Acclimatisation and Adaptive Capacity of Echinoderms in Response to Ocean Acidification and Warming

Acclimatisation and adaptive capacity of echinoderms in response to ocean acidification and warming

Kate Marie Sparks

Thesis submitted for the degree of Doctor of Philosophy University of Otago Department of Marine Science New Zealand December 2017

Abstract

Future ocean acidification and warming pose a substantial threat to the viability of some marine populations. In order to persist, marine species will need to acclimate or adapt to the forecasted changes. Recent research into adaptive capacity of marine species has identified mechanisms of non-genetic inheritance including trans-generational plasticity as important sources of resilience.

Based on literature indicating that echinoderms are tolerant to moderate increases in temperature and seawater pCO2, this study hypothesises three outcomes of long-term exposure to combined ocean acidification and warming:

1. Echinoderms possess the genetic capacity to adapt over long time-scales to

predicted levels of combined ocean acidification and warming.

2. Echinoderms possess the physiological capability to acclimatize to ocean

acidification and warming over long time-scales without a significant cost to

metabolic energy budget.

3. After long-term exposure to ocean acidification and warming, echinoderm parents

would alter the phenotype (Anticipatory Parental Effect) of their offspring to

increase fitness in the F1 generation in response to the environment to which the

parents were exposed.

Broadcast-spawning echinoderms from the phylum Echinodermata (Fellaster zelandiae;

Arachnoides placenta; Acanthaster spp.; Patiriella regularis; Odontaster validus) are used to investigate these hypotheses.

Adaptive capacity was investigated using a quantitative genetic approach to examine the response in gastrula-stage offspring of multiple half-sib families raised in fully crossed treatment combinations of temperature (ambient, +2.0, +4.0 °C) and pCO2 (ambient; 2x; 3x ambient ppm). Interactions between genotype and environment were tested using a permutational multivariate ANOVA and restricted error maximum likelihood calculations of variance, and the variance components used to generate genetic correlations (g*E) and broad- sense heritability estimates (h2).

Contributions of both genetic and environmental factors to the trait ‘normal development at gastrulation’ were examined for wild populations of all five echinoderm species. Complex interactions between sire (father) and dam (mother) identity and environmental factors were detected for most species, and the individual response to warming of between +2.5 - +3.0 °C and concurrent acidification equivalent to -0.4 pH units depended on phylogeny, genetic identity and environmental exposure history. For all species except O. validus, significant

[sire x environment] interactions were found. The response of Acanthaster spp. as well as that of both clypeasteroids (A. placenta and F. zelandiae) to ocean acidification depends on both maternal and paternal lineage, and differences in resilience among genotypes will affect future population response. Maternal lineage (dam identity) was an important factor in every species. For O. validus, the only genetic sources of variation among genotypes came from dam identity and [dam x environment] interactions. The hypothesis that marine populations possess the capacity to adapt to future ocean warming and acidification is therefore supported.

Physiological acclimation was investigated in two sea star species: P. regularis and O. validus. Adults were acclimated to combinations of elevated temperature and low pH for 12 months, during which time physiological indices of gonad and pyloric cecum growth rate, righting time and tissue inorganic content were measured every three months. At the same time respiration, energy absorption efficiency, and nitrogenous excretion rates were measured to estimate Scope for Growth (SfG) in all treatments.

Adult P. regularis from Dunedin were exposed to all combinations of three pH(T) levels

(ambient 8.10, 7.90 and 7.70) and two temperature levels (ambient and +2.50 °C above ambient). SfG in adult P. regularis was reduced by ~48 % through long-term exposure to pH

7.70, but elevated temperature at pH 7.70 enabled some physiological compensation for low

SfG i.e. an increase in SfG of ~45 % at pH 7.70/ elevated temperature. Reduced gonad size, pyloric indices and body organic content indicated that P. regularis can re-allocate energy away from gamete production at pH 7.70 in order to maintain homeostasis. Reduced parental investment in offspring due to energy reallocation may carry over into subsequent generations.

Adult O. validus from Antarctica were exposed to combinations of a +3.0 °C temperature and

-0.4 pH unit decrease for 9 months. There were no reductions in overall SfG of animals subjected to combined warm and acidified seawater. Exposure to elevated temperature in the absence of acidification significantly increased SfG by ~41 % relative to the control treatment. Maintenance of metabolic energy budget was sustained by an increase in food intake to compensate for increased respiration and excretion at elevated temperature and low pH. Animals had the same reproductive capacity in every treatment, indicating that stored

4 energy reserves are sufficient to maintain base metabolism as well as somatic growth and reproductive output. The hypothesis that adult echinoderms exposed to near-future climate- change conditions will be able to acclimate was supported for both species, however P. regularis may still be at risk of physiological distress over multiple generations as it maintains SfG by taking energetic investment away from gametes. Odontaster validus may be one of the most thermally-tolerant polar species and therefore may be one of the ‘winners’ of future climate change.

Trans-generational plasticity was investigated in F1 offspring of the P. regularis and O. validus parents used in the long-term acclimation study. F1 offspring were generated using a quantitative genetic breeding design and raised to the gastrula stage in all combinations of elevated temperature and pCO2, either in the same environment as their parents or a different environment. Parent acclimation increased the fitness of F1 gastrulae in response to elevated temperature in O. validus. Broad-sense heritability (h2) was 0.34, indicating that trait plasticity is partly attributable to genetic variation. There was little variation among offspring phenotype attributable to sire identity, but dam identity accounted for 5.79 % of variation among genotypes in the wild population and 23.27 % of variation among genotypes in the F1 population. Maternal identity accounted for 23.27 % of variation among genotypes in the F1 population. Strong maternal effects (m2) were observed at the gastrula stage (m2 = 0.51), indicating that maternal environmental history is a strong determinant factor in non-genetic inheritance. Maternal effects increased trans-generational fitness in response to elevated temperature, and, to a lesser extent, pCO2. These results suggests that O. validus has a pre- adaptive capacity to respond to ocean warming and acidification that may facilitate population persistence over the next century.

In P. regularis, parent adaptation to ocean acidification was maladaptive for F1 gastrulae, and this was only partially compensated for by F1 exposure to elevated temperature. Parent exposure to pH 7.70 for 12 months reduced offspring % normal development by ~ 13 % regardless of the offspring temperature or pH environment. Offspring exposure to pH 7.70 reduced % normal development by ~25 %. Among F1 offspring of parents at the end of the exposure period, 17.62 % of variation in % normal development between families was explained by dam identity, and strong maternal effects (m2 = 0.41) were detected. Broad- sense heritability was h2 = 0.06 at ambient pH and h2 = 0.16 at pH 7.70. Maladaptive TGP in response to low pH and strong maternal effects indicate intense natural selection for this population in the near future. The hypothesis that echinoderm parents can alter the fitness of their offspring in response to their own thermal or CO2 environment is supported for O. validus but not P. regularis.

This is the first study to combine long-term parent acclimatization to ocean acidification and warming with estimates of broad-sense heritability and trans-generational plasticity in the next generation, and can go some way towards answering questions about the effect of parent environment on non-genetic inheritance and evolutionary fitness. The result of this study have wider implications for predictive models of how Southern Ocean echinoderm species adapt to ocean acidification and warming, and supports the need to exercise caution when using models which predict the responses of temperate marine species from examining their related polar counterparts, or models which use parent environmental stability as a predictor of trans-generational responses to climate change.

Acknowledgments

I would like to thank my supervisors, Assoc. Prof. Miles Lamare, Prof. Abby Smith and Prof.

Steve Wing of the University of Otago, for your unending support, kindness and care throughout this process. Miles, you’ve been a sounding board, lab partner and endless supply of enthusiasm for science and discovery. Abby, you are one of my role models for female scientists, thank you for taking care of me and reminding me to eat, sleep and exercise. Steve, thank you for helping me along in the world of Antarctic science and your generous assistance collecting my sea stars.

I have had the opportunity to work with some amazing scientists and educators from other institutions, including Sven Uthicke, Shawna Foo, Antonio Aguera and Maria Byrne. Even though none of you are my supervisors, I still look up to you all and would like to express my sincere and grateful thanks to you.

Thank you to everyone with whom I have shared lab space at Otago over the past four years, in particular Sam Karelitz and Jennifer Chin. All your help and encouragement have kept me focussed, on track and able to laugh. Thank you to Doug Mackie, Reuben Pooley, Linda

Groenewegen and Dave Wilson, Tessa Mills, Steve Cutler and everyone else at PML and the

NZMSC for your encouragement and support in my experiments. Thank you also to

Shannon, Josie, Anna, Nadjejda, Sunkita, Katie and Esther, for the cheerleading and camaraderie.

Finally, to my wonderful partner Chris, the Clubb and Sparks families, and my friends Adam,

Brylie and Larissa, I would like to express my endless gratitude, love and thanks.

Table of Contents

Chapter Section Content Page Abstract 2 Acknowledgments 7 1 General Introduction, Literature Reviews and Hypotheses 16 1.1 Research question 16 a Rationale 16 1.2 Adaptive responses of marine invertebrates to ocean acidification and 17 warning a Ocean acidification and warming 17 b The effects of ocean acidification and warming on early life 19 stages of marine invertebrates c Adaptive capacity of marine invertebrates to ocean 22 acidification and warming d Trans-generational plasticity (TGP) and phenotypic plasticity 26 (PP) as strategies for short-term non-genetic adaptation to global ocean change e Quantitative genetics as a tool for assessing adaptive capacity 29 to ocean acidification and warming 1.3 Ocean acidification and warming in the southern ocean and Antarctic 32 marine ecosystems a Physical effects and biological responses 32 b Transgenerational plasticity and other adaptive responses to 33 ocean acidification and warming in Antarctic marine ecosystems 1.4 Physiological and bioenergetic responses of marine populations to 36 ocean acidification and warming over long timescales. a Allocation of metabolic energy within the bodies of marine 36 invertebrates b Energy storage in marine invertebrates, with a focus on 39 echinoderms c Physiological acclimation to ocean acidification and 41 warming; processes, timescales and vulnerabilities 1.5 Aims and hypotheses 45 2 Paternal identity influences response of Acanthaster spp. embryos 49 to ocean acidification and warming 2.1 Abstract 51 2.2 Introduction 52 2.3 Materials and methods 56 a Collection sites 56 b Experimental design 57 c Spawning and fertilization 59 d Experimental conditions 60 e Scoring development 61 2.4 Data analysis 62 2.5 Results 64 a Gastrula stage embryos 64

b Genetic correlations and heritability 69 2.6 Discussion 70 a Parental identity and embryo performance 70 b Effects of environmental variation on A. planci embryos 71 c Implications of adaptation for A. planci population dynamics 72 d Conclusion 73 3 Potential for adaptation in two sand dollar species, Arachnoides 74 placenta and Fellaster zelandiae, will be challenged by ocean acidification but not warming 3.1 Abstract 76 3.2 Introduction 76 3.3 Methods 81 a Animal collection 81 b Experimental temperature and pCO2 manipulations 82 c Spawning and fertilization 83 d Developmental scoring 85 3.4 Data analysis 85 3.5 Results 87 a Effects of environmental factors 87 b Heritability and genetic contributions 88 3.6 Discussion 90 a Parent identity and embryo performance 90 b Environmental selection 93 c Conclusions 95 4 Antagonistic effects of ocean acidification and warming on the 96 New Zealand common cushion sea star Patiriella regularis metabolic energy budget and reproductive capacity over a 12- month breeding cycle 4.1 Abstract 97 4.2 Introduction 98 4.3 Methods 105 a Animal collection and acclimation 105 b Experimental design and manipulations of temperature and 105 pH c Body size and tissue weight relationships 108 d Respiration rate and ammonia excretion 108 e Food absorption efficiency 109 f Righting time 110 4.4 Data analysis 111 4.5 Results 112 a Body weight indices 112 b Respiration rate 113 c Absorption and excretion of energy 116 d Scope for Growth 119 e Righting time 121 4.5 Discussion 122 a Mechanisms of maintaining Scope for Growth at high 122 temperature and low pH

b Ecological implications of acclimation to forecasted end-of- 126 century conditions c Conclusions 128 5 Physiological acclimation of the Antarctic cushion star, Odontaster 130 validus, to elevated temperature, but not low pH, over 12 months significantly alters metabolic energy budget 5.1 Abstract 132 5.2 Introduction 132 5.3 Materials and methods 137 a Animal collection and holding 137 b Experimental design and manipulations of temperature and 138 pH c Body size and tissue weight relationships 142 d Respiration rate and ammonia excretion 142 e Food energy absorption efficiency 143 f Righting time 144 5.4 Data analysis 144 5.5 Results 146 a Body weight indices 146 b Respiration rate 148 c Absorption and excretion of energy 148 d Scope for Growth 153 e Righting time 155 5.6 Discussion 156 a Physiological acclimation to ocean warming and acidification 156 b Energy allocation to reproductive output at high temperature 159 and low pH c Conclusions 161 6 Maladaptive trans-generational effects of exposure to low pH are 163 mitigated by both genetic and non-genetic factors for sea stars Patiriella regularis at elevated temperature 6.1 Abstract 165 6.2 Introduction 165 6.3 Methods 169 a Animal collection and acclimation to increased temperature 169 and low pH b Spawning and fertilization 170 c Experimental treatments and manipulations of temperature 171 and pH 6.4 Data analysis 172 6.5 Results 174 a Trans-generational plasticity 174 b Heritability and genetic contributions to tolerance 175 6.6 Discussion 180 a Maladaptive trans-generational plasticity 181 b Maternal effects 182 c Ecological implications of maladaptive plasticity 183 d Conclusions 184

