Next Article in Journal
Multi-Market Joint Trading of Distributed Resource Aggregators Considering a Carbon–Green Certificate Linkage Mechanism
Previous Article in Journal
Reinforcement Learning-Based Adaptive Control for a Permanent Magnet Synchronous Generator Connected to a Hybrid AC/DC Grid with Virtual Inertia Support
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Beyond Green Growth: Evaluating the Biophysical Feasibility of Rail-Centric Transport Systems in the European Union

1
Centre de Recerca Ecològica i Aplicacions Forestals (CREAF), 08193 Cerdanyola del Vallès, Spain
2
Department of Earth and Ocean Dynamics, Universitat de Barcelona, 08007 Barcelona, Spain
3
Institut de Ciències del Mar (CSIC), 08003 Barcelona, Spain
*
Author to whom correspondence should be addressed.
Sustainability 2026, 18(14), 7402; https://doi.org/10.3390/su18147402
Submission received: 8 June 2026 / Revised: 3 July 2026 / Accepted: 14 July 2026 / Published: 20 July 2026
(This article belongs to the Section Sustainable Transportation)

Abstract

Achieving the European Union’s (EU-27) 2050 climate neutrality goal requires a drastic reduction in transport emissions. This study utilizes the pymedeas2 integrated assessment model to evaluate trade-offs between technology-led and structure-led transitions under both continuous growth and steady-state economic paradigms. Our results reveal that relying primarily on private vehicle electrification falls short of emission targets. The baseline REF-G scenario—following current institutional roadmaps centred on rapid technological substitution and sustained economic growth—maintains a high final energy demand and requires a cumulative extraction of 2.35 Mt of lithium by 2050, claiming nearly 6.4% of current global proven reserves solely for European mobility. Conversely, combining a modal shift toward electrified rail with macroeconomic stabilization (RAIL-SSE) reduces transport final energy demand by 68% relative to the projected 2024 peak and decreases lithium requirements by 57%. This sufficiency-driven pathway achieves the deepest absolute climate mitigation, dropping residual transport emissions to approximately 90 MtCO2/year. Furthermore, despite the front-loaded costs of rail expansion, RAIL-SSE emerges as the least capital-intensive pathway, requiring a total investment of USD 19.83 trillion—a systemic saving of USD 7.27 trillion relative to the REF-G baseline. We conclude that reaching absolute sustainability in the EU transport sector necessitates a policy shift away from resource-intensive green growth strategies toward demand sufficiency and durable public infrastructure.

1. Introduction

The transport sector currently stands as the primary source of greenhouse gas (GHG) emissions in the European Union, accounting for approximately 25% of the total as of 2024 [1]. While other economic sectors have collectively achieved a 44% reduction since 1990, transport emissions have increased by 21% over the same period, structurally plateauing as efficiency gains are consistently offset by rising activity [2]. This persistent decoupling failure is driven by a combination of increasing road freight, growing car ownership, and a shrinking share of rail, leaving the sector 93% dependent on oil products [3,4].
To reverse this trajectory, current EU policy, anchored in the Green Deal and Fit for 55 packages, mandates a 90% reduction in transport-related emissions by 2050 [5,6]. This institutional strategy is predicated upon a green growth paradigm that seeks to achieve absolute decoupling through accelerated technological substitution. Such a transition is propelled primarily by mass electrification and systemic efficiency gains; yet, it largely overlooks the structural challenges posed by the continuous growth in mobility demand [7]. However, recent literature raises significant concerns regarding the biophysical feasibility of this techno-centric approach [8,9]. These studies argue that the rates of decoupling required are historically unprecedented and unlikely to be sustained, while the massive material requirements for battery production and renewable infrastructure pose severe risks of supply-chain bottlenecks and ecological shifting.
García-Olivares et al. [10,11] propose a strategic shift toward electrified collective transport as a primary mechanism to remain within absolute biophysical limits, arguing that high-capacity rail can provide essential mobility services with significantly lower energy and material demand compared to private road transport. Following this framework, this study operationalizes and quantifies the feasibility of a large-scale modal transition from road transport to electrified rail systems within the EU-27. Our primary objective is to quantify how prioritizing electrified rail can mitigate systemic constraints frequently overlooked in mainstream policy pathways, such as critical raw material bottlenecks in Lithium and the intensification of peak energy demand associated with universal private car electrification [12,13].
Finally, we investigate the frontier of sufficiency and post-growth strategies as a necessary complement to technological and modal shifts. As defined in the IPCC Sixth Assessment Report (AR6), sufficiency encompasses policy measures that avoid the demand for energy and materials while ensuring well-being within planetary boundaries [14]. To operationalize this, we explore scenarios specifically testing the impacts of a post-growth strategy where GDP per capita growth gradually decelerates to reach zero by 2040, remaining constant thereafter. We frame both continuous economic expansion and post-growth stabilization as distinct hypotheses to be objectively tested against the biophysical, material, and energy constraints of the European transport system.
To address these complexities, we integrate a new detailed transport module into the pymedeas2 framework [15]. While widely used institutional optimization tools like PRIMES or TIMES focus on cost-effective pathways, they frequently treat critical raw material availability as an exogenous factor, thereby ignoring potential supply-chain bottlenecks in Lithium, Nickel, or Platinum group metals [12,16]. Furthermore, empirical analysis suggests that mass electrification could induce non-negligible spikes in peak energy demand and operational costs, potentially exceeding current power grid capacities [13,17,18]. By contrast, pymedeas2 explicitly internalizes biophysical limits and dynamic feedback loops linking the economy, energy, and environment. This architectural design directly connects our theoretical framing to our methodology, allowing us to quantitatively stress-test the green growth paradigm by evaluating whether continuous economic expansion remains feasible when strict material and energetic constraints are endogenized. We first assess a Reference Scenario aligned with current institutional roadmaps, adopting the projections from the Clean Energy Technology Observatory (CETO) [19]. We contrast this with a Rail Scenario grounded in the material feasibility analysis of García-Olivares et al. [11], which prioritizes electrified collective rail systems over individual road transport to minimize the overarching mineral footprint.
This study makes three primary contributions to the fields of Integrated Assessment modeling (IAM) and Ecological Economics. First, it establishes a transparent framework for assessing transport decarbonisation within an IAM that treats material availability as a binding constraint. Second, it quantifies the trade-offs between a technology-led transition and a structure-led transition, specifically regarding energy demand, the related CO2 emissions, and critical material requirements. Finally, it derives policy-relevant insights for the EU, identifying the necessary conditions—such as managed GDP stabilization and structural modal shifts—under which a high-well-being mobility system can remain compatible with climate neutrality targets.

2. Materials and Methods

The methodology of this study is designed to quantify the biophysical and economic implications of structural transport shifts within the EU-27 through 2050. We employ a simulation-based framework to evaluate how different modal configurations and macroeconomic trajectories interact with finite material and energy resources. To achieve this, we define divergent decarbonization pathways—contrasting electrification-centric and rail-centric approaches—and subsequently test their robustness against material and energetic constraints. The stability of these results and the influence of key input parameters are further validated through a global sensitivity analysis, detailed in Appendix A.
The quantitative execution of these scenarios is performed using an upgraded version of the pymedeas2 modeling framework [15,20]. This tool internalizes the non-linear feedbacks between economic activity, transport demand (pkm and tkm), and the resulting mineral requirements for battery and infrastructure stocks. Specifically, the framework couples sectoral economic outputs with energy intensities to determine the total fuel consumption and subsequent CO2 emissions associated with each transport mode. While the model provides the computational core, the specific mathematical formulations of the transport module, the updated 49-sector economic aggregation based on EXIOBASE v3.9.6 [21], and the detailed causal chains are documented in Appendix B.

2.1. Scenarios Definition

To evaluate the trade-offs between technology-oriented and structure-oriented transition strategies, we define two primary decarbonization pathways: the Reference Scenario, which follows institutional technology-substitution roadmaps, and the Rail Scenario, which prioritises structural modal shifts to mitigate biophysical constraints. To further isolate the impact of economic scale, each pathway is evaluated under two macroeconomic variants: a continuous growth case, aligned with official projections, and a Steady-State case, characterized by a transition toward macroeconomic stabilization. This dual-track approach allows for a rigorous assessment of how technological choices and macroeconomic paradigms interact to shape energy requirements, CO2 emissions, monetary costs, and material demand through 2050. The fundamental differences in modal partition between the institutional paradigm and the rail transition are summarized in Figure 1. These configurations serve as the end-state targets for our simulations through 2050.

2.1.1. Reference Scenario: Institutional Green Growth (CETO 2025)

To establish a baseline trajectory compliant with institutional ambitions, we define a reference scenario based on the POTEnCIA model [22] and the CETO 2025 scenario [19], which operationalizes the targets of the European Climate Law [23,24]. As a proxy for the “Green Growth” paradigm, this pathway assumes absolute decoupling through rapid technological substitution and efficiency gains, projecting a 38% reduction in final energy consumption by 2050 despite continued economic expansion. While the strategy relies on pervasive road electrification—with battery electric vehicles exceeding 90% of the light-duty fleet—and the scale-up of renewable fuels, projections indicate a significant decarbonization lag where emissions do not fall below 1990 levels until 2032. We construct this reference within the pymedeas2 framework by adopting CETO technological and activity outputs as exogenous drivers, ensuring alignment with current EU climate policy while internalizing biophysical feedbacks for direct comparison with alternative structural pathways.
The Reference Scenario integrates institutional fuel-switching mandates for aviation and maritime sectors from ReFuelEU Aviation [25] and FuelEU Maritime [26] as projected in the CETO 2025 report [27]. Inland transport follows an electrification roadmap implementing the 2035 internal combustion engine phase-out, resulting in battery electric vehicles (BEVs) exceeding 90% of the passenger fleet by 2050. Freight dynamics reflect a heterogeneous transition with near-total electrification of Light Commercial Vehicles (LCVs) and a 2050 Heavy Goods Vehicle (HGV) mix of 66% BEVs and 20% hydrogen fuel cells [28]. Finally, rail activity growth is aligned with TEN-T expansion projections—84% for passengers and 68% for freight—to establish a consistent baseline for comparison with alternative modal shift strategies [29].

2.1.2. Alternative Pathway: Rail Scenario

The Rail Scenario operationalizes a structural shift from individual road mobility to electrified collective rail, building on the framework established by García-Olivares et al. [11]. Unlike technology-led substitution, this pathway targets the “decarbonization lag” inherent in the slow turnover of the European vehicle stock, which currently averages 12.5 and 14 years for passenger cars and heavy-duty vehicles, respectively [30]. The scenario simulates a fundamental reallocation of activity, projecting a 50–60% contraction of the private car fleet by 2050. Inland logistics follow an “All-Rail” logic, where road transport is capped at 15% of total ton-kilometers, reserved for last-mile distribution, while inter-urban freight is absorbed by electrified rail networks.
By 2050, rail transport reaches 100% electrification via overhead catenary systems, minimizing the need for large-scale battery storage. The remaining road fleet transitions entirely to electric propulsion for light-duty vehicles and buses. The freight transport segment (the 15% of activity remaining on road) operates on a mix of 90% battery or catenary-electric and 10% hydrogen fuel cells, with the latter strictly capped to preserve Platinum-group metal reserves.

2.1.3. Macroeconomic Variants: Growth vs. Steady-State Economy

To isolate the impact of structural modal shifts from the influence of total economic volume, each pathway described above is evaluated under two distinct macroeconomic trajectories. This dual-track approach generates four distinct scenarios (REF-G, REF-SSE, RAIL-G, and RAIL-SSE), allowing the pymedeas2 framework to capture the dynamic feedbacks between economic output, energy demand, and material constraints in a system-dynamics approach.
The first variant, the Growth trajectory (REF-G and RAIL-G), adopts the official macroeconomic projections aligned with the CETO 2025 and EU reference scenarios [31]. In this case, GDP per capita is assumed to follow a sustained growth trajectory through 2050. This variant serves as a “stress test” for technological and structural solutions: it evaluates whether deep electrification or radical modal shifts can remain biophysically feasible if the total volume of economic activity—and its associated mobility demand—continues to expand, as posited by proponents of green growth paradigms [7].
The second variant, the Steady-State Economy trajectory (REF-SSE and RAIL-SSE), implements a shift away from the paradigm of perpetual expansion [32]. In this trajectory, GDP per capita growth is parametrized to gradually decelerate, reaching zero by 2040 and remaining constant thereafter. This variant operationalizes the concept of sufficiency by reducing the activity pressure on the transport system, moving beyond mere technological efficiency [33]. By running the Rail Scenario under steady-state conditions (RAIL-SSE), we investigate the synergies between structural efficiency (trains) and demand-side sufficiency (macroeconomic stabilization), identifying the conditions necessary to stay within absolute planetary boundaries [34,35].

