Analysis of Sea Ice Timing and Navigability Along the Arctic Northeast Passage from 2000 to 2019

Journal of Marine Science and Engineering

Article Analysis of Sea Ice Timing and Navigability along the Arctic Northeast Passage from 2000 to 2019

Min Ji 1 , Guochong Liu 1,* , Yawen He 2, Ying Li 1 and Ting Li 1

1 Surveying and Spatial Information Institute, Shandong University of Science and Technology, Qingdao 266590, China; [email protected] (M.J.); [email protected] (Y.L.); [email protected] (T.L.) 2 Ocean and Spatial Information College, China University of Petroleum, Qingdao 266000, China; [email protected] * Correspondence: [email protected]; Tel.: +86-178-0626-3036

Abstract: The ablation of Arctic sea ice makes seasonal navigation possible in the Arctic region, which accounted for the apparent influence of sea ice concentration in the navigation of the Arctic route. This paper uses Arctic sea ice concentration daily data from January 1, 2000, to December 31, 2019. We used a sea ice concentration threshold value of 40% to define the time window for navigating through the Arctic Northeast Passage (NEP). In addition, for the year when the navigation time of the NEP is relatively abnormal, we combined with wind field, temperature, temperature anomaly, sea ice age and sea ice movement data to analyze the sea ice conditions of the NEP and obtain the main factors affecting the navigation of the NEP. The results reveal the following: (1) The sea ice concentration of the NEP varies greatly seasonally. The best month for navigation is September. The opening time of the NEP varies from late July to early September, the end of navigation is concentrated in mid-October, and the navigation time is basically maintained at more than 30 days. (2) The NEP was not navigable in 2000, 2001, 2003 and 2004. The main factors are the high amount of multi-year ice, Citation: Ji, M.; Liu, G.; He, Y.; Li, Y.; low temperature and the wind field blowing towards the Vilkitsky Strait and sea ice movement. The Li, T. Analysis of Sea Ice Timing and navigation time in 2012, 2015 and 2019 was longer, and the driving factors were the high temperature, Navigability along the Arctic weak wind and low amount of one-year ice. The navigation time in 2003, 2007 and 2013 was shorter, Northeast Passage from 2000 to 2019. and the influencing factors were the strong wind field blowing towards the Vilkitsky Strait. (3) The J. Mar. Sci. Eng. 2021, 9, 728. https:// key navigable areas of the NEP are the central part of the East Siberian Sea and the Vilkitsky Strait, doi.org/10.3390/jmse9070728 and the Vilkitsky Strait has a greater impact on the NEP than the central part of the East Siberian Sea. The main reason for the high concentration of sea ice in the central part of the East Siberian Sea Academic Editor: Lars (2000 and 2001) was the large amount of multi-year ice. The main reason for the high concentration Chresten Lund-Hansen of sea ice in the Vilkitsky Strait (2000 to 2004 and 2007, 2013) was the strong offshore wind in summer, −1 Received: 14 June 2021 all of which were above 4 m s , pushing the sea ice near the Vilkitsky Strait to accumulate in the Accepted: 27 June 2021 strait, thus affecting the opening of the NEP. Published: 1 July 2021 Keywords: Arctic; ice concentration; Northeast Passage; navigation window Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- iations. 1. Introduction In the past 30 years, Arctic sea ice has changed rapidly [1,2], with a warming rate of about 0.6 ± 0.07 ◦C per 10 years [3]. The annual average extent of Arctic sea ice also showed a decreasing trend, with the decreasing rate rising from 3% to 10% per 10 years [4], the Copyright: © 2021 by the authors. thickness of sea ice has become thinner [5,6], and the melting season has been extended [7]. Licensee MDPI, Basel, Switzerland. Studies have shown that, up to 2019, the average global temperature for 2015 to 2019 is This article is an open access article on track to be the warmest of any equivalent period on record, and the four lowest values distributed under the terms and of winter sea ice extent occurred in these five years [8]. Significantly climate warming conditions of the Creative Commons strengthened the ablation of polar sea ice and improved the ability of seasonal navigation Attribution (CC BY) license (https:// in the Arctic. The change of Arctic sea ice showed an apparent influence on the navigation creativecommons.org/licenses/by/ of the NEP. Factors such as solar radiation, temperature, wind field and other natural 4.0/).

J. Mar. Sci. Eng. 2021, 9, 728. https://doi.org/10.3390/jmse9070728 https://www.mdpi.com/journal/jmse J. Mar. Sci. Eng. 2021, 9, 728 2 of 26

factors can always affect the melting process and freezing time of sea ice. Therefore, the navigation time of the NEP shows highly uncertain. Among many Arctic routes, the NEP has significant advantages, which can save 15–50% of the voyage [9,10], greatly shortening the transportation distance from Western Europe to Northeast Asia. Smith et al. [11] obtained the best route in the Arctic seas based on the data of Arctic sea ice concentration and thickness in September. Rodrigues [12] used sea ice concentration data from 1979 to 2007 to study the sea ice coverage and the length of the ice-free season in the Arctic marginal area. Ruibo et al. [13] used remote sensing data from 1979 to 2012 to study the interannual and seasonal spatial variations of sea ice in the Northwest Passage (NWP), and found that the navigation period of the NWP increased by 32 days in 32 years. Dawei et al. [14] used NSIDC sea ice products to study the kinematic characteristics of sea ice in the Arctic Fram Strait and the northern part of the NEP. It was found that the opening period of the NEP has a significant relationship with the northward speed of the sea ice in the north. Polona et al. [15] used airborne sea ice thickness data from the floating ice areas in the southeast of the Laptev Sea in winter and found that the export volume of sea ice in winter strengthens the ablation of summer sea ice in the Laptev Sea. Chen et al. [16] look at future scenarios in a modeling experiment, but they also focused on the NEP and (at times) the Vilkitsky Strait. Zhiyuan Li [17] analyzed the characteristics of the Arctic ice conditions in summer and modified the risk index of polar business limit assessment to evaluate the navigation risk. Yangjun Wang [18] developed a new model to study the influence of sea ice conditions on the navigation of the NEP from 2020 to 2030, and found that the concentration and thickness of sea ice are the key factors for the navigation of icebreakers. Yu et al. [19] notes the importance of wind direction on determining whether sea ice is advected into or away from Vilkitsky Strait. Long Ma et al. [20] used sea ice concentration data from 2006 to 2015 and the first voyage route of the “Yongsheng” NEP in 2013 to obtain the 10-year navigation window of the NEP and analyze the navigation situation of the sea area through which the NEP passes. Xinqing et al. [21] used sea ice concentration data from 2002 to 2013 to study the sea ice distribution characteristics and navigation of the Vilkitsky Strait in the Arctic, which shows that the navigation time of the strait is basically more than 40 days per year, and the opening time varies from July to September. The ending time is relatively concentrated in October. Shang Meng et al. [22] used sea ice concentration data to analyze the sea ice variations and navigation conditions in the NEP, and concluded that the NEP is most likely to be opened in September. Most of the research is based on the data of sea ice concentration in the past 10 years or the study of navigation conditions in a small range of Arctic waters. The space-time span of the study is relatively small, and the abnormal years of channel opening are not analyzed. Sea ice concentration is a crucial factor for studying variations in sea ice conditions and route navigation. In this study, we used SSM/I (Special Sensor Microwave/Imager) and SSMIS (Special Sensor Microwave Imager/Sounder) sea ice concentration data from 2000 to 2019, combined with Arctic summer wind field, temperature, temperature anomalies, sea ice age and sea ice movement data, to analyze the characteristics of annual, seasonal and monthly sea ice variations in the NEP of the Arctic in the past 20 years, extracted the navigation window of the NEP in the past 20 years, and analyzed the ice conditions in the years of abnormal navigation. How to safely use the Arctic sea ice window to cross the Arctic is the focus of current research. Through the analysis of this article, we can have a more comprehensive understanding of the temporal and spatial variations of sea ice and navigation conditions of the NEP in the last 20 years, and provide certain references for the future development and utilization of the NEP.

2. Study Region, Data and Methods 2.1. Study Region We selected the NEP as the study area, which runs through the Arctic region and is one of the crucial Arctic routes. The NEP is a passage starting from the Norwegian Sea, passing through the Barents Sea, Kara Sea, Laptev Sea and East Siberian Sea to the Bering Strait. As J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 3 of 27

2. Study Region, Data and Methods 2.1. Study Region

J. Mar. Sci. Eng. 2021, 9, 728 We selected the NEP as the study area, which runs through the Arctic region3 ofand 26 is one of the crucial Arctic routes. The NEP is a passage starting from the Norwegian Sea, passing through the Barents Sea, Kara Sea, Laptev Sea and East Siberian Sea to the Bering Strait. As shown in Figure 1, the starting point of the route in this article we focused on is shownthe Bering in Figure Strait1, the (central starting geographic point of the coordinates: route in this 53.5° article N, 170.3° we focused E), with on a is total the Bering length of ◦ ◦ Strait5246.86 (central nautical geographic miles coordinates:: 0–1311.71 nautical 53.5 N, 170.3miles isE), the with Bering a total Strait, length 1311.71 of 5246.86–2098.74 nautical nau- miles:tical 0–1311.71 miles is nautical the East miles Siberian is the Sea, Bering 2098. Strait,74–28851311.71–2098.74.77 nautical nautical miles ismiles the is Laptev the East Sea, Siberian2885.77 Sea,–3410.46 2098.74–2885.77 nautical miles nautical is the miles Kara is the Sea, Laptev and 3410.46 Sea, 2885.77–3410.46–3935.14 nautical nautical milesmiles is the is theBarents Kara Sea. Sea, andBetween 3410.46–3935.14 November nautical and Apr milesil, the is NEP the Barents may be Sea. covered Between by November sea ice com- andpletely. April, In the May, NEP the may sea be ice covered on the by NEP sea icebegins completely. to melt, Inand May, the the melting sea ice continues on the NEP until beginsSeptember, to melt, when and the the melting sea ice continuesconcentration until on September, the NEP whenreaches the a seaminimum. ice concentration During the onArctic the NEP summer reaches, sea a minimum. ice concentrations During the decrea Arcticse summer, due to sea iceice concentrationsmelt. There can decrease be short dueperiods to sea iceof ice melt.-free There time can in the be shortNEP, periods which ofprovides ice-free convenience time in the NEP, to the which merchant provides ships conveniencescheduling to the the voyage, merchant and ships the scheduling commercial the value voyage, of the and NEP the commercial has become value increasingly of the NEPprominent. has become increasingly prominent.

FigureFigure 1. Schematic1. Schematic diagram diagram of of the the Arctic Arctic NEP. NEP. Section Section1 (triangle) 1 (triangle) is the is the Bering Bering Strait, Strait, Section Section2 2 (transverse(transverse line) line) is the is the East East Siberian Siberian Sea, Sea, Section Section3 (dotted-line) 3 (dotted-line) is the is the Laptev Laptev Sea, Sea, Section Section4 (pentacle) 4 (pentacle) is theis the Kara Kara Sea, Sea, and and Section Section5 (circle) 5 (circle) is the is the Barents Barents Sea. Sea.

