Jairo Quirós-Tortós, Luis Victor-Gallardo
Costa Rica is internationally recognized for its fight against climate change. In 2019, the country launched its National Decarbonization Plan (NDP), presenting concrete actions in the short, mid, and long terms to reach net-zero greenhouse gas (GHG) emissions by 2050. The decarbonization of the transport sector is critical to meet this target as it accounts for more than 50% of the country’s net emissions. Plus, it can benefit from its nearly 100% renewable electricity system and its abundant renewable sources to feed the future energy demand for transportation, thus increasing energy independence.
In Costa Rica, decarbonizing the transport sector means electrifying 85% and 95% of the public and private fleet by 2050, respectively, and deploying low- or zero-emissions technologies in the freight sector according to the evolution of the market. Decarbonization is more than electrification. Decarbonization also translates to changes in the way people travel. A pillar for decarbonization is to increase the use of public and nonmotorized transport (walking and cycling) to reduce congestion and accidents and increase productivity—the so-called cobenefits of decarbonization.
Studying the decarbonization of the transport sector needs to be coupled with the electricity sector. Integrated planning is critical to truly capturing the synergies between the electricity and transport sectors—the electrification of the transport sector must be done while preserving, or even increasing, the renewability of the electricity sector.
Planning the decarbonization of a country needs good governance. The NDP design was led by the Directorate of Climate Change of the Ministry of Environment and Energy. Decarbonizing the transport sector is an economic opportunity as the country would lower its fossil fuel bills, thus providing financial benefits for households and businesses that rely on vehicles for their activities. Moreover, there will be an important reduction in local pollution and an increment of productivity from more efficient transport, which in turn will translate into economic benefits to Costa Ricans. Establishing the aspirational goals of the NDP followed a participatory approach informed by science (bottom-up technologically rich models to design prospective scenarios), which can be adopted by other countries or regions to design energy policies to advance toward net-zero targets.
In this article, we present the process followed to support the creation of the decarbonization pathway for the transport and energy sectors presented in Costa Rica’s NDP. We discuss how stakeholders participated in the design of the energy system model and how it was then used to create scenarios (pathways that map our transformations toward the midcentury). We then present the development of the integrated energy system model to study the coupling between the electricity and transport sectors while discussing key policies toward the decarbonization of transport. We then analyze the relevance of capturing long-term uncertainties in the scenarios to identify possible risks and anticipate negative outcomes. Finally, we list some challenges ahead of the decarbonization of the transport sector so that countries can increase awareness and preparedness.
Decarbonizing the transport sector starts by producing policies toward a national goal. In Costa Rica, this was done using an innovative approach that combined qualitative and quantitative methods. We executed workshops with multiple stakeholders to capture ideas and concerns. The range of stakeholders varies per country, but generally, the process involves the ministers of energy, environment, transport, finance, and planning. Figure 1 shows the overall process followed and its connection to the formulation of climate change policies, which are the result of an agreement between stakeholders. A stakeholders-driven back-casting participatory approach was initially adopted to establish the policy packages in the decarbonized future.
Bilateral meetings were carried out first between high-level government actors (president, ministers and vice-ministers, and executive presidents of autonomous institutions), the private sector, civil society, and academia. These meetings not only allowed an understanding of the different priorities of the sectors and the definition of a national vision, but they also conveyed robustness to the process as key actors were involved from the genesis of the NDP. These meetings generated the base structure of the plan, consisting of 10 lines of action, eight cross-cutting strategies, and three periods of implementation. Table 1 summarizes the lines of action and cross-cutting strategies. The three periods of implementation are:
Table 1. A summary of the lines of action and cross-cutting strategies in the NDP.
As part of these high-level discussions, two scenarios were defined: 1) a business as usual (BAU) scenario projecting the current evolution of emissions caused by the growing demand for fossil fuels and ignoring the effects of future climate policies and 2) the NDP scenario that promotes public transport and electromobility consistent with net-zero emissions by 2050.
This participatory approach also helped define the narratives around the decarbonization pathway that were going to be quantitatively analyzed using modeling tools (formally discussed later). The modeling tools were then used to help actors gain perspective, clarity, and agreement about the transformations needed to reach decarbonization and inform the country’s energy policy.
Energy modeling—using mathematical models to generate insights about the future of energy systems—can provide policy support when based on rigorous analytics and good governance. It must capture the synergies between real-life-linked sectors. Without this sector coupling, the models may not represent the true transformations needed. For instance, modeling the electricity and transport sectors together allows the understanding of power grid needs caused by the ongoing electrification of the transportation fleet.
