A synthesis of the life history and ecology of juvenile Pacific herring in Prince William Sound,Alaska

Physical and biological variables affecting juvenile Pacific herring (Clupea pallasi) in Prince William Sound (PWS) from 1995 to 1998 were investigated as part of a multifaceted study of recruitment, the Sound Ecosystem Assessment (SEA) program. Though more herring larvae were retained in eastern PWS bays, ages‐0 and ‐1 herring used bays throughout PWS as nursery areas. Water transported into PWS from the Gulf of Alaska (GOA) contributed oceanic prey species to neritic habitats. Consequently, variations in local food availability resulted in different diets and growth rates of herring among bays. Summer food availability and possible interspecific competition for food in nursery areas affected the autumn nutritional status and juvenile whole body energy content (WBEC), which differed among bays. The WBEC of age‐0 herring in autumn was related to over‐winter survival. The limited amount of food consumption in winter was not sufficient to meet metabolic needs. The smallest age‐0 fish were most at risk of starvation in winter. Autumn WBEC of herring and winter water temperature were used to model over‐winter mortality of age‐0 herring. Differences in feeding and energetics among nursery areas indicated that habitat quality and age‐0 survival were varied among areas and years. These conditions were measured by temperature, zooplankton abundance, size of juvenile herring, diet energy, energy source (GOA vs. neritic zooplankton), WBEC, and within‐bay competition.     

Density‐ and size‐dependent mortality in fish early life stages

The importance of survival and growth variations early in life for population dynamics depends on the degrees of compensatory density dependence and size dependence in survival at later life stages. Quantifying density‐ and size‐dependent mortality at different juvenile stages is therefore important to understand and potentially predict the recruitment to the population. We applied a statistical state‐space modelling approach to analyse time series of abundance and mean body size of larval and juvenile fish. The focus was to identify the importance of abundance and body size for growth and survival through successive larval and juvenile age intervals, and to quantify how the dynamics propagate through the early life to influence recruitment. We thus identified both relevant ages and mechanisms (i.e. density dependence and size dependence in survival and growth) linking recruitment variability to early life dynamics. The analysis was conducted on six economically and ecologically important fish populations from cold temperate and sub‐arctic marine ecosystems. Our results underscore the importance of size for survival early in life. The comparative analysis suggests that size‐dependent mortality and density‐dependent growth frequently occur at a transition from pelagic to demersal habitats, which may be linked to competition for suitable habitat. The generality of this hypothesis warrants testing in future research.     

Ecosystem controls of juvenile pink salmon (Onchorynchus gorbuscha) and Pacific herring (Clupea pallasi) populations in Prince William Sound,Alaska

Five years of field, laboratory, and numerical modelling studies demonstrated ecosystem‐level mechanisms influencing the mortality of juvenile pink salmon and Pacific herring. Both species are prey for other fishes, seabirds, and marine mammals in Prince William Sound. We identified critical time‐space linkages between the juvenile stages of pink salmon and herring rearing in shallow‐water nursery areas and seasonally varying ocean state, the availability of appropriate zooplankton forage, and the kinds and numbers of predators. These relationships defined unique habitat dependencies for juveniles whose survivals were strongly linked to growth rates, energy reserves, and seasonal trophic sheltering from predators. We found that juvenile herring were subject to substantial starvation losses during a winter period of plankton diminishment, and that predation on juvenile pink salmon was closely linked to the availability of alternative prey for fish and bird predators. Our collaborative study further revealed that juvenile pink salmon and age‐0 herring exploit very different portions of the annual production cycle. Juvenile pink salmon targeted the cool‐water, early spring plankton bloom dominated by diatoms and large calanoid copepods, whereas young‐of‐the‐year juvenile herring were dependent on warmer conditions occurring later in the postbloom summer and fall when zooplankton was composed of smaller calanoids and a diversity of other taxa. The synopsis of our studies presented in this volume speaks to contemporary issues facing investigators of fish ecosystems, including juvenile fishes, and offers new insight into problems of bottom‐up and top‐down control. In aggregate, our results point to the importance of seeking mechanistic rather than correlative understandings of complex natural systems.     

