The importance of space and time when interpreting trophic ...

The importance of space and time when
interpreting trophic structure from stable isotopes
Master thesis in biology
Joan Holst Hansen
20062632
Marine Ecology, Department of Bioscience, Aarhus University
August 2011
The Greenland Institute
of Natural Resources
Greenland Climate
Research Center

Forord
Denne specialerapport indeholder en generel indledning, som beskriver specialeprojektet og teorierne
bag dette, samt et engelsk artikeludkast omhandlende fødenetstrukturen i det marine miljø ved
Grønland. Artiklen er formateret til indsendelse til tidsskriftet Marine Ecology Progress Series.
Specialet er lavet i samarbejde mellem Afdeling for Marin Økologi, Biologisk Institut, Aarhus
Universitet, Grønlands Naturinstitut og Grønlands Klimaforskningscenter. Specialeprojektet har
modtaget finansiel støtte af Grønlands Klimaforskningscenter og Forskningsrådet.
Peter Grønkjær og Jens Tang Christensen fra Aarhus Universitet var interne vejledere, og Rasmus Berg
Hedeholm og Kaj Sünksen fra Grønlands Naturinstitut var eksterne vejledere. Jeg vil hermed gerne
rette en stor tak til alle fire for gode råd og grundig vejledning gennem hele specialeforløbet. En stor
tak skal også lyde til Grønlands Naturinstitut samt hele besætningen på RV Pâmiut, som stillede
laboratorieudstyr til rådighed og hjalp med indsamlingen af data. Endvidere vil jeg gerne benytte
lejligheden til at udvise min store taknemmelighed til Rasmus Berg Hedeholm og Kaj Sünksen for at
bidrage til et fantastisk ophold i Nuuk.
Joan Holst Hansen
Aarhus Universitet, august 2011
1

Resumé
Specialet omhandler fødenetstrukturen i de grønlandske farvande belyst ved hjælp af stabile kulstof-
og kvælstofisotoper (δ 13C og δ 15N). δ 15N og δ 13C er et valideret og effektivt værktøj til økologiske
studier. δ 15N værdien bruges som et integreret mål for en organismes trofiske niveau, da δ 15N signalet
i prædatorer typisk beriges med 3-4‰ i forhold til deres bytte. Samme berigelse ses ikke for δ 13C
signalet. δ 13C bruges i stedet som et estimat for oprindelsen af den energi, der ernærer fødenettet
(bentisk vs. pelagisk) og kan endvidere relateres til vandmassernes oprindelse. Specialeprojektet
omhandler både en rumlig og en tidslig undersøgelse af det marine økosystem. Det rumlige forsøg blev
udført langs den grønlandske vest- og østkyst (59°N – 72°N). Tolv arter fra alle trofiske niveauer
(grønlandshaj, havkat, hellefisk, torsk, rødfisk, håising, lodde, polartorsk, rejer, krill, zooplankton og
POM) blev udvalgt til beskrivelse af fødenettet. Det tidslige forsøg blev udført i Nuuk Fjorden over en
periode på otte måneder fra april til november. Her blev isotopsignaturen undersøgt for ni arter
(hellefisk, helleflynder, torsk, rødfisk, håising, lodde, krill, copepoder og POM) fanget inden for et lille
geografisk område.
Undersøgelserne af isotopsignaturen viste, at der både er variationer på det rumlige og tidslige plan.
Der var en signifikant effekt i δ 15N på stor geografisk skala og studiet viste en parallelforskydningen af
hele fødenettet med breddegraden. δ 15N værdierne langs den grønlandske vestkyst ændres omkring
3‰ svarende til et helt trofisk niveau mellem det sydligste og nordligste undersøgte område. Det
tidslige forsøg viste en signifikant forskel i δ 13C med stigende værdier i sommerperioden. Ingen
signifikant forskel i δ 15N blev påvist i løbet af de otte måneder studiet varede.
Isotopsignaturen for de undersøgte arter varierede altså på både et stort og lille geografisk område.
Disse forskelle er af en størrelse som ikke alene kan skyldes ændret fødeadfærd, men sandsynligvis
skyldes variationer i de abiotiske faktorer, såsom forskelle i vandets fysiske og biologiske egenskaber
samt terrestrisk udledning til havet. Dette resultat har stor betydning for andre studier, som benytter
stabile isotoper (δ 15N og δ 13C) til belysning af økologiske spørgsmål, idet manglende hensyntagen til
rumlige og tidslige effekter kan have afgørende indflydelse på konklusionen.
2

Summary
The thesis is concerned with the structure of the marine food web in Greenland using stable carbon
and nitrogen isotopes (δ 13C and δ 15N). δ 13C and δ 15N is a validated and efficient tool when studying
food webs. The δ 15N value is used as an integrated measure of an organisms trophic level, because the
signal of δ 15N in predators are enriched by 3-4‰ compared to its prey. The same enrichment is not
true for δ 13C. The δ 13C value is instead used as a proxy of the origin of the energy source of the food
web (benthic vs. pelagic) and can also be related to the origin of the water masses. The thesis consists
of both a spatial and a temporal study of the marine ecosystem. The spatial study was conducted along
the Greenlandic West and East coast (59°N - 72°N). Twelve species (Greenlandic shark, wolffish,
Greenland halibut, Atlantic cod, redfish, American plaice, capelin, polar cod, shrimp, krill, copepods
and POM) were collected representing all trophic levels to describe the food web. The temporal study
was executed in the Nuuk Fjord during an eight month sampling period running from April through
November. In this study the isotopic signature was examined for nine species (Greenland halibut,
Atlantic halibut, Atlantic cod, redfish, American plaice, capelin, krill, copepod and POM) sampled
within a small geographic area.
Comparing stable isotope signatures in the arctic marine food web demonstrated both spatially and
temporally variations. On a large geographic scale the spatial study showed a significantly latitudinal
effect in δ 15N, revealing a parallel shift of the entire food web. δ 15N values along the West coast of
Greenland changed around 3‰ from south to north, corresponding to a full trophic level. In the
temporal study a significant difference in δ 13C values was demonstrated with increasing values during
the summer. Regarding δ 15N values no significant difference was shown during the sampling period.
Accordingly the isotopic signature of the examined species varied both on a small and large geographic
scale. These differences are of a size that not alone can be caused by chancing food preferences, but
most likely are due to abiotic factors such as differences in physical and biological properties of the
water masses as well as terrestrial inputs. This result will have a large influence on other ecological
studies using stable isotopes (δ 13C and δ 15N) since such spatial and temporal effects can have a crucial
influence on the conclusion if not included.
3

Indholdsfortegnelse
Dansk introduktion
FORORD .............................................................................................................................................................................................. 1
RESUMÉ .............................................................................................................................................................................................. 2
SUMMARY.......................................................................................................................................................................................... 3
INDHOLDSFORTEGNELSE .......................................................................................................................................................... 4
INTRODUKTION .............................................................................................................................................................................. 6
Arktiske fødekæder ................................................................................................................................................................. 6
Stabile isotoper ......................................................................................................................................................................... 7
Trofisk fraktionering ......................................................................................................................................................... 9
Regenereret produktion ................................................................................................................................................ 10
Begrænsninger ved brugen af stabile isotoper ..................................................................................................... 10
Det grønlandske system ..................................................................................................................................................... 11
Det grønlandske fødenet ............................................................................................................................................... 12
Oceanografiske forhold.................................................................................................................................................. 12
Formål med specialet .......................................................................................................................................................... 14
REFERENCER ................................................................................................................................................................................ 15
Article: Stable isotope variability in an arctic marine food web on a spatial and temporal scale
ABSTRACT ......................................................................................................................................................................................... 1
INTRODUCTION .............................................................................................................................................................................. 1
MATERIAL AND METHODS ........................................................................................................................................................ 3
Spatial Study............................................................................................................................................................................... 3
Field Sampling ..................................................................................................................................................................... 3
Temporal Study ........................................................................................................................................................................ 4
Field sampling ...................................................................................................................................................................... 4
Stable isotope preparation and analysis for both spatial and temporal studies ........................................... 5
Estimation of trophic level ................................................................................................................................................... 6
Estimation of relative stable carbon and nitrogen values....................................................................................... 7
Statistical analyses ................................................................................................................................................................... 7
RESULTS............................................................................................................................................................................................. 7
Spatial Study............................................................................................................................................................................... 8
Trophic level ......................................................................................................................................................................... 9
Temporal study ...................................................................................................................................................................... 10
Inshore-offshore comparison .......................................................................................................................................... 10
DISCUSSION ................................................................................................................................................................................... 11
4

Spatial Study............................................................................................................................................................................ 11
Temporal Study ..................................................................................................................................................................... 14
Inshore-offshore comparison .......................................................................................................................................... 15
CONCLUSION ................................................................................................................................................................................. 15
ACKNOWLEDGEMENTS............................................................................................................................................................ 15
REFERENCE ................................................................................................................................................................................... 17
TABLES AND FIGURES .............................................................................................................................................................. 23
5

Introduktion
Til belysning af økologiske spørgsmål er det helt centralt, at skabe en forståelse af det givne systems
struktur og interaktioner mellem de involverede arter. For at kunne gøre dette er det vigtigt at have en
fundamental viden om de implicerede organismers fødeadfærd og indbyrdes afhængighedsforhold.
Arktiske fødekæder
Kort beskrevet er en fødekæde betegnelsen for hvem der spiser hvem. En fødekæde beskriver hvordan
energien fra føden videregives fra primærproducenter gennem herbivorer og videre til carnivorer.
Økologen Charles Elton (1927) var en af de første, som definerede en simpel fødekæde og opdagede at
længden af en fødekæde ofte er begrænset til fire eller fem led (trofiske niveauer). Et eksempel på en
fødekæde er den arktiske pelagiske fødekæde: Fytoplankton bliver spist af zooplankton såsom
euphausider og copepoder. Disse zooplanktonarter er føde for en række carnivorer, heriblandt fisk,
blæksprutter og bardehvaler som til sidst ender som føde for topprædatorer, såsom spækhuggeren
(Fig. 1).
Figur 1: Skematiseret eksempel af et arktisk fødenet. Efter http://maps.grida.no/go/graphic/arctic-
pelagic-food-web
6

Flere faktorer er dog medvirkende til, at fødekæder aldrig er så simple som dette eksempel. En
fødekæde er ikke en isoleret enhed, men derimod sammenkædet med andre fødekæder i et større og
mere kompliceret fødenet. Endvidere kan en art prædere på flere trofiske niveauer, og aldersbaserede
fødeskift gør også at en fisk kan bevæge sig ”op gennem” fødekæden efterhånden som de bliver større.
Stabile isotoper
Stabile isotoper er ikke-radioaktive atomer af samme grundstof. Stabile isotoper har samme antal
protoner i kernen og derfor samme atomnummer, men har forskellige antal neutroner hvilket giver et
forskelligt massetal. Mange grundstoffer har mere end én stabil isotop. Dette gælder for eksempel for
grundstofferne kulstof (C), kvælstof (N), ilt (O), svovl (S) og hydrogen (H), som forekommer i
atmosfæren, jorden, planter og dyr. Eksempelvis akkumuleres kulstof og kvælstof i væv og knogler hos
dyr.
Stabile isotoper har store anvendelsesmuligheder og kan blandt andet bruges til bestemmelse af en
organismes forskellige fødekilder. Ikke kun i nulevende organismer, men også eksempelvis i fortidige
mennesker hvor man ved hjælp af kulstof- og kvælstofisotoper i knogler kan se om føden stammer fra
landjorden, havet eller søer (Szepanski et al. 1999; Arneborg et al. 1999). Andre anvendelses-
muligheder er iskernedatering, hvor iltisotoper bruges til at estimere tidligere tiders
temperaturvariationer (Johnsen et al. 2001), estimering af trofisk niveau (Peterson & Fry 1987) eller
identificering af forskellige biologiske processer (Lee et al. 2010).
Marine økosystemer rummer ofte store og komplekse fødenet (Fig. 1), hvilket komplicerer
evalueringen af disse. Ved hjælp af stabile isotoper, særdeles kulstof og kvælstof, er det dog muligt at
undersøge strukturen og dynamikken af marine fødenet (Peterson & Fry 1987; France 1995a; Vander
Zanden et al. 1999). Brugen af stabile kulstof- og kvælstofisotoper til undersøgelse af marine
økosystemer startede da DeNiro og Epstein (1978 og 1981) fandt en sammenhæng mellem ratioen af
den tunge og den lettere isotop af kulstof ( 13C/ 12C) og kvælstof ( 15N/ 14N) i en organisme og dens føde.
Der er i løbet af de seneste årtier bygget på denne viden, og brugen af stabile isotoper til evaluering af
fødenet er tiltaget kraftigt (Hobson & Wasenaar 1999). Forholdet mellem 13C/ 12C og 15N/ 14N betegnes
δ 13C og δ 15N, og udtrykkes i promille ved følgende notation:
δ ø
hvor X er 13C eller 15N og R er den tilhørende 13C/ 12C eller 15N/ 14N ratio.
7