2.2. Modeling Framework

To analyze these scenarios, a new transport module has been integrated into the pymedeas2 model, employing a hybrid top–down and bottom–up approach to maintain metabolic consistency with the broader economy. Its structural logic follows a sequential causal chain: macroeconomic output drives aggregate demand for passenger (pkm) and freight (tkm) activity; this demand is subsequently disaggregated by mode and fuel share according to specific policy scenarios. Finally, these modal splits are coupled with technology-specific energy intensities and material intensity factors to quantify mineral requirements, total energy demand, and its associated CO2 emissions, as detailed in Appendix B.1.

2.2.1. Transport Data Preprocessing and Demand Projection

Future transport activity trajectories are derived from econometric correlations with macroeconomic drivers, calibrated using historical data from the JRC-IDEES database [3]. We adopted a per-capita specification for passenger mobility and a GDP-centred model for freight to ensure mathematical consistency in the mature EU-27 economy. The projections follow a Constant Elasticity of Demand (CED) specification. To account for systemic inertia and temporal autocorrelation, the model incorporates a first-order autoregressive process ( ε t 1 ) derived from historical residuals. The resulting equations used for the system dynamics projections are:
PKM t = PKM 0 · exp γ ln GDP pc , t GDP pc , 0 + ϕ ε t 1
TKM t = TKM 0 · exp β ln GDP t GDP 0 + ϕ ε t 1
where γ and β represent the estimated income and freight elasticities, respectively. Parameters were estimated via Ordinary Least Squares (OLS) with Cochrane–Orcutt corrections, filtering pandemic-related anomalies (2020–2021) as exogenous shocks. The detailed regression framework, statistical diagnostics, and complete parameter tables are provided in Appendix C.

2.2.2. Material Demand and Economic Evaluation

The estimation of material requirements follows a two-step logic that translates transport demand into physical stocks. The required vehicle fleet size is first determined by applying vehicle intensity factors to the projected activity, accounting for historical occupancy rates and annual mileage. Subsequently, the modeling framework distinguishes between the active material stock in use and the annual demand required for maintenance and expansion. This mass-balance formulation, detailed in Appendix D.1, captures the total material throughput by accounting for both net fleet expansion and the replacement of degraded battery packs. Material intensity factors evolve dynamically through 2050 to reflect technological shifts, such as the transition toward Lithium Iron Phosphate (LFP) chemistries in passenger segments and the prioritization of high-nickel NCM 811 for heavy-duty logistics [36].
To evaluate the economic burden of the transition, we apply a physical capital replacement framework that disaggregates total annual investment into vehicle acquisition, charging infrastructure, and rail network expansion [11]. Vehicle investment covers both net expansion and the replacement of stock reaching the end of its operational life. The deployment of charging systems is parametrized following a high-utilization shared model that prioritizes fast-charging units to reduce the total material and spatial footprint of the network [37]. Finally, the structural shift toward rail necessitates significant investment in electrifying the remaining 46% of the current EU-27 network and constructing new tracks to absorb transferred road activity. The specific equations for material throughput and infrastructure costs across calculation methodology are documented in Appendix D.2. Crucially, our investment accounting captures only direct capital costs (fleet acquisition and primary track expansion). We explicitly exclude secondary expenses—such as maintenance, grid upgrades, and land acquisition—which require high-resolution spatial modeling outside our scope.

2.3. Data Integration and Calibration

The transport module is calibrated for the EU-27 using the JRC-IDEES 2023 database [3]. We aligned the model’s last historical year (2020) with empirical records of transport activity (pkm, tkm), vehicle stocks, and final energy demand (EJ). The calibration disaggregates the system by mode (m), fuel carrier (f), and technology (j).

2.3.1. Modal and Fuel Classification

Transport activity is disaggregated into passenger and freight sectors across the categories summarized in Table 1. Energy demand is partitioned into four primary carriers: liquids (gasoline, diesel, LPG, and biofuels), gases (natural gas and bio-methane), electricity, and plug-in hybrids (PHEV). For PHEVs, the liquid and electric components are isolated to facilitate accurate energy substitution accounting. Within the freight sector, we maintain a structural distinction between Light Commercial Vehicles (LCVs ≤ 3.5 t) and Heavy Goods Vehicles (HGVs > 3.5 t) to account for their divergent energy intensities and load-optimized efficiencies [30].

2.3.2. Model Calibration and Performance Parameters

The module is calibrated for the EU-27 using the JRC-IDEES 2023 database [3]. We aligned the model’s base year (2020) with empirical records of transport activity (A), vehicle stocks (V), and energy demand (E) for each mode (m) and fuel (f). From this dataset, we derive the fuel-specific energy intensity ( e m , f ) as the ratio of consumption to transport work:
e m , f = E m , f A m , f
To link transport demand with material requirements, we calculate the annual vehicle productivity ( P m , f ), representing the transport work performed per physical vehicle unit:
P m , f = A m , f V m , f
The derivation of P m , f enables differentiated stock estimations based on specific battery capacities—such as 80 kWh for LCVs versus 600 kWh for HGVs—providing a more granular assessment of mineral requirements than fleet-wide averages [11].

3. Results

The simulation of the various decarbonization pathways reveals that technological shifts alone are insufficient to meet the EU’s 90% transport emission reduction target. Our analysis indicates that deep decarbonization requires a combined strategy of significant modal shift, high electrification, and substantial reduction in total activity volumes.

3.1. Passenger and Freight Metabolic Transition

The transition matrices (Figure 2 and Figure 3) illustrate the divergent structural logic between institutional roadmaps and sufficiency-oriented strategies. In the passenger sector (Figure 2), Reference scenarios maintain a mobility system structurally reliant on individual vehicles, even with a near-total electrification of the light-duty fleet by 2050. In contrast, the Rail scenario implements a fundamental reallocation of activity, resulting in a 50–60% contraction of the private car fleet relative to baseline trends. By prioritizing high-speed electrified networks, rail becomes the primary carrier for medium-to-long distance travel by 2040.
Regarding freight logistics (Figure 3), the Reference pathways sustain high activity levels for heavy goods vehicles (HGVs), which continue to dominate long-haul transport. The Rail scenario applies an “All-Rail” logic to inland logistics, effectively phasing out HGVs from the strategic bulk of inter-urban cargo. Under this framework, road transport is strictly reserved for capillary and last-mile distribution, limited to approximately 15% of total tkm.
The vertical differentiation across both matrices highlights the impact of macroeconomic scale. While structural shifts improve systemic efficiency, the Steady-State (SSE) variants effectively mitigate the demand-side pressure, preventing the activity growth projected under continuous economic expansion.

3.2. Transport Final Energy Demand

The simulation of Final Energy Demand (FED) reveals a clear divergence between the transport sector’s metabolism (Figure 4c) and the total EU-27 energy footprint (Figure 4a). From a common 2024 peak of 11.8 EJ in transport and approximately 41 EJ for the total socio-economy, the Reference Scenario pathways demonstrate the inherent limitations of a transition centered exclusively on technological substitution. While the superior thermodynamic efficiency of electric power-trains helps reduce transport demand to 6.2 EJ in the growth variant or 5.4 EJ under demand stabilization, these sectoral gains are partially neutralized by sustained mobility growth, resulting in a total EU-27 FED that remains as high as 26 EJ in the REF-G case. This trend underscores that maintaining an individual road-based mobility paradigm, even when electrified, sustains a metabolic inertia that significantly limits the overall reduction potential of the total economy.
In contrast, the Rail Scenario pathways illustrate how structural modal shifts fundamentally reshape both sectoral and systemic energy trajectories. The RAIL-SSE variant achieves an accelerated contraction, reaching 3.8 EJ in the transport sector—a 68% reduction from the 2024 peak—which in turn drives the total EU-27 FED to a systemic low of approximately 21 EJ by 2050. This deep reduction suggests that prioritizing high-capacity electrified rail over road-based alternatives acts as a superior thermodynamic leverage point, where modal substitution provides higher energy savings.

3.3. Decarbonization Trajectories and Emissions

The emission figures reported in this assessment represent operational and upstream energy CO2 emissions, capturing direct vehicle operation and the carbon intensity of energy supply chains, while explicitly excluding the embodied carbon from vehicle manufacturing and large-scale infrastructure construction.
While all primary pathways achieve significant decarbonization by 2050, the trajectories for transport (Figure 4d) and the total EU-27 socio-economy (Figure 4b) diverge based on their specific structural and growth paradigms. In the transport sector, both Rail Scenario variants (RAIL-G and RAIL-SSE) achieve early action mitigation by bypassing the multi-decadal decarbonization lag inherent in private fleet turnover. However, the reduction in RAIL-G is partially limited by continued economic expansion, while the reference scenarios—even with the demand-side benefits of REF-SSE—remain fundamentally hampered by the high energy intensity of individual road-based mobility.
Consequently, the deepest absolute mitigation is achieved in the RAIL-SSE variant, where aggressive sectoral shifts and systemic sufficiency operate concurrently. Over the 2025–2050 time-frame, the institutional baseline (REF-G) results in cumulative transport emissions of 13.1 Gt CO2, while the RAIL-SSE scenario restricts them to 10.6 Gt CO2, reaching the lowest residual transport footprint of approximately 90 MtCO2/year. This sectoral result is matched by the total socio-economic trajectory, which reflects the broader impact of a steady-state economy where GDP per capita growth reaches zero by 2040. Ultimately, the RAIL-SSE profile demonstrates that absolute sustainability requires reducing the physical scale of the entire economy alongside technological and modal transitions.

3.4. Cumulative Material Requirements: The Lithium Wall

The biophysical feasibility of the evaluated pathways is primarily differentiated by the cumulative demand for Lithium (Figure 5). Under REF-G, the transition to a universal electric vehicle fleet, incorporating passenger cars, light commercial vehicles (LCVs), and heavy goods vehicles (HGVs), requires a cumulative extraction of approximately 2.35 million tons (Mt) of Lithium by 2050 [38]. These values represent the cumulative demand for metallic lithium. We acknowledge that the EU Battery Regulation (2023/1542) mandates ambitious lithium recovery targets of 50% by 2027 and 80% by 2031 [39]. However, due to the current lack of commercial-scale, high-efficiency closed-loop lithium recycling within the EU, and the inherent temporal lag between initial battery deployment and EoL availability, we explicitly frame our projections as a deliberately conservative upper bound for primary demand. Conversely, RAIL-SSE demonstrates a strategy for mineral parsimony, requiring only 1.02 Mt of Lithium over the same period, representing a 57% reduction relative to the baseline. By prioritizing catenary-powered rail and contracting the absolute volume of the private vehicle stock, this structural transition remains within safer biophysical boundaries while achieving superior decarbonization outcomes.

3.5. Economic Valuation: Capital Investment and Infrastructure Deployment

The economic assessment, summarized in Table 2, reveals a profound structural divergence in capital allocation strategies between the evaluated pathways. Total cumulative investment is governed by a fundamental trade-off: the high-volume, continuous renewal of short-lived private vehicle fleets in the Reference Scenarios versus the infrastructure-intensive expansion of a durable rail network in the Rail Scenarios. In the REF-G scenario, capital expenditure is primarily driven by the private passenger fleet, which requires an investment of USD 17.20 trillion. Even under demand stabilization (REF-SSE), maintaining an individual mobility paradigm necessitates massive financial outflows totalling USD 15.71 trillion for private cars alone.
Conversely, the Rail Scenarios facilitate a substantial reduction in total vehicle-related capital by redirecting investment from individual machinery to high-capacity collective assets. The 50–60% contraction of the private car fleet and the phase-out of long-haul road freight effectively offset the increased investment required for electrified rail rolling stock. While investment in locomotives increases to between 0.96 and 0.86 trillion in these scenarios, the total capital required for road-based freight vehicles is nearly half that of the REF-G. Physical deployment further illustrates this shift; the RAIL-G requires planned public investment to construct 451,322 km of new high-speed and standard tracks, or 364,032 km in the more efficient RAIL-SSE variant.
Energy infrastructure costs further reflect the efficiency gains of the high-utilization shared model implemented in the Rail Scenarios. While the Reference Scenarios necessitate a fragmented network of nearly 112 million charging posts to support universal private electric vehicle ownership, the Rail Scenarios optimize expenditure, requiring fewer than 59 million units. Cumulatively, the RAIL-SSE variant emerges as the most economical pathway over the simulation period with a total investment of USD 19.83 trillion. This represents a systemic saving of USD 7.27 trillion relative to REF-G, demonstrating that the savings from avoided private vehicle acquisition and optimized charging networks compensate for the multi-trillion dollar costs of new track construction.

