Skip to main content
European Commission logo
español español
CORDIS - Resultados de investigaciones de la UE
CORDIS
CORDIS Web 30th anniversary CORDIS Web 30th anniversary
Contenido archivado el 2024-06-18

Arctic Tipping Points

Final Report Summary - ATP (Arctic Tipping Points)

Executive Summary:
Roughly 50 % of the Arctic sea ice has already melted. Indications are that we indeed experienced a “tipping point” for the ice cover, in that it will not return to a year-round cover but will change to a purely seasonal cover of first-year ice. We also investigated Arctic ecosystem time series and trajectories. Analyses of these show that the accelerated melting of sea ice in 1996 was preceded by a period with increased fluctuations in sea ice cover as an early warning indicator of a potential tipping point. A drastic change in the bottom fauna community has been observed in two Svalbard fjords, consistent with increasing temperature.
Tipping points were detected in the growth and activity of all plankon communities within the expected range of warming for the Arctic, all falling tinto the rage of 4-6°C. Evidence suggests that enhanced heterotrophic processes combined with decreasing autotrophic processes in a warmer Arctic will result in a shift towards more heterotrophic microbial communities. A tipping point between the Atlantic species C. finmarchicus and the Arctic C. glacialis can be expected between temperatures of 5-6 °C, with major ramifications for the food webs supporting higher trophic levels such as fish.
Positive responses of the vegetation to warming and increased duration of the ice-free period may be applicable as indicators to monitor responses to climate change and tipping points. Early life-histories of marine benthic fauna, expected to be most sensitive to climatic change, show little effect from ocean acidification but significant performance reductions due to increased temperatures. Thus far it is clear that response thresholds are likely to be different even for taxa with similar ecological positions, further complicating predictions of climate impacts and if/where tipping points will be reached.
Increased melting of ice led to significant increase in modelled production for all productive groups, the most dramatic changes taking place for mesozooplankton. Temperature increase, and ice removal lead to a shift in the timing and strength of peak production. Transition to ice-free areas is abrupt and accompanied with large inter-annual variability. The most severe response to climate change predicted by the model is the disappearance of C. glacialis in the northern Barents Sea.
The aquaculture industry will adapt to increased water temperature by reallocating the farms. For fisheries, uncertainty regarding system dynamics includes both ecosystem and market dynamics. Objectives and uncertainty do however change, also due to climate change. It is increasingly important to develop harvest control rules, which are robust and resilient in a situation of great uncertainty, as well as developing system knowledge and clear management objectives. Climate driven changes reduces oil and gas exploration costs, thus making increased activity more likely. The oil and gas industry are able to tackle an increase in ice movement, and changes in ice structure, as a result of new technological improvements. Regional variability in future ice conditions will alter the cost levels of oil and gas activity for various Arctic regions. A robust framework for policy and management must be put in place that addresses the risk of tipping points and associated regime shifts. Existing management regimes in the Arctic are flexible and have a proven capacity to adapt to change. The project is characterised by extensive policy dissemination and outreach activities.

Project Context and Objectives:
Context
The concept that ecosystems do not respond smoothly to changing pressures has existed for the last 30 years, but the analyses of ecological tipping points and regime shifts remain conceptual in nature. In the same period statistical frameworks for analysing such phenomena have been developed in other disciplines, such as climate research and econometrics, and their use in ecology has been promoted recently through the THRESHOLDS integrated project. As a theoretical framework to explain regime shifts improves, it becomes apparent that regime shifts are triggered by exceeding a particular critical threshold in the driving variable. One example is how changes in climate drivers can force marine ecosystems into a new structural and functional state. Regime shifts are often persistent in that reverting the ecosystem to the original state often requires far greater changes in the driver than what prompted the initial shift. In the worst cases, reversion may prove impossible over managerial time scales. Anticipating where these tipping points are is critical in setting targets to conserve marine ecosystems. Unfortunately, most research on regime shifts has been retrospective in nature, focusing on the their detection and analysis after tipping points have been passed. The prediction of ecological thresholds remains elusive, a major bottleneck for the applicability of the concept.

There is mounting evidence that ecosystem response to certain types or magnitudes of extrinsic pressures (climate, human impacts, etc.) is often abrupt and non-linear, leading to a significant reorganization of system properties and processes. These ecosystem changes are known as regime shifts. Such non-linear responses are often initiated by qualitative changes in the structure or function of the ecosystem, and are so fundamental that the impacted ecosystems respond to new pressures in completely different manners than the original ecosystem did. Regime shifts arise, for instance, from the introduction of alien species or the loss of key species in the ecosystems. These changes can result in alterations of the most basic ecosystem parameters, including food-web structure, the flow of organic matter and nutrients through the ecosystem, or the patterns of space occupation, leading to a cascade of changes in the ecosystem. Climate drives both community structure and key organismal functions, so it is hardly surprising that regime shifts identified from marine ecosystems are often linked to climate.

Main objectives
ATP investigated the existence of climate-driven tipping points for key species and ecosystem processes in the Arctic Ocean, in particular its European sector. This was carried out through an analysis of available time-series data and coordinated experimental studies. The experimental evaluations were used to validate the thresholds identified from time-series analysis, and to postulate new climate-driven tipping points. Ecosystem models have tested these, and will help to formulate future trajectories of Arctic marine ecosystems under climate change scenarios that consider the possibilities of tipping points.

Workpackage 2
WP 2 provides the most complete set of data available on the variability of sea ice extent, concentration and thickness in the European Arctic, and of oceanographic parameters such as temperature and salinity profiles, and heat and freshwater fluxes.
It identifies trends and possible abrupt changes in oceanographic parameters, sea ice edge location, ice-covered area and sea ice thickness. Finally, WP2 predicts changes in the structure of the upper ocean and in the sea ice cover of the European Arctic in the forthcoming decades.

Workpackage 3
WP3 uses new statistical and analytical tools developed within the framework of the THRESHOLDS integrated project to extract regime shifts and tipping points from historical time series of ecological properties of the Arctic marine ecosystem. The possible causes of the regime shifts and tipping points identified will be ascertained by examining their consistency with climatic and anthropogenic forcing, as represented by time series of climate (collated in WP2) and human pressure (fisheries and gas and oil extraction derived from WP6) on the Arctic marine environment. The results will be transferred to WP5 to investigate the stability of regimes for the present and projected future climatic conditions.

Workpackage 4
WP4 supplies WP5 with climatic thresholds and tipping points for key Arctic ecosystem components and processes, as well as to validate those identified in WP3. Ecological modelling and experimentation work best in an iterative manner, where each informs subsequent activities by the other. Therefore, experiments to be conducted under WP4 will also be used to test climatic thresholds and tipping points hypothesised from model outputs (WP5).

Workpackage 5
WP5 aims at projecting the likely future trajectories of Arctic Marine Ecosystems. It runs the well-established coupled hydrodynamics-ecosystem model (SINMOD) with projected climatic forcing, and implementing the mechanisms leading to abrupt changes derived from the time-series and experimental analyses derived from WP3 and WP4, respectively, into the model. The ice-hydrodynamic model will be validated based on data from WP 2.
For the climatic forcing, high resolved regional climate simulations are applied (the regional climate model REMO). A control period of present/past climate (1950 to 2000) as well as two future climate scenarios for the period 2001 to 2100 (SRES A1B and B1) is available on a horizontal grid resolution of 0.22º (~25km). The incorporation of results from WP2, WP3 and WP4 into SINMOD will allow the development of future trajectories for Arctic marine ecosystems and formulate projections on when, under different scenarios. WP5 will find tipping points in the present ecosystem model using a 1D model, clarify the sensitivity of the ecosystem variables to climatic drivers using a 3D coupled model, identify when and where ecosystem tipping points may occur by using atmospheric input from climate model projections.

Workpackage 6
WP6 test the performance of different harvest control rules in fisheries under varying environmental conditions. It models optimal oil and gas exploitation strategies under uncertain prices and weather conditions. Finally, WP6 examines how institutions and policies for the management of living marine resources, tourism and petroleum development can cope with very rapid change in ecosystems driven by climate change.

Workpackage 7
WP7 aims at providing policy makers, managers, stake holders and the general public with an understanding of the ecological thresholds and regime shifts that may develop in the Arctic in response to climate change, and how the ecosystems will respond to EU targets for emissions. This information can then be used as a basis to refine policy targets, mitigate these impacts, take advantage of natural resilience already within the ecosystem, and to identify ways of promoting recovery. WP7 will inform the general public of the possible consequences of climate change on the Arctic ecosystem to help build support for policy frameworks reacting to these predictions.

Project Results:
Main Science and Technology results

1 Arctic climate changes and future projections and scenarios

Summary of progress towards objectives
We provide the most complete set of data available on the variability of sea ice extent, concentration and thickness in the European Arctic. We also identify trends and possible abrupt changes in oceanographic parameters, sea ice edge location, ice-covered area and sea ice thickness. Finally, we predict changes in the structure of the upper ocean and in the sea ice cover of the European Arctic in the forthcoming decades.

The Arctic is warming
Temperature records dating from pre-industrial times provide unambiguous evidence of the warming of our planet. Time-series of the global average temperature for the 1850-2010 period show that not only has the Earth been warming almost constantly since the beginning of the 20th century, the temperatures are now increasing faster than ever before. The average temperature of the Earth went up by 0.02°C/decade during the 1850-1949 period, at a rate of 0.12°C/decade from 1950 to 2010, and by 0.18°C/decade between 1991 and 2011.
All climate models selected by IPCC predict a continued warming of our planet. For the next two decades a warming of about 0.2°C/decade is projected for most emission scenarios but beyond 2030 the forecasts become strongly dependent on the scenario. In B1 the global average temperature is projected to rise by 1.1-2.9°C (average between models 1.8°C) until the end of the century. In A1B and A2 the increases are in the ranges 1.7-4.4°C (average 2.8°C) and 2.0-5.4°C (average 3.4°C), respectively. These variations are quoted with respect to the global average temperature during the 1980-1999 period.
All models predict that the Arctic amplification will persist in the 21st century but there are different estimates for its magnitude. Their average indicates that the Arctic will be warming at about twice the global average rate. The predictions for warming with the A1B scenario range from 2.8 to 7.8°C, with a median of 4.9°C. With B1 (A2) the median becomes 3.4°C (5.9°C). The warming will be much higher in the winter, with temperature increases in the interval 4.3-11.4°C (median 6.9°C) with the A1B scenario. In the summer, with the same scenario, the expected variation is between 1.2 and 5.3°C (median 2.1°C). A particularly high rise in temperatures is expected in the Central Arctic, with warming around 5-7°C in A1B. But the biggest changes are projected for the Barents Sea though this may simply be a consequence of the inability of the models to reproduce the temperatures and the ice cover of the last decades there.