Spatio‐temporal,ontogenetic and interindividual variation of age‐0 diets in a population of yellow perch

Dietary niches of fishes have traditionally been evaluated at the population level, with diet pattern central tendencies compared spatio‐temporally among habitats and populations. More recently, however, studies have emphasised the importance of within‐population diet variation and niche partitioning. Several studies have examined diets of young yellow perch (Perca flavescens) at the population level and have described an ontogenetic transition from zooplankton to benthic prey during the first year of life. However, independent of ontogenetic diet shifts, intrapopulation variation of young yellow perch diets remains largely unexplored. We quantified patterns of diet composition in age‐0 yellow perch collected from Saginaw Bay, Lake Huron, USA during July–October, 2009 and 2010. We observed substantial variation in diet composition among individuals across and within sites, but found relatively weak evidence indicating an ontogenetic diet shift. Zooplankton were the dominant prey for age‐0 yellow perch on most occasions, and individual diets were composed primarily of either zooplankton (e.g. Daphnia spp., Calanoida) or benthic (i.e. Chironomidae larvae, Chydoridae) prey. These patterns were not simply attributable to differences in prey availability and ontogenetic diet shifts, because a) not only diet composition, but also prey selectivity (Chesson's α) varied among sites and b) individual and spatial diet differences were evident independent of ontogeny. Within‐cohort differences in diet composition may be an important, but often overlooked, phenomenon with implications for cumulative trophic interactions and intracohort growth and survival among young fish.     

Density-dependent habitat selection and the ideal free distribution in marine fish spatial dynamics: considerations and cautions

Current methods and theory used in the study of the spatial dynamics of marine fish are problematic. Positive relationships between population abundance and occupied area are typically interpreted as evidence of density‐dependent habitat selection. However, both abundance and area may co‐vary with an un‐parameterized variable, such as a density‐independent effect. In addition, if density‐dependent habitat selection is present, population growth rates in optimal habitats would be expected to be lower than in marginal habitats. This same pattern can also evolve from a large‐scale, spatially autocorrelated change in a density‐independent factor. The theory underlying density‐dependent habitat selection, the ideal free distribution, can be tautological when no a priori information of how habitat suitability changes with density is known. In this case, an ideal free distribution can be defined for any pattern of habitat‐specific population growth rates. However, these problems are not insurmountable and solutions may be found by considering spatial variation in proxies of fitness and explicitly allowing for the relative importance of habitat selection (density dependent) and environmental (density independent) effects to vary with spatial scale.     

An investigation of the biological basis of recruitment,growth and adult survival rate variability of Pacific herring (Clupea pallasi) from British Columbia: a synthesis

I explored the biological basis of variation in recruitment (age 3 abundance), growth and age‐specific adult survival rate for the major populations [West Coast Vancouver Island (WCVI), Strait of Georgia, Central Coast, North Coast and Haida Gwaii] of Pacific herring (Clupea pallasi) that inhabit British Columbian waters. The analyses were based on a synthesis of time series of empirical observations of herring population characteristics (egg deposition, age‐specific abundance and size) and prey, competitor and predator biomass/abundance. Recruitment was not correlated among populations. Recruitment variability was explained for WCVI herring only, as a consequence of prey (the euphausiid Thysanoessa spinifera) biomass during August in each of the first 3 years of life, and the biomass of piscivorous Pacific hake (Merluccius productus) during the first year of life. Recruit mass and adult mass‐at‐age were correlated among populations and over ages within populations. Recruit mass was affected by T. spinifera biomass in August of the first and third years of life. Adult mass‐at‐age variability was determined mainly by size at the beginning of the growth season, but also by T. spinifera biomass in August. Age‐specific adult survival rates were not correlated among the five populations. Survival rates decreased with age; there were additional population‐specific effects of somatic mass and T. spinifera biomass in August. The analyses were repeated using physical oceanographic explanatory variables. Only recruit mass variation was explained significantly by physical oceanographic variables, and the biological‐based explanation of recruit mass variability accounted for more of the variation.     