Energy modeling enables understanding synergies between systems, e.g., electricity and transport, electricity and gas, and electricity, transport, and gas. While energy modeling offers policy insights for strategic planning, other challenges of integrating renewable energy sources and zero-emission vehicles will also have to be dealt with using other tools and studies (e.g., power flow tools to assess the effects on the electricity transmission and distribution networks). With adequate tools, it is possible to understand the extent to which the transport sector can support balancing grid operations. For instance, advanced power flow tools can help explain the benefits of using technologies such as vehicle-to-grid, smart charging, and dynamic end-user pricing on the grid to justify the deployment incentives or time-of-use-tariff to encourage users to connect their electric vehicle (EV) to charge or discharge the battery for grid-balancing needs.
Energy modeling allows an understanding of the macro effects in systems. Unlike electricity, gas, and transport modeling, where detailed network models are typically used, energy modeling can shed light on the big picture of systems and their overarching interactions to then perform detailed network assessments. For instance, once a country defines policy targets, power grid-planning studies can be done to identify bottlenecks in the transmission and distribution network and thus develop alternatives.
Many countries in Latin America have used integrated models to investigate the transformations needed to achieve net-zero emissions by midcentury. These models are commonly developed by local universities with international collaboration, including the financial support of multilateral development banks; they are built using local data and stakeholder engagement. While some countries are using licensed software for developing their energy policy, there is an ongoing appetite in the region to move toward open source and free tools as this increases transparency, leads to codevelopment with the international community, and reduces costs.
In Costa Rica, we use the open source platform called Open-Source Energy Modelling System (OSeMOSYS). We built, in collaboration with stakeholders (see the next section), the OSeMOSYS-CR—a Costa Rica energy system model. Figure 2 shows the reference energy system of Costa Rica. OSeMOSYS-CR does not perform power flows. Instead, it maps out the entire value chain of the energy system: from production or importation to consumption. It can capture energy flows between sectors, e.g., electricity for transport. Once the policy scenarios are defined (e.g., 85% of the electrification of the fleet), detailed tools are used to understand the extent to which transmission and distribution networks can cope with these penetrations or to assess the grid upgrades needed to enable that penetration.
Being an energy system optimization model, OSeMOSYS-CR finds a least-cost combination of technology use and investment as a function of relative costs of competing technologies, input–output relations, exogenous demands, and system constraints. Its energy supply chain logic allows the endogenous accounting of capacity, activity, emissions, and expenses disaggregated into the capital expenses, operation expenses, and external costs of the selected technology options.
The model was parametrized using local data provided by stakeholders (e.g., the energy balance) and international information available at well-known databases (e.g., costs of technologies and fuels). A combination of constraints is used to reflect the system’s physical, environmental, and societal boundaries and constitutes a scenario—a scenario represents the evolution of the system to a given planning horizon: 2050 in Costa Rica. We created a BAU scenario that expands the 2018 energy system to 2050 and an NDP reflecting road transport decarbonization based on the NDP until 2050.
Table 2 and Figure 3 present key transformations and targets, respectively, of the BAU and NDP scenarios. The decarbonization scenario implies the modal shifts and reductions in distance traveled promoted by compact cities. It involves an increase in motorized passengers using public transport that grows to 50% of motorized kilometers traveled by 2050 compared to often significantly lower figures (down to 40%) in the BAU scenario. Modal shift and lower distance traveled consequently imply a reduction in the demand for passenger kilometers in private transport (from 61% in 2015 to 40% in 2050). The NDP envisions that, by 2050, public transport should cater to most of the demand in metropolitan areas and that nonmotorized modes (including walking and cycling and reduced demand due to the digitalization of jobs and teleworking) should increase their contribution to 10% of the demand by 2050. The NDP also implies a technological pathway to zero or low emissions technologies reaching 85% and 100% in the public, private, institutional, and taxi fleets. Hydrogen-based buses and minibuses are also a part of this technological pathway to net-zero emissions, reaching a 10% penetration of the total fleet by midcentury. This pathway also involves a transformation in the freight sector, including battery- and hydrogen-based technologies. Biofuels also play an important role in emissions reduction, along with the reduced use of liquified petroleum gas (LPG) in the commercial and residential sectors. Finally, the electricity mix will reach 100% renewability by 2050 according to the NDP goals.
Table 2. The key transformational lines of actions in the NDP scenario for the energy sector.
