The ice-albedo mechanism, R1 (reinforcing, regional, well-established):
This reinforcing feedback mechanism is vital in maintaining the polar regime. The white surface of ice keeps the air temperature low and favours year-round ice cover due to high solar reflectivity. Summer melting and high precipitation rates in August, keep the ice thick in the central Arctic Ocean and in turn ensures that high reflectivity is maintained (Curry et al. 1995).
The wind-ocean mechanism, R2 (reinforcing, regional, well-established):
The large extent of sea ice in the polar regime during summer minimizes the amount of open water exposed to wind stress (Ardyna 2014). Cyclones that originate in open water reduce the ice concentration (Kriegsmann and Brümmer 2014) so minimum ocean-atmosphere interaction favours a polar regime.
Temperate Arctic Ocean
The ice-albedo mechanism, R1 (reinforcing, regional, well-established):
This reinforcing feedback mechanism can work in reverse to maintain the temperate regime. Under this scenario, the surface energy balance is not maintained due to an increase in the concentration of GHG in the atmosphere. An increase in surface air temperatures changes the characteristics of the sea ice. The sea ice cover thins due to advection and more wind turbulence (see below), melt ponds are formed and the lead fraction (amount of refrozen melt ponds) increases. Summer melting is more rapid as a result of the thinner ice, resulting in a larger area of open ocean. This darker surface results in a lower albedo, reflecting less solar radiation back into space further altering the energy balance. Winter freezing is not able to compensate the loss of sea-ice extent and thickness, thus promoting a year-on-year loss of sea ice (Curry et al. 1995).
The wind-ocean mechanism, R2 (reinforcing, regional, well-established):
The large extent of receding sea ice cover during summer has exposed more Arctic water to wind stress and resulted in an increase in stormy days over the last decade (Ardyna 2014). This might reduce the ability to regenerate lost sea ice and act to further reduce ice cover (Kriegsmann and Brümmer 2014). The result is a positive feedback that promotes more storm events.
The ice-sea surface temperature mechanism, R4 (reinforcing, regional, contested/speculative):
To give an example, the major oceanic heat inflow into the Canada Basin is the Pacific Summer Waters (PSW), a body of water that reduces sea-ice formation near coastal areas, strengthens upper ocean circulation, and thus increases heat transport and circulation in the north Arctic Ocean. With more transportation of heat the upper water columns are warmed further (Shimada 2006). This reinforcing feedback mechanism is contested but there is a growing body of evidence that this does apply to other sources of warm-water inflow like the THC in the east of the Arctic.
The Atmosphere-Ocean heat exchange, R3 (reinforcing, regional/global, contested):
The reduction in sea ice increases the ability of the Arctic Ocean to absorb solar radiation due to the darker surface (Curry et al. 1995) and has, in turn, led to a higher SST. The latent SST heat flux at the end of the summer results in warmer air temperatures and thus postpones the annual freeze up (Sirevaag et al. 2011). Because refreezing is delayed, precipitation decreases and thickness is reduced, exposing more ocean to solar absorption next year (Ardyna et al. 2014)
The Subpolar gyre feedback mechanism, R5 (reinforcing, regional, speculative): When the Subpolar gyre east of Southern Greenland in the Iceland Basin retreats it is replaced by the East North Atlantic Water (ENAW), bringing saline, warmer waters into the Arctic and speeding up the melting of ice in the east Arctic Ocean (Hátún et al. 2009; Falk-Petersen et al. 2007). This process flushes more freshwater south and seeks to further reduce the strength of the Subpolar gyre and potentially the THC (Aagaard and Carmarck 1989).
The biotic mechanism, B1, B2, B3, B4 (balancing, local/regional, contested):
Photosynthesis is enhanced when light is able to penetrate the upper water columns (see feedback mechanism R4) and when an increase nutrient upwelling from wind-driven vertical mixing occur (see feedback mechanism R3). These physical processes have increased the amount of double annual blooms in the Arctic Ocean and spread the amount of single annual blooms to previously flat patterns at the sea-ice boundary (Ardyna et al. 2014).