4. Discussion

This study applied the latest release of the pymedeas2 framework to assess the biophysical implications of the transport sector transition in the EU-27, aiming to evaluate which decarbonization pathways are most effective at minimizing structural dependencies on critical materials and overall energy demand. To achieve this, we contrasted four distinct scenarios: the institutional baseline (REF-G) focused on green growth and private fleet electrification; a rail-centric growth pathway (RAIL-G); a steady-state economy pathway maintaining road dominance (REF-SSE); and a sufficiency-driven pathway combining macroeconomic stabilization with rail expansion (RAIL-SSE). While all modeled pathways achieve significant decarbonization by 2050, the results demonstrate that institutional strategies like Reference Scenario remain unable to achieve the absolute decoupling of energy demand from transport activity. Although the superior thermodynamic efficiency of electric power-trains drives a downward trajectory in energy consumption across all scenarios—a trend consistent with official projections like the Clean Energy Technology Observatory [19]—these gains in Reference Scenario are partially neutralized by the sustained growth in transport volume. This confirms that a transition centred exclusively on fleet electrification maintains a high energy demand due to the metabolic inertia of persistent private vehicle usage, effectively limiting the scope of energy reduction to the efficiency of the delivery mechanism rather than the scale of the service provided.
In contrast, the RAIL-SSE scenario demonstrates that the integration of demand stabilization and structural modal shifts, specifically the transfer of long-distance activity to rail, acts as a superior thermodynamic leverage point. By bypassing the inherent energy intensity of individual road transport, these sufficiency-oriented pathways achieve an additional reduction in total energy consumption of approximately 39% relative to the reference. This divergence reveals that while technological substitution achieves a relative improvement, only the structural changes in Rail Scenario facilitate the absolute metabolic contraction required for biophysical sustainability. This outcome reinforces the theoretical critiques of the green growth paradigm offered by [7,8] while aligning with the demand-side mitigation strategies recently emphasized by the UNEP [40] and the European Scientific Advisory Board on Climate Change [41]. Both bodies increasingly emphasize that reaching climate neutrality necessitates structural reductions in energy service demand alongside rapid electrification.
The temporal dynamics of CO2 mitigation reveal a persistent decarbonization lag in Reference Scenario pathways, highlighting the tension between market-led technological change and climate urgency. In these scenarios, the rate of emission reduction is fundamentally constrained by the physical inertia of the vehicle stock—an effect directly linked to the 12.5-year average age of the European passenger fleet [30]. This constraint implies that even if sales of internal combustion engines (ICEs) were phased out rapidly, the existing fleet would continue to consume the remaining carbon budget for over a decade. To contextualize the severity of this lag, deducting 2020–2024 historical emissions from the sector’s 1.5 °C-aligned budget [42] leaves a remaining allocation of strictly 6.4 to 8.3 Gt CO2 from 2025 onward. Our modeling reveals that the REF-G trajectory drastically overshoots this biophysical limit by generating 13.1 Gt CO2 in cumulative transport emissions, whereas the RAIL-SSE scenario restricts them to 10.6 Gt CO2. While institutional frameworks often assume a near-linear decarbonization based on the market penetration of new electric vehicles [43], our results suggest that such projections frequently underestimate the “lock-in” effect of the current stock. This finding aligns with the warnings of Brand et al. [44], who argue that technological substitution alone is too slow to meet the 1.5 °C target, and reinforces observations by the European Environment Agency [45] that the ageing European fleet is becoming a structural barrier to rapid emission cuts; while other economic sectors are successfully decarbonizing, transport emissions remain high largely due to the slow turnover of the existing vehicle stock.
In contrast, the Rail Scenario serves as a temporal shortcut by decoupling decarbonization from the slow process of machine replacement. By enabling an immediate shift to existing electrified rail and reducing high-impact activities such as aviation, these pathways achieve deep systemic reductions long before the total electrification of the road fleet is physically possible. This structural intervention bypasses the multi-decadal turnover cycles that hamper Reference Scenario, delivering the early action mitigation [46] as essential for staying within biophysical limits. Although meeting the strictest 1.5 °C targets remains highly challenging even under the 10.6 Gt CO2 trajectory, this structural pathway converges significantly closer to biophysical sustainability than technology-led green growth. Consequently, our analysis indicates that structural and demand-side measures likely play a fundamental role in achieving the rapid mitigation required by the Paris Agreement, suggesting that a technology-only approach may face significant temporal risks that these interventions could help mitigate.
Beyond the temporal and energetic risks, the physical feasibility of a technology-centric transition is challenged by extreme mineral intensity, creating a significant material bottleneck in the Reference Scenario pathways. Our results indicate that the cumulative lithium requirement for the EU-27 passenger fleet in the REF-G scenario reaches 2.35 million tons by 2050. While this constitutes approximately 6.4% of the 37 million tons of current global proven reserves [47], the fact that a single region representing less than 6% of the global population would claim this entire “fair share” solely for its domestic mobility calls into question the geopolitical and biophysical viability of a global green growth strategy [48]. Although circular economy strategies, including hydro-metallurgical recycling and secondary material recovery, are often proposed to mitigate these constraints, their current industrial scalability remains highly uncertain [49]. Furthermore, due to the inherent temporal lag between the initial deployment of battery stocks and their eventual EoL availability, secondary supply can only satisfy a fraction of the rapidly growing demand through 2050 [36]. Thus, while essential for long-term sustainability, circularity serves as a complementary improvement rather than a systemic solution to the resource-intensive requirements of the REF-G pathway, justifying a conservative assessment of primary demand.
This disproportionate appropriation becomes even more critical when accounting for competing cross-sectoral demands. Industry forecasts project that electric mobility will directly compete for lithium with the rapidly expanding stationary energy storage sector, which is indispensable for stabilizing renewable grids and relies heavily on lithium-intensive LFP chemistry [43,50]. Furthermore, this techno-centric dependence offshores significant environmental burdens; the energy-intensive extraction and refining processes generate substantial CO2 emissions and ecological degradation outside the EU’s borders—typically in the Global South—a factor frequently externalised in institutional green growth accounting [38]. The aggressive electrification of private vehicles under REF-G effectively monopolizes the lithium required for both the EU’s own renewable grid and the basic energy needs of the rest of the world, suggesting that 1:1 vehicle replacement is structurally incompatible with a globally equitable transition. In contrast, the RAIL-SSE variant mitigates this risk by reducing cumulative lithium demand by 57%—to 1.02 Mt. This demonstrates that sufficiency and modal shifts act not only as transport mitigation strategies but as vital cross-sectoral resource-buffering mechanisms, essential to align the sector with the planet’s finite geological reality [38].

4.1. The Rail Transition as a Limit Case: Feasibility, Materiality, and Implementation Barriers

Despite the clear lithium demand advantages of the RAIL scenarios, the scale of infrastructure expansion required to absorb 60% of passenger and 76% of freight traffic presents profound real-world feasibility challenges. Our model indicates this shift would require an estimated 364,032 to 451,322 equivalent line-kilometers of new rail infrastructure. We explicitly frame these scenarios not as predictive forecasts, but as “limit cases” designed to analytically quantify the absolute biophysical boundaries of replacing road-dominated transport. While our macroeconomic framework captures the overarching capital and energy requirements, it does not account for micro-level operational factors such as traffic management, network capacity or specific operating modes. Furthermore, it does not account for protracted construction rates, specific land-use demand, or the complex spatial planning required to deploy new corridors. Absorbing such a massive reallocation of traffic would inevitably clash with protracted land-use planning processes and face significant hurdles regarding the social acceptance of extensive new construction projects [51].
Furthermore, this infrastructure-heavy pathway fundamentally shifts the material burden of the transition. While a rail-centred approach significantly reduces dependence on battery-grade lithium, it necessitates immense volumes of structural materials, such as copper for electrification and steel and cement for large-scale track construction. The associated embodied carbon represents an upfront emission penalty that may temporarily offset short-term climate gains [52,53]. However, this trade-off must be contextualized: the road-centric baseline is not exempt from these impacts, as it demands significant material inputs for highway maintenance and structural adaptation to accommodate heavier electric fleets. Consequently, rail expansion can trade the global supply-chain risks of battery metals for the localized, high-volume ecological impacts of heavy construction.
In light of these institutional, social, and physical barriers, our comparative analysis highlights the limitations of relying exclusively on either technology substitution or macro-structural modal shifts. While our aggregate EU-27 model abstracts from spatial specificities to identify absolute biophysical boundaries, realizing this transition requires unprecedented alignment across EU Member States [54]. In practice, transcending the binary between Green Growth and post-growth trajectories is essential. Policy must integrate sufficiency with systemic efficiency, optimizing transport architecture, dynamic flow management, and infrastructure utilization, alongside robust logistics reorganization and absolute transport demand reduction [55,56].

4.2. Macroeconomic Implications and Capital Reallocation

Finally, the divergence between these pathways translates into different macroeconomic requirements, where the Reference Scenarios represent a high capital expenditure transition that risks exacerbating the investment gap recently identified in assessments of European competitiveness [57]. By maintaining a growth-oriented activity model, this strategy necessitates a constant flow of capital into short-lived consumer goods—with the REF-G variant requiring USD 17.20 trillion for private vehicles alone, effectively crowding out investments in more durable public infrastructure.
In contrast, the Rail Scenarios suggest a shift in economic logic: by reducing the total volume of required machinery and prioritizing high-capacity rail, the transition moves from a consumption-heavy model toward a service-oriented one. While rail infrastructure investment rises to USD 6.16 trillion in the RAIL-SSE variant to support the construction of 364,032 km of new tracks, this expenditure is offset by the reduction in private vehicle acquisition costs to USD 7.50 trillion. Consequently, the RAIL-SSE pathway emerges as the most efficient strategy, requiring a total cumulative investment of USD 19.83 trillion, which represents a systemic saving of USD 7.27 trillion relative to the REF-G baseline. This alignment with post-growth frameworks [8,58] suggests that absolute sustainability is only achievable if the European economy is reoriented away from resource-intensive GDP growth and toward the provision of essential services with minimal biophysical and capital overhead.

5. Conclusions

This study demonstrates that, under the modeled assumptions of high-growth baselines, conventional technology-led pathways succeed in improving relative efficiency but face significant challenges in achieving the absolute energy reduction required for sustainability due to the metabolic inertia of transport demand within the EU-27. We frame our RAIL-SSE scenarios as normative limit-cases rather than predictive forecasts, designed to analytically quantify the biophysical boundaries of a post-growth transition. Consequently, our analysis indicates that transport decarbonization likely requires more than technological substitution; an absolute reduction in the physical scale of activity, quantified as the aggregate annual volume of p k m and t k m , appears to be a fundamental factor in achieving biophysical sustainability.
First, the temporal dynamics of CO2 mitigation reveal a persistent decarbonization lag, where the 12.5-year average age of the European passenger fleet acts as a structural barrier to rapid emission cuts. Assuming this current lifespan remains constant, the high average age of the European fleet creates a multi-decadal lag that market-driven electrification cannot overcome in time to meet climate targets. Only a large-scale modal shift toward electrified rail and demand-side management can bypass these vehicle turnover cycles, providing the immediate emission reductions necessary to remain within biophysical limits.
Second, universal private electrification presents significant challenges regarding resource equity under current demand growth assumptions and life-cycle material intensities. An EU-27 private electric fleet would require 2.35 million tons of lithium by 2050, claiming 6.4% of global reserves for only the transport sector of a region with only 6% of the world’s population. This concentration of resources is structurally incompatible with a globally equitable transition. Conversely, the RAIL-SSE scenario reduces cumulative lithium demand by 57%, aligning the sector with the reality of finite geological endowments.
Third, when assessed within the defined system boundaries that focus on direct capital requirements for fleet and basic network expansion, a structure-led transition proves to be economically superior to technology-led green growth. Despite the massive upfront capital required to expand and electrify the European rail network, the RAIL-SSE pathway emerges as the least capital-intensive strategy. By drastically reducing the continuous, high-volume acquisition of short-lived private vehicles and optimizing charging infrastructure, this sufficiency-driven scenario yields a systemic saving of over USD 7.27 trillion relative to the institutional baseline.
From a policy perspective, the divergence in metabolic and economic pathways indicates that absolute sustainability necessitates aligning EU policy with post-growth frameworks, prioritizing durable public infrastructure and service provision over resource-intensive GDP expansion. Future policy should integrate sufficiency as a core pillar alongside efficiency and renewables to mitigate the identified temporal and material risks and ensure a resilient transition within the sector’s absolute biophysical limits. To advance this paradigm, future research must expand this integrated framework to assess the socio-economic distributional impacts of sufficiency-based transitions, particularly concerning employment shifts and equity. Furthermore, deepening the analysis of circular economy strategies—such as secondary material recovery and alternative battery chemistries—is essential to further refine the mitigation of the material constraints identified in this study.

