ssrn-6359303

Energy Geopolitics in the Arctic: A Scenario-Based Strategic Analysis *Konstantinos Pappas1,2,3, Zainab Ashkanani1,2 1. Texas A&M Energy Institute, Texas A&M University, College Station, TX 77843, United States 2. Biological and Agricultural Engineering Department, Texas A&M University, College Station, TX 77843, United States 3. Borders & Migration Program, Mosbacher Institute for Trade, Economics, and Public Policy, Bush School of Government and Public Service, Texas A&M University, College Station, TX 77843, United States Email: *kostis.pappas@tamu.edu , Phone: +1 979-862-5927 Abstract The Arctic represents a critical theater where accelerating ice melting, vast resource potential, and great power competition converge to reshape global strategic dynamics. This paper provides a comprehensive, theory-driven analysis of Arctic geopolitics through the classical geostrategic frameworks. Mackinder’s Heartland, Spykman’s Rimland, and Mahan’s Sea Power theories, examining policies of key state- and non-state-actors. Russia and China leverage a fusion of Heartland and Sea Power doctrines to establish a formidable, sanctions-resistant strategic axis, contrasted with the Western coalition’s Rimland-based approach hampered by coordination challenges and investment deficits. Using scenario modeling, we project four plausible regional power distributions through 2035. Findings indicate that absent dramatic Western strategic commitment, Sino-Russian dominance of the Eastern Arctic is most probable, driven by irreversible first-mover advantages from operational expertise and multi-billion-dollar infrastructure investments. This creates a new geopolitical equilibrium with profound implications for global energy security and international relations. The study directly addresses a critical gap in energy strategy research by integrating geostrategic theory with quantitative energy infrastructure assessment, offering a predictive framework for how Arctic energy resources will be controlled, transported, and distributed under competing geopolitical alignments. These resources, comprising 13% of global undiscovered oil and 30% of natural gas reserves, represent a globally significant energy endowment whose governance will be shaped by the interplay of Eastern and Western bloc positioning. The analysis demonstrates that the Arctic energy competition carries direct implications for Sustainable Development Goal 7 (Affordable and Clean Energy), as the consolidation of energy supply routes under Eastern bloc control fundamentally reshapes global energy affordability, reliability, and access pathways. Keywords: Arctic, Energy Geopolitics, Energy Security, SDG7, Geostrategy, Sea Power, Scenario Analysis 1. Introduction: The Arctic as a new geopolitical frontier The Arctic is undergoing a rapid, multifaceted transformation driven by climate change and the unprecedented melting of sea ice [1, 2]. This physical reality is unlocking vast natural resources, including an estimated ninety billion barrels of undiscovered oil and 1,670 trillion cubic feet of natural gas, representing 13% and 30% of global undiscovered reserves, respectively, while simultaneously opening maritime trade routes that could reduce shipping distances by up to 42% [3]. The transformation of the Arctic from a region of “exceptional cooperation” to one of intensifying competition reflects both environmental realities and geopolitical tensions. Vessel traffic has increased 108% from 6.1 million nautical miles in 2013 to 12.7 million in 2024, and the Northern Sea Route cargo volume has grown from 1.5 million tons to a projected 35-40 million tons by 2025, creating economic facts on the ground that political declarations cannot easily reverse respectively [3, 4]. This paper contends that the current geopolitical dynamics are not random but are, consciously or not, modern manifestations of classical geostrategic theories. The strategies of key actors, i.e., Russia, China, the United States, and the European This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 1 of 22 -- Union, are re-engaging with the core tenets of Mackinder’s Heartland, Spykman’s Rimland, and Mahan’s Sea Power doctrines. The study employs a mixed-methods approach, first applying a theoretical framework to interpret policy and then, using a four-scenario model to project plausible future outcomes, based on current trends and quantitative data. By synthesizing these approaches, the paper moves beyond description to explain causal relationships and strategic logic, demonstrating that the unfolding competition is a predictable consequence of historical doctrines accelerated by environmental change and technological asymmetries. The primary aim of this research is to provide a comprehensive, theory-driven analysis of geopolitical competition in the Arctic, with a specific focus on its implications for global energy security and international power dynamics. To achieve this, the paper’s main objectives are to first apply and critically evaluate the applicability of classical geostrategic theories (Heartland, Rimland, and Sea Power) to the contemporary policies and actions of Russia, China, and the Western coalition [3]. The analysis will then proceed to analyze the strategic priorities of the key actors in the Arctic, i.e., Russia’s pursuit of energy dominance, China’s quest for energy diversification, the United States’ focus on security, and the European Union’s clean energy transition, and contextualize them within the broader geostrategic frameworks [1, 5]. The paper will then present a detailed scenario analysis for the period leading up to 2035, which will be supported by extensive quantitative data on infrastructure, investment, and trade volumes. Finally, this will lead to a discussion of the second- and third-order implications of these scenarios, including the role of governance vacuums, economic and technological asymmetries, and the paradoxical effects of international sanctions. Critically, this research positions Arctic competition as fundamentally an energy strategy question, rather than a purely geopolitical one. The Arctic’s energy resources constitute the primary driver of state investment and strategic positioning, and the outcomes of this competition will directly determine global energy supply diversification, pricing structures, and the feasibility of achieving SDG7 targets for affordable and clean energy access [6]. The geopolitical frameworks employed herein serve as analytical lenses, through which energy strategy decisions are interpreted, contributing to the growing body of literature that recognizes the inseparability of energy policy from geopolitical analysis [7, 8]. By framing the Arctic as a theater of energy strategy competition, this paper responds to calls within the energy research community for more integrated, forward-looking assessments that connect resource governance with broader sustainability objectives [9]. 2. Literature Review: Re-evaluating geostrategy in the 21st century Arctic 2.1. The Heartland Theory: Russia’s Arctic extension Harold Mackinder’s Heartland Theory, originally developed in 1904, posits that control of the Eurasian landmass, or “Heartland,” is the key to global power. Described as the “greatest natural fortress on Earth,” the Heartland’s inaccessibility by sea and its vast size were seen as formidable advantages. Mackinder initially applied this theory to Russia’s presence in the region and its ambitions to expand into periphery regions [3]. The Russian Arctic is home to massive natural gas reserves, with 83% of the nation’s 638 billion cubic meters of annual production originating from these territories [10]. Energy exports are a central pillar of Russia’s geopolitical leverage, comprising approximately 20% of its GDP [11, 12]. Control of the Northern Sea Route (NSR), which runs entirely through its internal waters, solidifies this Heartland-centric approach, transforming the NSR from a simple trade passage into a projection of Russia’s land-based power [13, 14]. The Russian approach to the Arctic is not a new or reactive policy but a long-standing strategic continuation of its Heartland doctrine, dating back to the Imperial Russian Ministry of Foreign Affairs’ territorial claims in 1916 [15]. This historical continuity suggests a deep-seated, systemic commitment that is far more resilient than the often-reactive and fragmented policies of Western nations. The extensive investment in its icebreaker fleet, numbering over 40 active vessels, including 10 nuclear-powered ships, and the modernization of Soviet-era military bases, are a modern-day execution of Mackinder’s principle [16, 17]. This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 2 of 22 -- The year-round operation of the 1,343 nautical mile Dudinka-Murmansk route since the 1978-79 winter season has provided over 45 years of continuous operational experience that cannot be purchased or quickly developed elsewhere. These capabilities are transforming the “natural fortress” from a defensive concept into a platform for proactive power projection. 