Each trophic level is interconnected via a series of balancing feedback mechanisms, where the abundance at each level depends on access to energy at lower trophic levels and the predation from higher trophic levels; however, the strength of the balancing feedback loops various according to the diversity of species on which they can predate and maintain a sufficient energy level (Bascompte et al. 2005). The amount of previously dominating zooplankton cold-loving species like Calanus hyberboreal and C. glacialis has been decreasing to the benefit of the smaller more warm-loving species C. finamarchicus (Falk-Petersen et al. 2007). The concentration of energy in the zooplankton constitute a vital role for the maintenance of upper trophic levels of fish species in the Arctic waters, which in turn again is the main source of energy for the sea birds and the mammals, like the aulk. The changes in the composition of the zooplankton species have impacts on the higher trophic levels through the balancing feedback mechanisms favouring some fish species like herring over others (Ibid.).
A northward movement of warm-water-loving species into the Arctic Ocean has been observed as the abiotic conditions change and the lower trophic species’ northward movements causes a rearrangement of the Arctic food web affecting fisheries yields (Wassmann et al. 2011). A large decrease of ice-dependent species has been documented along side with an increase in subarctic species. In the Barent sea an increase in cod recruitments have been observed as well as an increase in jellyfish abundance (ibid.). These observations signal a potential shift from a predator-dominated food web to a planktivore dominated food web (Niiranen et al. 2014), however, whether this will happen depends on the success of smaller cell communities to thrive at the expense of the larger cell communities; where the first mentioned is incapable of sustaining the higher trophic levels (Wassermann et al. 2011). It is currently difficult to predict whether small or larger plankton will be favoured in future.
The acidification mechanism, B5 (reinforcing, regional, contested):
The Arctic Ocean contributes 5-14% of the global balance carbon sinks from atmosphere, thus playing an important role in the global carbon cycle. The uptake of carbon alters the physics, biogeochemical and biological structures of the Arctic Ocean. Although these dynamics are not yet fully understood, zooplankton and other shell-forming species might be reduced due to increases in pH with profound implications on the entire Arctic ecosystem. It is, however, suggested that ocean acidification is seasonally mitigated by increases in primary productivity during summer, creating a balancing feedback mechanism for the biogeochemical changes in the Arctic Ocean (Bates and Mathis 2009).
The fishing mechanism B6 (balancing, regional, contested):
Increases in commercial fishing will affect the population of fishes in the Arctic Ocean. The North East Atlantic cod is an important commercial fish species and has previously been under stress from overfishing (Hjermann et al. 2006). Furthermore, it can be argued that the amount of commercial fishing in the Arctic Ocean depends on the abundance of fish in the area, legitimizing the balancing feedback loop B4. The changes in the foodweb in the Arctic Ocean might be vulnerable to changes in fishing patterns even further pushing a shift to planktivore-dominated foodweb (Niiranen et al. 2014).
Anthropogenic climate and land use change
Stroeve et al. (2007) showed that from 1953 to 2006, Arctic sea ice extent at the end of the melt season in September declined sharply, and while GHG loading represented 33-38% of the observed changes from 1956-2006, this grew to 47-57% for the period 1969-2006. Modelling has also gone on to show that the Arctic will likely be free of summer sea ice in less than 30 years (Wang and Overland 2012). Such research provides strong evidence for the impact of anthropogenic climate change in the highly vulnerable Arctic region.
The Arctic Ocean’s influx of meltwater and nutrient-rich runoff from land use change is transforming the region from an unproductive system to one that is experiencing heightened PP. While this is bringing short-term benefits in terms of PP and higher trophic level abundance, in the long term it is uncertain whether this will be able to abate ocean acidification and the loss of biodiversity that has been observed in the lower latitudes.
Natural modes of variability
There are two pulses of freshwater input into the Arctic Ocean: one in late-May, associated with the onset of riverine discharge, and one in September, from sea-ice melt and river runoff (Shadwick et al. 2013). As sea-ice extent continues to decline and precipitation in the Arctic changes, more freshwater will enter the Arctic Ocean. This increased stratification prevents PP, but as the THC is being driven by lower-latitude warming and the interaction between positive modes of variation and a weak Subpolar gyre, this potential loss has turned into a gain. However, as the NAO is highly sensitive to freshwater releases, it has been suggested that its influence may be lessened in the future (Masson-Delmotte et al. 2012). The NAO is expected to decline in 2025 (Ardyna et al. 2014).