Ecological processes influencing mortality of juvenile pink salmon (Oncorhynchus gorbuscha) in Prince William Sound,Alaska

Our collaborative work focused on understanding the system of mechanisms influencing the mortality of juvenile pink salmon (Oncorhynchus gorbuscha) in Prince William Sound, Alaska. Coordinated field studies, data analysis and numerical modelling projects were used to identify and explain the mechanisms and their roles in juvenile mortality. In particular, project studies addressed the identification of major fish and bird predators consuming juvenile salmon and the evaluation of three hypotheses linking these losses to (i) alternative prey for predators (prey‐switching hypothesis); (ii) salmon foraging behaviour (refuge‐dispersion hypothesis); and (iii) salmon size and growth (size‐refuge hypothesis). Two facultative planktivorous fishes, Pacific herring (Clupea pallasi) and walleye pollock (Theragra chalcogramma), probably consumed the most juvenile pink salmon each year, although other gadids were also important. Our prey‐switching hypothesis was supported by data indicating that herring and pollock switched to alternative nekton prey, including juvenile salmon, when the biomass of large copepods declined below about 0.2 g m^?3. Model simulations were consistent with these findings, but simulations suggested that a June pteropod bloom also sheltered juvenile salmon from predation. Our refuge‐dispersion hypothesis was supported by data indicating a five‐fold increase in predation losses of juvenile salmon when salmon dispersed from nearshore habitats as the biomass of large copepods declined. Our size‐refuge hypothesis was supported by data indicating that size‐ and growth‐dependent vulnerabilities of salmon to predators were a function of predator and prey sizes and the timing of predation events. Our model simulations offered support for the efficacy of representing ecological processes affecting juvenile fishes as systems of coupled evolution equations representing both spatial distribution and physiological status. Simulations wherein model dimensionality was limited through construction of composite trophic groups reproduced the dominant patterns in salmon survival data. In our study, these composite trophic groups were six key zooplankton taxonomic groups, two categories of adult pelagic fishes, and from six to 12 groups for tagged hatchery‐reared juvenile salmon. Model simulations also suggested the importance of salmon density and predator size as important factors modifying the predation process.     

Effects of Feeding Rate on Habitat Quality in Fish Rearing Ponds

Manufactured feeds are commonly added to earthen ponds to enhance growth and survival of juvenile fish. However, excessive feeding may decrease fish production efficiency and yields by causing hypoxia (dissolved oxygen, DO < 2 mg/L) and stimulating excessive phytoplankton, filamentous green algae, and vascular plant growth. In this study, we quantified the effects of manufactured feed addition (no feeding, 1%, or 3% body‐weight/day, BW/d) on DO and inorganic phosphorus (P) and nitrogen (N) concentrations, plant abundance, and invertebrate prey production in ponds stocked with age‐0 channel catfish, Ictalurus punctatus. We found that the 3% BW/d ponds had lower DO concentrations and greater infestation by filamentous green algae (Rhizoclonium spp.) as compared to the 1% BW/d and no‐feeding ponds. Using stable N and carbon (C) isotopes to trace the fate of feed‐derived N and C in ponds, as well as analysis of zooplankton abundance, we determined that the supplied feed did not support or enhance production of natural invertebrate prey. To improve fish production efficiency, we recommend that managers leverage natural prey support of growth during early life, then adjust feeding levels to enhance growth of older fish and maintain suitable habitat quality.     

The influence of ontogeny and prey abundance on feeding ecology of age-0 Lake Whitefish (Coregonus clupeaformis) in southeastern Lake Michigan

A shift towards oligotrophic conditions in Lake Michigan has led to concern that altered trophic pathways are leading to lower early life survival and recruitment for Lake Whitefish (Coregonus clupeaformis). This study evaluated ontogenetic shifts in age-0 Lake Whitefish diets and evaluated how feeding ecology and the amount of food eaten varied with prey abundance and composition at a site in southeastern Lake Michigan during 2014–2017. Although prey densities varied among years, cyclopoid copepods were overall the most abundant prey available. In turn, cyclopoids were the predominant prey item in diets each year, particularly for the smallest larval Lake Whitefish. However, there was a tendency for the importance of cyclopoids to decline somewhat in each diet index as fish grew and other prey such as calanoid copepods, Bosminidae, Daphniidae and/or chironomids increased in importance. High zooplankton abundance, especially high cyclopoid abundance, available to the small size groups of Lake Whitefish (<21 mm) in 2014 was associated with high food mass/fish, high number of zooplankton eaten/fish, and low incidence of empty stomachs compared with 2015–2017. As fish grew, the impact of food abundance on prey consumption diminished somewhat, indicating that the relationship between fish feeding ecology and the prey environment can change quickly with fish size during the early life period.     