2.2. The Rimland Theory: The Western Coalition’s strategy In contrast to Mackinder, Nicolas Spykman’s Rimland Theory, developed in 1942, argued that global power lay in controlling the coastal periphery of Eurasia, known as the “Rimland”. This framework emphasizes the importance of maritime proximity and access to the outside world, positioning a unified Rimland as a force capable of containing the Heartland’s influence. The North American and European Arctic powers, including the United States, Canada, Norway, Denmark (via Greenland), Sweden, and Finland, fit neatly into this theoretical framework. Their primary strategy is to form alliances, most notably NATO, to counter and contain potential Heartland influence [3, 18]. The United States’ foreign policy, as outlined in its National Strategy for the Arctic Region and the Department of Defense Arctic Strategy, is heavily focused on “Security/Defense” and “International Cooperation” with other Rimland states [19, 20]. While the Rimland approach is theoretically powerful, due to the combined economic and military strength of the Western alliance, the analysis reveals significant “coordination challenges” and “infrastructure gaps.” The coalition’s power remains latent and aggregated, whereas Russia’s is concentrated and deployed. The vast disparity in capabilities, such as Russia’s 20-to-1 operational advantage in icebreakers and its network of over 40 active Arctic military bases transforms the theoretical Rimland advantage into a practical disadvantage [17]. The legal dispute between the United States and Canada over the status of the Northwest Passage (NWP) further highlights the internal friction that hampers the coalition’s ability to present a unified and proactive Arctic strategy [17, 21]. 2.3. The Sea Power Theory: China’s maritime gambit Captain Alfred Thayer Mahan’s Sea Power Theory, articulated in 1890, posits that a nation’s ability to project power and secure global trade is dependent on its control of the seas (Britannica Editors, 2016). Despite not being an Arctic coastal state, China’s declaration as a “near-Arctic state” in its 2018 Arctic Policy White Paper is a clear and direct application of Mahan’s doctrine [22, 23]. China’s “Polar Silk Road” initiative, a maritime extension of its Belt and Road Initiative, is a strategic move to gain influence over Arctic maritime passages, specifically the NSR and the Transpolar Sea Route (TSR) [22, 23]. This strategy is driven by a desire to diversify its energy security and global trade routes, thus reducing its heavy dependence on vulnerable maritime chokepoints like the Strait of Malacca, through which 80% of its energy imports currently transit [10]. The key to China’s success in this endeavor is its investment strategy. By providing tens of billions of dollars in capital for Russian Arctic projects, such as the Yamal LNG facility and Arctic LNG 2, China is effectively buying influence and securing access to the very routes and resources that will define future maritime trade [24, 25]. Agent-based simulations of global LNG trade confirm the strategic rationale for this investment, projecting that under ice-free conditions the North-East Passage would channel 30–34 MTPA of LNG by 2035, predominantly of Russian origin and destined for Asian Pacific importers [26]. The partnership is compelling for China given that the NSR reduces the Shanghai-Rotterdam distance by 42% and saves approximately 14 days of transit time. This economic partnership creates a new and powerful geostrategic axis that combines Russia’s Heartland-based territorial control and operational expertise with China’s capital-driven Sea Power ambitions. The two nations established the Russian-Chinese Subcommission on Cooperation on the Northern Sea Route in June 2024, to support a joint venture to build ships and container lines [27]. This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 3 of 22 -- 2.4. Arctic Energy Strategy Literature: Gaps and Contributions The existing literature on Arctic energy development reveals several critical gaps that this study addresses. While substantial research has examined individual dimensions of Arctic geopolitics, resource assessments, shipping routes, or bilateral relationships, few studies have attempted to synthesize these elements within an integrated geostrategic framework that projects quantitative future outcomes. 2.4.1 Energy Strategy and Scenario Modeling in the Arctic Energy scenario research has evolved significantly over the past decade, with scholars increasingly recognizing that scenarios play a crucial role in energy policy-making, economic decisions, and public debate [28]. However, energy scenarios remain highly controversial: assumptions and methods are often not transparent, and the robustness of conclusions frequently remains unclear [28]. This methodological challenge is particularly acute in Arctic energy research, where geopolitical variables introduce additional uncertainty layers beyond standard techno-economic parameters. Early quantitative assessments of Arctic energy potential, such as the work by Leal on international oil company (IOC) strategies, demonstrated that, while Arctic resources appear comparatively attractive given undiscovered potential, drilling costs can be substantially higher due to remoteness and technical challenges [29]. Leal’s analysis of the five major IOCs revealed that these companies invested close to five billion dollars in the Arctic during the 2000s, yet Arctic production represented only 15% of their total oil output and 6% of gas output in 2012 [29]. This finding established an important baseline: despite significant resource endowments, commercial development faced substantial barriers that shaped strategic corporate behavior. At the intersection of energy trade modeling and Arctic route analysis, Meza et al. (2023) employed agent-based modeling to simulate global LNG trade under Arctic route scenarios, demonstrating that the North-East Passage primarily benefits Russian LNG exports to Asian markets while remaining marginal for other suppliers [26]. 2.4.2 Russian Energy Strategy and European Dependencies The role of Russian energy exports in shaping geopolitical relationships has received considerable scholarly attention. Sander’s scenario analysis of Russian natural gas in Europe concluded that absent very drastic policy interventions, Russian natural gas will continue to play a prominent role in the EU [30]. This finding proved prescient, as subsequent events demonstrated the profound dependencies that had developed. Konoplyanik’s earlier work on Russian gas adaptation highlighted the structural challenges facing Gazprom’s long-term gas export contracts within a tightening European market niche, noting that risks and uncertainties for oil-indexed contracts increased substantially under the EU’s Third Energy Package [31]. The literature on EU-Russia energy relations took a decisive turn following the 2022 invasion of Ukraine. Czyżak et al. examined Europe’s energy security transformation from Russian dependence toward renewable reliance, finding that gas price differences would remain prominent until 2026 due to tight global markets, but that beyond 2026, LNG expansion and Europe’s decarbonization commitments would lessen the price impact, making Russian gas economically irrelevant [28]. Their work represents a significant methodological advance, integrating multiple scenario pathways with quantitative projections, an approach we extend to Arctic-specific dynamics. 2.4.3 Sanctions, Energy Trade, and Geopolitical Realignment The impact of economic sanctions on global energy trade patterns has emerged as a critical research frontier. Zheng et al. employed complex network theory to analyze how sanctions on Russia reshaped global fossil energy trade, finding that sanctions promoted the transfer of energy control centers from Europe to Asia [32]. Their analysis demonstrated that to mitigate sanction impacts, Russia would shift its fossil-energy export focus to China and other Asian countries, a prediction that our empirical data on Arctic shipping This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 4 of 22 -- patterns strongly confirms. Their finding that crude oil trade has the largest potential impact, while natural gas trade experiences the largest direct loss, provides important context for understanding Sino-Russian Arctic LNG cooperation. Bakhsh et al. examined the broader relationship between geopolitics, economic complexity, and sustainable energy transition, arguing that geopolitical factors significantly influence national energy strategy choices [7]. Their work highlights how environmental governance and economic complexity mediate the relationship between geopolitical pressures and energy transition pathways, dynamics highly relevant to understanding why Western Arctic policies have diverged so substantially from Eastern approaches. 2.4.4 The Global Energy Transformation Context The Arctic energy competition unfolds against a backdrop of accelerating global energy transformation. Gielen et al.s influential analysis demonstrated that renewable energy can supply two-thirds of total global energy demand by 2050, contributing to the bulk of greenhouse gas emissions reduction needed to limit warming below 2°C [9]. Their finding that a six-fold acceleration of renewables growth is needed, with highest growth estimated for wind and solar PV, has profound implications for Arctic hydrocarbon development. If global energy demand increasingly shifts toward renewables, the strategic window for Arctic fossil fuel development narrows considerably, intensifying the urgency of current infrastructure investments and explaining the accelerated Sino-Russian partnership formation we document. 