7 Trans-generational plasticity in response to ocean acidification 185 and warming is strongly dependent on maternal identity in the Antarctic sea star Odontaster validus 7.1 Abstract 186 7.2 Introduction 187 7.3 Methods 190 a Animal collection 190 b Adult acclimation 191 c Spawning and fertilization 192 d Manipulations of experimental conditions 193 7.4 Data analysis 194 7.5 Results 195 a Trans-generational plasticity 195 b Heritability and maternal effects 197 7.6 Discussion 199 a Parent acclimation to elevated temperature 199 b Maternal investment in offspring 204 c Conclusions 205 8 Discussion: Standing genetic variation and trans-generational 206 plasticity among populations of echinoderms in response to ocean acidification and warming 8.1 Main findings relating to hypotheses 207 a Hypothesis 1 207 b Hypothesis 2 209 c Hypothesis 3 210 d Experimental limitations 212 8.2 Adaptive potential in invertebrates 215 a Adaptive capacity in response to multiple stressors 215 b The deterministic influence of parental environment and 219 physiological state c Wider context and recommendations for further work 222 8.3 Conclusions 225 9 References 227 10 Appendices 270 10.1 Appendices for chapter 2 270 a Detailed carbonate chemistry for experimental blocks 270 b Daily averaged pH over the course of the spawning and 271 experimental period c Water temperatures over the course of the experimental period 272 d Reference photos for developmental morphology of A. planci 273 embryos showing normal and abnormal development e Mean fertilization rates of A. planci embryos subjected to high 274 temperature and low pH f Results of PERMANOVA analysis performed on data from 275 gastrulae of Acanthaster spp. 10.2 Appendices for chapter 3 276 a Carbonate chemistry parameters for both A. placenta and F. 276 zelandiae experimental periods

b Results of PerMANOVA analysis on both A. placenta and F. 277 zelandiae embryo developmental data at gastrulation 10.3 Appendices for chapter 4 278 a Carbonate chemistry parameters throughout experimental 278 period b Body weight indices for adult P. regularis 280 c Ingestion, excretion and energy absorption of Parapercis 280 colias muscle tissue by Patiriella regularis 10.4 Appendices for chapter 5 281 a Body weight indices for adult Odontaster validus 281 b Feeding, excretion and energy absorption values 281 c Carbonate chemistry parameters 282 10.5 Appendices for chapter 6 283 a Contributions of fixed and random factors to variation among 283 F0 and F1 gastrulae of P. regularis 10.6 Appendices for chapter 7 284 a Contributions of fixed and random factors to variation among 284 F0 and F1 gastrulae of O. validus

List of Figures and Tables

Figure Title Page # 1.1 Thought process and conceptual outline for this thesis 47

1.2 Thesis workflow 48

2.1 Flowchart showing the experimental protocol for the modified North 58 Carolina II breeding design employed in this experiment to raise family lines of embryos of Acanthaster planci under different temperature and pCO2 regimes

2.2 The influence of temperature and the influence of a temperature x pCO2 65 interaction on gastrulation of Acanthaster spp. embryos subjected to different temperature and pCO2 regimes

2.3 Reaction norms for each of 12 sire and 6 dam Acanthaster spp. 66 genotypes across each temperature and pCO2 treatment at gastrulation

2.4 Log-response ratio for each of the 24 Acanthaster spp. genotypes (sire x 68 dam) at the gastrula stage

2.5 Relationships (percentage normal gastrulation) between embryo 69 performance at gastrulation for high vs. ambient temperature; high vs. ambient pCO2 ; high temperature vs. high pCO2 for the 24 Acanthaster spp. families examined in this study

3.1 The effect of elevated pCO2 on normal development of sand dollar 87 Arachnoides placenta 2-arm plutei

3.2 The effect of elevated pCO2 on mean normal development of sand 88 dollar Fellaster zelandiae 2-arm plutei

3.3 Reaction norms for sire x CO2 and dam x temperature for sand dollar 89 Arachnoides placenta at 2-arm pluteus stage

3.4 Reaction norms for male genotypes of sand dollar Fellaster zelandiae 90 across a CO2 gradient at 2-arm pluteus stage

4.1 Temperature and pH(T) in the aquaria used to acclimate adult Patiriella 107 regularis to all combinations of two temperature (ambient; elevated 2.5 °C above ambient) and three pH (ambient; 7.90; 7.70) treatments

4.2 Pyloric and gonad-somatic indices for Patiriella regularis adults 114 acclimated for 12 months to combinations of two temperature (ambient; elevated 2.5 °C above ambient) and three pH (ambient; 7.90; 7.70) treatments as well as the wild population sampled from Otago Harbour 4.3 Respiration rate of adult Patiriella regularis acclimated for 12 months 115 to combinations of two temperature (ambient; elevated 2.5 °C above ambient) and three pH (ambient; 7.90; 7.70) treatments as well as respiration rate in the wild population from Otago Harbour

4.4 Energy absorbed by adult Patiriella regularis exposed to combinations 117 of two temperature (ambient; elevated 2.5 °C above ambient) and three pH (ambient; 7.90; 7.70) treatments for 12 months as well as the wild population sampled from Otago Harbour

4.5 Nitrogenous excretion (NH3) by adult Patiriella regularis exposed to 118 two temperature (ambient; elevated 2.5 °C above ambient) and three pH (ambient; 7.90; 7.70) treatments for 12 months as well as the wild population sampled from Otago Harbour

4.6 Oxygen: Nitrogen ratios of adult Patiriella regularis over 12 months 119 exposure to combinations of two temperature (ambient; elevated 2.5 °C above ambient) and three pH (ambient; 7.90; 7.70) treatments as well as the wild population sampled from Otago Harbour

4.7 Scope for Growth Energy of adult Patiriella regularis after 12 months 120 exposure to combinations of two temperature (ambient; elevated 2.5 °C above ambient) and three pH (ambient; 7.90; 7.70) treatments as well as the wild population sampled from Otago Harbour

4.8 Turning time of adult Patiriella regularis over 12 months exposure to 121 two temperature (ambient; elevated 2.5 °C above ambient) and three pH (ambient; 7.90; 7.70) treatments as well as for a nearby wild population from Otago Harbour

5.1 Aquarium design for long-term acclimation of adult Odontaster validus 140 to elevated temperature and low pH

5.2 Mean weekly temperature and pH in aquaria over the 47-week 141 acclimation period

5.3 Gonad-somatic and pyloric indices for adult Odontaster validus 147 acclimated for 9 months to combinations of two temperature (ambient = 0.5 °C; elevated = 3.5 °C) and two pH (ambient = 8.15; low = pH 7.7) levels

5.4 Respiration rate of adult Odontaster validus acclimated for 9 months to 149 combinations of two temperature (ambient = 0.5 °C; elevated = 3.5 °C) and two pH (ambient = 8.15; low = pH 7.7) levels

5.5 Absorption of energy from food in adult Odontaster validus acclimated 151 for 9 months to combinations of two temperature (ambient = 0.5 °C; elevated = 3.5 °C) and two pH (ambient = 8.15; low = pH 7.7) levels

5.6 Nitrogenous excretion in adult Odontaster validus acclimated for 9 152 months to combinations of two temperature (ambient = 0.5 °C; elevated = 3.5 °C) and two pH (ambient = 8.15; low = pH 7.7) levels

5.7 Oxygen: Nitrogen (O:N) ratio in adult Odontaster validus acclimated 153 for 9 months to combinations of two temperature (ambient = 0.5 °C; elevated = 3.5 °C) and two pH (ambient = 8.15; low = pH 7.7) levels

5.8 Scope For Growth (SfG) in adult Odontaster validus acclimated for 9 154 months to combinations of two temperature (ambient = 0.5 °C; elevated = 3.5 °C) and two pH (ambient = 8.15; low = pH 7.7) levels

5.9 Righting time of adult Odontaster validus acclimated for 9 months to 155 combinations of two temperature (ambient = 0.5 °C; elevated = 3.5 °C) and two pH (ambient = 8.15; low = pH 7.7) levels

6.1 Normal development of F1 gastrulae is dependent on both parent and 176 offspring pH environment in sea star Patiriella regularis

6.2 Reaction norms for female Patiriella regularis genotypes from a wild 177 population from Otago Harbour, NZ across a temperature gradient and a pH gradient

6.3 Reaction norms for female Patiriella regularis genotypes acclimated for 178 12 months to the treatment described along the top of the graph columns

6.4 Reaction norms for male Patiriella regularis genotypes acclimated for 179 12 months to the treatments described along the top of the graph columns

6.5 Normal development at gastrulation for Patiriella regularis gastrulae is 180 dependent on the interaction between sire and dam identity across 144 families

7.1 Normal development of F1 Odontaster validus gastrulae is dependent on 196 both parent and offspring environment in response to elevated temperature and low pH

7.2 Dam identity interacts with temperature to influence offspring 198 performance among populations of Odontaster validus

8.1 Thesis workflow as it pertains to hypotheses 206

Table Title Page 1 Summary of results from Hypothesis 1 207 2 Summary of results from Hypothesis 2 209 3 Summary of results from Hypothesis 2 210

Chapter One: Literature Review, research aims and conceptual

structure

1.1 Research Question

Do populations of phylogenetically-related marine echinoderms from different latitudes and habitats possess the capacity to adapt to combined ocean acidification and warming? If so, what are the relative contributions of heritable plasticity and environmental exposure history of parents to this adaptive capacity?

a) Rationale

The rationale behind the current research question is that as more studies focus on the responses of echinoderm larvae to ocean acidification and warming, the question of adaptation versus acclimation becomes more relevant (Chirgwin et al. 2015; Donelson et al.

2017). It is important to know which species possess enough standing genetic variation that a sub-set of ‘tolerant’ individuals may become selected for in a near-future warm and acidified ocean scenario, and which species may become more vulnerable with each successive generation. Marine echinoderms are often referred to as ‘keystone’ or ‘critical’ species and large echinoderm populations can be indicative of disturbance in marine ecosystems.

Echinoderms under outbreak conditions can display rare strong effects in terms of ecological interactions. It is therefore relevant to characterise their responses as indicators of ecosystem function.

The Antarctic (and the Southern Ocean) have been called variously the ‘bellwether of change’ and the ‘canary in the coal mine’ – meaning that the effects of ocean acidification will be visible in Antarctic seawater before they affect lower-latitude oceans (Convey et al.

2003; McNeil and Matear 2008; Fabry et al. 2008; McClintock et al. 2011; Lenton et al.

2011). If the responses of marine species in the Antarctic environment are similar enough to those of phylogenetically related species in other ecosystems, it may be possible to predict in what ways temperate ecosystems will be affected, and therefore potentially put in place conservation and management measures. This will, however, only be possible if responses of

Antarctic species are very well-characterised, and if response types and mechanisms are conserved across phylogenetic and latitudinal gradients. It is therefore necessary to study

Antarctic and temperate species in tandem when addressing questions of environmental change.

1.2 Adaptive responses of marine invertebrates to ocean acidification and

warming.

a) Ocean acidification and warming

Human activities including burning of fossil fuels, deforestation, intensified agriculture and cement production have increased modern-day atmospheric CO2 levels to 400 ppm, up to 40

% higher compared to pre-industrial revolution levels (Raven et al. 2005; IPCC 2007;

Kutzbach et al. 2010). At the same time, global mean surface temperatures have increased by up to 1.5 °C (IPCC 2013). Up to 40 % of anthropogenic carbon dioxide is absorbed by the oceans, and the Southern Ocean in particular has been identified as a carbon sink (Sarmiento

17 et al. 1998). The mean pH of the ocean has dropped by approximately 0.1 pH units since the industrial revolution and is predicted to decrease by a further 0.3-0.5 pH units globally by the end of the 21st century (Feely et al. 2009; IPCC 2013). Absorption of atmospheric carbon dioxide by the global ocean initiates a chemical reaction which first forms carbonic acid

- + (H2CO3) which then dissociates to form bicarbonate ions (HCO3 ) and a hydrogen ion (H ).

The increase in concentration of hydrogen ions lowers seawater pH and concentration of bicarbonate ions, which are biologically important for mineralization and growth in a range of marine taxa. Low seawater pH and bicarbonate ion concentration can lead to changes such as lowering calcium carbonate saturation states, altering calcite/ aragonite ratios in seawater and ultimately changes to shell mineralogy for marine calcifiers which can increase their vulnerability to diseases and result in higher mortality rates (Ries 2011).

Carbon dioxide (CO2) is a greenhouse gas, so ocean warming is concurrent with acidification; as the upper ocean is a heat-sink for roughly 90 % of the heat energy generated by the Earth

(Bindoff et al. 2007; Levitus et al. 2012), and the increased absorption of carbon dioxide into the ocean as the temperature continues to warm creates a positive feedback-loop of warming and acidification (Cox et al. 2000; Fung et al. 2005). Decreased productivity of marine food webs, reduced species abundance and local extinctions, latitudinal range-shifts, increases in incidence of parasitism, disease and mortality, and changes to habitat use are among the known impacts of ocean acidification and warming on marine ecosystems (Fabry et al. 2008;

Hoegh-Guldberg and Bruno 2010; Pandolfi et al. 2011; Kroeker et al. 2013; Kroeker et al.

2017).

