from the leviathan gas field are enforced as per the 15-year agreement Scenario label based on enforcement/disregard of limits.
(NGS limited). The lower limit for production is not enforced for any of
Gas contract limits
the independent power producers (generation technologies) from the
relevant “take-or-pay” schemed power purchase agreement. The total with Without
model period cost is 25,112 million USD. PPA limits With 1 2
(BAU) (PPA limited)
3.3.4. Scenario four (unlimited) Without 3 4
(NGS limited) (Unlimited)
In this scenario the upper and lower limit for importing natural gas

7
M.S. Saadeh et al. Energy 281 (2023) 128237

Fig. 3. Total production by technology with IPP generation limits and with Fig. 6. Total production by technology with no IPP generation limits and no
limits on natural gas imports. limits on natural gas imports.

gas contract limitation is the only limit in the “NGS limited” scenario
while the “Unlimited” scenario has no limitations. It is also here
observed that wind and solar power production is increased in the
“Unlimited” scenario as compared to the “NGS limited” scenario. In
turn, natural gas production is reduced in the “Unlimited” scenario as
compared to the “NGS limited” scenario.
In the results in Fig. 6 of the “Unlimited” scenario all constraints on
natural gas imports as per the 15-year Leviathan gas field agreement, as
well as the constraints imposed by the PPA from IPP were disregarded.
The disregard for said constraints relieved the system from any obliga­
tion to use natural gas for electricity production. The results of this
scenario depicted a promising potential for renewable energy, particu­
Fig. 4. Total production by technology with IPP generation limits with no larly wind energy. Over time, it revealed a renewable energy penetration
limits on natural gas imports. exceeding 50%, and as we approach 2070, it predicts nearly 100% en­
ergy generation from wind energy.
Fig. 5 shows the optimal energy mix for the “NGS limited” scenario. However, it is crucial to note that such a high level of renewable
The optimal energy mix results for this scenario are identical to those of energy penetration is not achievable without substantial energy storage
the “BAU” scenario. This similarity means that limits imposed by either capacity. Furthermore, achieving this level of penetration with energy
scenario “BAU” or “NGS limited” and not by the other is irrelevant and storage is not yet feasible with the storage solutions currently available,
does not cause any restrictions. More elaboration on this similarity is as it is not economically efficient to store electricity at such a large scale.
presented later in the discussion section. Given that the primary purpose of this study was to assess the im­
Fig. 6 below shows the energy mix for the “Unlimited” scenario with pacts of fuel import contracts and power purchase agreements, we did
no upper or lower limits on neither Gas imports nor on power produced not incorporate energy storage in our modeling. However, the results
by technology. from the “unlimited” scenario asserts the need for further investigation
When comparing the results in Fig. 6 with those in Fig. 4, it is evident into incorporating energy storage and determining the maximum
that less activity is performed by natural gas plants in the “Unlimited” economically efficient potential of renewable energy, as well as the
scenario than in the “PPA limited” scenario. This is due to the disregard required storage capacity to achieve this potential. Energy storage ca­
of the power purchase agreement in the “Unlimited” scenario. The pacity and cost projections need to be used.
power purchase agreement limitation is the only limit in the “PPA Despite not incorporating energy storage due to current renewable
limited” scenario while the “unlimited” scenario has no limitations. It is energy penetration levels and current 30-year renewable energy outlook
also observed that wind and solar power cover the shortcoming in the not accounting for such high renewable energy penetration, the results
“Unlimited” scenario as compared to the “PPA limited” scenario. from our study indicate a need to further examine and potentially revise
When comparing the results in Fig. 6 with those in Fig. 5, it is evident the renewable energy outlook.
that less activity is performed by natural gas plants in the “Unlimited”
scenario than in the “NGS limited” scenario. This is due to the disregard 4.1. Total installed capacities
of the leviathan gas contract in the “Unlimited” scenario. The leviathan
Fig. 7 shows the total annual installed capacity for each technology
for the “BAU” scenario. It is evident from the figure that the installed
capacities of natural gas plants are all preexisting plants. This result
regarding the installed capacity of each technology is quite similar for all
simulated scenarios.
Fig. 8 below shows the total annual installed capacity for each
technology for the “PPA limited” scenario. These results for the total
annual installed capacities are also similar to those for the “BAU” sce­
nario, thus confirming the sufficiency of the power purchase agreement
provided that renewable energy investments live up to potential.
Fig. 9 below shows the total annual installed capacity for each
technology for the “NGS limited” scenario, it is observed that there is
similarity between Figs. 9 and 8. They are not exactly the same, but there
Fig. 5. Total production by technology with no IPP generation limits with is a striking resemblance. This similarity indicates that the natural gas
limits on natural gas imports. needed by the independent power producers to produce electricity as

