The invasive species problem, directly or indirectly, was addressed in several previous MARBENA conferences; probably that is why the number of participants in this particular session was not too high. In total, ten scientists (Fernando Boero, Emil Olafsson, Vadim Panov, Henn Ojaveer, Anna Occhipinti, Donatella Del Piero, Ahmet Kideys, Ferruccio Maltagliati, Bella Galil and Sergej Olenin) contributed with twenty messages. The importance of bioinvasion studies was not challenged in any of the messages, in opposite, the practical significance of this field of marine biodiversity research repeatedly has been stressed. A variety of expressed opinions and raised questions may be grouped conventionally into three main themes: 1) ecological and economic impacts of marine bioinvasions, 2) vectors of introductions, and 3) "difficult questions" in marine bioinvasion research.

Ecological and economic impacts of marine bioinvasions. A. Occhipinti, E. Olafsson and H. Ojaveer indicated that there is the obvious lack of studies that convincingly demonstrate ecological and economic impacts of marine bioinvasions. H. Ojaveer stated that invasive species are of greater concern in the US because the Americans got "some very bad exotics". Anna Occhipinti stressed that she does not "believe Europe is more resistant to species introduction", but she agreed that, possibly, bioinvasion "impacts are less visible as they are unknown, while biodiversity alterations and ecosystem impairment has usually been associated to other environmental issues, such as eutrophication, pollution etc". A. Kideys, F. Boero and B. Galil gave several examples of detrimental and rather "convincing" effects of biological invasions in the Black and Caspian Seas as well as in the Mediterranean. E. Olaffson pointed out that "the effects of an invader on native communities using correlative data alone must be taken with utmost caution". Natural environmental oscillations and other anthropogenic factors such as overfishing or eutrophication may cause detrimental effect on native biota and coincide in time with species invasions.

The invasion biology has now become a rapidly growing research area, but yet there are large gaps in our knowledge on bioinvasion impacts, especially in marine environment. The ecological and economic consequences of bioinvasions may be of interest not only from purely practical, but also from theoretical point of view helping to understand mechanisms of functioning of natural ecosystems and their interaction with socio-economic factors. In this respect it is worth to mention that MarBEF (Marine Biodiversity and Ecosystem Functioning, an EU Network of Excellence, www.marbef.org) is organizing a special workshop "Aquatic invasive species and the functioning of European coastal ecosystems" to be held in the end of January, 2005 at Alfred Wegener Institute (AWI), Island of Sylt in Germany (see "Events" at the MarBEF sites for more information). V. Panov mentioned also several other recent European initiatives, one of them being the new Strategic Targeted Research Project DAISIE (Delivering Alien Invasive Species Inventories for Europe), which will start next year.

Vectors of introductions. Bella Galil indicated that the increase in shipping-related invasions in the Mediterranean, documented in a recent series of CIESM Atlases on 'Exotic species (www.ciesm.org/atlas/), may be attributed to the increase in shipping volume throughout the region, changing trade patterns that result in new shipping routes, improved water quality in port environments, augmented opportunities for overlap with other introduction vectors, and … rising awareness and research effort. She also stressed that "with the IMO ban on organotin compounds in antifouling systems in force for close to 2 years, it is timely to check whether by concentrating our efforts on ballast transfer we are not missing on the fouling transport". Emil Olaffson, however, pointed out that man made rafting rubbish may be of much more importance than ballast water in transfer of especially smaller invertebrates and microorganisms. It was indicated (S. Olenin) that in spite of several recent comprehensive accounts the role of inland waterways in species transfer remains largely unknown, also much more attention should be paid to such vectors as aquarium and live food trade. Assembling various opinions Anna Occhipinti formulated the most urgent research questions as: 1) consideration of alternative transfer vectors, other than ballast waters, and 2) spreading dynamics from the inoculation point, incl. secondary transfer and vectors involved.

"Difficult questions" in marine bioinvasion research. Donatella Del Piero and Ferruccio Maltagliati pointed out that we should distinguish between "accidental invasions" and intentional species introductions, which are beneficial (some of them, e.g. supporting our food industry). In general, the problem of bioinvasions should not be politicized and by no means should parallels be made between xenophobia in human world and attempts to control unwanted species invasions.