Author Contributions

Conceptualization, E.A.C., R.S. and J.S.; methodology, E.A.C.; software, E.A.C. and R.S.; validation, E.A.C. and P.M.M.; formal analysis, E.A.C.; investigation, E.A.C.; resources, J.S. and R.S.; data curation, E.A.C. and P.M.M.; writing—original draft preparation, E.A.C.; writing—review and editing, E.A.C., P.M.M., R.S. and J.S.; visualization, E.A.C.; supervision, R.S. and J.S.; project administration, J.S.; funding acquisition, J.S. All authors have read and agreed to the published version of the manuscript.

Funding

The authors acknowledge the financial support of the Government of Catalonia (Generalitat de Catalunya) through the project PYMEDEASCAT, funded by the Climate Fund under the framework of the Catalan Climate Change Law. J.S. acknowledges the institutional backing of the ICM-CSIC, a Severo Ochoa Centre of Excellence (Grant CEX2024-001494-S, funded by AEI 10.13039/501100011033).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The pymedeas2 modeling framework is open-source and implemented in Python 3.13. The source code, including the specific transport module and scenario configurations used in this study, is available at https://github.com/Earth-and-Energy-Systems-Lab/pymedeas2 (accessed on 3 July 2026), version 1.1. Historical transport data and calibration parameters were derived from the JRC-IDEES 2023 database, publicly accessible through the European Commission’s Joint Research Centre.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
AR6Sixth Assessment Report
BEVBattery Electric Vehicle
CEDConstant Elasticity of Demand
CETO Clean Energy Technology Observatory
CHPCombined Heat and Power
CO2Carbon Dioxide
CRMCritical Raw Material
EJExajoule
EUEuropean Union
FEDFinal Energy Demand
GDPGross Domestic Product
GHGGreenhouse Gas
HGVHeavy Goods Vehicle
IAMIntegrated Assessment Model
IEAInternational Energy Agency
IOTInput–Output Table
IPCCIntergovernmental Panel on Climate Change
JRCJoint Research Centre
LCVLight Commercial Vehicle
LFPLithium Iron Phosphate
MtMillion Tonnes
NMCLithium Nickel Manganese Cobalt Oxide
PHEVPlug-in Hybrid Electric Vehicle
PKMPassenger-kilometer
RESRenewable Energy Sources
SSESteady-State Economy
TEN-TTrans-European Transport Network
TKMTon-kilometer
TPESTotal Primary Energy Supply

Appendix A. Global Sensitivity Analysis (GSA)

Appendix A.1. Methodology: The Morris Method

To ensure the robustness of the results presented in this study, we conducted a Global Sensitivity Analysis (GSA) following the Method of Elementary Effects, commonly known as the Morris Method [59]. This screening technique allows for the identification and ranking of input parameters that most significantly influence the model’s critical outputs, distinguishing between direct linear impacts and complex non-linear interactions or dependencies [60].
The analysis was implemented using the SALib library in Python [61]. Following the standard Morris design, we generated r = 50 trajectories with p = 4 grid levels. For each of the seven input parameters analyzed ( k = 7 ), the total computational cost followed the Morris equation N = r ( k + 1 ) , resulting in 400 simulations. This sample size was selected to ensure the convergence and stability of the sensitivity indices μ (mean absolute elementary effect) and σ (standard deviation) [60].

Appendix A.2. Input Parameters and Configuration

Seven key exogenous drivers were selected to capture the primary dimensions of uncertainty within the transport transition pathways. For each parameter, a variation range was established around the baseline value, consistent with uncertainty ranges used in similar biophysical modeling studies [20,62].
  • GDP Sensitivity Factor ( g d p _ s e n s ): Variation in annual GDP growth rates.
  • Activity Demand ( s e n s _ p k m _ t k m ): Scaling factor for total passenger (pkm) and freight (tkm) activity.
  • Passenger Efficiency ( s e n s _ p k m _ e f f ): Energy intensity improvements for passenger transport.
  • Freight Efficiency ( s e n s _ t k m _ e f f ): Energy intensity improvements for freight transport.
  • Renewable Expansion ( s e n s _ r e s ): Scaling factor for the deployment of renewable energy infrastructure.
  • Vehicle Efficiency ( s e n s _ v e h _ e f f ): Specific efficiency in the number of vehicles for tkm and pkm transport activity.
  • Rail Mode Share ( s e n s _ r a i l _ m o d e _ s h a r e ): Additive variation to the baseline percentage of transport activity shifted to electrified rail systems.

Appendix A.3. Results and Interpretation

The sensitivity of four primary output variables was assessed: Transport CO2 Emissions, Cumulative Lithium Requirements, Total Primary Energy Supply (TPES), and Total Final Energy Demand (FED). Table A1 summarizes the resulting indices for the most influential parameters in each category.
Table A1. Morris sensitivity indices ( μ and σ ) for the key transport output variables.
Table A1. Morris sensitivity indices ( μ and σ ) for the key transport output variables.
Target VariableInput Parameter μ σ
CO2 Emissions s e n s _ t k m _ e f f 1.0565.801
s e n s _ p k m _ e f f 0.7033.007
s e n s _ p k m _ t k m 0.6242.957
g d p _ s e n s 0.4310.286
s e n s _ r e s 0.2310.187
Lithium Requirements s e n s _ v e h _ e f f 1.0460.183
s e n s _ p k m _ t k m 0.9600.220
s e n s _ r a i l _ m o d e _ s h a r e 0.6140.142
g d p _ s e n s 0.5620.209
s e n s _ r e s 0.2150.206
TPES (EJ) g d p _ s e n s 23.0629.810
s e n s _ r e s 8.78710.282
s e n s _ p k m _ e f f 5.5452.619
s e n s _ p k m _ t k m 4.4562.456
s e n s _ t k m _ e f f 2.5582.310
Total FED g d p _ s e n s 9.0363.713
s e n s _ p k m _ e f f 6.4721.187
s e n s _ r e s 4.8183.908
s e n s _ t k m _ e f f 3.0960.902
s e n s _ p k m _ t k m 2.1241.000
The results identify the metabolic scale and systemic efficiency as the paramount drivers of environmental and material impacts. For CO2 emissions and Total FED, improvements in transport efficiency ( s e n s _ t k m _ e f f and s e n s _ p k m _ e f f ) alongside overall GDP growth exhibit the highest influence, confirming that the interplay between demand-side volume and energy intensity dictates the broader energy transition limits. For TPES, the g d p _ s e n s parameter remains the overwhelmingly dominant driver ( μ = 23.06 ) with a high σ ( 9.81 ), suggesting strong non-linear interactions within the energy-economy feedback loops.
Crucially, the analysis of Cumulative Lithium Requirements validates the core thesis of this study regarding material bottlenecks. The mineral demand is heavily dictated by the specific manufacturing efficiency of vehicles ( s e n s _ v e h _ e f f , μ = 1.04 ) and the absolute volume of transport demand ( s e n s _ p k m _ t k m , μ = 0.96 ). However, the Rail Mode Share parameter emerges as the third most critical driver ( μ = 0.61 ), surpassing even the influence of total GDP growth. This statistically confirms that structural interventions, specifically the degree to which mobility is shifted toward catenary-powered rail, directly and significantly determine the feasibility of staying within safe planetary boundaries for critical minerals. These findings reinforce the conclusion that achieving a sustainable transition requires prioritizing demand-side sufficiency and structural modal shifts over exclusive reliance on technological substitution [20,63].

Appendix B. pymedeas2 Improvements

Appendix B.1. Transport Module Architecture and Causal Logic

The transport module employs a hybrid top–down/bottom–up approach to internalize the biophysical and economic feedbacks of mobility within the system dynamics framework. The modeling sequence follows a four-stage causal chain:
1.
Demand Generation: Aggregate transport demand (pkm and tkm) is driven endogenously by macroeconomic variables: GDP and population. These variables are calibrated using econometric elasticities.
2.
Modal and Technological Allocation: Total demand is disaggregated into specific modes (road, rail, air, and maritime) and powertrain technologies (BEV, ICE, and Hydrogen) according to scenario-specific policy pathways.
3.
Metabolic Accounting: The model applies technology-specific energy intensities and material factors to calculate final energy demand, its associated greenhouse gas emissions, and critical raw material requirements (e.g., lithium for battery stocks).
4.
Systemic Feedback Loops: Variations in energy availability or climate impacts can retroactively constrain economic output, internalizing the biophysical limits to transport growth.

Appendix B.2. Economic Framework Integration

The economic module of pymedeas2 has been updated to work with the EXIOBASE database [21] for more detail in the economic sectors. The improved economic model consists of 49 sectors (see Table A11), including the land transport sector independently from the other transport categories (i.e., air transport and maritime transport).
The pymedeas2 model works with the sectoral energy consumption from the IEA world energy balances [64], which are published with low industrial resolution. To increase the resolution of the energy data to match the updated detail of the economic model, we implemented an adapted version of the methodology from [65]. It estimates the energy flows for each sector based on the EXIOBASE supply-and-use (SUT) tables. The IEA database energy sources are aggregated into five final fuel categories in pymedeas2: solid, liquid, gas, electricity and heat (see Table A13). The energy consumption for the transport sectors excludes international bunkers, which are not accounted for in this study.

Appendix C. Econometric Estimation and Statistical Analysis

To establish the empirical foundation for Equations (1) and (2), we developed an econometric framework to estimate the structural drivers of transport activity. All variables were normalized relative to the base year ( t 0 = 2000 ) to ensure dimensional homogeneity. The year 2020 served as the historical calibration year to validate the model’s baseline physical and energy outputs. To ensure continuity between historical trends and projected pathways, 2023 acts as the projection anchor (integrating the latest available statistical data). Projections for the 2024–2050 period (with 2024 marking the starting point for our forward-looking scenario simulations) were calculated by applying the estimated income elasticity to the relative growth of GDP per capita, using the 2023 anchored activity levels as the starting point for the extrapolation.

Appendix C.1. Econometric Specification

For passenger transport, a log-log per-capita regression model was fitted to the historical data:
ln P K M t P K M 0 = γ ln G D P p c , t G D P p c , 0 + ε t
where P K M t and G D P p c , t represent passenger-kilometers and GDP per capita, respectively. For freight transport, activity was modeled relative to total GDP:
ln T K M t T K M 0 = β ln G D P t G D P 0 + ε t
The residuals ( ε t ) were modeled as a first-order autoregressive process (AR(1)) to capture temporal autocorrelation: ε t = ϕ ε t 1 + u t [66].

