ICES Marine Science Symposia, 215: 221-21)6. 2002 Understanding the role of turbulence on fisheries production during the first century of ICES
Brian R. MacKenzie
MacKenzie, B. R. 2002. Understanding the role of turbulence on fisheries production during the first century of ICES. - ICES Marine Science Symposia, 215: 227-236.
Since its inception, ICES has been concerned with the effect of hydrography on the abundance and distribution of fish and fish catches. One of the earliest and most sig nificant oceanographic findings made by the ICES community was the influence of vertical mixing and turbulence on seasonal plankton production processes. This dis covery, acquired over several decades of investigation, led to three major theories of fish population regulation and demonstrates the underlying impact that turbulence has on seasonal plankton and fish production. More recently, moderate levels of turbu lence and upwelling have been shown to produce the highest recruitment among clu- peid populations inhabiting major upwelling areas. The mechanism responsible for this pattern is a balance between the positive and negative effects of both turbulence and upwelling on plankton production, larval feeding, and advective processes. In one ICES upwelling zone (Bay of Biscay), recruitment of a local clupeid is related to some of these processes. This knowledge is contributing to the ICES assessment process for this stock. Frontal zones on continental shelves within the ICES Area are also moder ately turbulent environments, may also have an impact on fish recruitment, and have received particular attention by colleagues within the ICES community. In future, an understanding of how turbulence affects fish and plankton production at upwelling and frontal zones and during storms could help justify including additional environ mental and ecosystem information in recruitment and catch prediction models.
Keywords: fish production, fronts. ICES, spring bloom, storms, turbulence, up welling.
Brian R. MacKenzie: Department o f Marine Ecology and Aquaculture, Danish Insti tute for Fisheries Research, Kavalergården 6, DK-2920 Charlottenlund, Denmark; tel: +45 33 96 34 03; fax: +45 33 96 34 34; e-mail: [email protected].
Introduction developed during the first 100 years of ICES. To be con sistent with the theme of this session of the Symposium But the great fisheries of temperate seas probably ("Variability at all scales and its effect on the ecosys depend upon the start o f production in turbulent waters. tem"), I will consider processes covering both large and - D. H. Cushing (1989). small scales.
Turbulence is an oceanic property which has a major influence on life in the sea. Phytoplankton species com position (Margalef, 1978), production rates (Lewis et Large-scale influences (decadal-century; a l, 1984), plankton foodweb structure (Kiørboe, 1993), whole basin) feeding rates (Dower et al., 1997; Kiørboe, 1997), and distributions (Lasker, 1975; Mackas et al., 1985; Seu- Arguably one of the most important influences of turbu ront et a l, 1999) have all been shown to co-vary with lence on fisheries production is its role in vertical mix turbulent intensity. Since turbulence is generated or sup ing and seasonal plankton production cycles in temper pressed by many different kinds of physical processes ate-boreal waters. In particular, turbulence generated by (e.g., tides, winds, buoyancy inputs, convection), the winds, tides, and convection is responsible for the re structure and functioning of aquatic ecosystems are plenishment of nutrients during the winter that allows commonly being perturbed by turbulent water motion of phytoplankton blooms to occur in the following spring various intensities and scales. (Cushing, 1975, 1989; Mann and Lazier, 1996). The ini In this paper, I describe how this turbulence affects tiation of these blooms depends partly on a relaxation of fish and fish production and how this understanding has turbulent mixing so that phytoplankton remain longer in 228 B. R. MacKenzie
Production o f larvae Production of larval food
Production o f eggs
Time
Figure 1. Schematic representation of Cushing’s match-mismatch hypothesis (see Cushing, 1990). The peaks illustrate the sea sonality of the abundance of fish eggs, larvae, and larval food. The striped area indicates temporal overlap between larvae and prey. the photic zone. The blooms lead to the seasonal in has been enormous. In fact, a century ago, many of the crease in zooplankton production and biomass which same workers who laid the organizational foundations becomes an important food source for fish larvae and for ICES were also instrumental in demonstrating that a juveniles, as well as for adult zooplanktivores (e.g., her seasonal plankton production pattern existed, that the ring, sprat; Cushing, 1975). seasonality itself depended on the availability, via verti The role of turbulence on the timing (i.e., initiation, cal mixing and turbulence, of nutrients for phytoplank delay, duration) of the spring plankton bloom in temper ton growth, and that the seasonality also depended on ate-boreal waters is central. Specifically, the reduction phytoplankton becoming exposed in the spring to prop in water column turbulence during the transition from er light and nutrient conditions. The fundamental obser winter to summer (e.g., due to reduced wind speeds and vations of phytoplankton abundance and distribution increasing temperatures) is critical for determining were made by scientists working in the late 1800s and when the onset of rapid phytoplankton growth occurs early 1900s (Mills, 1989). These scientists included after nutrients have been re-mineralized and re-supplied Hensen, Brandt, and Lohman in Kiel, Gran and Braarud to the surface layer during the autumn-winter mixing in Oslo and St. Andrews (Canada), Whipple in Boston, period (Mann and Lazier, 1996; Huisman et al., 1999). Riley in Woods Hole, and Sverdrup in Norway. Excessive turbulence in the spring will delay the initia One of the key roles that ICES played after its incep tion of the phytoplankton bloom because phytoplankton tion was the development of standardized techniques for cells will be mixed below the photic zone where light- measuring nutrient concentrations (Mills, 1989). These limitation will suppress growth rates. As a result, a techniques were spread among colleagues by