==== Front Ambio Ambio Ambio 0044-7447 1654-7209 Springer Netherlands Dordrecht 32720251 1362 10.1007/s13280-020-01362-4 Perspective Depletion of coastal predatory fish sub-stocks coincided with the largest sea urchin grazing event observed in the NE Atlantic http://orcid.org/0000-0001-6374-4275Norderhaug Kjell Magnus kjellmn@hi.no 12Kjell Magnus Norderhaug is a principal scientist at the Institute of Marine Research IMR, Norway and associate professor at the University of Oslo Norway. His research focus on human impact in coastal ecosystems. Nedreaas Kjell kjelln@hi.no 3Kjell Nedreaas is a principal scientist at the Institute of Marine Research, Norway. His research interests include fisheries dynamics and ecosystem management. Huserbråten Mats mats.huserbraaten@hi.no 4Mats Huserbråten is a Post-Doc at the Institute of Marine Research, Norway. His research interests include the effects of oceanography and climate on early life history of Gadoids. Moland Even even.moland@graduates.jcu.edu.au 56Even Moland is a principal scientist at the Institute of Marine Research, Norway, and associate professor at Centre for Coastal Research (CCR), University of Agder, Norway. His research interests include impacts of harvesting and protection on marine populations and ecosystems. 1 grid.10917.3e0000 0004 0427 3161Institute of Marine Research IMR, Norway, Nye Flødevigveien 20, 4817 His, Norway 2 grid.5510.10000 0004 1936 8921University of Oslo Norway, Oslo, Norway 3 grid.10917.3e0000 0004 0427 3161Institute of Marine Research, Norway, Nordnesgaten 33, 5005 Bergen, Norway 4 grid.10917.3e0000 0004 0427 3161Institute of Marine Research, Norway, Nordnesgaten 50, 5005 Bergen, Norway 5 grid.10917.3e0000 0004 0427 3161Institute of Marine Research, Norway, Nye Flødevigveien 20, 4817 His, Norway 6 grid.23048.3d0000 0004 0417 6230Centre for Coastal Research (CCR), University of Agder, Kristiansand, Norway 27 7 2020 27 7 2020 1 2021 50 1 163 173 15 1 2020 23 4 2020 29 6 2020 © The Author(s) 2020Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.In this contribution, we propose fishery driven predator release as the cause for the largest grazing event ever observed in the NE Atlantic. Based on the evolving appreciation of limits to population connectivity, published and previously unpublished data, we discuss whether overfishing caused a grazer bloom of the sea urchin (Strongylocentrotus droebachiensis) resulting in overgrazing of more than 2000 km2 kelp (Laminaria hyperborea) forest along Norwegian and Russian coasts during the 1970 s. We show that coastal fisheries likely depleted predatory coastal fish stocks through modernization of fishing methods and fleet. These fish were important predators on urchins and the reduction coincided with the urchin bloom. From this circumstantial evidence, we hypothesize that coastal predatory fish were important in regulating sea urchins, and that a local population dynamics perspective is necessary in management of coastal ecosystems. Keywords Coastal fisheriesFisheries managementKelp forestRegime shiftsSea urchin grazingInstitute of Marine Research, Norwayhttp://dx.doi.org/10.13039/501100012704Universitetet i Agderissue-copyright-statement© Royal Swedish Academy of Sciences 2021 ==== Body Introduction Hunting and fishing by humans have decimated large predator populations in many coastal areas globally (Jackson et al. 2001). Removal of predators high in the food web have caused ecosystem collapse with cascading effects through several food web levels. In the absence of predators, grazers may flourish, which in turn leads to overgrazing of the primary producers. Benthic primary producers like kelp and seagrasses are engineering species that provide ecosystem services, including habitats, that promote ecological stability (Teagle et al. 2017). When removed, desert-like barren grounds can prevail for decades (Ling et al. 2015). Recent development of molecular tools and increasingly sophisticated use in population genetics have brought new insight into reproductive patterns and connectivity between marine populations. Only during the last decade these tools have changed our understanding on population dynamics of important fish stocks including coastal Atlantic cod (Gadus morhua) (Jorde et al. 2007; Knutsen et al. 2011; Dahle et al. 2018). This understanding throw light on why management of coastal fish stocks in many cases has failed in preventing ecosystem collapse. Historically, a view prevailed of coastal stocks being supported and replenished by near inexhaustible offshore resources, with a constant surplus of larvae, and starvation during vulnerable stages being the most important impediment to stable recruitment (Hjort 1914). However, more complex models have been developed to understand recruitment variability and fluctuations in fish populations (see Houde 2008). Recently, molecular tools have facilitated the discovery that along Norway’s convoluted coastline, limited gene-flow may result in local populations (e.g., Jorde et al. 2007; Knutsen et al. 2011; Quintela et al. 2020). In coastal Atlantic cod, such patterns in population structure are driven by processes such as adult movement and spawning behavior, placement of propagules by spawners, retention of propagules by local hydrographical phenomena and local retention of fish larvae (Ciannelli et al. 2010; Skjaeraasen et al. 2011; Rogers et al. 2014; Huserbråten et al. 2018; Barth et al. 2019). Similar findings are presently underway for a suite of species, with coastal and fjordic affinities and widely different life histories. This shift in our appreciation of implicit vulnerabilities challenges traditional stock assessment models and classical management—assuming discrete populations within large administrational areas—calls for sub-stock-specific management recommendations (see e.g., Reiss et al. 2009; Lindegren et al. 2013; Dahle et al. 2018; Svedäng et al. 2018). Importantly, an appreciation of local-scale processes in maintenance of coastal fish populations has implications for assessing