Fisheries Research 238 (2021) 105894

Contents lists available at ScienceDirect

Fisheries Research

journal homepage: www.elsevier.com/locate/fishres

Niche overlap among anglers, fishersand cormorants and their removals of fish biomass: A case from brackish lagoon ecosystems in the southern Baltic Sea

Robert Arlinghaus a,b,*, Jorrit Lucas b, Marc Simon Weltersbach c, Dieter Komle¨ a, Helmut M. Winkler d, Carsten Riepe a,c, Carsten Kühn e, Harry V. Strehlow c a Department of Biology and Ecology of Fishes, Leibniz Institute of Freshwater Ecology and Inland Fisheries, Müggelseedamm 310, 12587, Berlin, Germany b Division of Integrative Fisheries Management, Faculty of Life Sciences, Humboldt-Universitat¨ zu Berlin, Invalidenstrasse 42, 10115, Berlin, Germany c Thünen Institute of Baltic Sea Fisheries, Alter Hafen Süd 2, 18069, Rostock, Germany d General and Specific Zoology, Institute of Biological Sciences, University of Rostock, Rostock, Germany e Mecklenburg-Vorpommern Research Centre for Agriculture and Fisheries, Institute of Fisheries, Fischerweg 408, 18069, Rostock, Germany

ARTICLE INFO ABSTRACT

Handled by Steven X. Cadrin We used time series, diet studies and angler surveys to examine the potential for conflict in brackish lagoon fisheries of the southern Baltic Sea in Germany, specifically focusing on interactions among commercial and Keywords: recreational fisheries as well as fisheries and cormorants (Phalacrocorax carbo sinensis). For the time period Fish-eating wildlife between 2011 and 2015, commercial fisheries were responsible for the largest total fish biomass extraction Commercial fisheries (5,300 t per year), followed by cormorants (2,394 t per year) and recreational fishers (966 t per year). Com­ Human dimensions mercial fishing dominated the removals of most marine and diadromous fish, specifically herring (Clupea Conflicts Recreational fisheries harengus), while cormorants dominated the biomass extraction of smaller-bodied coastal freshwater fish, spe­ cifically perch (Perca fluviatilis) and roach (Rutilus rutilus). Pike (Esox lucius) as large-bodied freshwater fish was the only species where recreational fisheries were responsible for the major fraction of the annual biomass extraction. A strong trophic overlap and hence a similar foraging niche was documented among commercial fishers and recreational anglers and among non-resident and resident anglers, indicating that the aversion expressed by anglers against commercial fisheriesin a survey had an objective underpinning related to resource competition. By contrast, the foraging niches of cormorants and of both fishers and anglers differed strongly as evidenced by largely non-overlapping sets of species that were caught and removed by cormorants and by commercial as well as recreational fishers. However, for individual species of commercial and recreational in­ terest, specificallyperch, cormorants were responsible for a major fraction of total biomass extraction, suggesting that at the individual fishspecies level competition with fishersand anglers may still occur. In an angler survey, respondents expressed a preference for cormorant control, indicating the existence of conflict between fisheries and cormorants. We recommend that conflicts in the lagoon fisheries be proactively managed, e.g., through improved communication, zoning, predator control and outreach. Further research should clarify the population- level impacts of cormorants on target species of commercial and recreational fisheries as well as the relative impact of commercial and recreational fisheries on selected species of joint interest.

1. Introduction interference (Jacob and Schreyer, 1980). Accordingly, stakeholders may perceive reduced satisfaction attributed to somebody else‘s behavior, Conflicts in fisheries are well documented (Bavnick, 2005; Bishop which produces outcomes that conflict with their own expectations and Samples, 1980; Kearney, 2001, 2002; Spijkers et al., 2018). From a (Arlinghaus, 2005). For example, when two types of fishers, such as social science perspective, conflictamong stakeholders is caused by goal commercial or recreational fishers, target the same fish species (i.e.,

* Corresponding author at: Department of Biology and Ecology of Fishes, Leibniz Institute of Freshwater Ecology and Inland Fisheries, Müggelseedamm 310, 12587, Berlin, Germany. E-mail address: [email protected] (R. Arlinghaus). https://doi.org/10.1016/j.fishres.2021.105894 Received 9 December 2020; Received in revised form 26 January 2021; Accepted 28 January 2021 0165-7836/© 2021 Elsevier B.V. All rights reserved. R. Arlinghaus et al. Fisheries Research 238 (2021) 105894 have the same goal), elevated landings of a particular species by one Mixed commercial and recreational fisheries are present in many fishertype might be perceived by the other group as preventing reaching coastal waters (Ihde et al., 2011; Hyder et al., 2018). While commercial their own goals (e.g., expected catch) in the future. Conflicts do not fishers (Pauly et al., 1998) and recreational fishers (Donaldson et al., necessarily involve direct interactions and are often value-based 2011) may both desire the catch and harvest of adult upper trophic level (Arlinghaus, 2005; Boucquey, 2017; Loomis and Ditton, 1993). More­ predators, cormorants have been described as opportunistic foragers over, conflicts in fisheries are often asymmetric, i.e., they may only be that mainly target smaller-bodied, highly abundant species of lower felt by one party (Manning, 1979, 1999; Martinson and Shelby, 1992). trophic levels (e.g., Hansson et al., 2018; Keller, 1995; Suter, 1997). Conflictsmay be founded on the behavior or even the mere existence of However, in the coastal environment of the Baltic Sea there are also other people, groups, or animals. Well-documented examples in fisheries low-trophic level fish that are intensively targeted by commercial fish­ are the conflictaround fish-eatingwildlife (Brehens et al., 2008; Cowx, eries, e.g., herring (Clupea harengus) or cyprinids such as roach (Rutilus 2003; Steffens, 2010) or conflicts among resident and non-resident an­ rutilus), where impacts of cormorants may be economically relevant. ¨ glers that differ in values and behaviours (Aas and Skurdal, 1996). Work by Hansson et al. (2018) and Ostman et al. (2013) has suggested Solving conflicts typically requires some form of managerial interven­ that cormorants in the Baltic Sea exert their greatest impact on the more tion, including facilitation, communication and conflict resolution accessible coastal fish species rather than on those regularly using (Redpath et al., 2013). offshore habitats. Within the sector of fisheries, conflicts between commercial and The Baltic Sea is an example where commercial fishing for herring, recreational fisheriesare common in situations of co-exploitation of fish cod (Gadus morhua) and other marine species is prevalent, but where the stocks (Boucquey, 2017; Kearney, 2001, 2002; Loomis and Ditton, total removals of marine fish by recreational fishing are non-negligible 1993). Often, both groups blame each other for reduced access to fishing (Hyder et al., 2018; Radford et al., 2018). In the Baltic Sea cormorants grounds, interference during fishing, overfishing, damaging or stealing have increased strongly in abundance and may thus lead to competition ¨ of gears or illegal harvest (Dorow and Arlinghaus, 2012; Quinn, 1988; for fish, particularly in inshore coastal habitats (Ostman et al., 2013; Sullivan, 2003). Beyond competition for fish, fisher-angler conflicts Hansson et al., 2018; Mustamaki¨ et al., 2014; Veneranta et al., 2020). often have a root cause in different interpretation of value, e.g., is a fish We studied the lagoon fisheries around the Island of Rügen in the generating more value when captured by a recreational angler or a southern Baltic Sea in Germany as a model coastal European ecosystem commercial fisher, and general discussions about the moral appropri­ where commercial and recreational fisheries as well as cormorants are ateness of the use of fisheries resources by either fishers or anglers all abundant and jointly exploit a shared multi-species resource (Boucquey, 2017). composed of both freshwater and marine species. Pike (Esox lucius), Another common conflict in fisheries is, broadly speaking, intersec­ pikeperch (Sander lucioperca) and perch (Perca fluviatilis) are examples of toral when conservation and fisheriessectors collide (Arlinghaus, 2006; stenohaline freshwater fish that are common in the lagoon ecosystems Brehens et al., 2008; Hansson et al., 2018; Marzano et al., 2013; Steffens, around Rügen, and in many areas of the Baltic Sea under oligo- to 2010). For example, as a result of European bird conservation regula­ mesohaline conditions. These species are targeted by commercial fish­ ¨ tions, cormorants (Phalacrocorax carbo sinensis) have strongly increased eries with varying intensity (Ostman et al., 2013; Winkler, 1990), and in abundance in many areas of Europe, in some cases causing depletion they are also prime targets of recreational fisheries (Basan, 1988; Lars­ ¨ ¨ of local fish populations (Dieperink, 1995; Jepsen et al., 2019; Ostman son et al., 2015; Ostman et al., 2013; Weltersbach et al., 2021). Concerns et al., 2013; Rudstam et al., 2004; Skov et al., 2014; Veneranta et al., that fishing and other forms of predation might impact stock dynamics 2020). Protection of the cormorants in Europe has been elevated to a of freshwater predators in the Baltic Sea have emerged (Olsen, 2019; ¨ highly symbolic conflict between bird conservationists and fishers, Ostman et al., 2013). Specifically,cormorant predation may negatively creating a polarized management dilemma due to the multitude of ac­ affect the abundance of certain coastal species in the Baltic Sea, such as ¨ tors involved, poor scientific basis and the pan-European nature of pikeperch or perch (Hansson et al., 2018; Ostman et al., 2013; Mus­ cormorant population dynamics and associated conservation regula­ tamaki¨ et al., 2014; Winkler et al., 2014). Yet, important knowledge tions (Brehens et al., 2008; Marzano et al., 2013). Current scientific gaps remain. For example, Hansson et al. (2018) presented zero pike knowledge indicates that ecological predation effects of cormorants removals for the entire ICES subdivision 24 by cormorants. This does not depend on local context and are sometimes non-detectable (Lyach, agree with local studies in this area, which reported substantial re­ 2020), but can also be substantial and affect fish abundance and movals of fish by cormorants (Winkler et al., 2014). size-structure of fish populations (Klimaszyk and Rzymski, 2016; The present study had four objectives. The first objective was to ¨ Ostman et al., 2013; Veneranta et al., 2020). describe the development and current size of the three forager pop­ Certain conflicts in fisheries, e.g., those among commercial and ulations (commercial fishers, recreational anglers, and cormorants), recreational fishers targeting the same resource (Bishop and Samples, with a special focus on recreational fisheries, to deal with the wide­ 1980; Kadagi et al., 2021; Kearney, 2001, 2002; Ngoc and Flaaten, 2010; spread lack of local data on recreational angling harvest in the Baltic Sea ¨ Samples, 1989), may be framed from a human ecological perspective (Ostman et al., 2013; Hyder et al., 2018). The second objective was to drawing on the ecological theory of competition for resources among analyze the niche overlap of the three forager populations from the foragers (Berkes, 1984; Monk et al., 2018). Competition occurs when at brackish coastal zones of the German Baltic Sea following the methods of least two species or stages within species desire to acquire the same Berkes (1984) and thereby examining the potential for between-group scarce resource (Begon et al., 1996). One way to understand the degree conflicts. The third objective was to present novel estimates on the of possible competition among foragers is to measure and compare the total fish removals for the three forager types following the example of trophic niches of the foragers (Berkes, 1984) and to additionally assess Hansson et al. (2018) but tailored to local scales. The fourth and final and evaluate total resource extraction by each forager (Hansson et al., objective was to examine whether the perceptions of the human stake­ 2018). In humans, competition between stakeholder groups may arise holder groups about each other and preferred management actions when their trophic niches overlap substantially in terms of the compo­ would match objective data about niche overlap and resource extrac­ sition of fish species that are targeted. To properly understand the tion. We hypothesized a strong foraging niche overlap among com­ conflictpotential between different forager groups that involve humans mercial and recreational fisheries as well as among resident and it is important to measure both trophic overlap among foragers and their non-resident recreational anglers due to similar target species, and lit­ total resource extraction in the same environment, and simultaneously tle overlap among cormorants and either fishers or anglers. Following to use stakeholder surveys to understand whether specificgroups indeed earlier research on eel (Anguilla anguilla) fisheriesin the same study area are perceiving conflictand desire to regulate or constrain the party with (Dorow and Arlinghaus, 2012), we also hypothesized that anglers would whom they are in conflict. have a negative attitude towards commercial fisheries and cormorants,

