Skip to main content

Forage Fish Interactions

Final Report Summary - FACTS (Forage Fish Interactions)

Executive summary:

removal of a forage fish has consequences for both predators and prey of forage fish. As everything is connected, every management action has a price which goes beyond the apparent, direct effect on the target species. The fishery on forage fish can therefore not be seen in isolation, as the immediate gain in profit from the fishery has to be discounted by the lowered potential for production of large piscivorous fish. Management actions on other species also influences forage fish, i.e. conservation efforts on marine mammals or sea birds have direct consequences for the predation pressure on forage fish.

the objective of the FACTS project has been to provide insight and quantitative advice on the ecosystem wide consequences of management actions directly or indirectly related to forage fish.

project Context and Objectives:
FACTS will develop and disseminate advice on the consequences of various forage fish harvest strategies to the ecosystem including their economic implications. FACTS research focuses on seven forage fish species (anchovy, herring, capelin, Norway pout, sardine, sandeel and sprat) that are a major natural resource to the European community and represent key elements in the functioning of marine ecosystems. FACTS will eliminate critical gaps in knowledge that currently exist concerning the impact of variations in forage fish populations (due to various drivers such as climate and fishing) on the trophodynamic structure and function of different European marine ecosystems.

to accomplish this goal, FACTS will answer the following questions:

1.What are the major short- and long-term drivers of changes in commercially and ecologically important forage fish populations within European waters?
2.What are the biological and economic consequences of changes in forage fish populations in terms of their prey, their competitors and their predators?
3.What are the biological and economic consequences of changes in predator populations on forage fish populations and their fisheries?
4.What is the role of forage fish species in maintaining biodiversity and ecosystem stability?

the Science and Technology (S&T) objectives of the FACTS project have been:

1.To quantify the trophic interactions between forage fish and the prey and predators, establish levels of competition between forage fish species within each case study region, and assess the impact of climate change on FACTS forage fish within European waters.
2.To quantify ecosystem responses to perturbations from human activities especially fisheries impact and environmental change.
3.To provide cost-benefit analyses and cost-effectiveness analyses of perturbations on forage fish populations, either through fishing or from predators (sea birds and marine mammals).
4.To develop case study specific and generic advice for ecosystem-based fisheries management of forage fish populations in European Seas.

in the objective description for theme 2 'Food, agriculture and fisheries, and biotechnology' (European Commission C(2008)4598 of 28 August 2008) in the work programme several social, environmental and economic challenges are listed. The FACTS project refers to the following:

challenge FACTS
growing demand for safer, healthier, higher-quality food coupled with the sustainable use and production of renewable bio-resources
FACTS will produce tools to assess the sustainability of resource use from a multiannual (medium and long term) and ecosystem perspective.
threats to the sustainability and security of agriculture, aquaculture and fisheries
FACTS will include cascading threats to sustainability, for example the effects of fisheries on small pelagic fish for large demersal fish, and FACTS will analyse sustainability from an economical point of view.

increasing demand for food production to take into account animal welfare and rural and coastal contexts
FACTS will, beyond fisheries management, assess the consequences of fisheries on small pelagic fish for bird- and marine mammals populations

specifically, the call text for topic KBBE-2009-1-2-14 states that

'The objective of the project is to establish the role of foraging fish , particularly with regards to ecosystem stability, which is relatively unknown and to establish the costs and benefits of maintaining high levels of foraging fish in an ecosystem.'

the Common Fisheries Policy (CFP) has three main objectives (according to the Commissions communication on the reform of the CFP from 2002):
-responsible and sustainable fisheries and aquaculture activities that contribute to healthy marine ecosystems
-an economically viable and competitive fisheries industry which will benefit the consumer
-a fair standard of living for those who depend on fishing activities

the FACTS project interprets these objectives as dependent on each other. However, despite the obvious need for fisheries management that accounts for the ecosystem as a whole, such management has not yet been implemented. This might be due to the lack of quantitative models that both mimic ecosystem effects of fishing and are applicable for management purposes. Another probable problem is that traditional fisheries reference points such as minimum acceptable spawning stock biomass or maximum sustainable yield have not yet been defined in an ecosystem context. Applying the maximum sustainable yield to a given species will often render it impossible to get the maximum sustainable yield for its predators or prey.

there exists, hence, a substantial need for improvement. This is foreseen in the Marine Strategy Framework directive: In view of the dynamic nature of marine ecosystems and their natural variability, and given that the pressures and impacts on them may vary with the evolvement of different patterns of human activity and the impact of climate change, it is essential to recognise that the determination of good environmental status may have to be adapted over time (taken from Directive 2008/56/EC (Marine Strategy Framework Directive)). The Marine Strategy Framework Directive constitutes the vital environmental component of the Union's future maritime policy, designed to achieve the full economic potential of oceans and seas in harmony with the marine environment. To his end, FACTS will develop the decision tools to assess the consequences at an ecosystem level of a given set of harvesting priorities, including their economic implications.

furthermore, we will make these decision tools available where they are needed: in the ICES and STECF working groups that delivering the basics for the recommendation of harvesting strategies to the Commission. This can be done because key participants in the FACTS project are active members of these working groups.

the FACTS project has the potential to bring ecosystem-orientated management in the European water a significant step ahead, by contributing to following aspects of the CFP:

CFP elements FACTS contributions
industrial fishing
impact of industrial fishing (mostly forage fish) on eco-systems
incorporating environmental concerns into fisheries management
impact of forage fish populations and their fisheries on ecosystem stability and resilience
improvement of scientific advice for fisheries management
implementation of FACTS in ICES (International Council for the Exploration of the Seas) and STECF assessment working groups
socio-economic consequences of multi-annual fisheries strategies
linkage of ecosystem and economic models of fisheries impact on ecosystems

project Results:

main Science and Technology (S&T) results (25)
climatic drivers (T1.1)
within each system, long-term time series analyses revealed significant correlations between either direct (temperature, winds) and/or indirect (prey field / trophodynamic) effects of climate on forage fish productivity (or distribution).