the functional role of predatory fish and the impact of fisheries on coastal ecosystems (Hammerschlag et al. 2019). This needs to be taken into account in sustainable fisheries management as past management actions have failed to adapt to prevailing biological processes (Francis et al. 2007; Svedäng et al. 2010; Cardinale et al. 2017). Although both functional diversity and redundancy is thought to be high for demersal fish in the Barents Sea region (Aune et al. 2018), limited research has been directed towards ecosystem functioning and possible harvest induced alterations in the Arctic. Aquatic ecosystems characterized by site-attached species with limited connectivity and strong biological interactions are likely to harbor less functional redundancy than more ‘open’ ecosystems (Teichert et al. 2017). Whether separated by physical, genomic or behavioral barriers (see Barth et al. 2019), subpopulations are likely more vulnerable to local depletion than large widespread fish stocks (André et al. 2016; Cardinale et al. 2017; Gunnarson et al. 2019). Consequently, coastal fisheries targeting local sub-stocks of predatory fish may have resulted in far greater ecosystem effects than anticipated historically. This recognition may throw light on one historic event that took place five decades ago and with major repercussions for coastal production and diversity. A large-scale bloom of sea urchins Strongylocentrotus droebachiensis along Norwegian and Russian coasts took place and some 2000 km2 kelp Laminaria hyperborea forests were grazed down and turned into a marine desert, or so-called barren ground (Sivertsen 1997). To date, the causes of this bloom of grazers remain largely unknown. Here, we present previously unpublished data and discuss the hypothesis that past coastal fisheries removed predatory fish controlling sea urchin populations and thus triggered an ecosystem collapse. No systematic collection of fisheries statistics for coastal stocks exist from the period before and during the grazing event. However, we have digitalized historic and previously unpublished fishery sales notes statistics (data provided by SSB, Statistics Norway) and combined this with anecdotal knowledge to increase the spatiotemporal resolution of these data. Together with target species’ diet data, these datasets provide the first opportunity to evaluate this hypothesis. Methods To identify potentially important green sea urchin predators among fishery target species, our main source of information is the extensive data compilation by Planque et al. (2014) to determine food web links in the Barents Sea. Their data preparation protocol categorize predator–prey relationships from 1 to 4. From this we identified four candidate fishery target species (Atlantic wolffish, spotted wolffish, Atlantic cod and haddock, see Table 1) that were listed as ‘category 1′ predators of green sea urchin. For three of these species (Atlantic wolffish, haddock and Atlantic cod) we obtained, digitalized and plotted landings data for the relevant coastal areas and time period, from 1950 to present (fishery sales notes statistics provided by SSB, Statistics Norway). Accounts of Norwegian fisheries and landings can be found dating back to 1866. Until 1977 these data were reported by coastal district and county where the catches were landed and are only available on paper. Since 1977, Norwegian landings statistics are available in electronic format, by county and statistical area, as well as inside/outside 12 nautical miles (nm) from the baseline. Since 2000, landings statistics have a much higher resolution, i.e., down to fishing vessel, port and company level. For Atlantic wolffish and haddock, the landing statistics per county are used until 1980, and per statistical area inside 12 nm during 1980–2018. Hence, there may be some catches from outside 12 nm included in the county-based 1950–1979 landings statistics, and direct comparison between before and after 1980 should be made with caution. To ‘ground-truth’ our interpretation of landings patterns for Atlantic wolffish, we conducted informal interviews with three fishers with first-hand experience from the fishery, in space and time. Two (R. Rånes, I.J. Husby) were selected from IMRs reference fleet (administered by author KN), and the third (A.J. Trondal) was recruited based on IMR local knowledge in Finnmark. Based on otolith shape, the proportion of coastal cod can be distinguished from northeast arctic cod (NEAC) (Rollefsen 1933). Otolith samples have been collected from cod landings since 1984 and were therefore available from 1984 to 2018. During 1914–1983, Norwegian landings statistics of cod are categorized differently. The category “bank cod” and “fjord cod” (or “other cod”) is presented as a proxy for coastal cod landings during 1950–1961. Between 1970 and 1983 landing statistics of cod are mainly categorized as “cod”, without distinguishing landings spatially, temporally or biologically and are therefore not presented. For all three fish species, Welch Two Sample t-tests were performed to compare landings before and after the sea urchin outbreak. Coastal fisheries and ecosystem collapse globally Blooms of grazers escaping predator control have been reported from other coastal areas (Estes et al. 1998; Jackson et al. 2001; Steneck et al. 2004) and have increased in frequency, mainly due to overfishing and climate change (Ling et al. 2015). Among the most well-known consequences are the hunting of sea otters in Alaska (Estes et al. 1998) and fisheries in the Gulf of Maine (Steneck et al. 2004) both leading to blooms of grazers and denuding of kelp forests. Industrial fisheries started in the western world from the 1930s, and by the 1960s had developed enough capacity to overfish stocks—resulting in cascading perturbations in many coastal food webs globally (Jackson et al. 2001). In Maine, developing fishing technology and onboard refrigeration improved the efficiency of the fleet to target spawning coastal cod and other predatory fish. By the 