2.2.2.2. Data Data 2.2.1. Sea Ice Concentration Data 2.2.1. Sea Ice Concentration Data In this article, we use the sea ice concentration data of SSM/I-SSMIS (Special Sensor In this article, we use the sea ice concentration data of SSM/I-SSMIS (Special Sensor Microwave/Imager–Special Sensor Microwave Imager/Sounder) provided by the National Microwave/Imager–Special Sensor Microwave Imager/Sounder) provided by the Na- Snow and Ice Data Center (NSIDC). The data set is Sea Ice Concentrations from Nimbus- tional Snow and Ice Data Center (NSIDC). The data set is Sea Ice Concentrations from 7 SMMR and DMSP SSM/I-SSMIS Passive Microwave Data (https://nsidc.org/data/nsidc- Nimbus-7 SMMR and DMSP SSM/I-SSMIS Passive Microwave Data 0051; accessed on 25 November 2020) [23], the version is Version 1. The spatial resolution of the data is 25 km, the temporal resolution is 1 day, and the time span is 1978–2020. The data are provided in the polar stereographic projection at a grid cell size of 25 × 25 km. The data set has a long time span, wide coverage and completeness, which is convenient for sea ice analysis for nearly 20 years. Since the data obtained in this article is in November 2020, we select the sea ice concentration data on the NEP of the Arctic from 1 January 2000 to 31 December 2019 in the data set for the analysis of this article. J. Mar. Sci. Eng. 2021, 9, 728 4 of 26

2.2.2. Auxiliary Data In this article, we use the seasonal average data of wind field, temperature, temperature anomaly, sea ice age and sea ice movement in the Arctic summer (July–September) for analysis, including the summer seasonal average maps of 2000, 2001, 2003, 2004, 2012, 2015 and 2019, as well as the monthly average maps for the July to September months of 2002, 2007 and 2013. The 10 m surface wind field map comes from the reanalysis grid data set ERA5 of the University of Maine’s reanalysis plotter (https: //climatereanalyzer.org/reanalysis/monthly_maps/; accessed on 25 November 2020) [24], the time resolution is 1 month and the grid resolution is 0.5◦ * 0.5◦. The time span is from January 1950 to May 2021, and the version of the dataset is Version 5. The temperature distribution of 2 m and the temperature anomaly of 2 m are from the reanalysis grid data set ERA5 (ESRL, https://psl.noaa.gov/repository/model/compare; accessed on 25 November 2020) [25] of the NOAA Earth System Research Laboratory, the time resolution is 1 month and the grid resolution is 0.25◦ * 0.25◦. The time span is from 1979 to 2021, and the version of the dataset is Version 5. The temperature anomaly here is based on the average value from 2000 to 2019. The sea ice age comes from the grid data set EASE- Grid Sea Ice Age (NSIDC, https://nsidc.org/data/NSIDC-0611/versions/4; accessed on 25 November 2020) [26] of National Snow and Ice Data Center, the spatial resolution is 25 km × 25 km, and the time resolution is 7 days. The time span is 1 January 1984 to 31 De- cember 2020, and the version of the dataset is Version 4. The sea ice motion comes from the grid data set Polar Pathfinder Daily 25 km EASE-Grid Sea Ice Motion Vectors (NSIDC, https://nsidc.org/data/NSIDC-0116/versions/4; accessed on 25 November 2020) [27] of NSIDC, with a spatial resolution of 25 km × 25 km. The time resolution is 7 days, and the time span is from 25 October 1978 to 31 December 2020. The version of the dataset is Version 4.

2.3. Methods 2.3.1. Sea Ice Concentration Extraction on the NEP We use the latitude and longitude coordinates of the sea ice concentration product itself and superimpose it with the NEP in the same coordinate system. Then, we perform interpolation analysis on the original sea ice concentration data of the NEP, and generate new interpolated raster data. Then, the average sampling point is extracted on the NEP, and the concentration value of the point is extracted from the interpolation data with the coordinates of the sampling point to obtain the sea ice concentration of the NEP.

2.3.2. Sea Ice Age and Sea Ice Movement We obtain the netCDF weekly data of sea ice age and sea ice movement for the required years from NSIDC. For the sea ice age, we use MATLAB to read the NetCDF data, and then superimpose the 13-week ice age data in the summer (July to September) of 2000, 2001, 2003, 2004, 2012, 2015 and 2019, and superimpose the monthly ice age in July (4 weeks), August (5 weeks) and September (4 weeks) of 2002, 2007 and 2013, and calculate the seasonal average ice age value in summer and monthly average ice age value. Finally, the average ice age obtained is rounded up to the nearest integer. For the sea ice movement, we use Matlab to read the u and v components of the ice motion data, and superimpose the values of u and v of the ice motion in the time span required in this article, to obtain the average values of u and v, respectively. When u is positive, it represents sea ice moving eastward. When v is positive, it represents sea ice moving northward. Formula (1) is used to get the speed and magnitude of the average sea ice movement. * p a = u2 + v2 (1) J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 5 of 27

For the sea ice movement, we use Matlab to read the u and v components of the ice motion data, and superimpose the values of u and v of the ice motion in the time span required in this article, to obtain the average values of u and v, respectively. When u is positive, it represents sea ice moving eastward. When v is positive, it represents sea ice moving northward. Formula (1) is used to get the speed and magnitude of the average sea ice movement.

J. Mar. Sci. Eng. 2021, 9, 728 5 of 26 ⃑푎 = √푢2 + 푣2 (1)

3. Sea Ice Concentration Analysis 3. Sea Ice Concentration Analysis 3.1.3.1. Annual Annual Average Average Variation Variation BasedBased on on the the SSMI/SSMIS SSMI/SSMIS sea sea ice ice concentration data data from from 1 January1 January 2000 2000 to 31 to De- 31 De- cembercember 2019, 2019, we we calculated calculated the the annual annual averageaverage sea sea ice ice concentration concentration of eachof each location location on on the theNEP NEP of ofthe the Arctic Arctic and and was was shown shown inin FigureFigure2. 2. According According to Figureto Figure2, we 2, can we more can more intuitivelyintuitively get get the the sea sea ice ice concentration concentration atat thethe navigable navigable nodes nodes in eachin each year, year, and and ﬁnd find the the choke points that affected the navigability of the NEP that year. choke points that affected the navigability of the NEP that year.

section1 section2 section3 section4 section5

The maximum annual average sea ice concentration in non-special years

The minimum annual average sea ice concentration

Figure 2. Average annual sea ice concentration of the NEP from 2000 to 2019. Among them, the red squares indicate that Figure 2. Average annual sea ice concentration of the NEP from 2000 to 2019. Among them, the red squares indicate that at at 1574 nautical miles, the sea ice concentration in 2001 and 2000 were the top two places, respectively. The black triangles 1574 nautical miles, the sea ice concentration in 2001 and 2000 were the top two places, respectively. The black triangles indicate that at 2885 nautical miles, the sea ice concentration in 2003 and 2000 were the top two places, respectively. The indicate that at 2885 nautical miles, the sea ice concentration in 2003 and 2000 were the top two places, respectively. The red red diamond indicates that at 2885 nautical miles, the sea ice concentration in 2012 was the lowest. diamond indicates that at 2885 nautical miles, the sea ice concentration in 2012 was the lowest.

As Asshown shown in in Figure Figure 2, thethe areas areas with with high high sea sea ice concentrationice concentration in the in NEP the from NEP 2000 from to 2000 to 20192019 were were mainlyconcentrated concentrated between between Sections Section2 and 2 and4, that Section is, 1311.71–3410.46 4, that is, 1311. nautical71–3410.46 nauticalmiles. miles. At Section At Section1, the annual 1, the annual average average sea ice concentration sea ice concentration of the NEP of increases the NEP with increases withdistance. distance. At At 1574 1574 nautical nautical miles, miles, it can it becan clearly be clearly seen thatseen the that sea the ice sea concentration ice concentration in in 20012001 was was anomalous, as as high high as as 90%, 90%, followed followed by by 2000. 2000. At SectionsAt Section2 ands 23 and, the 3 annual, the annual average sea ice concentration ﬁrst declines and then increases, around 2885 nautical miles, average sea ice concentration first declines and then increases, around 2885 nautical miles, the annual average sea ice concentration in 2003 was extremely high, followed by 2000, the whileannual the average sea ice concentrationsea ice concentration in 2012 reached in 2003 a recordwas extremely low, only high, 60%, whichfollowed was by the 2000, whilelowest the valuesea ice of theconcentration peak sea ice concentrationin 2012 reached in the a pastrecord 20 years. low, Atonly Sections 60%,4 which and5, thewas the lowestaverage value sea of ice the concentration peak sea iceon concentration the route decreases in the past until 20 it drops years. to At 0%. Section Combining 4 and theSection 5, theresults average of Figure sea ice2 with concentration the geography, on the it can route be seen decreas thates the until middle it drops section to of 0%. the Combining route the (aboutresults 1000–3000 of Figure nautical 2 with miles) the geography is located in, it high can latitudes, be seen closethat the to the middle North section Pole, and of the routethe (about sea ice 1000 concentration–3000 nautical is high, mile thes) risk is oflocated ship navigation in high latitudes, is greater. close The beginningto the North and Pole, end of the NEP are located in the mid to high latitudes, where the sea ice concentration is low and the navigation risk is smaller. In the recent 20 years annual average sea ice concentration in Figure2, it can be seen that the annual average sea ice concentration peaks on the NEP in 2000, 2001, 2003 and 2004 were all higher than 85%, and the annual sea ice concentration changed greatly. In 2012, J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 6 of 27

and the sea ice concentration is high, the risk of ship navigation is greater. The beginning and end of the NEP are located in the mid to high latitudes, where the sea ice concentra- J. Mar. Sci. Eng. 2021, 9, 728 6 of 26 tion is low and the navigation risk is smaller. In the recent 20 years annual average sea ice concentration in Figure 2, it can be seen that the annual average sea ice concentration peaks on the NEP in 2000, 2001, 2003 and 20152004 and were 2019, all higher the annual than average85%, and sea the ice annual concentration sea ice concentration changed relatively changed greatly, greatly. with In no considerable2012, 2015 and abrupt 2019, changes. the annual Therefore, average sea we selectice concentration seven special changed years includingrelatively g 2000,reatly 2001,, 2003,with 2004,no considerable 2012, 2015 andabrupt 2019 changes. for research Therefore, (here, we the select year seven of shorter special navigation years including periods is not2000, considered 2001, 2003, for 2004, the time2012, being). 2015 and Except 2019 forfor specialresearch years, (here, the the sea year ice of concentration shorter naviga- of the remainingtion periods 13 is years not considered is averaged for for the the time following being). Except analysis. for special years, the sea ice concentration of the remaining 13 years is averaged for the following analysis. 3.2. Seasonal Average Variation 3.2. Seasonal Average Variation Based on the daily average sea ice concentration data of the NEP, the average sea ice concentrationBased on variations the daily average of four seasonssea ice concentration in each year were data calculatedof the NEP, and the shownaverage in seaFigure ice 3. Inconcentration order to distinguish variations between of four seasons special in years each and year non-special were calculated years, and the shown monthly in Figure average sea3. In ice order concentration to distinguish of each between year special in the 7years special and yearsnon-special and the years, monthly the monthly average aver- sea ice age sea ice concentration of each year in the 7 special years and the monthly average sea concentration of the remaining 13 years are shown in Figure3. We divide the seasons: ice concentration of the remaining 13 years are shown in Figure 3. We divide the seasons: January–March is regarded as winter, April–June is regarded as spring, July–September is January–March is regarded as winter, April–June is regarded as spring, July–September regarded as summer, and October–December is autumn. is regarded as summer, and October–December is autumn.

FigureFigure 3. The3. The average average sea sea ice ice concentration concentration of of the the NEP NEP from from 20002000 toto 2019,2019, (a) is winter, ( b)) is is spring, spring, ( (cc)) is is summer, summer, and and (d (d) ) is autumn.is autumn. Among Among them, them, the red the boxred inbox (b )in represents (b) represents the variation the variation in the in averagethe average sea icesea concentration ice concentration in spring in spring in the in Laptevthe Laptev Sea. In (c), the red triangles indicate that the sea ice concentration in the East Siberian Sea in 2000 and 2001 was Sea. In (c), the red triangles indicate that the sea ice concentration in the East Siberian Sea in 2000 and 2001 was higher than higher than 60%, the blue diamond indicates that the sea ice concentration in the Vilkitsky Strait in 2003 was higher than 60%,80%, the and blue the diamond black box indicates indicates thatthat the sea ice ice concentration concentration in in the the Vilkitsky Vilkitsky Strait Strait in 2000, in 2003 2004 was and higher 2001 was than respectively 80%, and the blackabove box 50%. indicates that the sea ice concentration in the Vilkitsky Strait in 2000, 2004 and 2001 was respectively above 50%.

Laptev Sea (2098.74–2623.43 nautical miles) decreases. Compared with winter, the Laptev Sea (red box) has the largest amount of variation in sea ice concentration. In the spring, sea ice along the NEP absorbs solar radiation and begins to melt. During summer the sea ice continues to retreat northwards, and the sea ice concentration of the NEP drops to a minimum or even zero, and an ice-free section appears. The Bering Strait and the Barents Sea occasionally contain a small amount of ﬂoating ice of less than 20%. Except in 2000, 2001, 2003 and 2004, the sea ice concentration of the East Siberian Sea, the Laptev Sea and the Kara Sea in summer is generally less than 50%. In autumn (Figure3d), the Arctic sea ice enters the freezing period, and the sea ice expands southward. The sea ice concentration in each sections on the NEP gradually increases. The amount of sea ice in the NEP continues to increase, reaching its maximum in winter. Third, it can be seen in Figure3 that the sea ice concentration of the NEP in 2000, 2001, 2003 and 2004 was more abnormal than other years. Especially in summer (Figure3c), the sea ice concentration of the East Siberian Sea (1574 nautical miles) reached more than 60% in 2000 and 2001 (red triangle), which was at a high level, and the risk of navigation was high. In the Vilkitsky Strait (2623–2885 nautical miles), the sea ice concentration reached 80% in 2003 (blue diamond) and more than 50% in 2000, 2004 and 2001 (black box). In the summer of 2012, 2015 and 2019, the sea ice concentration of the NEP was lower, which was below 30%, and the risk of navigation was small. It can be seen from the above analysis that the seasonal variation of the average sea ice concentration of the NEP is quite different. The NEP is often covered by sea ice, and the inﬂuence of sea ice should be taken into account when sailing. In summer (July–September), the NEP has the least sea ice concentration, which is the best time for navigating ships, and marine operations are not easy in winter and spring.