Appendix C.2. Estimation Results and Diagnostics

The parameters were estimated using historical time-series (2000–2023) bound by the database limits. To isolate long-term structural elasticities from severe macroeconomic shocks, the anomalous COVID-19 pandemic years were identified using a standardized residual threshold of 2 σ . The income elasticity for passenger transport ( γ ) was estimated at 0.652 ( p < 0.001 , adjusted R 2 = 0.96 ), reflecting the saturation of individual mobility in high-income regions. For freight transport, the elasticity ( β ) was found to be 0.774 ( p < 0.001 , adjusted R 2 = 0.92 ), consistent with the relative decoupling of logistics from economic growth observed in the EU-27. Complete regression diagnostics are summarized in Table A2.
Table A2. Econometric regression parameters for passenger-kilometer ( p k m ) and ton-kilometer ( t k m ) demand projections in the EU-27.
Table A2. Econometric regression parameters for passenger-kilometer ( p k m ) and ton-kilometer ( t k m ) demand projections in the EU-27.
ParameterValueDescription/Diagnostic
Passenger Transport ( p k m )—GDP Per Capita Model
γ (Elasticity)0.6517Income elasticity of mobility
p k m 0 4747.18Base year activity (billion p k m , 2000)
G D P p c , 0 22,032.08Base year GDP per capita (1995USD, 2000)
R 2 0.9490Coefficient of determination
p-val ( γ ) 4.71 × 10 13 Statistical significance of elasticity
ϕ A R 1 0.4803First-order autoregressive coefficient
Freight Transport ( t k m )—GDP Model
β (Elasticity)0.7737Freight-income elasticity
t k m 0 1973.57Base year activity (billion t k m , 2000)
G D P 0 9.4402Base year total GDP (trillion 1995USD, 2000)
R 2 0.9209Coefficient of determination
p-val ( β ) 1.56 × 10 11 Statistical significance of elasticity
ϕ A R 1 0.2964First-order autoregressive coefficient

Appendix D. Material and Economic Formulations

Appendix D.1. Vehicle Stock and Material Demand

The estimation of material demand follows a two-step sequential logic that translates transport demand into primary material requirements. First, the required vehicle fleet size ( N m , f ) for each mode and technology is determined by applying the inverse of the productivity coefficient, defined here as the vehicle intensity ( σ m , f = 1 / P m , f ). This coefficient implicitly accounts for historical vehicle occupancy rates, load factors, and average annual mileage:
N m , f ( t ) = A m , f ( t ) · σ m , f
where N m , f is the number of vehicles for mode m and technology f, and A m , f is the mode-specific transport activity.
Second, the material modeling framework distinguishes between the active Material Stock in Use and the Annual Material Demand required for fleet maintenance and expansion. The total installed energy capacity of the fleet is calculated by assuming an average battery capacity per vehicle ( C a v g , m ) in kWh. The aggregate stock for a specific material k is then derived using material intensity factors ( I k ), expressed in kg of material per kWh of battery capacity:
S k ( t ) = m N m , e l e c ( t ) · C a v g , m · I k
The material intensity factors ( I k ) are technology-specific and reflect the chemical composition of the prevailing battery architectures, such as Lithium-ion Lithium Nickel Manganese Cobalt Oxide (NMC) or Lithium Iron Phosphate (LFP) (see Table A10). These shares evolve dynamically through 2050 following projected technological trends. For instance, the model simulates a transition in the passenger car segment toward LFP alternatives, reaching a 60% market share by 2050, while high-nickel chemistries like NCM 811 are prioritised for heavy-duty segments to meet specific energy density requirements [36] (see Table A6).
To accurately capture material throughput, the model calculates the Annual Primary Material Demand ( D k ( t ) ) by isolating gross physical inflows from net stock dynamics. Relying on net stock changes ( S k ( t ) ) would artificially underestimate primary resource requirements when vehicle retirements exceed new additions.
Consequently, the total annual primary material demand for a specific material k is defined strictly as the sum of materials required for gross new vehicle additions ( I n f l o w s k ( t ) ) and the maintenance of the existing active fleet ( R e p l a c e m e n t s k ( t ) ):
D k ( t ) = I n f l o w s k ( t ) + R e p l a c e m e n t s k ( t )
Gross inflows are driven by the required new vehicle sales ( S a l e s m , e l e c ( t ) ) and their corresponding material intensity ( I k ):
I n f l o w s k ( t ) = m S a l e s m , e l e c ( t ) · C a v g , m · I k
Replacements are calculated based on the existing stock and a technology-specific battery lifetime ( L b a t t ):
R e p l a c e m e n t s k ( t ) = S k ( t 1 ) L b a t t
The physical retirements of vehicles ( O u t f l o w s k ( t ) ) are determined by the fleet size and the average mechanical vehicle lifetime ( L v e h ):
O u t f l o w s k ( t ) = S k ( t 1 ) L v e h
The net change in the active fleet stock is the balance of these flows:
S k ( t ) = I n f l o w s k ( t ) O u t f l o w s k ( t )
Because this assessment models a primary-extraction-dominated pathway, circularity mechanisms are not endogenised. Therefore, O u t f l o w s k ( t ) are routed to an EoL scrap pool and do not provide a secondary supply stream to offset D k ( t ) . Even when S k ( t ) is negative due to intentional fleet contraction, D k ( t ) remains a strictly positive function of necessary fleet turnover and maintenance, ensuring that primary material bottlenecks are accurately represented [36,67].

Appendix D.2. Economic Valuation of the Transition

To evaluate the economic burden of the proposed transition, we follow the physical capital replacement framework established by [11], disaggregating total annual investment ( I t o t a l , t ) into vehicle fleet acquisition, charging infrastructure deployment, and rail network expansion. The investment in new vehicles ( I v e h , t ) accounts for both the net expansion of the fleet to meet growing transport demand and the replacement of existing stock reaching the end of its operational life. The cost is calculated as:
I v e h , t = m , f S m , f , t · P m , f
where S m , f , t represents the units sold for mode m and technology f at time t, and P m , f is the unit purchase price (see Table A3).
The deployment of the electrical charging system ( I c h a r g e , t ) is modeled based on the required density of charging points to ensure system functionality. For passenger vehicles, the number of required posts is a function of population density and average charging time, while for commercial vehicles, it depends on the energy throughput required for freight operations. The annual investment is given by:
I c h a r g e , t = N p o s t s , t · P p o s t
where N p o s t s , t is the annual increment of new charging stations and P p o s t represents the unit cost of installation, including hardware and grid connection. We distinguish between standard wallbox units for private use and high-power fast-chargers (e.g., 50–150 kW) for heavy-duty commercial logistics.
Charging infrastructure requirements are parametrized following a high-utilization shared model derived from Xylia et al. (2025) [37]. Private and depot-based settings—comprising residential, workplace, and commercial logistics hubs—are modeled with a ratio of 0.20 chargers per vehicle. Public infrastructure utilizes a lower ratio of 0.043, assuming a technological mix dominated by fast and super-fast charging units. This configuration prioritizes high unit occupancy and shared access, significantly reducing the total material and spatial footprint of the charging network compared to conventional slow-charging deployments.
The Rail Scenario necessitates a massive structural shift requiring the expansion and full electrification of the EU-27 rail network ( I r a i l , t ). The investment covers the electrification of the remaining non-electrified lines (approximately 46% of the current network) and the construction of new high-speed and medium-speed tracks to absorb the transferred road activity:
I r a i l , t = t y p e L n e w , t y p e , t · C k m , t y p e
where L n e w is the length of tracks added or upgraded, and C k m , t y p e represents the specific cost per kilometer. Following [11], we calculate weighted mean infrastructure costs that incorporate platforms, rails, signalling, and electrification across the diverse European orography.
Table A3. Consolidated economic parameters for the transport transition model (Base Year 2025) in USD.
Table A3. Consolidated economic parameters for the transport transition model (Base Year 2025) in USD.
Category/ComponentUnitAvg. Value (USD)Reference
Vehicle Purchase Price
Passenger Car (BEV)per unit43,500 [43]
Light Commercial (LCV)per unit53,000 [68]
Heavy Goods (HGV)per unit470,000 [68]
Electric Busper unit585,000 [69]
Electric Locomotiveper unit4,900,000 [70]
Charging Infrastructure
AC Wallbox (Private/Slow)per post1800 [71]
DC Fast Charger (Public/HGV)per post76,000 [71,72]
Rail Infrastructure
Electrification of existing linesper km1,300,000 [70]
New High-Speed Rail (HSR)per km25,500,000 [11]
New Standard Track (Double)per km15,840,000 [11]
New Single-Track Lineper km10,820,000 [11]

