TY - JOUR
AU - M, Brander, Keith
AB - Abstract The title of this paper is intended in the sense of both “seeing through things”, which requires critical, reflexive thinking and also in the sense of “seeing things through”, which requires tenacity and patience. I address some of the concerns that I have about how we think through, structure, and write about science and I introduce three major research areas that I have been involved in: (i) studies of population dynamics of fish in the Irish Sea that lead to work on plankton dynamics and marine ecosystems, but with the problems of marine policy and management in mind, (ii) the ICES/GLOBEC Cod and Climate Change programme, and (iii) global marine food production and the work of the IPCC. Introduction and acknowledgements Why are most scientific papers boring? We are rarely encouraged to write interesting papers; indeed many scientists and editors seem to believe that proper, serious science requires papers that are boring and barely understood, even by fellow specialists and certainly not by a wider public. Papers are written to present an argument and evidence in a carefully structured form, nevertheless, they are telling a story. It is hard to write that story in an interesting way, when confined by the kind of language, rigid structure, and impersonal style that most journals enforce. Poor arguments and self-delusion can persist in spite of a style of writing, which is supposed to safeguard “scientific rigour”. One of the joys of being asked to write a reflective piece like this, is that one can adopt a broader, more personal viewpoint, while struggling to avoid the creeping pomposity that us older scientists are prone to. However, much we may like to think that we are exploring and describing “reality” in an objective, neutral way, that does not simply reflect our viewpoint (as it is framed by our gender and belief systems), these inevitably play a part in our choice of problems that are worth working on and judgements about what constitutes an adequate scientific investigation (Ravetz, 1979). We can strive to be aware of our own particular “box”, to acknowledge our blinkers, and “see through” the flaws and limitations in our own thinking. There is a good discussion of the difference between reflective and reflexive thinking at https://www.researchgate.net/post/What_is_difference_beteween_reflexivity_and_reflectivity. My immersion in marine science began the summer before I went to university (UCNW, Bangor) in 1965, when Jakob Jakobsson employed me as a scientific assistant on an Icelandic research cruise. 50+ years is long enough to be aware of shifts in scientific fashion and perspective and I will write about some of the research areas that have waxed and waned, as changes in societal concerns led to changes in funding priorities. Many of the problems, such as effects of environmental variability on recruitment, that I became interested in as a PhD student, are hardly closer to being solved (Brander, 2009; Houde, 2009). Policy proposals put forward to the EU and FAO in the late 1970s for regional management councils (Brander, 1978), multispecies fisheries management, and ecosystem-based management (Brander, 1977a, b, 1980) are finally being adopted, so patience is a necessity if you want to see through ideas to fruition. There are many reasons why fisheries management has changed so slowly. One is that framing acceptable regulations (e.g. a catch quota regime) and getting international political agreement on legislation (e.g. the European Common Fisheries Policy) is a tough task and legislators are, not surprisingly, reluctant to change things quickly. During my PhD research at the Fisheries Laboratory, Lowestoft, on the population dynamics of cod in the Irish Sea, I quickly learned from discussions with fishermen that people often have radically different perspectives on the world. With my freshly acquired knowledge of fish population dynamics I would zealously expound the consequences of overfishing and the benefits of catch limits, but it takes more than this to persuade a fisherman of the long-term benefits of reducing his fishing effort, when he knows that competitors may not play by the rules and he has a mortgage payment on his boat due at the end of the month. We fisheries scientists might have made better progress if our training had included a grounding in micro-economics and social theory. In science, one never stops learning and being surprised, and it is a special pleasure to acknowledge all the colleagues who have made me change my mind, against stubborn resistance. David Cushing, was a major figure at Lowestoft in 1969 when I began there, and his unique blend of scientific insight, humour, hospitality, consideration for junior scientists, and rudeness were a great influence that I still miss. Bob Dickson took me for a pub lunch the day I arrived there and we carried on lunching almost every day for the next 20-odd years, while chewing over the latest issues in oceanography, climate science, and marine ecology. John Gulland hired me for two demanding and interesting contracts with FAO that broadened my experience of world fisheries issues and introduced me to some of the other authors in this series. My grasp of modelling and maths has always relied on help from more technically proficient colleagues, in particular John Shepherd, Bill Silvert, Bob Mohn, and Ken Haste Andersen. As well as publishing a number of papers together, we also spent many enjoyable evenings singing or playing music. Acknowledgements almost inevitably leave out those who most deserve it, so let me also thank all those who ought to be mentioned, but aren't. Science and philosophy Grammar is the shared system of rules that enables us to generate language, yet only a small proportion of the world population is aware that grammar exists—we just learn to speak and rarely delve into grammatical rules that we take for granted. Science is somewhat similar. Philosophers try to construct a rational framework for science, while sociologists look for the system of rules that scientists follow in generating scientific knowledge, but scientists just get on with their research, with little interest in the philosophy or sociology of science. The currently dominant view among scientists is that science proceeds by testing and falsifying hypotheses (e.g. Ellis and Silk, 2014). Most funding agencies and journal editors regard testable hypotheses as a necessary prerequisite for a research proposal to be funded or a paper published; falsifiability demarcates science. Two colloquia in London in 1965 and 1966 (I attended the latter) fired my continuing interest in the philosophy of science. They brought together the leading figures in the philosophy of science (Lakatos and Musgrave, 1970), many of whom contested the falsificationist view of science and pointed out its shortcomings. The reasons why falsificationism is inadequate as a method, as a demarcation criterion or as a solution to the problem of induction have been apparent to philosophers for at least 50 years (Lakatos and Musgrave, 1970), but in my experience scientists seem largely unaware or uninterested. This will no doubt provide fertile ground for students of philosophy, epistemology, and sociology for many years, but why should scientists be concerned? In my view, the dominant falsificationist view, while strongly held among scientists, remains naïve, and this has real and often pernicious consequences—would it be provocative to call it scientific fundamentalism? Observations depend on perception, instrumentation, and a background of theory (e.g. when I use a thermometer I assume that the theory of expansion is correct), therefore observations are at best not false. How can one falsify a hypothesis with an observation that is itself only not false? When we test a hypothesis it is always within the context of common theories, background knowledge, and more-or-less hidden assumptions. This means that not only the actual hypothesis, but also the background and assumptions, are being tested—indeed often the background is more interesting than the foreground, as I will try to show