Electronic Appendix to:Ehrich D, Ims RA, Yoccoz NG, Lecomte N, Killengreen ST, Fuglei E, Rodnikova AY, Ebbinge BS, Menyushina IE, Nolet BA, Pokorvsky IG, Popov IY, Schmidt NM, Sokolov AA, Sokolova NA, Sokolov VA. Can top predator diet as inferred from stable isotopes be used as indicator of shifting tundra ecosystem state?
Supplementary Methods 1: Sample Storage and Preparation for Stable Isotope Analysis
Fur samples were stored dry in paper or plastic bags; muscle and egg content samples were stored in 70% ethanol. All samples were transferred to the laboratory of the University of Tromsø for storage at room temperature and further analysis.
Muscle and egg content samples were prepared for stable isotope analysis as described in Ehrich and others (2011) and Feige and others (2012). Muscle samples were dried for at least 48 h at 60 °Cin 2 ml Eppendorf tubes and ground to a fine powder using a Mixer Mill (MM301; RetschGmbH & Co. Haan, Germany). Whole egg samples were homogenized before being driedfor at least 48 h at 60 °C. Fur samples were cleaned by washing them at least twice using a sonicator, and fat was chemically removed with a mixture of chloroform and methanol (2:1) according to Roth and Hobson (2000).Samples were then dried and clipped to small particles. All samples were weighed into tin capsules (precision ± 0.01 mg) before being sent to SINLAB (University of New Brunswick) for analysis in a CarloErba NC2500 elemental analyzer connected via continuous-flow to a Finnigan Mat Delta Plus isotope-ratio mass spectrometer.Lipids were extracted chemically from some samples as described in Ehrichand others(2011). For these samples we used δ15N values from samples processed without lipid extraction together withδ13C values after lipid extraction for further analyses. For all other samples, δ13C values were corrected for lipid content according to Ehrich and others (2011) if necessary. Isotope signatureswere represented as permil (‰) ratios referenced against Peedee belemnite carbonate (PDB) for δ13C and atmospheric nitrogen (AIR) for δ15N, according to the standard relationships [(13C/12Csample)/(13C/12CPDB) - 1]x 1000 and [(15N/14Nsample)/(15N/14NAIR) - 1]x 1000, respectively.The measurement error was assessed by repeating 87 samples comprising different species, tissues and sampling sites. The precision was found to be 0.28‰ on average for both isotopes (maximum errors were 3.77‰ for δ13C and 4.85‰ for δ15N most likely resulting from wrong labelling). This error rate includes both the variation introduced by imperfect homogenization of the samples, human error such as sample mixing, and errors of the spectrophotometer measurements (Bonin and others 2004).
Supplementary Methods 2: Assembling Groups of Prey for Plotting and Stable Isotope Mixing Models
Most prey samples were collected in 2007 and in 2008 and a few in later years. Resource groups used for plotting and in the mixing models included samples from different years and prey signatures were assumed to remain constant from year to year. Support for this assumption was assessed by plotting samples from different years for sites and prey species with large enough sample sizes. No consistent differences among years were observed. Differences among seasons were however observed for several herbivores such as reindeer, hare or some small rodents. Therefore only samples from the summer or fall were used to calculate means and included in further analyses. Means for groups including several species were obtained by first calculating means for each species, and then averaging these means to obtain an overall estimate for the group. Species names without specification of tissue refer to muscle, and Latin names for all species are provided in Table S1.
Wrangel
Biplot with all prey samples analyzed from Wrangel Island. For both species of lemmings only samples from the inland site on river Neyzvestnaya were used for analysis, as most fox samples were from this part of the island. Samples from the southern coast had somewhat higher δ15N values and were excluded from the analysis. The passerine sample with a very low δ15N signature (indicated by an arrow) was from an arctic redpoll (Acanthishornemanni), a species which is rather uncommon on Wrangel. This sample was excluded. Common eiders(Somateriamollissima) and waders were separated according to their isotopic signatures into marine or terrestrial: The ruddy turnstone (Arenariainterpes) was considered a marine prey, whereas grey plover (Pluvialissquatarola) was terrestrial. Only one common eider sample was considered terrestrial (highlighted by a black square). Marine samples are surrounded by a blue line.
Biplotwhich shows means and standard deviations for stable isotope signatures of prey species from Wrangel.For the mixing model analysis, preys were assembled in 5 groups: marine prey, large herbivores (reindeer and muskox), terrestrial birds, collared lemmings and siberian lemmings. The common eider sample with a terrestrial signature was placed together with the other terrestrial birds.