Also, F. Boero raised a very important ethical question of "how to convince the politicians of the usefulness of what we are doing in a serious way". In other words, how to show the importance of bioinvasion research without succumbing to "publicity rising hysteria" about "killing algae", "fish-eating comb jellies" and other mass media clichés. Fernando's "difficult" question was supported by Ferruccio Maltagliati, who ironically stated that "scientists' efforts are directed to extrapolate catastrophic predictions resulting from the introduction of a tiny worm in order to get funds for research..." To a certain extent this topic ("Importance of serious and honest dialog between science and policy makers, between scientists and general public") relates to all themes touched on in the present and previous MARBENA e-conferences, and, probably, it is worth a special discussion.

As stated in the introduction to the topic, one of the most important commercial fish species in the Baltic Sea is cod. Management of the Baltic cod hasn't been successful and this was discussed during the current e-conference. Henrik Sparholt stated that the most important problem for managing the Baltic cod stock is lack of enforcement of TACs and other management measures. The estimated illegal landings in 2000-2003 were 20-30% of the officially reported catch and independently of constant attempts to reduce fishing mortality at least over the last decade, fishing mortality has been not affected. He suggested that in order to build up the cod stock (the fishers are actually loosing around 100 million Euro per year due to the reduced productivity) a possible way forward could be a system of closed areas. This will take the focus from the control and the fleet capacity reduction, both issues which have politically proven difficult to achieve. The system of closured areas of about 30-50% of the Baltic Sea was suggested where hot spot juvenile areas and major parts of spawning grounds should be included. Arguments in favour of the proposed system: (1) easy to control; (2) easy for fishers to follow; (3) simple to explain; (4) transparent; (5) fair; (6) the fishers do not have to stop fishing right away, because 50-70% of the Baltic will still be open; (7) possibly even larger TAC's allowed (as discards and illegal landings are reduced); (8) safeguard benthos/ecosystem at the same time for the closed areas. Arguments against the proposed system: (1) low catches if fish learn and move into the closed areas; (2) difficult to justify precisely which areas and how much in total should be protected; (3) probably impossible to maximize yield and economic profit (the same can be said for the current situation) but the new system has the big advantage that the stocks will not suffer.

Discussion from the MARBENA earlier e-conference 'Sustainable Livelihoods and Biodiversity' held in April 2004 on 'European biodiversity action plan for fisheries: issues for non-target species' concluded that only the combination adequately designated and managed closed areas in combination with reduced fishing effort will achieve a sustainable ecosystem approach to management. This was reminded by Henn Ojaveer who further mentioned that, therefore, implementation of closed areas in the Baltic Sea may help cod to build up its stock, but is not necessarily sufficient to ensure sustainable management of marine living resources.

The proposed discussion topic on 'Why fish biodiversity in general is currently considered as an unimportant matter when managing fisheries?' was addressed by Pascal Lorance. It was stated that for the assessed stocks relationships between catches, effort and state of the stock are well understood. The effect of biomass removal is quantified and the effect of measures to change the level of exploitation can be forecasted. But there is nothing comparable for fish biodiversity. Considering biodiversity consisting of three components (i) diversity of sizes, age, genes, populations within a species; (ii) taxonomic diversity; (iii) diversity between ecosystems, we have little assessment of the fish biodiversity at local regional and global scales. Relatively more is known for taxonomic diversity of fish assemblages, number of fish species and their relative abundance, but even on this there is very little operational and only some general rules are understood. On the other hand our knowledge on the relationship between local fishing effort, type of fishing gear used and fish species diversity is rather limited. These relationships seem to be very complex. Finally, P. Lorance concluded that fish biodiversity is considered as an unimportant matter when managing fisheries because we have too little to offer to managers in terms of: (1) which changes have occurred in fish communities; (2) what are the reasons behind; (3) which further changes can be predicted; and (4) which measure are appropriate to prevent further changes (if necessary).

The EU has asked the member states to consider the establishment of Special Protected Zones (under the Habitat and/or Bird Directives) or Marine Protected Areas (in relation to OSPAR) in the open sea, in their EEZ, also outside the 12 miles zone. Germany has submitted its proposal for several areas, the other states are preparing their proposals. At the moment there is a lot of debate about the criteria which should be used to establish these areas. What are sandbanks, reefs, or gasseeps? Do MPAs protect individual species? Are MPAs necessary to protect biodiversity, and do they protect biodiversity?