3.3. Monthly Average Variation The NEP is feasible for navigation in summer. Since the sea ice cover minimum usually occurs in September each year, and it takes a while to grow the sea ice back, the sea ice concentration in the NEP is not high in October. Therefore, in this section, we put October in the summer for analysis. The spatial variability along the NEP is shown in Figure4. The NEP is in the freezing period from October to March of the following year, and it is in the melting process from April to September. Due to the inﬂuence of sea ice melting degrees, the inter-annual difference in the sea ice concentration of the NEP is great from July to October. It can be seen from Figure4 that the sea ice concentration of the NEP is the smallest in September, followed by August and July, and the sea ice concentration starts to increase in October. In July, the sea ice concentration of NEP is higher in each year, especially in 2000, 2001, 2003 and 2004 (red box), the sea ice concentration is above 70%, and the navigation risk is high. By August, the sea ice concentration is considerably reduced. Among them, in the East Siberian Sea (1574 nautical miles), the sea ice concentration in 2000 and 2001 (red triangle) was higher than 60%. Near the Vilkitsky Strait (2885 nautical miles), the sea ice concentration in 2003, 2000 and 2004 (blue square) were the top three, up to 50%, 70% and 80% respectively, which had a considerable impact on the navigation. In August 2012 and 2015, the monthly average sea ice concentration of the NEP was below 20%. In August 2019, the sea ice concentration of the NEP approached 0, which had little impact on navigation. In September, the sea ice in the NEP was sparse, and the NEP was in an open sea without ice for many years. In September of 2012, 2015 and 2019, the NEP entered an ice-free state, very feasible for navigation. However, in September 2000, 2001, 2003 and 2004, there was still a considerable amount of sea ice near the East Siberian Sea (green diamond) and the Vilkitsky Strait (blue box). The risk of navigation was relatively high. In October, the sea ice concentration of the NEP years began to increase in each year. Compared with other years, the sea ice concentration in 2000, 2001, 2003 and 2004 was extremely high, with the peak reaching 80%. The monthly average sea ice concentration along the NEP in October 2012 and 2019 was below 20%, while the 2015 and multi-year average sea ice concentration was slightly higher, with the peak at 40%. J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 7 of 27

Figure 3 reveals the variation in the average seasonal sea ice concentration of the NEP from 2000 to 2019. It can be seen from Figure 3 that the variations of sea ice concentration in the four seasons mainly reveal four characteristics. First, the ice section of the NEP is from 787 nautical miles to 3410 nautical miles, and there are perennial ice-free areas at the beginning and end of the route. Second, the sea ice concentration of the NEP has a relatively considerable seasonal characteristic. In winter (Figure 3a), the Arctic temperature is low and the sea is frozen. The sea ice concentration of the NEP reaches its maximum, which is above 95%, and the difference between the years is small. Entering spring (Figure 3b), the temperature in the northern hemisphere begins to rise, and the sea ice concentration in the Laptev Sea (2098.74–2623.43 nautical miles) decreases. Compared with winter, the Laptev Sea (red box) has the largest amount of variation in sea ice concentration. In the spring, sea ice along the NEP absorbs solar radiation and begins to melt. During summer the sea ice continues to retreat northwards, and the sea ice concentration of the NEP drops to a minimum or even zero, and an ice-free section appears. The Bering Strait and the Barents Sea occasionally contain a small amount of floating ice of less than 20%. Except in 2000, 2001, 2003 and 2004, the sea ice concentration of the East Siberian Sea, the Laptev Sea and the Kara Sea in summer is generally less than 50%. In autumn (Figure 3d), the Arctic sea ice enters the freezing period, and the sea ice expands southward. The sea ice concentration in each sections on the NEP gradually increases. The amount of sea ice in the NEP continues to increase, reaching its maximum in winter. Third, it can be seen in Figure 3 that the sea ice concentration of the NEP in 2000, 2001, 2003 and 2004 was more abnormal than other years. Especially in summer (Figure 3c), the sea ice concentration of the East Siberian Sea (1574 nautical miles) reached more than 60% in 2000 and 2001 (red triangle), which was at a high level, and the risk of navigation was high. In the Vilkitsky Strait (2623–2885 nautical miles), the sea ice concentration reached 80% in 2003 (blue diamond) and more than 50% in 2000, 2004 and 2001 (black box). In the summer of 2012, 2015 and 2019, the sea ice concentration of the NEP was lower, which was below 30%, and the risk of navigation was small. It can be seen from the above analysis that the seasonal variation of the average sea ice concentration of the NEP is quite different. The NEP is often covered by sea ice, and the influence of sea ice should be taken into account when sailing. In summer (July–Sep- tember), the NEP has the least sea ice concentration, which is the best time for navigating ships, and marine operations are not easy in winter and spring.

3.3. Monthly Average Variation The NEP is feasible for navigation in summer. Since the sea ice cover minimum usually occurs in September each year, and it takes a while to grow the sea ice back, the sea J. Mar. Sci. Eng. 2021, 9, 728 ice concentration in the NEP is not high in October. Therefore, in this section, we put Oc-8 of 26 tober in the summer for analysis. The spatial variability along the NEP is shown in Figure 4.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 8 of 27

FigureFigure 4. Time-sequence 4. Time-sequence map map of monthlyof monthly average average sea sea ice ice concentration concentration of of the the NEP NEP in in summer summer from from 2000 2000 to to 2019. 2019. In In ( a(),a), the the 70% line indicates that the peak value of monthly average sea ice concentration in July in 2000, 2001, 2003 and 2004 70% line indicates that the peak value of monthly average sea ice concentration in July in 2000, 2001, 2003 and 2004 was was respectively above 70%. In (b), the red triangles represent that in the East Siberian Sea, the sea ice concentration in respectively2000 and above2001 were 70%. the In (topb), two, the red and triangles the blue represent squares represent that in the that East in Siberianthe Vilkitsky Sea, theStrait sea, the ice concentrationsea ice concentration in 2000 in and 20012003, were 2000 the and top two,2004 andwere the the blue top squaresthree. Inrepresent (c), the green that diamonds in the Vilkitsky represent Strait, that the in sea theice East concentration Siberia Sea, the in2003, sea ice 2000 con- and 2004centration were the in top 2001 three. and 2000 In ( cwere), the the green top two, diamonds the blue represent square represents that in the that East in the Siberia Vilkitsky Sea, Strait the sea, the ice sea concentration ice concentra- in 2001tion and in 20002001,were 2003 theand top2004 two, were the the blue top squarethree. In represents (d), the 80% that line in in the Figure Vilkitsky 4 d indicates Strait, the that sea the ice peak concentration value of monthly in 2001, 2003average and 2004 seawere ice concentration the top three. in In October (d), the in 80% 2000, line 2001, indicates 2003 and that 2004 the was peak respectively value of monthly above 80%. average sea ice concentration in October in 2000, 2001, 2003 and 2004 was respectively above 80%. The NEP is in the freezing period from October to March of the following year, and it is in the melting process from April to September. Due to the influence of sea ice melting 4.degrees, Connectivity the inter Analysis-annual ofdifference the NEP in the sea ice concentration of the NEP is great from July to October. It can be seen from Figure 4 that the sea ice concentration of the NEP is The navigation window of a route refers to the navigation period during which the smallest in September, followed by August and July, and the sea ice concentration the sea ice concentration does not affect the safe navigation of ships on the route, and starts to increase in October. In July, the sea ice concentration of NEP is higher in each it reflects the sea ice conditions of the sea area which the route passes. According to year, especially in 2000, 2001, 2003 and 2004 (red box), the sea ice concentration is above Shibata et al.’s [28] description of sea ice concentration and ice and navigation conditions, 70%, and the navigation risk is high. By August, the sea ice concentration is considerably when the sea ice concentration is 10–30%, there is only a small amount of crushed ice on reduced. Among them, in the East Siberian Sea (1574 nautical miles), the sea ice concen- the sea surface, which is very sparse. Navigation is very smooth at this time. When the sea tration in 2000 and 2001 (red triangle) was higher than 60%. Near the Vilkitsky Strait (2885 icenautical concentration miles), the is 40–60%,sea ice concentration there is a large in 2003, amount 2000 of and unconnected 2004 (blue square) crushed were ice on the the top sea surface,three, up and to navigation50%, 70% and is more 80% difficultrespectively, at this which time. had Let a usconsiderable take the “Yongsheng impact on the Ship” nav- that crossedigation. the In ArcticAugust for 2012 the and first 2015, time the as monthly an example. average The sea icebreaking ice concentration capacity of of the the NEP cargo shipwas is below Arc4. 20%. Assuming In August that 2019, the the cargo sea shipice concentration sails independently, of the NEP without approached considering 0, which the assistancehad little ofimpact other on icebreakers, navigation. theIn September, threshold ofthe sea sea ice ice concentration in the NEP was through sparse, whichand the the “Yongsheng”NEP was in an can open pass sea is without 40%. Therefore, ice for many we setyears. the In threshold September of of sea 2012, ice 2015 concentration and 2019, at 40%.the NEP The algorithmentered an ofice navigable-free state,window very feasible is as for follows: navigation. the starting However, day in is Sep thetember first day when2000, the 2001, sea 2003 ice concentrationand 2004, there at was all pointsstill a considerable of the NEP isamount less than of sea 40% ice for near 3 consecutive the East days,Siberian and theSea ending(green diamond) day is the and sea icethe concentration Vilkitsky Strait at ( allblue points box). of The the risk NEP of is navigation greater than 40%was for relatively 3 consecutive high. In days. October, the sea ice concentration of the NEP years began to in- creaseFigure in each5 shows year. the Compared overall seawith ice other concentration years, the insea the ice NEP concentration from 2000 in to 2000, 2019. 2001, It can be2003 seen and that 2004 the was ice segmentextremely of high, the with NEP the isfrom peak 787.03reaching nautical 80%. The miles monthly to 3410.46 average nautical sea ice concentration along the NEP in October 2012 and 2019 was below 20%, while the 2015 and multi-year average sea ice concentration was slightly higher, with the peak at 40%.

4. Connectivity Analysis of the NEP The navigation window of a route refers to the navigation period during which the sea ice concentration does not affect the safe navigation of ships on the route, and it reflects the sea ice conditions of the sea area which the route passes. According to Shibata et al.’s [28] description of sea ice concentration and ice and navigation conditions, when the sea ice concentration is 10–30%, there is only a small amount of crushed ice on the sea surface, which is very sparse. Navigation is very smooth at this time. When the sea ice concentration is 40–60%, there is a large amount of unconnected crushed ice on the sea surface, and

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 9 of 27

navigation is more difficult at this time. Let us take the “Yongsheng Ship” that crossed the Arctic for the first time as an example. The icebreaking capacity of the cargo ship is Arc4. Assuming that the cargo ship sails independently, without considering the assistance of other icebreakers, the threshold of sea ice concentration through which the “Yongsheng” can pass is 40%. Therefore, we set the threshold of sea ice concentration at 40%. The algo- rithm of navigable window is as follows: the starting day is the first day when the sea ice J. Mar. Sci. Eng. 2021, 9, 728 concentration at all points of the NEP is less than 40% for 3 consecutive days, and the 9 of 26 ending day is the sea ice concentration at all points of the NEP is greater than 40% for 3 consecutive days. Figure 5 shows the overall sea ice concentration in the NEP from 2000 to 2019. It can miles,be seen and that there the ice is lesssegment than of 40% the ofNEP the is ice from in the787.03 rest nautical of the year.miles Therefore,to 3410.46 nautical we select from 1311.71miles, and nautical there milesis less than to 3410.46 40% ofnautical the ice in miles the rest to of study the year. the navigationTherefore, we window select from of the NEP, and1311.71 deﬁne nautical that whenmiles to the 3410.46 navigation nautical time miles is atto leaststudy 10the days, navigation the NEP window can beof the opened to ensureNEP, and the define practical that feasibilitywhen the navigation of navigation. time is at least 10 days, the NEP can be opened to ensure the practical feasibility of navigation.

