Ecology of Chaetognaths (Semi-Gelatinous Zooplankton) in Arctic Waters

Home , Chaetognatha, Cyclopoida, Parasagitta elegans, Sagittidae

Ecology of chaetognaths (semi-gelatinous zooplankton) in Arctic waters

Thèse

Jordan Grigor

Doctorat interuniversitaire en océanographie Philosophiæ doctor (Ph. D.)

Québec, Canada

Ecology of chaetognaths (semi-gelatinous zooplankton) in Arctic waters

Thèse

Jordan Grigor

Sous la direction de :

Louis Fortier, directeur de recherche

Résumé Les chaetognathes sont d’importants membres des communautés mésozooplanctoniques de l’Arctique en ce qui a trait à l'abondance et à la biomasse. Les chaetognathes de l’Arctique se répartissent en trois espèces principales qui sont considérées comme étant strictement carnivores : Eukrohnia hamata, Parasagitta elegans et Pseudosagitta maxima. Cette étude utilise un ensemble de données de filet planctoniques recueillies sur une période de 5 ans dans les régions européennes, canadiennes et de l'Alaska de l’Arctique (2007, 2008, 2012, 2013, 2014) et comprend un cycle annuel complet dans l'Arctique canadien (2007-2008), le but étant d’améliorer notre compréhension sur les distributions, les cycles de vie et les stratégies d'alimentation du E. hamata et du P. elegans. Dans la présente thèse, les points suivants seront abordés : (1) la stratégie d'alimentation et la maturité du P. elegans dans l'Arctique européen durant la nuit polaire en 2012 et 2013, (2) les cycles de croissance et de reproduction, les stratégies d'alimentation et les distributions verticales du E. hamata et du P. elegans dans l'Arctique canadien de 2007 à 2008, et (3) les différences spatiales dans les stratégies d'alimentation du E. hamata et du P. elegans à l'automne 2014. Afin d’étudier leurs stratégies d'alimentation, des analyses de contenu du tube digestif ainsi que des techniques biochimiques ont été utilisées. Dans l'Arctique canadien, le E. hamata et le P. elegans vivent tous deux pendant environ 2 ans. Le P. elegans colonise principalement les eaux épipélagiques, tandis que le E. hamata colonise principalement les eaux mésopélagiques. Dans cette région, P. elegans se reproduit en continue de l'été au début de l'hiver, dans la période de forte biomasse de copépodes, qui constituent ses proies, dans les eaux proches de la surface, un mode de reproduction basé sur l’apport immédiat d’énergie. Cependant, les résultats ont révélé que E. hamata a engendré des couvées distinctes dont on peut voir l’évolution au cours de fenêtres de reproduction séparées, à la fois durant les périodes de printemps-été et d’automne-hiver, ce qui suggère une reproduction basée sur les réserves. Les taux de prédation quotidiens évalués à partir des analyses du contenu du tube digestif sont généralement restés faibles pour les deux espèces de chaetognathes. Toutefois, pour E. hamata et P. elegans, les taux de prédation inférés en été-automne ont dépassé ceux de l’hiver-printemps. Des études d’alimentation ont révélé que E. hamata consomme de la matière organique particulaire (éventuellement des chutes de neige marine) tout au long de l'année, mais surtout en été, alors que le P. elegans se nourrit différemment. Les deux espèces

iii

sont caractérisées par une forte croissance estivale. La croissance hivernale du P. elegans était grandement restreinte, tandis que celle du E hamata l’était moindrement. En somme, les différences dans la façon dont les lipides et la neige marine sont utilisés par les deux espèces pourraient expliquer les différences dans leurs cycles de reproduction et leurs patrons de croissance saisonnière.

Abstract Chaetognaths are important members of Arctic mesozooplankton communities in terms of abundance and biomass. Despite this, the bulk of seasonal studies have focused on grazing copepods. Arctic chaetognaths comprise three major species which are thought to be strict carnivores: Eukrohnia hamata, Parasagitta elegans and Pseudosagitta maxima. This thesis uses datasets collected from plankton net sampling during five years in European, Canadian and Alaskan areas of the Arctic (2007, 2008, 2012, 2013, 2014) and includes a full annual cycle in the Canadian Arctic (2007-2008), the purpose being to improve our understanding of the distributions, life history and feeding strategies of E. hamata and P. elegans. The following topics are addressed: (1) the feeding strategy and maturity of P. elegans in the European Arctic during the polar night in 2012 and 2013; (2) the growth, breeding cycles, feeding strategies and vertical distributions of E. hamata and P. elegans, in the Canadian Arctic from 2007 to 2008; and (3) spatial differences in the feeding strategies of E. hamata and P. elegans in autumn 2014. To investigate feeding strategies, a combination of gut contents and biochemical techniques was used. In the Canadian Arctic, both E. hamata and P. elegans live for around 2 years. P. elegans mainly colonized epi-pelagic waters, whereas E. hamata mainly colonized meso-pelagic waters. In this region, P. elegans reproduced continuously from summer to early winter when copepod prey peak in near-surface waters. This is characteristic of income breeders. However, results for E. hamata revealed that this species spawned distinct and traceable broods during separate reproductive windows in both spring-summer and autumn-winter, suggesting capital breeding. Daily predation rates inferred from gut content analyses appeared to be generally low in the two chaetognath species, though inferred predation rates in summer-autumn exceeded those in winter-spring. Feeding studies revealed that E. hamata consumed particulate organic matter (possibly falling marine snow) throughout the year but especially in the summer, whereas P. elegans did not feed in this way. High summer growth seems to be a characteristic of both these species. Growth during winter was highly restricted in P. elegans, to a lesser extent in E. hamata. In summary, differences in how lipids and marine snow are utilised by the two species could explain differences in their breeding cycles and seasonal growth patterns.

Table of contents

Résumé iii

Abstract v

Acknowledgements xviii

Foreword xx

1. Chapter 1 – General Introduction 1

1.1 The Arctic Ocean 1 1.1.1 Physical environment 2 1.1.2 Arctic marine ecosystems 3 1.2 Zooplankton and their polar adaptations 5 1.3 Chaetognaths 8 1.3.1 Arctic species and distributions 9 1.3.2 Lifespans 10 1.3.3 Reproductive strategy 11 1.3.4 Oil vacuoles in Eukrohnia spp. 11 1.3.5 Feeding strategies and trophic importance 12 1.3.6 Fatty acids and stable isotopes 14 1.4 Climate change and other challenges for Arctic marine life 14 1.5 Study areas 17 1.6 Aims and objectives 18

2. Chapter 2 – Polar night ecology of a pelagic predator, the chaetognath Parasagitta elegans 20

2.1 Résumé 20 2.2 Abstract 21 2.3 Introduction 22 2.4 Method 23 2.4.1 Study area 23 2.4.2 Physical and biological environment 24 2.4.3 Zooplankton sampling 24 2.4.4 Sample processing 25 2.4.5 Gut content analyses 26 2.4.6 Food-containing ratio and feeding rate 26 2.4.7 Stable isotope analyses 27 2.4.8 Determination of trophic level 28 2.4.9 Fatty acid analyses 28 2.4.10 Mid-winter maturity status 29 2.5 Results 30 2.5.1 Chaetognath abundance and prey field 30

2.5.2 Gut contents 32 2.5.3 Body composition 34 2.5.4 Mid-winter maturity status 36 2.6 Discussion 36 2.6.1 Studies during the polar night 36 2.6.2 Feeding activity and rates 36 2.6.3 Energetics 38 2.6.4 Lipid profile 38 2.6.5 Reproduction 40 2.7 Concluding remarks 41

3. Chapter 3 – Growth and reproduction of the chaetognaths Eukrohnia hamata and Parasagitta elegans in the Canadian Arctic Ocean: capital breeding versus income breeding 42

3.1 Résumé 42 3.2 Abstract 43 3.3 Introduction 44 3.4 Method 46 3.4.1 Study area 46 3.4.2 Sampling 47 3.4.3 Chaetognath body size and sampler efficiency 48 3.4.4 Hatching and cohort development 49 3.4.5 Estimation of maturity and oil vacuole area 50 3.4.6 Vertical distributions 50 3.5 Results 51 3.5.1 Physical environment and primary production in the Amundsen Gulf 51 3.5.2 Physical environment and primary production in autumn (other sampling locations) 51 3.5.3 Abundances and vertical distributions of chaetognath species 52 3.5.4 Length distributions and sampler efficiency 53 3.5.5 Timing of reproduction 54 3.5.6 Eukrohnia hamata length cohorts and life cycle 55 3.5.7 Parasagitta elegans length cohorts and life cycle 62 3.6 Discussion 67 3.6.1 Chaetognath cohort interpretation and lifespans 67 3.6.2 Resource partitioning in the sympatric Eukrohnia hamata and Parasagitta elegans 67 3.6.3 The potential role of lipid reserves: contrasting growth in the two species 69 3.6.4 Maturation 71 3.6.5 Capital versus income breeding in a warming Arctic 72 3.7 Conclusions 73

4. Chapter 4 – Feeding strategies of arctic chaetognaths: are they really “tigers of the plankton”? 74

vii

4.1 Résumé 74 4.2 Abstract 75 4.3 Introduction 76 4.4 Method 77 4.4.1 Study areas 77 4.4.2 Sampling in the Amundsen Gulf 79 4.4.3 Sampling in the Chukchi Sea and Baffin Bay 80 4.4.4 Abundance of zooplankton 80 4.4.5 Gut contents 81 4.4.6 Fatty acids 82 4.4.7 Carbon and nitrogen 82 4.5 Results 84 4.5.1 Amundsen Gulf 84 4.5.1.1 Phenology of algae blooms 84 4.5.1.2 Zooplankton community 84 4.5.1.3 Visible prey items and predation rates 86 4.5.1.4 Lipid droplets and detritus 87 4.5.1.5 Scanning Electron Microscope observations 89 4.5.2 Chukchi Sea and Baffin Bay 90 4.5.2.1 In-vitro feeding observations 90 4.5.2.2 Lipid amounts 90 4.5.2.3 Fatty acid profiles: species differences 91 4.5.2.4 Fatty acid profiles: station and regional differences in Eukrohnia hamata and Parasagitta elegans 91 4.5.2.5 Carbon and nitrogen isotopes 94 4.6 Discussion 96 4.6.1 Prey items 96 4.6.2 Predation rates 97 4.6.3 Omnivory in arctic chaetognaths 98 4.6.4 Fatty acids and stable isotopes confirm gut contents 99 4.6.4.1 18:1 (n-9)/(n-7) ratios and Calanus copepod FATMs 99 4.6.4.2 Algal FATMs 100 4.6.4.3 δ13C values 100 4.6.4.4 δ15N values and inferred TLs 101 4.6.4.5 C/N ratios 101 4.6.5 Morphological explanations 102 4.7 Concluding remarks and future studies 102

5. Chapter 5 – General Conclusions 103

5.1 Resource use by arctic chaetognaths 103 5.2 Winter ecology in Svalbard 103 5.3 Life histories, habitats and spawning times 104 5.4 Food resources 105 5.5 Limitations of the study 107 5.5.1 Sampling limitations 107

viii

5.5.2 Stable isotopes and trophic levels 107 5.5.3 Length cohorts 108 5.6 Future research avenues 108 5.7 A note on climate change 109

Bibliography 110

Appendix A. Information on the zooplankton samples included in Chapter 2 126

Appendix B. Vertical profiles of temperature and salinity in Chapter 2 128

Appendix C. Stations sampled in Chapter 3 129

Appendix D. Vertical profiles of temperature, salinity and chlorophyll a in Chapter 3 131

Appendix E-1. Amundsen Gulf stations sampled in Chapter 4 132

Appendix E-2. Chukchi Sea and Baffin Bay stations sampled in Chapter 4 136

List of tables

Table 2.1 Total water-column abundances (ind. m-3) of a polar night zooplankton community (Rijpfjorden 2012), ordered according to abundance per taxonomic group. Net sampling (Multi-Plankton Sampler; 0.25-m2 opening, 0.2-mm mesh) was used. As larger chaetognaths may have avoided the smaller MPS net (see Grigor et al. 2014), chaetognath abundances presented here are likely to be underestimates. Mean abundances for species and copepod stages were first calculated over two hauls (one at midday and the other at midnight at various depth intervals, see Appendix A), and data were summed across all sampling depths. Copepod stages are CI-CV, AM (adult male) and AF (adult female). Functional (feeding) groups were extracted from Søreide et al. (2003). “Small Calanoida” comprised the following taxa: Acartia longiremis, Aetideidae CI-CIII, Bradyidius similis, Microcalanus spp. and Pseudocalanus spp. Only taxa with abundances of ≥0.1 ind. m-3 are shown. 31

Table 2.2 Stable carbon and nitrogen isotope values for Parasagitta elegans sampled by the MIK (3.14-m2 opening, 1.5-mm mesh) at various trawl depths (20, 30, 35, 60, 75 and 225 m) in Isfjorden and Rijpfjorden (2012). The average δ13C and δ15N composition (‰) in replicate samples (usually three but six from 225 m in Rijpfjorden) containing 25 pooled individuals (10-50 mm), average proportions of animal dry weight (DW) comprising carbon and nitrogen (%), and C/N weight ratios. All values are accompanied by standard deviations. Trophic levels (TL) were calculated for P. elegans from each fjord from mean δ15N (‰) values (see ‘Method’). 34

Table 2.3 Average fatty acid profile for Parasagitta elegans in 2012. Results are given as average percentages of the various fatty acids identified across all samples from Isfjorden and Rijpfjorden (see ‘Method’), alongside the standard deviations. Only fatty acids with mean percentages of =>0.5 ± 0.1 across both fjords are shown. In Isfjorden, the mean percentages of 15:0 FA and 18.2(n-6) FA differed between sampling depths (indicated by a † symbol, one-way ANOVA, P ˂ 0.05). In Rijpfjorden, the proportion of every fatty acid was similar between sampling depths (one-way ANOVA, P ˃ 0.05). 35

Table 3.1 Timing of peak development of maturity features in autumn and spring broods of Eukrohnia hamata and summer brood of Parasagitta elegans, and of the oil vacuole in E. hamata. Ages in months at peak development are given in parentheses. Monthly collections consisted of up to 397 and 434 E. hamata individuals from its autumn and spring broods respectively, and 177 P. elegans. 59

Table 4.1 Composition of the mesozooplankton (30 most abundant taxa) sampled in the Amundsen Gulf from November 2007 to July 2008, based on data from 70 Multinet hauls. *Taxa could not be identified to species level. 85

Table 4.3 Mean sample lipid, fatty acid amounts and fatty acid proportions (± 1 SD) in (a) Eukrohnia hamata and (b) Parasagitta elegans at different stations (stns) in autumn 2014. Sampling locations are abbreviated: CS, Chukchi Sea; SIF, Scott Inlet Fjord; GF, Gibbs Fjord; SBB, southern Baffin Bay. Also: number (n) of samples analysed and chaetognaths per sample, and chaetognath body lengths in mm. † Significant species differences (P ≤ 0.05). 92

Table 4.4 Mean sample carbon (C) and nitrogen (N) masses, C/N ratios, δ13C and δ15N values and derived trophic levels (TLs) (± 1 SD) in Eukrohnia hamata and Parasagitta elegans at different stations in autumn 2014. Sampling locations are abbreviated: CS, Chukchi Sea; SIF, Scott Inlet Fjord; GF, Gibbs Fjord; SBB, southern Baffin Bay. Also: number (n) of samples analysed and chaetognaths per sample, as well as body lengths of chaetognaths in mm. 95

List of figures

Figure 1.1 Bathymetric map of the central Arctic Ocean and several marginal seas (http://recherchespolaires.inist.fr/?L-ocean-Arctique-physiographie). 1

Figure 1.2 Water masses in the Arctic Ocean. Reproduced from Stein & Macdonald (2004).2

Figure 1.3 The planktonic food web showing protozoan and metazoan consumers at arrow caps. Adapted from Hopcroft et al. (2008). 4

Figure 1.4 Photographs of living holo-zooplankton captured in-situ by a zooplankton imager (Lightframe On-sight Key Species Investigation System; Schmid et al. 2016), during deployments in the Canadian Arctic (summer 2014). 6

Figure 1.5 Typical physiology of a chaetognath belonging to the family Sagittidae. Reproduced from Margulis & Chapman (2010). 8

Figure 1.6 Photographs of the heads and the bodies of the major arctic chaetognaths. (a) Eukrohnia hamata with characteristic marsupial sac and oil vacuole, (b) and (c) Parasagitta elegans and Pseudosagitta maxima without marsupial sacs and oil vacuoles. All images except P. maxima head from: http://www.arcodiv.org/watercolumn/Chaetognaths.html (courtesy of Russ Hopcroft). 9

Figure 1.7 Photograph of a chaetognath’s head taken by Scanning Electron Microscopy (left) with parts labelled (right). Reproduced from Bieri & Thuesen (1990). 12

Figure 1.8 Diagram illustrating the wide variety of sampling activities conducted during the Circumpolar Flaw Lead System Study (2007-2008) from the icebreaker CCGS Amundsen and at an ice camp, in order to study Arctic systems. Reproduced from Barber et al. (2010). 18

Figure 2.1 Map showing the locations of the stations sampled for chaetognaths in January 2012 and 2013. 24

Figure 2.2 Proportions (%) of Parasagitta elegans individuals per haul with gut contents (FCRmax), identifiable prey (FCRmin), and empty guts in each fjord. The horizontal line inside each boxplot shows the median of the proportions over multiple haul samples in a fjord. The lower and upper boxes show the lower and upper quartiles, respectively, and the vertical lines outside the boxes the differences between these quartiles and the lowest and highest proportions observed. Each dot represents an outlying data point. nhauls = numbers of hauls for each fjord. Hauls with ˂3 individuals analysed were not included. As only one haul was analysed for Rijpfjorden in 2013, full boxplots could not be shown. See Appendix A for numbers of individuals analysed per haul. 32

xii

Figure 2.3 Proportions (%) of feeding Parasagitta elegans individuals per haul in each fjord with different types of gut content. For details on the features of the boxplots and the data, see Figure 2.2. 33

Figure 2.4 Gut contents in ascending head width size classes: proportions of Parasagitta elegans (%) per haul with gut contents and of feeders with each gut content type. Includes all dissected specimens from Isfjorden (50 individuals, 4 hauls) and Rijpfjorden (152 individuals, 13 hauls) in 2012. For details on the features of the boxplots and the data, see Figure 2.2. ninds. = total numbers of individuals for each size class. 33

Figure 3.1 Bathymetric maps of the Canadian Arctic Ocean indicating the regions, and positions of stations (black circles) where chaetognaths were sampled from August 2007 to September 2008. Station IDs provided. 47

Figure 3.2 Illustrations and photographs of Eukrohnia hamata and Parasagitta elegans. (a) Diagrams of the two species indicating maturity features and the centrally-positioned oil vacuole in E. hamata. (b) Photographs of live E. hamata (top) and P. elegans (bottom) taken in-situ by a zooplankton imager (Schmid et al. 2016) in the Canadian Arctic. Specimens ~20 mm. (c) Photograph of oocytes in a stained P. elegans individual. (d) Photograph of ovaries in an E. hamata individual (imaged in-situ). (e) Photograph of tail sperm, seminal vesicles and seminal receptacles in a stained P. elegans individual. 49

Figure 3.3 Relative frequency of Eukrohnia hamata, Parasagitta elegans and Pseudosagitta maxima in relation to bottom depth at Amundsen Gulf stations. Multinet and square- conical net collections included. b) Weighted mean depths of E. hamata and P. elegans normalized relative to the bottom depth at Amundsen Gulf stations (see ‘Method’). Multinet collections. Standard deviation bars also given. The blue area indicates the Pacific Halocline (60-200 m) and the red area indicates the Atlantic Layer ˃ 200 m (Geoffroy et al. 2011). 53

Figure 3.4 Length frequency (mean % ± 1 SD) distributions of Eukrohnia hamata and Parasagitta elegans in the square-conical (S-C) net (1 m2 aperture, 200 µm mesh) and Multinet (0.5 m2 aperture, 200 µm mesh) in the Amundsen Gulf. Note the different scales for the two species. k is the number of collections. 54

Figure 3.5 Abundance (mean numbers m-2 + 1 SD) of newborn Eukrohnia hamata and Parasagitta elegans (body lengths of 2-4 mm) in monthly Multinet deployments from October to September. August 2007 was inserted between July and September 2008 to provide a complete composite of the annual cycle. Number of collections shown above bars. The solid horizontal bar above month labels indicates sampling in the Amundsen Gulf. 55

Figure 3.6 Monthly length frequency distributions of Eukrohnia hamata. Frequencies of newborns highlighted in orange. Visually identified length cohorts shown as normal distributions (in red) with red dots indicating the mean length. Each of the five cohorts is labelled with a capital E and a number from oldest (1) to youngest (5). Blue line is the

xiii

total distribution obtained by summing the modelled distributions. Chi-square values for the goodness-of-fit of the total distribution to the data are given. Sampling regions are abbreviated in each panel: PC, Parry Channel; AG, Amundsen Gulf; BB, Baffin Bay. k is the number of collections and n the number of length measurements included (see ‘Method’). 56

Figure 3.7 a) Monthly mean length (± 1 SD) of the five cohorts of Eukrohnia hamata starting from a major birth month; b) Composite growth trajectories of the autumn brood (born in October) and spring brood (born in April) assuming a 2-y lifespan; c) Growth-age curves of the autumn and spring broods. Circles show mean values of each characteristic (± 1 SD shown as ribbons). 58

Figure 3.8 Development of sexual features (a, oocyte number; b, ovary length; d, sperm load; e, seminal vesicle width; f, seminal receptacle diameter) and oil vacuole area (c) with length for the autumn and spring broods of Eukrohnia hamata. Circles show mean values of each characteristic (± 1 SD shown as ribbons). Vertical line indicates average length (15.4 mm) at one year of age. Maturity results for each brood were obtained from the analyses of up to 283 individuals from each 1 mm length class. 60

Figure 3.9 Mean frequency (± 1 SD) of Eukrohnia hamata with oil in vacuole and/or digestive tract by months. Number of water column collections shown inside bar (33 – 446 individuals per haul). 61

Figure 3.10 Abundances (ind. m-3) of Eukrohnia hamata age classes in discrete depth layers of the Amundsen Gulf, characterized by their mid-points (black squares). Top panels: autumn brood individuals. Bottom panels: spring brood individuals. Seabed shown in brown, unsampled sections of the water column in gray. See ‘Method’ for further details. 62

Figure 3.11 Monthly length frequency distributions of Parasagitta elegans. Frequencies of newborns highlighted in orange. Visually identified length cohorts shown as normal distributions (in red) with red dots indicating the mean length. Each of the three cohorts is labelled with a capital P and a number from oldest (1) to youngest (3). Blue line is the total distribution obtained by summing the modelled distributions. Chi-square values for the goodness-of-fit of the total distribution to the data are given. Sampling regions are abbreviated in each panel: NF, Nachvak Fjord; PC, Parry Channel; AG, Amundsen Gulf; BB, Baffin Bay. k is the number of collections and n the number of length measurements included (see ‘Method’). 63

Figure 3.12 a) Monthly mean length (± 1 SD) of the three cohorts of Parasagitta elegans starting from a major birth month; b) Composite growth trajectory of the single brood.64

xiv

Vertical line indicates average length (20.7 mm) at one year of age. Maturity results were obtained from the analyses of up to 63 individuals from each 1 mm length class. 65

Figure 3.14 Abundances (ind. m-3) of Parasagitta elegans age classes in discrete depth layers of the Amundsen Gulf, characterized by their mid-points (black squares). Seabed shown in brown, unsampled sections of the water column in gray. See ‘Method’ for further details. 66

Figure 4.1 Bathymetric maps of the Arctic Ocean, showing the positions and IDs of sampling stations (black circles) in (a) Amundsen Gulf, (b) north-eastern Chukchi Sea and (c) Baffin Bay (SIF = Scott Inlet Fjord; GF = Gibbs Fjord; SBB = southern Baffin Bay). Details of sampling stations are shown in Appendices E-1 and E-2. 79

Figure 4.2 Vertical distributions of chlorophyll a biomass (mg m-3) in the upper 200 m of the water column along the ship track from November 2007 to July 2008 in the Amundsen Gulf. Black dots indicate sampling depths. Details of sampling stations are shown in Appendix E-1. Chl a data were provided by Michel Gosselin (Université du Québec à Rimouski). 84

Figure 4.3 Average number of prey per chaetognath gut (npc) in square-conical (S-C) net and Multinet collections from November 2007 to August 2008 in the Amundsen Gulf. Numbers of individuals shown above data points. k is the number of collections. Inset bottom: photograph of a Parasagitta elegans specimen (12 mm) with a relatively large Pseudocalanus spp. copepod in the gut (from November 2 2007). 86

Figure 4.4 Photographs of substantial amounts of green macroaggregates (˃ 500 µm) in the guts of Eukrohnia hamata, but not in the guts of Parasagitta elegans. Specimens from the Amundsen Gulf on May 31 2008 (20-40 m depth), 20-30 mm. 87

Figure 4.5 Scanning Electron Microscope photographs of some items in Eukrohnia hamata guts (left box) and Parasagitta elegans guts (right box), in different months. Examples: hooks of a chaetognath, and mandibles of copepods (arrow caps). Specimens 20-30 mm. 89

Figure 4.6 Photographs of live Eukrohnia hamata individuals consuming green macroaggregates in-vitro in the Chukchi Sea. Specimens 20-30 mm. Bottom right: An inverted microscope photograph of phytoplankton such as Ceratium spp. in one gut. 90

Figure 4.7 Fatty acid biomarkers in Eukrohnia hamata and Parasagitta elegans at different stations in autumn 2014. Sampling locations are abbreviated: CS, Chukchi Sea; SIF, Scott Inlet Fjord; GF, Gibbs Fjord; SBB, southern Baffin Bay. From top panel to bottom panel: mean ratios of the carnivory biomarker 18:1 (n-9)/(n-7), mean proportions of the Calanus biomarkers ΣC20:1+C22:1 MUFA, and mean ratios of the algal biomarker 16:1/16:0 (± 1 SD). No bars – no data available. 94

Figure 4.8 Mean sample δ13C and δ15N values and corresponding trophic levels (± 1 SD) of Eukrohnia hamata and Parasagitta elegans. Station IDs shown next to data points. Results from Chukchi Sea stations inside red oval. Results from Baffin Bay region stations inside blue oval (locations are abbreviated: SIF, Scott Inlet Fjord; GF, Gibbs Fjord; SBB, southern Baffin Bay). 96

xvi

Dedicated to the memory of Helen McMahon, my granny, who made it possible for me to first pursue my interest in the Arctic regions. 23 August 1929 – 25 March 2015

Angelina Kraft

JP Aubé

xvii

Acknowledgements I am now nearing the end of my journey to gain a Ph. D. in Oceanography, and there are several people who have supported me whom I would like to acknowledge in this thesis. Firstly, I would like to thank my director of research, Dr. Louis Fortier for allowing me to realize this project on the enigmatic chaetognaths. I would also like to thank Louis for giving me opportunities to attend and present my research at several national and international conferences, to voyage to the Arctic/sub-Arctic on three separate research cruises, and for the many ways in which he helped me improve my experience in scientific writing. Thanks to Drs. Jean-Éric Tremblay, Guillaume Massé and Stéphane Plourde for evaluating the ‘dépôt initial’ of my thesis and thesis defense presentation, and providing useful suggestions for improvements. Thanks also to Dr. Maurice Levasseur for participating in my Ph. D. committee, and to Dr. Ladd Johnson for presiding over my defense presentation.

I express my gratitude to my colleagues Drs. Øystein Varpe (UNIS), Stig-Falk Petersen (Akvaplan-niva), Dominique Robert, Eric Rehm, Guillaume Massé and Marcel Babin (Université Laval), Roxane Barthélémy and Jean-Paul Casanova (Aix-Marseille Université), and Mrs. Ariane Beauféray (Université Laval), all of whom helped to shape the direction of this thesis. I would like to specifically acknowledge Dr. Øystein Varpe who has consistently shown a keen interest in my research, and been happy to offer constructive ideas and suggestions. Thanks to Drs. Catherine Lalande (Université Laval), and Tara Connelly (Memorial University of Newfoundland), as well as anonymous journal referees for Polar Biology and Journal of Plankton Research, for offering valuable comments on chapters.

I am grateful to the crew and scientific staff on expeditions with R/V Helmer Hanssen and CCGS Amundsen for research support. Dr. Gérald Darnis’ assiduity onboard CCGS Amundsen, and in the laboratory, was fundamental to the success of zooplankton sections of the CFL project. Mr. Carl Ballantine collected samples from Svalbard in 2013. I greatly appreciate the hard work of my student assistants; Mrs. Ariane Beauféray, Miss. Marianne Caouette, Miss. Vicky St-Onge, Mr. Joël-Fortin Mongeau and Mr. Pierre-Olivier Sauvageau, for helping me to analyse the vast number of chaetognaths included in this thesis. Sincere thanks also to the following individuals and lab groups for analysing fatty acids and stable

xviii

isotopes: Dr. Guillaume Massé, Miss Caroline Guilmette, Mr. Jonathan Gagnon, the Institute for Energy Technology and UNILAB in Norway. Thanks to Drs. Jean-Éric Tremblay and Michel Gosselin, and Miss Marjolaine Blais for contributing environmental datasets.

I would like to thank my close friends Moritz Schmid, Kai Shapiro and Sophie Regueiro for their moral support during the most challenging stages of my Ph. D. Many others, including Cyril Aubry, Kevin Gonthier, Arnaud Pourchez, Noémie Friscourt, Max Geoffroy and other members of the Fortier lab, have given me great memories from my time in Québec and Canada. I am grateful to my parents (Ali and Dave), sister Ara, as well as the rest of my family, who were ready to Skype at a moment’s notice and were always ready to offer support, advice and encouragement. Thanks to the School for Science and Math at Vanderbilt (Nashville, USA) for more recently giving me the opportunity to share my passion for the polar regions with the next generation of scientists.

Bursaries from Takuvik and Québec-Océan allowed me to keep the focus on all my doctoral activities and reach these final stages of my Ph. D. course.

xix

Foreword This doctoral thesis comprises a General Introduction (Chapter 1), three scientific articles (Chapters 2, 3 and 4), and a number of General Conclusions relating to the objectives of the thesis (Chapter 5). Chapter 2 is published in Polar Biology as part of a Special Issue on Polar Night ecology. Chapters 3 and 4 are currently in preparation for publication:

Chapter 2 Grigor J.J., Marais A., Falk-Petersen S., Varpe Ø. (2015) Polar night ecology of a pelagic predator, the chaetognath Parasagitta elegans. Polar Biology 38:87-98 (reproduced with permission of the publisher). Minor formatting changes have been made to the material published in Polar Biology, to ensure compatibility with the other chapters in this thesis (e.g. species names are written out in full the first time they are used in every new paragraph, and thereafter abbreviated to their shorter form. Citations follow the same format as in the other chapters).

Chapter 3 Grigor J.J., Schmid M.S., Fortier L. Growth and reproduction of the chaetognaths Eukrohnia hamata and Parasagitta elegans in the Canadian Arctic Ocean: capital breeding versus income breeding. This has been revised and recently re-submitted to Journal of Plankton Research.

Chapter 4 Grigor J.J., Barthélémy R.-M., Massé G., Casanova J.-P., Fortier L. Feeding strategies of arctic chaetognaths: are they really “tigers of the plankton”? This will be submitted to Journal of Plankton Research.

The aims and objectives of this thesis were designed by myself. All the analyses were performed by myself (or by students under my supervision). To collect material for the thesis, I participated in research cruises in winter 2012, autumn 2013 and autumn 2014. Chapters 2- 4 have benefited from the corrections and comments of my co-authors. Results of these three

chapters have been presented at the following national and international scientific conferences:

[9] Grigor J.J., Marais A.E., Schmid M.S., Varpe Ø., Fortier L. (2015) Ecology of arrow worms in the Arctic: are they really “tigers of the zooplankton”?? APECS Online Conference – New Perspectives in the Polar Sciences

[8] Grigor J.J., Varpe Ø., Marais A.E., Schmid M.S., Rehm E., Fortier L. (2014) Ecology of arrow worms in the Arctic: are they really “tigers of the zooplankton”?? Arctic Change Conference (Ottawa, Canada)

[7] Grigor J.J., Marais A.E., Schmid M.S., Rehm E., Fortier L. (2014) Arrow worm ecology in the Canadian Arctic. Arctic Change Conference (Ottawa, Canada)

[6] Grigor J.J., Marais A.E., Schmid M.S., Rehm E., Fortier L. (2014) Arrow worm ecology in the Canadian Arctic. Assemblée générale annuelle de Québec-Océan (Rivière- du-Loup, Canada)

[5] Grigor J.J., Marais A.E., Schmid M.S., Fortier L. (2013) Seasonal ecologies of chaetognaths (gelatinous zooplankton) in the Canadian Arctic. ArcticNet Annual Science Meeting (Halifax, Canada) *Awarded best poster in its category: Marine Natural Science

[4] Grigor J.J., Søreide J.E., Varpe Ø., Fortier L. (2013) Seasonal ecology and life history strategies of the chaetognath Parasagitta elegans in the Canadian and European Arctic: The story so far. Assemblée générale annuelle de Québec-Océan (Rivière-du-Loup, Canada)

[3] Grigor J.J., Søreide J.E., Varpe Ø. (2013) The annual routine of a predatory arctic chaetognath in a highly seasonal environment. Arctic Frontiers: Geopolitics and marine production in a changing Arctic (Tromsø, Norway)

xxi

[2] Grigor J.J., Søreide J.E., Varpe Ø. (2012) The annual routine of a predatory zooplankter in a highly-seasonal Arctic environment. Annual ArcticNet Science Meeting (Vancouver, Canada)

[1] Grigor J.J., Søreide J.E., Varpe Ø. (2012) The annual routine of a predatory zooplankter in a highly-seasonal Arctic environment. Québec Océan 10-yr conference: A reality check on oceans’ health (Montreal, Canada)

During my Ph. D., I participated as a co-author in the writing of one scientific article that used an underwater camera system, the Lightframe On-sight Keyspecies Investigation system (LOKI), to capture excellent images of zooplankton and document their fine-scale vertical distributions. The study also developed a model to automatically identify the species and stages of imaged animals:

[1] Schmid M.S., Aubry C., Grigor J.J., Fortier L. (2016) The LOKI underwater imaging system and an automatic identification model for the detection of zooplankton taxa in the Arctic Ocean. Methods in Oceanography 15:129-160

xxii

1. Chapter 1 – General introduction 1.1 The Arctic Ocean The Arctic area is characterized by strong seasonal cycles in solar radiation and surface sea- ice cover. The sun remains more than 12° below the horizon at the height of winter (the ‘polar night’) and does not set at the height of summer (the ‘midnight sun’). The Arctic Ocean covers an area of 14 million km2 (Huntington & Weller 2005) and is divided by a submarine ridge of continental crust (the Lomonosov Ridge) into two major basins, the Eurasian and Amerasian Basins (Figure 1.1). Vast continental shelves comprise 50 % of the Arctic Ocean area (Figure 1.1), and these are amongst its most biologically productive areas (Sakshaug 2003, see ‘Marine ecosystems’ section).

Figure 1.1 Bathymetric map of the central Arctic Ocean and several marginal seas (http://recherchespolaires.inist.fr/?L-ocean-Arctique-physiographie).

1.1.1 Physical environment Through dynamic and thermodynamic processes (movement, growth and melt), the extent of surface sea-ice increases to its maximum in March and declines thereafter towards its minimum in September (Stroeve et al. 2008). Sea-ice acts as a barrier between the underlying ocean and the atmosphere affecting heat exchange and the transmission of light that algae in the water require for primary production. Sea-ice is also a habitat for several specialized fish, seabird and mammal species, and influences global climate. Ice formation in key locations is important for the formation of cold and highly saline water. This briny water sinks, initiating deep-water currents that move into other oceans as an integral part of the ‘thermohaline circulation’ (Loeng et al. 2005). Surface water/ice follows two important surface currents: in the western Arctic the clockwise-turning Beaufort Gyre; and in the eastern Arctic the Transpolar Drift, which exports ice into the Atlantic (Stein & Macdonald 2004). Freshwater contributions from ice melt, precipitation and large non-freezing rivers (e.g. Ob and Yenisei in Eurasia, the McKenzie and Yukon in N. America), increase the buoyancy of surface waters (i.e. the Polar Mixed Layer). Great volumes of warm, relatively saline Atlantic water, also enter the Eurasian Basin through the deep (2600 m) Fram Strait (Figure 1.2) and the Barents Sea. Lower volumes of cold, relatively fresh Pacific water, are supplied to the Amerasian Basin through the 45 m-deep Bering Strait (Stein & Macdonald 2004, Figure 1.2).

Figure 1.2 Water masses in the Arctic Ocean. Reproduced from Stein & Macdonald (2004).