Taimyr
Biplot with all prey samples analysed from MysVostochny, western Taimyr. As no purely marine samples were collected on Taimyr, we used three samples of marine mammals from further west in the Kara Sea and two samples of common eider muscle from the Lena Delta area to indicate marine signatures for this site. Two muscle samples of goslings and on goose egg sample had unusually high δ15N values (indicated by the arrow). These were from a small island with a high concentration of nesting geese and gulls, leading to strong enrichment of the vegetation in δ15N (Erskine and others 1998). They were excluded from further analyses as outliers.
Biplot which shows means and standard deviations for stable isotope signatures of prey species from Taimyr. For the mixing model analysis, preys were assembled in 5 groups: collared lemmings, siberian lemmings, rock ptarmigan eggs, other birds and marine resources. River fish were considered a less likely major resource than the other prey groups, and were therefore not included in the mixing models.
Yamal:
Biplot with all prey samples collected in summer and fall analysed from Erkuta, southern Yamal. As the northern red-backed vole (Myodesrutilus) is a rare species at this site, it was excluded from further analyses.
Biplot which shows means and standard deviations for stable isotope signatures of prey species from Yamal.The bird categories (willow ptarmigan, passerine, wader, goose, duck and gull) refer to samples of both muscle and eggs. For the mixing model analysis, preys were assembled in 5 groups: collared lemmings, willow ptarmigan, reindeer, voles and birds, and marine resources. As mountain hare had isotopic values which were very close to collared lemmings and because we aimed at keeping lemmings in a separate group, we excluded hare as resource for the mixing model (see main text for discussion of this choice). River fish were considered a less likely major resource than the others, and were therefore not included in the mixing models.
Varanger:
Biplot with prey samples collected from Varanger Peninsula. As isotopic values of small rodents varied with season, only individuals trapped in the beginning of September are included here. Mountain hare samples were from animals shot in March, and were therefore not include in further analyses. As no marine samples were available from Varanger, samples from Svalbard further north in the Barents Sea were used to indicate marine isotope signatures. Ptarmigan include both willow (Lagopuslagopus) and rock ptarmigan (Lagopusmuta), which did not differ in average isotopic signatures.
Biplot which shows means and standard deviations for stable isotope signatures of prey species from Varanger. For the mixing model analysis, preys were assembled in 5 groups: Norwegian lemmings and tundra voles together, grey-sided voles, reindeer and ptarmigantogether, other birds, and marine resources.
Svalbard:
Biplot with prey samples analyzed from Svalbard.
Biplot which shows means and standard deviations for stable isotope signatures of prey species from Svalbard. For the mixing model analysis, preys were assembled in 3 groups: Geese, reindeer and rock ptarmigan, and marine resources.
Zackenberg:
Biplot with prey samples analyzed from Zackenberg, eastern Greenland.