Figure 5. The overall sea ice concentration of the NEP from 2000 to 2019; the black triangles indicate Figurethe start 5. andThe end overall positions sea ice of the concentration ice segment.of the NEP from 2000 to 2019; the black triangles indicate the start and end positions of the ice segment. 4.1. Navigation Window from 2000 to 2019 4.1. Navigation Window from 2000 to 2019 Figure 6 reveals the navigation window information of the NEP from 2000 to 2019. The Figureabscissa6 revealsis the location the navigation of the route, window and the information ordinate is the of number the NEP of from days 2000in the to 2019. Thecurrent abscissa year. Table is the 1 location is the navigation of the route, timetable and of the the ordinateNEP from is 2000 the to number 2019, including of days in the currentthe start year. and end Table dates1 is of the the navigation navigation timetableeach year and of the the NEPnumber from of 2000the NEP to 2019,available including theeach start year. and end dates of the navigation each year and the number of the NEP available each year.

Table 1. Navigation schedule of the NEP from 2000 to 2019.

Actual Start Date Actual End Date Actual Navigation Navigation Period Year of Navigation of Navigation Period (Days) Standard Deviation 2000 09-10 09-12 3 41.051 2001 – – 0 43.078 2002 08-17 09-19 33 26.219 2003 – – 0 38.239 2004 – – 0 41.286 2005 08-16 10-16 61 19.649 2006 08-14 09-23 40 21.343 2007 09-09 10-17 38 24.136 2008 08-12 10-05 54 27.383 2009 08-10 10-09 60 21.393 2010 08-01 10-09 69 17.11 2011 08-16 10-17 62 15.062 2012 07-31 10-21 82 13.895 2013 08-30 09-27 28 33.371 2014 08-12 10-11 60 21.773 2015 07-19 10-13 86 11.223 2016 08-08 10-21 74 14.26 2017 08-13 10-08 56 22.48 2018 09-03 10-13 40 24.777 2019 07-22 10-17 87 8.9387 J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 9 of 27

navigation is more difficult at this time. Let us take the “Yongsheng Ship” that crossed the Arctic for the first time as an example. The icebreaking capacity of the cargo ship is Arc4. Assuming that the cargo ship sails independently, without considering the assistance of other icebreakers, the threshold of sea ice concentration through which the “Yongsheng” can pass is 40%. Therefore, we set the threshold of sea ice concentration at 40%. The algo- rithm of navigable window is as follows: the starting day is the first day when the sea ice concentration at all points of the NEP is less than 40% for 3 consecutive days, and the ending day is the sea ice concentration at all points of the NEP is greater than 40% for 3 consecutive days. Figure 5 shows the overall sea ice concentration in the NEP from 2000 to 2019. It can be seen that the ice segment of the NEP is from 787.03 nautical miles to 3410.46 nautical miles, and there is less than 40% of the ice in the rest of the year. Therefore, we select from 1311.71 nautical miles to 3410.46 nautical miles to study the navigation window of the NEP, and define that when the navigation time is at least 10 days, the NEP can be opened to ensure the practical feasibility of navigation.

Figure 5. The overall sea ice concentration of the NEP from 2000 to 2019; the black triangles indicate the start and end positions of the ice segment.

4.1. Navigation Window from 2000 to 2019 Figure 6 reveals the navigation window information of the NEP from 2000 to 2019. The abscissa is the location of the route, and the ordinate is the number of days in the

J. Mar. Sci. Eng. 2021, 9, 728 current year. Table 1 is the navigation timetable of the NEP from 2000 to 2019, including10 of 26 the start and end dates of the navigation each year and the number of the NEP available each year.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 10 of 27

Figure 6.Figure The start 6. The and start end and time end and time duration and duration of the of navigation the navigation of the of NEP the NEP from from 2000 2000 to 2019. to 2019. The The blue blue area area indicates indicates the the start date of navigationstart date of the of navigationNEP that year, of the the NEP green that area year, indicates the green the area end indicates date, and the the end solid date, black and theline solid indicates black the line navigation indicates the duration of the currentnavigation year. duration of the current year. The navigation window calculated in this study has also been verified by related Table 1. Navigation schedule of the NEP from 2000 to 2019. studies. For example, Xinqing Li et al. [29] analyzed the sea ice variations of the NEP in the summerActual of 2014 Start and foundDate thatActual the openingEnd Date time Actual of the Navigation NEP in 2014 wasNavigation approximately Period Year from earlyof August Navigation to early October.of Navigation It just coincides Period with (Days the actual) Standard navigation Deviation results obtained in this article in 2014. Besides this, according to the report on ships sailing of the 2000 09-10 09-12 3 41.051 NEP from 2013 to 2019 in the International Shipping Network [30], the cargo ships sailing independently2001 in-- 2013, 2015, 2016, 2017-- and 2019 were all0 within the navigation43.078 window (Table2002 2). 08-17 09-19 33 26.219 2003Table 1 reflects-- the actual navigable-- time and standard0 deviation of the38.239 navigable period2004 of the NEP.-- It can be seen from Figure-- 6 and Table1 that0 the actual navigation41.286 period of2005 the NEP has08 a wide-16 range of fluctuations.10-16 The navigation61 start time ranges from19.649 late July to2006 early September,08-14 while the navigation09-23 end time is basically40 mid to late October.21.343 The most feasible2007 month09 for-09 navigation is September,10-17 followed by August.38 The standard24.136 deviation of the navigation period can reflect the degree of dispersion of the navigation period of 2008 08-12 10-05 54 27.383 various locations on the NEP. According to the standard deviation of the navigation period, 2009 08-10 10-09 60 21.393 2010 08-01 10-09 69 17.11 2011 08-16 10-17 62 15.062 2012 07-31 10-21 82 13.895 2013 08-30 09-27 28 33.371 2014 08-12 10-11 60 21.773 2015 07-19 10-13 86 11.223 2016 08-08 10-21 74 14.26 2017 08-13 10-08 56 22.48 2018 09-03 10-13 40 24.777

J. Mar. Sci. Eng. 2021, 9, 728 11 of 26

it can be seen that the navigation time of each position of the NEP in 2000, 2001, 2003 and 2004 is relatively discrete, while the standard deviation of the navigation period in 2012, 2015 and 2019 is relatively small, and the navigation time difference of each position is relatively small. Among them, the navigation period of each location on the NEP is the most concentrated in 2019.

Table 2. Summary of independent sailing cargo ships from 2013 to 2019.

State Year Ship Name Drive-in Time Departure Time Duration 2015 Yongsheng Bering Strait/20150728 Nordkapp/20150808 11 2016 Yongsheng Nordkapp/20160913 Bering Strait/20160922 9 Daan Bering Strait/20170902 Nordkapp/20170909 7 Tianle Nordkapp/20170902 Bering Strait/20170914 12 2017 Tianjian Bering Strait/20170910 Nordkapp/20170920 10 Tianfu Nordkapp/20170909 Bering Strait/20170920 11 Independent sailing Tianlu Bering Strait/20180915 Nordkapp/20180926 11 2018 Tianhui Bering Strait/20180926 Nordkapp/20181009 13 Tianxi Nordkapp/20190826 Bering Strait/20190905 10 Tianyou Bering Strait/20190905 Nordkapp/20190916 11 Tianhui Bering Strait/20190909 Nordkapp/20190920 11 2019 Daxiang Nordkapp/20190906 Bering Strait/20190916 10 Tianqi Nordkapp/20190912 Bering Strait/20190923 11 Tianen Nordkapp/20190921 Bering Strait/20191001 10 Datai Nordkapp/20190924 Bering Strait/20191005 11

In 2000–2019, the navigation period of the NEP in 2000 was only 3 days. Actually, it was not actual to navigate, so it was deemed unavailable. In addition to the year 2000, the NEP could not be opened in 2001, 2003 and 2004. The navigation time in 2012, 2015 and 2019 was longer, all over 80 days. The navigation time in 2002, 2007 and 2013 was shorter. In the non-navigable years, the key areas for navigation in 2000 were 1574 nautical miles and 2886 nautical miles. In 2001, they were 1574 nautical miles and 2623 nautical miles. At 1574 nautical miles is the central area of the East Siberian Sea. The ice concentration is higher than 50%, and it is at a high-risk node for navigation; while the locations of 2623 nautical miles and 2886 nautical miles are located in the eastern area (2623 nautical miles) and western area (2886 nautical miles) of the Vilkitsky Strait, respectively. These are the areas that were not navigable in 2003 and 2004. In the short navigable years, the key area for navigation in 2002, 2007 and 2013 is also the Vilkitsky Strait. For each special year, we use the seasonal average data of wind ﬁeld, temperature, temperature anomalies, sea ice age and sea ice movement in the Arctic summer for analysis. The summer seasonal average maps are used for 2000, 2001, 2003, 2004, 2012, 2015 and 2019 respectively, while the monthly average maps for July, August and September are used for 2002, 2007 and 2013.

4.2. Non-Navigable Years 4.2.1. 2000 and 2001 In 2000 and 2001, the areas where the NEP was not navigable were in the East Siberian Sea (1574 nautical miles) and the Vilkitsky Strait (2623 to 2886 nautical miles). The possibility of the opening of the NEP in these two years is extremely low. In 2000, there was only a three-day opening period, and the actual navigation was inconvenient. In the summer of 2001, the NEP did not open to navigation because of the high concentration of sea ice. According to Figure7a,d, in the summer of 2000, the East Siberian Sea, the Laptev Sea and the Kara Sea were all affected by the northeast wind, while a wind ﬁeld blowing from southwest to northeast was formed in the Barents Sea, and the wind ﬁeld in the Vilkitsky Strait was northerly. This was caused by the combined action of a low-pressure system in the central Arctic and high pressure over the European continent (not shown). In the J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 12 of 27

areas that were not navigable in 2003 and 2004. In the short navigable years, the key area for navigation in 2002, 2007 and 2013 is also the Vilkitsky Strait. For each special year, we use the seasonal average data of wind field, temperature, temperature anomalies, sea ice age and sea ice movement in the Arctic summer for analysis. The summer seasonal average maps are used for 2000, 2001, 2003, 2004, 2012, 2015 and 2019 respectively, while the monthly average maps for July, August and September are used for 2002, 2007 and 2013.

4.2. Non-Navigable Years 4.2.1. 2000 and 2001 In 2000 and 2001, the areas where the NEP was not navigable were in the East Sibe- rian Sea (1574 nautical miles) and the Vilkitsky Strait (2623 to 2886 nautical miles). The possibility of the opening of the NEP in these two years is extremely low. In 2000, there was only a three-day opening period, and the actual navigation was inconvenient. In the summer of 2001, the NEP did not open to navigation because of the high concentration of sea ice. According to Figure 7a,d, in the summer of 2000, the East Siberian Sea, the Laptev J. Mar. Sci. Eng. 2021, 9, 728 Sea and the Kara Sea were all affected by the northeast wind, while a wind field blowing12 of 26 from southwest to northeast was formed in the Barents Sea, and the wind field in the Vilkitsky Strait was northerly. This was caused by the combined action of a low-pressure summersystem ofin 2001,the central the Laptev Arctic Sea,and thehigh Kara pressure Sea andover the the Vilkitsky European Strait continent formed (not a windshown) ﬁeld. blowingIn the summer from northwest of 2001, the to Laptev southeast, Sea, the and Kara the Sea Barents and the Sea Vilkitsky was mainly Strait formed affected a wind by the field blowing from northwest to southeast, and the Barents Sea was mainly affected by southerly wind. This was caused by the combined action of the low-pressure system in the the southerly wind. This was caused by the combined action of the low-pressure system Laptev Sea and the high pressure over the Kara Sea (not shown). in the Laptev Sea and the high pressure over the Kara Sea (not shown).