Grundet diskrimination mod de tunge isotoper ( 13C og 15N) i organismers biokemiske reaktioner
akkumuleres de tunge isotoper op gennem fødekæden fra bytte til prædator i vævet. Berigelsen i
isotopsammensætning mellem bytte og prædator, kaldet trofisk fraktionering, er estimeret i flere
studier til værende 0-1‰ for δ 13C og 3-4‰ for δ 15N (DeNiro & Epstein 1978 og 1981; Peterson & Fry
1987; Post 2002; McCutchan et al. 2003; Sweeting et al. 2007a og 2007b).
Forskellen i fraktioneringsraten mellem kulstof og kvælstof skyldes forskelle i de stofskiftereaktioner i
organismen, som forårsager fraktioneringerne. Fraktioneringen af kulstof skyldes hovedsageligt
afgivelse af isotopisk set lettere CO2 i forhold til føden. Katabolisme af proteiner, lipider, og
kulhydrater bevirker en diskriminering mod den tungere 13C isotop når acetyl-grupper oxideres og
isotopisk let CO2 frigives fra organismen (Galimov 1985). Fraktioneringen i δ 13C er som tidligere nævnt
lille (0-1‰, Peterson & Fry 1987) men også variabel, og faktisk fandt DeNiro & Epstein (1978)
lignende δ 13C værdier mellem primærproducenter og primærkonsumenter, mens Sweeting et al.
(2007b) fandt en fraktionering for fisk på 1-2‰. På grund af den lille fraktionering i δ 13C mellem bytte
og prædator er stabile kulstofisotoper ikke ideelle til bestemmelse af trofisk niveau som kvælstof er,
men bruges i stedet til bestemmelse af organismers primære kulstofkilde (Vander Zanden &
Rasmussen 1999, Post 2002). Dette skyldes en rumlig variation i sammensætningen af kulstof
isotoper, hvorved det er muligt at anvende δ 13C til bestemmelse af organismers levested samt
eventuelle migrationsmønstre. Det er således muligt at undersøge om en organisme lever på land, i
ferskvand nær kysten eller i vandsøjlen (France 1995b) eller i stille eller turbulent vand (Osmond et al.
1981). Eksempelvis undersøgte Vander Zanden & Rasmussen (1999) isotopsignaturen hos
primærkonsumenter i søer og fandt en forskel i δ 13C på cirka 4‰ mellem den littorale zone og den
pelagiske zone, hvor primærkonsumenterne i den littorale zone havde mest berigede δ 13C værdier.
Desuden er det også muligt, at undersøge om en marin art er tilhørende en pelagisk vs. indenskærs
eller bentisk vs. udenskærs fødekæde (France 1995b; Lawson & Hobson 2000). Dette skyldes, at δ 13C
værdien i bentiske organismers fødekilde har en anden isotopisk sammensætning end de pelagiske
organismers, hvor pelagisk fytoplankton overvejende er kulstofkilden (Hobson 1999). Dette betyder at
bentiske organismer har mere berigede δ 13C værdier (mindre negative) end pelagiske organismer, da
bentiske primærproducenter lever i mindre turbulent vand og dermed bliver diffusionslaget større,
hvilket bevirker at mere 13C assimileres af planten. Dette isotopsignal kan følges i andre organismer på
højere trofiske niveauer i både den bentiske og pelagiske fødekæde.
Modsat kulstofisotoper er fraktioneringen i kvælstof stor og foregår under kvælstofudskillelse
(Peterson & Fry 1987), hvor den lettere 14N isotop reagerer hurtigere end den tungere 15N isotop.
Derved vil en prædator have en højere δ 15N værdi sammenlignet med dens føde. Fraktioneringen for
akvatiske organismer i δ 15N er 3-4‰ og en gennemsnitlig værdi på 3,4‰ er generelt accepteret (Post
2002). Der er dog ikke entydig konsensus omkring denne fraktioneringsgrad og flere studier har
8

foreslået en lavere kvælstoffraktionering. Sweeting et al. (2007a) fandt en fraktionering på 2,9‰ ved
analyse af hele fisk og McCutchan et al. (2003) fandt ved undersøgelse af tidligere studier en lavere
gennemsnitlig fraktioneringsværdi på 1,4‰ for organismer, som har levet af invertebrater, mens
konsumenter der fik en proteinholdig føde havde en fraktionering på 3,3‰.
Trofisk niveau
På grund af denne store fraktionering i δ 15N mellem prædatorer og deres bytte, er δ 15N ideel til
estimering af trofisk niveau. Dette kan gøres ved hjælp af følgende formel (Peterson & Fry 1987):
δ æ δ
δ
hvor δ 15Nprædator er δ 15N værdien for den pågældende art, Δδ 15N er fraktioneringen i δ 15N per trofisk
niveau, δ 15Nbasis er den gennemsnitlige δ 15N værdi af den valgte art til basisniveau og TLbasis er det
trofiske niveau for den art (eks. 1 for fytoplankton, 2 for zooplankton under antagelse af, at det er
herbivor zooplankton).
Eksempelvis estimerede Nilsen et al. (2008) trofisk niveau for en række invertebrater og fisk i en fjord
i det nordlige Norge. Dette blev gjort både ved hjælp af δ 15N værdier og ovenstående formel samt til
sammenligning ved brug af en økologisk model (ecopath). En ecopathmodel kan ved hjælp af relativ få
og grundlæggende input om økosystemets komponenter (biomasse, produktion, konsumption)
estimere trofisk niveau for hver komponent. Nilsen et al. (2008) fandt en god korrelation (r 2 = 0,72)
mellem trofisk niveau beregnet ud fra δ 15N og beregnet ved brug af ecopathmodellen. Dette verificerer
brugen af stabile kvælstofisotoper til estimering af trofisk niveau, uden man har en detaljeret viden om
den faktiske fødeadfærd af en organisme, hvilket kræver en relativ stor indsats og kendskab til det
pågældende system og dets arter.
Trofisk fraktionering
Fraktioneringen i både kulstof og kvælstof er som nævnt ovenfor ikke konstant og kan variere
afhængig af art, individets størrelse og kondition (Bode et al. 2007), temperaturen i det omgivende
miljø (Barnes et al. 2007), væksthastighed (Trueman et al. 2005), vævstype og vævets omsætningsrate
(Tieszen et al. 1983, Lorrain et al. 2002). Dette betyder blandt andet, at væv med en høj
omsætningsrate (måneder til år), såsom knogler og muskelvæv, integrerer isotopsignaturen fra føden
over en længere periode end lever- og gonadevæv, der har en høj omsætningsrate (dage til uger). Det
er derfor vigtigt, alt efter hvor lang en tidsperiode man er interesseret i at undersøge, at udvælge den
korrekte vævstype til undersøgelse af isotopsignaturen (O’Reilly et al. 2002, Post 2002). Udover den
9

høje omsætningsrate er væv som lever og gonader også ofte lipidholdige organer. Sammenlignet med
kulhydrater og proteiner indeholder lipider mindre 13C (Griffiths 1991; Sotiropoulos et al. 2004),
hvilket betyder at en varierende mængde lipid i de forskellige væv påvirker δ 13C værdierne (Lorrain et
al. 2002). Derfor anbefaler flere studier at lipidekstrahere, hvis man laver δ 13C analyser på lipidholdigt
væv (Post 2002; Sweeting et al. 2007b). Med hensyn til δ 15N er effekten af lipidekstrahering mere
uklar, da tidligere studier har vist et lidt tvetydigt resultat, med ingen eller lille effekt af
lipidekstrahering (Sweeting et al. 2006; Sweeting et al. 2007a; Søreide et al. 2006).
Regenereret produktion
I den eufotiske zone er de vigtigste former for uorganisk kvælstof ammonium (NH4), nitrat (NO3) og N2
gas (Michener & Kaufman 2008). δ 15N i primærproducenterne påvirkes af disse forskellige former af
uorganisk kvælstof, som findes i vandsøjlen, samt af forholdet mellem mængderne af ny og
regenereret kvælstof. Kvælstof som tilføres de frie vandmasser fra dybere vande, kvælstoffiksering og
atmosfæren betegnes som nyt kvælstof. Da kvælstofkredsløbet er et delvist lukket kredsløb vil dette
kvælstof atter indgå efter nedbrydning, nu som regenereret kvælstof. Regenereret kvælstof er således
den plantetilgængelige kvælstof (ammonium og urinstof) i den eufotiske zone, der bliver gendannet i
havet. Dette kan påvirke δ 15N værdierne, da regenereret produktion oftest har højere δ 15N værdier
end ny produktion. Eksempelvis fandt Ostrom et al. (1997) i det kolde marine miljø ud for Conception
Bay, Newfoundland, højere δ 15N i partikulært organisk materiale (POM), grundet en drastisk berigelse
i 15N angiveligt på grund af højere δ 15N i den regenererede produktion i området. Således var δ 15N
værdierne for ammonium i vandet højere end δ 15N for nitrat. Det er dog ikke altid regenereret
primærproduktionen giver berigede δ 15N værdier, som eksempelvis Needoba et al. (2006) der fandt
lavere δ 15N værdier. Det er derfor altid vigtigt, at tage forbehold for disse eventuelle forskelle ved
sammenligninger på tværs af studier.
Begrænsninger ved brugen af stabile isotoper
Brugen af stabile isotoper som et økologisk redskab indebærer en mængde fordele og muligheder.
Disse er dog baseret på flere antagelser (eksempelvis fraktionering) og følgeligt flere begrænsninger.
Når man bruger stabile kulstof og kvælstof isotoper til belysning af fødenetstruktur og energiflow,
bliver resultatet som tidligere nævnt et integreret estimat af det samlede fødeindtag. Dette leder til en
af de helt store begrænsninger, da det derved ikke er muligt direkte at udlede hvilke specifikke arter
den undersøgte organisme har konsumeret. Til dette skal bruges en analyse af maveindholdet, hvilket
dog kun giver et øjebliksbillede af den indtagne føde. En kombination af disse to metoder er derfor ofte
den optimale tilgang til en detaljeret fødeanalyse, ligesom fedtsyreanalyse kan supplere billedet af
fødeadfærden (Petursdottir et al. 2008)
10

En af de store udfordringer ved brugen af stabile isotoper er endvidere, at kunne sammenligne
resultaterne på tværs af forskellige økosystemer. Ved sammenligninger på tværs af økosystemer giver
δ 13C og δ 15N for én organisme alene ganske lidt information angående dens absolutte trofiske niveau
eller den ultimative kulstofkilde (Kling et al. 1992, Post et al. 2000). Dette skyldes, at der er betydelig
variation mellem økosystemer i δ 13C og δ 15N ved fødekædens basis (δ 13Cbasis og δ 15Nbasis). Uden et
korrekt estimat af δ 13Cbasis og δ 15Nbasis i hvert økosystem er det ikke muligt, at bestemme om en given
variation i δ 13C og δ 15N hos en organisme skyldes ændringer i fødenetstruktur og kulstofkilde eller
ændringer i δ 13C og δ 15N ved basis af fødekæden.
Variation i δ 13Cbasis og δ 15Nbasis skyldes forskelle i isotoprationen i det biologisk tilgængelige kulstof og
kvælstof ved basis af fødekæden og variation i fraktioneringsraten. De fleste primærproducenter i
marine økosystemer har en høj variation i δ 13C og δ 15N over tid, hvilket komplicerer deres direkte
anvendelighed som indikatorer af δ 13Cbasis og δ 15Nbasis for organismer højere oppe i fødekæden (Cabana
& Rasmussen 1996). På grund af sådanne forskelle i oprindelsen af kvælstof kan der forekomme både
rummelige og tidslige variationer i basisniveauet hos primærproducenterne og følgelig også på højere
trofiske niveauer.
For at mindske variationen i basisniveau, vil det ofte være mere ideelt at bruge længerelevende
organismer med en længere omsætningsrate i vævet. Dette kunne eksempelvis være muslinger, da
disse som regel lever i flere år og er forholdsvis stationære, hvilket gør dem mindre sensitive overfor
tidslige variationer. Alternativt kan herbivore zooplanktonarter bruges, da de ernærer sig direkte af
det tilgængelige fytoplankton. Det vil især være nødvendigt hvis man arbejder i ikke-kystnære
systemer hvor muslinger ikke er tilgængelige.
Det grønlandske system
Grønland har en mere end 2300 kilometer lang uafbrudt nord-sydgående kyststrækning og en række
faktorer er varierende grundet denne store breddegradsgradient. Eksempelvis er mængden af
lysindstrålingen i vandet meget varierende fra syd til nord, hvilket skyldes et faldende antal soltimer
og en stigende mængde is med breddegraden. Også temperaturen varierer langs denne
breddegradsgradient. Generelt er temperaturen faldende med stigende breddegrad, hvilket også er
tilfældet med lufttemperaturen i Grønland, men ikke nødvendigvis for temperaturen i havet.
Vandtemperaturen langs den Vestgrønlandske kyst er mere kompliceret og styres overordnet af to
forskelligt tempererede dominerende vandmasser (se afsnittet om oceanografiske forhold)
(Ribergaard 2011). Disse faktorer er medvirkende til, at også forårsopblomstringen af
primærproduktionen i de grønlandske farvande er varierende langs denne breddegradsgradient.
11