2.4.5 Research Gaps Addressed by This Study Despite this rich literature, several critical gaps remain that our study addresses. First, existing Arctic energy research has largely remained descriptive, cataloging capabilities and resources without projecting quantitative future outcomes. While scenario studies exist for European gas markets and global energy transitions, no comparable integrated scenario analysis has been conducted specifically for Arctic strategic positioning that combines geostrategic theory with quantitative machine learning methods. Second, the literature has treated geopolitical factors as exogenous constraints rather than endogenous variables within energy strategy models. Our approach explicitly operationalizes classical geostrategic theories, Mackinder’s Heartland, Spykman’s Rimland, and Mahan’s Sea Power doctrines, into quantifiable metrics that can be systematically analyzed and projected. Third, prior research has examined bilateral relationships (Russia-EU, Russia-China, US-Russia) in isolation rather than as an interconnected system of competitive dynamics. Our multi-scenario framework captures how actions by one actor influence the strategic position of others, revealing compound effects that bilateral analyses miss. Fourth, the paradoxical effects of Western sanctions on Arctic development remain underexplored. While Zheng et al. documented trade pattern shifts, no study has quantified how sanctions accelerated Sino-Russian partnership formation specifically in the Arctic context, creating the strategic axis that now dominates regional energy infrastructure [32]. By synthesizing geostrategic theory with quantitative scenario modeling and machine learning validation, this paper provides a methodological contribution applicable to energy strategy research beyond the Arctic context, while generating policy-relevant projections for one of the world’s most consequential emerging energy frontiers. This study also contributes to the achievement of Sustainable Development Goal 7 (Affordable and Clean Energy) by examining how Arctic energy developments will shape global LNG supply diversification, energy pricing dynamics, and the strategic calculus of energy-importing nations seeking to balance security considerations against decarbonization imperatives. This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 5 of 22 -- 3. Methodology 3.1 Methodological framework The rapidly evolving Arctic landscape presents methodological challenges that traditional analysis alone cannot address. This study employs a hybrid methodology synthesizing qualitative geopolitical analysis with machine learning techniques, projecting Arctic power dynamics from 2025 through 2035. Our approach operationalizes classical theories, Mackinder’s Heartland Theory, Spykman’s Rimland Theory, and Mahan’s Sea Power doctrine [33-37], into quantifiable metrics suitable for computational analysis while maintaining theoretical rigor. The methodology unfolds through five interconnected phases. First, we establish comprehensive baseline Arctic capabilities through systematic documentation of seventy-eight distinct variables, ensuring our analysis rests on solid empirical ground. Second, Multi-Criteria Decision Analysis (MCDA) [38-40] synthesizes these diverse metrics into strategic position assessments for four competing scenarios. Third, theory-guided feature engineering [41] transforms raw data into 47 meaningful features capturing dynamic geopolitical relationships. Fourth, an ensemble of six machine learning algorithms [42-45] trained on 5,000 synthetic samples explores theoretical parameter spaces beyond historical precedent. Finally, Monte Carlo simulation [46, 47] with 10,000 iterations quantifies uncertainty, acknowledging the inherent unknowns in geopolitical projection. 3.2 Data collection and baseline assessment 3.2.1 Data sources and triangulation Establishing an accurate baseline of current Arctic capabilities forms the foundation for meaningful projection of future scenarios [48]. We conducted a systematic examination of seventy-eight distinct geopolitical, economic, and infrastructural variables, each verified through multiple independent sources. Primary data derived from Arctic Council assessments, national strategy documents, the Arctic Ship Traffic Data System, International Energy Agency statistics, defense capability evaluations, and infrastructure investment records from national budgets and corporate reports. To ensure reliability, we implemented rigorous triangulation requiring each quantitative value to be confirmed by at least three independent sources [49]. Where discrepancies arose, we adopted median values while documenting uncertainty ranges for sensitivity analysis. For instance, estimates of Russia’s operational icebreaker fleet ranged from 37 to 46 vessels; we conservatively adopted forty fully operational ocean-going icebreakers. This triangulation achieved Cohen’s kappa [50] exceeding 0.85, indicating substantial inter-source agreement despite the challenging information environment. 3.2.2 Variable operationalization Transforming qualitative concepts into quantifiable metrics requires careful operationalization [51, 52]. A nuclear-powered icebreaker capable of year-round operation represents fundamentally different capability than a seasonal diesel vessel [53]. Our framework addresses these nuances through weighted indices reflecting actual operational capability. Icebreaker Capacity (IC) incorporates vessel numbers, capability classes, and operational readiness: IC = Σᵢ (nᵢ × wᵢ × oᵢ) where nᵢ represents vessels in class i, wᵢ denotes capability weight derived from operational performance data showing nuclear vessels operate three times more days annually than conventional vessels, Arc7-rated vessels operate twice as many days, yielding weights of 3.0, 2.0, and 1.0, respectively. The operational readiness factor oᵢ scales from 0 to 1 based on maintenance records and deployment data. Strategic Investment Intensity (SII) measures financial commitment relative to geographic responsibility: This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 6 of 22 -- SII = (I_committed / C_coastline) / max(I_committed / C_coastline) Here I_committed represents formally committed investment in billion USD from signed contracts and budget allocations, while C_coastline denotes Arctic coastline in thousand kilometers. The normalization factor ensures comparability across nations with different geographic scales, derived from the maximum observed investment-to-coastline ratio in our dataset. Operational Experience (OE) quantifies accumulated Arctic knowledge with temporal decay [54-56]: OE = Σₜ (yₜ × sₜ × e^(-λt)) where yₜ represents operational years, sₜ distinguishes year-round from seasonal operations based on ice navigation records, and the decay factor incorporates organizational learning theory showing knowledge half-life of approximately 16 years without practice, yielding λ=0.043. This formulation captures how Russia's 46-year continuous Northern Sea Route operation provides enduring advantage. 3.3 Multi-Criteria Decision Analysis 3.3.1 Criteria development and weighting Comparing diverse Arctic capabilities requires a structured synthesis of incommensurable factors [38, 39]. Rather than arbitrarily selecting criteria, we derived them through systematic analysis of 127 Arctic strategy documents from eight nations spanning 2010-2025. Using natural language processing to identify recurring strategic themes, we found four dominant clusters that nations consistently emphasize in their Arctic policies. The frequency of theme occurrence across documents, weighted by recency and source authority (official strategy documents weighted higher than policy papers), determined the relative importance of each criterion. To ensure robustness, we compared our document-derived weights against three alternative derivation methods. Principal component analysis [41] of the 78 baseline variables yielded similar factor loadings, with the first four components explaining 81% of variance and aligning with our identified criteria. Machine learning feature importance from a preliminary random forest model [42, 57] trained on historical Arctic development patterns corroborated the weight hierarchy. Expert survey results from 12 Arctic specialists, while not used for primary weight derivation, provided independent validation with a correlation coefficient of 0.91 between document-derived and expert-assessed weights [58]. 3.3.2 Strategic Position Index calculation Each scenario’s Strategic Position Index synthesizes multidimensional assessment through weighted linear aggregation [38, 40]: SPIᵢ = Σⱼ (wⱼ × Pᵢⱼ) where wⱼ represents the document-derived weight for criterion j and Pᵢⱼ denotes scenario i's normalized performance on criterion j. Component scores integrate multiple metrics using sub-weights derived from the same document analysis process: P_capability = 0.40×Infrastructure + 0.35×Icebreakers + 0.25×Experience P_momentum = 0.50×Investment_rate + 0.30×Project_pipeline + 0.20×Technology_advancement P_feasibility = (Governance_quality × Implementation_timeline)^0.5 P_synergy = (Resource_complementarity × (1 - Conflict_potential))^0.5 The multiplicative structure in feasibility and synergy scores reflects theoretical understanding that these factors interact non-linearly [59, 60]. Poor governance cannot be compensated by good timelines, and high conflict undermines any level of resource complementarity. These functional forms derive from game This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 7 of 22 -- theory and organizational behavior literature on coalition dynamics. 3.3.3 Sensitivity analysis To test robustness of our document-derived weights, we conducted a comprehensive sensitivity analysis [61], varying weights within confidence intervals determined by document analysis uncertainty. Monte Carlo simulation [46] with 10,000 iterations sampled weights from distributions centered on derived values with standard deviations reflecting measurement uncertainty from document interpretation. Scenario rankings remained stable in 94% of iterations, validating that our strategic assessments reflect robust underlying patterns rather than specific weight choices. 