meetings sequence of turbulence "events" (i.e., strong followed by and courses arranged by ICES and, for the first time, calm) is necessary to initiate the spring plankton bloom. enabled comparisons of phytoplankton abundance/dis This sequence of events (re-mineralization and strong tribution with nutrient concentrations (Mills, 1989). winter mixing that replenishes nutrients followed by Based on the empirical field observations, it was possi greatly reduced spring mixing that allows phytoplank ble to demonstrate that phytoplankton abundance was ton to remain in the photic zone) is a highly regular and associated with some of the variations in nutrient con predictable feature of temperate-boreal habitats. As a centrations. These observations were key to demonstrat result, these environments are characterized by strong ing that the spring bloom was related to vertical mixing seasonal pulses in plankton production, although their and stabilization processes (e.g., Gran and Braarud, exact timing, magnitude, and duration can vary greatly 1935; Riley, 1942; Sverdrup, 1953). from year to year. Nevertheless, at decadal-century time Notably one of the motivations to develop an interna scales and over large areas of the sea, they are consistent tional body for marine science investigations was the features that provide important environmental signals to desire to improve the possibility for predicting fluctua other biota in the ecosystem. tions in the important fish stocks of northern European The role of ICES in defining these links between tur waters. It was understood at the time that hydrographic bulence, plankton production, and fisheries production variability was part of the cause of such variations (e.g., Understanding the role of turbulence on fisheries production during the first century of ICES 229
1985/1986 1986/1987 1987/1988 90 60
60 40 0)Ï a)» £= F 30 20 Ü E 0 0 40 42 44 46 48 50 52 54 56 42 44 46 48 50 52 54 56 42 44 46 48 50 52 54 56
™ 2.4 a. O) »>
I 0.6 o 32 34 36 38 40 42 44 46 48 34 36 38 40 42 44 46 48 34 36 38 40 42 44 46 48
Figure 2. Total numbers of a newly settled temperate marine reef fish (Heteroclinus sp.) in Storm Bay, Australia, during three reproductive seasons and the abundance of chlorophyll a. Note that the scales on the ordinate axes are offset by 8 weeks corre sponding to the duration of the larval period before metamorphosis and settlement. Line breaks represent missing data. See Thresher et al. (1989) for details.
Hjort, 1914), but the mechanisms were only vaguely strate that fish recruitment was strongly correlated known and not quantified (e.g., owing to changes in with phytoplankton production peaks following storm- timing and location of migration routes, effects on egg induced mixing (Figure 2). Other colleagues have and larval mortality rates). However, as more knowl observed relationships between larval/0-group growth edge of the spring plankton bloom became available, and recruitment (Campana, 1996; Ottersen and Loeng, attention began to focus on the role of this event in fish 2000), but it is unclear to what extent the growth rates growth and survival. were driven by variations in temperature or food supply. Cushing (1975, 1990) observed that many of the most Nevertheless, the influence of turbulence on seasonal abundant fish species in temperate-boreal waters (e.g., plankton production and fish ecology is great. It seems cod, haddock, herring) spawned in the spring during or obvious that large parts of continental shelf regions shortly after the spring plankton production peak and would be considerably less productive if turbulence dur that spawning times were relatively stable compared ing winter were too weak to remix the water column and with the temporal variability in onset of the spring resupply the photic zone with nutrients. In such a case, plankton bloom. He and others before him (notably the productivity and fish biomass of these regions would Hjort, 1914) also noted that the larvae produced by probably decrease to much lower levels. these fish were poor swimmers, yet needed to find suf Alternatively, perhaps better evidence for the ficient food to grow and survive when their yolk match-mismatch theory could be found at the much reserves became depleted. Cushing suggested that one longer scale of life history adaptation (decades-cen- of the conditions for a strong year class was therefore turies) in terms of spawning time and location. Few overlap in time and space of larvae with their prey, and species in temperate-boreal latitudes spawn in winter this theory came to be known as the "match-mismatch" when phyto- and zooplankton concentrations are low, theory (Cushing, 1990; Figure 1). whereas most spawn later in the spring or summer when This theory has since formed the basis of numerous food concentrations are higher. However, even at this fisheries research programmes throughout the North scale, there are other explanations (e.g., the member/- Atlantic and farther abroad. Unfortunately it has proven vagrant theory of population maintenance; Sinclair, extremely difficult to verify because of various sam 1988) for the timing of spawning in fish populations. pling problems, measurement error (Heath, 1992; Leggett and deBlois, 1994; Cushing, 1995a), and the possibility that larvae and juveniles may not be food- Interannual influences limited (Sinclair, 1988). In particular, and perhaps somewhat surprisingly, there are relatively few studies Turbulence has major effects on fish production in which have directly compared plankton production, upwelling areas (Bakun, 1996). First, Lasker (1975, growth or condition of larvae or 0-group juveniles, and 1978) has shown that storms inhibit formation of phyto recruitment time-series. Thresher et al. (1989) appear to plankton patches required by first-feeding anchovy lar have conducted one of the few investigations to demon vae. This observation has led to the development of the 230 B. R. MacKenzie
0.30
M «
° 0.10
14 22 30 Calm periods per month
Figure 3. Average instantaneous natural mortality rate (M) of Figure 4. Theoretical relationship between recruitment and northern anchovy (Engraulis mordax) larvae in relation to environmental factors in upwelling areas (Cury and Roy, weighted average number of calm four-day periods per month 1989). during the spawning season for 13 years (Peterman and Brad ford, 1987).