The Plaice Box as example of an MPA? What is the Plaice Box? The North Sea flatfish fisheries generate considerable numbers of discards, especially of Plaice Pleuronectes platessa in coastal waters. To reduce discard mortality, a partially closed area was established (Plaice Box) in 1989 in the coastal waters along the continental coast. Since 1995, beam trawl fishery in this area was prohibited for vessels larger than 300 HP. Ever since, the plaice stocks have gone down, and fishermen claim that re-opening is justified, since it does not solve any problems.

Is the Plaice Box, with a reduced fishing effort, a representative example of a protected area? The following reactions were ventilated in the discussion:

Christian von Dorrien: I don't think that the Plaice Box has been a representative example for a protected area. The reason for this is mainly that the objective - to protect juvenile plaice - was not achievable, as long as small vessels are allowed in this area. And in addition, the fishing effort of the small vessels has increased substantially during that time, as far as I know. Although I personally think that MPAs could be a valuable measure to protect species and ecosystems (like, f. ex. deep sea coral reefs), it could be misleading to sell them as a panacea. At least, in temperate seas and for mobile species, like cod.

Michel Kaiser: Clearly the main problem with the plaice box was that it was a compromise to meet both Governmental and industry objectives. What it achieved was to exclude larger vessels from an area of the seabed leaving smaller vessels to fish without effort control (we know little about this effort as these vessels did not qualify for the vessel monitoring system). As long as chronic disturbance occurs on the seabed the system is likely to remain in an altered state with no opportunity to recover.

Mark Costello: There is an extensive literature on the effects of MPA on fish stocks. Where protection is complete. i.e. no fishing of any kind (including angling), numbers and size of individuals of the fished stocks always increase (seems obvious). However, most MPAs are not sufficiently protected or are far too small to produce measurable benefits to fisheries. My guess is that there has not been sufficient protection of the population, or the plaice box is too small in relation to the population and its mobility, or both.

Rob Witbaard: The Plaice Box was introduced as example of a proposed protected area in the south eastern North Sea, an area intended to protect juvenile plaice in the life stage before recruitment to the commercial exploitable stock. The support for this idea would never have been obtained if it did not contain the promise that such a measure, on the long term, would lead to a greater spawning stock and a sustainable yield. But is the economic perspective the only justified argument to designate a protected area?

Jan Geert Hiddink: Our studies in the North Sea show that the effect of the first ever bottom trawling disturbance in a pristine habitat on benthic biomass and species numbers is larger than the effect of trawling once in an area that is already heavily trawled. This means that a reduction in trawling intensity in the plaice box will only have a significant positive effect on benthic communities if this means that trawling intensity is reduced to a low level (e.g. a trawling intensity of less than once every two years). If bottom trawling was reduced from a high to an intermediate level, no positive effect of the plaice box on benthic communities can be expected.

Paolo Guidetti: The main problem is that most of the expected benefits of MPAs are always defined as 'potential' since robust data are still lacking, except for the well known evidence (but there are exceptions) that within MPAs there is a substantial recovery (eg in terms of density, biomass and size) of those fish species that elsewhere are strongly impacted by fishing. Scientific data about spillover are lacking for many regions, and very little is known about the potential benefits of an enhanced egg/larval dispersal due to the increased number and size of spawners within the MPAs. With regard to the benefits for fisheries, moreover, there are two points that would deserve to be considered: 1) e.g. in the Mediterranean fishing is exerted since long (centuries), so, why do we expect that recovery happens in a few years after having protected a negligible area compared to the areas open to fishing? 2) There are species, like the dusky groupers, which disappeared from many areas. They can live for decades, so that recovery needs decades, while many of the MPAs in the Mediterranean are far younger. All the above issues demonstrate the importance to include some (apparently obvious) life history traits (besides proper experimental designs) before drawing any conclusion about the potential effects of protection on species impacted by fishing.