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 13 of 27

FigureFigure 7. 7.The The average average wind wind ﬁeld field ( a(a),), averageaverage temperaturetemperature ( (bb)) and and temperature temperature anomalies anomalies (c) ( cof) ofthe the NEP NEP in inthe the Arctic Arctic summer (July–September) in 2000. The average wind field (d), average temperature (e) and temperature anomalies (f) of summer (July–September) in 2000. The average wind ﬁeld (d), average temperature (e) and temperature anomalies (f) of the NEP in the Arctic summer in 2001. The red circles indicate the throat areas that affected the navigation of the NEP in thethat NEP year, in the the Arcticsame below. summer in 2001. The red circles indicate the throat areas that affected the navigation of the NEP in that year, the same below. It can be seen from Figure 7b,e that the temperature in the East Siberian Sea and the VilkitskyIt can Strait be seen in 2000 from was Figure mainly7b,e around that the 0 temperature°C, which was in not the conducive East Siberian to the Sea ablation and the ◦ Vilkitskyof sea ice. Strait The intemperature 2000 was mainlyin the Laptev around Sea 0 wasC, which 2 °C , the was temperature not conducive in the to theKara ablation Sea ◦ ofwas sea around ice. The 2 °C temperature and the temperature in the Laptev in the Sea Barents was 2SeaC, was the 0 temperature–4 °C. Compared in the with Kara the Sea ◦ ◦ wasmulti around-year average, 2 C and the thetemperature temperature in the in East the Siberian Barents Sea, Sea the was western 0–4 C.side Compared of the Vilkit- with thesky multi-year Strait, the Laptev average, Sea the and temperature the Kara Sea in was the abnormally East Siberian low Sea, by about the western 2 °C, while side the of the ◦ VilkitskyBarents Sea Strait, had the not Laptev changed Sea much. and theAccording Kara Sea to wasFigure abnormally 7c,f, the temperature low by about of 2theC, East while theSiberian Barents Sea, Sea the had Laptev not changedSea and the much. Vilkitsky According Strait toin Figure2001 was7c,f, around the temperature −5 °C, and ofthe the temperature of the three sea areas was abnormally low compared with the average for many years, which had no effect on the ablation of sea ice. Among them, the temperature in the East Siberian Sea was about 1.2 °C lower, the temperature in the Laptev Sea was 2 °C lower, and the temperature in the eastern side of the Vilkitsky Strait was 0.8–1.6 °C lower. The temperature in the Kara Sea was slightly warmer, between 0–5 °C, and the temperature in the Barents Sea was around 5 °C. There was not much abnormal change in temperature in these two sea areas. It can be seen from Figure 8a,b that in the NEP in the summer of 2000 and 2001, the Bering Strait, the Laptev Sea, the Kara Sea and the Barents Sea were all dominated by one- year ice. However, in these two years, there was sea ice of more than 2 years in the central waters of the East Siberian Sea. Especially in 2000, there was a small amount of sea ice that was more than 5 years old. In the summer of 2000, in the western part of the Vilkitsky Strait, the sea ice was dominated by 2 to 3 years. In 2001, there was a large area of two- year ice in the eastern part of the Vilkitsky Strait. The multi-year ice layer is thick and the whole is firmer, and it is not easy to melt in summer, which easily causes the sea ice concentration here to increase. According to Figure 8c,d, in the summer of 2000 and 2001, certain sea ice movement occurred in the central waters of the East Siberian Sea. It was possible that under the influence of the Arctic wind field, sea ice was pushed to the central waters of the East Siberian Sea to accumulate, leading to a higher sea ice concentration here. There was a certain amount of sea ice movement in the western part of the Vilkitsky Strait in the summer of 2000 and the eastern part of the Vilkitsky Strait in the summer of 2001, which affected the navigation of the Vilkitsky Strait in that year.

J. Mar. Sci. Eng. 2021, 9, 728 13 of 26

East Siberian Sea, the Laptev Sea and the Vilkitsky Strait in 2001 was around −5 ◦C, and the temperature of the three sea areas was abnormally low compared with the average for many years, which had no effect on the ablation of sea ice. Among them, the temperature in the East Siberian Sea was about 1.2 ◦C lower, the temperature in the Laptev Sea was 2 ◦C lower, and the temperature in the eastern side of the Vilkitsky Strait was 0.8–1.6 ◦C lower. The temperature in the Kara Sea was slightly warmer, between 0–5 ◦C, and the temperature in the Barents Sea was around 5 ◦C. There was not much abnormal change in temperature in these two sea areas. It can be seen from Figure8a,b that in the NEP in the summer of 2000 and 2001, the Bering Strait, the Laptev Sea, the Kara Sea and the Barents Sea were all dominated by one-year ice. However, in these two years, there was sea ice of more than 2 years in the central waters of the East Siberian Sea. Especially in 2000, there was a small amount of sea ice that was more than 5 years old. In the summer of 2000, in the western part of the Vilkitsky Strait, the sea ice was dominated by 2 to 3 years. In 2001, there was a large area of two-year ice in the eastern part of the Vilkitsky Strait. The multi-year ice layer is thick and the whole is firmer, and it is not easy to melt in summer, which easily causes the sea ice concentration here to increase. According to Figure8c,d, in the summer of 2000 and 2001, certain sea ice movement occurred in the central waters of the East Siberian Sea. It was possible that under the influence of the Arctic wind field, sea ice was pushed to the central waters of the East Siberian Sea to accumulate, leading to a higher sea ice concentration here. There was a certain amount of sea ice movement in the western part of the Vilkitsky Strait

J. Mar. Sci. Eng. 2021, 9, x FOR PEERin REVIEW the summer of 2000 and the eastern part of the Vilkitsky Strait in the summer14 of of 2001,27 which affected the navigation of the Vilkitsky Strait in that year.

FigureFigure 8. The8. The distribution distribution map map of of sea sea ice ice age age inin thethe NEPNEP in the Arctic summer summer in in 2000 2000 (a ()a and) and 2001 2001 (b (),b and), and the the map map of sea of sea ice movement in the NEP in the Arctic summer in 2000 (c) and 2001 (d). ice movement in the NEP in the Arctic summer in 2000 (c) and 2001 (d). 4.2.2. 2003 and 2004 4.2.2. 2003 and 2004 In the summer of 2003 and 2004, the NEP was closed. The key area for navigation is the InVilkitsky the summer Strait. ofThe 2003 difference and 2004, is that the the NEP western was closed. area of The the keyVilkitsky area forStrait navigation had a ishigher the Vilkitsky sea ice concentration Strait. The difference in 2003, while is that it was the westernthe eastern area area of in the 2004. Vilkitsky Strait had a higherAccording sea ice concentration to Figure 9a, ind, 2003,in the while summer it was of 2003, the eastern the East area Siberian in 2004. Sea, the Laptev Sea,According the Kara Sea to and Figure Vilkitsky9a,d, in Strait the summerwere completely of 2003, controlled the East Siberian by the westerly Sea, the wind, Laptev Sea,while the the Kara Barents Sea andSea was Vilkitsky dominated Strait by were northwest completely winds. controlled This was caused by the by westerly the strong wind, whilelow-pressure the Barents field Sea only was in dominatedthe central region by northwest of the North winds. Pole This in the was summer caused of by 2003 the ( strongnot low-pressureshown). In the ﬁeld summer only inof the2004, central there was region a high of the-pressure North center Pole in in the the summer Kara Sea, of and 2003 the (not shown).pressure In on the the summer Eurasian of continent 2004, there was was relatively a high-pressure low (not shown) center in. Under the Kara this Sea, sea andlevel the pressurepressure on field, the the Eurasian southerly continent wind prevailed was relatively in the East low Si (notberian shown). Sea and Under the Barents this sea Sea level pressurein the summer ﬁeld, the of southerly2004, while wind the prevailednortherly inwind the prevailed East Siberian in the Sea Laptev and the Sea Barents and Kara Sea in theSea, summer and the ofcold 2004, air whilebrought the by northerly the northerly wind wind prevailed slowed in the the ablation Laptev of Sea sea andice. Kara Sea, and the cold air brought by the northerly wind slowed the ablation of sea ice.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 14 of 27

Figure 8. The distribution map of sea ice age in the NEP in the Arctic summer in 2000 (a) and 2001 (b), and the map of sea ice movement in the NEP in the Arctic summer in 2000 (c) and 2001 (d).

4.2.2. 2003 and 2004 In the summer of 2003 and 2004, the NEP was closed. The key area for navigation is the Vilkitsky Strait. The difference is that the western area of the Vilkitsky Strait had a higher sea ice concentration in 2003, while it was the eastern area in 2004. According to Figure 9a,d, in the summer of 2003, the East Siberian Sea, the Laptev Sea, the Kara Sea and Vilkitsky Strait were completely controlled by the westerly wind, while the Barents Sea was dominated by northwest winds. This was caused by the strong low-pressure field only in the central region of the North Pole in the summer of 2003 (not shown). In the summer of 2004, there was a high-pressure center in the Kara Sea, and the pressure on the Eurasian continent was relatively low (not shown). Under this sea level J. Mar. Sci. Eng. 2021, 9, 728 pressure field, the southerly wind prevailed in the East Siberian Sea and the Barents 14Sea of 26 in the summer of 2004, while the northerly wind prevailed in the Laptev Sea and Kara Sea, and the cold air brought by the northerly wind slowed the ablation of sea ice.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 15 of 27

Figure 9. The average wind field (a), average temperature (b) and temperature anomalies (c) of the NEP in the Arctic Figure 9. The average wind field (a), average temperature (b) and temperature anomalies (c) of the NEP in the Arctic summer (July–September) in 2003. The average wind field (d), average temperature (e), and temperature anomalies (f) of summer (July–September) in 2003. The average wind field (d), average temperature (e), and temperature anomalies (f) of the NEP in the Arctic summer in 2004. the NEP in the Arctic summer in 2004. It can also be seen from Figure 9b,e that the temperature in the East Siberian Sea and It can also be seen from Figure9b,e that the temperature in the East Siberian Sea and the Laptev Sea was basically between 0 °C◦ and 5 °C in◦ the summer of 2003, and there was theno Laptev abnormal Sea change was basically in the annual between average 0 Ctemperature and 5 C inof the two summer sea areas. of 2003, The te andmper- there was no abnormal change in the annual average temperature of the two sea areas. The ature in Kara Sea is between −5 °C and 0◦ °C in summer,◦ and that in Barents Sea is between temperature0 °C and 5 °C in. KaraThe temperature Sea is between anomaly−5 Cof andthe two 0 C sea in areas summer, is obviously and that negative. in Barents In Seathe is ◦ ◦ betweensummer 0 ofC 2003, and 5theC. temperature The temperature in the anomalyVilkitsky ofStrait the twowas searelatively areas islow, obviously around negative.−5 °C , Inwhich the summer also showed of 2003, negative the temperature anomaly compared in the Vilkitsky with the annual Strait was average relatively level, which low, around was ◦ −not5 C, conducive which also to the showed ablation negative of sea ice. anomaly In the summer compared of 2004 with (Figure the annual 9c,f), the average tempera- level, whichture in was the not East conducive Siberian Sea, to the the ablation Laptev ofSea sea and ice. the In Vilkitsky the summer Strait of was 2004 relatively(Figure9 lowerc,f), the temperaturethan the annual in the average East Siberian temperature. Sea, the The Laptev temperature Sea and in thethe VilkitskyEast Siberian Strait Sea was was relatively about lower0 °C , thanwhile the the annual temperature average in the temperature. Laptev Sea Thewas temperaturerelatively low, in between the East −5 Siberian and 0 °C Sea, ◦ wasand about the temperature 0 C, while in thethe temperatureVilkitsky Strait in was the betw Lapteveen Sea−5 and was 0 °C relatively. The temperature low, between in −the5 and Kara 0 ◦ SeaC, andhad thenot changed temperature significantly, in the Vilkitsky ranging Straitfrom 0 was °C to between 5 °C , while−5 the and tempera- 0 ◦C. The temperatureture in the Barents in the Kara Sea was Sea hadabnormally not changed higher significantly, than the annual ranging average from ranging 0 ◦C to from 5 ◦C, 5while °C to 10 °C . It can be seen from Figure 10a,b that a large area of 2 to 3 years of ice was both distributed in the western part of the Vilkitsky Strait in the summer of 2003 and the eastern part of the Vilkitsky Strait in 2004. Although there was still more two-year ice in the central part of the East Siberian Sea, the area of two-year ice decreased compared with 2000 and 2001. The Bering Strait, the Laptev Sea and the Barents Sea were all dominated by one- year ice. Due to the low-pressure center of the Arctic in 2003, sea ice was advected by the prevailing northwesterly winds from the Arctic Ocean into the northeastern Kara Sea, accumulating to the west of Vilkitsky Strait and thereby blocking the NEP there (Figure 10c). Similarly, due to the influence of wind, the sea ice of the Arctic Ocean moved and accumulated to the east of the Severnaya Zemlya in 2004, resulting in a higher sea ice concentration there and affecting the opening of the NEP, resulting in a higher sea ice concentration in the eastern part of the Vilkitsky Strait, and affected the opening of the NEP (Fig- ure 10d).

