Impact turns + answers – bfhmrs russia War Good
Climate change leads to increased access to arctic shipping routes – most recent models
Download 0.83 Mb.
|
Harbor Teacher Prep-subingsubing-Ho-Neg-Lamdl T1-Round3, Impact Turns Aff Neg - Michigan7 2019 BFHMRS
Climate change leads to increased access to arctic shipping routes – most recent models.Ng et al. 18 [Adolf K. Y., Professor of Transportation and Supply Chain Management at the Asper School of Business of the University of Manitoba, Jonathan Andrews, Centre for Earth Observation Science, University of Manitoba, David Babb, Associate Teaching Professor of Meteorology at PennState, Yufeng Lin, Research Assistant, University of Manitoba, PhD, University of Rhode Island, Department of Marine Affairs, “Implications of climate change for shipping: Opening the Arctic seas,” accessible online at https://onlinelibrary.wiley.com/doi/abs/10.1002/wcc.507, published 02/08/18] // BBM Climate change presents many impacts for human activities, including shipping. With more than 80% of globally traded cargo carried by ships (Ng & Liu, 2014), shipping and the maritime industry play an important role in the well-being of global and regional economies. A considerable volume of research investigates the negative impacts posed by climate change on transportation infrastructures. For example, the book by Ng et al. (2016) contains numerous cases on how climate change affects the operation of ports located in four continents, and how port stakeholders attempt to adapt to such challenges. Climate change impacts will likely have both positive and negative elements. In fact, some societies may benefit from the development and implementation of effective adaptation strategies and solutions (Ng et al., 2016). The United Nations Framework Convention on Climate Change supports this view and defines adaptation to climate change as follows: …Adjustments in ecological, social, or economic systems in response to actual or expected climatic stimuli and their effects or impacts...Changes in processes, practices, and structures to moderate potential damages or to benefit from opportunities associated with climate change. (UNFCCC, 2014) In this regard, Arctic shipping provides an example of potential climate-change related costs and benefits. Reduced ice cover may present opportunities as certain Arctic marine routes that have historically been covered by sea ice become navigable for part of the year. For example, a Russian tanker sailed through the Arctic without icebreaker assistance for the first time in August 2017 (Barkham, 2017). Although Larsen (2016) argues that the economic feasibility of Arctic shipping is underexplored, a growing body of literature is gradually filling the gap. It is not yet clear how climate change and Arctic shipping will impact the environment and indigenous people of the north, but recent research has pointed to a number of significant challenges. Conversely, Arctic “thawing” may also present certain positive implications, such as shorter routes (thus lower transportation costs) and increased accessibility to known resource deposits. If so it could trigger another phase of evolution in the global shipping industry that is characterized by shorter shipping distances through Arctic waters instead of conventional navigation routes via the major canals (e.g., Suez Canal, Panama Canal). This is especially true as the intense competitive landscape pressures cost-sensitive ship operators to reduce costs (and price) even further. There are currently three main trans-Arctic navigation options: the Northern Sea Route (NSR), the Northwest Passage (NWP), and the Trans-Polar Sea Route (TSR). The feasibility and economic viability of these Arctic routes are still under intense debate. Although some research has been undertaken (Cheng & Lee, 2011; Roh, 2011; Xu, Yin, Jia, Jin, & Ouyang, 2011), many studies have only addressed the potential cost savings of using Arctic routes compared to existing trans-continental shipping networks (e.g., between Asia and North America). In fact, many important aspects of Arctic shipping remain under-examined. For example, despite the potential benefits, some still argue there is a lack of real interest in Arctic shipping (e.g., according to Pierre & Olivier, 2015, only 53 transits via the Arctic were recorded in 2013) and that shipping stakeholders have yet to treat the Arctic routes as serious alternatives. Also, the opening of the Arctic waters will pose various environmental, social, and cultural