Advancing 100% renewable electricity in Tunisia: techno-economic pathways for integrating solar PV, hydropower, and pumped hydro energy storage

1 Introduction

Climate change is now one of the world’s most pressing issues, along with energy production and consumption. As a result, the global energy transition has become a crucial component of sustainable development, contributing to international efforts to combat climate change. The Paris Agreement aims to keep global temperatures from rising above 1.5 °C (Fan Xiaoyun et al., 2025) and emphasizes the urgency of adopting more advanced energy sources. Renewable energy sources, such as hydroelectricity, solar power and biomass, have significant potential to meet energy needs while lowering carbon emissions. The use of renewable energy has increased in recent years, with photovoltaic and solar energy being the main drivers of this change.

Due to technological advancements and lower costs, renewable energy sources accounted for approximately 92% (Zaman et al., 2025) of the new energy production capacity installed worldwide in 2024 (Ashraf et al., 2025). The world’s capacity for renewable energy has increased by 585 GW, with photovoltaics alone accounting for three quarters of this growth (Zaman et al., 2025), which highlights its contribution to capacity expansion. China and India have been at the forefront of this global boom, contributing 278 GW and 24.5 GW, respectively, mostly to large-scale solar systems.

However, the development of renewable energy in Southeast Asia is still inadequate. While Bangladesh and Pakistan continue to produce solar energy within the framework of regional strategies in favor of domestic energy, India has rapidly diversified into solar and geothermal energy. However, Tunisia continues to rely heavily on solar energy, with more than 6% of its installed capacity coming from photovoltaic systems (Imran et al., 2025). This limited diversification highlights its high susceptibility to climate change (Yan et al., 2025; Kishore et al., 2025).

The majority of Tunisia’s electricity production comes from natural gas centers, making the country’s energy supply heavily dependent on the importation of fossil fuels. Renewable energy sources, namely, solar and solar energy, are still just a small portion of the energy mix, highlighting the need for diversification. Figure 1 depicts Tunisia’s current energy mix and highlights its structural reliance on fossil fuels as a source of energy security, cost volatility, and emission reduction. Tunisia must also adopt renewable energy technology and lessen its reliance on gas imports in order to meet its carbon neutrality goals (Rauniyar et al., 2025; Lamichhane et al., 2025; Capodaglio, 2025).

The population of Tunisia is expected to increase from approximately 11.8 million in 2020 to approximately 13.5 million in 2050. This, along with economic growth and electrification, will result in a significant increase in the demand for electricity. According to national energy planning scenarios, the demand for electricity might reach between 60 and 70 TWh by 2040 in a moderate growth scenario and surpass 100 TWh in a strong growth scenario or as a result of public policies. Even though residential electricity consumption is still relatively low compared to global averages, demand has increased steadily over the past century due to growing urbanization, industrial activity, and increased use of electrical appliances (Wen et al., 2025).

Due to the scarcity of water resources, hydroelectricity only plays a minor role in Tunisia’s renewable energy generation, which now depends primarily on solar photovoltaic and solar energy. Photovoltaic solar energy production is subject to daily and seasonal changes, and solar energy production also varies according to meteorological circumstances. This variability, when combined with an inadequate storage capacity and a rigid system, highlights the system’s challenges with equilibrium (Zhu et al., 2025).

Because of this, Tunisia depends on electricity imports during periods of high use, particularly in summer, which exposes its economy to risks related to pricing and supply. These factors highlight the significance of increasing renewable energy generation capacity concurrently with investments in system reinforcement, storage solutions, and system flexibility to ensure long-term reliability and durability (Hatamkhani et al., 2025; Weiss, 2025).

Additionally, Tunisia’s heavy reliance on the production of electricity from fossil fuels raises a number of issues, including exposure to fluctuating combustible prices, risks associated with energy security, and increased greenhouse gas emissions. These limitations highlight the need to diversify the energy mix in order to increase system resilience and long-term durability. Given the good climate and abundant solar resources in Tunisia, photovoltaic energy is a classic diversification option. Due to Tunisia’s strong solar radiation and average horizontal global irradiance of 4.5–5.5 KWh/m²/day in most regions, solar photovoltaics is one of the country’s most promising renewable energy technologies. The initial national assessments estimate solar photovoltaic technology’s potential to several gigawatts, based on cautious assumptions about soil utilization and power density. More recent research indicates that Tunisia’s solar potential could significantly exceed 100 GW when considering a combination of distributed photovoltaic systems, toitures, and photovoltaic centers. However, the actual deployable capacity strongly depends on the availability of land, access to the energy, and system configuration. Despite its significant potential, solar photovoltaic energy is still underutilized (Tayyeh et al., 2026; Kakoulaki et al., 2026). In 2020, installed photovoltaic capacity would only make up a small portion of the overall amount of electricity produced, indicating a significant gap between available resources and their efficient use. Developing solar photovoltaic energy could play a significant role in reducing Tunisia’s reliance on imported natural gas and in preventing seasonal fluctuations and consumption peaks, particularly during the summer months when demand for electricity increases significantly due to climate change. However, the unpredictability of solar generation presents issues with system flexibility and reliability. It is also necessary to integrate technology for energy storage, demand management, and system reinforcement (Kaur et al., 2025; Fan Xinyi et al., 2025). In particular, battery-based energy storage systems (BESS) appear to be a promising and flexible option for Tunisia, given their low costs and compatibility with solar photovoltaic energy at both the system and decentralized production levels (Leal-Arcas, 2025; Ye et al., 2025). Stockage solutions can help reduce short-term variability, delay consumption points, and improve the system’s overall stability (Pata et al., 2025). From an energy policy perspective, Tunisia has historically prioritized the production of electricity from natural gas, with goals for renewable energy progressing more slowly than initially anticipated (Blakers et al., 2025; Bhatta and Prasad, 2025). If current national strategies and contributions determined at the national level set ambitious goals to increase the share of renewable energy in the electrical mix by 2030 and beyond, clear and coordinated paths for the widespread use of solar energy and the integration of storage remain limited (Rauf et al., 2025). This strategy runs the risk of maintaining the electrical industry’s long-term reliance on fossil fuels while underutilizing complementary renewable resources. As a result, a more diverse and strategic political approach is required, placing solar photovoltaic energy and energy storage at the center of Tunisia’s future economic development (Quaranta et al., 2025). Even though the current political frameworks serve as a foundation for the development of renewable energy sources, there is still a significant gap in the ability to integrate renewable resources in an optimal manner with flexible options in order to meet the growing demand for reliable and rentable electricity (Baccouche et al., 2025). Closing this gap is crucial to supporting Tunisia’s transition to a resilient and carbon-based economy. Therefore, this study aims to contribute to this goal by placing Tunisia’s energy transition within the broader context of global research on energy systems with strong renewable components and zero emissions (Bąk et al., 2025; Sakhraoui, 2025; Michailidis et al., 2025a).