The scenarios were then simulated to understand the evolution of sectorial emissions. We found that they can be reduced by midcentury to 7.2 metric tons of carbon dioxide equivalent (MtCO2eq) (Figure 4). The reduction occurs primarily with an increase in the renewability of the energy mix and a reduction in the use of fossil fuels (primarily in transportation as electricity in Costa Rica is 99% renewable). The electrification of the transport will require new renewable power plants and higher efficiency to produce electricity. The transition requires about 3 GW of additional installed capacity by 2050 compared to the BAU scenario (Figure 5). Their deployment is needed primarily after 2035 when massive deployment of zero-emission vehicles starts to take place. Deploying solar and wind technologies after 2035 implies that investments may not be significant since the costs of these renewable energy sources are decreasing in time (i.e., investments deferral).
The decarbonization process requires investments that are compensated with reduced operational costs. Deploying zero- or low-emission technologies may lead to higher investment costs today, but their operation is in almost all cases cheaper. In addition, the costs of zero-carbon technologies are dropping rapidly, whereas the BAU scenario is becoming more expensive and exposed to transition risks, including stranded assets (i.e., investments that then become a barrier to reach decarbonization because they have to be used until they are paid off). The cost of batteries for EVs has also seen a six-fold reduction in just eight years, which is expected to continue. Compared to the BAU scenario, the study finds that the NDP scenario requires additional investments of US${\$}$26.7 billion by 2050 that are contrasted with savings of US$29.7 billion by the same period (Figure 6), thus leading to a positive net financial benefit of US${\$}$2.9 billion [approximately 5% of current Costa Rica’s gross domestic product (GDP)].
Cobenefits could exceed the decarbonization investment costs. Decarbonizing the transport sector brings opportunities to improve mobility, reduce local air pollution, and improve the quality of life. Time lost in congestion and the cost of accidents are also expensive problems. In Costa Rica, it is estimated that time lost due to congestion, accidents, and the health impacts of local air pollution cost the country 3.8% of its GDP annually. Moving to efficient public transport systems and to EVs could be one of the greatest opportunities to support the transition to net-zero emissions while bringing substantial benefits to the economy and society. An effective urban transport system based on electric buses can cut congestion, accidents, and local pollution while taking advantage of renewable electricity and saving money. If the benefits of reduced congestion, accidents, and local pollution are included in the analysis, the net economic benefits of decarbonizing the transport sector in Costa Rica would represent about US$20.6 billion (Figure 6), which is equal to around 35% of 2019 GDP.
Investments made today enable the benefits in the mid and long term. The timeline of these socioeconomic benefits (Figure 7) indicates that investments surpass savings in the short term (2020–2030). The investments are related mainly to the initial roll-out of EVs in the private fleet, the electrification of light freight transport, the first phase of the passenger train deployment, and new renewable power plants. It also captures investments in the charging infrastructure. While investments are still needed in the mid and long terms, operational savings (highly related to lower fossil fuel consumption) always compensate capital costs, thus bringing net financial benefits. In these two periods, the investments in low-carbon freight trucks and their charging infrastructure are predominant. Furthermore, it includes the second phase of the electric passenger train and the transition to more efficient public transport while continuing the electrification of buses and private vehicles. In the energy sector, the renewable energy infrastructure continues to support the technological transition. If we add up the benefits in terms of health, congestion, and accidents, the socioeconomic benefits in the mid and long terms represent almost 13% and 26% of Costa Rica’s current GDP, thus highlighting that benefits are greater as the midcentury goal is reached.
The long-term magnitudes of costs and benefits can vary widely because of deep uncertainties that affect the estimation models. Deep uncertainties are a condition in which analysts and decision-makers do not know or agree upon the probability distribution of the uncertainties about key parameters in the model. For example, the unit cost of battery EVs (BEVs) in 2050 relative to 2022 can increase or decrease depending on how the market evolves and how the supply chain develops. Similarly, oil prices will depend on the future supply and demand, which will depend on derivative end-consumer markets and supply chain conditions. While publications like the International Energy Agency’s World Energy Outlook describe scenarios with some key associated variables by 2030 and 2050, parameters can vary significantly and change with each publication year after year, particularly with unforeseen shocks like the war between Russia and Ukraine. Moreover, the worldwide scenarios may not entirely apply to small- and low- or middle-income countries like Costa Rica as some more specific uncertainties may apply.