1960s, overfishing of large predatory fish including Atlantic wolffish, coastal Atlantic cod and haddock caused blooms of S. droebachiensis along the coast of Maine after more than 4 000 years of predatory fish domination (Steneck et al. 2004). The resulting overgrazing of kelp forests and formation of barren grounds lasted until fisheries decimated urchin populations in the mid-1990s (Steneck et al. 2013). As a consequence, kelp forests recovered. Predatory fish were still scarce, and a new apex predator expanded in the vacant niche, the crab Cancer borealis. The only large-scale grazing event in the NE Atlantic Blooms of green sea urchins S. droebachiensis were first observed by fishermen in the early 1970s along the coast of Mid- and North Norway. A 2 000 km2 large area formerly dominated by kelp L. hyperborea forests from Smøla (63° N) in the south to Finnmark (71° N) and into Russia in the north were denuded and turned into sea urchin dominated barren grounds (Sivertsen 1997). This constitutes the first known large-scale grazing event in the NE Atlantic from any available source going back almost 150 years (Norderhaug and Christie 2009). While kelp forests have recovered in the southern part of the barren grounds, sea urchins still dominate most of the denuded areas five decades later (Christie et al. 2018). Removal of kelp forests have severe implications for coastal production and diversity, provision of habitats and nursery areas for coastal fish (Filbee-Dexter and Scheibling 2014) and sequestration of carbon in the affected areas (Krause-Jensen and Duarte 2016). Atlantic wolfish, haddock, coastal cod and other coastal fish associated to kelp forests have remained at low levels in the grazed areas and are expected to impact seabirds (Christensen-Dalsgaard et al. 2017) and marine mammals (Bjørge 1995). In the period after the urchin bloom, several hypotheses were put forward, including regulation of sea urchin populations by the endoparasite Echinomermella matsi (Hagen 1987), amoebic disease (Christie et al. 1995), and temperature (Sivertsen 2006). These hypotheses were later rejected or questioned: Screenings revealed no evidence for microbial infections explaining mass mortality of sea urchins observed at 66°44′ N (Christie et al. 1995), and E. matsi could not account for local population crash episodes of S. droebachiensis (several local studies including, Stien et al. 1995). While S. droebachiensis is a cold-water species and prolific during cold periods, temperature alone cannot explain the outbreak (Anon. 2002; Norderhaug and Christie 2009). Thus, the drivers responsible for the “regime shift” have never been properly understood. Diets of predatory fish The importance of predators in regulating prey populations depend on biological interaction strength. The strong controlling impact S. droebachiensis has on kelp L. hyperborea is widely documented including Norwegian waters (Ling et al. 2015). In Maine and other coastal areas, the strong impact from predatory fish like coastal cod, haddock and Atlantic wolffish in regulating S. droebachiensis is well documented (Steneck et al. 2013). Also, in north Norwegian coastal waters available literature suggest these predatory fish are important predators on S. droebachiensis (Planque et al. 2014; Strand 2019). Benthic predatory fish are typically generalists and their prey will depend on what is available. Diets will therefore vary in time and space (Jiang and Jørgensen 1996). We do not know the diet of our focal fish species (Table 1) in the period prior to the grazing event, but S. droebachiensis may well have been a preferred food item because of its high nutritional value, especially when gonads are well developed (Tam et al. 2016). Today, the green sea urchin S. droebachiensis is considered the main prey of Atlantic wolffish (Anarhichas lupus) in the grazed region (Falk-Petersen et al. 2010), and also important prey for the spotted wolffish (Anarhichas minor, Simpson et al. 2013). Both wolffish species are vulnerable to harvest due to late sexual maturity (6–7 and 7–10 years for A. lupus and A. minor, respectively), site fidelity and life history attributes that confer vulnerability to overharvesting (see Table 1). Coastal Atlantic cod Gadus morhua is a typical opportunistic generalist predator with variable diet including fish and invertebrates, and S. droebachiensis is important prey in northern Norway fjord areas where they are abundant (Enoksen and Reiss 2018).Table 1 Fishery target species categorized as principal predators on green sea urchins S. droebachiensis (‘category 1′ according to Planque et al. 2014) Species Indices of vulnerability Gear Modernization References Antlantic wolffish Anarhicas lupus Internal fertilization, late maturing, low fecundity, paternal care of demersal eggs, homing to feeding and spawning grounds Longline Bottom trawlb Gillnet ABCD Eliassen et al. (1981), Keats et al. (1985), Falk-Petersen et al. (2010), Simpson et al. (2013) and Gunnarson et al. (2019) Spotted wolffish Anarhicas minor Internal fertilization, late maturing, low fecundity, paternal care of demersal eggs Longline Bottom trawlb Gillnet ABCD Eliassen et al. (1981), Gunnarson et al. (2008) and Simpson et al. (2013) Norwegian coastal cod Gadus morhua Spawning aggregation, spawning site fidelity, population structure Gillnet Longline Handline Bottom trawl Danish seine ABCE Jorde et al. (2007), Skjaeraasen et al. (2011), Dahle et al. (2018) and Enoksen and Reiss (2018) NEA haddock Melanogrammus aeglefinus Spawning aggregation, Population structurea Gillnet Longline Handline Bottom trawl Danish seine ABCE Jiang and Jørgensen (1996), Reiss et al. (2009), González-Irusta and Wright (2016) and Tam et al. (2016) Indices of vulnerability: biological, life cycle or life history attributes with consequences for the species’ vulnerability to harvesting. Gear: mode of capture/fishing gear directly or indirectly affecting the target species. Modernization: technological development and demand (1960s–80 s) affecting targeting of the