Changes in the Arctic Ocean
The oceans are getting warmer, yet, in general, more slowly than the atmosphere. The average ocean temperature rose at a pace of 0.02°C/decade between 1850 and 1949, then by 0.10°C/decade from 1950 to 2010 and, more recently, between 1990 and 2009, at an approximate rate of 0.15°C/decade. The Arctic Ocean is likely to be warming faster than the other oceans because of the greater increase in air temperatures at high latitudes and the strengthening of the incoming heat flux through Fram Strait and Bering Strait.
Observations show an increase of about 1°C in the temperature of the West Spitsbergen Current between 1997 and 2007. While this warming had an impact in the retreat of the ice edge in the European Arctic, its contribution to the decline of the volume of ice in other parts of the Arctic Ocean is unclear.
Ice and freshwater leave the Arctic Ocean through Fram Strait but there is some controversy about the recent variations in ice export. The decrease in sea ice export through Fram Strait is likely to have contributed to the observed increase of 20% in the freshwater content of the upper layer of the Arctic Ocean between the 1990s and the 2006-2008 period. The freshening of the Arctic Ocean was caused by a combination of the increase in runoff from the gigantic Siberian rivers, the significant acceleration in the melting of Arctic glaciers and the increase in precipitation.

Retreating of the sea ice
The sea ice edge has been retreating northwards since the late 1970s, faster in the summer than in the winter. The current value of the monthly averaged Arctic ice extent for the month of September is about two thirds of the typical value of the late 1970s. A linear fit to the time series for the period between 2001 and 2011 would lead to a rate of decline of 200,000 km2/year, about four times the value for the 1979-2000 period.
The length of the ice-free season can also be taken as a good indicator of the disappearance of the Arctic sea ice. Its average increase during the 1979-2006 period was 1.1 days/year while in the 2001-2007 period it was about five times higher. Two regions where the situation has been changing faster are the Barents Sea and the Greenland Sea. In the eastern (northern) sector of the Barents Sea the rate of increase was 2.9 days/year (3.0 days/year) in the 1979-2008 period and 10.0 days/year (18.5 days/year) in the 2001-2007 period. In the Greenland Sea the growth was 1.7 and 2.8 days/year, respectively.
These well-documented changes, which culminated with the record minimum of September 2007, are the combined effect of several factors. The increase in air temperatures delay ice formation in the autumn and favour surface melting in the summer. The advection of anomalously high ocean heat into the Arctic Ocean through Bering Strait led to enhanced melt from below in the summer, particularly in the Beaufort and Chukchi Seas in the summer of 2007. A high influx of warm Atlantic water entering the Arctic through Fram Strait may have had a similar effect in the European sector.
But perhaps the most interesting feature of our new Arctic is the persistence and strength of a dipole configuration formed by a high-pressure centre over the Beaufort Sea and a low-pressure centre over Siberia. This pressure field leads to low cloudiness, brings warm air (and water) from the Pacific and favours a rapid transport of ice from eastern Siberia towards Fram Strait. This recent acceleration of the Transpolar Drift was confirmed by the unexpectedly fast transit of the schooner Tara. Instead of the three years taken by Nansen’s Fram at the end of the 19th century, it took just 17 months.
The projections of most global climate models taken into account for an ice-free Arctic Ocean (at the end of the melt season) some time in the late 21st century. A much faster disappearance of the Arctic sea ice, conceivably as early as 2040 after identifying several periods of abrupt ice loss in simulations. Shortly after the possibility was pout forward of an ice-free Arctic during the summer in the next few years.

Thinning of the sea ice
Data obtained in the two most recent voyages of the submarine HMS Tireless, in the winters of 2004 and 2007, indicates that, in spite of the evidence from other sources that sea ice is thinning fast almost everywhere in the Arctic, this may not the case in Fram Strait and north of Greenland. The former is an area of very dynamic and heterogeneous ice conditions where the situation can vary significantly from one year to the next and from point to point. The latter contains some of the thickest multi-year ice in the Arctic and is predicted to be where the ice will be most resistant to melting.
Launched in January 2003, NASA’s ICESat measured sea ice freeboard between 2003 and 2009. It is concluded that the average ice thickness in the Arctic was reduced by 25% in three years, from 3.25m in the winter of 2005 to 2.45m in the winter of 2008. They also found that in the winter of 2006 the portion of the Arctic Ocean covered by first-year ice was higher than that covered by multi-year ice for the first time; That in the winter of 2007 the volume of first-year ice was higher than the volume of multi-year ice for the first time; and that in 2008 the volume of first-year ice was about twice the volume of multi-year ice. Thus, the thinning of the Arctic sea ice is above all a consequence of the transformation of a multi-year ice dominated Arctic Ocean into a first-year ice dominated one.

Significant results
• It is estimated that the average temperature of the area north of 65°N increased by 1.1°C in the past 50 years, in contrast to the global average of 0.6°C.
• All climate models selected by the IPCC predict a continued warming of our planet. For the next two decades a warming of about 0.2°C/decade is projected for most emission scenarios, but beyond 2030 the forecasts become strongly dependent on the assumed economic scenario.
• All models predict that the Arctic amplification will persist in the 21st century.
• The current value of the monthly averaged Arctic ice extent for the month of September is about two thirds of the typical value of the late 1970s. A linear fit to the time series for the period between 2001 and 2011 would lead to a rate of decline of 200,000 km2/year, about four times the value for the 1979-2000 period.
• The length of the ice-free season can also be taken as a good indicator of the disappearance of the Arctic sea ice. Its average increase during the 1979-2006 period was 1.1 days/year while in the 2001-2007 period it was about five times higher.
• There was a drop of more than 40% in sea ice thickness between the mid-1970s and the mid-1990s. Thus roughly 50 % of the Arctic sea ice b has already melted.
• The average ice thickness in the Arctic was reduced by 25% in three years, from 3.25m in the winter of 2005 to 2.45m in the winter of 2008.
• The thinning of the Arctic sea ice is above all a consequence of the transformation of a multi-year ice dominated Arctic Ocean into a first-year ice dominated one.
• The indications are that we indeed experienced a tipping point for the ice cover, in that it will not return to a year-round cover but will change to a purely seasonal cover of first-year ice, as currently found in the Antarctic.

2 Extraction of Arctic regime shifts and tipping points from time series records

Summary of progress towards objectives
We employed new statistical and analytical tools to analyse data sets from Arctic marine ecosystems. The ultimate aim was to identify relationships between ecological indicators and climate changes as well as human pressures in the Arctic marine ecosystems. The time series have been investigated for abrupt changes indicative of tipping points in responses to climate change and human pressure as part of an exploratory analysis. Biological responses to changing climatic conditions were examined for those data sets allowing such analyses. Several significant relationships to temperature were identified, but the majority of these are gradual rather than abrupt.

Loss of Arctic sea ice
The most drastic change for Arctic ecosystems is probably the loss of sea ice that even during the extreme negative phase of the Arctic Oscillation in 2009/2010 had a record year low ice extent. An abrupt change in the sea ice extent has been reported to have occurred in 1996, suggesting that the loss rate of sea ice increased by factor of four, although the exact timing of the change was not identified and tested by means of statistical methods.
A subset of the above data was examined for the existence of tipping points using several simple statistical models (Fig. 2.1.). The application of the statistical methods for detecting tipping points was exemplified using data on the Arctic sea ice extent downloaded from the National Snow and Ice Data Center. We used the annual extent in September (n=22, 1979-2010), which is the month with the minimum extent. While emphasizing that changes in the standard error should be considered in addition to changes in the mean, these analysis confirm that the loss of ice accelerated by a factor of 5 in 1996. Increases in random fluctuations were already observed in 1990, thus providing an early warning signal.

Regime shifts in functional relationships
The objective of this task is to investigate and model relationships between climate forcing and ecological responses. The most obvious climate forcing is temperature, however, changes in water masses, indicated through salinity and temperature signals, may also be important governing mechanisms underlying biological responses.

Macroalgae
An updated data set from Kongsfjorden was analysed together a similar data set from Smeerenburgfjord, Svalbard, revealing abrupt shifts in macroalgal cover in 1994 and 2000, respectively. Macroalgal cover increased from 8% to 40% in Kongsfjorden and from 3% to 26% in Smeerenburgfjord, most likely as a positive physiological response to increasing temperatures and enhanced light availability due to an extension of the sea ice free period. Temperature and the number of sea ice free days increased gradually during the period, suggesting a tipping point mechanism in the abrupt change in macroalgal cover.

Mesozooplankton
A 9-year zooplankton data set from the Nordic Seas, mostly covering the Norwegian Atlantic Current and its prolongation the West Spitsbergen Current, has been analysed. The time series was too short to investigate abrupt changes in Calanus spp. biomass, but relationships between biomass and environmental variables revealed threshold responses for C. glacialis biomass, which decreases drastically at temperatures above 6 °C. This may not change the Calanus distribution in the area with an expected water temperature increase of 2 °C within the present century, but C. glacialis may disappear from the shelf areas around Svalbard, and thus completely from the investigated area, if the temperature increases by 4 °C.

Effects of human pressure
The objective of this activity is to investigate if other human interventions, besides human impact on climate change, are interfering with arctic marine ecosystems. The most important human pressure beyond climate change is the potential overfishing.

Fish communities around Greenland
Recent data (1988-2008) from extensive fish surveys around Greenland have been analysed. The objective was to explore if the recent warming had led to changes in species abundance and focused on 16 abundant demersal fish species, constituting 66% of the total fish catch during the survey period. Preference temperatures of these species, determined from their presence in the trawls, ranged between 2.5 and 4.5 °C. Over the two decades of surveys with the most recent being approximately 1 °C warmer, Northern Wolffish, Atlantic Cod, and Atlantic Halibut have increased their abundances substantially in the 0-400 m depth zone, and in the 400-1000 m depth zone Atlantic Halibut and Greenland Halibut have increased their abundances. Overall, the decadal changes in fish abundances were higher for species with a higher temperature preference. Thus, this study shows that relatively small changes in temperature may precipitate species-specific shifts in abundance.