later. Hypothesis testing is undoubtedly a very important part of scientific method, but too often one reads hypotheses that are badly thought out or explained. For example, I have been asked more than once to review research proposals and papers that propose to test the hypothesis that latitude affects fish (distribution, metabolism, growth) but which do not specify the process involved. Am I expected to assume that latitude is a proxy for temperature or light regime or magnetic field or something else? How often do editors and funding agencies “see through” the hypothesis and ask by what process latitude is affecting fish? Can anyone send me an example where a funding agency really paid a scientist to carry out the same experiment at five different latitudes in order to test the effect (independently of temperature or light regime) on larval fish metabolism, growth, and mortality? I have read papers claiming to test the hypothesis that latitude affects fish, but have never seen one that specified what the latitude was actually doing to the fish. In case anyone thinks latitude is just a lazy but useful proxy for temperature, it’s not; for example the relationship between latitude and mean bottom temperature for the area inhabited by the 20 major cod stocks of the North Atlantic has a coefficient of determination (r2) of 0.01, i.e. there is no relationship (although you might be able to come up with regions where the fit is better). The hypocrisy of falsificationism becomes apparent when, for example, the distribution of a small number of species is observed to have moved equatorwards during the recent decades of global warming, thus apparently falsifying the hypothesis that warming will cause distributions to shift polewards. Do we reject the hypothesis? No, we look at the hidden assumptions (what are the processes? which way did the isotherms shift? did the species move into deeper, cooler water? is there something odd about these species?). Saving the hypothesis like this makes good sense and can be thought of as a search for the best current explanation, in which hypothesis testing plays a part. One of the main characteristics of biological hypotheses and models is that they are incomplete. Progress in understanding depends on finding out in what ways they are incomplete and when this incompleteness matters, not whether the hypothesis is true or false. The lesson is don’t throw out a hypothesis just because it was falsified under the conditions that you set out in your research proposal—it may be worth saving. Statistical testing adds another layer of fallibility to falsificationism (Anderson, 2010). The mantra of hypothesis testing has pernicious effects; for example, during the late 1980s the Natural Environment Research Council (NERC) in the United Kingdom decided that hypothesis testing should be used to demarcate real science from mere data collecting and that they would only fund hypothesis-driven research. One of several programmes to be given the chop under this clear-cut but insane criterion was the Continuous Plankton Recorder (CPR) programme, an outstanding long-term biological monitoring programme in marine science. Luckily Bob Dickson and I were in a position to secure sufficient funding to keep the programme running until sanity was restored and it rapidly became the jewel in the NERC marine monitoring crown (Brander et al., 2003). A similar situation arose with a US GLOBEC Georges Bank program that I reviewed in the 1990s, when the evaluation criteria stipulated that hypothesis-driven process studies scored higher than the (non-hypothesis driven) cruises needed to provide them with data. After the successful projects were announced it became obvious, even to the fundamentalists, that the process studies were not viable without data, so the funding for the cruises had to be clawed back from all the funded projects. The rationale for carrying out long-term monitoring can be cast in terms of hypothesis testing, but it is evident that well designed and carefully maintained observation programmes can have intrinsic value in giving us a picture of our changing world. They can generate new ideas and unexpected payoffs. Restricting science to a production-oriented straitjacket of hypothesis testing risks losing important discoveries and information about our world. This is true of the CPR programme, which has proved its value in relation to major societal concerns (eutrophication, biodiversity, climate change impacts, and marine litter, most recently plastic particles) that were unknown or did not exist when the survey was being designed in the 1920s. The classic example of unexpected payoffs is the discovery of black body radiation from the Big Bang (Penzias & Wilson, 1965). (Did Einstein really say “If we knew what we were doing it would not be called research, would it?”) The process of separating foreground (hypothesis) and background (assumptions) probably affects one of the remarkable, but widely ignored, characteristics of science and other human activities, namely over-confidence in predictions or projections. We forget to add in the uncertainties due to the assumptions when assessing confidence or risk (e.g. if tsunami waves are never higher than 4 m then Fukushima is safe). It seems that with the exception of weather forecasters and the insurance industry (whose forecasts come back to hit them if they get it wrong), we are poor at judging the reliability our own projections. I recommend Daniel Kahneman’s funny and perceptive article about financial advisors and how highly they rate and reward their own skill at selecting good investments (you can match their skill with a blindfold and a pin) (Kahneman, 2011). Lessons from the Irish Sea Two years into my PhD on the population dynamics of cod in the Irish Sea I was offered a civil service job at Lowestoft to carry out stock assessments of all fish species in the Irish Sea and to devise a strategy for managing them (Brander, 1977a). The Irish Sea is a fascinating place to study fisheries and fish population dynamics; small enough to carry out a research survey of fish and plankton in about a week, but very varied in bathymetry, substrate type, tidal mixing, and freshwater influence. The diversity of the fish fauna is high, but commercially the most valuable species is Dublin Bay prawn (Nephrops norvegicus), with other shellfish species such as scallops (both Chlamys and Pecten) also prominent. In the 1970s, the Irish Sea supported large but declining populations of several species of rays and other elasmobranchs. Trawl and seine fisheries in the Irish Sea were seasonally targeted at different species, and the landings from a single fishing trip or even a single haul usually included ten or more species (Brander, 1980). Diversity within catches and the technical and biological interactions between species raise a number of tricky management problems: (1) the level of scientific effort required to sample and assess all of these species is hard to justify for such small stocks, (2) regulations designed to optimise sustainable exploitation of one species often have undesirable consequences for other species in the mixed fishery, and (3) it is not obvious how to define and quantify single species Maximum Sustainable Yield (MSY) when the species interact biologically, so other management objectives need to be defined and agreed (Brander, 1988). A particularly serious example of unintended harm was the disappearance of the common skate (Raia batis, but now known to be two separate species) from the Irish Sea during the twentieth century. In the early 1900s, this species was frequently caught in many fisheries, but by the late 1970s, after a decade of regular fishing surveys in all parts of the Irish Sea, I had still never seen one, and realised that it had become locally extinct. I wrote this up and took it to Arthur Lee, the director of the Lowestoft Laboratory, who told me to send my paper to Nature (Brander, 1981) as there had just been a public outcry about the disappearance of the Large Blue butterfly in England and he thought that a similar conservation story about a marine species should be given