2 R. Arlinghaus et al. Fisheries Research 238 (2021) 105894 specifically assuming both groups to have a stronger impact on target recent years (LUNG, 2017). The vast majority of the animals are located stocks than recreational fishing. at the coastline (LUNG, 2017), and we studied the diet composition of the two largest colonies in M–V located in the study area (Fig. 1). 2. Material and methods The lagoon fisheries are regulated through state-level fisheries reg­ ulations that specify licensing requirements for both commercial fishers 2.1. Study area and anglers, monitoring, enforcement, and harvest and gear regulations, such as minimum-length limits for popular species (for both fishersand The study area (Fig. 1) encompasses classical and choked lagoon anglers), daily bag limits (for anglers), annual quotas for selected species ecosystems and estuaries around the island of Rügen, Germany, covering (for fishersonly), and protected seasons or spawning closures (for both a total area of 1.584 km2 (Biester, 1991; Schiewer, 2008). The lagoons fishers and anglers). Moreover, regulations exist related to nursery are eutrophic to polytrophic and offer a mix of freshwater, marine and grounds where fishing is entirely or partly prohibited to facilitate diadromous fish species (Winkler and Schroeder, 2003). Typical upper recruitment of fish. Additional regulations that affect fishing include trophic level fish are pike, pikeperch and perch. Other commonly tar­ nature conservation efforts. Two national parks, a biosphere reserve and geted marine and diadromous fishspecies are the western Baltic spring a multitude of smaller conservation areas exist in the study area (Fig. 1). spawning herring, eel, several flatfish species and various cyprinid In these areas, year-round the use of motorboats and the access for an­ species, such as bream (Abramis brama) and roach. glers is often restricted, limiting recreational fishing effort. In the core The lagoon ecosystems are traditionally used by commercial and protection zones, limited access is granted to a small number of com­ recreational fisheries (Basan, 1988; Winkler, 1990). The fisheries of mercial fisherswith historic fishingrights, and in the biosphere reserve Rügen have been relevant herring fisheriesfor centuries (Raillard, 2012) recreational anglers need a special permit to fishin certain areas. Many and have been lucrative commercial fisheriesdue to a fixedprice policy of these regulations are motivated by international and national bird in the German Democratic Republic. After the German reunification in protection (e.g., EU Bird Directive 2009/147/EC) and by conservation 1990 and a shift from a centralized to a market economy, the com­ regulations that protect certain animals and habitat types (e.g., the EU mercial fisheries sector has declined in size, and the number of recrea­ Habitats Directive 92/43/EEC). The conservation areas are a source of tional anglers has increased. Today a larger recreational fishery effort recurring conflicts, particularly due to restricted access for anglers stems from non-residents than from resident anglers in the German (Vogt, 2020). For most freshwater and other inshore fishthe total yield federal state of Mecklenburg-Western Pomerania (Mecklenburg-Vor­ in the lagoon fisheries is unregulated as there exist no quotas for com­ pommern in German, M–V) (Dorow and Arlinghaus, 2009). mercial fisheries and only daily bag limits to anglers that do not limit The cormorant population in the state of M–V has grown from about total annual harvest. Some regularly occurring marine species in the 2000 breeding pairs in the late 1980s to over 14,000 breeding pairs in lagoons (e.g., herring and cod) are quota-regulated within the European

Fig. 1. Overview of the study area with the various lagoons in northern Germany and locations (triangles) of the two largest cormorant colonies where cormorant pellets were collected for analysis of consumption patterns. Abbreviations of lagoons are shortcuts of local names of the lagoons and small inlets. SAB = Saaler Bodden, BOD = Bodstedter Bodden, G = Grabow, KB = Kubitzer Bodden, SB = Schaproder Bodden, VB = Vitter Bodden, WB = Wieker Bodden, BEZ = Bretzer Bodden, BEG = Breeger Bodden, KJB = Großer Jasmunder Bodden, KJB = Kleiner Jasmunder Bodden, S = Strelasund, GB = Greifswalder Bodden, P = Peenestrom, SH = Stettiner Haff.

3 R. Arlinghaus et al. Fisheries Research 238 (2021) 105894

Common Fisheries Policy. an iterative proportional fittingapproach (Battaglia et al., 2009; Gabler Our analysis compiled the best available data on commercial and et al., 1994; for details on the weighting procedure, see Lucas, 2018; recreational landings and of the removals by cormorants for the entire Weltersbach et al., 2021). Trip data were extrapolated to the total brackish waters in the lagoons around Rügen (Fig. 1). We chose the time population collected by the CATI survey to produce an estimate of total period between 2011 and 2015 to aid data comparison, and supple­ catch or harvest per year separately for the resident and non-resident mented our primary data collection with secondary data where appro­ angler group. As this approach involved values from two random dis­ priate. Hence, we analyzed a mix of publicly available data and primary tributions, that is the total number of anglers in the population and the data that we collected. diary-based catches, the 95% confidence intervals around the extrapo­ lated catch and harvest figures were estimated using bootstrapping. 2.2. Assessment of the recreational angling sector Specifically, empirical catch data, derived from weighted diary data (including zero values), were summed separately for the resident and In the study area, recreational fishing is open access but anglers non-resident group. To that end, a normal distribution of total catch was require a valid general fishing permit and an additional coastal fishing constructed drawing from the summated catch data with replacement permit of M–V. Harvest reporting is not mandatory to anglers. There­ and the normal distribution of the total number of anglers including fore, no regular recreational catch statistics exist, and the landings by their associated mean and standard deviation of the distribution (N = anglers must thus be estimated through randomized surveys. In 2014/ 5000 samples per resident group). The resulting distribution from the 2015, we conducted a representative, nationwide survey using 5000 samples was used to calculate 95% confidence intervals. computer-assisted telephone interviews (CATI) followed by a 1-year The estimation of the total numbers of anglers fishingin the coastal paper-and-pencil diary study with quarterly follow-ups to collect rec­ lagoons was based on selecting anglers from the nationwide sample that reational fishing data (see Weltersbach et al., 2021 for details). The had reported to have fishedat least one day in the past 12 months in the study approach by Weltersbach et al. (2021) in 2014/2015 followed an study area around Rügen. Similarly, catch estimation was based on diary earlier telephone-diary study in the same region for the time period data entries selecting only those trips that were conducted in the study 2006/2007 (Dorow and Arlinghaus, 2011). Both data sets contributed area. data to the present study. To calculate total catch in kg of each species the number of harvested The CATI survey in 2014/2015 was designed to identify marine fish was multiplied with an average weight per species. The average anglers in the German population including those fishing in coastal la­ weight per species was based on length data from a previous study goons and to recruit participants for the subsequent 1-year diary study examining recreational catches in the same coastal area (Dorow and intended to measure catch and effort. The CATI interviews were con­ Arlinghaus, 2011). The species-specific length-weight relationships ducted in nine out of 16 German federal states. For the other federal were taken from www.fishbase.org (Lucas, 2018). states, the angler incidences and angling effort were estimated using a set of reference federal states (Weltersbach et al., 2021). Random digit 2.3. Assessment of commercial landings dialing was used to generate telephone numbers and contact house­ holds, with selection probabilities proportional to the number of Commercial fishersare obliged to report landing data by species and households per municipality. Up to eight attempts were made to contact area either daily (boats > 8 m) or monthly (boats ≤ 8 m) to public households, after that the telephone numbers were considered control authorities. Commercial landing data for the coast of M–V for the quality-neutral failures. Altogether 358,411 telephone numbers where years 2014 and 2015 were provided by the State Officefor Agriculture, generated yielding a gross random sample of 73,213 valid telephone Food Safety and Fishery M–V (LALLF). For the purpose of our work, numbers, and a net random sample of 50,200 telephone interviews was these data were assumed to be reliable, but underreporting and other realized (68% screening response rate). Household size and number of issues may have occurred. recreational anglers in the household were determined. All people who had fishedat least once in the lagoons around Rügen during the last 12 2.4. Assessment of cormorant predation months preceding the interview were defined as costal lagoon anglers. All these persons, and in addition those who intended to go fishingin The number of cormorant breeding pairs was determined annually the area within the next 12 months, were asked to participate in the 1- by specialists from a M–V ornithological working group under the di­ year diary study. To increase the number of participants, coastal fishing rection of the State Officefor the Environment, Nature Conservation and permit holders of M–V, who voluntarily provided their telephone Geology of M–V. The experts count the breeding pairs through one to numbers to participate in scientificstudies, were additionally contacted. two randomly selected sampling days during the breeding season. Non- Permit holders who agreed to participate constituted an additional, non- breeding and non-resting birds are thus underrepresented. The results of representative boost sample. In the diaries, all anglers were asked to all breeding colonies in M–V are summarized in official reports pub­ report every fishing trip over a one-year observation period. For every lished by LUNG (2017). From these data, we extracted the counts for the trip, they were instructed to record exact information about the fishing two key colonies around the lagoons shown in Fig. 1. Officialdata for the area, the platform used (private boat, charter boat, shore/beach), the colonies around the island of Rügen were used as an estimate of the total target species and the number of fishcaught, harvested and/or released number of breeding pairs in the years 2011–2015. (independent of whether they were undersized or not). The diaries To estimate consumption, cormorant pellets were collected monthly offered visual and verbal stimuli to facilitate reliable records. To in the two largest cormorant breeding colonies near Rügen (Niederhof maintain the motivation to participate, retrieve diary data and to reduce and Peenemünde, Fig. 1) from April until October in 2011 and 2012. recall and panel attrition bias the participants were repeatedly contacted Fishbones extracted from cormorant pellets were used for fish species by telephone at quarterly intervals during the entire one-year observa­ identification. The original fish total length was estimated from bone- tion period. The data were collected between April 2014 and October fish length equations, and the fresh weight of individuals was calcu­ 2015 (Weltersbach et al., 2021). lated from species-specificlength-weight equations estimated separately For comparison, data were grouped into two strata: resident anglers from previously collected material. The overall cormorant diet by pellet (resident in M–V) and non-resident anglers (resident outside of M–V but was estimated by summing the numbers and estimated biomass of each within Germany). Similar to Dorow and Arlinghaus (2011) a design fish species. The resulting estimate of the amount of fish consumed by weight was applied to adjust for nonresponse bias depending on the the cormorants was based on the publicly available number of breeding household size. Non-representative data from the boost sample were pairs in all colonies around Rügen (LUNG, 2017). Predation by cormo­ also weighted with the representative data from the CATI screener using rant chicks were also taken into account and their consumption