species (see Notes at bottom of table) A = increased engine power; B = increased vessel size; C = introduction of nylon fiber; D = advent of market/demand, E = introduction of the hydraulic net hauler/line hauler aSpatial scale of population structure poorly known (Reiss et al. 2009) bWolffish are by-caught in bottom trawling, and bottom trawling is detrimental to wolffish habitat Haddock Melanogrammus aglefinus has a variable diet dominated by crustaceans and small echinoderms (Jiang and Jørgensen 1996; Tam et al. 2016) including S. droebachiensis. Being a choosier benthic feeder than Atlantic cod, some haddock individuals might specialize as urchin foragers and utilize periods with high abundance of post-settlement stage urchins (see Fig. 1). Together, Atlantic cod and haddock—would likely have exerted significant mortality on urchins throughout the affected area when abundant.Fig. 1 Green sea urchins Strongylocentrotus droebachiensis grazing kelp Laminaria hyperborea at Hammerfest (left side, Norway 71° N, photo: Stein Fredriksen) and green sea urchins in a haddock stomach sample (Photo: Hans Kristian Strand) Development of Norway’s coastal fisheries Prior to the 1930s Norway’s fisheries were still dominated by handline and other gear that could be hand-hauled by fishers from traditional- and smaller-motorized fishing vessels. By the onset of World War II, the fleet had entered a period of rapid technical development. Modern seine was introduced in the herring-, sprat- and saithe fisheries, and the first Norwegian trawlers had been operating since 1936. To bolster the coastal fisheries, Norwegian government offered loans and subsidies for boat building and repairs, which led to a sharp increase in the capacity (Gerhardsen 1945). Despite increased capacity, less fish was brought to market during the war as subsistence fishing dominated. Landings dropped from one million metric tons in 1940 to ≈ 600 thousand tons in 1945. With peace in 1945, the coastal fishery expanded rapidly with a surge in boat building and participation. The number of registered vessels increased from ≈ 30 thousand in 1945 to an all-time high of 41 433 vessels in 1960 (Steinset 2018). Introduction of nylon in the late 1950s, and the hydraulic power block in the 1960s gave single fishing vessels the ability to haul deeper and longer line-sets and gill-nets, as well as seines that previously required large teams (Benum 2015). The technological revolution resulted in an unprecedented and unchecked increase in fishing effort, that eventually drove the mighty spring spawning herring (Clupea harengus) to collapse in the late 1960s (Dragesund et al. 2008). Fisheries landing statistics digitalized and presented herein (described in the Method section) indicate depletion of predatory fish stocks in the decades before grazers bloomed and the coastal ecosystem collapsed. Landings of Atlantic wolffish (± SD) decreased significantly from 3268 (± 1187) tons before the grazer bloom (1950–1969) to 2215 (± 777) tons during the bloom (1970–1989) and to 869 (± 584) tons after sea urchins started retreating (1990–2017, Fig. 2, t = 8.35, df = 25.6, p < 0.0001 when comparing before and after). For haddock, annual landings were significantly reduced from 38 128 (± 11 014) tons before, to 36 561 (± 17 443) during and 27 376 (± 7324) tons after the bloom (t = 3.81, df = 30.7, p = 0.0006). The data, further, indicate increasing fisheries in the north as catches were reduced in the south. For Atlantic cod, coastal catches were significantly reduced from 59 932 (± 14 520) tons before (1950–1969) to 36 856 (± 13 279) tons after (1984–2017, t = 5.62, df = 38.8, p < 0.0001). For wolffish– primary predators on urchins—interviews with fishers conducted by the authors corroborate the pattern indicated by landings data. While demand and market for cod and haddock was well established for centuries, neither Atlantic nor spotted wolffish were targeted until the 1950s as they were considered unpalatable and unmarketable by Norwegian fishers and consumers. Throughout the 1950s and 1960s this changed as demand for these species drove prices up and opened a lucrative fishery that fishers could exploit during spring and summer, the off-season for the more established fisheries. Long-lines baited with squid were set in shallow water where wolffish were available—sometimes yielding catches where “every hook held a wolffish” when gear was deployed in virgin grounds (R. Rånes, pers. comm., see Methods). Atlantic wolffish landings peaked in the mid-1950s with well over 5000 metric tons annually. In the period 1962–1964, more than 3000 tons was still being landed annually in the affected areas, increasingly from the northernmost areas. By the late 1960s, Atlantic wolffish were gone from fjords and lagoons in the Bodø area (Nordland—Fig. 2, landings in orange), with no sign of recovery to date (R. Rånes, pers. comm.). Throughout the 1970s, wolffish species became high-end seafood also for Norwegian consumers (Eliassen et al. 1981). Over the period, the landings data indicated that stocks of coastal A. lupus underwent serial depletion from south to north. By the late 1980s, all but the northernmost areas were fished out. According to the complimentary interviews, the Nordland N—Troms S coastal section (Fig. 2, landings in light purple) yielded good catches from outer coastal areas on virgin grounds (< 100 m depth) during the 1980s to 90 s (I.J. Husby, pers. comm.). From Finnmark (Fig. 2, landings in purple), shallow coastal areas yielded good catches in a valuable spring fishery during the years 1997–1999 (A.J. Trondal, pers. comm., see also, Strand 2019). The conspicuous reduction in catches of all three predatory fish stocks during the 1980s was not related to regulation—as no or few regulations were in place until 1989—and therefore most likely resulting from stocks having been depleted throughout the historical fishing grounds. Other possible ecological reasons for the observed decline in landings include alterations in habitat quality or food availability. Bottom trawling would be the most likely driver of habitat alterations, especially detrimental to wolffish habitat (Table 1). Assessing the total impact of bottom trawling—also including international fishing vessels operating as near as 4 nmi from the Norwegian coast until the 1970s (A.H. Hoel, pers. comm.)