Ocean warming and acidification, although global in scope, do not affect all regions of the ocean with the same speed and severity. The upper 700 m of the ocean is the fastest-warming,

18 while the polar latitudes are the fastest acidifying (McNeil and Matear 2008; Lyman and

Johnson 2014), creating local heterogeneity in both the physical processes and their biological effects. Some areas, such as eastern Australia, Antarctica and the Southern Ocean, have been called ‘climate change hotspots’ or ‘bellwethers of change’ as their biogeochemical and physiochemical properties, and important role in the global biological carbon pump, make them likely to be among the first places to experience permanent change

(Fung et al. 2005; Fabry et al. 2009; Hobday and Lough 2011; Wenberg et al. 2011; Hobday and Pecl 2013; Hauck et al. 2015).

b) The effects of ocean acidification and warming on early life stages of marine

invertebrates

Ocean acidification and warming pose serious threats to marine organisms, especially those that utilise some form of calcium carbonate in their bodies. This group covers a range of taxa, but current research has focused on a few main groups: echinoderms, corals, molluscs and fish. Marine calcifiers rely on calcium carbonate to build and maintain structures such as shells, plates and teeth. The importance of carbonate in marine ecosystems, in particular the

CaCO3 polymorphs calcite and aragonite, has been widely demonstrated (e.g. by Feely et al.

2004; Orr et al. 2005; Hoffmann et al. 2010; Ries 2011). Dissolution of calcareous structures occurs when surface waters become under-saturated with respect to aragonite, which is linked to ocean acidification and has been suggested to occur before the end of the 21st century, although sooner at high latitudes (Orr et al. 2005). The rates of change associated with climate-forced biological processes are unprecedented, although somewhat altered by organism resistance and resilience (Brierley and Kingsford 2009).

Aragonite undersaturation negatively affects pelagic marine species such as pteropods, which exhibit significant developmental abnormalities including lack of calcification or cracked shell formation when grown in seawater undersaturated with aragonite (Comeau et al. 2010;

Bylenga et al. 2017). Analysis of a global dataset shows that up to 25 % of benthic marine communities experience aragonite undersaturation, which will increase to 57 % as ocean acidification continues, but benthic fauna appear acclimatised to their low-aragonite conditions and are able to survive the low carbonate and magnesium saturation states they experience (Lebrato et al. 2017). This indicates that adaptation to low carbonate saturation states is possible, although only in specific environments.

There is a rapidly-expanding body of literature that has identified that early embryonic and larval life history stages are the most vulnerable to the negative effects of ocean warming and acidification, particularly species with complex life histories including external fertilization and long duration pelagic larval stages (Pechenik 1987, 1991; Kroeker et al. 2013). Reviews including those by Dupont et al. (2010), Hoffmann et al. (2010), Kroeker et al. (2010), Byrne

(2011, 2012), Byrne and Przeslawski (2013), Przeslawksi et al. (2015), cover this subject in detail and provide comprehensive overviews of the current literature. General trends emerging from these meta-analytical papers include the resilience of fertilization and gamete interactions to ocean acidification and warming, high mortality and developmental abnormality in early larval stages, and negative carry-over effects from one stage to the next, sometimes called ‘downstream effects.’ These effects are often species-specific, meaning that close analysis is needed to compare how these broad trends translate into local ecosystems.

Effects of larval exposure to ocean acidification and warming are often interactive: synergistic, additive, or antagonistic. The observed reaction to a combination of stressors

20 often depends to a large extent on the exact stressor levels used, exposure time, ontogeny and species susceptibility. For example, fertilization of gametes from the sea urchin Heliocidaris erythrogramma is robust to a reduction in pH of 0.5 units (from pH 8.2 to 7.6), but development at the early juvenile stage is sensitive to changes in pH (from 8.1 to 7.4) and +4

°C warming in terms of skeletal development, survival and growth (Byrne et al. 2009, 2010,

2011; Wolfe et al. 2013). Likewise, a pH reduction from 8.10 – 7.79 warming did not affect larval mortality but a further reduction to pH 7.65 and 7.12 significantly increased mortality in the sand dollar Arachnoides placenta, and there was no interactive effect of ocean acidification and warming in this species (Gonzalez-Bernat et al. 2013; Karelitz et al. 2017).

Subtle non-lethal consequences of exposure to multiple stressors are hard to tease out due to the mechanisms of tolerance such as physiology, adaptation and plasticity (Gienapp et al.

2008; Somero 2011, 2012; Dam 2013), often made even harder to detect by the interaction between individual variation, local larval dispersal strategies and mortality bottlenecks

(Byrne 2011; Nanninga and Berumen 2014).

Experimental approaches that utilise different methodologies and treatment levels often produce different results, but inter-individual differences can also influence the results of studies on the same species. Although fertilization success at a population-level has been shown to be robust to ocean acidification, the success of gamete interactions differs between individuals, indicating that pair compatibility is an important determinant of successful genotypes in the context of responding to pH changes (Schlegel et al. 2012). Discrepancies among published literature studies often stem from differences in the methodology of individual researchers as well as inter-individual differences. Recent attempts to standardize the experimental approach, and recommendations of best practice for ocean acidification experiments, aim to minimise these differences to foster a more accurate and coherent

21 literature landscape (Dickson et al. 2007; Riebesell et al. 2010; Cornwall and Hurd 2016;

Watson et al. 2017).

c) Adaptive capacity of marine invertebrates under ocean acidification and warming

Ocean warming and acidification can impact the evolutionary potential of a population in a localised environment by both (a) presenting as a new selection pressure and (b) altering existing selection pressures (Chirgwin et al. 2015). Species response to climate change, if any, has been categorised into three distinct ‘options’ – (a) incremental acclimation a.k.a. phenotypic plasticity, allowing the population to persist in their current environment; (b) adaptive evolution, or (c) migration to a more favourable habitat (Holt 1990; Davis et al.

2005; Dam 2013). The inability to demonstrate one of these responses may increase a populations’ potential for local extinction.

The extent to which ocean acidification and warming will impact a population depends on both the standing genetic variation within a population as well as that populations’ ability to adapt (Sunday et al. 2011; Dam 2013; Foo and Byrne 2016). Environmental stress, viewed as a selection pressure, can increase both the genetic heritability and magnitude of stress- response traits (Jenkins et al. 1997) and therefore we expect ocean acidification and warming to produce heritable plastic responses from marine populations. The rates of change associated with ocean warming and acidification are rapid compared to rates of gene flow and mutation among marine invertebrate populations (Parmesan 2006; Kelly and Eernisse 2007).

Mutational (a.k.a. Darwinian) evolution may not be sufficient to keep pace with environmental change (Bürger and Lynch 1997) therefore the existence of heritable plasticity may become increasingly important in determining future phenotype (Jenkins et al. 1997;

Runcie et al. 2012; Collins et al. 2013). For example, heritable plasticity of cell division rates among microbial Ostreococcus spp.communities in response to ocean acidification (pCO2 levels 430, 1000 ppm) predicted the extent of adaptation and the extent of phenotypic evolution of a slow growth rate (Schaum and Collins 2014).

Latitudinal range shifts are one adaptive response that is consistently correlated with present- day or predicted ocean warming in over 80 % of marine species studies (Poloczanska et al.

2013). Range-restricted marine populations, such as those with narrow thermal tolerances or those already inhabiting polar regions, cannot take advantage of colder temperatures and therefore those populations must be able to adapt in order to survive. Some populations, despite being able to migrate, are not able to do so fast enough and therefore face the same adaptive challenges as polar species (Menéndez et al. 2006; Hoegh-Guldberg et al. 2007,

2008). Despite the relatively common observances of range shifts in response to ocean warming, there is little evidence that this will mitigate long-term negative effects at the species level, meaning that even migratory species will be subject to the selection pressures associated with ocean warming and acidification in the future (Parmesan 2006).

There are many types of inter-generational inherited effects which alter the phenotype of the offspring but are not DNA-based, some of which are discussed in this study. Possibly due to the difficulties in distinguishing true adaptive capacity from phenotypic plasticity (Gienapp et al. 2008), functional-genetic studies dealing with responses to acidification remain scarce.

Transcriptomic and heritable-trait studies that test the importance of parental influence can help to rectify this problem. One type of such trans-generational effects is known as

‘maternal effects,’ and ‘maternal provisioning’ which encompasses all material passed from

23 the mother to her offspring which increase its chances of survival. This is not limited to genetic material but also includes changes to the density of organic compounds like proteins and lipids which form the food source for the developing embryo, mRNA transcripts, and the rate at which the egg develops. It is known that maternal lineage affects egg quality and development, as the composition of the yolk sac, density, peptide mass, jelly coat size and toposome structural integrity are all factors in the survival of the larva (Perera et al. 2004;

Harvey et al. 2014; Foo 2015). Maternal effects have been detected in response to ocean acidification and warming in echinoderms such as Sterechinus neumayeri (Lister et al. 2017) and Dendraster excentricus (Chan et al. 2011) with the conclusion that maternal lineage is a significant contributor to differences in larval survival between individuals. This type of transgenerational effect is evident both before (Foo et al. 2012) and after (Chan et al. 2011) the MZT (maternal-zygotic transition) which is around gastrulation in echindoerms (Tadros and Lipshitz 2009). After the start of zygotic transcription,non-genetic maternal effects may be expected to decrease. The downstream effects of maternal lineage, however, still affect metrics such as growth, skeletogenesis and survival after the MZT, although environmental effects can take a more prominent role (Guillaume et al. 2016).

Another type of trans-generational effect is ‘epigenetics.’ Evolutionary fitness has traditionally been assumed to be wholly genetically-encoded. However, recent work on epigenetic inheritance has determined that sperm phenotype, which can be environmentally determined, is also an important transmitter of offspring fitness (Marshall 2015). Epigenetic effects are a subsection of trans-generational effects which are heritable, without altering the

DNA of an individual but still having an effect on the genome, such as DNA methylation, chromatin and histone modification, or small RNA interference (Feil and Fraga 2012; Burton and Metcalfe 2014). Epigenetic effects can be important mechanisms for climate change adaptation, as epigenetic effects should favour the local environment and could emerge as a

24 beneficial source of current plasticity and future adaptive potential in marine ecosystems

(Webster et al. 2013; Putnam et al. 2016). This thesis will not be examining epigenetic mechanisms, however as they are assumed to be happening constantly in natural situations they cannot be ignored in the types of experiments reproduced here.

Epigenetic evolutionary effects are understudied with respect to climate change in marine ecosystems, but have been proposed as a method for ‘assisted evolution’ and are an important source of phenotypic plasticity in response to ocean acidification and warming (van Oppen et al. 2015; Calosi et al. 2016; Hofmann 2017). The evolutionary implications of non-genetic inheritance pertaining to climate-change adaptation in marine invertebrates are largely undetermined (Bonduriansky and Day 2009; Bonduriansky et al. 2012; Crean et al. 2013), but some more sensitive scleractinian corals are able to modify their phenotype via DNA methylation but exhibit reduced calcification at low pH compared to less sensitive species

(Putnam et al. 2016). Recent advances including methylome sequencing of marine invertebrates has begun to detail the complex relationship between epigenetic and genomic controls on plasticity within marine ecosystems (Gavery and Roberts 2013). Epigenetic responses, which have been sometimes been seen as ‘noise’ in population-level experiments, could facilitate short-term plasticity in species with high adaptation potential as has been demonstrated in the Antarctic polychaete Spiophanes tcherniai in response to elevated temperature (Marsh and Pasqualone 2014).

Both sire and dam identity may influence success at gastrulation, as well as post-gastrulation development, as was shown by the response to both warming and acidification in the urchin

Centrostephanus rodgersii (warming +2.3 °C and +4°C; pH -0.26 and -0.46 units; Foo et al.

2012). Physiological tolerance to ocean acidification was therefore deemed an inheritable trait, although evidence of gene-level change (i.e. adaptation) was not conclusive due to intra- population phenotypic plasticity. The extent and magnitude of ocean acidification as an evolutionary selection pressure is therefore uncertain. More types of non-genetic inheritance, which are used in this thesis, are discussed in the following section 1.2d.

This field appears to be technology-driven, with novel developments in gene-sequencing and the “-omics” technologies allowing advancements in our understanding of the genetic basis for adaptation for climate change. Although fields such as transcriptomics can tell us a lot about biochemical responses to environmental change, the number of studies taking advantage of this opportunity are relatively small, and tend towards examination of only one environmental factor (Evans et al. 2015). Use of model organisms such as the sea urchin

Strongylocentrotus purpuratus, which has been used in a number of genomic studies to date, result in well-characterised responses for some species, while other taxa have few genetic studies.

Predicting evolutionary adaptation is an important challenge, especially in the face of potentially huge biodiversity losses (Byrne 2011; Wernberg et al. 2011). Recent studies have found genetic bases for many of the physical changes observed in response to ocean acidification and warming, such as the up-regulation of metabolic genes, and the down- regulation of calcification genes in Strongylocentrotus purpuratus raised under combinations of high temperature and high pCO2 (Todgham and Hofmann 2009; Stumpp et al. 2011;

Pespeni et al 2013). Adaptive capacity at the population level will rely not only on gene expression but also on heritable non-genetic plasticity, which may provide a mechanism for

26 survival in populations that do not currently possess the genetic variation required to selectively adapt to rapid change (Donelson et al. 2017).

d) Trans-generational plasticity (TGP) and phenotypic plasticity (PP) as strategies for

short-term non-genetic adaptation to global ocean change

Phenotypic plasticity is a broad term that defines the adjustment of phenotype, based on genotype, in response to external (i.e. environmental) change (Crozier and Hutchings 2014).

Different phenotypes can be produced from the same genotype in response to different environments (Schlichting and Piggliucci 1998). This phenotypic variation is important in the context of global climate change as in a rapidly changing environment, mean population phenotype often ‘lags’ behind the ‘optimum’ phenotype for a specific niche (Lande 2009) with natural selection always “playing catch-up” to minimise the lag and increase fitness in response to a new/altered environmental niche (as modelled by Bürger and Lynch 1997). The magnitude of the lag can determine the species’ vulnerability to extinction. To interpret

Bürger and Lynch’s lag-fitness model in the light of individual phenotypic evolution, as long as the environmental change is directional and the lag is always selecting towards an optimum, then rapid phenotypic plasticity is constantly occurring at the individual level.