J. Mar. Sci. Eng. 2021, 9, 728 15 of 26

the temperature in the Barents Sea was abnormally higher than the annual average ranging from 5 ◦C to 10 ◦C. It can be seen from Figure 10a,b that a large area of 2 to 3 years of ice was both distributed in the western part of the Vilkitsky Strait in the summer of 2003 and the eastern part of the Vilkitsky Strait in 2004. Although there was still more two-year ice in the central part of the East Siberian Sea, the area of two-year ice decreased compared with 2000 and 2001. The Bering Strait, the Laptev Sea and the Barents Sea were all dominated by one-year ice. Due to the low-pressure center of the Arctic in 2003, sea ice was advected by the prevailing northwesterly winds from the Arctic Ocean into the northeastern Kara Sea, accumulating to the west of Vilkitsky Strait and thereby blocking the NEP there (Figure 10c). Similarly, due to the inﬂuence of wind, the sea ice of the Arctic Ocean moved and accumulated to the east of the Severnaya Zemlya in 2004, resulting in a higher sea ice concentration there and affecting the opening of the NEP, resulting in a higher sea ice J. Mar. Sci. Eng. 2021, 9, x FOR PEERconcentration REVIEW in the eastern part of the Vilkitsky Strait, and affected the opening16 of of 27 the NEP (Figure 10d).

Figure 10. The distribution map of sea ice age in the NEP in the Arctic summer in 2003 (a) and 2004 (b), and the map of Figure 10. The distribution map of sea ice age in the NEP in the Arctic summer in 2003 (a) and 2004 (b), and the map of sea sea ice movement in the NEP in the Arctic summer in 2003 (c) and 2004 (d). ice movement in the NEP in the Arctic summer in 2003 (c) and 2004 (d). 4.3. Short Navigation Years 4.3. Short Navigation Years 4.3.1. 2002 4.3.1. 2002 The navigation period of the NEP of the Arctic was relatively short in 2002, only 33 days.The The navigation main reason period is that of the the navigation NEP of theended Arctic earlier was that relatively year, which short ended in 2002, in mid only- 33September. days. The The main decisive reason area is that affecting the navigation the navigation ended was earlier the western that year, part which of the endedVilkit- in mid-September.sky Strait. From TheFigure decisives 11–13, area it can affecting be seen the that navigation the northwest was wind the western prevailed part in ofthe the VilkitskyVilkitsky Strait. Strait Fromfrom July Figures to September 11–13, it can 2002, be t seenhe sea that ice theage northwest in the western wind part prevailed of the in theVilkitsky Vilkitsky Strait Strait was from basically July 4 to years September in July, and 2002, mainly the sea 2– ice4 years age in in A theugust western to September. part of the VilkitskyThe ice was Strait firm was and basically hard to melt 4 years in summer. in July, and In September mainly 2–4 2002, years the in sea August ice in the to September.western Thepart ice of was the ﬁrmVilkitsky and hardStrait to moved melt in eastward summer. from In September the Arctic 2002,Ocean the into sea the ice northeastern in the western partpart of of the the Vilkitsky Kara Sea ( StraitFigure moved 13c), blocking eastward the from western the e Arcticntrance Ocean of theinto Vilkitsky the northeastern Strait. part of the Kara Sea (Figure 13c), blocking the western entrance of the Vilkitsky Strait.

Figure 11. The average wind field (a), sea ice age (b) and sea ice movement (c) of the NEP in the Arctic in July 2002.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 16 of 27

Figure 10. The distribution map of sea ice age in the NEP in the Arctic summer in 2003 (a) and 2004 (b), and the map of sea ice movement in the NEP in the Arctic summer in 2003 (c) and 2004 (d).

4.3. Short Navigation Years 4.3.1. 2002 The navigation period of the NEP of the Arctic was relatively short in 2002, only 33 days. The main reason is that the navigation ended earlier that year, which ended in mid- September. The decisive area affecting the navigation was the western part of the Vilkit- sky Strait. From Figures 11–13, it can be seen that the northwest wind prevailed in the Vilkitsky Strait from July to September 2002, the sea ice age in the western part of the Vilkitsky Strait was basically 4 years in July, and mainly 2–4 years in August to September. J. Mar. Sci. Eng. 2021, 9, 728 The ice was firm and hard to melt in summer. In September 2002, the sea ice in the western16 of 26 part of the Vilkitsky Strait moved eastward from the Arctic Ocean into the northeastern part of the Kara Sea (Figure 13c), blocking the western entrance of the Vilkitsky Strait.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 17 of 27 FigureFigure 11. 11.The The average average wind wind ﬁeld field (a ),(a sea), sea ice ice age age (b ()b and) and sea sea ice ice movement movement ((cc)) of the NEP NEP in in the the Arctic Arctic in in July July 2002. 2002.

FigureFigure 12. 12.The The average average wind wind ﬁeld field ( a(a),), sea sea ice ice ageage ((bb)) andand sea ice ice movement movement (c (c) )of of the the NEP NEP in in the the Arctic Arctic in inAugust August 2002. 2002.

4.3.2. 2007 The year 2007 was the second lowest September sea ice area within the satellite observation era. However, the navigation time of the NEP in 2007 was relatively short, only 38 days. The reason was that the NEP was opened in September 2007. As can be seen from Figures 14–16, the NEP was continuously affected by northeast winds in July and August 2007. The sea ice moved west/northward under the inﬂuence of the wind ﬁeld. As the sea ice was blocked by the Severnaya Zemlya, it accumulated on its east side, resulting in a high sea ice concentration in this area. According to Figures 14b, 15b and 16b, there was a large area of three-year ice in the west of the Vilkitsky Strait from July to September 2007, which would delay the opening of the NEP. In September, although the NEP was still affected by the northwest wind, the wind decreased. Only the western part of the Vilkitsky Strait had a small amount of sea ice movement in the NEP.

Figure 13. The average wind field (a), sea ice age (b) and sea ice movement (c) of the NEP in the Arctic in September 2002.

4.3.2. 2007 The year 2007 was the second lowest September sea ice area within the satellite observation era. However, the navigation time of the NEP in 2007 was relatively short, only 38 days. The reason was that the NEP was opened in September 2007. As can be seen from Figures 14–16, the NEP was continuously affected by northeast winds in July and August 2007. The sea ice moved west/northward under the influence of the wind field. As the sea ice was blocked by the Severnaya Zemlya, it accumulated on its east side, resulting in a high sea ice concentration in this area. According to Figures 14b, 15b and 16b, there was a large area of three-year ice in the west of the Vilkitsky Strait from July to September 2007, which would delay the opening of the NEP. In September, although the NEP was still affected by the northwest wind, the wind decreased. Only the western part of the Vilkit- sky Strait had a small amount of sea ice movement in the NEP.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 17 of 27

J. Mar. Sci. Eng. 2021, 9, 728 17 of 26

Figure 12. The average wind field (a), sea ice age (b) and sea ice movement (c) of the NEP in the Arctic in August 2002.

J. Mar.Figure Sci. Eng. 13. 2021 The, 9 average, x FOR PEER wind REVIEW field (a ), sea ice age (b) and sea ice movement (c) of the NEP in the Arctic in September 2002.18 of 27 Figure 13. The average wind ﬁeld (a), sea ice age (b) and sea ice movement (c) of the NEP in the Arctic in September 2002. 4.3.2. 2007 The year 2007 was the second lowest September sea ice area within the satellite observation era. However, the navigation time of the NEP in 2007 was relatively short, only 38 days. The reason was that the NEP was opened in September 2007. As can be seen from Figures 14–16, the NEP was continuously affected by northeast winds in July and August 2007. The sea ice moved west/northward under the influence of the wind field. As the sea ice was blocked by the Severnaya Zemlya, it accumulated on its east side, resulting in a high sea ice concentration in this area. According to Figures 14b, 15b and 16b, there was a large area of three-year ice in the west of the Vilkitsky Strait from July to September 2007, which would delay the opening of the NEP. In September, although the NEP was still affected by the northwest wind, the wind decreased. Only the western part of the Vilkit- sky Strait had a small amount of sea ice movement in the NEP.

FigureFigure 14. 14.The The average average wind wind ﬁeld field (a (),a), sea sea ice ice age age ( b(b)) and and seasea iceice movementmovement ( (cc)) of of the the NEP NEP in in the the Arctic Arctic in in July July 2007. 2007.

4.3.3. 2013 The opening of the NEP in 2013 was relatively late, and it was navigable at the end of August. The key area for navigation was the western part of the Vilkitsky Strait. From Figures 17–19, it can be seen that in July and August 2013, the Vilkitsky Strait was affected by a strong westerly wind. Driven by the westerly wind, sea ice moved along the Arctic Ocean to the west of the Severnaya Zemlya, which easily caused the accumulation of sea ice here. In addition, there was a certain amount of three-year ice in the western part of the Vilkitsky Strait in July and August, which would also affect the opening of the NEP. In September, there was little sea ice movement in the Vilkitsky Strait, and the amount of multi-year ice in the Vilkitsky Strait was greatly reduced, which helped the opening of the NEP.

Figure 15. The average wind field (a), sea ice age (b) and sea ice movement (c) of the NEP in the Arctic in August 2007.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 18 of 27

J. Mar. Sci. Eng. 2021, 9, 728 18 of 26

Figure 14. The average wind field (a), sea ice age (b) and sea ice movement (c) of the NEP in the Arctic in July 2007.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 19 of 27 FigureFigure 15. 15.The The average average wind wind ﬁeld field (a (),a), sea sea ice ice age age ( b(b)) andand seasea iceice movement ( (cc)) of of the the NEP NEP in in the the Arctic Arctic in in August August 2007. 2007.

FigureFigure 16. 16.The The average average wind wind ﬁeld field ( a(),a), sea sea ice ice ageage ((bb)) andand sea ice ice movement movement (c (c) )of of the the NEP NEP in in the the Arctic Arctic in inSeptember September 2007. 2007.

4.3.3. 2013 4.4. Long Navigation Years 4.4.1. 2012The opening of the NEP in 2013 was relatively late, and it was navigable at the end of August. The key area for navigation was the western part of the Vilkitsky Strait. From FigureThes 17 area–19, coveredit can be byseen sea that ice in inJuly 2012 and wasAugust 45% 2013, less the than Vilkitsky the average Strait was over affected the past 2 30by years a strong [31]. westerly The minimum wind. Driven extent by of the Arctic westerly sea ice wind, in 2012 sea was ice moved 3.41 million along kmthe Arctic, and the NEPOcean was to openthe west to navigationof the Severnaya for 82 Zemlya, days during which the easily year, caused which the is accumulation the third longest of sea time sinceice here. the openingIn addition, of the there NEP was in a the certain study amount of this of article. three-year ice in the western part of the ItVilkitsky can be seenStrait from in July Figure and 20August,a that, which due to would the low-pressure also affect the system opening (not of shown) the NEP. across theIn September, NEP in the there summer was little of 2012, sea ice northerly movement wind in the prevailed Vilkitsky in Strait, the East and Siberian the amount Sea of and themulti Kara-year Sea, ice while in the theVilkitsky southwest Strait windwas greatly prevailed reduced, in the which Laptev helped Sea the and opening the Barents of the Sea. AccordingNEP. to Figure 20b,c, compared with the annual average, the temperature in the East Siberian Sea, the Laptev Sea and the Barents Sea was not signiﬁcantly abnormal. The temperature in the East Siberian Sea and the Laptev Sea in summer was between 0 and

Figure 17. The average wind field (a), sea ice age (b) and sea ice movement (c) of the NEP in the Arctic in July 2013.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 19 of 27

Figure 16. The average wind field (a), sea ice age (b) and sea ice movement (c) of the NEP in the Arctic in September 2007.

4.3.3. 2013 The opening of the NEP in 2013 was relatively late, and it was navigable at the end J. Mar. Sci. Eng. 2021, 9, 728 19 of 26 of August. The key area for navigation was the western part of the Vilkitsky Strait. From Figures 17–19, it can be seen that in July and August 2013, the Vilkitsky Strait was affected by a strong westerly wind. Driven by the westerly wind, sea ice moved along the Arctic 5 Ocean◦C, and to the the Barents west of Seathe reachedSevernaya 5–10 Zemlya,◦C. The which temperature easily caused on the the east accumulation side of the Karaof sea Sea andice thehere. Vilkitsky In addition, Strait th wasere was between a certain 0 and amount 5 ◦C, whichof three was-year abnormally ice in the highwestern compared part of to thethe multi-year Vilkitsky Strait average, in July promoting and August, the ablationwhich would of sea also ice affect in the the two opening sea areas. of the However, NEP. theIn temperatureSeptember, there in the was NEP little was sea generally ice movement warm in and the theVilkitsky wind wasStrait, light and in the the amount summer of of 2012,multi and-year the ice temperature in the Vilkitsky in the Strait NEP was was greatly about reduced, 0–5 ◦C. which helped the opening of the NEP.

