Submitted 19 December 2018Accepted 17 April 2019Published 15 May 2019

Corresponding authorChris Harrod chrisharrodlabnet

Academic editorDavid Nelson

Additional Information andDeclarations can be found onpage 17

DOI 107717peerj6968

Copyright2019 Pizarro et al

Distributed underCreative Commons CC-BY 40

OPEN ACCESS

Clarifying a trophic black box stableisotope analysis reveals unexpecteddietary variation in the Peruvian anchovyEngraulis ringensJessica Pizarro12 Felipe Docmac34 and Chris Harrod345

1 Facultad de Recursos Naturales Renovables Universidad Arturo Prat Iquique Chile2Departamento de Ecologiacutea Facultad de Ciencias Universidad Autoacutenoma de Madrid Madrid Spain3 Instituto de Ciencias Naturales Alexander von Humboldt Universidad de Antofagasta Antofagasta Chile4Universidad de Antofagasta Stable Isotope Facility (UASIF) Instituto Antofagasta Universidad deAntofagasta Antofagasta Chile

5Nuacutecleo Milenio INVASAL Concepcioacuten Chile

ABSTRACTBackground Small fishes play fundamental roles in pelagic ecosystems channellingenergy and nutrients from primary producers to higher trophic levels They supportglobally important fisheries in eastern boundary current ecosystems like the HumboldtCurrent System (HCS) of the SE Pacific (Chile and Peru) where fish catches are thehighest in theworld (per unit area) This production is associatedwith coastal upwellingwhere fisheries target small pelagic fishes including the Peruvian anchovy (Engraulisringens) The elevated biomass attained by small pelagics is thought to reflect their lowtrophic position in shortsimple food chains Despite their global importance large gapsexist in our understanding of the basic ecology of these resources For instance there isan ongoing debate regarding the relative importance of phytoplankton versus animalprey in anchovy diet and ecosystem models typically assign them a trophic position(TP) ofsim2 assuming they largely consume phytoplankton Recent work based on bothrelative energetic content and stable isotope analysis (SIA) suggests that this value is toolow with δ15N values indicating that anchovy TP is ca 35 in the Peruvian HCSMethods We characterised the trophic ecology of adult anchovies (n = 30) theirputative prey and carnivorous jack mackerel (n= 20) captured from N Chile SIA(δ13C and δ15N) was used to estimate the relative contribution of different putativeprey resources δ15N was used to estimate population level trophic positionResults Anchovies showed little variability in δ13C (minus187 to minus161h) but variedgreatly in δ15N (138 to 228h)mdashindividuals formed two groups with low or highδ15N values When considered as a single group mixing models indicated that anchovydiet was largely composed of zooplankton (median contribution 95 credibilitylimits) with major contributions of crustacean larvae (061 037ndash077) and anchovy(preflexion) larvae (015 002ndash034) and the assimilation of phytoplankton wasnegligible (005 0ndash022) The modal (95 credibility limits) estimate of TP for thepooled anchovy sample was 323 (293ndash358) overlapping with recent SIA-basedestimates from Peru When the two δ15N groups were analysed separately our resultsindicate that the lower δ15N group largely assimilated materials from crustacean larvae(073 042ndash088) with a TP of 291 (262ndash323) Mixing models suggested high δ15N

How to cite this article Pizarro J Docmac F Harrod C 2019 Clarifying a trophic black box stable isotope analysis reveals unexpecteddietary variation in the Peruvian anchovy Engraulis ringens PeerJ 7e6968 httpdoiorg107717peerj6968

anchovies were cannibalistic consuming anchovy preflexion larvae (055 011ndash074) Acarnivorous trophic niche was supported by high TP (379 348ndash416) mirroring thatof carnivorous juvenile jackmackerel (Trachurus murphyi 380 351ndash414) Our resultssupport recent conclusions regarding high TP values of anchovy from Peru and revealnew insights into their trophic behaviour These results also highlight the existenceof cryptic trophic complexity and ecosystem function in pelagic food webs classicallyconsidered as simple

Subjects Ecology Marine BiologyKeywords Trophic position Cryptic ecology Small pelagic fishes Food webs Mixing modelsChile S E pacific Humboldt current

INTRODUCTIONBoundary current ecosystems such as the Humboldt Current System (HCS) arecharacterised by high biological productivity driven by coastal upwelling of cold sub-surface nutrient-rich waters (Chavez amp Messieacute 2009) Food chains in such ecosystemshave been typically considered as simple and short (Ryther 1969) with high efficiency oftrophic transfer between trophic levels Here phytoplankton form the base of the food weband are consumed by zooplankton and small-bodied pelagic fishes such as the Peruviananchovy Engraulis ringens Jenyns 1842 (from hereon in anchovy) The small pelagic fishassemblage is typically dominated by one or a few species (Bakun 1996) which due totheir sheer abundance and biomass can exercise control over the trophic dynamics of thewhole ecosystem (Cury et al 2000)

Biological production in the HCS (Herrera amp Escribano 2006) is such that it supportsthe capture of more fish per unit area than any other environment in the world (Chavezet al 2008) Indeed production in the HCS is considered anomalous even among easterncurrent systems (Bakun ampWeeks 2008 Chavez amp Messieacute 2009) Industrial fisheries inPeru and Chile target anchovies and other pelagic fishes such as jack mackerel Trachurusmurphyi and sardines (Sardinops sagax) The anchovy stock in the region is the mostheavily exploited fish worldwide although annual catches vary considerably reflecting themarked effects of ENSO on stock size between 1990 and 2015 on average (medianplusmn IQR)7419295 plusmn 3329637 tonnes of anchovies were removed per year (FAO 2018)

Although anchovy (and other small pelagics) support globally important fisheries(Chavez amp Messieacute 2009 Kaumlmpf amp Chapman 2016) and play integral roles in structuringupwelling ecosystems (Cury et al 2000) many basic aspects of their ecology areunderstudied For instance information on their trophic ecology is limited and likelyconstrains our understanding of their role in the HCS and the utility of ecosystem modeloutputs Studies of anchovy diet based on stomach content analysis (SCA) of juvenilesand adults typically report that stomach contents are dominated (in numerical terms)by phytoplankton (Espinoza amp Bertrand 2008 Medina et al 2015 Ryther 1969) andanchovies have been classically considered as phytoplanktivores This low trophic position

Pizarro et al (2019) PeerJ DOI 107717peerj6968 222

has been proposed as an explanation for their great abundance in the HCS and theecosystemrsquos anomalous capacity to support such high production of pelagic fishes

Most studies of anchovy diet in the literature are based on counts of individual stomachcontents which is known to bias conclusions when prey of markedly different sizes areconsumed (Hyslop 1980) Not surprisingly microscopic food items such as diatoms can beextremely abundant leading them to dominate relative estimates of prey importance (upto 98Whitehead Nelson amp Wongratana 1998) even though zooplankton are commonlyreported from anchovy stomach contents Given the putative importance of phytoplanktonto their diet many researchers have assumed that anchovy have a trophic position of ca2ndash25 (Gueacutenette Christensen amp Pauly 2008 Pauly et al 1998) These values are commonlyused in ecosystem models such as Ecopath and anchovies are considered a classic low-trophic level species (Smith et al 2011) This estimate of anchovy TP has played animportant role in one of the largest controversies in modern fisheries sciencemdashthe issueof global fishing down the food web (Pauly et al 1998 Pauly Froese amp Palomares 2000)The global importance of the HCS anchovy fishery is such that if its statistics are includedit skews global estimates of the mean trophic level of catch downward