8
M.S. Saadeh et al. Energy 281 (2023) 128237

Fig. 7. Total installed capacity by technology with IPP generation limits and Fig. 10. Total installed capacity by technology with no IPP generation limits
with limits on natural gas imports. and no limits on natural gas imports.

“Unlimited” scenario and the “NGS limited” scenario. Even though both
scenarios have the same installed capacities, the “Unlimited” scenario
has less activity by said technologies due to the disregard of the levia­
than gas import agreement. Due to the match between activity and
installed capacity for wind and solar technologies, the capacity of said
technologies is higher in the “Unlimited” scenario to make up for
decreased electricity generation from natural gas due to the disregard of
the leviathan gas import agreement.

4.2. Total imported fuel

Fig. 11 below demonstrates the total fuel imports from the 15-year
Fig. 8. Total installed capacity technology with IPP generation limits with no
agreement and spot market purchases for the “BAU” scenario. Keeping
limits on natural gas imports.
in mind that this scenario was simulated with limits enforced for both
the 15-year natural gas contract and the power purchase agreements.
The natural gas import from NGS at henry hub price in the fuel mix
result means that the modeled NGS technologies representing the 15-
year natural gas contract was not sufficient to meet the demand. It is
evident that the 15-year natural gas agreement’s upper limit is insuffi­
cient to meet the demand. There is only one secured contract for natural
gas import implemented, that is the 15-year leviathan gas import con­
tract. There are other sources of natural gas imports. Natural gas can still
be imported from other sources such as the Aqaba port side regasifica­
tion facility. Those are short-term contracts and spot market purchases.
Very little information is available in the public domain in regard to the
short-term contracts and spot market purchases. In the model, short
term contracts and spot market purchases have been consolidated in one
Fig. 9. Total installed capacity by technology with no IPP generation limits technology that is named “HHIMPNGS”. Some of the IPP generation
with limits on natural gas imports. units are tri fuel units. Tri fuel units, in addition to natural gas, can use
heavy fuel and diesel. Diesel can be used for electrical power generation,
per the lower limit of the power purchase agreement is close in quantity but they are not implemented in the model as it rarely occurs. The last
to lower import limit as specified by the leviathan natural gas import occurrence of heavy fuel use for power generation occurred in 2012.
agreement. The lower limits of the power purchase agreement are Fig. 12 below shows the total imported fuel mix for the “PPA limited”
consistent with the lower limit of the 15-year leviathan natural gas scenario. Up until 2058 all fuel imports are NGS imports at Henry Hub
import agreement. price. Keeping in mind that this scenario did not enforce the limits of the
Fig. 10 below shows the total annual installed capacity for each 15-year leviathan gas contract and the optimum scenario was to import
technology for the “Unlimited” scenario. By comparing the results in
Fig. 10 to the results in Fig. 8, it is evident that the installed capacities for
natural gas technologies is the same in the “Unlimited” scenario and the
“PPA limited” scenario. Even though both scenarios have the same
installed capacities, the “Unlimited” scenario has less activity by said
technologies due to the disregard of the power purchase agreement
limits. Due to the match between activity and installed capacity for wind
and solar technologies, the capacity of said technologies is higher in the
“Unlimited” scenario. The increase in installed capacity for wind and
solar technologies is to make up for reduced activity by natural gas
technologies due to the disregard of the power purchase agreement
limit.
When comparing the results in Fig. 10 to those in Fig. 9, it is evident
that the installed capacities for natural gas technologies is the same in Fig. 11. Total fuel imports with IPP generation limits and with limits on nat­
ural gas imports.

9
M.S. Saadeh et al. Energy 281 (2023) 128237

Fig. 12. Total fuel imports with IPP generation limits with no limits on natural Fig. 14. Total fuel imports with no IPP generation limits and no limits on
gas imports. natural gas imports.