There is some uncertainty over the contributions of genetic variation and phenotypic plasticity to metabolic resilience, but it seems clear that any negative trans-generational effects of compromised reproductive capacity among marine populations will affect all individuals within a population to some extent (Pan et al. 2016; Lee et al. 2017). Although phenotypic plasticity has been suggested to be very important both in terms of adaptation to ocean warming and acidification (Gienapp et al. 2008) and ‘evolutionary rescue’ (Kopp and

Matuszewski 2014), there are instances where plasticity can be maladaptive. There are limits to the plasticity of some traits, especially those related to physiology and maintaining homeostasis e.g. body size (Ghalambor et al. 2007; Chevin et al. 2014). There are also limits to the effective timescales in which plasticity can impact population fitness (Nussey et al.

2007). High plasticity can be detrimental to population survival if it lessens the effectiveness of natural selection on adjusting mean phenotype to fit an altered niche (Morris and Rogers

2013; Nunney 2016), as has been observed for the Sydney rock oyster (Parker et al. 2017).

Phenotypic variation is proportional to the effective population size i.e. the population size needed for adaptation, assuming random mating within a wild type (Wright 1931). Population genetics often show a strong correlation between genetic variation over a wide geographic area, and life history strategies (Hart and Marko 2010; Keever et al. 2013). When Ne is small, genetic drift constrains adaptive response and therefore phenotypic adaptive capacity. Marine populations tend to possess high genetic and phenotypic variation (Reusch 2014), and therefore should have high adaptive potential. Plasticity was always assumed to be non- adaptive (e.g. by Visser 2008) and therefore an inadequate response to climate change, however, researchers such as Lande (2009) argued that phenotypic adaptation is possible in response to very rapid environmental change, followed by slower genetic ‘assimilation’ which comprises the underlying evolutionary component of adaptation, and this idea is becoming more prevalent in the climate change literature (Charmantier et al. 2008; Gomez-

Mestre and Jovani 2013; Merilä and Hendry 2014; Forsman 2015). Genotypic variation within a population is therefore likely to determine the magnitude and direction of evolutionary adaptation (Hughes et al. 2008).

Trans-generational plasticity (sometimes called Anticipatory Parental Effects (APEs), carry- over effects or multi-generational effects) (TGP) is the alteration of offspring (F1) phenotype in response to parent (F0) environment. It is a type of non-mutational genetic (i.e. epigenetic) process that can be influenced by environment as well as standing genetic variation.

Detecting adaptive trans-generational plasticity can be difficult, and the true extent of its influence is largely unknown, as only weak trans-generational effects in response to parental environmental heterogeneity are detectable across multiple marine and terrestrial taxa or may be underestimated or overestimated if the environmental effects on each successive generation are not additive (Shea et al. 2011; Uller et al. 2013; Engqvist and Reinhold 2016).

Of the observed types of plasticity, the most easily detectable are maternal effects such as provisioning, but the fitness consequences of maternal provisioning in their environmental context are not always clear (Räsänen and Kruuk 2007; Uller 2008; Shama et al. 2014;

Shama 2015; Malvezzi et al. 2015). It is also difficult to separate out APEs from parent condition-transfer effects which occur when the metabolic condition of the parent is compromised, preceding a negative trans-generational carry-over which is not necessarily genetic, nor is it true ‘maternal provisioning’ in the traditional sense of the word

(Bonduriansky and Crean 2017). Selection pressure acting on both the mother and the offspring in different ways may mean that TGP is maladaptive in some scenarios, and the costs to parents associated with alterations in egg energy or size may be too great, resulting in a reduction in offspring fitness in order to maintain parent survival which is termed a ‘selfish parental effect’ (Marshall and Uller 2007; Burgess and Marshall 2014; Jensen et al. 2014).

Any trans-generational effect that modifies the phenotype of F1 and subsequent marine larvae will also affect the phenotype of later ontogenetic stages, meaning that parental modification of offspring phenotype can have whole-life ramifications (aka ‘latent’ trans-generational

29 effects) (Pechenik 2006). This process is especially relevant when considering that the maladaptive plasticity seen in response to ocean warming in some species such as oysters may not be able to be modified (Morris and Rogers 2013; Parker et al. 2017). A critical consideration of experiments using environmental mismatch in order to identify TGP is that parent environment is only a good predictor of offspring performance when they will both experience the same type of environment (Marshall 2008). For K-strategist species with slow generational times typical of polar populations, this type of plasticity may not be adaptive and therefore the evolutionary implications of this type of plasticity can be of greater importance to survival than slower mutational evolution.

e) Quantitative genetics as a tool for assessing adaptive capacity to ocean acidification

and warming

Quantitative Genetics (QG) is one experimental approach used to measure phenotypic and genotypic diversity within a population that partitions trait expression into heritable and non- heritable components. QG practical technique often relies on variations of the North Carolina

II breeding design (Lynch and Walsh 1998). This design uses a system of breeding crosses, in which adults are crossed in all combinations, to give a range of families in the F1 generation which are either paternal or maternal half-sibs. Diallel crosses such as these have been called the ‘most direct way to estimate heritable variation in traits that are relevant to climate change adaptation’ (Munday et al. 2013). Both additive and non-additive effects of environmental selection on functional traits can be detected using this type of experimental design (Guinand et al. 2017). Using a modified factorial ANCOVA (analysis of co-variance) model, where the sire effect mean square is divided by the interaction mean square, the North

Carolina II design allows calculation of the additive genetic variance, (i.e. variance due to sire identity) which is assumed to be purely genetically-inherited.

Individual phenotype of an F1 generation individual (the kth offspring of an i x j mating where i=sire and j=mother) is defined as:

Z (ijk) = µ + S(i) + d(j) + I (ij) + e (ijk) where µ=mean population phenotype; S(i) and d(j) are the additive genetic effects of the ith sire and the jth dam (i.e. the ‘breeding values’ of the sire and dam); I(ijk) is the non-additive

(interaction) effect due to the combination of genes from parents; e(ijk) is the deviation of the observed phenotypes of the offspring from the models’ prediction (see Lynch and Walsh

1998 for more detail).

QG experiments generate reaction norms. Reaction norms are the usual way to determine adaptive fitness, as they represent a straightforward option for quantifying the interaction between genotype and environment and hence testing for Genotype x Environment (G*E) interactions (Nussey et al. 2007). They are interaction plots of change in a fitness-correlated trait (i.e. genotype) across a continuous environmental gradient (Pigliucci 2001). The slope of the line in these plots determines the phenotypic plasticity of an organism, where a straight line (slope = 0) denotes ‘phenotypic buffering’ (sensu Waddington 1942) and the steeper the slope of the reaction norm, the stronger the selection potential of that genotype. For worked examples of reaction norms, see work by Foo et al. (2012; 2014) and for theoretical examples see the work of Lande 2009; Chevin et al. 2014; Reusch 2014.

These plots usually represent linear (i.e. first order) responses, although some reaction norms, for example thermal envelope curves, model fitness across an environmental gradient representing the whole adaptive geographic range of a species’ distribution and therefore generally take a ‘bell curve’ shape. The non-zero slope of a reaction norm denoting phenotypic plasticity can also be seen as representing the genetic variation underlying plasticity, and therefore the effect of natural selection on genotypes is represented by the slope and variation among reaction norms (Schlichting and Pigliucci 1998; Whitman and

Agrawal 2009; Dam 2013).

1.3 Ocean acidification and warming in the Southern Ocean and Antarctic

marine ecosystems

a) Physical effects and biological responses.

The Southern Ocean and southern polar seas are especially sensitive to the negative effects of ocean acidification and warming (IPCC 2013). It is estimated that the Southern Ocean has absorbed 43 % of anthropogenic carbon dioxide and may be responsible for up to 75 % of global ocean heat uptake since the year 1861 (Frölicher et al. 2015). Rates of environmental change in the Antarctic are often faster than those recorded in other ecosystems across multiple parameters (Sancho et al. 2007; Peck et al. 2009; Constable et al. 2014; Gutt et al.

2015). Seasonal sea surface temperature and carbonate variation is also larger than for lower latitudes, for example the predicted winter under-saturation of aragonite in the Southern

Ocean is rapidly becoming one of the largest problems in the polar ocean and may affect ~85

% of the Southern Ocean within the next century (Christensen 2007; McNeil and Matear

2008; Hawkins and Sutton 2012; Constable et al. 2015; Gutt et al. 2015). Aragonite undersaturation is directly linked to sea ice melting, and changes to polar upwelling has increased circulation of low-aragonite, low-salinity seawater into the surface layers of the polar ocean in the Arctic Circle (Jutterström and Anderson 2005; Yamamoto-Kawai et al.

2009). The fast rate of change coupled with high climate variability mean that Antarctic biota are likely to experience tangible negative consequences before their counterparts in other regions.

Current research into the effect of ocean acidification and warming in Antarctic marine ecosystems includes: (a) investigation of the existence of environmental thresholds or

‘tipping points’ beyond which certain species would be subjected to irreversible physiological stress and mortality; (b) analysis of cellular, biochemical and organism level effects of ocean acidification on a wider range of species than has been studied, including non-calcifiers, different ontogenetic stages and climate change ‘winners’ as well as ‘losers’; and (c) the identification of ‘refuge ecosystems’ – areas with favourable conditions where animals forced outside their physiological tolerance zones could find respite (Dupont and

Thorndyke 2009). The response of early life-history stages (gametes, embryos, larvae and juveniles) to climate change in the Antarctic will be key to shaping future marine populations in the region. Antarctic fauna may be especially challenged due to their cold-adapted and stenothermal physiology, and the fact that polar seas are changing at a greater pace than at lower latitudes (Clark et al. 2007; Mulvaney et al. 2012).

The Antarctic seas are an environment that, as they warm and acidify, may present an even greater-than-usual physiological challenge (Feely et al. 2004). Long-term exposure to ocean

33 acidification and warming in Antarctic marine species has been shown to be deleterious to population reproductive capacity, physiology and population structure. For example, the urchin Sterechinus neumayeri was able to acclimate metabolic needs to a decrease in seawater pH of -0.5 pH units and a concurrent +2 °C temperature increase for 12 months, but reproductive capacity was compromised (Suckling et al. 2014). Conversely, metabolic rates increased in the Antarctic dragonfish Gymnodraco acuticeps when exposed to elevated water temperatures of +3 °C without any apparent metabolic compensation, resulting in increased thermal sensitivity of embryos that developed at high temperatures (Flynn and Todgham

2017).

b) Trans-generational plasticity and other adaptive responses to ocean acidification and

warming in Antarctic marine populations

Pelagic marine environments are likely to become warmer, more acidic and fresher over the next 100 years, and may provide refugia for poleward moving allochthonous marine populations as their current habitats warm beyond their tolerance limits (Walther et al. 2002;

Cheung et al. 2009; Constable et al. 2015; Deutsch et al. 2015). Not having the option to shift their range, polar marine species such as the sea star Asterias amurensis (Byrne et al. 2016) and many commercially-important species (Woodsworth-Jefcoats et al. 2017) must adapt to warming and acidification as it proceeds, or perish (Mieszkowska and Sugden 2016). This means that Antarctic populations must adapt to the predicted warming conditions for this century (+2.5°C this century; RCP 8.5; IPCC 2013) while remaining in the same geographical area. Their responses, therefore, must be looked at in terms of potential for adaptation and trans-generational effects resulting from adult exposure to acidification and warming over the life history of one generation. Adult Antarctic urchins Sterechinus

34 neumayeri can tolerate increased temperature and decreased pH (Morley et al. 2016) and larvae can tolerate a thermal range of -1.1 – 3.5 °C and a pH(T) range of 8.2 – 7.8 (Foo et al.

2016; Karelitz et al. 2017), this species will need to rely on existing phenotypic plasticity in the short term, rather than adaptation, to continue to tolerate these conditions, as they don’t have a wide range of tolerant genotypes in response to ocean acidification and warming (Foo et al. 2016). Antarctic marine populations have adapted to a stable environment over long time-scales (Clarke 1983, 1988, 1990; Peck and Conway 2000; Peck 2002; Peck et al. 2004) and therefore the same survival strategy has become commonplace across the whole population (Dunbar 1960; Clarke 1991; Tyler et al. 2000; Somero 2010), i.e. there is little phenotypic variation within these populations.

An exception to that assumption and example of an Antarctic species with a relatively wide thermal niche is the sea star Odontaster validus. O. validus may be one of the more thermotolerant Antarctic marine invertebrate species, with larval developmental thermal windows and thermal limits to normal adult physiology well characterised, with short term adult survival seen over a broad temperature range (0 - 9.0 °C Peck et al. 2008 17). In addition, work by Karelitz et al. (2016) characterised the thermal development window of the Antarctic echinoderms O. validus and S. neumayeri and demonstrated that the thermal windows of early developmental stages were not affected by exposure to low pH, and that O. validus embryos remained viable up to 8.5 °C during short-term (48 hr) experiments.

There are still uncertainties around the mechanisms whereby polar species have adapted, for example in terms of protein folding and genetic duplication, but some physiological functions of Antarctic marine species, including acclimation to temperature change, are significantly

35 slower than would be expected were the acclimation period determined completely by thermal environment (Peck 2016). This knowledge is the first step towards research that can link acclimation to rapid change into the physiological state of the adult population. In addition, the slower-than-expected thermal acclimation rates among polar marine species opens the question of whether trans-generational effects are able to compensate for such slow acclimation, and whether trans-generational plasticity is as viable and important in Antarctic species compared with temperate species and marine vertebrates (Donelson et al. 2017).