4 R. Arlinghaus et al. Fisheries Research 238 (2021) 105894 estimated separately from adults. To estimate total consumption, we with a guide (n = 180), (3) angling guides (n = 13), and (4) other (n = used seasonal per day consumption estimates from the literature, which 23, including 6 commercial fishers,1 scientist, 2 water police officers,4 ranged in daily per-capita fishconsumption from 350 g to 430 g during representatives of conservation agencies, 1 tourism representative, 4 the period March to October for adults and from 250 g to 350 g for chicks members of the fisheries authority, 1 canoeist, 2 fish wholesalers, 1 during the period May to October (Carrs et al., 1997). The daily con­ producer of angling equipment, and 1 representative of an angling as­ sumption was extrapolated by the number of days the breeding birds sociation). The resulting data are not representative for any of the rep­ stayed in the region following public counts. The estimated amount of resented stakeholder group but they allow firstcomparative insights into fish eaten by the cormorants was distributed across the individual fish the perceptions and beliefs held by the most committed stakeholders. species following the biomass proportions estimated from the pellet We were particularly interested in learning whether a high trophic investigations. overlap of anglers with either fishers or cormorants would be mirrored in the attitudes of stakeholders towards the role of the specificgroup in 2.5. Niche overlap among the three forager types stock development and whether the anglers perceived the specific possible competitor for fishto be adequately addressed by management. We estimated the diet overlap following Berkes (1984) to compare Given the ordinal nature of the dependent variables, we employed the species distribution of commercial fisheries landings with those of Kruskall-Wallis-tests (KW) and subsequent Dunn’s post-hoc tests and resident and non-resident anglers as well as cormorants. Specifically,for compared the average perceptions of the impact factors on the stock and all pair-wise comparisons we computed the simplifiedMorisita index as attitudes to management solutions among guides, guided anglers, other presented by Horn (1966): anglers, and other stakeholders. As some of the groups had low sample sizes, we used a liberal p value of p < 0.10 in the post-hoc tests, but a p ∑S 2 xiyi < i=1 , value of p 0.05 for the global KW test. ∑S ∑S 2 2 xi + yi i=1 i=1 3. Results

= = where S the total number of species caught, xi the proportion of the 3.1. Development of recreational fisheries total catch taken by “predator” x of a fish species i, and yi = the pro­ portion of the total catch taken by “predator” y of a given fishspecies i. The total number of general fishing permits (which are issued only The overlap coefficient varies from 0 when the catches are completely for residents of M–V and allow to use either inland or coastal waters distinct (containing no species in common) to 1 when the catches are depending on whether the permit holder buys a separate water-specific identical and hence trophic overlap is maximum (Berkes, 1984). angling permit, Fig. 2) in the state of M–V rose from about 40,000 in the Following Berkes (1984) we assumed an overlap index of more than 0.6 mid-1990s to close to 100,000 today, but has recently shown a satu­ to indicate substantial and practically relevant niche overlap, suggestive rating pattern (Fig. 2). By contrast, the total number of coastal fishing of high conflict potential among two foragers. We acknowledge this is permits required additionally to actually be able to fish at the coast of arbitrary but it follows conventions in the literature. M–V (which can be bought by both resident and non-resident fishing permit holders) has substantially increased from about 40,000 in 1990 2.6. Stakeholder survey on the perceived issues and solutions to pike to over 133,000 permits in 2017 (Fig. 2). After the peak in 2017, recently management the number of coastal fishing permits has been declining to about 120,000 coastal permits. We conducted a stakeholder survey to further our understanding of Roughly, half of the coastal fishingpermit holders in M–V also fished what issues and which parties were perceived as problematic and which at least once in the brackish lagoons in 2013/2014. In this fishing sea­ solutions stakeholders preferred to address. The survey consisted of a son, an estimated 48,700 (95% CI 42,300 58,000) anglers fishedin the short questionnaire with questions related to a model coastal fish spe­ lagoon system around the island Rügen at least once (Table 1). A third of cies, the northern pike. We assumed this species to be particularly tar­ these anglers were residents in the state of M–V, i.e., the majority (70%) geted by anglers and thus be a good indicator for how anglers perceive were non-resident anglers from other federal states. Resident and non- the impacts of commercial fishers, cormorants and other issues. Spe­ residents anglers combined exerted a fishing effort of about 390,000 cifically, we asked “Please think about the current situation: How – estimated angling days per year in the lagoon ecosystem in 2013/2014, positively or negatively – do the factors listed below affect the current of which the resident anglers contributed roughly half of the effort (47.4 stock size of pike in the Bodden area around Rügen?”. The five-point %, Table 1). Likert-type response scale ranged from” 1 = very negative”, “2 = Although the coastal fisheriesof M–V, including the lagoon systems, negative”, “3 = neither/nor”, “4= positive” to “5 = very positive”. Items have always attracted anglers, in particular the non-resident fishing rated by each respondent included the key conflictparties and issues (e. interest has increased in recent years (Fig. 2). This can be inferred from g. commercial fishing, recreational angling, cormorant predation). the relative stability of general fishing permits of M–V (which exclu­ Second, we asked respondents: “In your opinion, how suitable or sively reflect resident fishing interest), while the number of coastal unsuitable are the measures listed below to foster pike stocks around fishingpermits (which can be bought by both resident and non-resident Rügen?” Items included harvest and other regulations as well as in­ anglers) has strongly risen after the German reunification in 1990 terventions into the natural environment (e.g., habitat improvement). (Fig. 2). Further evidence can be derived from the officialstatistics of the Responses to this question were measured using a fivepoint Likert-type tourism fishingpermits that the state of M–V offers to persons lacking a rating scale from 1= “very unsuitable”, 2= unsuitable, 3 = neither/nor, general fishing permit (which in Germany requires passing an angling 4= suitable, to 5 = “very suitable” to conserve pike stocks around examination) and allowing them to fish for recreation in M–V for a Rügen. restricted period of 28 days (Fig. 3). The permit numbers rose from less The questionnaire was administered through a snowball technique to than 5,000 to more than 20,000 per year in about 10 years, and the bulk both resident and non-resident anglers, guided anglers, guides, com­ of this growth was due to non-resident tourists (Fig. 3). Today, the island mercial fishersand other stakeholders (e.g., scientists and agency staff). of Rügen is a well developed angling destination with harbors, rental Data were further collected during angling exhibitions in Stralsund and boats, and a professional guiding sector of about 50 part-time and full- Rostock, M–V, in 2019, and also through angling guides, achieving a time businesses offering recreational angling tours. total sample size of 258 completed questionnaires. We grouped the re­ spondents into four categories: (1) private anglers (n = 42), (2) anglers

5 R. Arlinghaus et al. Fisheries Research 238 (2021) 105894

Fig. 2. General fishing permits (only for resident anglers) and coastal fishing permits issued in the German federal state of Mecklenburg-Western Pomerania (data source: LALFF).

and in the lagoons (Fig. 5). For example, the total landings from the Table 1 lagoon fisheryhave decreased from 9,800 t in 2008 to 4,600 t in 2018. Origin, numbers and fishing effort of recreational anglers fishing in the lagoon Excluding herring, however, we see that commercial landings within the system around the island of Rügen, Germany, in the fishingseason 2013-2014 as lagoons have remained relatively stable over the period 1994–2019, i.e., estimated from a combined telephone-diary survey in 2014/2015 (Weltersbach et al., 2021). CI = confidence interval. n = estimated number of anglers or despite reduced fisher numbers, the harvesting pressure has largely fishing days. remained stable on inshore freshwater fish species (Fig. 5). Origin of anglers using the lagoon fisheriesaround Rügen 3.3. Development of the cormorant population size Federal state n of Lower Upper % of n of days % of anglers 95 % CI 95 % CI anglers days The breeding pairs of cormorants on the coastline of Pomerania Mecklenburg- 14,600 11,004 19,349 30.0 185,268 47.4 around the island of Rügen have shown exponential growth since the Western Pomerania 1950s and have somewhat stabilized with high interannual fluctuations Berlin 5949 3792 9330 12.2 31,627 8.1 in numbers since the 2000s (Fig. 6). The majority of the population is Saxony 5551 3501 8903 11.4 37,964 9.7 associated with the lagoons around Rügen (Fig. 6). For example, in 2015 Brandenburg 5118 3158 8290 10.5 24,083 6.2 there were 13,031 breeding pairs on the coast of Pomerania, of which Saxony-Anhalt 3191 1736 5859 6.6 13,408 3.4 Lower Saxony 2953 1560 5587 6.1 27,184 7.0 8,061 (67%) were foraging in the waters around Rügen. Schleswig- 2782 1435 5388 5.7 12,298 3.1 Holstein 3.4. Fish removals by forager type Thuringia 2580 1308 5088 5.3 19,066 4.9 Hamburg 1520 638 3618 3.1 9243 2.4 North Rhine- 1428 584 3528 2.9 9762 2.5 In the season 2013/2014, recreational anglers removed a total of 966 Westphalia t of fish from the lagoon systems (Table 2), roughly equally distributed Bavaria 1269 494 3300 2.6 8678 2.2 among resident and non-resident anglers. Herring and freshwater upper Baden- 793 246 2591 1.6 5423 1.4 trophic level species like pike, perch and to a smaller degree pikeperch Württemberg Hesse 634 174 2343 1.3 4339 1.1 constituted the bulk of the angler harvest, but the relative ranking Rhineland- 317 56 1823 0.7 2169 0.6 differed among resident and non-resident anglers. While pike (41% of Palatinate the harvest), herring (36%) and perch (9%) were the three dominant Bremen 0 0 1287 0.0 0 0.0 species in the harvest of non-resident anglers, herring (35%), perch Saarland 0 0 1239 0.0 0 0.0 Total 48,686 42,257 57,961 100 390,512 100 (25%) and pike (21%) were most relevant for resident anglers (Table 2). Across most species, recreational anglers only took a fraction of the fishas harvest and released the majority of fisheither due to regulations 3.2. Development of commercial fisheries (e.g., in the case of undersized fish) or because of ethical reasons. Be­ tween 2006/2007 and 2014/2015, the release rates substantially The commercial fisheryalong the coast of M–V is characterized by a increased for all coastal fishexcept roach (Fig. 7). The increase in release few large trawlers mainly fishing for herring or cod, and a larger rates was particularly high for pike (regulated via a minimum-size limit) segment of small-scale fishersusing passive gears along the coast and in and bream (no size limit in place), of which more than 60% of individual the lagoons. The number of commercial fisherson the coast of M–V has fish were released in the fishing season 2014/2015. The high release decreased since the German reunificationfrom 950 in 1990 to about 230 rates related mainly to freshwater fishes, while marine fishes, such as full-time fishers in 2019 (Fig. 4). The decrease in coastal commercial herring and garfish (Belone belone) or diadromous eel, were only rarely fishernumbers is mainly related to declines in those operating full time, returned to the water (Fig. 7). The release behavior of non-resident and while the number of part-time fishers has initially been rising and has resident anglers was largely similar (Fig. 8). then stabilized at around 100 part-time fishers( Fig. 4). The decrease in The average annual commercial fisheries landings in the lagoon full-time fishers was also visible for those fishers that also exploit the ecosystems were around 5,300 t in the years 2014 and 2015 (Fig. 5, lagoons (Fig. 4). In 2015, 177 full-time and 65 part-time fishers were Table 2). The bulk of these landings was composed of herring (68%), using the lagoon fisheries of Rügen at least once. Along with the followed by two freshwater cyprinids (bream and roach) with each decrease in fishernumbers, the annual commercial landings of fishof all about 10% of the total harvest. species have decreased after 2008 both in the coastal fisheriesin general The removals by cormorants ranged in-between the removals by

