Ice-covered areas of the Arctic were long presumed to be unproductive, and early scientific studies in the Arctic generally supported this paradigm (Nansen 1906). Evidence of human settlements in the high Arctic over several thousand years, however, conflicted with these early observations and constituted a paradox as to how human populations could subsist in regions considered to be biological deserts. Further investigations revealed the existence of productivity hot spots on par with some of the most productive places on earth, and provided the first indications of complexity and the importance of the links between ice, ocean, and land in Arctic ecosystems. During the last 20 years, national and international research efforts in the Arctic have sharply increased, culminating with the 3rd International Polar Year (IPY, 2007-2009). Highlights of the IPY work include cataloguing biodiversity from bacteria to top predators, documenting the importance of ice cover for a number of ecosystem processes, studying the relationships between physical and biotic processes on small spatial scales, describing the oceanography of previously poorly known areas, and investigating atmosphere-ice-ocean feedback relationships (e.g. Barber et al 2010, Bauerfeind et al 2009, Forest et al 2010). In addition, there have been significant developments in research infrastructure, including novel remote sensing technologies and algorithms for interpreting their output (Pabi et al 2008, Schofield et al 2010), enhanced oceanographic mooring networks, development of international databases, and long-term monitoring infrastructures. Conceptual models are now available for some physical, chemical, and biological changes expected with further withdrawal of ice, warming, acidification, and altered seasonality in light and nutrient distribution (Carmack & Wassmann 2006, Lovejoy et al 2007, Kirchman et al 2009, Kovacs et al 2012). Despite this, there are major and fundamental gaps hindering our ability to understand the Arctic as a single, linked system undergoing unprecedented change and in an earth science perspective. Perhaps the most obvious and largest of these known gaps is centred around the widely accepted paradigm that Arctic marine ecosystems are best compared with a marine desert during the long and dark polar night. Just as the paradigm of the Arctic Ocean being an unproductive biotope was refuted a hundred years ago, the prevailing view of the polar night as devoid of biological activity has recently been challenged (e.g. Berge et al 2009, 2011, 2012).
Climate change is leading to rapid shifts in Arctic ecosystem structure, resilience and function, which cannot be easily predicted by models based on our current understanding (Wassmann et al 2011). Rapid declines in sea ice, increased air and ocean temperatures, increased water-column stratification, and multiple physical, dynamic and chemical changes significantly alter the patterns of productivity at the base of marine food webs (Walsh 2008). Such changes are also anticipated to affect ecosystem structure and productivity higher in the food web. In addition, expected shifts in ocean currents and faunal migration patterns may substantially affect production of organisms at the top of the food web. Arctic marine ecosystem structure and productivity within the next decades will, therefore, be substantially different from what we observe today. Marine ecosystem processes are direct consequences of the complex behaviours and interactions between organisms, many of which are driven by the physical environment. Accordingly, a classical paradigm in Arctic marine ecology suggests that most biological processes stop during the polar night at high latitudes due to low food availability and the lack of light (Smetacek 2005, Piepenburg 2005). Recently, new research has challenged this assumption by presenting evidence of both diel vertical migration (DVM) of zooplankton (Berge et al. 2009) as well as bioluminescence levels indicative of biotic activity (Berge et al 2011) hitherto assumed to be absent during the polar night. Although the polar night at high latitudes is perceived as total darkness, new data indicate that Arctic organisms nevertheless may respond to light levels undetectable by the human eye. These recent results (Berge et al 2009, 2011) provide circumstantial evidence for both an endogenous and exogenous control of these poorly understood or previously unknown processes during the high Arctic polar night. During a recent expedition at 80° North during the darkest period of the polar night, five different species of seabirds were observed actively foraging at sea; little auk (Alle alle), black-legged kittiwake (Rissa tridactyla), northern fulmar (Fulmarus glacialis), black guillemot (Cepphus grylle) and brünnich's guillemot (Uria lomvia). These seabirds have not previously been reported to overwinter at this latitude, and it remains a mystery how these visual predators are able to detect their prey. These unexpected discoveries under the extreme conditions of the Arctic winter reflect the historically low levels of scientific investigations during polar night, and challenge our understanding of Arctic marine organisms and ecosystems. Not the least, based on the findings from both Berge et al (2009, 2012), Søreide et al (2010) and Wallace et al (2010), the current reduction of Arctic sea ice (Comiso et al 2008) is likely to have both a direct and indirect impact on marine organisms, their interactions and ultimately ecosystem processes. But without a more fundamental perception of Arctic ecosystem function, such impacts will remain largely impossible to understand and predict. Winter ecology of Arctic marine systems, then, is a largely new field of science with the potential for radically altering our fundamental perception of basic Arctic ecosystems processes, current state of the ecosystem and connections between the biosphere, hydrosphere and cryosphere within the Polar Region.