1.4 Physiological and bioenergetic responses of marine populations to ocean

acidification and warming over long timescales.

a) Allocation of metabolic energy within the bodies of marine invertebrates

Metabolic functions have been identified as a ‘unifying physicochemical parameter’ that

provides a good way to test the effects of hypercapnia and warming regardless of taxa and

environment (Pörtner 2008). Factors including energy intake, usage and allocation to

somatic growth, reproductive function and basal metabolism within the body can be

modelled, and significant changes to energy usage are indicative of internal homeostatic

disruption.

We can model the increased energetic demands that ocean acidification and warming may

have on marine invertebrates using a variety of experimental approaches (Stumpp et al.

2013; Vargas et al. 2013), such as Dynamic Energy Budget (DEB) modelling (Koojiman

1986; Agüera et al. 2017), aerobic-metabolic scope (e.g. Gräns et al. 2014; McArley et al.

2017) and Scope For Growth (SfG) (e.g. Stumpp et al. 2011). These approaches model growth and metabolism in the context of their environment including food supply, temperature, movement and stress tolerance using ratios of allometric measurements and partitioning of resource allocation. The estimation of change in these ratios when animals are exposed to ocean acidification and warming scenarios has become more widespread over the last decade, mostly through long-term acclimation studies on commercial species such as the Pacific oyster Crassostrea gigas (Bacher and Gangnery 2006; Pouvreau et al.

2006). Despite considerable intra-specific variation in the components of energy budget, simulations using physiological modelling approaches are generally accepted as capable of modelling population energetics in response to ocean acidification and warming (Ren and Schiel 2008; Gray et al. 2017).

In this study, I elected to use Scope for Growth as the method of estimating metabolic energy budget. SfG is easily-comparable throughout the literature and has been validated as a method of comparing between species and environments (Bayne and Widdows 1978;

Maltby et al. 1990). The Winberg equation (Winberg 1956), from which SfG is derived, is a measure of the proportion of consumed energy that is utilised by the body. Energy consumed during feeding is assimilated into the body and utilised for various functions including gonad development, somatic growth, metabolism, digestion and excretion. SfG is defined as ‘the difference between the energy input to an organism from it’s’ food and the output from respiratory metabolism’ (Naylor et al. 1979). It is essentially the energy available for somatic growth left over after ingestion, digestion and excretion have taken place, and can be used for processes such as feeding, gonad growth or locomotion.

The Winberg equation states that:

C – F = A +R+U+P

Where C= food energy consumed

A= energy assimilated by the individual (C x absorption efficiency)

F= energy lost in faeces

R= energy lost in respiration

U=energy lost in excretion of liquids

P= energy devoted to somatic or gonadal growth (Production)

Scope for Growth is related to the Winberg equation thus (Widdows and Johnson 1988);

SfG = C – F - R

If SfG is a positive value, energy is available for growth. This is expected in a healthy organism. However, if SfG is negative, then energy expenditure is greater than energy consumed, and individuals must use their stored energy reserves to complete necessary functions. This may mean taking energy away from gonadal or somatic growth to keep metabolism constant, and is characteristic of a stressed population.

The main advantage of measuring SfG is that is breaks down energy expenditure into its empirical components, so that the observer can see which physiological processes are

38 impacted by environmental stress. SfG is also widely applicable to most species and therefore makes possible direct comparisons among populations and species with different life history strategies.

b) Energy storage in marine invertebrates, with a focus on echinoderms

The pyloric caeca are large (~ 40 % body weight in fed Pisaster ochraceus Giese 1966) organs that are located along the arms in sea stars. Pyloric caeca have a lobed, ‘frilled’ external appearance that is thought to increase the surface area available for enzymatic secretion (Anderson 1953). The primary function of the pyloric caeca is to secrete enzymes which aid digestion, but other functions include steroid production, mucus secretion, and energy (usually glycogens and polysaccharides) storage (Anderson 1953). Energy reserves can be used by the animal in periods of starvation, but are also available for gametogenesis in the months immediately preceding the reproductive season, as found in Cosmasterias lurida

(Cossi et al. 2015) and Asterias rubens (Voogt and Dieleman 1984). It is commonly agreed that lipid content of the pyloric caeca is a good indicator of the energy reserves of an individual (e.g. by Giese 1966, Harrold and Pearse 1980, Lawrence 1987, Cossi et al. 2015,

Baeta et al. 2016). There is some evidence that energy reserves are prioritised for somatic growth over gonadal growth, as is the case for Pisaster giganticus (Harrold and Pearse 1980), although energy reserves must be partitioned to fulfil basic metabolic functions, movement, somatic growth and gametogenesis according to energy budget theory. The function of the pyloric caeca in providing energy to the animal during periods of starvation generally takes around 3 – 6 weeks without food in order to start (St Pierre and Gagnon 2015).

Energy is stored within the bodies of echinoderms primarily in the form of lipids, and those lipids are usually found within the pyloric caeca, as well as in the gonads and body walls. The contribution of pyloric caeca to overall body weight varies loosely with the gonad cycle

(Lawrence 1973; Jangoux and van Impe 1977), meaning that the pyloric index reflects energy available for growth only at certain times of year (Caballes et al. 2016) during the time when the animal is not building a gonad (i.e. not in the months immediately preceding the spawning period). The inverse relationship that Asteroids in particular show between pyloric and gonadal indices, as observed in Asterias rubens (Jangoux and Van Impe 1977),

Sclerasterias mollis (Barker and Xu 1991) and Patiriella pseudoexigua (Chen and Chen

1992), indicates that energy transference, in the form of yolk proteins (Alqaisi et al. 2016), from the pyloric caeca into the gonads, is the mechanism that enables a gonad to be built. The role of the pyloric caeca as a genesis for yolk proteins is currently being examined in a genetic context. For example, the major yolk protein precursor in female Patiriella regularis is thought to be vitellogenin, which is synthesised in the pyloric caeca, not the gonad (Alqaisi et al. 2016), but which must then be transferred to the ova in order to build oocytes, itself an energetically costly process. It is worth noting that only species within the Asteroidea express vitellogenins, while the Echinozoa express an unknown major yolk protein usually referred to as MYP, and that vitellogenin transcripts have only recently been identified (by Prowse and

Byrne 2012) in Parvulastra exigua. Not much is known about them beyond their relative abundance (Vtg mRNA transcripts can be up to 120x more abundant in the pyloric caeca than in other body parts) and that vitellogenin is different to other echinoderm MYPs as it is a lipid-transfer protein rather than a transferrin-like protein (Alqaisi et al. 2016). Proteins such as vitellogenin may be made in storage cells found within the ‘digestive epithelium’ layer of the pyloric caeca, which are thought to be large carbohydrate – protein complexes (as in the brooding sea star Leptasterias hexactis Chia 1969).

c) Physiological acclimation to ocean acidification and warming; processes, timescales

and vulnerabilities

Metabolic function and the ability of marine invertebrates to maintain physiological homeostasis can be affected by exposure to elevated temperature and pCO2, as well as various bioenergetic indices including growth rate, reproductive capacity, cellular function, respiration, feeding and acid-base balance (Lannig et al. 2010; Gaylord et al. 2015; Kroeker et al. 2017). Due to the link between increasing temperature and metabolic function in ectotherms, increased energetic demands at high temperatures can result in unsustainable energy deficits (Pörtner 2010). The sponge Cliona orientalis, for example, is unable to meet the increased metabolic demands of elevated temperature and pCO2 even in short-term exposures, which could alter its survival and function within coastal ecosystems (Fang et al.

2014). The energy required to overcome physiological stress increases proportionally to the magnitude and intensity of the stress, meaning that the larger the effect size of ocean warming and acidification, the greater the energetic deficit will become in metabolically- compromised species (Sokolova 2012). The interactive effect of multiple stressors may also critically affect the ability of some species to respond and recover (Gunderson et al. 2016).

Metabolic change, either hyperventilation or respiratory suppression, is one common response to elevated temperature and pCO2 which are outside the tolerance levels of marine organisms, particularly sessile species. An increase in respiration rate is one way to generate the diffusion gradients needed to expel excess CO2 from within the body and maintain internal pH (Melzner et al. 2009). However, not all species experience this type of metabolic reaction; there was no evidence of metabolic suppression in the oyster Crassostrea gigas in

41 response to a reduction in pH of 0.4 units over one month (Lannig et al. 2010). Even within the same species, different ontogenetic stages may respond in different ways. Larval O. validus respiration rates at bipinnaria stage have been shown not to differ within a temperature range of -0.5 to +2.0 °C (Peck and Prothero-Thomas 2002) and theoretical long- term adult acclimation up to 6 °C should be possible (Peck et al. 2008), but respiration in adult sea stars O. validus increased from -1.5 to 8.0 °C but decreased around 15 °C due to collapse of coelomic oxygen transport systems (Peck et al. 2008). Phenotypic expression within conspecifics may alter metabolic responses to ocean acidification, making population- level responses broad and varying (Rivest and Hofmann 2014). As body size and temperature are the two major controls on metabolic rate in marine invertebrates, broad models which do not take into account ontogeny, phylogeny and warming scenarios may severely underestimate the energetic cost of near-future ocean warming and acidification which is predicted to cause a combined metabolic change of up to 44 % (Carey et al. 2016).

Ocean acidification may have other physiological effects apart from reduction in respiration.

Internal pH regulation is a fundamental indicator of metabolic regulation and therefore physiological health. Exposure to elevated pCO2 can disrupt internal pH regulation and increase sensitivity to external pH change within the subphylum Ambulacraria

(Echinodermata), indicating a phylogenetic component to physiological tolerance which may enhance the potential for adaptation within group of related animals (Collard et al. 2013; Hu et al. 2017).

Elevated pCO2 delays the onset of feeding in larval mussels Mytilus californiatus and thus has a negative effect on the metabolic energy budget of the whole animal (Gray et al. 2017).

Juvenile Asterias rubens exposed to high pCO2 (1102, 3500 ppm) for 24 days resulting in lowered feeding, increased ammonia excretion and reduced SFG with no acclimation potential (Appelhans et al. 2014). As seen throughout the literature, the physiological energy budget of marine invertebrate larvae is vulnerable to disruption from ocean acidification and warming as their energy stores are often limited, derived from the egg yolk and unable to sustain the developing juvenile under the increased energy demands of high pCO2 (Stumpp et al. 2011; Pan et al. 2015; Gray et al. 2017). Some species are able to moderate their behavioural changes to increase feeding, but this leads to greater competition with conspecifics and other consumers (Nilsson et al. 2012; Gaylord et al. 2015).

Behavioural biomarkers of stress are, however, useful in conjunction with biochemical and other indices, but their use is assessing species responses to ocean acidification and warming is currently under-represented within the literature on invertebrates, but there is a growing body of studies dealing with behavioural responses in fishes (Munday 2014; Näslund et al.

2015; Nagelkerken and Munday 2016). Areas of research such as modelling the energy allocation involved in behavioural responses, species interactions and trans-generational behavioural alterations need more attention in this context. One study on juvenile fish

(Amphiprion melanopus) indicated that parent exposure to high pCO2 can alter offspring escape behaviour (Allan et al. 2014). Behavioural plasticity has been suggested to be as important as phenotypic plasticity in determining adaptive responses to global climate change, although more work is needed to indicate how the effects of such behavioural changes will scale up to the community and ecosystem level (Wong and Candolin 2014).

Behavioural changes are attributable to a combination of neurology, physiology, resource usage, environmental cues and interactions with other species (Schram et al. 2014;

Nagelkerken and Munday 2016) as well as exposure time. Righting time (the time taken for an inverted animal to turn itself right-side up) is correlated with temperature (Peck et al.

2008; Brothers and McClintock 2015) and therefore also metabolism and individual fitness

(Kleitman 1941; Manríquez et al. 2017). The sea star Luidia clathrata was unable to acclimate to low temperature over 30 days and showed a permanent increase in righting time

(Watts and Lawrence 1990). Although long-term studies are partially capable of circumventing this issue by ensuring that animals are fully acclimated before undertaking behavioural studies, righting time may not be a clear indicator of an individual’s physiological state due to the plasticity of behaviour (Challener and McClintock 2013;

Brothers and McClintock 2015; Wong and Candolin 2015; Nagelkerken and Munday 2016)

Reallocation of stored energy is one way for organisms to maintain physiological homeostasis during exposure to elevated temperature and pCO2 (Kuhnhold et al. 2017). Up to

84 % of available energy was able to be reallocated in Strongylocentrotus purpuratus at 800

µatm CO2 in order to survive, however there is a limit to the amount of energy which can be conserved this way (Pan et al. 2015). Energy reallocation can result in decreased reproductive capacity in order to maintain homeostasis (Mendo et al. 2016), or vice versa (Fitzer et al.

2012). The ability to reallocate stored energy in this way depends on both phylogeny and health, meaning that some species will be more successful in utilising this strategy than others.

1.5 Aims and hypotheses

The broad aim of the present study is to determine the heritable adaptive capacity of marine echinoderms in response to ocean acidification and warming in terms of their survival, tolerance and reproductive fitness. This general aim can be broken down into specific objectives, each corresponding to at least one experimental hypothesis. The colour coding for each hypothesis will be followed throughout the thesis and is used in Figs. 1 and 2.

Objective 1: To use Quantitative Genetics to assess adaptive capacity of wild populations of Echinoderms in response to ocean acidification and warming.

Hypothesis 1: Genotypic variation within wild populations

Wild populations of echinoderms from polar, temperate and sub-tropical latitudes exposed to combinations of high temperature and high pCO2 possess the capacity to adapt to a range of different high-temperature and high-CO2 environments.