Zooplanktivore fish body growth responses to browning-induced light limitation vary over ontogeny,but not with fish density

Ongoing climate change is leading to browning of many lakes and coastal areas, which can impair fish body growth and biomass production. However, whether and how effects of light limitation caused by browning on fish body growth vary over early ontogeny is unknown. In this study, we set up a mesocosm experiment to test whether roach (Rutilus rutilus) body growth responses to browning depend on body size, and if findings are robust over roach densities. We also studied a potential mechanism for size-specific responses by conducting an aquaria experiment to test if size-specific prey selectivity in roach changes with browning. We found that roach body growth responses to browning-induced light limitation vary over ontogeny (independent of roach density), negatively affecting body growth of young-of-the-year (YOY) but not of 1-year-old individuals. We also show that this difference in growth response is likely a consequence of browning-induced alterations in zooplankton community composition and variation in prey selectivity between YOY and 1-year-old fish. This suggests that we should account for the diverse effects of browning over fish ontogeny, mediated via altered prey composition and ontogenetic changes in prey preference, when assessing overall impacts of browning on aquatic ecosystems.     

Ontogenetic shifts in prey selection and foraging behaviour of larval and early juvenile alligator gar (Atractosteus spatula)

The early foraging ecology of alligator gar (Atractosteus spatula) is poorly documented, with little information available on young‐of‐year food habits or ontogenetic diet shifts. We conducted laboratory experiments to quantify prey selection and foraging behaviours of larval and early juvenile alligator gar (16–80 mm TL) simultaneously offered zooplankton, chironomid larvae and one of three densities of fish prey. The smallest size groups of alligator gar consumed zooplankton almost exclusively, but with increasing size, selection for zooplankton declined and selection for fish prey increased. At higher densities of fish prey, alligator gar exhibited lower selection for zooplankton and positively selected for fish at smaller sizes. Ingestion efficiencies for chironomids were considerably lower than for zooplankton or fish prey, resulting in low rates of consumption and negative selection for chironomids by all size groups of alligator gar. Fish prey elicited a different foraging response from alligator gar than zooplankton or chironomids, as alligator gar pursued and struck at fish over longer distances than for other prey types. With increasing size, alligator gar used a wider vertical range of the water column for foraging, changed their strike tactics and exhibited decreased handling times for zooplankton and fish. These observations indicate that alligator gar undergo several functional and behavioural changes during early ontogeny that facilitate a rapid transition to piscivory, but fish prey density strongly affects prey consumption patterns and the size at which alligator gar transition to piscivory.     

Plankton indices explain interannual variability in Prince William Sound herring first year growth

This study examines the relationships between first year growth of juvenile Prince William Sound herring, temperature and their food. We present time series of herring first year growth, determined from scale measurements as a proxy for herring length, water temperature and indices of multiple trophic levels of plankton obtained from Continuous Plankton Recorder (CPR) sampling on the adjacent Gulf of Alaska shelf. We show that there was a significant correlation between herring growth and water temperature, when the three warmest years were excluded (the mean July and August temperatures were greater than 12.5°C in 1989, 2004 and 2005). There were also strong, significant relationships between the abundance of appropriately sized (for first‐feeding herring) planktonic prey groups and herring growth. First year herring growth was greater in years with higher abundances of diatoms, microzooplankton and small mesozooplankton but not related to variability in abundance of larger mesozooplankton (such as euphausiids and large copepods). Furthermore, the strong interannual relationship between diatoms and herring growth held true even in the warmest years where the relationship between temperature and growth broke down. We also found seasonal timing and abundance changes in the plankton in warm years that would make the prey more abundant during the summer months immediately after metamorphosis of the herring larvae. We thus conclude that young‐of‐the‐year herring may grow better in warm years because the timing of key prey is a better match for their first feeding.     