for FACTS case study regions and target species, these can be briefly summarized as:
i. Bay of Biscay: Application of machine learning techniques predicted years of low, medium and high recruitment related to mean North Atlantic sea surface temperature and meridional momentum fluxes across offshore banks for sardine (Sardina pilchardus) and the first PCA component of de-trended climate indices, upwelling index and wind strength and direction in European anchovy (Engraulis encrasicolus).
ii. Barents Sea: Warming shifts the distribution of capelin (Mallotus villosus) northward likely due the direct effect of the species' relatively narrow range in temperatures (2 to 7) required for spawning and via changes in larval drift. Spring spawning herring have stronger year classes during warmer years due to changes in spawning and hatch times, drift regions and temperatures, and the magnitude of predation losses.
iii. Baltic Sea: Warming and decreased salinity have had positive effects on both sprat (Sprattus sprattus) and herring (Clupea harengus) via both direct (thermal tolerance in sprat) and/or indirect (increase in preferred prey species in both species) processes. An important, indirect effect is the reduction of cod recruitment due to climate-driven changes in hydrography.
iv. North Sea: Warming temperatures have markedly increased the productivity of anchovy (that appear to be relict populations that have persisted at low levels in the region) as well as improved conditions for life cycle closure of sardine in southern regions (e.g. German Bight) since the mid 1990s. Physiological foraging and growth models for larval herring and sprat suggested that the former has lower prey requirements at the same temperature but that sprat (spawning April - June) was predicted to be insensitive to differences in temperatures observed in the 1990s. Larval survival during the winter was predicted to constrain the adaptability of autumn spawning herring facing warming conditions.

climate-driven Regime Shifts: Climate-induced changes in the hydrography in the Baltic, North Seas recently caused an ecosystem regime shifts with changes at all trophic levels. The climate-driven mechanisms are more conspicuous in the Baltic and Barents Seas compared to the North Sea due to differences in the complexity of the food webs. In the Baltic, reduced salinity and warmer temperatures since the early 1990's favoured the expansion of Acartia spp and reduced the population of Pseudocalanus acuspes which increased the reproductive success of both sprat and herring. Recruitment environment relationships in some herring stocks predicted positive effect of temperature. Warming is positively correlated with herring recruitment but the mechanism is thought to be indirect via increased prey abundance and feeding success and survival of early life stages. Sprat are thought to benefit both directly (thermal tolerance) and indirectly (prey field) from warming. Interactions are apparent between Baltic herring and sprat. Large stock sizes of the latter are correlated with reduced adult condition in the former, an intrinsic factor also significantly correlated with herring recruitment success. Both herring and sprat have benefited from decreased predation pressure from cod. The complex changes in interaction strength among food web components are more difficult to disentangle in the North Sea where a higher number of species with different biogeographic affinities (e.g. Lusitanian vs. boreal) interact.

regional Processes Differ within Systems: Research on forage fishes in both the Baltic Sea (herring) and Barents Sea (capelin) highlight the importance of performing spatially-explicit analyses of climate impacts within marine systems; the strength of climate-driven changes in key processes vary spatially. In the Barents Sea, the strength of the increase and/or differences in zooplankton biomass in years of warmer temperature are not consistent across the system with more northern areas experiencing the largest differences due to hydrographic changes (and domination by Arctic versus Atlantic water masses / zooplankton communities). In the Baltic Sea, analyses of 5 herring stocks indicated regional differences with three of the five showing positive relationships between recruitment and temperature. In that case, salinity differences among the systems and the timing of herring spawning also likely contributed to differences. A second important factor is that the strength of climate-driven changes may depend upon stock size via density dependent regulation of zooplankton resources, the strength of which will also depend upon regional characteristics. Finally, differences between the southern and northern North Sea areas are apparent in terms of zooplankton community structure and water mass characteristics that will affect forage fish species composition and productivity.

prey (T1.2)
baltic Sea: The diets of all life stages (larvae, juveniles and adults) of sprat and herring have been well documented. These studies were reviewed with emphasis on the reliance of sprat and herring on calanoid copepods such as Pseudocalanus sp., Temora longicornis s and Acartia spp. Diet studies in combination with food consumption estimates from bioenergetics-based modelling and gastric evacuation studies support the assertion that both species can exhibit top-down control on zooplankton. Due to their high abundance, age-0 juveniles appear to be particularly important regulators of the zooplankton community. Predator consumption rates along with differences in feeding strategies (particulate versus filter feeding) and optimal temperatures (e.g. between 18 to 20°C for sprat) are highlighted. These basic, diet studies agree well with estimates of trophic position from food web modelling utilizing ECOPATH. In that model (parameterized for mid-1970's food web, suggest that both species mainly occupy the third and fourth trophic levels. Trophic transfer efficiency was estimated to be 12% with the bulk of energy flow coming directly from phytoplankton as opposed to being detritus-based.

north Sea: Diet and trophodynamic analyses in the North Sea indicate a more diverse role for forage fish species. This report includes new diets studies performed on sandeel, anchovy and sardine, and includes a review of previous work on sprat, herring and mackerel. Besides a reliance on a greater diversity of zooplankton, these diets studies shed light on an important difference in the trophodynamic role of forage fishes. As opposed to the Baltic Sea, where a lack of predator-prey overlap largely excludes fish eggs and larvae, 'traditional' North Sea forage fishes such as mackerel, herring may intensively feed upon fish early life stages. In the newly re-established clupeids (anchovy and sardine), anchovy is a particularly opportunistic feeder that often preys upon fish eggs and larvae. For example, 60% of anchovy stomachs collected from this species in the German Bight contained fish larvae (most likely sprat). The intensity of feeding upon eggs and larvae varies seasonally and spatially, indicating the importance of understanding forage fish movements and predator-prey overlap.

competition (T1.3)
an in situ process study on the respective role and quantitative impact of predation and hydrographic conditions on sprat and cod egg mortality in the Bornholm Basin, Baltic Sea, revealed that the intra- and interannual importance of both mortality factors changed after the large-scale environmental variation caused by the inflow of North Sea waters in early 2003.