—was beyond the scope of this study. It is possible that the fisheries-induced collapse of spring spawning herring in the late 1960s (see Dragesund et al. 2008), did confer changes in food availability to coastal predatory fish stocks in succeeding years. The consequences of the putative temporary loss of this trophic link would be worth exploring in future research. Lastly, the observed decline in landings could be affected by alteration in fisher behavior. Towards the end of the 1970s and for succeeding years, the Joint Russian-Norwegian Fisheries Commission agreed to introduce annual cod quotas, albeit insufficient to stop a positive trend in fishing mortality until the late 1980s (see Yaragina et al. 2011 and references therein). However, to the best knowledge of the authors, the degree to which this agreement brought about a shift in fisher behavior is unclear and does not preclude our conclusions.Fig. 2 Spatiotemporal development of coastal fisheries for a Atlantic wolffish (A. lupus), b haddock (M. eaglefinus) and c Atlantic coastal cod (G. morhua) shown as regional landings (darker color with increasing latitude) available for the period 1950–2018 (see Methods) within the area where sea urchins (S. droebachiensis) bloomed (Norderhaug and Christie 2009). The green lines show average landings (± SD) before, during urchin overgrazing and after sea urchins started to retreat. During 1970–1980 (dark gray period) urchin populations bloomed and barren ground area coverage peaked. During 1980–1990 (light gray period) sea urchins gradually retracted while kelp recovered in the southernmost part of the barren area. Only total landings data of cod including Barents Sea cod are available during 1970–1983 Discussion Herein we have summarized and evaluated available evidence on the potential historical role of local fisheries in depleting predatory fish stocks along coastal Norway (Fig. 3a). We have shown that these predatory fish are important predators on the green sea urchin S. droebachiensis, the principal grazer in the coastal ecosystem. We have also shown that the reduction of fish stocks coincided with the largest historical bloom of S. droebachiensis observed in coastal Norway for almost 150 years (Norderhaug and Christie 2009), resulting in large-scale denuding of kelp L. hyperborea forests into barren grounds during the 1970s.Fig. 3 Hypothesized drivers of the regime shift and idealized ecosystem-based adaptive management. a Prior to 1950, the fishing pressure did not deplete coastal demersal predators of sea urchins to levels that threatened functional redundancy. From 1950, an unregulated coastal fishery, rapid technological development of the fleet and increased price of wolffish led to overfishing, serial depletion and loss of ecosystem function (ecosystem overfishing, Murawski 2000). By 1980, the barren ground state reached its maximum spatial coverage (photos from the top: J Thormar, S Fredriksen, E Svensen). b Adaptive ecosystem-based management takes species- and sub-stock-specific vulnerabilities into account. Environmental monitoring (symbolized by gauges) on species, sub-stock, and ecosystem state and function is used to revise and tune management and protection tools (symbolized by fish-hook size). Conservation of sub-stock biomass and size-structure prevent urchin blooms and support healthy ecosystems dominated by kelp forests (photo: J Thormar) Based upon this—and the evolving understanding of limits to connectivity in coastal populations, we hypothesize that coastal predatory fish are important in regulating sea urchin populations and that overfishing has caused ecosystem collapse by releasing the principal grazer from predatory control. The resulting large-scale overgrazing of L. hyperborea kelp forest had dramatic consequences for the coastal ecosystem structure and hence also on its diversity, production, functional redundancy and other ecosystem services, with effects far exceeding the loss of kelp forests itself (Norderhaug and Christie 2009). While the evidence we present is circumstantial, the course of events is similar to other coastal areas and Norwegian ecosystems that seem to have collapsed in the same way (Estes et al. 1998; Jackson et al. 2001; Steneck et al. 2004). The realization of coastal fish stocks’ vulnerability to fishing may aid in understanding the failures of the past with implications for what can be defined as sound and sustainable management (see ‘Indices of vulnerability’ in Table 1). Small, semi-isolated populations, and particularly slow-growing and long-lived species, need strong regulation since the risk of local extinction for populations by random disturbances increases with isolation but decreases with population size and reproductive potential (see e.g., Begon et al. 2005). The combination of fish traits and the convoluted Norwegian coastal topography therefore calls for caution when it comes to fishing pressure (see Dahle et al. 2018). Coastal fisheries are integral in Norwegian culture and rural development policy. The layman’s view is that such fisheries—based on smaller vessels—is a sustainable part of a ‘natural’ utilization of coastal production by small communities, which do not pose a threat to fish stocks in the same way as large-scale fisheries by industrial trawlers. This was probably an important reason why the management historically focused on assessment and advice for offshore stocks. Until 1989, the management of coastal stocks mainly emphasized regulating vessel size and gear standards, rather than monitoring stock sizes and vulnerabilities (Hylen et al. 2008; Gullestad et al. 2014). This phenomenon in fisheries management, termed the ‘easy restriction syndrome’ by Cardinale