Globally, growing concerns about climate change and energy security have strongly accelerated the shift to 100% renewable energy systems, sparking a significant increase in scientific research. Numerous initiatives at the global level have shown that a full transition to renewable energy systems, mostly based on solar, solar energy, and energy storage technologies, is technically and economically feasible. For instance, BALAKRISHNAN, P. has shown that a global transition to an entirely renewable energy source by 2050 may be achieved at competitive costs thanks to a combination of photovoltaic, solar, and storage technologies (Balakrishnan, 2025; Sabry, 2025; Diop et al., 2025; Darmawan, 2025a). Subsequent research has extended this framework to the transportation, transportation, electricity, and desalination sectors, confirming the profitability of an entirely renewable global energy system. At the national level, similar modeling approaches for energy systems have been widely used to assess the paths towards 100% renewable energy in a number of countries, including Australia, China, Denmark, India, Ireland, Japan, Malaysia, and the Philippines (Michailidis et al., 2025b). Despite differences in geography, demand profiles, and resource availability, these studies consistently conclude that a significant portion of renewable energy sources can meet the demand for reliable electricity, provided they are supported by adequate storage and distribution infrastructures. One commonality among these studies is the use of low-coefficient optimization cadres with high temporal resolution, which simultaneously optimize production capacity, stocking technologies, and, in this case, transportation infrastructure while taking demand variability and resource intermittence into account (Ben Belgacem et al., 2025; Mansouri et al., 2025; Arribas et al., 2025). Recent review studies have further highlighted the rapid development of optimization techniques for integrated renewable energy systems. These studies emphasize that combining renewable generation with energy storage technologies through advanced optimization algorithms significantly improves system reliability, flexibility, and economic performance. They also demonstrate that techno-economic optimization has become an essential decision-support tool for identifying cost-effective renewable energy transition pathways under different resource, demand, and policy scenarios. These recent advances further support the methodological approach adopted in the present study, which combines least-cost optimization with scenario-based analysis to evaluate long-term renewable electricity pathways for Tunisia (Kenfack et al., 2024). Numerous regional studies demonstrate the strength of these modeling approaches. High-resolution models of large, interconnected systems have shown that solar and solar energy can meet the vast majority of electricity demand, with energy storage ensuring system reliability and minimizing losses. Additionally, it has been seen that regional interconnections significantly reduce the system’s global costs and storage requirements, which makes it possible to obtain current electricity costs in comparison to conventional production using fossil fuels (Guillot et al., 2025; Müller, 2025). Similar national models have shown that solar and solar power systems, when combined with suitable storage technologies, can meet future electricity demand at economically viable costs in a wide range of scenarios. Many of these rely on advanced optimization techniques, such as Differential Evolution (DE), to minimize system costs. The DE is particularly well-suited to complex, nonlinear, and multidimensional energy system problems, making it a powerful tool for investigating the best technological combinations under fluctuating demand and resource situations. The extensive use of this modeling framework in several countries highlights its adaptability, resilience, and transferability (SANCHEZ et al.). Despite these advancements, Tunisia is still somewhat underrepresented in the development of energy systems with strong renewable components and 100% renewable energy, particularly in countries using high-resolution and low-cost optimization cadres. Even though regional or continental assessments frequently include Tunisia’s agricultural sector, they typically rely on oversimplified hypotheses rather than detailed and specific analyses of the country’s demand dynamics, resource availability, and constraints. Given Tunisia’s strong solar potential, emerging solar potential, and growing electricity consumption, coupled with a strong reliance on imported natural gas, it is evident that national modeling efforts must be carried out. These kinds of evaluations are crucial for assessing how Tunisia may make the transition to a highly renewable energy system, or even an entirely renewable energy system, that is both rentable, reliable, and relevant to the political strategy (Miled et al., 2025; Hamdi et al., 2025; Darmawan, 2025b).

Building on similar international research frameworks, a number of studies have used energy modeling and optimization techniques to develop decarbonization pathways for national energy systems, proving the theoretical viability of high-renewable or fully renewable electricity systems under different assumptions of resource availability, demand growth, and storage integration (Aguayo et al., 2025). Though generally informative, many of these analyses lack continuous multi-year simulations with high temporal resolution, nonlinear least-cost optimization, and scenario-based representations of demand growth, technology costs, and trade or grid interconnection options—all of which are essential for strategic energy planning.

Instead of thorough, nation-specific least-cost system optimization, current research and planning initiatives in Tunisia have mostly concentrated on certain technical elements or desk studies of renewable potential (Kiehle et al., 2025). Techno-economic evaluations, for instance, have looked at hybrid systems for particular regions or industries, like solar and battery systems for remote locations, but they usually don't fully capture the dynamics of a national power system with hourly resolution and multi-decadal scenarios. Similar to this, more general policy and strategy documents describe renewable targets, but they typically do so in terms of capacity goals rather than optimal system pathways, leaving gaps in our knowledge of how to reliably and economically integrate high shares of variable renewables, such as solar PV and wind, with storage and interconnections (Haddad et al., 2025; Dora et al., 2025).

In fact, Tunisia’s contribution of renewable energy sources in the country’s electricity mix is still small, even if it is steadily rising. Renewable energy sources make up just a small portion of the country’s overall electricity generation, primarily from solar PV and, to a lesser extent, wind. However, national strategies have not yet been supported by comprehensive, high-resolution modeling studies that quantify least-cost pathways across various future demand and resource scenarios. National strategies envision a significant expansion of renewable capacity, with targets of 30%–35% of electricity from renewables by 2030 (Raza et al., 2025; Koundouri et al., 2025).

A national-level, high temporal resolution, least-cost optimization model that is adapted to Tunisia’s energy system features—such as its robust solar resource, developing wind capacity, current natural gas base, and possible regional electricity interconnections—is obviously needed to close this methodological and geographical gap. By identifying cost-optimal combinations of generation, storage, and trading under various future demand scenarios and policy targets, such a model would give investors and governments solid insights into viable routes toward deep decarbonization and energy security.

As far as the author is aware, this work offers the first comprehensive, non-linear, hourly least-cost optimization system at the national level. Unlike recent review studies that primarily synthesize optimization methodologies for hybrid renewable energy systems, the present work applies a high-resolution national-scale optimization framework to identify cost-optimal renewable electricity pathways specifically tailored to Tunisia’s energy system (Kenfack et al., 2024). The report also emphasizes the strategic importance of cross-border trade and diversification for affordability and dependability.

2 Research methodology

Using a net-load method intended to capture both seasonal and interannual variability in renewable resource availability and demand patterns, this work presents a high-resolution cost-optimization model that replicates the hourly balance between power supply and demand.

This paper is structured as follows: Section 2 establishes the methodological foundation of the study. It begins with the system representation which defines the boundaries of the modeled power system, its main components, and their interactions. This is followed by Section 2.2, which describes the data sources and processing procedures used to parameterize the model. Building on these inputs, Section 2.3 introduces the scenario framework and development, outlining how demand levels, resource availability, and policy or system constraints are combined to construct the analyzed scenarios. The modeling methodology and analytical framework are then presented in Section 2.4, which explains the mathematical structure of the model and the simulation workflow. The key model assumptions underlying all simulations are summarized in Section 2.5, providing transparency regarding simplifications and boundary conditions adopted in the analysis. Section 3 reports the analysis results. It starts with an overview of cost-optimal scenarios (3.1), followed by a detailed examination of energy balance and infrastructure planning across different demand scenarios (3.2). The composition of the levelized cost of electricity (LCOE) under various scenarios is analyzed in Section 3.3, while Section 3.4 presents a sensitivity assessment, focusing on the impact of variations in the discount rate. The implications of these findings are discussed in Section 4, where results are interpreted in terms of system design, economic feasibility, and policy relevance. Section 5 outlines the limitations of the study, highlighting data uncertainties and methodological constraints. Finally, Section 6 concludes the paper by summarizing the main insights and suggesting directions for future research and decision-making.

2.1 System representation

The structural design of this study was based on a nodal structure, with each node representing a major transition position within each Tunisian government. These studies examine the supply and demand for electricity within their respective administrative boundaries. Each governorate’s major transformation sites were identified using national electrical planning reports, and their geographical locations were mapped using the QGIS (Quantum Geographic Information System) software, as shown in Figure 2. The internal transportation system using alternating high tension that depends on Tunisian governors is represented by the red lines. Each governorat has been modeled as a distinct entity with unrestricted power flows inside its borders in order to simplify the power balance within governorats. Additionally, the transfrontal CAHT transportation lines that depend on Tunisia’s neighboring countries—Algeria and Libya—have been included and marked with black lines (Grigg, 2025). These connections were modeled as transportation links between the governors of Tunisia and neighboring electrical networks; the distances of the lines are specified in official planning documents, and each link aims to provide a maximum power transfer capacity in accordance with declared values. While intergovernmental connections were used to ensure the internal balance of power within the country, transfrontal connections were taken into consideration for scenarios involving the importation of electricity (Ahmad et al., 2025).