Objective 2: To expose adult echinoderms to combined ocean acidification and warming over

long time-scales. Acclimation to external stress factors is acknowledged as a factor in

determining survival. Physiological and reproductive measurements taken during this

exposure period will help determine acclimation potential over an ecologically-relevant period

of time.

Hypothesis 2: Metabolic energy budget

Adult echinoderms possess the physiological capability to acclimatize to ocean acidification and warming over long time-scales without a significant cost to metabolic energy budget. 45

Objective 3: To quantify the developmental success of the F1 generation of larvae from adults exposed to long-term ocean acidification and warming. Trans-generational plasticity and parental effects may be important in determining adaptive response of populations to rapid environmental change where genetic mutations are too slow to keep up with the pace of change. Non-genetic inheritance in response to climate change is a new and important area of research.

Hypothesis 3: Trans-generational responses to ocean acidification and warming

The range of genotypes within populations of F1 generation individuals will depend on the environment to which the parents are acclimated.

Figure 1.1: Thought process and conceptual outline for this thesis.

Figure 2: Thesis workflow.

Figure 1.2 Thesis workflow

Chapter Two: Paternal identity influences response of

Acanthaster spp. embryos to ocean acidification and warming

Photo: Adult Acanthaster spp. Photo credit: K. Sparks

Parts of this chapter have been published as: Sparks et al. (2017) paternal identity influences response of Acanthaster planci embryos to ocean acidification and warming. Coral Reefs 36

(1): 325-338. This manuscript is reproduced here with only minor edits and reformatting. A poster based on this chapter was shown at the 17th Annual Conference of Developmental

Morphology, Mexico 2014.

This chapter pertains to:

Hypothesis 1: Genotypic variation within wild populations

Wild populations of echinoderms will possess the adaptive capacity to combinations of high

st temperature and high pCO2 equivalent to those forecasted for the end of the 21 century in polar, subtropical and temperate locations.

Significance: Genotypic variation within a population drives the natural selection of tolerant individuals in response to an external stress factor. The existence of a range of ‘tolerant’ genotypes within wild populations will indicate that adaptation to ocean acidification and warming is possible. This has not been assessed in many species yet, so this work will contribute to a new and relevant field of research.

Contributions to this chapter:

Kate Sparks designed the experiment adapted from a protocol made by Shawna Foo and collected the data. Kate wrote, edited and revised this chapter for publication. Shawna Foo assisted with data collection. Shawna, Sven Uthicke and Maria Byrne read and provided feedback on an early draft of this chapter. Miles Lamare assisted with the data collection and setting up the experiment. He has read and edited this chapter. Two anonymous reviewers from the journal Coral Reefs provided additional feedback before publication.

Note on species names: Acanthaster spp. will be used to refer to the species commonly referred to as Acanthaster planci and Acanthaster cf. solaris due to recent reclassification.

Research was supported by the Australian Institute of Marine Science and carried out in the

National Sea Simulator (SeaSim). Water chemistry analysis was carried out by Stephen

Boyle at AIMS. The authors would like to thank SeaSim staff especially Andreas Severati for engineering and technical support in the SeaSim unit, and Michelle Liddy for assistance with data collection. The authors would like to acknowledge funding from a University of Sydney

Scholarship (S. Foo) and from the University of Otago (K. Sparks and M. Lamare). This is

Sydney Institute of Marine Science contribution number 187.

2.1 Abstract

The crown-of-thorns sea star Acanthaster spp. (aka the Crown of Thorns sea star; COTS) is a key predator of corals and has had a major influence on the decrease of coral cover across the

Indo-Pacific. To understand how this species may adapt to ocean warming and acidification, we used a quantitative genetic approach to examine the response in offspring of 24 half-sib

COTS families raised in fully crossed treatment combinations of temperature (27, 29 and 31

°C) and pCO2 (450 ppm and 900 ppm) to the gastrulation stage (26 h post fertilisation).

Interactions between genotype and environment were tested using a permutational multivariate ANOVA and restricted error maximum likelihood calculations of variance.

Increased temperature (from 29 to 31 °C) reduced normal gastrulation from ~65% to ~30%.

The extent to which each genotype was affected depended on sire identity, which explained

15% of variation. pCO2 did not significantly influence development at gastrulation. To explore the importance of individual mating pairs, response ratios were calculated for offspring of each family across all treatments. Response ratios demonstrated that the majority of genotypes experienced the highest percentage of normal development to gastrulation in the control treatment, and that family (sire × dam) is important in determining the response to ocean warming and acidification. A positive genetic correlation (overall r*G = 0.76) from sire × environment interactions, however, indicated that individuals which develop ‘better’ at

51 both high temperature and high pCO2 may cope better with near-future predicted warm and acidified conditions for eastern Australia.

2.2 Introduction

Tropical coral reef communities such as Australia’s Great Barrier Reef (GBR) are vulnerable to environmental damage, including from cyclones, bleaching, disease, pollution and the crown-of-thorns sea star (Hoegh-Guldberg 1999; Bellwood et al. 2004), which together have accounted for a ~50% coral cover loss on the GBR over the past 27 years (De’ath et al.

2012). Shallow reef communities may generally have low resilience to ocean warming and acidification, leading to increased coral mortality (Anthony et al. 2011) with associated declines in macroinvertebrate communities living around the reef (Fabricius et al. 2014). In combination with these physical and chemical factors, extensive coral predation by COTS is a leading cause of coral reef decline (Dulvy et al. 2004). Acanthaster populations follow periodic large-scale density increases (‘outbreaks’) (Uthicke et al. 2009), with three major outbreaks recorded on the GBR since 1962 (Moran et al. 1988, Uthicke et al. 2009). During the 1962 outbreak, up to 80% of shallow scleractinian corals died (Pearson and Endean

1969). Based on modelled data of three stressors (cyclones, bleaching and Acanthaster spp. outbreaks), Acanthaster spp. predation accounted for up to 42% loss of coral cover on the

GBR since 1985 (De’ath et al. 2012).

The causes and mechanisms of Acanthaster spp. outbreaks are still debated (see Pratchett et al. 2014 for review), but may be influenced by an increase in food availability for their larvae associated with coastal eutrophication (Brodie et al. 2005; Fabricius et al. 2010; but see

Roche et al. 2015) and the phenotypic plasticity and robustness of these larvae (Wolfe et al.

2015 a, b). The response of larvae to increased food supply may be enhanced by warmer sea temperatures (Uthicke et al. 2015), although near-future warming (to 30 °C) results in increased abnormality and smaller larvae (Kamya et al. 2014; Lamare et al. 2014). As the pattern of outbreaks is likely to be influenced by future ocean warming and acidification

(Uthicke et al. 2013, 2015; Lamare et al. 2014), the interaction between Acanthaster spp. predation and these environmental changes needs to be understood in order to accurately model possible future climate-change effects on the ecosystem.

The responses of larvae to the combined stressors of ocean acidification and warming vary with latitude, taxon and environment (see Byrne 2010, 2011; Przeslawski et al. 2015 for reviews). Recent studies by Gianguzza et al. (2014), Hardy and Byrne (2014) and Garcia et al. (2015) investigated the combined effects of warming and acidification on echinoderm larvae using gametes pooled from multiple parents. Sea star larvae placed in warmer and decreased pH conditions from fertilisation displayed similar responses across taxa. These responses include smaller body size (Odontaster validus in Gonzalez-Bernat et al. 2013;

Patiriella regularis in Byrne et al. 2013; Asterias rubens in Appelhans et al. 2014), and increased mortality (O.validus in Gonzalez-Bernat et al. 2013; Meridiastra calcar in Nguyen et al. 2012). Post-settlement stages appear to be more resilient to ocean acidification, as seen in Parvulastra exigua (Nguyen and Byrne 2014), and juvenile Pisaster ochraceus, which even demonstrated an increase in both growth rate and feeding rate under high pCO2 regimes

(Gooding et al. 2009).

Acanthaster spp. larvae develop in the plankton for around 10–24 d (Uthicke et al. 2015).

Outside their optimum temperature and pH range, the larvae of this species have a greater

53 risk of increased abnormality and mortality (Kamya et al. 2014; Lamare et al. 2014), and ultimately reductions in settlement success (Uthicke et al. 2013). The pattern of Acanthaster spp. outbreaks is therefore likely to be influenced by the degree to which their larvae can adapt to future ocean warming and acidification.

The rate at which climate change affects the environment may be so rapid that traditional

(Darwinian) evolution cannot keep pace with the changes necessary to maintain optimum fitness within an environmental niche (Bürger and Lynch 1997). If rapid adaptation to a fast- paced environmental change such as ocean warming or acidification is required, it has been suggested that standing genetic variation within wild populations may be a more important adaptation mechanism than mutational change (Lande 2009; Hoffmann and Sgro 2011).

Quantitative genetic studies, which characterise genetic variation within populations, can be used to estimate the evolutionary potential of a species (Hoffmann et al. 2005; Dahlhoff et al.

2008; Foo and Byrne 2016). Quantifying this potential requires measurements of a trait among one or many populations under different selection regimes (Hofmann and Sgro 2011).

In such experiments, the responses of individuals of known genotypes under different selection scenarios can be quantified (Gienapp et al. 2008; Hoffmann and Sgro 2011).

Although quantitative genetic experiments have limitations (see Gienapp et al. 2008), they are one method used to quantify adaptive potential and determine the additive genetic variance of a trait. Quantitative genetic experiments have been used to quantify species’ adaptive capacity in response to predicted global ocean change using both single-stressor

(e.g., acidification; Kelly et al. 2013; Sunday et al. 2014), and multiple-stressor experiments

(warming and acidification; Foo et al. 2012, 2014; Malvezzi et al. 2015) although there are inherent difficulties in determining whether observed changes are due to phenotypic plasticity or genetic adaptation (see Merilä and Hendry 2014).

Quantitative genetic studies have increased our understanding of how planktonic larvae are provided for by their parents, and while the process of maternal provisioning is well- established (e.g., Byrne et al. 1999, 2008; Villinski et al. 2002; Falkner et al. 2006; Prowse et al. 2008; Marshall et al. 2010) the effects of sire variation on larval survival are less well studied in the context of climate change in marine environments. Recent QG experiments have demonstrated that additive genetic variation is important in determining individual responses to environmental change. For example, in the sea urchin Pseudoboletia indiana, sire × environment interactions had a significant effect on offspring performance at gastrulation (Foo et al. 2014). Other genetic studies, such as that by Pespeni et al. (2013) on the sea urchin Strongylocentrotus purpuratus, characterise changes in gene expression in response to environmental stressors. Both quantitative and traditional genetic approaches demonstrate that rapid adaptation to environmental change is possible (Sunday et al. 2011;

Pespeni et al. 2013).

Given the importance of understanding the role of Acanthaster spp. in coral reef functioning, the aim of this study was to estimate additive genetic variance and adaptive potential of

Acanthaster spp. embryos and gastrulae in the response to near-future ocean warming and acidification scenarios (‘business-as-usual’ scenario for the year 2100; IPCC 2007). We hypothesised that interactions between parent identity and environmental selection pressures from increased temperature and pCO2 would influence the performance of offspring at gastrulation, i.e., that that successful development to gastrulation would depend on sire identity. Experimental treatments were chosen to reflect near-future warming and acidification scenarios for eastern Australian waters, which predict that by the year 2050 the pH could decrease to 7.90 (from present mean of 8.10; Orr et al. 2005) and sea surface

55 temperature could rise to 2 °C warmer than the 1990 average (mean summer temperatures

22–28 °C; Lough and Hobday 2011).

We hypothesised that populations of Acanthaster spp. would have the capacity to adapt to ocean acidification and warming. We used genotype × environment interactions to determine the additive genetic variance of the offspring response to an environmental stressor (Lynch and Walsh 1998), with the caveat that some variance due to sire may have a phenotypic environmental (epigenetic) component (Crean et al. 2013). Estimation of the additive genetic portion of Acanthaster spp. larval response to ocean acidification and warming was used to provide insight into how this species’ distribution range may shift and whether the population dynamics of this species may change in the future.

2.3 Materials and methods

a) Collection sites

Adult Acanthaster spp. were collected using SCUBA from Arlington (16°43'S, 146°02'E) and

Agincourt (16°02'S, 145°51'E) reefs, eastern Australia, in November 2013 and transported to the National Sea Simulator (SeaSim) at the Australian Institute of Marine Science (AIMS) in

Townsville, where they were kept in flow-through tanks at 27 ± 1.0 °C (SE; n = 12) prior to experiments. Ambient sea temperature at the time of the collection was 26 °C. Sea surface temperature on the GBR during the spawning season of Acanthaster spp. (Dec/Jan; Babcock

1990; Babcock and Mundy 1992) ranges from 26.0 to 28.4 °C (annually 22.0–28.4 °C;

Lamare et al. 2014; Shaw and McNeil 2014).

b) Experimental design: the North Carolina II breeding design

Quantitative genetic experiments using a modified version of the North Carolina II design

(Lynch and Walsh 1998; Foo et al. 2012) were conducted as three consecutive runs (‘blocks’) over a 7-d period (see Fig. 2.1 for experimental design). In each block, gametes from four males (sires) and two females (dams) were fertilised in all combinations to create eight half- sib offspring genotypes. Each block was replicated three times using different individuals each time, resulting in 24 unique ‘families’ of known parentage. This number exceeds the recommended 20 genotypes needed to detect genetic effects in response to treatment

(Munday et al. 2013).