NGS at Henry Hub price, this means that the limits imposed by the 15-
year leviathan gas contract; more specifically the lower limit is not the
best scenario for Jordan. It is important to note that since the total
technology annual activity of the NGS plants are above their lower
limits, that means the consumption of natural gas is above the lower
limit of the power purchase agreement as well. But the import did not
come from the 15-year agreement, it came from NGS at Henry Hub price.
From the aforementioned arguments it can be concluded that the issue
revealed in this scenario with the contract is not the limit as quantity of
imported NGS, but with limit on quantity at contract price.
Fig. 13 shows the total imported fuel mix for the “NGS limited”
scenario. The results in Fig. 13 are identical to those in Fig. 11 due to the
fact that the imported fuel as per the 15-year agreement is more than the
natural gas quantity needed to comply with the lower limit of the power
purchase agreement made with the independent power producers.
Fig. 14 below shows the total imported fuel mix for the “Unlimited”
scenario. When comparing Fig. 14 with Fig. 12, the main difference is Fig. 15. Total cost in million USD for the duration (2015–2070) for the
that there are more gas imports in scenario two due to the enforcement different scenarios.
of the power purchase agreement limit. When comparing Fig. 14 with
Fig. 13, it observed that in the “Unlimited” scenario there is less gas 4.4. Discussion
import in general and most of the imported gas is at Henry Hub price.
This is due to the disregard of leviathan gas field import agreement. The financial short-term and long-term consequences have been
investigated. It is evident based on the results of scenarios one through
4.3. Model period cost four, that the natural gas obtained through the leviathan gas field con­
tract within it’s upper limit is not enough to meet the demand. It is also
Different scenarios have been investigated. Total costs over the evident from the results that the pricing of the leviathan gas contract is
entire modeled period for the different scenarios are presented in Fig. 15 not financially efficient. Yet, a secure steady supply of natural gas en­
below. The least expensive energy production mix has been evaluated. hances energy security and shields Jordan from the volatility of market
As is expected, the least expensive scenario is the “Unlimited” scenario price and from the turbulence in the region.
with no limits on natural gas imports nor on power production. A long- It is highly recommended to find a new source for natural gas. For
term natural gas import contract is in place for the purpose of ensuring a that specific reason the floating storage regasification facility option was
steady supply of natural gas. It is also observed that the effect of the implemented as it is a quick, cost effective, temporary solution. There
power purchase agreements on the model period cost exceeds the effect are plans to replace the floating storage regasification facility with a
of the natural gas import contract. floating storage facility and an onshore regasification facility to reduce
cost through a tie-in point to the Jordanian gas transmission pipeline
network. The expedition of the replacement and the expedition of the
full construction of the Arab pipeline will expedite Jordan reaching
energy hub status.
The total annual installed capacity results for all simulated scenarios
are quite similar. It is observed that the levels of installed capacities of
all natural gas plants are steady. The variation in installed capacities
occurs with wind and solar generation technologies. Having steady
levels for installed capacity of all natural gas plant means all these plants
are pre-existing plants. And the variation of installed capacities for wind
and solar technologies indicates the high potential and low cost of those
technologies. The power purchase agreements with the independent
power producers are sufficient considering the potential of wind and
solar energy. It is important to keep in mind that current power purchase
agreements are only sufficient if wind and solar technologies are suffi­
ciently invested in.
Fig. 13. Total fuel imports with no IPP generation limits with limits on natural
As mentioned in the introduction Jordan aims to become a regional
gas imports.

10
M.S. Saadeh et al. Energy 281 (2023) 128237

technologies is crucial to achieve the most financially efficient energy

system. It is also important to diversify the sources of natural gas imports
in Jordan through multiple long-term agreements with multiple sources.
Having multiple long-term agreements with multiple sources increases
energy security.
It is clear from Fig. 16 that the projection for the future prices of
natural gas has changed significantly. The results of the simulation will
vary drastically if different projections are used. It is highly recom­
mended that future long-term gas import contracts be relatively shorter.
Shorter long-term gas import contracts give decision makers the op­
portunity to reassess revise previous contracts and make a more
informed decision for future contracts based on simulations such as the
one conducted in this study. To summarize, shorter long-term gas import
agreements give the decision makers the opportunity to tailor future
contracts in terms of pricing, lower limits, and upper limits to be in the
Fig. 16. The change in future price projections with time. best interest of Jordan.