3.4 Feature engineering and synthetic data generation We constructed forty-seven features through systematic application of six theoretical frameworks: infrastructure lock-in effects from economic geography theory [59, 62] (eight features capturing cumulative investment stocks and utilization ratios); experience premiums from organizational learning theory [54, 56] (six features quantifying accumulated operational wisdom through weighted years and efficiency curves); network effects modeling infrastructure value scaling through port connectivity and Metcalfe's law applications (seven features); temporal dynamics incorporating differential decay rates for infrastructure longevity and skill degradation (nine features); geopolitical interactions operationalizing game-theoretic concepts including strategic substitutability and sanction exposure (ten features); and climate-infrastructure coupling linking ice retreat rates to route viability (seven features). To address the absence of historical precedent for the Arctic transformation, we generated 5,000 synthetic training samples through structural causal simulation, where X ~ f(B, Θ, ε), with B representing empirically grounded 2025 baselines, Θ denoting theoretically bounded parameter distributions, and ε introducing realistic stochastic variation [64]. Sample size was determined through convergence analysis, with scenario probability estimates stabilizing at a coefficient of variation below 0.05. Complete feature descriptions and generation methodology are provided in Supplementary Material S1. 3.5 Machine Learning ensemble architecture We selected six complementary machine learning algorithms based on theoretical properties and empirical performance: three gradient boosting variants (XGBoost [43], LightGBM [44], CatBoost [45]), random forests [42], extra trees [70], and a multilayer perceptron neural network [71]. Each captures different aspects of the prediction task through distinct mathematical approaches, from sequential error correction to bootstrap aggregation to layered non-linear transformations. Predictions were combined using inverse error weighting, where models with lower prediction errors receive proportionally higher influence. Validation employed nested cross-validation [74] with an outer five-fold split for unbiased performance estimation and inner three-fold validation for hyperparameter selection. Feature importance analysis confirmed that infrastructure variables show highest predictive power, aligning with theoretical emphasis on material capabilities. Complete model specifications, hyperparameter configurations, and validation details are provided in Supplementary Material S2. 3.6 Uncertainty Quantification and Robustness Analysis 3.6.1 Monte Carlo simulation Geopolitical projections inherently involve deep uncertainty from incomplete knowledge about system dynamics, random variations in human decision-making, and necessary model simplifications [64]. Rather than providing false precision, we employ Monte Carlo simulation [46, 47] with 10,000 iterations to generate probability distributions honestly representing plausible future ranges. This approach explores the This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 8 of 22 -- full space of parameter uncertainty through systematic sampling: P(scenario) = ∫ P(scenario|θ,M) × P(θ,M|D) dθdM where scenario probabilities integrate over parameter and model uncertainty weighted by posterior probabilities given available data [77]. This Bayesian formulation ensures our projections reflect fundamental uncertainty about Arctic system dynamics rather than just sampling variation. 3.6.2 Confidence Intervals and Sensitivity Confidence intervals employ bias-corrected accelerated bootstrap [69] accounting for distribution asymmetry common in bounded probability estimates. Global sensitivity analysis through Sobol indices [61] decomposes output variance into first-order effects from individual parameters and higher-order interactions, revealing that infrastructure variables contribute most to prediction uncertainty while providing targets for strategic focus or intelligence gathering to reduce projection uncertainty. 3.7 Software implementation and reproducibility All analyses employed Python 3.11.5 with established scientific libraries, ensuring reproducibility. Scikit- learn [41] provided random forests and neural networks. XGBoost [43], LightGBM [44], and CatBoost [45] implemented gradient boosting variants. NumPy [78] and SciPy handled numerical computations. Pandas facilitated data manipulation. Complete code, processed datasets, and documentation are available at https://github.com/SAAAHco/scenario-analysis-framework, enabling verification, extension, and adaptation of our methodology [49]. 3.8 Limitations and Assumptions Our methodology assumes continuity in fundamental geopolitical structures absent major discontinuities like military conflict or revolutionary technology. Linear aggregation in MCDA [38] may miss threshold effects requiring minimum capabilities across dimensions. Document analysis for weight derivation assumes stated priorities reflect actual strategic intent. Data limitations force estimation for classified variables. Model projections beyond training distributions require careful interpretation as uncertainty increases with departure from current conditions [38]. 3.9 Ethical Considerations This research maintains ethical boundaries by excluding operational military details while providing strategic insight for informed policy [48]. Analysis received institutional review board exemption confirming no human subjects’ involvement. While acknowledging Arctic Indigenous peoples’ critical stakes, our state-centric analysis does not claim to represent their perspectives, highlighting the need for participatory research approaches in future work. 4. Results: Four Futures for the Arctic The following analysis presents four plausible scenarios for the Arctic, grounded in theoretical frameworks and quantified through machine learning analysis. The strategic position indices presented for each scenario represent the 2025 baseline distribution calculated using Multi-Criteria Decision Analysis (MCDA) methodology: Russian Arctic Consolidation (37.6%), Sino-Russian Partnership (44.2%), Western Coalition (5.3%), and Fragmented Competition (12.8%). These indices weight infrastructure capability (46.7%), strategic momentum (27.7%), operational feasibility (16.0%), and partnership synergy (9.6%) based on current deployed assets and verified investments (Figure 1). Machine learning projections to 2035, developed through ensemble methods incorporating forty-seven engineered features from empirical Arctic data, reveal how these baseline positions evolve under current trajectories. This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 9 of 22 -- Figure 1: Strategic Position Index Distribution (2025 Baseline). Bar graph displaying the strategic position index percentages for four Arctic scenarios (Sino-Russian Partnership, Russian Arctic Consolidation, Western Coalition, and Fragmented Competition) in 2025 and 20235. 4.1. Comparative Performance Analysis Across Strategic Metrics The heatmap visualization (Figure 2) reveals a profound bifurcation in Arctic capabilities. Eastern scenarios demonstrate near-complete dominance across all six strategic dimensions, with the Sino-Russian Partnership scoring 0.85-0.98 across Infrastructure, Experience, Investment, Geography, Resilience, and Partnership Synergy. In stark contrast, the Western Coalition exhibits catastrophic deficiencies in Infrastructure (0.15) and Investment (0.10), the two lowest scores across all despite a collective GDP exceeding $50 trillion. The compound nature of these advantages creates self-reinforcing feedback loops: Eastern capabilities accumulate synergistically while Western deficiencies deepen absent intervention. The detailed scenario-by-scenario analysis is provided in Supplementary Material S3. This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 10 of 22 -- Figure 2. Comparative Scenario Performance Across Strategic Metrics. Heatmap of normalized performance scores (0-1 scale) for four Arctic scenarios across six strategic dimensions. Color intensity represents capability strength from critical deficiency (red) to strategic dominance (green). 4.2. Scenario A: Russian Arctic Consolidation This scenario, which holds a 37.6% strategic position index based on current infrastructure and operational capabilities (calculated using MCDA methodology detailed in Section 4), sees Russia leveraging its unparalleled advantages to achieve regional dominance. The nation controls approximately 24,140 kilometers of Arctic coastline, roughly 53% of the total Arctic littoral [79]. The cornerstone of this dominance is its icebreaker fleet, which consists of approximately forty vessels, including ten nuclear- powered icebreakers, the largest such fleet in the world. Russia’s newest Project 22220-class icebreakers generate 60 megawatts of propulsion power, enabling year-round Arctic navigation that no other nation can currently match [80, 81]. The economic foundation for this consolidation is built on its vast hydrocarbon reserves, with Arctic territories producing 83% of Russia’s natural gas. These resources feed into the NSR, where transport costs are $2.2 per MMBtu, creating a 21% cost advantage over traditional European routes [25]. This economic logic drives massive infrastructure investment, as evidenced by the ~$5 billion invested in the port of Murmansk since 2004, with capacity expanding by 82% to 52 million tonnes. Russia’s deliberate, long-term strategy, demonstrated by its continuous operational experience on the NSR since the winter of 1978-79 [82], has created a non-transferable asset that reinforces its first-mover advantage. The completion of three new Project 22220 nuclear icebreakers between 2021 and 2024, with seven more planned, demonstrates a committed investment that is difficult for other nations to match. 