Possible cascading effects
We could not deliver direct evidence for possible cascading effects since there are no long-term monitoring data sets covering several trophic levels. However, several of the studies carried out have suggested indirect evidence for cascading effects. Temperature increases and reduced ice cover will alter the seasonality of primary production and lead to a shift from ice algae to pelagic phytoplankton. This may consequently induce a shift from C. glacialis to C. finmarchicus because of differences in their life strategies. Temperature increases and the resulting loss of sea ice similarly lead to improved light conditions during larger parts of the year. We have shown that this may in consequence lead to deeper distributions of macroalgae. We have has also documented significant relationships between temperature and certain fish species around Greenland, and although these relationships are not due to a direct causal linkage the underlying trophic cascading effects are still unsettled.

Significant results
• The accelerated melting of sea ice in 1996 was preceded by a period with increased fluctuations in sea ice cover as an early warning indicator of a potential tipping point.
• The responses of the Arctic zooplankton community to temperature are not abrupt, except for Calanus glacialis which abundance is significantly reduced with temperatures above 6 °C.
• A drastic change in the bottom fauna community has been observed in two Svalbard fjords, consistent with increasing temperature.
• Increasing temperature around Greenland has shifted the fish community towards more temperate species.

3 Experimental exploration of climatic tipping points for Arctic marine ecosystem components

Summary of progress towards objectives
We aimed to supply the modellers with climatic thresholds and tipping points for key Arctic ecosystem components and processes, as well as to validate those identified above. Ecological modelling and experimentation work best in an iterative manner, where each informs subsequent activities by the other. Therefore, experiments to be conducted here will also be used to test climatic thresholds and tipping points hypothesised from model outputs.

Experimental evaluation of thresholds and tipping points for planktonic Arctic organisms and processes

Phytoplankton communities and UV impacts
We examined the response of polar phytoplankton from Arctic Ocean to increasing seawater temperature, under the exposure of natural levels of UVR, and to test whether there is a synergetic effect between temperature and UVR that amplify the phytoplankton responses.
Phaeocystis pouchetii colonies dominated the community in experiment 1, although diatoms of the genera Chaetoceros sp., Thalassiosira sp., and the flagellated Dynobrium sp. mainly formed the community of experiment 2. The abundance of Phaeocystis pouchetii tended to show a general increase with time between the treatments except for the treatment at the highest temperature and receiving full solar radiation. Thalassiosira sp. showed generally a trend to decrease its abundance, and although the response of Dynobrium sp. was not very constant between treatments, Chaetoceros sp. showed always an increase in abundance in all the treatments performed. The differences or similarities between taxonomic groups encountered in the experiments were more evident when calculating the taxon averaged growth rates, in example showing that Chaetoceros sp. presented the highest average growth rate .
Increasing temperature has different effects between groups. For example, in the population of P. pouchetii, although growth was experienced in all treatments, the growth rates seem to decrease with increasing temperature. Besides these differences, the response to temperature increments varied for some species when exposed, or not, to full solar radiation. For example, the decrease in the growth rate experienced for P. pouchetii with increasing temperature was more important when exposed to full radiation than when UV radiation was removed (Fig. 3.1).The statistical analysis of the effect of increasing temperature, to be exposed or not to UVR, and their synergy indicated that the responses varied with the group, but most of them were not statistically significant. When considering only the effect due to increased temperature, the value of the intercept a1, representing the growth rate predicted when temperature approached the 0ºC, was significant for some of the taxa as Phaeocystis sp. The growth under UVR has resulted in a negative effect on the growth rates of P. pouchetii and Thalassiosira. The presence of UVR may also affect the slope of the response to temperature.
The chlorophyll a concentration shows an almost significant negative effect of temperature on growth rate. The effect of UVR was not significant, nor the synergy of both factors. The analysis of the response of phytoplankton when considering the total community, as indicated by changes in the chlorophyll a concentrations showed a negative effect to increased temperature.

Microbial food-webs
To evaluate experimentally the response of Arctic microbial plankton communities to increasing water temperatures and to determine possible tipping points in this response. Microbial planktonic communities form the base of the Arctic food web. Thus, evaluating the response of Arctic plankton communities to increasing water temperature is crucial to predict the consequences of the predicted warming scenarios on the functioning of the Arctic ecosystem. Determining whether this response will be smooth of whether there are thresholds for sudden change (tipping points) in the response of Arctic microbial planktonic communities is essential for the planning of adaptation and mitigation strategies. We set up an experimental system designed to evaluate the response of Arctic microbial food webs at 7 evenly separated temperature steps over the range of 1.5 to 10.5°C, encompassing the range of temperatures expected along the 21st century. These experiments were carried out in July 2009 at the facilities of the UNIS university centre in Svalbard. Two experiments were conducted where different microbial communities were assayed, the first one representing typical Arctic water from the Barents Sea while the second one represented the Atlantic influenced water masses found in the Greenland Sea West of Svalbard (Fig 3.2). The microbial communities were incubated in duplicate 20 litre flasks at the different temperatures under field light conditions for 10-15 days in order to allow the communities to respond to the temperature treatment. Hence, the responses evaluated here have two components (1) a physiological component, reflecting the effect of temperature of metabolic processes; and (2) a community component, reflecting the effect of temperature on community composition and biomass.
The response of the microbial communities was monitored by following the responses of the individual compartments of the microbial food web. Several variables were measured to characterize, not only the amount of biomass in each compartment (Chlorophyll a, abundance of bacteria, viruses, flagellates and ciliates), but also the relevant processes mediated by these organisms (bacterial production, bacterial ectoenzyme activity, bacterial mortality by protistan grazing and viral lysis, community respiration and net community production) and the possible changes in the structure of the viral and bacterial communities.
Phytoplankton (phototrophs) showed a clear response to temperature in both experiments, with phytoplankton biomass estimated as chlorophyll a concentrations decreasing abruptly at temperatures >6°C (Fig 3.2). Moreover, the composition of the phytoplankton community changed from one dominated by diatoms at low temperature towards a phototrophic community dominated by smaller flagellate cells at the highest temperatures tested. Conversely, bacterial abundance and production increased abruptly at temperatures above 4-6°C in the same experiments (Fig 3.3). Viral and protistan abundance also showed marked increases at temperatures >5°C. These changes were also reflected in a marked increase in protistan mediated bacterial mortality while viral mediated mortality showed no significant changes over the whole range of temperatures tested (Fig 3.4). Significant decreases in the utilization of carbohydrates by the bacterial community were observed at the highest temperatures tested while no significant tipping point could be detected for protein utilization. Analysis of bacterial community structure detected several bacterial phylotypes reactive to increasing temperatures, with some phylotypes decreasing their relative importance in the community and others becoming more dominant. These changes resulted in marked shifts in community structure revealing clear differences between the bacterial community growing at temperatures >5°C and those growing at lower temperatures (Fig 3.5).
These responses were also evident when looking at more integrative measurements of the metabolism of the planktonic community like community respiration and net community production (Fig 3.6). Net Community Production (NCP) decreased with increasing temperatures in good agreement with the reported decrease in biomass of primary producers. However, NCP values normalized to chlorophyll also showed an abrupt decrease between 4-6°C indicating a physiological threshold between those temperatures. Conversely, Community Respiration (CR) showed a marked increase with temperature, with a tipping point between 4-6°C where the community respiration roughly doubled its value. This confirmed the observations of enhanced growth and activity observed for heterotrophic bacteria and protists.
Our results indicate in that warming in Arctic waters will lead to enhanced heterotrophic processes and impaired autotrophic processes, leading more heterotrophic microbial planktonic communities. The stable values measured in the higher temperature treatments may not be so realistic, but nevertheless they provide evidence of the existence of a different regime at temperatures >4-6°C, confirming the validity of the thresholds ant tipping points found in this study.
Future temperature predictions (IPCC A1B scenario, Fig. 3.7) indicate that a much larger area of the Arctic Ocean will be crossing the 4-6°C threshold in summer shifting a large part of the ice-free Arctic ocean towards a more heterotrophic regime. The consequences of this regime shift at around 4-6°C are that in a warmer Arctic, the metabolic balance of the microbial planktonic communities which form the base of the Arctic food web will become more heterotrophic, diminishing the role of the Arctic ocean as a carbon sink. Moreover, the associated changes in the species composition of phytoplankton and the channelling of increasing amounts of carbon through the microbial loop will likely affect the quality and quantity of food available for key zooplankton species, which in turn will have affect fish stocks and the overall functioning of the Arctic ecosystem.

Temperature thresholds or tipping points for community metabolic balance in plankton communities in the Arctic.
We tested experimentally the response of the metabolism of Arctic plankton communities when exposed to temperatures spanning from 1-10°C using a gradient experimental design. Two types of communities were tested, open-ocean Arctic communities from water collected in the Barents Sea and warm- Atlantic water influenced fjord communities from water collected in the Svalbard fjord system.
Metabolic rates increased as predicted by metabolic theory, however these results suggest a temperature threshold of 5°C, beyond which the metabolism of plankton communities shifts from autotrophic to heterotrophic. Barents Sea communities showed a much clearer threshold response to temperature manipulations than fjord communities. These results suggest that at temperatures beyond 5°C arctic plankton communities could switch as acting as sink of anthropogenic CO2 to acting as a source. This threshold temperature is well with-in the range of temperatures predicted by the IPCC by 2050 suggesting that this important ecosystem service could be vulnerable to the effects of increasing temperatures in the future.