"stable-ocean" theory of recruitment variability. Second, ditions. Fish larvae will, therefore, have difficulty find Peterman and Bradford (1987) have shown that storm ing high concentrations of prey required for high growth frequency directly affects larval anchovy mortality rates and survival rates (Lasker, 1975, 1978). Moreover, the (Figure 3). Third, Cury, Roy, and Bakun have shown that high upwelling intensity favours advection of eggs and recruitment of pelagic clupeids in several of the world’s larvae away from the coastal area and hence loss of off major upwelling zones is optimal at intermediate levels spring from the population (Roy, 1998). Both the low of upwelling and turbulence intensity (Cury and Roy, and high turbulence/upwelling regimes are, therefore, 1989; Bakun, 1996; Figure 4). In many of the systems predicted to result in low recruitment. studied, the turbulence and upwelling are strongly cor In contrast, an intermediate level of turbulence and related (Cury and Roy, 1989) because both are estimat upwelling appears to produce the oceanographic condi ed indirectly from windspeed. tions that lead to highest recruitment (Figure 4). The role of upwelling and turbulence in these systems Intermediate intensities of turbulence and upwelling is believed to be related to general principles of biolog result in moderately high growth rates of phytoplankton ical (plankton) production in marine ecosystems (Cury because the cells experience optimal light and nutrient and Roy, 1989; Bakun, 1996). Years with low upwelling conditions. The higher mixing intensity results in a sim and turbulence have low phytoplankton production pler foodweb with a higher proportion of diatoms than because nutrients are not mixed into the surface layer at in the calmer situation. The simpler foodweb and high high rates. As a result, the base of the foodweb is small er concentration of phytoplankton stimulates zooplank er, and there is less food for zooplankton whose produc ton production rates, which in turn increase zooplankton tion rates remain low. Moreover, we now know that in concentrations. Since the turbulence intensity is moder low turbulence situations, pelagic foodwebs will be ate, some of the phytoplankton and zooplankton can dominated by smaller phytoplankton species (Margalef, remain aggregated in patches, thereby providing larvae 1978; Kiørboe et al., 1988) and have more trophic links and juveniles with relatively high concentrations of due to the microbial food web (Kiørboe, 1993, 1997). prey. The moderate upwelling intensity will help to This means that the foodweb is less efficient at transfer retain eggs and larvae in coastal areas where survival is ring energy from phytoplankton to zooplankton (Cush probably higher (Roy, 1998). Lastly, the moderate tur ing, 1989; Legendre, 1990; Kiørboe, 1993), and that bulence regime is also highly advantageous for the feed zooplankton production rates and abundances will be ing of individual larvae on their prey (see below). lower. As a result, it is assumed that larvae will have Much of this work has been conducted outside the lower feeding, growth, and survival rates during these geographical framework of ICES because most of the years. ecosystems lie beyond the North Atlantic. However, the In comparison, when upwelling intensity and turbu biological oceanographic basis (e.g., vertical mixing lence are too strong, the phytoplankton will be light- and plankton production, role of patchiness in larval limited owing to excessive vertical mixing. Zooplankton feeding success, advection) for the statistical analyses concentrations will still be low, and zooplankton distri have been discussed and presented in various ICES fora butions will be more homogeneous than in calmer con (e.g., Lasker, 1978; Parsons et al., 1978). Understanding the role of turbulence on fisheries production during the first century of ICES 231
Relative Velocity3D (mm s'1)
Figure 5. Probability of successful pursuit of cod larvae preying on live copepod nauplii and copepodites in water having different turbulent intensities as expressed by the relative velocity of prey during individual pursuit events. The solid line is a regression fit weighted for the number of observations in each relative velocity interval (sample sizes are given above bars). These relative velocities are common in the upper layer of the sea. See MacKenzie and Kiørboe (2000) for details.