J. Mar.Figure Sci.Figure Eng. 17. 2021 17.The, The9, x average FOR average PEER wind wind REVIEW ﬁeld field ( a(),a), sea sea ice ice age age ((bb)) andand seasea ice movement ( (cc) )of of the the NEP NEP in in the the Arctic Arctic in inJuly July 2013. 2013. 20 of 27

FigureFigure 18. 18.The The average average wind wind ﬁeld field ( a(),a), sea sea ice iceage age ((bb)) andand sea ice movement movement ( (cc) )of of the the NEP NEP in in the the Arctic Arctic in inAugust August 2013. 2013.

Figure 21a reveals that compared to 2000, the amount of multi-year ice in the NEP was substantially lower in 2012, and the ﬁrst-year ice accounted for a larger proportion [32]. In addition, the sea ice in the NEP in summer of 2012 was mostly ﬁrst-year ice, which usually melts completely earlier than thicker multi-year ice. It can be seen from Figure 21b that in the summer of 2012, the sea ice movement along the NEP was relatively slight, and the accumulation was naturally small, which indirectly promoted the navigation of NEP.

Figure 19. The average wind field (a), sea ice age (b) and sea ice movement (c) of the NEP in the Arctic in September 2013.

4.4. Long Navigation Years 4.4.1. 2012 The area covered by sea ice in 2012 was 45% less than the average over the past 30 years [31]. The minimum extent of Arctic sea ice in 2012 was 3.41 million km2, and the NEP was open to navigation for 82 days during the year, which is the third longest time since the opening of the NEP in the study of this article. It can be seen from Figure 20a that, due to the low-pressure system (not shown) across the NEP in the summer of 2012, northerly wind prevailed in the East Siberian Sea and the Kara Sea, while the southwest wind prevailed in the Laptev Sea and the Barents Sea. According to Figure 20b,c, compared with the annual average, the temperature in the East Siberian Sea, the Laptev Sea and the Barents Sea was not significantly abnormal. The temperature in the East Siberian Sea and the Laptev Sea in summer was between 0 and 5

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 20 of 27

J. Mar. Sci. Eng. 2021, 9, 728 20 of 26

Figure 18. The average wind field (a), sea ice age (b) and sea ice movement (c) of the NEP in the Arctic in August 2013.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 21 of 27

°C, and the Barents Sea reached 5–10 °C. The temperature on the east side of the Kara Sea and the Vilkitsky Strait was between 0 and 5 °C, which was abnormally high compared to

the multi-year average, promoting the ablation of sea ice in the two sea areas. However, FigureFigure 19. 19.The The average average wind windthe ﬁeld field temperature (a (),a), sea sea ice ice age agein the ( b(b)) andNEPand seasea was iceice generally movement warm ( (cc)) of of theand the NEP NEPthe inwind in the the Arctic was Arctic light in inSeptember Septemberin the summer 2013. 2013. of 2012, and the temperature in the NEP was about 0–5 °C. 4.4. Long Navigation Years 4.4.1. 2012 The area covered by sea ice in 2012 was 45% less than the average over the past 30 years [31]. The minimum extent of Arctic sea ice in 2012 was 3.41 million km2, and the NEP was open to navigation for 82 days during the year, which is the third longest time since the opening of the NEP in the study of this article. It can be seen from Figure 20a that, due to the low-pressure system (not shown) across the NEP in the summer of 2012, northerly wind prevailed in the East Siberian Sea and the Kara Sea, while the southwest wind prevailed in the Laptev Sea and the Barents Sea. According to Figure 20b,c, compared with the annual average, the temperature in the East Siberian Sea, the Laptev Sea and the Barents Sea was not significantly abnormal. The temperature in the East Siberian Sea and the Laptev Sea in summer was between 0 and 5

FigureFigure 20. 20.The The average average wind wind ﬁeld field ( a(a),), average average temperaturetemperature ( (bb),), and and temperature temperature anomalies anomalies (c) ( cof) ofthe the NEP NEP in the in the Arctic Arctic summer in 2012. summer in 2012. Figure 21a reveals that compared to 2000, the amount of multi-year ice in the NEP 4.4.2.was substantially 2015 lower in 2012, and the first-year ice accounted for a larger proportion [32].The In addition, opening the time sea of ice the in NEPthe NEP in2015 in summer was 86 of days, 2012 thewas secondmostly first longest-year time ice, which since the openingusually melts of the completely NEP from earlier 2000 to than 2019. thicker multi-year ice. It can be seen from Figure 21b that Asin the shown summer in Figureof 2012, 22 thea, sea since ice movement the Arctic along had only the NEP one was high-pressure relatively slight, center and (not shown)the accumulation in Greenland was naturally in 2015, thesmall, East which Siberian indirectly Sea, promoted the Kara the Sea navigation and the Barents of NEP.Sea were prevailing northeasterly winds, while the Laptev Sea was completely easterly winds. According to Figure 22b,c, the temperature of the NEP in 2015 was basically between 0 and 5 ◦C. Compared with the multi-year average, the temperature in the East Siberian Sea and the Kara Sea was abnormally higher by about 1.2 ◦C, while the temperature in the

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 22 of 27

J. Mar. Sci. Eng. 2021, 9, 728 21 of 26

◦ J. Mar. Sci. Eng. 2021, 9, x FOR PEERNovosibirskiye REVIEW Islands was about 0.8 C lower, and the temperature in the other sea22 of areas 27 had little change.

Figure 21. The sea ice age (a) and sea ice movement (b) of the NEP in the Arctic summer in 2012.

4.4.2. 2015 The opening time of the NEP in 2015 was 86 days, the second longest time since the opening of the NEP from 2000 to 2019. As shown in Figure 22a, since the Arctic had only one high-pressure center (not shown) in Greenland in 2015, the East Siberian Sea, the Kara Sea and the Barents Sea were prevailing northeasterly winds, while the Laptev Sea was completely easterly winds. Ac- cording to Figure 22b,c, the temperature of the NEP in 2015 was basically between 0 and 5 °C . Compared with the multi-year average, the temperature in the East Siberian Sea and the Kara Sea was abnormally higher by about 1.2 °C, while the temperature in the Novo- Figure 21. The sea ice age (a) and sea ice movement (b) of the NEP in the Arctic summer in 2012. Figuresibirskiye 21. The Islands sea ice was age about (a) and 0.8°C sea ice lower, movement and the (b )temperature of the NEP in in the the Arctic other summer sea areas in 2012.had little4.4.2 .change. 2015 The opening time of the NEP in 2015 was 86 days, the second longest time since the opening of the NEP from 2000 to 2019. As shown in Figure 22a, since the Arctic had only one high-pressure center (not shown) in Greenland in 2015, the East Siberian Sea, the Kara Sea and the Barents Sea were prevailing northeasterly winds, while the Laptev Sea was completely easterly winds. Ac- cording to Figure 22b,c, the temperature of the NEP in 2015 was basically between 0 and 5 °C . Compared with the multi-year average, the temperature in the East Siberian Sea and the Kara Sea was abnormally higher by about 1.2 °C, while the temperature in the Novo- sibirskiye Islands was about 0.8°C lower, and the temperature in the other sea areas had little change.

Figure 22. The average wind ﬁeld (a), average temperature (b) and temperature anomalies (c) of the NEP in the Arctic summer in 2015.

It can be seen from Figure 23a that the sea ice distributed in the NEP was mostly one-year ice in summer of 2015. Although sea ice movement had occurred along the NEP, it can be seen from Figure 23b that the sea ice movement had a relatively small effect, in particular, in the two key areas of the NEP—the central part of the East Siberian Sea and the Vilkitsky Strait—where sea ice movement was very small, and the possibility of sea ice accumulation was also low.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 23 of 27

Figure 22. The average wind field (a), average temperature (b) and temperature anomalies (c) of the NEP in the Arctic summer in 2015.

It can be seen from Figure 23a that the sea ice distributed in the NEP was mostly one- year ice in summer of 2015. Although sea ice movement had occurred along the NEP, it J. Mar. Sci. Eng. 2021, 9, 728 can be seen from Figure 23b that the sea ice movement had a relatively small effect22, i ofn 26 particular, in the two key areas of the NEP—the central part of the East Siberian Sea and the Vilkitsky Strait—where sea ice movement was very small, and the possibility of sea ice accumulation was also low.

Figure 23. The sea ice age (a) and sea ice movement (b) of the NEP in the Arctic summer in 2015. Figure 23. The sea ice age (a) and sea ice movement (b) of the NEP in the Arctic summer in 2015.

4.4.3.4.4.3 2019. 2019 TheThe global global ocean ocean heat heat capacitycapacity and sea level level had had reached reached new new highs highs in in 2019 2019,, the the sea sea iceice area area was was relatively relatively small, small, andand thethe opening time time of of the the NEP NEP had had reached reached the the longest longest openingopening period period in in history history(87 (87 days).days). ItIt can can be be seen seen from from Figure Figure 2424aa thatthat the northeast wind wind prevailed prevailed in in the the East East Siberian Siberian Sea,Sea, the the Laptev Laptev Sea Sea was was dominated dominated byby the westerly wind, wind, and and the the Kara Kara Sea Sea and and the the Barents Barents SeaSea were were dominated dominated by by the the northwestnorthwest wind in 2019. 2019. Due Due to to the the average average sea sea level level pressure pressure overover the the Arctic Arctic Ocean Ocean waswas muted-theremuted-there was no no strong strong center center of of action action (not (not shown) shown),, the the wind in the NEP was generally weak in 2019, which had little effect on the accumulation wind in the NEP was generally weak in 2019, which had little effect on the accumulation of sea ice. It can be seen from Figure 24b,c that the temperature in the NEP in 2019 was of sea ice. It can be seen from Figure 24b,c that the temperature in the NEP in 2019 was warmer, especially in the Laptev Sea, the Vilkitsky Strait and the Kara Sea, which were warmer, especially in the Laptev Sea, the Vilkitsky Strait and the Kara Sea, which were abnormally 2 °C higher than the annual average. Among them, the temperature in the abnormally 2 ◦C higher than the annual average. Among them, the temperature in the East Siberian Sea was between 0–5 °C , the temperature in the Laptev Sea was even up to East Siberian Sea was between 0–5 ◦C, the temperature in the Laptev Sea was even up to 5 °C , and the temperature in the Vilkitsky Strait, the Kara Sea and the Barents Sea was also 5 ◦C, and the temperature in the Vilkitsky Strait, the Kara Sea and the Barents Sea was also between 0 and 5 °C , which played a role in improving navigation. between 0 and 5 ◦C, which played a role in improving navigation. From the sea ice age map (Figure 25a) and sea ice movement map (Figure 25b) of the NEP in the summer of 2019, it can be seen that the extent of multi-year ice in the Arctic is greatly reduced. The sea ice in the NEP was mostly one-year ice, which is basically seasonal, with thin ice thickness and easy melting in summer. In the summer of 2019, the sea ice movement on the NEP had little effect, which further promoted the opening of the NEP.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 24 of 27

J. Mar. Sci. Eng. 2021, 9, 728 23 of 26 J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 24 of 27

Figure 24. The average wind field (a), average temperature (b) and temperature anomalies (c) of the NEP in the Arctic summer in 2019.

From the sea ice age map (Figure 25a) and sea ice movement map (Figure 25b) of the NEP in the summer of 2019, it can be seen that the extent of multi-year ice in the Arctic is greatly reduced. The sea ice in the NEP was mostly one-year ice, which is basically sea- FigureFigure 24. 24.The The average average wind wind ﬁeld field ( a(),a), average average temperaturetemperature ( b) and temperature temperature anomalies anomalies (c) ( cof) ofthe the NEP NEP in inthe the Arctic Arctic summer in 2019. sonal, with thin ice thickness and easy melting in summer. In the summer of 2019, the sea summer in 2019. ice movement on the NEP had little effect, which further promoted the opening of the NEP.From the sea ice age map (Figure 25a) and sea ice movement map (Figure 25b) of the NEP in the summer of 2019, it can be seen that the extent of multi-year ice in the Arctic is greatly reduced. The sea ice in the NEP was mostly one-year ice, which is basically seasonal, with thin ice thickness and easy melting in summer. In the summer of 2019, the sea ice movement on the NEP had little effect, which further promoted the opening of the NEP.

FigureFigure 25. 25.The The seasea ice age ( a)) and and sea sea ice ice movement movement (b ()b of) of the the NEP NEP in inthe the Arctic Arctic summer summer in 2019. in 2019.

5. Discussion and Conclusions 5.1. Discussion

The sea ice movement of the NEP is due to the action of the wind field, and the wind Figure 25. The sea ice age (a) and sea ice movement (b) of the NEP in the Arctic summer in 2019. field is caused by the pressure gradient force. The cloud cover variability can also have an influence on wind, but cloud cover has a greater thermodynamic impact on sea ice.

