Shipping traffic through the Arctic Ocean: Spatial distribution, seasonal variation, and its dependence on the sea ice extent

Abstract

The reduction in sea ice cover with Arctic warming facilitates shipping through remarkably shorter shipping routes. Automatic identification system (AIS) is a powerful data source to monitor Arctic Ocean shipping. Based on the AIS data from an online platform, we quantified the spatial distribution of shipping through this area, its intensity, and the seasonal variation. Shipping was heterogeneously distributed with power-law exponents that depended on the vessel category. We contextualized the estimated exponents with the analytical distribution of a transit model in one and two dimensions. Fishing vessels had the largest spatial spread, while narrower shipping routes associated with cargo and tanker vessels had a width correlated with the sea ice area. The time evolution of these routes showed extended periods of shipping activity through the year. We used AIS data to quantify recent Arctic shipping, which brings an opportunity for shorter routes, but likely impacting the Arctic ecosystem.

Keywords: Earth sciences; Global change; Oceanography.

Graphical abstract

Figure 1

Arctic shipping density between January 2020 and April 2022 Shipping density is computed as the aggregated shipping transit intensity (time spent transiting) over all the vessels, at each $0.1^{\text{°}}\times 0.1^{\text{°}}$ cell, divided by the cell area. Top panel represents the total shipping traffic, while bottom panels include the traffic for passenger, tanker, cargo, and fishing vessel categories.

Figure 2

Shipping density distributions for total shipping (left) and each vessel category (right) These heavy-tailed distributions fit to a power-law distribution $pdf\left(x\right)\sim x^{-\alpha }$ for $x>x_{\mathit{min}}$. The slope, obtained from fitting with the Python package powerlaw, is represented in the black lines.

Figure 3

Route detection on shipping traffic For every cell, given its longitude, we computed the relative shipping density as the shipping density (as represented in Figure 1), divided by the average shipping density for that longitude, computed over cells with non-null density. This highlighted some hotspots at each longitude, which we associate with the shipping routes. Left panel represents the total shipping, while right panels are broken down into vessel categories.

Figure 4

Time evolution of the shipping transit intensity through the Arctic Ocean The black line represents the total shipping transit intensity, while the colored lines stand for the vessel categories.

Figure 5

Time evolution of the shipping traffic per longitude The top panel represents the average shipping density over cells with non-null shipping at each longitude, while the bottom represents the sectional length of these cells. White entries represent the absence of traffic. The gray lines stand for the longitudes that we chose as the most representative for measuring the Northeastern and the Northwest Passage routes being, respectively, 150° and −90°.

Figure 6

Correlation between shipping width and sea ice area We selected two specific longitudes that displayed a seasonal behavior: −90° (West, Northwest Passage route) and 150° (East, Northeastern route), computing the shipping width L (Figure 5). We computed the Pearson correlation of the shipping width with the non-zero values of the sea ice area in different Arctic seas, obtaining the maximum absolute value of the correlation for the Beaufort Sea (West, $C=-0.90$) and the Kara Sea (East, $C=-0.84$). The main plots represent the shipping width as a function of the sea ice area, while the insets depict their temporal evolution. Points represent data, while the dashed curves are exponentially decreasing fits, $L=A{\mathrm{e}}^{-BI}$, with $B_{\text{NWP}}=7.68\times {10}^{-6}$ km⁻² and $B_{\text{NER}}=4.29\times {10}^{-6}$ km⁻² for, respectively, the Northwest Passage and the Northeastern route.

Figure 7

Time evolution of the shipping width at two specific longitudes: −90° (West, Northwest Passage route) and 150° (East, Northeastern route), to characterize the evolution of shipping traffic The local shipping width is computed as, for each constant longitude and each month, the cross-sectional length of the cells that display non-null shipping traffic. Data between 2010 and 2015 have been extracted from a different database with resolution of 0.25°, number of unique vessels per month. For comparability, this figure includes our analyzed dataset considering grid cells of 0.25° side instead of those of 0.1° side.