6 R. Arlinghaus et al. Fisheries Research 238 (2021) 105894

Fig. 3. The tourist angling permits (28 days valid) sales in Mecklenburg-Western Pomerania (M–V) between 2005 and 2018 (data source: LALFF).

Fig. 4. Number of commercial fishers operating in the coastal fishery of Mecklenburg-Western Pomerania (M–V) and specifically in the lagoon fisheries since the German reunification (1990) separated by main and part-time occupation. The number of fishers targeting fish in the lagoon fisheries indicates enterprises that reported at least one fish landed from the corresponding lagoon areas (data source: LALFF).

Fig. 5. Total commercial landings (in 1,000 t) of coastal fisheries in M–V per year by species and for lagoon and non-lagoon areas (data source: LALFF).

7 R. Arlinghaus et al. Fisheries Research 238 (2021) 105894

impact: natural predators, commercial fisheries, non-resident anglers and finally resident anglers (Table 5). The ranking was again entirely different for the “other stakeholder” group, who considered non- resident anglers to have the greatest negative impact, followed by nat­ ural predators, resident anglers and finally commercial fisheries (Table 5). Post-hoc tests revealed that guided anglers perceived com­ mercial fisheries significantly more negative for the pike stock than private anglers and other stakeholders. The impact on pike of both resident and non-resident anglers was perceived as significantly more negative by the “other stakeholder” group compared to the angler groups (Table 5). Finally, the impact of natural predators such as cor­ morants on the pike stock was seen as least impactful by the “other stakeholder” group, but post-hoc tests failed to identify group-specific differences. Fig. 6. Number of cormorants (breeding pairs) on the coast of Pomerania over The stakeholder groups also significantlydiffered in their attitudes to time, in total and specifically for the lagoons of the island of Rügen (Fig. 1). In various management actions that targeted commercial fisheries, recre­ the initial time series the lagoons were equivalent to total coastal numbers. ational fishing or control of natural predators (such as cormorants) aiming to promote the pike stock. A few characteristic patterns emerged commercial and recreational fisheries. The estimated total fish con­ (Table 6). Anglers, both guided and private, considered actions to reduce sumption was on average 2,394 t per year in the years 2011 and 2012 commercial fishing mortality most suitable, followed by control of (Table 2). During this time, cormorants mainly consumed perch (26%) natural predators and control of recreational fishing. Guides also and roach (24%). Also, other smaller-bodied non-game species, such as considered controls on commercial fishing to be the most well-suited stickleback (19%), gobids (7%) or ruffe (5.0%) were very important in approach, followed by implementation of annual recreational fishing the diet, collectively representing 32% of the biomass removed by cor­ quotas for pike and predator control. By contrast, the “other stake­ morants. Among the freshwater predators pikeperch (10%) represented holder” group perceived controls of recreational fishingand commercial a substantial fraction of total biomass removals by cormorants, whereas fishing as equally suited, followed by predator control. The attitudes pike was very rare in the diet of cormorants (0.3%). towards increased control of commercial fishing were significantly less At the species level (Table 3), commercial fisherieswere responsible pronounced in the “other stakeholder” group compared to anglers and for the majority (close to or above 50%) of the total removals for most of guides (Table 6). Also, both angler groups and guides perceived controls the marine and diadromous species, specificallyherring, eel, sea trout/ of recreational fishing mortality as significantly less suitable than the salmon (due to possible misidentificationlumped together), garfishand “other stakeholder” group (Table 6). Similarly, the reduction of natural cod. Among the freshwater fish, commercial fishing was the primary predators was considered of higher suitability by the anglers and guides source of mortality with close to or more than 50% of total removals for than by the “other stakeholder” group. bream and pikeperch. Cormorants dominated the total removals of perch and roach, but also quite substantially contributed to the harvest 4. Discussion of flatfish species and were primarily responsible for the removal of “other species” (non-game species like stickleback, gobies etc.). Both study hypotheses were supported. We detected a high potential Together with rudd (whose harvest volume was overall small and highly for conflictsamong commercial and recreational fisheriesdue to strong likely biased due to species misidentificationwith roach), northern pike trophic overlap. By contrast, limited trophic overlap was found for both (77% of total removals) was the only species where the harvest by rec­ fisheryforms and cormorants when the entire species mix harvested by reational fisheries constituted the largest mortality source of the three the foragers was analyzed. Yet, for individual species, individual for­ forager groups we examined, particularly among non-resident anglers agers, either fishers, anglers, or cormorants, were dominating the total (Table 3). biomass removals, suggesting at the species-specificlevel the dominant forager may be perceived as forcing strong competition with others. Using the special case of coastal pike management, we also revealed that 3.5. Trophic overlap recreational fishers and recreational angling guides preferred the regu­ lation of commercial fishingand cormorants over the implementation of The analysis of the trophic overlap index (Table 4) revealed that the regulations for recreational fisheries. As the trophic overlap among species composition of commercial fisheries landings overlapped anglers and natural predators was not high and anglers were responsible strongly with recreational fisheriesharvest of both subgroups of anglers. for the highest harvest of pike, the results suggest anglers were gener­ Also, there was a strong overlap among resident and non-resident an­ alizing perceptions that the other two forager types (commercial fish­ glers in their foraging niche. By contrast, overlap of the foraging niche of eries and cormorants) were more impactful than the anglers themselves commercial and recreational fisheries with cormorants was generally across species and preferred to regulate the real or perceived competi­ low and below the cut-off value of 0.6. tors for resources, similar to findings of other work in north-eastern Germany (Dorow and Arlinghaus, 2012) and beyond (Arlinghaus, 3.6. Stakeholder perceptions of issues and solutions using the example of 2006). northern pike We found recreational angling interest has increased in coastal sites of M–V after the German reunification. On first sight, the pattern The survey revealed that recreational fishingguides, guided anglers, appeared broadly consistent with the life-cycle of inland fisheries( Cowx private anglers and other stakeholders (six of which were commercial et al., 2010; Smith, 1986). The conceptual model predicts that with fishers) significantly varied in what they perceived to be the main increasing economic development and wealth commercial fisher sources of negative influenceon the pike population around the island of numbers decrease and recreational anglers increase. Today, Rügen (Table 5). Guides and guided anglers perceived commercial non-resident anglers are more numerous than resident anglers on the fisheriesas being the most negative factor for the pike stock, followed by coast of M–V, indicating the potential of angling tourism for regional natural predators (such as cormorants) and resident anglers. Private development in the area (Dorow and Arlinghaus, 2009). Both groups of anglers had a different ranking about the sources of greatest negative anglers, resident and non-residents, were found to preferentially remove

8 R. Arlinghaus et al. Fisheries Research 238 (2021) 105894 other of t of 32.0 2.8 3.4 26.9 0.3 subgroup % total 10.0 0.0 0.1 24.4 and presented (in 2012 (CI). average by are and 2 4,6%) total 70.7 2,393.8 81.2 643.4 8.2 Removals biomass) 2011 Cormorants Total 238.2 584.8 1.8 765.5 intervals of % removals cernua , The of angler 0.6 0.9 67.8 0.4 3.9 1.5 % total 5.2 10.7 0.0 8.5 0.0 0.4 0.0 0.0 0.0 0.0 0.0 0.1 confidence 2015 biomass) with of and t ( Gymnocephalus importance. 5,354.3 29.8 46.4 3,630.3 18.8 211.1 79.5 Commercial fisheries Total 277.7 573.2 0.0 454.8 2.4 21.5 0.6 0.0 2.1 1.8 0.2 4.1 recreational (in 2014 ruffe The expressed of diet. 0.5 3.5 35.6 0.0 17.4 31.4 Landings average % total 8.5 1.3 0.2 1.2 0.1 0.1 0.0 0.1 0.0 0.1 0.0 0.0 thus 7.4%), recreational Rügen. or are of % and 95 cormorant ( Gobiidae , island 7.7 49.2 466.7 0.0 225.7 395.2 up CI 111.5 29.6 6.7 19.6 2.5 2.7 0.0 2.5 0.0 2.5 0.0 total the survey the commercial gobies % no angling within 95 of around 19%), 2.0 23.3 259.6 0.0 127.8 239.0 low CI 61.4 3.5 0.0 5.7 0.0 0.5 0.0 0.2 0.0 0.0 0.0 pooled species telephone-diary fisheries 966.2 4.4 34.1 344.0 0.0 168.0 303.3 81.8 Recreational Total 12.9 2.0 11.4 0.9 1.1 0.0 0.9 0.0 1.4 0.0 aculeatus , fish reported lagoon of combined thus all 0.8 2.1 35.8 0.0 9.4 41.3 % total 8.0 1.3 0.4 0.5 0.2 0.1 0.0 0.0 0.0 0.0 0.0 0.0 the are ( Gasterosteus across smaller-bodied on and % of 95 based 7.0 20.4 268.3 0.0 71.2 267.2 up CI 62.4 12.1 6.4 5.2 2.5 1.7 0.0 0.6 0.0 0.0 0.0 pellets range stickleback are a as cormorants % anglers and 95 such cormorant includes 1.5 3.6 109.5 0.0 28.5 154.9 low CI 23.9 2.2 0.0 0.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 calculations fishers the species in 489.2 3.8 10.4 175.0 0.0 46.0 202.0 Non-resident Total 39.3 6.4 1.9 2.7 0.9 0.6 0.0 0.2 0.0 0.0 0.0 angler cormorants The irrelevant of commercial separable of 0.1 5.0 35.4 0.0 25.6 21.2 % total 8.9 1.4 0.0 1.8 0.0 0.2 0.0 0.2 0.0 0.0 0.2 0.0 easily category anglers, % population. not recreationally 95 “ other ” 1.2 36.9 251.4 0.0 182.3 147.1 up CI 61.5 22.6 0..4 17.1 0.0 1.7 0.0 2.3 0.0 0.0 2.6 were and forager the recreational % by turbot) that given 95 and anglers any 0.2 15.3 112.3 0.0 81.0 71.3 low CI 29.3 0.0 0.0 3.4 0.0 0.3 0.0 0.1 0.0 0.0 0.0 tons) Note commercially plaice within metric 477.0 0.6 23.7 168.9 0.0 122.0 101.3 Resident Total 42.5 6.5 0.1 8.7 0.0 0.8 0.0 0.8 0.0 0.0 0.9 anglers. (in species flounder, small-bodied, per of of 1 removals maximus (dab, non-resident platessa 2014/2015 flesus maraena biomass range in limanda anguilla brama species a of and lucioperca rutilus harengus belone morhua salar trutta frequency t fluviatilis name lucius misidentified species biomass fractions. (in biomass) Platichthys Belone Clupea Coregonus Perca Esox Sander Latin Abramis Pleuronectes Rutilus Gadus Scardinius erythrophthalmus Salmo Salmo Scophthalmus Incl. Various Anguilla Limanda total includes resident of relative flatfish t of small for the subgroup (in spp. group four with by 2 trout The This year) species Total Flounder Garfish Herring Maraena Perch Pike Harvest Species Pikeperch Bream Plaice Roach Cod Flatfish Eel Dab Rudd Salmon Sea Turbot Other 1 2 species represent Table Compilation separately

