Background and Rational
Current understanding of Changes in the North Sea Ecosystem
Atmosphere-Ocean Physics and the North Sea
Changes in physical attributes of the ecosystem have been examined and mechanisms proposed. A number of recent studies have clearly demonstrated the close link between the North Atlantic atmospheric variability, represented by the NAO, and hydrodynamic changes occurring on the Northwest-European Shelf. The process is not unidirectional but contains feedbacks between large-scale North Atlantic atmospheric patterns and cross-shelf patterns of North Sea water circulation. The former influences the latter but the latter also modifies the former. Long-term changes in datasets of physical factors in the North Sea have been empirically correlated with the NAO. Two major possible transfer mechanisms have been discussed:
- the direct atmospheric forcing and its effects on stratification and
- advection of Atlantic water through the English Channel and the northern North Sea entrance.
Recently, Sundby and Drinkwater (2007) reported that the major salinity anomalies associated with decadal-scale climate pattern of the North Atlantic can result from variability in volume flux induced by the NAO. Variability in these fluxes will also impact advection of planktonic fauna. The abundance of Calanus finmarchicus in the North Sea can be linked to the advection of cool Calanus-rich water masses from the Norwegian Sea (Sundby 2000). Hence, the decadal-scale variability of the ecosystem features of the North Sea seems to be linked primarily to wind forcing and flux variability of water masses. The multi-decadal climate variability, as manifested by the Atlantic Multidecadal Oscillation (AMO), forces the marine ecosystem differently from the NAO. The AMO is a thermal signal that seems to have a more profound and large-scale impact on ecosystems and habitat ranges of marine organisms in the North Atlantic.
Changes in North Sea Lower Trophic Levels
Although changes in North Sea species diversity have been documented at all trophic levels, marine plankton are particularly good indicators of changes in climate and eutrophication since
- these species are directly linked to bottom-up forcing via nutrient dynamics,
- planktonic species are short-lived and hence respond quickly to the prevailing abiotic conditions,
- responses of species are not directly impacted by exploitation (e.g., fishing pressure), and
- non-linear responses of plankton communities may amplify subtle environmental changes (Taylor et al. 2002; Hays et al. 2005).
The major source of information on published long-term trends in lower trophic levels (phytoplankton and zooplankton) in light of environmental change is the data series provided by the Continuous Plankton Recorder (CPR) (Corten & Lindley 2003, Beaugrand 2003). In terms of phytoplankton, the long-term increased phytoplankton biomass in the North Atlantic parallels the development of temperature caused by the North Atlantic Oscillation (NAO) (Reid et al. 1998, Edwards et al. 2001). Along with the increased phytoplankton biomass, increased abundance of dinoflagellates and decreased abundance of diatoms have been reported for most of the Northeast Atlantic regions (Edwards et al. 2002) caused mainly by reduced stratification and low nutrient availability in offshore areas favouring dinoflagellates (Leterme et al. 2005). This climate-driven change is most pronounced in the Northern North Sea and the adjacent southern North Atlantic (Leterme et al. 2006).
The response of zooplankton to environmental change is more complex due to a mixture of both direct (e.g., temperature, salinity, water column stability, advection) and indirect (via changes in trophodynamics via species dominance and/or phenological shifts in prey availability) impacts. Direct impacts have been inferred from correlations between changes in zooplankton community characteristics (e.g., species composition and diversity of the copepod community) and changes in hydroclimate including a positive relationship between copepod diversity and the NAO index (Beaugrand 2003). Consistent with the prediction that continued warming will drive species ranges toward the poles, strong biogeographical shifts in calanoid copepod assemblages in the North Sea and adjacent waters have occurred with a northward extension of more than 10° latitude of warm-water species and a decrease in the number of colder-water species (Beaugrand et al. 2003). A shift in dominance has been reported between Calanus finmarchicus and C. helgolandicus, congeners with markedly different thermal preferences. Although the former and latter are positively related to cold and warm conditions, respectively (Beaugrand 2003), the shift is not merely driven by ecophysiological preferences but is also influenced by changes in climate-driven patterns of advection of C. finmarchicus from their overwintering areas west of the Shetland Isles, (Heath et al. 1999).
Changes in North Sea Upper Trophic Levels
Identifiying drivers of change in upper trophic levels is more complex due to a number of potential drivers including commercial explotation. The North Sea fish community has undergone pronounced changes in the last six decades. While the 1950s and 1960s were characterised by huge pelagic stocks of herring and mackerel, in the 1970s these collapsed largely due to overfishing and stocks of demersal gadoids exploded (gadoid outburst). Since the 1970s, large-scale commercial fisheries on sandeel and Norway pout developed and exploitation of demersal fisheries steadily increased which reduced the share of large predators in the system. During the 1980s and 1990s, herring stocks recovered and large horse mackerel populations entered into the southern and northern North Sea in spring and summer.