Fig. 2.1 Flowchart showing the experimental protocol for the modified North Carolina II breeding design employed in this experiment to raise family lines of embryos of Acanthaster spp. under different temperature and pCO2 regimes. The left-hand side shows the factorial dam × sire breeding cross design and the 24 offspring genotypes generated. The right-hand side shows an example of the treatment of embryos from a single cross after fertilisation. This set-up (right hand side) was repeated for each offspring family line across three experimental runs or ‘blocks’, where each block consisted embryos from eight half-sib families

c) Spawning and fertilisation

Mature gonads were removed by biopsy. Ovaries were placed in the ovulatory hormone 10-3

M 1-methyladenine and released eggs were transferred to filtered sea water (0.22 µm). At this point, egg density was determined in five 100-µL aliquots taken from the egg suspension.

Sperm was pipetted ‘dry’ from the testes and stored at 4 °C for no more than 1 h prior to fertilisation. Dry sperm concentration, determined from three haemocytometer counts, was

1x1010 ± 35,000 sperm mL-1(averaged from n=4 males). Fertilisation was carried out separately for each family in each block in all combinations of the three temperature (control

= 27 °C, 29 °C, 31 °C) and two pCO2 levels (control = 450 ppm, high = 900 ppm; pH = 8.05 and 7.90 respectively) in 250 mL beakers (Fig. 2.1).

Eggs were acclimated to their treatments for 20 min prior to fertilisation, while sperm from each sire was added individually to separate filtered seawater from the same treatment, and insemination of eggs made in all treatment combinations. The volume of sperm added to each container was calculated to achieve a final sperm density of 105 sperm mL-1 (resulting sperm: egg ratio of 200:1). After 10 min, fertilisation was checked by examining for a fertilisation envelope (ranging between 75% and 86% of eggs). At this point, to remove excess sperm

80% of the water from the fertilisation containers was decanted off and replaced to the same volume with fresh filtered seawater from the respective treatment following the methods of

Foo et al. (2014).

One hour post fertilisation, all the eggs from each cross were divided evenly among falcon tubes with meshed windows of a pore size (63 µm) to prevent loss of the eggs (~180 µm egg diameter; Hoegh-Guldberg and Pearse 1995). The mesh windows were 45 x 6 mm and there

59 were four windows in each falcon tube, giving a total area for water exchange of 10.9 cm2

(~5% of tubes’ total surface area). The mean flow rate of seawater through the tubes was ~4.4 mL min-1 (n = 15, σ2 = 0.29 mL min-1), measured using dye dissolution over time.

Three tubes from each cross were selected randomly and placed into a treatment container

(20 L) of fresh 0.22-µm filtered seawater of either ambient or elevated pCO2 seawater. Two of each container were then incubated for 26 h in controlled temperature ‘water jackets’ set at either 27 °C, 29 °C or 31 °C. Containers of the same pCO2 treatment were incubated within each temperature water jacket. Thus each water jacket represented one treatment, with six water jackets in total (i.e., 3 temperatures and 2 pCO2 levels). Temperature treatments were arranged randomly within the same room so that water jackets of the same temperature were not adjacent. Each container was aerated at a constant rate using a separate air-tube and air- stone depending on pCO2 treatment (450 ppm or 900 ppm CO2) Each temperature/pCO2 treatment combination was replicated three times in each block (Fig. 2.1).

During scoring, embryos from all three falcon tubes were counted and data were pooled to represent a replicate of the treatment. If any one sample did not contain embryos, data from the remaining two samples from that tube were pooled for the replicate.

d) Experimental conditions

Experiments were carried out in multiple 20-L recirculating tanks. The temperature and pCO2 in each tank were stabilised over a 5-d period prior to the start of the experiments.

Temperature was controlled to within a range of ±0.5 °C of a set temperature inside the tanks throughout the experiment (Appendix 10.1), monitored using HOBO temperature loggers

(Onset HOBO Pendant data/alarm logger 64K UA-001-64, Massachusetts) at 5-min intervals

60 and periodically checked throughout the experimental period with a handheld YSI temperature probe (Xylem, Queensland). Pure CO2 concentration and injection rate was digitally controlled to maintain the pCO2 at either 450 or 900 ppm, with pH spectrophotometrically determined thrice daily (morning, noon, evening) using m-cresol purple (Sigma-Aldrich batch 2303-01-7) using methods described by Dickson et al. (2007).

Technical replication ensured accuracy of each measurement through taking three subsamples from each treatment at each time point, as well as multiple time points on the same day. pH standards and blank seawater samples were also run through the spectrometer as part of each time point sampling protocol. A 250-mL water sample fixed with 100 µL saturated HgCl2 taken from each treatment was analysed by potentiometric titration and calibrated against certified reference materials CRM126 using analytical method V INDTA 3C. One sample was taken from each treatment in each block, but daily pH variation was mainly documented through the m-cresol pH measurements. Total alkalinity, total inorganic carbon, temperature and salinity were used to calculate pCO2 and carbonate saturation states using the software

CO2SYS (Lewis and Wallace 1998) using the seawater dissociation constants proposed by

Mehrbach et al. (1973) and as refitted by Dickson and Millero (1987). Detailed treatment levels are shown in Chapter 10.1.

e) Scoring development

Samples of at least 50 embryos from each falcon tube were removed at 26 h (≈gastrulation stage) post fertilisation, fixed in a 2% buffered formalin seawater solution and their morphology examined microscopically. ‘Normal’ gastrulae followed developmental schedules described for A. planci by Uthicke et al. (2015). Gastrulae were defined as abnormal if they deviated from normal morphological criteria, e.g. of symmetry and cell wall

61 integrity. Individuals that were disintegrating, had a grainy internal appearance, displayed arrested development or were of an unusually small or large size relative to the rest of the sample were deemed ‘abnormal.’ It is unknown to what extent morphological abnormalities such as size negatively impact development at later stages. However, the scoring criteria were established to provide consistent definitions of ‘abnormality’ between samples and blocks.

Reference photos of developmental classification are given in Chapter 10.1.

2.4 Data analysis

For analysis of the cleavage and gastrulation stages, percentage data were arc-sine transformed prior to testing as raw data were in ‘count’ form. After transformation, residual vs. fitted value plots as well as a Shapiro–Wilk test indicated that assumptions of normality and homoscedasticity were met. A pairwise Tukey’s post hoc test was performed to further analyse significant results for differences between levels of a factor, and for interactions between factors.

Percent normal development data for both the cleavage and gastrulation stages were analysed using a permutational multivariate ANOVA (PERMANOVA), using pseudo-F statistics and p-values derived from Monte-Carlo simulations as implemented in the Permanova+ extension of Primer 6 (Clark and Gorley 2006). Temperature and pCO2 were fixed factors, experimental block was a random factor with dam and sire as random factors nested within experimental blocks. F-statistics were calculated using an average of 9999 permutations of the raw data.

Variance components for random factors were calculated using a reduced maximum likelihood (REML) analysis in the lmerTest package of R. Fixed factors (block, temperature, pCO2 ) were tested for significance using the PERMANOVA but their variances were not

62 calculated from the REML analysis. Therefore, the fixed factors were not used to explain any percent variation (Appendix 10.1). This is consistent with the approach used in a number of studies (e.g., Terlizzi et al. 2007; Clark et al. 2013; Busby et al. 2014; Harris 2015; Lamit et al. 2015) and is applied here because variance estimations from fixed effects using REML may lead to generation of negative variance components (Searle et al. 1992).

Reaction norms were created for each genotype. Log-response ratios (lnRR) for the gastrulation trait were calculated for each dam × sire offspring family in each environmental condition to show the change in developmental normality for each dam × sire pair throughout each of the six experimental treatments. Log-response ratios were calculated as the mean effect of an experimental treatment divided by the mean effect in the control treatment.

Variance components from the REML analysis were used to calculate genetic correlations for the gastrulation trait, r*G (Eisen and Saxton 1983; Foo et al. 2012). In this case, overall r*G indicated correlation between additive sire variance and the environment where r*G(T) represents the genetic correlation for the gastrulation trait across the temperature environment and r*G(pH) represents the genetic correlation for the same trait across the pH (pCO2) environment. The overall r*G is the genetic correlation for the gastrulation trait averaged over both the temperature and pCO2 environments (see Eisen and Saxton 1983 for detailed calculation steps). R2 values derived from linear regression were used to show the association between the two factors.

Broad-sense heritability (h2) was calculated across environmental gradients. It was defined as the ratio of the additive genetic portion of a trait (VA), in this case normal development, to the

63 phenotypic portion of that trait (VB) present within a population (Lynch and Walsh 1998;

Malvezzi et al. 2015) thus:

푉퐴 ℎ2 = 푉퐵

2.5 Results

a) Gastrula stage embryos

The percent of normal gastrulation after 26 h decreased as temperature increased (pseudo-F =

6.54, df = 2, p = 0.026) although this was only significant between the 29 °C and 31 °C treatment (Tukey’s post hoc test p = 0.01; Fig. 2.2). The proportion of individuals successfully completing gastrulation at 31 °C was reduced by up to 41% compared to that at

27 °C (Fig. 2.2a). However, the response to temperature was also dependent on pCO2 (pCO2

× temperature interaction; pseudo-F = 5.77, df = 2, p = 0.023; Table 2; Fig. 2.2b). This interaction is most likely caused by slightly higher percent gastrulation at ambient pCO2 in the two lowest temperature treatments (27 °C, 29 °C), whereas at 31 °C, more individuals in the higher pCO2 treatment reached gastrulation after 26 h (mean 28.52% gastrulation at 31 °C in the ambient pCO2 treatments vs. 31.51% gastrulation at 31 °C in the high pCO2 treatment;

Tukey’s post hoc test p = 0.02; Fig. 2.2). Appendix 10.1 contains the output of the

PerMANOVA analysis.

Of the random factors, experimental block explained 17.54% of variation (pseudo-F = 2.74, df = 2, p = 0.004). Sire identity also explained a significant (pseudo-F = 23.30, df = 9, p =

0.007) proportion of variation among offspring (15.05%), with individual sire outcomes ranging from 10.31 to 88.8% across all treatments while dam identity was not significant.

Fig. 2.2 The influence of temperature (a) and the influence of a temperature × pCO2

interaction (b) on gastrulation of Acanthaster spp. embryos subjected to different temperature

and pCO2 regimes. In (b) black bars indicate ambient pCO2 and grey bars indicate high pCO2.

Letters indicate levels of significant differences among treatments calculated using a pairwise

Tukey’s test (p < 0.001) Error bars indicate one standard deviation. n=75 observations per

treatment.

Reaction norms (Fig. 2.3) for the percentage of normal gastrulae ranged from 32.24 to 97.4% among sire genotypes in control temperatures (Fig. 2.3). The range of sire responses also narrowed as temperature increased. Overall, the spread of responses to increasing pCO2 was not large, and did not vary with treatment (Fig. 5). The reaction norms clearly show that the difference in gastrulation can be attributed to sire identity. For example, while offspring of some sires showed a decline in gastrulation between 29 and 31 °C, offspring of other sires

65 showed no change or even an increase (Fig. 2.3). In contrast to this, increased pCO2 did not affect any specific genotypes at gastrulation, as shown by the parallel and near-horizontal slopes of the reaction norms in Fig. 2.3 (pCO2 plots only).

Fig. 2.3 Reaction norms for each of 12 sire (a, b) and 6 dam (c, d) Acanthaster spp. genotypes

across each temperature and pCO2 treatment at gastrulation. Data represent the percentage of

normally developed individuals in a particular treatment. n=3 for all data points

The dam × sire interaction explained 2.37% of offspring variation (pseudo-F = 3.12, df = 9, p

= 0.001). The largest proportion of variation, however, was explained by the significant four- way interaction dam × sire × temp × pCO2, which accounted for 22.81% of variation in the gastrulation trait (pseudo-F = 3.184 df = 18, p = 0.001) and highlighted the importance of individual mating pairs (Fig. 2.4). In light of this, the individual factors or lower-level interactions should not be over-interpreted. In total, the factors and interactions examined in this experiment explained 74.84% of variation among individuals at gastrulation.

The four-way interaction between parental pair, pCO2 and temperature was illustrated for the gastrulation trait of each dam × sire cross in each experimental treatment level (Fig. 2.4) using the log-response ratio (lnRR). Both dam and sire interacted with temperature and pCO2 to produce variable responses across progeny, which can be small (e.g., pair 9: range = -0.19–

0.21) or large (e.g., pair 6: range -2.85–0.75), and depended on temperature and pCO2 treatment as well as sire × dam interaction (22.81% of offspring variation at gastrulation).

Overall, the highest gastrulation success for 20 families were observed in ambient CO2 conditions (450 ppm), and for 19 families in ambient temperatures (27 °C); the other five families developed better or equally well at 29 °C. No families showed highest percent normal development at either high temperatures or high pCO2. The control treatment

67 represented the conditions in which developmental success was highest for the majority of families.

Fig. 2.4 Log-response ratio for each of the 24 Acanthaster spp. genotypes (sire x dam) at the

gastrula stage. X-axis represents pair number in block order. Data points represent the log-

response ratio of the difference in pair performance (normal development) between the mean

percentage normal development for that pair’s offspring and those individuals’ percentage

normal development in each of six experimental treatments of varying temperature and pCO2

combinations. Closed symbols represent low pCO2 treatments (450 ppm) and open symbols

represent high pCO2 treatments (900 ppm). Pairs are numbered sequentially according to block order, so that pairs 1–8 are from Block 1, 9–16 from Block 2, and 17–24 from Block 3;

and then within block, by order of pairing. Therefore, pair 1 is the offspring of dam 1 × sire 1;

pair 2 is 1 × 2, pair 3 is 1 × 3 etc. (within each block there are two females and four males).