Appendix E. Additional Data Tables

Table A4. Assumed mean battery capacities ( C a v g ) by vehicle segment and technology used by pymedeas2 model.
Table A4. Assumed mean battery capacities ( C a v g ) by vehicle segment and technology used by pymedeas2 model.
Vehicle SegmentTechnologyCapacity (kWh)Reference
Passenger CarBEV (Electric)66[36]
Passenger CarPHEV (Hybrid)12[36]
Light Commercial (LCV)BEV (Electric)80[43]
Heavy Goods (HGV)BEV (Electric)600[73]
BusesBEV (Electric)450[74]
Table A5. Historical Market Shares of Battery Chemistries in the EU-27 (1995–2023). Light and heavy commercial vehicle data reflect early fleet adoption trends [36,43].
Table A5. Historical Market Shares of Battery Chemistries in the EU-27 (1995–2023). Light and heavy commercial vehicle data reflect early fleet adoption trends [36,43].
Segment/Technology1995201520162017201820192020202120222023
Cars (Passenger)
NCM 11185%85%84%80%60%40%15%5%2%0%
LFP1%1%1%1%2%3%5%15%28%32%
NCM 6220%0%0%5%20%40%55%40%20%15%
NCM 8110%0%0%0%0%2%10%25%38%43%
NCA14%14%15%14%18%15%15%15%12%10%
Buses
NCM 11195%85%80%60%40%20%5%0%0%0%
LFP5%15%20%25%30%35%40%42%45%48%
NCM 6220%0%0%15%30%45%50%43%35%22%
NCM 8110%0%0%0%0%0%5%15%20%30%
Light Trucks (LCV)
NCM 11195%85%80%60%40%20%5%0%0%0%
LFP5%5%5%5%5%5%5%15%25%35%
NCM 6220%10%15%35%55%70%80%60%40%25%
NCM 8110%0%0%0%0%5%10%25%35%40%
Heavy Trucks (HGV)
NCM 11195%85%80%60%40%20%5%0%0%0%
LFP5%5%5%5%5%5%10%15%20%30%
NCM 6220%10%15%35%55%75%80%65%40%20%
NCM 8110%0%0%0%0%0%5%20%40%50%
Table A6. Projected market shares of battery chemistries by vehicle segment in the EU-27 (2025–2050). Passenger car projections follow the LFP scenario from Xu et al. [36], while commercial segments integrate demand forecasts from recent industry literature [43,75].
Table A6. Projected market shares of battery chemistries by vehicle segment in the EU-27 (2025–2050). Passenger car projections follow the LFP scenario from Xu et al. [36], while commercial segments integrate demand forecasts from recent industry literature [43,75].
Segment/Technology202520302035204020452050
Cars (Passenger)
LFP45.0%60.0%60.0%60.0%60.0%60.0%
NCM 81145.0%35.0%35.0%35.0%35.0%35.0%
NCM 6225.0%0.0%0.0%0.0%0.0%0.0%
NCA5.0%5.0%5.0%5.0%5.0%5.0%
Buses
LFP50.0%60.0%65.0%75.0%80.0%85.0%
NCM 81150.0%40.0%35.0%25.0%20.0%15.0%
Light Trucks (LCV)
LFP40.0%55.0%60.0%65.0%70.0%75.0%
NCM 81155.0%45.0%40.0%35.0%30.0%25.0%
NCM 6225.0%0.0%0.0%0.0%0.0%0.0%
Heavy Trucks (HGV)
LFP40.0%55.0%65.0%70.0%75.0%80.0%
NCM 81160.0%45.0%35.0%30.0%25.0%20.0%
Table A7. Adapted Transport Mode Share Projections for the EU-27 (2020–2050) based on the CETO Scenario.
Table A7. Adapted Transport Mode Share Projections for the EU-27 (2020–2050) based on the CETO Scenario.
CategoryMode2020202520302035204020452050
Freight (tkm)Road Heavy71.96%70.20%67.40%63.50%59.80%55.40%51.50%
Road Light3.61%3.75%3.80%3.80%3.80%3.80%3.80%
Rail15.82%17.40%20.10%24.10%28.30%32.50%36.50%
Maritime8.61%8.65%8.70%8.60%8.10%8.30%8.20%
Air (Domestic)0.01%0.00%0.00%0.00%0.00%0.00%0.00%
Passenger (pkm)Road Bus6.87%7.75%8.32%8.94%9.60%10.30%11.06%
Road Car85.52%81.30%78.50%74.20%68.80%63.10%58.60%
Rail6.51%9.10%11.80%15.60%20.80%26.10%30.34%
Maritime0.00%0.00%0.00%0.00%0.00%0.00%0.00%
Air (Domestic)1.10%1.85%1.38%1.26%0.80%0.50%0.00%
Table A8. Projected fuel shares by transport mode for the EU-27 Rail Scenario (2025–2050).
Table A8. Projected fuel shares by transport mode for the EU-27 Rail Scenario (2025–2050).
ModeFuel Type202520302035204020452050
RailLiquid18.50%14.80%11.10%7.40%3.70%0.00%
Electricity81.50%85.20%88.90%92.60%96.30%100.00%
Road Heavy (HGV)Liquid94.50%80.00%65.00%50.00%35.00%20.00%
Gas2.00%5.00%8.00%10.00%12.00%15.00%
Electricity3.50%15.00%27.00%40.00%53.00%65.00%
Road Light (LCV)Liquid92.00%75.00%55.00%35.00%15.00%5.00%
Gas3.00%5.00%5.00%5.00%5.00%5.00%
Electricity5.00%20.00%40.00%60.00%80.00%90.00%
MaritimeLiquid98.00%93.00%88.00%83.00%78.00%73.00%
Gas2.00%7.00%12.00%17.00%22.00%27.00%
Air (Domestic)Liquid100.00%100.00%100.00%100.00%100.00%100.00%
Table A9. Projected installed capacity for renewable sources and storage (TW) in the EU27 (2020–2050).
Table A9. Projected installed capacity for renewable sources and storage (TW) in the EU27 (2020–2050).
Variable [TW]2020202520302035204020452050
Hydro0.10610.10720.10870.10870.10870.10870.1087
Geothermal elec.0.00090.00100.00170.00430.00670.00750.0084
Solid Bioenergy0.02070.01990.01920.01860.01580.01460.0117
Oceanic energy0.00030.00030.00050.00050.00050.00060.0006
Onshore wind0.16270.21730.31150.40590.47850.52990.5957
Offshore wind0.01460.02400.08310.14970.18910.21290.2212
Solar PV0.12730.33770.61790.84341.17181.49581.7393
CSP0.00230.00230.00520.00540.00540.00640.0064
Pumped Hydro (PHS)0.04520.04740.05350.05520.05520.05560.0556
Table A10. Material intensity factors for different battery architectures (kg/kWh). Calculated from [36].
Table A10. Material intensity factors for different battery architectures (kg/kWh). Calculated from [36].
MaterialNCM 111LFPNMC Leg.NMC 811NCA
Copper (Cu)0.75300.85350.74140.73610.7449
Nickel (Ni)0.35120.00000.52750.67060.6706
Lithium (Li)0.14280.10040.12690.11360.1037
Table A11. Economic sectors aggregation used in pymedeas2, with 49 industries, 4 final consumption and 2 value added categories. The corresponding sector number from the EXIOBASE classification is also added (c.f. Table A12).
Table A11. Economic sectors aggregation used in pymedeas2, with 49 industries, 4 final consumption and 2 value added categories. The corresponding sector number from the EXIOBASE classification is also added (c.f. Table A12).
No.Sector NameExio.No.Sector NameExio.
Industries 30Metallurgy72–82
1Accommodation, food and beverage services11731Other business activities133
2Activities of membership organizations157, 16132Other personal service activities159
3Agriculture, livestock and related services1–1733Paper and paper products industry52–54
4Air transport12334Postal and telecomunication activities125
5Chemical industry58–6335Private households with employed pers.160
6Coke and petroleum refineries56–5736Public admin, defense and social sec.134
7Computer activities and info services13137Real estate activities129
8Construction111–11238Rec., cultural and sporting activities158
9Education13539Rental activities130
10Electricity, gas, steam and air-cond.94–10940Research and development132
11Extractive industries, mining and quarrying20–3441Retail trade, exc. motor vehicles114, 116
12Financial intermediation12642Sale and repair of motor vehicles113
13Fishing and aquaculture1943Sanitation, waste management and decont.92–93, 137–156
14Food, beverages and tobacco industries35–4644Support activities for financial interm.128
15Forestry and logging1845Textile, clothing, leather and footwear47–49
16Graphic arts and recorded media5546Warehousing and support activities124
17Healthcare and social work activities13647Water collection, treatment and supply110
18Insurance and pension funds12748Wholesale trade and intermediation115
19Land transport; pipeline transport118–12049Wood and cork industry50–51
20Manuf. computer, electronic and optical85, 87–88
21Manuf. metal products, exc. machinery83Value Added
22Manuf. motor vehicles, trailers891Labor Compensation
23Manuf. of machinery and equipment n.e.c.842Capital Compensation
24Manuf. of other non-metallic mineral prod.65–71
25Manufacture of electrical equipment86Final Demand
26Manufacture of furniture; manuf. n.e.c.911Household Consumption
27Manufacture of other transport equipment902Government Expenditure
28Manufacture of rubber and plastic products643Gross Fixed Capital Formation
29Maritime and inland water transport121–1224Changes in Inventories
Table A12. Complete list of the 163 EXIOBASE industries. For more detail, see [21].
Table A12. Complete list of the 163 EXIOBASE industries. For more detail, see [21].
#Sector#Sector#Sector#Sector
1Cultivation rice42Sugar refining83Reproc. other nonfer124Inland water trans.
2Cultivation wheat43Proc. food prod nec84Casting of metals125Air transport
3Cultiv. cereal nec44Manuf. beverages85Manuf. fabric. metal126Support/aux trans.
4Cultiv. veg/fruit45Manuf. fish prod.86Manuf. machinery nec127Post and telecom
5Cultiv. oil seeds46Manuf. tobacco prod87Manuf. office/comp.128Financial intermed.
6Cultiv. sugar crops47Manuf. textiles88Manuf. elec. mach.129Insurance/pension
7Cultiv. plant fiber48Manuf. apparel/fur89Manuf. communic. eq.130Aux. financial act.
8Cultiv. crops nec49Manuf. leather prod90Manuf. med/opt inst131Real estate act.
9Cattle farming50Manuf. wood prod.91Manuf. motor vehicle132Renting machinery
10Pigs farming51Reproc. wood mat.92Manuf. other trans.133Computer activities
11Poultry farming52Pulp93Manuf. furniture nec134RandD
12Meat animals nec53Reproc. paper/pulp94Recycling waste135Other business act.
13Animal products nec54Paper95Recycling bottles136Public admin/defense
14Raw milk55Publishing and print.96Elec. prod: coal137Education
15Wool and silk cocoons56Manuf. coke prod.97Elec. prod: gas138Health and social work
16Manure trt. (conv.)57Petroleum Refinery98Elec. prod: nuclear139Incineration: Food
17Manure trt (biogas)58Proc. nuclear fuel99Elec. prod: hydro140Incineration: Paper
18Forestry and logging59Plastics, basic100Elec. prod: wind141Inciner. Plastic
19Fishing and services60Reproc. plastic101Elec. prod: oil142Inciner. Metals
20Mining coal and peat61N-fertiliser102Elec. prod: biomass143Inciner. Textiles
21Extr. crude petrol.62P- and other fert.103Elec. prod: solar PV144Inciner. Wood
22Extr. natural gas63Chemicals nec104Elec. prod: solar th145Inciner. Oil/Haz
23Extr. petro/gas nec64Manuf. rubber/plast105Elec. prod: tide/wav146Biogas. food waste
24Mining uranium/thor65Manuf. glass prod.106Elec. prod: geotherm147Biogas. paper
25Mining iron ores66Reproc. glass107Elec. prod: nec148Biogas. sewage
26Mining copper ores67Manuf. ceramic108Transmission of elec149Compost food waste
27Mining nickel ores68Manuf. bricks/tiles109Dist/trade elec.150Compost paper/wood
28Mining alum. ores69Manuf. cement/lime110Manuf/dist gas151Waste water, food
29Mining precious met70Reproc. ash/clinker111Steam/hot water152Waste water, other
30Mining Pb, Zn, Sn71Manuf. non-metal nec112Water supply/dist.153Landfill: Food
31Mining other nonfer72Manuf. iron and steel113Construction154Landfill: Paper
32Quarrying stone73Reproc. steel114Reproc. constr. mat.155Landfill: Plastic
33Quarrying sand/clay74Prod. precious met.115Sale/rep motor veh.156Landfill: Inert/Met
34Mining chem/fert nec75Reproc. precious met116Retail auto fuel157Landfill: Textiles
35Proc. meat cattle76Aluminium product.117Wholesale trade158Landfill: Wood
36Proc. meat pigs77Reproc. aluminium118Retail trade159Membership orgs nec
37Proc. meat poultry78Prod. Pb, Zn, Sn119Hotels and restaurants160Rec/cult/sport act.
38Meat products nec79Reproc. Pb, Zn, Sn120Rail transport161Other service act.
39Proc. veg oils/fats80Copper production121Other land transport162Private households
40Proc. dairy prod.81Reproc. copper122Pipeline transport163Extra-territor. orgs
41Processed rice82Prod. other nonfer.123Sea/coast transport
Table A13. Aggregation of the IEA fuels into the five fuel categories used in pymedeas2.
Table A13. Aggregation of the IEA fuels into the five fuel categories used in pymedeas2.
No.Energy CarrierNo.Energy Carrier
Electricity23Bio jet kerosene
1Electricity24Other liquid biofuels
2Hydro25Fuel oil
3Wind26Refinery gas
4Solar photovoltaics27Petroleum coke
5NuclearGases
6Electricity1Gas works gas
Heat2Coke oven gas
1Heat3Blast furnace gas
2Geothermal4Other recovered gases
3Solar thermal5Natural gas
Liquids6Biogases
1Crude oilSolids
2Natural gas liquids1Gas coke
3Refinery feedstocks2Municipal waste (non-renewable)
4Additive/blending components3Hard coal (if no detail)
5Other hydrocarbons4Brown coal (if no detail)
6Crude/NGL/feedstocks (if no detail)5Anthracite
7Ethane6Coking coal
8Liquefied petroleum gases (LPG)7Other bituminous coal
9Motor gasoline excl. biofuels8Sub-bituminous coal
10Aviation gasoline9Lignite
11Gasoline type jet fuel10Patent fuel
12Kerosene type jet fuel excl. biofuels11Coke oven coke
13Other kerosene12Coal tar
14Gas/diesel oil excl. biofuels13BKB
15Naphtha14Peat
16White spirit and SBP15Peat products
17Lubricants16Oil shale and oil sands
18Bitumen17Industrial waste
19Paraffin waxes18Municipal waste (renewable)
20Other oil products19Primary solid biofuels
21Biogasoline20Non-specified primary biofuels and waste
22Biodiesels21Charcoal