Combined effects of temperature and CO2 concentration on community metabolism in plankton communities in the Arctic.
We tested experimentally the response of Arctic plankton community metabolism to a increased of temperatures 1, 4 and 10°C in combination with an increase in concentration of CO2 from 380ppm to 1000ppm using a factorial experimental design. The response was tested on Arctic plankton communities from water collected in a Barents Sea fjord community.
The results form this experiment suggest that at CO2 concentrations similar to the present, primary production rates are enhanced beyond 5°C as expected from previous experimentation in objective 1. However at higher CO2 concentrations (i.e. those expected in 2100) gross primary production rates are only enhanced at lower temperatures but are not significantly affected by an increase in temperature. Current theory suggests that primary production should be enhanced in a high CO2 scenario, however we find that that enhancement only takes place at lower temperatures. Thus, this two-factor experiment (temperature and CO2) has significant implications for future predictions of increased primary production in a warmer and a more CO2 ¬rich Arctic. The results of this experiment may alter the predictions for metabolic balance in future climate change scenarios. The outcome of this experiment is currently in preparation for publication.

Mesozooplankton
Mesozooplankton metabolism and recycling of the nutrient pool
Increased temperatures resulted in a change in the stoichiometry of the metabolic products (C:N:P). Respiration was the metabolic activity less dependent on temperature, followed by NH4-N and PO4-P excretion, ensuing an inverse relation between temperature and C:N, C:P and N:P metabolic quotients (Fig 3.8). The significant effects of temperature on the stoichiometry of the excretion products would contribute to modify the nutrient pool available for phytoplankton.

Shift in species composition
Calanoid copepods comprise about 58-97 % of marine zooplankton biomass in high latitudes, and are dominated by the genus Calanus, whose life cycle is closely tuned to the seasonally pulsed food availability in the Arctic. Two species of this genus were studied in ATP: the Arctic shelf species Calanus glacialis, and the more southern Calanus finmarchicus, which has its core habitat in the Atlantic Ocean. The two species co-occur in Arctic shelf waters, but C. glacialis dominates. A major question connected to the future warming in the Arctic, is whether the boreal C. finmarchicus will expand northwards and eventually outcompete its arctic sibling, C. glacialis. In order to approach these questions, temperature-dependent responses were experimentally tested for the two species with regard to ingestion, respiration, egg production, egg development and adult survival.

Feeding and respiration
Ingestion rates in the Arctic C. glacialis, indicated that a short-term response to increased temperatures was a speeding up of the physiological process, whereas the long-term effect involved a modification of this response with lower rates at the highest temperatures, probably due to a combination of physiological and behavioural factors.
The carbon demand of individual mesozooplankton organisms will be influenced by temperature because increased respiration rates leads to higher carbon turnover within the organisms. For C. glacialis, functional responses of phytoplankton ingestion, and respiration rates to temperature were both dome-shaped in the experiments performed. Respiration increased with temperature until 6°C, whereas the highest ingestion rates were observed at 5 or 7 °C when fed algal cultures, and at 2.5° when fed a natural suspension of food. The difference between ingestion rates and respiration rates (metabolic balance) of C. glacialis decreased with increasing temperatures, and above a threshold of approximately 6°C, the metabolic respiratory demands of C. glacialis exceeded that of carbon ingestion. This indicates that C. glacialis is not able to consume more food at higher temperatures, leading to a poorer physiological state at temperatures above 6 ° C.

Reproduction and egg hatching success
A key characteristic of the genus Calanus in the Arctic is their capacity to build up energy reserves during the short productive period, and then overwinter at depth before resurfacing to reproduce in spring. Calanus finmarchicus is slightly smaller than C. glacialis, and has its core distribution in Atlanic water masses, whereas the latter is a true Arctic shelf water species. Life history of the two differs with respect to the ability of using stored energy to fuel gonad maturation and early reproduction, which is most pronounced in C. glacialis.
Egg production was measured for both species in April-May, and rates increased linearly with temperature for well-fed C. finmarchicus, while C. glacialis was negatively affected at temperatures higher than 5°C (Fig. 3.9). Further, egg production in C. finmarchicus showed a strong response to food availability, whereas food had no significant effect on egg production in C. glacialis. In lipid-storing copepods, feeding and reproduction can be separated by a time lag of different duration. C. glacialis has one of the highest lipid reserves among zooplankton, and in short-term egg production experiments it may show virtual independence of feeding conditions.
There were significant differences in egg hatching time in different temperature treatments (the higher temperature, the faster eggs’ development), but the eggs’ development rates were not different for Calanus glacialis and C. finmarchicus in any of the tested experimental temperature. There was no evidence of the effects of temperature on the hatching success (proportion of hatched eggs in different temperature treatments).

Survival
Increasing temperatures had a negative effect on adult survival (non-predatory mortality) of both C. finmarchicus and C. glacialis, with highest survival at 0°C, and lowest at 10°C (Fig. 3.10). There was no significant difference between the two species within the time window of the experiments, which was about 4 weeks.
Overall, the results from the experimental studies on mesozooplankton point to a potential for the boreal species, C. finmarchicus, to do better than its Arctic congener, C. glacialis, if temperatures during the reproductive period rise above 6 °C in a spring bloom situation. This would result from higher egg production rates, faster egg development times, and a more efficient utilisation of food in this species under such conditions.

Natural gradients to develop relationships between climatic forcing and ecosystem processes

Benthic flora
A unique multi-decadal time series (1980-2010) documented dramatic structural changes in the rocky bottom communities of two arctic fjords of the Svalbard archipelago, Norway, concurrent with increasing sea water temperatures and decreasing sea ice cover. After more than a decade of relative stability during the 1980s and early 1990s, structural variability in the composition and relative abundance of taxa increased and was followed by abrupt five-eight fold increase in macroalgal cover, most likely induced by longer ice-free periods and higher seawater temperature. The abrupt, substantial, and persistent nature of the changes indicated a catastrophic regime shift, and the study thereby provide empirical support for tipping points in arctic rocky-bottom communities associated with climate change.
A 13-year time series from northeastern Greenland also demonstrated marked increases in the production of the kelp Saccharina latissima in response to increased duration of the open water period. Longer open-water periods explained up to 60% of the variation in the annual production of kelp, which increased by 0.1 g C per day of increase in the annual open water period (Fig 3.11).
The intertidal macroalga Ascophyllum nodosum, distributed as far north as 69 ºN on Greenland’s west coast, was studied at and close to this northern limit (64-69 °N) at sites representing different water temperatures and sea-ice coverage in time and space. Average annual tip growth of the populations was found to be faster in years with highest sea temperature and least sea-ice, as evaluated retrospectively over the past decade based on morphological features.

At all investigated Greenlandic sites, annual growth rates of A. nodosum tips were low, as compared to growth rates further south compiled from the international literature, and growth rates increased southwards by 0.80 cm per ºC increase in summer seawater temperature (Fig. 3.12). However, longevity and a standing biomass of several kg DW m-2, matching that of populations growing further south, enabled an area production comparable to further south in spite of the slow growth of the tips. Laboratory studies also indicated that warming stimulated the growth of intertidal fucoids.
All results are thus in accord and suggest a northward expansion and an increased importance of intertidal macroalgae in a warmer future.

The northernmost documented populations of eelgrass (Zostera marina) in Greenland grow in inner branches of the Nuuk Fjord system (64 ºN), where temperatures are relatively high. These populations were first recorded 100-200 years ago and our studies confirm that the populations are still present and that additional populations exist. The Greenlandic populations produce fewer leaves per shoot per year than populations further south, but this is balanced by larger and long-lived shoots, implying that the area production and the biomass match that of populations further south (Fig. 3.13). The reproductive success and, thus, the spreading capacity of the Greenlandic meadows, is limited, as indicated by no ripe fruits, seeds or seedlings and a high degree of clonality, probably due to the short, relatively cold summers. Lab. studies also indicate that warming stimulates eelgrass growth and all results thereby suggest a northward expansion and increased importance of eelgrass in a warmer future.

Benthic fauna
Effects of climate change on bivalve reproduction and early life-history
Early developmental stages of marine organisms, especially in those species where development is external from the parent (either in the plankton or at the sea floor), are likely to be particularly sensitive to changing environmental conditions. Limited energy reserves, few obvious defences, and little control over their physical environment even on micro scales suggest that these stages will be vulnerable to changes in the physical environment caused by climate change. For example, many bivalve molluscs shed gametes into the water column, and increased temperatures and reduced pH values expected over the coming decades can influence survival, developmental rate, energy budgets, and calcification. These then alter the length of time larvae are exposed to predators and the condition they have when they eventually settle to the sea floor. We studied many of these processes under controlled laboratory conditions where temperature and pH (CO2 content of incubation water) were varied. We ran experiments on both Arctic and temperate bivalve species.
Sperm motility and swimming speed are often directly linked with fertilization success in marine bivalves. Increased CO2 content led to a significant reduction in both (approximately 60 and 30%, respectively), suggesting ocean acidification may negatively impact bivalve reproduction. In addition, high levels of intraspecific variation in these parameters were observed. This implies that, even if bivalve species can survive under new climatic conditions, increased CO2 can be a selection force that acts to reduce genetic diversity, with possible implications for long-term sustainability of the species. A third finding related to sperm response to changing environmental conditions was that sperm half-life in Arctic species was considerably longer in cold (current) conditions than in water heated by 4?C. Shorter sperm half-life in a warmer Arctic may result in reduced fertilization success in Arctic bivalves, perhaps favoring reproduction in boreal species. All three results suggest reduced reproductive success for bivalves under conditions of warmer and more CO2-rich waters.
We found no significant effects of near-future (2100) ocean acidification levels on post-fertilization stages, including survival, growth rates, calcification, and respiration, of the temperate bivalve Mytilus galloprovincialis. On the other hand, increased temperature had strong effects, including decreased growth and mortality. Calcification rates of the Arctic bivalve, Macoma calcarea, however, were affected by both temperature and CO2 level (as calcite saturation state). Calcification was lower at lower temperatures and at higher CO2 levels, and there was an apparent threshold in CO2 content below which calcification rates declined very steeply. This threshold level (?-calcite of approximately 1.5) is similar to that found in several other species of marine calcifiers.