More recently however there has been an important reassessment of anchovy dietEspinoza amp Bertrand (2008) rather than counting stomach contents focused on theenergetic content of prey and clearly showed that zooplankton (mainly copepods andeuphausiids) were the principle source of energy to anchovies Beyond highlightinga key issue of using counts of stomach contents to estimate TP their results had majorimplications regarding our understanding of how this globally important marine ecosystemfunctions (eg Balloacuten et al 2011) Given the biomass of anchovies in the HCS and theirdominant role in the food web changes in our understanding of their putative diet hassubsequent impacts on how we interpret the movement of energy and nutrients throughthe food web with consequences for qualitative and quantitative models of HCS functionand resource management

SCA has long been the gold standard for assessing what fish consume (Hynes 1950Hyslop 1980) but clearly can bias our understanding of how fish direct the flow ofmaterials through a food web (Espinoza amp Bertrand 2008 Hyslop 1980) An alternative tothe snapshot of recently ingested prey provided by SCA is to take a biochemical approachto assessing diet such as stable isotope analysis (SIA) or fatty acid analysis (Nielsen et al2018) The advantages of SIA in particular are that the technique provides information onprey assimilation over longer temporal scales than SCA (weeksndashmonths) with the perioddepending on the tissue sampled (Thomas amp Crowther 2015) By combining analysis ofcarbon and nitrogen stable isotope ratios of consumers and their putative prey it is possibleto characterise the source of energy and nutrients assimilated by a consumer (Parnell et al2013) Furthermore if isotope values are available for both the consumer and the base ofthe food web (Vander Zanden amp Rasmussen 1996) it is possible to estimate the long termtrophic position at which a consumer feeds (Quezada-Romegialli et al 2018)

Recent work in central Chile and Peru using SIA has provided further evidence thatprevious assumptions regarding anchovy trophic position were wrong Huumlckstadt Rojasamp Antezana (2007) compared POM and anchovy δ15N values from central Chile and

Pizarro et al (2019) PeerJ DOI 107717peerj6968 322

estimated anchovy TP as 36 In a wide-ranging recent study from the northern HCSEspinoza et al (2017) compared anchovy (and other taxa) δ15N values relative to copepodδ15N to estimate TP They estimated that anchovy TP ranged between 34 and 37 Bothof these studies provide evidence that anchovy TP is at least a trophic level above valuestypically used for the species in trophic models (Gueacutenette Christensen amp Pauly 2008 Paulyet al 1998) and challenge the concept of anchovies as a low trophic level fish (Smith et al2011) Espinoza et alrsquos (2017) estimates for anchovy TP are important as they support theauthorsrsquo previous findings from SCA indicating that phytoplankton although abundantnumerically in stomachs did not make a major contribution to assimilated diet Howevertheir and Huumlckstadt Rojas amp Antezanarsquos (2007) estimates were based on a simple modelwhich does not include isotopic variation in trophic discrimination or in the baselineitself (see Quezada-Romegialli et al 2018 for a description of the issue) The latter point isparticularly important in the Espinoza et al (2017) case as the use of copepods as a baselinelikely introduces considerable error compared to eg the use of a primary producer giventhe range of trophic strategies displayed by pelagic marine copepods (Giesecke amp Gonzaacutelez2004) and the uncertainties in allocating a TP for mixed samples of copepods

It is becoming increasingly apparent that foodwebs associatedwith upwelling ecosystemscan be more complex and dynamic than previously thought (Docmac et al 2017 Espinozaamp Bertrand 2008) As such there is a need for improved understanding of how thesesystems function in order to inform and update existing trophic models used to explainthe flow of energy and nutrients as well as to allow an informed management of a globallyimportant fishery Here we examine the trophic ecology of adult anchovies from northernChile using stable isotope ratios of anchovies and their putative prey to estimate the relativerole of phytoplankton and other prey and to provide robust estimates of anchovy trophicposition Given the debate over the trophic ecology of anchovy we compared their stableisotope values with that of juvenile jack mackerel a known pelagic carnivore (Alegre et al2015 Orrego amp Mendo 2015)

MATERIALS amp METHODSStudy areaSamples were collected between 2030primeS and 2130primeS off the coast of northern Chile(Fig 1) during the Austral winter of 2008 This area is characterized for having persistentwinds that permit year-round upwelling (although upwelling often strengthens duringSpringndashSummer) generating the intrusion of cold nutrient rich sub-surface waters alongthe shore (Thiel et al 2007)

During the study period neutral environmental conditions were present (ie non-ENSO) Sea surface temperatures ranged between 152 and 175 C salinities variedbetween 346 and 349 and surface dissolved oxygen concentrations between 42 and 67ml lminus1 In general terms the study area was characterized by a low-magnitude permanentupwelling more marked around a latitude of 2110primeS (Fuenzalida et al 2009)

Pizarro et al (2019) PeerJ DOI 107717peerj6968 422

-23

-22

-21

-20

-19

-18

-71 -70 -69 -68

Tacna

Arica

Iquique

Tocapilla

Mejillones

Antofagasta

N

BOLIVIA

CHILE

PERU

Figure 1 Study area and position of sampling stations off the coast of N ChileDark markers represent the location where putative prey were cap-tured while the light grey markers show the capture location of the anchovy and jack mackerel used in the study

Full-size DOI 107717peerj6968fig-1

Sample collectionFish were captured (Fig 1) at 2057primeSndash7022primeW (adult anchovies) and 2059primeSndash7023primeW(juvenile jack mackerel to allow comparisons with a known carnivorous fish) by the fishingvessel Atacama V using a commercial purse seine (mesh 15 mm) Once captured fishwere frozen at minus20 C until further analysis Permission to undertake field sampling was

Pizarro et al (2019) PeerJ DOI 107717peerj6968 522

provided by the Chilean Subsecretariacutea de Pesca through Resolucioacuten Exenta No 2205 (21August 2008)

Samples of putative planktonic prey (phytoplankton and zooplankton) were collectedduring daylight hours following a grid design along four longitudinal transects (latitudinaldistance between transects 15 nautical miles) located perpendicular to the N Chileancoast (Fig 1) Each longitudinal transect included four stations located at 1 3 5 and 10nautical miles from the coast resulting in a total of 16 sampling sites Phytoplankton wascollected via vertical tows (from 50 m) of a phytoplankton net (20 microm mesh size 20 cmopening) Zooplankton was collected through vertical tows (max depth 100 m but variedwith sampling station depth) of a WP-2 net (300 microm mesh size 50 cm opening) Sampleswere immediately frozen at minus20 C and following transfer to the laboratory placed in aminus80 C freezer

Laboratory analysisOnce defrosted total length (TL plusmn1 mm) and blotted wet mass (plusmn01 g) were estimatedfor anchovy (n= 30) and jack mackerel (n= 20) Stomachs were removed for analysis ofstomach contents but on inspection a large majority of anchovy stomach contents werefound to be in an advanced stage of digestion limiting the utility of SCA in this case andwe do not consider these data further in this study

A sample of dorsal muscle was excised and treated with chloroformmethanol (21) toextract lipids (Bligh amp Dyer 1959) Samples were then oven dried (60 C for 48 h) andhomogenised prior to analysis of carbon and nitrogen stable isotope values

Prior to SIA phytoplankton samples were pre-filtered through a 200 microm sieveto remove zooplankton and large detritus Zooplankton were sorted into differentcomponents anchovy eggs and larvae unidentified fish eggs amphipods appendicularianschaetognaths copepods crustacean larvae euphausiids ostracods pelagic polychaetessalpidae and siphonophora Each of these groups were pooled to obtain sufficient mass forSIA