impacts to surrounding regions, including a heightened risk of maritime accidents and environmental harm due to the lack of infrastructural support. Given the complex nature of the Arctic, what is the right approach for further research? This paper examines Arctic shipping through a collaborative effort between researchers from different disciplines. Section 2 discusses the feasibility of shipping in the Arctic marine environment from the physical perspective. It examines the impact of climate change on sea ice and marine weather and considers the resultant consequences for Arctic shipping accessibility. Section 3 discusses the challenges of Arctic shipping from the socio-economic and environmental perspective. It reviews the major research investigating the economic feasibility of diverting ships from conventional shipping routes to the Arctic routes. Also, it reviews the attitudes of shipping stakeholders, as well as the social and other considerations that affect the prospect of Arctic shipping. Finally, Section 4 concludes the paper and identifies the major gaps that need to be addressed in future research. FEASIBILITY OF ARCTIC SHIPPING: PHYSICAL PERSPECTIVE The Arctic marine environment under the influence of climate change In its Fifth Assessment Report (AR5), the Intergovernmental Panel on Climate Change (IPCC) estimated (with high confidence) that between 1880 and 2012 the global mean temperature increased roughly 0.85C (IPCC, 2013). Moreover, temperatures in the polar regions have increased at a significantly greater rate than the lower latitudes due to a number of mechanisms collectively referred to as “Polar Amplification” (ACIA, 2005; Masson-Delmotte et al., 2013). Warming temperatures in the Arctic have driven a rapid decline in sea ice over the past 50 years and have triggered changes in Arctic weather systems (IPCC, 2013), with considerable implications for marine transportation. Sea ice: extent, thickness, timing, and motion Sea ice is the greatest physical constraint on Arctic shipping. Shippers transiting Arctic waters can either avoid sea ice by confining their activities to the seasonal open water periods in certain Arctic regions, or shippers can travel within sea ice using vessels with some measure of ice strengthening. A vessel’s “ice-strengthening” is the product of many components in the vessel’s design (e.g., hull strength, thickness). The world’s shipping fleet contains a broad range of ice-strengthened vessels, varying from “Open Water” (OW) vessels capable of traveling in ice up to 15 cm thick to “Polar Class” vessels that can travel in sea ice several meters thick (Transport Canada, 2010). In recent decades, vessels have been built according to numerous different ice-capacity classification systems, resulting in a patchwork of vessel designs and regulations (Government of Canada, 1985; Guard, 2012). For example, Table 1 shows a reference of suitable ice thickness for various vessel types within the Canadian classification system. It is now recommended that new ice-strengthened vessels be built according to international guidelines set out in the Requirements Concerning Polar Class by the International Association of Classification Societies (IACS) at the behest of the International Maritime Organization (International Association of Classification Societies, 2011; Transport Canada, 2009). Ice-strengthened vessels designed according to an earlier classification system may apply for an equivalency in the IACS system. With respect to shipping, Arctic sea ice is primarily described by referring to ice extent (or area) and ice thickness. It is important to understand the meaning of these variables and how they are observed. Ice extent and area both relate to the geographic distribution of sea ice: ice extent refers to the total area with sea ice concentration above a selected threshold (typically 15%), while ice area refers to the total ocean area actually covered with sea ice (Cavalieri & Parkinson, 2012; Vaughan et al., 2013). Area-based ice measurements are also used to examine the timing of sea ice in areas with non-continuous ice coverage, producing time-related variables such as breakup and freezeup dates, or melt-, open water-, and ice-season length. Satellite-based observations, from both visible and microwave based platforms, provide a continuous data record of the geographic distribution of sea ice, and thus all of the aforementioned variables, for 1979 to the present (Stroeve & Notz, 2015; Vaughan et al., 2013). Ice