energy hub both electrical and hydrocarbon. It is important to note that CRediT authorship contribution statement
the simulation includes uncapped renewable energy technologies. Jor­
dan achieving energy hub position will help increase energy security and Mahmood Shihadeh Saadeh: Software, Formal analysis, Writing –
reduce overall cost. original draft, Visualization, Supervision. Zakariya Dalala: Data cura­
All the results and discussion above were made based on simulation tion, Project administration, Resources, Funding acquisition. Osama
of a detailed model. Simulations can only be as accurate as the data Saadeh: Conceptualization, Validation, Investigation, Resources. Taco
entered into the model. The pricing projection data used for Henry Hub Niet: Methodology, Writing – review & editing.
pricing were obtained from the U.S. Energy Information Administra­
tion’s Annual Energy Outlook 2022 [29]. Fig. 16 show the Henry Hub
Declaration of competing interest
pricing projections per the Annual Energy Outlook published in 2015
through 2022. It is evident that the projections vary significantly even
The authors declare that they have no known competing financial
for seven consecutive years.
interests or personal relationships that could have appeared to influence
Despite the contributions of this study, it is important to acknowl­
the work reported in this paper.
edge its limitations. The most crucial limitation is related to the lack of
energy storage incorporation in the model. Given the study’s focus on
Data availability
the assessment of fuel import contracts and power purchase agreements,
the incorporation of energy storage was deemed outside the scope of this
The authors do not have permission to share data.
analysis. This decision, while suitable for the study’s objectives, restricts
the ability to fully explore and optimize scenarios with high renewable
Acknowledgments
energy penetration. Furthermore, the use of relatively broad time slices,
while practical for the current model, might not be sufficient when
The development of this paper was partially supported by the project
modeling the complexities associated with the utilization of energy
“Policy dialogue and knowledge management on low emission devel­
storage systems. Consequently, the conclusions drawn about the optimal
opment strategies in the MENA region” (DIAPOL-CE), implemented by
energy mix and the feasibility of high renewable energy penetration
the Deutsche Gesellschaft für INTERNATIONALE ZUSAMMENARBEIT
might not fully account for the potential dynamics and benefits of
(GIZ) GmbH on behalf of the German Federal Ministry for the Envi­
incorporating energy storage into the system.
ronment, Nature Conservation and Nuclear Safety (BMU). This work was
also partially supported by the Deanship of Scientific Research at the
5. Conclusion
German Jordanian University. The Authors also thank the Unit of Energy
Systems at KTH and the OpTIMUS community for their support.
Fig. 5, Fig. 9, and Fig. 13 show the results for the “NGS limited”
scenario. When comparing the results for the “BAU” scenario and the
results of the “NGS limited” scenario it is noticed that they are identical. References
Any limits imposed by one of these scenarios and not by the other is
[1] Mogren Olsson J, Gardumi F. Modelling least cost electricity system scenarios for
irrelevant and does not cause any restrictions. When comparing both Bangladesh using OSeMOSYS. Energy Strategy Rev Nov. 2021;38:100705. https://
scenarios, the power purchase agreement limit is the limit imposed by doi.org/10.1016/j.esr.2021.100705.
the “NGS limited” scenario and not imposed by the “BAU” scenario. It is [2] Dhakouani A, Gardumi F, Znouda E, Bouden C, Howells M. Long-term optimisation
model of the Tunisian power system. Energy Dec. 2017;141:550–62. https://doi.
concluded that the purchased gas as per the 15-year agreement exceeds org/10.1016/j.energy.2017.09.093.
the natural gas quantity needed to comply with the lower limit of the [3] Awopone A, Zobaa A. Analyses of optimum generation scenarios for sustainable
power purchase agreement made with the independent power power generation in Ghana. AIMS Energy Feb. 2017;5:193–208. https://doi.org/
10.3934/energy.2017.2.193.
producers. [4] Taliotis C, Rogner H, Ressl S, Howells M, Gardumi F. Natural gas in Cyprus: the
A secure steady supply of natural gas along with long term power need for consolidated planning. Energy Pol Aug. 2017;107. https://doi.org/
purchase agreements provided a path toward energy security and 10.1016/j.enpol.2017.04.047.
[5] Schmidt J, Cancella R, Pereira Jr A. An optimal mix of solar PV, wind and hydro
created the necessary investment incentive for the generation sector in
power for a low-carbon electricity supply in Brazil. Renew Energy Jan. 2016;85.
Jordan. The advantage of the aforementioned contracts is clear, but as https://doi.org/10.1016/j.renene.2015.06.010.
demonstrated throughout the results and discussion section their aspects [6] García-Gusano D, Suárez-Botero J, Dufour J. Long-term modelling and assessment
in terms of limits and pricing are not always in the best interest of of the energy-economy decoupling in Spain. Energy Mar. 2018;151:455–66.
https://doi.org/10.1016/j.energy.2018.03.102.
Jordan. It is important to diversify the energy mix to achieve a more [7] Choi D, Ahn Y-H, Choi D. Multi-criteria decision analysis of electricity sector
financially efficient energy system. Investing in wind and solar energy transition policy in Korea. Energy Strategy Rev May 2020;29:100485. https://doi.
org/10.1016/j.esr.2020.100485.