References

  1. European Commission; Joint Research Centre (JRC); International Energy Agency (IEA). EDGAR (Emissions Database for Global Atmospheric Research) Community GHG Database, Version EDGAR_2025_GHG; European Commission, JRC (Datasets): Ispra, Italy, 2025. [Google Scholar]
  2. Crippa, M.; Guizzardi, D.; Pagani, F.; Banja, M.; Muntean, M.; Schaaf, E.; Quadrelli, R.; Risquez Martin, A.; Taghavi-Moharamli, P.; Köykkä, J.; et al. GHG Emissions of All World Countries—2025 Report; Technical Report JRC143227; Publications Office of the European Union: Luxembourg, 2025. [Google Scholar] [CrossRef]
  3. Rozsai, M.; Jaxa-Rozen, M.; Salvucci, R.; Sikora, P.; Gea Bermudez, J.; Neuwahl, F. JRC-IDEES-2023. European Commission, Joint Research Centre (JRC) [Dataset]. 2025. Available online: http://data.europa.eu/89h/1f0b480c-6d21-4d95-897d-20c7ca33df6f (accessed on 13 July 2026).
  4. Transport and Environment. Europe’s Transport Sector Set to Make Up Almost Half of the Continent’s Emissions in 2030; Policy Analysis; Transport and Environment: Brussels, Belgium, 2024. [Google Scholar]
  5. European Commission. The European Green Deal; COM(2019) 640 Final; European Commission: Brussels, Belgium, 2019. [Google Scholar]
  6. European Commission. Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions. ’Fit for 55’: Delivering the EU’s 2030 Climate Target on the Way to Climate Neutrality; European Commission: Brussels, Belgium, 2021. [Google Scholar]
  7. Hickel, J.; Kallis, G. Is Green Growth Possible? New Political Econ. 2020, 25, 469–486. [Google Scholar] [CrossRef]
  8. Parrique, T.; Barth, J.; Briens, F.; Kerschner, C.; Kraus-Polk, A.; Magee, A.; Spangenberg, J.H. Decoupling Debunked: Evidence and Arguments Against Green Growth as a Sole Strategy for Sustainability; Technical report; European Environmental Bureau (EEB): Brussels, Belgium, 2019. [Google Scholar]
  9. Wiedmann, T.; Lenzen, M.; Keyßer, L.T.A.; Steinberger, J.K. Scientists’ warning on affluence. Nat. Commun. 2020, 11, 3107. [Google Scholar] [CrossRef] [PubMed]
  10. García-Olivares, A.; Solé, J.; Osychenko, O. Transportation in a 100% renewable energy system. Energy Convers. Manag. 2018, 158, 266–285. [Google Scholar] [CrossRef]
  11. García-Olivares, A.; Solé, J.; Samsó, R.; Ballabrera-Poy, J. Sustainable European Transport System in a 100% Renewable Economy. Sustainability 2020, 12, 5091. [Google Scholar] [CrossRef]
  12. Valero, A.; Valero, A.; Calvo, G.; Ortego, A. Material bottlenecks in the future development of green technologies. Renew. Sustain. Energy Rev. 2018, 93, 178–200. [Google Scholar] [CrossRef]
  13. Alcover, E.; Mas, B.; Martínez-Moll, V.; Rosselló, J.L.; Roca, M.; Canals, V. Energetic and economic analysis of the Electric Vehicles charge impacts on public parking lots. Renew. Energy Power Qual. J. (REandPQJ) 2020, 18, 741–746. [Google Scholar] [CrossRef]
  14. Creutzig, F.; Niamir, L.; Bai, X.; Callaghan, M.; Cullen, J.; Díaz-José, J.; Figueroa, M.; Grübler, A.; Lamb, W.; Leip, A.; et al. Demand, Services and Social Aspects of Mitigation. In Climate Change 2022: Mitigation of Climate Change. Contribution of Working Group III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change; Cambridge University Press: Cambridge, UK; New York, NY, USA, 2022. [Google Scholar] [CrossRef]
  15. Solé, J.; Samsó, R.; García-Ladona, E.; García-Olivares, A.; Ballabrera-Poy, J.; Madurell, T.; Turiel, A.; Osychenko, O.; Álvarez, D.; Bardi, U.; et al. modeling the renewable transition: Scenarios and pathways for a decarbonized future using pymedeas, a new open-source energy systems model. Renew. Sustain. Energy Rev. 2020, 132, 110105. [Google Scholar] [CrossRef]
  16. Michaux, S.P. Assessment of the Extra Capacity Required of Alternative Energy Electrical Power Systems to Completely Replace Fossil Fuels; Open File Report 42/2021; Geological Survey of Finland (GTK): Espoo, Finland, 2021. [Google Scholar] [CrossRef]
  17. Muratori, M. Impact of uncoordinated plug-in electric vehicle charging on residential power demand. Nat. Energy 2018, 3, 193–201. [Google Scholar] [CrossRef]
  18. Alsharif, A.; Tan, C.; Ayop, R.; Al Smin, A.; Ali Ahmed, A.; Kuwil, F.; Khaleel, M. Impact of Electric Vehicle on Residential Power Distribution Considering Energy Management Strategy and Stochastic Monte Carlo Algorithm. Energies 2023, 16, 1358. [Google Scholar] [CrossRef]
  19. Neuwahl, F.; Wegener, M.; Salvucci, R.; Jaxa-Rozen, M.; Gea Bermudez, J.; Sikora, P.; Rózsai, M. Clean Energy Technology Observatory: POTEnCIA CETO 2025 Scenario; Publications Office of the European Union: Luxembourg, 2025. [Google Scholar] [CrossRef]
  20. Alcover Comas, E.; Guardia Calsina, P.; Samsó, R.; Solé, J. Exploring decarbonisation strategies in the Southern-European region of Catalonia through a multi-scenario assessment. Energy Strategy Rev. 2026, 64, 102184. [Google Scholar] [CrossRef]
  21. Stadler, K.; Wood, R.; Bulavskaya, T.; Södersten, C.J.; Simas, M.; Schmidt, S.; Usubiaga, A.; Acosta-Fernández, J.; Kuenen, J.; Bruckner, M.; et al. EXIOBASE 3: Developing a Time Series of Detailed Environmentally Extended Multi-Regional Input-Output Tables. J. Ind. Ecol. 2018, 22, 502–515. [Google Scholar] [CrossRef]
  22. Mantzos, L.; Wiesenthal, T. POTEnCIA Model Description, Version 0.9; Publications Office of the European Union: Luxembourg, 2016. [CrossRef]
  23. European Union. Regulation (EU) 2021/1119 of the European Parliament and of the Council of 30 June 2021 establishing the framework for achieving climate neutrality and amending Regulations (EC) No 401/2009 and (EU) 2018/1999 (European Climate Law). Off. J. Eur. Union 2021, L 243, 1–17. [Google Scholar]
  24. European Commission. Securing Our Future Europe’s 2040 Climate Target and Path to Climate Neutrality by 2050 Building a Sustainable, Just and Prosperous Society; Technical Report COM(2024) 63 final; European Commission: Brussels, Belgium, 2024. [Google Scholar]
  25. European Parliament and Council. Regulation (EU) 2023/2405 on ensuring a level playing field for sustainable air transport (ReFuelEU Aviation). Off. J. Eur. Union 2023, L 243, 1–30. [Google Scholar]
  26. European Parliament and Council. Regulation (EU) 2023/1805 on the use of renewable and low-carbon fuels in maritime transport (FuelEU Maritime). Off. J. Eur. Union 2023, L 243, 48–100. [Google Scholar]
  27. Buffi, M.; Motola, V.; Rejtharova, J.; Hurtig, O.; Scarlat, N.; Mountraki, A.; Georgakaki, A.; Salvucci, R.; Schade, B.; Rózsai, M.; et al. Clean Energy Technology Observatory: Renewable Fuels of Non-Biological Origin in the European Union—2025 Status Report on Technology Development, Trends, Value Chains and Markets; Publications Office of the European Union: Luxembourg, 2025. [Google Scholar] [CrossRef]
  28. European Parliament and Council. Regulation (EU) 2024/1610 amending Regulation (EU) 2019/1242 as regards strengthening the CO2 emission performance standards for new heavy-duty vehicles and integrating reporting obligations, amending Regulation (EU) 2018/858 and repealing Regulation (EU) 2018/956 (Text with EEA relevance). Off. J. Eur. Union 2024, L 243, 1–67. [Google Scholar]
  29. European Commission. Revised TEN-T Regulation and Infrastructure for Sustainable Mobility: Impact Assessment and Projections; Technical report; Directorate-General for Mobility and Transport (DG MOVE): Brussels, Belgium, 2024. [Google Scholar]
  30. ACEA. Report: Vehicles on EUROPEAN Roads 2025; Technical report; European Automobile Manufacturers’ Association: Brussels, Belgium, 2025. [Google Scholar]
  31. European Commission. EU Reference Scenario 2024: Energy, Transport and GHG Emissions—Trends to 2050; Technical report; Directorate-General for Energy, DG Climate Action and DG Mobility and Transport: Brussels, Belgium, 2024. [Google Scholar]
  32. Daly, H.E. Steady-State Economics: Second Edition with New Essays; Island Press: Washington, DC, USA, 1991. [Google Scholar]
  33. Zell-Ziegler, C.; Thema, J.; Best, B.; Wiese, F.; Lage, J.; Schmidt, A.; Toulouse, E.; Stagl, S. Enough? The role of sufficiency in European energy and climate plans. Energy Policy 2021, 157, 112483. [Google Scholar] [CrossRef]
  34. Steffen, W.; Richardson, K.; Rockström, J.; Cornell, S.E.; Fetzer, I.; Bennett, E.M.; Biggs, R.; Carpenter, S.R.; de Vries, W.; de Wit, C.A.; et al. Planetary boundaries: Guiding human development on a changing planet. Science 2015, 347, 1259855. [Google Scholar] [CrossRef] [PubMed]
  35. O’Neill, D.W.; Fanning, A.L.; Lamb, W.F.; Steinberger, J.K. A good life for all within planetary boundaries. Nat. Sustain. 2018, 1, 88–95. [Google Scholar] [CrossRef]
  36. Xu, C.; Dai, Q.; Gaines, L.; Hu, M.; Tukker, A.; Steubing, B. Future material demand for automotive lithium-based batteries. Commun. Mater. 2020, 1, 99. [Google Scholar] [CrossRef]
  37. Xylia, M.; Olsson, E.; Macura, B.; Nykvist, B. Estimating charging infrastructure demand for electric vehicles: A systematic review. Energy Strategy Rev. 2025, 59, 101753. [Google Scholar] [CrossRef]
  38. Valero, A.; Valero, A.; Calvo, G. The Material Limits of Energy Transition: Thanatia; Springer: Cham, Switzerland, 2023. [Google Scholar] [CrossRef]
  39. European Parliament and Council of the European Union. Regulation (EU) 2023/1542 of the European Parliament and of the Council of 12 July 2023 concerning batteries and waste batteries, amending Directive 2008/98/EC and Regulation (EU) 2019/1020 and repealing Directive 2006/66/EC. Off. J. Eur. Union 2023, L 191, 1–117. [Google Scholar]
  40. United Nations Environment Programme. Emissions Gap Report 2024: No More Hot Air … Please! Technical report; UNEP: Nairobi, Kenya, 2024. [Google Scholar]
  41. European Scientific Advisory Board on Climate Change. Towards EU Climate Neutrality: Progress, Policy Gaps and Opportunities; Technical report; European Environment Agency: Copenhagen, Denmark, 2024.
  42. Plötz, P.; Wachsmuth, J.; Gnann, T. Net-Zero-Carbon Transport in Europe Until 2050—Targets, Technologies and Policies for a Long-Term EU Strategy; Technical report; Fraunhofer Institute for Systems and Innovation Research (ISI): Karlsruhe, Germany, 2021. [Google Scholar]
  43. International Energy Agency. Global EV Outlook 2026: Growing Sales Amid an Energy Crisis; Technical report; IEA: Paris, France, 2026. [Google Scholar]
  44. Brand, C.; Götschi, T.; Dons, E.; Gerike, R.; Anaya-Boig, E.; Avila-Palencia, I.; de Nazelle, A.; Gascon, M.; Gaupp-Berghausen, M.; Iacorossi, F.; et al. The climate change mitigation impacts of active travel: Evidence from a longitudinal panel study in seven European cities. Glob. Environ. Change 2021, 67, 102224. [Google Scholar] [CrossRef]
  45. European Environment Agency. Trends and Projections in Europe 2024; EEA Report 08/2024; EEA: Copenhagen, Denmark, 2024. [Google Scholar]
  46. Grubler, A.; Wilson, C.; Bento, N.; Boza-Kiss, B.; Krey, V.; McCollum, D.L.; Rao, N.D.; Riahi, K.; Rogelj, J.; De Stercke, S.; et al. A low energy demand scenario for meeting the 1.5 °C target and sustainable development goals without negative emission technologies. Nat. Energy 2018, 3, 515–527. [Google Scholar] [CrossRef]
  47. U.S. Geological Survey. Mineral Commodity Summaries 2026; Technical report; U.S. Geological Survey: Reston, VA, USA, 2026. [CrossRef]
  48. Goonan, T.G. Lithium Use in Batteries; Technical Report 1371; U.S. Geological Survey: Reston, VA, USA, 2012. [CrossRef]
  49. Harper, G.; Sommerville, R.; Kendrick, E.; Driscoll, L.; Slater, P.; Stolkin, R.; Walton, A.; Christensen, P.; Heidrich, O.; Lambert, S.; et al. Recycling lithium-ion batteries from electric vehicles. Nature 2019, 575, 75–86. [Google Scholar] [CrossRef] [PubMed]