thickness refers to the thickness of ice (and sometimes snow) between the liquid ocean and the atmosphere. Numerous techniques have been used to measure ice thickness in recent decades, including Upward Looking Sonar (ULS) on submarines and moorings, airborne electromagnetic (EM) instruments, and altimetry from aircraft and satellites (Lindsay & Schweiger, 2015; Stroeve & Notz, 2015). Despite the many techniques, the current record of sea ice thickness remains incomplete, with gaps in spatial and temporal coverage and relatively high uncertainty (Stroeve & Notz, 2015). For example, satellite altimetry observations are limited to the period between November and April and do not provide measurements during spring and summer when the data could be most applicable to Arctic shipping. Arctic sea ice is complex and dynamic, with considerable variation over space and time (i.e., “spatiotemporal variation”). Arctic sea ice extent typically varies by roughly 250% over the course of the year, with a maximum extent in March and a minimum in September (Vaughan et al., 2013); Figure 1 shows the peripheral Arctic seas with seasonal ice coverage that surround the continuously ice-covered Central Arctic. Two of the trans-Arctic shipping routes, the NWP and NSR, pass through multiple peripheral seas, while the projected TSR passes through Fram Strait and the Central Arctic (Figure 1). Like ice extent, ice thickness follows an annual cycle, a product of thermodynamic and dynamic thickening of the ice pack during winter and the spring onset of melt (Landy, Babb, Ehn, Theriault, & Barber, 2016). At any given time, the Arctic ice pack consists of a mixture of first-year ice (FYI)—ice that has not survived one summer and typically grows to a maximum thickness of 2 m, second-year ice (SYI)—ice that has survived one summer, and multiyear ice (MYI)—ice that has survived at least two summers and which varies in thickness from roughly 2–10 m (Haas, Hendricks, Eicken, & Herber, n.d.). Figure 2 shows how older (and thicker) sea ice is concentrated in the central Arctic. Generally, older sea ice is thicker because it has undergone multiple winter growth seasons (Tschudi, Stroeve, & SJ, 2016). Finally, the Arctic ice cover can be further differentiated between “mobile pack ice” that drifts under the influence of surface winds, ocean currents, the Coriolis force, and tides, and “landfast ice” that is immobilized by the shore or seafloor. Taken together, Figures 1 and 2 indicate a central Arctic region with continuous ice-cover and older, thicker ice surrounded by seasonally ice-covered waters. While this has been the case throughout the satellite record (1979–present), the Arctic ice pack is undergoing rapid change. Satellite records clearly display negative trends in ice extent, area, and duration (Comiso, 2012; Stroeve, Markus, Boisvert, Miller, & Barrett, 2014; Vaughan et al., 2013). For example, between 1979 and 2016 the September minimum ice extent declined at a significant rate of −87 200 km2 or −13.3% per decade (National Snow and Ice Data Center, 2016). Moreover, the rate of ice loss has accelerated through this period and record minima in ice extent occurred during 2007 and again in 2012 (Vaughan et al., 2013). For context, the September 2012 record minimum ice extent of 3.41 million km2 was only 54% of the average minimum extent between 1981 and 2010 (Liu, Babanin, Zieger, Young, & Guan, 2016). Underlying the reductions in the spatial extent of sea ice are significant trends towards a thinner and younger ice pack (Comiso, 2012; Kwok & Rothrock, 2009; Lindsay & Schweiger, 2015; Maslanik, Stroeve, Fowler, & Emery, 2011). Why is Arctic ice declining? Average air temperatures are rising in the Arctic, affecting sea ice formation, growth, persistence, and movement throughout the year. Also, warmer temperatures and other climate changes appear to drive a series of “feedback loops” that cause accelerated, nonlinear Arctic ice loss (Masson-Delmotte et al., 2013; Stroeve et al., 2014). The ice-albedo feedback loop, for example, is a major contributor to Arctic ice loss and can be summarized as follows: reduced sea ice coverage increases the proportion of exposed OW, which lowers the surface albedo and increases the absorption of solar radiation in the OW area; in turn, increased solar absorption warms the ocean surface and leads to increased melt of the remaining ice pack, thereby exposing more areas of OW and feeding the cycle (Parkinson, 2014; Serreze & Barry, 2011). The ice-albedo feedback loop