4.3. Scenario B: Sino-Russian Strategic Partnership Building on Scenario A, this outcome posits that the relationship between Russia and China evolves into a deep, symbiotic partnership, with a 44.2% strategic position index representing the highest current strategic advantage based on combined deployed capabilities and verified investments (MCDA calculation, Section This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 11 of 22 -- 4). China provides the critical financial and technological investment, while Russia provides the physical route and operational expertise. The partnership is a clear demonstration of complementary capabilities. China’s financial flows have already transformed Russian Arctic capabilities, with the China National Petroleum Corporation and Silk Road Fund taking a 20% stake in Yamal LNG, while Chinese shipyards have constructed 15 specialized Arc7 LNG carriers at a total cost of $5 billion [10, 25, 83]. For China, which is dependent on imports for 43% of its natural gas consumption, Russian Arctic supplies via secure routes represent energy security worth hundreds of billions in reduced strategic vulnerability [84, 85]. The surge in Chinese crude oil and gas tanker traffic in the Arctic in 2024, representing increases of 77% and 66% respectively from 2023, demonstrates the partnership’s resilience despite 22,230 sanctions levied against Russia since 2022 [11, 25, 86]. 4.4. Scenario C: Western Arctic Coalition This scenario, assessed at a 5.3% strategic position index reflecting the lowest current strategic position due to capability gaps (MCDA calculation, Section 4), envisions the Western powers successfully mobilizing their latent capabilities to challenge Sino-Russian dominance. The coalition’s aggregate power is immense, with a collective GDP exceeding $50 trillion versus Russia’s $2 trillion [11]. The European Union contributes to this economic power but demonstrates the coordination challenges inherent in the Western approach. Despite investing ~$428 million in Arctic research and ~$315 million through Interreg cooperation projects, the EU’s engagement remains fragmented across multiple directorates, lacking the unified strategic direction of Russian or Chinese Arctic programs [87]. The recent accession of Finland and Sweden to NATO means seven of the eight Arctic Council members now belong to the military alliance [17]. Initiatives like the ICE Pact, which aims to produce 70 to 90 icebreakers, signal a growing strategic awareness [88]. The US Arctic Alaska Petroleum Province holds mean estimates of undiscovered, technically recoverable oil and gas resources of nearly 30 billion barrels of oil, about 179 trillion cubic feet of no associated gas, and 40 trillion cubic feet of associated gas [89, 90]. However, the data reveals significant “coordination challenges” and “infrastructure gaps” that undermine this scenario’s plausibility. The US Polar Security Cutter program, for example, is delayed until 2029 and is 60% over budget according to Congressional Budget Office assessments. Canada's only Arctic deepwater port with rail access, Churchill, handles limited tonnage compared to Murmansk's approximately 52 million tonnes annual capacity [91]. The total investment required for the West to achieve parity is estimated at $70-$90 billion, a figure that is currently uncommitted and exceeds most nations’ entire annual defense budgets [92]. While the European Investment Bank has provided ~$1.8 billion for Arctic renewable energy projects since 1994, this represents less than 40% of Russia's investment in the single port of Murmansk (~$5 billion), illustrating the scale disparity [87]. 4.5. Scenario D: Fragmented Competition With a 12.8% strategic position index representing a governance vacuum scenario (MCDA calculation, Section 4), this is the default outcome defined by institutional failure. The suspension of the Arctic Council in 2022 following Russia’s invasion of Ukraine created a void in multilateral cooperation [93, 94]. While the Council resumed limited work in July 2022, it was on “projects that do not involve the participation of the Russian Federation” [95]. The human and environmental consequences of this scenario are significant. Indigenous communities, such as the 73 Alaska Native villages at risk from climate change and requiring costly relocations, face a fragmented and inadequate response from multiple jurisdictions [17]. Environmentally, the increase in vessel traffic generates pollution, with ships in the Norwegian and Barents seas producing approximately 13,000 metric tons of oily sludge annually [82]. The absence of unified search and rescue capabilities means vessels operating beyond coastal areas face response times measured in days, not hours. Furthermore, Arctic oil extraction costs averaging $75 per barrel for Russian offshore projects, compared to approximately $43-50 for conventional offshore deepwater production, mean projects require sustained high prices [96, 97]. This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 12 of 22 -- 4.6. Quantitative Validation Through Machine Learning Analysis The four scenarios presented above, while grounded in empirical data and theoretical frameworks, represent static snapshots of current strategic positions. To project these dynamics forward and validate the trends of Arctic power distribution, a comprehensive machine learning analysis was conducted using ensemble methods that incorporated forty-seven engineered features derived from the infrastructure, investment, and operational metrics detailed throughout this analysis. The model ensemble, combining gradient boosting algorithms with neural network architectures trained on 5,000 synthetic samples, achieved a mean squared error of 0.0034, demonstrating robust predictive reliability.. Projections reveal a striking consolidation of Arctic power toward the Eastern bloc by 2035, with the Sino- Russian Partnership scenario ascending from 44.2% to 67.8%, driven by compounding first-mover advantages and infrastructure lock-in effects (Table 1). The Russian Consolidation scenario declines from 37.6% to 27.3%, reflecting the economic constraint of maintaining Arctic dominance without sustained Chinese partnership. Western-aligned scenarios experience near-total collapse: Fragmented Competition contracts to 2.0% and Western Coalition to 2.9%, with narrow confidence intervals indicating high model certainty. Monte Carlo uncertainty analysis (10,000 iterations) confirms these projections, with combined Eastern Arctic control maintaining a 95% confidence interval ranging from 79.4% to 110.6% (Table 2, Figure 3). The temporal decay functions prove particularly illuminating: Western capabilities deteriorate at 14% annually without active deployment, while Russian operational expertise depreciates at only 4.3% due to continuous utilization since 1978-79. The detailed interpretive analysis of these projections is provided in Supplementary Material S4. Table 1: Projected Strategic Position Index Changes (2025-2035). Comparative data showing strategic position index values for four Arctic scenarios in 2025 and 2035, with calculated percentage point changes and directional trend indicators. Scenario 2025 2035 Change Trend Sino-Russian Partnership 44.2% 67.8% +23.6% Russian Consolidation 37.6% 27.3% -10.3% Fragmented Competition 12.8% 2.0% -10.8% Western Coalition 5.3% 2.9% -2.4% Monte Carlo uncertainty analysis comprising 10,000 iterations, varying key parameters within their empirically observed ranges, confirms the robustness of these projections (Figure 2). Russian icebreaker fleet was varied between 40 and 57 vessels [107], Chinese Arctic investment between $12 and $90 billion [24, 108], Western commitment between $3 and $27 billion [109, 110], sanctions effectiveness following a beta distribution (α=2, β=5), and coordination friction between 0.8 and 1.5. This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 13 of 22 -- Figure 3. Violin plots with embedded box plots illustrate probability distributions for four Arctic geopolitical scenarios derived from 10,000 Monte Carlo simulations with systematically varied input parameters. The violin width represents probability density at each value, while internal box plots show quartiles and medians. Vertical whiskers extending from each box indicate the data range within 1.5 times the interquartile range. Table 2: Statistical Uncertainty Analysis of 2035 Projections. Statistical measures from Monte Carlo uncertainty analysis (n=10,000 iterations) showing mean projections, standard deviations, and 95% confidence interval bounds for four Arctic scenarios in 2035. Scenario Mean Std Dev CI Lower CI Upper Sino-Russian Partnership 66.2% 4.1% 60.3% 75.7% Russian Consolidation 28.8% 4.2% 19.1% 34.9% Fragmented Competition 1.9% 0.1% 1.7% 2.2% Western Coalition 3.1% 0.1% 2.8% 3.4% 4.7. Strategic Threshold Analysis To address the threshold effects acknowledged as a limitation of linear MCDA aggregation (Section 3.8), the sensitivity framework established in Sections 3.3.3 and 3.6.2 was extended to identify minimum Western commitment levels required to materially alter scenario probabilities. A two-dimensional This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 14 of 22 -- parameter sweep (984 investment × icebreaker combinations), single-variable sensitivity of eight parameters, and Monte Carlo threshold identification (n = 10,000 iterations, 2,050,000 total predictions) were conducted under stochastic background conditions (Supplementary Material S5). Consistent with the Sobol decomposition finding that infrastructure variables dominate prediction uncertainty (Section 3.6.2), icebreaker fleet size (3.04 pp) and committed funds (2.95 pp) exhibit the largest marginal effects, while aggregate investment shows minimal impact (0.26 pp). The analysis reveals a structural ceiling: even the most favorable combination tested ($80 billion, 20 icebreakers) yields only 7.3% Western Coalition probability with a 10.0:1 Eastern advantage. Monte Carlo simulation identifies a median requirement of $42 billion and 10 icebreakers to reach the 5% threshold, with the 10% threshold unachievable under any tested combination. 