two different approaches are being utilized to assess the potential niche overlap of anchovy, sprat and herring in the North Sea. The first summarizes all available information on diets of the three species in the North Sea based upon gut content analysis of juvenile and adult life stages collected at fronts, well mixed and stratified waters (primarily in the German Bight, southern North Sea, an important nursery ground for forage fishes). That analysis quantifies potential overlaps (as mentioned in The second method estimates realized overlaps by examining both prey niche as well as spatial overlap of sprat, anchovy and herring.

the second method utilizes two main approaches:
1) optimal foraging sub-routines implemented within individual-based models for each of the three species, and
2) field data on juvenile and adult distributions – particularly during spring and summer.

although anchovy exhibits more diet flexibility and omnivory than either sprat or herring, the three species often consume the same prey sizes / species. The degree of spatial overlap will be critical for the strength of competitive interactions. Physiological differences leads to different life history scheduling of these species in the North Sea which minimizes competition during the early larval phase but not during late-juvenile phases (particularly for sprat and anchovy).

distribution and food overlap of sprat and sardine larvae were investigated during late spring 2003 on two transects covering the full range of environmental conditions in the German Bight. Larvae co-occurred on all stations investigated. The results on food overlap confirmed that sprat and sardine larvae share a wide range of prey types. Competition for food is possible and gut fullness and feeding success were similar in both species. However, direct competition seemed to be partly avoided by different preferences in ambient conditions.

the importance of large bivalve molluscs (red) as a prey item for plaice, dab and haddock has declined dramatically since the early 20th Century, whereas sandeels (yellow) have increased in importance, as have polychaete worms (green) to plaice and crabs (orange) to haddock. The importance of sandeels has also increased in whiting and grey gurnard stomachs (not shown), and these changes reflect real changes that are known to have occurred in the availability of certain prey items over the 100 year period.

long-term changes in the relative importance of forage fish species in the North Sea have been revealed by comparing recently digitised diet data for predatory fishes between the early 1900s, 1950s, and early 2000s. There was very limited evidence, in particular, of sandeels being eaten in the 1900s (when hard-shelled molluscs were still abundant in predator diet), but sandeels became increasingly important in the 1950s and are now hugely important especially for whiting. Meanwhile the relative importance of herring has decreased.

the diet of anchovy (Engraulis encrasicolus) in the North and Baltic Seas was studied using stomach analysis on samples from four sampling events in different areas. Zooplanktivory was confirmed; the most frequent prey items (in over 40% of stomachs) were copepods, malacostracan larvae and fish larvae. In the Baltic Sea, Paracalanus spp. And Pseudocalanus spp. Were important in relative terms; in the German Bight, Temora spp. Dominated the stomach contents. Relative abundances of prey items varied with area more than absolute abundance or presence absence of items. Moreover, the level of resolution of prey categories influenced which prey categories were considered to be most important in driving variability in stomach content. Anchovy diet is broad across the seasons, years and areas sampled, suggesting that it is not a specialist feeder in the North Sea. The similarity of diet between anchovy and other clupeids, as well as anchovy consumption of larval fish, makes the new increased anchovy population a potential intraguild predator of commercial species like herring.

in addition to the analysis of the analysis of intraguild predation between anchovy and herring in the North Sea the issue has been analysed at two other level: 1) generic relevance and 2) anchovy - sardine in the Bay of Biscay. For the first case a paper is in press in Marine Biology, whereas a second paper is in preparation.

T1.4 Predators (T1.4)
A number of manuscripts produced within FACTS, focusing more specifically on predator–prey links between forage fish and predatory fishes, seabirds and marine mammals, with emphasis on quantifying removals by predators. These predator-prey relationships were studied in the 4 case study areas of FACTS : the Barents, Baltic and North Seas, and the Bay of Biscay.

T2.1 Parameterization and improvements of models (T2.1)
the Barents Sea can be broadly characterized as being divided into two regions; a north-eastern, Arctic water dominated area, and a south-western, Atlantic waters dominated area. The north-eastern area is relatively close to the ice edge, with cold waters and dominated by cold water adapted species. Forage fish are predominantly polar cod and feeding capelin, with relatively few predators. The south-eastern region has warmer waters, with a more diverse assemblage of fish including young Norwegian Spring Spawning herring. The main predator species (including cod) are predominantly in the south-western region, with only a small fraction of the populations feeding in the north-eastern area.

the direct and indirect effects of changes in environment, fishing and top predators (T2.2)
all the multispecies Fmsy values for Eastern Baltic cod, Central Baltic herring and Baltic sprat are higher than the single species values. Particularly for cod and sprat higher Fs give very similar yields on the long term and will give lower SSBs and in some cases risks of stock decline to the 'lower biomass' reference points (that is a first suggestion for a lower SSB to avoid impaired recruitment). Model results indicate that although higher Fs on Eastern Baltic cod give little increase in cod yield, a higher cod F gives higher yields from Baltic sprat and Central Baltic herring. As current modelling for Fmsy does not include any structural uncertainty, risks of stock decline and impaired cod recruitment will be higher than those estimated. The presence of year-year constraints in change in cod TAC increases the variability in stock size and the increases are greater in a multi-species system (for detailed results, please cf. STECF 2012; in this report only the main results will be presented). The present distribution pattern, with a limited distribution range for cod (concentrated in the southern area) and basin wide distribution for herring and sprat (but mainly concentrated in the northern areas, at least in some seasons), implies that an increase in F on cod, not necessarily will result in increasing Baltic wide clupeid stock sizes. Conversely a decrease in F on cod will not necessarily result in a decrease of the Baltic clupeid stock size if it will not be accompanied by a cod expansion to northern areas. However, cod cannibalism will be higher and limited growth of cod due to food deprivation will become a bigger problem.

the role of forage fish on ecosystem stability and biodiversity (T2.3)