et al. (2017), refers to using the least controversial restriction without basis in data. The failure of taking the modernization of this previously unmanaged fishing fleet into consideration may have been catastrophic for several small and/or local fish stocks. To this day, the predatory fish stocks has failed to recover (Fig. 2), and similar to the Gulf of Maine coast and facilitated by ocean warming, predatory crab populations expand and seem to have taken over the niche as apex predators (Fagerli et al. 2014; Christie et al. 2018). In our view, there is an urgent need for bridging the gap between research and policy. Successful management of coastal fish stocks hinges on good information on life history, reproductive strategy, population structure (connectivity and source sink dynamics), and functional role in the ecosystem (Fig. 3b, see also Francis et al. 2007). When such information is available—which is now true for a growing list of species and local populations, it should have tangible bearing on management advice and policy. To exemplify herein, we have summarized ‘Indices of vulnerability’ for the highlighted target species (see Table 1). Such traits must be considered carefully when designing harvest rules, where conservation of old-growth age structure should be a priority. Moreover, management needs to be implemented at relevant spatial scales, i.e., matching the geographical distribution of a population. We call for adaptive management systems including sub-stock monitoring and the use of management security fuses such as marine protected areas (MPAs). Adaptive management can be seen as a learning process with stock monitoring as an important part (Curtin and Prellezo 2010). Stakeholder involvement and local ownership is vital to increase society’s understanding of managing local stocks and ecosystems, and quantitative data from both monitoring and effectiveness of MPAs must be used. By now, MPAs that impose strict limitations on fishing activities (e.g., no-take marine reserves) are proven as efficient management tools for species with limited home ranges and may therefore be effective also for northern temperate coastal fish stocks (Fenberg et al. 2012; Fernández-Chacón et al. 2015). MPAs also serve as reproductive refugia, effectively conserving spawning biomass and promoting recruitment if overfishing occurs despite other management efforts. Importantly, protection may confer trophic changes and ecosystem stability. Indeed, reversal of urchin dominance has been demonstrated in no-take MPAs with restored predator populations and size-structure in the Mediterranean (Guidetti and Sala 2007), Tasmania (Edgar et al. 2009) and California (Hamilton and Caselle 2015). Long-term management success requires a holistic approach and whole-ecosystem perspective (Francis et al. 2007; Houde 2008; Curtin and Prellezo 2010). Barren and kelp forest states are stable, while shifts between them are discontinuous (Ling et al. 2015). Understanding of threshold and feedback factors preventing kelp recovery should therefore be part of management strategies (Filbee-Dexter and Scheibling 2014). Recovery of kelp forests on the southern border of the barren ground area since the 1990s (Norderhaug and Christie 2009) suggest that this is achievable when conditions are favorable. Further work to explore the large-scale grazing event described, and the hypothesis put forward herein, would benefit from an ecosystem-modeling approach in which these dynamics might be simulated. Shared understanding of lessons learned, and inclusion of novel management and research tools in effective adaptive management systems are necessary to increase our understanding, avoid future ecosystem collapses and to restore kelp forests and depleted coastal fish stocks. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Acknowledgements Open Access funding provided by Institute Of Marine Research. Two anonymous referees provided insightful comments that significantly improved the quality of this contribution. We thank IMR colleagues Å. Bjordal, A.H. Hoel and senior advisor A. Haaland at Norwegian Fisheries Museum, both for sharing knowledge on Norway’s fisheries history. Grethe Kuhnle, Directorate of fisheries, Norway provided valuable help in extracting fisheries statistics. Funding to EM was provided by IMR through the Coastal Ecosystems research program (Project 14 914), and by University of Agder through the priority research center allocation to Centre for Coastal Research (CCR). ==== Refs References André C Svedäng H Knutsen H Dahle G Jonsson P Ring A-K Sköld M Jordem PE Population structure in Atlantic cod in the eastern North Sea-Skagerrak-Kattegat: early life stage dispersal and adult migration BMC Research Notes 2016 9 63 10.1186/s13104-016-1878-9 26843185 Anon. 2002. Nedbeiting av tareskog i Norge. Report to Ministry of Fishery and Coastal Affairs. (in Norwegian, English summary). Barth JMI Villegas-Ríos D Freitas C Moland E Star B André C Knutsen H Bradbury I Disentangling structural genomic and behavioural barriers in a sea of connectivity Molecular Ecology 2019 28 1394 1411 10.1111/mec.15010 30633410 Begon M Townsendand CR Harper JL Ecology: From Individuals to Ecosystems 2005 London Wiley-Blackwell Benum, E. 2015. Industrialisering av fiske og jordbruk (in Norwegian). University of Oslo, Norgeshistorie.no. Retrieved 20 December, 2019, from https://www.norgeshistorie.no/oljealder-og-overflod/artikler/1940-industrialisering-av-fiske-og-jordbruk.html. Bjørge A Evans PGH Raga JA How persistent are marine mammal habitats in an ocean of variability? Marine Mammals , Biology and Conservation 1995 New York Springer Science 63 92 Cardinale M Svenson A Hjelm J The “easy restriction” syndrome drive local fish stocks to extinction: The case of the management of Swedish coastal populations Marine Policy 2017 83 179 183 10.1016/j.marpol.2017.06.011 Christensen-Dalsgaard S Mattisson J Bekkby T Gundersen H May R Rinde E Lorentsen SH Habitat selection of foraging chick-rearing European shags in contrasting