Effects of stocking density and natural food availability on the extensive cage culture of pejerrey (Odontesthes bonariensis) in a shallow Pampean lake in Argentina

An experiment was conducted for 80 days at La Salada de Monasterio Lake (Buenos Aires, Argentina) to assess the effect of stocking density and natural food availability on the growth and production of zooplanktivorous juveniles of pejerrey (Odontesthes bonariensis) in extensive cage culture. Ten cages were installed and stocked with 33‐day‐old fish, at three density treatments: 25, 50 and 75 ind. m^?3. Zooplankton was analysed in terms of abundance, biomass and diversity considering three size classes. Caged pejerrey diet was assessed regularly. The pejerrey gut contents composition was clearly different from cage zooplankton, showing a trend to contain bigger components. Reared fish exhibited a tendency to diversification of the diet and to change the mean prey size depending on fish length and stocking density. Regression models showed a positive and direct effect of the bigger zooplankton biomass on fish growth rates, which were also inversely affected by the availability of smaller zooplankton. The results demonstrate that stocking density and available zooplankton, in both quantity and quality, are key factors in regulating extensive cage culture of pejerrey. Handling stocking densities in a dynamic way it is possible to maximize fish growth, biomass harvest or fish number according to the goals of production.     

Density‐dependent diet composition of juvenile Atlantic salmon (Salmo salar)

Abstract – Decreases in body growth with increasing population density may be caused by reduced prey delivery rate. However, changes in food quality because of an increasing inclusion of suboptimal prey in the diet may also contribute to such effects. Here, we test for density‐dependent diet composition by creating spatial variation in Atlantic salmon young‐of‐the‐year (YOY) density in three replicate streams and obtain detailed information on individual positions (±1 m) and diet. Diet breadth with respect to prey size increased with increasing local density for the two most common prey types (Chironomidae and Ephemeroptera). For the largest prey type (Ephemeroptera), there was also an increase in mean prey size with increasing density, suggesting that YOY salmon preferentially utilise the smaller prey among those available. According to optimal foraging theory, changes in diet with increasing local density are likely to come at an energetic cost and hence may contribute to the commonly observed density‐dependent growth of juvenile salmonids.     

Environmental variability controls recruitment but with different drivers among spawning components in Gulf of St. Lawrence herring stocks

The factors affecting herring recruitment are still poorly understood, complicating the prediction of stock dynamics and the choice of operational management strategies. We investigated effects of intrinsic (SSB) and extrinsic factors (physical and biological environments, including competition and predation) on recruitment of the spring and fall spawning components of each of the two herring stocks occurring in the Gulf of the St. Lawrence between 1971 and 2014. Effects of potential explanatory factors on recruit (age 2) abundance were tested using Generalized Additive Models. Model fit was significantly improved by incorporating both physical and biological environmental variability, but effects of herring SSB and predation were not significant. Indices of zooplankton abundance and phenology explained more variance in recruitment than physical indices. Our results emphasize the dominance of bottom‐up processes over SSB in the regulation of herring recruitment. Environmental variability did not seem to act uniformly on the recruitment of either stock or their respective spawning components. A long‐term trend of decreasing recruitment in spring spawners was associated with a long‐term decline in abundance of cold water copepods. In fall spawners, optimal recruitment was dependent on warmer environmental conditions combined with an adequate supply (species composition and phenology) of zooplankton. These results provide the first empirical evidence that spring and fall spawning herring are adapted to contrasting environmental conditions and shed light on the potential mechanisms linking herring recruitment to key zooplankton community characteristics and phenology. Management strategies can be improved by incorporating this new knowledge on environmental drivers of herring recruitment.     

Predation dynamics of mackerel on larval and juvenile anchovy: is capture success linked to prey condition?

We tested whether the predation dynamics of chub mackerel Scomber japonicus and spotted mackerel S. australasicus on young anchovy Engraulis japonicus relates to individual growth characteristics of the prey and could account for the growth-selective survival predicted by recruitment hypotheses. Juvenile and adult mackerel were sampled along with their young anchovy prey field in 2004 (juvenile mackerel and larval anchovy) and 2005 (adult mackerel and juvenile anchovy) off the Pacific coast of Honshu, Japan. The recent 5-day mean growth rate of larval and juvenile survivors and prey found in the stomach of mackerel was estimated from the otolith microstructure. No significant difference was found between the recent growth of larval or juvenile survivors and that of preyed individuals. We conclude that despite a relatively small body size, the high activity level and predation skills displayed by mackerel prevent fast-growing larvae and early juveniles from benefitting in terms of the expected survival advantage over slow-growers. Hence, growth-selective predation mortality of larval fish would depend on the feeding ecology of the predator rather than predator size. Selection for fast growth is more likely to occur under predation pressure from invertebrate organisms and small pelagic fish specialized on zooplankton, such as herring and anchovy.     