In the Amerasian Basin, the Atlantic Water cools and becomes denser, plunging below the Pacific water to depths ˃200 m, where it resides for ~30 years (Stein & Macdonald 2004, Figure 1.2). The inflows of Pacific water in the Amerasian Basin and of Atlantic water in the Eurasian basin create salinity gradients (haloclines) that protect sea-ice from warm waters below (Winsor & Bjork 2000). Arctic deep water (cold and relatively saline), can occur below 900 m in the deep areas of the basins, where it has a residence time of 75-300 years (Stein & Macdonald 2004, Figure 1.2).

1.1.2 Arctic marine ecosystems Arctic marine ecosystems are considered less complex than most temperate and tropical systems, with lower productivity and biodiversity in general (Loeng et al. 2005). In Arctic marine ecosystems, extreme cycles in solar illumination and sea-ice cover lead to strong seasonal cycles in primary production by autotrophic algae, with a considerable reduction in winter (Ji et al. 2013). However, biological hotspots occur where a good nutrient supply to surface waters supports high primary productivity. For instance, large blooms of phytoplankton in some parts of the Chukchi Sea near Alaska support sizable communities of invertebrates, both in the water column and on the seabed (Hopcroft et al. 2004, Grebmeier et al. 2006, Nishino et al. 2016). Furthermore, polynyas (permanently open water areas otherwise surrounded by sea-ice) are viewed as biological ‘oases’, with phytoplankton blooms occurring in polynyas earlier in the season than in any surrounding areas (Ingram et al. 2002, Hannah et al. 2009). This provides less-seasonally restricted access to food for grazers and in turn their predators. Polynyas can be formed by the upwelling of warm seawater which prevents ice from forming (i.e. sensible heat polynya), and/or by the influence of wind in driving ice away (i.e. latent heat polynya; Hannah et al. 2009). The large North Water Polynya in northern Baffin Bay (Canadian Arctic) is mainly a latent heat polynya in winter and spring, however, sensible heat is also important for its growth in late spring (Ingram et al. 2002).

In ice-covered seas, a bloom of under-ice algae is initiated during spring which is then succeeded by a bloom of pelagic phytoplankton when the ice retreats in the spring/summer.

Some locations can have another distinct phytoplankton bloom in autumn (Ardyna et al. 2014). Increases in the occurrence of autumn blooms across the Arctic may be a symptom of global warming (see ‘Climate change and other challenges for Arctic marine life’ section). On average, phytoplankton production is higher (12 to 50 g C m-2 yr-1) than ice algae production (5 to 10 g C m-2 yr-1). However, ice algae production exceeds that of phytoplankton in locations with thick multi-year ice cover (Legendre et al. 1992, Leu et al. 2011). Bloom phenology varies with latitude and timing of open water (Zenkevich 1963, Falk-Petersen et al. 2009, Tremblay et al. 2012), and the magnitude of a phytoplankton bloom during the ice-free period is determined by nitrogen availability (Tremblay & Gagnon 2009), and also influenced by zooplankton grazing (e.g. Banse 2013).

Arctic marine ecosystems contain at least 5000 species of invertebrate animal taxa from 24 phyla (Hoberg et al. 2013 and references therein), 91 % of which are associated with the seabed (i.e. ‘benthic’), 8 % with the water column (i.e. ‘pelagic’) and 1 % with the sea-ice (i.e. ‘sympagic’). Copepods belonging to the genus Calanus (Figure 1.3) are known as ‘key species’ in the Arctic seas. They convert carbohydrates and proteins taken up from the algae into high-energy lipids (wax esters) that comprise the main energy source within the food web (e.g. Falk-Petersen et al. 2009).

Figure 1.3 The planktonic food web showing protozoan and metazoan consumers at arrow caps. Adapted from Hopcroft et al. (2008).

Amongst the vertebrates, up to 250 fish species reside in the Arctic area defined by the Conservation of Arctic Flora and Fauna (an Arctic Council Working Group), for part or all of the year (Hoberg et al. 2013). The polar cod Boreogadus saida is an endemic species which can channel up to ~75 % of energy flow from zooplankton to vertebrate predators (Welch et al. 1992). In addition, 35 species of mammals and 60 species of seabirds are observed in Arctic waters, with the Baffin Bay little auk population considered to be the largest seabird population in the world (Hoberg et al. 2013). Heterotrophic microbes and protists (Figure 1.3) remineralize dissolved organic carbon (˂0.45 µm) excreted from these consumers into inorganic forms of carbon which can then be taken up again by protozoans (Hopcroft et al. 2008).

1.2 Zooplankton and their polar adaptations Zooplankton are aquatic animals with body sizes ˂0.002 mm to ˃200 mm which have limited ability to control their horizontal positions against water currents. As zooplankton have little capacity to evade warming waters and are typically not harvested by humans, variations in their distribution or phenology may strongly herald climate change effects (Richardson 2008). Zooplankters include herbivores, omnivores and carnivores, and are food for a wide diversity of invertebrate and vertebrate predators. Zooplankton fecal pellets can be an important source of carbon sequestration (e.g. Wassmann et al. 1999). However, fecal pellets excreted in epi-pelagic waters are typically removed rapidly in surface waters. This is due to degradation by bacteria, protozooplankton and copepods (Morata & Seuthe 2004, Turner 2015). Copepods, chaetognaths and other zooplankters are classified as ‘holoplankton’ taxa (that drift in the sea throughout their entire life cycle). In contrast, a variety of fish, decapods and other species are plankton only when they are larvae (‘meroplankton’), before then continuing life as nekton or benthos (Stübner et al. 2016). Some zooplankton species are capable of migrating considerable distances through the water column over day-night, seasonal or ontogenetic cycles. These behaviors may be associated with finding food, avoiding predation, or gaining an energy bonus from resting in colder waters, and they alter the spatial distributions of food, materials and energy (Longhurst et al. 1984, Haney 1988).

Arctic zooplankton biomass is typically dominated by Calanus copepods, followed by non- copepod taxa such as amphipods and chaetognaths (Figure 1.4). Zooplankton abundance is dominated by small cyclopoid and calanoid copepods including Oithona similis and Pseudocalanus spp., amongst others (e.g. Søreide et al. 2003, Hopcroft et al. 2004, Darnis & Fortier 2014). For stable co-existence, similar species must occupy slightly different ecological niches, efficiently partition available food and share habitat resources (Gause 1934, Ross 1986).

Due to reduced growth rates in colder temperatures, zooplankton and other polar invertebrates often have longer lives and they reach maturity at older ages than those at lower latitudes (Hoberg et al. 2013). High-latitude zooplankton display a range of adaptations to a seasonal food source associated with, for instance, the timing and extent of growth and

reproduction, diapause, lipid content and feeding strategy (e.g. Conover 1988, Falk-Petersen et al. 2009, Varpe 2012, Grigor et al. 2014).

For grazers, such as Calanus glacialis, timing their growth and reproduction to coincide with the ice algae bloom can be of the utmost importance, as it would enable offspring to hatch ~3 weeks later during the phytoplankton bloom (Søreide et al. 2010, Leu et al. 2011, Daase et al. 2013). The high-quality food provided during this latter bloom allows for critical early growth of copepod offspring, without risk of starvation. Such a ‘match’ between the development of primary production and copepod offspring benefits copepod populations and the food web in general. However, in cases where offspring hatch before or after the peak of the phytoplankton bloom, they miss the phytoplankton food needed for healthy growth (Søreide et al. 2010, Leu et al. 2011). Such a ‘mismatch’ scenario has been shown to drastically reduce Calanus biomass (Leu et al. 2011).

Storage lipids are important for Arctic organisms, in which they may be used to fuel one or more of the following processes; basal metabolism, growth, maturation and reproduction (Kattner et al. 2007). Wax esters (comprising one fatty acid and one fatty alcohol) are the main long-term energy deposit in zooplankton, whereas triacylglycerols (comprising three hydrocarbon chains on a glycerol backbone) provide energy for activities in the short term (Lee et al. 2006). In winter, Calanus copepods and other grazers survive by entering diapause in relatively deep waters during which time accumulated wax ester stores are metabolized for energy. Capital breeders (e.g. C. hyperboreus) use their wax esters to fuel reproduction, and can produce offspring earlier in the season giving them a longer time to develop their energy reserves, required for surviving the next winter. Reproductive fitness is also improved (Varpe et al. 2007). Income breeders need recent food to reproduce, and mixed capital- income breeders use a combination of the two strategies (e.g. Sainmont et al. 2014a). Carnivores are assumed to have a low capacity for long-term energy storage (Hagen 1999), however, this may not be true for some species. Carnivorous and omnivorous zooplankters, like amphipods (Figure 1.4) are also assumed to feed opportunistically year-round (Hagen 1999).

1.3 Chaetognaths Chaetognaths (Figure 1.5) are a phylum of semi-gelatinous (dry weight ˃5 % wet weight; Larson 1986) marine mesozooplankton. They are represented worldwide by at least 127 pelagic and benthic species from 23 genera with body lengths 2-120 mm (e.g. Bone et al. 1991, Foster 2011). The name ‘chaetognath’ (Chaeto = bristle, gnath = jaw) relates to the presence of hooks on the head (Figure 1.5). Chaetognaths are also known as ‘arrow worms’ due to their lean appearance and rapid motion. These coelomates have existed from the Middle Cambrian and possibly before then (Walcott 1911). The global biomass of chaetognaths has been estimated at 10-30 % of that of copepods (Bone et al. 1991). The majority of species are residents of the upper 200 m of the water column (i.e. epi-pelagic; Pierrot-Bults & Nair 1991).

Figure 1.5 Typical physiology of a chaetognath belonging to the family Sagittidae. Reproduced from Margulis & Chapman (2010).

Chaetognaths occur in the diets of zooplankton such as amphipods (Gibbons et al. 1992), jellyfish (e.g. Zavolokin et al. 2008) and other chaetognaths (e.g. Pearre 1982), and recently have received attention as the preferred prey for the larvae of tropical fish and commercially

important decapods (e.g. Sampey et al. 2007, Saunders et al. 2012). In the Arctic, fish such as the polar cod (Rand et al. 2013), seabirds such as the little auk (Mehlum & Gabrielsen 1993) and baleen whales (Pomerleau et al. 2012), feed on chaetognaths. Suthers et al. (2009) described chaetognaths as “the tigers of the plankton”. Animals in the phylum are generally thought to be opportunistic, but strict carnivores (Marazzo et al. 1997). For further details, see ‘Feeding strategies and trophic importance’ section.

1.3.1 Arctic species and distributions In contrast to the high chaetognath diversity observed in other seas (e.g. De Souza et al. 2014), only three major species (Eukrohnia hamata, Parasagitta elegans and Pseudosagitta maxima) are frequently reported in Arctic plankton surveys (e.g. Kramp 1939, Buchanan & Sekerak 1982, Sameoto 1987, Figure 1.6). Another two species have been detected in Arctic bathy-pelagic surveys; Heterokrohnia involucrum (Dawson 1968) and H. mirabilis (Kapp 1991). Kosobokova & Hopcroft (2009) reported that chaetognaths represented, on average, about 13 % of the zooplankton biomass in the Canadian Arctic in summer 2005.

Parasagitta elegans (previously Sagitta) is restricted to the North Atlantic, North Pacific and Arctic Oceans, with its lower distribution limit at 41°N and a northern limit short of the central Arctic basin (e.g. Bogorov 1946, Bieri 1959, Grainger 1959). It is generally an epi- pelagic, neritic species throughout most of its range (e.g. Terazaki 2004). Terazaki (2004) described P. elegans as the best-studied chaetognath species. In contrast to P. elegans, Eukrohnia hamata and Pseudosagitta maxima are cosmopolitan species occurring in all oceans, with predominantly meso- or bathy-pelagic affinities (e.g. Bigelow 1926, Bieri 1959, Cheney 1976, Thuesen et al. 1993). All three species occasionally co-occur at epi-pelagic depths in the Arctic (e.g. Alvarino 1964, Hopcroft et al. 2005, Bieri 1959). The life cycle and feeding strategy of E. hamata is more thoroughly studied in the Southern Ocean (e.g. Øresland 1990, Øresland 1995, Froneman & Pakhomov 1998, Froneman et al. 1998, Kruse 2009, Kruse et al. 2010), and few studies exist on the ecology of P. maxima in the Arctic (see Sameoto 1987).

Some chaetognaths may migrate long distances vertically through the water column. Grigor et al. (2014) reported seasonal vertical migrations (SVM) in a fjord population of Parasagitta elegans in the European Arctic. In this Svalbard fjord, the youngest age class resided near the surface during the summer phytoplankton bloom, whereas all three age classes mostly remained below 60 m from September to April (seasonal migrations closely mirrored those of Calanus prey). Daase et al. (2016) also reported diel (day-night cycle) vertical migrations (DVM) in a population of Eukrohnia hamata in another Svalbard fjord, towards the end of the midnight sun period. Migrations involved upward movements of E. hamata to surface waters, apparently to feed at night, and downward movements to deeper waters, apparently to avoid predators in well-lit waters during the day.

1.3.2 Lifespans The life history characteristics of chaetognath species (e.g. life span, size at maturity, time and number of spawning cycles per year) differ throughout their geographical ranges in response to variations in temperature, food quantity and quality (Pearre 1991, Alvarino 1990, Terazaki 2004). In temperate seas, both Eukrohnia hamata and Parasagitta elegans may

have lifespans of less than a year (e.g. Russell 1932, Terazaki & Miller 1986, Sameoto 1971, Zo 1973, King 1979, Terazaki & Miller 1986). In the Arctic, two-year lifespans have been suggested for E. hamata (Sands 1980, Sameoto 1987, Pearre 1991); one to three-year lifespans have been suggested for P. elegans (e.g. Dunbar 1940, McLaren 1961, Timofeev 1995, Welch et al. 1996, Grigor et al. 2014); and the larger Pseudosagitta maxima may live for four to five years (Sameoto 1987).

1.3.3 Reproductive strategy Chaetognaths are hermaphrodites, able to reproduce by both cross and self-fertilization. Male gonads in the tail region (testes and seminal vesicles, Figure 1.5) produce and transmit sperm, whilst female gonads in the trunk region (ovaries and seminal receptacles, Figure 1.5) produce ova and receive sperm from a neighbor (Alvarino 1992). Although cross-fertilization is more common due to sperm maturing earlier than oocytes, this strategy makes self- breeding possible when access to mates is limited (Alvarino 1992). The eggs of chaetognaths hatch a few days after oocytes are fertilized (e.g. Kotori 1975). Hatching takes place inside a marsupial sac in Eukrohniidae (Figure 1.6a) and in open water in Sagittidae (Alvarino 1968, Kotori 1975, Hagen 1985). Kuhl (1938) suggested that chaetognaths die immediately after spawning. However, several other authors have disputed this finding, instead showing iteroparity and continued spawning after laying their first batch of eggs (Conway & Williams 1986, Grigor et al. 2014). Whilst Eukrohnia hamata is reported to breed year-round in the Southern Ocean (Øresland 1995), breeding of Parasagitta elegans in the Arctic mainly coincides with peak abundance of zooplankton food (e.g. Kramp 1939, Dunbar 1940, Grainger 1959, McLaren 1961, Dunbar 1962, Grigor et al. 2014).

1.3.4 Oil vacuoles in Eukrohnia spp. The middle section of the gut in Eukrohnia spp. contains an elliptical oil vacuole (Figure 1.6a), also described in earlier reports as an ‘oil droplet’ (Øresland 1990, Froneman & Pakhomov 1998, Froneman et al. 1998, Kruse et al. 2010, Giesecke & Gonzalez 2012, Pond 2012). Such an oil vacuole is not present in Sagittidae species (e.g. Figure 1.6b and Figure 1.6c). Oil vacuoles in Eukrohnia species may store wax esters. Wax esters seem to occur in

much higher amounts in Eukrohnia hamata compared to Parasagitta elegans and Pseudosagitta maxima (e.g. Lee & Hirota 1973, Falk-Petersen et al. 1987, Donnelly et al. 1994, Lee et al. 2006, Connelly et al. 2012, Connelly et al. 2016). Large wax ester supplies in Eukrohnia chaetognaths could suggest a potential for metabolic reduction/capital breeding, as occurs in overwintering zooplankton, though the centralized position of the oil vacuole in the body also hints at a buoyancy control function (Pond 2012).

1.3.5 Feeding strategies and trophic importance Zooplankton prey includes a wide range of ichthyoplankton and zooplankton, particularly copepods (e.g. Terazaki 2004). Prey is sensed mechanically; chaetognath hairs detect vibrations at distances of 1-3 mm (Feigenbaum & Reeve 1977, Goto & Yoshida 1981). Specific prey types can be differentiated by their unique movements (Tonnesson & Tiselius 2005). The eyes of chaetognaths lack lenses capable of forming images but they do contain a pigment that detects light direction (Pearre 1973). Therefore, their eyes may help them to detect prey outside the maximum distance for hair detections (Goto & Yoshida 1981). Dorso- ventral undulating movements are used to dart at prey, which is then seized by the hooks and punctured by rows of small teeth (Figure 1.7). The locomotion appendages, as well as the feeding appendages, differ between chaetognath species, for instance Eukrohnia hamata lacks anterior teeth (Furnestin 1965), present in P. elegans (Terazaki 1993). Relatively large items are consumed by expansions of the head (Bieri & Thuesen 1990). Chaetognaths may paralyze prey by secreting a bacterial-produced neural poison (tetrodotoxin) from vestibular ridges (Thuesen & Kogure 1989, Bieri & Thuesen 1990, Figure 1.7).

Figure 1.7 Photograph of a chaetognath’s head taken by Scanning Electron Microscopy (left) with parts labelled (right). Reproduced from Bieri & Thuesen (1990).

In the waters of Nova Scotia (Canada), Sameoto (1971, 1972) suggested that the abundant Parasagitta elegans could remove considerable fractions of the copepod standing stock per day (36 % to ˃100 % depending on season). Welch et al. (1992) suggested that Eukrohnia hamata and Pseudosagitta maxima may ingest up to 51 % of the copepod biomass in Lancaster Sound annually, but noted that their conclusions for these chaetognaths were “little better than guesses” (Welch et al. 1992). The above estimates of predation rates or proportion of prey standing stock removed, were based on observations of gut contents in dead chaetognaths, and specifically, methodology which treated visible lipid droplets in their bodies as evidence of recent prey digestion (e.g. Sameoto 1971, Sameoto 1972, Sameoto 1987). This approach was also adopted for E. hamata in the Southern Ocean (Froneman & Pakhomov 1998). Including lipids in predation rate calculations may be suitable for species without considerable capacity for lipid storage. For Eukrohnia species it seems there is a risk of confusing diagnostic lipid stores with oil droplets from copepods, which is likely to have occurred in at least the Sameoto (1987) study (Øresland 1990). Several other polar studies on E. hamata, as well as P. elegans and P. maxima, have reported much lower predation rates when lipid droplets were excluded from the estimates (e.g. Sameoto 1987, Øresland 1995, Froneman & Pakhomov 1998, Brodeur & Terazaki 1999, Bollen 2011, Giesecke & Gonzalez 2012).

Whilst chaetoganths are generally recognized as predators, several studies have also observed algae (e.g. Alvarino 1965, Boltovskoy 1981, Alvarez-Cadena 1993, Kruse et al. 2010) or high amounts of algae biomarkers (e.g. Philp 2007, Connelly et al. 2014) in chaetognath guts, possibly indicating that these opportunistic feeders also ingest algae in seawater or graze them from a substrate. Chaetognaths belonging to the genera Archeterokrohnia and Heterokrohnia reside in the nepheloid layer, where they are reported to feed on organic matter and bacteria from sediments (Casanova 1986). In an interesting recent study, Casanova et al. (2012) suggested facultative osmotrophy (ingestion of dissolved organic matter in ambient seawater) as the main mode of nutrition in chaetognaths.

1.3.6 Fatty acids and stable isotopes Fatty acid profiles and stable isotope signals can also reveal information about consumer diets, trophic positions, as well as the main primary producers in the food web. Chaetognaths are easily stressed and damaged in plankton sampling nets. Potential food loss from damaged guts or stress-induced regurgitation and defecation, as well as cod-end feeding, can all bias gut content analyses and predation rate estimates (see Baier & Purcell 1997 and references therein). In contrast to gut contents, fatty acid and stable isotope signals can persist for weeks or months (Dalsgaard et al. 2003, Arim & Naya 2003). Fatty acid analysis is based on the premises that (a) fatty acids are produced by specific organisms, and (b) the fatty acids undergo limited breakdown and transformation in consumers (Dalsgaard et al. 2003, Arim & Naya 2003). Nitrogen isotopes are used to assess trophic position, based on the premise that the 15N/14N ratio undergoes a constant level of change between trophic levels (Minagawa & Wada 1984, Michener & Schell 1994). Previous studies have used this technique to infer that Eukrohnia hamata may feed more omnivorously than Parasagitta elegans in the Arctic (Søreide et al. 2006, Hop et al. 2006). Carbon isotopes can indicate the carbon source in a food web since the 13C/12C ratio differs little between a consumer and its food (Hobson & Welch 1992).

1.4 Climate change and other challenges for Arctic marine life Over millions of years, Arctic marine ecosystems have been shaped by profound environmental changes that include transitions between glacial periods and interglacial periods, in addition to large variations in sea level (Darby et al. 2006). However, the 2-3 ºC rise in mean annual air temperatures since 1950 (Chapman & Walsh 2003) has had an anthropogenic source (greenhouse gas emissions, and CO2 in particular). Sea surface temperatures are increasing 2-3 times as fast as in the temperate oceans (known as ‘Arctic amplification’), as elevated influx of warm Atlantic waters causes major heat advection (Spielhagen et al. 2011). Overland et al. (2014) predicted further warming by up to 13 °C prior to the autumn of 2100 if greenhouse gas emissions continue unabated. Consistent with these changes, conditions in the Arctic Ocean are becoming more similar to those in temperate oceans (referred to as ‘Atlantification’, see Wassmann et al. 2006).

Arctic seas are acidifying twice as fast as temperate seas due to enhanced CO2 solubility in colder waters (Bates et al. 2011). This shift could directly affect zooplankters such as the pteropod mollusc Limacina helicina, a prominent component of Arctic zooplankton communities (Gannefors et al. 2005). L. helicina precipitates CaCO3 to build a shell required for its survival. However, experiments suggest that the lower pH values expected to occur by the end of the century will be associated with a drastic 28 % decrease in shell calcification rate and the death of these animals (Comeau et al. 2009).

Previous studies have reported decreases in both the annual mean extent of Arctic sea-ice by 3.5-4.1 % per decade (1979-2012; IPCC 2014), and mean sea-ice thickness at the end of the melt season by 1.6 m (1958-2000; Kwok et al. 2009) as well as a 5-day increase in the length of the summer melt-season (1979-1996; Smith 1998). Further sea-ice loss would be disastrous for species that depend on sea-ice to live, hunt and breed, such as some invertebrates and vertebrates including the polar bear. Ice algae would also lose their substrate (Loeng et al. 2005). Despite this, some phytoplankton taxa may benefit because a major barrier to sunlight access would be removed. Whilst oceanic primary production has recently decreased at lower latitudes, Pabi et al. (2008) reported an overall ~30 % Arctic increase in oceanic primary production between 1998 and 2006. Ardyna et al. (2014) also noted that phytoplankton blooms in autumn are becoming more common throughout the Arctic, as reduced ice cover allows winds to stir the water, breaking down any existing water column stratification, and making nutrients previously at depth available to phytoplankton near the surface. Another consequence of the recent sea-ice melt combined with an increased river discharge (Peterson et al. 2002) has been an increase in water column stratification (McLaughlin & Carmack 2010). This has created a scenario in which nutrients in shallow water can become more quickly depleted by its algal residents. Li et al. (2009) have already reported a corresponding shift from larger cells (i.e. diatoms) to smaller algal cells (i.e. pico- phytoplankton) which can extract available nutrients more efficiently. Such changes in the composition or biomass of primary producers could result in unpredictable ‘bottom-up’ effects that ripple through planktonic grazers to higher trophic levels. In the future, possible loss of ice algae blooms and changes in the timing of phytoplankton blooms could cause

more ‘mismatch’ scenarios in the development of primary producers and grazers, with consequences for entire food webs (Søreide et al. 2010, Leu et al. 2011).

Changes in zooplankton abundances may also affect the survival of consumers which selectively take larger lipid-rich taxa. Falk-Petersen et al. (2007) suggested that warming in the Barents Sea (+2-3 ºC) could result in the replacement of two large lipid-rich species of copepods (Calanus glacialis and C. hyperboreus) by a smaller, lipid-poor North Atlantic congener species (C. finmarchicus). This may adversely affect specialist feeding seabirds whilst benefiting herring and baleen whales, herring predators. Global warming is also associated with northern movements of previously southern fish and mammals. There are recent reports of northern movements of NE Atlantic Cod in the Barents Sea (e.g. Kjesbu et al. 2014) and of killer whales in the Canadian Arctic (e.g. Higdon et al. 2012). As particularly voracious predators on seals and fish, killer whales could exert top-down effects on lower trophic levels, with unpredictable effects on zooplankton (Ferguson et al. 2010).

Arctic seas provide a wide range of ecosystem services for sustenance, recreation and the well-being of people living within and outside the Arctic (e.g. petroleum products, fish, animal furs, tourist destinations). Seabed beyond the Arctic Circle is thought to contain 30 % and 13 % of the world’s undiscovered oil and gas respectively, with the majority of undiscovered oil located in Arctic Alaska (Bird et al. 2008). With ice-free waters expected to occur as early as September 2045 (Laliberté et al. 2016), these supplies will become increasingly accessible and may be exploited as global population grows. Potential oil spills are a concern because Arctic ecosystems are frail and Arctic seas present significant containment and recovery challenges. The ecological impacts of previous large spills in sub- Arctic waters have been considerable (e.g. Exxon Valdez 1989 in the Gulf of Alaska), with spilt oil persisting in marine habitats for decades after spillages (e.g. Peterson 2001). Shipping has also increased: with a change in the number of recorded vessels travelling through the Northern Sea Route from 4 in 2010 to 71 in 2013 (Linstad et al. 2016). This also exposes ecosystems to increased pollution, and so careful regulation placed on activities will be critical to maintaining Arctic biodiversity. Worldwide increases in jellyfish populations in recent years have been a concern for human health and marine operations such as fishing,

drilling etc. These observations may be related to increasing temperatures, low populations of forage fish (competitors for food) and the presence of humans, amongst other factors (e.g. Brotz et al. 2012, Purcell 2012). Brotz et al. (2012) concluded with high certainty that jellyfish numbers will increase in the Southern Ocean but it is unclear how they will fare in the Arctic where less restricted access offers new prospects for human activity. Similarly, few predictions have been made into how polar chaetognaths might respond to climate change and other environmental pressures.

1.5 Study areas This thesis includes depth-integrated and depth-stratified sampling of zooplankton in several different regions of the Arctic Ocean. The major sampling regions were the Svalbard archipelago in the European Arctic (Chapter 2), and Amundsen Gulf, located in the south- eastern Beaufort Sea of the Canadian Arctic (Chapters 3 and 4). Additional collections from other regions of the Canadian Arctic (the Labrador Sea, Baffin Bay and the Canadian archipelago) were included in the analyses for Chapters 3 and 4. Finally, collections from the Alaskan part of the Chukchi Sea were included in Chapter 4.

The Svalbard archipelago: The first of our main sampling regions; the Svalbard archipelago in Arctic Norway, is a meeting point for Arctic water and Atlantic water (Strömberg 1989). We studied chaetognaths in several fjords on the west and north coasts of Svalbard. Parasagitta elegans is the dominant chaetognath species in the Barents Sea (Falkenhaug 1991) and along the west coast of Svalbard (Grigor et al. 2014, Hirche et al. 2015), but Eukrohnia hamata dominates to the north in regions with pack ice and reduced influence of Atlantic water (Daase & Eiane 2007, Daase et al. 2016). Several Arctic fjords including those in Svalbard and Baffin Bay are particularly useful for zooplankton studies due to having ‘sills’, seabed intrusions that reduce advection of water masses, and associated plankton communities (e.g. Syvitski et al. 1990, Bell & Josenhans 1997, Arnkvaern et al. 2005, Grigor et al. 2014). Svalbard fjords in recent years have been the focus of novel polar night research cruises led by Norwegian universities to investigate pelagic biology and winter ecology (Lønne et al. 2014).

The Amundsen Gulf: The second of our main sampling regions; the Amundsen Gulf in the southeastern Beaufort Sea, was the site of a unique overwintering sampling campaign in which an icebreaker was used to sample zooplankton and other ecosystem components year- round (Figure 1.8). This formed a key part of the Circumpolar Flaw Lead (CFL) System Study of 2007-2008. Several studies on copepod ecology have already been published from this time (e.g. Forest et al. 2010, Wold et al. 2011a, Darnis & Fortier 2014).

1.6 Aims and objectives The central aim of this doctoral thesis is to improve our understanding of the ecology of two major arctic chaetognath species; Eukrohnia hamata and Parasagitta elegans (Pseudosagitta maxima was generally excluded from the thesis because of its rarity or absence at the sampling locations). Specifically, the thesis aims at improving our understanding of these chaetognaths’ feeding strategies and life cycles in the European Arctic, Canadian Arctic and Alaskan Arctic. To that end, the thesis comprises three main chapters (2-4), presented as scientific articles. The seasonal adaptations of arctic chaetognaths are interesting given the

high seasonality in the abundance of known prey (copepods) in epi-pelagic waters, which could require these predators to carry out seasonal vertical migrations (SVM), or be able to quickly switch their food sources. Chapters 2-4 each comprise winter sampling, addressing the knowledge gap of biological processes during the polar night (Lønne et al. 2014). This will allow us to gain a more holistic understanding of how chaetognaths in the Arctic could react to climate change and other environmental pressures. The inclusion of locations with different environmental conditions at the same time of the year allowed me to investigate possible differences in life cycles and feeding strategies between populations.

The objective of Chapter 2 was to shed light on the polar night ecology of Parasagitta elegans in fjords of the Svalbard archipelago, where P. elegans is previously reported to be an abundant chaetognath species. Feeding ecology and reproductive dynamics were examined in four different fjords during January 2012 and January 2013. Gut content analyses were used to assess diets, feeding rate and potential impacts of P. elegans on copepod populations. To gain further insights on the trophic positions of chaetognaths, stable isotope and fatty acid trophic marker (FATM) analyses were carried out.

The objective of Chapter 3 was to describe the life histories of Eukrohnia hamata and Parasagitta elegans in the Canadian Arctic, mainly in the Amundsen Gulf, over a full annual cycle (August 2007 to August 2008). Growth and reproductive dynamics were inferred from body length, and the condition of oil vacuoles and maturity features. I tested the hypothesis that E. hamata is a capital breeder that spawns during most of the year, whilst P. elegans is an income breeder, reproducing exclusively during periods of high prey availability.

The objective of Chapter 4 was to understand how food resources are utilized year-round by co-existing Eukrohnia hamata and Parasagitta elegans in the Canadian and Alaskan Arctic. In the Amundsen Gulf, gut contents were examined to reveal diets from November 2007 to August 2008. FATM and stable isotope signatures of E. hamata and P. elegans in two distant regions (the Alaskan portion of the Chukchi versus Sea and Baffin Bay) were compared, based on collections made in autumn 2014.

2. Chapter 2 – Polar night ecology of a pelagic predator, the chaetognath Parasagitta elegans

2.1 Résumé

Les routines annuelles et l'écologie saisonnière des espèces de zooplancton herbivores sont relativement bien connues en raison de leur couplage étroit avec leur source de nourriture à variation saisonnière, la production primaire. Quant aux niveaux trophiques planctoniques supérieurs, ces interactions saisonnières sont moins bien comprises. Le présent chapitre traitera de l’alimentation des chaetognathes dans les écosystèmes des fjords du haut-Arctique au milieu de l’hiver. Les chaetognathes sont des prédateurs planctonophages très présents dans les mers de haute latitude. Nous avons étudié l’espèce commune, le Parasagitta elegans, autour de l'archipel du Svalbard (78-81ºN) au cours des hivers 2012 et 2013. Nos échantillons étaient constitués de spécimens (longueur de corps : 9-55 mm), échantillonnés dans trois fjords, dont le contenu du tube digestif (n = 903), la présence d’isotopes stables, la composition en acides gras et l'état de maturité (n = 352) ont été étudiés. Environ un quart des spécimens présentaient du contenu dans leur tube digestif, principalement des gouttelettes de lipide et de débris chitineux, alors que seulement 4 % d’entre eux contenaient des proies identifiables, principalement des copépodes Calanus spp. et Metridia longa. La teneur élevée en δ15N du P. elegans et son niveau trophique moyen de 2.9 a confirmé son profil de carnivore. Dans la même perspective, le profil d'acides gras [en particulier ses niveaux élevés de 20:1 (n-9) et 22:1(n-11)] a confirmé la consommation de Calanus. Des observations de gonades non-développées dans la plupart des P. elegans de grande taille, ainsi que l'absence de petits spécimens ˂10 mm, a suggéré que la reproduction n'a pas commencé si tôt dans l'année. Le taux d'alimentation moyenne du P. elegans selon les fjords et les années était de 0,12 proie/ind.-1/jour-1, ce qui est faible par rapport aux mesures du taux d'alimentation printanier et estival dans des environnements de haute latitude. Nos résultats suggèrent une activité d'alimentation réduite pendant l'hiver. Par ailleurs, la prédation du P. elegans semble avoir peu d'effets sur la mortalité des copépodes.

2.2 Abstract

The annual routines and seasonal ecology of herbivorous zooplankton species are relatively well known due to their tight coupling with their pulsed food source, the primary production. For higher trophic levels of plankton, these seasonal interactions are less well understood. Here, we study the mid-winter feeding of chaetognaths in high-Arctic fjord ecosystems. Chaetognaths are planktivorous predators which comprise high biomass in high-latitude seas. We investigated the common species Parasagitta elegans around the Svalbard archipelago (78-81ºN) during the winters of 2012 and 2013. Our samples consisted of individuals (body lengths 9-55 mm) from three fjords, which were examined for gut contents (n = 903), stable isotopes, fatty acid composition, and maturity status (n = 352). About a quarter of the individuals contained gut contents, mainly lipid droplets and chitinous debris, whilst only 4 % contained identifiable prey, chiefly the copepods Calanus spp. and Metridia longa. The δ15N content of P. elegans, and its average trophic level of 2.9, confirmed its carnivorous position and its fatty acid profile [in particular its high levels of 20:1 (n-9) and 22:1 (n-11)] confirmed predation on Calanus. Observations of undeveloped gonads in many of the larger P. elegans, and the absence of small individuals ˂10 mm, suggested that reproduction had not started this early in the year. Its average feeding rate across fjords and years was 0.12 prey ind.-1 day-1, which is low compared to estimates of spring and summer feeding in high- latitude environments. Our findings suggest reduced feeding activity during winter and that predation by P. elegans had little impact on the mortality of copepods.

2.3 Introduction The Arctic is a highly seasonal environment, with conditions in the marine pelagic varying greatly between summer and winter. The extreme cycles in solar illumination and sea-ice cover create strong seasonality in primary production (Ji et al. 2013), which affects food availability for organisms at higher trophic levels (Varpe 2012). Food availability for herbivores such as Calanus copepods is severely reduced during the polar night. Many grazers survive winter food shortages by resting at depth and sustaining themselves on storage lipids accumulated when food was last available (e.g. Conover 1988). Although pelagic omnivores and carnivores may experience a less-pulsed food source than for the herbivores, they may also display adaptations to a seasonal food source (e.g. Newbury 1971, Choe et al. 2003, Kraft et al. 2013). However, for many predators we lack knowledge of their feeding opportunities and activities during winter.

The chaetognath Parasagitta elegans (Verrill 1873) is highly represented in Arctic mesozooplankton communities, numerically and in biomass terms (e.g. Søreide et al. 2003, Hopcroft et al. 2005). This typically neritic species (Bieri 1959) can exert high predation pressure on Arctic copepods and compete with larval fish (e.g. Sameoto 1987). Some authors have proposed that chaetognaths are not particularly sensitive to seasonality (e.g. Hagen 1999), suggesting that their varied diets (reviewed by Terazaki 2004) and non-visual food search should allow them to find adequate nutrition throughout the year. Lipid-rich copepods, often preferred prey, diapause in large densities and in an unalert state during winter, so they may be easy prey for some carnivores (Darnis et al. 2012). However, some chaetognath species at high latitudes are found to contain considerable amounts of lipids, including wax esters, which may be used for storage (e.g. Lee 1974, Kruse et al. 2010). Furthermore, winter growth rates of P. elegans in the Arctic is lower than those in spring and summer (Grigor et al. 2014), and these observations could indicate reduced winter feeding and resting strategies in this chaetognath.