Climate impacts on growth rates of Arctic bivalves
Periodic (here, annual) growth lines laid down in the hard skeletons of marine organisms offer a record of both life processes, such as growth rate, and environmental conditions at the time of carbonate deposition, such as temperature and salinity. Before any of these data can be evaluated, however, the periodicity of the growth bands must be determined. We showed that growth bands for two species of marine bivalves (Serripes groenlandicaus and Clinocardium ciliatum) are annual for both species in two fjords. This suggests that growth bands can also be used to back-calculate relative ecological conditions (e.g. primary production) over the several decade life-spans of these clams.Warmer more food-rich, Kongsfjorden supported considerably higher growth Clinocardium but not Serripes, suggesting that small changes in temperature (in this case approximately 3?C) may lead to significant responses in some species, but not others. We used these findings to study both temporal and spatial variability in growth patterns in one of these species (Serripes). Strong variability in growth from 11 locations around Svalbard was correlated with large-scale climatic drivers (related to air-pressure differences), and the local effects of those climatic conditions (temperature and sea ice extent). Insights derived from these data are valuable in linking climatic conditions with ecosystem-relevant processes on various scales of space and time. It is intriguing that such a clear link between climate and Serripes growth was evident considering that elevated temperature and phytoplankton stock size (apparent food) were not found to influence growth in the study. Either some artifact of the experimental treatment (collection, marking, and redeploying in baskets attached to mooring lines) affected the results in the growth band validation study, or some other factor related to climate is responsible.

Trophic interactions among three demersal fish species
Climate change is predicted to impact ecosystems is many ways, but the expansion of boreal taxa into the Arctic has been suggested to be particularly important. In recent years, boreal relatives of the polar cod (Boreogadus saida), an important species in Arctic food webs that helps transfer energy from zooplankton to seabirds and marine mammals, have been observed in greater numbers. These species feed exclusively in the pelagic zone as larvae and juveniles, but as they mature derive increasingly greater proportions of their diets from benthic taxa, including other demersal fish species. Juveniles often co-occur around Svalbard, and their similar sizes when collected together, particularly in the pelagic zone, suggest the potential for competition. Dietary studies revealed relatively low overlap among polar cod and each of the two related species (Atlantic cod and haddock), and also little overlap between the two boreal species. These results suggest that at current population sizes, the boreal species do not compete with their Arctic relative. Juvenile polar cod, however, were found in large schools near the sea floor, and food from this habitat may be poorer than zooplankton that characterize diets of pelagic fish. The boreal species become considerably larger than polar cod, and as adults prey on all size classes of polar cod. Continued increases in their population sizes may affect polar cod populations, with unknown consequences for ecosystem function.

Development of early-warning indicators of Climatic Thresholds and Tipping Points for Arctic phytoplankton using genomic markers of climate-driven stress
Comparative metatranscriptomic responses to warming in Arctic and N. Atlantic eukaryotic microbial communities.
Eukaryotic microbial communities lie at the base of Arctic food webs. In order to understand the potential impacts of predicted climate warming scenarios on these communities, we applied cutting-edge metatranscriptomic analyses for the first time in a comparative experimental study on intact communities from Arctic and Atlantic water masses. Summarizing the main results in terms of biological processes and pathways, several major differences were observed between the communities. Energy-transducing photosynthetic and antenna protein transcripts were down-regulated with increasing temperature in the Arctic, but remained constant in the N. Atlantic community, with potential impacts on energy production. Increased transcript levels for thiamine metabolism as well as pantothenate and CoA biosynthesis were observed in the N. Atlantic with increasing temperature, in contrast to the Arctic. This may reflect differences in the ability of the respective communities to synthesize and metabolize proteins, carbohydrates and fats at higher temperatures. Supporting this were signs temperature-induced cellular stress on the Arctic diatom community, as pathways for ubiquitin-mediated proteolytic activity, stress-responsive p53 signalling and apoptotic pathways (indicating e.g. DNA damage) were all up-regulated.
In summary, the effects observed on transcription at the community-level through a comparative metatranscriptomic approach have allowed insight into the potential effects of warming on community function, highlighting in particular the dangers and stresses imposed on Arctic communities. Clearly, potential threats at higher trophic levels through bottom-up effects of warming on microbial communities should not be ignored.

Effects of warming on the Arctic copepod Calanus glacialis, a transcriptomic study.
Climate change may dramatically affect key members of Arctic marine communities. However, the limited application of genetic tools has so far prevented comprehensive investigation of molecular processes underlying stress responses and ecological adaptation. To begin to address this, we performed the first large-scale transcriptomic analysis of the keystone Arctic copepod C. glacialis submitted to projected temperature trends (0, 2.5 5, 7.5 and 10°C).
Overall, lack of extensive transcriptional change in response to short-term temperature treatments suggests a considerable acclimation capacity, at least at the cell physiological level. The majority of the transcripts were assigned to two KOG (EuKaryotic Orthologous Groups) functional categories; cytoskeleton and energy production and conversion. KEGG (Kyoto Encyclopedia of Genes and Genomes) analysis revealed a broad range of metabolic (e.g. energy metabolism, amino acid metabolism) and organismal systems pathways (e.g. circulatory system). Among the tested temperatures, 5 and 10°C samples contained the largest number of up-regulated transcripts assigned to KOG categories. We identified more than 1000 potential simple sequence repeats (SSRs) and a subset of 24 microsatellite sequences were chosen as candidate markers for population genetic analysis. Our results provide extensive transcriptomic information (e.g. identification of new genes, gene expression related with abiotic stress) that will greatly contribute to future research including the comparative analysis of other Calanus sp. transcriptomes (e.g. C. finmarchicus) as well as the identification of novel genetic markers.

Significant results
• Tipping points were detected in the growth and activity of microbial communities within the expected range of warming for the Arctic.
• Heterotrophic bacteria grow faster and shift their use of different components of the dissolved organic carbon pool in response to increasing temperature.
• The structure of the heterotrophic bacterial community is rapidly changed by moderate warming.
• Transfer of bacterial biomass to higher trophic levels also increases in response to warming.
• Net community production decreases abruptly at the same point, while comunity respiration increases.
• The biomass of primary producers decreases with increasing temperature.
• Experimental evidence suggests that enhanced heterotrophic processes combined with decreasing autotrophic processes in a warmer Arctic will result in a shift towards more heterotrophic microbial communities.
• Identified positive responses of the benthic vegetation to warming and increased duration of the ice-free period may be applicable as indicators to monitor responses to climate change and tipping points.
• Arctic rocky-bottom communities show tipping points in terms of abrupt increases in macroalgal cover associated with longer ice-free periods and higher water temperature.
• Annual production, abundance and depth extension of kelps along Arctic shorelines increase in response to longer open-water periods and higher water temperatures.
• Kelp belts form an extensive underwater forest of meter-long macroalgae along major parts of Greenland’s coast, in strong contrast to the tiny plants on land in the high-Arctic
• Intertidal macroalgae.
• Biomass and production of sub-Arctic intertidal algae are high but warming stimulates growth suggesting an increased importance northwards in a warmer future.
• Eelgrass populations at the northern distribution limit have high biomass and area production but limited reproductive capacity; warming likely stimulates their expansion and importance in the north.

4 Future trajectories of Arctic ecosystems

Summary of progress towards objectives
We aim at projecting the likely future trajectories of Arctic Marine Ecosystems. We do so by running the well-established coupled hydrodynamics-ecosystem model (SINMOD) with projected climatic forcing, and implementing the mechanisms leading to abrupt changes derived from the time-series and experimental analyses derived above into the model. The ice-hydrodynamic model will be validated based on data above.

Model extensions and ecological model sensitivity to climatic drivers

1D sensitivity tests
The parameters for temperature-dependent ingestion rates in the ecosystem component of the model were analyzed. Simultaneous perturbations of the parameter for Calanus finmarchicus/C. glacialis and the parameters for the other functional groups were made. Temperature itself was also perturbed by simply increasing temperature throughout the entire water column at all time steps by ?T degrees.
The analyzes were done for two locations within the Barents sea: one dominated by Atlantic water masses (Atlantic station) and one dominated by Arctic water (Arctic station). At the Arctic water station removal of ice cover was used as a forcing (at ?T = 0) in addition to temperature increases. Annual primary production and Calanus spp. and ciliate production were studied as functions of the parameters and changes in temperature. Aspects of timing were considered, e.g. timing of maximal primary production. Changes in the distribution of primary production between diatoms and flagellates, and the fractions of diatoms and ciliates of the diets of C. finmarchicus/C. glacialis were also studied.
Nonlinear responses in primary production, ciliate production and Calanus spp. production to parameter and temperature perturbations were found at the Atlantic station. In the case of phytoplankton and ciliates, these responses were quite small. The maximal change in primary production found with the simultaneous perturbation of three parameters (Calanus ingestion parameter, photosynthetic response to temperature and temperature itself) was about 5 % from the “standard”. Changes in ciliate production were a bit bigger, while with a 4-degree temperature increase, annual C. finmarchicus production increased by 0 to 200 % within the parameter space.
Responses to parameter changes at the Arctic station were small for phytoplankton, ciliates and C. glacialis alike. Effects of increasing temperatures and removal of ice cover were significant: Annual primary production increased from about 53 to 110 g C m-2 year-1 with removal of the ice cover, and further to 119 g C m-2 year-1 with a temperature increase of four degrees. Ciliate production increased from 1.9 to 11.8 g C m-2 year-1 with removal of the ice cover. C. glacialis production went up from 4.3 to 7.6 g C m-2 year-1 when the ice cover was removed. A 1-degree temperature increase resulted in a further minor production increase, while a 1.3 degree increase lead to complete annihilation of the species.
Annual C. finmarchicus production was a lot more sensitive to parameter perturbations than the other variables, indicating that one should be careful with the parameter choices.
The annihilation of C. glacialis at the Arctic station with a small temperature increase suggests that one should try to find the correct temperature thresholds for survival or extinction of this species. However, it must be pointed out that complete annihilation is a 1D-model artefact: In the 3D hydrodynamic and ecological model, C. glacialis populations might be advected into the area from other regions. C. glacialis may be (partially) replaced by C. finmarchicus populations advected into the area as well, as temperatures increase.