Body size trends along vertical and thermal gradients of chum salmon in the Bering Sea during summer

Relationships between the vertical distribution and thermal habitat, and body size of chum salmon Oncorhynchus keta were studied in the Bering Sea in summer using trawl surveys at various depths. Chum salmon abundance decreased with increasing depth, but the patterns of decrease differed between size groups. The abundance of small salmon fell rapidly with depth, whereas that of large salmon decreased gradually to 40 m depth, and abruptly below that. The average fork length of chum salmon collected from each trawl correlated positively with trawl net depth and negatively with water temperature. Since the optimal temperature for growth decreases with body size in this species, the observed body size‐related vertical habitat use by chum salmon may indicate size‐dependent thermal preferences.     

Total population density during the first year of life as a major determinant of lifetime body‐length trajectory in marble trout

Abstract – The conditions experienced early in life can strongly influence life‐history trajectories in a variety of animal species. Here, we use data from four isolated populations of the endangered stream‐dwelling salmonid marble trout (Salmo marmoratus Cuvier 1817) living in the Soca and Idrijca river basins (Slovenia) to explore the influence of the total density experienced during and after the first year of life by marble trout year‐classes on body length of marble trout through the lifetime. Analyses were performed by pooling together the stream‐specific datasets to cover a wider range of densities. Mean body length of marble trout year‐classes through the lifetime (from age 1+ to 5+) was negatively related to total density of marble trout during the first year of life. The relationship between density during the first growth period and body length through the lifetime was well described by negative power curves. Total population density after the first year of life was not correlated with body length, thus suggesting that body growth trajectories are heavily determined early in life. Given size‐dependent sexual maturity and egg production in marble trout, the relationship between density early in life and lifetime individual growth may have strong implications in terms of population dynamics and regulation of population size.     

The inverse life‐history problem,size‐dependent mortality and two extensions of results of Holt and Beverton

In 1958, Sidney Holt developed a model to determine the optimal mass at which to harvest a cohort of fish having von Bertalanffy growth and experiencing constant natural mortality. Holt and Ray Beverton then gave a life‐history interpretation to the analysis, from which Beverton developed a theory of Growth, Maturity, and Longevity (GML) that allows one to predict quantities such as age at maturity or relative size at maturity using life‐history parameters. I extend their results in two ways. First, keeping the original formulation, in which the rate of natural mortality is constant, I show how one can invert Beverton's result to determine the rate of natural mortality from life‐history data. I illustrate this inverse method with data on three species of tuna and compare the estimates with those based on tagging. Second, I extend Beverton's GML theory to include size‐dependent mortality. I explore previously published mortality models and introduce a new mortality function that has size‐independent and size‐dependent components. I show that the new size‐dependent mortality function leads to the prediction that age at maturity depends upon asymptotic size (as well as the other life‐history parameters), something that Beverton's original theory lacked. I illustrate this extension with a simple example, discuss directions for future work and conclude that nearly 60 years on these contributions of Holt and Beverton continue to lead us in new and exciting directions.     

Biogeography of pelagic food webs in the North Pacific

The tufted puffin (Fratercula cirrhata) is a generalist seabird that breeds throughout the North Pacific and eats more than 75 different prey species. Using puffins as samplers, we characterized the geographic variability in pelagic food webs across the subarctic North Pacific from the composition of ~10,000 tufted puffin meals (~56,000 prey items) collected at 35 colonies in the Gulf of Alaska (GoA) and Aleutian Archipelago. Cluster analysis of diet species composition suggested three distinct forage fish communities: (i) in the northern GoA, multiple age‐classes of coastal and shelf residents such as capelin, sand lance and herring dominated the food web, (ii) in the western GoA to eastern Aleutians, the shelf community was dominated by transient age‐0 walleye pollock, and (iii) in the western Aleutians, shelf‐edge and mesopelagic forage species such as squid, lanternfish, and Atka mackerel were prevalent. Geographic patterns of abundance of capelin and sand lance in tufted puffin diets were corroborated by independent research fisheries and diets of piscivorous fish, indicating that puffin diets reflect the local abundance of forage species, not just selection of favored species. Generalized additive models showed that habitat characteristics predict, in a non‐linear fashion, forage species distribution and abundance across two large marine ecosystems. We conclude that major biogeographic patterns in forage fish distribution follow gradients in key habitat features, and puffin diets reflect those patterns.