There are few upwelling zones within the ICES Area and Lough, 1987; Munk et al., 1995, 1999). Their im and, therefore, the influence of these findings on fish portance for both plankton production and aggregation, production and management there may be limited to an and as young fish nursery areas, has received much improved general understanding of how hydrographic attention among scientists within and outside the ICES conditions might influence fish stocks. However, in one community (Kiørboe et al., 1988; Munk et al., 1995, of the larger ICES upwelling zones (Bay of Biscay), it 1999; Nakata et a i, 1995; St. John and Lund, 1996; has recently been shown that recruitment to the anchovy MacKenzie and Werner, 1997) and also within its Cod stock is strongly related to windspeed and upwelling and Climate Change Programme (ICES, 1994). As a intensity in ways consistent with the patterns seen by result, many researchers have presented findings at Bakun and colleagues (Borja et al., 1998; ICES, 2000). ICES Statutory Meetings and Annual Science Con Notably for the Biscay anchovy stock, this environmen ferences on how frontal processes affect the distribution tal information was used to provide inputs to the short and production of both fish and plankton (e.g., term catch predictions and to justify a reduction in fish Christensen et al., 1985; Kiørboe and Johansen, 1986; ing effort in 2000 to conserve the stock (ICES, 2000). Munk et al., 1986; Richardson et al., 1986). How This step indicates that knowledge of oceanographic ever, the influence of fronts on fish recruitment is still processes affecting the stock may have important prac undetermined, but since their formation, intensity, lo tical benefits for fisheries management. It also suggests cation, and duration are partly climatically determined that ICES has improved its understanding of how hydro (LeFevre, 1986; Bowers and Simpson, 1987), it is pos graphy and, in particular, upwelling and turbulence sible that the mixing and turbulence associated with affects fish production in at least one of its stocks. fronts could contribute to fish production. The issue Turbulence also affects plankton, and probably also remains unresolved. Consistent with this possibility, fish, production in other ways at interannual time scales. however, are statistical analyses of the influence of var For example, frontal zones (Simpson and Hunter, 1974; ious environmental factors on fish recruitment. These LeFevre, 1986) appear to have some (but certainly not studies frequently show that turbulence-related variables all) of the same turbulence and plankton production (e.g., windspeeds, stratification, freshwater run-off, characteristics of upwelling zones (Cushing, 1995b; salinity) can have significant impacts on recruitment in MacKenzie, 2000). These areas often have more larvae ICES fish stocks (Shepherd et al., 1984; Svendsen et al., and 0-group juveniles than neighbouring areas (Buckley 1995). 232 B. R. MacKenzie
C/) c o '■*—> •*—>0 3 O o k_ Figure 6. Total number of citations of seven papers investigating the role of turbulence on plankton and fish. Citations are from the ISI Science Citation Index database covering more than 3500 scholarly journals. Small-scale (individual-level) influences encountered prey. Models that estimate the success of pursuit in turbulent conditions show that pursuit success In recent years, the influence of small-scale turbulence deteriorates with increases in turbulence (Matsushita, on plankton has become a research focus. One part of 1991; MacKenzie et al., 1994; Kiørboe and Saiz, 1995). this focus has been the role of turbulence on feeding One of these models has been validated with experi rates between planktonic predators and prey. Rothschild mental observations which showed that pursuit success and Osborn (1988) developed a theory for encounter decreased with increasing turbulence (MacKenzie and rates between predators and prey which demonstrated Kiørboe, 2000; Figure 5), consistent with model predic that turbulent water motion at small scales (cm-mm) tions. can increase the apparent prey concentration perceived In the field, these theories are difficult to evaluate by planktonic predators. This result means that en because of sampling errors (MacKenzie, 2000). How counter rates are higher in turbulent water than in calm ever, Sundby and Fossum (1990) and Sundby et al. water, if other factors (especially food concentration) (1994) have shown that gut contents of cod larvae in remain unchanged. crease significantly with windspeed up to ca. 10 m s'1. These findings relate particularly well to fish larvae. This suggests that turbulence may also affect feeding The typical size of fish larvae (ca. 3-20 mm) places rates in nature. However, a large compilation of field them in a size range where the small eddies of decaying studies which considered various biological responses turbulence still have sufficient energy to affect relative to turbulence (i.e., gut contents, RNA/DNA ratios, velocities between two separated particles (e.g., a larva otolith growth rates) showed that there was no overall and its prey). Notably, at very small size scales (i.e., pattern in how turbulence affected these responses below the Kolmogorov scale; Sanford, 1997), turbulent (MacKenzie, 2000). Reasons for the