2.2 Data sources and processing

2.2.1 Electricity load data

The 2021 and 2022 electricity consumption forecasts are from the Tunisian Ministry of Energy and Gas (STEG). The purpose of these time series was to create a nationwide profile of the hourly demand for electricity. Due to a lack of demand hours at the state level, the national demand for electricity has increased for each state with the help of population-based scale factors (Wen et al., 2025).

In order to account for future growth in the need for electricity, three demand scenarios have been developed based on forecasts of electricity consumption per resident, in accordance with national energy transition objectives and development trajectories. These scenarios were modeled using Tunisia’s 2050 population forecasts, which were based on the Banque Worldwide’s estimates of population growth. In this study, the demand for electricity represents the national agricultural charge profile, encompassing all consumer sectors. Although the sectoral demand for electricity has not been explicitly modeled, it is implicitly included in the global demand issue of the national STEG system. This approach maintains a consistent temporal demand profile across all governors while providing a coherent and flexible framework for projecting future electricity demand (Zhu et al., 2025).

2.2.1.1 Low demand scenario (2MWh per capita per year)

This scene is based on Tunisia’s national energy planning strategies and policies, which favor a steady increase in electricity consumption in the residential, institutional, transportation, and industrial sectors. In the Tunisian environment, a modest increase in electricity consumption reflects a gradual shift toward electrification and the replacement of fossil fuels, particularly in final user sectors like transportation and transportation. This scenario represents a sensible development trajectory characterized by limited economic growth and the implementation of energy-related policies (Trabelsi et al., 2025a).

2.2.1.2 Medium demand scenario (5MWh per capita per year)

This scenario is in line with the levels of electricity consumption seen in middle-aged countries, where the average annual consumption of electricity per resident is between 4 and 6 MWh. It represents a trajectory of sustained economic growth and accessible electrification throughout the European sectors. This trajectory aligns with Tunisia’s long-term development goals and the major global trends in energy transition and increased reliance on electricity (Kahoui and Sahed, 2025).

2.2.1.3 High demand scenario (9MWh per capita per year)

This scenario envisions a future in which Tunisia achieves nearly complete electrification of all its end use sectors, including residential, industrial, and transportation uses. These levels of demand are similar to those seen in developed and heavily regulated economies that follow carbon neutrality trajectories. This scenario is based on strong economic growth, ambitious electrification policies, and widespread adoption of electrical technologies (Zuhaer et al., 2025).

Together, these three scenarios show a wide range of plausible future electricity demand trajectories and allow for the assessment of the system’s resilience, adaptability, and investment requirements from both cautious and ambitious perspectives.

2.2.2 Solar energy resource data

Due to Tunisia’s abundant solar potential and the significant decline in costs at the global level, solar photovoltaic energy is regarded as the main new technology. In order to understand the temporal and spatial characteristics of solar availability, this study combines long-term irradiance measurements with a selection of representative sites for each province. Solcast has been used to provide solar radiation hours (2013–2022) with various parameters such as global horizontal irradiance (GHI), ambient temperature, relative humidity, and vent direction and speed. Twenty-three solar locations were chosen in order to account for the geographical variance in solar resource availability. The model’s ability to account for daily variability, seasonal trends, and interannual changes is made possible by the use of time series spanning 10 years (Malakar and et, 2026). This provides a foundation for the assessment of energy production and storage needs. Figure 3 shows the locations chosen throughout Tunisia for the collection of enrollment hours.

2.2.3 System advisory model

The European Consultative Model (SAM), a computer program developed by the National Renewable Energy Laboratory (NREL) of the United States, has been used to convert energy production hours in order to obtain photovoltaic production hour profiles.

The PVWatts performance model was used to simulate a conventional photovoltaic system with a fixed inclination using system concept parameters and specific input data. The hourly simulated photovoltaic productions were then normalized by dividing them by the system’s installed capacity in order to obtain capacitance factor profiles. These normalized profiles represent the portion of production that is available at any given time and have been used as entry constraints in the optimization model (Quaranta et al., 2025). Figure 4 shows the global work flow of SAM inside this process.

2.2.4 Hydropower generation capacity

The modeling approach took into account two hydropower capacity scenarios—Reference Hydro and Ambitious Hydro—to represent varying degrees of hydropower development in Tunisia by 2050. The Reference Hydro scenario include both projects that have been formally proposed or approved for development as well as hydropower facilities that are now in operation, mostly small-scale installations connected to dams and water management infrastructure. In line with Tunisia’s current hydropower portfolio and short-term growth goals, this scenario reflects a cautious outlook.

By adding potential hydropower projects found in national water and energy planning studies as well as locations being considered for future growth, the Ambitious Hydro scenario expands on the Reference Hydro example. Subject to hydrological, environmental, and water-use constraints, this scenario represents a more hopeful pathway in which hydropower potential is further exploited through the growth of small and medium-sized plants (Weiss, 2025; Tayyeh et al., 2026).

Data about hydropower projects, such as installed capacity and geographic coordinates, were gathered from national sources in the water and energy sectors. The Quantum Geographic Information System (QGIS) program was used to process and map the spatial data, then the governorate boundaries of Tunisia were superimposed. In order to estimate generation levels under both hydropower scenarios, annual energy generation statistics for active hydropower facilities were obtained from national utility reports. Table 1 shows the installed hydropower capacity distribution at the governorate level for each scenario, and Figure 5 shows the spatial distribution of hydropower project locations.

TABLE 1. Governorate-level reference and high-ambition hydropower capacity scenario.

2.2.5 Pumped hydro energy storage data

The model took into account Pumped Hydro Energy Storage (PHES) as a well-established and economical choice for long-term, large-scale electricity storage. Based on worldwide evaluations of greenfield PHES potential, which identify possible areas based on topography, elevation variations, and reservoir configuration, the research is predicated on the existence of appropriate off-river PHES sites in Tunisia. These possible locations include a broad variety of storage capacities and are divided into pricing groups based on the approximate building costs per unit of energy stored (Pata et al., 2025).

Important physical and techno-economic characteristics, such as hydraulic head, storage capacity, and the separation between higher and lower reservoirs, are used to categorize sites. Low-cost locations with advantageous features and more expensive locations with less ideal configurations are examples of cost classes. A conservative but accurate estimate of Tunisia’s PHES potential was obtained for this study using techno-economic factors typical of relatively low-cost sites. Both PHES’s installed power capacity and energy storage capacity were considered optimization factors in the model (Bhatta and Prasad, 2025). Table 2 summarizes the primary cost and performance assumptions for PHES.

TABLE 2. Model input parameters for technology costs, discount rate, efficiency, and project lifetime.

2.2.6 Techno-economic parameters

Table 2 summarizes all pertinent expenses, which were gathered from national policy papers, market reports, and pertinent literature. These prices include capital expenditure, fixed and variable operation and maintenance (O&M) costs, round-trip efficiency of PHES, and transmission costs. With off-peak pricing based on STEG reports, it was believed that PHES charging (pumping) would take place during off-peak hours.

The average Power Purchase Agreement (PPA) rate given by STEG, which represents the true cost of electricity purchased from active hydropower plants, was used to calculate the cost of hydropower generating. This strategy guarantees complete cost recovery, including capital payback, O&M, and project risk, and accounts for the expenses borne by independent power producers (IPP). Project costs are largely location-dependent due to Tunisia’s relatively small and site-specific hydropower potential, local topography, and hydrological unpredictability. According to the International Renewable Energy Agency (IRENA), which points out that regulated tariffs and PPAs boost investor confidence and lower financing costs, using PPA-based tariffs is in line with market realities. The application of PPA-based generation costs has been supported by earlier research conducted in Tunisia and other settings (Ashraf et al., 2025).

A sensitivity analysis was conducted by altering expenses by ±25% in order to account for variability in hydropower costs (Section 3.4). Although real rates may differ depending on market conditions and state incentives, a 5% discount rate was chosen to reflect average financing conditions for renewable energy projects in Tunisia. This presumption is consistent with national and international financial frameworks that prioritize affordable funding for renewable energy initiatives (Imran et al., 2025). To account for funding concerns, sensitivity analysis was also carried out across a range of discount rates (Mustafa et al., 2026; Ma et al., 2026).