Det grønlandske fødenet
Det grønlandske fødenet er som andre arktiske fødenet domineret af relativt få arter (Roy et al. 1998,
Allen et al. 2002). Eftersom marine organismer direkte eller indirekte er afhængige af det første led i
fødekæden, primærproduktionen, har den rumlige og tidslige variation af primærproduktionen stor
betydning for hele fødekæden. Primærproduktionen i de grønlandske farvande er afhængig af
mængden af sollys, men selvfølgelig også af mængden af tilgængelige næringsstoffer. Dette betyder, at
der er store sæsonmæssige variationer i primærproduktionen (Fig. 2). Denne store tidslige variation i
primærproduktionen har som tidlige nævnt ikke kun betydning for organismer, som lever direkte af
primærproduktionen. En stor del af primærproduktionen ender som føde for eksempelvis copepoder,
som udgør grundlaget for fødenettene i de frie vandmasser, mens en anden del af
primærproduktionen synker til havbunden og derved bidrager til at ernærer den bentiske fødekæde.
Sæsonvariationen i primærproduktionen har altså stor betydning for hele fødenettet også for
organismer på højere trofiske niveauer.
Figur 2: Årlig variation i primærproduktionen ved Nuuk (64°N) i 2007 (Jensen & Rasch 2008).
Oceanografiske forhold
De hydrografiske forhold omkring den sydlige del af Grønland er karakteriseret af den Øst- og
Vestgrønlandske Strøm (Fig. 3). Den kolde og lavsaline Østgrønlandske Strøm fra Polarhavet mødes
med den tempererede Irmingerstrøm i Danmarksstrædet mellem Grønland og Island.
Irmingerstrømmen er en gren af den Nordatlantiske Strøm, der udgøres af varmt og højsalint vand.
Disse to havstrømme samles under vedvarende opblanding langs den grønlandske vestkyst i den
Vestgrønlandske Strøm (Ribergaard 2010, Fig. 3), hvor den tungere og varmere Irmingerstrøm
placeres under den Østgrønlandske Strøm. Den gradvise opblanding af de to vandmasser langs den
grønlandske vestkyst resulterer i en opvarmning overfladevandet og vice versa for bundvandet.
12

Omkring Diskoøen (69°N) har den Vestgrønlandske Strøm aftaget i styrke og forgrener sig mod David
Strædet, hvilket dermed resulterer i at vandets overfladetemperatur falder fra Diskoøen og længere
nordpå. Dog varierer sammensætningen af komponenterne i den Vestgrønlandske Strøm fra år til år,
hvilket er en medvirkende faktor til temperaturvariationer i vandmasserne langs den grønlandske
kyst (Ribergaard 2010). Sådanne udsving i temperatur mellem år kan have stor betydning for
organismer på alle trofiske niveauer i det marine økosystem. Det er blandt andet vist, at væksten hos
flere arter i grønlandsk farvand påvirkes af disse rumlige forskelle i vandmassernes abiotiske faktorer.
Hos hellefisk (Reinhardtius hippoglossoides) og lodde (Mallotus villosus) ses ændringer i væksten
grundet temperaturvariationerne i vandmasserne. Således har hellefisk en faldende vækstrate med
stigende breddegrad (Sünksen 2009), mens det modsatte vækstmønster ses for lodden (Hedeholm
2010) konsistent med det inverse temperaturmønster som følge af opblandingen. Udover disse
overordnede mønstre giver den lokale bundtypografi og meget dybe fjorde en yderligere variation i
vandmassernes input til forskellige områder (Mortensen et al. 2011).
Figur 3: Havstrømme ved Grønlands Vest- og Østkyst (Pedersen & Smidt 2000).
13

Formål med specialet
Kun ganske få studier har tidligere undersøgt den marine fødekæde langs Grønland. Størstedelen af
disse har brugt mere traditionelle metoder, såsom maveanalyser. Enkelte analyser med stabile
isotoper er blevet udført til belysning af den marine fødekæde langs vestkysten, men i disse studier
har eventuelle ændringer i basisniveau langs breddegradsgradienten ikke været medtaget. Der er altså
så vidt vides ingen tidligere studier, som har undersøgt eventuelle tidslige eller rumlige forskelle i
isotopsignalet i den marine fødekæde ved Grønland. Det er netop disse problemstillinger jeg vil belyse
i specialeprojektet. Resultaterne fra det rumlige forsøg langs den grønlandske vest- og østkyst vil give
et detaljeret billede af det marine fødenet på et stort geografisk område, samt klarlægge om en
ændring af basisniveau langs kysten er forårsaget af opblanding af vandmasser. Resultaterne fra det
tidslige forsøg udført i Nuuk fjorden fra april til november, vil kunne belyse organismers eventuelle
fødeskift samt ændringer i basisniveau i løbet af perioden.
Resultaterne fra dette projekt har også stor betydning for andre studier, som benytter stabile isotoper
(δ 13C og δ 15N) til belysning af økologiske spørgsmål. Dette gør sig ikke kun gældende for studier udført
i det arktiske miljø, men for alle studier som undersøger fødenetstrukturer, da manglende
hensyntagen til rumlige og tidslige effekter kan have afgørende indflydelse på konklusionen.
14

Referencer
Allen AP, Brown JH, Gillooly JF (2002) Global biodiversity, biochemical kinetics, and the energetic-
equivalence rule. Science 297:1545-1548
Arneborg J, Heinemeier J, Lynnerup N, Nielsen HL, Rud N, Sveinbjörnsdóttir ÁE (1999) Changes of diet
of the Greenland Vikings determined from stable carbon isotope analysis and 14C dating of their
bones. Radiocarbon 41:157-168
Barnes C, Sweeting CJ, Jennings S, Barry JT, Polunin NVC (2007) Effect of temperature and ration size
on carbon and nitrogen stable isotope trophic fractionation. Funct Ecol 21:356-362
Bode A, Alvarez-Ossorio MT, Cunha ME, Garrido S, Peleteiro JB, Porteiro C, Valdés L, Varela M (2007)
Stable nitrogen isotope studies of the pelagic food web on the Atlantic shelf of the Iberian
Peninsula. Prog Oceanogr 74:115-131
Cabana C, Rasmussen JB (1996) Comparison of aquatic food chains using nitrogen isotopes. Proc Natl
Acad Sci 93:10844-10847
DeNiro MJ, Epstein S (1978) Influence of diet in the distribution of carbon isotopes in animals.
Geochim Cosmoshim Acta 42:495-506
DeNiro MJ, Epstein S (1981) Influence of diet in the distribution of nitrogen isotopes in animals.
Geochim Cosmoshim Acta 45:341-351
Elton CS (1927) Animal ecology. Sidgwick & Jackson, London, UK.
France RL (1995a) Differentiation between littoral and pelagic food webs in lakes using carbon
isotopes. Limnol Oceanogr 40:1310-1313
France R (1995b). Carbon-13 enrichment in benthic compared to pelagic algae: food web implications.
Mar Ecol Prog Ser 124:307-312
Galimov EM (1985) The biological fractionation of isotopes. Academic Press, Orlando, FL.
15

Griffiths H (1991) Applications of stable isotope techonology in physiological ecology. Funct Ecol
5:254-269
Hedeholm R (2010) The importance of small pelagic fishes to the energy flow in marine ecosystems:
the Greenlandic capelin. Ph.D. dissertation. Aarhus University, Aarhus, Denmark.
Hobson KA (1999) Tracing origins and migration of wildlife using stable isotopes: a review. Oecologia
120:314-326
Hobson KA, Wasenaar LI (1999) Stable isotope ecology: an introduction. Oecologia 120:312-313
Jensen LM, Rasch M (2008) Nuuk ecological research operations 1 st annual report, 2007. Copenhagen,
Danish Polar Centre, Danish Agency for Science, Technology and Innovation, Ministry of Science
Technology and Innovation.
Johnsen SJ, Dahl-Jensen D, Gundestrup N, Steffensen JP, Clausen HB, Miller H, Masson-Delmotte V,
Sveinbjörnsdottir AE, White J (2001) Oxygen isotope and palaeotemperature records from six
Greenland ice-core stations: Camp Century, Dye-3, GRIP, GISP2, Renland and NorthGRIP. J
Quaternary Sci 16:299-307
Kling GW, Fry B, O Brien WJ (1992) Stable isotopes and plantonic trophic structure in arctic lakes.
Ecology 73:561-566
Lawson JW, Hobson KA (2000) Diet of harp seals (Pagophilus groenlandicus) in nearshore northeast
Newfoundland: inferences from stable-carbon (δ 13C) and nitrogen (δ 15N) isotope analyses. Mar
Mamm Sci 16:578-591
Lee P, Wahjudi PN, Xu J, Go VL (2010) Tracer-based metabolomics: Concept and practices. Clinical
Biochemistry 43:1269-1277
Lorrain A, Paulet YM, Chauvaud L, Savoye N, Donval A, Saout C (2002) Differential δ 13C and δ 15N
signatures among scallop tissues: implications for ecology and physiology. J Exp Mar Biol Ecol
275:47-61
16

McCutchan JH Jr., Lewis WM Jr., Kendall C, McGrath CC (2003) Variation in trophic shift for stable
isotope ratios of carbon, nitrogen, and sulfur. Oikos 102:378-390
Mortensen J, Lennert K, Bendtsen J, Rysgaard S (2011) Heat sources for glacial melt in a sub-Arctic
fjord (Godthåbsfjord) in contact with the Greenland Ice Sheet. J Geophys Res C 116.
Michener RH, Kaufman L (2008) Stable isotope ratios as tracers in marine food webs: an
updateisotopes in ecology and enviromental science, Oxford, UK.
Needoba JA, Marchetti A, Henry MF, Harrison PJ, Wong CS, Johnson WK, Pedersen TF (2006) Stable
notrigen isotope dynamics of a mesoscale iron enrichment experiment in the NE Subarctic
Pacific. Deep-Sea Res Part II 53:2214-2230
Nilsen M, Pedersen T, Nilssen EM, Fredriksen S (2008). Trophic studies in a high-latitude fjord
ecosystem – a comparison of stable isotope analyses (δ 13C and δ 15N) and trophic-level estimates
from a mass-balance model. Can J Fish Aquat Sci 65:2791-2806
O’Reilly MC, Hecky RE, Cohen AS, Plisnier PD (2002) Interpreting stable isotopes in food webs:
Recognizing the role og time averaging at different trophic levels. Limnol Oceanogr 47:306-309
Osmond CB, Valaane N, Haslan SM, Uotila P, Roksandic Z (1981) Comparisons of δ 13C values in leaves
of aquatic macrophytes from different habitats in Britain and Finland: some implications for
photosynthesis processes in aquatic plants. Oecologia 50:117-124
Ostrom NE, Macko SA, Deibel D, Thompson R (1997) Seasonal variation in the stable carbon and
nitrogen isotope biogeochemistry of a coastal cold ocean environment. Geochim Cosmochim
Acta 61:2929-2942
Pedersen SA, Smidt ELB (2000) Zooplankton distribution and abundance in West Greenland waters,
1950-1984. J Northwest Atl Fish Soc 26:45-102
Peterson BJ, Fry B (1987) Stable isotopes in ecosystem studies. Annu Rev Ecol Syst 18:293-320
17

Petursdottir H, Gislason A, Falk-Petersen S, Hop H, Svavarsson J (2008) Trophic interactions of the
pelagic ecosystem over the Reykjanes Ridge as evaluated by fatty acid and stable isotope
analyses. Deep-Sea Res Part II 55:83-93
Post DM, Pace ML, Hairston Jr NG (2000) Ecosystem size dertermines fodd-chain length in lakes.
Nature 405:1047-1049
Post DM (2002) Using stable isotopes to estimate trophic position: models, methods, and assumptions.
Ecology 83:703-718
Ribergaard MH (2011) Oceanographic Investigations off West Greenland 2010. Danish Meteorological
Institute Centre for Ocean and Ice: 44 pp.
Roy K, Jablonski D, Valentine JW, Rosenberg G (1998) Marine latitudinal diversity gradients: Tests of
causal hypotheses. Proc Natl Acad Sci USA 95:3699-3702
Sotiropoulos MA, Tonn WM, Wassenarr LI (2004) Effects of lipid extraction on stable carbon and
nitrogen isotope analyses of fish tissues: potential consequences for food web studies. Ecol
Freshw Fish 13:155-160
Sweeting CJ, Polunin NVC, Jennings S (2006) Effects of chemical lipid extraction and arithmetic lipid
correction on stable isotope ratios of fish tissues. Rapid Commun Mass Spectrom 20:595-601
Sweeting CJ, Barry J, Barnes C, Polunin NVC, Jennings S (2007a) Effects of body size and environment
on diet-tissue δ 15N fractionation in fishes. J Exp Mar Biol Ecol 340:1-10
Sweeting CJ, Barry JT, Polunin NVC, Jennings S (2007b) Effects og body size and environment on diet-
tissue δ 13C fractionation in fishes. J Exp Mar Biol Ecol 352:165-176
Sünksen K, Stenberg C, Grønkjær P (2009) Temperature effects on growth of juvenile Greenland
halibut (Reinhardtius hippoglossoides Walbaum) in West Greenland waters. J Sea Res 64:125-132
Szepanski MM, Ben-David M, Van Ballenberghe V (1999) Assessment of anadromous salmon resources
in the diet of the Alexander Archipelago wolf using stable isotope analysis. Oecologia 120:327-
335
18