In this article, neither cloud cover nor the effect of cloud cover on sea ice ablation has been considered.

J. Mar. Sci. Eng. 2021, 9, 728 24 of 26

5.2. Conclusions Sea ice concentration is a crucial factor that affects the variations in sea ice conditions and navigation conditions. We use the Arctic sea ice concentration data from the summer of 2000–2019, sea ice movement and sea age and for years after 2010, as well as thickness and atmospheric information such as near-surface wind vector and near-surface air tem- peratures, to analyze the sea ice variations and navigation conditions of the NEP and draw the following conclusions: (1) On the whole, the ice segment in the NEP is from 787 nautical miles to 3410 nautical miles. There are perennial ice-free areas at the beginning and end of the NEP. (2) The seasonal characteristics of sea ice concentration along the NEP are very obvious. In spring and summer, the sea ice along the NEP gradually melts. At the end of summer, the sea ice concentration on the NEP reaches a minimum, or even 0. The sea ice concentration on the NEP begins to increase in autumn and reaches its maximum in winter. The NEP has the smallest sea ice concentration in September, followed by August and July. In July, the sea ice concentration is high in each year on the NEP, and by August, the sea ice concentration is greatly reduced. In September, the sea ice in the NEP is sparse, and it is basically in the open sea without ice. In September of 2012, 2015 and 2019, the sea ice concentration of the NEP had approached 0, making it very suitable for navigation. However, in September of 2000, 2001, 2003 and 2004, there was still more sea ice near the East Siberian Sea and the Vilkitsky Strait, and the risk of navigation was high. In October, the sea ice concentration along the NEP began to increase. The opening time of the NEP varies from late July to early September, and the end time of navigation is concentrated in mid-to-late October. The most suitable month for navigation is September. Except for the years when navigation is not available, the navigable duration in summer is basically maintained at more than 30 days, and the navigation duration in the past ten years has reached 60 days. (3) During 2000–2019, the sea ice concentration of the NEP was relatively high in 2000, 2001, 2003 and 2004, and the NEP was not navigable. The sea ice concentration in 2012, 2015 and 2019 was relatively low and the navigation time was longer than 80 days. The opening time of the NEP in 2002, 2007 and 2013 was shorter and did not exceed 38 days. The choke points for navigation of the NEP were the central part of the East Siberian Sea (1574 nautical miles) and the Vilkitsky Strait (2623–2886 nautical miles), where the ice situation is complex and the sea ice concentration is relatively high. Moreover, the inﬂuence of the Vilkitsky Strait on the NEP is greater than that of the central part of the East Siberian Sea. (4) The main reason for the high sea ice concentration in the central part of East Siberian Sea (2000 and 2001) is that the area is the only area distributed with multi-year ice in the NEP in spring and summer, and the wind here is small in summer and the sea ice is not easy to move. Besides this, the temperature at the place is abnormally low in the summer of these two years, which is not conducive to the ablation of sea ice, leading to the high sea ice concentration in this area. (5) The main reason for the high sea ice concentration in the Vilkitsky Strait (2000, 2001, 2003, 2004 and 2002, 2007 and 2013) is the wind ﬁeld and sea ice movement. Here, sea ice is prone to movement under the strong wind in summer. In the Vilkitsky Strait region, this is the persistence of sea ice drift directions favoring blocking either the western or the eastern entries. In addition, there is multi-year ice near the Vilkitsky Strait, which also blocks navigation. (6) The sea ice concentration was low in 2012, 2015 and 2019, and the navigation time of the NEP was long. In these three years, the temperature of the NEP was relatively high. The average temperature in summer was between 0–5 ◦C. The wind was light, which would not promote the movement and accumulation of sea ice, nor cause the choke point of the NEP to be blocked. In addition, the NEP in the summer of these three years was dominated by one-year ice, which was easy to melt in summer, greatly extending the navigation period. J. Mar. Sci. Eng. 2021, 9, 728 25 of 26

Author Contributions: Conceptualization, M.J. and G.L.; Data curation, Y.H.; Investigation, G.L.; Methodology, G.L.; Supervision, T.L.; Validation, Y.H. and Y.L.; Writing—original draft preparation, G.L.; Writing—review and editing, M.J. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the National Natural Science Foundation of China (grant number 41471330); the Special Study on the Third Land Survey in Shandong Province (I) (grant number Y220004202000004_001); the National Science Fund subsidized project (grant number 41976184); and the Major scientific and technological innovation projects in Shandong Province (grant number 2019JZZY020103). The APC was funded by the National Natural Science Foundation of China (grant number 41471330). Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: The sea ice concentration data is from the SSM/I-SSMIS data provided by the National Snow and Ice Data Center (NSIDC): https://nsidc.org/data/nsidc-0051; accessed on 25 November 2020. The wind field maps were acquired by the University of Maine’s reanalysis plotter: https://climatereanalyzer.org/reanalysis/monthly_maps/; accessed on 25 November 2020. The temperature maps and the temperature anomaly maps were obtained from Earth System Research Laboratory: https://psl.noaa.gov/repository/model/compare; accessed on 25 November 2020. The sea ice age data came from NSIDC: https://nsidc.org/data/NSIDC-0611/versions/4; accessed on 25 November 2020. The sea ice motion data came from NSIDC: https://nsidc.org/data/NSIDC-0116 /versions/4; accessed on 25 November 2020. Conflicts of Interest: The authors declare no conflict of interest.

References 1. Serreze, M.C.; Holland, M.M.; Stroeve, J. Perspectives on the Arctic’s Shrinking Sea-Ice Cover. Science 2007, 315, 1533–1536. [CrossRef] 2. Otsuka, N.; Izumiyama, K.; Furuichi, M. Study on feasibility of the northern sea route from recent voyages. In Proceedings of the International Conference on Port and Ocean Engineering under Arctic Conditions (POAC), Espoo, Finland, 9–13 June 2013. 3. Comiso, J.C.; Hall, D.K. Climate trends in the Arctic as observed from space. Wiley Interdiscip. Rev. Clim. Chang. 2014, 5, 389–409. [CrossRef] 4. Stroeve, J.; Holland, M.M.; Meier, W.; Scambos, T.; Serreze, M. Arctic sea ice decline: Faster than forecast. Geophys. Res. Lett. 2007, 34, 09591. [CrossRef] 5. Rothrock, D.A.; Percival, D.B.; Wensnahan, M. The decline in arctic sea-ice thickness: Separating the spatial, annual, and interannual variability in a quarter century of submarine data. J. Geophys. Res. Space Phys. 2008, 113, 05003. [CrossRef] 6. Kwok, R.; Untersteiner, N.; Hafemeister, D.; Kammen, D.; Levi, B.G.; Schwartz, P. The Thinning of Arctic Ice. AIP Conf. Proc. 2011, 1401, 220–231. [CrossRef] 7. Markus, T.; Stroeve, J.C.; Miller, J. Recent changes in Arctic sea ice melt onset, freezeup and melt season length. J. Geophys. Res. Ocean. 2009, 114, C12024. [CrossRef] 8. Intergovernmental Panel on Climate Change. Landmark United in Science Report Informs Climate Action Summit; IPCC: New York, NY, USA, 2019. 9. He, S.; Ping, Y.; Zhang, W. The inﬂuence of the Arctic routes on China’s trade potential: An empirical study based on the stochastic frontier gravity model. Int. Trade Issues 2013, 8, 3–12. [CrossRef] 10. Council, A. Arctic Marine Shipping Assessment Report 2009; Protection of the Arctic Marine Environment Working Group: Ottawa, ON, Canada, 2009. 11. Smith, L.C.; Stephenson, S.R. New Trans-Arctic shipping routes navigable by midcentury. Proc. Natl. Acad. Sci. USA 2013, 110, E1191–E1195. [CrossRef][PubMed] 12. Rodrigues, J. The rapid decline of the sea ice in the Russian Arctic. Cold Reg. Sci. Technol. 2008, 54, 124–142. [CrossRef] 13. Lei, R.; Xie, H.; Wang, J.; Leppäranta, M.; Jónsdóttir, I.; Zhang, Z. Changes in sea ice conditions along the Arctic Northeast Passage from 1979 to 2012. Cold Reg. Sci. Technol. 2015, 119, 132–144. [CrossRef] 14. Gui, D.; Pang, X.; Lei, R.; Zhao, X.; Wang, J. Changes in sea ice kinematics in the Arctic outﬂow region and their asso-ciations with Arctic Northeast Passage accessibility. Acta Oceanol. Sin. 2019, 38, 101–110. [CrossRef] 15. Itkin, P.; Krumpen, T. Winter sea ice export from the Laptev Sea preconditions the local summer sea ice cover and fast ice decay. Cryosphere 2017, 11, 2383–2391. [CrossRef] 16. Chen, J.; Kang, S.; Chen, C.; You, Q.; Du, W.; Xu, M.; Zhong, X.; Zhang, W.; Chen, J. Changes in sea ice and future accessibility along the Arctic Northeast Passage. Glob. Planet. Chang. 2020, 195, 103319. [CrossRef] J. Mar. Sci. Eng. 2021, 9, 728 26 of 26

17. Li, Z.; Ringsberg, J.W.; Rita, F. A voyage planning tool for ships sailing between Europe and Asia via the Arctic. Ships Offshore Struct. 2020, 15, S10–S19. [CrossRef] 18. Wang, Y.; Liu, K.; Zhang, R.; Qian, L.; Shan, Y. Feasibility of the Northeast Passage: The role of vessel speed, route plan-ning, and icebreaking assistance determined by sea-ice conditions for the container shipping market during 2020–2030. Transp. Res. Part E Logist. Transp. Rev. 2021, 149, 102235. [CrossRef] 19. Yu, M.; Lu, P.; Li, Z.; Li, Z.; Wang, Q.; Cao, X.; Chen, X. Sea ice conditions and navigability through the Northeast Passage in the past 40 years based on remote-sensing data. Int. J. Digit. Earth 2021, 14, 555–574. [CrossRef] 20. Ma, L.; An, L.; Zhang, X.; Zheng, Z.; Li, Z.; Chen, G. The Northeast Arctic Route 2006-2015 Navigation Window Dataset. J. Glob. Chang. Data 2019, 3, 244–251, 354–361. [CrossRef] 21. Li, X.; Ci, T.; Ros, H.; Hui, F.; Cheng, X. Spatial-temporal changes and seaworthiness analysis of sea ice in the Northeast Passage of the Arctic Velikitsky Strait. Polar Res. 2015, 27, 282–288. [CrossRef] 22. Comiso, J.C.; Parkinson, C.; Gersten, R.; Stock, L. Accelerated decline in the Arctic sea ice cover. Geophys. Res. Lett. 2008, 35, 8–13. [CrossRef] 23. National Snow and Ice Data Center. NSIDC. Available online: https://nsidc.org/data/nsidc-0051 (accessed on 25 November 2020). 24. University of Maine’s Reanalysis Plotter. Available online: https://climatereanalyzer.org/reanalysis/monthly_maps/ (accessed on 25 November 2020). 25. Earth System Research Laboratory. ESRL. Available online: https://psl.noaa.gov/repository/model/compare (accessed on 25 November 2020). 26. NSIDC. Available online: https://nsidc.org/data/NSIDC-0611/versions/4 (accessed on 25 November 2020). 27. NSIDC. Available online: https://nsidc.org/data/NSIDC-0116/versions/4 (accessed on 25 November 2020). 28. Shibata, H.; Izumiyama, K.; Tateyama, K.; Enomoto, H.; Takahashi, S. Sea-ice coverage variability on the Northern Sea Routes, 1980–2011. Ann. Glaciol. 2013, 54, 139–148. [CrossRef] 29. Li, X.; Cheng, X.; Hui, F.; Zhai, M.; Zhang, Y. Analysis of ice conditions in the Northeast Passage of the Arctic in summer 2014. Polar Res. 2016, 28, 87–94. [CrossRef] 30. International Shipping Network. Available online: http://www.eworldship.com (accessed on 25 November 2020). 31. Perovich, D.K.; Grenfell, T.C.; Richter-Menge, J.A.; Light, B.; Tucker, W.B.; Eicken, H. Thin and thinner: Sea ice mass balance measurements during SHEBA. J. Geophys. Res. Space Phys. 2003, 108, 8050. [CrossRef] 32. Polyakov, I.V.; Walsh, J.E.; Kwok, R. Recent Changes of Arctic Multiyear Sea Ice Coverage and the Likely Causes. Bull. Am. Meteorol. Soc. 2012, 93, 145–151. [CrossRef]