2.3 Scenario framework and development

A structured scenario framework was created based on two crucial elements in order to assess Tunisia’s possible future energy system pathways: the amount of hydroelectric capacity that is currently available and the anticipated levels of electricity demand. In order to capture the impact of linkages with neighboring countries like Algeria and Libya, each scenario was further examined under various assumptions on cross-border electricity commerce. The whole scenario structure included in the modeling is shown in Table 3 (Adhikary et al., 2025; Ben Belgacem et al., 2025). NoImport scenarios take into account a completely domestic supply, whereas Import scenarios assume cross-border electricity imports.

TABLE 3. Least-cost modeling scenario framework.

The electricity demand levels considered in this study (2, 5, and 9 MWh per capita per year by 2050) were selected to represent three plausible long-term electrification pathways for Tunisia rather than official demand forecasts. They correspond to low-, medium-, and high-demand scenarios that reflect different rates of economic development, technological adoption, and electrification of end-use sectors.

The 2 MWh per capita per year scenario represents a conservative transition pathway in which electricity demand increases moderately through gradual improvements in living standards and limited electrification beyond current trends.

The 5 MWh per capita per year scenario represents an intermediate pathway characterized by accelerated deployment of renewable electricity and significant electrification of transport, residential cooking, commercial activities, and industrial processes. This scenario is considered the most representative of Tunisia’s medium-term energy transition objectives.

The 9 MWh per capita per year scenario represents an ambitious deep-electrification pathway consistent with a highly decarbonized economy in which electricity becomes the dominant final energy carrier across transport, industry, buildings, and other end-use sectors. Such demand levels are commonly adopted in international long-term energy system studies to evaluate infrastructure requirements under net-zero and 100% renewable electricity scenarios.

These scenarios were developed to investigate the robustness of the proposed optimization framework under different future electricity demand trajectories rather than to predict a single future demand level. They therefore provide a comprehensive basis for evaluating the sensitivity of the optimal energy mix, storage requirements, transmission capacity, and Levelized Cost of Electricity (LCOE) to different degrees of future electrification.

2.4 Modeling methodology and framework

The model’s main goal is to identify the least expensive electrical system configuration that uses exclusively renewable energy sources to consistently meet electricity demand at every provincial node throughout the day. Equation 1 defines the network-level LCOE, which the model minimizes.

L C O E = T o t a l a n n u a l i z e d s y s t e m c o s t T o t a l e l e c t r i c i t y g e n e r a t e d (1)

The optimization problem was solved using the Differential Evolution (DE) algorithm, a population-based stochastic optimization technique that has proven highly effective for solving nonlinear, non-convex, and high-dimensional optimization problems encountered in energy system planning. DE was selected because of its robustness, simplicity, and ability to efficiently search the global solution space without requiring gradient information.

The objective of the optimization is to minimize the total levelized cost of electricity (LCOE) while satisfying the hourly electricity demand over the entire simulation horizon. The optimization problem is formulated as shown in Equation 2:where (C_PV), (C_Hydro), (C_PHES), (C_Transmission), (C_Imports), and (C_O&M) denote the annualized costs associated with photovoltaic generation, hydropower, pumped hydro energy storage, transmission infrastructure, imported electricity, and operation and maintenance, respectively, while (E_Demand) represents the total annual electricity demand supplied by the system.

min ⁡ L C O E = C P V + C H y d r o + C P H E S + C T r a n s m i s s i o n + C I m p o r t s + C O & M E D e m a n d (2)

The optimization decision variables include the installed photovoltaic capacity, PHES power capacity, PHES storage capacity, and cross-border transmission capacity. Hydropower capacity is treated as a fixed input according to the selected hydropower development scenario (Reference Hydro or Ambitious Hydro).

The optimization is subject to several constraints, including hourly power balance, storage charging and discharging limits, storage energy balance, transmission capacity limits, generation capacity limits, and non-negativity constraints for all decision variables. Candidate solutions violating any constraint are considered infeasible and are handled through a penalty-function approach that increases the objective function value proportionally to the magnitude of the constraint violation.

The DE algorithm employs the classical DE/rand/1/bin mutation strategy. After preliminary numerical testing, the algorithm parameters were selected to provide a satisfactory balance between convergence speed and solution quality. The population size was set to 100 individuals, the mutation factor was fixed at 0.8, and the crossover probability was set to 0.9.

The optimization process was terminated when either the maximum number of 500 generations was reached or when the relative improvement of the objective function remained below (10⁻⁶) over 50 consecutive generations. To verify solution robustness, the optimization was repeated several times using different randomly generated initial populations. All independent runs converged to nearly identical objective-function values, indicating stable convergence and good robustness of the proposed optimization framework.

The model allows for the investigation of both seasonal/interannual trends (such as wet-dry hydro patterns) and short-term variability (such as daily solar cycles) throughout a 10-year simulation period (2013–2022) at an hourly resolution. As covered in Section 2.2.1, electricity demand was scaled to estimated 2050 population and per capita consumption scenarios. The DE algorithm, which is appropriate for high-dimensional, nonlinear energy systems, was used to solve the optimization.

Cross-border transmission constraints, PHES power and energy capacity, and solar PV capacity are examples of decision factors. As explained in Section 2.2.5, two hydropower scenarios used hydropower capacities as fixed inputs.

The modeling employed a net load balancing approach, in which net load is determined by subtracting demand from total generation from renewable sources and, if relevant, imports. The fundamental signal for running storage systems, primarily for PHES charging and discharging, as well as for controlling spills and electrical imports, is this net load.

The model effectively reflects demand patterns and renewable generating variations based on net load decisions, allowing for the economical use of imports and storage.

The following is a summary of the model’s primary balancing equations.

2.4.1 Net load modeling equation

After deducting generation from PV, hydro, and imports (where imports are permitted), Equation 3 determines the net load at each hour, which represents the remaining demand.

N e t l o a d t = D e m a n d t − P V t − H y d r o t − I m p o r t s t (3)

2.4.2 PHES energy discharge

PHES energy discharge under positive net load is governed by Equation 4, subject to restrictions imposed by stored energy and available capacity.

D i s c h a r g e P H E S = min ⁡ ( max 0 , N e t l o a d t , P H E S p o w e r C a p , S t o r a g e P H E S t − 1 Δ t (4)

2.4.3 PHES energy charge

PHES charging, which happens when the net load is negative (i.e., surplus energy), is determined by Equation 5; charging is limited by the PHES system’s round-trip efficiency and available storage capacity.

C h a r g e P H E S t = min ⁡ ( − 1 x min ⁡ ( 0 , N e t l o a d t , P H E S p o w e r C a p , P H E S m a x S t o r a g e − S t o r a g e P H E S t − 1 μ P H E S (5)

2.4.4 PHES storage level update

Based on hourly charging and discharging actions, Equation 6 updates the stored energy in PHES.

S t o r a g e P H E S t = S t o r a g e P H E S t − 1 − D i s c h a r g e P H E S t x Δ t + C h a r g e P H E S t x μ P H E S (6)

2.4.5 Final system balancing

System-wide balance is guaranteed by Equation 7. We use the import scenario permitted or not allowed option in the model. The model assigns the remaining demand to imports from outside sources when the import flag is raised. In this scenario, the optimizer makes an economical decision based on which option lowers the total system cost: importing power or increasing domestic PV and PHES capacity. On the other hand, the model tries to fully balance demand using domestic generation and PHES, regardless of cost, when the import flag is off. A shortfall is the result of any unfulfilled demand.

I m p o r t t = D e m a n d t + C h a r g e P H E S t + S p i l l a g e t − P V t − H y d r o t − E x t e r n a l t − D i s c h a r g e P H E S t − D e f i c i t t (7)

The convergence behavior of the Differential Evolution algorithm was monitored throughout the optimization process by recording the evolution of the objective function at each generation. A rapid reduction in the objective function was generally observed during the early generations, followed by gradual convergence toward a stable optimum. No significant improvement was observed after convergence, confirming that the selected parameter settings provide a reliable and computationally efficient optimization framework for the proposed energy system model.