  50. Gregoir, L.; van Acker, K. Metals for Clean Energy: Pathways to Solving Europe’s Raw Materials Challenge; Technical report; KU Leuven: Leuven, Belgium, 2022. [Google Scholar]
  51. Flyvbjerg, B. Policy and Planning for Large-Infrastructure Projects: Problems, Causes, Cures. Environ. Plan. B Plan. Des. 2007, 34, 578–597. [Google Scholar] [CrossRef]
  52. Hausberger, L.; Cordes, T.; Gschösser, F. Life Cycle Assessment of High-Performance Railway Infrastructure, Analysis of Superstructures in Tunnels and on Open Tracks. Sustainability 2023, 15, 7064. [Google Scholar] [CrossRef]
  53. Knabl, D.; Landgraf, M. In-Depth Lifecycle Assessment of Ballasted Railway Track and Slab Track Considering Varying Subsoil Conditions. Sustainability 2023, 15, 12974. [Google Scholar] [CrossRef]
  54. Boira, J.V.; Berzi, M. Liberalization, Trans-European Corridors and EU Funds: A New Scenario in the Relationship between Rail Networks and Mediterranean Cities. Land 2023, 12, 1986. [Google Scholar] [CrossRef]
  55. Creutzig, F.; Roy, J.; Lamb, W.F.; Azevedo, I.M.L.; de Bruin, W.B.; Dalkmann, H.; Edelenbosch, O.Y.; Geels, F.W.; Grubler, A.; Hepburn, C.; et al. Towards demand-side solutions for mitigating climate change. Nat. Clim. Change 2018, 8, 260–263. [Google Scholar] [CrossRef]
  56. Jarre, M.; Noussan, M.; Campisi, E. Avoid–Shift–Improve: Are Demand Reduction Strategies Under-Represented in Current Energy Policies? Energies 2024, 17, 4955. [Google Scholar] [CrossRef]
  57. Draghi, M. The Future of European Competitiveness; Technical report; European Commission: Brussels, Belgium, 2024. [Google Scholar] [CrossRef]
  58. Jackson, T. Post Growth: Life After Capitalism; Polity Press: Cambridge, UK, 2021. [Google Scholar]
  59. Morris, M.D. Factorial sampling plans for preliminary computational experiments. Technometrics 1991, 33, 161–174. [Google Scholar] [CrossRef]
  60. Saltelli, A.; Ratto, M.; Andres, T.; Campolongo, F.; Cariboni, J.; Gatelli, D.; Saisana, M.; Tarantola, S. Global Sensitivity Analysis: The Primer; John Wiley and Sons: Hoboken, NJ, USA, 2007. [Google Scholar] [CrossRef]
  61. Herman, J.; Usher, W. SALib: An open-source Python library for sensitivity analysis. J. Open Source Softw. 2017, 2, 97. [Google Scholar] [CrossRef]
  62. Samsó, R.; De Blas, I.; Perissi, I.; Martelloni, G.; Solé, J. Scenario analysis and sensitivity exploration of the MEDEAS Europe energy-economy-environment model. Energy Strategy Rev. 2020, 32, 100582. [Google Scholar] [CrossRef]
  63. Hickel, J.; Brockway, P.; Kallis, G.; Keyßer, L.; Lenzen, M.; Slameršak, A.; Steinberger, J.; Ürge-Vorsatz, D. Urgent need for post-growth climate mitigation scenarios. Nat. Energy 2021, 6, 766–768. [Google Scholar] [CrossRef]
  64. International Energy Agency (IEA). World Energy Outlook 2025 Extended Dataset; Licence: Terms of Use for Non-CC Material; International Energy Agency (IEA): Paris, France, 2025. [Google Scholar]
  65. Rasul, K.; Schmidt, S.; Hertwich, E.G.; Wood, R. EXIOBASE energy accounts: Improving precision in an open-sourced procedure applicable to any MRIO database. J. Ind. Ecol. 2024, 28, 1771–1785. [Google Scholar] [CrossRef]
  66. Hyndman, R.J.; Athanasopoulos, G. Forecasting: Principles and Practice, 2nd ed.; OTexts: Melbourne, Australia, 2018. [Google Scholar]
  67. Neidhardt, M.; Mas-Peiro, J.; Schulz-Moenninghoff, M.; Pou, J.O.; Gonzalez-Olmos, R.; Kwade, A.; Schmuelling, B. Forecasting the Global Battery Material Flow: Analyzing the Break-Even Points at Which Secondary Battery Raw Materials Can Substitute Primary Materials. Appl. Sci. 2022, 12, 4790. [Google Scholar] [CrossRef]
  68. Danielis, R.; Niazi, A.M.K.; Scorrano, M.; Masutti, M.; Awan, A.M. The Economic Feasibility of Battery Electric Trucks: A Review of the Total Cost of Ownership Estimates. Energies 2025, 18, 429. [Google Scholar] [CrossRef]
  69. Ghotge, R.; van Rooij, D.; van Breukelen, S. Total Cost of Ownership of Electric Buses in Europe. World Electr. Veh. J. 2025, 16, 464. [Google Scholar] [CrossRef]
  70. Research and Markets. Electric Locomotives Market Size, Share and Forecast; Technical report; Research and Markets: Dublin, Ireland, 2026. [Google Scholar]
  71. Nicholas, M.; Hall, D. Lessons Learned on Charging Infrastructure Costs and Deployment; Technical report; International Council on Clean Transportation (ICCT): Washington, DC, USA, 2022. [Google Scholar]
  72. Wood, E.; Borlaug, B.; Moniot, M.; Lee, D.-Y.; Ge, Y.; Yang, F.; Liu, Z. The 2030 National Charging Network: Estimating U.S. Light-Duty Demand for Electric Vehicle Charging Infrastructure; Technical Report NREL/TP-5400-85654; National Renewable Energy Laboratory (NREL): Golden, CO, USA, 2023. [CrossRef]
  73. Hao, H.; Geng, Y.; Tate, J.E.; Liu, Z.; Sun, X.; Mu, Z.; Sun, L.; Zhao, F. Impact of transport electrification on critical metal sustainability with a focus on the heavy-duty segment. Nat. Commun. 2019, 10, 5398. [Google Scholar] [CrossRef] [PubMed]
  74. Wagner, D.; Walther, G. Techno-economic analysis of mixed battery and fuel cell electric bus fleets: A case study. Appl. Energy 2024, 376, 124285. [Google Scholar] [CrossRef]
  75. Mordor Intelligence. Electric Bus Battery Market Size and Share Analysis—Growth Trends and Forecasts (2024–2029); Market research report; Mordor Intelligence: Telangana, India, 2024. [Google Scholar]
Figure 1. Structural archetypes for the 2050 transport system. Panel A illustrates the passenger modal split (pkm), highlighting the transition from private vehicle dominance in the Reference Scenario to a rail-centric collective system. Panel B displays the freight modal split (tkm), where the Rail Scenario applies an “all-rail” logic to long-haul logistics, reserving road transport (LCV) primarily for capillary distribution. Note that while absolute activity volumes vary between Growth (-G) and Steady-State (-SSE) variants, these modal shares define the structural logic of each pathway.
Figure 1. Structural archetypes for the 2050 transport system. Panel A illustrates the passenger modal split (pkm), highlighting the transition from private vehicle dominance in the Reference Scenario to a rail-centric collective system. Panel B displays the freight modal split (tkm), where the Rail Scenario applies an “all-rail” logic to long-haul logistics, reserving road transport (LCV) primarily for capillary distribution. Note that while absolute activity volumes vary between Growth (-G) and Steady-State (-SSE) variants, these modal shares define the structural logic of each pathway.
Sustainability 18 07402 g001
Figure 2. Evolution of Passenger Transport Demand and Modal Share (2020–2050). The projections illustrate distinct transition pathways, contrasting institutional technology substitution with structural modal shifts under two macroeconomic paradigms.
Figure 2. Evolution of Passenger Transport Demand and Modal Share (2020–2050). The projections illustrate distinct transition pathways, contrasting institutional technology substitution with structural modal shifts under two macroeconomic paradigms.
Sustainability 18 07402 g002
Figure 3. Evolution of Freight Transport Demand and Modal Share (2020–2050). The projections illustrate a structural shift, replacing road-based long-haul freight with electrified rail networks under the two macroeconomic scenarios.
Figure 3. Evolution of Freight Transport Demand and Modal Share (2020–2050). The projections illustrate a structural shift, replacing road-based long-haul freight with electrified rail networks under the two macroeconomic scenarios.
Sustainability 18 07402 g003
Figure 4. Evolution of Final Energy Demand (FED) and CO2 emissions (2020–2050). (a) Total EU-27 FED, capturing the systemic effects of transport modal shifts and macroeconomic stabilization; (b) Total CO2 emissions, showing the synergy between modal shifts and macroeconomic stabilization; (c) Transport sector’s FED evolution, contrasting the gradual reduction in Reference pathways with the Rail variats; (d) Transport-related emissions, highlighting the decarbonization lag of technology-centric scenarios.
Figure 4. Evolution of Final Energy Demand (FED) and CO2 emissions (2020–2050). (a) Total EU-27 FED, capturing the systemic effects of transport modal shifts and macroeconomic stabilization; (b) Total CO2 emissions, showing the synergy between modal shifts and macroeconomic stabilization; (c) Transport sector’s FED evolution, contrasting the gradual reduction in Reference pathways with the Rail variats; (d) Transport-related emissions, highlighting the decarbonization lag of technology-centric scenarios.
Sustainability 18 07402 g004
Figure 5. Cumulative Lithium requirements for EV batteries through 2050. The results contrast the material intensity of private fleet electrification (REF-G and REF-SSE) with the mineral parsimony of structural modal shifts (RAIL-G and RAIL-SSE).
Figure 5. Cumulative Lithium requirements for EV batteries through 2050. The results contrast the material intensity of private fleet electrification (REF-G and REF-SSE) with the mineral parsimony of structural modal shifts (RAIL-G and RAIL-SSE).
Sustainability 18 07402 g005
Table 1. Classification and scope of transport modes included in the model.
Table 1. Classification and scope of transport modes included in the model.
CategoryModeScope and Inclusions
PassengersPrivate RoadIndividual cars and light-duty passenger vehicles
Collective RoadUrban and interurban bus services
RailHigh-speed, regional, and metropolitan networks
Domestic AirInternal flights within the modeled region
MaritimeRegional ferries and passenger shipping
FreightRoad HeavyHeavy goods vehicles (>3.5 t)
Road LightLight commercial vehicles (≤3.5 t)
RailElectric and non-electric cargo trains
MaritimeShort-sea shipping and inland waterways
AirDomestic cargo and logistical flights
Table 2. Summary of cumulative investment and physical infrastructure deployment (2025–2050).
Table 2. Summary of cumulative investment and physical infrastructure deployment (2025–2050).
Indicator (Total 2025–2050)REF-GREF-SSERAIL-GRAIL-SSE
Capital Investment (Trillion USD)
Private Vehicles17.2015.718.137.50
Buses0.780.710.780.71
Freight Vehicles (LCV + HGV)7.126.394.143.78
Rail Rolling Stock (Locomotives)0.200.170.960.86
Charging Infrastructure ( I c h a r g e )1.671.520.880.81
Rail Infrastructure (Tracks) ( I r a i l )0.140.147.606.16
Total Investment ( I t o t a l )27.1024.6422.5019.83
Physical Deployment
New Rail Lines (km)00451,322364,032
Rail Electrification (km)108,784108,784108,784108,784
AC Charging Posts (Millions)91.8283.6848.3944.43
DC Fast Chargers (Millions)19.7417.9910.419.56
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Alcover Comas, E.; Martínez Marín, P.; Samsó, R.; Solé, J. Beyond Green Growth: Evaluating the Biophysical Feasibility of Rail-Centric Transport Systems in the European Union. Sustainability 2026, 18, 7402. https://doi.org/10.3390/su18147402

AMA Style

Alcover Comas E, Martínez Marín P, Samsó R, Solé J. Beyond Green Growth: Evaluating the Biophysical Feasibility of Rail-Centric Transport Systems in the European Union. Sustainability. 2026; 18(14):7402. https://doi.org/10.3390/su18147402

Chicago/Turabian Style

Alcover Comas, Enric, Pau Martínez Marín, Roger Samsó, and Jordi Solé. 2026. "Beyond Green Growth: Evaluating the Biophysical Feasibility of Rail-Centric Transport Systems in the European Union" Sustainability 18, no. 14: 7402. https://doi.org/10.3390/su18147402

APA Style

Alcover Comas, E., Martínez Marín, P., Samsó, R., & Solé, J. (2026). Beyond Green Growth: Evaluating the Biophysical Feasibility of Rail-Centric Transport Systems in the European Union. Sustainability, 18(14), 7402. https://doi.org/10.3390/su18147402

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop