New insights into projected Arctic sea road: operational risks, economic values, and policy implications

Abstract

As Arctic sea ice continues to retreat, the seasonally navigable Arctic expected by mid-century or earlier is likely to facilitate the growth of polar maritime and coastal development. Here, we systematically explore the potentials for opening of trans-Arctic sea routes across a range of emissions futures and multi-model ensembles on daily timescales. We find a new Transpolar Sea Route in the western Arctic for open water vessels starting in 2045 in addition to the central Arctic corridor over the North Pole, with its frequency comparable to the latter during the 2070s under the worst-case scenario. The emergence of this new western route could be decisive for operational and strategic outcomes. Specifically, the route redistributes transits away from the Russian-administered Northern Sea Route, lowering the navigational and financial risks and the regulatory friction. Navigational risks arise from narrow straits that are often icy choke points. Financial risks arise from the substantial interannual sea ice variability and associated uncertainty. Regulatory friction arises from Russian requirements imposed under the Polar Code and Article 234 of the UN Convention on the Law of the Sea. These imposts are significantly reduced with shipping route regimes that enable open water transits wholly outside Russian territorial waters, and these regimes are revealed most accurately using daily ice information. The near-term navigability transition period (2025-2045) may offer an opportunity for maritime policy evaluation, revision, and action. Our user-inspired evaluation contributes towards achieving operational, economic and geopolitical objectives and serves the goal of planning a resilient, sustainable, and adaptive Arctic future.

Supplementary information: The online version contains supplementary material available at 10.1007/s10584-023-03505-4.

Keywords: Arctic maritime access; Geopolitics; Model uncertainty; Navigation risk and cost; Scenario analysis; Sea ice variability.

Fig. 1

Projected change in shipping season length under varying levels of anthropogenic emissions. a Interannual season length variability for 2015–2079. b Decadal season length distribution through time. The vertical boxplot in a shows the rate of annual change in season length and the associated uncertainty range for the corresponding level of emissions. The horizontal boxplot in panel b displays the interquartile and range of season length over the respective decade for 14 CMIP6 simulations. The boxplot is defined as line inside the box, median; colored box, interquartile range (IQR) from 25th percentile (Q1) to 75th percentile (Q3); upper and lower whiskers, bounds of Q3 + 1.5 × IQR and Q1‒1.5 × IQR, respectively. Note that in panel b, the frequency of zero value is separated from the rest as a single bin in the leftmost of the histogram; and each emissions scenario is colored in the same way as that shown in a

Fig. 2

Geographical distribution of projected trans-Arctic shipping routes under the SSP5-8.5 forcing scenario during 2065–2074. The navigation routes of Northeast Passage (NEP), Northwest Passage (NWP), Central Arctic Route (CAR), and Transpolar Sea Route (TSR) are labeled in white. The transit density is colored in blue shading and based on the proportion of total transits that falls within a certain radius (unit: km) over the decade. Each transit is attached to equal weight in line density calculation. The background Arctic map consists of North Pole (star symbol), land areas (brown), internal waters (dark blue), territorial seas (defined as 12 nautical miles (nm) from a nation’s baseline; darker colored as per the inset legend), exclusive economic zones (EEZs; defined as 200 nm extending from a nation’s baseline, thin crosshatched area), and claims to the extended continental shelf (thick hatched area) as of 23 May 2019. The March 2021 submissions for extensions of continental shelf made by Russia are not shown

Fig. 3

Quantification of navigational and financial risks. a Monthly safe navigation index (SNI) across scenarios and decades. The monthly mean SNI values averaged for each model are displayed in dots. Boxplot is solely applied to months with no less than four models available. A SNI closes to 1 indicates low-risk navigation. For visualization, the SNI of June of 2035–2044 has the same bound limit as in other time periods, and a full view of it is shown in inset. b Risk of measuring additional cost of variance with multi-model ensemble (MME) mean. It is quantified by the ratio of additional variance cost estimated by applying MME mean to that resulting from model uncertainty across high- to low-emissions pathways. The dashed gray line indicates where the ratio is one. Above the dashed line, the cost of interannual variability is greater than that of model structural uncertainty and vice versa

Fig. 4

Levels of Arctic navigability beyond the Russian territorial waters under the four SSPs scenarios over the period 2015–2079. Each dot represents the level of confidence in individual years. The color line indicates the change of confidence using a quadratic regression fitted to the navigation probability series, which explains 79%, 62%, 48%, and 20% of the variance in navigation probability from very high- to low-emissions scenarios. The level of confidence corresponds to the IPCC convention