Despite the deep uncertainties that can affect the real implementation of a decarbonization plan, the plan is a valuable guide with sectoral goals to reach through other policy instruments. Policymakers can find ways to make the plan more robust. For instance, they can adapt policies that meet the desired objectives as new information becomes available or as the situation changes. Alternatively, they can define policies that perform well, compared to the alternatives, over a wide range of plausible futures.
One approach is Robust Decision Making (RDM), one of several methodologies that support decision making under deep uncertainty. It is a set of concepts, approaches, and tools that use simulations to explore and find scenarios of interest. The underlying paradigm consists of using exploratory modeling and scenario discovery algorithms to understand which model inputs best explain good outcomes. Applications of RDM to Costa Rica’s energy planning can be found in the literature: the cost-benefit analysis of the decarbonization plan and a subsequent research study on robust decarbonization pathways. This section summarizes how the RDM method can be applied to make transport decarbonization policy robust, following the lessons learned from the Costa Rican case. It also presents and discusses the policy insights that previous studies have generated for Costa Rica.
Figure 8 shows a framework that can generate insights for a robust energy decarbonization policy. There are three fundamental actors: decision-makers, stakeholders, and citizens. The goals, objectives, and preferences of decision-makers and stakeholders feed the design of policy levers to implement in both public and private settings. The policies transform the systems involved in energy decarbonization, which inherently include the electricity and transport systems. Due to the policy levers’ effects, these systems will have technical, economic, and environmental metrics that can be monitored. An additional actor, the decision supporter, could be defined to help the decision-makers better understand possible futures through quantitative computational modeling.
External forces can also affect the metrics of the system. Since energy systems have long-lived infrastructure requirements, their planning must consider multiple years ahead. The horizon expands further when considering decarbonization objectives because investments in the next few years can determine whether a country or region reaches an emissions reduction goal in three decades. Thus, deep uncertainties become relevant as an external force affecting system outcomes.
While decarbonization aims to reduce emissions from supplying societal energy demands, even if defined as reaching economy-wide net-zero emissions, policies that affect energy and transport have other development objectives, such as
Hence, a decarbonization plan must also meet those objectives, and the policies become multiobjective. If a plan does not help advance those central energy planning goals, the decarbonization process may not advance. Furthermore, pursuing misaligned strategies that are too expensive would also prevent any significant progress.
The RDM method adopted in Costa Rica followed four steps:
The RDM process (steps 1 to 4) is flexible because each step will be customized to the context of a country or region. For example, when defining levers in step 1, different options will be available depending on geographic factors, like the availability of hydropower potential to increase renewable power generation. Additionally, an assortment of modeling tools, including energy systems models, economic models, and climate models, may be employed to generate simulations with quantitative estimations; in some cases, it might be necessary to develop new tools from the ground up. The choice of modeling features can be selective, based on the problem framing, and it’s not always essential to employ all the engineering modeling capabilities to address specific policy questions. High-level estimates with well-documented caveats can generate insights that will start to steer the formation of a policy. Only when a specific technical issue needs a more sophisticated model would it be necessary to increase the complexity of the simulations.
Table 3 shows an XLRM matrix for the Costa Rican case studies. The cost-benefit analysis of Costa Rica’s decarbonization plan considered many uncertainties because it is a plan with objectives by 2050. The classification of the uncertainties, as in Table 2, eases the interpretability of the analysis. Not all possible uncertainties were included in the case studies, for example, the variation of other renewables’ capacity factors or load dependencies on climate variation. These were analyzed in subsequent studies.
Table 3. XLRM matrix from the Costa Rican case studies.
The XLRM matrix shows five relevant metrics of interest as expressed by stakeholders during the cost-benefit analysis of the decarbonization plan. However, not all metrics were covered in that study but were analyzed in subsequent research. Crucially, the metrics show that energy decarbonization success transcends emission reductions. The economic metrics describe whether decarbonization will generate economic benefits, low end-user prices, and a relatively low financial effort to transform the system (capital expenditure requirement). Each metric has a merit that other metrics do not cover directly.
The cost-benefit analysis of the decarbonization plan found that high GDP growth can increase emissions significantly because of high transport growth. If the electrification of transport does not rise, emissions reductions will be challenging to achieve. It also found that a combination of low use of public transportation, high freight transport demand, and expensive EVs would make transport decarbonization have low economic benefits. If transport electrification costs are high for end users, then the government will have to emphasize its support of public transport use, cycling, and walking as alternatives to private transport for citizens. The benefits of decarbonization are greater under high economic growth because the cost of keeping the existing fossil fuel-based transport system can be high, considering that internal combustion engine vehicles (ICEVs) would have higher maintenance costs than their EV counterparts.