Bay of Biscay: OSMOSE (Object-oriented Simulator of Marine ecOSystems Exploitation) is a multispecies and Individual-based model (IBM) which focuses on fish species. We have implemented the model to the ecosystem of the Bay of Biscay in order to investigate trophic interactions between different species. This model assumes opportunistic predation based on spatial co-occurrence and size adequacy between a predator and its prey (size-based opportunistic predation). It is therefore a suitable tool to explore processes such as competition, intraguild predation or cannibalism. Variability in fishing pressure is used to test the sensitivity of the system to changes in external forcing.

generic: The size-spectrum model developed in 2.1 has been optimized and implemented as a web application (see online). The model quantifies how management actions targeted on one group of species propagates to the rest of the community.

the website has open access, and it is possible for anyone to analyse the impact of fishing on the whole ecosystem based on three different fleets:
(i) an 'industrial' fleet targeting fish with an asymptotic size less than 100 g;
(ii) a pelagic fleet targeting small pelagics with an asymptotic size in the range 100 g to 4 kg;
(iii) a demersal fleet targeting fish with an asymptotic size larger than 4 kg.

we have initiated work on a systematic characterization of the ecosystem response to fishing on forage fish with the 'industrial' fishing fleet.

the role of forage fish on ecosystem stability and biodiversity (T2.3)
for the Baltic Sea, an Ecopath with Ecosim Baltic Proper food–web model was developed to simulate and better understand trophic interactions and their flows. The model enables the quantification of the flows through the food-web from primary producers to top predators including fisheries over time. The model is able to explain 51% of the variation in biomass of multiple trophic levels and to simulate the regime shift from a cod dominated to a sprat dominated system. Using a spatially disaggregated statistical food-web model, we applied a metacommunity perspective on source-sink dynamics and examples of management alternatives for Baltic Sea fish stocks. Fitted to area specific time-series of multiple abiotic and biotic variables our analysis showed clear regional patterns of inter-population exchange between all species and areas, dependent on local species interactions (density dependence, competition and predation), commercial fishing and climate effects (e.g. temperature, salinity and oxygen) on recruitment and survival.

economics into ecosystem models (T3.1)
bay of Biscay: A bioeconomic model has been developed for the Basque purse seiner fleet which catches a range of pelagic fish. Since Basque purse seiner fleet shares the resources of the Bay of Biscay, others fleets have to be included in the model. Given that, the model besides being multistock, is a multifleet model. The compromise between data and objective resulted in the selection of five species and two principal fleets. Other fleets are included in the model but in more general terms. The temporal resolution of the model has been defined on the basis of the highest resolution possible that will allow us to model in a realistic way the behaviour of the fleets. Although fishermen take the fishing decision day by day, a daily resolution was not possible due to the lack of data, thus a monthly resolution was established.

the following functions have been defined: With respect to biomass dynamic population two possible functions has been considered; Biomass Dynamic Population Growth (Pella and Tomlinson Model) and Age Structured Population Growth function. In the economic side of the model, an effort model which simulates the short term behaviour of the sequential pelagic fleet has been created, two price models have been considered (fixed price and price dynamic which depends on the catches), production function is defined according to Cobb Douglas production function. We assume here that the catch in the middle of the season as in Pope's approximation. Costs functions have been also defined, separating fixed costs (are fixed along the period) and variable costs that depend on the effort allocated to each species and on the type of vessel. And finally a Capital function has been included (currently only 'Fixed Capital Function' has been is set and the 'Capital Dynamic' will be developed).

combining all functions cited above, the model structure has been built. This model, named FLBEIA, has been developed in FLR. It is an R package developed to conduct Bio-Economic Impact Assessment under Management Strategy Evaluation framework. The package is built on top of existing FLR packages, which provides function to condition the simulations, to run them and to analyse the results.

the novelty of the current model lies in the fact that the biological side interacts with economic side with a monthly resolution and also lies in inclusion of covariables operating model that will allow us to include relationships between species. In that sense, the choice that makes each skipper of the target species will affect not only to this species, but also to those species that are related to the target species. For this reason fishermen behaviour or fishing pattern has been very well defined. To this end the involved fleets, especially Basque fleets, have been analysed in detail to find out what their fishing behaviour is month by month; when the target species are available for those fleets, how many days per month each fleet operates, when they change the fishing gear, etc. With this exhaustive analysis of the fleet a Sequential Fisheries Behaviour model which reproduces the Basque pelagic fleets’ behaviour has been developed.

baltic Sea: Traditional fisheries economic models have been criticized by biologists, especially if results are obtained applying simple biomass models. Biological stock assessment models, on the other hand, are more sophisticated with regard to biological content and include the population age structure but rarely take into account economic objectives. To overcome these shortcomings, we included a full age-structured population model based on the SMS in a bio-economic optimization problem.

north Sea: An external economic module to the SMS model has been constructed. This module takes the stock biomasses resulting from running the SMS model and evaluates the economic outcomes for the relevant fleets given these biomasses. The economic cost benefit module asks how the fleets fishing forage fish in the North Sea, and additionally the total Danish fishing fleet, will reallocate effort optimally when/if the fishing mortalities for forage fish are changed. By optimally is meant that the model optimizes the fleet profits by reallocating effort between fleets and species when faced with changing fishing patterns for forage fish. The cost benefit module added externally to the SMS model answers the question 'if the fishing patterns for forage fish are changed, what will be the optimal reallocation of fisheries among fleets and species for the fleets involved in the forage fish fisheries'. The module answers this question by maximizing the total fleet profit with respect to effort allocation in and between fishing fleets. Additionally the module asks what the tradeoffs will be for the total Danish fishing fleet of changing the forage fish fishery patterns.

generic: The generic ecosystem model was revised and formulated to fit economic analysis. A side effect of this work was a significant improvement in the performance of the program in form of increased speed. Traditionally, there is a difference in how fishery is modeled from an economic production perspective and how fishery is modeled in biological models. FACTS succeeded in formulating the fishery in such a way so the two approaches can be joined.