Phytoplankton and zooplankton samples were washed with milli-Q water and thenfiltered through pre-combusted (450 Ctimes 4 h) GFF filters (Whatmann 07 microm pore size47 mm diameter) Filters were subsequently oven dried (60 C for 48 h) then acidified bydirectly applying HCl (1N) for 24 h to remove inorganic carbon and re-dried (Carabel etal 2006) Zooplankton samples were then scraped off filters prior to homogenisation Fishand zooplankton samples were homogenised with an agate mortar and pestle and weighedinto tin capsules (sample mass sim05 mg) Phytoplankton samples were run from sectionsof GFF filters

Analysis of δ13C and δ15N were conducted at the Colorado Plateau Stable IsotopeLaboratory in Northern Arizona University (USA) using a Costech ECS4010 elementalanalyser coupled to a Delta Plus Advantage isotope ratio mass spectrometer in continuousflow mode via a ConFlo III interface Isotopic abundances are expressed in δ notation (h)using the formula δ13C or δ15N = [(Rsample ndash R standard)Rstandard] times103 where Rsample

is 13C12C or 15N14N Rstandard is Vienna Pee Dee Belemnite for δ13C and atmosphericnitrogen for δ15N NIST 1547 (peach leaves) were used as an internal laboratory working

Pizarro et al (2019) PeerJ DOI 107717peerj6968 622

standard to check on measurement reproducibility throughout each run (analyticalprecision of le01h for δ13C and le 02h for δ15N) A number of peach leaf standardsthat varied in mass (from 05 to 6 mg) were also included at the end of the run to checkon linearity Analytical data were normalized to international standards using calibratedreference standards with known δ13C (IAEA CH6 amp IAEA CH7) or δ15N (IAEA N1 amp IAEAN2) values

Data analysisData were plotted in the form of an δ15N-δ13C scatterplot to provide a visual representationof the distribution of isotopic values for the different taxa examined and to examine putativedifferences between anchovy and jack mackerel Individual data were plotted for fish whileputative prey were plotted as means plusmn SD Size differences between the fish specieswere examined using the Welchrsquos t -test We examined potential relationships between fishisotope values and their individual bodymass using Spearmanrsquos rank correlation coefficientBothWelchrsquos t-tests and correlationswere conducted in SYSTAT131 (SYSTAT RichmondCA USA) Unless otherwise reported summary statistics reflect mean plusmn 1 SD

In order to examine whether the two fish species differed isotopically we comparedδ15N-δ13Ccentroids of anchovies and jackmackerel using a permutation-basedmultivariateanalysis of variance (PERMANOVA npermutations= 9999) based on a Euclidean similaritymatrix of untransformed δ15N-δ13C data (Anderson Gorley amp Clarke 2008) Due to anapparent sub-structuring within the anchovies based on δ15N values (see below) wealso tested for differences between the two observed groups of anchovies using a similarapproach PERMANOVA analyses were conducted in PRIMER with PERMANOVA 7013(Anderson Gorley amp Clarke 2008 Clarke amp Gorley 2015)

We used the R-based (version 350 R Core Team 2018) Bayesianmixingmodel SIMMRto estimate the relative contribution of seven different key putative prey groups to anchovyand jack mackerel assimilated diet (Parnell 2016) These groups were selected reflectingtheir abundance in zooplankton hauls and literature descriptions of anchovy diet (Espinozaamp Bertrand 2008 Medina et al 2015) SIMMR was run in RStudio (version 11447RStudio Team 2016) using default settings for the number of iterations burn in andMarkov chain Monte Carlo (MCMC) chains Putative prey were grouped a priori asanchovy eggs anchovy larvae copepods crustacean larvae euphausiids unidentified fisheggs and phytoplankton We used mean plusmn SD trophic discrimination factors from Post(2002) (113C = 04 plusmn 13h 115N = 34 plusmn 10h) and included information on preyelemental concentrations

We estimated TP of anchovy and jack mackerel in R using the Bayesian packagetRophicPosition (Quezada-Romegialli et al 2018) using the Onebaseline model Brieflythe tRophicPositionmodel includes isotopic variation in the baseline indictor the consumerand in the trophic discrimination in order to provide a robust estimation of consumer TPat the population level We included the phytoplankton δ15N data to provide the baseline(λ= 1) and assumed a trophic discrimination factor (115N) of 34 plusmn 1h(Post 2002)

Pizarro et al (2019) PeerJ DOI 107717peerj6968 722

-22 -21 -20 -19 -18 -17 -16 -15

12

14

16

18

20

22

δ13C (permil)

δ15 N

(permil)

Phytoplankton

Fish eggs (unidentified)

Crustacean larvaeCopepods

Euphausiids Anchovy eggs

Anchovy larvae

Figure 2 Stable isotope δ15N - δ13C biplot showing individual values for adult anchovies (open trian-gles) and juvenile jack mackerel (filled triangles) captured during the current studymdashputative prey val-ues are shown as mean (plusmnSD) Note the considerable inter-individual differences in δ15N shown by an-chovies Fish δ13C values reflect lipid-extracted tissues while putative prey were not treated as consumersassimilate carbon from different biochemical compounds in their prey including lipids

Full-size DOI 107717peerj6968fig-2

RESULTSIsotopic values for the different taxa analysed are shown in Fig 2 Phytoplankton hada mean δ13C value of minus186h but showed considerable variation between samplinglocations (SD = 16h) Phytoplankton mean δ15N was 124 (plusmn 13h) reflecting theinfluence of 15N-enriched upwelling-derived NO3 and apparently formed the basalresource for consumers from overlying trophic levels Carbon stable isotope values forthe different putative zooplankton prey classes (anchovy eggs and larvae unidentified fisheggs copepods crustacean larvae euphausiids) ranged between minus203 and minus179h forδ13C and 151 to 197h for δ15N

Carbon and nitrogen stable isotope values were estimated from 30 adult anchovies (28male 2 female) Anchovy TL varied between 137 and 161 cm (mean TL= 148plusmn 06 cm)

Pizarro et al (2019) PeerJ DOI 107717peerj6968 822

Lipi

d-tre

ated

δ13

C (permil

)

δ15N

(permil)

10 20 30 40 50 60 70 80Mass (g)

-19

-18

-17

-16

10 20 30 40 50 60 70 8010

15

20

25

Mass (g)

A) B)

Figure 3 Scatterplot showing a lack of any obvious relationship between (A) δ13C and (B) δ15N andfish mass in anchovy (low δ15N= red triangles high δ15N= blue triangles) and juvenile jack mackerel(black triangles)

Full-size DOI 107717peerj6968fig-3

and blotted wet mass between 139 and 257 g (195 plusmn 29 g) Juvenile jack mackerel (onemale one female 18 indeterminate) TL varied between 185 and 222 cm (195 plusmn 11 cm)and mass between 432 and 798 g (537 plusmn 98 g)

Anchovy lipid-free δ13C values varied between minus187 and minus161h with a mean ofminus171 plusmn 06h Variation in δ15N values was more marked with individual anchoviesranging between 138 and 228h and a mean δ15N of 197 (plusmn19)h Examination ofanchovy δ15N values revealed the presence of two apparent groups differing in their δ15Nvalues one group was relatively 15N enriched (216plusmn 09h) and the second was made upof 15N-depleted individuals (185 plusmn 13h)

Jack mackerel lipid-free δ13C values ranged between minus184 and minus176h (mean δ13Cminus180 plusmn 02h) while their δ15N values ranged between 183 and 222h with a meanδ15N of 216 (plusmn 09)h apparently similar to that of the high δ15N group of anchovies

There was no evidence of correlation between the isotopic composition and mass(Fig 3) of anchovy (δ13C RS=minus002 P = 094 δ15N Rs= 011 P = 056) Jack mackerelδ13C was similar across the size range examined (Rs =minus027 P = 027) but showeda positive relationship between mass and δ15N (δ15N Rs = 055 P = 001) Anchovieswere significantly smaller than jack mackerel (Welchrsquos t test of mass t281 =minus151P lt 0001) but there was no evidence for any size difference between the two anchovygroups (t172= 070 P = 0493)

Isotopic differencesPERMANOVA indicated that δ13C-δ15N centroids differed between the anchovy andjack mackerel (Pseudo-F149 = 206 P = 00001) When anchovies are considered astwo different groups (high and low δ15N) there was again strong evidence for isotopicdifferences between the three groups of fish compared low δ15N adult anchovies high δ15N

Pizarro et al (2019) PeerJ DOI 107717peerj6968 922

adult anchovies and juvenile jackmackerel (Pseudo-F249= 452P = 00001) Pairwise testshowever revealed significant overlap between juvenile jack mackerel and the individualsin the high δ15N anchovy group (P = 011) Anchovies from the low δ15N group wereisotopically different from both (P = 00001 in both cases)

A univariate PERMANOVA showed significant differences in δ13C between groups(Pseudo-F249= 303 P = 00001) with higher δ15N anchovies being ca 06h depleted in13C compared to individuals from the lower δ15N group A similar analysis showed thatthe three groups of fish differed in their δ15N values (Pseudo-F249= 479 P = 00001) andthat the group of anchovies lower in δ15Nwas 15N depleted by ca 3h (close to one trophiclevel given a TDF of 34h) compared to the higher δ15N anchovies and the jack mackerel(Pairwise P = 00001) δ15N values overlapped between the high δ15N anchovy group andthe jack mackerel (P = 074) suggesting a similar feeding mode

Mixing modelsMixing models results (Fig 4 Table 1) indicate that all fish examined were carnivorouswith little evidence for large-scale contributions by phytoplankton When considered asa single group anchovies largely assimilated carbon and nitrogen from crustacean larvae(median estimate= 61) and anchovy larvae (15) with phytoplankton making a minorcontribution (5) When considered separately the anchovy group with high δ15N valueshad a large contribution of anchovy larvae in their diet (55) while the low δ15N anchovygroup were estimated to have assimilated the majority of their carbon and nitrogen fromcrustacean larvae (73) Mixing model results indicated that anchovy larvae (39)unidentified fish eggs (19) and anchovy eggs (15) all made major contributions to theassimilated diet of jack mackerel Comparison between the mixing model results for thehigh δ15N group of anchovies and jack mackerel indicated that they had broadly similarforaging modes

Trophic positionBayesian estimates of TP supported the mixing model results with credibility limits for allfish including TP 3 (Fig 5) Modal (95 credibility limits) estimated TP for the pooledanchovy sample was 323 (293ndash358) The low δ15N anchovy group had a lower estimatedTP (291 262ndash323) Estimates of TP for the high δ15N anchovy group (379 348ndash416)and jack mackerel (380 351ndash414) were highly similar

DISCUSSIONOur study aimed to further our understanding of the trophic ecology of the Peruviananchovymdasha fish that supports the most productive fishery in the world Existing dataon anchovy diet like that of other pelagic fishes is largely derived from the analysis ofstomach contents and has generally concluded that anchovy production in the studyregion is fuelled by the consumption of phytoplankton (Medina et al 2015) and copepods(Castillo et al 2002 Castillo et al 2011) Our results based on stable isotope valuesindicate that the contribution of phytoplankton to anchovy somatic tissues is minimal andthat zooplankton represents the main food source assimilated by anchovies Our results

Pizarro et al (2019) PeerJ DOI 107717peerj6968 1022

B) Jack mackerel

000 025 050 075 100

D) High δ15N anchovy

000 025 050 075 100Proportion

A) All anchovy

Anchovy eggs

Anchovy larvae

Copepods

Euphausiids

Fish eggs (unidentified)

Phytoplankton

Crustacean larvae

000 025 050 075 100

C) Low δ15N anchovy

000 025 050 075 100Proportion

Anchovy eggs

Anchovy larvae

Copepods

Euphausiids

Fish eggs (unidentified)

Phytoplankton

Crustacean larvae

Figure 4 Visual representation of SIMMRmixing model results Boxplots showing the distribution ofposterior estimates of contribution of the different putative prey to assimilated diet to (A) all anchovy (B)jack mackerel (C) low δ15N anchovies and D) high δ15N anchovies

Full-size DOI 107717peerj6968fig-4

follow those of Espinoza and colleagues (Espinoza amp Bertrand 2008 Espinoza et al 2017)who have elegantly demonstrated that a reliance on stomach content data has resulted in asignificant misinterpretation of anchovy trophic ecology and by extension the food web ofthe HCS Beyond our reaffirmation of Espinoza et alrsquos observations (2017) regarding thetrophic position of anchovy we have also identified several previously unknown featuresof the pelagic food web of the Chilean HCS suggesting that the trophic ecology of thePeruvian anchovy and ecosystem function in the HCS are both more complex than istypically considered

The HCS has several characteristic features associated with elevated fisheries productionUsing stable isotopes of anchovies jack mackerel and their putative prey as well as thekey pelagic primary producer phytoplankton we were able to highlight the influence ofupwelling throughout the food chain in theChileanHCS Phytoplankton andhigher trophiclevels were naturally-labelledwith heavy nitrogen associatedwith upwelling (Casciotti 2016Docmac et al 2017 Reddin et al 2015) and results indicate that phytoplankton was thelikely carbon source fuelling upper trophic levels in the HCS We used phytoplankton

Pizarro et al (2019) PeerJ DOI 107717peerj6968 1122

Table 1 Summary statistics (median (credibility limits)) for estimated contribution of different puta-tive prey to the assimilated diet of anchovy (all anchovy combined low δ15N and high δ15N groups) andjack mackerel estimated using SIMMR

Estimated proportional contribution to diet

Putative prey All anchovy High δ15N anchovy Low δ15N anchovy Jack mackerel

Anchovy eggs 006(001ndash026)

008(001ndash036)

004(001ndash015)

015(002ndash036)

Anchovy larvae 015(002ndash034)

055(011ndash074)

003(001ndash013)

039(018ndash059)

Copepods 003(001ndash01)

004(001ndash014

003(001ndash013)

005(001ndash013)

Crustacean larvae 061(037ndash077)

006(001ndash023)

073(042ndash088)

009(001ndash023)

Euphausiids 002(0001ndash008)

004(001ndash015)

002(0004ndash008)

005(001ndash014)

Unidentified fish eggs 003(0001ndash010)

009(001ndash044)

002(0004 ndash007)

019(005ndash034)

Phytoplankton 005(0001ndash022)

005(001ndash017)

007(001ndash036)

006(001ndash018)

2

3

4

All anchovy

Jackmackerel

Low δ15Nanchovy

High δ15Nanchovy

Trop

hic

posit

ion

323(293 minus 358)

379(348 minus 416)

380(351 minus 414)

291(262 minus 323)

Modal TP95 CL

Figure 5 Estimates of trophic position for anchovies and jack mackerel Bars show the modal TP (cir-cle)plusmn95 credibility limits calculated relative to a phytoplankton baseline (TP1) using tRophicPosition

Full-size DOI 107717peerj6968fig-5

collected across the survey area both as a putative food source in mixing models and asa baseline for the estimation of trophic position Phytoplankton δ13C values varied byca 3h between locations likely reflecting local differences in primary production ratesdue to variation in upwelling intensity CO2 concentration phytoplankton growth rateetc (Magozzi et al 2017) and possibly community composition (Fry amp Wainright 1991)It is notable that several taxa examined here were notably 15N-enriched relative to their

Pizarro et al (2019) PeerJ DOI 107717peerj6968 1222

counterparts in the northern HCS as reported by Espinoza et al (2017) (eg copepods by75h euphausiids by 76h anchovies by 76h and jack mackerel by 41h) This likelyreflects regional differences in upwelling intensity

Anchovies have typically been considered to have homogeneous diets and thereforetrophic position (Medina et al 2015) Using SIA however we have shown the existenceof considerable inter-individual differences in δ15N a robust indicator of trophicposition (Post 2002) The differences between the two approaches presumably reflectsan overestimation on the contribution of phytoplankton due to using count data (Espinozaamp Bertrand 2008) as well as the different timescales associated with SCA (hours) comparedwith SIA (months) (Nielsen et al 2018 Thomas amp Crowther 2015) but highlights thepower of SIA to identify cryptic individual variation (Bolnick et al 2002 Harrod etal 2005)

The observed variability in δ15N values for adult anchovies very likely reflects differencesin individual feeding strategies We identified two trophic groups one relatively enrichedin 15N associated with the consumption of anchovy larvae and another depleted in 15Nthat largely consumed crustacean larvae Anchovies from the first group had similar δ15Nvalues to juvenile jack mackerel implying that they fed at a similar TP however theydiffered in δ13C values likely indicating resource segregation Anchovies in the lower δ15Ngroup had a modal (95 credibility interval) estimated TP of 291 (262ndash323) indicatingsome omnivory but a largely carnivorous diet Anchovies in the higher δ15N group hada modal value of 379 (348ndash416) which overlapped completely with carnivorous jackmackerel (380 (351ndash414)) indicating that they are functionally similar Support for ahigh TP for some anchovy is provided by Espinoza et al (2017) who also showed overlapbetween jack mackerel δ15N values and anchovies with high δ15N values from southernPeruvian latitudes

When the two anchovy groups were pooled estimated TP (323 (293ndash358)) was ca 1TP higher than values estimated from stomach contents data and widely used in modellingstudies (Gueacutenette Christensen amp Pauly 2008 Pauly et al 1998) However these pooledestimates effectively overlap with values from other studies using stable isotope analysissuch as Huumlckstadt Rojas amp Antezana (2007) working in central Chile (anchovy TP = 36)and Espinoza et al (2017) from the Peruvian part of the HCS (34ndash37)

Our sample of anchovy was relatively small (n= 30) with individuals falling into twobroad groups based on δ15N values It is unknown whether the patterns in inter-individualvariability in anchovy δ15N values we have shown are typical of the species or even thegenus However in their Fig 5 Espinoza et al (2017) show both a similar range of δ15Nvalues (12ndash20h) as well as possible evidence for different δ15N groups from anchoviesfrom latitudes ca 300ndash350 km north of where our samples were collected It is possible thatfurther intensive sampling may reveal that anchovy feed along a δ15N continuum ratherthan in discrete groups as we suggest It is likely that this is worthy of further study in otherpopulations and Engraulis species

An alternative explanation for the existence of two groups of anchovy differing inδ15N (which we associate with differences in TP) is that the sample included fish fromdifferent geographical locations where baseline δ15N differed but that had actually fed

Pizarro et al (2019) PeerJ DOI 107717peerj6968 1322

at similar TPs Baseline δ15N varies considerably along the Pacific coast of Chile andPeru reflecting differences in the intensity of upwelling and the associated denitrificationprocess De Pol-Holz and colleagues (2009) reported a marked SouthndashNorth gradient of15N enrichment in marine sediments along the Chilean coast We have shown elsewhere(Docmac et al 2017) that consumer δ15N increases along a South-North gradient in thestudy region and that δ13C co-varies spatially across the same scale Given that the twoputative groups of anchovy had similar δ13C values we feel it is unlikely that they haddifferent origins and that we are confident in our identification of the two groups feedingat different trophic positions

There was no evidence for any relationships between individual body mass and δ13Cor δ15N in anchovy as seen in the northern HCS by Espinoza et al (2017) who examinedfish from a larger size range In our study jack mackerel δ13C was similar across the sizegradient but showed a significant but small (06h) positive shift in δ15N from the smallest(432 g) to the largest (798 g) individual examined In contrast to our results Espinozaet al (2017) again in a wider size range showed that jack mackerel δ15N was negativelyrelated with individual size in the northern HCS In order to have a reliable picture ofontogenetic shifts in δ13C and δ15N in N Chile it is clear that larger samples are requiredboth in terms of the number and of the size of individual fish

Our estimates of anchovy consumption patterns and TP using stable isotope ratios arebased on several key assumptions For example they are reliant on the use of representativevalues for the different prey groups and the isotopic baseline (phytoplankton) used toestimate TP as well as for trophic discrimination (isotopic differences between prey andthe consumer) With regard to our estimates of putative prey and the isotopic baseline wefollowed a standardised sampling protocol across a considerable area (gt2000 km2) whichis likely representative for the wider region Such wide-scale sampling should be sufficientto account for spatial differences in phytoplankton and prey δ15N associated with variationin in upwelling intensity and hence the relative enrichment in 15N associated with thisprocess (Reddin et al 2015)

However upwelling intensity (and the relative amount of 15N-enrichment) in theregion also varies on a temporal scale (Herrera amp Escribano 2006) raising the possibilityof a disconnect between phytoplankton (which rapidly reflect isotopic changes) andanchovies whose tissue δ15N values may reflect conditions in an earlier time periodeg when phytoplankton δ15N was different than during the study period Such isotopicdecoupling has been reported elsewhere (OrsquoReilly et al 2002) and if not identified canpotentially confuse estimates of fish TP We do not think this is a major issue in ourstudy as the presence of favourable winds year-round means than coastal upwelling occursthroughout the year in the study region (Palma Escribano amp Rosales 2006 Thiel et al2007) This suggest that there is unlikely to be a marked seasonal shift in δ15N values at thebase of the food web that could lead to us underestimating baseline δ15N Furthermoreanchovy growth is rapid in the study region and continues throughout the year (Cerna ampPlaza 2016) lowering the risk of seasonal decoupling from a dynamic isotopic baseline

In terms of TDF we used a commonly applied δ15N TDF of 34 plusmn 10h to allow directcomparisons with Espinoza et al (2017) This value represents a mean estimated from

Pizarro et al (2019) PeerJ DOI 107717peerj6968 1422

many different habitats feeding strategies and taxa (Post 2002) Studies of carnivorous fishoften use a smaller δ15N value of 29 plusmn 03h following McCutchan et al (2003) Use ofthese values would increase our estimates of anchovy TP Beyond the actual mean valuesused it is also important to note that our estimates of TP were calculated following arobust Bayesian approach that includes error in both diet-tissue TDF and baseline δ15N(Quezada-Romegialli et al 2018) Our approach however relies on the assumption thatTDFs are additive and this has been criticised by some authorsCaut Angulo amp Courchamp(2009) suggested that TDFs decrease with greater concentrations of 15N in the diet andHussey et al (2014) developed a model to estimate what they refer to as scaled-TP Theirmodel includes a threshold (named δ15 Nlim) estimating the situation where δ15N in thediet is such that trophic discrimination is expected to be zero δ15 Nlim is calculated basedon estimates for the slope (β0) and the intercept (β1) of a negative relationship betweenconsumer TDF and the δ15N of their food based on experimental feeding studies Using theBayesian median estimates provided by Hussey and colleagues based on a meta-analysisδ15 Nlim has a value of 219h Many consumers in the HCS of N Chile including benthicrockfish (Docmac et al 2017) and the pelagic fishes detailed here (jack mackerel and highδ15N anchovies) have δ15N values greater than this threshold This not only prevents thecalculation of scaled TP estimates in these naturally 15N-enriched consumers but highlightsthat the relationship between experimentally-derived TDF and dietary δ15N at the heartof the scaled-TP approach does not extend to natural ecosystems where δ15N is naturallyhigh Although this does not discount the Hussey et al (2014) method it does highlightthat it is not universally applicable and raises the need for more experimental studies

Our mixing model results reflect an attempt to estimate the relative contribution ofseven different putative prey groups using only two different isotopes Although Bayesianmixing models can provide useful results where the number of potential sources is greaterthan the number of markers used (Phillips et al 2014) the performance of our modelmight have been limited by both the large number of prey groups and the fact thatsome of our prey groups were isotopically similar (See Fig 2) even though these groupsdiffered taxonomically (and functionally) We attempted to counter this by includinginformation on prey elemental concentration in the model (ie we used a concentration-dependent model) which will likely be most useful in distinguishing between animal andphytoplankton prey (Phillips et al 2014) Given these caveats we feel confident that ourmixing model results (Fig 4) support our conclusions that 1 anchovy are not assimilatingsignificant amounts of C and N from phytoplankton 2 that the anchovy populationincludes individuals feeding on two broad groups of animal prey ie consumptionof crustacean larvae by the low δ15N anchovy group and ichthyoplankton (eggs andpreflexion-stage anchovy larvae) by the high δ15N anchovy group (as well as juvenile jackmackerel)

Our data indicate that anchovy diet is bothmore variable than is generally considered andthat some individuals can feed at high trophic positions It has become increasingly clear thatpopulations of consumers can include individuals following specialised trophic strategies(Bolnick et al 2003) The inclusion of different trophic specialisms within anchovypopulations may contribute to their productivity by limiting intraspecific competition

Pizarro et al (2019) PeerJ DOI 107717peerj6968 1522

and support the unusually high fish production associated with the HCS (Bakun ampWeeks2008 Chavez amp Messieacute 2009) The cannibalism and predation of eggs and larvae we suggestfor the high δ15N anchovy group is commonly reported in clupeiform fishes (Alheit 1987Hunter 1981) allowing consumption of an energy-rich prey (Konchina 1991) The studyarea is recognised as a spawning and nursery area for fish particularly anchovies (PalmaEscribano amp Rosales 2006 Palma Pizarro amp Flores 1992) Given that anchovies and otherfishes can spawn throughout the year in the region this potentially results in a permanentsupply of early life stage prey to predators but may represent an important driver of naturalmortality with potential implications for future recruitment success

CONCLUSIONSWe used a stable isotope approach to examine the trophic ecology of one of the worldrsquosmost important (but apparently understudied) fishes in an extremely productive coastalupwelling zone Our work has revealed several important findings with implicationsfor how we understand and manage this important resource We have shown thatlike previous studies using stable isotopes from both central Chile (Huumlckstadt Rojasamp Antezana 2007) and Peru (Espinoza et al 2017) that anchovy TP has likely beenconsiderably underestimated Our estimates of anchovy TP differ from the classic value(22) used in much of the fisheries literature (Gueacutenette Christensen amp Pauly 2008 Paulyet al 1998) and support Espinoza et alrsquos (2017) observation that the classical hypothesisthat short and efficient food chains drive secondary production in this ecosystem (Ryther1969) is no longer valid We also have shown that the trophic ecology of individualanchovies is far more complicated than previously consideredmdashindividuals captured inthe same net fell into two broad trophic groups even though they were largely of a similarsize These groups differed in trophic position and estimated diet revealing previouslyunrecognised trophic variation in a fish that has long been considered to indiscriminatelyconsume phytoplankton (ie TPsim22) Given the similarities with the data presentedby Espinoza et al (2017) from anchovies further north in their distribution (which alsoshowed considerable variation in δ15N) it is clear that any model assuming that anchovyare feeding at a low TP is suspect and management decisions based on such models shouldbe reconsidered

Our use of stable isotope analysis to characterise ecosystem function in the pelagiczone of the HCS has shown that the system is more complex than previously thoughtThis parallels a recent study where using a similar approach we challenged long-heldassumptions regarding energy supply in coastal kelp forest ecosystems in the north of Chile(Docmac et al 2017) The current study highlights the utility of stable isotope analysis asa means to reveal cryptic ecological variation in systems that are generally considered welldescribed We are currently examining amino acid δ15N of different pelagic fishes fromnorthern Chile including anchovy and jack mackerel to examine whether we see similarestimates of TP (McClelland amp Montoya 2002) and levels of individual variation as seenhere with isotope analysis of bulk materials

Pizarro et al (2019) PeerJ DOI 107717peerj6968 1622

ACKNOWLEDGEMENTSWe thank the handling editor and three different reviewers for useful comments on anearlier version of the manuscript

ADDITIONAL INFORMATION AND DECLARATIONS

FundingThe study was supported by the Project FIP 2007-45 lsquolsquoEfectos de la variabilidad de la capa demiacutenimo de oxiacutegeno en la distribucioacuten y la abundancia de los principales recursos pesquerosde la zona nortersquorsquo ChrisHarrod is supported byNucleoMilenio INVASAL funded byChilersquosgovernment program Iniciativa Cientiacutefica Milenio from the Ministerio de EconomiacuteaFomento y Turismo Felipe Docmac is supported by the Rector of the Universidad deAntofagasta Jessica Pizarro is supported by the Vicerrectoriacutea de Investigacioacuten Innovacioacuteny Postgrado of the Universidad Arturo Prat The funders had no role in study design datacollection and analysis decision to publish or preparation of the manuscript

Grant DisclosuresThe following grant information was disclosed by the authorsEfectos de la variabilidad de la capa demiacutenimo de oxiacutegeno en la distribucioacuten y la abundanciade los principales recursos pesqueros de la zona norte Project FIP 2007-45Chilersquos government program Iniciativa Cientiacutefica Milenio from the Ministerio deEconomiacutea Fomento y TurismoVicerrectoriacutea de Investigacioacuten Innovacioacuten y Postgrado of the Universidad Arturo Prat

Competing InterestsThe authors declare there are no competing interests

Author Contributionsbull Jessica Pizarro conceived and designed the experiments performed the experimentsanalyzed the data contributed reagentsmaterialsanalysis tools prepared figures andortables authored or reviewed drafts of the paper approved the final draftbull Felipe Docmac analyzed the data contributed reagentsmaterialsanalysis tools preparedfigures andor tables authored or reviewed drafts of the paper approved the final draftbull Chris Harrod analyzed the data contributed reagentsmaterialsanalysis tools preparedfigures andor tables authored or reviewed drafts of the paper approved the final draft

Animal EthicsThe following information was supplied relating to ethical approvals (ie approving bodyand any reference numbers)

Permission to sample was provided by the Chilean Subsecretariacutea de Pesca (ResolucioacutenExenta N 2205 21 August 2008) and by the Chilean Navyrsquos Hydrographic andOceanographic Service (SHOA N 1327024317VRS 21 June 2008)

Pizarro et al (2019) PeerJ DOI 107717peerj6968 1722

Data AvailabilityThe following information was supplied regarding data availability

Stable isotope data and associated code for SIMMR mixing models and tRophicPostionare available in the Supplemental File

Supplemental InformationSupplemental information for this article can be found online at httpdxdoiorg107717peerj6968supplemental-information

REFERENCESAlegre A Bertrand A EspinoM Espinoza P Dioses T NtildeiquenM Navarro I Simier M

Meacutenard F 2015 Diet diversity of jack and chub mackerels and ecosystem changes inthe northern Humboldt Current system a long-term study Progress in Oceanography137299ndash313 DOI 101016jpocean201507010

Alheit J 1987 Egg cannibalism versus egg predation their significance in anchoviesSouth African Journal of Marine Science 5467ndash470 DOI 102989025776187784522694

AndersonMJ Gorley RN Clarke KR 2008 PERMANOVA+ for PRIMER guide tosoftware and statistical methods Plymouth PRIMER-R

Bakun A 1996 Patterns in the ocean ocean processes and marine population dynamicsSan Diego CA USA and La Paz Mexico California Sea Grant in cooperation withCentro de Investigaciones Biologicas del Noroeste

Bakun AWeeks SJ 2008 The marine ecosystem off Peru what are the secrets of itsfishery productivity and what might its future hold Progress in Oceanography79290ndash299 DOI 101016jpocean200810027

BalloacutenM Bertrand A Lebourges-Dhaussy A Gutieacuterrez M Ayoacuten P Grados DGerlotto F 2011 Is there enough zooplankton to feed forage fish populationsoff Peru An acoustic (positive) answer Progress in Oceanography 91360ndash381DOI 101016jpocean201103001

Bligh EG DyerWJ 1959 A rapid method of total lipid extraction and purificationCanadian Journal of Biochemistry and Physiology 37911ndash917 DOI 101139y59-099

Bolnick DI Svanbaumlck R Fordyce JA Yang LH Davis JM Hulsey CD Forister ML2003 The ecology of individuals incidence and implications of individual special-ization American Naturalist 1611ndash28 DOI 101086343878

Bolnick DI Yang LH Fordyce JA Davis JM Svanbaumlck R 2002Measuring individual-level resource specialization Ecology 832936ndash2941DOI 1023073072028

Carabel S Godinez-Dominguez E Verisimo P Fernandez L Freire J 2006 Anassessment of sample processing methods for stable isotope analyses of marinefood webs Journal of Experimental Marine Biology and Ecology 336254ndash261DOI 101016jjembe200606001

Casciotti KL 2016 Nitrogen and oxygen isotopic studies of the marine nitrogen cycleAnnual Review of Marine Science 8379ndash407DOI 101146annurev-marine-010213-135052

Pizarro et al (2019) PeerJ DOI 107717peerj6968 1822

Castillo J Coacuterdova J Saavedra A Espejo M Gaacutelvez P Barbieri MA Osses J Reyes HArriagada G Barriacutea P Gili R Oliva E Brieba C 2002 Evaluacioacuten del reclutamientode anchoveta en la I y II Regiones temporada 2001ndash2002 Informe Final ProyectoFIP No 2001ndash11 Instituto de Fomento Pesquero Valparaiacuteso Available at httpwwwsubpescacl fipa613articles-88966_informe_finalpdf

Castillo J Saavedra A Leiva F Reyes H PizarroM Espiacutendola F Castati V Lang CHernaacutendez CE Leiva B Cerna F Loacutepez A Herrera L Claramunt G Oliva EMoreno P MedinaM 2011 Evaluacioacuten hidroacuacutestica del reclutamiento de laanchoveta en la XV I y II Regiones antildeo 2010 Informe Final Proyecto FIP No 2009ndash02 Instituto de Fomento Pesquero Chile Valparaiacuteso Available at httpwwwsubpescacl fipa613articles-89269_informe_finalpdf

Caut S Angulo E Courchamp F 2009 Variation in discrimination factors (δ15N andδ13C) the effect of diet isotopic values and applications for diet reconstructionJournal of Applied Ecology 46443ndash453 DOI 101111j1365-2664200901620x

Cerna F Plaza G 2016 Daily growth patterns of juveniles and adults of the Peruviananchovy (Engraulis ringens) in northern ChileMarine and Freshwater Research67899ndash912 DOI 101071MF15032

Chavez FP Bertrand A Guevara-Carrasco R Soler P Csirke J 2008 The northernHumboldt current system brief history present status and a view towards the futureProgress in Oceanography 7995ndash105 DOI 101016jpocean200810012

Chavez FP Messieacute M 2009 A comparison of eastern boundary upwelling ecosystemsProgress in Oceanography 8380ndash96 DOI 101016jpocean200907032

Clarke KR Gorley RN 2015 PRIMER v7 user manualtutorial Plymouth PRIMER-ECury P Bakun A Verheye HM Shannon LJ Crawford RJM Jarre A Quintildeones RA

2000 Small pelagics in upwelling systems patterns of interaction and structuralchanges in wasp-waist ecosystems ICES Journal of Marine Science 57603ndash618DOI 101006jmsc20000712

De Pol-Holz R Robinson RS Hebbeln D Sigman DM Ulloa O 2009 Controls onsedimentary nitrogen isotopes along the Chile margin Deep Sea Research Part IITopical Studies in Oceanography 561042ndash1054 DOI 101016jdsr2200809014

Docmac F ArayaM Hinojosa IA Dorador C Harrod C 2017Habitat couplingwrit large pelagic-derived materials fuel benthivorous macroalgal reef fishes in anupwelling zone Ecology 982267ndash2272 DOI 101002ecy1936

Espinoza P Bertrand A 2008 Revisiting Peruvian anchovy (Engraulis ringens) tropho-dynamics provides a new vision of the Humboldt current system Progress inOceanography 79215ndash227 DOI 101016jpocean200810022

Espinoza P Lorrain A Meacutenard F Cherel Y Tremblay-Boyer L Arguumlelles J Tafur RBertrand S Tremblay Y Ayoacuten P Munaron J-M Richard P Bertrand A 2017Trophic structure in the northern Humboldt Current system new perspectives fromstable isotope analysisMarine Biology 16486 DOI 101007s00227-017-3119-8

Food and Agriculture Organization of the United Nations (FAO) 2018 Fisheries andaquaculture software FishStatJmdashsoftware for fishery statistical time series RomeFAO Fisheries and Aquaculture Department

Pizarro et al (2019) PeerJ DOI 107717peerj6968 1922

Fry BWainright SC 1991 Diatom sources of 13C-rich carbon in marine food websMarine Ecology-Progress Series 76149ndash157 DOI 103354meps076149

Fuenzalida R Escribano R Oliva ME Garceacutes J Rosales S Claramunt G Morales CHerrera L Santander E RojoM Pizarro P Carrasco C Moreno P Azocar C 2009Efectos de la variabilidad de la capa miacutenimo de oxiacutegeno (CMO) en la distribucioacuteny la abundancia de los principales recursos pesqueros de la zona norte InformeFinal Proyecto FIP No 2007ndash45 Depto Ciencias del Mar Universidad Arturo PratIquique Available at httpwwwsubpescacl fipa613articles-89212_informe_finalpdf

Giesecke R Gonzaacutelez HE 2004Mandible characteristics and allometric relations incopepods a reliable method to estimate prey size and composition from mandibleoccurrence in predator guts Revista Chilena De Historia Natural 77607ndash616DOI 104067S0716-078X2004000400004

Gueacutenette S Christensen V Pauly D 2008 Trophic modelling of the Peruvian upwellingecosystem towards reconciliation of multiple datasets Progress in Oceanography79326ndash335 DOI 101016jpocean200810005

Harrod C Grey J McCarthy TK Morrissey M 2005 Stable isotope analyses providenew insights into ecological plasticity in a mixohaline population of European eelOecologia 144673ndash683 DOI 101007s00442-005-0161-x

Herrera L Escribano R 2006 Factors structuring the phytoplankton community in theupwelling site off El Loa River in northern Chile Journal of Marine Systems 6113ndash38DOI 101016jjmarsys200511010