Søreide JE, Hop H, Carroll ML, Falk-Petersen S, Hegseth EN (2006) Seasonal food web structures and
sympagic-pelagic coupling in the European Arctic revealed by stable isotopes and a two-source
food web model. Prog Oceanogr 71:59-87
Tieszen LL, Boutton TW, Tesdahl KG, Slade NA (1983) Fractionation and turnover of stable carbon
isotopes in animal tissues: implications for δ 13C analysis of diet. Oecologia 57:32-37
Trueman CN, McGill RAR, Guyard PH (2005) The effect of growth rate on tissue-diet isotopic spacing in
rapidly growing animals. An experimental study with Atlantic salmon (Salmon salar). Rapid
Commun Mass Spectrom 19:3239-3247
Vander Zanden MJ, Rasmussen JB (1999) Primary consumer δ 13C and δ 15N and the trophic position of
aquatic consumers. Ecology 80:1395-1404
Vander Zanden MJ, Shuter BJ, Lester N, Rasmussen JB (1999) Patterns of food chain length in lakes: a
stable isotope study. Am Nat 154:406-416
19

Stable isotope variability in an arctic marine food web on a
spatial and temporal scale
Joan Holst Hansen 1, Rasmus Berg Hedeholm 2,3, Kaj Sünksen 2, Jens Tang Christensen 1, Peter Grønkjær 1
1Marine Ecology, Department of Bioscience, Aarhus University, 8000 Aarhus C, Denmark
2Greenland Institute of Natural Resources, PO box 570, 3900 Nuuk, Greenland
3Greenland Climate Research Centre, PO box 570, 3900 Nuuk, Greenland
Abstract
Stable isotopes of carbon (δ 13C) and nitrogen (δ 15N) were used to examine trophic structures in an
arctic marine food web on both a spatial and temporal scale. 12 species in total were examined around
the Greenlandic West and East coast, from primary producers through Greenland shark. There was a
significant latitudinal effect on δ 15N values, with a difference of 2‰ to 4‰ between the most southern
and northern areas. The temporal study was conducted on nine species in the Nuuk Fjord during an
eight month long sampling period. A significant temporal effect was seen in δ 13C values, with enriched
values during the summer period. δ 15N values did not change significantly with time, but an increasing
tendency occurred at the end of the period (November). These significant differences in isotopic
signatures on both a large and small geographic scale are most likely unrelated to feeding, but rather
due to different physical and biological properties of the water masses, which would be reflected in the
trophic signatures at the base of the food web and thereby also in higher trophic levels. Generally the
results illustrate the importance of space and time when interpreting trophic structure from stable
isotopes.
Key words: δ 15N, δ 13C, Greenland, latitudinal effects, marine food web
Introduction
Analyses of nitrogen and carbon stable isotopes are a validated and commonly applied method to
describe food web structure in marine ecosystems (e.g. Peterson & Fry 1987; Post 2002). The stable
isotope ratios of carbon (δ 13C) and nitrogen (δ 15N) provide a time-integrated measure of an organisms
trophic position and feeding ecology covering periods ranging from weeks to months, compared to the
1

traditional “snap-shot” picture provided by stomach content analyses, and have the potential to track
energy flow through food webs (Hobson & Welch 1992; Hobson et al. 1995; Post 2002). The method is
based on the principle that heavier isotopes ( 13C and 15N) accumulates from prey to predator (i.e.
fractionation). The fractionation of nitrogen in consumers is on average between 3-4‰ relative to
their diet (Post 2002; Søreide 2006). Thus stable nitrogen isotopes provide a good estimate of a
consumers trophic position given a known baseline δ 15N. Stable carbon isotopes are a less useful index
of trophic position because the trophic fractionation of 13C typically is less than 1‰ (DeNiro & Epstein
1978 and 1981; Vander Zanden & Rasmussen 2001; Post 2002). The stable carbon isotope ratio (δ 13C)
can however be useful when evaluating sources of primary production in marine systems as well as
general patterns of inshore or benthic versus offshore or pelagic feeding preferences (France 1995a;
Lawson & Hobson 2000). Benthic and inshore organisms in marine ecosystems in general have more
enriched δ 13C values as opposed to the more depleted δ 13C values of pelagic and offshore organisms
(France 1995a). Thus, the integrated use of δ 13C and δ 15N can provide valuable information on both
food web structure and carbon source.
Trophic level is estimated relative to a chosen food web base (δ 15Nbase). This allows for comparison of
segregated samples (spatially and/or temporally) in spite of differences in δ 15Nbase. Therefore, if a
proper δ 15Nbase estimate is not available it is not possible to determine if variation in δ 15N values
between systems reflects differences in food web structure or baseline variation. δ 15Nbase values are
predominantly influenced by nitrogen limitation (Pennock et al. 1996), inter-specific differences in
isotope fractionation (Needoba et al. 2003) and form of nitrogen assimilated (new primary production
(nitrate) vs. regenerated production (ammonium), Ostrom et al. 1997). Because of these influencing
factors both the baseline and the absolute nitrogen values will vary in time and space due to changing
chemical and biological conditions (Cabana & Rasmussen 1996; Post 2002) and changes can be seen
on a small geographic scale (Pentoja et al. 2002). The water masses in the coastal areas around
Greenland are complex and consist of two main currents. These are the cold low saline East Greenland
Current from the Polar Sea and the temperate saline Irminger Current which originates from the
Atlantic Ocean. The two currents meet at the southern area of the Greenland East coast, with the
heavier Irminger Current subducting the relatively low saline East Greenland current. The water
masses round Cape Farewell, forming the West Greenland Current were the two water masses
gradually mix flowing north (Ribergaard 2011).
Petursdottir et al. (2008) showed that the δ 15N value of Calanus finmarchicus caught in June in the
Irminger Current of the Reykjanes Ridge south of Iceland was (mean ± SE) 3.5 ± 0.1‰, while Søreide
et al. (2006) found a δ 15N value of 6.4 ± 0.2‰ in September for C. finmarchicus in the Fram Strait west
2

of Svalbard. This clearly demonstrates that water masses around Greenland have different physical
and/or biological properties, which is reflected in shifts in the isotopic baseline dependent on the
degree of mixture between water masses. This in turn can be affected by large scale circulation
patterns or local area bathymetry such as sill fjords. Hence differences in the isotopic baselines would
be expected to vary on both large (Tamelander et al. 2009) and small geographical scale (Tamelander
et al. 2006). Therefore, it is hypothesized that a significant latitudinal gradient in δ 15N values would be
found as illustrated in capelin by Hedeholm (2010) in Greenland. As Greenland has an extensive north-
south directed coast line with a number of relative large fjord systems with glacier run off that differ
from offshore regions (Mortensen et al. 2011) it is an ideal place to study effects of geographical
separation on food web structure and shifts in δ 15N and δ 13C values irrespective of changes in trophic
position.
The main objectives of this study were threefold. 1) Provide a novel detailed description of the marine
food web on a large geographic scale along the Greenland coast with a quantification of the effect
differences in sampling site may have on isotopic signatures. 2) Temporal variation in stable isotope
structure and finally 3) importance of local area variation was described by comparing samples from
the Nuuk fjord and adjacent offshore waters.
Material and Methods
The study has both a spatial and temporal component. The species and size categories selected for
sampling in both aspects of the study were based on their ecological relevance in the Greenlandic
ecosystem while also attempting to span as much as possible of the ecosystems trophic range.
Spatial Study
Field Sampling
All samples were collected from the trawler RV Pâmiut (722 GRT) from June to August 2010 during
the annual stratified-random bottom trawl surveys carried out by the Greenland Institute of Natural
Resources in Greenlandic waters. Samples included seven teleost fish species: wolffish (Anarhichas
lupus and Anarhichas minor), Atlantic cod (Gadus morhua (small (25-35cm) and large (45-55cm))),
polar cod (Boreogadus saida) American plaice (Hippoglossoides platessoides), capelin (Mallotus
villosus), Greenland halibut (Reinhardtius hippoglossoides), redfish (Sebastes mentella (small (15-
25cm) and large (35-45cm))) and Greenland shark (Somniosus microcephalus), shrimps (Pandalus
3

orealis), krill (Thysanoessa raschii), copepods (Calanus finmarchicus and C. glacialis) and filtered POM
giving a total of 14 sample groups (Table I and II).
Fish were sampled in six predetermined areas at depths between 88 and 1 446 meter along the
Greenlandic coast from Upernavik (72˚N) in west to Tasiilaq (66˚N) in east (Fig. 1). For each of the
predetermined areas it was attempted to sample five individuals of each sample group (Table I). To
obtain the best representation of the within area variation, individuals were sampled from as many
catches within each area as possible. Whenever, regardless of the six predetermined areas, a
Greenland shark was caught the total length, weight and sex was determined and a muscle sample was
taken near the dorsal fin before it was released.
Shrimps, krill, copepods and filtered POM were collected at eleven different sites separated by
approximately two latitudinal degrees (Fig. 1 and Table II). Shrimps and krill were taken from trawl
hauls whereas copepods and filtered POM were sampled using a 500 µm plankton net and Niskin
water bottle (5 L), respectively. The plankton net was lowered to 150 m and then slowly retrieved (20
m · min -1) while the vessel was moving at 1 knot. This was done in the evening to minimize copepod
gut content (Lampert & Taylor 1985; Head & Harris 1987). The content from the cod end was first
filtered though a 2 cm sieve and afterwards a 200 µm sieve to remove jellyfish, chaetopods etc. and
retain the larger copepods, respectively. The copepods were kept frozen on a filter until further
analysis. The Niskin bottle was lowered to ten and twenty meters, respectively. Two liters from each
depth was filtered though a 200 µm filter to remove large organisms (i.e. jellyfish, amphipods etc).
Three liters of this water was then filtered onto a pre-combusted (450°C for 24 hours) GF/F filter (47
mm in diameter). All samples from all groups were immediately frozen at -20°C.
A CTD profile was obtained at approximately every latitudinal degree along the Greenlandic west coast
(60-71°N). Unfortunately the CTD malfunctioned on all East coast stations and at 62°N, 65°N and 66°N
on the West coast.
Temporal Study
Field sampling
Samples were collected in the Nuuk Fjord (64˚N, Fig. 1) four times during an eight-month period from
April to November 2010 and included six fish species (Atlantic cod (small and large), redfish, American
plaice, Atlantic halibut, Greenland halibut and capelin), krill, copepods (C. finmarchicus) and filtered
POM (Table III). Most fish species (except Greenland halibut) and phytoplankton were sampled at the
same location, but logistic circumstances entailed that copepods and to some extent krill had to be
sampled at a different site within the fjord (Fig. 1). Most fish species were caught using fishing rods,
whereas undigested capelin were taken from cod stomachs immediately upon capture. Greenland
4

halibut were bought from the local fish market and were caught in the fjord using a long-line set at 300
– 500 meters. Krill and copepods were collected at the inner part of the fjord with a 2 m in diameter
and 600 µm mesh size MIK net and a MultiNet (Hydrobios) equipped with five 300 µm net. Samples of
filtered POM were collected similarly to the spatial study.
Stable isotope preparation and analysis for both spatial and temporal studies
Total length, weight and sex of the fish were determined in the laboratory. Length of shrimps and krill
were measured from the post-orbital notch to the posterior margin of the carapace and from the post-
orbital notch to the posterior end of the uropods, respectively. The weight of shrimp, krill and
copepods were noted.
Fish samples were prepared by removing white muscle tissue (10.31 ± 4.26 g wet weight, mean ± SD)
dorsally from both sides of the fish, posterior to the dorsal fin which ensured that no bones were
present in the sample. All skin was subsequently removed. White muscle tissue was used for stable
isotope analysis because it best represents the isotopic signature of fish (Rounick and Hicks 1985;
Hesslein et al. 1993) and does not require the removal of inorganic carbonates (Pinnegar & Polunin
1999). Muscle tissue was also dissected from shrimp and krill after the chitinous exoskeleton and the
gut were removed. All samples were freeze-dried (Freeze dryer ALPHA 1-2/LD plus) to constant mass
at -60°C for 24 hours. The dry samples were kept in an exsiccator containing silica gel until further
analysis.
In order to make stable isotope results from the lipid-rich copepods (Lee et al. 2006) comparable to
other groups, lipids were removed using a chloroform-methanol solution (2:1) (Søreide et al. 2007). In
order to confirm that this was necessary an initial experiment was performed at the Greenland
Institute of Natural Resources. It was found that removing lipids had a significant effect on δ 13C values
and these significant values were then comparable to other groups. Whereas the removal of
exoskeleton (using 10% HCL) had no significant effect compared to the non-treated samples on
neither δ 13C nor δ 15N values.
The dried muscle tissue from shark, fish, shrimps and krill was homogenized using a mortar and
pestle. Approximately 1 mg (dry weight, dw) of sample (1.14 ± 0.10 dw, mean ± SD) was weighed into
pre-weighted tin capsules (5x9 mm). Copepods with no visible stomach content were selected, and
packed into the tin capsules as whole individuals. At least three copepods collected at the same time
and place were pooled in one sample to obtain sufficient material for isotopic analysis. Filtered POM
(3.09 ± 0.05 mg dw, mean ± SD) was carefully removed from the filter with needles to minimize the
amount of GF/F filter mixed into the sample, and weighed into tin cups.
5