We report here on the winter feeding ecology of Parasagitta elegans in the European high Arctic. We used three different methods to assess its diet and trophic level: gut content

analysis, which records recent feeding, as well as stable isotope and fatty acid trophic marker (FATM) analyses, which provide additional information on its feeding history, over time frames of weeks to months (Graeve et al. 2005). We calculated feeding rates and possible impact on copepod populations. Although we did not expect mid-winter reproduction (e.g. Kramp 1939, Grigor et al. 2014), we also examined their level of maturity, allowing us to infer how close P. elegans is to reproducing, which links to how actively individuals are expected to feed.

2.4 Method 2.4.1 Study area Chaetognaths were collected from waters around Svalbard (78-81ºN) during the winters of 2012 and 2013, during the ARCTOS ‘Polar Night Cruises’ (8 to 21 January 2012 and 9 to 18 January 2013).

The Svalbard archipelago is a meeting place for warm Atlantic water sourced from more southern regions and colder Arctic water from the north. The northward moving West Spitsbergen Current deposits Atlantic water into fjords on the west coast, with one branch finally turning north-east at the top of Spitsbergen and affecting conditions in the Arctic Ocean (Saloranta & Haugen 2001). We sampled three fjords in 2012: Rijpfjorden (station R3), Isfjorden proper, and Adventfjorden, a tributary fjord of Isfjorden (station ISA, see Figure 2.1). The latter two locations are hereafter referred to as ‘Isfjorden’. Another fjord, Kongsfjorden, was sampled in 2013, and Rijpfjorden was re-sampled. See Appendix A for full details on the sampling activities. Isfjorden (78ºN, 14ºE) and Kongsfjorden (79ºN, 11.5ºE) are influenced by Atlantic water masses and the outer parts are ice free for much of the year (Svendsen et al. 2002, Nilsen et al. 2008), causing important seasonal changes in biodiversity and animal populations in these fjords. Rijpfjorden, in contrast (80ºN, 22ºE), is influenced by Arctic water, typically remaining ice-covered from December/February to July (Wallace et al. 2010).

Figure 2.1 Map showing the locations of the stations sampled for chaetognaths in January 2012 and 2013.

2.4.2 Physical and biological environment At all stations, vertical profiles of salinity, temperature, density, and fluorescence were obtained using a CTD (SBE 9) and processed following standard Sea Bird Electronics (SBE) procedures. In 2012, a steep thermocline was observed in Rijpfjorden between ~70 and 100 m, over which temperature rose to 2 ºC and then fell again to 0.5 ºC at ~200 m (Appendix B). At the mouth of Isfjorden, the thermocline was much weaker, with temperature increasing gradually to 2 ºC at ~200 m. At Isfjorden (station ISA), temperature varied little with depth (Appendix B). At all stations, salinity varied with depth in similar ways to temperature. A strong halocline occurred in Rijpfjorden only, at the same depths as the thermocline (Appendix B). Fluorescence was very low (˂0.6 g l-1) at all stations in January 2012.

2.4.3 Zooplankton sampling Horizontal trawls of the large Methot Isaac Kidd gear (MIK, 3.14-m2 opening, 1.5-mm mesh) were used to obtain large numbers of Parasagitta elegans for the various dietary analyses (Appendix A). MIK sampling depths in Rijpfjorden and Isfjorden (in 2012) captured all water

masses in the fjords (melt water, Arctic water/mixed Fram Strait water and Arctic deep water). In contrast, most MIK sampling for chaetognaths in 2013 (Kongsfjorden and Rijpfjorden) was carried out in the upper 20 m (Appendix A). This sampling decision was taken to reduce haul time, as chaetognaths can easily become damaged and stressed during net sampling, which poses several problems for gut content analyses (Baier & Purcell 1997).

In Rijpfjorden in 2012, vertical hauls of the smaller Multi-Plankton Sampler gear (MPS, 0.25-m2 opening, 0.2 mm mesh) were also performed to collect zooplankton community data (see also Daase et al. 2014) and to source smaller Parasagitta elegans for gut content analyses (sampled strata: 260-200 m, 200-100 m, 100-50 m, 50-20 m, 20-0 m; see Appendix A for details). The MPS captured higher densities of smaller P. elegans (10-19 mm) in the upper 20 m than the MIK (0.5 ± 0.4 compared to 0.1 ± 0.1 ind. m-3), but fewer individuals ≥20 mm (1.8 ± 1.7 compared to 5.9 ± 8.2 ind. m-3). Individuals ˂10 mm were not captured by either gear in Rijpfjorden or Isfjorden in 2012. Grigor et al. (2014) similarly showed that the MPS captured the smaller size fraction of P. elegans more efficiently than a larger ‘WP3’ net with a 1-m2 opening. MPS samples collected from Isfjorden ISA on 27 January 2012 were also analysed for abundance data.

2.4.4 Sample processing Upon retrieval of each MIK haul, the cod end was immediately transferred to a marked bucket and diluted up to the 10 l mark. A 0.6 l sub-sample was taken for community analysis and fixed in 4 % buffered formalin-seawater solution. Webster et al. (2013) show presence- absence data for zooplankton in Rijpfjorden from the community samples from 2012. From a second 0.6 l subsample, 100 Parasagitta elegans were randomly picked out for gut content analyses (Appendix A) and preserved in 4 % buffered formalin-seawater solution. On the 2012 cruise, 75-150 P. elegans were also picked out of this second sub-sample for stable isotope analyses (n individuals = 675, Appendix A), and 30 individuals were randomly picked out for fatty acid trophic marker (FATM) analyses (n individuals = 290, Appendix A). These samples were frozen at -80 ºC until the end of the cruise and then stored at -80 ºC for further processing.

2.4.5 Gut content analyses A total of 202 individuals from 2012 (body lengths 10-55 mm) and 701 from 2013 (body lengths 9-42 mm) were dissected and analysed for gut contents. In 2012, a minimum of 10 individuals were dissected from each MIK haul. In 2013, we examined the guts of all 100 individuals preserved from each MIK haul; 3-10 smaller individuals were sourced from each MPS haul (Appendix A).

Parasagitta elegans individuals were measured to the nearest millimeter (excluding the caudal fin) and stained with a solution of lignin-red, used to stain tissues of crustacean prey in their guts (Falkenhaug 1991), and borax carmine, capable of highlighting the gonads of prey (Pierce 1941), diluted 50 times in water. Head widths were measured to the nearest 0.01 millimeter under the binocular microscope. All visible gut contents were described. Detected prey was identified to the lowest possible taxonomic level and photographed using a digital camera connected to a dissecting microscope (Leica DFC 320).

2.4.6 Food-containing ratio and feeding rate The food-containing ratio (FCR) is the proportion of individuals in a population with food in their gut. Some only include chaetognaths containing identifiable prey in the FCR calculation

(FCRmin, e.g. Falkenhaug 1991, Kruse et al. 2010). However, this approach may not account for loss of prey items from guts during net towing, due to damage or stress-induced regurgitation (Baier & Purcell 1997, see ‘Discussion’). Other studies also included individuals with other signs of recent prey digestion (gut lipids and debris, e.g. Sameoto 1987, Froneman & Pakhomov 1998), which may give an upper estimate of the proportion feeding

(FCRmax). FCRmax measurements may not be appropriate for some chaetognath species, such as those that contain oil vacuoles in the centre of their bodies. This includes another Arctic species, Eukrohnia hamata (Möbius 1875), in which oil vacuoles may be for storage or buoyancy (Øresland 1990, Pond 2012, pers. obs.). However, these oil vacuoles do not occur in Parasagitta elegans. We calculated both FCRmax and FCRmin for P. elegans from each haul. We also calculated feeding rates (FR: no. of prey items consumed ind.-1 day-1) according

to Eq. (2.1). Only individuals containing identifiable prey were included in the FR calculation.

푛 x 24 FR = prey (2.1) tdig where nprey = mean no. of identifiable prey per chaetognath, and tdig = digestion time in hours.

Feigenbaum (1982) suggested a tdig of 10.2 h for Parasagitta elegans from a laboratory study on starved specimens from Vineyard Sound, maintained at 0 ºC. As the water temperature in all our study fjords was close to 0 ºC (Appendix B), we used Feigenbaum’s tdig estimate.

Proportions of individuals per haul containing different gut types were analysed using Kruskal–Wallis tests.

2.4.7 Stable isotope analyses As the ratios of 13C/12C and 15N/14N in organisms at different trophic levels tend to differ considerably, stable isotope analyses can be used to examine marine food web structure and the transfer of energy between trophic levels (McConnaughey & McRoy 1979). Stable isotope analyses were performed on our 2012 samples at the Institute for Energy Technology in Kjeller, Norway, using similar methods to Søreide et al. (2006). Three replicate samples each containing 25 Parasagitta elegans were analysed from seven of the MIK hauls, whilst six replicate samples of 25 inds. were analysed from an eighth MIK haul at 225 m in Rijpfjorden (Appendix A). All P. elegans had body lengths within the range of 10-50 mm. The whole body of each chaetognath was used, except lipids and non-dietary carbon (i.e. carbonates), which were removed before analyses by Soxhlet extraction with CH2Cl2. Lipids have notably lower levels of 13C than proteins and carbohydrates (van Dongen et al. 2002), and their removal reduces a main source of measurement variability between individuals (Hobson & Welch 1992). C/N ratios were expressed as the deviation from standards in ppt (‰) according to Eq. (2.2) (Søreide et al. 2006).

δX=[(푅sample/푅standard) − 1]×1000 (2.2) where X = 13C or 15N and R = the corresponding ratios 13C/12C or 15N/14N. International standards, Pee Dee Belemnite for δ13C (PDB: USGS 24), and atmospheric air for δ15N (IAEA-N-1 and 2) were used to determine R. Carbon and nitrogen composition were expressed as percentages of animal dry weight.

2.4.8 Determination of trophic level The trophic level (TL) of Parasagitta elegans in each fjord was calculated as the difference between its average δ15N content and that of the winter food web baseline, assuming a constant fractionation between trophic levels [Eq. (2.3), Søreide et al. 2006].

δ15N 푃.푒푙푒푔푎푛푠− δ15N 퐶.푔푙푎푐푖푎푙푖푠 TL =∝ + (2.3) ∆N where α = trophic level of the food web baseline and DN = the trophic enrichment factor for δ15N (average amplification) per trophic level. We took the abundant grazer Calanus glacialis (α = 2) to represent the winter food web baseline. This decision was made because Parasagitta elegans are thought to be carnivores that typically do not feed on primary production but feed often on Calanus, and primary production is anyway low in winter (pers. comm.: J Søreide).

We used a δ15N value for Calanus glacialis of 9.6 ± 0.2 ‰, estimated for individuals from 300 m in Svalbard fjords during December (Søreide et al. 2008). We used Δ = 3.4 ‰, also determined for the European Arctic by Søreide et al. (2006).

2.4.9 Fatty acid analyses Groups of primary producers and some herbivores produce fatty acids that are unique to them, known as fatty acid trophic markers (FATMs, Dalsgaard et al. 2003). Higher consumers possess a lower capacity to synthesise their own fatty acids, or to modify those received from their prey. Therefore, when FATMs persist in a predator for some time with little modification, they can be used to quantify its diet (Dalsgaard et al. 2003).

Determination of the fatty acid signatures of Parasagitta elegans from Isfjorden and Rijpfjorden (in 2012) were performed on triplicate samples from nine MIK hauls (Appendix A) at UNILAB, Tromsø, Norway, using similar methods to Wold et al. (2011b). Each sample contained 10 pooled individuals with dry weights between 0.001 and 0.05 g. Total lipid was extracted in 15 ml 2:1 chloroform–methanol with butylated hydroxytoluene (BHT) (Folch et al. 1957). Each sample was supplemented with a known amount of the fatty acid 21:0, as an internal standard, and trans-methylated in methanol containing 1 % sulphuric acid with toluene for 16 h at 50 ºC. The relative compositions (%) of FA methyl esters were determined on an Agilent 6890 N gas chromatograph, equipped with a fused silica, wall-coated capillary column with an Agilent 7683 injector and flame ionisation detection. The fatty acid methyl esters were identified and quantified by gas chromatography. Results are given as relative percentages of the various fatty acids identified in specimens from the two fjords. Differences in average fatty acids proportions between sampled depths in Isfjorden and Rijpfjorden were analysed using one-way ANOVA. Differences in the fatty acid profile between fjords was analysed using the Mann-Whitney U test (sum rank test).

2.4.10 Mid-winter maturity status Chaetognaths are hermaphrodites possessing male and female gonads (Alvarino 1992). The male gonads (testes and seminal vesicles) produce and secrete sperm, whilst the female gonads (ovaries and seminal receptacles) produce ova and receive sperm from a sexual partner (Alvarino 1992). Mature Parasagitta elegans are characterised by advanced ovaries containing large oocytes, high sperm loads filling the tail section of the animal, and pairs of pronounced seminal receptacles and vesicles (Russell 1932, Choe et al. 2003). We described the maturity of 29 of the largest individuals (40-51 mm, 8×1 mm size classes), and in 323 smaller individuals (13-39 mm, 26×1 mm size classes) in MIK samples from all fjords (both years, Appendix A). Measured individuals were stained with borax carmine solution to highlight their gonads (Pierce 1941). Maturity assessments were made under the binocular microscope following the methods of Grigor et al. (2014). Ovaries were measured to the nearest 0.1 mm, and individuals showing advanced ovary development were taken as those with ovaries ≥5.4 mm, whilst shorter ovaries ˂5.4 mm were considered to be poorly developed. The volume of loose sperm in the tail was estimated to the nearest 25 %. Low

sperm loads could suggest that sperm is filling the tail or has already been released. However, when sperm has already been released, remaining sperm appears more sparsely distributed. Therefore, spent individuals can be separated from those which have not yet secreted sperm. The size of the seminal receptacles was also described (see Grigor et al. 2014 for further details).

2.5 Results 2.5.1 Chaetognath abundance and prey field In Rijpfjorden in 2012, the most abundant zooplankton taxa based on the MPS sampling were the larger calanoid copepod Calanus finmarchicus and the smaller copepods Pseudocalanus spp. (Calanoida) and Oithona similis (Cyclopoida) (Table 2.1). Parasagitta elegans was the most abundant chaetognath, but overall chaetognaths were also less abundant than the non- copepod taxa Oikopleura spp. (Tunicata) and Beroe cucumis (Ctenophora) (Table 2.1).

Average Parasagitta elegans abundances in Rijpfjorden MIK trawls ranged from 2.4 ± 1.2 ind. m-3 (225 m) to 14.7 ± 7.0 ind. m-3 (75 m), comparing relatively well with the MPS results. For further details on the polar night zooplankton community in Rijpfjorden, see Webster et al. (2013), Kraft et al. (2013), and Daase et al. (2014).

Taxonomic group Functional (feeding) group Abundance (ind. m-3) Cyclopoida (Copepoda) Omnivores 1467.6 Small Calanoida (Copepoda) Omnivores 1257.5 Calanus spp. (Copepoda) CI-AF Herbivores 881.5 Metridia longa Omnivore 78.6 Tunicata Variable between species 36.9 Ctenophora Carnivores 23.5 Limacina helicina (Pteropoda) Herbivore 17.3 Parasagitta elegans (Chaetognatha) Carnivore 5.5 Clione limacina (Pteropoda) Carnivore 3.9 Other Mollusca Variable between species 3.9 Isopoda Variable between species 3.4 Eukrohnia hamata (Chaetognatha) Carnivore 1.4 Travisiopsis sp. (Polychaeta) Carnivores 1.2 Harpacticoida (Copepoda) Variable between species 0.9 Echinodermata Variable between species 0.6 Hydrozoan medusae Variable between species 0.6 Euphasiacea Variable between species 0.3 Apherusa glacialis (Amphipoda) Herbivore 0.1 Carnivorous Calanoid copepods Carnivores 0.1 Pseudomma truncatum (Mysidacea) Omnivore 0.1 Ostracoda Variable between species 0.1

2.5.2 Gut contents Observed gut contents were copepods, chitinous debris, lipid droplets, and other detritus that could not be identified. The proportions of Parasagitta elegans with gut contents (FCRmax) varied between fjords and years (Kruskal-Wallis test, P < 0.01), ranging from 9 to 53 % (Figure 2.2). Some individuals contained two or more gut content types. Lipid droplets were the most common gut observation overall (Figure 2.3 and Figure 2.4), observed in 106 (11.7 %) of the 903 chaetognaths. A total of 33 individuals (3.7 %) contained identifiable copepod prey, including Calanus finmarchicus, Metridia longa, and harpacticoid copepods.

Proportions containing prey (FCRmin) also varied between fjords and years (Kruskal-Wallis test, P ˂ 0.05), ranging from 0 to 10 % (Figure 2.2). FRs ranged from 0.00 to 0.24 prey ind.- 1 day-1 (mean = 0.12 prey ind.-1 day-1). Amongst feeding P. elegans, median per haul proportions with prey varied with fjord (Kruskal-Wallis test, P ˂ 0.005), but proportions with other gut contents did not (P ˃ 0.05, Figure 2.3). The relationship between chaetognath head width and body length was linear (y = 0.031x + 0.37, R2 = 0.79), and prey was identified in individuals with head widths ≥0.87 mm (Figure 2.4). Per haul proportions of P. elegans with gut contents increased with head width size class (Kruskal-Wallis test, P ˂ 0.05, Figure 2.4), but amongst feeders, proportions with each gut content type did not vary with head width (P ˃ 0.05, Figure 2.4).

Figure 2.2 Proportions (%) of Parasagitta elegans individuals per haul with gut contents

(FCRmax), identifiable prey (FCRmin), and empty guts in each fjord. The horizontal line inside each boxplot shows the median of the proportions over multiple haul samples in a fjord. The lower and upper boxes show the lower and upper quartiles, respectively, and the vertical lines outside the boxes the differences between these quartiles and the lowest and highest

proportions observed. Each dot represents an outlying data point. nhauls = numbers of hauls for each fjord. Hauls with ˂3 individuals analysed were not included. As only one haul was analysed for Rijpfjorden in 2013, full boxplots could not be shown. See Appendix A for numbers of individuals analysed per haul.

Figure 2.2. ninds. = total numbers of individuals for each size class.

2.5.3 Body composition Values for carbon and nitrogen isotopes varied little between fjords and sampling depths (Table 2.2). The δ13C content of Parasagitta elegans ranged from -22.0 ± 0.3 ‰ (Rijpfjorden) to -21.5 ± 0.2 ‰ (Isfjorden), whilst δ15N content ranged from 12.5 ± 0.3 % to 12.9 ± 0.1 ‰ (both Isfjorden). Average trophic level (TL) based on the δ15N content was 2.9, and average C/N ratio over fjords and depths was 3.1 (Table 2.2).

Location Trawl depth (m) δ13C (‰) δ15N (‰) DW%C DW%N C/N TL Mouth of Isfjorden 250 -21.7±0.0 12.5±0.3 48.4±0.5 15.6±0.4 3.1±0.0 2.9 Isfjorden (ISA) 30 -21.5±0.2 12.9±0.1 48.4±0.4 15.5±0.1 3.1±0.0 3.0 -"- 35 -21.5±0.1 12.8±0.2 47.9±1.1 15.4±0.4 3.1±0.0 2.9 -"- 60 -21.6±0.1 12.9±0.1 48.3±0.8 15.5±0.4 3.1±0.0 3.0 Rijpfjorden 20 -22.0±0.0 12.5±0.2 48.7±0.3 15.8±0.1 3.1±0.0 2.9 -"- 75 -21.8±0.2 12.5±0.2 48.0±0.2 15.4±0.0 3.1±0.0 2.9 -"- 75 -22.0±0.3 12.6±0.3 48.9±0.4 15.7±0.2 3.1±0.0 2.9 -"- 225 -21.8±0.1 12.5±0.1 48.6±0.4 15.8±0.2 3.1±0.0 2.9

Parasagitta elegans fatty acid signature did not differ between Isfjorden and Rijpfjorden (Mann-Whitney U test, P ˃ 0.05). Similarly, percentage composition of almost all fatty acids did not differ between sampling depths in the two fjords (one-way ANOVA, P ˃ 0.05). The high levels of 18:1 (n-9) compared with those of 18:1 (n-7) strongly indicate carnivory. The Calanus fatty acid marker 20:1 (n-9) and 22:1 (n-11) were both abundant. The dinoflagellate marker docosahexaenoic acid (DHA) 22:6 (n-3) was recorded in interestingly high amounts,

whilst eicosapentaenoic acid (EPA) 20:5 (n-3) was recorded in moderate to high amounts (Table 2.3).

Mean % Isfjorden Mean % Rijpfjorden Fatty acid (n samples = 12) (n samples = 17) 14:0 FA 5.02±2.09 4.13±0.62 14:1 (n-5) FA 0.54±0.06 0.52±0.08 † 15:0 FA 0.72±0.07 0.73±0.06 16:0 FA 11.91±1.16 12.65±1.47

16:1 (n-5) FA 2.97±0.41 3.30±0.39 16:1 (n-7) FA 7.13±0.27 6.05±0.33 18:0 FA 1.09±0.13 1.18±0.20 18:1 (n-7) FA 1.95±0.28 1.74±0.18 18:1 (n-9) FA 5.35±0.65 5.82±0.54 18:2 (n-6) FA 1.28±0.11† 1.35±0.09

18:3 (n-3) FA 1.11±0.11 1.37±0.07 18:4 (n-3) FA 1.76±0.63 2.35±0.35 20:1 (n-9) FA 14.80±1.02 11.46±1.71 20:4 (n-3) FA 0.82±0.10 0.99±0.07 20:5 (n-3) FA (EPA) 11.41±0.68 12.73±0.75 22:1 (n-11) FA 6.77±1.38 5.69±0.71

22:1 (n-7) FA 0.47±0.02 0.78±0.30 22:5 (n-3) FA 0.60±0.14 0.69±0.15 22:6 (n-3) FA (DHA) 18.01±2.33 18.81±1.99 24:1 (n-9) FA 2.88±0.21 3.81±0.62

2.5.4 Mid-winter maturity status Amongst the largest individuals in the population (body lengths 40-51 mm), 55 % had well- developed seminal receptacles, whereas 93 % had relatively high sperm volumes in their tails (50-100 %) and all had advanced ovaries (≥5.4 mm). In smaller individuals (lengths 13-39 mm), 45 % had well-developed seminal receptacles and 42 % had advanced ovaries; 80 % had synthesized relatively high sperm volumes (filling 50-100 % of the tail area).

2.6 Discussion 2.6.1 Studies during the polar night Studies of plankton ecology in the high Arctic are typically restricted to spring and summer, due to the logistical difficulties of sampling during winter. The polar night cruise, of which our study formed a part, offered unique possibilities for a better understanding of polar night marine ecology (see Berge et al. 2012, Webster et al. 2013, Daase et al. 2014).

We show that the common chaetognath Parasagitta elegans remains an active carnivore during the polar night. This arrow worm is not in a dormant and non-feeding state although its feeding rates may be considerably lower than in spring and summer, and reproduction is absent (or occurring at very low rates) at this time of year. These observations add to the increasing awareness of activity levels and ecological interactions of pelagic organisms during the polar night.

2.6.2 Feeding activity and rates Copepods are abundant in Svalbard fjords during winter (Table 2.1) and as a non-visual predator Parasagitta elegans should, as opposed to fish and other visual predators, be able to encounter and catch them also during the polar night. The behaviour of some copepods during the polar night (i.e. diapausing in an unalert state) could make them particularly susceptible to predators such as chaetognaths (Darnis et al. 2012). About a quarter of the P. elegans individuals showed signs of recent feeding (mainly lipid droplets). The body of P. elegans does not contain a centralised oil vacuole, suggesting that all lipid droplets in P. elegans guts remained from recently digested prey, yet only 4 % contained identifiable prey.

Based on Feigenbaum’s (1982) digestion time estimate of 10.2 h at 0 ºC (likely longer in colder waters), our findings would suggest that most individuals had not fed for several hours before their capture. In contrast, Falkenhaug (1991) identified prey in 36 % of the P. elegans population. collected from the Barents Sea in summer. Our average estimate of per capita feeding in P. elegans was 0.12 prey ind.-1 day-1, corresponding to 0.66 prey ind.-1 m-3 consumed per day by the P. elegans population (based on the abundance data in Table 2.1). Their predation impact on Calanus is therefore low given the high Calanus abundances (Table 2.1, see also Daase et al. 2014). Calanus are assumed to be in a dormant, non-feeding state during winter (Falk-Petersen et al. 2009). Daase et al. (2014) reported on Calanus from the same cruise in 2012 and found that the Calanus finmarchicus population in Rijpfjorden comprised mainly copepodite stage CV and some CIVs, which could indicate overwintering. However, these authors also noted that “the bulk of the C. finmarchicus and C. glacialis population was found close to the surface and not at greater depth where they presumably should overwinter”, suggesting they were not in an unalert dormancy phase, at least not in January. If not, these copepods may be more alert to the presence of predators than assumed, allowing them to avoid or escape chaetognaths.

Our FR estimate is similar to that obtained for immature specimens in northern Sweden in autumn-winter (0.2 prey ind.-1 day-1) and much lower than reported for spring-summer (0.7- 0.9 prey ind.-1 day-1, Øresland 1987). These findings suggest that feeding rates for Parasagitta elegans at high-latitudes drop during winter, corresponding well with the reduced growth rates observed during winter (Grigor et al. 2014), and the lack of reproduction at this time of year (this study). In less seasonal environments, energetic requirements may vary less between the seasons, accounting for relatively higher feeding rates at lower latitudes, also during winter and early spring (e.g. up to 1.33 prey ind.-1 day-1 in Vineyard Sound, Feigenbaum 1982).

However, care should be taken when comparing estimates of feeding rates between studies, because each study detects prey with various levels of precision. For example, we did not search for mandible remains in guts to detect further signs of recent copepod digestion, as in

other studies (e.g. Falkenhaug 1991, Giesecke and Gonzalez 2004). Feeding estimates in chaetognaths are also highly sensitive to sampling methodology. As well as the risk of damage to their fragile bodies, stressed chaetognaths may also regurgitate gut contents, leading to underestimates of true feeding rates. Baier & Purcell (1997) estimated prey loss from guts of ~50 % when tows were longer than 2 min. Such short hauls can be unfeasible, especially when using large trawl nets such as the MIK, and when sampling populations in deeper waters. Further studies should therefore utilize new zooplankton imaging devices for observing chaetognaths in the water column without necessarily capturing them in nets. Optical methods are promising avenues (e.g. Schulz et al. 2010, Sainmont et al. 2014b).

2.6.3 Energetics Stable isotope and FATM analyses confirmed the position of Parasagitta elegans as a predator during the polar night. From its average trophic level (TL) of 2.9, it can be classified as a carnivore according to a trophic model devised for the European Arctic (Søreide et al. 2006, in which carnivores had TLs between 2.9 and 3.3). δ15N (an indicator of protein content) and δ13C (an indicator of organic matter content) values varied little with station or depth, and agree well with values from the Barents Sea in March (Søreide et al. 2006). δ15N values were, however, lower than reported from the Bering Sea (14.7 ± 0.7 ‰, Lovvorn et al. 2005), suggesting that P. elegans in the sub-Arctic Pacific typically feed higher up the food chain (also see ‘Lipid profile’ section of ‘Discussion’). Many factors affect the elemental composition of zooplankton species, including age, size maturity status, reproductive strategies, as well as variations in the composition and growth rates of primary producers (Ikeda 1974), yet for P. elegans in the Barents Sea, δ15N values seem to vary little throughout the year. Søreide et al. (2006) reported δ15N values ranging from 11.9 ± 0.2 ‰ in spring to 12.2 ± 0.1 ‰ in winter. δ13C values varied slightly more between spring (-19.3 ± 0.8 ‰) and winter (-20.8 ± 0.3 ‰).

2.6.4 Lipid profile The high levels of the FATMs 20:1 (n-9) and 22:1 (n-11) confirm that Parasagitta elegans is part of the Arctic Calanus-based food web (Falk-Petersen et al. 2007, Wold et al. 2011b).

Furthermore, the high level of 22:6 (n-3) indicates that the base of the food chain is dominated by dinoflagellates, which can be very abundant in waters north of Svalbard in spring-summer (Hegseth & Sundfjord 2008). This lipid signature agrees well with previous results from the European sub-Arctic (Falk-Petersen et al. 1987). Both the presence of green detritus in P. elegans guts, and the surprisingly high levels of the dinoflagellate fatty acid marker docosahexaenoic acid, may suggest that some omnivory and/or detritivory occurs. Non- carnivorous feeding in chaetognaths has been suggested by Casanova et al. (2012). In the Canadian Arctic, Eukrohnia hamata and Pseudosagitta maxima have been observed ingesting green detritus under the microscope (pers. obs.). Eukrohnia species in Antarctica were found to contain higher amounts of 16:0, reflecting feeding on the Antarctic copepods including Rhincalanus gigas (Kruse et al. 2010). We suggest that winter feeding is captured in these results, following the suggestions of Graeve et al. (2005) that a new signal (source of food) will be visible relatively quickly (~4-8 days after feeding). If P. elegans starves during autumn and early winter, and if 22:6 (n-3) undergoes little transformation or metabolism (Dalsgaard et al. 2003), it remains possible that this signal is retained from feeding on Calanus during a previous summer or autumn bloom (Falk-Petersen et al. 1990, Wold et al. 2011b). However, this study and others (e.g. Feigenbaum 1982, Øresland 1987, Falkenhaug 1991) suggest a low likelihood of long-term fasting in P. elegans.

An important difference between the two main chaetognath species in Arctic waters, Eukrohnia hamata and Parasagitta elegans, is that the former typically possess oil vacuoles in the centre of their bodies, whilst the latter do not (Øresland 1990, Pond 2012, pers. obs.). Previous studies that considered the presence of oil vacuoles in this species to reflect recent feeding may therefore have overestimated feeding rates (e.g. Sameoto 1987, Froneman and Pakhomov 1998). The role of these vacuoles in E. hamata (e.g. buoyancy, storage) is not yet clear (Pond 2012), but unlike P. elegans, E. hamata has a cosmopolitan range. In the Arctic, both species commonly occur at epi-pelagic depths (Bieri 1959), but in the North Pacific, E. hamata populations are commonly found residing deeper, and throughout a wider depth range than P. elegans (e.g. Bieri 1959, Alvarino 1964). Maintaining such wide vertical distributions would certainly require a strong control of buoyancy, which could be offered by having centrally positioned oil vacuoles (Pond 2012). Lipid reserves form a central

component of the life history of polar zooplankton species (Falk-Petersen et al. 2009, Varpe et al. 2009). In Antarctica, 34 % of examined E. bathyantarctica individuals were found to contain oil droplets in summer compared to 57 % of individuals in winter. This species also contained relatively high amounts of the fatty acid 18:1 (n-9), which is found in storage lipids (Kruse et al. 2010). If the primary role of oil vacuoles is storage, this could suggest that survival and possibly reproduction in E. hamata is less dependent on concurrent food intake than in P. elegans. Studies of the extent of capital breeding (Varpe et al. 2009) in E. hamata are therefore needed.

2.6.5 Reproduction Small Parasagitta elegans <10 mm were absent from MPS samples from January, and whilst many individuals above 20 mm had synthesised high sperm loads, many of the largest P. elegans specimens in the nets (40-51 mm) still lacked well developed seminal receptacles, suggesting that they were not fully mature (Russell 1932, Choe et al. 2003). Grigor et al. (2014) frequently observed pronounced receptacles in individuals ≥20 mm from February to May, after which they disappeared. In the Arctic, breeding of P. elegans generally takes place in spring and summer (Kramp 1939, Ussing 1939, Bogorov 1940, Grigor et al. 2014); this is in contrast to Eukrohnia hamata, which may also reproduce in winter (pers. obs.). Saito & Kiørboe (2001) showed that P. elegans <5 mm in the North Sea fed almost exclusively on prey <350 μm in length. In Svalbard, the prey in this size range is represented by small cyclopoids such as Oithona similis, as well as Calanus nauplii and the young Calanus stages. In winter, individuals may be waiting for increased food input before they reproduce or for suitable food for the young and newborns to feed on (in terms of size and availability). By investing energy into maturation during winter, egg hatching can be scheduled to coincide with the reproduction of a wide range of copepod prey in spring and summer, and the buoyancy of chaetognath eggs may also allow them to hatch in shallow waters (Hagen 1999 and references therein) amongst their grazing prey. In another fjord on Svalbard, large numbers of P. elegans eggs became visible in the water column in March 2003, suggesting that conditions for reproduction begin to improve at this time of year (Hirche & Kosobokova 2011).

2.7 Concluding remarks Knowledge of an organism’s activity level and foraging ecology outside of the windows of the main primary production period (spring–summer) is key to establishing an understanding of the full annual routine (Varpe 2012). Here, we have focused on the polar night feeding ecology of a predatory zooplankter, adding to recent work on the life history and vertical distribution (Grigor et al. 2014). We found the chaetognath Parasagitta elegans to feed during the polar night, but with feeding rates lower in winter than have been reported from other seasons. As growth rates (e.g. Dunbar 1962, Grigor et al. 2014) may also decrease at this time and reproduction does not seem to occur, individuals should have lower-energy requirements and require less food. Although several copepod taxa may rest at depth in an unalert state during the Arctic winter, our study shows that mortality on copepods caused by P. elegans in the water column is low at this time of year. This study did, however, not sample the zone immediately above the seafloor, the hyperbenthic zone, where high densities of chaetognaths are known to aggregate during winter (Choe & Deibel 2000), and where abundances of resting copepods may also peak. The activities of P. elegans in this zone, as well as those of the other Arctic species Eukrohnia hamata and Pseudosagitta maxima, require more attention.

3. Chapter 3 – Growth and reproduction of the chaetognaths Eukrohnia hamata and Parasagitta elegans in the Canadian Arctic Ocean: capital breeding versus income breeding

3.1 Résumé

Dans les mers arctiques, la production primaire et l’accessibilité de la nourriture du zooplancton varient fortement durant la courte période estivale. Nous avons testé l'hypothèse selon laquelle Eukrohnia hamata et Parasagitta elegans, deux chaetognathes arctiques similaires et sympatriques, partitionnent les ressources selon différentes stratégies de reproduction. Les deux espèces avaient des longévités naturelles similaires d'environ 2 ans. Espèce mésopélagique, E. hamata engendre deux couvées distinctes à l'automne et au printemps. La production de jeunes coïncide avec une baisse de la fréquence d’E. hamata présentant des réserves lipidiques visibles, ce qui est caractéristique d’une reproduction basée sur les réserves. La croissance s’est avérée positive d'avril à janvier et négative en février- mars. La croissance et la maturation étaient similaires pour les deux cohortes. Les réserves lipidiques contenues dans une vacuole permettent à E. hamata de se reproduire et de croitre en dehors de la courte saison de productivité. Espèce plus néritique, P. elegans produit une couvée en été-automne, durant la production biologique maximale dans les eaux près de la surface, un mode de reproduction basé sur l’apport immédiat d’énergie. Cependant, avec le réchauffement climatique réchauffe, un bloom phytoplanctonique automnal pourrait favoriser la couvée d'été-automne chez P. elegans.

3.2 Abstract

In Arctic seas, primary production and the availability of food to the zooplankton are strongly pulsed over the short productive summer. We tested the hypothesis that Eukrohnia hamata and Parasagitta elegans, two similar and sympatric arctic chaetognaths, partition resources through different reproductive strategies. The two species had similar natural longevities of around 2 years. The meso-pelagic E. hamata spawned two distinct broods in autumn and spring. Offspring production coincided with drops in the frequency of E. hamata with visible lipid reserves, characteristic of capital breeders. Growth was positive from April to January and negative in February and March. Growth and maturation were similar for the two broods. Storage reserves contained in an oil vacuole may allow E. hamata to reproduce and grow outside the short production season. The neritic P. elegans produced one brood in summer- autumn during peak production in near-surface waters, characteristic of income breeders. However, as the Arctic warms, the development of an autumn phytoplankton bloom could favour the summer-autumn brood of P. elegans.

3.3 Introduction Arrow worms or chaetognaths form a phylum of gelatinous zooplankton represented by ~200 species worldwide, with biomass estimated at 10-30 % that of copepods (Bone et al. 1991). They can represent a significant fraction of the diet of many animals such as amphipods (e.g. Gibbons et al. 1992), other chaetognaths (e.g. Pearre 1982), seabirds (e.g. Mehlum and Gabrielsen 1993), and are the preferred prey of the larvae of some tropical fish and commercially important decapods (e.g. Sampey et al. 2007, Saunders et al. 2012).

Chaetognaths are hermaphrodites. Spermatogonia bud off from testes parallel to the tail walls to produce spermatocytes, which are transmitted to a conspecific to fertilise its oocytes (Alvarino 1992, Bergey et al. 1994). Each chaetognath possesses one pair of seminal vesicles on the tail segment and one pair of seminal receptacles at the posterior ends of the oviducts. Two individuals first come into contact with each other, before sperm is passed from the seminal vesicles of one individual to the seminal receptacles of the other (Goto & Yoshida 1985). Hatching strategies vary amongst species. For instance, Parasagitta elegans releases buoyant eggs that hatch near the surface where small prey is abundant (Kotori 1975, Hagen 1985). Eukrohnia hamata broods its eggs in an external sac from which the young later swim out, often at depth (Alvarino 1968 and references therein).

In contrast to the high diversity of chaetognaths in other seas (e.g. De Souza et al. 2014), only three species are frequently reported in Arctic plankton surveys. Parasagitta elegans is a neritic species, often peaking in abundance in epi-pelagic waters; Eukrohnia hamata is abundant in meso-pelagic and deep waters; and Pseudosagitta maxima, growing much larger (up to 90 mm) than the other species (40-45 mm), is typically bathy-pelagic but may also occur near the surface in the Arctic (Bieri 1959, Alvarino 1964, Terazaki & Miller 1986, Sameoto 1987). A fourth species, Heterokrohnia involucrum has also been recorded at bathy- pelagic depths, so far only in the Arctic Ocean (Dawson 1968). Whilst their contribution to higher trophic levels is poorly documented, arctic chaetognaths may consume an important fraction of copepod biomass. Although admittedly little better than a guess, Welch et al.

(1992) estimate that chaetognaths in Lancaster Sound ingest 51 % of the copepod biomass annually (164 of 319 kJ m-2 yr-1).

At high arctic latitudes, resource availability is strongly pulsed seasonally, with maximum phytoplankton and zooplankton biomass occurring during the short ice-free season in late summer and autumn (e.g. Falk-Petersen et al. 2007, Falk-Petersen et al. 2009, Søreide et al. 2010). Here, the co-existence of similar species may depend on the seasonal partitioning of resources, which can be achieved through differences in vertical distribution, diet and reproductive strategies. For instance, the reproductive strategies of the two dominant arctic herbivorous copepods differ markedly in ice-covered waters. The large capital-breeder Calanus hyperboreus reproduces in winter, fuelling egg production with lipids accumulated in summer (e.g. Conover 1967, Hirche & Niehoff 1996, Pasternak et al. 2001). Calanus glacialis may combine lipid stores and high-quality food to fuel reproduction (e.g. Conover 1988, Sainmont et al. 2014a). Income breeders do not synthesise reserves as fuel for reproduction at a later time. Peak reproduction occurs in spring-summer in the epi-pelagic Parasagitta elegans (e.g. Kramp 1939, Sameoto 1971, Grigor et al. 2014), which suggests income breeding with the summer maximum in prey abundance fuelling reproduction. By contrast, year-round breeding in Gerlache Strait, Antarctica (Øresland 1995), and the accumulation of lipids and wax esters in an oil vacuole (Pond 2012) suggest capital breeding fuelled by lipid reserves in E. hamata.

We describe the life histories of Eukrohnia hamata and Parasagitta elegans over an annual cycle based on weekly sampling of zooplankton in the Amundsen Gulf (Beaufort Sea) from August 2007 to July 2008, complemented by collections from other areas of the Canadian Arctic in autumn of the two years. In particular, growth of size cohorts, reproduction, and oil reserves are tracked to explore the hypothesis that E. hamata is a capital breeder that spawns during most of the year, whilst P. elegans is an income breeder, reproducing during periods of high prey availability.

3.4 Method 3.4.1 Study area Sampling was conducted on-board the CCGS Amundsen, primarily from October 2007 to August 2008 in the Beaufort Sea (Figure 3.1). To complete the annual cycle of observations, additional collections from the Labrador Sea, the Canadian archipelago and Baffin Bay were included in the analyses (Figure 3.1). In addition to the 27 stations sampled at weekly or higher resolution in the Amundsen Gulf region (69-72°N, 121-131°W) from November 2007 to August 2008, two stations were sampled in Nachvak Fjord in August 2007, two stations in Parry Channel in October 2007, and two stations in northern Baffin Bay in September 2008 (Figure 3.1 and Appendix C). Amundsen Gulf, our main sampling region, is a 400 km long, 170 km wide channel with a maximum bottom depth of ~630 m that connects the south- eastern part of the Beaufort Sea to the Canadian archipelago. Amundsen Gulf is typically covered with sea-ice from October to early June (Barber & Hanesiak 2004). Three water masses are detected in the Amundsen Gulf (Geoffroy et al. 2011 and references therein); the Pacific Mixed layer (PML; 0-60 m), the Pacific Halocline (PH; 60-200 m) and the Atlantic layer (AL; ˃ 200 m). Nachvak Fjord is a pristine 45-km long sill-fjord in Nunatsiavut, Northern Labrador, with a maximum depth of 210 m (Bell & Josenhans 1997, Simo-Matchim et al. 2016). The deep semi-enclosed basin of Baffin Bay lies further north, and is connected via Parry Channel to the Beaufort Sea in the west (Figure 3.1).

3.4.2 Sampling In the Amundsen Gulf, estimates of weekly profiles of temperature and salinity between November 2007 and July 2008 are from Geoffroy et al. (2011), and those of chlorophyll a (chl a) biomass are from Darnis & Fortier (2014). Salinity, and temperature profiles of the water column at single stations in Nachvak Fjord, Parry Channel and Baffin Bay, were obtained using a conductivity-temperature-depth (CTD) system. We also collected seawater samples from 9-10 depths in the upper 100 m, using the CTD rosette equipped with twenty- four 12 1 Niskin-type bottles (OceanTest Equipment). Subsamples (500 ml) for the determination of total chlorophyll a (chl a) were filtered onto 25 mm Whatman GF/F filters

(nominal porosity of 0.7 µm). Chl a was measured using a Turner Designs 10-AU fluorometer, after 24 h extraction in 90 % acetone at 4 °C in the dark without grinding (acidification method of Parsons et al. 1984).

Two similar samplers were used to collect zooplankton: a square-conical net with a 1 m2 opening area, 200 µm mesh and a rigid cod-end, and a Hydrobios® Multinet with 9 individual opening and closing nets with 0.5 m2 openings, 200 µm mesh and rigid cod ends. At 12 stations, the 1 m2 net was deployed cod-end first (non-filtering) to 10 m above the seabed and hauled back to the surface at a speed of 0.5 m s-1. At 22 stations, the Multinet was deployed to 10 m above the seabed and retrieved at 0.5 m s-1, sampling sequentially over pre- set depth intervals, the overall thickness of which varied from 100 to 520 m. Typically, 3×20- m depth strata were sampled from 10 m above the seafloor upward, and 3×20-m depth strata were sampled from 60 m depth to the surface. The remaining interval (from 70 m above the bottom to 60 m below the surface) was then divided into three equal sampling layers (Appendix C). Samples were fixed in 4 % formaldehyde-seawater solution buffered with sodium borate.

3.4.3 Chaetognath body size and sampler efficiency Chaetognaths were sorted from the samples, identified to species and when possible, measured (nearest mm) from the top of the head to the tip of the tail, excluding the caudal fin. Heavily damaged or broken individuals were excluded, but those missing heads only (~1 mm) were included in the analyses, as positive identification remained possible (Figure 3.2) and length could be estimated. Out of 8179 chaetognaths sorted, 7245 chaetognaths (5918 Eukrohnia hamata and 1298 Parasagitta elegans were included. Twenty-nine (29) Pseudosagitta maxima present in the samples were not analysed further. The relative efficiency of the Multinet and square-conical net at capturing chaetognaths was assessed by comparing the size frequency distribution of E. hamata or P. elegans in the two nets.

3.4.4 Hatching and cohort development The occurrence of newborns (2-4 mm) in Multinet collections was used as an index of recent hatching. The subsequent growth of the different cohorts was inferred visually from successive monthly length frequency distributions in collections from both samplers. Lengths from the Multinet collections were included twice in the histograms to give equal weight to this sampler (0.5 m2 aperture) and the square-conical net (1 m2 aperture). Each cohort was detected by visual means, and the MIXdist library in RTM (Du 2002) was used to estimate its mean length (arithmetic mean), standard deviation and length range in each month. Monthly

growth rates of each cohort were calculated from changes in mean length. In most cases, a cohort was clearly defined in the month length frequency distributions. In some cases, the interpretation of cohorts was aided by two assumptions. First, based on published estimates of growth (Welch et al. 1996, Choe & Deibel 2000, Grigor et al. 2014), we assumed that monthly growth of a cohort could not exceed 7 mm mo-1. Second, the possibility that a cohort was absent from the population in one month if clearly present during the preceding or following month in the time series was excluded. Solutions aimed to minimise cohort length overlaps. Goodness-of-fit was tested using the Chi-square statistic. Parasagitta elegans data for August are from Nachvak Fjord in 2007 due to the scarcity of this species in the Amundsen Gulf collections in that month in 2008.

3.4.5 Estimation of maturity and oil vacuole area Eukrohnia hamata and Parasagitta elegans were stained with a Borax Carmine solution to highlight gonads (Pierce 1941, Grigor et al. 2014). Oocytes (Figure 3.2) in single ovaries were counted and diameters measured under the stereomicroscope. Ovary length, diameter of the externally protruding part of the seminal receptacles and width of the seminal vesicles (Figure 3.2) were measured to the nearest 0.01 mm. The distribution of spermatogonia/spermatocytes in tails was characterised as absent, spermatogonia present near walls only, or spermatocytes dispersed throughout the tail. The width of seminal vesicles was measured to the nearest 0.01 mm. The presence or absence of an oil vacuole (Figure 3.2) was noted in 2089 individuals of E. hamata. The width and height of the vacuole were measured to the nearest 0.01 mm under the stereomicroscope. Area was calculated as π×0.5width×0.5height. The presence of oil in the vacuole and/or oil droplets in the gut was noted.

3.4.6 Vertical distributions 19 Multinet collections (typically 9 samples per haul), 1-2 per month from August 2007 to September 2008, were analysed to estimate the abundances of Eukrohnia hamata and Parasagitta elegans populations and age classes in discrete layers of the water column (ind. m-3). Contour plots were produced in the R computing environment (R Development Core

Team 2008), using Akima bilinear interpolation performed vertically for every meter of the water column. Assuming a lognormal distribution, species abundances A were log- transformed as log10(A+1) prior to interpolation, also serving to preserve abundance A = 0 (Legendre & Legendre 2012). Vertical distributions of populations were also characterized by their weighted mean depths in the water column [Zm, m, Eq. (3.1a)] with standard deviations [Zs, m, Eq. (3.1b)] as described by Dupont & Aksnes (2012). Zm was normalized between 0 and 1 to identify the positions of animals relative to a station’s bottom depth.

푛 푛 푍푚 = ∑푖=1 푤푖 푑푖푧푖/ ∑푖=1 푑푖 푧 (3.1a)

푛 2 푛 2 푍푠 = √(∑푖=1 푤푑1 푧1 / ∑푖=1 푑1푧1) − 푊푀퐷 (3.1b)

Where wi is the abundance of animals in depth interval i, di is the thickness of interval i, and zi is its mid-point.

3.5 Results 3.5.1 Physical environment and primary production in the Amundsen Gulf From November 2007, onwards, the PML in the Amundsen Gulf had temperatures of -1.6 to -0.6 ºC, but warmed to ˃ 9 ºC at some stations in June and July 2008 on top of a strong thermocline. Throughout the study, temperatures in the PH were -1.6 to 0 ºC, and temperatures in the AL were ~0 ºC. Salinities were 30-32 psu in the PML, 32-34.6 psu in the PH, and ˃ 34.6 psu in the AL (Geoffroy et al. 2011). Ice algae was first detected in late March in the PML and peaked in late April and early May (~10 mg m-3), when it was succeeded by a surface phytoplankton bloom. In July, a subsurface chl a maximum (SCM) occurred at ~40 m (16 mg m-3), and the phytoplankton penetrated the PH (Darnis & Fortier 2014).

3.5.2 Physical environment and primary production in autumn (other sampling locations) In August 2007, a steep thermocline was observed in Nachvak Fjord between the surface and 60 m, with temperature decreasing from 3.4 °C to -1.6 °C (Appendix D). Chl a concentrations peaked at 20 m (2.8 µg L-1). In October 2007, temperatures at our Parry Channel station rose

from ˂-1 °C at the surface to 0.4 °C at ~45 m, fell to ˂ 1 °C at 60 m, and then rose again from -1.5 °C at 200 m to 0.7 °C at 500 m. Chl a concentrations remained below 0.5 µg L-1 at sampled depths. In September 2008, temperatures at our Baffin Bay station rose from ˂-1 °C at the surface to ˃ 0.3 °C at ~85 m, fell to ˂ 1.1 °C at 100 m, with smaller fluctuations in temperature in deeper waters. Chl a concentrations peaked at 45 m (1.1 µg L-1). Strong haloclines were observed in the upper 100 m at all three stations (Appendix D).

3.5.3 Abundances and vertical distributions of chaetognath species Eukrohnia hamata was present at all stations in the Amundsen Gulf with abundances ranging from 56 ind. m-2 in late July to 894 ind. m-2 in mid-April, based on Multinet collections. Parasagitta elegans was less abundant than E. hamata at all stations in Amundsen Gulf (16 ind. m-2 in November to 244 ind. m-2 in December), and in Parry Channel (320 ind. m-2) and Baffin Bay (64 ind. m-2) in October and September respectively. Pseudosagitta maxima occurred in only 4 Multinet collections from the Amundsen Gulf, typically in the Atlantic Layer, with maximum abundances of 4 ind. m-2. Relative to P. elegans, the abundance of E. hamata increased with station depth (Figure 3.3a). E. hamata occurred primarily in the deep AL of the Amundsen Gulf (˃ 200 m, Figure 3.3b), based on weighted mean depth, except on two dates in late May and late July when the E. hamata population was centred in the PH (60-200 m). In Parry Channel (October) and Baffin Bay (September), the E. hamata population remained ˃ 200 m. In the Amundsen Gulf, seasonal vertical migration (SVM) was most pronounced in P. elegans, which, based on weighed mean depth, resided primarily in the PH from December to March, rose into the PML (0-60 m) in March/April during the development of ice algae, and returned to the PH in late June (Figure 3.3b). P. elegans was 39 times more abundant than E. hamata in Nachvak Fjord in August (462 ind. m-2). At Nachvak Fjord, Parry Channel and Baffin Bay, the P. elegans population remained ˂ 200 m (Figure 3.3b).

Figure 3.3 a) Relative frequency of Eukrohnia hamata, Parasagitta elegans and Pseudosagitta maxima in relation to bottom depth at Amundsen Gulf stations. Multinet and square-conical net collections included. b) Weighted mean depths of E. hamata and P. elegans normalized relative to the bottom depth at Amundsen Gulf stations (see ‘Method’). Multinet collections. Standard deviation bars also given. The blue area indicates the Pacific Halocline (60-200 m) and the red area indicates the Atlantic Layer ˃ 200 m (Geoffroy et al. 2011).

3.5.4 Length distributions and sampler efficiency Eukrohnia hamata in the Amundsen Gulf ranged in length from 2 to 40 mm (≤34 mm in other sampling regions) and Parasagitta elegans from 2 to 42 mm (up to 45 mm in other sampling regions). The length frequency distributions of each species were not significantly different (Kolmogorov-Smirnov test, P ˃ 0.1) between the two samplers (Figure 3.4). The few E. hamata and P. elegans ≥ 40 mm were caught solely by the larger-aperture square- conical net.

3.5.5 Timing of reproduction The hatching seasons of the two chaetognath species were approximately complementary over the annual cycle (Figure 3.5). Apart from the occurrence of newborns in October, Eukrohnia hamata hatched from December to July with the production of newborns ramping up from low values in December-January to a maximum in April, and stopping after July (Figure 3.5). Except for their absence in our collections in November and January, Parasagitta elegans newborns were observed continuously from July to February, with hatching peaking in December (Figure 3.5).

3.5.6 Eukrohnia hamata length cohorts and life cycle Four cohorts were distinguished in all months except August and September 2008, when there were three (Figure 3.6). The cohort of largest/oldest chaetognaths was labelled E1 (Eukrohnia 1), the intermediate-sized cohorts E2 and E3, and the cohort of autumn (October) newborns E4. The E1 cohort of large chaetognaths increased little in length from October to March and disappeared from the collections in April. The length-frequency distributions of intermediate cohorts E2 and E3 shifted to larger length quasi-monotonically throughout the annual cycle from October to September. Cohort E2 disappeared from the collections in August. Initially, the length frequency distribution of cohort E4 progressed slowly to longer lengths or even regressed as some limited numbers of newborns were added to the cohort

starting in December. A new cohort E5 hatched in April (Figure 3.6). Cohort E5 hardly increased in average length from April to July, as newly hatched chaetognaths were constantly added to it, but the width of the adjusted Gaussian distribution increased (declining kurtosis).

labelled with a capital E and a number from oldest (1) to youngest (5). Blue line is the total distribution obtained by summing the modelled distributions. Chi-square values for the goodness-of-fit of the total distribution to the data are given. Sampling regions are abbreviated in each panel: PC, Parry Channel; AG, Amundsen Gulf; BB, Baffin Bay. k is the number of collections and n the number of length measurements included (see ‘Method’).

A similar pattern of annual growth was observed for cohorts E1 to E5 of Eukrohnia hamata (Figure 3.7). After a period of latency of 3-4 months, the mean length of cohorts E4 and E5 of newly hatched E. hamata started to increase (Figure 3.7a). Apart from this initial latency in younger animals, cohorts E2-E4 increased more or less regularly in mean length from March to June. The mean length of all cohorts declined from January to February and this decline persisted until March in cohorts E3 and E4 (Figure 3.7a).

The average length of cohort E4 and E5 in September corresponded to the average length of cohort E2 and E3 in October respectively (Figure 3.7a), strongly suggesting the co-existence of an autumn brood (October) and a spring brood (April) of Eukrohnia hamata. Connecting the corresponding cohorts created a composite of the growth trajectory of the assumed autumn and spring broods (Figure 3.7b). In all months, the standard deviations around average length of the presumed autumn and spring broods were distinct. The E1 cohort of large chaetognaths likely represented the final months of life of the spring brood (Figure 3.7b). The growth trajectory of the autumn brood terminated after 22 months with the disappearance of cohort E2 from the collections. The growth trajectory of the spring brood indicated a slightly longer life cycle of two years (24 months). Mean length at a given age did not differ significantly (t-test, t = 43 df = -0.36, P = 0.72) between the autumn and spring broods (Figure 3.7c). For the autumn brood, growth was significantly faster in the second year of life than in the first (ANCOVA, P ˂ 0.01). In the spring brood, growth rates were not significantly different between years one and two (ANCOVA, P ˃ 0.05).

The autumn and spring broods of Eukrohnia hamata matured in different seasons (Table 3.1) but at similar body lengths (Figure 3.8). In the autumn brood, all variables peaked in the second year of life (16 to 23 months of age) from the winter (January-March) to July (Table

3.1). Maturity was somewhat less protracted in the spring brood, with ovary length, width of seminal vesicles and diameter of the protruding seminal receptacles first peaking in August of the second year of life at 17 months of age (Table 3.1). Maturity indices and oil vacuole area peaked at different times from November (age 20 months) to March (age 24 months). In both broods, oocyte number, ovary length and oil vacuole area started to increase by the end of the first year of life (Figure 3.8a, Figure 3.8b and Figure 3.8c). Some chaetognaths presented dispersed sperm in the tail during their first year, but sperm frequency started to increase systematically around the 21th month of life (Figure 3.8d). Seminal vesicle width, and sperm receptacle diameter began to develop by the 21th month of life as well (Figure 3.8e and Figure 3.8f).

Characteristic Month(s) of peak maturity (age in months) E. hamata E. hamata P. elegans

autumn brood spring brood summer brood Number of oocytes Jan-Jul (16-23) Mar (24) Feb (20) Length of ovaries Mar-Jul (18-23) Aug (17), Jan (22) Jan-Aug (19-26) Area of oil vacuole Mar-Jul (18-23) Dec-Jan (21-22) - Fraction of inds. with dispersed sperm Mar-Jul (18-23) Nov (20) Feb-Mar (20-21), Aug (26) Width of seminal vesicles Feb-Jul (17-23) Aug (17), Feb (23) Aug (14), Aug (26) Diameter of protruding seminal receptacles Mar-Jul (18-23) Aug (17), Feb (23) Jan (19)

A vacuole was observed in 99 % of Eukrohnia hamata (n = 2089), the exception being in newborns 2-3 mm long from the spring brood. Overall, oil was observed in the vacuole or the digestive tract of 72 % of the E. hamata examined. The fraction of the population with oil in the vacuole or the body cavity peaked in February (85 %), declined slowly from February to April, then faster to a minimum in June (54 %) before peaking again in September (Figure 3.9). Low percentages were observed in October (72 %) and December (51 %).

Figure 3.9 Mean frequency (± 1 SD) of Eukrohnia hamata with oil in vacuole and/or digestive tract by months. Number of water column collections shown inside bar (33 – 446 individuals per haul).

In the Amundsen Gulf, individuals that rose into the PML during summer were mostly second-year members of the autumn brood and first-year members of the spring-brood (Figure 3.10).

3.5.7 Parasagitta elegans length cohorts and life cycle Two cohorts were distinguished in all months except August 2007 and July 2008, when there were three (Figure 3.11). The cohort of largest/oldest chaetognaths was labelled P1 (Parasagitta 1), the intermediate-sized cohort P2, and the cohort of summer (July) newborns P3. Cohort P1 (monthly mean lengths >32 mm) disappeared from the collections by September. The length-distribution of P2 (monthly mean lengths = 21-33 mm) shifted to larger sizes from July to September and then remained constant until April. That of P3

(monthly mean lengths = 5-19 mm) increased from July to November, remained constant or shifted to smaller sizes from December to April, and then increased again starting in May (Figure 3.11). A decrease in the length of cohort P3 in December may not be realistic, because P. elegans reproduced in December (Figure 3.5). However, any winter cohort produced was not detected in the size-frequency distributions from subsequent months (Figure 3.11).

goodness-of-fit of the total distribution to the data are given. Sampling regions are abbreviated in each panel: NF, Nachvak Fjord; PC, Parry Channel; AG, Amundsen Gulf; BB, Baffin Bay. k is the number of collections and n the number of length measurements included (see ‘Method’).

The growth trajectories of the three cohorts over the annual cycle indicated relatively fast growth from April to September/October followed by stagnation over the winter months and even a slight regression in length from January to April (Figure 3.12a). Assembling the growth trajectories of the individual cohorts of Parasagitta elegans (Figure 3.12a) revealed a single summer brood, which terminated at 2.2 years with the disappearance of cohort P1 from the collections (Figure 3.12b).

Figure 3.12 a) Monthly mean length (± 1 SD) of the three cohorts of Parasagitta elegans starting from a major birth month; b) Composite growth trajectory of the single brood.

Figure 3.13 Development of sexual features (a, oocyte number; b, ovary length; c, sperm load; d, seminal vesicle width; e, seminal receptacle diameter) with length in Parasagitta elegans. Circles show mean values of each characteristic (± 1 SD shown as ribbons). Vertical line indicates average length (20.7 mm) at one year of age. Maturity results were obtained from the analyses of up to 63 individuals from each 1 mm length class.

In Parasagitta elegans, oocyte number, ovary length, the occurrence of dispersed sperm, and seminal receptacle diameter first peaked in January or February at ages 19 or 20 months (Table 3.1). Occurrence of sperm peaked again in August (26 months). The width of seminal

vesicles reached a maximum in August early in the second year of life (14 months) and again in August at 26 months of age (Table 3.1). As P. elegans increased in length, the presence of oocytes and perceptible ovaries were the first sign of maturation, both indices starting to increase at 7 mm (Figure 3.13a and Figure 3.13b). Dispersed sperm was first detected around 8 mm and was present in nearly 100 % of individuals >35 mm (Figure 3.13c). Seminal vesicles and receptacles started to develop around 17 mm in length (Figure 3.13d and Figure 3.13e).

In the Amundsen Gulf, similar SVM behaviour was detected in first- and second-year individuals. Based on weighted mean depths, the typically resided in the PH from December to March, ascended in February, were distributed between the PML and the PH from March to November, and descended in December (Figure 3.14).

3.6 Discussion 3.6.1 Chaetognath cohort interpretation and lifespans Terazaki & Miller (1986) reported three separate broods of Eukrohnia hamata (spring, summer and autumn) in the north Pacific with lifespans of 8-10 months. Two-year lifespans were suggested from the existence of two distinct size cohorts in summer in northern Norway (Sands 1980), and Baffin Bay (Sameoto 1987). In the present study, E. hamata autumn and spring broods (staggered by 6 months) also had lifespans of 2 years.

Parasagitta elegans lifespans in Arctic waters ranged from one year in a land-locked fjord (McLaren 1961), to three years in the Barents Sea (Falkenhaug 1993) and Svalbard (Grigor et al. 2014). Two-year lifespans were suggested in Disko Bay (Dunbar 1940) and Lancaster Sound (Welch et al. 1996). In this study, a single annual summer-autumn brood had two-year life cycles in the Amundsen Gulf, Parry Channel and Baffin Bay. In the Amundsen Gulf, a possible fourth cohort that contained offspring in late December may have quickly perished in the lower part of the Pacific Halocline. McLaren (1969) found that first-feeding P. elegans preyed upon Pseudocalanus nauplii. Darnis & Fortier (2014) reported that Pseudocalanus spp. CI resided relatively deeper in the PH/AL in December compared to January. Admittedly, cohort interpretation for P. elegans could also have been hampered by relatively low sample sizes. In Nachvak Fjord (August 2007), the capture of several individuals up to 45 mm long could indicate a slightly longer lifespan there, which could complicate using these data to fill in missing data from the Amundsen Gulf time series. Differences could be due to lower growth rates, genetic differences or different feeding strategies between populations (Pearre 1991, Øresland 1995). The relatively shallow waters of Nachvak Fjord also contained low numbers of Eukrohnia hamata, which could reduce competition for resources.

3.6.2 Resource partitioning in the sympatric Eukrohnia hamata and Parasagitta elegans In arctic seas where many planktonic species depend on a single brief annual bout of primary production in summer, morphologically similar species can co-occur thanks to subtle differences in the ecological niches occupied. For instance, in the Beaufort Sea the morphologically identical larvae and juveniles of polar cod (Boreogadus saida) and ice cod

(Arctogadus glacialis) share the same spatio-temporal distribution, hatching season and growth rate, but a slightly larger size at hatch results in A. glacialis being longer and feeding on different prey at a given date than B. saida (Bouchard et al. 2016). As well, the herbivorous calanoid copepod Calanus glacialis is morphologically almost identical to its sympatric congener C. hyperboreus except for a smaller size starting at the copepodite IV developmental stage. Within their sympatric distribution, C. glacialis is preferentially distributed over the shelf while C. hyperboreus concentrates over the slope and deeper basin (Darnis et al. 2008 and references therein). The two species also differ in their vertical distribution and ontogenetic migration (Darnis & Fortier 2014).

Our results support the notion that the morphologically similar and sympatric chaetognaths Eukrohnia hamata and Parasagitta elegans also partition resources in arctic seas. First, although the two species generally overlapped in time and space, observations confirm the more pelagic distribution of E. hamata relative to the neritic P. elegans (Figure 3.3) as reported in previous studies (Hopcroft et al. 2005, Kosobokova & Hopcroft 2009). Darnis et al. (2008) identified three zooplankton assemblages in southeastern Beaufort Sea: the Shelf (43-182 m depth range), Polynya (250-537 m) and Slope (435-1080 m) assemblages. The pelagic E. hamata thus associated with the Polynya assemblage dominated by Calanus hyperboreus and the Slope assemblage where omnivores and carnivores are abundant (Darnis et al. 2008). The neritic P. elegans preferentially associated with the Shelf assemblage, which is strongly dominated by the neritic copepods Pseudocalanus spp. (Darnis et al. 2008). Interestingly, McLaren (1969) found that first-feeding P. elegans preyed upon Pseudocalanus nauplii.

Second, the two species produced offspring in mostly opposite seasons, Eukrohnia hamata mainly from February to July with a second brood in the autumn (October, December); and Parasagitta elegans from July to February (Figure 3.5). Consistent with our observations, E. hamata breeds mostly in spring and summer in the waters off west Greenland and northern Norway, with reports of continuous low reproduction rates during autumn and winter in west Greenland and Gerlache Strait, Antarctica (Kramp 1939, Sands 1980, Øresland 1995). Eukrohnia young are retained in the folded lateral fins of the adults and released at meso-

bathy-pelagic depths (Alvarino 1968, Kruse 2009, Terazaki et al. 2013). Emergence at meso- pelagic levels in the Amundsen Gulf suggests that they may feed on sinking particles, protozoans, and copepod nauplii present at depth (Makabe et al. 2016), and gut content observations in 2007 and 2008 in the same region suggest that E. hamata ingested substantial amounts of large, amorphous macroaggregates (marine snow) throughout the year, while P. elegans did not feed in this way (unpublished results).

Summer-autumn reproduction in Parasagitta elegans also agrees with previous observations in Arctic waters (e.g. Kramp 1939, Dunbar 1940, Grainger 1959, McLaren 1961, Sameoto 1971, Grigor et al. 2014). A similar breeding season from July to February in the Canadian sub-Arctic led Dunbar (1962) to propose that the breeding season of P. elegans is determined by temperature rather than food availability. However, the epi-pelagic hatching of the buoyant eggs of P. elegans (Hagen 1985) matches the emergence of the nauplii and CI of the copepods Pseudocalanus spp. and Oithona similis in the surface layer from July to January (Darnis & Fortier 2014), potentially ensuring suitably-sized prey (Saito & Kiørboe 2001) to newborns for the length of the hatching period. At body lengths ˂ 15 mm, first-year P. elegans participated in the autumn/winter migrations from the PML to the PH (Figure 3.14), possibly in pursuit of several species of seasonally migrating copepods (Darnis & Fortier 2014).

We suggest that the generally asynchronous reproduction and depth distributions of Eukrohnia hamata and Parasagitta elegans may reduce potential competition for food during the early life of arctic chaetognaths.

3.6.3 The potential role of lipid reserves: contrasting growth in the two species In Eukrohnia hamata, a large fraction (72 %) of total lipids are in the form of the long-term energy stores steryl esters and wax (Connelly et al. 2012). The lipids contained in the oil vacuole could have roles in growth, reproduction and/or buoyancy (e.g. Kruse et al. 2010, Pond 2012, Grigor et al. 2015), although Øresland (1990) doubted a storage role from their small size. By comparison, lipids in Parasagitta elegans contain low amounts of wax and

steryl esters (˂1-8 % of total lipids), and short-term lipid stores such as triacylglycerol are present only in low to moderate amounts (Falk-Petersen et al. 1987, Lee et al. 2006, Connelly et al. 2012, Connelly et al. 2016). Lipids gained from copepod prey by P. elegans during spring-summer feeding may be rapidly utilized for reproduction (Choe et al. 2003).

Some studies have reported reduced growth in winter in arctic and subarctic chaetognaths (Dunbar 1962, Grigor et al. 2014), and others have not (Welch et al. 1996). Slow winter growth has been attributed to poor feeding as indicated by gut content analyses (e.g. Øresland 1987, Falkenhaug 1991, Grigor et al. 2015). In the present study, both Eukrohnia hamata and Parasagitta elegans exhibited slower growth in winter, but the annual growth pattern differed between the two species. In E. hamata growth was variable but positive over ten months from March to January (1.55 and 1.82 mm mo-1 on average for the autumn and spring broods respectively), followed by a 1 to 2-month period of negative growth in late winter (January to February-March, Figure 3.8b). In P. elegans, fast growth from April to September or November (2.1 mm mo-1 in year 1 and 3.7 mm mo-1 in year 2 of life), was followed by a long period of winter stagnation from September-November to April (Figure 3.12b).

The migration of some copepod prey to meso-pelagic depths in the autumn could decrease prey availability for epi-pelagic chaetognaths in winter (Hagen 1999). The seasonal vertical migrations in Parasagitta elegans suggest that conscious efforts were made to reduce starvation (Figure 3.3 and Figure 3.14), but if winter food access was a problem, arrested growth from late autumn to April in this species may reflect the absence of lipid stores and oil vacuole. On the other hand, the return of migrating copepods to epi-pelagic depths as early as February (Darnis & Fortier 2014) could decrease prey availability for Eukrohnia hamata in spring, and this species carried out relatively limited seasonal migrations (Figure 3.3 and Figure 3.10). Sullivan (1980) suggested that E. hamata requires less prey than P. elegans. The morphology, growth and vertical distribution of E. hamata suggests that long- term energy stores enable this species to limit starvation at depths ˃ 200 m to a brief interlude in late winter and spring. However, contradicting this interpretation, the frequency of E. hamata presenting oil reserves peaked from January to March during the months of negative growth in length, and started to decline in April-May (Figure 3.9) coinciding with peak

reproduction in April (Figure 3.5). The rebuilding of lipid reserve frequency from June to September was followed by drops in October and then December (Figure 3.9), which both coincided with short bouts of reproduction (Figure 3.5). The observed negative correlation between lipid frequency in the population and offspring production is consistent with the suggestion by Båmstedt (1978) that part of the lipid content of large specimens of E. hamata is drawn upon during reproduction, thus lowering the average individual lipid proportion in the population. We conclude that E. hamata invests its lipid reserves primarily into reproduction rather than growth, which may contribute to a loss of length in February and March when prey is likely scarce in deep waters.

3.6.4 Maturation Dispersed sperm in the tail early in the first year of life was the first sign of maturation in Eukrohnia hamata, confirming the protandric nature of the species (Terazaki & Miller 1982). All other maturation traits and the oil vacuole developed in the second year of life. In the North Pacific, only 3 days separated egg release from the appearance of juveniles (Kotori 1975). Accordingly, the successive maturation episodes in spring and autumn (Table 3.1) coincided with the occurrence of juvenile E. hamata in net collections. These two maturation cycles at 6-mo intervals could be interpreted as a sign of iteroparity but actually reflect the asynchronous maturation of the two broods, which supports the suggestion that E. hamata is semelparous and dies after a single reproduction effort in the second year of life (Kuhl 1938).

Dunbar (1940) and Grigor et al. (2014) suggested an age-at-maturity of one year for Parasagitta elegans in the Canadian and European Arctic respectively. Consistent with this, first-year P. elegans (3-20 mm) could not reproduce as they lacked seminal receptacles and vesicles. However, oocyte numbers and the frequency of dispersed sperm started to increase at 9-12 mm corresponding to an age of ca. 6 months. Contrary to Eukrohnia hamata, sexual maturation occurred from January to August in P. elegans, several months prior to the emergence of the annual brood from July to December.

3.6.5 Capital versus income breeding in a warming Arctic Capital breeding based on wax esters accumulated in summer to fuel reproduction in winter so as to match the emergence of offspring with the spring-summer maximum in biological production has been documented in several arctic zooplankton taxa (e.g. Kattner et al. 2007 for a review). Arctic cod (Boreogadus saida), a pivotal forage fish in the arctic pelagic food web, also spawn in winter with the buoyant eggs hatching in the surface layer before the onset of the spring-summer primary production (e.g. Bouchard & Fortier 2011). In the present study, Eukrohnia hamata relied on wax ester reserves to fuel reproduction. But contrary to other arctic capital breeders such as Calanus hyperboreus and B. saida that produce a single spring brood, E. hamata produced distinct broods in the autumn and the spring. Much of the reproduction occurred in deep waters (Figure 3.10). The comparison of growth trajectories and maturity at length shows a similar developmental rate in the two broods, which suggests that E. hamata is impervious to differences in temperature regime or food availability between the two seasons. The distribution of E. hamata in the meso-pelagic (Atlantic) layer where temperature is constant and where copepod prey resides for most of the year (Darnis & Fortier 2014) may explain the parallel existence of two sub-populations that hatch, develop, and reproduce separated by a 6-month time-lag.

The alternative strategy to capital breeding is income breeding in which reproduction is fueled by immediate food availability in spring-summer with the production of offspring coinciding with peak production in summer. By contrast to Eukrohnia hamata, the epi- pelagic Parasagitta elegans, which accumulates few lipid reserves, produced a single brood in summer, typical of an income breeder. This strategy, which increases in frequency towards lower latitudes (e.g. Kattner et al. 2007), is also used for instance by the Pacific sand lance (Ammodytes hexapterus) a recent subarctic invader of the Beaufort Sea (Falardeau et al. 2014). The end result of the remarkably different reproduction strategies of the two species is the staggering of three distinct broods of morphologically and ecologically similar chaetognaths: the E. hamata spring cohort in April, the P. elegans summer cohort in July, and the E. hamata autumn cohort in October. The sequestration of lipids thanks to the early development of an oil vacuole may enable young E. hamata at meso-pelagic depths to overcome potential food shortages in spring and autumn, while the only survival window for

the lipid-poor P. elegans would be during peak availability of prey in the surface layer in summer.