2.5 Model assumptions

In order to preserve computational viability and guarantee relevance for long-term planning, a number of assumptions were made during modeling. Along with hydropower and PHES, solar PV was seen as the next-generation source. The existing data environment unique to Tunisia, the solar PV-dominated system, and the focus on medium-to long-duration energy balancing where PHES is both technically feasible and economical are all reflected in the optimization choices (Istiqomah, 2025).

In Tunisia, wind energy is already utilized to help produce electricity. Numerous wind farms that are connected to the national grid and have been in operation since the 2000s are found throughout the nation, mostly in the northern and northeastern regions. Wind power is a component of Tunisia’s aim to diversify its energy mix and develop renewable energies, even though its contribution is still smaller than that of solar and natural gas.

Due to the lack of disaggregated data, the normalized hourly demand profile for electricity was assumed to be the same across all provinces. For the forecast year 2050, it was believed that the hourly demand pattern would not change (Quaranta et al., 2025). However, future electrification of industries, transportation, and cooking may alter consumption patterns, particularly during peak hours. However, due to a lack of end-use data, these variations were excluded. Rather, the current hourly demand profile was scaled by three per capita electricity consumption. This premise offers a transparent and consistent foundation for evaluating system configurations under various scenarios of demand growth.

According to this concept, imports of electricity were only taken into account when domestic production and storage were unable to keep up with demand (Ucok, 2026). Because of its consistent contribution to Tunisia’s present power system, run-of-river (RoR) hydropower was modeled as baseload. Pumped hydro energy storage is the only energy storage technology covered in this study; batteries and other short-duration storage technologies were not included because of their higher prices and limited scalability for medium-to long-duration applications. When appropriate, the Tunisian Dinar is used to express all monetary values.

To ensure model transparency, several simplifying assumptions were adopted based on current data availability and the scope of this study. Their potential influence on the optimization results should therefore be acknowledged.

Wind energy was excluded from the optimization framework because the objective of this study was to evaluate the long-term role of solar photovoltaic (PV), hydropower, and pumped hydro energy storage (PHES) under a consistent and well-characterized resource database. Although Tunisia possesses significant wind resources, high-resolution hourly wind generation profiles covering all governorates were not available with sufficient temporal and spatial consistency for the considered planning horizon. Including uncertain wind generation profiles could introduce additional uncertainty into the optimization results. Nevertheless, integrating wind power would likely increase resource complementarity with solar PV, reduce seasonal variability, decrease storage requirements, and potentially lower the overall system LCOE. Therefore, future studies should incorporate wind energy as more comprehensive datasets become available.

A uniform hourly electricity demand profile was assumed for all governorates because detailed regional load data are currently unavailable. This assumption provides a consistent basis for comparing alternative development scenarios while avoiding biases associated with incomplete regional demand information. Although regional consumption patterns may differ due to industrial activity, climate conditions, and population density, the adopted assumption is considered appropriate for a national-scale long-term planning study. More detailed regional demand modelling could influence the geographical allocation of generation and storage assets but is expected to have only a limited effect on the overall least-cost system configuration.

PHES was selected as the only storage technology because it currently represents the most mature and economically competitive large-scale solution for medium- and long-duration energy storage. Unlike battery energy storage systems, which are generally more suitable for short-duration applications, PHES offers significantly larger storage capacities, longer operational lifetimes, and lower lifetime costs for balancing high shares of renewable electricity. Although hybrid storage configurations combining PHES and batteries may further improve operational flexibility, their inclusion would substantially increase model complexity and require additional technology-specific cost and performance assumptions. Future research should investigate such hybrid storage systems as battery costs continue to decline.

Hydropower generation was treated as an exogenous input under both the Reference Hydro and Ambitious Hydro scenarios because future hydropower expansion is primarily determined by national resource availability, hydrological constraints, and long-term infrastructure planning rather than by the optimization itself. This assumption allows the study to isolate the contribution of solar PV, PHES, and electricity imports under different hydropower development pathways. If hydropower capacity were considered an optimization variable, the optimal technology mix and corresponding LCOE could change depending on future investment costs, water availability, and climate conditions. Consequently, the presented scenarios should be interpreted as representative planning pathways rather than unique optimal solutions.

3 Analysis results

3.1 Overview of cost-optimal scenarios

Different hydropower development levels (Reference and Ambitious), cross-border energy import conditions, and electricity demand (2 and 9 MWh per capita by 2050) were used to analyze twelve least-cost optimization scenarios (Kafle et al., 2026). The system-level characteristics of each scenario, such as installed capacities, generating mix, storage needs, interconnection size, and the resulting LCOE, are summarized in Table 4. Overall, the scenario structure illustrates how Tunisia’s least-cost renewable energy transition is influenced by interactions between regional electricity commerce and the increase of local resources.

TABLE 4. An overview of the system requirements for the twelve least expensive scenarios in 2050.

The scenarios vary in terms of hydropower assumptions (Reference Hydro: A; Ambitious Hydro: B), import conditions (No Import and Import), and demand level (expressed as 1 (2 MWh per capita per year), 2 (5 MWh per capita per year), and 3 (9 MWh per capita per year).

3.2 Energy balance and infrastructure planning for different demand scenarios

The resulting generation and capacity mix for each of the twelve scenarios is shown in Figures 6, 7 under two hydropower availability scenarios, import conditions, and changing demand levels. Figures 6a, 7a illustrate no-import scenarios, while Figures 6b, 7b illustrate import scenarios. The model maintains constant hydropower capacities in each scenario, while optimized solar PV, PHES, and, where enabled, electricity imports are used to meet rising demand levels. The system grows more dependent on solar PV as demand increases and hydropower’s contribution of electricity generation decreases. Without altering the hydropower capacity, imports provide a flexible option that partially replaces domestic solar PV and PHES capacity (Zhang et al., 2025).

With just minor solar PV and PHES additions for diurnal balancing, hydropower generation under the reference hydro scenario satisfies the majority of the annual demand under low demand (2 MWh per capita per year).

PV and PHES requirements are further decreased by permitting imports. Greater hydropower output (about 171 TWh/year) under the ambitious hydro scenario (B1) almost eliminates the need for PV and PHES, producing excess energy that might support seasonal electricity exports. Additionally, the import-enabled situation is preferable in low demand reference hydro scenarios because it allows us to import and export.

The system shifts to a combined solar-hydro design at the medium demand level (5 MWh per capita per year). In these situations, PHES balances demand while PV generation makes up a sizable portion of yearly consumption.

In reference hydro instance (A2), allowing power imports lowers PV and PHES capacity requirements by about 21% and 41%, respectively, showing a balance medium demand configuration (Trabelsi et al., 2025b). Additionally, the massive hydro output under the ambitious hydro scenario (B2) substantially lessens the need for solar and storage, particularly in import-enabled scenarios.

PV takes over as the main component of the system at a high demand level of 9 MWh per capita per year, with PHES offering vital balancing support. The model’s upper bound configuration under land, grid, and investment constraints is represented by the A3-NoImport case, which calls for PV capacity of about 346 GW and PHES capacity of 92 GW.

However, in A3-Import, PV capacity drops to about 235 GW and PHES capacity to 53 GW, indicating how regional commerce can save costs. In an ambitious hydro, import-enabled scenario, the same demand can be satisfied with just 134 GW of PV and 37 GW of PHES. This demonstrates that the latter, which combines hydropower expansion with cross-border flexibility to minimize infrastructure requirements, is the most realistic and viable route.