inconsistencies motion is greatly reduced owing to viscosity (Sanford, (especially sampling deficiencies), and potential ways 1997), and little effect on encounter rate is expected. to overcome them, are detailed in full elsewhere (Mac Turbulent dissipation in the upper layer of the sea is Kenzie, 2000). More work at small scales (cm-m) is mainly related to windspeed (Oakey and Elliott, 1982; needed in situ (e.g., direct behavioural observations, MacKenzie and Leggett, 1993) and in shallow areas also improved understanding of the dynamics of patchiness to tidal currents (Simpson and Hunter, 1974; Simpson et under changing turbulence conditions) to resolve some al., 1996). Fish larvae encounter rates with prey as pre of these issues. dicted from models can be several-fold higher than in These relatively new findings of the roles of turbu calm situations (Dower et al., 1997). Direct laboratory lence at small scales in fish ecology can contribute to observation of cod larvae feeding on live copepods the oceanographic basis of statistical relationships be showed that these predictions agree well with the obser tween turbulence-related variables and fish recruitment vations (MacKenzie and Kiørboe, 1995). However, en or growth. For example, the overall ingestion rate counter is only one part of the complete predation response of larval fish to changes in turbulence is dome sequence; predators must also pursue and capture their shaped (MacKenzie et al., 1994). This response shape is the 1S1 Science Citation Index database covering more than 3500 scholarly journals. 3500 1S1than more covering the Indexdatabase Citation Science Figure 7. Number of 7.of Number Figure citations per year for seveninvestigating from papers Citations are how andfish. affects plankton turbulence Number of citations Number of citations Number of citations Number of citations 20 24 20 24 28 28 12 16 20 12 16 24 28 12 16 20 24 28 16 12 4 0 4 0 8 4 8 0 8 4 0 8 92 96 90 94 98 92 96 2000 1996 1992 1988 1984 1980 1976 1972 92 96 90 94 98 92 96 2000 1996 1992 1988 1984 1980 1976 1972 92 96 90 94 98 92 96 2000 1996 1992 1988 1984 1980 1976 1972 92 96 90 94 98 92 96 00 92 96 90 94 98 92 96 2000 1996 1992 1988 1984 1980 1976 1972 2000 1996 1992 1988 1984 1980 1976 1972 Understanding the rote o f turbulence on fisheries production during the first century o f f ICES o century the first during production fisheries on f turbulence o rote the Understanding iro ta. 1988 al.et Kiørboe Margalef 1978 Margalef Lasker 1975Lasker Simpson & Hunter 1974 Hunter &Simpson J i i A A v a A A A A y \ y w m RA \ 20 24 28 20 24 12 16 12 28 16 20 24 28 12 16 4 4 0 0 8 8 4 0 8 92 96 90 94 98 92 96 2000 1996 1992 1988 1984 1980 1976 1972 92 96 90 94 98 92 96 2000 1996 1992 1988 1984 1980 1976 1972 uhn 1989 Cushing Cury & Roy 1989RoyCury & ohcid&Obm 98 A 1988 Osbom & Rothschild • J J N V \ 233 234 B. R. MacKenzie similar to the dome-shaped response of recruitment in Nevertheless, the citation data show that these papers sardine and anchovies to turbulence and upwelling in have been widely read and used by many other col their habitats (Bakun, 1996). The influence of turbu leagues (Figures 6 and 7). In particular, the papers by lence on feeding of individual fish larvae is probably a Lasker (1975), Rothschild and Osborn (1988), Cushing factor that contributes to these statistical patterns along (1989), and Cury and Roy (1989) are notable because with the plankton production and advection mecha they deal explicitly with fish and fisheries, whereas the nisms outlined above. other papers consider hydrography or lower trophic lev Interestingly, copepods appear to live at the border els. Nearly all of the papers have been cited on average between the viscous and inertial worlds, as defined by about 10 times per year (Figure 7). the Reynolds number (Naganuma, 1996). Copepods feed mainly in the viscous regime (Re <1), but when attacked by predators, they can escape at velocities in A future role for ICES in evaluating the inertial regime (Re >1). Hence, copepods are one of the few taxa which live at the border of the viscous-iner how turbulence affects fish and fish tial worlds, and this may be a reason why they are so production? abundant in the ocean (Nanaguma, 1996). Copepods also live in the transition regime between laminar and ICES has been very supportive of studies investigating turbulent flow (Re = 1-2000). This Reynolds range is biological-physical interactions and how these interac occupied by fish larvae (e.g., a 5-mm larva swimming at tions might influence fish populations (e.g., the Balti 1 body length s_l has Re = 25). Hence, predation of more Symposium on Recruitment Dynamics of Exploi copepods by fish larvae represents a link between lower ted Marine Populations: Physical-Biological Inter and higher trophic levels, and between photosynthetic actions in 1997, Cod and Climate Programme, Working and fisheries production (Naganuma, 1996). These links Group on Recruitment Processes). A large amount of are possible because copepods live at the border of im work remains to be done to fully understand how fish portant fluid dynamic regimes and because fish larvae and pelagic ecosystems respond to turbulence (see, for themselves begin life at the laminar-turbulent regime example, reviews by Kiørboe, 1991; Bakun, 