A total of 397 samples were analyzed in the spatial study (Table I and II) and 118 samples in the
temporal study (Table III). Stable carbon and nitrogen isotope analyses were performed at the UC
Davis Stable Isotope Facility in California, USA. Samples were combusted in a PDZ Europa ANCA-GSL
elemental analyzer interfaced to a PDZ Europa 20-20 isotope ratio mass spectrometer (Sercon Ltd.,
Cheshire, UK). The crimped tin capsules were introduced via a solid autosampler and combusted at
1000°C in a reactor packed with chromium oxide and silvered cobaltous/cobaltic oxide. Following
combustion, oxides are removed in a reduction reactor (reduced copper at 650°C). A post-reactor gas
chromatography (GC) column was kept at 65°C for separation of evolved N2 and CO2 before entering
the IRMS. Stable isotopes are expressed in a δ notation as the deviation from international standards
in parts per thousand (‰) according to the formula:
where X is 13C or 15N and R is the corresponding ratio 13C/ 12C or 15N/ 14N.
δ
Standards for δ 13C and δ 15N were calibrated against Vienna PeeDee Belemnite and atmospheric air,
respectively.
Estimation of trophic level
Trophic level (TLN) was estimated for all groups based on the following formula:
δ δ
δ
where δ 15Nconsumer is the δ 15N of the species in question, Δδ 15N is the enrichment in δ 15N per trophic
level, δ 15Nbase is the average δ 15N of the group chosen as the base of the food web and TLbase is the
trophic level of that group. Copepods (C. finmarchicus and C. glacialis) were chosen as the base
assuming a strictly herbivorous diet (e.i. TLbase = 2, Hobson & Welch 1992; Søreide et al. 2006; Nilsen
et al. 2008). Filtered POM was not used as the values showed some irregularities (see results and
discussion). A Δδ 15N of 3.4‰ was applied for invertebrates which were based on analyzes of samples
from the Barents Sea (Søreide et al. 2006) and a separate Δδ 15N value of 3.2‰ for fish (Sweeting et al.
2007). The two different enrichment factors indirectly indicate variations in enrichment correlated to
body size, diet (Fry et al. 1999), growth rate (Trueman et al. 2005) and tissue catabolism (Kurle &
Worthy 2001; Olive et al. 2003).
6

Estimation of relative stable carbon and nitrogen values
In order to detect changes in the isotopic signal across latitudes for the entire food web, comparable
relative values of δ 13C and δ 15N were calculated. A single relative value was first estimated for each
species in each area according to the formula:
A single relative value was then calculated for each predetermined area as the mean of all species
within that area:
where is the relative value of δ 13C or δ 15N in any of the predetermined areas referred to as i.
is the mean of either δ 13C or δ 15N for a given species in one of the areas. N is the number of areas. In
this study N always equals six because there are six predetermined study areas. n is the number of
species in area i, which in this study equals ten because if a species is not caught in every area it is not
included in this calculation since this would bias the result.
Statistical analyses
All analyses were carried out using the statistical computing program R (R Development Core Team
2011). Standard parametric test were preceded by test for assumptions. When these were violated the
data were either transformed or non-parametric statistics were applied. Filtered POM was excluded
from all statistical analyzes as the measured δ 15N values were unusually high in some of the sampling
stations (see discussion).
Results
A total of 515 samples were taken along the Greenland coast from 72°N on the west coast to 66°N on
the east coast (N = 397) and in the Nuuk Fjord (N = 118) during 2010. Stable carbon and nitrogen
isotope ratios were analyzed for all samples representing 13 species and 15 sampling groups (Tables
I-III).
7

Spatial Study
Twelve species and 14 sampling groups were analyzed for δ 13C and δ 15N (Table I and II). Average
stable isotope composition of the analyzed species ranged from -25.3‰ to -15.2‰ in δ 13C and from
3.3‰ to 17.2‰ in δ 15N. The most depleted δ 15N values (mean ± SE) were found in filtered POM (6.43
± 1.40‰) and copepods (7.76 ± 0.57‰) while Greenland shark (16.68 ± 0.25‰) followed by wolffish
(13.35 ± 0.79‰) and Greenland halibut (12.87 ± 0.39‰) were the most enriched in δ 15N. The strictly
benthic organisms (i.e. Greenland shark, wolffish, American plaice and shrimp) were enriched in δ 13C
(-19.22‰ to -15.63‰) compared with strictly pelagic organisms (copepod, krill, polar cod, capelin
and Greenland halibut) which ranged from -21.26‰ to -20.29‰ (Fig. 2). There were large variations
in δ 15N values for filtered POM among the six different sampling areas. δ 15N values ranged from (mean
± SE) 10.77 ± 1.57‰ in area 1 to 3.26 ± 0.19‰ in area 6. It is also remarkable that the δ 15N values of
wolffish in all cases were higher on the West coast compared to δ 15N values of Greenland halibut, but
on the East coast the opposite pattern was found (Fig. 2). Furthermore, the mean δ 13C values of polar
cod on the East coast were depleted 2.1‰ compared to the mean values on the West coast (Fig. 2).
There were significant regional differences in δ 15N values among the six study areas along the
Greenlandic coast (Two-way ANOVA, F5,351 = 99.58, P < 0.0001) irrespective of the also significant
species effect (Two-way ANOVA, F13,351 = 91.41, P < 0.0001). Species specific tests showed a significant
effect of area (i.e. latitude) on all species (P ≤ 0.026) except for Greenland shark (P = 0.35), which
could be due to its limited number of areas included (N = 3, Table I). For most species the δ 15N values
decreased from north (area 1) to south (area 4) on the West coast and increased slightly on the East
coast (area 5 and 6). A clear example of this pattern was seen in shrimps (Fig. 3). However, a few
species did diverge (small cod, capelin, polar cod and wolffish) from this general pattern. For instance,
the δ 15N values of capelin did not decline gradually from north to south. Rather, the δ 15N value (mean ±
SD) of capelin declined significantly (Kruskal-Wallis, χ 2 = 19.01, P < 0.0001) from a high value in area 1
and 2 (12.64 ± 0.12‰) to a lower mean value in area 3, 4, 5 and 6 (9.58 ± 0.77‰). Also, wolffish
showed a different pattern in δ 15N. The values of δ 15N did not increase on the East coast, but rather
continued to decline from high values (mean ± SD) in the most northern area on the West coast (16.11
± 0.44‰) to the lowest value in the northern area on the East coast (11.25 ± 0.51‰). Comparing δ 15N
values on the West coast separately for each species, it was found that the δ 15N values for most species
did not differ significantly among two adjacent areas, but for most species there was a significant
difference between areas which were not adjacent (Table IV).
Similar to δ 15N values there was a significant effect of area on δ 13C values in most species (ANOVA, F4-5,
18-34 ≤ 57.42, P ≤ 0.04), but not in capelin (ANOVA, F5,24 = 0.53, P = 0.75) and small redfish (Kruskal-
8

Wallis, χ 2 = 9.28, P = 0.10). For most species δ 13C values were declining from area 3 to 6 whereas no
clear trend was found for the northern area on the West coast. An example of this pattern of δ 13C along
the coast was seen in shrimps (Fig. 3).
In order to reveal the difference between areas across all species, relative values of δ 15N and δ 13C were
calculated. The regional differences were also reflected in the relative δ 15N values calculated for each
area which differed significantly (ANOVA, F5,54 = 28.86, P < 0.0001, Fig. 4). On the West coast the
highest relative δ 15N value were in the most northern study area (area 1) being 29% higher than the
most southern study area (area 4). Hence, along the investigated 1300 km gradient on the Greenlandic
West coast there was a positive relationship between latitude and relative δ 15N values (linear
regression, F1,38 = 141.5, r 2 = 0.79, P < 0.0001, Fig. 4). On the East coast the two relative δ 15N values did
not differ significantly from each other (Tukey’s post hoc test, P = 0.55), neither did the relative values
from the East coast (areas combined) differ from the most southern area on the West coast (Tukey’s
post hoc test, P > 0.97) (Fig. 4 and Table VI). The relative δ 13C values also differed significantly
between areas (ANOVA, F5,54 = 5.92, P = 0.0002), but this was only caused by area 3 having a
significant lower relative δ 13C value than areas 1, 5 and 6 (Tukey’s post hoc test, P < 0.02, Fig.4). The
relative δ 13C values of area 1, 2 and 3 increased with latitude, like the relative values of δ 15N. However,
the relative δ 13C value of area 4 is approximately the same as the relative δ 13C value of area 5 on the
East coast. There is an almost linear relationship among area 1, 2 and 3 (linear regression, F1,28 = 10.37,
r 2 = 0.27, P = 0.003) and between area 4, 5 and 6 (linear regression, F1,28 = 5.80, r 2 = 0.17, P = 0.02,
Fig.4).
Trophic level
Greenland shark was the apex predator with a mean trophic level (mean ± SE) of 4.77 ± 0.22 followed
by wolffish (3.74 ± 0.16) and Greenland halibut (3.60 ± 0.18, Table V). There was a significant
difference in trophic level between species (ANOVA, F12,341 = 84.75, P < 0.0001) and there was also an
area effect (ANOVA, F5,341 = 31.91, P < 0.0001). However, this effect was primarily caused by West and
East coast differences, with 77% (37 of 48) of the significant species comparisons between areas being
West and East coast related. The only species not being significantly different between the coasts were
krill (ANOVA, F5,34 = 1.46, P = 0.23). On the west coast only small cod (ANOVA, F2,12 = 13.17, P >
0.0009), polar cod (ANOVA, F3,16 = 15.67, P < 0.0001) and shrimp (ANOVA, F3,16 = 13.09, P > 0.0001)
showed a significant difference between areas. The estimated trophic level of polar cod and shrimp
were both significantly higher in area 3 and 4 compared to area 1 and 2 (Tukey’s post hoc test, P ≤
0.03). The same pattern was not found for small cod. Here only area 3 had a higher value of trophic
level, which was significantly different from both area 2 and 4 (Tukey’s post hoc test, P ≤ 0.01).
9

Temporal study
118 muscle samples divided between nine species and ten sampling groups were analyzed for δ 13C
and δ 15N during April, June, August and November 2010 in the Nuuk Fjord (Table III). Average isotopic
values ranged from -16.20 to -21.66 in δ 13C and from 8.40 to 14.83 in δ 15N (Fig. 5).
Mean δ 15N values differed significantly between species (ANCOVA, F8,96 = 15.76, P < 0.0001) but there
was no effect of time-of-catch (ANCOVA, F1,96 = 0.16, P = 0.69). However, a significant interaction term
(ANCOVA, F8,96 = 2.97, P = 0.005), indicates that some species do vary in δ 15N during the eight month
long sampling period. Thus, δ 15N values for small cod increased from 12.84‰ in April to 14.17‰ in
November (ANOVA, F1,16 = 13.23, P = 0.002). Similarly, large redfish (ANOVA, F1,18 = 9.74, P = 0.006)
and krill (ANOVA, F1,13 = 6.66, P > 0.03) increased from 13.95‰ to 14.83‰ and 9.53‰ to 10.16‰,
respectively during the sampling period, but this trend was not true for all species. Atlantic halibut and
large cod showed almost no variation while American plaice showed a decrease in the middle of the
sampling period (Fig 6).
δ 13C values varied both between species (ANCOVA, F8,104 = 46.30, P > 0.001) and as a result of time-of-
catch (ANCOVA, F1,104 = 11.12, P = 0.001) with a tendency of increasing values. There was no
significant interaction (ANOVA, F8,96 = 0.44, P = 0.90) (Fig. 6). This general tendency is for example
shown by small cod and krill where δ 13C values (mean ± SD) of small cod increasing from the lowest
value in June (-18.73 ± 0.32‰) to the highest value in November (-17.77 ± 0.34‰). Likewise, krill
showed a slight increase from -20.26 ± 0.54‰ to -19.66 ± 0.17‰ (mean ± SD) during the sampling
period.
Inshore-offshore comparison
Comparing species caught in- (Nuuk Fjord) and offshore (area 3) at approximately the same time (June
and July 2010) there was a significant area effect (i.e. in- vs. offshore) with regard to both δ 13C
(ANOVA, F1,51 = 9.36, P = 0.004) and δ 15N (ANOVA, F1,51 = 13.03, P = 0.0007) values. Also, there was a
significant species effect in both δ 13C (ANOVA, F7,51 = 31.93, P < 0.0001) and δ 15N (ANOVA, F7,51 =
22.51, P < 0.0001). In general values were highest inshore, however significant interactions terms
(δ 13C: ANOVA, F7,51 = 7.81, P < 0.0001 and δ 15N: ANOVA, F7,51 = 3.27, P = 0.006) show that the pattern
is not similar for all species. In the case of δ 13C values, this was caused by copepods being the only
value higher offshore compared to inshore (Fig.7). Regarding δ 15N values, they were for all species
higher inshore compared to offshore, but not significantly in all species (Fig. 7). It is noteworthy that
δ 15N values (mean ± SE) for copepod (9.98 ± 0.18‰) and krill (9.96 ± 0.19‰) inshore were
approximately the same, while there was a difference of 2.3‰ in δ 15N values between copepod (7.26 ±
10

0.28‰) and krill (8.92 ± 0.46‰) offshore. Values of δ 13C were likewise higher (i.e. less negative) for
all species, except copepods, inshore compared to offshore (Fig. 7). Especially δ 13C values (mean ± SE)
of American plaice caught inshore (-16.20 ± 0.48‰) were higher than δ 13C values offshore (-18.95 ±
0.13‰).
Comparing trophic level for each species caught in- and offshore there was an area effect (Kruskal-
Wallis, χ 2 = 11.08, P = 0.0009) as well as a species effect (Kruskal-Wallis, χ 2 = 43.90, P < 0.0001). The
mean trophic level were for all species higher offshore compared to inshore, which properly is a result
of higher δ 15N baseline values (i.e. copepods) offshore. Hence, the compared food webs is fairly
similarly in composition (Fig. 2 and 5). Even though the estimated mean trophic level were higher
offshore compared to inshore only four out of the eight species sampled differed significantly (krill,
ANOVA, F1,8 = 8.00, P = 0.022; American plaice, ANOVA, F1,6 = 34.05, P = 0.001; small cod, ANOVA, F1,8 =
38.51, P = 0.0003 and large cod, ANOVA, F1,8 = 6.59, P = 0.033, Fig 8).
Discussion
Spatial Study
δ 15N values differed significantly between six predetermined areas along the Greenland West and East
coast being positively correlated with latitude. For all species (except Greenland shark) δ 15N values
increased between 2‰ and 4‰ (Fig. 2) from the most southern area to the northern study area on
the West coast, which is equivalent to a full trophic level (Post 2002). On the East coast the overall
pattern was the same (Fig. 3), but a few species have a reversed pattern of decreasing δ 15N values with
latitude (i.e. wolffish, polar cod, small cod and Greenland halibut, Fig. 2). These represent very
different ecosystem components, and the reason for the discrepancy is unknown but could be related
to migration patterns or actual feeding differences between areas. The same latitudinal gradient was
not evident in δ 13C values. The general pattern of δ 13C values were not gradually increasing but instead
divided in two with increasing values from area 1 to 3 on the West coast and decreasing values from
area 4 to 6. A rather large difference of up to 1.5‰ appears between area 3 and 4 (Fig. 3).
The latitudinal gradients can be explained by a multitude of factors. This includes a change in food
consumption with changing latitude (Sweeting et al. 2005), but given the consistent shift across the
entire food web in especially δ 15N values the explanation must be of a more fundamental nature.
Supporting this is the fact that the estimated trophic level for each species does not change
consistently with latitude, but stays fairly constant across study areas (Table V). In addition to some
11

ase line induced coastal differences in trophic level, only three species (shrimp, polar cod and small
cod) differed significantly in trophic level between some of the West coast areas (Table V). So although
it is not possible to totally disregard an effect of small feeding differences across latitudes, we do not
believe they contribute significantly to the large latitudinal shifts in δ 15N values.
More likely, changes are a result of changing isotope values at the base of the food web with changing
degree of latitude which then cascades through the food web. The baseline isotope values are
potentially influenced by physical, chemical and biological properties of the currents (Pantoja 2002),
terrestrial input (Sherwood & Rose 2005) and primary producer species composition and bloom
progression (Tamelander et al. 2009). The combined effect of these direct and indirect effects on the
primary produces will change the values on all higher trophic levels.
The currents along the Greenlandic west coast are a mixture of the cold low saline East Greenlandic
Current coming from the north and the temperate saline Irminger Current branching of from the Gulf
Stream (Ribergaard 2011). The two currents meet at the southern Greenland East coast and then
merge gradually, homogenizing the water masses moving north along the West coast. If the water
masses differ in isotope signal in the biological available nutrients this will be reflected in POM and
consequently in higher trophic levels. Hence, the parallel shift of the food web may well be related to
shifts in the relative contributions of the two dominant water masses around Greenland.
Two previous studies conducted in the two main currents dominating the waters around Greenland
(East Greenland Current and the Irminger Current) supports the notion of differing water masses
influencing δ 15N and δ 13C (Table VII). Sarà et al. (2009) measured stable isotopes in the Irminger
Current on the Reykjanes Ridge south of Iceland, whereas Hobson et al. (1995) performed stable
isotope analyses in the East Greenland Current (Table VII). Comparing species, or ecologically similar
species, between areas demonstrates relatively large differences in δ 15N values, with species caught in
the Northeast Water Polynya (East Greenland Current) having similar or higher values than those
caught in the Irminger Current. However, all mean δ 15N values in the most northern area on the West
coast (area 1) were all higher compared to mean δ 15N values in the northern area on the East coast
(area 6) opposite of what intuitively to expect given the simplified current pattern described here,
suggesting a complex mixing pattern or some other contributing factor. However, given the differences
in water masses, it surely in some way contributes to the pattern described here.
This gradual mixing of water masses could also possibly be related to the migrational pattern of the
sampled species. Wolffish is considered a stationary species (Riget & Messtorff 1988) and displays a
very large latitudinal gradient (3.2‰) whereas the more migratory behavior seen in for instance
Greenland shark (Skomal & Benz 2004) could serve to dilute the effect of latitudinal related
12

differences in δ 15N values give little north-south difference (0.8‰). However, too little is known about
the migration of the sampled species to make any conclusions.
The parallel shift in δ 15N values with latitude across groups could also be related to the length of the
productive open-water period (Blicher et al. 2007). In high-Arctic areas this period is identical with
the ice-free period, while diminishing day lengths have a pronounced effect further south were ice-free
periods are longer. The productive open-water period is here defined according to Blicher et al. (2007)
as the annual number of days with open water and a minimum day length of six hours. There is a
significant relationship between the productive open-water period and the relative δ 15N values in the
four West coast areas and a similar tendency in the two East coast areas (Fig. 9). Hence, the shorter the
productive open-water period is this higher is the relative δ 15N values. A possible explanation could be
that a longer period with ice cover results in an increasing primary production intensity, whereby
nutrients are exhausted at a faster rate which ultimately could lead to more regenerated production
and subsequently increasing δ 15N values (Wainright & Fry 1994). Another explanation could be that a
longer ice cover period stabilizes the water column, either due to the ice cover or the outlet of
freshwater which would cause a stronger stratification thereby decreasing possible upwelling also
leading to more regenerated primary production (Wu et al. 1997).
Exploring the results from the POM analyses should help in interpreting the results. We have
nevertheless chosen to exclude them from most analyses as the variation in mean POM δ 15N values
between areas was large, ranging from 3.3‰ to 10.8‰, and often exceeded other groups being
biologically impossible (Fig. 2). However, given the low intra-area variation in isotope values and
general high analyses precision, measurement error does not appear to be a factor. Rather, some POM
samples were taken close to or over the shelf edge (Fig. 1) and δ 15N values in these samples were
generally higher (mean 8.5-12.1‰) compared to the values of POM sampled on the shelf (3.3-6.5‰).
CTD profiles taken simultaneously with POM samples clearly demonstrate that different water masses
are present (Fig. 10), but whether these cause the differences in δ 15N values is unknown, as other
factors such as changes in POM composition or seasonal progression could give similar differences
(Tamelander et al. 2009). Thus, as there clearly is some spatial effect on the results on both a large and
small spatial scale (see also the inshore-offshore section) so whatever the ultimate cause of this
latitudinal related pattern in isotopic values (irrespective of feeding behavior) is, it must be taken into
account when addressing hypotheses by means of stable isotope analysis covering a noticeable
geographical range (Møller, 2006, Riget et al. 2007), especially when considering a few number of
species that does not reveal any geographical effect.
Lastly, the vast latitudinal gradient in the present study is naturally associated with a similar
temperature gradient. Barnes et al. (2007) showed in a laboratory study that rising temperature can
have a negative effect on δ 15N values. In Greenland benthic fish will generally experience colder
13

temperatures moving north (Sünksen et al. 2009) while the reverse may be the case for the strictly
pelagic species such as capelin, at least on the West coast (Hedeholm 2010) but as the potential effect
of temperature is small (i.e. -0.06‰ per 1°C for δ 15N assuming linearity, Barnes et al. 2007) compared
to the differences in isotope values and temperature observed here (Fig. 2 and 10) it offers little
explanation in the present study.
Temporal Study
During the eight month sampling period from April to November there was an overall effect of Day-of-
year on δ 13C. The most enriched and depleted values of δ 13C were obtained in June and August and
April and November, respectively (Fig. 5 and 6) with only small cod showing a different pattern.
Greenland halibut and large redfish experienced the largest peak in δ 13C with a difference of about
1‰ in the summer period. The increasing δ 13C values in the summer period are consistent with other
studies investigating primary production, invertebrates and fish (Sarà et al. 2002; Vizzini & Mazzola
2003). This increase in δ 13C during the summer period could indicate a shift in feeding preference
towards a more benthic diet. However, generally evidence of isotopic temporal variability is confined
to relatively short-lived primary producers and consumers (e.g. Nordström et al. 2009; Vizzina &
Mazzola 2003) and the finding of temporal changes of this magnitude are unexpected, as muscle
turnover is relatively slow and feeding induced changes exceeds that normally associated with small
feeding shifts (Bootsma et al. 1996). Hence, a substantial influence from differences in the isotopic
baseline also appears to be of importance in the inshore area which could be due to differences in
water masses, currents and terrestrial input example due to increasing glacial ice melt.
Although there was no significant time effect on δ 15N irrespective of species, the general pattern of
mean δ 15N showed an increasing trend at the end of the sampling period (Fig. 5 and 6) perhaps
indicating a beginning of a winter peak similar to δ 13C. Given the time laps in the response of muscle
tissue (weeks to months, Hesslein et al. 1993) this could be caused by an intensified summer feeding
period following a winter period with lower feeding intensity. In order to demonstrate this potentially
annual isotopic pattern a longer and maybe more intense sampling period must be conducted.
Like in the spatial study a large variation occurred in δ 15N values of POM. The largest difference
occurred between the first two sampling periods with a variation of 4‰. Again this large different in
isotopic signatures of primary production is most likely caused by changes in physical and chemical
properties of the water masses during the sampling period, hence the water masses and conditions of
the currents in the Nuuk Fjord change between summer and winter and increasing freshwater runoff
during summer (Mortensen et al. 2011). It is however also possible that the large difference is due to
variation in species composition of the primary production in the different sampling periods.
14

Inshore-offshore comparison
When comparing stable isotope data measured offshore (area 3) and inshore (Nuuk Fjord) almost all
comparable species (except copepods) had higher δ 13C and δ 15N values inshore than offshore. This is
in agreement with the observation that inshore species have a more enriched δ 13C values compared to
the more depleted offshore species, as benthic primary production generally exhibit less δ 13C
fractionation during carbon fixation than do pelagic phytoplankton (Vander Zanden & Rasmussen
1999, Fig. 7). The pattern in δ 15N values agrees with those of Sherwood & Rose (2005) which on a
similar geographic scale also found lower δ 15N values offshore, but is in contrast with result
represented by France 1995b. One explanation of the higher δ 15N values inshore could be a high
degree of vertical mixing caused by physical processes involving ocean currents, wind and tides
leading to high levels of nutrient availability and productivity in the inshore waters (Sherwood & Rose
2005). Because of the relatively large difference in the base of the food web (i.e. copepods) between
inshore and offshore the estimated trophic level offshore is shifted upwards for all species, but
maintains the same relative pattern (Fig. 8).
Conclusion
Comparing stable isotope signatures in the arctic marine food web demonstrated large differences on
both a large and small geographic scale. The spatial study showed a significantly latitudinal effect,
revealing increasing δ 15N values with latitude. Also a temporal effect was demonstrated, even though
only significant in δ 13C. In order to accurately demonstrate the importance of a temporal change in
δ 15N values a more balanced and intensive sampling may be required. These differences in isotopic
signature both spatial and temporal are probably caused by abiotic factors, differences in physical and
biological properties of the water masses as well as terrestrial inputs. These results will have a large
influence on comparative studies if such differences are not considered when studying isotopic
signatures.
Acknowledgements
The authors acknowledge the staff at the Greenland Institute of Natural Resources for permission to
use laboratory facilities and sampling assistance. Also, we are thankful to the staff at RV Pâmiut and
Sanne Kjellerup and Rasmus Swalethorp (DTU Aqua, DK) for sampling assistance. The study received
15

financial support from the Greenland Institute of Natural Resources and the Danish Agency for
Science, Technology and Innovation and is a part of the Greenland Climate Research Centre.
16