works on a seasonal and an inter-annual timescale: warmer surface waters delay fall freeze-up and shorten the ice growth season, thereby preconditioning the ice pack of the subsequent year to be thinner, less extensive and more susceptible to earlier breakup (Serreze & Barry, 2011). Parkinson (2014) used satellite observations to examine the length of the ice- and melt-seasons in the northern hemisphere for 1979–2013. Figure 3, taken from Parkinson (2014), suggests that for both 1979 and 2013 the ice season was longest in the Central Arctic and the Canadian Arctic Archipelago, followed by the Beaufort, East Siberian, and Kara seas, and then by the Baffin Bay and the Chukchi, Barents, and Bering seas. In addition, Figure 3 suggests a change in the length of the ice season between 1979 and 2013. While the central Arctic remains ice covered throughout the year, the summer ice pack is retreating northwards and new or longer OW seasons have been observed in peripheral Arctic seas (Galley et al., 2016; Parkinson, 2014; Serreze, Crawford, Stroeve, Barrett, & Woodgate, 2016). However, regional variability persists within these long-term trends. For example, Melia, Haines, and Hawkins (2016) note that during the then record September sea ice minimum of 2007 the NSR remained impassable because sea ice protruded from the main pack towards Russia. Cavalieri and Parkinson (2012) used the satellite record to examine the change in ice extent and area in the northern hemisphere over 1979–2010. Put briefly, they found significant negative trends in ice extent and area for most regions over annual, seasonal, and monthly time frames (Cavalieri & Parkinson, 2012). For the northern hemisphere as a whole, annual average ice extent declined at a significant rate of −4.1% decade, with seasonal trends ranging from −2.6% in the winter to −7.9% in the summer; different regions exhibited considerable variation in trend magnitude, though nearly all regions displayed significant negative trends (Table 2) (Cavalieri & Parkinson, 2012). The authors also noted that trends for ice extent were often significantly more negative for 1979–2010 versus 1979–2006, indicating an accelerating decline (Cavalieri & Parkinson, 2012). In a related study, Stroeve et al. (2014) used the satellite record to examine changes in the timing of Arctic sea ice between 1979 and 2013. Their results suggest that, for most regions in the northern hemisphere, the arrival of continuous ice melt (“Melt Onset”) became significantly earlier and the arrival of continuous freeze-up (“Freeze Onset”) became significantly later (Table 3) (Stroeve et al., 2014). These results indicate the “Arctic Opening”—the major expansion of OW area and duration of the OW area in the Arctic (Table 3)—which has garnered considerable public interest of late. While it is becoming increasingly possible for OW vessels to operate in the Arctic (e.g., the route to and from the Port of Churchill, Andrews, Babb, McKernan, Horton, & Barber, 2016), a greater area and a longer season are available to vessels with some degree of ice-strengthening. For ice-strengthened vessels, the navigability of ice-covered waters is dependent on ice thickness. As discussed, there are numerous methods for measuring ice thickness and there is no continuous spatiotemporal record. Satellite based observations of ice thickness have the advantage of providing spatially complete and repetitive surveys of the Arctic ice pack, which can provide information on the seasonal thickening of the ice pack over the winter. Figure 4 presents satellite based estimates of ice thickness for 2013 following fall freeze-up (November) and at the end of winter (March) from Ricker et al. (2014); the estimates presented are in rough agreement with those produced by other authors and methods (Haas & Howell, 2015; Lindsay & Schweiger, 2015). Figure 4 indicates that the thickest ice in the Arctic is located in the central Arctic just north of the Canadian Arctic Archipelago and that ice becomes generally thinner with distance away from that location. The observational record of ice thickness is subject to large gaps in both space and time that are underscored by intercomparison issues between observational methods (Stroeve & Notz, 2015). Fortunately, the record of MYI distribution in the Arctic (which can be used as a proxy for thickness) extends back to 1981. Maslanik et al. (2011) and Comiso (2012) present conclusions on ice thickness trends based on examinations of MYI extent: for example, for the