cost-benefit analyses (T3.2)
A bioeconomic model for a North Sea fishery has been coupled to recruitment parameters and species interactions matrices for the multi species Schaefer model estimated with the SMS model. The bioeconomic model includes detailed economic income and cost structure evaluation for each fleet segment in the Danish North sea fleet, and additionally the non- Danish fleet segments targeting forage fish in the North Sea has been added to the model, also including detailed income and cost structure. The model performs optimisations of the total net income of both the Danish fleet and the non-Danish forage fishery fleet, given the forage fish management scenarios, by reallocating effort between fleet segments. As such the model asks what will be the economically optimal fleet structure, given proposed management scenarios, and in this case what will be the costs and net income of the Danish North Sea fleet and the overall forage fishery fleet in the North sea. Two management scenarios have been investigated; increasing the forage fish fishing mortalities one by one to (or close to) the fishing mortality in maximum sustainable yield (FMSY) and decreasing (restricting) the forage fish fishing mortalities one by one to below half of the fishing mortality in status quo (2007).

we compute optimal management under three cod stock scenarios:
(a) a reference case, using the situation in 2008,
(b) a 'low cod' case using historically low predator abundance values, and
© a 'high cod' case using historically high predator stock estimates.

cost of conservation (T3.3)
for the Barents Sea, we fit a multi-species dynamic model to time-series population and diet data in the Barents Sea, and investigate the emergent effects of marine mammal predation on fisheries for cod, capelin and herring. Scenarios in which marine mammals are reduced in number are run, and the effects on fisheries under contrasting regimes of management are explored. We also investigate the consequences of these different fishing scenarios for the marine mammal populations.

once the GADGET model had been fitted to the fish, fishery and marine mammal data, the following scenarios were investigated using simulations in order to explore the costs of marine mammal predation in this system. (a) The original model, fitted to time series fish survey and stomach data for the Barents Sea (the „base case? scenario) (b) All conditions remain as for the original model runs, but the abundance of harp seals was reduced by 25% in all years © All conditions remain as for the original model runs, but the proportion of the minke whale migrating into the Barents Sea each summer was reduced by 25% in all years.

marine mammal predation is an important factor in the Barents Sea ecosystem, where the absolute levels of consumption may be considerable (for example harp seals may consume up to 13% of the total cod biomass in a given year). This suggests the potential for the seals to act as serious competitors with the fishery. Increased marine mammal populations inevitably result in increased net fish consumption, but their predation on cod, which is itself an important predator in the system, can also diminish cod predation pressure on fish at lower trophic levels and this effect may compensate for increases in marine mammal predation: this can be the case for example in the cod-capelin-minke whale interaction. Equally the impact of predation on forage fish can have indirect impacts on the cod, in addition to direct predation from marine mammals. The scenario experiments also suggest that marine mammals as generalist predators may play an important role in controlling and possibly stabilizing oscillatory dynamics in the system, particularly those involving cod.

to investigate the impacts of fisheries on marine mammal populations, we explored the consequences of changes in management for fish stocks and their dependent predators. The Gadget multispecies model was used to examine the possible effects of a non-size-selective fishing regime ('flat fishing') on the forage fish and predators in the Barents Sea. In scenario

(d) a forecast approach was adopted. We predicted the effects on fish stocks of a continuation of current fishing strategy for 2011-2020 and contrasted this with 'flat fishing' at a plausible level of F=0.2 for the same time period.

integrating process and generic knowledge (T4.1)
due to their tight coupling to the dynamics of lower trophic levels and their high rates of reproduction and growth, forage fish are excellent bio-indicators of climate change worldwide and are extremely important to food web dynamics and fisheries catches and food security world-wide. The trophodynamic and commercial role of forage fishes highlights that a variety of pressures interact to affect forage fish species. These pressures include changes in 1) bottom-up processes due to climate-driven changes in physical-biological coupling within lower trophic level examined in FACTS 1.1 as well as 2) top-down processes due to both climate and fisheries impacts on the distribution and abundance of competitors (intraguild) and predators of forage fishes.

system-specific recommendations (T4.2)

(i) Baltic Sea
the predominant forage fishes are sprat and herring and the commercially most relevant predatory fish species is Baltic cod.

Q1 Does the strength of the impact of forage fish on the Baltic Sea food web differ between sub-systems and how do such regional differences influence advice on ecosystem-based management? - Baltic cod feed primarily on sprat and to a lesser extent on herring (and if these are scarce, cannibalise on juvenile cod); in turn, herring and sprat feed on cod eggs. Previously, spatial overlap between these three species was extensive implying tight trophic links, but this has changed and currently, most herring and sprat are outside of the predatory reach of cod (too far north). This implies wide regional differences in trophic links.

Q2 How important are density-dependent feedback loops between forage fish, their prey and predators and how do these influence ecosystem stability? - The currently limited spatial overlap between sprat/herring and cod implies limited density-dependent feedback loops between these forage fish and predator species, unless cod expand into more northerly areas and overlap with herring and sprat becomes substantial. If cod have limited spatial overlap with key prey forage fish, higher local cod densities will imply higher cannibalism.

Q3 What are the economic costs and benefits (for fisheries and other ecosystem services) of changes in the management of Baltic Sea forage fish, taking into account interactions with top predators (cod and seals) and climate induced ecosystem changes? - The cod fishery is the economically most important fishery in the Baltic. Under strong predator-prey coupling, economically optimal multispecies management would lead to a large cod stock, a medium-sized herring stock, but a very low sprat stock. Increases in sprat target SSB, e.g. in order to avoid recruitment impairment or to take ecosystem considerations into account, would have to be paid in terms of forgone profits from the cod fishery. Under current spatial de-coupling, the shadow price of fishing sprat in the south-western part (the area where overlap with the cod stock exists) might be high, as growth and reproduction of cod is negatively impacted by food shortage. Quantification is, however, currently not possible. Anticipated climate change will increase sprat stock productivity owing to higher recruitment success. Related higher potential exploitation rates are enforced due to lower productivity in the cod stock, resulting in lower predation mortality.