A complementary analysis found that Costa Rica’s desirable outcomes depend on very affordable BEVs and energy infrastructure unit costs, including renewable generation, energy storage, charging stations, and electrical grid reinforcements. If fossil fuels are expensive, then decarbonization will be more beneficial. Since Costa Rica is a technology importer, it will depend on international efforts to keep costs low (real prices) through robust and well-funded supply chains. The country is also a fuel importer, and a cost penalization on fuel prices (e.g., a carbon tax) in the producing (exporting) countries could accelerate its decarbonization.
Table 4 shows a more exhaustive compilation of drivers and uncertainties that explain desirable and undesirable outcomes. A desirable outcome is, for example, a driver that would result in low emissions and high socioeconomic benefits. An undesirable outcome refers to drivers or uncertainties that may result in either high emissions or low socioeconomic benefits. From the perspective of the RDM methodology, these undesirable outcomes are termed risks because they may pose a challenge to the political acceptability and feasibility of decarbonization efforts due to their negative implications. Each column corresponds to one outcome (desirable or undesirable, in 2030 or 2050). Each row has the drivers mapped in Table 3 that proved to be relevant for the five analyzed metrics. The values under each column indicate the magnitude, and the associated symbol (“<” or “>”) specifies the direction of the relation for each driver (specified in each row with the corresponding units). For instance, “GDP growth (%) < 3.5” implies that the driver is associated with GDP growth rates of less than 3.5%. All the values under a column must be considered concurrently to explain desirable or undesirable outcomes for each year. The results in Table 4 were generated from analyzing 4,001 simulations of OSeMOSYS-CR with the patient rule induction method algorithm (i.e., a scenario discovery algorithm). The decision supporters should translate the findings of the analytics into insights that can help decision-makers formulate robust policies.
Table 4. Concurrent combinations of uncertainties and levers that produce desirable and undesirable outcomes.
Once the desirable outcomes are identified, decision-makers can establish robust policies for transport decarbonization. In Costa Rica, the analysis highlighted the need to invest in public transport. Increasing the share of passenger kilometers covered by public transport allows the economy to grow without an increase in emissions and transport costs typically associated with fossil fuel-based private vehicles. Consequently, the country started piloting BEVs and fuel-cell EVs (FCEVs) for public transport to understand the routes where the technology could be deployed. The government also changed the regulation and increased the concession duration from seven to 15 years for those bus companies that would deploy zero-emission vehicles. These are policy levers that could make a significant difference as they show political willingness. Currently, there are also ongoing efforts to pilot 10 FCEV heavy trucks by 2025. The analysis also supported decision-makers’ understanding of other levers like increasing telework, nonmotorized transport, ridesharing, city densification, and logistic hubs that could reduce the demand for passenger and freight kilometers, thus leading to benefits. If those kilometers do not grow substantially, it will be easier to make them run on renewable electricity than fossil fuels. Hence, robust energy decarbonization needs to enable both large-scale electrification and other technological policies that can control transport demand. For example, electric trucks may arrive late in Central America, and freight rail can help reduce costs and emissions. The technological policies can increase the productivity of Costa Rican businesses and reduce transport and energy costs for households; these combined effects can help the country advance development goals. For example, decarbonization can also generate more jobs than the BAU scenario. A subsequent study found that decarbonizing the energy sector (electricity, industry, and transport) could translate to 135,000 additional jobs.
While modeling insights can help build robust plans, their progress needs to be monitored and possible contingencies defined. Moreover, decarbonization strategies need to transform into actionable road maps for specific sectors. The strategic level of the decarbonization plan and its cost-benefit analysis was the first stage in a continuous analytical research effort to understand the best course forward for new problems arising as the energy system decarbonizes. Making the plan robust should be an iterative process that adapts it as new information becomes available. A robust plan would suggest new actions for decision-makers in an ever more specific manner.
One robust plan should consider the long-term fiscal effects of decarbonization as the current energy system generates government revenues. In Costa Rica, transport decarbonization would make the government not receive 0.41% of GDP on average between 2023 and 2050 relative to a BAU scenario because about a fifth of current revenues are associated with fuel consumption from and a property tax on internal combustion vehicles. A BAU scenario would perpetuate this revenue into the future, and decarbonizing would imply an unperceived revenue that could significantly affect public finances. Therefore, the country needs to prepare for an eventual reform that redistributes decarbonization benefits: about 1.49% of GDP in the same period.