It is well-established that trends during short periods of a decade or so can be dominated by natural variability (Hawkins and Sutton 2009). For instance, the timing of the subpolar bloom does show distinct decadal-scale periodicity, which is found to be correlated with the NAO index (Henson et al. 2009). When these variations are plotted over a longer period of time, they appear as oscillations over the upward trend of global warming.
Physical vertical mixing
A delay in the freeze-up due to poleward moving warm water in the summer and an increase in the exposure of the sea surface to wind stress, has enhanced wind-driven vertical mixing and a second bloom event for waters surrounding Greenland and Siberia (Ardyna et al. 2014). Nutrient input from Siberian land use change may further increase Arctic PP, and therefore bloom events, in the future.
Summary of Drivers
|#||Driver (Name)||Type (Direct, Indirect, Internal, Shock)||Scale (local, regional, global)||Uncertainty (speculative, proposed, well-established)|
|1||Climate change||External, direct, slow (delayed)||Global||Well-established|
|2||Land-use change||External, indirect, slow||Regional||Proposed|
|3||NAO and AMO||Internal, direct, slow (multidecadal)||Regional||Well-established|
|4||Thermohaline conveyor||Internal, direct, slow||Global||Well-established|
|5||Wind-driven vertical mixing||Internal, direct, fast||Regional||Contested|
|6||Acidification||External, direct, slow (delayed)||Global||Contested|
Shift from polar to temperate PP patterns:
- CO2 concentration in the atmosphere: Several tipping points are connected to the concentration of carbon dioxide in the atmosphere, and might occur at different concentration levels. 1) Arctic sea ice and the GIS are tipping points which the IPCC (2014) said should not be crossed. Some have assigned imprecise thresholds to these such as Kriegler et al (2009) where they show that between a 2.7°C and 4.7°C warming above preindustrial temperatures there is between a 15% and 90% chance the GIS will tip and the system will collapse. However these type of assessments are infrequent and currently lacking the complexity needed to give more accurate predictions.
- 2) The IPCC also state that tipping points apply to deforestation. 3) The point when the exposure of open water is extended to allow warm air to stir up nutrients and create biannual blooms usually characteristic of temperate zones. 4) The point when the sea ice cover has retreated or thinned enough as to allow light to penetrate the water column and support the Arctic biannual blooms.
- Subtropical water masses in the Arctic sea: The threshold when SST rises and external modes of variability are positive, making the Subpolar gyre weaken and allow subtropical Atlantic water to start flowing into the Arctic sea.
Shift from temperate to polar PP patterns:
- CO2 concentration in the atmosphere: The point when the amount of CO2 is reduced to below a certain threshold, halting a further rise in temperatures and allowing the Arctic ice sheet to recover some of its lost sea ice or stabilise in its reduced state, thereby improving albedo reflectivity and reducing storminess.
- Strong subpolar gyre: The threshold when SST is lowered and external modes of variability are negative, making the gyre stronger and inhibiting the northward migration of subtropical Atlantic waters.
- The THC: In addition to the dynamics surrounding the subpolar gyre, there might be tipping points in other natural modes of variability influencing the Arctic Ocean. One example is the potential collapse of THC, which has been shown in model simulations (Rahmstorf and Ganopolsky 1999), that would entail a European climate significantly colder and less productive than at present or even pre-industrial times.
The Arctic climate and its PP patterns during the last few decades have undergone vast and potentially irreversible changes (Trembley et al., 2011; Ardyna et al., 2014; Lenaerts et al. 2013). As such, management needs to emphasis adaptation over mitigation in the immediate term. On a longer time-scale, it could theoretically be possible to enhance resilience and revert the regime shift through the following actions:
- Cutting anthropogenic emissions (global, well-established): Anthropogenic emissions are a significant driver behind this regime shift (Dicks et al. 2011) and reducing them is of paramount importance to prevent sea-ice tipping points. Year-round ice cover helps to sustain the albedo effect (Dicks et al. 2011) and minimize wind-driven vertical mixing (Ardyna et al. 2014), the two most important dynamics in this system.
- Stopping deforestation (global/regional, well-established): Just as it is important to cut emissions of GHG, it is for the same reasons important to preserve the biosphere’s ability to absorb GHG.