These trends highlight how commerce and domestic resources interact. PHES and cross-border electricity imports can significantly minimize Tunisia’s vulnerability to dry season shortages due to its heavy hydro seasonality (Ghouil, 2025; Adejoh, 2025). In conclusion, solar PV is still a vital energy source in all scenarios, even though the expansion of hydropower and regional connections lower the amount of infrastructure investment needed. Even while ambitious hydropower development lessens the need for large-scale PV growth under lower demand levels, solar PV is still essential under high demand scenarios, confirming its vital role in achieving a dependable, 100% renewable electricity grid in Tunisia (Shrestha et al., 2025; Oyenuga et al., 2025). The overall system costs, which are further examined in Section 3.3, are directly impacted by these capacity and generation patterns.

3.3 LCOE composition under different scenarios

The LCOE for each of the 12 simulated scenarios is displayed in Figure 8, with a distinction made between those that permit energy imports (Figure 8b) and those that do not (Figure 8a). Total LCOE values vary from 164.08TND/MWh to 263.7TND/MWh across all scenarios, with no import scenarios often showing higher values. These figures, which acknowledge that actual system costs may change in response to shifting market or financing conditions, are the least-cost results obtained from the model’s optimization under current cost assumptions for established technologies including solar PV, PHES, and domestic hydropower (Nejla et al., 2025).

The B1 scenario, which combines ambitious hydropower expansion with modest per capita demand (2 MWh/year), is the most expensive condition. LCOE is approximately 334.02TND/MWh in both import and no import scenarios, mostly due to the high levelized cost of balancing (LCOB approximately 190.45TND/MWh). Such expansion would be unfeasible unless Tunisia could export significant surplus electricity. In these scenarios, there is substantial generation capacity that exceeds demand needs, leading to significant underutilization of hydropower output and requiring investment in PHES to absorb surplus (Al-Mahmodi et al., 2025). The lowest LCOE, approximately 164.08TND/MWh, is achieved by flexible import access and A1 import, which indicates low demand with reference hydro.

Since greater PV and PHES capacity are needed to handle peak and seasonal imbalance in no-import scenarios, LCOE typically rises with increasing demand. Under no-import scenarios, balancing costs account for a sizable portion of overall LCOE, especially at high demand levels like A3-NoImport, when LCOB makes up about 68% of the total. LCOE estimates are noticeably lower in scenarios including electricity imports. Lower storage needs are made possible by the inclusion of flexible imports, which lowers LCOB at all demand levels. For example, scenario A3-Import achieves an LCOE of 222.68TND/MWh, which is over 21% less than its equivalent without import. This demonstrates how regional connections may balance solar seasonality and hydro shortfalls at a lower cost.

Although import-enabled scenarios often offer reduced LCOE, the generation cost (MW) is still comparable to no-import scenarios since imports partially replace domestic generation but do not completely eliminate the need for significant expenditures in hydro and solar power. Overall, the findings indicate that while increasing domestic hydro capacity can somewhat cut costs, allowing energy imports greatly reduces system balancing costs and increases affordability. However, import-free situations are still technically possible, but at a higher cost because they require more storage and PV capacity.

3.4 Sensitivity assessment: Sensitivity to discount rate

Sensitivity analysis was carried out using reference hydro power development for the A2-Import scenario medium demand level (5 MWh per capita per year). With balanced domestic generation and cross-border imports, this scenario was chosen as a representative middle range situation. To evaluate the influence on LCOE (baseline 195.98TND/MWh), six important economic factors are changed one at a time (OAT) by ±24% around their baseline values because the least-cost optimization depends on a number of assumptions. These factors include the cost of hydropower, the discount rate, the cost of solar PV, the cost of imports, the cost of PHES capital and charging, and the cost of high voltage AC (HVAC) transmission lines. Figure 9 displays the LCOE variation (Mohammed et al., 2025).

The analysis shows that the LCOE is most sensitive to the cost of solar PV; a 24% decrease in PV costs lowers the LCOE to around 181.074TND/MWh, whereas a 24% increase in cost rises the LCOE to about 210.96TND/MWh. This emphasizes how crucial PV investment prices are to Tunisia’s renewable energy transition. As a result, ±24% fluctuation accounts for future module price decreases while capturing short to medium-term cost uncertainty. Reducing solar PV expenses will be a top objective in order to minimize the total cost of the system. Additionally, the system cost is significantly impacted by the discount rate. The significance of financing conditions is highlighted by the fact that a 24% increase in the discount rate increases LCOE by about 6.9%.

However, changes in HVAC transmission costs have a negligible impact on LCOE, indicating that transmission costs account for a small portion of the overall system cost.

According to the model, LCOE is very sensitive to the discount rate. In order to generate funding scenarios ranging from concessional to commercial, the sensitivity analysis explored discount rates between 2% and 10%. Concessional public or climate-aligned lending is represented by the 3%–5% range, whereas private sector or commercial borrowing with greater perceived danger. These findings demonstrate the crucial role of finance in system costs and are consistent with earlier studies (Mohammed et al., 2025; Zidan et al., 2025) for concessional to commercial lending. When the discount rate increases from 2% to 10%, the system LCOE goes from 178.73TND/MWh to 246.12TND/MWh, resulting in a cost escalation of almost 38% (Figure 10).

This behavior is consistent with worldwide trends in which capital-intensive systems dominated by solar PV and PHES become extremely sensitive to financing conditions because their energy output generates future cash flows but their total expenses occur at the beginning. This demonstrates that technological cost reductions are just as significant as concessional and blended financing (Zhan et al., 2025). Large-scale PV and PHES deployment would become more competitive and cheap if discount rates were kept below 4.9%. Sensitivity analysis demonstrates that in order to make Tunisia’s renewable energy future both affordable and dependable, low-cost solar deployment, low-cost finance, and effective hydropower utilization are essential.

Although the present sensitivity analysis focuses on economic parameters, other sources of uncertainty, including climate variability, hydrological conditions, renewable resource availability, electricity demand growth, and electricity import availability, may also influence the optimization results. These aspects are discussed in the Limitations section and identified as priorities for future research.

4 Discussion

With an Levelized Cost of Electricity (LCOE) ranging from 164.08TND/MWh to 263.7TND/MWh across scenarios with a balanced mix of solar photovoltaic (PV), hydropower, Pumped hydro energy storage (PHES), and regional connections, the modeling result demonstrates that a 100% renewable energy system is both technically feasible and economically competitive. This demonstrates that Tunisia can use tried-and-true renewable technology to provide a steady supply of electricity. In a low-demand, ambitious hydro scenario, when a significant hydropower surplus results in underutilization and increased balancing costs, the top bound LCOE of 263.7TND/MWh occurs. This result is not due to rising technology costs, but rather to greater curtailment and PHES cycling.

Even if all of Tunisia’s electricity comes from renewable sources, the grid is still reliant on Run-of-river (RoR) hydropower, which causes imbalance. Only a small portion of all primary energy use is still made up of electricity. Despite the fact that 99% of people have access to electricity, per capita power use is modest at 0.5 MWh, with the majority of use being restricted to simple lighting. As electrification spreads into transportation, manufacturing, cooking, and heating, there will likely be a big rise in demand, necessitating a major expansion of the infrastructure for generation, storage, and transmission. According to the model’s findings, hydropower can mostly meet low electricity demand, but as demand increases, solar PV becomes the system’s primary energy source, with PHES helping to control daily and seasonal fluctuations.

The results are in line with earlier research (Baccouche et al., 2025), which highlights the crucial role that solar PV will play in a nation’s future transition to renewable energy 344 GW of PV and 90 GW of PHES with a total storage capacity of 2,445 GWh are needed for domestic supply under the high demand reference hydropower case (A3 NoImport). These figures, however, indicate maximum capacities; they provide a more realistic and economical route to complete electrification when paired with hydropower augmentation (B3) and electricity import.

The most scalable renewable energy source is solar photovoltaics (PV), but large-scale deployment necessitates coordinated land-use, grid, and funding frameworks for practical implementation. However, the installed capacity of renewable energy sources other than hydroelectric power is limited by the existing energy policy to 10%–1% of the overall hydro capacity. Although this hydropower-centric strategy has previously made energy expansion easier, model estimates indicate that by 2050, solar PV will contribute between 40% and 60% of total generation for cost-effectiveness and reliability. There is significant solar potential throughout grassland and barren areas, according to spatial evaluations (Ye et al., 2025), and rooftop and agrivoltaic systems can increase capacity without competing with land.