(ii) Barents Sea
the forage fishes of highest commercial importance are capelin (exploited within the Barents Sea) and Norwegian spring-spawning herring (residing in the Barents Sea as juveniles, but not exploited until fish have left and migrated into the Norwegian Sea); the third forage fish, polar cod, mainly occurs very far north and is not exploited at any great scale.

Q1 Is the difference in the strength of the bottom-up effects between the capelin collapses due to differences in the amount of other fish prey available to the predators? - This hypothesis was supported by analysis of combined survey-haul and acoustics data sources on abundance dynamics of capelin, cod, and the potential alternative prey species for cod; as well as cod stomach contents data. These revealed that during the 1980s capelin collapse, alternative cod prey species were scarce, cod cannibalism was high, and condition poor; the cod stock declined. During two later capelin collapses (mid-1990s and mid-2000s) alternative prey species were available, and impacts on the cod stock were mild.

Q2 Why was there good recruitment of capelin in the Barents Sea in some years even though there was a large amount of herring was present? - From the 1970s to early 2000s close negative relationships between capelin and herring abundance in the Barents Sea were documented (unexpected as herring prey on juvenile capelin), but the relationship appears to have broken down. This remains not fully understood but may be attributable to local differences in spatial distribution and/or management of capelin fishing pressure in association with cod dynamics.

Q3 Why do herring leave the Barents Sea as 3-year-olds in some years, while in other years, most stay until age 4-5? - Differences in the age at migration likely relate to significant between-year-class differences in juvenile herring growth rates, which are both density- and temperature-dependent. Year-classes can moreover be different in the proportions that as juveniles reside in the Barents Sea (where growth tends to be slow), more southerly Norwegian fjords (fast growth), or the Norwegian Sea itself (variable juvenile growth rates).

(iii) Bay of Biscay

the predominant forage fishes are sardine, anchovy and sprat, which show various trophic interactions including food competition and intraguild predation. Horse-mackerel and mackerel also occur in large numbers. Hake is a very important predatory fish that preys on these forage fishes.

Q1 Is increasing fishing pressure on sardines and sprat likely to help restore the anchovy? - These species interact through diet overlap (food competition) and intraguild predation. On the basis of a Mixed Trophic Impact analysis of the levels of food competition, it is most likely that increasing fishing pressure on sardine and sprat will not to help restore the anchovy population in the Bay of Biscay. This differed, however, from the result derived from analysing intraguild predation between these species. Sardine prey extensively on anchovy eggs in the entire area, and sprat do so in the Gironde area. Hence if increased fishing pressure on anchovy or sprat leads to reduced spatial overlap with anchovy and hence less egg predation, a positive effect on anchovy stock abundance may be expected.

Q2 Does forage fish overfishing modify the stability of the Bay of Biscay ecosystem? - Different ecosystem model types and structures were tested to examine stability of the Bay of Bay ecosystem, and the following management recommendations for small pelagics were derived based on conclusions that were consistent across model structures and types (i) a rise (decline) in primary productivity will drive an increase (decrease) in small pelagic fish abundance; (ii) an increase in piscivorous demersal fish abundance (hake) will create a high risk of decline in small pelagic fish abundance. Simultaneous single-pressure management measures might conflict. The fisheries for hake, anchovy and sardine need to be considered together.

Q3 What is the most cost effective way to increase the profitability of the pelagic fishery with limited effects on the ecosystem as a whole? - Recently the French forage fish fishery consisted primarily of pelagic trawlers and purse seiners. The profits of the French purse seine fleet were negative during 2004-2006 when the anchovy fishery was closed, but the fishery quickly recovered thereafter on its own even before the reopening of the fishery in 2010. The French pelagic trawl fleet remained highly profitable in all years despite the closure. Given the success of this fleet it is hard to conceive how the profits could be much further improved through additional management intervention. Anchovy discards were very low (less than 1% by value). Thus landing the discarded anchovy would not have improved the profits of the French pelagic fleet, and a discard ban is therefore unlikely to improve the profitability of the pelagic fishery.

(iv) North Sea

the North Sea has a broad portfolio of forage fish species: herring is primarily targeted by human consumption fisheries; sandeel, sprat and Norway pout are targeted by extensive industrial fisheries; and anchovy and sardine only occur at low biomass levels and are only exploited by very small-scale (but increasing) fisheries.

Q1 What are the costs and benefits of maintaining small-mesh fisheries on herring, sprat, sandeel and Norway pout in the North Sea - FACTS scientists created a bioeconomic cost-benefit model consisting of an economic module for the North Sea forage fishing fleet, added to a multispecies model. This suggested that relative to a 2007 'status quo', the North Sea forage fishing fleet gained (higher net income) when F for forage fish species was increased to near or equal to FMSY. In modelling scenarios where F values for forage fish species were limited below ½F2007, the total North Sea FF fleet obtained a lower contribution margin and lower net income relative to the 2007 'status quo'.

Q2 How would reductions in herring, sprat, sandeel and Norway pout populations impact seabirds, marine mammals, and predatory fish? - Seabirds are most dependent on forage fish and vulnerable to forage fish reductions, owing to specialised diet needs (high proportion of forage fish ind diet) and distributional constraints (seabird foraging is limited to vicinity of colonies, especially during the breeding season). They are followed by marine mammals, and predatory fish are considered less vulnerable to local forage fish depletions owing to greater flexibility to switch between prey species. Amongst different forage fish species, sandeel is crucial prey to the greatest diversity of top predators (especially shag, great skua, Sandwich tern, puffin, guillemot, kittiwake, minke whale, harbour and grey seal).