Biplot which shows means and standard deviations for stable isotope signatures of prey species from Zackenberg. For the mixing model analysis, preys were assembled in 4 groups: collared lemmings; muskox, hare and rock ptarmigan; other terrestrial birds; and marine resources. Fish were considered a less likely major resource than the others, and were therefore not included in the mixing models.
Table S1: Mean Stable Isotope Signatures of Prey Species from Each Site with Standard Deviations are Given in ‰
Species / n / Mean δ13C / ± / SD / Mean δ15N / ± / SDWrangel
Lemmussibiricusportenkoi / 14 / -26.9 / ± / 0.8 / 8.3 / ± / 3.3
Dicrostonyxvinogradovi / 21 / -26.1 / ± / 0.8 / 2.5 / ± / 1.3
Ovibosmoschatus / 2 / -23.7 / ± / 0.7 / 4.4 / ± / 0.6
Rangifertarandus / 6 / -23.5 / ± / 0.4 / 5.9 / ± / 0.8
Odobenusrosmarus / 2 / -17.7 / ± / 0.9 / 17.9 / ± / 3.9
Chen caerulescensM / 4 / -25.8 / ± / 1.1 / 7.1 / ± / 0.6
Chen caerulescensE / 12 / -24.9 / ± / 1.2 / 7.1 / ± / 0.8
SomateriamollissimaM / 7 / -19.8 / ± / 2.2 / 14.4 / ± / 3.8
SomateriamollissimaE / 3 / -22.1 / ± / 3.1 / 13.4 / ± / 2.4
ArenariainterpesM / 3 / -16.9 / ± / 1.0 / 12.8 / ± / 0.9
PluvialissquatarolaM / 3 / -24.7 / ± / 0.2 / 5.2 / ± / 0.5
PlectrophenaxnivalisM / 8 / -24.9 / ± / 0.3 / 5.9 / ± / 0.8
Acanthishornemanni M / 1 / -22.7 / 0.7
Oncorhynchusgorbusha / 1 / -19.7 / 11.3
Taimyr
Lemmussibiricus / 21 / -27.3 / ± / 0.3 / 5.9 / ± / 1.0
Dicrostonyxtorquatus / 4 / -27.1 / ± / 0.3 / 2.5 / ± / 0.9
Anser albifronsE / 5 / -26.0 / ± / 0.6 / 7.0 / ± / 0.3
BrantaberniclaM / 3 / -24.7 / ± / 0.8 / 16.4 / ± / 5.0
BrantaberniclaE / 13 / -26.3 / ± / 1.6 / 9.5 / ± / 3.0
BrantaruficollisE / 1 / -26.2 / 7.5
ClangulahyemalisM / 1 / -23.5 / 9.4
SomateriaspectabilisE / 1 / -25.6 / 7.2
ArenariainterpresE / 1 / -25.5 / 8.7
Calidrisalpina E / 3 / -23.5 / ± / 0.6 / 6.5 / ± / 0.4
Calidrisferruginea E / 1 / -22.5 / 6.3
Calidrisminuta M / 4 / -23.7 / ± / 1.7 / 8.1 / ± / 1.4
Calidrisminuta E / 3 / -24.5 / ± / 0.2 / 7.3 / ± / 0.5
Charadriushiaticula E / 1 / -24.0 / 8.2
Phalaropusfulicarius E / 1 / -24.7 / 7.2
Pluvialisfulva M / 1 / -23.0 / 6.5
Pluvialisfulva E / 5 / -23.3 / ± / 0.7 / 6.7 / ± / 0.7
Pluvialissquatarola E / 5 / -23.8 / ± / 1.5 / 6.7 / ± / 0.5
Larusheuglini E / 10 / -25.1 / ± / 0.6 / 9.2 / ± / 1.2
Larushyperboreus E / 2 / -23.4 / ± / 0.5 / 12.2 / ± / 2.1
Lagopus muta E / 3 / -24.5 / ± / 0.1 / 2.2 / ± / 0.4
Anthuscervinus M / 1 / -25.1 / 8.7
Calcariuslapponicus M / 4 / -24.9 / ± / 0.5 / 7.5 / ± / 0.6
Eremophila alpestris M / 6 / -25.0 / ± / 0.9 / 7.2 / ± / 3.3
Lusciniasvecica M / 1 / -24.7 / 8.3
Oenantheoenanthe M / 1 / -24.2 / 7.9
Plectrophenaxnivalis M / 5 / -25.4 / ± / 0.6 / 6.7 / ± / 0.8
Plectrophenaxnivalis E / 3 / -24.4 / ± / 0.4 / 7.9 / ± / 1.1
Coregonusautumnalis / 2 / -24.5 / ± / 0.9 / 13.1 / ± / 0.1
Coregonuslavaretus / 3 / -25.4 / ± / 1.3 / 12.0 / ± / 1.1
Coregonusmuksun / 3 / -26.9 / ± / 0.2 / 13.4 / ± / 0.5
Coregonusnasus / 2 / -22.4 / ± / 2.5 / 11.0 / ± / 0.5
Lotalota / 1 / -25.9 / 14.6
Yamal
Dicrostonyxtorquatus / 18 / -26.3 / ± / 0.5 / 1.3 / ± / 1.4
Microtusgregalis / 20 / -26.5 / ± / 0.4 / 4.9 / ± / 1.6
Microtusmiddendorffii / 20 / -25.5 / ± / 0.7 / 5.8 / ± / 1.1
Lepustimidus / 8 / -25.9 / ± / 0.3 / 2.4 / ± / 0.6
Rangifertarandus / 9 / -23.4 / ± / 0.5 / 3.1 / ± / 0.3