9 R. Arlinghaus et al. Fisheries Research 238 (2021) 105894

Fig. 7. Comparison of the release rates reported in diaries of anglers in the lagoon fisheriesaround the island of Rügen in two different study periods (data for 2006/ 2007 taken from Dorow and Arlinghaus, 2011).

Fig. 8. Comparison of release rates by resident and non-resident anglers in the lagoon fisheries around the island of Rügen in 2014/15.

Table 3 Species-specific total removals and relative frequencies by recreational anglers (resident and non-resident shown separately and pooled), commercial fishers and cormorants in the lagoon ecosystems of M–V, Germany in the period 2011 to 2015 on average per year. For Latin species names, see Table 2.

Overall removals (in t of biomass) Resident anglers Non-resident anglers Recreational anglers total Commercial fisheries Cormorants Species Total % of species total % of species total % of species total % of species total % of species total

Bream 586.17 1.1 1.1 2.2 97.8 0.0 Cod 5.07 0.0 17.3 17.3 47.2 35.5 Eel 22.93 3.6 2.7 6.2 93.8 0.0 Flatfish spp. 107.73 0.7 5.3 5.9 28.4 65.6 Garfish 80.47 29.4 13.0 42.4 57.6 0.0 Herring 4,055.44 4.2 4.3 8.5 89.5 2.0 Maraena 18.81 0.0 0.0 0.0 100.0 0.0 Perch 1,022.52 11.9 4.5 16.4 20.6 62.9 Pike 390.96 25.9 51.7 77.6 20.3 2.1 Pikeperch 597.73 7.1 6.6 13.7 46.5 39.9 Roach 1,050.95 0.8 0.3 1.1 43.3 55.6 Rudd 1.06 78.2 21.8 100.0 0.0 0.0 Salmon 2.07 0.0 0.0 0.0 100.0 0.0 Sea trout 2.69 34.3 0.0 34.3 65.7 0.0 Other species 769.64 0.0 0.0 0.0 0.5 99.5

10 R. Arlinghaus et al. Fisheries Research 238 (2021) 105894

Table 4 or that the fraction of undersized fish has increased, increasing regula­ Modified Morisita index of trophic overlap among three forager populations. tory catch-and-release practice. All three species just mentioned (i.e., Index values of > 0.6 are considered relevant and indicative of competition (in pike, pikeperch and perch) are regulated by minimum-length limits and boldface). daily bag limits, such that increased regulatory catch-and-release might Foragers Group A Foragers Group B Overlap Coefficient occur if the fishsize declines due to high fishingmortality (Ferter et al., Commercial fishers Cormorants 0.18 2013). The latter is more likely for pikeperch than for pike because Anglers (pooled) Commercial fishers 0.70 pikeperch is heavily targeted by commercial fisheries and thus size Anglers (pooled) Cormorants 0.29 truncation may occur in the region (Ahrens et al., 2020). By contrast, Resident anglers Commercial fishers 0.72 commercial landings of pike are low and anglers rarely catch fishbelow Resident anglers Cormorants 0.40 the minimum-size limit of 50 cm in the lagoons. Thus, the roughly two Non-resident anglers Commercial fishers 0.65 Non-resident anglers Cormorants 0.18 third release rate for pike are most likely due to voluntary releases of Resident anglers Non-resident anglers 0.88 harvestable fish.Despite the high release rates, the total removals of pike by anglers were much larger than those by commercial fisheries. We thus conclude that anglers are today the major mortality factor for adult herring and otherwise focus on adults of upper trophic level fishin their pike around Rügen. However, this is the only species for which anglers harvest, as known for recreational anglers in general (Donaldson et al., represent the bulk of total removals, for all others either commercial 2011). However, residents and non-residents differed in the relative fishers or cormorants dominate. ranking of freshwater species harvested: while piscivorous pike featured Supporting the life-cycle of fisheries,commercial fisheriesnumbers, relatively high among non-residents, this species was of less importance at least those operating in full-time, have decreased over time in the to residents who instead focused on herring and omnivorous perch. lagoon fisheriesof Rügen, suggesting decreasing profitabilityand other These differences may reflect variation in culinary tastes but can also structural changes affecting the desirability of commercial fishing op­ suggest differences in fishtargeting behavior. Pike in general belongs to erations (Doring¨ et al., 2020). However, the number of part-time fishers the main target species of German anglers and is also intensively tar­ has remained quite constant in the coastal and lagoon fisheriesof M–V. geted in inland waters (Ensinger et al., 2016). Among non-residents the Total commercial yields of quota-regulated fish (herring and cod) have lagoons are particularly well known as a habitat for catching large pike decreased, while the yield of non-quota regulated freshwater fish has (Koemle et al., unpublished), which might explain why pike was so remained relatively stable in the lagoon fisheries. The latter indicates prominent in the composition of the harvest among non-residents. Pike, that the reduced numbers of commercial fishers have become more perch and pikeperch are also prime recreational fishingtargets in other ¨ effective over time, and the relative stability of the lagoon fishyields for areas of the Baltic Sea, e.g., in Sweden (Ostman et al., 2013). species other than herring does not support the idea of a dwindling Harvest composition does not equal general species preferences, commercial fishing pressure in the region. Yet, the bulk of the com­ because anglers are known to engage in voluntary catch-and-release mercial landings is composed of the low trophic level herring, which has behavior or are obliged to release fish for regulatory reasons. We been the “bread and butter” fishfor the livelihood of commercial fishing found that the release rates of most freshwater fish targeted in the in the region for centuries (Raillard, 2012). Commercial landings today lagoon areas have increased over time, particularly pike, pikeperch and still dominate the total removals for most marine and diadromous spe­ perch. The increase in release rates either indicates that key freshwater cies as well as selected cyprinids and salmonids, which reflects the fishare today often targeted for non-consumptive reasons in the lagoons

Table 5 Summary statistics for perceived factors influencingthe pike stock in lagoons around the Island of Rügen answered by two groups of anglers, angling guides and other stakeholders using a five-point Likert-type scale (1 = very negative, 5 = very positive). Significances of Dunn’s post-hoc tests are indicated at the 10% level, where different letters indicate significantdifferences. K–W indicates the global Kruskall-Wallis-test for significantdifferences. * indicate level of significance(* p < 0.05, *** p < 0.001).

Others (n = 23) Guided anglers (n = 180) Guides (n = 13) Private anglers (n = 42) K–W

Factors mean sd mean sd mean sd mean sd P-value

Commercial fisheries 2.45ac 0.69 1.65b 0.75 1.75bc 0.62 2.00c 0.79 *** Resident angling 2.32a 0.75 2.75b 0.70 2.42ab 0.67 3.03c 0.83 *** Non-resident angling 1.85a 0.67 2.94b 0.83 3.00b 0.95 2.62b 0.99 *** Abundance of natural predators 2.20 1.11 1.66 0.74 1.83 0.83 1.87 0.88 *

Table 6 Summary statistics for attitudes to management measures promoting the pike stock in the lagoons around the Island of Rügen by two groups of anglers, angling guides and other stakeholders, assessed on a five-pointLikert-type scale (1 = very unsuitable, 5 = very suitable). Significancesof Dunn’s post-hoc tests are indicated at the 10% level, where different letters indicate significant differences. K–W indicates the global Kruskall-Wallis-test for significant differences. * indicate levels of sig­ nificance (** p < 0.01, *** p < 0.001).

Other (n = 23) Guided anglers (n = 180) Guides (n = 13) Private anglers (n = 42) K–W

mean sd mean sd mean sd mean sd P-value

Commercial fishing regulations Introduce annual quotas for fishers 3.81a 1.03 4.51ab 0.86 4.67b 0.65 4.16b 1.26 *** Limit gillnets 3.59a 1.18 4.57b 0.75 4.55b 0.69 4.29b 0.90 *** Angling regulations Introduce annual quotas for anglers 3.80 1.24 3.74 1.10 4.09 0.70 3.45 1.33 Limit number of anglers 3.53a 1.23 2.85a 1.11 2.50ab 1.18 2.22b 0.96 *** Introduce harvest tags 3.82a 1.19 3.85a 1.12 3.36ab 1.12 2.63b 1.33 *** Steer angling tourists 4.09a 0.61 3.32b 1.02 3.82ab 0.87 3.24b 1.34 *** Regulation of the natural environment Reduce predators 3.45a 1.32 4.25b 1.01 3.92ab 1.16 3.82ab 1.31 **