3D sensitivity tests
An analysis of the SINMOD ecological model in 3D has also been performed to further study the importance of the temperature effect upon zooplankton production. The results from the 1D run also show high sensitivity for Arctic copepods, i.e. Calanus glacialis, as well as in the seasonal timing of microzooplankton production. The temperature dependent C. glacialis mortality rate has been tested by reducing the temperature threshold by 2 and 4 °C. The resulting differences (Fig. 4.1) in biomass are compared to the standard simulation (Fig. 4.1a b) representing 1998. The seasonal average distribution patterns show very high sensitivity to the change in the temperature condition for the dominant Arctic zooplankton species with strong reductions up to 100% in the northern and eastern Barents Sea in winter and summer (Fig. 4.1c d). The areas where sensitivity of mortality is the most import are located in the eastern Arctic Ocean stretching from the north eastern Barents Sea to the Chukchi Sea and connected through along shelf slope advection. Sensitivity of mortality is positively correlated to the distribution pattern C. glacialis biomass, which explains that sensitivity in the spatial mortality pattern is related to main flow pattern in the eastern Arctic. But sensitivity varies from West (Barents Sea) to East (Chukchi Sea) with lower values along the Siberian shelf slope.

Model tipping points and regional climate scenarios
Evaluation of the physical model (ice-hydrodynamics) is important for reliable biological simulations. The 3D coupled ice-hydrodynamic model was validated against the physical data sets provided in 2. Modeled ice-cover area fraction, and satellite measured ice fraction for the Barents Sea is shown in Fig 4.1. The modeled and measured ice fraction (ice cover) compares well, especially after 1997.
Considering the entire Barents Sea, global warming caused by climate scenario B1 or A1B will both have negative consequences for the ecosystem. Only slightly for primary production, but more significantly for the secondary production. Production of these important fish food species will decline to half of today’s level in both scenarios. Thus, neither of the two policy options will be effective in avoiding exceeding a tipping point for the Barents Sea (Fig. 4.3). Considering the entire Arctic shelves, global warming caused by climate scenario B1 or A1B will both result in increased primary and secondary production. The differences between scenario B1 and A1B are small. Neither policy decision B1 nor A1B will be sufficient to maintain production patterns as they are today (Fig. 4.4). Considering the entire Arctic Basin, global warming caused by climate scenario B1 or A1B will have mostly positive consequences. The differences in B1 and A1B are not significant. The decision of a B1 policy will not make a difference with regard to the production patterns, compared to A1B (Fig. 4.4). Global warming caused by climate scenario B1 or A1B will both have positive and negative consequences. There will be a decrease in primary and secondary production in todays fishing regions, but productivity will in crease in the eastern Barents Sea, the Kara Sea and along the Siberian shelf. Decisions to reduce greenhouse gas emissions to live up to climate scenario B1 will not change this.

Significant results
• The model predicts strongly decreasing ice extent during the 21st century.
• Transition to ice-free areas is abrupt and is accompanied with large interannual variability.
• Changes in ice conditions suggest a great impact on the physical environment of the Arctic Ocean by altering water mass composition.
• There is a great variability in primary production in the present seasonal ice zone. There will also be changes in the geographical position of seasonal ice zone and less variability in areas that no longer has a seasonal ice cover.
• The model predicts that annual primary production decreases in areas dominated by Atlantic Water. This change is mainly driven by decrease of nutrient content of inflowing Atlantic Water and by reduced winter mixing caused by increased thermal stratification.
• The most severe response to climate change predicted by the model is the disappearance of the Artic zooplankton key species, Calanus glacialis, in the northern Barents Sea. The change occurs abruptly as a response to a large increase in water temperature mainly due to changes in water mass composition in this area.

5 Socio-economic opportunities and risks emerging from climate-driven impacts

Summary of progress towards objectives.
We test the performance of different harvest control rules in fisheries under varying environmental conditions. We model optimal oil and gas exploitation strategies under uncertain prices and weather conditions. And we examine how institutions and policies for the management of living marine resources, tourism and petroleum development can cope with very rapid change in ecosystems driven by climate change.

Harvest control rules based on indicators capable of handling major ecosystem perturbations.
The aquaculture industry will adapt to increased water temperature by reallocating farms. A necessary condition for such a reallocation is that the government adapts the aquaculture management laws to allow for such an allocation. Temperature increase in Norwegian waters is likely to influence productivity for salmonid aquaculture. As sites are experiencing different temperature regimes, the impact on productivity will vary. The spatial distribution of Norwegian salmon farming depends on both institutional and physical conditions. At present, the authorities restrict the general location of farms, but in the case of large productivity differences, this regime may be liberalized. Our findings indicate substantially differing effects on the spatial distribution of production and hence value added and employment. In the baseline scenario, production growth is relatively equal among the northern, middle and southern parts of Norway. Increased temperature shifts production considerably north, both in the case of stable and liberalized management.
Harvest control rules based on indicators capable of handling major ecosystem perturbations. Management strategies, based on precautionary approaches to management, which are already introduced in Arctic fisheries, address knowledge gaps and uncertainties related to biological factors like growth, recruitment and distribution. The management procedures are now based on Harvest Control Rules (expert systems) rather than model performance, hence less vulnerable to unexpected changes and system perturbations. The performance of such expert systems have been investigated by varying spatially distributed biomasses and fishing activities in cellular automata models mimicking possible dynamics in situations of climate change.

Optimal oil and gas exploitation under uncertainties regarding price, weather and ice conditions and policy change.
Climate driven changes that have resulted in a significant reduction of Arctic ice mass reduces oil and gas exploration costs, thus making increased activity more likely. First, a continued ice reduction will naturally create possibilities of new shipping routes in the summer. Second, ice-free summers will create longer drilling seasons, which could increase the rate at which new fields become developed. On the other hand, a different ice structure could prove to be more difficult to deal with for the oil companies than previously expected. Fragile Arctic ice (i.e. an increase in first and second year ice) has been shown to be more moveable by strong winds. This represents a potential risk for the oil and gas industry, as increased ice movement could interrupt drilling.
The oil and gas industry are able to tackle an increase in ice movement, and changes in ice structure, as a result of new technological improvements. According to the oil and gas companies, relatively new technology enables oil rigs and vessels to handle larger movements of ice.
Arctic ice changes will alter the cost levels of oil and gas activity for various Arctic regions. The retreat of the sea ice and its decline in thickness are projected to continue throughout the 21th century, eventually leading to a seasonal ice cover and an Arctic Ocean almost totally free of ice during the summer. More specific, the Arctic ice extent varies from region to region, and consequently changes caused by climate changes will alter the cost of operating in different Arctic regions.

The politics of tipping points.

Socio-economic opportunities and risks emerging from climate-driven impacts
Empirical and theoretical evidence suggest that the impacts of regime shifts on human wellbeing can be substantial because it leads to significant changes in the ecosystem services produced. If unexpected, regime shifts can lead to sudden drops in social welfare. Even when they are expected and optimally managed, regime shifts generate substantial redistributions of welfare between different stakeholder groups. Small management errors could also have large impact on the aggregated long run social welfare if they cause a regime shift.
Addressing the risk of regime shifts requires setting up and using a robust framework for policy and management that accounts for potential regime shifts and tipping points. Such a framework involves spelling out the different strategies available including strategies to avoid a shift, adapt to a shift and gather more information about the shift. Further, making informed decisions requires highlighting the trade-offs between these different strategies in short and long run and the welfare implications associated with the different strategies for different stakeholders. Given the large degrees of uncertainty associated with regime shifts it can be a good strategy to invest in a mix of all three kinds of strategies adaptation/mitigation/information gathering.
Gathering information about regime shifts requires systematic programs for monitoring, modelling and prediction. Given the limited resources available it is fundamental to gather information systematically and efficiently through monitoring programs. Researchers must collaborate to define, which variables to monitor and for how long, how to prioritize among these and agree on common ways to collect data so that they can be compared across time, space and systems. Long time series are particularly useful because they can help identify relevant slow variables. Data analysis must account for the possibility of regime shifts and test for that, using appropriate tools. Data that has been systematically collected can be used to build relevant dynamic system models, which include the potential for regime shifts.
Omitting slow variables might mask key dynamics when modelling systems with tipping points. How precise should a model be? A way to answer such a question is to compare the predictions of three different models of the same phenomenon. A simple, one dimensional model only represents fish dynamics; an intermediate, two dimensional system represents fish dynamics and floating vegetation and a complex, three dimensional system represents fish dynamics, floating vegetation and bottom vegetation, which could also be a slow variable compared to the others. It turns out that all three models can predict regime shifts for some ranges of parameter values. In addition there is a range of parameter values for which following a limit cycle is an optimal strategy but this only occurs in the three dimensional model and lower dimensional models could not predict this phenomenon.
The behaviour of resource users who collectively manage a resource with threshold depends on their capacity to cooperate and on the initial state of the system. If resource users cooperate they will manage the resource optimally and a regime shift will only occur if it is an optimal outcome. If resource users do not cooperate their management will not be optimal and they will either exploit the resource too much risking a suboptimal regime shift or too little. In all cases the outcome depends on the initial conditions and very small changes may cause an undesirable regime shift or prevent a desirable one.
Linked systems of people and nature, called social-ecological systems are increasingly understood as complex adaptive systems. Understanding their high level of complexity requires interdisciplinary approaches, including integrative models to support policy. Such models should include and combine strategic interactions among economic agents, non-convexities induced by nonlinear feedbacks, separate spatial and temporal scales and spatiotemporal dynamics, and allowance of alternative time scales. Issues like spatial scaling and uncertainty are also very important. Modelling along these lines increases model complexity and analysis costs, however, empirical observations suggest that realism requires moving towards more complex nonlinear systems with such characteristics. The design of economic policies that do not take complex adaptive systems characteristics into account might lead to erroneous results and undesirable states of social ecological systems with even higher associated welfare costs.

Ecosystem based management
Ecosystem-based oceans management is a useful strategy for confronting the challenges posed by rapid climate change. The prospects of rapid climate change and the potential existence of tipping points in marine ecosystems where non-linear change may result from their being overstepped, raises the question of strategies for coping with ecosystem change. There is broad agreement that the combined forces of climate change, pollution and increasing economic activities necessitates more comprehensive approaches to oceans management, centring on the concept of ecosystem-based oceans management. The Norwegian experience in introducing ecosystem-based oceans management emphasize how climate-change can be addressed in the plan, as it is seen as a major long-term driver of ecosystem change. Understanding the direct effects of climate variability and change on the ecosystem and indirect effects on human activities is essential for adaptive planning to be useful in the long-term management of the marine environment.
Existing management regimes are flexible and have a proven capacity to adapt to change. Facing the possible future of tipping points in marine ecosystems, a salient question is how existing management regimes cope with rapid change. The existing management regime at all levels of governance in Norway has been subject to major perturbations over the last decades. While the existing management regime has proved quite flexible in adopting to change, but that it is important to retain and develop regime elements that preserves and enhances the resilience and adaptability. The questions concerning the consequences of climate change for fisheries are not something we can provide definite answers to. A main conclusion of the Arctic Climate Impact Assessment (ACIA) regarding fisheries was that good management of fisheries is important to confront the challenges posed by climate change. By reducing fishing effort and fish mortality, stocks will become more resilient to change. And we also know that our management systems over the past decades have proved capable of adapting to substantial changes in the marine environment.