1996; (Re ~25), but grow into the turbulent regime (Re >2000) Dower et al., 1997; Marrasé et al., 1997; Sanford, 1997; during ontogeny. MacKenzie, 2000) and to identify which of the many Many of the investigations of small-scale turbulence oceanographic processes that turbulence influences have been motivated by ICES findings, presentations have the most impact on fish growth and survival. More (e.g., Parsons etal., 1978; Roy, 1993), and study groups. work is needed, and continued support from ICES for Since variations in the intensity of turbulence are partly combined biological-physical oceanographic investiga related to climatic factors, the investigations are rele tions is needed. vant to the ICES/GLOBEC Working Group on Cod and It is still too early, in many cases, to link the conse Climate Change and to the Working Group on Recruit quences of changes in turbulence or other environmen ment Processes. They may also help to interpret envi tal variables to entire fish populations. However, as such ronmentally based models of recruitment (Bakun, 1996). links become available, it may become desirable for ICES to assist with the incorporation of this information into new fish population and assessment models. One Impact of turbulence-related studies on role for ICES in future could be to assist with the devel opment of more ecologically based models. Notably, the marine science community some encouraging steps in this direction are being taken (e.g., Study Group on Incorporation of Environmental Several papers showing how turbulence affects fish, fish Processes into Assessment Models). production, and pelagic ecosystems have had major impacts on the fisheries/biological oceanographic com munities. One measure of this impact is how frequently the studies have been cited by subsequent workers. I Conclusions have conducted a citation analysis of some of these papers to evaluate their total number of citations and The most important role that ICES has had in under annual citation rates. The citation data are compiled standing how turbulence affects fish production was in from the Institute for Scientific Information’s Science identifying the existence of the spring plankton produc Citation Index, and the analysis includes data until the tion bloom and the factors controlling it. This was done end of 1999. According to SCI (http://www.isinet.com/ in the earliest decades of ICES’ existence and set the products/citation/citsci.html), the index covers 3500 of stage for developing at least three major theories of the world’s leading scholarly journals. recruitment and population dynamics: the match-mis I have chosen seven papers for the analysis. The match theory, the member/vagrant theory, and the sta- choice no doubt reflects the personal bias of the author, ble-ocean theory. ICES has also supported research pro and apologies go to those whose papers are excluded. grammes on other topics more or less directly related to Understanding the role o f turbulence on fisheries production during the first century o f ICES 235 turbulence and fisheries by generally encouraging stud Gran, H. H.. and Braarud, T. 1935. A quantitative study of the ies of biological-physical interactions. These studies phytoplankton in the Bay of Fundy and the Gulf of Maine. have identified several of the major production charac Journal of the Biological Board of Canada, 1: 279-^-67. teristics of upwellings, frontal zones, and storms, and Heath, M. R. 1992. Field investigations of the early life histo ry stages of marine fish. Advances in Marine Biology, 28: their potential importance to fish production. 2-174. Hjort, H. 1914. Fluctuation in the Great Fisheries of Northern Europe Reviewed in the Light of Biological Research. Rap ports et Procès-Verbaux des Réunions du Conseil Interna Acknowledgements tional pour l'Exploration de la Mer, 20: 1-228. Huisman, J., Oostveen, P. v., and Weissing, F. J. 1999. Critical 1 thank Carina Anderberg, librarian at the Danish depth and critical turbulence: two different mechanisms for Institute for Fisheries Research (DIFRES), for biblio the development of phytoplankton blooms. Limnology and graphic assistance with the citation analyses and two Oceanography, 44: 1781-1787. ICES. 1994. Report of the ICES/GLOBEC Cod and Climate reviewers for constructive comments and suggestions. Change AGGREGATION Workshop, Charlottenlund, Den This study was funded partly by DIFRES and the EU mark, 22-24 August 1994. ICES CM 1994/A: 10. 13 pp. (FAIR contract no. CT98-3959). ICES. 2000. Report of the Working Group on the Assessment of Mackerel, Horse Mackerel, Sardine and Anchovy, 14-23 September 1999. ICES CM 2000/ACFM:5. 546 pp. Kiørboe, T. 1991. Pelagic fisheries and spatio-temporal vari References ability in zooplankton productivity. Proceedings of the 4th International Conference on Copepoda. Bulletin of the Bakun, A. 1996. Patterns in the Ocean: Ocean Processes and Plankton Society of Japan, Special Volume 1991: 229-249. Marine Population Dynamics. California Sea Grant College Kiørboe, T. 1993. Turbulence, phytoplankton cell size, and the System, NOAA and Centro de Investigaciones Biolögicas structure of pelagic food webs. Advances in Marine Biology, del Noroeste, La Paz, Mexico. 