References
Barnes C, Sweeting CJ, Jennings S, Barry JT, Polunin NVC (2007) Effect of temperature and ration size
on carbon and nitrogen stable isotope trophic fractionation. Funct Ecol 21:356-362
Blicher ME, Rysgaard S, Sejr MK (2007) Growth and production of sea urchin Strongylocentrotus
droebachiensis in a high-Arctic fjord, and growth along a climatic gradient (64 to 77°N). Mar Ecol
Prog Ser 341:89-102
Bootsma HA, Hecky RE, Hesslein RH, Turner GF (1996) Food partitioning among Lake Malawi
nearshore fishes as revealed by stable isotope analyses. Ecology 77:1286-1290
Cabana G, Rasmussen JB (1996) Comparison of aquatic food chains using nitrogen isotopes. Proc Natl
Acad Sci 93:10844-10847
DeNiro MJ, Epstein S (1978) Influence of diet on the distribution of carbon isotopes in animals.
Geochim Cosmochim Acta 42:495-506
DeNiro MJ, Epstein S (1981) Influence of diet on the distribution of nitrogen isotopes in animals.
Geochim Cosmochim Acta 45:341–351
France R (1995a). Carbon-13 enrichment in benthic compared to pelagic algae: food web implications.
Mar Ecol Prog Ser 124:307-312
France R (1995b) Stable Nitrogen Isotopes in Fish: Literature Synthesis on the Influence of Ecotonal
Coupling. Est Coast Shelf Sci 41:737-742
Fry B, Mumford PL, Tam F, Fox DD, Warren GL, Havens KE, Steinman AD (1999) Trophic position and
individual feeding historied of fish from Lake Okeechobee. Can J Fish Aquat Sci 56:590-600
Head EJH, Harris LR (1987) Copepod feeding patterns before and during a spring bloom in Bedford
Bassin, Nova Scotia. Mar Ecol Prog Ser 40:221-230
17

Hedeholm R (2010) The importance of small pelagic fishes to the energy flow in marine ecosystems:
the Greenlandic capelin. Ph.D. dissertation. Aarhus University, Aarhus, Denmark.
Hesslein RH, Hallard KA, Ramlal P (1993) Replacement of sulfur, carbon, and nitrogen in tissue of
growing broad whitefish (Coregonus nasus) in response to a change in diet traced by δ 34S, δ 13C
and δ 15N. Can J Fish Aquat Sci 50:2071–2076
Hobson KA, Welch HE (1992) Determination of trophic relationships within a high Arctic marine food
web using δ 13C and δ 15N analysis. Mar Ecol Prog Ser 84:9–18
Hobson KA, Ambrose Jr WG, Renaud PE (1995) Sources of primary production, benthic-pelagic
coupling, and trophic relationships within the Northeast Water Polynya: insights from δ 13C and
δ 15N analysis. Mar Ecol Prog Ser 128:1-10
Kurle CM, Worthy GAJ (2001) Stable isotope assessment of temporal and geographic differences in
feeding ecology of northern fur seals (Callorhinus ursinus) and their prey. Oecologia 126:254-
265
Lampert W, Taylor BE (1985) Zooplankton grazing in eutrophic lake: implications of diel vertical
migration. Ecology 66:68-82
Lawson JW, Hobson KA (2000) Diet of harp seals (Pagophilus groenlandicus) in nearshore northeast
Newfoundland: inferences from stable-carbon (δ 13C) and nitrogen (δ 15N) isotope analyses. Mar
Mamm Sci 16:578-591
Lee RF, Hagen W, Kattner G (2006) Lipid storage in marine zooplankton. Mar Ecol Prog Ser 307:273-
306
Mortensen J, Lennert K, Bendtsen J, Rysgaard S (2011) Heat sources for glacial melt in a sub-Arctic
fjord (Godthåbsfjord) in contact with the Greenland Ice Sheet. J Geophys Res C 116.
Møller P (2006) Lipids and stable isotopes in marine food webs in West Greenland - Trophic relations
and health implications. Ph.D. dissertation. National Environmental Research Institute.
Denmark.
18

Needoba JA, Waser NA, Harrison PJ, Calvert SE (2003) Nitrogen isotope fractionation in 12 species of
marine phytoplankton during growth on nitrate. Mar Ecol Prog Ser 225:81-91
Nilsen M, Pedersen T, Nilssen EM, Fredriksen S (2008). Trophic studies in a high-latitude fjord
ecosystem – a comparison of stable isotope analyses (δ 13C and δ 15N) and trophic-level estimates
from a mass-balance model. Can J Fish Aquat Sci 65:2791-2806
Nordström M, Aarnio K, Bondsdorff E (2009) Temporal variability of a benthic food web: patterns and
processes in an low-diversity system. Mar Ecol Prog Ser 378:13-26
Olive PJV, Pinnegar JK, Polunin NVC, Richards G, Rachel W (2003) Isotope trophic-step fractionation: a
dynamic equilibrium model. J Anim Ecol 72:608-617
Ostrom NE, Macko SA, Deibel D, Thompson R (1997) Seasonal variation in the stable carbon and
nitrogen isotope biogeochemistry of a coastal cold ocean environment. Geochim Cosmochim
Acta 61:2929-2942
Pennock JR, Velinsky DJ, Ludlam JM, Fogel ML (1996) Isotopic fractionation of ammonium and nitrate
during uptake by Skeletonema costatum: Implications for δ 15N dynamics under bloom
conditions. Limnol Oceanogr 41:451-459
Pentoja S, Repeta DJ, Sachs JP, Sigman DM (2002) Stable isotope constraints on the nitrogen cycle of
the Mediterranean Sea water column. Deep-Sea Res Part I 49:1609-1621
Peterson BJ, Fry B (1987) Stable isotopes in ecosystem studies. Annu Rev Ecol Syst 18:293-320
Petursdottir H, Gislason A, Falk-Petersen S, Hop H, Svavarsson J (2008) Trophic interactions of the
pelagic ecosystem over the Reykjanes Ridge as evaluated by fatty acid and stable isotope
analyses. Deep-Sea Res Part II 55:83-93
Pinnegar JK, Polunin NVC (1999) Differential fractionation of δ 13C and δ 15N among fish tissues:
implications for the study of trophic interactions. Funct Ecol 13:225–231
Post DM (2002) Using stable isotopes to estimate trophic position: models, methods, and assumptions.
Ecology 83:703-718
19

R Development Core Team (2011). R: A language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-
project.org/.
Ribergaard MH (2011) Oceanographic Investigations off West Greenland 2010. Danish Meteorological
Institute Centre for Ocean and Ice: 44 pp.
Riget F, Møller P, Dietz R, Nielsen TG, Asmund G, Strand J, Larsen MM, Hobson KA (2007) Transfer of
mercury in the marine food web of West Greenland. J Environ Monitor 9:877-883
Riget F, Messtorff J (1988) Distribution, abundance and migration of Atlantic wolffish (Anarhichas
lupus) and spotted wolffish (Anarhichas minor) in West Greenland waters. J Northw Atl Fish Sci
12:13-20
Rounick JS, Hicks BJ (1985) The stable carbon isotope ratios of fish and their invertebrate prey in four
New Zealand Rivers. Freshw Biol 15:207–214
Sarà G, Vizzini S, Mazzola A (2002) The Effect of Temporal Changes and environmental Trophic
Condition on the Isotopic Composition (δ 13C and δ 15N) of Atherina boyeri (Risso, 1810) and
Gobius niger (L., 1758) in a Mediterranean Coastal Lagoon (Lake of Sabaudia): Implications for
Food Web Structure. Mar Ecol 23:352-360
Sarà G, De Pirro M, Sprovieri M, Rumolo P, Halldórsson HP, Svavarsson J (2009) Carbon and nitrogen
stable isotope inventory of the most abundant demersal fish captured by benthic gears in
southwest Iceland (North Atlantic). Helgol Mar Res 63:309-315
Sherwood GD, Rose GA (2005) Stable isotope analysis of some representative fish and invertebrates if
the Newfoundland and Labrador continental shelf food web. Est Coast Shelf Sci 63:537-549
Skomal GB, Benz GW (2004) Ultrasonic tracking of Greenland sharks, Somniosus microcephalus, under
Arctic ice. Mar Biol 145:489-498
Sweeting CJ, Jennings S, Polunin NVC (2005) Variance in isotopic signatures as a descriptor of tissue
turnover and degree of omnivory. Funct Ecol 19:777-784
20

Sweeting CJ, Barry J, Barnes C, Polunin NVC, Jennings S (2007) Effects of body size and environment on
diet-tissue δ 15N fractionation in fishes. J Exp Mar Biol Ecol 340:1-10
Sünksen K, Stenberg C, Grønkjær P (2009) Temperature effects on growth of juvenile Greenland
halibut (Reinhardtius hippoglossoides Walbaum) in West Greenland waters. J Sea Res 64:125-132
Søreide JE, Hop H, Carroll ML, Falk-Petersen S, Hegseth EN (2006) Seasonal food web structures and
sympagic-pelagic coupling in the European Arctic revealed by stable isotopes and a two-source
food web model. Prog Oceanogr 71:59-87
Søreide JE, Tamelander T, Hop H, Hobson KA, Johansen I (2007) Sample preparation effects on stable C
and N isotope values: a comparison of methods in Arctic marine food web studies. Mar Ecol Prog
Ser 328:17-28
Tamelander T, Renaud PE, Hop H, Carroll ML, Ambrose WG, Hobson KA (2006) Trophic relationships
and pelagic-benthic coupling during summer in the Barents Sea Marginal Ice Zone, revealed by
stable carbon and nitrogen isotope measurements. Mar Ecol Prog Ser 310:33-46
Tamelander T, Kivimäe C, Bellerby RGJ, Renaud PE, Kristiansen S (2009) Base-line variations in stable
isotope values in an Arctic marine ecosystem: effects of carbon and nitrogen uptake by
phytoplankton. Hydrobiologia 630:63-73
Trueman CN, McGill RAR, Guyard PH (2005) The effect of growth rate on tissue-diet isotopic spacing in
rapidly growing animals. An experimental study with Atlantic salmon (Salmon salar). Rapid
Commun Mass Spectrom 19:3239-3247
Vander Zanden MJ, Rasmussen JB (1999) Primary consumer δ 13C and δ 15N and the trophic position of
aquatic consumers. Ecology 80:1395-1404
Vander Zanden MJ, Rasmussen JB (2001) Variation in δ 15N and δ 13C Trophic Fractionation:
Implications for Aquatic Food Web Studies. Limnol Oceanogr 46:2061-2066
21

Vizzini S, Mazzola A (2003) Seasonal variations in the stable carbon and nitrogen isotope ratios
( 13C/ 12C and 15N / 14N) of primary producers and consurmers in a western Mediterranean coastal
lagoon. Mar Biol 142:1009-1018
Wainright SC, Fry B (1994) Seasonal Variation of the Stable Isotopic Compositions of Coastal Marine
Plankton from Woods Hole, Massachusetts and Geoges Bank. Estuaries 3:552-560
Wu J, Calvert SE, Wong CS (1997) Nitrogen isotope variations in the subarctic northeast Pacific:
relationships to nitrate utilization and trophic structure. Deep-Sea Res Part I 44:287-314
22

Table I. Sampling from six areas along the Greenland coast (Fig. 1). Mean total length, standard deviation (SD) and number (N) sampled are
illustrated for each species in each area.
Species
Size interval
(cm)
Area 1 Area 2 Area 3 Area 4 Area 5 Area 6
70-72°N
West coast
1-5 July
67-69°N
West coast
16 June–11 July
63-65°N
West coast
11 June-21 July
59-61°N
West coast
25-31 July
Total length (mean cm ± SD (N))
59-62°N
East coast
5-8 August
63-66°N
East coast
10-29 August
Anarhichas sp. (Wolffish) 30-60 47 ± 9 (4) 42 ± 4 (5) 33 ± 1 (5) 52 ± 5 (5) 50 ± 0 (1) 48 ± 8 (4)
Boreogadus saida (Polar cod) 5-20 12 ± 3 (5) 11 ± 1 (5) 12 ± 0 (5) 15 ± 2 (5) 11 ± 0 (1) 12 ± 0 (5)
Gadus morhua (Atlantic cod), small 25-35 27 ± 0 (5) 31 ± 1 (5) 28 ± 0 (5) 28 ± 3 (5) 29 ± 2 (5)
Gadus morhua (Atlantic cod), large 45-55 47 ± 1 (5) 48 ± 3 (6) 46 ± 0 (5) 50 ± 2 (5) 49 ± 0 (5)
Hippoglossoides platessoides (American plaice) 20-40 29 ± 1 (5) 31 ± 2 (5) 25 ± 1 (5) 23 ± 1 (5) 25 ± 2 (5) 30 ± 0 (5)
Mallotus villosus (Capelin) 10-15 11 ± 0 (5) 13 ± 0 (5) 13 ± 1 (5) 12 ± 1 (5) 14 ± 1 (5) 12 ± 0 (5)
Reinhardtius hippoglossoides (Greenland halibut) 35-45 39 ± 2 (5) 40 ± 0 (5) 45 ± 3 (5) 45 ± 7 (4) 43 ± 3 (2) 41 ± 1 (5)
Sebastes sp. (Redfish), small 15-25 17 ± 0 (5) 17 ± 8 (5) 16 ± 8 (5) 16 ± 0 (5) 23 ± 2 (4) 22 ± 1 (5)
Sebastes sp. (Redfish), large 35-45 34 ± 1 (3) 38 ± 2 (5) 38 ± 2 (5) 38 ± 3 (5) 39 ± 2 (5) 37 ± 1 (5)
Somniosus microcephalus (Greenland shark) Any 270 ± 0 (1) 445 ± 0 (1) 449 ± 17 (6)
23