marine environments Marine Biology 2017 164 196 10.1007/s00227-017-3227-5 Christie H Leinaas HP Skadsheim A Skjoldal HR Hopkins C Erikstad KE Leinaas HP Local patterns in mortality of the green sea urchin, Strongylocentrotus droebachiensis , at the Norwegian coast Ecology of Fjords and Coastal Waters 1995 Amsterdam Elsevier 573 584 Christie H Gundersen H Rinde E Filbee-Dexter K Norderhaug KM Pedersen T Bekkby T Gitmark J Can multitrophic interactions and ocean warming influence large-scale kelp recovery? Ecology and Evolution 2018 9 2847 2862 10.1002/ece3.4963 Ciannelli L Knutsen H Olsen EM Espeland SH Asplin L Jelmert A Knutsen JA Maintenance of small-scale genetic structure in a marine population in relation to water circulation and egg characteristics Ecology 2010 91 2918 2930 10.1890/09-1548.1 21058552 Curtin R Prellezo R Understanding marine ecosystem based management: A literature review Marine Policy 2010 34 821 830 10.1016/j.marpol.2010.01.003 Dahle G Quintela M Johansen T Westgaard J-I Besnier F Aglen A Jørstad KE Glover KA Analysis of coastal cod (Gadus morhua L.) sampled on spawning sites reveals a genetic gradient throughout Norway’s coastline BMC Genetics 2018 2018 19 42 Dragesund O Østvedt OJ Toresen R Nakken O Norwegian spring-spawning herring:history of fisheries, biology and stock assessment In Norwegian Spring -Spawning Herring and Northeast Arctic Cod —100 Years of Research and Management 2008 Trondheim Tapir Academic Press 41 82 Edgar G Barret NS Stuart-Smith RD Exploited reefs protected from fishing transform over decades into conservation features otherwise absent from seascapes Ecological Applications 2009 19 1967 1974 10.1890/09-0610.1 20014571 Eliassen JE Venvik T Perger S Kristiansen S Steinbit (wolffish) ‘Ufesk’ -fra agn til delikatesse 1981 Oslo Arrow Team 74 83 Enoksen SE Reiss H Diet of Norwegian coastal cod (Gadus morhua ) studied by using citizen science Journal of Marine Systems 2018 180 246 254 10.1016/j.jmarsys.2017.06.006 Estes JA Tinker MT Williams TM Doak DF Killer whale predation on sea otters linking oceanic and nearshore ecosystems Science 1998 282 473 476 10.1126/science.282.5388.473 9774274 Fagerli CW Norderhaug KM Christie H Pedersen MF Fredriksen S Predators of the destructive sea urchin grazer Strongylocentrotus droebachiensis on the Norwegian coast Marine Ecology Progress Series 2014 502 207 218 10.3354/meps10701 Falk-Petersen I-B Kanapathippila P Primicerio R Hansen TK Size-, locality- and seasonally related feeding prefeences of common wolffish (Anarhichas lupus L.) from north-Norwegian waters Marine Biology Research 2010 6 201 212 10.1080/17451000903078614 Fenberg PB Caselle J Claudet J Clemence M Gaines S García-Charton JA Gonçalves E Grorud-Colvert K The science of European marine reserves: status, efficacy and future needs Marine Policy 2012 36 1012 1021 10.1016/j.marpol.2012.02.021 Fernández-Chacón A Moland E Espeland SH Olsen EM Demographic effects of full vs. partial protection from harvesting: inference from an empirical before–after control-impact study on Atlantic cod Journal of Applied Ecology 2015 52 1206 1215 10.1111/1365-2664.12477 Filbee-Dexter K Scheibling RE Sea urchin barrens as alternative stable states of collapsed kelp ecosystems Marine Ecology Progress Series 2014 495 1 25 10.3354/meps10573 Francis RC Hixon MA Clarke ME Murawski SA Ralston S Ten commandments of ecosystem-based fisheries scientists Fisheries 2007 32 217 233 10.1577/1548-8446(2007)32[217:TCFBFS]2.0.CO;2 Gerhardsen, G.M. 1945. Våre fiskerier under og etter krigen. Sak og samfunn nr. 7, Bergen: Chr. Michelsens Institutt for Vitenskap og Åndsfrihet (in Norwegian with English abstract). González-Irusta JM Wright PJ Spawning grounds of haddock (Melanogrammus aeglefinus ) in the North Sea West of Scotland Fisheries Research 2016 183 180 191 10.1016/j.fishres.2016.05.028 Guidetti S Sala E Community-wide effects of marine reserves in the Mediterranean Sea Marine Ecology Progress Series 2007 335 43 56 10.3354/meps335043 Gullestad P Aglen A Bjordal Å Blom G Johansen S Krog J Misund OA Røttingen I Changing attitudes 1970–2012: Evolution of the Norwegian management framework to prevent overfishing and to secure long-term sustainability ICES Journal of Marine Science 2014 71 173 182 10.1093/icesjms/fst094 Gunnarson Á Sólmundsson J Björnsson H Sigurðsson G Pampoulie C Migration patterns and evidence of homing in the Atlantic wolffish (Anarhichas lupus ) Fisheries Research 2019 215 69 75 10.1016/j.fishres.2019.03.001 Hagen NT Sea urchin outbreaks and nematode epizootics in Vestfjorden, Northern Norway Sarsia 1987 72 213 229 10.1080/00364827.1987.10419719 Hamilton SL Caselle JE Exploitation and recovery of a sea urchin predator has implications for the resilience of southern California kelp forests Proceedings of the Royal Society B 2015 282 20141817 10.1098/rspb.2014.1817 25500572 Hammerschlag N Schmitz OJ Flecker AS Lafferty KD Sih A Atwood TB Gallagher AJ Irschick DJ Ecosystem function and services of aquatic predators in the anthropocene Trends in Ecology & Evolution 2019 34 369 383 10.1016/j.tree.2019.01.005 30857757 Hjort J Fluctuations in the great fisheries of northern Europe viewed in the light of biological research Rapports et procès-verbaux des reunions Conseil permanent international pour l’exploration de la mer 1914 20 1 228 Houde ED Emerging from Hjort’s shadow Journal of Northwest Atlantic Fishery and Science 2008 41 53 70 10.2960/J.v41.m634 Huserbråten MBO Moland E Albretsen J Cod at drift in the North Sea Progress in Oceanography 2018 167 116 124 10.1016/j.pocean.2018.07.005 Hylen A Nakken O Nedreaas K Nakken O Northeast Arctic cod: fisheries, life history, stock fluctuations and management Norwegian Spring -Spawning Herring and Northeast Arctic Cod —100 Years of Research and Management 2008 Trondheim Tapir Academic Press 83 118 Jackson JBC Kirby MX Berger WH Bjorndal KA Botsford LW Bourque BJ Bradbury RH Cooke R Historical overfishing and the recent collapse of coastal ecosystems Science 2001 293 629 637 10.1126/science.1059199 