Q3 Should forage fish populations be managed as a core component of the ecosystem and what is their role in ecosystem stability and dynamics? - Motivations for managing forage fish populations as a core ecosystem component rather than in isolation as single species, include the multitude of links highlighted in the FACTS project: with climatic variability, plankton dynamics, many species of predatory fish, seabirds and marine mammals, as well as the various trophic linkages between the different forage fish species or different life stages. Foodweb models suggest that reducing fishing mortality on forage fish may not necessarily lead to larger stocks of piscivorous fish, especially if their early life-stages compete with forage fish for zooplankton resources. In complex systems such as the North Sea, changes in the impact of fisheries on forage fish may have potentially complex (and perhaps unanticipated) consequences on other commercially and/or ecologically important species.

model responses to different management scenarios (T4.3)
within this deliverable, we evaluated the different ecosystem modelling approaches with respect to their ability to realistically reflect the ecosystem dynamics in response to different management scenarios of forage fish and hence their usefulness as operational tools for ecosystem-based fisheries assessment and management activities. This work includes comparing the different modelling approaches applied in FACTS with respect to their ability to capture dynamical properties as influenced by differences in trophodynamic complexity (e.g. biodiversity and the strength of trophic coupling). The work also involves identifying key metrics (model outputs) that can be used as endpoints for weighing potential management options.

the results of this exercise are given in the bulletin list below. Each bulletin point corresponds, according to the nature of the deliverable, to an already published manuscript or a manuscript in press:

1.In five well-studied ecosystems, we found that fishing these species at conventional maximum sustainable yield (MSY) levels can have large impacts on other parts of the ecosystem, particularly when they constitute a high proportion of the biomass in the ecosystem or are highly connected in the food web. Halving exploitation rates would result in much lower impacts on marine ecosystems while still achieving 80% of MSY.

2.We demonstrated how the biological ensemble modelling approach makes it possible to evaluate the relative importance of different sources of uncertainty in future species responses, as well as to seek scientific conclusions and sustainable management solutions robust to uncertainty of food-web processes in the face of climate change.

3.Qualitative modelling provided robust conclusions regarding the effects of pelagic fisheries on the bentho-demersal food chain and the effects of demersal fisheries on the pelagic food chain.

4.We proposed a typology of methods and approaches that are currently used, or could possibly be used for making large-scale ecosystem comparisons.

5.We combined life-history invariants, metabolic scaling and size-spectrum theory to develop a general size- and trait-based theory for demography and recruitment of exploited fish stocks. Larger species have a higher egg production per recruit than small species. This means that density dependence is stronger for large than for small species and has the consequence that fisheries reference points that incorporate recruitment do not obey metabolic scaling rules. Even though small species have a higher productivity than large species their resilience towards fishing is lower than expected from metabolic scaling rules.

future Research Needs (T4.4)
during the FACTS project and the FACTS symposium, following 10 points of major research needs in forage fish ecology and management have been identified:

1) Do all populations / stocks share the same (perhaps unique) ability to rebound from extremely low stock sizes? Can our exploitation patterns / intensity affect this trait - putting ecosystem structure and function at risk? Various harvest strategies need to be tested including estimations of risk which account for the unique properties of forage fish.

2) As the productivity of forage fish is dependent on life cycle closure, what habitats and their connections are keys to the survival and reproduction of exploited stocks?

3) Can indicators based upon aggregate measures across the ecosystem be appropriate for management at the local or sub-regional levels? What is the appropriate spatial scale of aggregation needed to effectively manage forage fish stocks in an ecosystem context?

4) There is a need to investigate different regulation systems and associated incentives applied to fisheries on forage fish within the context of the ecosystem approach.

5) Better empirical data streams have to be developed to detect indirect effects of forage fish fisheries on predators, ideally in an adaptive management framework. For example, up to date stomach contents data are desperately needed. This issue includes also the need for data assimilation methods suitable for different types of data and models.

6) There is a need for testing the robustness of management via biomass set asides or exploitation rates, including spatial issues. Furthermore, how to include those in bio-economic models? Compensation? Payment for ecosystem services?

7) There is a need to include the market implications of changes in forage fish stocks (e.g. value chain analyses). Evaluate the role of green fisheries and conservation pressures.

8) We need to develop tools to simultaneously explore all management strategies within GES. Especially, there is a need to investigate whether the current governance structures can handle the trade-offs in the ecosystem?

9) We need to investigate if ecosystem and economic risk can be reduced by by management of the portfolio (guilds) of stocks in an ecosystem approach.

10) We need to develop a suite of modelling tools that include not only complex (end-to-end), but models of intermediate complexity that are more tractable and practical.

potential Impact:

potential impact and main dissemination activities

the project has addressed the work programme topic FP7-KBBE-2009-1-2-14 'Sustainable use of seas and oceans: importance of foraging fish in the ecosystem' under area 2.1.2 'Increased sustainability of all production systems (agriculture, forestry, fisheries and aquaculture); plant health and crop protection', activity 2.1 'Sustainable production and management of biological resources from land, forest and aquatic environments' of Cooperation Programme Theme 2 'Food, Agriculture and Fisheries, and Biotechnology', call: FP7-KBBE-2009-3. As requested in the work programme, the project has assessed the importance of forage fish in European ecosystems and economy. This assessment can be used to weigh the cost and benefits of different levels of exploitation of the forage fish resources and hence provide a basis for decision makers to integrate forage fish dynamics and their exploitation into the ecosystem approach of the CFP. This has brought ecosystem orientated management in the European waters a significant step ahead.