11 R. Arlinghaus et al. Fisheries Research 238 (2021) 105894 market value driven by the culinary preferences of the consumers. also cormorants as having a strong negative impact on the pike stock. However, in recent times cyprinids are also sold as stocking material to This finding is noteworthy as our removal statistics showed that spe­ angling clubs (Arlinghaus et al., 2015), which is creating a new market cificallyfor pike anglers were by far the largest source of mortality. This, for commercial fishing of cyprinids. however, seems not to be known or appreciated by anglers, who instead Our cormorant consumption study suggested that this species is an believed that commercial fishing and cormorants had greater negative opportunistic feeder that mainly targets smaller-bodied species in pro­ impacts and should be preferentially regulated. These findings mirror portion to their abundance. This findingis mirrored by many diet studies research in other recreational fisheries, which repeatedly found that of cormorants (e.g., Keller, 1995; Ovegård et al., 2017; Suter, 1997; anglers tended to perceive other resource users to be more impactful and Trayler et al., 1989; VanDeValk et al., 2002) and contrasts with the in need of greater regulation than the anglers themselves (Arlinghaus, mainly adult fishes targeted by both commercial and recreational fish­ 2006; Dorow and Arlinghaus, 2011). ¨ eries in the study area and in other areas of the Baltic Sea (Ostman et al., We found limited trophic overlap among commercial/recreational 2013). Cormorants around Rügen strongly preyed upon smaller-bodied fisheries and cormorants and a very low relative mortality induced by non-game species, but the predation rates on some flatfish species, cormorants on pike. Yet, anglers and guides considered control of cor­ roach, perch and pikeperch were also high relative to the removals by morants to be important for the improvement of the pike stocks. This anglers and commercial fisheries. Overall, however, low trophic prey finding generally agrees with earlier work that has shown that anglers species, such as roach, stickleback or gobies, dominated the cormorant have quite negative perceptions of cormorants (Dorow et al., 2009), ¨ diet, similar to the case in other Baltic areas (Ostman et al., 2013). The even if the species often consumes organisms of little fisheries value. dominance of roach and perch in the cormorant diet on the coast of Two reasons may explain this phenomenon. First, anglers might not be Pomerania has also been shown in earlier studies (Preuß, 2002) and is aware that cormorants preferentially forage on species and size classes well known from freshwater systems (Keller, 1995; Ovegård et al., 2017; that are not the targets of anglers. Second, as elaborated above, Suter, 1997; VanDeValk et al., 2002). However, as the fishcommunities cormorant predation on smaller size classes may still affect target stock shift, also the cormorant diets shift. Accordingly, in recent times round size and yield potential of selected species that are predominantly or ¨ goby (Neogobius melanostomus) has become the third most important heavily consumed by cormorants, such as perch (Ostman et al., 2013, food species of the cormorant in the lagoons (Oesterwind et al., 2017). 2016; Veneranta et al., 2020) or pikeperch (Mustamaki¨ et al., 2014), Our work extended the earlier large-scale compilation of cormorant diet where the removal of juveniles might affect recruitment into the fishery data by Hansson et al. (2018), by producing local estimates for the (Winkler et al., 2014). Thus, although appearing as a conflict fueled Rügen area. These estimates are important because most coastal fish primarily by erroneous beliefs about the trophic niche of cormorants at have localized dynamics and thus local mortality caused by cormorants, firstsight, it might well be that for selected species cormorants do have a e.g., on perch, can have far-reaching impacts on the productivity of a negative impact on fishing quality and that the resulting aversion by fishery,as shown recently in the Baltic Sea of Sweden (Veneranta et al., fishers and anglers may indeed have an empirical underpinning. These ¨ 2020) and implied previously for selected Swedish sites (Ostman et al., species are likely to encompass the freshwater coastal species pikeperch ¨ 2013). Yet, because cormorants hunt smaller-bodied, often juvenile or­ (Winkler et al., 2014) and perch (Ostman et al., 2013; Skov et al., 2014; ganisms, the trophic overlap with commercial and recreational fisheries Veneranta et al., 2020), and less likely pike, although more studies are was found to be rather low. needed to corroborate this speculation for the Rügen area. This does not mean, however, that cormorants have no impact on the A few limitations of our study should be taken into account when yield of selected species. Because small fishmay eventually also recruit interpreting the results. First, the commercial and recreational landings into the fishery,the indirect predation impacts of cormorants on species data depend on self-reports. Although the commercial landings data are such as perch or pike in the Baltic Sea through consumption of juvenile also monitored by the authorities, illegal, unregulated and unreported fish can quickly escalate compared to when only the direct removal of landings are still possible (Sampson, 2011), which would likely lead to ¨ ¨ biomass is considered (Ostman et al., 2013). Ostman et al. (2013), for conservatively biased commercial landing estimates. Data entries in the example, calculated that in one of the two sites they studied the direct diaries used to estimate recreational fisheries catches may also be un­ predation impact by cormorants on perch rose from about 10% to 34% reliable or biased (e.g., as a result of over- or underreporting for several lost yield when indirect impacts through the selective predation on reasons, Hartill and Edwards, 2015). Regarding cormorants, the total small, juvenile fish were considered. Because the cormorants around consumption is likely an underestimate as the sampling did not cover Rügen consumed large quantities of (juvenile) roach and perch, non breeding birds. Second, the mortality induced by the three forager ¨ following the calculations of Ostman et al. (2013), we assume that re­ types might be larger than assessed in our work due to post-release and ductions of fisheries yields are very likely relevant for these species, in by-catch mortality (Kerns et al., 2012). Third, some species may have contrast to, for example, pike, which very rarely occurred in cormorant been misidentified (Chizinski et al., 2014) and mistakenly classified diets around Rügen. (which is likely to happen in the case of, e.g., roach/rudd, flatfish, Our work revealed large potential for conflict among commercial sprat/herring or sea trout/salmon). In consequence, some of the data and recreational fisheries and among resident and non-resident anglers might be biased. However, this effect is unlikely to be relevant for originating from high trophic overlap as a consequence of similar species roach/rudd as rudd is rarely present in the lagoons around Rügen. preferences. The potential for conflictis fueled by the dominance of total Fourth, we derive the impact of the three forager populations by removals of specific fish by one sector, showing the following charac­ removal estimates in terms of biomass. Clearly, cormorants in particular teristic pattern: commercial fisheriesdominated the removals of marine differ from human fisheries in that they tend to target smaller-bodied and diadromous fish, commercially and recreationally valuable pike­ organisms, which are still under strong compensatory mortality regu­ perch and generally less valuable lower trophic cyprinids, while angler lation (Allen et al., 1998). Therefore, catch in numbers may be consid­ harvest dominated for pike. The perception by anglers that commercial erably higher than catch in weight. The true relative impact of the three fisheries may affect their well-being negatively through excessive har­ forager types can only be revealed by constructing fish population vest was therefore corroborated. Further aggravating the conflict,some models with appropriate density-dependence to see whether elevated of the interviewed stakeholders believed that the recent quota re­ mortalities in certain life stages have population-level consequences. As ductions in the commercially very attractive herring and cod fisheries a further limitation, our stakeholder survey was not representative of the (ICES, 2020a, 2020b) may have enhanced commercial fishing pressure total population of lagoon anglers and likely reflected the views of the on freshwater coastal fishin M–V that are also desired by anglers (Vogt, most involved and avid anglers, which might have biased the results 2020). The stakeholder survey data of our model species coastal pike (Connelly et al., 2000). It has previously been reported that convenience confirmed that anglers negatively perceived commercial fisheries and samples may reveal different preferences and attitudes to management

12 R. Arlinghaus et al. Fisheries Research 238 (2021) 105894 than random samples of anglers (Hunt et al., 2010). Specifically, we and cormorants, which is supported by related social science research in assume that more avid, specialized anglers were overrepresented in the the region (Vogt, 2020). This conflict cannot be explained by a strong sample, who have been shown before to differ in their perceptions of trophic overlap among commercial and recreational fisheries and cor­ fishery developments and preferences (e.g., Dorow and Arlinghaus, morants. Instead it may be explained by species-specific effects of cor­ 2012). Nonetheless, aversion of anglers in relation to both commercial morants on valuable species such as perch, pikeperch and perhaps pike fisheriesand cormorants seems to be commonplace among the majority and that anglers had a negative perception of cormorants in general, but of German anglers as has been repeatedly shown using different forms of further population dynamic research and discourse analysis is war­ surveys (Arlinghaus and Mehner, 2003; Arlinghaus, 2006; Brehens et al., ranted. Conflictsreduce the satisfaction of the involved parties and thus 2008; Dorow et al., 2009). Thus, despite the non-representative nature should be proactively managed (Arlinghaus, 2005). Managers can use of our stakeholder survey, we assume that our main conclusions are tools such as spatial or temporal zoning, allocation of fish stocks, robust. Yet, stakeholder surveys using probabilistic sampling designs are cormorant culling and control, and improved communication and in­ needed in the future to test whether our results extend to the population formation to manage these conflicts (Redpath et al., 2013). Some of level. Finally, a limitation of our niche overlap index is that it only these actions will be difficultto implement for social or legal reasons (e. represented the composition of the fish consumption, not the absolute g., cormorant control or zoning) and may also be constrained by deeply amount of resource removal by each forager. Therefore, it is possible engrained stakeholder aversions. Therefore, building trust between the that two foragers show little dietary overlap overall, but given unequal different stakeholder groups through improved communication will be a species-specificamounts of resource extraction one forager might still be key to address the conflict in the region. competitively affected by another despite low trophic overlap. For this reason, we presented both the total resource extractions and the Author contribution breakdown by species and forager type of total removals and the dietary overlap. Noting that the total and relative resource removals by cor­ RA, CK, HVS conceptualization of study, RA, MSW, DK, HMW, HVS morants were substantial for selected species (i.e., species that are also data generation, RA, JL, MSW, DK, HMW, CR data analysis, RA, JL, of large commercial and recreational importance like perch and pike­ MSW, DK, HMW, CR, CK, HVS interpretation of results, RA writing of perch), our findings of overall limited overall niche overlap should not paper, with all others involved in revisions. be misinterpreted as total absence of negative impact of cormorant predation on either commercial or recreational fisheries. Quite the Declaration of Competing Interest contrary, the perch and pikeperch yield lost to cormorant predation may be similarly relevant on the German Baltic coast as previously reported The authors report no conflict of interest. ¨ for certain sites in Sweden (Ostman et al., 2012, 2013). Therefore, further work in this area is warranted. Equally important is more work Acknowledgements on the relative population dynamical consequences of commercial and recreational fisherieson key species targeted by both user groups (e.g., We thank the European Union through the European Maritime and perch, pikeperch or pike). Fisheries Fund, and the State of Mecklenburg-Vorpommern (Germany) (Grant MV-I.18-LM-004, B 730117000069) for funding of the Bod­ 5. Conclusions and implications denhecht project. The 2014/2015 telephone-diary survey was con­ ducted in the framework of the German marine angling program of the Our work has shown that there is a solid basis for conflictin terms of Thünen Institute of Baltic Sea Fisheries and co-funded by the European competition for the same resources among commercial and recreational Commission’s Data Collection Framework (DCF) and the State Officefor fisheriesin the Rügen area and among resident and non-resident anglers. Agriculture, Food Safety and Fishery M–V (LALLF). We thank all survey Therefore, reported conflictsamong these stakeholders (Vogt, 2020) are respondents and guides for help with the survey and Thomas Richter and backed up by objective data. The potential for conflictis aggravated by Thomas Schaarschmidt from LALFF for excellent help with data com­ the fact that there is often one fisheryform that strongly dominates the pilations. We thank Henry Hansen for help with Fig. 1 and the reviewers total removals for a given species or species group. With the exception of for constructive feedback. pike, where anglers take by far the largest fraction, many other valuable fish such as pikeperch and marine and diadromous species (eel, salmo­ References nids) are primarily taken by commercial fisheries in the study area. Thus, despite the decline in commercial enterprises in the Rügen lagoon Aas, Ø., Skurdal, J., 1996. Fishing by residents and non-residents in a rural district in fisheriesover time and a rise of recreational angler numbers, specifically Norway: subsistence and sport-conflict or coexistence? Nord. J. Freshw. Res. 72, 45–51. non-resident anglers, commercial fisheries continue to dominate the Ahrens, R., Allen, M.S., Walters, C., Arlinghaus, R., 2020. Saving large fish through total fishextraction for many fishspecies. This can be perceived as goal harvest slots outperforms the classical minimum-length limit when the aim is to interference and explain the negative attitudes of anglers towards achieve multiple harvest and catch-related fisheries objectives. Fish Fish. 21, 483–510. commercial fisheries in the region. The high trophic overlap among Allen, M.S., Miranda, L.E., Brock, R.E., 1998. Implications of compensatory and additive commercial and recreational fisheries further fuels the perception of mortality to the management of selected sportfish populations. Lakes Reserv. Res. anglers that commercial fisheriesare heavily impacting on target stocks. Manag. 3, 67–79. Arlinghaus, R., 2005. A conceptual framework to identify and understand conflicts in We thus conclude that the conflict among commercial and recreational recreational fisheriessystems, with implications for sustainable management. Aquat. fisheriesis not just perceptual in nature, but has also an empirical basis Resour. Cult. Dev. 1, 145–174. related to the competition for fish. In most cases, where commercial Arlinghaus, R., 2006. Overcoming human obstacles to conservation of recreational – fishingyield dominates the harvest, anglers may be inferior competitors. fishery resources, with emphasis on central Europe. Environ. Conserv. 33, 46 59. Arlinghaus, R., Mehner, T., 2003. Management preferences of urban anglers: habitat The situation is different for pike, but this conflictis asymmetric as only rehabilitation versus other options. Fisheries 28, 10–17. few commercial fishers are complaining about declining pike catches Arlinghaus, R., Cyrus, E.-M., Eschbach, E., Fujitani, M., Hühn, D., Johnston, F., Pagel, T., and do not perceive anglers as a direct competitor for pike (Vogt, 2020). Riepe, C., 2015. Hand in Hand für eine nachhaltige Angelfischerei: Ergebnisse und Empfehlungen aus fünf Jahren praxisorientierter Forschung zu Fischbesatz und Clearly, there are other reasons than direct competition for fishthat will seinen Alternativen. Berichte des IGB 28. additionally feed into the conflict among commercial and recreational Basan, U., 1988. Wir Angeln in Bodden und Haffe (2. Aufl.). Sportverlag, Berlin. fisheriesin the region, for example issues of distributional justice (Vogt, Battaglia, M.P., Hoaglin, D.C., Frankel, M.R., 2009. Practical considerations in ranking survey data. Surv. Pract. 2, 1–10. 2020). Bavnick, M., 2005. Understanding fisheries conflicts in the South – a legal pluralist Our study also revealed that there is a strong conflictamong anglers perspective. Soc. Nat. Resour. 18, 805–820.