3.7 Conclusions Based on our observations from August 2007 to September 2008 in the Canadian Arctic, the processes of growth and reproduction are more seasonally-restricted in Parasagitta elegans compared to Eukrohnia hamata. Length and maturity data for the two species from other years would help to confirm if this is a general trend. As the Arctic warms and the ice-free season lengthens, wind mixing increasingly triggers a second phytoplankton bloom in the autumn, with potentially major impacts on arctic marine ecosystems (Ardyna et al. 2014). If the availability of suitable prey at the first-feeding stage is critical to the survival of young chaetognaths, we hypothesize that this autumn bloom would more directly and quickly benefit the income-breeding P. elegans as copepod prey in epi-pelagic waters would increase.

4. Chapter 4 – Feeding strategies of arctic chaetognaths: are they really “tigers of the plankton”?

4.1 Résumé

Les chaetognathes, aussi connus sous le nom de « tigres du plancton » jouent un rôle important dans les communautés mésozooplanctoniques, en termes d'abondance et de biomasse. Bien que traditionnellement considérés comme étant strictement carnivores, des études récentes suggèrent l’usage de stratégies non -carnivores au sein du phylum. Les stratégies d'alimentation des chaetognathes de l’Arctique sont particulièrement intéressants comptes tenus de la forte saisonnalité dans les abondances et les distributions de proies connues (copépodes). Le présent chapitre traiter des stratégies d'alimentation saisonnières de deux principales espèces de l’Arctique : Eukrohnia hamata de la zone mésopélagique, ainsi que Parasagitta elegans de la zone épipélagique. Le contenu du tube digestif de spécimens récoltés au printemps, en été et en hiver au sud-est de la mer de Beaufort suggère de faibles taux de prédation chez toutes les espèces (0-0,27 proie ind. d-1). Toutefois, les taux de prédation de E. hamata et de P. elegans étaient plus élevés au printemps-été par rapport à l'automne-hiver. E. hamata mange des macroagrégats ne contenant pas de proies (˃500 µm), probablement de la neige marine, tout au long de l'année. Ce mode d'alimentation précédemment non répertorié a été confirmé pour E. hamata par des observations d'alimentation in-vitro dans la mer de Chukchi à l'automne, mais n’a pas été observé chez P. elegans. Des quantités étonnamment élevées d’acides gras marqueurs de diatomées en automne dans les spécimens de E. hamata suggèrent fortement que les diatomées ont été consommées directement, tandis que des ratios inférieurs à 18:1 (n-9)/(n-7), et valeurs inférieures de δ15N et δ13C comparativement au P. elegans ont tous suggéré son régime alimentaire plus omnivore. Nous suggérons que par la consommation de neige marine, les pelotes fécales du E. hamata pourraient contribuer davantage à la séquestration du carbone qu'on ne le pensait auparavant.

4.2 Abstract

Chaetognaths, also known as the “tigers of the plankton” are important components of mesozooplankton communities, in terms of abundance and biomass. Although traditionally considered to be strict carnivores, recent studies suggest non-carnivorous strategies occur within the phylum. The feeding strategies of arctic chaetognaths are particularly interesting given the strong seasonality in the abundance of known food items, chiefly copepods. This chapter addresses the seasonal feeding strategies of two major Arctic species; the meso- pelagic Eukrohnia hamata and the epi-pelagic Parasagitta elegans. Gut contents of specimens collected in spring, summer and winter in the south-eastern Beaufort Sea suggested low predation rates in the two species (0 – 0.27 prey ind. d-1), but predation rates in E. hamata and P. elegans were higher in spring-summer compared to autumn-winter. E. hamata ate non-prey macroaggregates (˃ 500 µm, probably marine snow) throughout the year. This previously unreported feeding mode was confirmed for E. hamata by in-vitro feeding observations in the Chukchi Sea in the autumn. High levels of diatom fatty acid markers in E. hamata sampled in autumn, strongly suggested that diatoms were consumed directly, whilst lower 18:1 (n-9)/(n-7) ratios, δ15N and δ13C, compared to P. elegans confirmed its more omnivorous diet. We suggest that by containing algal-rich contents, the fecal pellets of E. hamata could contribute more to carbon sequestration than those of strict carnivores.

4.3 Introduction Chaetognaths, the so-called “tigers of the plankton” (Suthers et al. 2009), are semi-gelatinous zooplankters considered to be primary carnivores in the marine food web (Reeve 1970). Reported prey includes fish larvae, copepods and other chaetognaths (e.g. Alvarino 1965, Sullivan 1980, Brodeur & Terazaki 1999). In some locations, chaetognath populations could control copepod standing stocks (e.g. Sameoto 1972, Williams & Collins 1985). The fecal pellets of chaetognaths are said to be large and carbon-rich, and, considering the high abundances of chaetognaths in the ocean, these pellets could be important food for zooplankton in the water column or in the flux of carbon to the benthos (Dilling & Alldredge 1993, Giesecke et al. 2010).

The presence of algal cells (Alvarino 1965, Alvarez-Cadena 1993, Marazzo et al. 1997, Kruse et al. 2010), green detritus and phaeophytin (Philp 2007, Grigor et al. 2015) in chaetognath guts hints at non-carnivorous feeding. However, such vegetal remains are generally interpreted as originating in the guts of digested animal prey. Based on various aspects of chaetognath nutrition, notably in-vitro ingestion of seawater and active digestive processes in gut cells in the absence of visible gut content, Casanova et al. (2012) suggested that dissolved and fine particulate organic matter were major food sources for chaetognaths. Several studies on the polar chaetognaths Eukrohnia hamata, Parasagitta elegans and Pseudosagitta maxima have reported on ˃90 % of individuals lacking visible gut prey (e.g. Sameoto 1987, Øresland 1995, Froneman & Pakhomov 1998, Brodeur & Terazaki 1999, Bollen 2011, Giesecke & Gonzalez 2012, Grigor et al. 2015). For many polar zooplankters, the ability to switch between carnivorous and non-carnivorous feeding (e.g. detritivory or filtering) is an important adaptation to seasonal cycles in food availability (e.g. Auel et al. 2002, Hirche et al. 2003, Norkko et al. 2007, Søreide et al. 2008). Could the polar chaetognaths utilize similar seasonal adaptations, even though they do not obviously exhibit adaptations to filter phytoplankton?

Gut content analysis provides information on recent feeding but the resulting interpretation of trophic relationships can be flawed if sampling or handling protocols damage the gut,

induce regurgitation or defecation, or if feeding is biased by the concentration of predator and prey in the sampling gear (Baier & Purcell 1997 and references therein). Additional and likely less-biased longer-term information on trophic dynamics can be gained from the analysis of fatty acids and stable isotopes. Some fatty acids specific to a given prey typically persist in predators for weeks to months with little or no breakdown or transformation, and can be useful trophic markers (Dalsgaard et al. 2003, Arim & Naya 2003, Grigor et al. 2015). Nitrogen isotopes are routinely used to infer trophic levels due to a step-wise increase in 15N/14N ratio (δ15N) with every ascending trophic level (Minagawa & Wada 1984, Michener & Schell 1994). Carbon isotopes can be used to infer the source of primary productivity, given close similarities in the 13C/12C ratios (δ13C) of an animal and its food (Hobson & Welch 1992).

This study aims at understanding how food resources are utilized year-round by co-existing Eukrohnia hamata and Parasagitta elegans in the Canadian and Alaskan Arctic. In the south- eastern Beaufort Sea, gut contents are examined to reveal diets from November 2007 to August 2008. In addition, we compare the fatty acid and stable isotope signatures of E. hamata and P. elegans in the north-eastern Chukchi Sea (Alaskan Arctic) and Baffin Bay (Canadian Arctic), based on collections made in autumn 2014.

4.4 Method 4.4.1 Study areas Sampling surveys were carried out on-board CCGS Amundsen between from 2007 to autumn 2008 in the Beaufort Sea, and in autumn 2014 in the Chukchi Sea and Baffin Bay (Figure 4.1). 127 stations were sampled at weekly (or higher) resolution, in the Amundsen Gulf (69- 72°N, 120-131°W) between November 2007 and August 2008 (Figure 4.1a and Appendix E- 1). 6 stations were sampled in the north-eastern Chukchi Sea (71-76°N, 144-168°W) in September 2014 (Figure 4.1b and Appendix E-2). 3 stations were sampled in the Baffin Bay region (67-72°N, 61-73°W) in early October 2014; single stations in Scott Inlet Fjord, in Gibbs Fjord and southern Baffin Bay (Figure 4.1c and Appendix E-2).

The Amundsen Gulf is a bridge between the south-eastern part of the Beaufort Sea to the Canadian archipelago. This 400-km long gulf has a width of 170 km and a maximum depth ~630 m, and sea-ice cover between October and early June (Barber & Hanesiak 2004, Geoffroy et al. 2011). Three water masses prevail in the region; the nutrient-poor Pacific Mixed Layer (PML, 0-50 m depth; salinity ˂ 31.6 psu), the Pacific Halocline (PH, 50–200 m depth; 32.4-33.1 psu), and lastly the Atlantic Layer (AL, ˃200 m depth; ˃34 psu) (Carmack & Macdonald 2002). The Chukchi Sea (Alaska’s northernmost shelf sea) is a relatively shallow environment characterised by hotspots of high primary productivity stimulated by the northward invasion of nutrient-rich Pacific water (Hopcroft et al. 2004, Weingartner et al. 2005). Sinking phytoplankton may only be partly exploited by zooplankton, supporting benthic animals (Hopcroft et al. 2004 and references therein). On the eastern side of the Canadian archipelago lies the semi-enclosed basin of Baffin Bay, with numerous fjords along the east coast of Baffin Island. A unique hydrocarbon seep and chemolithic community occurs in Scott Inlet Fjord (DFO 2015), but the plankton communities in these fjords are poorly studied.

Figure 4.1 Bathymetric maps of the Arctic Ocean, showing the positions of sampling stations (black circles) in (a) Amundsen Gulf, (b) north-eastern Chukchi Sea and (c) Baffin Bay (SIF = Scott Inlet Fjord; GF = Gibbs Fjord; SBB = southern Baffin Bay). Details of sampling stations are shown in Appendices E-1 and E-2.

4.4.2 Sampling in the Amundsen Gulf The biomass of chlorophyll a (˃0.7 µm) was used as an indicator of algae blooming. At 54 stations (Appendix E-1), seawater samples were collected from 7-12 depths at using the CTD rosette equipped with twenty-four 12 1 Niskin-type bottles (OceanTest Equipment). Methods for the determination of chlorophyll a (chl a) from seawater subsamples are outlined in Chapter 3 of this thesis.

At 10 stations, zooplankton were sampled using a large square-conical net with a 1 m2 opening area, 200 µm mesh and a 2-L rigid cod-end. At 72 stations, zooplankton were sampled using a Hydrobios® Multinet, comprising 9 nets with 0.25 m2 apertures, 200 µm meshes and 2-L rigid cod-ends. Individual nets open and close sequentially to sample zooplankton from pre-selected depth intervals (Appendix E-1). Samplers were deployed to a depth of 10 m above the seabed, cod-end(s) first (non-filtering), and then hauled back to the surface at a constant velocity of 0.5 m s-1. For Multinet sampling, the upper 60 m and lower 60 m of the sampled water column were divided into three 20-m depth layers, and the remainder was divided into three equal layers. Samples were fixed in 4 % buffered formaldehyde-seawater solution.

4.4.3 Sampling in the Chukchi Sea and Baffin Bay Square-conical nets (1 m2 opening areas, 2-L rigid cod-ends) were deployed to collect chaetognaths for fatty acid and stable isotope analyses (Appendix E-2). At 7 stations, epi- pelagic chaetognaths were sampled by oblique trawls of 500/750 µm-mesh nets at 90 m. The ship speed was approximately 1 m s-1, and the cable angle was ~60º. At 6 stations, meso- pelagic chaetognaths were sampled in vertical tows of 200/500 µm-mesh nets (see Amundsen Gulf sampling protocol). A maximum of 30 Eukrohnia hamata or Parasagitta elegans individuals were randomly removed from each sample, and kept frozen at -80 ºC for further processing.

4.4.4 Abundance of zooplankton A total of 70 Multinet collections in the Amundsen Gulf (typically 9 samples per haul), 3 to 11 per month from November 2007 to July 2008, were analysed to estimate the abundance of main zooplankton species. Formalin-preserved samples were sieved into two size fractions; 0.2-1 mm and >1 mm. From each fraction, known subsamples were taken until 300 copepods were removed, and all animals in the subsamples were counted and identified to the lowest possible taxonomic level, typically species or stage. Abundances of species present in the water column were calculated (where possible), summing abundances of appropriate taxa in the two size fractions.

4.4.5 Gut contents A total of 21 net collections in the Amundsen Gulf were used to analyze chaetognath gut contents. 4742 chaetognaths were included in gut content analyses (4078 Eukrohnia hamata with body lengths between 2 and 39 mm and 654 Parasagitta elegans with body lengths between 2 and 42 mm). 10 Pseudosagitta maxima also present in the samples (15-65 mm) were not analysed further. Each individual chaetognath was stained with a solution of Borax Carmine to highlight potential prey tissues (Sameoto 1987). Gut contents were observed under the stereomicroscope (maximum magnification = 11.5×). Prey items were identified to the lowest taxonomic level possible, although accurate identifications of prey species were often difficult due to their augmented state of digestion. Prey items in the mouths of chaetognaths were ignored, as they are a likely reflection of the concentration of predator and prey in the sampling gear, and therefore biased. The average number of prey per chaetognath gut (npc) was estimated for each collection for the two species. Daily predation rates (DPR; number of prey items consumed ind.-1 d-1) were calculated after Bajkov (1935) [Eq. (4.1)].

npc x 24 DPR = (4.1) tdig

Where tdig = digestion time in hours at a suitable temperature (here ~0 ºC). For Eukrohnia hamata, we used 11 h, determined for this species in the Southern Ocean (Giesecke & Gonzalez 2012). For Parasagitta elegans, we used 10.2 h, determined for this species in Massachusetts (Feigenbaum 1982).

To detect the presence of different food items, such as algae, the color of lipid droplets and detritus in the guts of 2118 Eukrohnia hamata and 201 Parasagitta elegans was also described. A limited subset of individuals (3 E. hamata and 3 P. elegans from single collections in November, February, May and July) were re-analysed using Scanning Electron Microscopy (SEM), to detect other evidence of recent feeding only visible at these high magnifications.

4.4.6 Fatty acids To determine the fatty acid profiles of Eukrohnia hamata in autumn (2014), we analysed samples of 5 pooled individuals from 1 station in the Chukchi Sea and 3 in the Baffin Bay region. We also analysed samples of 5 pooled Parasagitta elegans individuals from 3 stations in the Chukchi Sea and 2 in the Baffin Bay region (Appendix E-2). For E. hamata, three samples were analysed per station. For P. elegans, 1-3 samples were analysed per station. Samples were freeze-dried and an internal standard (5β cholanic acid, 1 µg) was added. Total lipids were extracted using a mixture of dichloromethane and methanol (2/1; 5 ml; 3×15 minute ultrasonications). The total lipid fraction was dried with N2 and MeOH-H2O KOH (80/20; 5 %; 3 ml) and was then added to the extract before heating at 90 ºC for 2 hours. The fraction containing the non-saponifiable lipids was obtained by liquid-liquid extraction, dried over Na2SO4, and derivatised using BSTFA (70 ºC, 30 minutes) prior to gas chromatography–mass spectrometry (GC-MS) analysis. The saponification mixture was acidified with HCl (10N; 2 ml) and the fatty acids extracted with hexane. Fatty acids were methylated (BF3MeOH; 20/80; 80 ºC; 30 minutes) prior to identification and quantification of 20 fatty acid methyl esters previously reported in Grigor et al. (2015), plus 20:1 (n-7) by GC-MS. Results are given as relative percentages of the various fatty acids identified in specimens at each station. A higher 18:1 (n-9)/(n-7) ratio was used to indicate a greater tendency towards carnivory (Falk-Petersen et al. 1990, Wang et al. 2015). Higher proportions of the monounsaturated FAs Σ20:1+22:1 MUFA were taken to indicate a greater contribution of Calanus copepods in the diet (Falk-Petersen et al. 1987, Wang et al. 2015). A higher ratio of 16:1/16:0 was used to indicate a greater contribution of diatoms compared to flagellates (Nelson et al. 2001, Wang et al. 2015).

4.4.7 Carbon and nitrogen To determine the stable isotope signatures of Eukrohnia hamata in autumn (2014), we analysed individuals from 3 stations each in the Chukchi Sea and Baffin Bay region. We also analysed Parasagitta elegans samples from 4 stations in the Chukchi Sea and 2 in the Baffin Bay region (Appendix E-2). These samples each contained up to 5 individuals. For E. hamata, 3-6 samples were analysed per station. For P. elegans, 1-5 samples were analysed per station. Tin-wrapped samples were combusted at a localized temperature up to 1800 °C

using an ECS 4010 Elemental Analyser/ZeroBlank Autosampler (Costech Analytical Technologies). Carbon and nitrogen masses were determined to the nearest microgram.

Gases produced (N2 and CO2) after oxidation/reduction and water removal were separated using an internal GC column. Isotope ratios were measured by on-line continuous-flow isotope ratio mass spectrometry (IRMS) with a Thermo Electron Delta Advantage spectrometer operating the continuous-flow mode (Thermo Electron ConFlo III). Five of each of the standards (USGS40 and USGS41, Qi et al. 2003) were analysed at the beginning and end of each run. One standard was run for every 12 samples to check for combustion and correct any instrumental drift; isotope ratio errors were ±0.006 or better.

δ13C values were calculated for samples as changes in sample ratios of 13C/12C from those in the international standard Vienna Pee Dee Belemnite (13C/12C). δ15N values were calculated as changes in sample ratios of 15N/14N from those in the standard AIR (15N/14N), as in [Eq. (4.2)].

13 15 δ C or δ N = [(푅sample/푅standard) − 1]×1000 (4.2) Where R = ratio of 13C/12C or 15N/14N

Trophic levels (TLs) of samples were calculated as changes in δ15N values from that of a typical food-web baseline (TL = 1 in Eq. 4.3). We used particulate organic matter (POM) as the baseline, in line with other arctic studies (e.g. Iken et al. 2005, Bergmann et al. 2009, Roy et al. 2015).

15 15 (δ Nsample−δ Nbaseline) TL = 1 + 15 (4.3) δ Nenrichment per TL 15 Where 1 = TL of POM. We used a δ Nbaseline value of 6.8 ‰, previously reported for shallow water POM (˂50 m) in the North Water Polynya in late spring/summer of 1998 (Hobson et al. 2002). It was assumed that δ15N values increased by 3.8 ‰ with every ascending trophic level (Hobson & Welch 1992, Hobson et al. 2002, Connelly et al. 2014).

4.5 Results 4.5.1 Amundsen Gulf 4.5.1.1 Phenology of algae blooms Based on chlorophyll a biomass, a bloom of ice algae was first detected in late March in the PML (0-60 m), peaked at these depths in late April and early May (chl a biomass ˃ 5 mg m- 3), and succeeded by a surface bloom of phytoplankton (10 mg chl a m-3). In July, phytoplankton penetrated the PH (60-200 m), and phytoplankton biomass peaked in mid-July in the PML (˃ 16 mg chl a m-3; Figure 4.2).

Figure 4.2 Vertical distributions of chlorophyll a biomass (mg m-3) in the upper 200 m of the water column along the ship track from November 2007 to July 2008 in the Amundsen Gulf. Black dots indicate sampling depths. Details of sampling stations are shown in Appendix E- 1. Chl a data were provided by Michel Gosselin (Université du Québec à Rimouski).

4.5.1.2 Zooplankton community A total of 119 zoo- and icthyoplankton species and 57 other taxa not identifiable to the species level were detected in the Amundsen Gulf collections from November 2007 to July 2008. Numerically, the mesozooplankton was dominated by the usual guild of copepods (Oithona similis, Triconia borealis, Metridia longa, Calanus glacialis, Microcalanus pygmaeus, Microcalanus pusillus, Pseudocalanus elongatus, Pseudocalanus minutus and Calanus hyperboreus), which were present at all stations (Table 4.1). Eukrohnia hamata was also

present at all stations in the Amundsen Gulf with abundances ranging from 4 m-2 in late March to 3318 m-2 in late July. Parasagitta elegans was less abundant than E. hamata at all stations in Amundsen Gulf (0 m-2 in December to 1927 m-2 in June).

Table 4.1 Composition of the n hauls Mean number m-2 Date of peak Species / taxon mesozooplankton (30 most present (± 1 SD) abundance abundant taxa) sampled in the Oithona similis 70 30475±14290 27/01/08 Amundsen Gulf from November Triconia borealis 70 17583±12191 17/04/08 Metridia longa 70 15402±10758 05/03/08 2007 to July 2008, based on data Calanus glacialis 70 10255±6702 27/01/08 from 70 Multinet hauls. *Taxa Microcalanus pygmaeus 70 4818±3750 27/01/08 could not be identified to species Microcalanus pusillus 67 4603±3260 19/04/08 level. Pseudocalanus elongatus 70 4085±3241 27/01/08 Pseudocalanus minutus 70 3431±2874 05/12/07 Cyclopina sp.* 51 1968±2308 10/06/08

Calanus hyperboreus 70 1962±1565 19/04/08 Radiolarians* 66 1053±2262 27/05/08 Scolecithricella minor 70 798±752 18/04/08

Triconia parila/notopus 18 787±1745 01/05/08 Frittilaria sp.* 18 648±3756 06/05/08

Limacina helicina 66 608±964 01/07/08 Clione limacina 68 606±563 03/03/08 Spinocalanus longicornis 66 540±758 27/01/08 Pseudocalanus acuspes 59 467±1305 27/01/08 Eukrohnia hamata 70 456±565 23/07/08 Microcalanus sp.* 18 381±2957 25/07/08 Pseudocalanus sp.* 27 307±830 01/07/08 Boroecia maxima 69 295±274 19/04/08 Bivalves* 27 282±938 23/11/07 Aetideopsis rostrata 55 277±443 19/04/08 Oikopleura sp. 43 273±660 01/07/08

Dimophyes arctica 50 273±386 05/03/08

Paraeuchaeta glacialis 70 253±519 28/06/08 Aglantha digitale 68 252±308 27/07/08 Gaetanus tenuispinus 57 212±260 18/04/08 Parasagitta elegans 68 173±339 10/06/08

4.5.1.3 Visible prey items and predation rates Prey organisms were detected in 0.9 % of Eukrohnia hamata and 0.8 % of Parasagitta elegans in the Amundsen Gulf. Copepods comprised 97 % of total prey items in E. hamata and all detected prey items in P. elegans. Identifiable prey taxa were Oithona similis, Calanus spp., a female Pseudocalanus spp. (Figure 4.3), and a chaetognath in one E. hamata individual. Multiple copepods (2 or 3) were detected in five chaetognaths. Daily predation rates (DPRs) were in the range 0-0.27 prey ind. d-1 for E. hamata, and 0-0.09 prey ind. d-1 for P. elegans. In general, predation rates were higher on dates in November and December compared to dates in spring and summer. Average npc values were generally higher in chaetognaths from Multinet collections compared to square-conical net collections, but note that collections from the latter gear were often considerably larger (Figure 4.3).

4.5.1.4 Lipid droplets and detritus Lipid droplets occurred in the body cavities of 72 % of 3408 analysed Eukrohnia hamata. Amounts were substantial though not quantified. The fraction of the E. hamata population with oil in the body cavity peaked between mid-January and early-May (74-87 %), thereafter declining quickly to below 50 % in late-May and early-June. Low percentages were observed in December (51 %). Droplets were typically yellow in color between late November and early March (72-90 % of droplets), but from then until late July were typically green in color (58-92 % of droplets). In addition to occurring in the guts of only 4 % of 497 Parasagitta elegans, lipid droplets were smaller and less abundant in this species. Whilst detritus (mainly crustaceous debris) was observed in the guts of 12 % of Parasagitta elegans, 38 % of Eukrohnia hamata contained detritus in their guts, with 79 % of these containing thick green macroaggregates ˃ 500 µm (see Figure 4.4). The fraction of the E. hamata population containing detritus in the gut peaked in late May (71 %). Detritus in E. hamata were typically green in color on all 11 sampling dates (56-100 %). The fractions of the population containing detritus and green macroaggregates were lower in early May (25 % and 14 % respectively) and in early June (28 % and 25 %).

Green-colored lipid droplets and macroaggregates were observed in both first-year and second-year Eukrohnia hamata individuals. The fraction of the second-year individuals containing such material was ≥ 50 % on all 11 sampling dates from November to July. The occurrence of such gut contents in the first-year animals increased steadily throughout winter as the animals grew. (Table 4.2).

Table 4.2 Frequency of first- and second-year Eukrohnia hamata individuals with green lipid droplets or macroaggregates in guts. Separate results are presented for the autumn (October) and spring (April) broods of E. hamata (see Chapter 3), from approximate month of hatching (no gut content data were available for October). Month Autumn brood Month Spring brood n % with n % with green guts green guts Age 0-1 Apr 15 73 Nov 82 17 Early May 14 100 Dec 49 35 Late May 12 58 Jan 1 100 Jun 33 61 Feb 33 21 Jul 6 67 Mar 106 13 Nov 135 67 Apr 12 8 Dec 97 65 Early May 46 48 Jan 83 88 Late May 14 50 Feb 20 90 Jun 75 45 Mar 100 79 Jul 78 72 Age 1-2 Apr 3 67 Nov 159 49 Early May 14 71 Dec 86 57 Late May 7 71 Jan 51 80 Jun 29 52 Feb 67 73 Jul 16 94 Mar 74 82 Nov 50 56 Apr 4 0 Dec 33 79 Early May 35 31 Jan 4 75 Late May 27 22 Feb 19 100 Jun 104 31 Mar 26 100 Jul 47 40

4.5.1.5 Scanning Electron Microscope observations Spherical structures resembling bacteria were observed in Eukrohnia hamata guts in samples from November and May (Figure 4.5). The apparent hooks of a chaetognath, and copepod mandibles and filaments, were observed in Parasagitta elegans guts in samples from November and May. Individuals of both species also contained unidentified amalgamated structures in November, February, May and July.

4.5.2 Chukchi Sea and Baffin Bay 4.5.2.1 In-vitro feeding observations At station CS1 in the Chukchi Sea (Figure 4.1b), in-vitro ingestion of green macroaggregates was observed for two living Eukrohnia hamata individuals enclosed in a petri dish (Figure 4.6). One Pseudosagitta maxima individual also observed to consumed marine snow in a petri dish (not shown). Guts of E. hamata in Chukchi Sea samples contained algal cells including those of the Ceratium spp. (dinoflagellate) and diatoms (inverted microscopy, Figure 4.6).

4.5.2.2 Lipid amounts Lipid amounts (expressed as a fraction of sample dry weights) did not vary between Eukrohnia hamata and Parasagitta elegans (Kruskal-Wallis ANOVA, P = 1.00). In each species, lipid amounts were also similar between stations (E. hamata: One-way ANOVA, P = 0.12, P. elegans: Kruskal-Wallis ANOVA, P = 0.35).

4.5.2.3 Fatty acid profiles: species differences Of the 21 fatty acid methyl esters quantified, proportions of 13 varied between Eukrohnia hamata and Parasagitta elegans (Kruskal-Wallis ANOVA, P ˂ 0.05, Table 4.3). Not accounting for station differences, amounts of Calanus markers (ΣC20:1+C22:1) were statistically higher in E. hamata samples (≤30.29 %) than P. elegans samples (≤13.40 %) (Kruskal-Wallis ANOVA, P = 0.00, Table 4.3 and Figure 4.7). Diatom markers 16:1 (n-7) and 20:5 (n-3) were dominant fatty acids in both species (Table 4.3). Amounts of 20:5 (n-3) were similar in samples of the two species (11.12 % - 25.82 % in E. hamata, 8.40 - 22.76 % in P. elegans), but 16:1 (n-7) amounts were higher in E. hamata (23.42 - 48.93 %) compared to P. elegans (12.88 - 20.44 %). Some P. elegans samples contained high amounts of the flagellate marker 16:0 (≤25.69 %), and in general, amounts of 16:0 were lower in E. hamata samples (3.38 - 15.06 %). Corresponding 16:1/16:0 ratios were higher in E. hamata (˃3 versus ˂2 in P. elegans) (Kruskal-Wallis ANOVA, P = 0.00, Figure 4.7). The bacterial marker 18:1 (n-7) occurred in moderate amounts in both species, but the animal marker 18:1 (n-9) occurred in lower amounts in E. hamata (2.48-5.33 %) compared to P. elegans (7.15 - 11.55 %. Corresponding 18:1 (n-9)/(n-7) ratios were higher in P. elegans (˃1 versus ˂1 in E. hamata) (Kruskal-Wallis ANOVA, P = 0.00, Figure 4.7).

4.5.2.4 Fatty acid profiles: station and regional differences in Eukrohnia hamata and Parasagitta elegans For Eukrohnia hamata, proportions of only four fatty acids; 14:0, 18:2(n-6), 18:4(n-3) and 20:1 (n-9) differed between stations (One-way ANOVA, P ≤ 0.05). 18:1 (n-9)/(n-7) ratios did not differ between regions (P = 0.43) or stations (P = 0.38). Similarly, 16:1/16:0 ratios did not differ between regions (P = 0.65) or stations (P = 0.29). For Parasagitta elegans, proportions of all fatty acids were similar between sampling stations (Kruskal-Wallis ANOVA, P ˃ 0.05). Kruskal-Wallis ANOVAs also showed that P. elegans in the two regions had different 18:1 (n-9)/(n-7) ratios (P = 0.02), and different 16:1/16:0 ratios (P = 0.07). These differences, however, were not detected at the station level (P = 0.15 and P = 0.33), which may be an artefact of low sample sizes (n = 1) at two stations (Table 4.3b).

Fatty acid CS4 SIF GF SBB Overall average (a) E. hamata n samples 3 3 3 3 12 n individuals per sample 4-5 5 5 4-5 Body lengths (mm) 10-32 9-32 20-32 10-29 Lipids % DW 41.99±6.57 40.31±6.52 48.68±1.41 37.28±4.08 42.06±6.16 FA mg g-1 ind.-1 24.34±10.39 37.73±18.49 47.60±10.92 20.47±7.30 32.54±15.51 14:0† 1.32±0.63 0.86±0.42 1.19±0.34 2.81±0.66 1.55±0.90

14:1 (n-5)† 0.20±0.06 0.14±0.05 0.22±0.05 0.13±0.06 0.17±0.06

15:0† 0.10±0.02 0.07±0.04 0.09±0.03 0.16±0.06 0.11±0.05

16:0† 5.78±0.95 6.62±4.80 4.96±1.19 11.25±3.30 7.15±3.62

16:1 (n-5)† 0.48±0.07 0.42±0.28 0.70±0.17 0.89±0.37 0.62±0.29

16:1 (n-7)† 26.03±3.40 32.05±6.88 29.53±5.42 36.12±12.66 30.93±7.74

18:0† 0.76±0.22 0.82±0.51 0.50±0.14 1.04±0.22 0.78±0.33 18:1 (n-7) 10.21±5.46 7.22±2.58 10.43±2.93 6.22±0.18 8.52±3.44 18:1 (n-9)† 3.03±0.50 3.62±1.48 3.20±0.57 3.44±0.52 3.32±0.78

18:2 (n-6)† 0.84±0.12 0.54±0.10 0.33±0.28 0.86±0.26 0.64±0.29 18:3 (n-3) 0.33±0.06 0.28±0.16 0.08±0.07 0.22±0.09 0.23±0.13 18:4 (n-3) 1.88±0.33 0.61±0.35 1.01±0.67 1.58±0.39 1.27±0.64 20:1 (n-7) 0.47±0.10 0.72±0.29 0.52±0.17 0.54±0.11 0.56±0.19 20:1 (n-9)† 10.41±2.01 13.51±3.44 14.05±0.59 7.23±0.69 11.30±3.34 20:4 (n-3) 0.10±0.02 0.16±0.12 0.14±0.08 0.10±0.08 0.12±0.08 20:5 (n-3) 18.13±6.80 15.04±3.67 16.41±0.36 15.39±6.79 16.24±4.56 22:1 (n-7) 0.22±0.12 0.21±0.22 0.17±0.13 0.34±0.08 0.24±0.14 22:1 (n-11)† 6.66±2.91 10.35±4.46 9.58±0.60 4.70±1.66 7.82±3.36 22:5 (n-3) 0.53±0.34 0.60±0.54 0.53±0.20 0.35±0.15 0.50±0.31 22:6 (n-3)† 12.20±3.70 5.99±0.64 6.22±2.74 6.32±4.29 7.68±3.83

24:1 (n-9)† 0.31±0.07 0.17±0.06 0.15±0.12 0.30±0.13 0.23±0.11

Table 4.3 (cont.)

Fatty acid CS1 (CS) CS3 (CS) CS4 (CS) SIF SBB Overall average (b) P. elegans n samples 2 3 1 1 2 9 n individuals per sample 5 3-5 5 4 5 Body lengths (mm) 20-28 18-34 20-27 16-24 14-27 Lipids (% DW) 29.22±20.91 40.49±2.94 41.33 38.00 58.29±11.39 41.66±13.54 FA mg g-1 ind.-1 25.95±10.80 41.85±2.67 9.78 32.35 50.71±5.49 20.88±10.03 14:0† 2.80±0.03 2.49±0.29 2.32 3.66 3.65±0.40 2.93±0.60

14:1 (n-5)† 0.02±0.00 0.01±0.00 0.01 0.01 0.01±0.00 0.01±0.00

15:0† 0.40±0.01 0.43±0.08 0.54 0.26 0.24±0.03 0.38±0.11

16:0† 13.88±0.53 17.57±7.04 20.89 23.03 16.62±2.67 17.51±4.69

16:1 (n-5)† 3.18±0.39 2.90±0.51 3.55 2.83 2.41±0.42 2.92±0.49

16:1 (n-7)† 14.69±0.49 16.63±2.90 20.44 14.67 14.19±1.84 15.86±2.57

18:0† 1.01±0.06 1.19±0.44 1.66 1.23 1.17±0.02 1.20±0.29 18:1 (n-7) 5.83±0.82 6.45±2.50 9.22 13.67 9.46±1.48 8.09±2.95 18:1 (n-9)† 8.08±0.06 8.67±2.50 11.29 9.42 8.53±0.45 8.88±1.60

18:2 (n-6)† 1.03±0.18 1.55±0.06 1.88 0.78 0.86±0.02 1.23±0.41 18:3 (n-3) 0.33±0.11 0.31±0.14 0.29 0.15 0.14±0.17 0.26±0.13 18:4 (n-3) 1.01±0.16 1.25±0.49 0.78 0.51 0.79±0.23 0.96±0.37 20:1 (n-7) 0.56±0.04 0.59±0.08 0.77 0.45 0.61±0.07 0.59±0.10 20:1 (n-9)† 6.63±0.51 7.19±1.04 6.75 7.73 6.50±1.86 6.92±0.95 20:4 (n-3) 0.10±0.04 0.10±0.05 0.06 0.03 0.10±0.03 0.09±0.04 20:5 (n-3) 17.72±0.16 13.98±5.72 8.40 12.33 19.37±4.80 15.20±4.93 22:1 (n-7) 0.35±0.04 0.21±0.08 0.26 0.13 0.27±0.02 0.25±0.08 22:1 (n-11)† 3.11±0.59 2.70±0.69 1.67 3.22 3.84±1.31 2.99±0.91 22:5 (n-3) 0.51±0.23 0.32±0.20 0.15 0.25 0.64±0.24 0.41±0.23 22:6 (n-3)† 18.18±0.63 15.00±7.97 8.52 5.35 10.16±4.85 12.84±6.24

24:1 (n-9)† 0.59±0.17 0.47±0.13 0.55 0.30 0.46±0.05 0.49±0.13

4.5.2.5 Carbon and nitrogen isotopes In the Baffin Bay region, δ13C values were similar in the two species (Kruskal-Wallis ANOVA, P = 0.34). In the Chukchi Sea region, δ13C values in Parasagitta elegans remained similar as in Baffin Bay, but δ13C values in Eukrohnia hamata dropped significantly below those of P. elegans (Kruskal-Wallis ANOVA, P = 0.00). Interestingly, the lowest mean δ13C values in the study (-23.1 ± 0.3 ‰), and the highest mean values in the study (-19.2 ± 0.3 ‰), respectively occurred in E. hamata and P. elegans sampled at the same Chukchi Sea station (Table 4.4 and Figure 4.8). In the Baffin Bay region, mean TLs were 2.5-2.7 for E.

hamata and 2.9-3.1 for P. elegans. In the Chukchi Sea, mean TLs were 3.0-3.3 for E. hamata and 3.5-3.7 for P. elegans (Table 4.4 and Figure 4.8).

n samples Body lengths C N δ13C δ15N Station (n inds. per C/N TL (mm) (% DW) (% DW) (‰) (‰) sample) Eukrohnia hamata CS2 6 (4-5) 5-32 46.0±5.8 14.3±1.8 3.2±0.0 -23.1±0.3 14.4±0.5 3.0±0.1 CS4 4 (4-5) 4-32 40.3±5.3 13.0±1.5 3.1±0.1 -22.7±0.4 15.5±0.2 3.3±0.1 CS5 4 (5) 10-32 47.8±0.6 14.9±0.2 3.2±0.0 -22.6±0.3 14.4±0.2 3.0±0.1 SIF 3 (2-5) 9-32 42.0±4.9 13.4±1.3 3.1±0.1 -19.8±0.3 13.4±0.2 2.7±0.1 GF 4 (4-5) 5-31 46.9±1.2 14.8±0.4 3.2±0.0 -20.1±0.3 12.9±0.6 2.6±0.1 SBB 5 (3-5) 4-29 39.6±5.4 12.5±1.6 3.2±0.0 -20.0±0.2 12.6±0.4 2.5±0.1 Parasagitta elegans CS1 5 (5) 19-35 42.8±1.9 13.7±0.6 3.1±0.1 -19.2±0.3 16.8±0.9 3.6±0.2 CS2 5 (5) 17-32 36.2±10.8 11.5±3.4 3.2±0.0 -19.3±0.2 16.3±0.3 3.5±0.1 CS3 5 (1-5) 6-34 43.9±2.1 14.1±0.6 3.1±0.1 -19.9±0.7 17.1±0.4 3.7±0.1 CS4 4 (1-5) 6-27 45.2±2.1 14.3±0.5 3.2±0.1 -20.0±0.4 16.7±0.7 3.6±0.2 SIF 1 (4) 16-24 30.4 9.7 3.1 -19.8 14.9 3.1 SBB 4 (1-5) 7-27 44.2±3.4 14.1±0.8 3.1±0.1 -19.3±0.8 14.2±0.3 2.9±0.1

4.6 Discussion 4.6.1 Prey items Kruse et al. (2010) reported a high degree of diet flexibility in Eukrohnia hamata from the Southern Ocean, detecting copepods, jellyfish, radiolarians, tintinnids, and diatoms in their guts. However, copepods comprised almost all prey for Parasagitta elegans in the Barents Sea (Falkenhaug 1991). In this study, except for the meagre evidence of cannibalism in two chaetognaths (see Figure 4.5), copepods were the only prey for P. elegans and E. hamata in the Amundsen Gulf. Copepods taken were the same as in earlier studies; Pseudocalanus spp., Oithona spp. and Calanus spp. (e.g. McLaren 1969, Øresland 1990, Falkenhaug 1991). Small

chaetognaths could ingest relatively large copepods compared to their head width (Figure 4.3).