11
M.S. Saadeh et al. Energy 281 (2023) 128237

[8] Rady Y, Rocco M, Serag-Eldin MA, Colombo E. Modeling for power generation [23] Verme P. Subsidy reforms in the Middle East and North Africa region: a review.
sector in developing countries: case of Egypt. Energy Sep. 2018;165. https://doi. Washington, DC: World Bank; Jul. 2016. https://doi.org/10.1596/1813-9450-
org/10.1016/j.energy.2018.09.089. 7754. Working Paper.
[9] Bazilian M, et al. Energy access scenarios to 2030 for the power sector in sub- [24] Atamanov A, Jellema J, Serajuddin U. Energy subsidies reform in Jordan: welfare
saharan Africa. Util Pol Oct. 2011;20. https://doi.org/10.2139/ssrn.1949612. implications of different scenarios. Washington, DC: World Bank; Jun. 2015.
[10] Howells M, et al. OSeMOSYS: the open source energy modeling system: an https://doi.org/10.1596/1813-9450-7313. Working Paper.
introduction to its ethos, structure and development. Energy Pol Oct. 2011;39: [25] Inchauste G, Victor DG. The political economy of energy subsidy reform.
5850–70. https://doi.org/10.1016/j.enpol.2011.06.033. Washington, DC: World Bank; 2017. https://doi.org/10.1596/978-1-4648-1007-7.
[11] Welsch M, Howells M, Bazilian M, DeCarolis JF, Hermann S, Rogner HH. Modelling [26] Gamba MA. New energy sources for Jordan: macroeconomic impact and policy
elements of smart grids – enhancing the OSeMOSYS (open source energy modelling considerations. International Monetary Fund; 2015.
system) code. Energy Oct. 2012;46(1):337–50. https://doi.org/10.1016/j. [27] Fund IM. Jordan: staff report for the 2012 article IV consultation. International
energy.2012.08.017. Monetary Fund; 2012.
[12] Gardumi F, et al. From the development of an open-source energy modelling tool to [28] Glachant J-M, Finon D, de Hauteclocque A. Competition, contracts and electricity
its application and the creation of communities of practice: the example of markets. Edward Elgar Publishing, Books; 2011. https://econpapers.repec.org/
OSeMOSYS. Energy Strategy Rev Apr. 2018;20:209–28. https://doi.org/10.1016/j. bookchap/elgeebook/14165.htm. Accessed: October. 24, 2021. [Online].
esr.2018.03.005. Available:.
[13] Niet T, Shivakumar A, Gardumi F, Usher W, Williams E, Howells M. Developing a [29] “Annual energy outlook - U.S. Energy information administration (EIA).” https://
community of practice around an open source energy modelling tool. Energy www.eia.gov/outlooks/aeo/index.php (accessed March. 12, 2022).
Strategy Rev May 2021;35:100650. https://doi.org/10.1016/j.esr.2021.100650. [30] Camos D, Bacon R, Estache A, Hamid MM. Shedding light on electricity utilities in
[14] Das P, Gaur A, Jain A, Bhakar R, Mathur J. Long-term energy system planning the Middle East and North Africa: insights from a performance diagnostic. World
considering short-term operational constraints. Energy Strategy Rev Aug. 2019;26. Bank Publications; 2017.
https://doi.org/10.1016/j.esr.2019.100383. [31] UPDATE 2-Jordan aims to supply Lebanon with electricity by year-end -minister.
[15] Weinstein Miles. Future scenarios for energy security and sustainable desalination Reuters; Oct. 07, 2021. https://www.reuters.com/article/lebanon-crisis-electricit
in Jordan. Stockholm, Sweden: M.S., KTH Royal Institute of Technology; 2019 y-idAFL1N2R30CT. Accessed: October. 24, 2021. [Online]. Available:.