prominence. The Times newspaper wrote an article about it and I did a radio interview, but although publicity can help with conservation issues, we are still struggling with how to protect large, long-lived, vulnerable species such as the common skate, while allowing fishing to continue for other species in the areas where they co-occur (Dedman et al., 2016). The story of the Large Blue butterfly gives grounds for optimism about conservation efforts. It became extinct in England by 1979, but has since been successfully re-introduced (https://butterfly-conservation.org/679-899/large-blue.html). An ICES assessment group for the Irish Sea (Figure 1) carried out multispecies assessments that explored the consequences of biological interactions between species and technical interactions between fisheries (ICES, 1978). We put forward a number of management proposals, including restrictions on fishing effort and a limit on total demersal fish catches. For several years in the 1980s, the agreed Total Allowable Catch (TAC) for cod in the Irish Sea took into account the predation by cod on Nephrops; the models projected that high cod catches would reduce the predation mortality and result in higher catches of Nephrops, and the joint revenue for the two species, which were largely caught by the same fishing fleets would increase (Brander and Bennett, 1986). This is one of the few examples where fisheries management has deliberately attempted a form of ecological engineering. Fisheries have always been an inadvertent and unplanned experiment in ecological engineering, so my response to those who question whether we should be attempting ecological engineering is that we already do so and the real choice is between thinking it through (e.g. ecosystem-based fisheries management—EBFM) or pretending it is not happening. Figure 1. View largeDownload slide The Irish Sea and Bristol Channel Working Group harmonizing a new approach to fisheries management. The artist is Joop de Veen (also on tuba). The other members are Iwan Davies (clarinet), David Bennett (guitar), John Pope (drums), Jacques Guégen (trumpet), Mike Sissenwine (trombone), Paul Hillis (piano), David Griffith (alto sax), Terry Hutley (double bass), Rudy de Clerck (tenor sax), and me (clarinet and also violin, while kicking Schaeffer the dog, who is peeing a perfect parabola in the corner). Figure 1. View largeDownload slide The Irish Sea and Bristol Channel Working Group harmonizing a new approach to fisheries management. The artist is Joop de Veen (also on tuba). The other members are Iwan Davies (clarinet), David Bennett (guitar), John Pope (drums), Jacques Guégen (trumpet), Mike Sissenwine (trombone), Paul Hillis (piano), David Griffith (alto sax), Terry Hutley (double bass), Rudy de Clerck (tenor sax), and me (clarinet and also violin, while kicking Schaeffer the dog, who is peeing a perfect parabola in the corner). The metrics that are widely applied in managing fisheries and ecosystems could benefit from “seeing through”. Since fish exist within predator-prey systems, the yield of each species depends on the abundance of its predators and prey (Brander and Mohn, 1991); MSY is defined only if several shaky assumptions are made (principally that natural mortality, growth, and the relationship between spawners and recruits are all fixed). The continuing reliance on MSY as a target for sustainable fisheries management and international agreements, such as the Johannesburg Declaration, shows the attraction of misleadingly simple concepts. MSY is quite easy to explain and appears to provide a neutral “scientific” objective for achieving sustainable fisheries. The enthusiasm for enshrining MSY as an objective in management policies and international agreements comes mainly from politicians and negotiators, but behind the simple “three-letter panacea” (Mesnil, 2012) lurks a panoply of compromises and trade-offs over how much protection is to be given to vulnerable species and ecosystems, which groups of fishermen are to benefit, and what levels of risk are acceptable in balancing yields and variability in stocks. An example of this messy process could be seen when establishing the new EU Management Strategy Framework Directive, where scientists wrote that the choice of how to set biomass targets was a policy decision and policy makers replied that more scientific work was needed (EU Council and Presidency, 2010)! Recent work with Ken Haste Andersen and Lars Ravn-Jonsen uses a size and trait-based model to investigate the definition and dynamics of yields and rent in a multispecies system with conservation constraints (Andersen et al., 2015). The aim was to explore what management policies come closest to fulfilling the often conflicting objectives of maximizing food production, while protecting biomass of vulnerable groups. It is encouraging to see that similar models (Mackinson et al., 2018) are now being used to explore multiannual management plans for the North Sea and that these North Sea models share a number of the structural features (cod–Nephrops interactions, the role of whiting, and how to deal with “choke” species) that we tried to formulate in the Irish Sea analysis 30 years ago. Fishing is of course not the only pressure that humans put on marine ecosystems. The 14th European Marine Biology Symposium in 1979 had “Protection of Life in the Sea” as its title and began with a deeply pessimistic statement from the chairman, Otto Kinne: “ecological research and its wise application have become the basic prerequisites for civilized human life on this planet to continue beyond the next few decades or centuries. I think that most of us would agree that our chances for long-term survival are slim, the difficulties and problems are overwhelming, and our abilities for comprehending complex living systems and for critically applying the knowledge gained are very modest at the most.” A total of 38 papers in the symposium publication were about pollution (by heavy metals, radioactivity, oil, pesticides, industrial and domestic pollution, and other sources), 13 papers were about environmental evaluation and monitoring, and 10 about management of areas, species, and ecosystems. The latter included my paper on “Fisheries management and conservation in the Irish Sea” (Brander, 1980), which was the only paper at the symposium dealing with the need to protect life in the sea from fishing and I ended it with a call for ecosystem-based management. I cherish an astonishingly unperceptive statement in Kinne’s summing up: “It seems that in many cases factors other than pollution determine fish abundance…”! However, scientific concerns and fashions move on and the word “pollution” does not appear even once among the titles for the 52nd European Marine Biology Symposium in 2017. This is not just a random change in scientific fashion; with the benefit of hindsight it seems that Kinne’s pessimism about applying our knowledge was misplaced in this instance, since a great and successful international effort has succeeded in reducing pollution, but by no means eliminating this threat to life in the sea (plastics being the latest worry and of course CO2 should probably be included as pollution). Where fisheries are regulated by single species quotas, managers generally require annual projections for each stock for the year ahead. Carrying out a single species fish stock assessment is in many ways a satisfying form of biological book keeping. The ICES Irish Sea and Bristol Channel group was set up in 1978 to carry out both single species and multispecies assessments (Brander, 1977b). A congenial group of colleagues would meet annually, often in exotic Copenhagen, and work hard together for a week. After some years of this I began to wonder how good our projections were and was surprised to find that nobody had ever done such a retrospective evaluation. It turned out that our projections were not measurably better than telling the managers that the stock in the year ahead will be the same as the average of the last 5 years (i.e. a bit like telling people that the weather tomorrow will be the same as today—a weatherman who did this would probably