4.6.2 Predation rates Our detections of visible prey in only 0.9 % of 4078 Eukrohnia hamata and 0.8 % of 654 Parasagitta elegans (and the corresponding daily predation rates ˂0.3 prey ind.-1 d-1), provide little evidence that either chaetognath species is an important predator on copepods in the Amundsen Gulf. For P. elegans daily predation rates in Arctic Ocean populations may be lower than those at lower latitudes (e.g. 5.50 ind.-1 d-1 in the North Sea; Saito & Kiørboe 2001); this is likely an effect of temperature on digestion rates/gut transit times. However, little information exists on gut transit times in arctic chaetognaths. Lipids inside or outside the oil vacuole of 63 % of E. hamata may not necessarily reflect recent prey consumption (Øresland 1990, Kruse et al. 2010), so excluding these from predation rate estimates is wise. Sameoto (1987) recorded visible prey in just 0.3 % of 2000 E. hamata during autumn 1983 in Baffin Bay, but also reported that lipid “globules” in 63 % of E. hamata resembled copepod oil sacs. Interpreting these as evidence of recent prey consumption led Sameoto (1987) to suggest predation rates of 1.8 prey ind. d-1. Based on this earlier estimate, Welch et al. (1992) suggested that chaetognaths annually ingest 51 % of the copepod biomass in the Lancaster Sound Region of the Canadian Arctic (164 of 319 kJ m-2 yr-1). However, we suspect that oil vacuoles in E. hamata individuals were wrongly identified as copepod oil sacs in the Sameoto (1987) study, and suggest that the estimates of Sameoto (1987) and Welch et al. (1992) are erroneous.

Øresland (1995) reported little seasonal change in the low predation rates of Eukrohnia hamata (0.3-0.7 prey ind.-1 d.-1) between austral summer and winter in the Gerlache Strait, Antarctica. Differences in the seasonal behaviour of Arctic and Antarctic copepods may affect the seasonal feeding opportunities for meso-pelagic chaetognaths such as E. hamata (Chapter 3). Our observation of an apparent decline in winter predation in Parasagitta elegans is consistent with earlier studies in the European Arctic (Øresland 1987, Falkenhaug 1991, Bollen 2011, Grigor et al. 2015). Several studies on this species also reported reduced

winter growth in arctic and sub-arctic seas (Dunbar 1962, Grigor et al. 2015, Chapter 3 of this thesis). These observations could suggest a winter decline in predation opportunities, though individuals >12.5 mm can consume a wide variety of prey size classes > 250 μm (Saito & Kiørboe 2001), and by migrating from the PML to the PH in winter, it seems that P. elegans pursued several copepods which showed similar SVM behaviour (see Darnis & Fortier 2014). Therefore, rather than prey access, changes in energy requirements may be the main reason for the winter reductions in predation and growth rates. Some studies have detected diurnal differences in predation rates (e.g. Sullivan 1980, Feigenbaum 1982, Sameoto 1987, Brodeur & Terazaki 1999), a factor not accounted for in the present study.

For the fragile chaetognaths, an apparent scarcity of prey in guts could also be an artefact of sampling and observation protocols. Baier & Purcell (1997) suggested prey was lost in 50% of chaetognaths owing to regurgitation and defecation during net sampling, so a subsequent study on gut contents of Mediterranean chaetognaths doubled npc estimates to account for this potential loss (Duro & Saiz 2000). Inference of recent copepod prey from their undigested mandible remains could also be useful. In a study from the Barents Sea (Falkenhaug 1991), detections of such mandibles at magnifications of up to 50× suggested that 35.7 % of Parasagitta elegans individuals had recently consumed small copepod prey. In this study, Scanning Electron Microscope (SEM) detections of chaetognath hooks, and copepod mandibles and filaments in the chaetognath guts, confirmed the value of high power microscopy for chaetognath feeding studies (Figure 4.5).

4.6.3 Omnivory in arctic chaetognaths Although vegetal remains in guts are generally interpreted as originating in the guts of digested animal prey, gut contents suggested Eukrohnia hamata is more omnivorous than Parasagitta elegans. Specifically, the presence of vast amounts of green macroaggregates and green lipid droplets in E. hamata guts (Figure 4.4), and also the ingestion of green macroaggregates by living E. hamata individuals (Figure 4.6), are two observations offering strong evidence for omnivory. The absence of these observations in P. elegans from the same sampling locations, times and depths, suggest reduced overlap in their feeding strategies.

Particulate organic matter sourced in the photic zone during spring and summer sinks in the water column as marine snow, collecting mucus, detritus, protists, bacteria, algae, other organisms and inorganic material (Alldredge & Silver 1988). Marine snow is identified as a food source for meso-pelagic copepods (e.g. Dagg 1993, Dilling et al. 1998, Sano et al. 2013), euphausiids (Dilling & Brzezinski 2004), polychaete larvae (Bochdansky & Herndl 1992), and fish (Larson & Shanks 1996). Small meso-pelagic copepods and protists which colonize marine snow may be an important food source for feeding zooplankton.

For polar zooplankton, omnivorous and detritivorous feeding strategies may be useful to buffer against seasonal prey shortages (Norkko et al. 2007). For instance, the arctic krill species Thysanoessa inermis may similarly be omni-carnivorous to support growth and reproduction (Søreide et al. 2006). Eukrohnia hamata mostly resided at meso-pelagic depths in the Amundsen Gulf, except for some individuals which entered epi-pelagic depths, the Pacific Halocline, in summer (Chapter 3 of this thesis). As well as adults, juveniles ate marine snow (Table 4.2), suggesting a role of omnivory in stimulating early growth. ROV dives have documented aggregations of arctic chaetognaths near the seabed (L Fortier, unpublished results), where dissolved and fine particulate organic matter occurs in high concentrations (Iken et al. 2005). In this area, E. hamata may sit-and-wait for marine snow to sink down, occasionally also capturing the passing copepod. Other chaetognaths in the genera Archeterokrohnia and Heterokrohnia, known to feed on organic matter and bacteria from sediments (Casanova 1986), are also closely associated with the nepheloid layer.

4.6.4 Fatty acids and stable isotopes confirm gut contents 4.6.4.1 18:1 (n-9)/(n-7) ratios and Calanus copepod FATMs 18:1 (n-9)/(n-7) ratios ˃ 1 indicate that Parasagitta elegans are typically carnivores. However, the lower 18:1 (n-9)/(n-7) ratios seen in P. elegans in Baffin Bay (Figure 4.7) compared the Chukchi Sea are a sign that detritus does contribute more to its diet in the former region. Moreover, other studies on P. elegans (e.g. Øresland 1987, Falkenhaug 1991) do not plead in favor of evident carnivory. Unlike in P. elegans, 18:1 (n-9)/(n-7) ratios in E. hamata populations from different regions were consistently low (~0.2), characteristic of detritivores such as mud-ingesting seastars (e.g. Howell et al. 2003). Admittedly, care should

be taken when interpreting 18:1 (n-9)/(n-7) ratios in animals, since 18:1 (n-9) FA could be synthesized by some primary producers, not only zooplankton. In addition, 18:1 (n-9)/(n-7) ratios may fluctuate with lipid content (Dalsgaard et al. 2003). Higher amounts of Σ20:1+22:1 MUFA in E. hamata than P. elegans in Scott Inlet Fjord, could indicate that in certain environments, like fjords, E. hamata takes Calanus copepods more frequently than P. elegans, which may focus on smaller taxa (e.g. Falkenhaug 1991). Partitioning of food resources, including zooplankton prey, could allow for the co-existence of the two chaetognath species in shallow environments.

4.6.4.2 Algal FATMs Differences in the amounts of 16:0 and Σ16:1 in Eukrohnia hamata and Parasagitta elegans represent interesting observations of the study. An earlier study conducted in the hyperbenthic zone of the Beaufort Sea during autumn 2003 and summer 2004 (Connelly et al. 2014) reported that E. hamata contained the diatom 16:1 (n-7) in similar proportions (21.4 ± 4.9 %) to an omnivorous mysid (33.0 ± 5.3 %); (30.93 ± 7.74 in our study), whilst lower proportions of 16:1 (n-7) occurred in hyperbenthic P. elegans (15.9 ± 2.6 %); (14.5 ± 2.6 in our study). 16:1/16:0 ratios in hyperbenthic E. hamata were amongst the highest of all the zooplankters analysed by Connelly et al. (2014), whilst proportions of 16:1 (n-7) in E. hamata exceeded those in all analysed copepods. Arguably, the high amounts of the diatom markers 16:1 (n-7) and 20:5 (n-3) in E. hamata in both our study (Table 4.3a), and that of Connelly et al. (2014), reflect direct consumption of diatoms in seawater, consistent with our in-vitro ingestion observations (Figure 4.6).

4.6.4.3 δ13C values Several studies have reported a general enrichment in the δ13C values of zooplankton taxa, including copepods and chaetognaths, as longitude decreased in the Canadian Arctic (e.g. Dunton et al. 1989, Saupe et al. 1989, Schell et al. 1989, Hobson & Welch 1992, Hobson et al. 2002). We detected such δ13C enrichment for Eukrohnia hamata but not for Parasagitta elegans (Figure 4.8). The reason for this difference is unclear, but according to Σ20:1+22:1 MUFA results, the contribution of Calanus copepods to the diet of E. hamata was greater in

100

the west than in the east (not the case for P. elegans). This, or other regional differences in diets may contribute to the observed change in δ13C values in E. hamata.

4.6.4.4 δ15N values and inferred TLs Consistent with other arctic studies presenting stable isotope data for chaetognaths, our δ15N observations (Figure 4.8) suggest that Eukrohnia hamata generally feeds at a lower trophic level (TL) than Parasagitta elegans (Søreide et al. 2006, Pomerleau et al. 2011, Connelly et al. 2014). Mean trophic level estimates were lowest for E. hamata in the Baffin Bay region (2.5-2.7), and a full trophic level higher for P. elegans in the Chukchi Sea (3.5-3.7). Pelagic animals with TLs of 2.4-2.8 are omnivores, and those with TLs of 2.9-3.3 carnivores, according to one food web model (Søreide et al. 2006). However, in our method to estimate TLs of chaetognaths, a potential weakness was assigning a constant δ15N value (6.8) to the POM baseline, at all sampling stations. δ15N estimates for POM (or any baseline) could change considerably with sampling location and time, as well as differences in the constituents of the POM, amongst other factors. For comparison, δ15N values for POM in the Amundsen Gulf between September and October 2011 (˂ 100 m) ranged between 2.2 and 7.2 ‰ (J-É Tremblay, unpublished results).

4.6.4.5 C/N ratios Interestingly, the presumably lower capacity of Parasagitta elegans to maintain a wax ester supply (Connelly et al. 2012), did not cause its C/N ratios to differ from those of Eukrohnia hamata (Table 4.4). Mean C/N ratios for hyperbenthic chaetognaths in the Beaufort Sea during summer and autumn were 6.6 ± 0.8 for E. hamata and 5.2 ± 0.7 for P. elegans (Connelly et al. 2012), whereas those of the pelagic chaetognaths in this study were 3.1-3.2. This is similar to those of pelagic P. elegans in the Barents Sea in early summer (3.2; Ikeda & Skjoldal 1989). A C/N ratio decrease in the pelagic populations could reflect enhanced availability of nitrogen above the nepheloid layer (Connelly et al. 2012).

101

4.6.5 Morphological explanations Consistent with the above evidence that Eukrohnia hamata feeds at a lower trophic level than Parasagitta elegans, this species has fewer posterior teeth and fewer hooks than P. elegans, and lacks anterior teeth which exist in their counterpart (Furnestin 1965, Terazaki 1993). Whilst P. elegans hooks are unserrated, the hooks in young E. hamata are serrated (serrations disappear with age). We suspect that these serrated hooks may be particularly useful for grabbing falling particles, early in the life cycle.

4.7 Concluding remarks and future studies Owing to the large size and presumably high abundance of chaetognath fecal pellets, chaetognath pellets could be important in the flux of carbon to the benthos (Dilling & Alldredge 1993). However, Giesecke et al. (2010) suggested that a zooplankton-based diet in taxa such as Parasagitta spp., Solidosagitta spp. and Pseudosagitta spp. cause their pellets to sink at a slower velocity (5-10 ×), compared to those of other zooplankters known to mainly consume algae. This could certainly limit the export of fecal-pellet carbon to depth, as the likelihood of full pellet breakdown in the water column would rise. It is possible that the pellets of omnivorous Eukrohnia hamata would sink faster, and make more of a contribution to carbon sequestration. As carbon sequestration by zooplankton may be important in regulating global climate (Turner 2015), this is an interesting avenue for future studies. Instead of using intrusive nets, underwater traps that rapidly kill invertebrates with toxic chemicals may help to reduce prey loss, allowing us to obtain a better impression of recent feeding. Food choice experiments would help to evaluate preferred food items for E. hamata and other species.

102

5. Chapter 5 – General Conclusions 5.1 Resource use by arctic chaetognaths The main purpose of this Ph. D. thesis was to shed light on the ecology of two major chaetognath species that reside in Arctic waters (Eukrohnia hamata and Parasagitta elegans), with a particular focus on how the species use and share habitat and food resources, allowing them to co-exist. The bulk of studies on zooplankton ecology in Arctic waters have focused on grazing copepods such as Calanus spp., whereas many aspects of chaetognath ecology have received relatively less attention. The data used in Chapters 2-4 were based on chaetognath samples collected during five years (2007, 2008, 2012, 2013, 2014) in European, Canadian and Alaskan portions of the Arctic, including a full annual cycle in the Canadian Arctic (2007-2008). This enabled the acquisition of detailed information on feeding habits (Chapters 2 and 4), and growth and reproductive dynamics (Chapter 3), over broad spatial and temporal scales. Relative to many other mesozooplankton, chaetognaths were abundant components at all locations and times sampled. The following sections re-cap upon how P. elegans had the most specialized requirements (its distribution and life cycle appeared to be closely associated with copepod prey), whereas E. hamata showed greater flexibility for food options and spawning times. Co-existence of the similar chaetognaths E. hamata and P. elegans in the Arctic is possibly due to reduced temporal and spatial overlaps in their habitats and feeding strategies; differences that reduce competition between species (Ross 1986).

5.2 Winter ecology in Svalbard Insights into the activities of chaetognaths during the polar night are presented in all three manuscripts. Winter ecology is considered a “new field of science” (Lønne et al. 2014), and the polar night sampling dimension of all three articles represents a particularly important step forward in our understanding of zooplankton communities. The first of the three articles (Chapter 2) added to our understanding of Parasagitta elegans ecology during the Svalbard polar night. We showed reduced winter activity through a low predation rate and the absence of reproduction (both contrasting spring-summer reports in the European Arctic). Reduced predation in winter offers one explanation for the low winter growth rates reported in the same area (Grigor et al. 2014). This could possibly be due to reduced access to prey (although

103

P. elegans individuals >12.5 mm can consume many prey size classes > 250 μm, Saito & Kiørboe 2001), or reduced energy requirements for winter activities, which may be more likely.

5.3 Life histories, habitats and spawning times Based on our annual time series in the Canadian Arctic, mostly the Amundsen Gulf, we predicted similar maximum body sizes (40-45 mm), lifespans (~2 years), and ages at maturity (1 year+) for Eukrohnia hamata and Parasagitta elegans in this region of the Arctic (Chapter 3). These results suggested that egg laying in these species did not kill chaetognaths, at least not immediately (contradictory to Kuhl 1938).

Our observations in the Amundsen Gulf showed that the bulk of Eukrohnia hamata individuals remained in the Atlantic layer (below ~200 m) in all seasons (Chapter 3). E. hamata bred during all seasons including winter at these meso-pelagic depths, though most of its spawning occurred in spring. Hatching earlier in the season would give E. hamata a longer time to feed and develop energy reserves required for the next winter, and this would lead to higher fitness and a competitive advantage over chaetognaths born later in the season with little time to store lipids (Varpe et al. 2007). In stark contrast, the bulk of Parasagitta elegans individuals was restricted to shallower, epi-pelagic depths. P. elegans bred in the shallower water masses (mostly the Pacific Mixed Layer) later in the season, during summer and autumn, showing an overlap in both distribution and breeding phenology with specific copepods (e.g. Calanus glacialis, Darnis & Fortier 2014). Offspring gain because they hatch near high numbers of suitable copepod prey, ensuring healthy early growth.

I linked the less seasonally restricted spawning cycle of Eukrohnia hamata (compared to Parasagitta elegans), to it being a capital breeder that can draw upon its considerable supply of pre-accumulated wax esters (e.g. Lee & Hirota 1973, Falk-Petersen et al. 1987, Donnelly et al. 1994, Lee et al. 2006, Connelly et al. 2012, Connelly et al. 2016), as its main reproductive fuel. Capital breeding in E. hamata could also be supported by apparent declines in lipid amounts in the oil vacuoles when winter breeding occurred (Chapter 3). Based on our

104

interpretations, concurrent food could be more useful for sustenance or to allow E. hamata to grow a little in winter (Varpe et al. 2007, Varpe 2012, Sainmont et al. 2014a). Peaks in the frequency of E. hamata presenting oil reserves during the months of negative growth in length (from January to March), suggest that reserves were not used to fuel growth.

Parasagitta elegans appears to be an income breeder dependent on high prey availability to reproduce, perhaps because it lacks the storage mechanisms seen in its counterpart (e.g. Lee & Hirota 1973, Falk-Petersen et al. 1987, Donnelly et al. 1994, Lee et al. 2006, Connelly et al. 2012, Connelly et al. 2016). Income breeding in P. elegans could also be supported by the elevated rates of growth and apparent predation in summer-autumn compared to winter- spring (Chapters 3 and 4) and by seasonal overlaps in the habitats of P. elegans and small copepods such as Calanus glacialis and Pseudocalanus spp. (Darnis & Fortier 2014, Chapter 3). Although some seasonal vertical migration was apparent in populations of both chaetognath species in the Amundsen Gulf, the clearest ascent and descent signals were seen in P. elegans. Ascensions by many P. elegans in the spring-time may have covered distances close to 200 m. The downward migration of P. elegans in the water column during autumn, to depths not much deeper than 200 m, also appeared to follow Calanus copepods which overwintered here (Darnis & Fortier 2014). These copepods are likely to have fueled P. elegans maturation in winter-spring.

5.4 Food resources The two species ate copepods and chaetognaths, however predation rate estimates in both Chapters 2 (Parasagitta elegans) and 4 (Eukrohnia hamata and Parasagitta elegans) provided little evidence that either species controlled populations of copepods. Although proportions of analysed P. elegans individuals with visible prey in guts remained below 5 % in both feeding chapters, the higher proportions in Svalbard (4 %) compared to the Amundsen Gulf (˂ 1 %) could be due to differences in digestion times, prey abundances, or body conditions of the analysed chaetognaths.

105

In some ways, the feeding strategies of Eukrohnia hamata and Parasagitta elegans differed markedly. Our observations also documented thick particulate organic matter (POM) as a food source for E. hamata across much of the year in the Amundsen Gulf (and during our limited samplings in the Chukchi Sea). The same gut contents were not apparent in P. elegans over the entire annual cycle in the Amundsen Gulf (2007-2008). Our observations follow previous assertions that fine POM and dissolved organic matter (DOM) are important food sources for chaetognaths (Casanova et al. 2012), as well as stable nitrogen results suggesting E. hamata to be more omnivorous than P. elegans (Søreide et al. 2006, Pomerleau et al. 2011, Connelly et al. 2014). The most likely source of these macroaggregates (˃500 µm) is falling marine snow (grabbed or gulped in seawater), or consumption of material on the seabed. There were high abundances of E. hamata near the seabed in the Amundsen Gulf, possibly in the nepheloid layer, where some chaetognath species are known to ingest organic matter from sediments (Casanova 1986). Grazing of algal cells from ice is unlikely given the meso- pelagic distribution of E. hamata. Copepods attached to particles may be consumed with marine snow, and both components provide nutritional benefits for a chaetognath. Morphological characters in E. hamata, such as the absence of anterior teeth (Furnestin 1965), provide additional and indirect evidence for such omnivory. The head and mouthpart enhancements of P. elegans (more teeth and hooks) are evidence of a more predatory life cycle. In-vitro observations also showed this feeding mode in Pseudosagitta maxima, and to investigate this further, the gut contents of more individuals would need to be analysed.

In the Chukchi Sea (autumn sampling), our observations of macroaggregate consumption by Eukrohnia hamata in-vitro are useful, but limited by sample size. However, these observations are reflected by E. hamata fatty acid and stable isotope signatures. In particular, the high amounts of diatom fatty acid markers in this species offers strong evidence that diatoms were consumed directly. In Parasagitta elegans, flagellate / dinoflagellate markers were mixed in with relatively lower amounts of diatom markers (also in Svalbard, see Chapter 2), and could reflect a direct or an indirect source of algae. However, 18:1 (n-9)/(n- 7) ratios, δ15N values, and δ13C values were all higher in P. elegans than in E. hamata (especially in the Chukchi Sea), again reinforcing our conclusions that P. elegans tends more towards carnivory.

106

Marine snow could be a convenient food source for chaetognaths located below the subsurface chlorophyll maximum, avoiding the need for considerable migrations in the water column to obtain high-quality food. Such food could be critical for the early survival of the ≥90 % of Eukrohnia hamata newborns that hatched at meso-pelagic depths in the months of December, April, June and July. The specialized serrated hooks in young E. hamata could help them to grab particles (Furnestin 1965). Interestingly, this feature does not remain in the adults.

5.5 Limitations of the study 5.5.1 Sampling limitations Chapters 3 and 4 were based mainly on sampling in the Amundsen Gulf, a restricted region of the Canadian Arctic. However, since the Amundsen Gulf was not sampled during the autumn, in Chapter 3 we included length data for Eukrohnia hamata and Parasagitta elegans from other parts of the Canadian Arctic (the Baffin Bay area and Canadian archipelago) to infer their growth dynamics in autumn. We have concluded that life histories of chaetognaths may have slightly different in these regions. Ideally, future seasonal studies should aim to remain in single sampling regions with restricted water mass advection to ensure single populations are persistently sampled (e.g. Grigor et al. 2014).

5.5.2 Stable isotopes and trophic levels In Chapters 2 and 4, we extracted lipids to stabilize carbon isotope values. For a given species, comparisons between the carbon isotope signatures values reported in different studies can be difficult, as other methods do not always take this additional step (see El- Sabaawi et al. 2009). Trophic levels were inferred by comparing δ15N values of our samples with those previously estimated for a food web baseline (we chose POM). We discussed how differences in the choice of the baseline, and its assigned δ15N value, could alter trophic level estimates in this and other studies. As for the chaetognaths, δ15N values of any baseline vary with e.g. location and season. POM is also a mixture of various materials, with a mixture of δ15N values.

107

5.5.3 Length cohorts Seasonal growth dynamics and lifespans of Eukrohnia hamata and Parasagitta elegans were predicted from body length cohorts, identified by visual means in monthly length frequency distributions (Chapter 3). We noted that cohorts were visually clearer (and therefore growth and lifespan interpretations were likely to be better), when species produced a single annual brood or had restricted breeding seasons. Methods that depend less on visual identifications of cohorts could help to reduce subjectivity, for verifications of growth rates and lifespan.

5.6 Future research avenues This study identified important differences in how two different chaetognath species utilize available resources for co-existence in Arctic waters. Another interesting avenue for further study could be the investigation of the feeding ecology of the two Heterokrohnia species that reside at bathy-pelagic depths in the Arctic (Dawson 1968, Kapp 1991). In the future, analyses of the amounts and composition of lipids in chaetognaths throughout a full annual cycle (as done here for gut contents) would help to clarify how reserves are utilized.

For gut content analyses of chaetognaths sampled in nature, the use of sampling techniques that are less obtrusive than nets (sediment traps, light traps and others) can be expected to reduce the sampling biases that corrupt predation rate estimates (Baier & Purcell 1997). Since small, rapidly digested copepods (e.g. Oithona spp.) may be key chaetognath prey, guts must be scanned for small body parts such as copepod mandibles (Falkenhaug 1991). This will certainly require high power microscopy such as Scanning Electron Microscopy (SEM), which showed promising results in Chapter 4. Information on gut passage time and real predation rates may also be obtained by incubating chaetognaths in large containers alongside different prey, providing they can be kept alive, which can be a challenge. Such experiments could check if Eukrohnia hamata show selectivity for different sizes and types of marine snow particles, or if this mode of feeding is mostly opportunistic. E. hamata remain part-time carnivores, so what is the energetic benefit of a diet based mainly on marine snow versus one based mainly on zooplankton prey? Is marine snow frequently consumed in other E. hamata

108

populations? Analyses of fatty acids and stable isotopes throughout a full annual cycle could also provide insights into diet switching.

The role of arctic chaetognaths in mediating biogeochemical fluxes (their role in the biological pump) should be re-considered. The fecal pellets of chaetognaths are large, carbon-rich and produced year-round due to continued feeding. Large fecal pellets tend to sink faster when loaded with algal contents and marine snow, and may be more likely to reach seafloor depths than those of strict carnivores. This is especially the case when excreted at meso-pelagic depths (Dilling & Alldredge 1993, Giesecke et al. 2010), so the fecal pellets of Eukrohnia hamata could contribute more to carbon sequestration than previously thought. A similar conclusion was reached by Casanova et al. (2012), based on their suggestions that chaetognaths make use of the DOM products of viruses and bacteria.

5.7 A note on climate change Climate change can affect the populations of grazers by influencing the phenology of primary production and taxonomic composition of primary producers (Søreide et al. 2010, Leu et al. 2011). Our observations suggest that any future changes in copepod abundances, in response to climate change, would have more severe effects on Parasagitta elegans than Eukrohnia hamata, this being due to its greater reliance on copepods. Increased primary production in ice-free waters (Pabi et al. 2008) could benefit E. hamata due to increased availability of copepods, protozoans or freshly produced marine snow. Specifically, as the ice-free season lengthens, the autumn brood of E. hamata (which currently comprises smaller numbers than its main spring brood) may benefit from the increased food availability provided by new phytoplankton blooms in autumn. On the other hand, P. elegans may also benefit, now encountering more herbivorous copepod prey at epi-pelagic depths. Changes in chaetognath abundances or distributions are expected to affect ecosystems through ‘bottom-up’ and/or ‘top-down’ effects.

109

Bibliography

Alldredge AL, Silver MW (1988) Characteristics, dynamics, and significance of marine snow. Progress in Oceanography 20:41-82 Alvarez-Cadena JN (1993) Feeding of the chaetognath Sagitta elegans Verrill. Estuarine Coastal and Shelf Science 36:195-206 Alvarino A (1964) Bathymetric distribution of chaetognaths. Pacific Science 18:64-82 Alvarino A (1965) Chaetognaths. Oceanography and marine biology. An annual review 3:115-194 Alvarino A (1968) Egg pouches and other reproductive structures in pelagic Chaetognatha. Pacific Science 22:488-492 Alvarino A (1990) 12: Chaetognatha. In: Adiyodi KG, Adiyodi RG (eds) Reproductive biology of invertebrates: IV: sexual differentiation and behaviour. John Wiley & Sons, Chichester, pp 255-282 Alvarino A (1992) 22: Chaetognatha. In: Adiyodi KG, Adiyodi RG (eds) Reproductive biology of invertebrates. V. Sexual differentiation and behaviour. John Wiley & Sons, Chichester, pp 425-470 Ardyna M, Babin M, Gosselin M, Devred E, Rainville L, Tremblay J-É (2014) Recent Arctic Ocean sea ice loss triggers novel autumn phytoplankton blooms. Geophysical Research Letters 41:6207-6212 Arim M, Naya DE (2003) Pinniped diets inferred from scats: analyses of biases in prey occurrences. Canadian Journal of Zoology 81:67-73 Arnkvaern G, Daase M, Eiane K (2005) Dynamics of coexisting Calanus finmarchicus, Calanus glacialis and Calanus hyperboreus populations in a high-Arctic fjord. Polar Biology 28:528-538 Auel H, Harjes M, da Rocha R, Stübing D, Hagen W (2002) Lipid biomarkers indicate different ecological niches and trophic relationships of the Arctic hyperiid amphipods Themisto abyssorum and T. libellula. Polar Biology 25:374-383 Baier CT, Purcell JE (1997) Effects of sampling and preservation on apparent feeding by chaetognaths. Marine Ecology Progress Series 146:37-42 Båmstedt U (1978) Studies on the deep-water pelagic community of Korsfjorden, western Norway: seasonal variation in weight and biochemical composition of Chiridius armatus (Copepoda), Boreomysis arctica (Mysidacea), and Eukrohnia hamata (Chaetognatha) in relation to their biology. Sarsia 63:145-154 Banse (2013) Reflections about chance in my career, and on the top-down regulated world. Annual Review of Marine Science 5:1-19 Bajkov AD (1935) How to estimate the daily food consumption of fish under natural conditions. Transactions of the American Fisheries Society 65:288-289 Barber DG, Hanesiak JM (2004) Meteorological forcing of sea ice concentrations in the southern Beaufort Sea over the period 1979 to 2000. Journal of Geophysical Research. doi:10.1029/2003JC002027 Barber DG, Asplin MG, Gratton Y, Lukovich J, Galley RJ, Raddatz RL, Leitch D (2010) The International Polar Tear (IPY) Circumpolar Flaw Lead (CFL) system study: overview and the physical system. Atmosphere-Ocean 48:225-243

110

Bates NR, Cai WJ, Mathis JT (2011) The ocean carbon cycle in the western Arctic Ocean: distributions and air-sea fluxes of carbon dioxide. Oceanography 24:186-201 Bell T, Josenhans H (1997) The seismic record of glaciation in Nachvak Fiord, Northern Labrador. In: Davies TA, Bell T, Cooper AK, Josenhans H, Polyak L, Solheim A, Stoker MS, Stravers JA (eds) Glaciated continental margins: an atlas of acoustic images. Chapman & Hall, London, pp 190-193 Berge J, Varpe Ø, Moline MA, Renaud PE, Daase M, Falk-Petersen S (2012) Retention of ice-associated amphipods: possible consequences for an ice-free Arctic Ocean. Biology Letters 8:1012-1015 Bergey MA, Crowder RJ, Shinn GL (1994) Morphology of the male system and spermatophores of the arrow worm Ferosagitta hispida (Chaetognatha). Journal of Morphology 221:321-341 Bergmann M, Dannheim J, Bauerfeind E, Klages M (2009) Trophic relationships along a bathymetric gradient at the deep-sea observatory HAUSGARTEN. Deep-Sea Research I 56:408-424 Bieri R (1959) The distribution of the planktonic Chaetognatha in the Pacific and their relationship to the water masses. Limnology and Oceanography 4:1-28 Bieri R, Thuesen EV (1990) The strange worm Bathybelos. American Scientist 78:542-549 Bigelow HB (1926) Plankton of the offshore waters of the Gulf of Maine. Bulletin of the Bureau of Fisheries 40:1-509 Bird KJ, Charpentier RR, Gautier DL, Houseknecht DW, Klett TR, Pitman JK, Moore TE, Schenk CJ, Tennyson ME, Wandrey CR (2008) Circum-Arctic resource appraisal – estimates of undiscovered oil and gas north of the Arctic Circle: U.S. Geological Survey Fact Sheet 2008-3049, pp 1-4. Available, http://pubs.usgs.gov/fs/2008/3049/, retrieved 1 March 2017 Bochdansky AB, Herndl GJ (1992) Ecology of amorphous aggregations (marine snow) in the Northern Adriatic Sea. III: zooplankton interactions with marine snow. Marine Ecology Progress Series 87:135-146 Bogorov BG (1940) On the biology of Euphausiidae and Chaetognatha in the Barents Sea. Bulletin of Moscow Society of Naturalists 4:3-18 Bogorov V (1946) Zooplankton collected by the Sedov expedition 1937-1939. Trudy Sedov Expedition 3:336-370 Bollen M (2011) Feeding ecology of the arctic chaetognath Parasagitta elegans. MSc thesis, University of Bremen Boltovskoy J (1981) Atlas del zooplancton del Atlantico Sudocddental y mitodos de trabajo con el zooplancton marino. Publicación Especial del INIDEP, Mar de Plata (in Spanish) Bone Q, Kapp H, Pierrot-Bults AC (1991) The biology of chaetognaths. Oxford University Press, New York Bouchard C, Fortier L (2011) Circum-arctic comparison of the hatching season of polar cod Boreogadus saida: a test of the freshwater winter refuge hypothesis. Progress in Oceanography 90:105-116 Bouchard C, Mollard S, Robert D, Suzuki K, Fortier L (2016) Contrasting the early life histories of sympatric polar cod Boreogadus saida and ice cod Arctogadus glacialis in southeastern Beaufort Sea. Polar Biology 39:1005-1022

111

Brodeur RB, Terazaki M (1999) Springtime abundance of chaetognaths in the shelf region of the northern Gulf of Alaska, with observations on the vertical distribution and feeding of Sagitta elegans. Fisheries Oceanography 8:93-103 Brotz L, Cheung WL, Kleisner K, Pakhomov E, Pauly D (2012) Increasing jellyfish populations: trends in large marine ecosystems. Hydrobiologia 690:3-20 Buchanan RA, Sekerak AD (1982) Vertical distribution of zooplankton in Eastern Lancaster Sound and Western Baffin Bay, July-October 1978. Arctic 35:41-55 Carmack EC, Macdonald RW (2002) Oceanography of the Canadian Shelf of the Beaufort Sea: a setting for marine life. Arctic 55:29-45 Casanova J-P (1986) Quatre nouveaux chaetognathes atlantiques abyssaux (genre Heterokrohnia). Description, remarques éthologiques et biogéographiques. Oceanologica Acta 9:469-477 Casanova J-P, Barthélémy R-M, Duvert M, Faure E (2012) Chaetognaths feed primarily on dissolved and fine particulate organic matter, not on prey: implications for marine food webs. Hypotheses in the Life Sciences 2:20-29 Chapman WL, Walsh JE (2003) Observed climate change in the Arctic, updated from Chapman and Walsh, 1993: Recent variations of sea ice and air temperatures in high latitudes. Bulletin of the American Meteorological Society 74:33-47 Cheney J (1976) The spatial and temporal abundance patterns of chaetognaths in the western North Atlantic Ocean. Ph. D. thesis, Woods Hole Oceanographic Institution Choe N, Deibel D (2000) Seasonal vertical distribution and population dynamics of the chaetognath Parasagitta elegans in the water column in the hyperbenthic zone of Conception Bay, Newfoundland. Marine Biology 137:847-856 Choe N, Deibel D, Thompson RJ, Lee SH, Bushell VK (2003) Seasonal variation in the biochemical composition of the chaetognath Parasagitta elegans from the hyperbenthic zone of Conception Bay, Newfoundland. Marine Ecology Progress Series 251:191-200 Comeau S, Gorsky G, Jeffree R, Teyssie JL, Gattuso JP (2009) Impact of ocean acidification on a key Arctic pelagic mollusc (Limacina helicina). Biogeosciences 6 :1877-1882 Connelly TL, Deibel D, Parrish CC (2012) Elemental composition, total lipid content, and lipid class proportions in zooplankton from the benthic boundary layer of the Beaufort Sea shelf (Canadian Arctic). Polar Biology 35:941-957 Connelly TL, Deibel D, Parrish CC (2014) Trophic interactions in the benthic boundary layer of the Beaufort Sea shelf, Arctic Ocean: combining bulk stable isotope and fatty acid signatures. Progress in Oceanography 120:79-92 Connelly TC, Businski TN, Deibel D, Parrish CC, Trela P (2016) Annual cycle of lipid content and lipid class composition in zooplankton from the Beaufort Sea shelf, Canadian Arctic. Canadian Journal of Fisheries and Aquatic Sciences 73:1-12 Conover RJ (1967) Reproductive cycle, early development, and fecundity in laboratory populations of the copepod Calanus hyperboreus. Crustaceana 13:61-72 Conover RJ (1988) Comparative life histories in the genera Calanus and Neocalanus in high- latitudes of the northern hemisphere. Hydrobiologia 167:127-142 Conway DVP, Williams R (1986) Seasonal population structure, vertical distribution and migration of the chaetognath Sagitta elegans in the Celtic Sea. Marine Biology 93:377- 387 Daase M, Eiane K (2007) Mesozooplankton distribution in northern Svalbard waters in relation to hydrography. Polar Biology 30:969-981