5. Discussion: The convergence of theory and practice in a contested Arctic 5.1. The New Geostrategic Axis: Heartland and sea power converged The most profound finding of this analysis is the emergence of a new geostrategic axis formed by the Sino- Russian partnership. With a combined strategic position index of 81.8% (44.2% + 37.6%), the Eastern Arctic advantage represents a 4.5:1 ratio over Western/fragmented positions (Section 4.5). This partnership represents a dynamic convergence of Mackinder’s Heartland (Russia’s landmass, vast resources, and operational control of the NSR) and Mahan’s Sea Power (China’s maritime ambitions, financial capital, and technological investment). This unified front is proving to be a highly effective and sanctions-resistant force in the Arctic. The West’s Rimland-based, coalition-focused approach struggles to counter this unified strategy because its strength is fragmented, and its actions are often reactive. The disparate and often conflicting interests of the Western allies make it difficult to translate their collective economic and military power into a coherent, proactive Arctic strategy. The result is a widening gap in committed capabilities, favoring the more unified, albeit ideologically distinct, actors in the region. 5.2. The paradox of Western sanctions and the first-mover advantage The imposition of Western sanctions on Russian energy following the 2022 invasion of Ukraine had minimal long-term impact on its Arctic ambitions, instead catalyzing a strategic pivot with unforeseen consequences. With energy exports comprising approximately 20% of Russia’s GDP, sanctions initially caused a temporary halt in Russian energy exports to Europe [11, 12]. Russia’s response was not to cease its operations but to redirect its energy exports from Europe to Asia, finding willing buyers in China and India [113, 114]. This redirection created an immediate and accelerated need for new, secure routes and a sanctions-resistant financial and technological ecosystem. China, with its capital and technology, was well- positioned to fill the void left by Western companies [24]. As a result, the sanctions, rather than isolating Russia and halting its Arctic projects, have paradoxically accelerated the very Sino-Russian partnership that is creating a powerful, alternative energy and trade system. Global LNG trade simulations of a European ban scenario confirm this dynamic, showing that Russian Arctic LNG exports increase from 38.4 to 50 MTPA when the North-East Passage is available, effectively converting the Arctic corridor into a sanctions- resilience mechanism [26]. 5.3. The critical window of opportunity: Infrastructure lock-in The window for the West to fundamentally alter the Arctic’s strategic trajectory is rapidly closing. The analysis identifies the “2025–2027 window” as a period of maximum strategic flexibility before irreversible infrastructure investments create path dependencies. The urgency of this window is established by converging infrastructure timelines: Russia is set to complete its Project 22220 fleet of nuclear icebreakers, China will have significantly expanded its fleet of Arc7 LNG carriers, and the first US Polar Security Cutter is delayed until at least 2028 at 60% over budget. By the time Western assets begin entering service, the This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 15 of 22 -- Eastern bloc will have achieved operational lock-in across the Northern Sea Route corridor. The threshold analysis (Section 4.7) quantifies the scale of commitment required within this window. Monte Carlo simulation (n = 10,000 iterations) establishes that achieving even a 5% Western Coalition probability, the minimum level constituting marginal strategic relevance, requires a median commitment of approximately $42 billion and 10 operational icebreakers (IQR: $0–62 billion, 7–15 icebreakers), requirements that dwarf current appropriations ($0 committed, 2 icebreakers operational). Critically, the 10% probability threshold is not achievable under any tested parameter combination, establishing a structural ceiling on Western strategic repositioning. The model’s temporal decay functions illuminate why each year of delay is consequential: Western capabilities deteriorate at 14% annually without active deployment, while Russian operational expertise depreciates at only 4.3% due to continuous utilization since 1978–79. The sensitivity analysis further reveals that fleet size is a more consequential variable than aggregate investment (3.04 pp vs. 0.26 pp), suggesting that the icebreaker gap, rather than the investment gap, constitutes the binding constraint on Western Arctic strategy. This “lock-in” effect means that after this critical period, it will become exponentially more difficult and expensive for the West to catch up. The most plausible future is therefore not a sudden shift in power but a continuous reinforcement of existing advantages, making a blend of Scenarios B and D the most likely long-term outcome. 5.4. Implications for global energy strategy and SDG7 The findings of this analysis carry profound implications for global energy strategy that extend well beyond Arctic borders. The projected consolidation of Arctic energy infrastructure under Sino-Russian control by 2035 threatens to create a structurally bifurcated global energy system, where access to approximately 22% of the world’s undiscovered conventional hydrocarbon reserves and the most efficient LNG shipping routes is mediated through a single geopolitical bloc [26]. This outcome directly challenges the objectives of SDG7, which calls for ensuring access to affordable, reliable, sustainable, and modern energy for all [6]. When energy supply routes are concentrated under the control of politically aligned actors, the affordability and reliability dimensions of SDG7 become contingent upon geopolitical alignment rather than market mechanisms or development need. Sampedro et al.’s (2024) analysis of the European gas supply crisis demonstrated that Russian energy leverage created subregional disparities in energy costs across the EU, with Central and Eastern European countries bearing disproportionate burdens, a pattern that Arctic route consolidation could replicate on a global scale [8]. Furthermore, the Arctic energy competition unfolds against the backdrop of a global energy transformation in which natural gas is positioned as a critical transition fuel [9]. The control of Arctic LNG production and shipping routes therefore determines not only fossil fuel access but also the pace and equity of the clean energy transition itself. Nations dependent on Arctic-sourced LNG for their transition from coal to gas to renewables may find their decarbonization pathways constrained by geopolitical gatekeeping. The IRENA (2024) assessment of energy security in the context of the energy transition warns precisely against such concentration risks, arguing that the geopolitics of the renewable era demands diversified supply chains and multilateral governance frameworks, both of which the current Arctic trajectory undermines [115]. The implications extend to critical mineral access as well, as the Arctic holds significant deposits of rare earth elements essential for renewable energy technologies [90]. The same infrastructure lock-in dynamics that govern fossil fuel access in the Arctic will increasingly apply to the minerals required for batteries, wind turbines, and solar panels, creating a compounding energy strategy challenge that demands urgent Western policy recalibration. 6. Conclusion: A new geopolitical equilibrium The analysis concludes that the Arctic’s future is defined by a contested rather than a cooperative environment. The most probable outcome is a blend of Russian Consolidation and Fragmented Competition, where the Sino-Russian partnership dominates the Eastern Arctic and Northern Sea Route, while Western efforts remain fragmented and focused on the North American sphere. The evidence suggests that classical geostrategic theories are not outdated but are, in fact, powerful interpretive tools for This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 16 of 22 -- understanding contemporary state behavior. The confluence of Heartland, Rimland, and Sea Power dynamics explains the current competition. The key lesson is that the struggle for Arctic influence is not about potential but about deployed capabilities and committed investment. The data show that the nations that have invested billions and decades into the region are now reaping the rewards, leaving others with a difficult and costly choice. The transformation is not future speculation but present reality: 12.7 million nautical miles sailed, 21.2 million tons of Arctic LNG produced, ~ $5 billion invested in single ports, and 1,781 unique vessels operating in the Arctic Polar Code Area. The decisions made in the 2025-2027 window will determine whether the Arctic becomes a domain of managed competition or dangerous fragmentation, with consequences for global trade, energy security, and international order that will reverberate for generations. These findings yield direct implications for policymakers. Western governments must replace multi-decade procurement timelines with emergency funding mechanisms before the 2025-2027 window closes. NATO Arctic members should establish a unified coordination structure to convert latent economic power into deployed capability. The current sanctions policy requires recalibration, as broad economic restrictions have paradoxically accelerated the Sino-Russian partnership; targeted measures on Arctic-specific chokepoints may prove more effective. Energy security planners must incorporate the emerging Sino- Russian controlled Northern Sea Route into long-term supply diversification strategies. The Arctic Council’s paralysis demands alternative multilateral frameworks for environmental protection and indigenous community support that can function despite great power competition. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not- for-profit sectors. CRediT Author Statement Konstantinos Pappas: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Writing - Original Draft, Writing - Review & Editing, Supervision, Project administration, Funding acquisition. Zainab Ashkanani: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data Curation, Writing - Original Draft, Writing - Review & Editing, Visualization. Data Availability The data and code supporting this study are available in a public repository. The machine learning and Monte Carlo framework for scenario-based geopolitical analysis, including training data and analysis scripts, can be accessed at: https://github.com/SAAAHco/scenario-analysis-framework Declaration of Generative AI and AI-assisted Technologies in the Manuscript Preparation Process During the preparation of this work the author(s) used Grammarly in order to improve grammar, clarity, and readability of the text. After using this tool, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the published article. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. References 1. Reynolds, J. and K. Taylor. The frontier is the front line: On climate resilience for infrastructure and supplies in Canada's Arctic (Issue Brief). 2025 May 30, 2025]; Available from: This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 17 of 22 -- https://www.atlanticcouncil.org/in-depth-research-reports/issue-brief/the-frontier-is-the-front- line-on-climate-resilience-for-infrastructure-and-supplies-in-canadas-arctic/. 2. Arctic Climate Impact Assessment, Arctic Climate Impact Assessment (ACIA) scientific report. 2005, Arctic Council. 3. Mirza, M.N. and S. Ayub, Heartland, Rimland, and the Grand Chessboard: Deciphering the Great Power Politics in Central Asia. UWJSS, 2022. 5(1): p. 187–204. 4. Chowdhury, S.K. and A.H. Kafi, The Heartland theory of Sir Halford John Mackinder: justification of foreign policy of the United States and Russia in Central Asia. J. Liberty & Int'l Aff., 2015. 1: p. 58. 5. Arctic Council. Arctic states. n.d. [cited 2025 November 24, 2025]; Available from: https://arctic- council.org/about/states/. 6. IRENA, I., Tracking SDG 7: The Energy Progress Report 2024. 2024. 7. Bakhsh, S., et al., Strategy towards sustainable energy transition: The effect of environmental governance, economic complexity and geopolitics. Energy Strategy Reviews, 2024. 52: p. 101330. 8. Sampedro, J., et al., Energy system analysis of cutting off Russian gas supply to the European Union. Energy Strategy Reviews, 2024. 54(PNNL-ACT-SA-10794). 9. Gielen, D., et al., The role of renewable energy in the global energy transformation. Energy strategy reviews, 2019. 24: p. 38–50. 10. Šerić, M. Heartland geopolitical theory: An idea that dictates contemporary geopolitics – Analysis. 2023; Available from: https://www.eurasiareview.com/09042023-heartland-geopolitical-theory- an-idea-that-dictates-contemporary-geopolitics-analysis/. 11. Yermakov, V., Follow the Money: Understanding Russia’s oil and gas revenues. Oxford Institute for Energy Studies, March, 2024. 12. Fenton, N., Down But Not Out: The Russian Economy Under Western Sanctions. 2025: Center for Strategic and International Studies (CSIS). 13. Gatopoulos, A. What is behind Russia's interest in a warming Arctic? 2022; Available from: https://www.aljazeera.com/features/2022/3/28/what-is-behind-russias-interest-in-a-warming- arctic. 14. Rumer, E., R. Sokolsky, and P. Stronski, Russia in the Arctic: A critical examination. Vol. 29. 2021: Carnegie Endowment for International Peace Washington, DC. 15. Araldsen, O. The Soviet Union and the Arctic. in US Naval Institute. 1967. 16. Breum, M., The deep roots of Russia’s Arctic ambitions—and what they mean for future stability in the region. Retrieved from Arctic Today: https://www. arctictoday. com/the-deep-roots-of- russias-arctic-ambitions-and-what-they-mean-for-future-stability-in-the-region, 2019. 17. Elliot-Meisel, E., Reconsidering the Northwest Passage: Arctic sovereignty and continental security in U.S.-Canada relations. 2024, North American and Arctic Defence and Security Network. 18. Bykova, A. NATO has always been an Arctic Alliance (Part II). 2024 November 24, 2025]; Available from: https://www.thearcticinstitute.org/nato-arctic-alliance-part-ii/. 19. The White House, National Strategy for the Arctic Region. 2022, The White House. 20. U.S. Department of Defense, 2024 Department of Defense Arctic Strategy. 2024, U.S. Department of Defense. 21. Arctic Portal. Northwest Passage. n.d. November 24, 2025]; Available from: https://arcticportal.org/shipping-portlet/shipping-routes/northwest-passage. 22. Dou, E., A New Cold War? China Declares Itself a ‘Near-Arctic State’. The Wall Street Journal, 2018. 23. Tirziu, A.G. China's cultural conquest of the Arctic. 2025; Available from: https://www.gisreportsonline.com/r/china-arctic-culture/. 24. Edstrøm, A., G.R.T. Hauksdóttir, and P.W. Lackenbauer. Cutting through narratives on Chinese Arctic investments. 2025; Available from: https://www.belfercenter.org/research-analysis/china- arctic-investments. This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 18 of 22 -- 25. Yermakov, V., Arctic LNG 2: The litmus test for sanctions against Russian LNG. 2024, Oxford Institute for Energy Studies. 26. Meza, A., et al., Relevance and potential of the Arctic Sea Routes on the LNG trade. Energy Strategy Reviews, 2023. 50: p. 101174. 27. World Nuclear News. Russia and China discuss Northern Sea Route development. 2024; Available from: https://www.world-nuclear-news.org/articles/russia-and-china-discuss-northern-sea-route- development. 28. Ah-Voun, D., C.K. Chyong, and C. Li, Europe's energy security: From Russian dependence to renewable reliance. Energy Policy, 2024. 184: p. 113856. 29. Ermida, G., Strategic decisions of international oil companies: Arctic versus other regions. Energy Strategy Reviews, 2014. 2(3-4): p. 265–272. 30. Mitrova, T., T. Boersma, and A. Galkina, Some future scenarios of Russian natural gas in Europe. Energy strategy reviews, 2016. 11: p. 19–28. 31. Konoplyanik, A.A., Russian gas at European energy market: Why adaptation is inevitable. Energy Strategy Reviews, 2012. 1(1): p. 42–56. 32. Zheng, S., et al., Preliminary study on the global impact of sanctions on fossil energy trade: Based on complex network theory. Energy for Sustainable Development, 2022. 71: p. 517–531. 33. Mackinder, H.J., Democratic ideals and reality. 1962: Diane Publishing. 34. Mackinder, H.J., The geographical pivot of history, in The Structure of Political Geography. 2017, Routledge. p. 162–169. 35. Spykman, N.J., The geography of the peace. 1944: Harcourt, Brace. 36. Spykman, N.J., America's strategy in world politics: the United States and the balance of power. 2017: Routledge. 37. Mahan, A.T., The influence of sea power upon history, 1660-1783. 2011: Read Books Ltd. 38. Saaty, T.L., Decision making with the analytic hierarchy process. International journal of services sciences, 2008. 1(1): p. 83–98. 39. Saaty, T.L., Analytic hierarchy process, in Encyclopedia of operations research and management science. 2013, Springer. p. 52–64. 40. Hwang, C.-L. and K. Yoon, Multiple attribute decision making: methods and applications: a state- of-the-art survey. (No Title), 1981. 41. Hastie, T., R. Tibshirani, and J. Friedman, The elements of statistical learning. 2009, Springer series in statistics New-York. 42. Breiman, L., Random forests. Machine learning, 2001. 45(1): p. 5–32. 43. Chen, T. and C. Guestrin. Xgboost: A scalable tree boosting system. in Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining. 2016. 44. Ke, G., et al., Lightgbm: A highly efficient gradient boosting decision tree. Advances in neural information processing systems, 2017. 30. 45. Prokhorenkova, L., et al., CatBoost: unbiased boosting with categorical features. Advances in neural information processing systems, 2018. 31. 46. Metropolis, N. and S. Ulam, The monte carlo method. Journal of the American statistical association, 1949. 44(247): p. 335–341. 47. Robert, C.P., G. Casella, and G. Casella, Monte Carlo statistical methods. Vol. 2. 1999: Springer. 48. King, G., R.O. Keohane, and S. Verba, Designing social inquiry: Scientific inference in qualitative research. 2021: Princeton university press. 49. Freese, J. and D. Peterson, Replication in social science. Annual Review of Sociology, 2017. 43(1): p. 147–165. 50. Cohen, J., A coefficient of agreement for nominal scales. Educational and psychological measurement, 1960. 20(1): p. 37–46. 51. De Block, D. and B. Vis, Addressing the challenges related to transforming qualitative into quantitative data in qualitative comparative analysis. Journal of Mixed Methods Research, 2019. 13(4): p. 503–535. This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 19 of 22 -- 52. Sibul, G., P. Schütz, and K. Fagerholt, Arctic route planning under ice uncertainty: A risk-averse stochastic shortest path problem. Transportation Research Part E: Logistics and Transportation Review, 2026. 205: p. 104507. 53. Bukharin, O., Russia's nuclear icebreaker fleet. Science & Global Security, 2006. 14(1): p. 25–31. 54. Argote, L., Organizational learning: Creating, retaining and transferring knowledge. 2012: Springer Science & Business Media. 55. Ioffe, S. and C. Szegedy. Batch normalization: Accelerating deep network training by reducing internal covariate shift. in International conference on machine learning. 2015. pmlr. 56. Nonaka, I., The knowledge-creating company, in The economic impact of knowledge. 2009, Routledge. p. 175–187. 