These findings demonstrate that large-scale PV expansion is achievable with the help of favorable investment policies and integrated land-use and energy planning.

PHES offers balance in addition to solar PV. greater than 2,800 eligible off-river PHES sites were found by the Global PHES Atlas, with a combined potential storage capacity of almost 50,000 GWh (Bhatta and Prasad, 2025). This is significantly greater than our modeling upper bound storage need of about 2,444 GWh under high demand scenarios. Furthermore, earlier research demonstrates PHES’s contribution to peak load balancing and its economic viability in favorable market circumstances (Zhan et al., 2025). When compared to conventional reservoir systems, off-river PHES development also lessens the social and environmental effects. Therefore, the key to promoting PHES development is giving priority to feasibility studies and enabling policies. PHES sites that can supplement solar and RoR hydropower should be given priority by the government, and they should be incorporated into national energy plans through financial support and supportive legislation.

Electricity imports give the electrical system flexibility, even if PHES is crucial for balancing. According to our projections, strategic imports can further ease the strain on domestic PHES energy capacity by 40%–80%. While maintaining energy independence, a moderate interconnection capacity of 10.5–12 GW might result in significant savings. Similar trends have been noted in least-cost optimization studies in Bolivia, Malaysia, and Southeast Asia (Darmawan, 2025a; Michailidis et al., 2025b; Müller, 2025; Sanchez et al., 2025; Miled et al., 2025), where external connectivity lower storage requirements by 50%–80% without sacrificing system dependability. BANERJEE, Subhanil, GANGULY, Arnab, and KONER, Souren. 's study (Gharbi et al., 2025; Banerjee et al., 2025), which shown that cross-border energy cooperation in South Asia reduces system costs and storage requirements by utilizing solar and hydro resources, further supports the outcome.

These revelations underscore the strategic significance of regional grid connectivity for accomplishing a robust and economical renewable energy transition in Tunisia.

The dynamic interaction between governmental aims and techno-economic factors in Tunisia’s renewable energy transition is reinforced by sensitivity analysis. The significance of having access to inexpensive and favorable financing mechanisms is highlighted by the system costs’ predominant sensitivity to solar PV costs and discount rates. PV module prices have decreased by more than 80% since 2010 and are predicted to drop by a further 30%–40% by 2030 (Ashraf et al., 2025), according to International Renewable Energy Agency (IRENA), indicating that further cost reductions might drastically cut the total cost of the system.

These results are consistent with Tunisia’s long-term net-zero strategy, the Third Nationally Determined Contribution (NDC), and the Sixteenth Development Plan. These policy frameworks emphasize transmission network expansion, concessional financing, and renewable energy diversification. Revising energy mix targets and introducing dedicated regulatory and financial mechanisms for PHES development would further align national policy with the least-cost optimization results. Moreover, concessional financing from bilateral and multilateral institutions could reduce the effective discount rate to below 5%, thereby significantly improving project feasibility. All things considered, these results show the necessity of legislative initiatives aimed at reducing investment risk and encouraging low-interest financing for the growth of renewable energy.

Lastly, expanding electrification in transportation, cooking, and manufacturing is crucial to reducing reliance on fossil fuels and conventional biomass and fully using renewable production. This analysis assumes deep electrification in the end-use sectors and predicts high demand scenarios.

Tunisia’s energy strategy should prioritize increasing the number of electric vehicles, encouraging electric cooking (e.g., through appliance subsidies and rural electrification), and providing incentives for electric industrial operations. These actions will increase the affordability and dependability of a totally renewable electricity grid while also lowering emissions and reliance.

Although the optimization results demonstrate that large-scale deployment of solar PV and PHES is technically feasible and economically attractive under the analyzed scenarios, successful implementation will also depend on several non-technical factors. Large-scale PV deployment requires careful land-use planning to minimize conflicts with agriculture, biodiversity, and protected areas while maximizing the use of rooftops, degraded land, and other low-conflict locations. Similarly, PHES development requires comprehensive environmental impact assessments, water resource management, and appropriate site selection to reduce ecological impacts. Streamlined permitting procedures and stable regulatory frameworks will be essential to accelerate project development and reduce investment delays. In addition, substantial reinforcement and modernization of the transmission network will be required to integrate geographically distributed renewable generation and maintain system reliability. Access to affordable financing through public investment, international climate finance, and public-private partnerships will also play a critical role in reducing investment risks and lowering the cost of capital. Finally, public acceptance and stakeholder engagement will be essential for ensuring the successful implementation of large-scale renewable energy projects. Addressing these institutional, environmental, financial, and social challenges alongside techno-economic optimization will significantly enhance the feasibility of Tunisia’s long-term renewable energy transition.

Overall, the analyzed scenarios suggest that a diversified combination of solar PV, hydropower, PHES, and strategic electricity imports could provide a technically feasible and economically attractive pathway toward a highly renewable electricity system in Tunisia under the assumptions adopted in this study. The modeling results provide planners and policymakers with useful information to prioritize flexible and affordable options that can improve long-term energy security, lessen supply shortages during the dry season, and meet rising electricity demand. The results help achieve sustainable development goals (SDGs) in addition to national energy planning. In particular, SDG 7 (clean and affordable energy for all) is aligned with the study’s cost-effective routes for a renewable energy transition. By lowering dependency on conventional biomass and fossil fuels, increasing the use of renewable energy contributes to SDG 13 (Climate action) and helps Tunisia achieve its zero emission target (Lamichhane et al., 2025).

The sensitivity of the optimization results to the adopted modeling assumptions should also be acknowledged. The exclusion of wind energy, the use of a uniform hourly demand profile, the exclusive consideration of PHES for energy storage, and the predefined hydropower scenarios inevitably influence the optimal system configuration and the resulting LCOE. However, these assumptions were intentionally selected to maintain model transparency, ensure consistency with currently available data, and focus the analysis on the dominant renewable technologies considered in Tunisia’s long-term energy transition strategy. Future studies incorporating additional renewable technologies, regional demand diversity, alternative storage options, and climate-dependent hydrological variability are expected to provide a more comprehensive assessment of the optimal energy mix.

Although the present study focuses on long-term least-cost optimization rather than short-term operational forecasting, the credibility of the proposed model can be assessed by comparing its overall trends with previous studies and national energy planning objectives.

The optimization results indicate that solar photovoltaic (PV) becomes the dominant electricity generation technology as electricity demand increases, while pumped hydro energy storage (PHES) plays a crucial role in balancing the variability of renewable generation. These findings are consistent with previous optimization studies on 100% renewable electricity systems, which similarly identify solar PV as the primary generation source and PHES as one of the most cost-effective technologies for medium- and long-duration energy storage.

Furthermore, the reduction in system costs observed when cross-border electricity imports are permitted is in agreement with earlier regional energy system studies demonstrating that increased interconnection capacity improves system flexibility, reduces storage requirements, and lowers the overall Levelized Cost of Electricity (LCOE).

The projected expansion of solar PV capacity and the increasing importance of energy storage are also consistent with Tunisia’s long-term renewable energy transition objectives, which promote large-scale deployment of renewable energy technologies, enhancement of transmission infrastructure, and regional electricity market integration.

Although the present study employs a different optimization framework and scenario assumptions than previous investigations, the close agreement in the overall evolution of the optimal energy mix provides additional confidence in the robustness of the proposed model. Differences in the absolute values of installed capacities and LCOE are expected because of variations in demand assumptions, technology costs, hydropower availability, and system boundaries. Table 5 summarizes the comparison between the main findings of the present study, previous renewable energy optimization studies, and Tunisia’s national energy planning objectives. The comparison highlights a strong convergence regarding the dominant role of solar PV, the strategic importance of PHES for system flexibility, and the cost-reduction benefits associated with regional electricity interconnections, thereby providing additional evidence of the validity and robustness of the proposed optimization framework. Nevertheless, the consistency of the principal conclusions with published literature and national planning objectives supports the validity of the adopted optimization framework.