13 R. Arlinghaus et al. Fisheries Research 238 (2021) 105894

Begon, M., Harper, J.L., Townsend, C.R., 1996. Ecology. Individuals, Populations and Zolubas, T., Strehlow, H.V., 2018. Recreational sea fishing in Europe in a global Communities. Blackwell Scientific Publications, Oxford. context - participation rates, fishing effort, expenditure, and implications for Berkes, F., 1984. Competition between commercial and sport fishermen: an ecological monitoring and assessment. Fish Fish. 19, 225–243. analysis. Hum. Ecol. 12, 413–429. ICES, 2020a. ICES Advice on Fishing Opportunities, Catch, and Effort Baltic Sea and ¨ Biester, E., 1991. Okologische Veranderungen¨ in Boddengewassern¨ und mogliche¨ Folgen Greater North Sea Ecoregions - Herring (Clupea Harengus) in Subdivisions 20-24, in der Fischerei. Arbeiten des Deutschen Fischereiverbandes 52, 1–28. Spring Spawners (Skagerrak, Kattegat, and Western Baltic). ICES, Copenhagen, Bishop, R.C., Samples, K.C., 1980. Sport and commercial fishing conflicts: a theoretical Denmark. analysis. J. Env. Econ. Manage. 7, 220–233. ICES, 2020b. ICES Advice on Fishing Opportunities, Catch, and Effort in the Baltic Sea Boucquey, N., 2017. ‘That’s my livelihood, it’s your fun’: the conflicting moral Ecoregion - Cod (Gadus Morhua) in Subdivisions 22-24, Western Baltic Stock economies of commercial and recreational fishing. J. Rural Stud. 54, 138–150. (western Baltic Sea. ICES, Copenhagen, Denmark. Brehens, V., Rauschmayer, F., Wittmer, H., 2008. Managing national ‘problem’ species: Ihde, T.F., Wilberg, M.J., Loewensteiner, D.A., Secor, D.H., Miller, T.J., 2011. The why pan-European cormorant management is so difficult. Environ. Conserv. 53, increasing importance of marine recreational fishing in the US: challenges for 53–63. management. Fish. Res. 108, 268–276. Carrs, D.N., Bevan, R.M., Bonetti, A., Cherubini, G., Davies, J., Doherty, D., El Hili, A., Jacob, G.R., Schreyer, R., 1980. Conflictin outdoor recreation: a theoretical perspective. Feltham, M.J., Grade, N., Granadeiro, J.P., Gremillet, D., Gromadzka, J., Harari, Y.N. J. Leis. Res. 12, 368–380. R.H., Holden, T., Keller, T., Lariccia, G., Mantovani, R., Mc Carthy, T.K., Mellin, M., Jepsen, N., Flavio,´ H., Koed, A., 2019. The impact of cormorant predation on Atlantic Menke, T., Mirowska-Ibron, I., Muller, M., Musil, P., Nazirides, T., Suter, W., salmon and Sea trout smolt survival. Fish. Manag. Ecol. 26, 183–186. Trauttmansdorf, J., Volponi, S., Wilson, B., 1997. Techniques for assessing Kadagi, N.I., Wambiji, N., Belhabib, D., Ahrens, R.N.M., 2021. Ocean safaris or food: cormorant diet and food intake: toward a consensus view. Supplemento alle Ricerche characterizing competitive interactions between recreational and artisanal billfish di Biologia della Selvaggina 26, 197–230. fisheries on the coast of Kenya. Ocean Coast. Manage. 291, 105432. Chizinski, C.J., Martin, D.R., Pope, K.L., 2014. Self-confidenceof anglers in identification Kearney, R.E., 2001. Fisheries property rights and recreational/commercial conflict: of freshwater sport fish. Fish. Manag. Ecol. 21, 448–453. implications for policy developments in Australia and New Zealand. Mar. Pol. 25, Connelly, N.A., Brown, T.L., Knuth, B.A., 2000. Assessing the relative importance of 49–59. recall bias and nonresponse bias and adjusting for those biases in statewide angler Kearney, R.E., 2002. Co-management: the resolution of conflictbetween commercial and surveys. Hum. Dimens. Wildl. 5, 19–29. recreational fishers in Victoria, Australia. Ocean Coast. Manag. 45, 201–214. Cowx, I.G., 2003. Interactions Between Birds and Fish: Implications for Management. Keller, T., 1995. Food of cormorants Phalacrocorax carbo sinensis wintering in Bavaria, Blackwell Science, Oxford. southern Germany. Ardea 83, 185–192. Cowx, I.G., Arlinghaus, R., Cooke, S.J., 2010. Harmonizing recreational fisheries and Kerns, J.A., Allen, M.S., Harris, J.E., 2012. Importance of assessing population-level conservation objectives for aquatic biodiversity in inland waters. J. Fish Biol. 76, impact of catch-and-release mortality. Fisheries 37, 502–503. 2194–2215. Klimaszyk, P., Rzymski, P., 2016. The complexity of ecological impacts induced by great Dieperink, C., 1995. Depredation of commercial and recreational fisheries in a Danish cormorants. Hydrobiologia 771, 13–30. fjord by cormorants, Phalacrocorax carbo sinensis. Shaw. Fish. Manag. Ecol. 2, Larsson, P., Tibblin, P., Koch-Schmidt, P., Engstedt, O., Nilsson, J., Nordahl, O., 197–207. Forsman, A., 2015. Ecology, evolution, and management strategies of northern pike Donaldson, M.R., O’Connor, C.M., Thompson, L.A., Gingerich, A.J., Danylchuk, S.E., populations in the Baltic Sea. Ambio 44, 451–461. Duplain, R.R., Cooke, C.J., 2011. Contrasting global game fish and non-game fish Loomis, D.K., Ditton, R.B., 1993. Distributive justice in fisheries management. Fisheries species. Fish. 36, 385–397. 18, 14–18. Doring,¨ R., Berkenhagen, J., Hentsch, S., Kraus, G., 2020. Small-scale fisheries in Lucas, J.S., 2018. Recreational Fishing in the Bodden Waters of the Southern Baltic Sea - Germany: a disappearing profession? In: Pascual-Fernandez,´ J.J., Pita, C., Touristic Potential and Implications for Management. M.Sc. Thesis. Humboldt- Bavinck, M. (Eds.), Small-Scale Fisheries in Europe: Status, Resilience and University zu Berlin. Governance. Springer, pp. 483–502. LUNG (Landesamt für Umwelt, Naturschutz und Geologie), 2017. Kormoranbericht Dorow, M., Arlinghaus, R., 2009. Angelbegeisterung und anglerische Fischertrage¨ in Mecklenburg-Vorpommern 2017. LUNG, Güstrow, Germany. Binnen- und Küstengewassern¨ Mecklenburg-Vorpommerns unter besonderer Lyach, R., 2020. The effect of fishing effort, fish stocking, and population density of Berücksichtigung des Aals (Anguilla anguilla). Fischerei Fischmarkt Mecklenburg- overwintering cormorants on the harvest and recapture rates of three rheophilic fish Vorpommern 9, 36–46. species in central Europe. Fish. Res. 223, 105440. Dorow, M., Arlinghaus, R., 2011. A telephone-diary-mail approach to survey recreational Manning, R.E., 1979. Behavioral characteristics of fishermenand other recreationists on fisheries on large geographic scales, with a note on annual landings estimates by four Vermont rivers. Trans. Am. Fish. Soc. 108, 536–541. anglers in Northern Germany. Am. Fish. Soc. Symp. 75, 319–344. Manning, R.E., 1999. Studies in Outdoor Recreation: Search and Research for Dorow, M., Arlinghaus, R., 2012. The relationship between personal commitment to Satisfaction (2nd Edn). Oregon State University Press, Corvallis. angling and the opinions and attitudes of German anglers towards the conservation Martinson, K.S., Shelby, B., 1992. Encounter and proximity norms for salmon anglers in and management of the European eel Anguilla anguilla. N. Am. J. Fish. Manag. 32, California and New Zealand. N. Am. J. Fish. Manag. 12, 559–567. 466–479. Marzano, M., Carss, D.N., Cheyne, I., 2013. Managing European cormorant-fisheries Dorow, M., Beardmore, B., Haider, W., Arlinghaus, R., 2009. Using a novel survey conflicts: problems, practicalities and policy. Fish. Manag. Ecol. 20, 401–413. technique to predict fisheries stakeholders’ support for European eel (Anguilla Monk, C.T., Barbier, M., Romanczuk, P., Watson, J.R., Alos,´ J., Nakayama, S., anguilla L.) conservation programs. Biol. Conserv. 142, 2973–2982. Rubenstein, D.I., Levin, S.A., Arlinghaus, R., 2018. How ecology shapes exploitation: Ensinger, J., Bramick,¨ U., Fladung, E., Dorow, M., Arlinghaus, R., 2016. a framework to predict the behavioural response of human and animal foragers Charakterisierung und Perspektiven der Angelfischerei in Nordostdeutschland along exploration–exploitation trade-offs. Ecol. Lett. 21, 779–793. Potsdam–Sacrow. Schriften des Instituts für Binnenfischerei e.V. Band 44 (2016). Mustamaki,¨ N., Bergstrom,¨ U., Ådjers, K., Sevastik, A., Mattila, J., 2014. Pikeperch Institut für Binnenfischerei e.V., Potsdam–Sacrow. (Sander lucioperca L.) in decline - high mortality of three populations in the northern Ferter, K., Weltersbach, M.S., Strehlow, H.V., Vølsta, J.H., Alos,´ J., Arlinghaus, R., Baltic Sea. Ambio 43, 325–336. Armstrong, M., Dorow, M., De Graaf, M., Van der Hammen, T., Hyder, K., Levrel, H., Ngoc, Q.T.K., Flaaten, O., 2010. Protected areas for conflictresolution and management Paulrud, A., Radtke, K., Rocklin, D., Sparrevohn, C.R., Veiga, P., 2013. Unexpectedly of recreational and commercial fisheries. Mar. Resour. Econ. 25, 409–426. high catch-and-release rates in European marine recreational fisheries: implications Oesterwind, D., Bock, C., Forster,¨ A., Gable, M., Henseler, C., Kotterba, P., Menge, M., for science and management. ICES J. Mar. Sci. 70, 1319–1329. Myts, D., Winkler, H.M., 2017. Predator and prey: the role of the round goby Gabler, S., Hoffmeyer-Zlotnik, J.H.P., Krebs, D., 1994. Gewichtung in der Umfragepraxis. Neogobius melanostomus in the western Baltic. Mar. Biol. Res. 13, 188–197. Westdeutscher Verlag, Opladen, Germany. Olsen, J., 2019. Past and current trends of coastal predatory fishin the Baltic Sea with a Hansson, S., Bergstrom,¨ U., Bonsdorff, E., Hark¨ onen,¨ T., Jepsen, N., Kautsky, L., focus on perch, pike, and pikeperch. Fishes 4, 7. ¨ ¨ Lundstrom,¨ K., Lunneryd, S.-G., Ovegard, M., Salmi, J., Sendek, D., Vetemaa, M., Ostman, O, Bergenius, M.A.J., Bostrom,¨ M.K., Lunneryd, S.-G., 2012. Do cormorant 2018. Competition for the fish - fish extraction from the Baltic sea by humans, colonies affect local fishcommunities in the Baltic Sea? Can. Can. J. Fish. Aquat. Sci. aquatic mammals, and birds. ICES J. Mar. Sci. 75, 999–1008. 69, 1047–1055. ¨ ¨ Hartill, B.W., Edwards, C.T.T., 2015. Comparison of recreational harvest estimates Ostman, O., Bostrom,¨ M., Bergstrom,¨ U., Andersson, J., Lunneryd, S.-G., 2013. Estimating provided by on-site and offsite surveys: detecting bias and corroborating estimates. competition between wildlife and humans – a case of cormorants and coastal Can. J. Fish. Aquat. Sci. 72, 1379–1389. fisheries in the Baltic Sea. PLoS One 8, e83763. ¨ Horn, H.S., 1966. Measurement of "overlap" in comparative ecological studies. Am. Nat. Ovegård, M.K., Ohman, K., Mikkelsen, J.S., Jepsen, N., 2017. Cormorant predation 100, 419–424. overlaps with fish communities and commercial-fishery interest in a Swedish lake. Hunt, L.M., Gonder, D., Haider, W., 2010. Hearing voices from the silent majority: a Mar. Freshw. Res. 68, 1677–1685. comparison of preferred fish stocking outcomes for Lake Huron by anglers from Pauly, D., Christensen, V., Dalsgaard, J., Froese, R., Torres Jr., F., 1998. Fishing down representative and convenience samples. Human Dim. Wildl. 15, 27–44. marine food webs. Science 279, 860–863. Hyder, K., Weltersbach, M.S., Armstrong, M., Ferter, K., Townhill, B., Ahvonen, A., Preuß, D., 2002. Nahrungsokologische¨ Untersuchungen zum Einfluß des Kormorans Arlinghaus, R., Baikov, A., Bellanger, M., Birzaks, J., Borch, T., Cambie, G., De Phalacrocorax carbo sinensis auf die Fischerei im Küstenbereich Vorpommerns. Graaf, M., Diogo, H.M.C., Dziemian, Ł., Gordoa, A., Grzebielec, R., Hartill, B., Naturschutz Mecklenburg-Vorpommern 45, 57–67. Kagervall, A., Kapiris, K., Karlsson, M., Kleiven, A.R., Lejk, A.M., Levrel, H., Quinn, S.P., 1988. Effectiveness of restricted areas in reducing incidental catches of game Lovell, S., Lyle, J., Moilanen, P., Monkman, G., Morales-Nin, B., Mugerza, E., fish in a gill-net fishery. N. Am. J. Fish. Manag. 8, 224–230. Martinez, R., O’Reilly, P., Olesen, H.J., Papadopoulos, A., Pita, P., Radford, Z., Radford, Z., Hyder, K., Zarauz, L., Mugerza, E., Ferter, K., Prellezo, R., Strehlow, H.V., Radtke, K., Roche, W., Rocklin, D., Ruiz, J., Scougal, C., Silvestri, R., Skov, C., et al., 2018. The impact of marine recreational fishingon key fishstocks in European Steinback, S., Sundelof,¨ A., Svagzdys, A., Turnbull, D., Van Der Hammen, T., Van waters. PLoS One 13, e0201666. https://doi.org/10.1371/journal.pone.0201666. Voorhees, D., Van Winsen, F., Verleye, T., Veiga, P., Vølstad, J.-H., Zarauz, L.,