Erika Washburn: I support the comment about life histories and also point out the importance of potentially extending the baseline back in time with historical and archaeological information. This can dramatically affect our understanding of a healthy ecosystem. Please see the following link for the Census of Marine Life's History of Marine Animal Populations program, which includes work in the North Sea: http://www.cmrs.dk/Default.asp?ID=1 Or for social science tools for marine protected areas: http://www.noaanews.noaa.gov/stories2004/s2340.htm

Filip Volckaert: It is clear that man has a hard time to exploit rather complex ecosystems in a sustainable way. With our poor understanding of the marine ecosystem we thought that monitoring (poorly defined!) target species stocks one-by-one and year-by-year would suffice to control the fish resource. Alas that regime has failed. The food web has shrunk in size through the elimination of the top predators, the microbial loop has benefited from the modified ecosystem, populations of fish have disappeared all together, the sediment surface has been modified by trawling and the enhanced input of organic matter, eutrofication has modified coastal ecosystems and on top of that a shifting climate has shaken the ecosystem. At the moment, multi-year ecosystem-based multi-species management seems an alternative. Fairly large sized MPAs with a "natural delimination" have to be implemented for long periods to come. Amongst others, the Southern Bight of the North Sea (the major spawning ground of the North Sea), the Friesian Front, the Shetland Islands, the Doggerbank and all estuaries represent ecologically logical choices.

Sabine Christiansen: Experimental areas (as for example also proposed by ICES in 1995) are urgently required to improve our understanding of ecosystem processes with and without human pressure - in this sense, MPAs (which by definition serve the conservation of marine life, not one group in particular) will serve a double purpose: the immediate release of pressure in the areas concerned, and a test field for learning about whether and what alterations occur after removing direct impacts, about the time scales involved and the size and connectivity of areas needed. In 2003, the governments of the North East Atlantic states, as contracting parties to OSPAR, signed up for "establishing an ecologically coherent network of well-managed MPAs " by 2010. So far there is not a single MPA in the NE Atlantic offshore, which is managed sufficiently to reduce human impacts. Time is running out.

Rob Witbaard: The necessity of marine protected areas to protect macrobentic organisms/communities is without doubt because it is evident that they suffer from the side effects of marine resource exploitation. The function of an MPA as tool to protect marine biodiversity in the sense of keeping species richness high may be somewhat controversial because we not really know what the meaning of such biodiversity is. Does biodiversity have a functional role. We neither know whether designation of MPA's really lead to a conservation of biodiversity. But if they do protect biodiversity, one may ask the question whether a unique habitat with only a few, locally rare, species is of less value than a habitat with a manifold of species. In other words the habitat uniqueness may be important as well. But would we be willing to protect such habitat? Could we make a separation between conserving habitats and conserving species or is the linkage between both so strong that we cannot separate between them? Is it anyhow necessary to make such a separation?

Han Lindeboom: In the case of the southern North Sea, we know a lot about the direct effects of beam trawling. It is huge and the short-term effects are easy to prove. We do not know enough about the long-term effects, but if one catches or kills large parts of the population continuously, it must have an effect.

Now, an argument can be, that we do not want these effects in all places at all times. Maybe we should want to create areas that develop as they can develop without a huge damaging fishing pressure. If that becomes a goal, managerial it is easy to set aside an area and let it do its own developing. In that case we only have to monitor the fishing pressure. As long as that is zero we have reached our goals.

In the North Sea local biodiversity has decreased due to human activities like fisheries. If we do not create proper Marine Protected Areas and implement a rigid regime of protection, the deterioration of the North Sea ecosystem will continue and local biodiversity will further decrease.

Introduction
The marine ecosystems most threaten are those situated in coastal zones. Roughly half of the shorelines of the continents are threatened by development. In some continents, the percentage of degraded coasts is much greater. In Europe, it is considered that 86% of the coastal perimeter is at risk (moderate and high risk), which also means that the habitats and the species to which they are associated are also at risk. Along with the occupation and destruction of habitats and that of pollution, one of the other major threats to marine biodiversity is the excessive exploitation of resources. Fish makes up 16% of the world's supply of proteins for human use.