57. Fisher, A., C. Rudin, and F. Dominici, All models are wrong, but many are useful: Learning a variable's importance by studying an entire class of prediction models simultaneously. Journal of Machine Learning Research, 2019. 20(177): p. 1–81. 58. Linstone, H.A. and M. Turoff, The delphi method. Vol. 1975. 1975: Addison-Wesley Reading, MA. 59. North, D.C., Institutions, institutional change and economic performance. 1990: Cambridge university press. 60. Williamson, O.E., The economics of organization: The transaction cost approach. American journal of sociology, 1981. 87(3): p. 548–577. 61. Saltelli, A., et al., Global sensitivity analysis: the primer. 2008: John Wiley & Sons. 62. Arthur, W.B., Competing technologies, increasing returns, and lock-in by historical events. The economic journal, 1989. 99(394): p. 116–131. 63. George, A.L. and A. Bennett, Case studies and theory development in the social sciences. 2005: mit Press. 64. Der Kiureghian, A. and O. Ditlevsen, Aleatory or epistemic? Does it matter? Structural safety, 2009. 31(2): p. 105–112. 65. Geweke, J., Evaluating the accuracy of sampling-based approaches to the calculation of posterior moments. 1991, Federal Reserve Bank of Minneapolis. 66. Dietterich, T.G. Ensemble methods in machine learning. in International workshop on multiple classifier systems. 2000. Springer. 67. Zhou, Z.-H., Ensemble methods: foundations and algorithms. 2025: CRC press. 68. Giannakas, F., et al. Xgboost and deep neural network comparison: The case of teams’ performance. in International conference on intelligent tutoring systems. 2021. Springer. 69. Efron, B. and R.J. Tibshirani, An introduction to the bootstrap. 1994: Chapman and Hall/CRC. 70. Geurts, P., D. Ernst, and L. Wehenkel, Extremely randomized trees. Machine learning, 2006. 63(1): p. 3–42. 71. Nair, V. and G.E. Hinton. Rectified linear units improve restricted boltzmann machines. in Proceedings of the 27th international conference on machine learning (ICML-10). 2010. 72. Abadi, M., et al. {TensorFlow}: a system for {Large-Scale} machine learning. in 12th USENIX symposium on operating systems design and implementation (OSDI 16). 2016. 73. Bergstra, J. and Y. Bengio, Random search for hyper-parameter optimization. The journal of machine learning research, 2012. 13(1): p. 281–305. 74. Kohavi, R. A study of cross-validation and bootstrap for accuracy estimation and model selection. in Ijcai. 1995. Montreal, Canada. 75. Caruana, R., et al. Ensemble selection from libraries of models. in Proceedings of the twenty-first international conference on Machine learning. 2004. 76. Gneiting, T. and A.E. Raftery, Strictly proper scoring rules, prediction, and estimation. Journal of the American statistical Association, 2007. 102(477): p. 359–378. 77. Kass, R.E., et al., Markov chain Monte Carlo in practice: a roundtable discussion. The American Statistician, 1998. 52(2): p. 93–100. 78. Kotz, S., N. Balakrishnan, and N.L. Johnson, Continuous multivariate distributions, Volume 1: This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 20 of 22 -- Models and applications. Vol. 1. 2019: John wiley & sons. 79. Arctic Council. The Russian Federation. n.d. November 24, 2025]; Available from: https://arctic- council.org/about/states/russian-federation/. 80. Yu, L., J. Hao, and W. Chen, Nuclear Power Technology and Its Applications, in Nuclear Technology and Safety. 2025, Springer. p. 111–140. 81. Pincus, R. The icebreaker gap doesn't mean America is losing in the Arctic. 2019 November 24, 2025]; Available from: https://warontherocks.com/2019/11/the-icebreaker-gap-doesnt-mean- america-is-losing-in-the-arctic/. 82. Arctic Council, Arctic marine shipping assessment 2009 report (2nd printing). 2009, Protection of the Arctic Marine Environment (PAME) working group. 83. Vakhrusheva, K., The status and future projections of LNG production and use in the Russian Arctic. 2024, Bellona Foundation. 84. The State Council of the People's Republic of China. China's natural gas consumption up 7.6 pct in 2023. 2024; Available from: https://english.www.gov.cn/archive/statistics/202402/11/content_WS65c88aacc6d0868f4e8e3f68. html. 85. U.S. Energy Information Administration. The United States remained the world's largest liquefied natural gas exporter in 2024. 2025; Available from: https://www.eia.gov/todayinenergy/detail.php?id=64844. 86. Protection of the Arctic Marine Environment, The increase in Arctic shipping: 2013-2024. 2024, Protection of the Arctic Marine Environment. 87. European External Action Service, The EU in the Arctic. n.d. 88. Peck, M. The Arctic great game won't be won in U.S. shipyards. 2024 [cited November 24, 2025; Available from: https://foreignpolicy.com/2024/09/03/arctic-nato-russia-us-china-icebreakers/. 89. Houseknecht, D.W., K.J. Bird, and C.P. Garrity, Assessment of undiscovered petroleum resources of the Arctic Alaska Petroleum Province. 2012, U.S. Geological Survey. 90. Christiansen, F.G. Can geopolitics unlock Greenland's rare earths? 2025 November 24, 2025]; Available from: https://spectrum.ieee.org/greenland-rare-earth-minerals. 91. Congressional Budget Office, The cost of the Coast Guard's Polar Security Cutter. 2024, Congressional Budget Office. 92. Peter G. Peterson Foundation. The United States spends more on defense than the next 9 countries combined. 2024; Available from: https://www.pgpf.org/article/the-united-states-spends-more-on- defense-than-the-next-9-countries-combined/. 93. State, U.D.o., Joint Statement on Arctic Council Cooperation Following Russia’s Invasion of Ukraine. Office of the Spokesperson, Press Releases, 2022. 94. Simpson, B., The rise and sudden fall of the Arctic Council. Foreign Policy, 2023. 31. 95. Cunningham, A. Shifting Ice: How the Russian invasion of Ukraine has changed Arctic Circle governance and the Arctic Council's path forward. 2024 November 24, 2025]; Available from: https://www.thearcticinstitute.org/shifting-ice-russian-invasion-ukraine-arctic-circle-governance- arctic-councils-path-forward/. 96. Eurasian Ventures. The economy of breakeven oil prices. 2020 November 24, 2025]; Available from: https://eurasianventures.com/the-economy-of-breakeven-oil-prices/. 97. Offshore Energy. Offshore deepwater oil production one of cheapest sources of new supply as costs reach new low, Rystad says. 2020 November 24, 2025]; Available from: https://www.offshore- energy.biz/offshore-deepwater-oil-production-one-of-cheapest-sources-of-new-supply-as-costs- reach-new-low-rystad-says/. 98. Interesting Engineering. Yakutia: Russia unveils 8th nuclear icebreaker vessel to rule seas. 2025 November 24, 2025]; Available from: https://interestingengineering.com/transportation/russia- nuclear-powered-icebreaker-yakutia. 99. Śliwa, Z. and A. Sosnowski. China's Polar Silk Road: Long game or failed strategy? 2024 November 24, 2025]; Available from: https://www.thearcticinstitute.org/china-polar-silk-road- This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 21 of 22 -- long-game-failed-strategy/. 100. TASS. Traffic volumes along Northern Sea Route to exceed 100 mln tons by 2025 - ministry. 2022 November 24, 2025]; Available from: https://tass.com/economy/1540681. 101. Maritime Executive. Russia marks completion of fourth Project 22220 nuclear-powered icebreaker. 2024 November 24, 2025]; Available from: https://maritime- executive.com/article/russia-mark-s-completion-of-fourth-project-22220-nuclear-powered- icebreaker. 102. High North News. Russia to earn $160bn in taxes from Northern Sea Route by 2035, Arctic region accounts for 7.5 percent of GDP. 2025 November 24, 2025]; Available from: https://www.highnorthnews.com/en/russia-earn-160bn-taxes-northern-sea-route-2035-arctic- region-accounts-75-percent-gdp. 103. Trading Economics. Russia GDP. 2025 November 24, 2025]; Available from: https://tradingeconomics.com/russia/gdp. 104. Congressional Budget Office, Costs to build Coast Guard's polar security cutters continue to grow. 2024. 105. Bertelsen, R.G. No. 2 | Introducing the Arctic Infrastructure Inventory. 2020; Available from: https://www.wilsoncenter.org/blog-post/no-2-introducing-arctic-infrastructure-inventory. 106. Arts, S., et al., Leveraging the Joint Capacity of Arctic Allies and Industry Partners. 2025. 107. Studies, I.I.f.S. Icebreakers and ice-capable patrol ships. 2022; Available from: https://www.statista.com/chart/33823/icebreakers-and-ice-capable-patrol-ships/. 108. Affairs, U.S.H.C.o.F., China's influence in the Arctic. 2022. 109. Canada, P.S.a.P. Canada signs new partnership agreement with United States and Finland to produce Arctic and polar icebreakers. 2024 November 29, 2025]; Available from: https://www.canada.ca/en/public-services-procurement/news/2024/11/canada-signs-new- partnership-agreement-with-united-states-and-finland-to-produce-arctic-and-polar- icebreakers.html. 110. Defense, B. GOP reconciliation bill includes over $9 billion for new icebreaker ships. 2025 November 29, 2025]; Available from: https://breakingdefense.com/2025/05/gop-reconciliation- bill-includes-over-9b-for-new-icebreaker-ships/. 111. Argote, L., S.L. Beckman, and D. Epple, The persistence and transfer of learning in industrial settings. Management science, 1990. 36(2): p. 140–154. 112. Benkard, C.L., Learning and forgetting: The dynamics of aircraft production. American Economic Review, 2000. 90(4): p. 1034–1054. 113. Hedlund, S. Russian energy politics at a crossroads. 2025; Available from: https://www.gisreportsonline.com/r/russia-energy-2/. 114. Katinas, P. and V. Raghunandan. October 2025 — Monthly analysis of Russian fossil fuel exports and sanctions. 2025; Available from: https://energyandcleanair.org/october-2025-monthly- analysis-of-russian-fossil-fuel-exports-and-sanctions/. 115. IRENA, Geopolitics of the energy transition: Energy security. 2024, International Renewable Energy Agency: Abu Dhabi. This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=6359303 Preprint not peer reviewed -- 22 of 22 --