the project has done so by:
-Providing knowledge basis for policy: The project has provided important knowledge tools for the Common Fisheries Policy, the Marine Strategy Framework Directive (adopted in June 2008) and future EU policies related to ecosystem based management of fisheries. It has enhanced the competencies of key scientific personnel involved with international scientific advice thus reinforcing the implementation of ecosystem based management principles.
-Significantly advancing the knowledge of forage fish population dynamics in European waters: By compiling existing, long-term data on the abundance of various forage fish species and their prey and filling gaps in existing knowledge with new analyses of gut contents and stable isotopes of forage fish, the project has quantified the trophic interactions between forage fish and their prey and further investigate the competition among forage fish species. Changes in potential habitat characteristics and the prey requirements have been quantified using physiological-based habitat modelling to determine likely areas of distribution.
-Estimating the consequences of forage fish population size and structure on top predators: The project has used multispecies models to investigate the effect of forage fish exploitation on the diet and food intake and, through bioenergetic models, the effect on reproductive potential of top predators. This has enabled to identify the ecological requirements and the economic costs for fisheries on forage fish and their predators of sustaining or reaching a specific reproductive potential of top predators along.
-Estimating economic effects of forage fish exploitation using integrated economic analysis of ecosystem models: The project has estimated the economic effects of forage fish exploitation through a model describing economic gain of fishing while using ecosystem models to describe the abundance of both forage fish and interactions with other components of the ecosystem. Integration of a model of species interactions and a model of market price of landings and costs of the fishery is unique and has provided an estimate of economic consequences which takes account of species interactions in contrast to traditional bio-economic single species models.
-Estimating the effect of forage fish exploitation on biodiversity and ecosystem stability: By quantifying the interconnections between forage fish and community level metrics, the project has estimated the relationship between changes in forage fish abundance and biodiversity and ecosystem stability. This will allow assessing the effect of future exploitation plans on these ecosystem characteristics and hence will allow the benefit of exploitation by fisheries to be weighed against the risk of ecosystem instability and regime shifts and declining biodiversity.
-Delivering general rules for forage fish management in an ecosystem orientated management: By analysing several ecosystems as well as generic models, the project has provided the basis for deriving general rules applicable to forage fish fisheries in European waters. This type of general knowledge required internationally coordinated investigations of forage fish dynamics in a range of different ecosystems rather than local studies. The general rules will be particularly valuable in areas where the population dynamics of forage fish are less well known or where the knowledge of the effects of forage fish abundance on top predator dynamics is limited.

the project has provided benefits for scientists, scientific advisors, policy makers, and stakeholders such as fishermen, NGO's and the European public:
-Scientists have benefited from establishing a scientific network of associated partners connecting forage fish research throughout Europe: Not only has knowledge been built up within the institutes participating in the project and disseminated to the scientific community, but a network connecting forage fish research throughout Europe has been formed through the collaboration with associated partners.
-Scientific advisors have benefited from making the quantitative knowledge available where it is needed: The key participants in the FACTS project are active members of a range of ICES working groups. The results produced in the FACTS project have therefore been introduced directly in the ICES working groups and contribute to the scientific advice given by ICES concerning the effects of exploitation of forage fish on a range of different economic and ecosystem levels.
-Policy makers have benefited from the tools and concepts developed in FACTS: The critical transition from a fisheries management system that in its theoretical roots is based on the dynamics of single exploited fish populations ignoring ecosystem effects.
-Stakeholders, such as fishermen, NGOs and the public have benefited from operationalising the ecosystem management: more effective and user-friendly management programmes that are better tailored to specific ecosystem related questions, e.g. the costs of conservation, have integrateed the dialogue between the different users of the European marine ecosystems and in the end enable a more effective use of European marine resources, while maintaining them in a healthy status.

the EU is committed to the contents of the Johannesburg World Summit on Sustainable Development Plan of Implementation, including targets to restore depleted fish stocks by 2015. To this end, FACTS has advanced the state of the art in development and application of strategic medium- to long-term fisheries management plans and their ecosystem effects. The project has contributed directly to the successful implementation of the CFP and the Marine Strategic Framework Directive with respect to ecosystem management.

the project has strengthened the European cooperation in this area of marine resource management, by creating a forum on forage fish, their ecological role and economic importance considering different harvesting strategies and climate change. The cross fertilisation of expertise across Europe has put Europe internationally in the fore front of fisheries and marine science related to forage fish being of fundamental importance for integration of the ecosystem approach into fisheries management and the success of the CFP.

the consortium has undertaken the following actions for using and disseminating knowledge to all Member States during the lifetime of the project and afterwards:
-Establishment of the project website during the first six months of the project, under the project's acronym with an extensive public section pen to all Member States.
-Publications in scientific popular press
-Publications in scientific journals
-Publications in daily/weekly press, specialized magazines and practitioner journals
-Issuing of press releases to local, national or international press at suitable occasions
-Organization of media events such as press releases, conferences, workshops, information days, for example on the occasion of a project meeting, as well as ensuring access to such events via webcasts.
-Production and dissemination of information dedicated to appropriate media, e.g. a printed brochure, and newsletters
-Participation at conferences, such as the ICES Annual Science conference.

the dissemination to advisory committees has been through direct person mediated knowledge transfer as scientists participating in the FACTS projects are members of several ICES advisory working groups including ACOM, WGSAM, WGFE, HAWG, WGMME, WGIAB and ICES Advice Drafting Groups as well as STECF subgroups. This has ensured that results are introduced in scientific advice with the shortest possible delay as FACTS scientists will bring working documents and presentation to the upcoming working group meetings.

the consortium has acknowledged that a pan-European approach to the operationalisation of the ecosystem management would be desirable, and attempts to come closer to this necessary broadness by building a network of associated partner institutes. Furthermore, similar initiatives are undertaken in the Northwest Atlantic. In order to guarantee an optimal flow of ideas and methods, the Consortium decided to include also a Canadian institute as associated partner.

communication and exchange of information among Consortium- and associated partners will be:
-Intensive face-to-face communication through the regular meeting and workshops. The FACTS project foresees funding of the travel expenses for additional partners.
-Direct communication of all project partners through electronic mail as and when necessary.
-Through the partners' website (extranet) operational within the first six months, under the projects acronym, for distribution and management of project documents that will be made available in electronic format to the Consortium and associated partners only.

internal project workshops have allowed the different participants to have a general view on the progress on the project. The organisation of the workshops and project meetings in the different participating countries has allowed the project partners:
-To better understand the specific national needs and barriers with respect to ecosystem management in specific regions against European and global background
-To estimate under what conditions the results of the project can be transferred into management actions with a view to assessing the probability and time line of introducing an ecosystem orientated management in that region.

list of Websites: