General enquiries on this form should be made to:randd.defra.gov.uk/Document.aspx?Document=10587_MF1202... · Web viewMarine Ecology Progress Series 253, 1–16. Hall SJ, Collie JS,

General Enquiries on the form should be made to:Defra, Procurements and Commercial Function (Evidence Procurement Team)E-mail: [email protected]

Evidence Project Final Report

EVID4 Evidence Project Final Report (Rev. 06/11) Page 1 of 24

mailto:[email protected]

NoteIn line with the Freedom of Information Act 2000, Defra aims to place the results of its completed research projects in the public domain wherever possible. The Evidence Project Final Report is designed to capture the information on the results and outputs of Defra-funded research in a format that is easily publishable through the Defra websiteAn Evidence Project Final Report must be completed for all projects.

This form is in Word format and the boxes may be expanded, as appropriate.

ACCESS TO INFORMATIONThe information collected on this form will be stored electronically and may be sent to any part of Defra, or to individual researchers or organisations outside Defra for the purposes of reviewing the project. Defra may also disclose the information to any outside organisation acting as an agent authorised by Defra to process final research reports on its behalf. Defra intends to publish this form on its website, unless there are strong reasons not to, which fully comply with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.Defra may be required to release information, including personal data and commercial information, on request under the Environmental Information Regulations or the Freedom of Information Act 2000. However, Defra will not permit any unwarranted breach of confidentiality or act in contravention of its obligations under the Data Protection Act 1998. Defra or its appointed agents may use the name, address or other details on your form to contact you in connection with occasional customer research aimed at improving the processes through which Defra works with its contractors.

Project identification

1. Defra Project code MF1202

2. Project title

A Strategic Evaluation of Ecosystem Models in Support of Fisheries Management (STEEM)

3. Contractororganisation(s)

Centre for Environment, Fisheries & Aquaculture Science (CEFAS)Pakefield Road,Lowestoft.SuffolkNR33 0HT

54. Total Defra project costs £ 204,529(agreed fixed price)

5. Project: start date................ 01/04/2008

end date................. 31/03/2011


6. It is Defra’s intention to publish this form. Please confirm your agreement to do so....................................................................................YES NO (a) When preparing Evidence Project Final Reports contractors should bear in mind that Defra intends that

they be made public. They should be written in a clear and concise manner and represent a full account of the research project which someone not closely associated with the project can follow.Defra recognises that in a small minority of cases there may be information, such as intellectual property or commercially confidential data, used in or generated by the research project, which should not be disclosed. In these cases, such information should be detailed in a separate annex (not to be published) so that the Evidence Project Final Report can be placed in the public domain. Where it is impossible to complete the Final Report without including references to any sensitive or confidential data, the information should be included and section (b) completed. NB: only in exceptional circumstances will Defra expect contractors to give a "No" answer.In all cases, reasons for withholding information must be fully in line with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.

(b) If you have answered NO, please explain why the Final report should not be released into public domain

Executive Summary7. The executive summary must not exceed 2 sides in total of A4 and should be understandable to the

intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together with any other significant events and options for new work.

The primary aim of this project has been to test, refine and critically evaluate the usefulness of different modelling frameworks for addressing ecosystem and multispecies considerations in fisheries science and management. It has attempted to assess how important the interactions are between predators and their prey in comparison with other sources of uncertainty in fisheries models, and to help predict ecosystem knock-on effects of management actions. The project has primarily focused on two case study regions, (i) the North Sea, and (ii) the Celtic Sea/western-approaches.

This project has brought together fisheries-related multispecies and ecosystem modelling activities that were previously scattered across Defra and EU programmes. Much of the work has been carried out in collaboration with partner institutes across Europe through participation in relevant EU Framework 6 and 7 programmes, most notably: SIZEMIC (2007-2011), Incofish (2005-2008), MEECE (2008-2012), IMAGE (2008-2012), UNCOVER (2006-2010) and FACTS (2010-2012). The funding provided by Defra was ‘matched’ against activities under each of these programmes to provide significant ‘added value’.

This work has contributed to the ongoing debate on ‘ecosystem-based approaches to fisheries management’ (EAF), as well offering insight into the practicability of commitments to maintain or restore all fish stocks simultaneously to levels that can produce the maximum sustainable yield (MSY). A wide diversity of very detailed modelling work has been conducted over the past 3 years using a range of different ecosystem and multispecies modelling frameworks. Some of the main conclusions from the work conducted under this project are:

1. It is now possible to construct exceedingly complex models of whole ecosystems but these are not always the most appropriate tool for evaluating fishery trade-offs or consequences of management action.

2. A wide variety of tools are available to answer different types of question - most management questions can be tackled using one or other of the methodologies trialed in the MF1202/STEEM project.

3. Significant progress has been made with regard to constructing models for the Celtic Sea, but the information and approaches still lag behind those in the North Sea.

4. Ecosystem and multispecies models are most useful for ‘strategic’ planning (i.e. asking ‘what if’ questions) over longer time-scales, rather than giving predictions in the short to medium term.

5. Several modelling frameworks are available that can take account of spatial patterns of predator-prey


overlap. Such techniques (e.g Ecospace or ATLANTIS) can be used to predict the indirect ecosystem consequences of introducing offshore MPAs, as well as the resulting fishery displacement.

6. Multispecies modelling frameworks have been used to investigate trade-offs between fleet-segments, and consequently impacts on revenues, profits and costs in the fishing industry.

7. Models can be used to investigate the relative influence of climate change as well as fishery management on fish stocks and the wider ecosystem.

8. No model will ever provide all of the answers required.

A particularly important output of this project has been the construction of two, fully peer-reviewed, ‘Ecopath-with-Ecosim’ (EwE) models; one for each case study region. Mackinson & Daskalov (2007) constructed a model for the North Sea, which incorporates 68 functional groups and 12 fleet categories defined by the EU Data Collection Regulations. Parameterisation for each functional group was peer-reviewed by a panel of independent experts. In 2008-2010 this model was used to: (a) carry out spatial evaluations of proposed and existing marine protected areas, (b) to evaluate the potential multispecies implications of managing to single species MSY, and (c) to examine the impact of subsidies on the profitability of North Sea fisheries over the past 20 years. During the latter period of the MF1202 work programme, efforts were primarily focussed on construction of a detailed (64 box) EwE model for the Celtic Sea. This region had largely been neglected in earlier multispecies modelling initiatives, and it was necessary to construct a new model from scratch. This model has since been used to investigate the impact of fisheries management policies (e.g. elimination of discards) and climate change on seabirds in the region. Initial results have revealed diverse simulated consequences for seabirds as a result of different fisheries management regimes.

An important achievement in 2009/2010 has been the ‘coupling’ of the North Sea EwE model to biogeochemical models including GOTM-ERSEM. This has involved a major reprogramming effort (supported in-part by the EU FP7 project ‘MEECE’). The ultimate aim has been to enable predictive modelling of ‘bottom up’ climate effects as well as ‘top-down’ fishing effects on the functioning of the whole marine ecosystem.

Throughout the MF1202/STEEM project a number of different, size-based modelling approaches have been developed and applied in the North and Celtic Seas. In 2009, Blanchard et al. published a detailed description of a coupled benthic-pelagic size-spectra model applied to fish and invertebrate communities in the North Sea. The approach has been used to assess the theoretical consequences of external sources of mortality (e.g. harvesting) as well as changes in productivity at the base of the food-web. Removal of large predators resulted in steeper predator size-spectra and concomitant increases in their prey (small fish and detritivores). The model predictions were remarkably consistent with observed patterns in ‘real world’ ecosystems. A derived size-based model for 12 important North Sea fish species has been used (under the EU project IMAGE) to ’test-drive’ fish indicators, to see how they will be affected by changes in fishing mortality and future climate change. In particular, the model has been utilized to study the “large fish indicator” (LFI), a concept initially put forward by OSPAR but now considered to be vitally important for operationalizing the ‘ecosystem approach’ within the revised EU Common Fisheries Policy (CFP).

By considering a whole range of different modelling frameworks we have attempted to enhance and expand the tool-box on which Defra can draw and to demonstrate what is possible using these methods. We have built capability and have demonstrated that we now have a suite of tools with which we are able to address most issues of concern to policy-makers. The considerable progress made under this project has enabled Cefas to build further partnerships and to contribute to a wide range of newly agreed EU research initiatives, most notably: EuroBasin (2010-2014), VECTORS (2011-2015), MYFISH (2012-2016), and GAP2 (2011-2015). Development work (for example using the ’state-of-the-art’ ATLANTIS framework), as well as application of the full suite of models developed under MF1202, will be continued under the Defra Strategic Evidence & Partnership Fund (SEPF) Project ’Physics to Fish’ (2012-2016).


Figure 1. Schematic representation of the work programme under MF1202 (STEEM).

Project Report to Defra8. As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with details of

the outputs of the research project for internal purposes; to meet the terms of the contract; and to allow Defra to publish details of the outputs to meet Environmental Information Regulation or Freedom of Information obligations. This short report to Defra does not preclude contractors from also seeking to publish a full, formal scientific report/paper in an appropriate scientific or other journal/publication. Indeed, Defra actively encourages such publications as part of the contract terms. The report to Defra should include: the objectives as set out in the contract; the extent to which the objectives set out in the contract have been met; details of methods used and the results obtained, including statistical analysis (if appropriate); a discussion of the results and their reliability; the main implications of the findings; possible future work; and any action resulting from the research (e.g. IP, Knowledge Exchange).

1.0 Technical and Scientific Objectives (as set out in the contract)

The overall aim of the STEEM/MF1202 project was to “test, refine and critically evaluate the usefulness of different models for addressing ecosystem and multispecies considerations in fisheries science and management”. Broadly speaking the programme followed the original work-plan and schedule, as illustrated in figure 1, concentrating on North Sea modelling activities during the first two years but with greater emphasis on the Celtic Sea during years 2 and 3 (figure 2).


Figure 2. Case study areas that were the major focus of STEEM/MF1202 modelling activities.

The specific scientific and technical objectives of the project were: Objective A. Develop a priority list of multispecies issues and draft policy/management scenarios, with specific

emphasis on the North Sea. Collate a list of generic ‘type 1’ model-structure questions to be addressed throughout the project life-time and an overview of the different model types available to the project team.

Objective B. Agree upon the scenarios to be tested, through consultation with Defra policy makers, but also Regional Advisory Councils (RACs) and stakeholders. Provide a revised list of case-study specific (‘Type 2’) and generic modelling-related (‘Type 1’) questions to be addressed in the first 18 months of the project.

Objective C. Parameterize, test and validate appropriate models, with specific emphasis on the North Sea. Ensure that the models are capable of evaluating the management options/scenarios agreed in objectives A and B.

Objective D. Evaluate initial model outputs and offer a comparison of model strengths and weaknesses. Objective E. Through consultation with Defra policy makers, but also Regional Advisory Councils (RACs) and

stakeholders, provide a revised list of case-study specific (‘Type 2’) and generic modelling-related (‘Type 1’) questions to be addressed in the second 18 months of the project, with specific emphasis on the Celtic Sea.

Objective F. Parameterize, test and validate appropriate models, with specific emphasis on the Celtic Sea. Ensure that the newly re-formulated/constructed models are capable of evaluating the management options/scenarios agreed in objective E.

Objective G. Evaluate model outputs and provide a comparison of model strengths and weaknesses. Objective H. Detailed reappraisal of ‘type 1’ (generic) and ‘type 2’ (case-study-specific) questions. Provision of

‘decision matrix’ tables, outlining which modeling approaches are/were most well-suited to answering particular questions, their data requirements, and shortcomings.

Objective I. Provide final project report, discuss outcomes with industry and Defra, highlight data deficiencies and suggest areas for future research.

2.0 Extent to which stated objectives have been met

All objectives of the project have been met (at least in part) although it should be recognized that the work suffered some delays (see table 1) and considerable disruption as a result of key staff moving to other institutes in the UK and overseas. Cefas attempted to address this situation through establishing strategic partnerships with these institutes in order to gain access to the same staff members, most notably at Imperial College, University of Sheffield and University of Plymouth (and latterly with the Institute of Biodiversity and Ecosystem Research, Bulgarian Academy of Sciences). The following text provides a brief commentary on each objective and table 1 provides an overview:

Table 1. Summary of progress in relation to targets.

Milestone Number

Objective Target date Achieved in full?

Achieved on time?


1.0 List of candidate multispecies issues and draft policy/management scenarios.[Objective A]

30/06/2008 Partially Partially (see below)

2.0 Consultation to define key multispecies questions. Revised question list. [Objective B]

30/09/2008 Yes Yes

3.0 Model parameterisation & characterisation - North Sea. [Objective C]

31/03/2009 Yes Yes

4.0 Scenario evaluation - North Sea. Model sensitivities, and initial outputs. [Objective D]

30/09/2009 Yes Yes

5.0 Consultation to revise list of key multispecies questions and derive scenarios for the Celtic Sea [Objective E]

30/09/2009 Yes Yes

6.0 Model parameterisation & characterisation - Celtic Sea [Objective F]

31/07/2010 Yes (with Plymouth University)

Partially

7.0 Scenario evaluation - Celtic Sea. Model sensitivity, and conclusions re Celtic Sea scenarios. [Objective G]

30/11/2010 Yes (with Plymouth University)

No

8.0 Evaluation/reappraisal of 'type 1' (generic) and 'type 2' (case study-specific) questions. Decision matrix tables. [Objective H]

31/01/2011 Partially Partially

9.0 Final report - strengths & weaknesses of different modelling approaches, data deficiencies and future research. [Objective I]

31/03/2011 Yes No

Objective A (originally scheduled for 30/06/2008) was only partially achieved in 2008, but will be instead revisited in August 2009 when the project team arranged a workshop at Defra with relevant policy makers, to devise a list of scenarios that could be tested using the now fully-parameterised North Sea models. This meeting also contributed to Objective E. Much of the activity within year 1 concentrated on further development and refinement of modelling approaches, with little focus on comparative studies.

Objective B was achieved through dialogue with the North Sea RAC (Regional Advisory Council), partly supported by an additional FIFG grant, but also through involvement in several EU projects that made use of MF1202 ‘matched funding’, notably the FP6 projects PROTECT, UNCOVER, and IMAGE. A full description of the stakeholder dialogue process can be found in Mackinson et al. (2009) [Can. J. Fish. Aquat. Sci. 66: 1107–1129].

Objective C was the primary focus of activity in year 1 of the project. A variety of different models were developed and applied in the North Sea, however the major focus was directed at completion of a fully-documented and validated EwE model (see Mackinson and Daskalov 2007), as well as a coupled size-structured food-web model (see Blanchard et al. 2009), both of which were completed in late 2008 (on schedule).

Objective D was achieved in full. Throughout 2009 the North Sea EwE model was refined as necessary and spatially validated. It was then used to address a number of applied research questions and scenarios, most notably the multispecies implications of introducing MPAs or offshore wind-farms, as well as the complexities associated with trying to achieve maximum sustainable yield (MSY) for all species simultaneously. At the October 2008 ICES Working Group on Multispecies Assessment Methods (WGSAM), a comparison was made between projections from the North Sea EwE model and those from SMS/MSVPA of fish stocks in the North Sea (see below).

Objective E was achieved (on schedule) via a small workshop held at Defra on 7 th July 2009 with fishery policy-makers. The purpose of this meeting was to report on progress so far, and to consider modelling-related questions to be addressed in the second 18 months of the project. This workshop did not raise any immediate concerns or a need for changes in plans, and so work continued to parameterise models for the Celtic Sea region.

Objective F. Development of a new Celtic Sea EwE model proceeded more slowly than originally envisaged, and although initiated at Cefas – building on earlier work by John Pinnegar, the construction work was completed by Valentina Lauria, initially through a 6 month travel scholarship based at Cefas and subsequently at University of Plymouth as part of her PhD studies. Detailed model validation and ‘fitting’ was undertaken and a draft Cefas Technical Report has now been prepared to accompany those already available for the North Sea, Irish Sea and English Channel.

Objective G. Given that a complete Celtic Sea EwE model was not available until late 2010, the running of fisheries


management and environmental change scenarios/simulations was not possible until early 2011 (and only completed after the end of the project). A major focus has been on the evaluation of ‘bottom up’ implications of a potential discard ban on seabird populations in the region, but a range of other management questions were also considered, including (a) an MSY scenario where fishing mortality for herring was set to 0.25 from 2015 and held constant until 2035, (b) a “Status quo” or No-change scenario, (c) a High fishing impact followed by a no-take scenario, (d) a scenario of increasing sea temperature , and (e) a scenario combining fisheries and climate effects.

Objective H. The full suite of multispecies and ecosystem modelling tools available across Europe were reviewed by MF1202/STEEM project staff at ICES WGSAM meetings in October 2009 and 2010. A wide variety of different approaches have been applied in the region (see Table 3), each with different data requirements, strengths, weaknesses and abilities. A summary table (Table 4) is provided in section 4.0 of this report, and this analysis has revealed that different types of model are useful for different applications, and no model will ever provide all of the answers required.

Objective I. The STEEM/MF1202 project team discussed outputs of the completed work at two scheduled meetings with Defra on 23rd March 2011, and again on 4th July 2011 - as well as final reporting at the ‘Annual Cefas Marine Fisheries R&D monitoring Meeting’ on 20th July 2011. However, completion and preparation of the final report (this document) was subsequently delayed due to additional personnel changes/losses and difficulties in contacting those staff who had left the organisation. Ideas for future work, and the need to retain specialist modelling skills, were discussed at both workshops, and this culminated in a new Defra concept note (‘Physics to Fish’) that was later submitted to the Strategic Evidence & Partnership Fund (SEPF), as well as successful bids for further EU research contracts (e.g. EuroBasin (2010-2014), VECTORS (2011-2015), MYFISH (2012-2016), and GAP2 (2011-2015)).

3.0 Details of the methods used and the results obtained.

In the following paragraphs we attempt to summarise the methods used, the different modelling approaches that have been developed/applied and some illustrative examples of the results obtained.

‘Ecopath with Ecosim’ (EwE) in the North Sea

‘Ecopath with Ecosim’ (EwE) is essentially a food-web model, and includes all fluxes between biological components of the system from detritus and bacteria up to whales. Within Europe, Cefas remains a centre of expertise in the application EwE models and Cefas scientists have constructed EwE models for the North, Irish, Mediterranean, Barents, Beiring and Black Seas as well as the English Channel and Gulf of Florida. Cefas has been at the centre of efforts to explore the sensitivities and capabilities of EwE models for many years (e.g. Mackinson et al. 2003; Pinnegar et al. 2005) it has also had a key role in programming the software and ‘test driving’ new developments. Under the Defra-funded contract MF0323 work was begun to construct a detailed, fully peer-reviewed ecosystem model of the North Sea, but this work was only completed (in 2008) under the MF1202/STEEM project, after 6 years of intense endeavour.

Model construction and validation (figure 3) are all documented in the Cefas Technical Report, authored by Mackinson and Daskalov in 2007. In summary the model comprises 68 functional groups (including 44 fish groups) and 12 fleet categories defined by the EU Data Collection Regulations. A critical step has been to ensure quality control. Accordingly, parameterisation for each functional group was peer-reviewed by a panel of 35 independent experts, both inside and external to Cefas. The model was set up to replicate the situation in 1991 (although a 1973 variant was also prepared) since this made best use of the detailed information available on fish diets (1991 ‘ICES Year of the Stomach’) and catch and discard information by specific fishing fleet segments (STECF 1991 data). Another reason for choosing 1991 (and 1973) is that constructing a model in the past offers the opportunity to calibrate or ‘tune’ the model to changes that have actually been observed since that time, i.e. survey and assessment data from 1973-2008.


Figure 3. Strategic approach guiding construction, testing and application of the North Sea model.

In the early stages of MF1202/STEEM, particular effort was directed towards time series fitting and ‘tuning’ the model. Time series fitting was used to estimate the vulnerability (v) of each prey species to its predators. The process is a lengthy iterative procedure that uses optimisation algorithms to improve the goodness of fit between model predictions and observed data by making adjustments to the vulnerability parameters. The main source of time series data on the fishing mortality and relative fish biomasses was single-species stock assessment data and survey data from ICES International Bottom Trawl Surveys. In addition we collated environmental time series data on NAOI, GSI, temperature and phytoplankton colour index and examined the correlations with herbivorous and carnivorous zooplankton biomass (data from CPR, SAHFOS) – uncorrected and corrected (Pitois and Fox, 2006). On the whole, good fits were obtained to observed data (see plots in Mackinson and Daskalov 2007) and it was found that the best overall fits (to all groups, as opposed to the best for any single group) were derived by using a combination of fishing and environmental drivers.

A comparative exercise was performed (under the EU FP6 Incofish project) with other EwE models worldwide (see Mackinson et al. 2009a). This analysis revealed that in the North Sea, with a long history of heavy exploitation, it is fishing that emerges as the dominant driving force. Elsewhere, notably in the Irish Sea, East China Sea and Southern Humboldt, the degree of improvement obtained by adding primary production forcing was more than twice that obtained when fishing alone was used as a driver. Once all ‘tuning’ and ‘fitting’ had been completed it was possible to begin to use the model to evaluate management trade-offs. This was achieved through dialogue with the North Sea RAC (Regional Advisory Council) partly supported by an additional FIFG grant (see Mackinson et al 2009b). This high-profile and very important work in collaboration with the fishing industry focused its attention on the challenge of developing long-term management plans for the ‘‘mixed-demersal fishery’’ that targets cod, haddock, and whiting in the North Sea.

For each of the target species, the authors ran a simulation, where fishing mortality rate (F) of that species was incremented or decreased slowly, while holding all other F values constant at Ecopath base values. FMSY for the species was taken to be the F that resulted in maximum catch for the particular species. Comparisons were made of the ecosystem model’s equilibrium predictions of FMSY and MSY for cod, haddock and whiting separately when species interactions were either ‘turned off’ or ‘turned on’. When species interactions are ‘turned off’, the biomass of other groups was held constant and thus food availability and predation impacts are constant; in effect the ecosystem model mimics a single species stationary assessment with the biomass of the harvested group responding only to changes in F.

Results indicate that it is not possible to achieve yields equivalent to the ‘single species’ MSYs predicted by the model for cod, haddock and whiting, when individual species target FMSY’s are applied simultaneously. i.e. the stated objective of the EU “to manage fisheries (independently) to achieve MSY by 2015” may be an ecological impossibility. When F MSY

targets for cod, haddock and whiting are implemented together (a mixed-fishery approach) the predicted MSY of cod was higher than its corresponding individual species MSY, but that of whiting and haddock was lower. Cod and haddock are broadly compatible in their responses to changes in fishing effort. It is the contradictory response of whiting that is central to the trade-off of the mixed demersal fishery. Whiting biomass declines as a consequence of any recovery in cod (because of increased predation on whiting by cod). Indirect consequences of cod recovery were also noted in terms of negative impacts on sandeels and certain flatfish.


Mackinson et al (2009b) attempted to establish what happens to yield and revenue in each of the main fisheries (demersal and pelagic trawlers) as a result of implementing different MSY scenarios. The principal fisheries appear to be at odds with one another. Catches of demersal trawls are negatively affected in a relatively small way by changes in the effort of pelagic trawlers, although results indicate that harvesting the prey species does not have a strong influence on the productivity of predators (cod and whiting in particular) through bottom-up effects (Fig. 4a). The results also suggest that pelagic trawl catches of juvenile cod, haddock, and whiting are not large enough to create a significant direct impact. Conversely, pelagic trawl fisheries benefit from higher levels of demersal trawl effort (Fig. 4c) because of the positive impact that culling of predators (demersal species) has on their prey (pelagic species), i.e. ‘prey release’. Incompatibility between mixed-fishery and ecosystem-scale considerations exemplifies the difficult conceptual and practical challenges faced when moving toward an ecosystem approach (see Mackinson et al 2009b).

Figure 4. Response curves for the principal fisheries (demersal and pelagic) taking mixed-demersal fish (from Mackinson et al 2009b).

A crucial step to helping the EU and the relevant countries to reduce harmful fisheries subsidies is to demonstrate through modelling, the damaging impact that these subsidies have on the health of the ecosystem and on the economic and social well-being of the fishing sector. The MF1202/STEEM project team worked with the Scottish Association for Marine Science (SAMS) to develop and apply the North Sea EwE model to investigate the impact that subsidies can have on both the long-term biomass of important fish species and the possible profit from fisheries (see Heymans et al. 2011).

The proportion of the landings and discards of each species taken by each fleet, as reported by STECF (Scientific, Technical and Economic Committee for Fisheries) from 2003 to 2007, was used to update the distribution of landings and discards among the 12 modelled fleets. Current information on the ex vessel price (Euro/tonne) of each species to‐ each fleet and economic performance of each fleet was obtained from the 2008 EU Annual Economic Report. Hindcast simulations were run with the fixed and variable costs of fishing subsidised and not subsidised. The differences in gross revenue and profit were recorded in 1000s of Euros per square kilometre. Two future policy optimisation scenarios were performed (using a Davidson Fletcher Powell non linear routine to improve an objective function by changing‐ ‐ ‐ relative fishing rates iteratively): (a) maximising economic return, and by contrast; (b) maximising the ecological stability of the ecosystem. The economic optimisation scenario aimed to maximise the total profit over all fleets, even if this meant operating some fleets unprofitably to act as controls on less valued species that compete/predate on more valued ones. The ecological stability scenario maximised the longevity weighted summed biomass over all the‐ ecosystem groups. The study found that while removing subsidies might reduce the total catch and revenue in the short-term, in the longer-term this action increases the overall profitability of the fishery and total biomass of commercially important species (see Heymans et al. 2011).

At the October 2008 ICES Working Group on Multispecies Assessment Methods (WGSAM), and in part funded through participation in the EU FP6 project ‘UNCOVER’ (Understanding the Mechanisms of Stock Recovery), a comparison was made between projections from the North Sea EwE model and those from SMS/MSVPA of fish stocks in the North Sea. Estimated Spawning Stock Biomass (SSB) trajectories from the North Sea SMS model (parameterised by University of Hamburg) were compared to SSB trajectories from the EwE model parameterised by Cefas. The EwE model was initially tuned to results of 4M, the deterministic version of SMS. Therefore, the historical SSB trajectories (1990 to 2005) showed similarities in the general abundance between both models. The absolute estimates of historic SSB values, were however, sometimes quite distinct. e.g. for cod SMS estimated a SSB of around 90 thousand tonnes in 1995, while EwE estimated an SSB of 230 thousand tonnes. Predictions from 2006 to 2030 were carried out with both models assuming a constant fishing mortality at precautionary level (Fpa) for all stocks. SMS and EwE came to different conclusions


especially in the short to mid term. In contrast, the long term equilibria estimated for the different stocks were often‐ ‐ quite similar according to both models (see figure 5).

Figure 5. Comparative time series (1990-2030) of spawning stock biomass predicted by EwE (pink) and SMS/MSVPA (dark blue) for whiting, saithe and cod in the North Sea (from ICES 2008).

In conclusion, this is the first time a comparison has been carried out between the stochastic multi species assessment model SMS and the ecosystem model EwE. It is encouraging that both models reached similar equilibria in the long run, although the differences between both models in the short to mid-term were often substantial and this is somewhat worrying. In general, EwE dynamics tended to be more dampened and tended to reach equilibria faster. This may be caused by the larger number of trophic links included in the EwE model.

A major achievement in 2009/2010 has been the ‘coupling’ of the North Sea EwE model to biogeochemical models including GOTM-ERSEM. This has involved a major reprogramming effort (supported in-part by the EU FP7 project ‘MEECE’). The ultimate aim has been to enable predictive modelling of ‘bottom up’ climate effects as well as ‘top-down’ fishing effects on the functioning of the whole marine ecosystem. Investigating how ecosystems respond to combined pressures necessitates that processes at lower and higher levels (LTL and HTL respectively) are linked in some way. This is the basis of the present fashion for ‘end-to-end’ modelling, whereby the connection of physics all the way through to fish (or even fisheries) is made. The principal challenges of coupling LTL models with HTL models involves reconciling discrepancies in how the different approaches handle and represent important processes at time and spatial scales. Here, as a proof of concept, the North Sea EwE model (Mackinson and Daskalov 2007) was coupled to GOTM-ERSEM, a 1D model parameterised for a stratified site in the central North Sea (the Oyster Ground 54º 24’ N, 4º 3’ E). The plugin manager that is part of version 6 of EwE was used to create a link with GOTM (General Ocean Turbulance Model; Bolding & Burchard 2007). The link is a dynamic one with GOTM being loaded from within EwE at runtime and set up so that data can be exchanged in both directions with the Ecosim component of EwE whilst the model is running. GOTM is implemented in FORTRAN with a Python Front end.

Python is an interpreted language. We use a ‘.net’ version of the Python interpreter as a dynamic library (dll) within a EwE plugin to run the python scripts for the front end. ‘Python.net’ allows the Python interpreter to access the variables of the whole ‘.net’ system including those of EwE. When the Python interpreter launches GOTM, both EwE and GOTM share a common process and can share information. The two programs run in a synchronized way, with each part run in a separate blocking thread. All thread handling is carried out by the middleware layer which is ‘called’ by both Ecosim and GOTM (so that the plugin is multithreaded). GOTM controls the semaphore to stop and start Ecosim and vice-versa. GOTM runs for a month of simulation and then releases the Ecosim thread, which is stopped at the start of the Ecosim timestep, before itself blocking. When Ecosim reaches the end of the timestep it unblocks GOTM and stops itself.

The North Sea EwE model consists of 67 functional groups, which are both pelagic and benthic. In order to be consistent with ERSEM, this 67 was extended by two new detritus groups (separating particulate organic matter (POM) into pelagic and benthic components and adding a new faecal POM group). ERSEM passes to EwE the biomass of the pelagic functional groups, whereas the amount of offtake by predators as a proportion of biomass is returned to ERSEM from EwE. The version of EwE used has been adapted to run at a timestep of a single day, rather than a month, this is important since Productivity / Biomass ratios for Phytoplankton are of the order of 286 per year.

The first attempt to link the two models was not successful. Pelagic groups that feed off mesozooplankton overexploited the mesozooplankton during the winter period causing death of the plankton populations and consequent run-down in productivity of the pelagic part of the system. This was a result of the Ecosim model having being calibrated primarily on summer plankton levels. In practice consumption is very much lower in winter months when the plankton population is lower but also the metabolic rate of the predatory fish is very much reduced. In addition to the reduced metabolism which follows an exponential temperature to metabolism law resulting from the Ahhrenius equation, there may be behavioural changes that reduce prey consumption still further. For example sand-eels, which were observed to be one of the major predators of mesozooplankton in summer, switch to being buried in the sediment between November and April (Englehard et al. 2008) and so have zero consumption of plankton in winter


months. To fix these changes it was necessary to add a forcing function to reduce predation of pelagic sources in winter months. Two forcing functions were used – one a purely metabolic model using a Q10=2.0 rule of temperature, with maximum temperature having a multiplier based on the original EwE model of 1.0, the other fitted the observations on sand eel populations of Englehard et al. (2008). In the longer-term it is hoped that the Cefas team will be able to modify the already ‘state of the art’ plugin developed to connect EwE with GOTM-ERSEM into a ‘universal adaptor’ that will enable EwE models to be connected to biogeochemical models more broadly.

A relatively early action in the MF1202/STEEM project was to create a spatial-explicit variant of the North Sea EwE model (known as ‘Ecospace’), in order to investigate the consequences of proposed marine protected areas and spatial management actions. This work was partly funded by the EU FP6 project Incofish. While Ecospace relies on the Ecopath mass-balance approach for most of its parameterisation, it uses a cell based format (in this case ICES statistical rectangles of 1º Latitude by 0.5º Longitude) to describe the two-dimensional, spatial distribution of species under the influence of biotic and abiotic factors. Inputs include: (i) movement rates of fauna (used to calculate changes in species distribution) (ii) the vulnerability settings (top-down vs. bottom-up control) required for Ecosim (iii) habitat preferences (the influences of physical variables on spatial distribution of a species) (iv) spatial distribution of fishing effort and (v) vulnerability to predators in the various specified habitats (Walters et al. 1999).

In the North Sea model, a range of habitat basemaps were developed using temperature, stratification, depth and sediment. Comparison of the habitat maps with the distribution of fish recorded from the IBTS surveyed indicated that the simplest division of habitat could be achieved using 4 depth categories only. During model testing, it was later found to be necessary to create an additional coastal habitat on the east coast of Scotland to reflect the distribution of species in this particular area. Relative densities of species recorded from the International Bottom Trawl Survey (IBTS) trawl survey data and North Sea benthos surveys were used to make initial assignments of the species to each habitat type. Minor modifications were made to these assignments during the process of parameterising the model so that the equilibrium distribution was broadly consistent with the 10 year average distribution of species (1985-1995) recorded from survey data.

The distribution of fishing fleet activity is specified in Ecospace by assigning fleets to habitats, i.e. defining in which habitat(s) a fishing fleet may operate, the costs of fishing based on distance from port and whether a given fleet may operate within a restricted area. Fisheries restricted areas (e.g. MPAs) can be assigned by not allowing certain fleets to operate in them. During the simulation, the fishing mortality rates (F) of the fleets are distributed using a simple ‘gravity model’ where the proportion of the total effort allocated to each cell is assumed proportional to the sum over groups of the product of the biomass, the catchability, and the profitability of fishing the target groups (Caddy, 1975; Hilborn and Walters, 1987).

Our preliminary investigations of existing and proposed MPAs in the North Sea focussed on 5 spatial management scenarios: (a) the CFP plaice box, (b) the CFP Firth of Forth sandeel box, (c) the 2001 CFP cod closure area, (d) proposed Special Areas of Conservation under the EU Habitats Directive, and (e) closure of all ‘round 1, 2 and 3’ windfarm’ sites in the North Sea. Figure 6 shows the impact of SACs (closed to all fishing) on sandeel biomass, but also on the biomass of plaice. The biomass of sandeels within the MPA increased by 6.2% compared to simulations without the MPA. Many predators also increased in abundance within the MPA (e.g. adult whiting by 15.8%, spurdog by 23%). Total North Sea catch of all fish increased by 3.2% under this scenario. However, catch of sandeel vessels declined by 0.5% since this is one of the major fisheries that would be affected by closure of the Dogger Bank region. Total catch of sandeels however, remained virtually unchanged (+0.4%) because much of the displaced sandeel fishing effort was projected to move towards the Firth of Forth. Biomass of plaice increased by 12% within the MPA. Overall catches of North Sea plaice and dab increased by 5 and 18% respectively under this scenario. North Sea demersal trawl catches increased by 1.8% and profits in the demersal trawl fleet increased by 27%.


Figure 6. Simulated impact of proposed SACs (difference compared to model run without MPA) closed to all fishing on (a) sandeel biomass, and (b) plaice biomass. Inset map shows the cells closed to fishing pressure (green).

Figure 7 shows outputs with regard to the ‘Forth of Forth sandeel box’ – an area along the Scottish coast where sandeel fishing has been banned since 2000. The fisheries closure had a significant effect on sandeel biomass which increased locally by 10% in the area. The effects on sandeel also translated into consequences for important predators, for example seabird biomass increased by 1% locally, but decreased elsewhere, notably in the southern North Sea (due to displaced sandeel fishing effort). Within the ‘sandeel box’ many higher trophic level fish predators (especially whiting) also increased in biomass. However, The catch of sandeel across the whole North Sea system declined by 0.5%, and profits in the sandeel trawl fleet decreased by 1.5%.

Some conclusions from this important area of work are:

1. Displaced fishing effort is a major problem. In the model, an increase in the target species was often observed within the MPA, but catches were higher elsewhere, and thus the stock as a whole suffered when an MPA was introduced.

2. There is some evidence of ‘top down’ and ‘bottom up’ trophic cascade effects, for example an increase in seabirds in the Firth of Forth when sandeels were protected.

3. Indirect consequences of introducing a new MPA can be difficult to predict and can seem counter-intuitive because of the complex interplay between multiple predators, preys, competitors and fishing fleets (see Le Quesne et al. 2008).

4. Tools are now available to start to ask ‘what if’ questions with regard to spatial management actions. In the present analysis we were able to compare and contrast, implications of potential fishery exclusion areas resulting from expansion of the renewable energy sector, introduction of conservation measures and fishery management practices.

Figure 7. Simulated direct and indirect impact of the Firth of Forth sandeel fishery closure (difference compared to model run without MPA) on (a) sandeel biomass, and (b) seabird biomass. Inset map shows the cells closed to fishing pressure (green).

‘Ecopath with Ecosim’ (EwE) in the Celtic Sea

Construction work on a EwE model for the Celtic Sea was completed by Valentina Lauria, initially through a 6 month travel scholarship funded by MF1202/STEEM and based at Cefas, but subsequently at University of Plymouth as part of her PhD studies. Detailed model validation and ‘fitting’ was undertaken and a draft Cefas Technical Report has now been prepared to accompany those already available for the North Sea, Irish Sea and English Channel. The Celtic Sea EwE model comprises 64 functional groups: 3 mammals, 6 groups of seabird, 34 groups of fish, 15 invertebrates, 2 microbial groups, 1 primary producer (phytoplankton) and 3 detritus groups (figure 8).

The model has been constructed to represent the 5 year period spanning 1989-1993. This period coincides with a major change in the oceanography and zooplankton communities (Beaugrand et al. 2002; Pitois and Fox, 2006) of the Celtic Sea and in addition, detailed fleet based information is only available (through STECF ICES) for 1991, as well as fish stomach content data (Pinnegar et al. 2003). International landings of fish and shellfish were obtained from the ICES Fishstat plus (ICES area VII f-j) database and an average from 1989-93 was calculated for each country (UK, Ireland, France, Spain, Belgium, Denmark, Germany and Netherland). Unfortunately these data were not broken down into gear


categories, and for this reason all fishery landings were allocated to one of two gear types (either pelagic or demersal) for each country fishing in the area. Comprehensive discards data were not available for the Celtic Sea during the 1989-93 study period, however for some groups, quantities discarded were calculated using the rate of retention and discards available in literature (e.g. Rochet et al., 2002; Borges et al., 2005; Perez et al., 1996).

Figure 8. Schematic representation of the Celtic Sea EwE model in terms of relative biomass and major energy flows within or among Functional Groups (FGs). The horizontal axis is aligned with the estimated trophic level of each functional group.

The model is currently being used to investigate the impact of fisheries management policies and climate change on seabird populations in the region. A major focus has been on the evaluation of ‘bottom up’ implications of a potential discard ban on seabird populations, but a range of other management questions were also considered, including (a) an MSY scenario where fishing mortality for herring was set to 0.25 from 2015 and held constant until 2035, (b) a “Status quo” or No-change scenario, (c) a High fishing impact followed by a no-take scenario, (d) a scenario of increasing sea temperature, and (e) a scenario combining fisheries and climate effects.

When fishing pressure on pelagic fish was altered in the system, the most rapid response was among the target species which showed a biomass change within 1-2 years. Indirect responses of other groups (e.g. seabirds) occurred over longer time scales (5 years). Under the MSY scenario the application of constant fishing mortality (F=0.25) for pelagic fish from 2015 resulted in an overall increase in seabird biomass. Offshore divers (guillemot, puffin, razorbill) showed the most striking increase (with an increase in the total biomass of 77.8 %.), although surface feeders (gannet, kittiwake, fulmar) also benefited and increased by 19 % after 20 years. By contrast, under the ‘discard ban’ scenario, there was a decline of 51.02% in the biomass of gulls in comparison to the situation when discards were kept in the model. As a secondary effect of gull decline, offshore divers increased their biomass in the system by 32.77 % as a result of competitive release. Under the increasing of sea temperature scenario, the biomass of all seabird groups declined in the system by 2025. However, the magnitude of decline in the different seabird groups varied markedly with surface feeders declining by 35.74%, while offshore divers declined by 56.93 % and gulls by 57.28 %. An additional scenario involving combination of the indirect effect of climate change and a 50% reduction in fishing effort from 2015 to 2025 also showed an overall decline in the biomass of seabirds in the system over time.

Size-based modelling approaches

Throughout the MF1202/STEEM project and in collaboration with a number of EU Framework projects (e.g. IMAGE, DEEPFISHMAN and ESF funded SIZEMIC) a number of different, size-based modelling approaches have been developed and applied in the North and Celtic Seas.

In 2009, Blanchard et al. published a detailed description of a coupled benthic-pelagic size-spectra model applied to fish and invertebrate communities in the North Sea (figure 9). The approach was used to assess the theoretical consequences of external sources of mortality (e.g. harvesting) as well as the ‘bottom up’ impact of changes in plankton productivity. Removal of large predators by the fishery resulted in steeper predator size-spectra and concomitant increases in their prey (small fish and detritivores). The model predictions were remarkably consistent with observed patterns in ‘real world’ ecosystems.


Figure 9. Conceptual illustration of two size structured communities with trophic interactions resulting in growth and mortality. The pelagic community consists of predators feeding on increasingly larger prey, as they themselves grow larger. Animals in the benthic zone share and compete for the same food: sinking detrital particles that are comprised of phyto-detritus, faeces and dead animals.

Several general conclusions can be drawn from this study. First, differences in spectrum slopes (between the benthic and pelagic food-chains)) were apparent in empirical analyses and the slopes for pelagic predator communities were steeper than those for detritivore communities that share energy. Second, a range of slopes could arise in the model depending on (i) the extent of food web coupling, (ii) different levels of productivity at the base of the food web (usually mediated by the climate), and (iii) fishing. Lower levels of productivity should result in truncated size spectra that are less linear and, in areas where predator coupling is stronger, the slope of the detritivore spectrum should be steeper. These findings suggest several testable hypotheses for future work and highlight the need for an ecosystem approach to understanding the effects of exploitation (see Blanchard et al. 2009 for further details).

Castle et al. (2011) took the model developed by Blanchard et al. (2009) under MF1202/STEEM and has subsequently created a spatially-explicit version of this model to predict how the active movement and passive transport of individuals can influence individual growth and size spectra. Active movements comprise ‘prey-seeking’ behaviour, with individual organisms moving locally towards areas with high concentrations of favoured prey, and ‘predator-avoiding’ behaviour, with prey moving away from areas of high predator density. Passive transport represents the effects of turbulent mixing on small individuals. The model was used to explore the individual and community effects of biotic and abiotic processes and their interactions, and to predict how energy from local sources of primary production is propagated through the food web. In areas of high phytoplankton abundance, community size-spectrum slopes were shallower and larger individuals were present, whereas in low production areas, slopes were steeper and size spectra truncated. Efforts are currently underway to combine the model developed by Blanchard et al. (2009) and elaborated upon by Castle et al. (2011) with complex biogeochemical models including POLCOMS-ERSEM (at PML) and GOTM-ERSEM (at Cefas) to predict the ‘bottom up’ consequences of climate change and plankton variability (see van Leeuwen & Blanchard, 2009).

A derived size-based model (based on Andersen & Pedersen, 2009) has been used to ’test-drive’ fish indicators, to see how they will be affected by changes in fishing mortality and future climate change. It is broadly similar to the size-based multispecies model of Hall et al. (2006), but with food-dependent as opposed to predetermined growth. The model was applied within the EU project ‘IMAGE’ (but with ‘matched funding’ support from MF1202/STEEM). In particular, the model has been utilized to study the “large fish indicator” (LFI), a concept initially put forward by OSPAR but now considered to be vitally important for operationalizing the ‘ecosystem approach’ within the revised EU Common Fisheries Policy.

The authors (Blanchard et al. in prep) parameterised the model for 12 important fish species, accounting for nearly 90% of the total abundance of all fish species sampled by research trawl surveys in the North Sea. Each species is characterized by a set of parameters detailing its physiological, life history and foraging traits, which were obtained from the literature and existing datasets. To capture realistic levels of abundance and size distributions for each species, the model was calibrated to abundance data obtained from international trawl surveys and by adjusting the least known biological parameter, the maximum reproductive output (Rmax). The model also included size and species-selective fishing mortality based on best available estimates from international fish stock assessments.


Table 2. Community-level indicators obtained from the unexploited and 1985-1995 fished models and from empirical estimates based on survey data from the same time period. The last two rows contain model results for setting all Fs at the MSY for all estimated for each species (fished in isolation) without species interactions (single species) and with species interactions included (food web).

Community slopeMean Weight Mean Wmax LFI No collapsed

Total Biomass totYield

Model - Unexploited -0.86 113.24 692.62 0.44 0 518436836 0

Model - 1985:1995 -1.38 82.98 1316.08 0.12 1 458633220 18953464

Observed - 1985-1995 -1.33 76.65 2232.65 0.12 0.91 38918968 975595

Model - MSY single species -1.08 84.50 1281.48 0.25 0.00 458036770 17023734

Model - MSY food web -1.07 88.14 1347.62 0.24 0.00 469093795 17170906

The validated model was used to establish baseline and target levels of indicators, which can inform policy (for example setting indicators and targets that are required for the MSFD by 2012). Table 2 gives the estimated indicator values from the model in the absence of fishing, or with fishing representative of the 1985-1995 period. It is important to contrast whether or not potential community-level or population-level indicator targets might be achieved in comparison with fisheries management objectives, such as the fishing mortality that would maximize yields FMSY. Using the calibrated model, Blanchard et al. (in prep) explored the conditions under which the fisheries management and conservation targets could be reached, in particular the authors carried out two fishing scenarios: single targeted species versus mixed fisheries (targeting different subsets of the community simultaneously).

Figure 10. Relative contribution of each of the 12 modelled species to the total size composition (black) of the North Sea fish community (from Blanchard et al. in prep).

The results from this modelling study show that FMSY required to meet single species MSY targets did not substantially differ with inclusion of food web interactions if species were assumed to be fished in isolation, suggesting changes in yield are more strongly determined by the life-histories than species interactions. However, applying the Fs required to simultaneously achieve MSY for all species (food security objective) prevented the Large Fish Indicator target (LFI=0.3) from being achieved (biodiversity and food web integrity objective). The trait-based size-spectra model developed under MF1202/STEEM has also been applied to fish communities in the Celtic Sea (by Finlay Scott and Julia Blanchard) as part of Defra- project MF1001 and comparative work (with the Celtic Sea EwE) model may be carried out in the future if time and resources allow.

Other multispecies modelling approaches applied under MF1202

A number of other multispecies/ecosystem modelling approaches were explored (at least in a preliminary sense) over the duration of the MF1202/STEEM project, notably (1) an Ecospace model of the English Channel, (2) a multispecies GADGET model of the Celtic Sea, and (3) a simple two species model of cod-Nephrops interactions in the Irish Sea. In each case this work was primarily carried out under associated research contracts (ALSF MEPF 08/P37, EUROBASIN, DAPSTOM3 MF1109) – but with some matched support provided by MF1202/STEEM. Each of these is briefly described


below.

A spatially dynamic EwE model of the Eastern English Channel was developed under the Marine Aggregate Levy Sustainability Fund (MALSF) and used to investigate the possible effects that changes in species abundance and distribution as a result of aggregate extraction might have on food webs and fisheries. The model was based on an earlier EwE model developed at IFREMER and Cefas, and described by Villanueva et al. (2009). The model was calibrated by comparing to time series and spatial survey data (the French GOV trawl survey). This spatial representation of the Eastern Channel ecosystem was then used to simulate and explore the ecosystem effects of aggregate dredging (see Daskalov & Mackinson 2011). Changes in the populations of benthic organisms and the food web were modelled assuming a range of potential mortalities induced by dredging impact. In the first scenario it was assumed that dredging causes mortality on zoobenthos alone. In the second scenario it was assumed that dredging causes direct mortality on both zoobenthos and demersal fish. The decrease in total zoobenthos and total fish biomass in the dredged area were 4.7% and 4.9% respectively, when applying 30% mortality on fish and zoobenthos. Within the dredged area notable decreases were detected in catches of otter and beam trawls, dredges, pots and nets, whilst catches increased for pelagic trawlers.

A preliminary, 3 species (cod, whiting, blue-whiting) model, building on an approach known as 'Gadget', was developed for the Celtic Sea by scientists from Cefas and IFREMER in 2000-2004 (Trenkel et al. 2004). GADGET is a flexible, length-based, modelling framework (see Begley & Howell 2004) which can be used for multi-species, multi-area and multi-fleet simulations. Length-based modelling approaches are particularly useful because predators and fishing gears, typically select their 'prey' on the basis of size, rather than age (as is assumed in models such as SMS/MSVPA). Towards the end of the MF1202/STEEM project (in 2010/2011) and in association with the EU FP7 project EuroBasin, the 3 species Celtic Sea model was revisited with a view to updating time-series, perhaps introducing new predator species (hake and megrim) and improving the parameterisation of the original blue-whiting model. This work is still underway (see section 6 of this report) and will be completed over the next two years.

An approach that has become commonplace elsewhere in the world has been to extend single-species stock assessment models to include some element of predation mortality, where it is known that a particular predator has a direct impact on the species of concern. Examples include capelin in the Barents Sea, linked to cod; northern fur seals in the walleye pollack assessment for the eastern Bering Sea. As part of the Defra-funded DAPSTOM-3 project (which ended in March 2011) we endeavoured to construct an extended single-species population model for Nephrops (Norway lobster) in the Irish Sea, but with an explicit predator-prey link to cod (see MF1109 report). In order to characterise the level of interaction between cod and Nephrops it was necessary to estimate the number of cod that exist within the key Nephrops grounds (west of the Isle of Man) at different times of the year. The analysis suggests that the quantity of Nephrops consumed has declined steadily since 2003 to around 150 tonnes/quarter (yearly totals for 2003-2007: 1.56, 1.08, 0.94, 0.68, 0.61 thousand tonnes), and this has largely been associated with a decline in the size of the Irish Sea cod stock (particularly the number of large individuals). Given the large size of the Nephrops stock, this represents a relatively low mortality rate (especially compared to the 8.4 thousand tonnes removed by fisheries in 2007). Assuming that the Irish Sea cod stock were allowed to recover to the stated MSY spawning biomass target of 10,000 tonnes (from 1,658 tonnes in 2007), then this would imply a much greater level of predation pressure on Nephrops in the future than is currently the case, but this would still represent a much smaller toll than is imparted by the fishery.

4.0 Discussion of the results and their reliability

A wide diversity of multispecies and ecosystem models have been developed or applied under the MF1202/STEEM project. They have been used to investigate a wide range of very applied research questions, and the results reported here give a good impression of “what each approach can do”. In the North Sea many different variants of the same EwE model have been applied, as well as several different size-based modelling approaches. In the Celtic Sea, the focus has been somewhat narrower but has nevertheless included the development of EwE models and size-based approaches (together with the Defra project MF1002).

Table 3. Summary of the multispecies/ecosystem modelling approaches that have been applied in the North Sea and Celtic Sea. (*models that have been applied as part of MF1202/STEEM).

Model Type/Name Spatially-explicit (Y or N)

Application so far

North SeaEwE model (Mackinson & Daskalov 2007)* No To evaluate multispecies MSY and fisheries trade-offs.

To consider the indirect impact of fisheries subsidies.Coupled EwE-GOTM/Ersem* No To evaluate ‘bottom up’ impact of climate change and

‘top down’ impact of fisheries on the ecosystem.


Ecospace (spatial variant of EwE model)* Yes To examine the indirect consequences of proposed and existing marine protected areas (MPAs)

Size-based ecosystem model (Blanchard et al. 2009)*

No To investigate the impact of climate-driven changes in productivity and fishing pressure on the ecosystem

Spatial variant of size-based ecosystem model (Castle et al. 2011)*

Yes To investigate the impact of climate-driven changes in productivity and fishing pressure on the ecosystem

Species-specific size-based model (Blanchard et al (in prep)*

No To examine the impact of fishing and climate on key fish community indicators (under MSFD & CFP)

ATLANTIS (cross-sector model framework – being developed under VECTORS)

Yes To investigate spatial consequences of MPAs, climate change, eutrophication and wind-farms etc.

SMS/MSVPA/4M (Stochastic Multi-species Assessment Model)

No To provide natural mortality estimates for input to stock-assessments. To test fish community indicators. To investigate MSY and fisheries trade-offs.

Multispecies Individual-based models (e.g. OSMOSE and modelling at Marine Scotland)

Yes To examine the indirect consequences of marine protected areas (MPAs) and fisheries trade-offs.

Celtic Sea & English ChannelEwE model – Celtic Sea (Lauria et al)* No To investigate MSY and fisheries trade-offs. Impact of

climate and fisheries (discard ban) on seabirds.EwE model – Celtic Sea (Pinnegar et al) No Preliminary investigation of the Celtic Sea food-web.

Species-specific size-based model (under MF1001)*

No To examine the impact of fishing and climate on key fish community indicators (under MSFD & CFP)

GADGET model of cod, whiting, blue whiting (EuroBasin)*

No To examine influence of climate change and predation pressure on blue whiting population

EwE model – western channel (Araujo et al., 2005)

No To examine influence of climate change and fisheries on the marine foodweb. Fisheries trade-offs

EwE model – eastern channel (Villanueva et al. 2009; Standford and Pitcher 2004)

No To examine influence of climate change and fisheries on the marine foodweb. Fisheries trade-offs

Ecospace – eastern channel (MALSF)* Daskalov & Mackinson 2011

Yes To examine the influence of aggregate extraction on marine food-webs and fisheries. Fisheries trade-offs

Extended single-species model (cod-Nephrops in the Irish Sea)*

No To examine the impact of predation by cod on the Irish Sea Nephrops (Norway lobster) population

Plagányi (2007) in a report to the FAO, reviewed the relative merits of different multispecies modelling approaches in terms of answering fisheries management questions. This review included a description of each approach, a comparison in terms of levels of complexity, their functionality, advantages, disadvantages and limitations. The author concluded that the models available differ markedly in terms of levels of complexity. Most of the approaches considered may be categorized as ‘Minimum Realistic Models’ (MSMs), with only EwE and ATLANTIS representing the full ecosystem (size-based approaches such as those of Blanchard et al. 2009 were not considered by Plagányi 2007).

Consequently for broad-scale objectives related to the structure and functioning of the ecosystem, Plagányi (2007), concluded that ECOPATH/ECOSIM (or Ecopace) might be the only tool that is readilly available; other models may be more appropriate for more specific questions. Unlike EwE, individually tailored approaches such as MRMs (including the cod-Nephrops model described above) or SMS/MSVPA have more flexibility in modelling the dynamics of specific marine predators, but usually ignore any ‘bottom up’ effects that changing prey populations may have on the predators themselves. Fulton and Smith (2004) strongly recommend that ideally a suite of different “minimum-realistic” ecosystem models should be constructed and their results compared. However, given limited person-power and pressure to produce results, it is important first to engage in discussions regarding which are the preferred modelling approaches to be pursued in each context.

Plagányi (2007) provided a summary table featuring advantages, disadvantages and limitations of each method, as well as notes on the ease of presentation of model outputs and the user-level of programming and mathematical skills required. In table 4 (below) we have taken this basic format and have tailored it to the modelling approaches that have been applied in the North and Celtic Seas (as part of MF1202/STEEM and other projects. It was originally envisaged that the MF1202/STEEM project would involve considerably more parallel running and comparison of model outputs from the same case-study region. Unfortunately, with the exception of the comparative model runs of EwE and SMS/MSVPA described above, and various studies comparing single-species and multi-species formulations of the same model (with links ‘switched on’ or ‘switched off’) this proved very difficult given personnel changes during the course of the project. This being said, efforts are still underway with regard to comparisons of the various models applied in the Celtic Sea.

Plagányi (2007), presented a useful ‘decision tree’ to show the classification and key differences between multispecies-ecosystem modelling approaches. Objective H of the MF1202/STEEM project called for ‘decision matrix’ tables, outlining which modelling approaches are/were most well-suited to answering particular questions. A combination of tables 3 and 4 (MF1202/STEEM specific), along with figure 11 (generic) fulfil this function.


[SEE EMBEDDED TABLE 4]

The past three decades have witnessed explosive growth in the number and variety of multispecies models directed at answering fisheries questions (Hollowed et al., 2000, Whipple et al., 2000; Plagányi 2007). Having reviewed the relative merits of the approaches available it is probably fair to agree with Chatfield (1995) and say that “All models are wrong, but some are useful”. Naturally there is a large number of very specific questions that models have been constructed to address and every (good) model is useful in the context for which it has been designed.

Fulton et al. (2003) argued that complex models have often acquired a poor reputation, primarily because of two factors. First, the models are often so large that they may not be cost-efficient, with most modelling resources spent in development rather than application. Second, complexity introduced for the sake of completeness accomplishes nothing if the resulting model is actually of poor quality (hence our focus on detailed peer-review of each functional group in the North Sea EwE model). While modern computing power makes ever-more complex ecosystem models attractive as computational restraints are lifted, this does not solve the problems of uncertain model specification, parameterization and system understanding. Studies have tended to indicate that the relationship between model detail and performance is non-linear (Costanza and Sklar, 1985). Too much complexity can lead to too much uncertainty and render the model’s dynamics and predictions difficult to interpret. Too little detail results in models that do not produce realistic behaviours (Fulton et al., 2003). Thus, there may be an ‘optimum’ level of model complexity depending on the situation, that may be substantially below the maximum possible (Fulton et al., 2003). Different types of model are useful for different applications, and no model will ever provide all of the answers required. One clear indication that the models developed under MF1202/STEEM are having a significant wider impact beyond the modelling group at Cefas, is that the Mackinson & Daskalov (2007) Ecospace model for the North Sea has recently been used by the Netherlands Environmental Assessment Agency (PBL) to explore the impact of spatially explicit changes in human use on the Dutch coast (see Kooten & Klok 2011).


Figure 11. Decision tree showing the different multispecies and ecosystem modelling approaches that have been applied to fisheries questions, highlighting the main differences and similarities (from Plagányi 2007)

5.0 Main implications of the findings

There have been many calls within the EU and at the international level to move towards ‘ecosystem-based approaches to fisheries management’ (EAF), and this is often interpreted as necessitating a greater understanding of interactions between commercial species since exploitation of one organism can have knock-on consequences for the wider food-web. Ecosystem and multispecies models are one means of exploring potential indirect consequences of management actions and the recent expansion in the number and variety of modelling approaches available (see figure 11) reflects this more intense interest in food-web dynamics and ecosystem effects.

The 2008 European Marine Strategy Framework Directive (2008/56/EC) includes a requirement for Member States to work to achieve ‘Good Environmental Status’ (GES) by 2015. This is defined by eleven qualitative descriptors, one of which (descriptor 4) explicitly focuses on “Food Webs” (specifically “All elements of marine food webs, to the extent that they are known, occur at normal abundance and diversity and levels capable of ensuring the long-term abundance of the species and the retention of their full reproductive capacity “). The 2009 Green Paper on ‘Reform of the EU Common Fisheries Policy’ also recognises that “Scientific knowledge and data are of vital importance to the CFP, because policy decisions must be based on robust and sound knowledge on the level of exploitation that stocks can sustain, of the effects of fishing on marine ecosystems and on the impacts of changes such as climate change”. The FAO Code of Conduct on Responsible Fishing requires that “States should assign priority to undertake research and data collection in order to improve scientific and technical knowledge of fisheries including their interaction with the [wider] ecosystem”.

In the MF1202/STEEM project we have endeavoured to apply and where necessary develop new ecosystem and multispecies modelling techniques, these have been successfully used to address key questions of importance to Defra and to the fishing industry (as evidenced by our collaborative working with the North Sea RAC). By considering a whole range of different modelling frameworks (Table 3) we have attempted to enhance and expand the tool-box on which Defra can draw and to demonstrate what is possible using these methods. We have built capability and have demonstrated that we now have a suite of tools with which we are able to address most issues of concern to policymakers. Defra’s Evidence and Innovation Strategy (2005) classified multispecies issues as ‘high priority’, needing continued support and commitment. This was reiterated in the ‘Statement of Needs’ for the Sustainable Marine Fisheries division (priority E10), and the Conservation of Salmon & Freshwater Fish Stocks, and Whaling division, which specifically called for “modelling ….. marine ecology and in particular feeding relationships and species interactions” (priority E10). The work carried out here is consistent with science requirements of the ICES Strategic Plan, “new exploitation strategies that take account of complexity (such as trophic interactions) and uncertainty (such as effects of natural variability and climate change) need to be evaluated” and “robust exploitation strategies for living marine resources, taking into account ecosystem complexity and uncertainty need to be designed.”

Based on our work a number of conclusions become apparent:

1. A wide variety of tools are available to answer different types of question - most management questions can be tackled using one or other of the methodologies trialed in the MF1202/STEEM project.

2. It is now possible to construct exceedingly complex models of whole ecosystems but these are not always the most appropriate tool for evaluating fishery trade-offs or consequences of management action.

3. Significant progress has been made with regard to constructing models for the Celtic Sea, but the information and approaches still lag behind those in the North Sea.

4. Ecosystem and multispecies models are most useful for ‘strategic’ planning (i.e. asking ‘what if’ questions) over longer lime-scales, rather than giving predictions in the short to medium term.

5. The modelling work has demonstrated that single-species approaches may offer unrealistic expectations with regard to long-term yield or rates of stock recovery.

6. Several modelling frameworks are available that can take account of spatial patterns of predator-prey overlap. Such techniques (e.g Ecospace or ATLANTIS) can be used to predict the indirect ecosystem consequences of introducing offshore MPAs, as well as the resulting fishery displacement.

7. Multispecies modelling frameworks have been used to investigate trade-offs between fleet-segments, and consequently impacts on revenues, profits and costs in the fishing industry.

8. Models can be used to investigate the relative influence of climate change as well as fishery management on


fish stocks and the wider ecosystem.9. No model will ever provide all of the answers required.