Huumlckstadt LA Rojas CP Antezana T 2007 Stable isotope analysis reveals pelagicforaging by the Southern sea lion in central Chile Journal of Experimental MarineBiology and Ecology 347123ndash133 DOI 101016jjembe200703014

Hunter JR 1981 Feeding ecology and predation of marine fish larvae In Lasker R edMarine fish larvae morphology ecology and relations to fisheries Seattle University ofWashington Press

Hussey NE MacNeil MA McMeans BC Olin JA Dudley SFJ Cliff GWintner SPFennessy ST Fisk AT 2014 Rescaling the trophic structure of marine food websEcology Letters 17239ndash250 DOI 101111ele12226

Hynes HBN 1950 The food of fresh-water sticklebacks (Gasterosteus aculeatus andPygosteus pungitius) with a review of methods used in studies of the food of fishesJournal of Animal Ecology 1936ndash58 DOI 1023071570

Hyslop EJ 1980 Stomach contents analysismdasha review of methods and their applicationJournal of Fish Biology 17411ndash429 DOI 101111j1095-86491980tb02775x

Kaumlmpf J Chapman P 2016Upwelling systems of the world a scientific journey to the mostproductive marine ecosystems Cham Springer International Publishing

Konchina YU 1991 Trophic status of the Peruvian anchovy and sardine Journal ofApplied Ichthyology 3159ndash72

Magozzi S Yool A Vander Zanden HBWunder MB Trueman CN 2017 Usingocean models to predict spatial and temporal variation in marine carbon isotopesEcosphere 8e01763 DOI 101002ecs21763

Pizarro et al (2019) PeerJ DOI 107717peerj6968 2022

McClelland JWMontoya JP 2002 Trophic relationships and the nitrogen isotopic com-position of amino acids in plankton Ecology 832173ndash2180DOI 1018900012-9658(2002)083[2173TRATNI]20CO2

McCutchan JH LewisWM Kendall C McGrath CC 2003 Variation in trophicshift for stable isotope ratios of carbon nitrogen and sulfur Oikos 102378ndash390DOI 101034j1600-0706200312098x

MedinaM Herrera L Castillo J Jaque J Pizarro N 2015 Alimentacioacuten de la anchoveta(Engraulis ringens) en el norte de Chile (1825primendash2540primeS) en diciembre de 2010 LatinAmerican Journal of Aquatic Research 4346ndash58 DOI 103856vol43-issue1-fulltext-5

Nielsen JM Clare EL Hayden B Brett MT Kratina P Gilbert MTP 2018 Diet tracingin ecology method comparison and selectionMethods in Ecology and Evolution9278ndash291 DOI 1011112041-210X12869

OrsquoReilly CM Hecky RE Cohen AS Plisnier PD 2002 Interpreting stable isotopesin food webs recognizing the role of time averaging at different trophic levelsLimnology and Oceanography 47306ndash309 DOI 104319lo20024710306

Orrego H Mendo J 2015Haacutebitos alimenticios del jurel Trachurus murphyi (Nichols) enla zona nor-central del mar peruano Ecologiacutea Aplicada 14103ndash113DOI 1021704reav14i1-287

PalmaW Escribano R Rosales S 2006Modeling study of seasonal and inter-annualvariability of circulation in the coastal upwelling site of the El Loa River off northernChile Estuarine Coastal and Shelf Science 6793ndash107 DOI 101016jecss200511011

PalmaW Pizarro J Flores C 1992 Co-ocurrencia distribucioacuten y abundancia de losestados tempranos de Engraulis ringens y Sardinops sagax (Pisces Clupeiformes)en un aacuterea de surgencia costera en el norte de Chile Investigaciones Cientiacuteficas yTecnoloacutegicas serie Ciencias del Mar 212ndash30

Parnell A 2016 simmr a stable isotope mixing model Available at https cranr-projectorgwebpackages simmr indexhtml

Parnell AC Phillips DL Bearhop S Semmens BXWard EJ Moore JW Jackson ALGrey J Kelly DJ Inger R 2013 Bayesian stable isotope mixing models Environ-metrics 24387ndash399 DOI 101002env2221

Pauly D Christensen V Dalsgaard J Froese R Torres Jr F 1998 Fishing down marinefood webs Science 279860ndash863 DOI 101126science2795352860

Pauly D Froese R Palomares ML 2000 Fishing down aquatic food webs AmericanScientist 8846ndash51

Phillips DL Inger R Bearhop S Jackson AL Moore JW Parnell AC Semmens BXWard EJ 2014 Best practices for use of stable isotope mixing models in food webstudies Canadian Journal of Zoology 92823ndash835 DOI 101139cjz-2014-0127

Post DM 2002 Using stable isotopes to estimate trophic position models methods andassumptions Ecology 83703ndash718DOI 1018900012-9658(2002)083[0703USITET]20CO2

Quezada-Romegialli C Jackson AL Hayden B Kahilainen KK Lopes C Harrod C2018 tRophicPosition an R package for the Bayesian estimation of trophic position

Pizarro et al (2019) PeerJ DOI 107717peerj6968 2122

from consumer stable isotope ratiosMethods in Ecology and Evolution 91592ndash1599DOI 1011112041-210X13009

R Core Team 2018 R a language and environment for statistical computing Vienna RFoundation for Statistical Computing

Reddin CJ Docmac F OrsquoConnor NE Bothwell JH Harrod C 2015 Coastal upwellingdrives intertidal assemblage structure and trophic ecology PLOS ONE 10e0130789DOI 101371journalpone0130789

RStudio Team 2016 RStudio integrated development for R Boston RStudio IncRyther JH 1969 Photosynthesis and fish production in the sea Science 16672ndash76

DOI 101126science166390172Smith ADM Brown CJ Bulman CM Fulton EA Johnson P Kaplan IC Lozano-

Montes H Mackinson S Marzloff M Shannon LJ Shin Y-J Tam J 2011 Impactsof fishing lowndashtrophic level species on marine ecosystems Science 3331147ndash1150DOI 101126science1209395

SYSTAT Software Inc 2009 SYSTAT Rcopy 131 Richmond CA USA Available at https systatsoftwarecom

Thiel M Macaya EC Acuna E ArntzWE Bastias H Brokordt K Camus PA CastillaJC Castro LR Cortes M Dumont CP Escribano R FernandezM Gajardo JAGaymer CF Gomez I Gonzalez AE Gonzalez HE Haye PA Illanes JE Iriarte JLLancellotti DA Luna-Jorquerai G Luxoroi C Manriquez PH Marin V MunozP Navarrete SA Perez E Poulin E Sellanes J Sepulveda HH StotzW Tala FThomas A Vargas CA Vasquez JA Vega JMA 2007 The Humboldt CurrentSystem of northern and central Chile In Gibson RN Atkinson RJA Gordon JDMeds Oceanography and marine biology Vol 45 Boca Raton CRC PressmdashTaylor ampFrancis Group 195ndash344

Thomas SM Crowther TW 2015 Predicting rates of isotopic turnover across theanimal kingdom a synthesis of existing data Journal of Animal Ecology 84861ndash870DOI 1011111365-265612326

Vander ZandenMJ Rasmussen JB 1996 A trophic position model of pelagic foodwebs impact on contaminant bioaccumulation in lake Trout Ecological Monographs66451ndash477 DOI 1023072963490

Whitehead PJP Nelson GJ Wongratana T 1998 FAO species catalogue Volume 7Clupeoid fishes of the world (Suborder Clupeoidei) An annotated and illustratedcatalogue of the herrings sardines pilchards sprats anchovies and wolf-herringsPart 2 Engraulididae FAO Fish Synopsis 125305ndash579

Pizarro et al (2019) PeerJ DOI 107717peerj6968 2222