[Online]. Available: https://kth.diva-portal.org/smash/get/diva2:1372882/FULLT [32] Al-Khalidi S. Lebanon to get Egyptian gas via Syria in plan to ease crisis. Reuters;
EXT01.pdf. [Accessed 9 March 2022]. Sep. 08, 2021. https://www.reuters.com/business/energy/egypt-says-it-hopes-e
[16] Gresat A, Sukchai S, Thanarak P. Electricity generation mix scenarios simulation xport-gas-supply-lebanon-with-power-soon-2021-09-08/. Accessed: October. 24,
and optimization in Jordan up to 2050 using LEAP software. J Renew Energy Smart 2021. [Online]. Available:.
Grid Technol May 2018;13(2). Art. no. 2, https://ph01.tci-thaijo.org/index.php/ [33] Ismail A. ‘Arab Alliance’ meets in Iraq in first visit in decades for an Egyptian
RAST/article/view/111533. March. 10, 2022. [Online]. Available:. leader. Reuters; Jun. 27, 2021. : https://www.reuters.com/world/middle-east/sisi-
[17] Ministry of Energy and Mineral Resources, “Jordan energy sector strategy for 2020- makes-first-visit-iraq-by-egyptian-leader-decades-2021-06-27/. Accessed: October.
2030,” Energy Sect Strat. http://memr.gov.jo (accessed October. 21, 2021). 24, 2021. [Online]. Available.
[18] Saadeh O, Dalala Z, Niet T, Ramos EP, Howells M. Technoeconomic data adopted [34] de Moura GNP, Legey LFL, Howells M. A Brazilian perspective of power systems
for the development of a long-term electricity supply model for the Hashmite integration using OSeMOSYS SAMBA – south America Model Base – and the
Kingdom of Jordan. Data Brief Jun. 2020;30:105391. https://doi.org/10.1016/j. bargaining power of neighbouring countries: a cooperative games approach.
dib.2020.105391. Energy Pol Apr. 2018;115:470–85. https://doi.org/10.1016/j.enpol.2018.01.045.
[19] Little D. Pipeline politics: America, TAPLINE, and the Arabs. Bus Hist Rev 1990;64 [35] Pinto de Moura GN, Loureiro Legey LF, Balderrama GP, Howells M. South America
(2):255–85. https://doi.org/10.2307/3115583. power integration, Bolivian electricity export potential and bargaining power: an
[20] Alshwawra A, Almuhtady A. Impact of regional conflicts on energy security in OSeMOSYS SAMBA approach. Energy Strategy Rev Sep. 2017;17:27–36. https://
Jordan. Int J Energy Econ Pol Feb. 2020;10:45–50. https://doi.org/10.32479/ doi.org/10.1016/j.esr.2017.06.002.
ijeep.9031. [36] Pupo-Roncallo O, Campillo J, Ingham D, Hughes K, Pourkashanian M. Large scale
[21] Abu-Rumman G, Khdair AI, Khdair SI. Current status and future investment integration of renewable energy sources (RES) in the future Colombian energy
potential in renewable energy in Jordan: an overview. Heliyon Feb. 2020;6(2): system. Energy Nov. 2019;186:115805. https://doi.org/10.1016/j.
e03346. https://doi.org/10.1016/j.heliyon.2020.e03346. energy.2019.07.135.
[22] World Bank Group. Hashemite kingdom of Jordan promoting poverty reduction [37] Groppi D, Kumar Pinayur Kannan S, Gardumi F, Astiaso Garcia D. Optimal
and shared prosperity: systematic country diagnostic. Washington, DC: World planning of energy and water systems of a small island with a hourly OSeMOSYS
Bank; Feb. 2016. https://doi.org/10.1596/23956. model. Energy Convers Manag Jan. 2023;276:116541. https://doi.org/10.1016/j.
enconman.2022.116541.

12