period 1981–2011, Comiso (Comiso, 2012) reports a decline of −15.6% and −17.5% in winter (December–February) MYI extent and area, respectively. Maslanik et al. (2011) report that the proportion of March ice extent consisting of MYI declined from 75% in the 1980s to 45% in 2011. On a regional scale, Galley et al. (2016) observed the accelerating replacement of old ice by FYI in the Beaufort Sea during winter. Both Comiso (2012) and Maslanik et al. (2011) conclude that the decline in MYI translates into thinner ice in the Arctic both overall and in several regions. Direct measurements of ice thickness also indicate a declining trend. Lindsay and Schweiger (2015) combined subsurface, airborne, and satellite observations in a comprehensive study of Arctic ice thickness and calculated a decline in average Arctic-basin ice thickness between 2000 and 2012. These results support those of Kwok and Rothrock (2009)) who combined submarine and satellite thickness estimates and reported a decline in ice thickness from 1958 to 1976 (submarine) to 1993–1997 (submarine) to 2003–2008 (satellite). Finally, Tilling, Ridout, Shepherd, and Wingham (2015) used 5 years of Cryosat-2 (satellite) ice thickness observations to highlight the high degree of inter-annual variability in ice thickness. This variability underlies the trend towards thinner sea ice and reflects the multitude of processes that dictate the extent and thickness of the Arctic ice pack (Tilling et al., 2015). While the broad-level concentration, thickness, and age of an ice pack can provide some indication of the pertinent ice conditions for shipping activities, one must be careful not to oversimplify. Dynamic forces or glacial-source “ice hazards” can present considerable risks within an apparently navigable ice pack. For example, dynamically deformed first year ice can reach thicknesses of 5–7 m in Hudson Bay (Landy et al., 2016) and large ridge networks in Hudson Strait can beset ships operating in the area (Mussels, Dawson, & Howell, 2016). Sea ice is dynamically thickened throughout the Arctic but particularly so north of Greenland and the Canadian Archipelago, where the Beaufort Gyre and Transpolar Drift Stream converge and create the thickest sea ice in the world (Bourke & Garrett, 1987; Melling, 2002). Ice hazards can also be created when large glacial ice features known as “ice islands” and smaller icebergs calve from ice shelves and become entrained within the mobile ice pack. For example, ice hazards originating from Ellesmere Island can be transported southwards through the Beaufort and Chukchi Seas and across potential transport routes and known reserves of offshore oil and gas (Barber et al., 2014). The ice pack in the Baffin Bay and the Greenland Sea also contain potentially hazardous ice features of glacial and dynamic origins (Peterson, Prinsenberg, & Pittman, 2009). In contrast, other peripheral seas of the Arctic Ocean, specifically those along the NSR (East Siberian, Laptev, Kara, and Barents Seas) contain very little multiyear sea ice (Figure 4) and very few, if any, glacial ice hazards. The risk posed by sea ice and ice hazards entrained within the ice pack is compounded by ice drift (Barber et al., 2014). Within the Arctic there is a mean field of motion that is highlighted by the clockwise Beaufort gyre in the Western Arctic and the Transpolar Drift Stream, which advects sea ice from the Eastern Arctic across the North Pole and out of the Arctic through the Fram Strait (Colony & Thorndike, 1984). Large-scale surface pressure patterns, ocean currents, the Coriolis force, and an underlying field of internal stresses within the ice pack dictate the mean field of motion (Rigor, Wallace, & Colony, 2002). As a result of declining sea ice extent and thickness, the Arctic ice pack has become more mobile, allowing ice drift speeds to increase and the ice pack to behave more dynamically (Rampal, Weiss, & Marsan, 2009). This may result in a greater number of dynamically created ice hazards drifting at higher speeds, which would be of consequence for Arctic shipping. We should also note the risk posed by large ice hazards that melt out of their surrounding ice pack and persist in regions of otherwise OW, potentially posing a threat for marine operations. On that note, Barber et al. (2014) outlined the need for increased accuracy in forecasting high frequency oscillations and local eddy driven ice motion to assist with the prediction of ice hazard drift. Arctic weather Sea ice is the main determinant of