Since the benefits of transport decarbonization are greater than the unreceived fiscal revenue relative to the BAU scenario, gradually modifying the tax structure could be helpful. For example, existing BEV subsidies should be removed once the uptake of EVs is high. Moreover, taxing vehicle kilometers through tolls to account for driving externalities would cover EVs, compensating partially for the drop in revenue. These reform strategies would have to be introduced once transport electrification is high; a robust plan should consider them in advance because of the political process required to implement them.
Finally, the transformation will advance if the zero-emission technologies enable productivity increase for Costa Ricans and the public perceives it so. Moreover, the insights generated for a robust plan must translate into specific policy instruments and investments that will yield returns to investors and have sustainable business models. Future work must focus on generating quantitative insights for specific actions (deriving from the strategic plans) that benefit stakeholders and citizens. For example, accessing the financing for specific transport demand management investments is a challenge that the government and the private sector need to solve in the next few years to unlock a robust decarbonization process.
Costa Rica is a small country with big ambitions. Decarbonizing the transport sector is critical to meet its goal of net-zero emissions by 2050 as this sector currently accounts for about 51% of the emissions.
This article has shared the experience of Costa Rica shaping its energy policy to enable transformational changes to fight against climate change. Countries worldwide can follow the process by starting with good governance and science-based studies to support the decision-making process.
Countries need to establish goals that would suffice multiple objectives. Having workshops or bilateral meetings with the different stakeholders could bring perspectives that may not be foreseen by energy-related experts (e.g., the impact of decarbonization on revenue). It would also help to create a sense of ownership that helps its implementation.
Assessing the costs and benefits of decarbonizing the transport sector necessitates innovative approaches and tools that transcend the scope of traditional power system planning studies, including but not limited to power flow analysis, as it calls for strategic foresight. RDM is one of such possible tools as it provides tools to deal with uncertainty. Many countries have an array of tools, but these are predominantly tailored to unravel the techno-economic facets of the electricity sector, often overlooking the broader financial or economic considerations that are critical in the decision-making process necessary to achieve objectives as ambitious as deep decarbonization.
The design of energy policy in countries will vary; however, some recommendations for better coupling this with the energy modeling efforts are as follows:
The works discussed here have been supported by the Interamerican Development Bank, the World Bank Group, the French Development Agency, and the United Nations.
“National decarbonization plan of Costa Rica 2018 - 2050,” Government of Costa Rica, Costa Rica, 2019. [Online] . Available: https://unfccc.int/sites/default/files/resource/NationalDecarbonizationPlan.pdf
D. G. Groves et al., “The benefits and costs of decarbonizing costa Rica’s economy: Informing the ‘Implementation,’ of Costa Rica’s national decarbonization plan under uncertainty,” Inter-American Development Bank, Washington, DC, USA, 2020. [Online] . Available: https://publications.iadb.org/en/benefits-and-costs-decarbonizing-costa-ricas-economy-informing-implementation-costa-ricas-national
G. Godínez-Zamora et al., “Decarbonising the transport and energy sectors: Technical feasibility and socioeconomic impacts in Costa Rica,” Energy Strategy Rev., vol. 32, Nov. 2020, Art. no. 100573, doi: 10.1016/j.esr.2020.100573.
L. Victor-Gallardo and J. Quirós-Tortos, “Techno-economic comparison of centralized and distributed power generation to support large-scale transport electrification in Costa Rica,” Transport Policy, vol. 131, pp. 120–138, Feb. 2023, doi: 10.1016/j.tranpol.2022.12.009.
V. A. J. Marchau, W. E. Walker, P. J. Bloemen, and S. W. Popper, Eds, “Decision making under deep uncertainty: From theory to practice,” Rand Corporation, Santa Monica, CA, USA, 2019. [Online] . Available: https://www.rand.org/pubs/external_publications/EP67833.html
L. F. Victor Gallardo, “Robust energy system planning for decarbonization under technological uncertainty: From transport electrification to power system investments,” M.Sc. thesis, Universidad de Costa Rica, Faculty of Eng. San Jose, Costa Rica, 2022. [Online] . Available: https://www.kerwa.ucr.ac.cr/handle/10669/87273
Jairo Quirós-Tortós is with the University of Costa Rica, San Jose, Montes de Oca 11501 Costa Rica.
Luis Victor-Gallardo is with the University of Costa Rica, San Jose, Montes de Oca 11501 Costa Rica.
Digital Object Identifier 10.1109/MPE.2023.3308234
Date of current version: 19 October 2023
1540-7977/23©2023IEEE