The pronounced changes in the upper trophic levels observed during the 1990s are associated with a regime shift in the late 1980s that affected all trophic levels from phytoplankton, benthos to fish and bird populations. The regime shift was most likely due to changes in physical conditions associated with changes in atmospheric forcing as evidenced by the sudden increase of the NAOI (e.g. Beaugrand 2004, Weijermann et al. 2005, Alheit et al. 2005). The most recent period is characterised by, among other trends, strongly decreasing stocks of cod and whiting, successive recruitment failures in herring, Norway pout and sandeel, recruitment failures in sea-bird populations and a recolonisation of the North Sea by spawning populations of anchovy and sardine (Beare et al. 2004, Alheit et al. 2007).
Superimposed upon changes due to exploitation strategies are those due to climate-drivers. Perry et al. (2005) have described a general northward shift of two thirds of the commercially important stocks related to the general warming trend in the most recent decades. A similar warming trend and documented northward shifts in fish distribution were reported in the 1930s and 1940s. These historical and more recent distributional changes correlate well with the phase of the AMO. Recent trends in the AMO were strongly correlated to a number of biotic changes including latitudinal shifts in the peak spawning area of Arcto-Norwegian cod (Sundby & Nakken 2008) and are quite possibly related to the Gadoid Outburst of the North Sea in the 1960s.
Removal of key predators driven by either changes in fishing pressure and/or climate can result in dramatic changes in the foodweb structure via trophic cascades (Frank et al. 2005). A specific example for the North Sea would be the dramatic declines in herring and mackerel in the 1960s resulting in decreased predation pressure on fish larvae and zooplankton that may have contributed to the subsequent increase of sandeel and even seabird populations. In turn, bottom-up control of key predator species must also be considered as evidenced by the correlation between seasonal prey availability and recruitment failure of North Sea cod (Beaugrand et al. 2003). Furthermore, trophic links will also be modulated by differential rates of distributional shifts that result in altered spatial overlap among species, thereby disrupting interactions and potentially compounding the decoupling effects of climate-driven changes in phenology (Edwards & Richardson 2004).
Simulating Drivers of the North Sea Ecosystem using Physical-Biological Models
Long-term climate-driven physical processes have been simulated in the North Sea using hydrodynamic models (e.g., Schrum et al. 2003, see review by Lenhart & Pohlmann 2004). Models have undergone detailed assessment and validation, including the ability of models to describe observed low-frequency climatic variability in relation to atmospheric forcing and the impact of this on key hydrodynamic parameters such as SST, stratification, and salinity (e.g., Janssen et al. 2001). Coupled 3-D physical-biological models have also been developed by ECODRIVE partners that allow reconstruction of climate driven (high and low frequency) changes in the dynamics of bottom-up processes. These include NPZD (nutrient, phytoplankton, zooplankon and detritus) models (for review see Moll & Radach 2003) and bio-physical models of key species at the tertiary level (e.g., larval & juvenile fish) to assess the relative importance of various mechanisms proposed to influence recruitment success (Kühn et al. In Press).
Finally, North Sea trophodynamic relationships have been modeled for decades using 0-D ecosystem models (e.g., ECOSIM, Andersen & Ursin 1977) and, more recently, historical changes in upper trophic levels examined using detailed multispecies (4M/SMS) assessment models. The suite of models developed by ECODRIVE partners for the North Sea allows ECODRIVE to evaluate the links between changes in atmospheric forcing and ecosystem components by using derived proxiesbased on simulated and observed processes and drivers.
Predicting Drivers of North Sea Ecosystem Change (ECODRIVE)
ECODRIVE will address several deficiencies that exist in previous research exploring changes in the North Sea ecosystem. First, ECODRIVE utilizes the longest available time-series data (e.g., 50 yr CPR survey) to address historical as well as recent change in physical and biological components of the system exploring the effects of decadal (NAO) to multidecadal drivers (AMO). Largely “correlative” and/or “descriptive” research is replaced by a mechanistic understanding of change by including detailed knowledge on ecophysiological (abiotic) and trophodynamic (biotic) processes impacting key species and assemblages of the North Sea foodweb.
It is expected that progress made toward understanding (and disentangling) the multiple drivers of ecosystem change will largely result from the use of advanced bio-physical models that include more accurate and precise parameterizations of important biological processes as well as the utilization of higher level statistical models (Drinkwater et al. 2005, Stenseth et al. 2004). ECODRIVE improves and utilizes these models to assess the impact of various drivers (climate, exploitation, eutrophication) on historical changes as well as the potential for future ecosystem changes in the North Sea Ecosystem.