not keep the job for long) (Brander, 1987). I don’t know whether fisheries projections are better now than they were 35 years ago, but Management Strategy Evaluation methodology has been developed in order to examine not only the reliability of data and assessments, but also the performance of implemented regulations in relation to the objectives. This type of reflexive, empirical approach identifies which parts of the system need to improve, and may be the best way to manage complex biological–social interactive systems like fisheries (Brander, 2003). In 1977, the European Community and the European Commission were establishing their legitimacy (disputed by the USSR) as a political entity with the power to manage the fish stocks in European waters. A small group of scientists was assembled in Brussels to advise on the establishment of a Common Fisheries Policy. We advised that limiting the level of fishing effort (number and power of fishing boats) should be the main instrument of management, but were told that although this was undoubtedly a good idea, it would have to wait, because political considerations dictated that member states would negotiate about tons of fish, not about fishing effort and we therefore had to devise a TAC regulation and to quantify the sharing of fish catches by creating a key derived from historic catches. Based on my experience with the Irish Sea I presented a proposal to Eamon Gallagher, the first Director General of the EU Fisheries Directorate, for a radically different approach to management using a cooperative regional management structure. The proposal was founded on the view that problems of implementing sustainable management of fisheries have more to do with how to develop effective, acceptable institutions and regulations than with improving stock assessments. This proposal was also impaled on the spike of political expediency, but many of its features [which were published in outline in an FAO paper on the effects of extending fisheries jurisdictions (Brander, 1978) and subsequently in Shepherd et al., (1985)] are now being revived and I am optimistic that within the next 20 years most of the suggestions that I made to Gallagher in September 1977 will have been adopted. Having become dissatisfied with the adequacy (skill) and value (implementation) of stock assessments, I was fortunately able to shift to a more research-oriented role at Lowestoft, investigating the causes of interannual variability in recruitment and differences in fish productivity (Brander and Dickson, 1984). Plankton dynamics plays a critical role in both recruitment and fish productivity, but the processes are complex and have proved very difficult to understand (Brander et al., 2001). I think there are at least two great problems in linking fish dynamics to plankton processes. One is that critical events happen at very fine space scales (often at ocean fronts or vertical discontinuities) and short time scales (timing of spring blooms over days to weeks), therefore it is difficult to carry out field studies. The second is that factors and processes that govern recruitment and productivity of a particular population are not necessarily the same from year to year. Identifying periods when density dependence may occur is vital to studies of recruitment and here some work on the early life of cod larvae in the Irish Sea shows how rapidly the dynamics can change as the developing larvae interact with their planktonic food, but also with competing species of larvae, suggesting that multispecies density dependence is critical (Thompson and Harrop, 1991). Sampling plankton and resolving distributions and interactions at relevant scales is a very tough challenge (Brander and Thompson, 1989; Nichols and Brander, 1989; Thompson & Harrop, 1991; Brander et al., 1993) and for a few years in the early 1990s I was fortunate to be provided with ship time, great colleagues, and a decent budget to study biota from picoplankton to fish. We had lengthy cruises in the Irish Sea during the period of the spring plankton bloom with scientists from Plymouth, Menai Bridge and a team from the United States, using everything from particle counters, flow cytometers, and 21 frequency echosounders to a range of profiling multinet plankton samplers and fishing gear (Nichols et al., 1993; Coombs et al., 1994; Burkart et al., 1995; Prestidge and Taylor, 1995; Conway et al., 1997). I commissioned the first Scanfish undulating wing, with optical plankton counter, fluorimeter, CTD, etc., and a data acquisition capability that resolved spatial detail about two orders of magnitude greater than we had previously been capable of (Brown et al., 1996). It took years to work up the data and the amount of information was overwhelming. One of the lessons that I learned was that the pattern of distribution of biota at scales from picoplankton to fish larvae is coherent with the pattern of physical and chemical changes in the water at vertical scales of cms and at horizontal scales not much more than that. This is fascinating, but also mindblowing, when you are trying to integrate and synthesize the information up to scales that are relevant to fisheries management. By 1994 staff changes at Lowestoft pushed me into taking a job as government Fisheries Science Adviser in London, but I think I had probably also realised that making scientific headway on the plankton dynamics influencing fish recruitment was beyond my scientific capability. This was also when the UK GLOBEC programme got underway and a decision was made to fund a North Atlantic study of Calanus rather than the Irish Sea research that I had initiated. Sometimes you have to accept that you are losing battles and that it is better to go away and perhaps return later with stronger allies, better weapons, and fresh enthusiasm. Demersal fisheries productivity (as measured by catch per unit area) in the Irish Sea is roughly half of that in the North Sea (Brander & Dickson, 1984). Pelagic fish productivity is also about half, although comparisons are tricky because in many years the catch locations of pelagic fish, particularly mackerel, have been misreported. Various hypotheses were put forward to account for fisheries productivity differences, but the most likely explanation is lower plankton productivity, which in turn is mainly due to differences in both the horizontal and vertical transport of nutrients. The horizontal component affects the large-scale supply from the adjacent ocean and the vertical component, due to differences in bathymetry and tidal mixing, affects the plankton bloom dynamics. I don’t think we are yet confident that we understand the causes of differences in fisheries productivity, such as between the North Sea and the Irish Sea and this has always worried me for several reasons. One is that it reveals the gulf between our understanding (and control) of the processes determining marine food production compared with terrestrial food production. A second is that this gulf in understanding does not seem to generate much scientific interest. A third is that there is now great interest in projecting future fisheries productivity under climate change which, according to the Fifth IPCC report (Pörtner et al., 2014) we can do with a high degree of confidence. Since the models being used for the IPCC projections have not been tested to see whether they can reproduce either the temporal history or the geographic differences in fisheries productivity that are typical of shelf areas such as the North Sea and the Irish Sea, the IPCC projections seem to me like a classic example of overconfidence; I will return to this issue. I remember a wise, but not particularly successful, old economist explaining to me that to be a really successful economist you had to be clever enough to understand the models and stupid enough to believe them. One of the aims of this series of articles is to distil pearls of wisdom from retrospection on a scientific career, so I offer some further thoughts on models, starting with a definition from Solomon (not the Old Testament king, the ecologist) that I have always cherished. He defined a model as a