TABLE 5. Comparison of the proposed optimization results with previous studies.

5 Limitations

Although this analysis offers Tunisia a thorough least-cost optimization of routes to 100% renewable energy, it is crucial to recognize a number of limitations. As mentioned in Section 2.5, wind power was excluded from the optimization due to data and geographic constraints. Demand side flexibility and short-term grid stability requirements like frequency regulation, inertia, and ramping limitations have not been included in this model. Incorporating these factors could lower storage requirements and peak loads in subsequent analyses. Additionally, short-term storage, such batteries, which may work in tandem with PHES to support electric vehicle charging infrastructure or manage intra-day balance, was left out.

Seasonal reliability may be impacted by future climate unpredictability because hydropower generation is based on historical generation data. Additionally, the study bases the profile of future electricity consumption on existing demand trends. Instead of using end-use sectors, this analysis examines the whole national power consumption.

The three demand levels (2.5 and 9 MWh per capita per year) show growing electrification for transportation, industrial, and residential purposes. However, future demand patterns may be altered by electrifying these industries and implementing demand-side management and energy-efficient appliances (Gharbi et al., 2025). Future study can incorporate these modifications’ potential effects on generation dispatch patterns and storage requirements. Furthermore, whereas cross-border electricity imports provide cost and flexibility benefits, they also increase reliance on outside markets and geopolitical factors. Therefore, future modeling should assess how resilient renewable systems are to changing trade situations.

Another limitation concerns the representation of uncertainty. The sensitivity analysis performed in this study focuses on key economic parameters, including investment costs and discount rates, because these variables have the strongest direct influence on the levelized cost of electricity. However, other sources of uncertainty may also influence the optimal system configuration. These include future climate variability, changes in hydrological conditions affecting hydropower availability, interannual photovoltaic resource variability, uncertainty in long-term electricity demand growth, technology learning curves that may further reduce renewable energy and storage costs, and the future availability of cross-border electricity imports. Although these factors were not explicitly modeled in the present study, they could modify the optimal generation mix and infrastructure requirements. Incorporating probabilistic or stochastic representations of these uncertainties would provide a more comprehensive assessment of system robustness.

Consequently, the results presented in this study should be interpreted as scenario-based planning outcomes rather than deterministic forecasts, since future technological, economic, climatic, and policy developments may influence the actual evolution of Tunisia’s electricity system.

6 Conclusion

Despite having substantial potential for renewable energy, hydropower currently dominates Tunisia’s electrical system with little diversification. However, the realization of these scenario-based pathways will require supportive regulatory frameworks, adequate financing mechanisms, transmission network expansion, efficient permitting procedures, environmental safeguards, and strong stakeholder engagement. The development of hydropower is still prioritized under current regulations, while other renewable energy sources like solar PV and energy storage technologies receive relatively less attention. By electrifying important end-use sectors like transportation, heating, cooking, and industrial, this analysis suggests that Tunisia could achieve a highly renewable, and potentially 100% renewable, electricity system by 2050 under the assumptions and scenarios considered in this study. With the help of PHES and cross-border imports that provide flexibility to handle daily and seasonal variability in generation and demand fluctuations. According to the modeling results, the twelve least-cost modeling scenarios consistently identify solar PV as the dominant technology in the optimized electricity mix across most future scenarios. The LCOE can vary from 164.08TND/MWh to 263.7TND/MWh based on import scenarios, hydropower supply, and demand levels. These scenario-based results suggest that prioritizing large-scale solar PV deployment, together with PHES development and transmission expansion, could facilitate a cost-effective renewable electricity transition.

To guarantee a dependable and sustainable energy transition, long-term energy planning should also incorporate sectoral electrification, storage capacity growth, and transmission network extension. The results provide a scenario-based decision-support framework that may assist policymakers in evaluating alternative long-term energy transition strategies. Additionally, by encouraging access to clean energy and reducing emissions, this study contributes to the accomplishment of Sustainable Development Goal (SDG) 7 and 13. Other emerging nations with low per capita electricity consumption and an abundance of renewable resources can also benefit from the findings, which offer a route toward a robust and sustainable energy future.

It should be emphasized that these findings represent scenario-dependent optimization outcomes based on the assumptions adopted in this study. They should therefore be interpreted as plausible long-term planning pathways rather than definitive predictions of Tunisia’s future electricity system.

7 Future research directions

Future research should build upon this work by incorporating a wider range of renewable energy technologies and system flexibilities into the modeling framework. Key options such as wind power, concentrated solar power, battery energy storage, and green hydrogen production merit closer examination to capture a more diversified and resilient electricity mix for Tunisia. Integrating these technologies would enable a more nuanced assessment of their complementary roles in mitigating short-term variability and addressing long-term seasonal storage needs.

In addition, further studies should deepen the exploration of sector coupling by explicitly modeling the electrification of transport, industry, and residential heating and cooling. Such an approach would yield a more realistic representation of future electricity demand profiles and peak load dynamics, while clarifying their implications for grid stability and infrastructure investment. Within this context, demand-side management, smart grid deployment, and energy efficiency measures warrant particular attention as cost-effective alternatives to extensive supply-side expansion.

Another promising avenue lies in spatial and grid-level analysis. Future work could integrate regional resource distribution, transmission bottlenecks, and cross-border interconnections with neighboring countries to evaluate the technical and economic feasibility of higher levels of electricity trade. This would enhance understanding of Tunisia’s potential role within a broader North African or Mediterranean renewable energy system.

Expanding uncertainty and risk analysis is also essential. Beyond sensitivity testing of economic parameters, future research could employ stochastic modeling to account for climate variability, hydrological uncertainty, fuel price fluctuations, and policy shifts. Such methods would strengthen the robustness of long-term energy transition pathways.

Future work should also extend the uncertainty analysis by integrating stochastic optimization and probabilistic scenario approaches. Such methods would allow simultaneous assessment of multiple uncertainty sources, including climate change impacts on solar and hydrological resources, future electricity demand trajectories, technology cost learning, and the reliability of regional electricity imports. Incorporating these uncertainties would improve the robustness of long-term planning decisions and provide policymakers with more resilient renewable energy transition pathways under a wider range of future conditions.

Finally, future studies should incorporate social, institutional, and policy dimensions, including regulatory frameworks, financing mechanisms, and public acceptance of large-scale renewable projects. Considering these non-technical factors alongside techno-economic modeling would provide more holistic guidance for policymakers and support the practical realization of a fully renewable electricity system in Tunisia and other countries with comparable development trajectories.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Glossary

CAHT Corporate Human Resources and Training CAPEX Capital Expenditure DE Differential Evolution GDP Gross Domestic Product GHI Global Horizontal Irradiance GW GigaWatt GWH GigaWatt-Hour HVAC High Voltage Alternating Current IPP Independent Power Producer IRENA International Renewable Energy Agency KWh KiloWatt-hour KWh/m ^**2** /day Kilowatt-hours per square meter per day LCOB Levelised Cost of Balancing LCOE Levelised Cost of Electricity LCOG Levelised Cost of Generation MTND Million Tunisian Dinar MW MegaWatt MWh MegaWatt-hour NDC Nationally Determined Contribution NREL National Renewable Energy Laboratory O&M Operation and Maintenance OAT One-at-a-Time PHES Pumped Hydro Energy Storage PPA Power Purchase Agreement PV Photovoltaic QGIS Quantum Geographic Information System RoR Run-of-River SAM System Advisory Model SDG Sustainable Development Goals STEG Tunisian Company of Electricity and Gas TWh TeraWatt-hour TND Tunisian Dinar

Footnotes

Edited by: Solomon Giwa, Olabisi Onabanjo University, Nigeria

Reviewed by: Medani Bhandari, Akamai University, United States

Armel Zambou Kenfack, Laval University, Canada