Fishing has reduced numerous fish populations to very low levels. Those affected are to be found in the most varied marine ecosystems: fish that live on continental platforms such as halibuts and cod, submarine seamount fish, such as some species of redfish, including the orange roughy, and fish of vast pelagic distribution such as swordfish, albacore, and bluefin tuna. In some regions of the oceans, over-fishing has reduced stocks to half of their original maximum amount. It seems difficult to find refuge for marine life. The North-Atlantic Ocean, with its long time tradition of fishing, is now a threaten sea in need of urgent ideas and good practices in view of a return to a sustainable state.

The Azores case
The Azores, the most isolated archipelago of the North Atlantic, consists of nine volcanic islands and several small islets, forming three groups along a tectonic zone in the middle of the Atlantic, is of considerable conservation and marine biological interest with an interesting experience on conservation of marine habitats and species related with regional, national and European instruments, the most relevant being the Habitats Directive for the Natura 2000 network.

Natura 2000 and beyond
The Habitats Directive has led to large numbers of Marine Protected Areas being established by EU Member States and over a relatively short period of time (compared to progress with MPAs before the Directive came into force).

This has required considerable effort in the areas such as data collection, developing supporting legal regimes, provision of information and opening up opportunities for stakeholder participation. In the EU, the Natura 2000 network is likely to be where those working on MPAs are most likely to concentrate their effort for at least the next decade. This raises some important questions. Will the Natura 2000 network provide what we want from a European MPA network? Is our concentration on Natura 2000 distorting/limiting what we might achieve from MPAs? Are we losing sight of other equally important marine conservation objectives by concentration on Natura 2000?

Will decision-makers get enough from the science being done at these sites, to enable them to implement the marine conservation agenda required by international frameworks such as the Biodiversity Convention?

Certainly, there is lots of good management activity marine SACs (e.g in the UK, Azores, Madeira, Canaries, etc) - so they have been (mainly) worth it. However, the effort may be now blindly pursuing Habitats Directive sites without considering if the categories of habitat in the Directive are meaningful or if the Directive can identify threatened and declining marine habitats and species comprehensively for MPA's - especially now that it is being applied offshore.

The definition of biogeographic boundaries is an exercise that is not complete here in Europe. An operational network of MPAs might serve to define a biogeographic network of MPAs, reaching then a goal that is relevant for management. The identification of discontinuities in the distribution of species and habitat types is essential to properly manage biodiversity.

The Habitats Directive was designed, with recognised insufficiencies, for inshore marine and, in fact, never intended for offshore. At the same time it is partiaaly distorted in terms of habitat (and species) covered. There is a need to go beyond Natura 2000, OSPAR has done some good thinking - let's use it.

Natura 2000 in the Azores: strengths and weaknesses So far, the conservation of the resources and habitats inside the Azorean waters has certainly benefited from the precautionary policy of exploitation and maintenance of the artisanal character in many fisheries and small capacity of the one with a more industrial character. Regarding coastal conservation, however, the establishment of the first MPAs (previous to the implementation of the Habitats or Birds Directives) hasn't done much for crediting them near the fishermen as a management tool. The detailed regulations were never issued for most of the areas and the few measures set by their designation acts have been poorly and insufficiently enforced. Implementation of Natura 2000 improved, but did not solved the issue.

A wider and more thorough habitat surveys were and are still are required if a region wants to ensure that they focus their conservation efforts on the sites that are really important due to their dimensions, representativeness or major role in maintaining ecosystem functioning. However nature conservation has also an aspect of opportunism and that has been used to designate sites in the absence of broader assessments. Even when broader surveys find out new sites that turn out to be more relevant, the initial ones will still have a role to play on the conservation network. This is because from the time when they were designated, some of them have become flagship sites for certain features (e.g., either because they are more easily accessible to the public or because certain features have been studied more deeply in that particular location) and societal support for conservation efforts could be affected if they appear to be dropped.

The transformation of the "paper MPAs" into a living, practical, credible and successful management tool (be their goals driven by fishery resource management, conservation of particular feature or preservation of recreational seascape values). With more or less depth, management plan proposals have already been elaborated for all of the Natura 2000 sites in the Azores. They mustn't be forgotten on a shelf or be used just as papers to flaunt before the European Commission. Logistics are not even too complicate for enforcing most of the Natura 2000 marine sites and previously existing MPAs in the Azores provided their coastal nature and easy access from neighbouring towns and villages. Management bodies need now to work close enough to their scientific partners, but also to main other stakeholders.