representation in simplified or metaphorical form of aspects of a functioning system, and he added that “they are devices to assist thinking”. This makes it clear that we should regard models as tools; they may fail to help us to understand if they are badly explained, too complicated, wrong or because we are too stupid (or lack the necessary logical or mathematical skill). My advice is to be honest if you don’t understand a model and to avoid using (applying) that model until you do understand it, because until then you will be unaware of its shortcomings and limitations. Many models [e.g. relating vertebral count to temperature (Brander, 1979)] are phenomenological rather than “mechanistic”, i.e. underlying processes are implicit rather than explicit. My advice is not to trust them until you are satisfied that there are reasonable implicit processes that can explain cause and effect and that the data are of sufficient quality to earn your trust. A remarkable example of an old and widely cited phenomenological model that has recently been called into question by carefully reviewing the data is Bergmann’s Rule, which states that individuals within a species are smaller in warmer environments (Riemer et al., 2018). The ICES/GLOBEC Cod and Climate Change programme The origins of GLOBEC (Global Ocean Ecosystem Dynamics) and of the Cod and Climate Change (CCC) programme go back to a series of workshops in the late 1980s initiated by Brian Rothschild (Rothschild, 2015). The aim of GLOBEC, which ran as a major international cooperative programme through to 2010, was no less than to understand the structure and functioning of marine ecosystems. CCC, which ran from 1990 to 2008, was the first regional component of GLOBEC and was designed to show how variability in plankton affects the recruitment and population dynamics of the very well-studied North Atlantic cod stocks. The ICES Cod and Climate Change Symposium in 1993 resulted in a publication containing 60 papers, which gives a measure of the wealth of information about cod and its response to environmental variability, on which CCC was able to build. The papers from this symposium also illustrate the sense in which “climate” was being used by the contributors in the early 1990s. It was rarely used in the sense of anthropogenic climate change, but rather to mean past variability in temperature, salinity, windfields, vertical mixing at scales from inter-annual to centennial and how this variability affected distribution, recruitment, growth, biomass, and resilience of cod under exploitation. One paper addressed the anthropogenic component and concluded from an analysis of 1400 years of Fennoscandian tree rings that summer temperature would rise by 0.9–1.5°C from 1990 to 2030, but that we would not be confident of the cause of this increase (natural or anthropogenic) until after 2020 (Dickson et al., 1994). None of the papers assessed current effects of anthropogenic climate change or projected future impacts. Nowadays the term “climate change” is primarily used in relation to global warming rather than natural variability. CCC ran until 2010 and by the end of the programme its emphasis had shifted away from the effects of interannual climate variability and onto decadal variability and anthropogenic climate change impacts. The number of papers published on marine climate impacts has been doubling roughly every 5 years since 1990 (Brander et al., 2013). The second, related, change in direction that took place over the lifetime of CCC, was a shift away from the “recruitment problem”, which investigates survival during early life of cod and the factors determining year-class strength, towards more general investigations of changes in growth, reproductive output, distribution, mortality, and prey species caused by climate change (Houde, 2009). Twelve enjoyable years (1996–2008) working at ICES as coordinator of the CCC programme, organizing and facilitating 17 workshops, 9 theme sessions and 3 symposia, convinced me that cooperative, cross-disciplinary studies add much value to scientific programmes. The reasons and evidence for this, as well as information about the structure and planning of the CCC programme, are set out in a chapter of a book on “Atlantic cod: the bio-ecology of the fish” edited by George Rose (Wiley, in press), so I will not repeat them here. By 2008 the funding for my coordinator post at ICES, which came from several different countries, had been insecure for some time. When ICES decided that they no longer wanted to support it I was very fortunate to be invited to move to DTU, Aqua, at the time located in Charlottenlund Castle, just north of Copenhagen. I thus began my career as an academic at the age of 61 and still have an emeritus position there. Marine food production and world food security Concern about climate impacts on marine systems and food production from the sea has been growing rapidly since the first IPCC report was published in 1990 (Tegart et al., 1990) and it is remarkable how much we have learned since then. In the 1990 IPCC report a brief section on fisheries cited analogies with past periods of warming (such as 1925–1945 in the Northern Hemisphere) as providing the most reliable evidence of likely future effects on fish stocks. The second IPCC report in 1995 (Bruce et al., 1995) included much more detail and concluded: Globally, overfishing and diverse human stresses on the environment will probably continue to outweigh climate-change impacts for several decades. However, fishing impacts are usually reversible—whereas climate-change and habitat-loss impacts are not (at least within the IPCC scenario)—and the overfishing problem may well be solved within the time horizon of this IPCC assessment. We would probably agree with this judgement and there are even welcome indications that their optimism about solving the overfishing problem was not totally unfounded (Brander et al., 2018). The 1995 IPCC report concluded that: “In some cases, species may move poleward, but there may be little change evident to someone living in the middle range of a given ecosystem.” Such poleward movement of fish species was in fact already occurring at the time; for example annual records showed two sub-tropical species of dory extending their distribution 1000 km northwards along the European shelf edge between 1963 and 1995 (Brander et al., 2003). The changes in distribution of both rare and common species were so striking that in 2004 the whole front page of one of Denmark’s national newspapers featured an article about new fish species in Danish waters and such evidence and public interest, continues to accumulate. The third IPCC report in 2001 (McCarthy et al., 2001) agreed with previous reports that “future saltwater fisheries production is likely to be about the same as at present, though changes in distribution could affect who catches a particular stock.” The report describes large changes in fisheries due to multidecadal climate variability and regime shifts and comments that the “key to understanding the direction of change for world fisheries is the ability to incorporate decadal-scale variability into general circulation models (GCMs).” The increasing role of aquaculture in meeting world demand for fish was noted as were likely interactions between aquaculture and wild capture production. A workshop on Detection and Attribution in New York in 2003 was my first direct involvement with the IPCC and it introduced me to several of the major figures in climate change research and gave me the opportunity to see a magical production of Rameau’s Les Boreades performed by the Paris Opera. Coverage of marine systems and fisheries in the fourth IPCC report (Parry et al., 2007) was spread over three main chapters (1, 4, 5), plus several regional chapters (e.g. 15 Polar regions, 16 Small Islands) and special topics (Coral reefs, Megadeltas), which made it hard to find and also gave the mistaken impression that the biology of the oceans was neglected. I learned a huge amount from my fellow authors of the “Food, Fibre and Forest Products” chapter (Easterling et al., 2007) about research in