112

Daase M, Falk-Petersen S, Varpe Ø, Darnis G, Søreide JE, Wold A, Leu L, Berge J, Philippe B, Fortier L (2013) Timing of reproductive events in the marine copepod Calanus glacialis: a pan-Arctic perspective. Canadian Journal of Fisheries and Aquatic Sciences 70:871-884 Daase M, Varpe Ø, Falk-Petersen S (2014) Non-consumptive mortality in copepods: occurrence of Calanus spp. carcasses in the Arctic Ocean during winter. Journal of Plankton Research 36:129-144 Daase M, Hop H, Falk-Petersen S (2016) Small-scale diel vertical migration of zooplankton in the High Arctic. Polar Biology 39:1213-1223 Dagg MJ (1993) Sinking particles as a possible source of nutrition for the calanoid copepod Neocalanus cristatus in the subarctic Pacific Ocean. Deep-Sea Research I 40:1431-1445 Dalsgaard J, St. John M, Gerard K, Muller-Navarra D, Hagen W (2003) Fatty acid trophic markers in the pelagic marine environment. Advances in Marine Biology 46:227-318 Darby DA, Polyak L, Bauch HA (2006) Past glacial and interglacial conditions in the Arctic Ocean and marginal seas – a review. Progress in Oceanography 71:129-144 Darnis G, Barber DG, Fortier L (2008) Sea ice and the onshore-offshore gradient in zooplankton assemblages in early autumn in south-eastern Beaufort Sea. Journal of Marine Systems 74:994-1011 Darnis G, Robert D, Pomerleau C, Link H, Archambault P, Nelson RJ, Geoffroy M, Tremblay J-É, Lovejoy C, Ferguson S, Hunt BV, Fortier L (2012) Current state and trends in Canadian Arctic marine ecosystems: II. Heterotrophic food web, pelagic- benthic coupling, and biodiversity. Climatic Change 115:179-205 Darnis G, Fortier L (2014) Temperature, food and the seasonal vertical migration of key arctic copepods in the thermally stratified Amundsen Gulf (Beaufort Sea, Arctic Ocean). Journal of Plankton Research 36:1092-1108 Dawson JK (1968) Chaetognaths from the Arctic Basin, including the description of a new species of Heterokrohnia. Bulletin of the Southern California Academy of Science 67:112-124 De Souza CS, Luz JAG, Mafalda PO (2014) Relationship between spatial distribution of chaetognaths and hydrographic conditions around seamounts and islands of the tropical southwestern Atlantic. Anais da Academia Brasileira de Ciências 86:1151-1165 DFO (2015) Ecologically and biologically significant areas. Canada's Eastern Arctic biogeographic region. DFO Canadian Science Advisory Secretariat Science Advisory Report 2015/049, pp 1-18. Available, http://www.dfo-mpo.gc.ca/csas- sccs/publications/sar-as/2015/2015_049-eng.pdf, retrieved 1 March 2017 Dilling L, Alldredge AL (1993) Can chaetognath fecal pellets contribute significantly to carbon flux? Marine Ecology Progress Series 92:51-58 Dilling L, Wilson J, Steinberg DK, Alldredge AL (1998) Feeding by the euphausiid Euphausia pacifica and the copepod Calanus pacificus on marine snow. Marine Ecology Progress Series 170:189-201 Dilling L, Brzezinski MA (2004) Quantifying marine snow as a food choice for zooplankton using stable silicon isotope tracers. Journal of Plankton Research 26:1105-1114 Donnelly J, Torres JJ, Hopkins TL, Lancraft TM (1994) Chemical composition of Antarctic zooplankton during austral autumn and winter. Polar Biology 14:171-183 Du J (2002) Combined algorithms for fitting finite mixture distributions. M. Sc. thesis, McMaster University

113

Dunbar MJ (1940) On the size distribution and breeding cycles of four marine planktonic animals from the Arctic. Journal of Animal Ecology 9:215-226 Dunbar MJ (1962) The life cycle of Sagitta elegans in Arctic and subarctic seas, and the modifying effects of hydrographic differences in the environments. Journal of Marine Research 20:76-91 Dunton KH, Saupe SM, Golikov AN, Schell DM, Schonberg SV (1989) Trophic relationships and isotopic gradients among arctic and subarctic marine fauna. Marine Ecology Progress Series 56:89-97 Dupont N, Aksnes DL (2012) Effects of bottom depth and water clarity on the vertical distribution of Calanus spp. Journal of Plankton Research 34:263-266 Duro A, Saiz E (2000) Distribution and trophic ecology of chaetognaths in the western Mediterranean in relation to an inshore-offshore gradient. Journal of Plankton Research 22:339-361 El-Sabaawi L, Dower JF, Kainz M, Mazumder A (2009) Characterizing dietary variability and trophic positions of coastal calanoid copepods: insight from stable isotopes and fatty acids. Marine Biology 156:225-237 Falardeau M, Robert D, Fortier L (2014) Could the planktonic stages of polar cod and Pacific sand lance compete for food in the warming Beaufort Sea? ICES Journal of Marine Science 71:1956-1965 Falkenhaug T (1991) Prey composition and feeding rate of Sagitta elegans var. arctica (Chaetognatha) in the Barents Sea in early summer. Polar Research 10:487-506 Falkenhaug T (1993) Population structure and metabolism of Sagitta elegans var. arctica in the Barents Sea. In: Moreno I (ed) Proceedings of the 2nd international workshop of chaetognaths. Universitat de les Illes Balears, Mallorca, pp 107-119 Falk-Petersen S, Sargent JR, Tande K (1987) Food pathways and life strategy in relation to the lipid composition of sub-Arctic zooplankton. Polar Biology 8:115-120 Falk-Petersen S, Hopkins CCE, Sargent JR (1990) Trophic relationships in the pelagic, Arctic food web. In: Barnes M, Gibson RN (eds) Trophic relationships in the marine environment. Proceedings of 24th European Marine Biology Symposium. Aberdeen University Press, Aberdeen, pp 315-333 Falk-Petersen S, Timofeev S, Pavlov V, Sargent JR (2007) Climate variability and possible effects on Arctic food chains. The role of Calanus. In: Ørbæk JB, Tombre T, Kallenborn R, Hegseth EN, Falk-Petersen S, Hoel AH (eds) Arctic-alpine ecosystems and people in a changing environment. Springer Verlag, Berlin, pp 147-166 Falk-Petersen S, Mayzaud P, Kattner G, Sargent JR (2009) Lipids and life strategy of Arctic Calanus. Marine Biology Research 5:18-39 Feigenbaum DL, Reeve MR (1977) Prey detection in the Chaetognatha: response to a vibrating probe and experimental determination of attack distance in large aquaria. Limnology and Oceanography 22:1052-1058 Feigenbaum DL (1982) Feeding by the chaetognath, Sagitta elegans, at low temperatures in Vineyard Sound, Massachusetts. Limnology and Oceanography 27:699-706 Ferguson SH, Higdon J, Chmelnitsky EG (2010) The rise of killer whales as a major Arctic predator. In: Ferguson S, Loseto L, Mallory M (eds) A little less Arctic: top predators in the world’s largest northern inland sea, Hudson Bay. Springer, New York, pp 117-136 Folch J, Lees M, Sloane-Stanley GH (1957) A simple method for isolation and purification of total lipids from animal tissue. Journal of Biological Chemistry 226:497-509

114

Forest A, Galindo V, Darnis G, Pineault S, Lalande C, Tremblay J-É, Fortier L (2010) Carbon biomass, elemental ratios (C:N) and stable isotopic composition (δ13C, δ15N) of dominant calanoid copepods during the winter-to-summer transition in the Amundsen Gulf (Arctic Ocean) Journal of Plankton Research 33:161-178 Foster L (2011) Chaetognaths (arrow worms). In: Lewbart GA (ed) Invertebrate Medicine: II. Wiley-Blackwell, Oxford, pp 355-364 Froneman PW, Pakhomov EA (1998) Trophic importance of the chaetognaths Eukrohnia hamata and Sagitta gazellae in the pelagic system of the Prince Edward Islands (Southern Ocean). Polar Biology 19:242-249 Froneman PW, Pakhomov EA, Perissinotto R, Meaton V (1998) Feeding and predation impact of two chaetognath species, Eukrohnia hamata and Sagitta gazellae, in the vicinity of Marion Island (Southern Ocean). Marine Biology 131:95-101 Furnestin ML (1965) Variations morphologiques des crochets au cours du développement dans le genre Eukrohnia. Revue des Travaux de l'Institut des Pêches Maritimes 29:275- 284 Gannefors C, Boer M, Kattner G, Graeve M, Eiane K, Gulliksen B, Hop H, Falk-Petersen S (2005) The Arctic sea butterfly Limacina helicina: lipids and life strategy. Marine Biology 147:169-177 Gibbons MJ, Stuart V, Verheye HM (1992) Trophic ecology of carnivorous zooplankton in the Benguela. South African Journal of Marine Science 12:421-437 Geoffroy M, Robert D, Darnis G, Fortier L (2011) The aggregation of polar cod (Boreogadus saida) in the deep Atlantic layer of ice-covered Amundsen Gulf (Beaufort Sea) in winter. Polar Biology 34:1959-1971 Giesecke R, Gonzalez HE (2004) Mandible characteristics and allometric relations in copepods: a reliable method to estimate prey size and composition from mandible occurrence in predator guts. Revista Chilena de Historia Natural 77:607-616 Giesecke R, Gonzalez HE, Bathmann UV (2010) The role of the chaetognath Sagitta gazellae in the vertical carbon flux of the Southern Ocean. Polar Biology 33:293-304 Giesecke R, Gonzalez HE (2012) Distribution and feeding of chaetognaths in the epipelagic zone of the Lazarev Sea (Antarctica) during austral summer. Polar Biology 35:689-703 Goto T, Yoshida M (1981) Oriented light reactions of the arrow worm Sagitta crassa Tokioka. Biological Bulletin 160:419-431 Goto T, Yoshida M (1985) The mating sequence of the benthic arrow worm Spadella schizoptera. Biological Bulletin 169:328-333 Gause GF (1934) The struggle for existence. Williams & Wilkins, Baltimore Graeve M, Albers C, Kattner G (2005) Assimilation and biosynthesis of lipids in Arctic Calanus species based on 13C feeding experiments with a diatom. Journal of Experimental Marine Biology and Ecology 317:109-125 Grainger EH (1959) The annual oceanographic cycle at Igloolik in the Canadian Arctic: 1: the zooplankton and physical and chemical observations. Journal of the Fisheries Research Board of Canada 16:453-501 Grebmeier JM, Cooper LW, Feder HM, Sirenko BI (2006) Ecosystem dynamics of the Pacific-influenced Northern Bering and Chukchi Seas in the Amerasian Arctic. Progress in Oceanography 71:331-361

115

Grigor JJ, Søreide JE, Varpe Ø (2014) Seasonal ecology and life history strategy of the high- latitude predatory zooplankter Parasagitta elegans. Marine Ecology Progress Series 499:77-88 Grigor JJ, Marais A, Falk-Petersen S, Varpe Ø (2015) Polar night ecology of a pelagic predator, the chaetognath Parasagitta elegans. Polar Biology 38:87-98 Hagen W (1985) On distribution and population structure of Antarctic chaetognaths. Meeresforsch 30:280-291 Hagen W (1999) Reproductive strategies and energetic adaptations of polar zooplankton. Invertebrate Reproduction & Development 36:25-34 Haney JF (1988) Diel patterns of zooplankton behaviour. Bulletin of Marine Science 49:583- 603 Hannah CG, Dupont F, Dunphy M (2009) Polynyas and tidal currents in the Canadian Arctic archipelago. Arctic 62:83-95 Hegseth EN, Sundfjord A (2008) Intrusion and blooming of Atlantic phytoplankton species in the high Arctic. Journal of Marine Systems 74:108-119 Higdon JW, Hauser DDW, Ferguson SH (2012) Killer whales (Orcinus orca) in the Canadian Arctic: distribution, prey items, group sizes, and seasonality. Marine Mammal Science 28:93-109 Hirche H-J, Niehoff B (1996) Reproduction of the Arctic copepod Calanus hyperboreus in the Greenland Sea-field and laboratory observations. Polar Biology 16:209-219 Hirche H-J, Fetzer I, Graeve M, Kattner G (2003) Limnocalanus macrurus in the Kara Sea (Arctic Ocean): an opportunistic copepod as evident from distribution and lipid patterns. Polar Biology 26:720-726 Hirche H-J, Kosobokova KN (2011) Winter studies on zooplankton in Arctic seas: the Storfjord (Svalbard) and adjacent ice-covered Barents Sea. Marine Biology 158:2359- 2376 Hirche H-J, Laudien J, Buchholz F (2015) Near-bottom zooplankton aggregations in Kongsfjorden: implications for pelago-benthic coupling. Polar Biology. doi:10.1007/s00300-015-1799-4 Hoberg EP, Kutz SJ, Cook JA, Galaktionov K, Haukisalmi V, Henttonen H, Laaksonen S, Makarikov A, Marcogliese DJ (2013) Arctic biodiversity assessment: status and trends in Arctic biodiversity. CAFF, Akureyri Hobson KA, Welch HE (1992) Determination of trophic relationships within a high arctic marine food web using δ13C and δ15N analysis. Marine Ecology Progress Series 84:9-18 Hobson KA, Fisk A, Karnovsky N, Holst M, Gagnon J-M, Fortier M (2002) A stable isotope (δ13C, δ15N) model for the North Water food web: implications for evaluating trophodynamics and the flow of energy and contaminants. Deep-Sea Research II 49:5131-5150 Hop H, Falk-Petersen S, Svendsen H, Kwasniewski S, Pavlov V, Pavlova O, Søreide JE (2006) Physical and biological characteristics of the pelagic system across Fram Strait to Kongsfjorden. Progress in Oceanography 71:182-231 Hopcroft RR, Kosobokova KN, Pinchuk AI (2004) Zooplankton community patterns in the Chukchi Sea during summer 2004. Deep-Sea Research II 57:27-39 Hopcroft RR, Clarke C, Nelson RJ, Raskoﬀ KA (2005) Zooplankton communities of the Arctic’s Canada Basin: the contribution by smaller taxa. Polar Biology 28:198-206

116

Hopcroft RR, Bluhm BA, Gradinger RR (2008) Arctic Ocean synthesis: analysis of climate change impacts in the Chukchi and Beaufort Seas with strategies for future research: II. North Pacific Research Board, Anchorage, pp 1-184 Howell KL, Pond DW, Billett DSM, Tyler PA (2003) Feeding ecology of deep-sea seastars (Echinodermata: Asteroidea): a fatty acid biomarker approach. Marine Ecology Progress Series 255:193-206 Huntington H, Weller G (2005) Ch. 1: An Introduction to the Arctic Climate Impact Assessment. Arctic Climate Impact Assessment, Cambridge University Press, pp 1-19 Ikeda T (1974) Nutritional ecology of marine zooplankton. Memoirs of the Faculty of Fisheries Hokkaido University 22:1-97 Ikeda T, Skjoldal HR (1989) Metabolism and elemental composition of zooplankton from the Barents Sea during early Arctic summer. Marine Biology 100:173-183 Iken K, Bluhm BA, Gradinger R (2005) Food web structure in the high Arctic Canada Basin: evidence from δ13C and δ15N analysis. Polar Biology 28:238-249 Ingram R, Bacle J, Barber D, Gratton Y, Melling H (2002) An overview of physical processes in the North Water. Deep-Sea Research Part II 49:4893-4906 IPCC (2014) Summary for Policymakers. In: Edenhofer O, Pichs-Madruga R, Sokona Y, Farahani E, Kadner S, Seyboth K, Adler A, Baum I, Brunner S, Eickemeier P, Kriemann B, Savolainen J, Schlömer S, von Stechow C, Zwickel T, Minx JC (eds) Climate change 2014: mitigation of climate change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA Ji R, Jin M, Varpe Ø (2013) Sea ice phenology and timing of primary production pulses in the Arctic Ocean. Global Change Biology 19:734-741 Kapp H (1991) Redescription of Heterokrohnia mirabilis Ritter-Zfihony, 1911 (Chaetognatha). Helgolander Meeresunters 45:253-262 Kattner G, Hagen W, Lee RF, Campbell R, Deibel D, Falk-Petersen S, Graeve M, Hansen BW, Hirche H-J, Jónasdóttir SH, Madsen ML, Mayzaud P, Muller-Navarra D, Nichols PD, Paffenhöfer G-A, Pond DW, Saito H, Stübing D, Virtue P (2007) Perspectives on marine zooplankton lipids. Canadian Journal of Fisheries and Aquatic Sciences 64:1628- 1639 King KR (1979) The life history and vertical distribution of the chaetognath, Sagitta elegans, in Dabob Bay, Washington. Journal of Plankton Research 1:153-167 Kjesbu OS, Bogstad B, Devine JA, Gjøsæter H, Howell D, Ingvaldsen RB, Nash RDM, Skjæraasen JE (2014) Synergies between climate and management for Atlantic cod fisheries at high latitudes. Proceedings of the National Academy of Sciences of the United States of America 111:3478-3483 Kosobokova KN, Hopcroft RR (2009) Diversity and vertical distribution of mesozooplankton in the Arctic’s Canada Basin. Deep-Sea Research II 57:96-110 Kotori M (1975) Newly-hatched larvae of Sagitta elegans. Bulletin of the Plankton Society of Japan 21:113-114 Kraft A, Berge J, Varpe Ø, Falk-Petersen S (2013) Feeding in Arctic darkness: mid-winter diet of the pelagic amphipods Themisto abysorrum and T. libellula. Marine Biology 160:241-248 Kramp PL (1939) The Godthaab Expedition 1928: Chaetognatha. Meddelelser om Grønland 80:1-38

117

Kruse S (2009) Biology of meso- and bathy-pelagic chaetognaths in the Southern Ocean. Ph. D. thesis, University of Bremen Kruse J, Hagen W, Bathmann U (2010) Feeding ecology and energetics of the Antarctic chaetognaths Eukrohnia hamata, E. bathypelagica and E. bathyantarctica. Marine Biology 157:2289-2302 Kuhl W (1938) Chaetognatha. In: Bronns HG (ed) Klassen und Ordnungen des Tierreichs: IV. Akademische Verlagsgesellschaft, Leipzig Kwok R, Cunningham GF, Wensnahan M, Rigor I, Zwally HJ, Yi D (2009) Thinning and volume loss of the Arctic Ocean sea ice cover: 2003-2008. Journal of Geophysical Research Oceans. doi:10.1029/2009JC005312 Laliberté F, Howell SEL, Kushner PJ (2016) Regional variability of a projected sea ice-free Arctic during the summer months. Geophysical Research Letters 43:256-263 Larson RJ (1986) Water content, organic content, and carbon and nitrogen composition of medusae from the Northeast Pacific. Journal of Experimental Marine Biology and Ecology 99:107-120 Larson ET, Shanks AL (1996) Consumption of marine snow by two species of juvenile mullet and its contribution to their growth. Marine Ecology Progress Series 130:19-28 Lee RF, Hirota J (1973) Wax esters in tropical zooplankton and nekton and the geographical distribution of wax esters in marine copepods. Limnology and Oceanography 18:227- 239 Lee RF (1974) Lipid composition of the copepod Calanus hyperboreus from the Arctic Ocean. Changes with depth and season. Marine Biology 26:313-318 Lee RF, Hagen W, Kattner G (2006) Lipid storage in marine zooplankton. Marine Ecology Progress Series 307:273-306 Legendre L, Ackley SF, Dieckmann GS, Gullicksen B, Horner R, Hoshiai T, Melnikov IA, Reeburgh WS, Spindler M, Sullivan CW (1992) Ecology of sea ice biota part 2: global significance. Polar Biology 12:429-444 Legendre P, Legendre LF (2012) Numerical ecology. III. Elsevier Leu E, Søreide JE, Hessen DO, Falk-Petersen S, Berge J (2011) Consequences of changing sea-ice cover for primary and secondary producers in the European Arctic shelf seas: timing, quantity, and quality. Progress in Oceanography 90:18-32 Li WKW, McLaughlin FA, Lovejoy C, Carmack EC (2009) Smallest algae thrive as the Arctic Ocean freshens. Science 326:539 Linstad H, Bright RM, Strømman AH (2016) Economic savings linked to future Arctic shipping trade are at odds with climate change mitigation. Transport Policy 45:24-30 Loeng H, Brander K, Carmack E, Denisenko S, Drinkwater K, Hansen B, Kovacs K, Livingston P, McLaughlin F, Sakshaug E (2005) Ch. 9: Marine Systems. Arctic Climate Impact Assessment, Cambridge University Press, pp 453-538 Longhurst A, Sameoto D, Herman A (1984) Vertical distribution of Arctic zooplankton in summer: eastern Canadian archipelago. Journal of Plankton Research 6:137-168 Lønne OJ, Falk-Petersen S, Berge J (2014) Introduction to the special issue on polar night studies conducted onboard RV Helmer Hanssen in the Svalbard area. Polar Biology 38:1-3 Lovvorn JR, Cooper LW, Brooks ML, De Ruyck CC, Bump JK, Grebmeier JM (2005) Organic matter pathways to zoo-plankton and benthos under pack ice in late winter and

118

open water in late summer in the north-central Bering Sea. Marine Ecology Progress Series 291:135-150 Makabe R, Hattori H, Sampei M, Darnis G, Fortier L, Sasaki H (2016) Can sediment trap- collected zooplankton be used for ecological studies? Polar Biology. doi:10.1007/s00300-016-1900-7 Margulis L, Chapman M (2010) Kingdoms and domains: an illustrated guide to the phyla of life on Earth: IV. Academic Press, Philadelphia Marazzo A, Machado CF, Nogueira CSR (1997) Notes on feeding of Chaetognatha in Guanabara Bay, Brazil. Journal of Plankton Research 19:819-828 McConnaughey T, McRoy CP (1979) 13C label identifies eel-grass (Zostera marina) carbon in an Alaskan estuarine food web. Marine Biology 53:263-269 McLaren IA (1961) The hydrography and zooplankton of Ogac Lake, a landlocked fjord on Baffin Island. Ph. D. thesis, Yale University McLaren IA (1969) Population and production ecology of zooplankton in Ogac Lake, a landlocked fiord on Baffin Island. Journal of the Fisheries Research Board Canada 26:1485-1559 McLaughlin FA, Carmack EC (2010) Deepening of the nutricline and chlorophyll maximum in the Canada Basin interior, 2003–2009. Geophysical Research Letters. doi:10.1029/2010GL045459 Mehlum F, Gabrielsen GW (1993) The diet of high-arctic seabirds in coastal and ice-covered, pelagic areas near the Svalbard archipelago. Polar Research 12:1-20 Michener RH, Schell DM (1994) Stable isotope ratios as tracers in marine aquatic food webs. In: Lajtha K, Michener R (eds) Stable isotopes in ecology and environmental science. Blackwell Scientific, Boston, pp 138-157 Minagawa M, Wada E (1984) Stepwise enrichment of 15N along food chains: Further evidence and the relation between δ15N and animal age. Geochimica et Cosmochimica Acta 48:1135-1140 Morata N, Seuthe L (2004) Importance of bacteria and protozooplankton for faecal pellet degradation. Oceanologia 56:565-581 Nelson MM, Mooney BD, Nichols PD, Phleger CF (2001) Lipids of Antarctic Ocean amphipods: food chain interactions and the occurrence of novel biomarkers. Marine Chemistry 73:53-64 Newbury TK (1971) Adaptations of chaetognaths to subarctic conditions. Ph. D. thesis, McGill University Nilsen F, Cottier F, Skogseth R, Mattsson S (2008) Fjord-shelf exchanges controlled by ice and brine production: the interannual variation of Atlantic Water in Isfjorden, Svalbard. Continental Shelf Research 28:1838-1853 Nishino S, Kikuchi T, Fujiwara A, Hirawake T, Aoyama M (2016) Water mass characteristics and their temporal changes in a biological hotspot in the southern Chukchi Sea. Biogeosciences 13:2563-2578 Norkko A, Thrush SF, Cummings VJ, Gibbs MM, Andrew NL (2007) Trophic structure of coastal Antarctic food webs associated with changes in sea ice and food supply. Ecology, 88:2810-2820 Øresland V (1987) Feeding of the chaetognaths Sagitta elegans and S. setosa at different seasons in Gullmarsfjorden, Sweden. Marine Ecology Progress Series 39:67-79

119

Øresland V (1990) Feeding and predation impact of the chaetognath Eukrohnia hamata in Gerlache Strait, Antarctic Peninsula. Marine Ecology Progress Series 63:201-209 Øresland V (1995) Winter population structure and feeding of the chaetognath Eukrohnia hamata and the copepod Euchaeta antarctica in Gerlache Strait, Antarctic Peninsula. Marine Ecology Progress Series 119:77-86 Overland JE, Wang M, Walsh JE, Stroeve JC (2014) Future Arctic climate changes: adaptation and mitigation timescales. Earth's Future 2 68-74 Pabi S, van Dijken GL, Arrigo KR (2008) Primary production in the Arctic Ocean, 1998– 2006. Journal of Geophysical Research Oceans. doi:10.1029/2007JC004578 Parsons TR, Maita Y, Lalli CM (1984) A manual of chemical and biological methods for seawater analysis. Pergamon Press, Toronto Pasternak A, Arashkevich E, Tande K, Falkenhaug T (2001) Seasonal changes in feeding, gonad development and lipid stores in Calanus finmarchicus and C. hyperboreus from Malangen, northern Norway. Marine Biology 138:1141-1152 Pearre S (1973) Vertical migration and feeding in Sagitta elegans Verrill. Ecology 54:300- 314 Pearre S (1982) Feeding by Chaetognatha: aspects of inter- and intra-specific predation. Marine Ecology Progress Series 7:33-45 Pearre S (1991) Growth and reproduction. In: Bone Q, Kapp H, Pierrot-Bults AC (eds) The biology of chaetognaths. Oxford University Press, Oxford, pp 61-75 Peterson CH (2001) The “Exxon Valdez” oil spill in Alaska: acute, indirect and chronic effects on the ecosystem. Advances in Marine Biology 39:1-103 Peterson BJ, Holmes RM, McClelland JW, Vörösmarty CJ, Lammers RB, Shiklomanov AI, Shiklomanov IA, Rahmstorf S (2002) Increasing river discharge to the Arctic Ocean. Science 298:2171-2173 Philp K (2007) Ningaloo Reef as a plankton filter: changes in the size spectrum and community structure of zooplankton across a fringing reef. Ph. D. Thesis, University of Western Australia Pierce EL (1941) The occurrence and breeding of Sagitta elegans Verrill and Sagitta setosa J. Miller in parts of the Irish Sea. Journal of the Marine Biological Association of the United Kingdom 25:113-124 Pierrot-Bults AC, Nair VR (1991) Distribution patterns in Chaetognatha. In: Bone Q, Kapp H, Pierrot-Bults AC (eds) The biology of chaetognaths. Oxford University Press, Oxford, pp 86-116 Pomerleau C, Winkler G, Sastri AR, Nelson RJ, Vagle S, Lesage V, Ferguson SH (2011) Spatial patterns in zooplankton communities across the eastern Canadian sub-Arctic and Arctic waters: insights from stable carbon (δ13C) and nitrogen (δ15N) isotope ratios. Journal of Plankton Research 33:1779-1792 Pomerleau C, Lesage V, Ferguson SH, Winkler G, Petersen SD, Higdon JW (2012) Prey assemblage isotopic variability as a tool for assessing diet and the spatial distribution of bowhead whale Balaena mysticetus foraging in the Canadian eastern Arctic. Marine Ecology Progress Series 469:161-174 Pond DW (2012) The physical properties of lipids and their role in controlling the distribution of zooplankton in the oceans. Journal of Plankton Research 34:443-453 Purcell JE (2012) Jellyfish and ctenophore blooms coincide with human proliferations and environmental perturbations. Annual Review of Marine Science 4:209-235

120

Qi H, Coplen TB, Geilmann H, Brand WA, Bohlike JK (2003) Two new organic reference materials for δ13C and δ15N measurements and a new value for the δ13C of NBS 22 oil. Rapid Communications in Mass Spectrometry 17:2483-2487 R Development Core Team (2008) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna Rand KM, Whitehouse A, Logerwell EA, Ahgeak E, Hibpshman R, Parker-Stetter S (2013) The diets of polar cod (Boreogadus saida) from August 2008 in the US Beaufort Sea. Polar Biology 36:907-912 Reeve MR (1970) The biology of chaetognaths. I. Quantitative aspects of growth and egg production in Sagitta hispida. In: Steele JH (ed) Marine foods chains. Oliver and Boyd, Edinburgh, pp 168-189 Richardson AJ (2008) In hot water: zooplankton and climate change. ICES Journal of Marine Science 65:279-295 Ross ST (1986) Resource partitioning in fish assemblages: a review of field studies. Copeia 1986:352-388 Roy V, Iken K, Gosselin M, Tremblay J-É, Belanger S, Archambault P (2015) Benthic faunal assimilation pathways and depth-related changes in food-web structure across the Canadian Arctic. Deep-Sea Research I 102:55-71 Russell FS (1932) On the biology of Sagitta. The breeding and growth of Sagitta elegans Verrill in the Plymouth area, 1930–1931. Journal of the Marine Biological Association of the United Kingdom 18:131-146 Sainmont J, Anderson KH, Varpe Ø, Visser AW (2014a) Capital versus income breeding in a seasonal environment. The American Naturalist 184:466-476 Sainmont J, Gislason A, Heuschele J, Webster C, Sylvander P, Wang M, Varpe Ø (2014b) Inter- and intra-specific diurnal habitat selection of zooplankton during the spring bloom observed by Video Plankton recorder. Marine Biology 161:1931-1941 Saito H, Kiørboe T (2001) Feeding rates in the chaetognath Sagitta elegans: effects of prey size, prey swimming behaviour and small-scale turbulence. Journal of Plankton Research 23:1385-1398 Sakshaug E (2003) Primary and secondary production in the Arctic Seas. In: Stein R, Macdonald RW (eds) The organic carbon cycle in the Arctic Ocean. Springer, Berlin, pp 57-81 Saloranta TM, Haugen PM (2001) Interannual variability in the hydrography of Atlantic water northwest of Svalbard. Journal of Geophysical Research 106:13931-13943 Sameoto DD (1971) life history, ecological production, and an empirical mathematical model of the population of Sagitta elegans in St. Margaret's Bay, Nova Scotia. Journal of the Fisheries Research Board of Canada 28:971-98 Sameoto DD (1972) Yearly respiration rate and estimated energy budget for Sagitta elegans. Journal of the Fisheries Research Board of Canada 29:987-996 Sameoto DD (1987) Vertical distribution and ecological significance of chaetognaths in the Arctic environment of Baffin Bay. Polar Biology 7:3170-328 Sampey A, McKinnon AD, Meekan MG, McCormick MI (2007) Glimpse into guts: overview of the feeding of larvae of tropical shorefishes. Marine Ecology Progress Series 339:243-257 Sands NJ (1980) Ecological studies on the deep-water community of Korsfjorden, Western Norway: population dynamics of the chaetognaths from 1971-1974. Sarsia 65:1-12

121

Sano M, Maki K, Nishibe Y, Nagata T, Nishida S (2013) Feeding habits of mesopelagic copepods in Sagami Bay: Insights from integrative analysis. Progress in Oceanography 110:11-26 Saunders MI, Thompson PA, Jeffs AG, Sӓwstrӧm C, Sachlikidis N, Beckley LE, Waite AM (2012) Fussy feeders: phyllosoma larvae of the Western Rocklobster (Panulirus cygnus) demonstrate prey preference. PLoS ONE. doi:10.1371/journal.pone.0036580 Saupe SM, Schell DM, Griffiths WB (1989) Carbon-isotope ratio gradients in western arctic zooplankton. Marine Biology 103:427-432 Schell DM, Barnett BA, Vinette KA (1989) Carbon and nitrogen isotope ratios in zooplankton of the Bering, Chukchi and Beaufort seas. Marine Ecology Progress Series 162:11-23 Schmid MS, Aubry C, Grigor J, Fortier L (2016) The LOKI underwater imaging system and an automatic identification model for the detection of zooplankton taxa in the Arctic Ocean. Methods in Oceanography. doi:10.1016/j.mio.2016.03.003 Schulz J, Barz K, Ayon P, Ludtke A, Zielinski O, Mengedoht D, Hirche H-J (2010) Imaging of plankton specimens with the Lightframe On-sight Key Species Investigation (LOKI) System. Journal of the European Optical Society - Rapid Publications 5:1-9 Simo-Matchim AG, Gosselin M, Blais M, Gratton Y, Tremblay J-É (2016) Seasonal variations of phytoplankton dynamics in Nunatsiavut fjords (Labrador, Canada) and their relationships with environmental conditions. Journal of Marine Systems 156:56-75 Smith DM (1998) Recent increase in the length of the melt season of perennial Arctic sea ice. Geophysical Research Letters 25:655-658 Søreide JE, Hop H, Falk-Petersen S, Gulliksen B, Hansen E (2003) Macrozooplankton communities and environmental variables in the Barents Sea marginal ice zone in late winter and spring. Marine Ecology Progress Series 263:43-64 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. Progress in Oceanography 71:59-87 Søreide JE, Falk-Petersen S, Hegseth EN, Hop H, Carroll ML, Hobson KA, Blachowiak- Samolyk K (2008) Seasonal feeding strategies of Calanus in the high-Arctic Svalbard region. Deep-Sea Research II 55:2225-2244 Søreide JE, Leu E, Berge J, Graeve M, Falk-Petersen S (2010) Timing of blooms, algal food quality and Calanus glacialis reproduction and growth in a changing Arctic. Global Change Biology 16:3154-3163 Spielhagen RF, Werner K, Sørensen SA, Zamelczyk K, Kandiano E, Budeus G, Husum K, Marchitto TM, Hald M (2011) Enhanced modern hat transfer to the Arctic by warm Atlantic water. Science 331:450-453 Stein R, Macdonald RW (2004) The organic carbon cycle in the Arctic Ocean. Springer, Heidelberg Stroeve J, Serreze M, Drobot S, Gearheard S, Holland M, Maslanik J, Meier W, Scambos T (2008) Arctic sea ice extent plummets in 2007. Eos, Transactions American Geophysical Union 89:13-14 Strömberg J-O (1989) Northern Svalbard Waters. In: Rey L, Alexander V (eds) Proceedings 6th Conference of the Comité Arctique International 13-15 May 1985, p 402-426