11474098 Jiang W Jørgensen T The diet of haddock (Melanogrammus aeglefinus L.) in the Barents Sea during the period 1984-1991 ICES Journal of Marine Science 1996 53 11 21 10.1006/jmsc.1996.0002 Jorde PE Knutsen H Espeland SH Stenseth NC Spatial scale of genetic structuring in coastal cod Gadus morhua and geographic extent of local populations Marine Ecology Progress Series 2007 343 229 237 10.3354/meps06922 Keats W South GR Steele DH Reproduction and egg guarding by Atlantic wolffish (Anarhichas lupus : Anarhichidae) and ocean pout (Macrozoarces americanus : Zoarcidae) in Newfoundland waters Canadian Journal of Zoology 1985 63 2565 2568 10.1139/z85-382 Knutsen H Olsen EM Jorde PE Espeland SH André C Stenseth NC Are low but statistically levels of genetic differentiation in marine fishes ‘biologically meaningful’? A case study of coastal Atlantic cod Molecular Ecology 2011 20 768 783 10.1111/j.1365-294X.2010.04979.x 21199035 Krause-Jensen D Duarte CMI Substantial role of macroalgae in marine carbon sequestration Nature Geoscience 2016 10 737 742 10.1038/ngeo2790 Lindegren M Waldo S Nilsson PA Svedäng H Persson A Towards sustainable fisheries of the Öresund cod (Gadus morhua ) through sub-stock-specific assessment and management recommendations ICES Journal of Marine Science 2013 70 1140 1150 10.1093/icesjms/fst042 Ling SD Scheibling RE Johnson CR Rassweiler A Shears N Connell SD Salomon A Norderhaug KM Global regime-shift dynamics of catastrophic sea urchin overgrazing Philosophical Transactions B 2015 370 20130269 10.1098/rstb.2013.0269 Murawski SA Definitions of overfishing from an ecosystem perspective ICES Journal of Marine Science 2000 57 649 658 10.1006/jmsc.2000.0738 Norderhaug KM Christie HC Sea urchin grazing and kelp re-vegetation in the NE Atlantic Marine Biology Research 2009 5 515 528 10.1080/17451000902932985 Planque B Primicerio R Michalsen K Aschan M Certain G Dalpadado P Gjøsæter H Hansen C Who eats whom in the Barents Sea: a food topology from plankton to whales Ecology 2014 95 1430 10.1890/13-1062.1 Quintela M Kvamme C Bekkevold D Nash RDM Jansson E Sørvik AG Taggart JB Skaala Ø Genetic analysis redraws management boundaries for the European sprat Evolutionary Applications 2020 00 1 17 10.1111/eva.12942 Reiss H Hoarau G Dickey-Collas M Wolff WJ Genetic population structure of marine fish: mismatch between biological and fisheries management units Fish and Fisheries 2009 10 361 395 10.1111/j.1467-2979.2008.00324.x Rogers LA Olsen EM Knutsen H Stenseth NC Habitat effects on population connectivity in a coastal seascape Marine Ecology Progress Series 2014 511 153 163 10.3354/meps10944 Rollefsen G The otoliths of the cod. Preliminary report. Fiskeridirektoratets Skrifter Serie Havundersøkelser (Report of the Norwegian Fisheries and Marine Investigations) 1933 4 3 14 Simpson, M. R., G. D. Sherwood, L. G. S. Mello, C. M. Miri and D. W. Kulka. 2013. Feeding habits and trophic niche differentiation in three species of wolffish (Anarhichas sp.) inhabiting Newfoundland and Labrador waters. DFO Canadian Science Advisory Secretariat Research Document Report 2013/056. Sivertsen K Geographical and environmental factors affecting the distribution of kelp beds and barren grounds and changes in biota associated with kelp reduction at sites along the Norwegian coast Canadian Journal of Fisheries and Aquatic Sciences 1997 54 2872 2887 10.1139/f97-186 Sivertsen K Overgrazing of kelp beds along the coast of Norway Journal of Applied Phycology 2006 18 599 610 10.1007/s10811-006-9064-4 Skjaeraasen JE Meager JJ Karlsen Ø Hutchings JA Fernö A Extreme spawning site fidelity in Atlantic cod ICES Journal of Marine Science 2011 68 1472 1477 10.1093/icesjms/fsr055 Steinset, T. A. 2018. Det har gått i bølger. SSB Analyse 2018/13: Norsk fiskeri gjennom 80 år. Retrieved 20 December, 2019, from https://www.ssb.no/jord-skog-jakt-og-fiskeri/artikler-og-publikasjoner/det-har-gatt-i-bolgjer. Steneck RS Vavrinec J Leland AV Accelerating trophic level dysfunction in kelp forest ecosystems of the Western North Atlantic Ecosystems 2004 7 323 332 10.1007/s10021-004-0240-6 Steneck RS Leland A McNaught D Vavrinec J Ecosystem flips, locks and feedbacks: The lasting effects of fisheries on Maine’s kelp forest ecosystem Bulletin of Marine Sciences 2013 89 31 55 10.5343/bms.2011.1148 Stien A Halvorsen O Leinaas HP Skjoldal HR Hopkins C Erikstad KE Leinaas HP No evidence of Echinomermella matsi (Nematoda) as a mortality factor in a local mass mortality of Strongylocentrotus droebachiensis (Echinoidea) Ecology of Fjords and Coastal Waters 1995 Amsterdam Elsevier 585 592 Strand, H. K. 2019. Porsangerfjorden 2.0. IMR Report, 2019-7 (in Norwegian with English abstract). Svedäng H Stål J Cardinale M Consequences of subpopulation structure on fisheries management: cod (Gadus morhua ) in the Kattegat and Öresund (North Sea) Reviews in Fisheries Science 2010 18 139 150 10.1080/10641260903511420 Svedäng H Barth JMI Svenson A Jonsson P Jentoft S Knutsen H André C Local cod (Gadus morhua ) revealed by egg surveys and population genetic analysis after longstanding depletion on the Swedish Skagerrak coast ICES Marine Journal of Marine Science 2018 76 418 429 10.1093/icesjms/fsy166 Tam JC Link JS Large SI Bogstad B Bundy A Cook AM Dingsør GE Dolgov AV A trans-Atlantic examination of haddock Melanogrammus aeglefinus food habits Journal of Fish Biology 2016 88 2203 2218 10.1111/jfb.12983 27145075 Teagle H Hawkins SJ Moore PJ Smale DA The role of kelp species as biogenic habitat formers in coastal marine ecosystems Journal of Experimental Marine Biology and Ecology 2017 492 81 98 10.1016/j.jembe.2017.01.017 Teichert N Lepage M Sagouis A Borja A Chust G Ferreira MT Pasquaud S Schinegger R Functional redundancy and sensitivity of fish assemblages in European rivers, lakes and estuarine ecosystems Scientific Reports 2017 7 17611 17622 10.1038/s41598-017-17975-x 29242556 Yaragina NA Aglen A Sokolov M Jakobsen T Ozhigin K Cod The Barents Sea —Ecosystem, Resources, Management 2011 Trondheim Tapir Acadademic Press 224 270