Erignathusbarbatus / 1 / -20.7 / 14.9
Phocahispida / 1 / -22.0 / 14.8
Anser albifrons M / 1 / -26.6 / 6.1
Anser sp. M / 1 / -23.9 / 9.7
Anas acuta E / 1 / -26.2 / 7.0
Anas crecca E / 1 / -25.6 / 6.7
Anas penelope E / 1 / -25.8 / 7.3
Clangulahyemalis M / 7 / -26.0 / ± / 2.3 / 8.8 / ± / 0.7
Clangulahyemalis E / 2 / -26.7 / ± / 2.2 / 8.7 / ± / 0.7
Melanittanigra M / 1 / -25.2 / 9.8
Calidristemmincki M / 1 / -24.8 / 8.3
Phalaropuslobatus M / 1 / -29.0 / 7.2
Philomachuspugnax M / 1 / -27.1 / 8.5
Pluvialisapricaria E / 1 / -22.0 / 5.4
Tringaglareola E / 1 / -26.3 / 4.6
Larussp. M / 1 / -27.0 / 8.1
LagopuslagopusM / 3 / -24.8 / ± / 0.4 / 1.7 / ± / 1.3
LagopuslagopusE / 8 / -24.5 / ± / 0.5 / 2.3 / ± / 1.2
AnthuscervinusM / 6 / -25.6 / ± / 1.3 / 5.6 / ± / 1.3
AnthuscervinusE / 1 / -24.3 / 5.0
AnthuspratensisM / 3 / -25.3 / ± / 0.8 / 6.1 / ± / 0.1
CalcariuslapponicusM / 6 / -24.9 / ± / 1.4 / 6.2 / ± / 1.2
CarduelisflammeaM / 2 / -23.8 / ± / 0.1 / 2.5 / ± / 0.7
Lusciniasvecica M / 3 / -26.0 / ± / 0.3 / 7.0 / ± / 0.9
Phyloscopuscollybita M / 2 / -26.4 / ± / 0.3 / 5.5 / ± / 0.4
Ripariariparia M / 1 / -27.3 / 7.8
Turdusiliacus M / 1 / -23.6 / 5.2
Coregonuspeled / 1 / -27.5 / 9.4
Coregonuspischian / 1 / -29.2 / 10.2
Eleginusnavaga / 1 / -20.5 / 13.8
Esoxlucius / 2 / -26.7 / ± / 0.1 / 11.8 / ± / 0.7
Thymallusthymallus / 1 / -27.6 / 11.5
Varanger
Lemmuslemmus / 78 / -27.8 / ± / 0.7 / 1.6 / ± / 0.7
Microtusoeconomus / 40 / -27.5 / ± / 0.6 / 2.8 / ± / 1.3
Myodesrufocanus / 66 / -26.5 / ± / 0.6 / 1.0 / ± / 1.0
Lepustimidus / 6 / -26.6 / ± / 0.3 / 3.6 / ± / 0.2
Rangifertarandus / 13 / -25.2 / ± / 1.1 / 2.3 / ± / 1.4
Anser fabalis M / 1 / -24.8 / 2.5
Melanittanigra E / 6 / -23.8 / ± / 0.8 / 7.3 / ± / 0.2
Pluvialisapricaria E / 1 / -24.8 / 4.1
Lagopuslagopus M / 16 / -25.6 / ± / 0.7 / 1.2 / ± / 0.7
Lagopus muta M / 6 / -25.1 / ± / 0.9 / 1.1 / ± / 1.2
Anthuspratensis M / 8 / -25.4 / ± / 1.7 / 4.9 / ± / 0.6
Lusciniasvecica M / 1 / -26.0 / 5.9
Oenantheoenanthe M / 1 / -24.3 / 5.8
Svalbard
Rangifertarandus / 39 / -25.7 / ± / 0.3 / 4.2 / ± / 1.3
Anser brachyrhynchus M / 2 / -28.6 / ± / 0.5 / 15.0 / ± / 2.4
Anser brachyrhynchus E / 4 / -27.4 / ± / 0.5 / 21.0 / ± / 3.7
Brantaleucopsis M / 2 / -27.9 / ± / 0.0 / 9.9 / ± / 0.6
Brantaleucopsis E / 2 / -27.8 / ± / 0.1 / 11.6 / ± / 2.6
Somateriamollissima M / 1 / -20.0 / 8.9
Somateriamollissima E / 1 / -18.8 / 10.3
Alle alle M / 6 / -20.8 / ± / 0.9 / 11.2 / ± / 0.3
Fraterculaarctica M / 1 / -20.6 / 12.5
Rissa tridactyla E / 10 / -19.0 / ± / 0.8 / 11.6 / ± / 1.5
Uria lomvia M / 1 / -19.8 / 12.1
Uria lomvia E / 10 / -19.7 / ± / 1.6 / 11.6 / ± / 0.8
Lagopus muta M / 7 / -26.6 / ± / 0.6 / 4.0 / ± / 1.5
Clupeidfish / 1 / -19.8 / 12.6
Crab / 1 / -18.8 / 7.0
Zackenberg
Dicrostonyxgroenlandicus / 3 / -26.2 / ± / 0.6 / -0.4 / ± / 0.9
Lepusarcticus / 1 / -23.4 / 1.8
Ovibosmoschatus / 7 / -24.2 / ± / 0.5 / 3.1 / ± / 0.7
Erignathusbarbatus / 1 / -19.8 / 16.7
Pusasp. / 1 / -19.2 / 18.6
Anser sp. E / 1 / -24.7 / 8.3
Clangulahyemalis E / 3 / -19.3 / ± / 0.2 / 10.3 / ± / 0.8
Arenariainterpes E / 2 / -24.5 / ± / 0.3 / 7.6 / ± / 0.4
Calidrisalpina E / 2 / -22.7 / ± / 1.8 / 5.9 / ± / 0.4
Calidriscanutus E / 4 / -22.0 / ± / 1.0 / 4.1 / ± / 0.8
Larushyperboreus E / 1 / -24.6 / 8.2
Lagopus muta E / 1 / -23.0 / -0.2
Salvelinusalpinus / 5 / -21.6 / ± / 0.2 / 14.6 / ± / 0.5