14 R. Arlinghaus et al. Fisheries Research 238 (2021) 105894

Raillard, S., 2012. Die See- und Küstenfischerei Mecklenburgs und Vorpommerns 1918 Trayler, K.M., Brothers, D.J., Wooler, R.D., Potter, I.C., 1989. Opportunistic foraging by bis 1960. Oldenbourg, München. three species of cormorants in an Australian estuary. J. Zool. 218, 87–98. Redpath, S.M., Young, J., Evely, A., Adams, W.M., Sutherland, W.J., Whitehouse, A., VanDeValk, A.J., Adams, C.M., Rudstam, L.G., Forney, J.L., Brooking, T.E., et al., 2002. Amar, A., Lambert, R.A., Linnell, J.D.C., Watt, A., Guti´errez, R.J., 2013. Comparison of angler and cormorant harvest of walleye and yellow perch in Oneida Understanding and managing conservation conflicts. Trends Ecol. Evol. (Amst.) 28, Lake, New York. Trans. Am. Fish. Soc. 131, 27–39. 100–109. Veneranta, L., Heikinheimo, O., Marjomaki,¨ T.J., Durif, C., 2020. Cormorant Rudstam, L.G., van de Valk, A.J., Adams, C., Coleman, J.T.H., Forney, J.L., Richmind, M. (Phalacrocorax carbo) predation on a coastal perch (Perca fluviatilis) population: E., 2004. Cormorant predation and the population dynamics of walleye and yellow estimated effects based on PIT tag mark-recapture experiment. ICES J. Mar. Sci. perch in Oneida Lake. Ecol. Appl. 14, 149–163. Vogt, A., 2020. Understanding Stakeholder Conflictsin Coastal Fisheries: Evidence From Samples, K.C., 1989. Assessing recreational and commercial conflicts over artificial the Brackish Lagoons Around Rügen, Germany. M.Sc. Thesis. Humboldt-University fishery habitat use: theory and practice. Bull. Mar. Sci. 44, 844–852. zu Berlin. ¨ Sampson, D.B., 2011. The accuracy of self- reported fisheriesdata: oregon trawl logbook Weltersbach, M.S., Riepe, C., Lewin, W.-C., Strehlow, H.V., 2021. Okologische, sozial und fishing locations and retained catches. Fish. Res. 112, 59–76. okonomische¨ Dimensionen des Meeresangelns in Deutschland. Braunschweig: Schiewer, U. (Ed.), 2008. Ecology of Baltic Coastal Waters. Springer-Verlag, Berlin. Johann Heinrich von Thünen-Institut. Thünen Report 83. https://doi.org/10.3220/ Skov, C., Jepsen, N., Baktoft, H., Jansen, T., Pedersen, S., Koed, A., 2014. Cormorant REP1611578297000. predation on PIT-tagged lake fish. J. Limnol. 73, 177–186. Winkler, H.M., 1990. Fishery in shallow estuaries of the GDR, possibilities and Smith, C.L., 1986. The life cycle of fisheries. Fisheries 11, 20–25. limitations. Limnologica 20, 195–198. Spijkers, J., Morrison, T., Blasiak, R., Cumming, G.S., Osborne, M., Watson, J., Winkler, H., Schroeder, H., 2003. Die Fischfauna der Ostsee, Bodden und Haffe. Meer ¨ Osterblom, H., 2018. Marine fisheries and future ocean conflict. Fish Fish. 19, und Museum 17, 25–35. 798–806. Winkler, H.M., Myts, D., Lüttkemoller,¨ E., Groger,¨ J., 2014. Ernahrung¨ des Kormorans Steffens, W., 2010. Great cormorant – substantial danger to fish populations and fishery und sein Einfluss auf die Fischbestande¨ der Küstengewasser¨ Vorpommerns. in Europe. Bulg. J. Agric. Sci. 16, 322–331. Sachbericht zum Fordervorhaben¨ LFI-LU-FA-09-12 “Populationsanalyse und Sullivan, M., 2003. Active management of walleye fisheries in Alberta: dilemmas of Erprobung von Maßnahmen zur Reduzierung des Bruterfolges beim Kormoran managing recovering fisheries. N. Am. J. Fish. Manag. 23, 1343–1358. (Phalacrocorax Carbo Sinensis) in M–V sowie Untersuchungen über seinen Einfluss Suter, W., 1997. Roach rules: shoaling fishare a constant factor in the diet of cormorants auf freilebende Fischbestande¨ . Phalacrocorax carbo in Switzerland. Ardea 85, 9–27.

15