323 pp. 29: 1-72. Borja, A., Uriarte, A., Egana, J., Motos, L., and Valencia, Kiørboe, T. 1997. Small-scale turbulence, marine snow forma V 1998. Relationships between anchovy (Engraulis en- tion, and planktivorous feeding. Scientia Marina, 61 (Sup crasicolus) recruitment and environment in the Bay of Bis plement I): 141-158. cay (1967-1998). Fisheries Oceanography, 7 (3/4): 375- Kiørboe, T., and Johansen, K. 1986. Studies of a larval herring 380. (Clupea harengus) patch in the Buchan area. IV Zooplank Bowers, D. G., and Simpson, J. H. 1987. Mean position of tidal ton distribution and productivity in relation to hydrographic fronts in European shelf seas. Continental Shelf Research, 7: features. Dana, 6: 37-51. 35-44. Kiørboe, T.. Munk, P., Richardson, K., Christensen, A., and Buckley, L. J., and Lough, R. G. 1987. Recent growth, bio Paulsen, H. 1988. Plankton dynamics and larval herring chemical composition, and prey field of larval haddock growth, drift and survival in a frontal area. Marine Ecology (Melanogrammus aeglefinus) and Atlantic cod (Gadus Progress Series, 44: 205-219. morhua) on Georges Bank. Canadian Journal of Fisheries Kiørboe, T., and Saiz, E. 1995. Planktivorous feeding in calm and Aquatic Sciences, 44: 14-25. and turbulent environments with emphasis on copepods. Campana, S. 1996. Year-class strength and growth rate in Marine Ecology Progress Series, 122: 135-145. young Atlantic cod Gadus morhua. Marine Ecology Pro Lasker, R. 1975. Field criteria for survival of anchovy larvae: gress Series, 135: 21-26. the relation between inshore chlorophyll maximum layers Christensen, V, Heath, M., Kiørboe, T., Munk, P., Paulsen, H., and successful first feeding. Fishery Bulletin, 73: 453-462. and Richardson, K. 1985. Investigations on the relationship Lasker, R. 1978. The relation between oceanographic condi of herring larvae, plankton production and hydrography at tions and larval anchovy food in the California Current: Aberdeen Bank. Buchan area, September 1984. ICES CM identification of factors contributing to recruitment failure. 1985/L:23. 22 pp. Rapports et Procès-Verbaux des Réunions du Conseil Inter Cury, P., and Roy, C. 1989. Optimal environmental window national pour l’Exploration de la Mer, 173: 212-230. and pelagic fish recruitment success in upwelling areas. LeFevre, J. 1986. Aspects of the biology of frontal systems. Canadian Journal of Fisheries and Aquatic Sciences, 46: Advances in Marine Biology, 23: 163-299. 670-680. Legendre, L. 1990. The significance of microalgal blooms for Cushing, D. H. 1975. Marine Ecology and Fisheries. Cam fisheries and for the export of particulate organic carbon in bridge University Press, Cambridge. 278 pp. oceans. Journal of Plankton Research, 12: 681-699. Cushing, D. H. 1989. A difference in structure between ecosys Leggett, W. C., and deBlois, E. 1994. Recruitment in marine tems in strongly stratified waters and in those that are only fishes: is it regulated by starvation and predation in the egg weakly stratified. Journal of Plankton Research, 11: 1-13. and larval stages? Journal of Sea Research, 32: 119-134. Cushing, D. H. 1990. Plankton production and year class Lewis, M. R., Horne, E. P. W., Cullen, J. J., Oakey, N. S., and strength in fish populations: an update of the match/mis Platt, T. 1984. Turbulent motions may control phytoplankton match hypothesis. Advances in Marine Biology, 26: 249-293. photosynthesis in the upper ocean. Nature, 311: 49-50. Cushing, D. H. 1995a. A comment on Leggett & DeBlois. Mackas, D. L., Denman, K. L., and Abbott, M. R. 1985. Plank Marine Ecology Progress Series, 128: 305-306. ton patchiness: Biology in the physical vernacular. Bulletin Cushing, D. H., 1995b. Population Production and Regulation of Marine Science, 37: 652-674. in the Sea: A Fisheries Perspective. Cambridge University MacKenzie, B. R. 2000. Turbulence, larval fish ecology and Press, Cambridge. 354 pp. fisheries recruitment: a review of field studies. Oceano- Dower, J. F., Miller, T. J., and Leggett, W. C. 1997. The role of logica Acta, 23: 357-375. microscale turbulence in the feeding ecology of larval fish. MacKenzie, B. R., and Kiørboe, T. 1995. Encounter rates and Advances in Marine Biology, 31: 169-220. swimming behavior of pause-travel and cruise larval fish 236 B. R. MacKenzie predators in calm and turbulent laboratory environments. Peterman, R. M., and Bradford, M. J. 1987. Wind speed and Limnology and Oceanography, 40: 1278-1289. mortality rate of a marine fish, the northern anchovy, MacKenzie, B. R., and Kiørboe, T. 2000. Larval fish feeding Engraulis mordax. Science, 235: 354-356. and turbulence: a case for the downside. Limnology and Richardson, K., Heath, M., and Pihl, N. J. 1986. Studies of a Oceanography, 45: 1-10. larval herring (Ctupea harengus L.) patch in the Buchan MacKenzie, B. R., and Leggett, W. C. 1993. Wind-based mod area. 1. The distribution of larvae in relation to hydrograph els for estimating the dissipation rates of turbulent energy in ic features. Dana, 6: 1-10. aquatic environments: empirical comparisons. Marine Eco Riley, G. H. 1942. The relationship of vertical turbulence and logy Progress Series, 94: 207-216. spring diatom flowering. Journal of Marine Research, 5: MacKenzie, B. R„ Miller, T. J., Cyr, S., and Leggett, W. C. 67-87. 