Significant results
• The aquaculture industry will adapt to increased water temperature by reallocating farms.
• Harvest control rules based on indicators capable of handling major ecosystem perturbations.
• Significant reduction of Arctic ice mass reduces oil and gas exploration costs, thus making increased activity more likely.
• Arctic ice changes will alter the cost levels of oil and gas activity for various Arctic regions. The impacts of regime shifts on human well-being can be substantial
• Addressing the risk of regime shifts requires setting up and using a robust framework for policy and management that accounts for potential regime shifts and tipping points
• Gathering information about regime shifts requires systematic programs for monitoring, modeling and prediction
• Tipping points imply that regulation using prices may not produce the same outcome as regulation using quotas, even if the regulator has perfect information
• Resource users tend to behave more carefully when they exploit collectively an ecosystem if they know about tipping points that could trigger a less productive regime
• If management influences the probability of a regime shift that affects ecosystem dynamics, it is optimal to behave precautionary
• The behavior of resource users who collectively manage a resource with threshold depends on their capacity to cooperate and on the initial state of the system
• Modeling and management of social ecological systems should address strategic interactions among economic agents, tipping points, spatiotemporal dynamics with alternative time scales and uncertainty.
• Ecosystem-based oceans management is a useful strategy for confronting the challenges posed by rapid climate change

Potential Impact:
Potential impact, main dissemination activities and exploitation of results
Summary of progress towards objectives
We aimed at providing policy makers, managers, stake holders and the general public with an understanding of the ecological thresholds and regime shifts that may develop in the Arctic in response to climate change, and how the ecosystems will respond to EU targets for emissions. This information can then be used as a basis to refine policy targets, mitigate these impacts, take advantage of natural resilience already within the ecosystem, and to identify ways of promoting recovery. Furthermore, it informs the general public of the possible consequences of climate change on the Arctic ecosystem to help build support for policy frameworks reacting to these predictions.

Our project was initiated with emphasis on important outreach effort to disseminate facts of changes in the Arctic marine ecosystem, policy needs, and raise awareness. These efforts included:

• Organisation of events for the public, including debates and colloquia, where ATP scientists informed the public of changes in the Arctic marine ecosystem, policy needs, and the contribution of ATP
• International press releases
• Intense Web page activity
• Podcasts from ATP participants describing the goals of ATP and their individual roles and expectations in the project
• News pieces broadcasted in news shows in national newspapers, radio and TV stations, including interviews with ATP scientists.
• Blogs series published in connection with ATP events, posted on the ATP or other relevant (web sites, broadcasting companies as well as in major national newspapers
• Production of outreach materials: fridge magnets, flyers, posters, etc.
• Joint science-events with artists (photographers, writers, sculptors) and journalists (art exhibitions, participation on cruises)
• Production of a book “Artic Tipping Points” conveying the notion of abrupt changes in the Arctic through photographs, short texts, and poems, including an introduction to the science of Arctic Tipping Points.
• Inclusion of ATP materials in postgraduate teaching courses and an ATP summer school
• ATP featured in a career workshops such as Young Scientists Forum in Tromsø (Jan. 2011 and 2012)
• Organization of the 2011 Arctic Frontiers Conference (Tromsø, Norway) with Arctic Tipping Points as the leading theme.
• Production of the special issue The Arctic in the Earth System perspective: the role of tipping points, in AMBIO 41 (2012)
• Production of a documentary on Arctic Tipping Points.Shorter and promontory movies were produced
• Production of radio interviews and video clips on Arctic Tipping Points, and high-quality professional photos provided for free to reporters, TV stations and journalists.
• Specific ATP outreach cruise

Project results and potential general impact
The project has collated and synthesized all evidence for impacts of climate change on the Arctic marine ecosystem according to plans and schedules. These have been summarized in various published and in preparation papers, including a full summary of the ATP results, to be published after the majority of ATP publications have been appeared in the international literature. An important integration result is the volume The Arctic in the Earth System perspective: the role of tipping points (AMBIO, vol 41 (2012)), planned by the ATP management team and written and edited in part by ATP partners. ATP results were synthesized and collated in an extraordinary meeting in Esporles and Cap Salines, Mallorca in November 2010 and 2011, respectively, as the total ATP results became available. The published ATP evidence has been delivered as input to the forthcoming IPCC report by the deadline March 15 2012.
Five important papers with general potential impact are worthwhile mentioning:

Wassmann, P., C.M. Duarte, S. Agustí and M.K. Sejr. (2011). Footprints of climate change in the Arctic Marine Ecosystem. Footprints of climate change in the Arctic Marine Ecosystem. Biological Global Change, 17 (2): 1235-1429. DOI 10.1007/s00300-010-0839-3

Wassmann, P. (2011). Arctic Marine Ecosystems in an Era of Rapid Climate Change. Progress in Oceanography 90: 1-17. DOI: 10.1016/j.pocean.2011.02.002

Wassmann, P., Lenton, T. (2012). Arctic tipping points in the Earth System perspective. AMBIO 41(1):1-9. DOI 10.1007/s13280-011-0230-9

Duarte, C.M. Agusti, S, Wassmann, P., Arrieta, J.M. Alcaraz, M., Coello, A., Marbà N., Hendriks,, I.E. Holding, J., García-Zarandona, I., Kritzberg, E., Vaqué, D. (2012). Tipping elements in the Arctic marine ecosystem. AMBIO 41(1): 44-55. DOI 10.1007/s13280-011-0224-7

Duarte, C.M. Lenton, T., Wadhams, P., Wassmann, P. (2012). Abrupt climate change in the Arctic. Nature Climate Change 2: 60-63. doi:10.1038/nclimate1386

Dissemination of results to the scientific community
The ATP results were conveyed at the Arctic Frontiers conference (www.arctic-frontiers.com/) in January 2011 (> 1000 participants). In 2011 the ATP leadership had a dominating influence and position onto the organization of the dominating, recurrent conference on pan-Arctic perspective. The entire conference was entitled Arctic Tipping Points and composed of a policy conference and a scientific conference, entitled The Arctic in the Earth System perspective: the role of tipping points. The science conference was organized by ATP. The scientific conference, organized by the ATP scientists Paul Wassmann, Carlos Duarte, Dorte Krause-Jensen & Elisabeth Halvorsen, was divided into 3 parts:

1. A joint and multi disciplinary first day with 10 invited speakers ending with a panel and plenum debate;
2. Four parallel sessions with connected, interwoven and interdisciplinary themes: (Ice-ocean-atmosphere interactions in the Arctic; Marine ecosystems and fisheries; Socioeconomic and institutional perspectives; People of the North; included is a poster session) on the second and third day;
3. A summary session ending with a plenum debate.

Part of the ATP results were disseminated to the scientific public through a total of 36 publications so far, in particular through volumes such as

• Polar Biology - Special issue ‘Impacts of Climate Warming on Polar Marine and Freshwater Ecosystems’, edited by Agustí, S, Sejir, M., Duarte, C.
• Progress in Oceanography Arctic Marine Ecosystems in an Era of Rapid Climate Change, edited by P. Wassmann.
• AMBIO, The Arctic in the Earth System perspective: the role of tipping points, edited by P. Wassmann & T. Lenton.

The volumes play an particular important role because they do not only presemt essential evidence from the ATP project, but they place them into an international and pan-Arctic context. As a consequence, ATP results obtain the widest possible spread.

Dissemination to interested parties and stake holders
This activity took place towards the end of the project and included stakeholders meeting in Copenhagen, Nuuk and Tromsø. The program office traveled to Copenhagen (Denmark) and Nuuk (Greenland) in order to meet stakeholders and representatives of the Greenland Government and the Circumpolar Inuit Council. A major stakeholder meeting inviting stakeholders from Arctic oil and gas, fisheries and tourism interest was initiated during the Arctic Frontiers conference in January 2012 in Tromsø (Norway). Also present were the members of the ATP advisory board. The main ATP results were presented and the advisors board commented upon their experience within ATP. Questions by stakeholders were answered.
During a press conference in Tromsø (January) the international press present at the Arctic Frontiers conference was addressed with summaries of ATP results. Questions were answered. The Project Office represented the project inside the frame of the EU FP7 projects CLAMER (Climate Change Impacts on the Marine Environment: Research Results And Public Perception) and ACCESS (Arctic Climate Change, Economy and Society) during the final CLAMER meeting in Brussels (September 2011) and the ACCESS kick off meeting in Paris (April 2011).
Several interviews were given which often resulted in exchange of formulations and opinions, which contributed to a better understanding of ATP relevant information among interested parties and stakeholders.

Dissemination of results to relevant policy actors
We informed actors in the policy world about the scientific results emanating from the Arctic Tipping Points project. This was mainly done through IPPC relevant ATP publications were summarized and submitted to two IPCC committees. And through the formulation of a White paper and an overview on adaptation and mitigation strategies.
Dissemination during the lifetime of ATP was obviously limited, as the dissemination of results necessarily comes for the most after the conduct of research and publication of its result. We have however written or been involved in a range newspaper chronicles and interviews addressing issues central of the Arctic Tipping Points project: the international efforts to address climate change, and new approaches to oceans management to confront the effects of climate change. A number of political articles dealing with ATP issues were published in Norwegian newspapers: Barentshavet blir til Nordsjøen (Fiskeribladet); Klimaforandringer i Arktis: Overmot, desperasjon eller håp? (Klima); Når naturen tipper over (NTB); Torsken kan bli russisk (NRK web side); Varmere hav gir mindre mat (Aftenposten); Marinøkologiske utfordringer i Polhavet: toget som gikk (Klima); Arktis kan være i vippepeosisjon (Klima); The North Pole is on thin ice (http://sciencenordic.com/north-pole-thin-ice).
ATP was also interviewed by Norwegian, Swedish as well as German broadcasting companies (e.g. Fremtidig fiskerier i Barentshavet and Meeresökologische Konsquenzen des Klimawandels). ATP also participated in meetings such the final CLAMER meeting in Brussels (September 2011) and the ACCESS kick off meeting in Paris (April 2011).
For more details regarding newspaper contributions, see Press Rom on the ATP website (http://www.euatp.org/index.php?option=com_content&view=section&layout=blog&id=7&Itemid=2).