Δ13C values of muscle samples have been corrected for lipid content arithmetically according to (Ehrich and others 2011). For birds, egg (E) and muscle (M) values are reported separately. Δ13C values of egg samples were obtained after chemical extraction of fat or using arithmetical correction according to Ehrich and others (2011). The samples from Taimyr have been previously published in (Feige and others 2012) and some of the samples from Varanger have been published in (Killengreen and others 2011).
Figure S1. Proportions of different groups of prey in the diet of arctic foxes at six study sites over several years as estimated by Bayesian isotope mixing models (R package siar). White points represent the mode of the posterior probability distribution and boxes the 50, 75 and 95% probability intervals respectively. The probability intervals represent both individual variability and uncertainty resulting from the MCMC estimation. The dotted line shows the prior proportions (equal proportions assumed as priors throughout). The following prey groups were used (for details about prey groups see section supplementary methods 2 above): cL=collared lemming, sL=Siberian lemming, L=Norwegian lemming, Mo=tundra vole, Mr=grey-sided vole, V=voles, rP= rock ptarmigan, wP=willow ptarmigan, P= rock and willow ptarmigan, Geese, B=other birds, M=muskox, R=reindeer, MPH=muskox, rock ptarmigan and hare, and Mar=marine resources. Lemming years are highlighted in bold.
Figure S2. Size of the isotopic niche of arctic foxes at six arctic study sites in different years estimated as Bayesian ellipses (SIBER, (Jackson and others 2011)). The white dot shows the mode of the posterior distribution and the boxes show 50, 75 and 95% posterior probability intervals. Lemming years are highlighted in bold.
Supplementary references:
Bonin A, Bellemain E, Eidesen PB, Pompanon F, Brochmann C, Taberlet P. 2004. How to track and assess genotyping errors in population genetics studies. Molecular Ecology 13: 3261-3273.
Ehrich D, Tarroux A, Stien J, Lecomte N, Killengreen ST, Berteaux D, Yoccoz N. 2011. Stable isotope analysis: modelling lipid normalization for muscle and eggs from arctic mammals and birds. Methods in Ecology and Evolution 2: 66-76.
Erskine PD, Bergstrom DM, Schmidt S, Stewart GR, Tweedie CE, Shaw JD. 1998. Subantarctic Macquarie Island - a model ecosystem for studying animal-derived nitrogen sources using N-15 natural abundance. Oecologia 117: 187-193.
Feige N, Ehrich D, Popov IY, Broekhuizen S. 2012. Monitoring Least Weasels after a Winter Peak of Lemmings in Taimyr: Body Condition, Diet and Habitat Use. Arctic 65: 273-282.
Jackson AL, Inger R, Parnell AC, Bearhop S. 2011. Comparing isotopic niche widths among and within communities: SIBER - Stable Isotope Bayesian Ellipses in R. Journal of Animal Ecology 80: 595-602.
Killengreen ST, Lecomte N, Ehrich D, Schott T, Yoccoz N, Ims RA. 2011. The importance of marine vs. human-induced subsidies in the maintenance of an expanding mesocarnivore in the arctic tundra. Journal of Animal Ecology 80: 1049-1060.
Roth JD, Hobson KA. 2000. Stable carbon and nitrogen isotopic fractionation between diet and tissue of captive red fox: implications for dietary reconstruction. Canadian Journal of Zoology 78: 848-852.
1