122

Stübner E, Søreide JE, Reigstad M, Marquardt M, Blachowiak-Samolyk K (2016) Year- round meroplankton dynamics in high-Arctic Svalbard. Journal of Plankton Research. doi:10.1093/plankt/fbv124 Sullivan BK (1980) In situ feeding behavior of Sagitta elegans and Eukrohnia hamata (Chaetognatha) in relation to the vertical distribution and abundance of prey at Ocean Station “P”. Limnology and Oceanography 25:317-326 Suthers I, Dawson M, Pitt K, Miskiewicz AG (2009) Coastal and marine zooplankton: diversity and biology. In Suthers IM, Rissik D (eds) Plankton, a guide to their ecology and monitoring for water quality. CSIRO Publications, Melbourne, pp 181-222 Svendsen H, Beszczynska-Møller A, Hagen JO, Lefauconnier B, Tverberg V, Gerland S, Ørbæk JB, Bischof K, Papucci C, Zajaczkowski M, Azzolini R, Bruland O, Wiencke C, Winther J-G, Dallmann W (2002) The physical environment of Kongsfjorden- Krossfjorden, an Arctic fjord system in Svalbard. Polar Research 21:133-166 Syvitski JPM, LeBlanc WG, Cranston RE (1990) The flux and preservation or organic carbon in Baffin Island fjords. Geological Society, London, Special Publications 53:177-199 Terazaki M, Miller CB (1982) Reproduction of meso- and bathypelagic chaetognaths in the genus Eukrohnia. Marine Biology 71:193-196 Terazaki M (1993) Deep-sea adaptation of the epipelagic chaetognath Sagitta elegans in the Japan Sea. Marine Ecology Progress Series 98:79-88 Terazaki M (2004) Life history strategy of the chaetognath Sagitta elegans in the world oceans. Coastal Marine Science 29:1-12 Terazaki M, Miller CB (1986) Life history and vertical distribution of pelagic chaetognaths at ocean station P in the subarctic Pacific. Deep-Sea Research I 33:323-337 Terazaki M, Takahashi KT, Odate T (2013) Zonal variations in abundance and body length of chaetognaths in the 140ºE seasonal ice zone during the austral summer of 2001/02. Polar Science 7:39-47 Thuesen EV, Kogure K (1989) Bacterial production of tetrodotoxin in four species of Chaetognatha. Biological Bulletin 176:191-194 Thuesen EV, Numachi K, Nemoto T (1993) Genetic variation in the planktonic chaetognaths Parasagitta elegans and Eukrohnia hamata. Marine Ecology Progress Series 101:243- 251 Timofeev SV (1995) Parasagitta elegans Verrill (chaetognaths) in waters of Spitsbergen archipelago. Oceanology 34:788-791 Tonnesson K Tiselius P (2005) Diet of the chaetognaths Sagitta setosa and S. elegans in relation to prey abundance and vertical distribution. Marine Ecology Progress Series 299:177-190 Tremblay J-É, Gagnon J (2009) The effects of irradiance and nutrient supply on the productivity of Arctic waters: a perspective on climate change. In: Nihoul JCJ, Kostianoy AG (eds) Influence of climate change on the changing Arctic and Sub-Arctic conditions. Elsevier, USA, pp 73-93 Tremblay J-É, Robert D, Varela DE, Lovejoy C, Darnis G, Nelson RJ, Sastri AR (2012) Current state and trends in Canadian Arctic marine ecosystems I: Primary production. Climatic Change 115:161-178 Turner JT (2015) Zooplankton fecal pellets, marine snow, phytodetritus and the ocean’s biological pump. Progress in Oceanography 130:205-248

123

Ussing HM (1939) Biology of some important plankton animals in fjords of East Greenland. Meddelelser om Grønland 100:1-108 van Dongen BE, Schouten S, Sinninghe Damsté JS (2002) Carbon isotope variability in monosaccharides and lipids of aquatic algae and terrestrial plants. Marine Ecology Progress Series 232:83-92 Varpe Ø (2012) Fitness and phenology: annual routines and zooplankton adaptations to seasonal cycles. Journal of Plankton Research 34:267-276 Varpe Ø, Jorgensen C, Tarling GA, Fiksen Ø (2007) Early is better: seasonal egg fitness and timing of reproduction in a zooplankton life-history model. Oikos 116:1331-1342 Varpe Ø, Jørgensen C, Tarling GA, Fiksen Ø (2009) The adaptive value of energy storage and capital breeding in seasonal environments. Oikos 118:363-370 Walcott CD (1911) Cambrian geology and paleontology II: Middle Cambrian annelids. Smithson M. Sc. College Bulletin 57:109-144 Wallace MI, Cottier FR, Berge J, Tarling GA, Griffiths C, Brierley AS (2010) Comparison of zooplankton vertical migration in an ice-free and a seasonally ice-covered Arctic fjord: an insight into the influence of sea ice cover on zooplankton behavior. Limnology & Oceanography 55:831-845 Wang SW, Budge SM, Iken K, Gradinger RR, Springer AM, Wooller MJ (2015) Importance of sympagic production to Bering Sea zooplankton as revealed from fatty acid carbon stable isotope analyses 518:31-50 Wassmann P, Hansen L, Andreassen I, Wexels Riser C, Urban-Rich J (1999) Distribution and sedimentation of faecal pellets on the Nordvestbanken shelf, northern Norway, in 1994. Sarsia 84:239-252 Wassmann P, Reigstad M, Haug T, Rudels B, Carroll ML, Hop H, Gabrielsen GW, Falk Petersen S, Denisenko SG, Arashkevich E, Slagstad D, Pavlova O (2006) Food webs and carbon flux in the Barents Sea. Progress in Oceanography 71:232-287 Webster C, Varpe Ø, Falk-Petersen S, Berge J, Stübner E, Brierley A (2013) Moonlit swimming: vertical distributions of macrozooplankton and nekton during the polar night. Polar Biology. doi:10.1007/s00300-013-1422-5 Weingartner TJ, Aagard, Woodgate R, Danielson S, Sasaki Y, Cavalieri D (2005) Circulation on the north-central Chukchi Sea shelf. Deep-Sea Research II 52:315-317 Welch HE, Bergmann MA, Siferd TD, Martin KA, Curtis MF, Crawford RE, Conover RJ, Hop H (1992) Energy flow through the marine ecosystem of the Lancaster Sound region, Arctic Canada. Arctic 45:343-357 Welch HE, Siferd TD, Bruecker P (1996) Population densities, growth, and respiration of the chaetognath Parasagitta elegans in the Canadian high Arctic. Canadian Journal of Fisheries and Aquatic Science 53:520-527 Williams R, Collins NR (1985) Chaetognaths and ctenophores in the holoplankton of the Bristol Channel. Marine Biology 85:97-107 Winsor P, Bjork G (2000) Polynya activity in the Arctic Ocean from 1958 to 1997. Journal of Geophysical Research 105:8789-8803 Wold A, Darnis G, Søreide JE, Leu E, Phillipe B, Fortier L, Poulin M, Kattner G, Graeve M, Falk-Petersen S (2011a) Life strategy and diet of Calanus glacialis during the winter– spring transition in Amundsen Gulf, south-eastern Beaufort Sea. Polar Biology 34:1929- 1946

124

Wold A, Jæger I, Hop H, Gabrielsen GW, Falk-Petersen S (2011b) Arctic seabird food chains explored by fatty acid composition and stable isotopes in Kongsfjorden, Svalbard. Polar Biology 34:1147-1155 Zavolokin AV, Glebov II, Kosenok NS (2008) Distribution, quantitative composition, and feeding of jellyfish in the Western Bering Sea in summer and autumn. Russian Journal of Marine Biology 34:461-467 Zenkevich L (1963) Biology of the seas of the USSR. Georg Allen and Unwin, London Zo Z (1973) Breeding and growth of the chaetognath Sagitta elegans in Bedford Basin. Limnology and Oceanography 18:750-756

125

Appendix A.

Information on the zooplankton samples included in Chapter 2. Sampling date (UTC), station name, latitude, longitude and bottom depth, sampling gear (MIK or MPS), sampling time (UTC), sampling depths (m). Sampling was horizontal (MIK) in cases where the sampling depth is given by one number, and vertical (MPS) when given by a lower and an upper depth. The ‘Analyses performed’ column indicates if haul individuals were analysed for gut contents (GC), data on abundance, maturity, as well as for stable isotope and fatty acid composition. Each X indicates that 75 Parasagitta elegans from the haul were analysed for stable isotopes (3 x 25 individuals; see Table 2.2), while each O indicates that 30 individuals were analysed for fatty acid trophic markers (3 x 10 individuals; see Table 2.3). The last column gives numbers and body lengths (mm) of individuals dissected for gut content analyses, where relevant.

Date Latitude Longitude Bottom depth Time Sampling depth(s) n P. elegans dissected : Station Gear Analyses performed (UTC) (°N) (°E) (m) (UTC) (m) length range (mm) 12/01/2012 Rijpfjorden (R3) 80.19 22.16 282 MIK 2353 20 GC, XO, abundance 10 : 18-38.5 -"- -"- 80.19 22.11 211 MIK 1146 75 GC, XO, abundance, maturity 40 : 17-55 -"- -"- 80.19 22.15 280 MIK 1220 225 GC, XXO, abundance, maturity 30 : 21-44.5 -"- -"- 80.19 22.16 284 MPS 1310 260-200 Abundance -"- -"- 80.19 22.16 284 MPS 1310 200-100 Abundance -"- -"- 80.19 22.16 284 MPS 1310 100-50 Abundance -"- -"- 80.19 22.16 284 MPS 1310 50-20 Abundance -"- -"- 80.19 22.16 284 MPS 1310 20-0 Abundance -"- -"- 80.19 22.16 284 MPS 1350 256-224 GC 10 : 22.5-33 -"- -"- 80.19 22.16 284 MPS 1350 192-160 GC 6 : 24-31.5 -"- -"- 80.19 22.16 284 MPS 1350 160-128 GC 3 : 21.5-29.5 -"- -"- 80.19 22.16 284 MPS 1350 128-0 GC 10 : 15-41 13/01/2012 Rijpfjorden (R3) 80.19 22.16 278 MIK 1104 20 Abundance -"- -"- 80.19 22.16 275 MIK 0029 75 GC, XO, abundance 11 : 22-44 -"- -"- 80.19 22.16 278 MIK 1136 75 Abundance only -"- -"- 80.19 22.16 278 MIK 1218 225 O, abundance -"- -"- 80.19 22.16 281 MPS 0105 260-200 Abundance -"- -"- 80.19 22.16 281 MPS 0105 200-100 Abundance -"- -"- 80.19 22.16 281 MPS 0105 100-50 Abundance -"- -"- 80.19 22.16 281 MPS 0105 50-20 Abundance -"- -"- 80.19 22.16 281 MPS 0105 20-0 Abundance -"- -"- 80.19 22.16 281 MPS 0157 256-224 GC 8 : 18-32 -"- -"- 80.19 22.16 281 MPS 0157 224-192 GC 4 : 23-28

126

-"- -"- 80.19 22.16 281 MPS 0157 192-128 GC 10 : 20-31.5 -"- -"- 80.19 22.16 281 MPS 0157 128-20 GC 10 : 17-34.5 -"- -"- 80.19 22.16 284 MPS 0157 20-0 GC 1 : 10 17/01/2012 Mouth of Isfjorden 78.20 15.50 268 MIK 1957 250 GC, XO, maturity 20 : 20.5-42 17/01/2012 Isfjorden ISA 78.16 15.32 98 MIK 2208 35 GC, XO, maturity 10 : 18-46 -"- -"- 78.16 15.33 98 MIK 2238 60 GC, XO, maturity 10 : 20-46.5 18/01/2012 Isfjorden ISA 78.16 15.33 80 MIK 1142 30 GC, XO, maturity -"- -"- 78.16 15.33 80 MIK 1207 60 GC, maturity 10 : 25-47 27/01/2012 Isfjorden ISA 78.16 15.33 80 MPS 1717 25-0 Abundance only -"- -"- 78.16 15.33 80 MPS 1727 65-25 Abundance only 12/01/2013 Rijpfjorden (R3) 80.18 22.17 267 MIK 1249 20 GC, maturity 100 : 14-40 16/01/2013 Kongsfjorden 78.58 11.53 327 MIK 0952 30 GC 100 : 8-30 -"- -"- 78.58 11.52 309 MIK 1145 20 GC 100 : 16-41 -"- -"- 78.58 11.53 303 MIK 1255 20 GC, maturity 100 : 17-36 -"- -"- 78.58 11.50 298 MIK 1322 100 GC, maturity 100 : 12-42 -"- -"- 78.58 11.54 308 MIK 1634 20 GC 100 : 15-40 17/01/2013 Kongsfjorden 79.03 11.27 280 MIK 0549 20 GC 100 : 16-41

127

Appendix B.

Vertical profiles of temperature and salinity in Chapter 2. a) Rijpfjorden on 13th January 2012; b) the mouth of Isfjorden on 17th January 2012; and c) Isfjorden station ISA on 18th January 2012. Profiles were obtained by CTD downcast.

128

Appendix C.

Stations sampled in Chapter 3 for information on chaetognath body size and life history characteristics; dates, regions (NF = Nachvak Fjord, PC = Parry Channel, BB = Baffin Bay, AG = Amundsen Gulf), positions, bottom depths, sampler used (Mn = Multinet, SCn = square-conical net), sampling strata and sample splits analysed.

Station Date Bottom Sampler Sample Region Position Sampling strata (m) ID (dd.mm.yy) depth (m) used split 600/602 02.08.07 NF 59° 5.25' N, 63° 26.20' W 215 Mn 0-50, 50-100, 100-150, 150-195 1 302 07.10.07 PC 74° 9.30' N, 86° 17.64' W 529 Mn 0-20, 20-40, 40-60, 61-160, 160-300, 300-460, 460-480, 480-500 0.5 308 09.10.07 PC 74° 8.19' N, 103° 7.30' W 346 SCn 0-335 1 1600 02.11.07 AG 71° 40.77'N, 130° 44.21' W 452 SCn 0-358 1 437 23.11.07 AG 71° 43.81' N, 126° 41.37' W 440 Mn 3-21, 21-40, 41-60, 61-161, 161-261, 261-360, 361-380, 381-396, 397-412 1 D3 30.11.07 AG 71° 2.39' N, 123° 58.46' W 320 SCn 0-310 1 D7 14.12.07 AG 71° 26.15' N, 125° 53.12' W 455 Mn 0-21, 21-40, 41-60, 61-150, 151-250, 251-350, 351-370, 372-390, 391-411 0.5 D12-A 26.12.07 AG 71° 14.96' N, 124° 27.10' W 289 Mn 10-30, 30-50, 50-70, 70-120, 120-170, 170-220, 220-240, 240-260, 260-280 1 D17-9 18.01.08 AG 71° 32.50' N, 125° 0.20' W 216 SCn 0-206 1 D17-I 21.01.08 AG 71° 36.20' N, 125° 9.40' W 242 Mn 10-20, 21-40, 40-60, 60-97, 97-133, 134-171, 172-191, 191-210, 211-233 1 D19-B 04.02.08 AG 71° 4.61' N, 124° 49.02' W 354 Mn 2-21, 22-41, 41-60, 61-125, 126-195, 196-264, 265-286, 286-306, 307-325 0.5 D19-E 10.02.08 AG 71° 3.93' N, 124° 46.15' W 365 Mn 0-20, 20-40, 40-60, 60-130, 130-200, 200-280, 300-320, 320-340 1 D26-12 28.02.08 AG 70° 55.79' N, 123° 51.10' W 359 SCn 0-345 1 D26-E 01.03.08 AG 70° 50.40' N, 123° 36.00' W 443 Mn 60-160 1 D27-B 03.03.08 AG 70° 48.87' N, 123° 53.50' W 352 Mn 4-22, 22-41, 41-62, 62-131, 132-206, 207-280, 280-300, 301-340 0.5 D29-27 19.03.08 AG 70° 54.50' N, 123° 28.60' W 401 SCn 0-385 1 D36-8 08.04.08 AG 71° 17.60' N, 124° 30.60' W 260 SCn 0-250 1 D41-D 18.04.08 AG 70° 37.90' N, 123° 57.90' W 499 Mn 0-20, 20-40, 40-60, 60-185, 185-309, 309-340, 376-404, 404-493 1 D41-E 18.04.08 AG 70° 37.30' N, 121° 55.40' W 500 Mn 0-20, 20-40, 40-60, 60-181, 181-303, 303-425, 424-445, 445-465, 465-485 1 D43-F 01.05.08 AG 70° 48.93 'N, 124° 13.33' W 447 Mn 11-20, 21-40, 40-60, 61-161, 161-263, 263-365, 365-385, 385-405, 405-428 1 405b 20.05.08 AG 70° 39.54' N, 122° 52.74' W 521 Mn 11-19, 20-40, 40-60, 60-179, 179-306, 307-430, 431-449, 450-470, 470-497 0.5 6010 26.05.08 AG 71° 31.86' N, 129° 34.20' W 696 SCn 0-684 1 D45-A 31.05.08 AG 71° 13.07' N, 124° 40.94' W 275 Mn 10-20, 20-40, 40-60, 60-105, 105-150, 150-195, 195-215, 215-235, 235-255 1 405c 02.06.08 AG 70° 37.22 'N, 123° 11.01' W 549 Mn 10-20, 20-40, 40-60, 60-197, 197-333, 433-470, 472-490, 490-510, 510-530 0.5 F7-5 08.06.08 AG 69° 49.45' N, 123° 37.97' W 78 SCn 0-70 1 1208-A 28.06.08 AG 71° 3.84' N, 126° 2.66' W 407 Mn 1-24, 24-40, 40-60, 60-156, 157-245, 246-336, 337-356, 357-372, 372-395 0.5 405 21.07.08 AG 70° 41.70' N, 122° 55.47' W 599 SCn 0-580 1

129

405 21.07.08 AG 70° 41.92' N, 122° 55.91' W 592 Mn 20-40, 40-60 0.5 CA16-07 23.07.08 AG 71° 47.62' N, 126° 29.13' W 439 Mn 1-21, 21-40, 41-60, 61-161, 162-261, 261-371, 371-391, 392-411, 411-433 1 CA05-07 25.07.08 AG 71° 18.19' N, 127° 34.80' W 231 Mn 1-20, 40-61, 61-82, 82-121, 122-160, 181-202, 202-220 1 CA18-07 03.08.08 AG 70° 41.74' N, 122° 54.94' W 591 SCn 0-580 1 108 14.09.08 BB 76° 15.98' N, 74° 34.85' W 436 SCn 0-436 1 118 17.09.08 BB 77° 18.62' N, 76° 38.66' W 480 Mn 0-20, 40-60, 60-177, 177-294, 294-410, 410-430, 430-450, 450-470 1

130

Appendix D.

Vertical profiles of temperature, salinity and chlorophyll a (chl a) in Chapter 3: Nachvak Fjord on 2 August 2007; Parry Channel on 7 October 2007; and Baffin Bay on 17 September 2008. Chl a data were provided by Michel Gosselin (Université du Québec à Rimouski).

131

Appendix E-1. Amundsen Gulf stations sampled in Chapter 4: abbreviated station IDs, full station IDs, dates, positions, bottom depths, sampler used and sample depths. Properties measured are abbreviated: chl a = chlorophyll a biomass; zoop = zooplankton abundance; gc = chaetognath gut contents; vd = chaetognath vertical distributions. SCn = square-conical net, Mn = Multinet, CTD = conductivity-temperate-depth device. See Appendix C for the complete list of Multinet sampling intervals.

Station ID Full Date Bottom Position Sampler Sample depths (m) Properties measured (our study) station ID (dd.mm.yy) depth (m) 1 1600 02.11.07 71° 40.77' N, 130° 44.21' W 452 SCn 0-358 gc 2 1606 03.11.07 71° 33.47' N, 125° 41.51' W 345 CTD 0, 3, 5, 7, 11, 18, 24 chl a 3 1908 05.11.07 71° 8.64' N, 124° 21.17' W 259 CTD 0, 10, 20, 26, 30, 50, 50, 60, 75, 100 chl a 4 405 19.11.07 70° 37.30' N, 123° 0.09' W 640 CTD 10, 40, 80, 80, 120, 240, 500 chl a 5 1100 19.11.07 71° 2.23' N, 123° 15.11' W 281 CTD 10, 30, 60, 80, 100, 267 chl a, zoop 6 437 22.11.07 71° 44.37' N, 126° 38.85' W 420 CTD 10, 30, 50, 100, 200, 414 chl a 6 437 23.11.07 71° 43.81' N, 126° 41.37' W 440 Mn 3-412 zoop, vd, gc 7 D1 28.11.07 70° 25.85' N, 126° 27.44' W 285 CTD 10, 20, 40, 100, 160, 275 chl a 8 D3 30.11.07 71° 2.39' N, 123° 58.46' W 320 SCn 0-310 zoop, gc 9 D4 02.12.07 71° 43.93' N, 125° 33.80' W 247 CTD 10, 40, 60, 90, 160, 232 chl a 10 D5 05.12.07 71° 24.50' N, 124° 50.09' W 217 Mn 2-205 zoop 11 D5 07.12.07 71° 18.30' N, 124° 47.26' W 239 CTD 10, 50, 70, 110, 170, 232 chl a 12 D7 10.12.07 71° 16.22' N, 125° 13.76' W 323 CTD 10, 40, 80, 120, 190, 315 chl a 13 D7 13.12.07 71° 16.22' N, 125° 13.76' W 323 CTD 10 chl a 14 D7 14.12.07 71° 26.15' N, 125° 53.12' W 455 Mn 0-411 zoop, vd, gc 15 D8 16.12.07 71° 24.84' N, 126° 4.69' W 440 CTD 10, 30, 100, 150, 200, 441 chl a 16 D9 18.12.07 71° 46.22' N, 125° 57.94' W 247 Mn 2-230 zoop 17 D12-A 26.12.07 71° 14.96' N, 124° 27.10' W 289 Mn 10-280 vd, gc 18 D12 27.12.07 71° 12.91' N, 124° 28.63' W 280 CTD 10, 40 chl a 19 D12-B 28.12.07 71° 19.68' N, 124° 49.80' W 222 Mn 40-211 zoop 20 D14 03.01.08 71° 14.93' N, 124° 32.20' W 290 CTD 10, 40 chl a, zoop 21 D14-D 06.01.08 71° 32.60' N, 125° 32.10' W 343 Mn 1-333 zoop 22 D14-F 09.01.08 71° 31.60' N, 125° 42.90' W 360 Mn 10-347 zoop 23 D17 17.01.08 71° 31.53' N, 124° 57.97' W 215 CTD 10, 40 chl a 24 D17-E 18.01.08 71° 32.20' N, 124° 59.70' W 217 Mn 10-208 zoop 25 D17-9 18.01.08 71° 32.50' N, 125° 0.20' W 216 SCn 0-206 gc 26 D17-H 21.01.08 71° 36.20' N, 125° 9.40' W 242 Mn 10-232 zoop

132

27 D17-I 21.01.08 71° 36.20' N, 125° 9.40' W 242 Mn 10-233 zoop, vd, gc 28 D17-J 21.01.08 71° 36.20' N, 125° 9.40' W 241 Mn 9-231 zoop 29 D17-K 21.01.08 71° 36.20' N, 125° 9.40' W 241 Mn 9-232 zoop 30 D17-L 21.01.08 71° 36.20' N, 125° 9.40' W 241 Mn 10-232 zoop 31 D19 24.01.08 71° 11.12' N, 125° 8.87' W 319 CTD 10, 40 chl a 32 D19 26.01.08 71° 6.65' N, 124° 56.70' W 312 CTD 10, 40 chl a 33 D19-D 27.01.08 71° 5.50' N, 124° 54.00' W 317 Mn 8-309 zoop 34 D19 03.02.08 71° 4.63' N, 124° 49.00' W 354 CTD 12, 30, 40, 160, 250, 300 chl a 35 D19-B 04.02.08 71° 4.61' N, 124° 49.02' W 354 Mn 2-325 zoop, vd 36 D19-D 08.02.08 71° 4.62' N, 124° 49.01' W 357 Mn 2-331 zoop 37 D19-E 10.02.08 71° 3.93' N, 124° 46.15' W 365 Mn 0-340 vd, gc 38 D19 11.02.08 71° 4.24' N, 124° 47.14' W 371 CTD 5, 12, 50, 60, 160, 250, 350 chl a 39 D19-G 12.02.08 71° 8.91' N, 125° 7.53' W 347 Mn 1-316 zoop 40 D19 14.02.08 71° 27.13' N, 126° 26.94' W 456 CTD 12 chl a 41 D22 18.02.08 71° 18.65' N, 124° 29.80' W 261 CTD 12, 20, 50, 63, 121, 201, 246 chl a, zoop 42 D24-A 23.02.08 71° 26.47' N, 125° 39.80' W 418 Mn 0-400 zoop 43 D26 25.02.08 70° 56.04' N, 123° 55.36' W 357 CTD 12, 20, 60, 80, 150, 200, 342 chl a, zoop 44 D26-12 28.02.08 70° 55.79' N, 123° 51.10' W 359 SCn 0-345 gc 45 D26-D 29.02.08 70° 50.97' N, 123° 39.92' W 425 Mn 2-405 zoop 46 D27 01.03.08 70° 47.43' N, 123° 4.16' W 370 CTD 1, 12, 40, 70, 150, 200, 363 chl a 47 D33-4 03.03.08 70° 48.66' N, 123° 54.90' W 390 CTD 0, 2, 5, 10, 25 chl a 48 D27-B 03.03.08 70° 48.87' N, 123° 53.50' W 352 Mn 4-340 zoop, vd, gc 49 D29-A 05.03.08 71° 0.80' N, 123° 23.30' W 236 Mn 1-230 zoop 50 D29-B 05.03.08 71° 2.52' N, 123° 26.34' W 249 Mn 3-230 zoop 51 D29-C 05.03.08 71° 3.61' N, 123° 27.11' W 252 Mn 2-240 zoop 52 D29-D 05.03.08 71° 4.02' N, 123° 27.13' W 242 Mn 2-230 zoop 53 D29-E 06.03.08 71° 3.94' N, 123° 27.00' W 270 Mn 2-230 zoop 54 D29-F 06.03.08 71° 4.30' N, 123° 27.01' W 236 Mn 3-255 zoop 55 D36 06.03.08 71° 3.97' N, 123° 27.05' W 250 CTD 0, 2, 5, 10, 25 chl a 56 D36 09.03.08 71° 1.09' N, 123° 50.69' W 323 CTD 0, 2, 5, 10, 25 chl a 57 D29 10.03.08 71° 2.32' N, 123° 54.65' W 320 CTD 12, 20, 40, 65, 150, 200, 310 chl a, zoop 58 D29 17.03.08 70° 54.48' N, 123° 28.62' W 401 CTD 2, 5, 10, 50 chl a 59 D31 19.03.08 70° 54.73' N, 123° 1.19' W 444 CTD 0, 10, 25 chl a 60 D29-27 19.03.08 70° 54.50' N, 123° 28.60' W 401 SCn 0-385 gc 61 D29M 19.03.08 70° 54.50' N, 123° 28.60' W 401 Mn 1-386 zoop 62 D33C 28.03.08 71° 3.90' N, 121° 47.20' W 187 Mn 5-181 zoop

133

63 D35A 02.04.08 71° 4.00' N, 121° 56.30' W 213 Mn 0-203 zoop 64 D36C 08.04.08 71° 17.60' N, 124° 30.70' W 259 Mn 0-251 zoop 65 D36-8 08.04.08 71° 17.60' N, 124° 30.60' W 260 SCn 0-250 gc 66 D38 11.04.08 71° 15.52' N, 124° 37.23' W 259 CTD 0, 2, 5, 10, 25 chl a 67 D40 15.04.08 70° 47.89' N, 122° 27.95' W 471 CTD 0, 9, 15, 25, 37, 60, 78, 100 chl a 68 D41 16.04.08 70° 46.65' N, 122° 10.66' W 528 CTD 0, 2, 5, 10, 25 chl a 69 D41C 17.04.08 70° 44.20' N, 122° 8.10' W 541 Mn 2-527 zoop 70 D41-D 18.04.08 70° 37.90' N, 123° 57.90' W 499 Mn 0-493 vd, gc 71 D41-E 18.04.08 70° 37.30' N, 121° 55.40' W 500 Mn 0-485 zoop, vd 72 D41-F 18.04.08 70° 36.50' N, 121° 53.30' W 500 Mn 0-465 zoop 73 D41-G 18.04.08 70° 37.30' N, 121° 51.60' W 514 Mn 0-500 zoop 74 D41 19.04.08 70° 36.52' N, 121° 52.32' W 508 CTD 0, 2, 5, 10, 25 chl a 75 D41-H 19.04.08 70° 35.90' N, 121° 51.00' W 515 Mn 0-500 zoop 76 D41-I 19.04.08 70° 35.90' N, 121° 51.00' W 519 Mn 0-504 zoop 77 D41-J 19.04.08 70° 35.90' N, 121° 51.00' W 519 Mn 0-505 zoop 78 D43-A 23.04.08 70° 36.20' N, 122° 13.00' W 490 Mn 0-475 zoop 79 D43 26.04.08 70° 35.25' N, 122° 26.13' W 550 CTD 0, 2, 5, 10, 25 chl a 80 D43-C 28.04.08 70° 41.33' N, 123° 1.15' W 570 Mn 11-552 zoop 81 D43-1 29.04.08 70° 43.06' N, 123° 17.10' W 459 CTD 0, 2, 5, 10, 25 chl a 82 D43-F 01.05.08 70° 48.93' N, 124° 13.33' W 447 Mn 11-428 zoop, vd, gc 83 D43-2 02.05.08 70° 50.49' N, 124° 26.53' W 475 CTD 0, 2, 5, 10, 25 chl a 84 D43-H 04.05.08 70° 59.99' N, 125° 52.06' W 407 Mn 10-380 zoop 85 D43-3 05.05.08 71° 10.56' N, 126° 18.12' W 420 CTD 0, 2, 5, 10, 25 chl a 86 1020a 06.05.08 71° 1.24' N, 127° 3.10' W 254 CTD 0, 5, 11, 17, 26, 40, 45, 54, 100 chl a 87 6010 06.05.08 72° 31.58' N, 129° 34.50' W 699 Mn 11-680 zoop 88 O2 12.05.08 69° 59.03' N, 126° 2.91' W 205 CTD 11, 18, 26, 40, 54, 75, 100 chl a, zoop 89 405b 19.05.08 70° 39.55' N, 122° 50.74' W 537 CTD 0, 5, 11, 16, 18, 27, 37, 75, 100 chl a 90 405b 20.05.08 70° 39.54' N, 122° 52.74' W 521 Mn 11-497 zoop, vd 91 1011 22.05.08 70° 42.35' N, 124° 0.02' W 455 CTD 0, 6, 13, 21, 30, 32, 51, 69, 75, 100 chl a, zoop 92 6010 26.05.08 71° 31.86' N, 129° 34.20' W 696 SCn 0-684 gc 93 9008 27.05.08 74° 19.91' N, 126° 59.48' W 347 Mn 10-331 zoop 94 D46 30.05.08 71° 13.17' N, 124° 41.11' W 273 CTD 0, 5, 10, 25 chl a 95 D45-A 31.05.08 71° 13.07' N, 124° 40.94' W 275 Mn 10-255 vd, gc 96 405 02.06.08 70° 37.37' N, 123° 11.21' W 546 CTD 0, 10, 14, 22, 35, 53, 73, 100 chl a 97 405c 02.06.08 70° 37.22' N, 123° 11.01' W 549 Mn 10-530 zoop, vd, gc 98 F7-A 07.06.08 69° 49.45' N, 123° 37.97' W 78 Mn 1-78 zoop

134

99 F7-5 08.06.08 69° 49.45' N, 123° 37.97' W 78 SCn 0-70 gc 100 405B 10.06.08 70° 39.85' N, 123° 0.04' W 562 CTD 0, 7, 12, 19, 30, 36, 46, 60, 75, 100 chl a, zoop 101 FB6 15.06.08 69° 58.95' N, 126° 5.18' W 205 CTD 0, 10, 25 chl a 102 FB6-A 16.06.08 69° 58.84' N, 126° 5.30' W 204 Mn 3-190 zoop 103 1216-A 23.06.08 70° 36.69' N, 127° 25.13' W 236 Mn 1-226 zoop 104 1200-A 28.06.08 71° 32.72' N, 124° 20.30' W 207 Mn 1-197 zoop 105 1208 28.06.08 71° 3.87' N, 126° 4.34' W 400 CTD 0, 7, 12, 18, 22, 34, 45, 60, 75, 100 chl a 106 1208-A 28.06.08 71° 3.84' N, 126° 2.66' W 407 Mn 1-395 zoop, vd 107 421-A 01.07.08 71° 27.95' N, 133° 53.78' W 1104 Mn 1-1080 zoop 108 435-A 02.07.08 71° 4.48' N, 133° 47.54' W 300 Mn 2-290 zoop 109 6006-A 04.07.08 72° 39.47' N, 128° 21.56' W 224 Mn 1-211 zoop 110 2010-A 06.07.08 75° 7.28' N, 120° 22.89' W 405 Mn 1-405 zoop 111 410 08.07.08 71° 42.34' N, 126° 29.26' W 395 CTD 0, 12, 29, 33, 54, 75, 80, 100, 108, 200, 388 chl a 112 410-A 08.07.08 71° 41.96' N, 126° 29.18' W 404 Mn 1-392 zoop 113 416 10.07.08 71° 42.16' N, 126° 7.42' W 236 CTD 0, 12, 21, 33, 52, 75, 78, 80, 100, 108 chl a 114 1100 11.07.08 71° 2.66' N, 123° 16.36' W 265 CTD 0, 7, 12, 18, 29, 46, 60, 75, 100, 200 chl a 115 1100-A 11.07.08 71° 2.58' N, 123° 15.58' W 270 Mn 2-261 zoop 116 403-10A 20.07.08 70° 5.46' N, 120° 8.90' W 410 CTD 5, 8, 14, 28, 35, 54, 73, 100, 200, 400 chl a 117 405-10A 21.07.08 70° 41.15' N, 122° 54.08' W 584 CTD 4, 9, 16, 26, 37, 41, 62, 75, 84, 100, 200, 400 chl a 118 405 21.07.08 70° 41.70' N, 122° 55.47' W 599 SCn 0-580 zoop, gc 119 437 23.07.08 71° 41.64' N, 126° 36.01' W 441 CTD 3, 11, 18, 29, 46, 55, 70, 75, 95 chl a 120 CA16-07 23.07.08 71° 47.62' N, 126° 29.13' W 439 Mn 1-433 zoop, vd, gc 121 408 25.07.08 71° 18.83' N, 127° 36.15' W 200 CTD 0, 9, 15, 25, 39, 55, 59, 75, 80, 100, 192 chl a 122 CA05-07 25.07.08 71° 18.19' N, 127° 34.80' W 231 Mn 1-220 zoop, vd 123 CA08-07 27.07.08 71° 8.73' N, 126° 8.12' W 402 Mn 1-393 zoop 124 CA04-07 30.07.08 71° 5.57' N, 133° 45.24' W 335 Mn 1-327 zoop 125 2008-10A 02.08.08 71° 19.95' N, 126° 13.09' W 442 CTD 5, 8, 13, 21, 51, 70, 75, 100, 200 chl a 126 405-10A 03.08.08 70° 40.03' N, 122° 59.01' W 536 CTD 0, 5, 8, 13, 21, 32, 43, 75, 100, 200 chl a 127 CA18-07 03.08.08 70° 41.74' N, 122° 54.94' W 591 SCn 0-580 gc

135

Appendix E-2.

Chukchi Sea and Baffin Bay stations sampled in Chapter 4: abbreviated station IDs (see Figure 4.1b and Figure 4.1c), full station IDs, dates, positions, bottom depths, sampler used and sampling depths. For square-conical net sampling, selected mesh size and tow type are also shown (O = oblique tow, V = vertical tow). FA = fatty acids, SI = stable isotopes. *Analysis performed on Eukrohnia hamata, +analysis performed on Parasagitta elegans (see ‘Method’).

Station ID Full Bottom Sampler Sampling Parameters Date Position (our study) station ID depth (m) (SCn mesh, tow type) depths (m) measured Chukchi Sea CS1 1034 13.09.14 71° 54' N, 154° 58' W 379 SCn (750 µm, O) 0-90 FA+, SI+ CS2 1030 14.09.14 72° 13' N, 153° 58' W 2081 SCn (750 µm, O) 0-90 SI+ CS3 1085 16.09.14 75° 3' N, 167° 8' W 254 SCn (750 µm, O) 0-90 FA+, SI+ CS4 1100 18.09.14 75° 4' N, 161° 16' W 1985 SCn (750 µm, O) 0-90 FA+, SI+ CS4 1100 18.09.14 75° 4' N, 161° 16' W 1985 SCn (200 µm, V) 0-1972 FA*, SI* CS6 1115 20.09.14 73° 54' N, 147° 11' W 3773 SCn (200 µm, V) 0-999 SI* CS7 1130 22.09.14 72° 36'’ N, 144° 44' W 3234 SCn (200 µm, V) 0-1000 SI* Scott Inlet Fjord SIF PCBC-2 01.10.14 71° 5' N, 71° 50' W 693 SCn (500 µm, O) 0-90 FA+, SI+ SIF PCBC-2 01.10.14 71° 5' N, 71° 50' W 693 SCn (500 µm, V) 0-685 FA*, SI* Gibbs Fjord GF Gibbs-B 01.10.14 70° 46' N, 72° 14' W 440 SCn (500 µm, V) 0-430 FA*, SI* Southern Baffin Bay BB 180 03.10.14 67° 28' N, 61° 42' W 214 SCn (500 µm, O) 0-90 FA*+, SI*+

136