Arctic marine accessibility and thus far Arctic weather has garnered far less research attention. However, there is a growing interest in Arctic weather as sea ice declines and Arctic shipping opportunities increase. Researchers have begun to assemble a climatology of Arctic weather, but complex ice-ocean–atmosphere dynamics, shortage of in situ weather measurements, and rapid climate change in the Arctic make this an extremely difficult task. The following will focus on efforts to characterize Arctic winds, waves, and storms—weather elements particularly relevant to shipping. Most analyses of Arctic winds are at least somewhat dependent on satellite altimetry or reanalysis data (the product of in situ measurements and climate models). This is a result of the relatively low number of weather stations in the Arctic. The different data sources for Arctic winds generally produce similar results, but there is as yet no consensus on the precise character or trends of Arctic winds and estimates carry considerable uncertainty. Arctic winds appear to average roughly 5–8 m per second (m s−1 ) (~10–16 knots) during the shipping season: satellite altimetry-based estimates from Liu et al. (2016) indicate an Arctic-wide summer (August–September) average wind speed of 8 m s−1 , while reanalysis data suggest average summer (July–September) wind speeds clustered around 5–8ms−1 in the “Beaufort-Chukchi-Bering Seas” region (Wang, Feng, Swail, & Cox, 2015). Altimetry data suggest an overall tendency towards lower wind speeds in the Arctic over 1996–2015 (Liu et al., 2016), whereas Stopa, Ardhuin, and Girard-Arduin (2016)) report both subtle positive and negative trends in wind speeds throughout the Arctic between 1992 and 2014 based on reanalysis data (Figure 5). Other research based on reanalysis data indicates moderately rising wind speeds overall since at least 1970 (Hakkinen, Proshutinsky, & Ashik, 2008; Spreen, Kwok, & Menemenlis, 2011; Wang et al., 2015). Collectively, the literature indicates that different Arctic regions exhibit opposing trends for wind speed and that results retain considerable uncertainty (Aksenov et al., 2017; Hakkinen et al., 2008; Spreen et al., 2011). For example, Stopa et al. (2016)) report that average wind speeds are increasing over the Chukchi, Laptev, Kara and Barents Seas and the Baffin Bay, while wind speeds are declining over the central Arctic, and Beaufort and Norwegian Seas (Figure 5). Conversely Spreen et al. (2011) found the strongest positive trends of +1–2% per decade over the central Arctic. Changing sea ice patterns appear to be influencing the wave climate of the Arctic. Simply speaking, reductions in sea ice create larger areas of OW and increase fetch, which is the area of water over which wind can act to create waves (Liu et al., 2016; Thomson & Rogers, 2014). This may have implications for vessels and coastal shipping infrastructures, which could be exposed to new, more energetic wave regimes and more rapid coastal erosion (Barnhart, Overeem, & Anderson, 2014; Thomson & Rogers, 2014; Vermaire et al., 2013). Liu et al. (2016) report relatively few significant trends in wave height for the Arctic between 1996 and 2015, whereas Stopa et al. (2016) found an increase in wave heights in much of the Arctic between 1992 and 2014 (Figure 5). In fact, Stopa et al. (2016) found that the only coherent statistically significant negative trend in wave height was observed in the Nordic Greenland Sea. They also reported that anomalously high wave height trends in places like the Beaufort and East Siberian Seas were driven by the variability in the location of the summer marginal ice zone (Stopa et al., 2016). On a regional level, Francis, Panteleev, & Atkinson, 2011) and Stopa et al. (2016) found a significant increase in wave heights of 1.5–2 cm yr−1 in the Chukchi Sea that was correlated with gradual ice retreat. Thomson and Rogers (2014) discussed the role of a reduced ice cover on wave heights in the Beaufort Sea, though they focus on the interaction of waves with ice floes in the marginal ice zone. The literature on Arctic weather contains considerable discussion of polar lows, or Arctic storms, but analyses tend to be focused towards the inter-relationships between Arctic storminess and sea ice dynamics or weather in temperate climates. It is difficult to use this research to gauge the potential shipping-related implications of Arctic storminess. On a more general level, research suggests that Arctic storm frequency is increasing and