- Geoengineering (local/regional/global, proposed): a highly controversial means of altering the climate, most of the proposed schemes lack economic viability, entail vast ecological risks, are ethically challenging or consist of all three pitfalls. Two schemes with traction are 1) stratospheric aerosols, which would increase the amount of sunlight reflected by the atmosphere, thus counteracting the albedo effect (Heckendorn et al. 2009); and 2) wind-power driven sea-ice spraying, whereby the polar ice sheet is artificially thickened by pumping up sea water from below and spraying on top (Zhou and Flynn 2005). Both of these schemes, at present, lack satisfactory cost-benefit analyses and risk assessments.
Ecosystem service impacts
The table below explains the impacts of the potential regime shift to an increase in PP in the context of the Arctic Ocean ES and HWB. Until recently, there has been little research on how changes in the Arctic will impact indigenous communities. This explains why we are uncertain whether changes in PP will have a positive or negative impact on ES and HWB.
As stated below (section 31), the future of fisheries will depend on which species of phytoplankton dominates the system. An increase in picoplankton limits the ability of higher trophic levels to persist as it is nanoplankton that are better suited to transfer carbon to higher trophic levels (Li et al. 2009). This situation could affect all the stakeholders listed in the table below. It is less certain what the impact will be on wild food and products.
In terms of climate regulation, the global direction of change would be negative as the Arctic Ocean could shift from a carbon sink to source, depending on how the dynamics unfold in the future (Bates & Mathis 2009). Nevertheless, more data is needed to understand how ES (such as air quality, climate regulation, water purification and pest & disease regulation) and HWB could be affected by an increase in PP. In addition, more open water would lead to an increase in storm events, which enhance vertical mixing and promote the production of PP, but we speculate that fisheries, both large-scale communities and local communities within the area, could be negatively impacted by the increased frequency and intensity of natural hazards.
Subsistence resource users would be negatively affected by a decline in all the cultural services they currently enjoy. According to the Arctic Council (2013), indigenous communities have a highly adapted way of living and their capacity to respond to changes is therefore limited - they have a low resilience. Furthermore, due to the uniqueness of the Arctic ecosystem, we argue that cultural services, such as cognitive and educational, would also diminish. For non-Arctic residents, we assume that increased PP in the system (as explained in the above sections) and thus the increase of some mammals (eg. whales and seals), will improve tourism and recreation and aesthetic values, depending on the management/policy decisions made.
Summary of Ecosystem Service impacts on different User Groups
||References (if available)|
|Feed, Fuel and Fibre Crops||0||0|
|Fisheries||+/-||?||Li et al. 2009|
|Wild Food & Products||0||?|
|Air Quality Regulation||?||0|
|Climate Regulation||-||0||Bates and Mathis 2009|
|Soil Erosion Regulation||0||0|
|Pest & Disease Regulation||?||?|
|Protection against Natural Hazards||-||?|
|Recreation||+/-||?||Arctic Council 2013|
|Aesthetic Values||+/-||?||Arctic Council 2013|
|Cognitive & Educational||-||?||Arctic Council 2013|
|Spiritual & Inspirational||?||?|
Uncertainties and unresolved issues
Climate change is a significant driver of warming in the Arctic Ocean (Dicks et al. 2011). However, it is uncertain to what extent anthropogenic sources are able to influence natural modes of variability and thereby produce highly unpredictable outcomes.
Several parameters, like sea ice cover, river runoff and modes of variability, differ greatly across temporal and spatial scales, making it improbable that the Arctic Ocean will respond homogenously to change. Small differences in stratification, for instance, might favour picoplankton over nanoplankton or vice versa, sending cascading effects up through the food web and eventually impacting ES and HWB. A greater supply of carbon could enhance food web biomass, but as mentioned, a shift to smaller planktonic communities will inhibit this potential as they are a less nutritious food source. More research is needed in the area of food web dynamics in the Arctic Ocean before any concrete conclusions can be drawn.
Increased CO2 concentrations in the atmosphere leads to ocean acidification, but it is uncertain to what degree PP will be able to negate this process. Either way, phytoplankton species are integral to the functioning of the Arctic Ocean biological community and could play a significant role in the global carbon budget.