agriculture, forestry, land-use, and animal husbandry. Terrestrial biological science is based on thousands of years of farming experience and universal reliance on controlled experiments, whereas the sea is an unfamiliar place, capture fisheries are like blindfolded hunting and almost no experiments are possible. One of our meetings was held at a research station near Clermont-Ferrand in France, where field experiments on the effects of temperature and CO2 change were being carried out under controlled conditions. Thousands of similar experiments have been performed worldwide on the main food crops (maize, wheat, and rice) and also on forests, to give a huge body of empirical information about likely future yields. Based on these, our chapter produced a world map of expected impacts of climate change on regional yields of crops, livestock and forestry by 2050. I was asked to provide similar projections for regional fisheries yields, but declined as we had no reliable projections and I thought that making educated guesses based on the expected global changes in primary production (increasing at high latitudes and declining at low latitudes) would give a mistaken impression of our state of knowledge. The best global projections of new primary production (NPP) indicated an increase of 0.7–8.1% by 2050 with ΔTglobal ∼1.5–3°C (Sarmiento et al., 2004), however the list of caveats in trying to turn such NPP figures into projections of fisheries yields was long and daunting (Brander, 2015). In the circumstances it seemed best to be modest about our ability to make fisheries yield projections, as there was still a great deal to be modest about. The fifth IPCC report (IPCC, 2014) included a huge chapter on the oceans (Pörtner et al., 2014), but with fisheries systems again dealt with in several other chapters as well. Unlike the previous reports, this one provided such confident and detailed projections of future potential fisheries yields that the global map showing them was used extensively in launching the report and in media coverage (http://www.bbc.com/news/science-environment-26814742). The new projections used the same NPP projections as the 2007 report, but added an oxygen- and capacity-limited thermal tolerance (OCLTT) model together with a bioclimate envelope model and claimed that these gave medium to high confidence in the results, which was a higher level of confidence than was claimed for terrestrial crop projections. During the open review process that preceded the publication of the report I questioned whether such high confidence could be justified for a number of reasons and I am still sceptical (Brander, 2015). The global climate models represent ocean dynamics well, but are not yet able to reproduce shelf sea conditions, which is where most global fisheries occur (Bopp et al., 2013). Plankton are still poorly represented, with zooplankton often missing (Mitra et al., 2014) and important processes such as remineralization of phytoplankton remain uncertain. The validity and universality of the OCLTT model used to parameterise physiological processes is being increasingly questioned (Lefevre et al., 2017; Jutfelt et al., 2018). We need to be able to justify our confidence in models and projections more openly and in greater detail than was the case with the fifth IPCC report and that requires setting out the consequences of the background assumptions fully and using more than one modelling approach and one scenario. Paradigms, words and world views Language plays a huge and widely ignored role in how we do science and how we construct and view our world. Certain buzz-words, such as “robust” and “paradigm”, have increased in frequency in scientific papers over the years (Figure 2). “Robust” began it’s scientific career in a specific statistical context, to denote results that remain valid in spite of moderate deviation from underlying assumptions (typically about distribution). It seems to have become the adjective of choice for saying “hey, I did a good job here”, which is fine if you have actually carried out testing to substantiate this, but not if you are just delivering yourself a pat on the back. Figure 2. View largeDownload slide Frequency (occurrence per million) of particular words in all English texts in Google Ngram (about 5 million texts). “Paradigm” has increased by roughly 16-fold in frequency since the mid-1960s; “robust” by over 2.5 times since 1975. Note, however, that Ngram does not separate scientific papers from other texts. Figure 2. View largeDownload slide Frequency (occurrence per million) of particular words in all English texts in Google Ngram (about 5 million texts). “Paradigm” has increased by roughly 16-fold in frequency since the mid-1960s; “robust” by over 2.5 times since 1975. Note, however, that Ngram does not separate scientific papers from other texts. The usage of “paradigm” in science stems principally from the work of Thomas Kuhn (one of the participants at the 1960s London colloquia) who distinguished between what he called “normal” science that solved problems within an accepted framework of theory, ideas, and methods and “revolutionary” science that overthrew the accepted framework and replaced it with a new “paradigm” that entailed a new way of seeing the world, or re-interpreting a field of science (Kuhn, 1962). Revolutionary science is clearly a sexier and more career-enhancing thing to do and, not surprisingly, scientists now frequently use the word and claim that their work represents a paradigm shift. Since Kuhn’s original work used the term “paradigm” in at least 21 different senses, ranging from “analogy” to “world view” (Masterman, 1970), it is not surprising that its meaning has become diffuse and diluted over time. Nevertheless, the increasing frequency with which it is used probably says more about the dilution of meaning rather than a shift in science from being mostly “normal” to being in a constant state of revolution. It would be foolish to propose that words should have a fixed meaning, but I believe we have a responsibility to think carefully about the meaning of the words we choose and should justify the use of strong adjectives, such as “robust”. The discipline of explaining the sense in which we are using words, so that others can understand and not be misled, helps us to avoid misleading ourselves. As a final thought about world views and changing scientific fashions, I spent much of my career helping to build the case for the inclusion of environmental factors (including climate) in fisheries assessments and management strategies, but have recently begun to warn against overdoing it! The CCC programme tried to get stock assessment scientists to include environmental effects, but resistance was strong and we largely failed for a variety of reasons, some of which were good. We acknowledged that environmental factors should only be included if they could be shown to have a significant effect, if the process by which this happened was reasonably well understood and if the required environmental data (e.g. mean temperature, climatic index) were available on time and at a reasonable cost. Research on environmental drivers and climate change has become a huge topic in science and sometimes dominates the public discourse about sustainability, impacts, and adaptation to the extent that the role that fishing plays in determining the fate of fish stocks risks being overlooked. The COP15 meeting in Copenhagen in 2009 was the point in time when I sensed that the issue of climate change shifted from being an inconvenient truth to being a convenient scapegoat; environmentalists from all over the world came, proclaiming that their problems with declining fish stocks, loss of forests, depleted water resources, coastal degradation, and many more, were attributable to climate change, which in a few cases may have been true. For fisheries problems it was evident that overfishing and damage to habitats (e.g. due to increased sediment outflow from de-forested areas) were far more immediate and locally tractable drivers than climate and that over-attribution to climate meant that these drivers were overlooked. The equity issues that