Convey results to the general public
The homepage of ATP-project (http://www.eu-atp.org) served as the basic tool to convey results to the general public. Here the general public finds a wide range of easily accessible, updated information. ATP had more than 80 entries - among other blogs from the ATP field cruises and experiments giving the readers a popular insight in the scientific work of ATP. Since launched, the web page www.eu-atp.org the page has been visited by 6 368 unique visitors making 10 794 visits and making 31 412 page views. Visitors come from 96 countries most from Norway, Spain, Greenland, US, Denmark, UK, France, Germany, Poland and Canada. Interestingly, US Internet users rank high as visitors to the project’s web page, which cannot be assigned to media campaigns, since these did not reach the US. Instead, we believe the interest of US users derives from activities conducted at meetings in the US, driving the attention of students to the ATP project. Most visitors came directly to the site or via Google. Next via the ARCTOS website, El Pais and Natur in Greenland.
ATP produced the spectacular, 250 pages long, coffee table book “Arctic Tipping Points” (3000 copies). The notion of abrupt changes in the Arctic were conveyed through first class photographs, short texts, and poems, including an introduction to the science of climate change and Arctic Tipping Points.
The project’s web page was the central hub of making use of mass media (TV, radio, press, general public) to maximise the dissemination of project results and news. Press releases have been conducted, with an uneven reception in different countries (most hits in the Spanish speaking media, where ATP was featured in all prime-time news of major TV stations in Spain and through CNN+ in South America). Several interviews and shorter movies were produced during the Arctic Frontiers conference Arctic Tipping Points in January 2011 and during the 3rd ATP cruise which was dedicated to outreach and research. News articles telling about the ATP projects and results from the work have been written in Spanish, Polish Norwegian and Danish newspapers.
Teaching material for young school kids at 4. -6. grade and for pupils at high school level have been produced and published on the web page for electronic teaching material under the Danish Ministry of Education (see i.e.: http://www.emu.dk/elever4-6/natfag/gronland/tang.html; http://www.emu.dk/elever4-6/natfag/gronland/index.html and http://www.emu.dk/gym/fag/bi/klimaarktis/index.html). It is the hope to translate this material to Greenlandic and English.
An extensive sampling of high-level photographs has been taken by ATP employed and associated artists. Artists on board the ATP cruises have produced art inspired by the scientific work. This work is reflected in the Arctic Tipping Point book as well as the exhibition “Creatures”-
ATP attempted to finalize the project through an outreach event entitled Climate Change in Ultima Thule. The event was planned to take place at the premises of FBBVA in the centre of Madrid, but ATP was not allowed to use these premises before April 2012, implying an extension of the contract. Later the FBBVA annulled the offer and ATP organized the outreach event at the excellent premises of Wissenschaftsforum in central Berlin. The event had to be cancelled because the EU did not accept an extension of the contract. The planned final event in a central European capital, accompanied by exhibitions of photographs, slide shows, movies, talks to the public and a press conference was thus not possible. As an alternative the final event was carried out in Tromsø at the Arctic Frontiers conference in January 2012. Outreach products (book, blogs, teaching material for school kids, extracts from the children book and comic strip, slide shows, movies and the photo exhibition “Creatures”) were presented, as well as a talk to the general Tromsø public hold.

A summarising overview of the activities conducted include:

• Organisation of events for the public, including debates and colloquia, where ATP scientists informed the public of changes in the Arctic marine ecosystem, policy needs, and the contribution of the Arctic Tipping Points project.
• Press releases coordinated across countries.
• Podcasts from ATP participants describing the goals of ATP and their individual roles and expectations in the project, posted on the ATP web as well as in El País newspaper.
• News pieces broadcasted in news shows in national newspapers, radio and TV stations, including interviews with ATP scientists.
• Blogs series published in connection with ATP events (cruises, experiments, meetings), posted on the ATP web site as well as in major national newspapers (El País and La Vanguardia, in Spain). Blogs were written by ATP scientists, as well as senior science reporters participating in ATP cruises.
• Production of outreach materials: fridge magnets, flyers, posters, etc.
• Joint science- Events with artists (photographers, writers, sculptors) and journalists
• Production of a book “Arctic Tipping Points” conveying the notion of abrupt changes in the Arctic through photographs, short texts, and poems, including an introduction to the science of Arctic Tipping Points.
• Inclusion of ATP materials in postgraduate teaching courses (course “Impacts of Climate Change in Polar Ecosystems”, International Univ. Menéndez Pelayo-CSIC, coordinator S. Agustí, CSIC, Spain).
• Preparation of the pan-Arctic PhD course Young Scientists Forum, to take place parallel to the Arctic Frontiers Conference in January 2011
• ATP featured in a career workshop on Opportunities for Polar Research for graduate students at the ASLO Aquatic Sciences meeting in Nice (Feb 2009).
• Organization of the 2011 Arctic Frontiers Conference (Tromsø, Norway) with Arctic Tipping Points as the leading theme, and plans to produce a special issue of the journal AMBIO with the keynote lectures.
• Production of a 50’ documentary on Arctic Tipping Points produced and featured at Cuatro TV station in Spain and CNN+. The documentary featured ATP scientists and was filmed during ATP 2010 cruise, including the 2010 experimental campaign.
• Production of video clips on Arctic Tipping Points, and high-quality professional photos provided for free to reporters, TV stations and journalists, reproduced in major TV stations.
• Preparation of an ATP outreach cruise, funded by additional funds provided by the Spanish Ministry of Science and Technology, to be conducted in May 2011.

Using funds allocated to add funding to the outreach component by the Spanish Ministry of Science and Innovation (120,000 €, granted to C.M. Duarte for this activity and additional support by UoT), ATP’s WP7 organized an outreach cruise to bring together artists and communicators with scientists to report on the changes taking place in the Arctic. The cruise took place in late May 2011.

We are also progressing with three additional products:
• a comic strip on Arctic Tipping Points
• a book for children to better understand the relevance of climate in the Arctic.

Training of young scientists on Arctic climate change and tipping points
Training of young scientists has proceed through three main axis:

1. Continued training of Ph.D. students in the project (> 12 students conducting their Ph.D. thesis, in total or part, in the framework of ATP).
2. Coordination of the 2011 ATP Cruise with training of Ph.D. students from UoT and CSIC.
3. Young Scientist Forum, associated with the Arctic Frontiers conference. In association with APECS
4. ATP summer school “Climate change, marine ecosystems and tipping points in the Arctic Ocean” (P. Wassmann, UoT, Norway). An international event with PhD students from all over Europe and several invited marine arctic ecologists.

Significant results
Project results and potential, general impact
• Wassmann, P., C.M. Duarte, S. Agustí and M.K. Sejr. (2011). Footprints of climate change in the Arctic Marine Ecosystem. Footprints of climate change in the Arctic Marine Ecosystem. Biological Global Change, 17 (2): 1235-1429. DOI 10.1007/s00300-010-0839-3
• Duarte, C.M. Lenton, T., Wadhams, P., Wassmann, P. (2012). Abrupt climate change in the Arctic. Nature Climate Change 2: 60-63. doi:10.1038/nclimate1386.
• Duarte, C.M. Lenton, T., Wadhams, P., Wassmann, P. (2012). Abrupt climate change in the Arctic. Nature Climate Change 2: 60-63. doi:10.1038/nclimate1386

Dissemination of results to the scientific community
• Arctic Frontiers conference, January 2011
• Arctic Marine Ecosystems in an Era of Rapid Climate Change. Progress in Oceanography 90 (2011). Edited by P. Wassmann. See DOI: 10.1016/j.pocean.2011.02.002
• The Arctic in the Earth System perspective: the role of tipping points in AMBIO 41 (2012). Edited by Wassmann & Lenton, for an overview, see DOI 10.1007/s13280-011-0230-9.

Dissemination to interested parties and stake holders
• Stakeholders meeting with Circumpolar Inuit Council and Greenland Government in Copenhagen and Nuuk, November 2011.
• Stakeholder meeting with representatives from Arctic oil and gas, fisheries and tourism interest, the ATP advisory board and the ATP core team at Arctic Frontiers conference in January 2012 in Tromsø (Norway).
• Representation at EU FP7 projects CLAMER (Climate Change Impacts on the Marine Environment: Research Results And Public Perception) and ACCESS (Arctic Climate Change, Economy and Society) during meetings in Brussels and Paris, April and September 2011.

Dissemination of results to relevant policy actors
• IPPC relevant ATP publications summarized and submitted to two IPCC committees.
• White paper/informative guide describing the effectiveness of different policy options in avoiding exceeding tipping points for Arctic ecosystems
• Report on adaptation and mitigation strategies

Convey results to the general public
• The book Arctic Tipping, edited by C. Duarte & P. Wassmann. http://www.fbbva.es/TLFU/microsites/artic/ATPweb.html
• Intense coverage of ATP in major media: TV, Newspapers and radio.
• Results presented at Norwegian and Nordic science web sites
• Talk to general public “Tipper Polhavet”, Tromsø, January 2012
• Invited talk to general public “Tipper Polhavet”, Århus, November 2012, three evenings, about 8000 listeners

Training of young scientists on Arctic climate change and tipping points
• Young Scientist Forum, associated with the Arctic Frontiers conference, 2010 and 2011.
• ATP summer school “Climate change, marine ecosystems and tipping points in the Arctic Ocean” (P. Wassmann, UoT, Norway).

List of Websites:
www.eu-atp.org

Prof. Paul Wassmann
University of Tromsø
Faculty for Biosciences, Fisheries and Economics
Department for Arctic and Marine Biology

+4777644459
paul.wassmann@uit.no
final1-226248-1008106-figures-publishable-summary.pdf