researchers have identified a decreasing tendency in sea level pressure over the Arctic (Sepp & Jaagus, 2011). Reanalysis data suggests a significant increase in cyclonic activity over the Arctic region between 1948 and 2002, driven in part by an increase in the number of pre-formed cyclones entering the region (Hakkinen et al., 2008; Sepp & Jaagus, 2011). The increase in cyclonic activity is most marked for the summer season (Sepp & Jaagus, 2011). Also, reanalysis data suggests that storms in the Arctic region have become stronger and more intense (Sepp & Jaagus, 2011). Projections for Arctic accessibility: Sea ice along trans-Arctic shipping routes The scientific community has developed and refined climate models to predict future climate change and the resultant consequences for Arctic sea ice. Climate models consist of a series of inter-connected mathematical equations that attempt to simulate the complex mechanisms of the earth-atmosphere–ocean system. They are run, or “forced,” using estimates for future atmospheric concentrations of greenhouse gases and aerosols; these estimates are themselves complex entities, based on predictions of climate feedback mechanisms and society’s future emissions. Two series of forcing estimates, or “forcing scenarios,” have been created and adopted by the climate modeling community since 2000: the Special Report on Emissions Scenarios (SRES) of 2000, followed by the Representative Concentration Pathways (RCP) scenarios of 2011 (Cubasch et al., n.d.; IPCC, 2013; Van Vuuren et al., 2011). For the latter, four pathways have been defined according to the radiative forcing of the atmosphere in 2100 relative to 1750 as measured in watts per meter-squared (W m−2 ): RCP 2.6, 4.5, 6.0, and 8.5. When the RCP scenarios were established, RCP 2.6 was considered a “low” forcing scenario, RCP 4.5 and 6.0 were considered “medium,” and RCP 8.5 was considered “high” or “very high” (Cubasch et al., n.d.; Van Vuuren et al., 2011). Since that time, global emissions have exceeded the estimates in RCP 8.5, suggesting this scenario may, in fact, be somewhat conservative as the upper estimate of future radiative forcing by 2100 (Aksenov et al., 2017; Barnhart, Miller, & Overeem, 2016). Because the task of projecting future sea ice conditions is so complex, it is not surprising that current projections contain considerable uncertainty. The uncertainty in sea ice projections arises from uncertainty in emissions, uncertainty within the climate models themselves, and the inherent variability in natural systems such as Arctic sea ice (Melia, Haines, & Hawkins, 2015; Stroeve & Notz, 2015). Researchers generating sea ice projections apply a number of methods in order to minimize or characterize a model’s uncertainty. These include, for example, using large numbers of models and model simulations in order to capture the range of possible future scenarios, and the method of “model weighting,” which involves pre-selecting only those models that meet certain specific criteria (Stroeve & Notz, 2015). Despite these methods, uncertainty remains high for sea ice projections (Figure 6) (Stroeve & Notz, 2015; Swart, Fyfe, Hawkins, Kay, & Jahn, 2015). For example, Jahn, Kay, Holland, & Hall, 2016) estimate that recent projections for the arrival of a sea-ice free Arctic summer (<1 million km2 of sea ice) produced by models from the Coupled Model Intercomparison Project Phase 5 (CMIP5), including most of the leading models in the field, carry a prediction uncertainty of over two decades. Awareness of this uncertainty is important when considering the current projections for sea ice and Arctic navigability. A number of recent articles present sea ice projections that can be used to qualitatively assess the prospects for future Arctic shipping: Barnhart et al. (2016) used the Community Earth System Model-Large Ensemble (CESM-LE) forced with RCP 8.5 to project the future prevalence of OW (ice concentrations below 15%) in the Arctic. The CESM-LE produced 30 separate simulations (or ensembles) based on slight differences in the model’s initial conditions, with a resolution of 100 km (Barnhart et al., 2016). They used the model to compare projected conditions to a simulated “pre-industrial” state, and found that the Arctic OW season will experience considerable growth (Figure 7) (Barnhart et al., 2016): for example, by 2050 the Arctic coastline and much of the Arctic ocean will experience an additional 60 days of OW each year, and many other sites will have more than 100 additional