need to be tackled in relation to climate, which are the main stumbling block in reaching effective agreements on climate action, became overloaded by trying to blame climate change for all the ills of the world. There are signs that climate change science has been distorted by over-attribution, as research funding has poured in and new journals have proliferated. A personal story, that could probably be echoed many times, concerns a neighbour, a Malaysian student who just got her PhD for work on intercropping in tropical agriculture. Her supervisor has money for post-doctoral research, but it has to be spent on climate change, which is not her topic. Of course, climate impacts are a pervasive and immediate concern and we need to ramp up effective mitigation as rapidly as possible, but unfortunately the old pressures have not gone away and still need to be tackled. For fisheries these include overcapacity, overfishing, habitat degradation, pollution, introduced species, and loss of biodiversity and tackling these will incidentally also help to make exploited species more resilient to climate impacts. A paper attributing the collapse of the Gulf of Maine cod stock to climate change, published in Science recently (Pershing et al., 2015) and widely reported in public news media (Lavelle, 2015), is a striking example of the prominence that the climate narrative can achieve. It is not surprising that scientists want to write about climate impacts and that the major high profile journals are eager to publish on the topic, but the dangers of over-attribution are real. The recovery of the North Sea cod stock provides a fascinating contrast with the collapse in the Gulf of Maine stock and implicates fishing rather than climate as the principal (but probably not only) driver of observed biomass changes in both cases, since their temperature history over the past three decades has been very similar (Brander, submitted). Three words and concepts that seem to recur with increasing frequency in texts on sustainability and environmental management are attribution, equity, and accountability. A recent invitation to contribute a book chapter on accountability in marine governance might even call on the professional expertise of my whole family, with sections on biology and management of natural resources; valuation and principles of accounting for ecosystem goods and services; rights, international agreements and legal frameworks and training and games dealing with rights, equity, property, and the environment. Since one cannot be an expert in all these fields I recommend an alternative, which is to marry and raise them, but it takes about 40 years. Footnotes †Food for Thought articles are essays in which the author provides their perspective on a research area, topic, or issue. They are intended to provide contributors with a forum through which to air their own views and experiences, with few of the constraints that govern standard research articles. This Food for Thought article is one in a series solicited from leading figures in the fisheries and aquatic sciences community. The objective is to offer lessons and insights from their careers in an accessible and pedagogical form from which the community, and particularly early career scientists, will benefit. The International Council for the Exploration of the Sea (ICES) and Oxford University Press are pleased to make these Food for Thought articles immediately available as free access documents. References Andersen K. H. , Brander K. , Ravn-Jonsen L. 2015 . Trade-offs between objectives for ecosystem management of fisheries . Ecological Applications , 25 : 1390 – 1396 . Google Scholar Crossref Search ADS PubMed Anderson T. R. 2010 . Progress in marine ecosystem modelling and the “unreasonable effectiveness of mathematics.” Journal of Marine Systems , 81 : 4 – 11 . Google Scholar Crossref Search ADS Bopp L. , Resplandy L. , Orr J. C. , Doney S. C. , Dunne J. P. , Gehlen M. , Halloran P. , et al. 2013 . Multiple stressors of ocean ecosystems in the 21st century: projections with CMIP5 models . Biogeosciences , 10 : 6225 – 6245 . Google Scholar Crossref Search ADS Brander K. M. 1977a . The management of the Irish Sea fisheries—a review . Laboratory Leaflet, MAFF Directorate of Fisheries Research, Lowestoft , 36 : 40 . Brander K. M. 1977b . The assessment of Irish Sea and Bristol Channel fisheries . ICES Paper Report CM 1977/F: 26. 7 pp. Brander K. M. 1978 . The effect of 200 mile limits on fisheries management in the Northeast North Atlantic. FAO Fisheries Technical Paper . Rome , 183 , p. 19 . Brander K. M. 1980 . Fisheries management and conservation in the Irish Sea . Helgolander Meeresunters , 33 : 687 – 699 . Google Scholar Crossref Search ADS Brander K. M. 1981 . Disappearance of common skate Raia batis from Irish Sea . Nature , 290 : 48 – 49 . Google Scholar Crossref Search ADS Brander K. M. 1987 . How well do Working Groups predict catches? Journal du Conseil International Pour l’Exploration de la Mer , 43 : 245 – 252 . Google Scholar Crossref Search ADS Brander K. M. 1988 . Multispecies fisheries of the Irish Sea. In Fish Population Dynamics , 2 nd edn., pp. 303 – 328 . Ed. by Gulland J.A. . Wiley, Chichester. Brander K. M. 2003 . What kinds of fish stock predictions do we need and what kinds of information will help us to make better predictions? Scientia Marina , 67 : 21 – 33 . Google Scholar Crossref Search ADS Brander K. 2009 . Summing up—symposium on reproductive and recruitment processes of exploited marine fish stocks . Journal of Northwest Atlantic Fishery Science , 41 : xx – xxiv . Google Scholar Crossref Search ADS Brander K. 2015 . Improving the reliability of fishery predictions under climate change . Current Climate Change Reports , 1 : 40 – 48 . Google Scholar Crossref Search ADS Brander K. Submitted. Sustainable management depends on correct attribution—contrasting the state of Gulf of Maine and North Sea cod . Nature Sustainability . Brander K. M. , Bennett D. B. 1986 . Interactions between Norway Lobster (Nephrops Norvegicus) and Cod (Gadus morhua L.) and Their Fisheries in the Irish Sea . Canadian Special Publication in Fisheries and Aquatic Science , 92 , pp. 269 – 281 . Brander K. M. , Blom G. , Borges, M. F., Erzini, K., Henderson, G., MacKenzie, B. R., Mendes, H., et al. 2003 . Changes in fish distribution in the eastern North Atlantic: are we seeing a coherent response to changing temperature? ICES Marine Science Symposia , 219 : 261 – 270 . Brander K. M. , Dickson R. R. 1984 . An investigation of the low level of fish production in the Irish Sea. Rapports et Procés-Verbaux des R‚Unions du Conseil International Pour l’Exploration de la Mer , 183 , pp. 234 – 242 . Brander K. M. , Dickson R. R. , Edwards M. 2003 . Use of Continuous Plankton Recorder information in support of marine management: applications in fisheries, environmental protection, and in the study of ecosystem response to environmental change . Progress in Oceanography , 58 : 175 – 191 . Google Scholar Crossref Search ADS Brander K. M. , Dickson R. R. , Shepherd J. G. 2001 . Testing the match-mismatch hypothesis on North Atlantic cod stocks, using a 1-D zooplankton production model . ICES Journal of Marine Science , 58 : 962 – 966 . Google Scholar Crossref Search ADS Brander K. M. , Milligan S. P. , Nichols J. H. 1993 . Flume tank experiments to estimate the volume filtered by highspeed plankton samplers and to assess the effect of net clogging . Journal of Plankton Research , 15 : 385 – 401 . Google Scholar Crossref Search ADS Brander K. , Neuheimer A. , Andersen K. H. , Hartvig M. 2013 . Overconfidence in model projections . ICES Journal of Marine Science , 70 : 1065 – 1068 . Google Scholar Crossref Search ADS Brander K. M. , Mohn R. K. 1991 . Is the whole always less than the sum of the parts? ICES Marine Science Symposia , 193 : 117 – 119 . Brander K. M. , Thompson A. B. 1989 . Diel differences in avoidance of three vertical profile sampling gears by herring larvae . Journal of Plankton Research , 11 : 775 – 784 . Google Scholar Crossref Search ADS Brander K. M. , Cochrane, K., Barange, M., and Soto, D. 2018 . Climate change