Species
Table II: Sampling from eleven stations along the Greenland coast. Mean length, standard deviation (SD) and number (N) sampled are
illustrated for each species at each location. Only N is illustrated for Calanus sp. because the individuals were not measured. At all stations the
krill species was Thysanoessa raschii except at station 10 were it was Meganyctiphanes norvegica.
Station 1
71°N
West coast
4 July
Station 2
69°N
West coast
15 June-11 July
Station 3
66°N
West coast
Station 4
64°N
West coast
17-18 July
Station 5
62°N
West coast
22-23 July
Station 6
61°N
West coast
24-25 July
Length (mean cm ± SD (N))
Station 7
60°N
West coast
28 July
Station 8
60°N
East coast
5-6 August
Station 9
62°N
East coast
7 August
Station 10
64°N
East coast
11 August
24
Station 11
66°N
East coast
18-20 August
Calanus sp. N = 3 N =3 N =3 N =3 N =3 N =3 N =3 N =3 N =3 N =3 N =3
Krill 2.0 ± 0.2 (5) 1.5 ± 0.1 (5) 1.8 ± 0.1 (5) 1.9 ± 0.1 (5) 1.9 ± 0.3 (5) 1.8 ± 0.1 (5) 1.9 ± 0.1 (5) 1.9 ± 0.1 (5) 1.8 ± 0.1 (5) 1.8 ± 0.1 (5)
Filtered POM N = 2 N = 2 N = 2 N = 2 N = 2 N = 2 N = 2 N = 2 N = 2 N = 2 N = 2
Pandalus borealis 2.6 ± 0.1 (5) 2.3 ± 0.1 (5) 2.6 ± 0.1 (5) 2.5 ± 0.1 (5) 2.1 ± 0.2 (5) 2.2 ± 0.2 (5) 2.4 ± 0.4 (4) 2.7 ± 0.2 (5)

Table III. Samples caught in the Nuuk Fjord. All fish species were sampled at the same location except R. hippoglossoides, while krill and
copepods were sampled in a nearby bay (see text and Fig. 1).
Species
Position
(°N;°W)
Size interval
(cm)
Sampling depth
(m)
April June August November
Length (mean cm ± SD (N))
Gadus morhua (Atlantic cod), small 64.25;50.88 30-40 5-20 35 ± 1 (3) 35 ± 3 (5) 34 ± 4 (5) 42 ± 2 (5)
Gadus morhua (Atlantic cod), large 64.25;50.88 50-60 5-20 54 ± 2 (5) 55 ± 3 (4) 55 ± 3 (5) 51 ± 2 (5)
Hippoglossus hippoglossus (Atlantic halibut) 64.25;50.88 35-55 8-15 38 ± 0 (1) 48 ± 10 (2) 49 ± 0 (2) 45 ± 0 (1)
Hippoglossoides platessoides (American plaice) 64.25;50.88 30-40 8-15 39 ± 1 (5) 36 ± 2 (2) 34 ± 5 (5) 37 ± 2 (5)
Mallotus villosus (Capelin) 64.25;50.88 10-15 13 ± 0 (4) 12 ± 1 (2)
Reinhardtius hippoglossoides (Greenland halibut) 64.34;51.16 35-50 300-500 Unknown (2) Unknown (2) Unknown (2)
Sebastes sp. (Redfish), large 64.25;50.88 35-45 60-80 40 ± 1 (5) 41 ± 3 (5) 39 ± 4 (5) 42 ± 2 (5)
Calanus finmarchicus 64.45;50.25 N = 3 N = 3
Thysanoessa raschii 64.45;50.25 1.4 ± 0.1 (5) 1.4 ± 0.1 (5) 1.2 ± 0.1 (5)
Filtered POM 64.21;51.22 10 + 20 N = 1 N = 1 N = 1 N =1
25

Table IV. Species specific comparisons of δ 15N values in the different areas (see Fig 1). indicate a significant difference (P < 0.05), ns indicates
no significant difference between the species in the compared study areas and – indicate species absence in one/both areas. Shaded section is
comparisons between west and east coast.
Species
Comparison of different areas
1 vs. 2 1 vs. 3 1 vs. 4 2 vs. 3 2 vs. 4 3 vs. 4 1 vs. 5 1 vs. 6 2 vs. 5 2 vs. 6 3 vs. 5 3 vs. 6 4 vs. 5 4 vs. 6 5 vs. 6
Anarhichas sp. (Wolffish) ns ns ns ns ns ns
Boreogadus saida (Polar cod) ns ns ns ns ns ns ns ns ns ns
Gadus morhua (Atlantic cod), small - - - ns - - ns
Gadus morhua (Atlantic cod), large - - - ns - - ns ns ns
Hippoglossoides platessoides (American plaice) ns * ns ns ns ns ns
Mallotus villosus (Capelin) ns ns ns ns ns ns ns
Reinhardtius hippoglossoides (Greenland halibut) ns ns ns ns ns ns ns ns ns ns ns ns ns ns
Sebastes sp. (Redfish), small ns ns ns ns ns
Sebastes sp. (Redfish), large ns ns ns ns ns ns ns ns
Calanus sp. (Copepod) ns ns ns ns ns ns ns ns ns ns ns ns ns ns
Thysanoessa raschii (Krill) ns ns ns ns ns ns
Pandalus borealis (Shrimp) ns ns ns ns ns ns ns ns
26

Table V. Trophic level estimated according to Nilsen et al. (2008) and Søreide et al. (2006). Calanus sp.
were used as baseline in trophic level estimation. The mean trophic level is an unweighted mean
across areas.
Species
Trophic Level
Area 1 Area 2 Area 3 Area 4 Area 5 Area 6 Mean
Calanus sp. (Copepods) 2.00 2.00 2.00 2.00 2.00 2.00 2.00
Thysanoessa raschii (Krill) 2.42 2.13 2.49 2.35 2.48 2.33 2.37
Pandalus borealis (Shrimp) 2.56 2.62 2.95 2.91 3.02 2.61 2.78
Mallotus villosus (Capelin) 3.01 3.09 2.86 2.98 2.90 2.49 2.89
Sebastes sp. (Redfish), small 2.93 3.00 3.16 2.89 3.29 2.75 3.01
Boreogadus saida (Polar cod) 2.79 2.98 3.56 3.38 3.68 2.70 3.18
Gadus morhua (Atlantic cod), small 3.43 3.76 3.28 3.91 2.44 3.37
Sebastes sp. (Redfish), large 3.70 3.49 3.45 3.94 2.99 2.98 3.43
Hippoglossoides platessoides (American plaice)
3.48 3.71 3.69 3.60 3.28 3.04 3.47
Gadus morhua (Atlantic cod), large
3.75 3.75 3.83 3.57 2.97 3.57
Reinhardtius hippoglossoides (Greenland halibut)
3.28 3.32 3.74 3.51 4.42 3.30 3.60
Anarhichas sp. (Wolf fish)
4.01 3.99 3.85 3.66 4.00 2.98 3.74
Somniosus microcephalus (Greenland shark)
4.34
4.88 5.09
4.77
27

Table VI. Comparison of the relative δ 13C and δ 15N values in study areas along the Greenlandic West
and East coast. indicate a significant difference (P < 0.05) while ns indicate no difference between
the two areas in question (P > 0.05).
Area 2 Area 3 Area 4 Area 5 Area 6
δ 13C δ 15N δ 13C δ 15N δ 13C δ 15N δ 13C δ 15N δ 13C δ 15N
Area 1 ns ns
Area 2 ns ns ns ns
Area 3 ns ns
Area 4 ns ns ns ns
Area 5 ns ns
28

Table VII. A comparison of different ecosystems in different water masses, based on the current study and literature review. Data
from Iceland is by Sarà et al. (2009), Northeast Water Polynya is by Hobson et al. (1995) and Greenland Area 1 and 6 is from the
current study (Fig. 1). a indicate standard error instead of standard deviation, - indicate data not available. All values from Hobson
et al. (1995) are lipid extracted.
Species
Iceland,
Irminger Current
Northeast Water Polynya,
East Greenland Current
Greenland, Area 1,
West Greenland Current
Stable isotope values (‰ ± SD)
Greenland, Area 6,
East Greenland Current
δ 15N δ 13C δ 15N δ 13C δ 15N δ 13C δ 15N δ 13C
Anarhichas sp./Icelus bicornis 13.1 ± 1.7 -16.5 ± 0.6 13.7 ± 0.4 -21.0 ± 0.2 16.2 ± 0.4 -18.5 ±0.9 11.3 ± 0.5 -18.5 ± 0.5
Boreogadus saida 13.7 ± 0.0 -21.6 ±0.5 12.2 ±0.5 -20.4 ± 0.4 10.4 ± 0.6 -22.6 ± 0.2
Gadus morhua (Atlantic cod), small 12.8 ± 0.1 -17.0 ± 0.5 - - - - 9.5 ± 0.7 -20.3 ± 0.1
Hippoglossus platessoides/Lycodes rossi 12.0 ± 0.7 -17.5 ± 0.3 15.3 ± 0.6 -20.5 ± 0.3 14.4 ±1.0 -19.5 ± 0.2 11.4 ± 0.2 -19.4 ± 0.2
Mallotus villosus/Micromesistus poutassou 10.8 ± 0.1 -22.5 ± 0.3 13.1 ± 0.4 -21.3 ±0.2 12.9 ±0.1 -20.5 ±0.1 9.7 ±0.9 -20.7 ± 0.5
Sebastes sp. (redfish), small 11.6 ± 0.1 -18.6 ± 0.1 14.5 ± 0.0 -21.4 ± 0.0 12.7 ±0.3 -20.4 ± 0.1 10.5 ± 0.1 -20.6 ± 0.3
Calanus sp. 3.5 ± 0.1 a -20.3 ± - a 8.2 ± 0.0 -22.3 ± 0.0 9.7 ±0.0 -22.1 ± 0.4 8.1 ± 0.4 -21.2 ±0.2
Krill/Themisto sp. 11.0 ± 0.3 -17.9 ± 0.2 10.4 ± 0.0 -24.2 ± 0.0 11.1 ± 0.8 -20.8 ±0.4 9.2 ±1.1 -20.9 ± 0.3
POM 5.2 ± 0.5 -21.1 ± 0.4 5.0 ±0.0 -28.4 ±0.0 10.8 ± 2.2 -25.1 ± 0.2 3.3 ± 0.3 -25.3 ± 0.3
29

Figure 1. The southern part of Greenland. The offshore squares represent the study areas in which all
fish species were sampled. The gray circles mark the stations where shrimps, krill, copepod, POM and
CTD-measurements (CTD-measurements were only sampled on the West coast) were sampled. The
line indicates the 500 meter depth contour and thereby the shelf edge. The small map shows the Nuuk
fjord system with Nuuk city, fish (except Greenland halibut) and POM station (open square), sampling
station of copepod and krill (black circle) and the sampling site of Greenland halibut (black triangle)
marked.
30

Figure 2. δ 13C and δ 15N (‰) values (mean ± SE) along the Greenlandic coast for the analyzed species:
Filtered POM (POM), copepod (COP), krill (KRI), shrimp (SHR), capelin (CAP), polar cod (POL), small
redfish (RES), large redfish (REL), American plaice (APL), small cod (ATS), large cod (ATL), Greenland
halibut (GHL), wolffish (WOF) and Greenland shark (GSK).
31

Figure 3. The general pattern of δ 15N (‰) and δ 13C values (mean ± SE) along the Greenlandic west
(area no. 1, 2, 3 and 4) and east coast (area no. 5 and 6), represented by shrimp. Note the difference in
scale on the y-axis.
32

Figure 4. The calculated relative values (± SE) of δ 13C and δ 15N against latitude (°N). Note the
difference in scale on the y-axis.
33

Figure 5. δ 13C and δ 15N (‰) values (mean ± SE) in the Nuuk Fjord for the analyzed species: Filtered
POM (POM), copepod (COP), krill (KRI), capelin (CAP), small cod (ATS), large cod (ATL), redfish (large)
(REL), American plaice (APL), Greenland halibut (GHL), Atlantic halibut (AHL)).
34

Figure 6. Mean index values of δ 13C and δ 15N for each sampling date in the Nuuk Fjord during the
sampling period in 2010. Mean index values are calculated based on the mean values of each species in
each sampling period divided by the mean of the given species at all sampling dates, and then a mean
value of all species at each sampling dates were calculated. The mean index values of δ 13C and δ 15N is
in this way centered around one. All δ 15N data points are displaced two days from the original
sampling day. Error bars represent standard error.
35

Figure 7. Inshore and offshore comparison of δ 15N and δ 13C (‰) values (mean ± SE). indicate
significant difference (P ≤ 0.05).
36

Figure 8. Mean trophic level (± SE) comparison among species inshore and offshore. indicate a
significant difference (P < 0.05).
37

Figure 9. Relative δ 15N values of the study areas offshore plotted against productive open-water
period.
38

Figure 10. CTD-profiles from the west coast showing the temperature and salinity with depth.
39