Conclusions
I could relish on what has been achieved so far in the Azores with all the planning documents for Natura 2000 already finished and a good practice of collaboration between administration and scientific institutions. However in practical terms the work to provide a robust protection to the SACs still lies head.

Three very important steps are missing for the correct implementation of MPAs in the Azores, and should be concluded until 2006 for the case of Natura 2000 sites:

• Public discussion
• Active management
• Enforcement and management.

Participants
Based on the contributions made by the participants at the at this section of the MARBENA e-conference: Ferdinando Boero, Fernando Tempera, Frederico Cardigos, Jorge Fontes, Keith Hiscock, Ricardo Serrão Santos, Simon Claus and Susan Gubbay.

The Wadden Sea is a shallow estuarine area between the North Sea and the coasts of Denmark, Germany and the Netherlands. About 50% of this area consists of tidal flats harbouring rich stocks of shellfish and very large numbers of shorebirds. Commercial exploitation of shellfish focuses on blue mussels (Mytilus edulis) and cockles (Cerastoderma edule). The exploitation of shellfish led to the disappearance of some of the more important natural values of the Dutch Wadden Sea around 1990. After a series of years without recruitment of young cockles and mussels, and with continued fishing, almost all cockle and mussel beds had disappeared. Shellfish feeding birds showed a strongly increased mortality. Therefore, the Dutch government decided to develop a new fisheries policy for the exploitation of the Dutch Wadden Sea. A research programme EVA I and EVA II was started in 1993 to determine if the new management measures enabled commercial shellfish exploitation in a nature conservation area without causing significant damage to the area.

Under the new rules for fishing shellfish, three main objectives were formulated, viz.

• permanently closing for fisheries of 26% of the total area of tidal flats in the Wadden Sea in order to restore lost types of ecosystems, such as seagrass beds and mussel beds
• reservation of a sufficient amount of food for shellfish-feeding birds in years with low stocks of shellfish
• introduction of a system of co-management in the fisheries in order to create awareness by the fishermen and to avoid control problems.

The EVA project concluded that the goal of restoring the area of stable intertidal mussel beds to between 2000 and 4000 ha had been achieved. Restoration of seagrass beds could have reached its objectives, but in practice new seagrass beds sometimes were fished away accidentally. It was concluded that the food reservation policy failed to provide sufficient guarantees against food shortages due to fishery, in part because food reservation levels were not set high enough.

In the final report all relevant information from the EVA study as well as from studies published elsewhere, were brought together. The EVA team did not draw any policy conclusions; this was considered the domain of the policy makers.

At the end of the day the Dutch government decided to terminate the cockle fisheries as per 1 January 2005 since it was concluded that cockle dredging could not be transformed into a sustainable fishery without damage to the conservation values of the Wadden Sea. For the mussel culture it was concluded that for this fishery prospects for sustainability existed. The industry is given 15 years for it's further transformation into a more environmentally friendly activity.

The Dutch approach was based on three pillars: areas closed for fisheries (MPAs), catch limitations, and co-management of the fishery by the government and the fishermen. The closed areas demonstrated well how tidal flats habitats could develop without fisheries but they were not needed as source areas for recolonization of the fished flats. There is also no indication that they enhanced the harvest of the shellfish fisheries in the areas open for fisheries. Catch limitations on cockles did not serve to protect the stock of cockles but rather to protect other organisms, such as shellfish-feeding birds. The system of co-management in general worked well. The fishermen, in their fishing plans, took care to obey all measures set by the government and in general complied with these rules. However, in the end it appeared that part of the rules were wrong.

The EVA studies showed that the Dutch government was willing to invest heavily (€ 3.5 million) in scientific research to solve the societal problem of conflicting interests between shellfish exploitation and biodiversity conservation. It also showed a willingness to organize these investigations in such a way that the general scientific requirements of transparancy, peer review and publication were met. In the end far-reaching decisions on the future of the shellfish industry were based on the EVA studies. However, the role of science was minimized in the design of the new policy, which is up for evaluation in 2010.