implications for fisheries and aquaculture. In Climate Change Impacts on Fisheries and Aquaculture: A Global Analysis . Ed. Phillips, B., Ramirez, M. Wiley-Blackwell , Chichester . pp. 45 – 63 . Brown J. , Brander, K. M., Fernand, L., and Hill, A. E. 1996 . Scanfish: high performance Towed undulator . Sea Technology , 37 : 23 – 27 . Bruce J. , Lee H. , Haites E. 1995 . Climate Change 1995—Economic and Social Dimensions of Climate Change Contribution of Working Group III to the Second Assessment Report of the Intergovernmental Panel on Climate Change . Cambridge University Press, Cambridge . Burkart C. A. , Kleppel G. S. , Brander K. , Holliday D. V. , Pieper R. E. 1995 . Copepod and barnacle nauplius distributions in the Irish Sea: relation to springtime hydrographic variability . Journal of Plankton Research , 17 : 1177 – 1188 . Google Scholar Crossref Search ADS Conway D. V. P. , Coombs S. H. , Smith C. 1997 . Vertical distribution of fish eggs and larvae in the Irish Sea and southern North Sea . ICES Journal of Marine Science , 54 : 136 – 147 . Google Scholar Crossref Search ADS Coombs S. H. , Robins D. B. , Conway D. V. P. , Halliday N. C. , Pomroy A. J. 1994 . Suspended particulates in the Irish Sea and feeding conditions for fish larvae . Marine Biology , 118 : 7 – 15 . Google Scholar Crossref Search ADS Dedman S. , Officer, R., Brophy, D., Clarke, M., and Reid, D. G. 2016 . Towards a flexible decision support tool for MSY-based marine protected area design for skates and rays . ICES Journal of Marine Science , 74 : 576 – 587 . Dickson R. R. , Briffa K. R. , Osborn T. J. 1994 . Cod and climate: the spatial and temporal context . ICES Marine Science Symposia , 198 : 280 – 286 . Easterling W. E. , Aggrawal, P. K., Batima, P., Brander, K. M., Erda, L., Howden, S. M., Kirilenko, A. et al. 2007 . Food, fibre and forest products. In Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change , pp. 273 – 313 . Ed. by van der L P. J. , Parry C. E. H. M. L. , Canziani O. F. , Palutikof J. P. . Cambridge University Press , Cambridge . Ellis G. , Silk J. 2014 . Scientific Method: defend the Integrity of Physics . Nature , 516 : 321 – 323 . Google Scholar Crossref Search ADS PubMed EU Council and Presidency , 2010 . Challenges for the good environmental status of the marine environment—Information from the Presidency and from the Commission, 10545/10. Houde E. D. 2009 . Emerging from Hjort’s shadow . Journal of Northwest Atlantic Fishery Science , 41 : 53 – 70 . Google Scholar Crossref Search ADS ICES , 1978 . Report of the Irish Sea and Bristol Channel Working Group, Lowestoft. ICES Document CM 1978/G: 6. 84 pp. IPCC , 2014 . Climate change 2014: impacts, adaptation, and vulnerability. Part A: global and sectoral aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change . Ed. by Field C. et al. . Cambridge University Press , Cambridge . Jutfelt F. , Norin, T., Ern, R., Overgaard, J., Wang, T., McKenzie, D. J., Lefevre, S. et al. 2018 . Oxygen- and capacity-limited thermal tolerance: blurring ecology and physiology . The Journal of Experimental Biology , 221 . Available at: http://jeb.biologists.org/content/221/1/jeb169615.abstract. Kahneman D. 2011 . Don’t blink! The hazards of confidence. New York Times Magazine. Available at: http://www.nytimes.com/2011/10/23/magazine/dont-blink-the-hazards-of-confidence.html (last accessed 29 March 2018). Kuhn T. 1962 . The Structure of Scientific Revolutions . University of Chicago Press , Chicago . Lakatos I. , Musgrave A. (Ed.) 1970 . Criticism and the Growth of Knowledge: Proceedings of the International Colloquium in the Philosophy of Science, London, 1965. Cambridge University Press, Cambridge. Available at: https://www.cambridge.org/core/books/criticism-and-the-growth-of-knowledge/74C4E5A7D52C5D11F65849CCD0710012 (last accessed 29 March 2018). Lefevre S. , McKenzie D. J. , Nilsson G. E. 2017 . Models projecting the fate of fish populations under climate change need to be based on valid physiological mechanisms . Global Change Biology , 23 : 3449 – 3459 . Mackinson S. , Platts M. , Garcia C. , Lynam C. 2018 . Evaluating the fishery and ecological consequences of the proposed North Sea multi-annual plan . PLoS One , 13 : e0190015. Google Scholar Crossref Search ADS PubMed Masterman M. 1970 . The nature of a paradigm. In Criticism and the Growth of Knowledge . Ed. by Lakatos I. , Musgrave A. . Cambridge University Press , London . McCarthy J. J. , Canziani, O. F., Leary, N. A., Dokken, D. J., White, K. S. 2001 . Climate Change 2001: Impacts, Adaptation, and Vulnerability Contribution of Working Group II to the Third Assessment Report of the Intergovernmental Panel on Climate Change . Cambridge University Press, Cambridge . Mesnil B. 2012 . The hesitant emergence of maximum sustainable yield (MSY) in fisheries policies in Europe . Marine Policy , 36 : 473 – 480 . Google Scholar Crossref Search ADS Mitra A. , Castellani C. , Gentleman W. C. , Jónasdóttir S. H. , Flynn K. J. , Bode A. , Halsband C. , et al. 2014 . Bridging the gap between marine biogeochemical and fisheries sciences; configuring the zooplankton link . Progress in Oceanography , 129 : 176 – 199 . Google Scholar Crossref Search ADS Nichols J. H. , Haynes, G. M., Fox, C. J., Milligan, S. P., Brander, K. M., and Chapman, R. J. 1993 . Spring plankton surveys of the Irish Sea in 1982, 1985, 1987, 1988 and 1989: hydrography and the distribution of fish eggs and larvae. Directorate of Fisheries Research, Fisheries Research Technical Report, 95. Nichols J. H. , Brander K. M. 1989 . Herring larval studies in the west-central North Sea . Rapports et Procès-verbaux des Réunions du Conseil International pour l’Exploration de la Mer , 191 , pp. 160 – 168 . Parry M. L. , Canziani, O. F., Palutikof, J. P., van der Linden, P. J., Hanson, C. E. 2007 . Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change . Cambridge University Press, Cambridge . Penzias A. A. , Wilson R. W. 1965 . A measurement of excess antenna temperature at 4080 Mc/s . Astrophysical Journal , 142 : 419 – 421 . Google Scholar Crossref Search ADS Prestidge M. C. , Taylor A. H. 1995 . A modelling investigation of the distribution of stratification and phytoplankton abundance in the Irish Sea . Journal of Plankton Research , 17 : 1397 – 1420 . Google Scholar Crossref Search ADS Pörtner H.-O. , Karl, D., Boyd, P. W., Cheung, W., Lluch-Cota, S. E., Nojiri, Y., Schmidt, D. N. et al. 2014 . Ocean systems. In Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel of Climate Change , pp. 411 – 484 . Ed. by Field C. B. ., et al. Cambridge University Press , Cambridge, United Kingdom and New York, NY . Ravetz J. R. 1979 . Scientific Knowledge and Its Social Problems . Oxford University Press , Oxford . Riemer, K., Guralnick, R. P., and White, E. P. 2018. No general relationship between mass and temperature in endothermic species eLife, 7: e27166, doi.org/10.7554/eLife.27166. Rothschild B. J. 2015 . On the birth and death of ideas in marine science . ICES Journal of Marine Science , 72 : 1237 – 1244 . Google Scholar Crossref Search ADS Sarmiento J. L. , Slater R. , Barber R. , Bopp L. , Doney S. C. , Hirst A. C. , Kleypas J. , et al. 2004 . Response of ocean ecosystems to climate warming . Global Biogeochemical Cycles , 18 : n/a , Google Scholar Crossref Search ADS Shepherd J. G. , Brander K. M. , Garrod D. J. 1985 . Towards improved procedures for fisheries management . ICES Document CM 1985/G: 73. 14 pp . Tegart W. J. M. , Sheldon G. W. , Griffiths D. C. 1990 . Report Prepared for Intergovernmental Panel on Climate Change by Working Group II , Canberra . Thompson A. B. , Harrop R. T. C. 1991 . Feeding dynamics of fish larvae on Copepoda in the Western Irish Sea, with particular reference to cod Gadus morhua . Marine Ecology Progress , 68 : 213 – 223 . Google Scholar Crossref Search ADS © International Council for the Exploration of the Sea 2018. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
TI - Seeing through
JF - ICES Journal of Marine Science
DO - 10.1093/icesjms/fsy045
DA - 2018-10-01
UR - https://www.deepdyve.com/lp/oxford-university-press/seeing-through-n9LFCV9lGP
SP - 1536
VL - 75
IS - 5
DP - DeepDyve
ER -