6.0 Possible future work

The considerable progress made under this project has enabled Cefas to build further partnerships and to contribute to a wide range of newly agreed EU research initiatives, most notably: EuroBasin (2010-2014), VECTORS (2011-2015), MYFISH (2012-2016), and GAP2 (2011-2015). Development work (for example using the ’state-of-the-art’ ATLANTIS framework), as well as application of the full suite of models developed under MF1202/STEEM, will be continued under the Defra Strategic Evidence & Partnership Fund (SEPF) Project ’Physics to Fish’ (2012-2016). Comparative modelling work (running different models in parallel) will continue in the near future. In particular it is intended that the size-based model (based on Blanchard et al., in prep) currently being constructed for the Celtic Sea under the Defra project MF101 will be compared and contrasted with the EwE model developed under MF1202/STEEM, following a meeting with French colleagues in May 2012.

Major aspirations for the future will include: (1) to utilise the various models developed under MF1202/STEEM to evaluate further fishery trade-offs suggested by the industry – this will be facilitated through workshops under the auspices of the EU ‘GAP 2’ project; (2) to utilise the existing North Sea EwE model to simulate possible ecosystem-wide consequences of ocean acidification; (3) to carry out simulations using the species-specific size-based model (Blanchard et al., in prep) in order to compare fish community indicators/targets under proposed management regimes with those at the beginning of the 20th Century; (4) to work with partners under the EU FP7 project VECTORS to construct a new ATLANTIS model for the North Sea; (5) to explore the utility of Dynamic Energy Budget (DEB) models for answering multispecies questions.

7.0 Cited References

Andersen KH, Pedersen M (2009) Damped trophic cascades driven by fishing in marine ecosystems. Proceedings of the Royal Society of London, Series B. 277:795-802.

Begley J, Howell D (2004) An overview of Gadget, the Globally applicable Area-Disaggregated General Ecosystem Toolbox. ICES CM 2004/FF: 13.Beaugrand G, Reid P, Ibanez F, Lindley A, Edwards M (2002) Reorganization of North Atlantic marine copepod biodiversity and climate. Science, 296:

1692-1694.Blanchard JL, Andersen KH, Scott F, Hintzen NT, Piet G, Jennings S (in prep) Trade-offs between biodiversity and food security objectives for the North

Sea elucidated from fish community dynamics. Manuscript in preparation.Blanchard JL, Jennings S, Law R, Castle MD, McCloghrie P, Rochet M-J, Benoît E (2009) How does abundance scale with body size in coupled size-

structured food webs? Journal of Animal Ecology 78 , 270–280Bolding K, Burchard H. (2007). GOTM – General Ocean Turbulance Model. www.gotm.net/index.php.Borges L, Rogan E, Officer R (2005) Discarding by the demersal fishery in waters around Ireland. Fishery Research, 76: 1-13. Caddy JF (1975) Spatial model for an exploited shellfish population, and its application to the Georges Bank scallop fishery. Journal of the Fisheries

Research Board of Canada, 32: 1305–1328.Castle MD, Blanchard JL, Jennings S (2011) Predicted effects of behavioural movement and passive transport on individual growth and community size

structure in marine ecosystems. Advances in Ecological Research. 45, 41–66.Chatfield C (1995) Model uncertainty, data mining and statistical inference. Journal of the Royal Statistical Association 158, 419–466.Costanza R, Sklar FH (1985) Articulation, accuracy and effectiveness of mathematical models: a review of freshwater wetland applications. Ecological

Modelling 27, 45–68.Daskalov GM, Mackinson S (2011) Modelling possible food-web effects of aggregate dredging in the eastern English Chanel (MEPF 08/P37). Marine

Aggregate Levy Sustainability Fund (MALSF), March 2011.65pp.Englehard GH, van der Koij J, Bell ED, Pinnegar JK, Blanchard JL, Mackinson S, Righton D (2008), Fishing mortality versus natural predation on diurnally

migrating sandeels, Ammodytes marinus. Marine Ecology Progress Series, 369:213-227.Fulton EA, Smith ADM (2004) Lessons learnt from the comparison of three ecosystem models for Port Phillip Bay, Australia. African Journal Marine

Science 26: 219 – 243Fulton EA, Smith ADM, Johnson CR (2003) Effect of complexity on marine ecosystem models. Marine Ecology Progress Series 253, 1–16.Hall SJ, Collie JS, Duplisea DE, Jennings S, Bravington M, Link J (2006) A length-based multispecies model for evaluating community responses to

fishing. Canadian Journal of Fisheries and Aquatic Sciences; 63:1344-1359.Heymans J J, Mackinson S, Sumaila UR, Dyck A, Little A. (2011) The Impact of Subsidies on the Ecological Sustainability and Future Profits from North

Sea Fisheries. {PLoS} {ONE}. 6:e20239.Hilborn R, Walters, CJ (1987) A general model for simulation of stock and fleet dynamics in spatially heterogeneous fisheries. Canadian Journal of

Fisheries and Aquatic Sciences 44:1366-1370.Hollowed AB, Bax N, Beamish R, Collie J, Fogarty M, Livingston P, Pope J, Rice JC (2000) Are multispecies models an improvement on single-species

models for measuring fishing impacts on marine ecosystems? ICES Journal of Marine Science 57, 707–719.ICES (2008) Report of the Working Group on Multispecies Assessment Methods (WGSAM). 16-10th October, ICES Headquarters, Copenhagen. ICES

CM 2008: RMC06. (Meeting co-chaired by John Pinnegar).Kooten T van, Klok, C (2011) The Mackinson-Daskalov North Sea EcoSpace model as a simulation tool for spatial planning scenarios. Wageningen,

Statutory Research Task Unit for Nature and the Environment. WOt-werkdocument 249. 98 pp. http://edepot.wur.nl/185174Le Quesne WJF, Arreguin-Sanchez F, Albanez-Lucero M, Cheng H, Cruz Escalona VH, Daskalov G, Ding ., Gonzalez Rodriguez E, Heymans JJ, Jiang H,

Lercar, D, Lopez-Ferreira C, Lopez-Rocha JA, Mackinson S, Pinnegar J (2008) Analysing ecosystem effects of selected marine protected areas with Ecospace spatial ecosystem models. Fisheries Centre Research Reports 16(2). Fisheries Centre, University of British Columbia.

Mackinson S, Daskalov G (2007) An ecosystem model of the North Sea to support an ecosystem approach to fisheries management: description and parameterisation. Sci. Ser. Tech Rep., Cefas Lowestoft, 142: 196pp.

Mackinson S, Blanchard JL, Pinnegar JK & Scott R (2003) The importance of predator-prey functional response in models exploring whale-fishery interactions. Marine Mammal Science, 19, 661-681.

Mackinson S, Daskalov G, Heymans JJ, Neira S, Aranciba H, Zetina-Rejón M, Jiang H, Cheng HQ, Coll M, Arreguin-Sanchez, F, Keeble K, Shannon L


http://www.gotm.net/index.php

(2009a) Which forcing factors fit? Using ecosystem models to investigate the relative influence of fishing and changes in primary productivity on the dynamics of marine ecosystems. Ecological Modelling, 220: 2972-2987.

Mackinson S, Deas B, Beveridge D, Casey J (2009b) Mixed-fishery or ecosystem conundrum? Multispecies considerations inform thinking on long-term management of North Sea demersal stocks. Canadian Journal of Fisheries & Aquatic Sceience 66: 1107–1129.

Perez N, Trujillo V, Pereida P (1996) Discards of the trawl and long line Spanish fleet in Ices subarea VII in 1994. ICES CM/Mini: 8. International Council for the Exploration of the Sea. Copenaghen, Denmark.

Pinnegar JK, Blanchard JL, Mackinson S, Scott RD & Duplisea DE (2005) Aggregation and removing weak-links in food-web models: system stability and recovery from disturbance. Ecological Modelling, 184, 229-248.

Pinnegar JK, Trenkel VM, Tidd AN, Dawson WA, Du Buit MH (2003) Does diet in Celtic Sea fishes reflect prey availability? Journal of Fish Biology, 63 (supplement A), 197-212.

Pitois S, Fox CJ (2006) Long-term changes in zooplankton biomass concentration and mean size over the Northwest European shelf inferred from Continuous Plankton Recorder data. ICES J. Mar. Sci., 63: 785-798

Plagányi É (2007) Models for an ecosystem approach to fisheries. FAO Fisheries Technical Paper. Food and Agriculture Organization, Rome, 108 pp. Rochet MJ, Peronnet I, Trenkel VM (2002) An analysis of discards from the French trawler fleet in the Celtic Sea. ICES journal of Marine Science, 59 :

538-552.Trenkel VM, Pinnegar JK & Tidd AN (2004) Can multispecies models be expected to provide better assessments for Celtic Sea groundfish stocks? ICES

CM 2004/ FF:05.van Leeuwen S, Blanchard JL (2009) "Linking physical processes to higher trophic levels: results from coupling hydrological, biogeochemical and size-

based food web models". ASLO Aquatic Sciences Meeting, Nice, France (February 2009).Villanueva C, Ernand B, Mackinson S (2009) Trophic Network, Chapter 6. pp548-562, In Carpentier A, Martin CS, Vaz S (Eds.), 2009. Channel Habitat

Atlas for marine Resource Management, final report / Atlas des habitats des ressources marines de la Manche orientale, rapport final (CHARM phase II). INTERREG 3a Programme, IFREMER, Boulogne-sur-mer, France. 626 pp.

Walters C, Pauly D, Christensen V (1999) Ecospace: predictions of mesoscale spatial patterns in trophic relationships of exploited ecosystems, with emphasis on the impacts of marine protected areas. Ecosystems, 2: 539–554.

Whipple SJ, Link J, Garrison LP, Fogarty MJ (2000) Models of predation and fishing mortality in aquatic ecosystems. Fish and Fisheries 1, 22–40.


References to published material9. This section should be used to record links (hypertext links where possible) or references to other

published material generated by, or relating to this project.