days by that time (Barnhart et al., 2016). Their results also suggest that many regions of the Arctic will have OW for half of the year by mid-century (Figure 7) (Barnhart et al., 2016). Laliberté, Howell, and Kushner (2016) used 42 models and 91 simulations from CMIP5 forced with RCP 8.5 to examine regional variability in sea ice projections. They present projections for the arrival of ice-free conditions (sea ice concentration below 15% in 94% of the region) for the months of June to October (Figure 8). Their results suggest that Hudson Bay and eastern Arctic waters such as the Kara, East-Siberian, Laptev, and Chukchi Seas will experience longer periods of ice-free conditions arriving earlier in the century, while the central Arctic, the Canadian Arctic Archipelago, and to a lesser extent the Beaufort Sea will be slower to become ice-free during June to October (Laliberté et al., 2016). Khon, Mokhov, and Semenov (2017) used projections from five CMIP5 models (selected via model weighting) forced with RCP 4.5 and 8.5 to examine the evolution of navigability along a pre-defined NSR in the Eastern Arctic. They applied a 15% ice concentration threshold to define “open water” and concluded that end-to-end OW conditions (80–90% OW) have emerged on the NSR in recent decades, and that these conditions may occur for as many as 77–78 days in 2016–2025, 101–118 days in 2046–2055, and 125–192 days in 2090–2099 (Khon et al., 2017). When considering the shipping implications of the sea ice projections in the articles discussed above, it is important to distinguish between sea ice extent or concentration and sea ice thickness. Because the historical record of ice thickness is less rigorous than the record for ice area, efforts to project future Arctic sea ice often focus on ice area. To wit, the above articles present projections in the form of concentration or extent; however, it is sea ice thickness that often directly determines whether an ice-covered region is navigable (Table 1) (Transport Canada, 2010). Furthermore, areas with “open water” may still contain small pieces of thick ice that could present a hazard. Some recent research takes a more direct and quantitative approach than the above articles in assessing future Arctic navigability. The authors of the following articles have directly modeled the future navigability of Arctic regions or trade routes by applying sea ice projections to transport models: Melia et al. (2016) modeled the future navigability in the regions of three Arctic shipping routes, the NWP, the NSR, and TSR, by applying detailed sea ice projections to a shipping model. Sea ice simulations were obtained from a number of bias-corrected CMIP5 models selected via model weighting; each model had multiple ensembles and models were forced with RCP 2.6, 4.5, and 8.5 (Melia et al., 2016). Their shipping model simulated the fastest shipping route from either Rotterdam or New York, to Yokohama for Polar Class 6 (PC6—suitable for ice up to 1.2 m thick) and OW (suitable for ice up to 15 cm thick) vessels based on projections for sea ice thickness and the navigability regulations of the Canadian Arctic Ice Regime Shipping System (AIRSS). Amongst other results, the authors show that Arctic OW (or PC6) transits are possible for at least 30% (PC6–90%) of Septembers between 2015 and 2030, for at least 58% (PC6–94%) by 2045–2060, and for at least 68% (PC6–97%) by 2075–2090 (Melia et al., 2016); these minima were obtained via RCP 2.6 and are likely highly conservative. Their simulations also show OW vessels using the NSR and NWP for 2015–2030 while PC6 vessels commonly accessed the TSR, and OW vessels accessing the TSR in some years in 2045–2060 under RCP 8.5 (Figure 9) (Melia et al., 2016). In a similar study to Melia et al. (2016), Stephenson and Smith (2015) projected optimal Arctic routes between two North Atlantic ports. They obtained ice projections from a subset of CMIP5 models, selected via model weighting, forced under RCP 4.5 and 8.5; they then applied these ice projections to a transport navigability model based on the Arctic Ice Regime Shipping System. Their results forecast the density of optimal routes in the Arctic for both OW and PC6 vessels traveling between the two North Atlantic ports in the years 2011–2035 and 2036–2060 (Figure 10) (Stephenson & Smith, 2015). These results suggest that even under RCP 4.5 emissions, OW vessels will be able to access the TSR by 2036–2060 (Stephenson & Smith, 2015). Download 0.83 Mb. Share with your friends:
|