1994. Evidence for a dome-shaped relationship between Rothschild, B. J., and Osborn, T. R. 1988. Small-scale turbu turbulence and larval fish ingestion rates. Limnology and lence and plankton contact rates. Journal of Plankton Re Oceanography, 39: 1790-1799. search, 10: 465^174. MacKenzie, B. R., and Werner, F. E. 1997. Introduction. Roy, C. 1993. The optimal environmental window hypothesis: Theme Session on Intermediate-Scale Physical Processes a non-linear environmental process affecting recruitment and Their Influence on the Transport and Feeding Environ success. ICES CM 1993/L:76. 13 pp. ment of Fish. ICES Journal of Marine Science, 54: 159. Roy, C. 1998. An upwelling-induced retention area off Sene Mann, K. H., and Lazier, J. R. N. 1996. Dynamics of Marine gal: a mechanism to link upwelling and retention processes. Ecosystems: Biological-Physical Interactions in the Oceans. South African Journal of Marine Science, 19: 89-98. 2nd ed., Blackwell Science, Boston, Massachusetts, USA. St. John, M. A., and Lund, T. 1996. Lipid biomarkers: linking 394 pp. the utilisation of frontal plankton biomass to enhanced con Margalef, R. 1978. Life-forms of phytoplankton as survival dition of juvenile North Sea cod. Marine Ecology Progress alternatives in an unstable environment. Oceanologica Acta, Series, 131: 75-85. 1: 493-509. Sanford, L. P. 1997. Turbulent mixing in experimental ecosys Marrasé, C., Saiz, E., and Redondo, J. M. (Eds.). 1997. tem studies. Marine Ecology Progress Series, 161: 265-293. Lectures on plankton and turbulence. Scientia Marina, 61 Seuront, L., Schmitt, F., Lagadeuc, Y., Schertzer, D., and Love- (Supplement 1). 238 pp. joy, S. 1999. Universal multifractal analysis as a tool to Matsushita. K. 1991. How do fish larvae of limited motility characterize multiscale intermittent patterns: example of encounter nauplii in the sea? Proceedings of the 4th Inter phytoplankton distribution in turbulent coastal waters. national Conference on Copepoda. Bulletin of the Plankton Journal of Plankton Research, 21: 877-922. Society of Japan, Special Volume 1991: 251-270. Shepherd, J. G., Pope, J. G., and Cousens, R. D. 1984. Mills, E. L. 1989. Biological Oceanography: An Early History, Variations in fish stocks and hypotheses concerning their 1870-1960. Cornell University Press, Ithaca, New York. 378 pp. links with climate. Rapports et Procès-Verbaux des Munk, P., Christensen, V., and Paulsen, H. 1986. Studies of a Réunions du Conseil International pour l'Exploration de la larval herring (Ctupea harengus L.) patch in the Buchan Mer, 185: 255-267. area. II. Growth, mortality and drift of larvae. Dana, 6: 11-24. Simpson, J. H., and Hunter, J. R. 1974. Fronts in the Irish Sea. Munk, P., Larsson, P. O., Danielsson, D. S., and Moksness, E. Nature, 250: 404-406. 1995. Larval and small juvenile cod Gadus morhua concen Simpson. J. H., Crawford, W. R., Rippeth, T. P., Campbell, A. trated in the highly productive areas of a shelf break front. R., and Cheok, J. V. S. 1996. The vertical structure of turbu Marine Ecology Progress Series, 125: 21-30. lent dissipation in shelf seas. Journal of Physical Oceano Munk, P., Larsson, P. O., Danielsson, D. S., and Moksness, E. graphy, 26: 1579-1590. 1999. Variability in frontal zone formation and distribution Sinclair. M. 1988. Marine Populations. Washington Sea Grant of gadoid fish larvae at the shelf break in the northeastern Program, Seattle, Washington, USA. 252 pp. North Sea. Marine Ecology Progress Series, 177: 221-233. Sundby, S.. Ellertsen, B., and Fossum. P. 1994. Encounter rates Naganuma, T. 1996. Calanoid copepods: linking lower-higher between first-feeding cod larvae and their prey during mod trophic levels by linking lower-higher Reynolds numbers. erate to strong turbulent mixing. ICES Marine Science Sym Marine Ecology Progress Series, 136: 311-313. posia, 198: 393-405. Nakata, K., Zenitani, H., and Inagake, D. 1995. Differences in Sundby, S., and Fossum. P. 1990. Feeding conditions of Arcto- food availability for Japanese sardine larvae between the Norwegian cod larvae compared with the Rothschild- frontal region and the waters on the offshore side of Osborn theory on small-scale turbulence and plankton con Kuroshio. Fisheries Oceanography, 4: 68-79. tact rates. Journal of Plankton Research, 12: 1153-1162. Oakey, N. S., and Elliott, J. A. 1982. Dissipation within the sur Svendsen, E., Agien, A.. Iversen, S. A., Skagen, D. W., and face mixed layer. Journal of Physical Oceanography, 12: Smedstad, O. 1995. Influence of climate on recruitment and 171-185. migration of fish stocks in the North Sea. Canadian Special Ottersen, G., and Loeng, H. 2000. Covariability in early Publication of Fisheries and Aquatic Sciences, 121: 641 growth and year class strength of Barents Sea cod, haddock 653. and herring: the environmental link. ICES Journal of Marine Sverdrup, H. O. 1953. On conditions for the vernal blooming Science, 57: 339-348. of phytoplankton. Journal du Conseil International pour Parsons, T. R., Jansson, B.-O., Longhurst, A. R., and Sætersdal, l’Exploration de la Mer, 18: 287-205. G. (Eds.). 1978. Marine ecosystems and fisheries oceanog Thresher, R. E., Harris. G. P., Gunn, J. S., and Clementsen, L. raphy. Rapports et Procès-Verbaux des Réunions du Conseil A. 1989. Phytoplankton production pulses and episodic set International pour l’Exploration de la Mer, 173. 240 pp. tlement of a temperate marine fish. Nature, 341: 641-643.