(Earliest to most recent)

Mackinson S, Daskalov G (2007). An ecosystem model of the North Sea to support an ecosystem approach to fisheries management: description and parameterisation. Sci. Ser. Tech Rep., Cefas Lowestoft, 142: 196pp. www.cefas.defra.gov.uk/publications/techrep/tech142.pdf

Le Quesne WJF, Arreguin-Sanchez F, Albanez-Lucero M, Cheng H, Cruz Escalona VH, Daskalov G, Ding H, Gonzalez Rodriguez E, Heymans JJ, Jiang H, Lercari D, Lopez-Ferreira C, Lopez-Rocha JA, Mackinson S, Pinnegar J (2008) Analysing ecosystem effects of selected marine protected areas with Ecospace spatial ecosystem models. Fisheries Centre Research Reports 16(2). Fisheries Centre, University of British Columbia. http://fisheries.ubc.ca/sites/fisheries.ubc.ca/files/pdfs/fcrrs/16-2.pdf

Pinnegar JK, Blanchard JL (2008) Long-term shifts in the feeding preferences of North Sea fish over the past 100 years: old data and new modelling approaches. ICES CM 2008 F:06 www.ices.dk/products/CMdocs/CM-2008/F/F-2008.pdf

ICES (2008a) Report of the Working Group on Multispecies Assessment Methods (WGSAM). 16-10th October, ICES Headquarters, Copenhagen. ICES CM 2008: RMC06. (Meeting co-chaired by John Pinnegar). www.ices.dk/products/CMdocs/CM-2008/RMC/WGSAM08.pdf

ICES (2008b) Report of the Workshop on historical data on fisheries and fish (WKHIST). ICES Headquarters, Copenhagen. ICES CM 2008: RMC04. www.ices.dk/products/CMdocs/CM-2008/RMC/wkhist08.pdf

Jennings S, Mélin F, Blanchard JL, Forster RM, Dulvy NK, Wilson RW (2008) Global-scale predictions of community and ecosystem properties from simple ecological theory. Proceedings of the Royal Society B., 275, 1375–1383. http://rspb.royalsocietypublishing.org/content/275/1641/1375

Blanchard JL, Law R, Castle MD, Barnes C, Jennings S (2008) Predator-prey mass ratios and the variability of marine fish populations. ICES CM 2008 F:03. Presentation at ICES Annual Science Conference, Halifax, Canada, September 2008. http://www.ices.dk/products/CMdocs/CM-2008/F/F-2008.pdf

ICES (2009) Report of the Working Group on Multispecies Assessment Methods (WGSAM). 5–9 October 2009, ICES Headquarters, Copenhagen. ICES CM 2009/RMC:10 (Meeting co-chaired by John Pinnegar). www.ices.dk/reports/SSGSUE/2009/wgsam09.pdf

Daskalov GM, Mackinson S, Cheng HQ, Pinnegar JK (2009) Evaluation of the usefulness of Marine Protected Areas (MPAs) for management of recovery of fish stocks and ecosystems in the North Sea. In: Palomares, M.L.D., Morissette, L., Cisneros- Montemayor, A., Varkey, D., Coll, M., Piroddi, C. (eds.), Ecopath 25 Years Conference Proceedings: Extended Abstracts, pp. 50-51. Fisheries Centre Research Reports 17(3). Fisheries Centre, University of British Columbia. 167 pp. ftp://ftp.fisheries.ubc.ca/FCRR/17-3.pdf

Mackinson S (2009) Ecospace: has its time come? In: Palomares, M.L.D., Morissette, L., Cisneros-Montemayor, A., Varkey, D., Coll, M., Piroddi, C. (eds.), Ecopath 25 Years Conference Proceedings: Extended Abstracts, pp. 49. Fisheries Centre Research Reports 17(3). Fisheries Centre, University of British Columbia. 167 pp. ftp://ftp.fisheries.ubc.ca/FCRR/17-3.pdf

Mackinson S, Daskalov G, Heymans JJ, Neira S, Aranciba H, Zetina-Rejón M, Jiang H, Cheng HQ, Coll M, Arreguin-Sanchez F, Keeble K, Shannon L (2009a) Which forcing factors fit? Using ecosystem models to investigate the relative influence of fishing and changes in primary productivity on the dynamics of marine ecosystems. Ecological Modelling, 220: 2972-2987. www.sciencedirect.com/science/article/pii/S0304380008004961

Beecham JA, Mackinson S, Aldridge J (2009) Dynamic linking of Ecosim and the GOTM biogeochemical model using plugins. In: Palomares, M.L.D., Morissette, L., Cisneros-Montemayor, A., Varkey, D., Coll, M., Piroddi, C. (eds.), Ecopath 25 Years Conference Proceedings: Extended Abstracts, pp. 134-135. Fisheries Centre Research Reports 17(3). Fisheries Centre, University of British Columbia 167 pp. ftp://ftp.fisheries.ubc.ca/FCRR/17-3.pdf

Mackinson S, Deas B, Beveridge D, Casey J (2009b) Mixed-fishery or ecosystem conundrum? Multispecies considerations inform thinking on long-term management of North Sea demersal stocksCanadian Journal of Fisheries & Aquatic Scence, 66: 1107–1129. www.nrcresearchpress.com/doi/pdf/10.1139/F09-057

Blanchard JL, Jennings S, Law R, Castle MD, McCloghrie P, Rochet M-J, Benoît E (2009) How does abundance scale with body size in coupled size-structured food webs? Journal of Animal Ecology 78 , 270–280. http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2656.2008.01466.x/abstract

van Leeuwen S, Blanchard JL (2009) "Linking physical processes to higher trophic levels: results from coupling hydrological, biogeochemical and size-based food web models". ASLO Aquatic Sciences Meeting, Nice, France (February 2009)

Forster RM, Blanchard JL, Jennings S (2009) Remote sensing and the ecosystem approach: integrating


phytoplankton ecology, benthic processes and food web models. ASLO Aquatic Sciences Meeting, Nice, France (February 2009)

Law R, Plank MJ, James A, Blanchard JL (2009) Size-spectra dynamics from stochastic predation and growth of individuals. Ecology: Vol. 90, No. 3, pp. 802-811. www.esajournals.org/doi/pdf/10.1890/07-1900.1

Reeves, S.A., Reveill, A.S. and Pinnegar, J.K. (2009) The past, present and future of technical measures in the Common Fisheries Policy. ICES CM 2009/R:04 www.ices.dk/products/CMdocs/CM-2009/CM2009.pdf

ICES (2010) Report of the Working Group on Multispecies Assessment Methods (WGSAM). 16-10th October 2010, San Sebastian, Spain. ICES CM 2010/SSGSUE:05. www.ices.dk/reports/SSGSUE/2010/wgsam10.pdf

Beecham JA , Bruggeman J, Aldridge JN, Mackinson S (2010) Linking Biogeochemical and Upper Trophic Level Models using an Xml based Semantic Coupler. ICES CM 2010/ L02. www.ices.dk/products/CMdocs/CM-2010/L/L0210.pdf

Blanchard JL, Law R, Castle MD, Jennings, S (2010) Coupled energy pathways and the resilience of size-structured food webs. Theoretical Ecology, 4: 289-300. www.springerlink.com/content/3389515172386243/

Blanchard JL, Coll M, Trenkel VM, Vergnon R, Yemane D, Jouffre, D, Link J, Shin Y-J (2010) Trend analysis of indicators: a comparison of recent changes in the status of marine ecosystems around the world. ICES Journal of Marine Science. 67: 732–744. http://icesjms.oxfordjournals.org/content/early/2010/01/07/icesjms.fsp282.short

Shin Y-J, Shannon L J, Bundy A, Coll M, Aydin K, Bez N, Blanchard JL, Borges MF, Diallo I, Diaz E, Heymans J J, Hill L, Johannesen E, Jouffre D, Kifani S, Labrosse P, Link JS, Mackinson S, Masski H, Möllmann C, Neira S, OjaveerH, ould Mohammed Abdallahi K, Perry I, Thiao D, Yemane D, Cury PM (2010) Using indicators for evaluating, comparing, and communicating the ecological status of exploited marine ecosystems. 2. Setting the scene. ICES Journal of Marine Science, 67: 692–716. http://icesjms.oxfordjournals.org/content/67/4/692.abstract

Blanchard JL (2011) Body size and ecosystem dynamics: an introduction. Oikos,. 120(4):481-482 (Editorial) http://onlinelibrary.wiley.com/doi/10.1111/j.1600-0706.2010.19564.x/abstract

Yvon-Durocher G, Reiss J, Blanchard JL, Ebenman B, Perkins D, Reuman DC, Thierry A, Woodward G, Petchey OL (2011) Across ecosystem comparisons of size structure: methods, approaches, and prospects. OIKOS. 120(4):550-563. http://onlinelibrary.wiley.com/doi/10.1111/j.1600-0706.2010.18863.x/abstract

Woodward G, Blanchard JL, Lauridsen R, Edwards FK, Jones JI, Figueroa D, Warren P, Petchey OL (2010) Individual-based food webs: species identity, body size and sampling effects. Advances in Ecological Research 43: 211-266. http://www.sciencedirect.com/science/article/pii/B978012385005800006X

Mackinson S, Wilson D, Galiay P, Deas B (2010) Engaging Stakeholders in Fisheries and Marine Research. Marine Policy 35 (1) 18-24. http://www.sciencedirect.com/science/article/pii/S0308597X10001375

Shin Y-J, Bundy A, Shannon LJ, Simier M, Coll M, Fulton EA, Link JS, Jouffre D, Ojaveer H, Mackinson S, Heymans JJ (2010) Can simple be useful and reliable? Using ecological indicators for representing and comparing the states of marine ecosystems. ICES Journal of Marine Science, 67(4): 717-731. http://icesjms.oxfordjournals.org/content/67/4/717.abstract

Smith ADM, Brown CJ, Bulman CM, Fulton EA, Johnson P, Kaplan IC, Lozano-Montes H, Mackinson S, Marzloff M, Shannon LJ, Shin Y-J, Tam J (2011) Impacts of fishing low trophic level species on marine ecosystems. Science 333 (6046) 1147-1150. http://www.sciencemag.org/content/333/6046/1147

Heymans J J, Mackinson S, Sumaila U R, Dyck A, Little A (2011) The Impact of Subsidies on the Ecological Sustainability and Future Profits from North Sea Fisheries. {PLoS} {ONE}. 6:e20239. www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0020239

Castle MD, Blanchard JL, Jennings S (2011) Predicted effects of behavioural movement and passive transport on individual growth and community size structure in marine ecosystems. Advances in Ecological Research. 45, 41–66. www.sciencedirect.com/science/article/pii/B9780123864758000022

Daskalov GM, Mackinson S (2011) Modelling possible food-web effects of aggregate dredging in the eastern English Chanel (MEPF 08/P37). Marine Aggregate Levy Sustainability Fund (MALSF), March 2011.65pp. www.cefas.defra.gov.uk/media/463480/mepf%2008%20p37_final%20report.pdf

Beecham J.A., Bruggeman J., Aldridge J. N. and Mackinson S. (submitted) A Coupled Model of Upper and Lower Trophic Levels in a Marine Ecosystem. Journal of Marine Systems.


Documents

General enquiries on this form should be made to:randd.defra.gov.uk/Document.aspx?Document=10587_MF1202... · Web viewMarine Ecology Progress Series 253, 1–16. Hall SJ, Collie JS,