Pairs 1–4, for example, share the same dam, but all have different sires.

b) Genetic correlation and heritability

Overall, there was a positive genetic correlation for development at gastrulation (overall r*G

= 0.76) across both the temperature and pCO2 variables, suggesting that individuals which perform well in response to one stressor (high temperature) are more likely to perform well in response to the second stressor (high pCO2). The correlations between temperature and genotype (Fig. 2.5a), and pCO2 and genotype (Fig. 2.5b), were also both positive (r*G(T)=

0.87 and r*G(pH)= 0.49, respectively), indicating that individuals that performed well in the control environment were more likely to perform well in their respective high temperature and high pCO2 environments (Fig. 2.5c). Broad-sense heritability was 0.44 (+/- 0.01 SE) for the normal development trait at elevated temperature and 0.31 (+/- 0.04 SE) at ambient temperature.

Fig. 2.5 Relationships (percentage normal gastrulation) between embryo performance at

gastrulation for (a) high vs ambient temperature; (b) high vs ambient pCO2; (c) high

temperature vs high pCO2 for the 24 Acanthaster spp. families examined in this study. This

figure illustrates (a) r*G(T) (p = 0.04); (b) r*G(pH) (p = 0.05) and (c) overall r*G (p = 0.01). Data

points represent percentage normal gastrulation for individual genotypes from n = 3 counts

2.6 Discussion

a) Parental identity and embryo performance

This study investigated the adaptive potential of early developmental stages of Acanthaster spp. in response to combined ocean acidification and warming. We examined the influence of male and female genotype on population response to a range of environments, and whether or not genotype confers increased resilience to elevated sea temperatures and greater pCO2. In general, we showed that early development is significantly affected by warmer (up to +4 ºC) temperatures, but not by pCO2 levels, as previously found (Kamya et al. 2014; Lamare et al.

2014). Paternal genome transcripts are not expressed in the offspring until around blastulation

(Crain and Bushman 1983) and therefore it is not surprising that sire identity accounted for a larger proportion of variation at gastrulation (15.05%) than at the 16-cell cleavage stage

(7.22%). Importantly, based on the sire × environment interactions, there is potential for adaptation to ocean acidification and warming in this species.

An interesting feature of Acanthaster spp. development in this experiment was the interaction between to sire × dam × temperature × pCO2 (22.81% variation) at gastrulation.

Not only was parent identity influential overall, but maternal influence interacted with other factors at gastrulation. As the zygotic genome is not fully switched on until near blastrulation

(Tadros and Lipshitz 2009), maternal influence might be expected to be more prominent during pre-gastrulation development than afterwards. Maternal provisioning of eggs can buffer environmental stress (Hamdoun and Epel 2007), which was also expected to influence developmental responses to both stressors in this study. Although we did not measure egg

70 quality, our results suggest that differences in provisioning between individual females did not account for a significant proportion of developmental variation. A possible explanation for this observation is that levels of maternal provisioning are relatively high and equal among females in this population. Egg diameter ~180 µm (Hoegh-Guldberg and Pearse 1995) is relatively large for a planktotrophic species (Emlet et al. 1987). Therefore, with relatively well endowed eggs, maternal lineage may not confer significant developmental advantage at gastrulation.

b) Effects of environmental variation on Acanthaster spp. embryos

The observed differences in the response of the progeny among Acanthaster spp. pairs to combined changes in warming and elevated pCO2 indicate that parent compatibility will be an important factor in determining the response to combined ocean warming and acidification for future recruiting cohorts of Acanthaster spp.. This compatibility is likely to be mediated by the Asteroid sperm bindin – egg bindin receptor system (Patiño et al. 2016). While the majority of genotypes in this study responded negatively to increased temperature, there were genotypes that were both more susceptible (up to 25% lower normality than average) and more robust (up to 20% higher than average) to a 4 °C warming. At 31 °C, most genotypes

(15 of 24) showed an additive deleterious response to increasing pCO2. In this study, increasing temperature to 29 °C did not appear to make a difference to most individuals’ performance, whereas an increase to 31 °C consistently decreased the response ratio, regardless of the pCO2 treatment. This is consistent with 29 °C being near the optimum temperature for development (Kamya et al. 2014; Lamare et al. 2014). The strong genetic correlation between genotype and temperature found in this study, along with the number of

71 significant interactions between temperature and the other variables, indicates that increased temperature may cause a ‘bottleneck of mortality’ at gastrulation in this species.

c) Implications of adaptations to ocean acidification and warming for Acanthaster

spp. population dynamics

Echinoderm species are known to undergo population fluctuations that Uthicke et al. (2009) broadly categorised into three models. Acanthaster spp. populations follow their

‘Acanthaster–Asterias’ model, which states that the population is characterised by constant fluctuations which do not lead to a large reduction in genetic diversity. Instead, the population is subject to continuous natural selection pressure and a continuous flow of alleles.

Population-level genotypic changes, such as a change in allele frequency and the selection for more ‘tolerant’ genotypes as observed in the present study, may therefore mean that

Acanthaster spp. will be able to adapt to near-future climate-change conditions

Outbreaks of Acanthaster spp. on the GBR are intermittent but extensive events (>1500 individuals km–2 indicate outbreak; Moran and De’ath 1992). The east Australian Acanthaster spp. population outbreaks are thought to start in the northern regions of the GBR and spread south along the eastern Australian coastline where they encounter progressively colder sea temperatures (Pratchett 2005). Acanthaster spp., therefore, requires a degree of genetic variation to overcome isothermic clines which could otherwise be a barrier to dispersal. The range of genotypes displayed in this study includes those that perform better in colder as well as warmer temperatures. Indeed, at the population level, the thermal window of the early development stages and larvae appears to encompass cooler temperatures to around 25.5 ºC,

72 and populations of Acanthaster spp. occur in water temperatures as low as 24 °C (Lamare et al. 2014).

Additive genetic (paternal) variation at both individual and population levels has been shown to be more important than previously thought for the adaptive potential of this species in response to ocean acidification and warming. Responses observed in this study from a range of genotypes, and the influence of sire genotype × environment interactions in determining offspring performance, should enable a sub-set of Acanthaster spp. to benefit from the predicted warming and acidification in the oceans around eastern Australia. The predicted

1.48 ºC temperature rise before the end of this century (Hobday and Pecl 2014 is still within the thermal development window for this species (Johnson and Babcock 1994; Lamare et al.

2014), even though development at high temperature is constrained by elevated pCO2 as is the case at 31 ºC. Therefore, while difficult to predict, Acanthaster spp. appears to have the adaptive capacity to colonise new regions of eastern Australia as predicted warming of that area continues.

d) Conclusion

The potential for Acanthaster spp. to continue to periodically decimate coral reef ecosystems will likely be determined by the response of this species’ developmental stages to changing environmental conditions. The capacity to tolerate both high temperatures and high pCO2 is a trait that is dependent on both genetic identity and environment. The presence of tolerant genotypes in Acanthaster spp. indicates that this species may be able to adapt to future environmental changes on the GBR, and indicated that a sub-set of the population may perform as well or better in near-future predicted conditions as in present-day conditions.

Chapter Three: Potential for adaptation in two sand dollar

species, Arachnoides placenta and Fellaster zelandiae, will be

challenged by ocean acidification but not warming.

Photo: Adult Arachnoides placenta during experimental spawning. Photo credit: K. Sparks.

This chapter forms a manuscript currently in preparation as: Sparks KM, Foo SA, Uthicke S,

Byrne MA and Lamare MD ‘Potential for adaptation in two sand dollar species, Arachnoides placenta and Fellaster zelandiae, will be challenged by ocean acidification but not warming.’

Target publication: Marine Ecology Progress Series (2017 impact factor 2.29)

This chapter pertains to:

Hypothesis 1: Standing genetic variation within wild populations

Wild populations of Echinoderms from polar, temperate and sub-tropical latitudes exposed to combinations of high temperature and high pCO2 possess the capacity to adapt to a range of

different high-temperature and high-CO2 environments.

Contributions to this chapter:

Kate Sparks designed the experiment based on an earlier protocol created by Shawna Foo.

Kate, Shawna, Miles Lamare and Sven Uthicke built the experimental system at AIMS and carried out the experiment. Kate and Shawna collected the data. KS analysed the data and wrote this chapter. Miles, Shawna, Sven and Maria gave feedback on an early draft copy.

Kate and Miles received funding from the University of Otago. Sven was funded by AIMS.

Maria and Shawna were funded by the University of Sydney and the University of Otago.

The authors acknowledge the contributions of Michelle Liddy and Andreas Severati for their assistance with the SeaSim monitoring and checking.

3.1 Abstract

Marine ecosystems will likely experience elevated pCO2 and temperature in the next century, and much of the resilience of these ecosystems will depend on the adaptive responses of populations. We investigated the adaptive responses of two clypeasteroid echinoids (sand dollars), Arachnoides placenta from tropical Australia and Fellaster zelandiae from temperate New Zealand, to forecasted near-future increases in seawater temperature and pCO2. We used a quantitative genetic experimental design to determine the adaptive capacity of normal development among male and female genotypes at the 2-arm pluteus stage to a

+3.0 °C temperature increase and 2x current pCO2 levels. Elevated pCO2 decreased the mean percentage of normal development of early plutei by 20.3 % in F. zelandiae and by 24.6 % in

A. placenta, and was driven by the [sire x pCO2] interaction which accounted for 2.14 % of variation among A. placenta plutei and 4.27 % of variation among F. zelandiae plutei.

Maternal lineage (dam identity) accounted for 4.73 % of variation among A. placenta. There was no effect of increased temperature and no interaction between temperature and pCO2.

Heritability estimates and genetic correlations for normal development at the pluteus stage were low for both species. Our results indicate that warming at the level tested will not exert strong selective pressure in these populations. The response of clypeasteroids to ocean acidification depends on both maternal and paternal lineage, and differences in resilience among genotypes will affect future population response.

3.2 Introduction

Temperature and pH in marine environments are naturally variable but long term trends indicate global sea surface warming and acidification (Harley et al. 2006). Ocean acidification is predicted to decrease the mean and extreme ocean surface pH by up to 0.3 units by the end of the 21st century (IPCC 2013). Despite experiencing large daily variations in pH and temperature, adjusting to elevated means and extremes will present a physiological and evolutionary challenge to marine communities (Wenberg et al. 2011; Melzner et al. 2013;

Bakun et al. 2015). Echinoderms have been shown to be vulnerable to combined ocean acidification and warming (meta-analysis by Harvey et al. 2013). Nevertheless adaptation to environmental stressors depends on a combination of phylogeny and life exposure history.

Clypeasteroids (sand dollars) may be relatively tolerant to elevated temperature and pCO2, making them likely to persist in a changing ocean. Fertilization and embryonic development of the sand dollar Arachnoides placenta are robust to decreases in pH from 8.1 – 7.8

(Gonzalez-Bernat et al. 2013) and in general, the thermal development window for sand dollar larvae is wide, for example Dendraster excentricus larvae are competent from 7-26 °C

(Bingham et al. 1997).

Range shifts, expansions and local extinctions may be likely for species that are not able to adapt to the rapidly changing conditions. The potential for pelagic larval dispersal to be altered by future predictions of ocean warming and acidification depends on both changes in ocean currents, and to physico-chemical properties of those water masses (Wilson et al.

2016). Near-future ocean acidification and warming represent stressors for selection that may decrease variability in the genomes of marine populations (Hoffmann and Sgrò 2011).

Although phenotypic plasticity may facilitate dispersal and short-term responses to environmental changes, populations must possess the capacity to adapt in order to survive in the long-term (Gienapp et al. 2008; Hendry et al. 2008; Lande 2009; Hansen et al. 2012).

Due to the observed variability, coastal marine populations may be more robust to predictions of future acidification and warming than previously thought (Byrne et al. 2010; Ross et al.

2011). The sea urchin Pseudoboletia indiana from coastal Australia has the capacity to pass on heritable tolerance to elevated temperature through sire x environment interactions (Foo et al. 2014), whereas Sterechinus neumayeri, a sea urchin from the Antarctic ocean, has a reduced adaptive ability due to a reduction in the amount of genetic variation in response to ocean warming and acidification (Foo et al. 2016). Such species-specific differences are important as adaptation to climate change will require not only plasticity and tolerance, but also heritability of fitness-correlated traits over short timescales and over generations

(Bernhardt and Leslie 2013).

Both sire (= fathers) and dam (=mothers) identity are important with respect to adaptive capacity to environmental stressors. Several recent studies have shown that the relative importance of these factors varies among marine invertebrates (Byrne 2012; Cruz et al. 2013;

Gibbin et al. 2016).). The capacity of the sea urchin Strongylocentrotus purpuratus to locally adapt to 1000 ppm CO2 is dependent on the genetic identity of both sire and dam. Here it has been observed that abundant genetic variation conferred significant advantages in the present- day habitat (Kelly et al. 2013). Additive genetic variance, the main contribution of the sire to offspring genotype, has been shown to be a significant factor in determining the response of many coastal echinoderm species to future ocean changes (Foo et al. 2014; Campbell et al.

2016; Delorme and Sewell 2016). For example, paternal identity influences the adaptive capacity to tolerate conditions under predicted future ocean warming and acidification in the sea star Acanthaster cf. solaris (Sparks et al 2016). For the sea urchin Sterechinus neumayeri, maternal identity determines adaptive capacity through alterations in maternal provisioning and heritable maternal effects in response to warming and acidification, such as decreases in fecundity and egg size in response to ocean acidification (Suckling et al. 2015).

Epigenetic effects including maternal embryo provisioning can be a strong factor in determining the success of future generations of marine invertebrates.

We investigated the presence of adaptive responses of development of two clypeasteroid echinoids (sand dollars) Arachnoides placenta from tropical Australia and Fellaster zealandiae from temperate New Zealand in response to increased temperature and acidification. Arachnoides placenta is found in sandy inter-tidal throughout the western Indo-

Pacific, from Taiwan to eastern Australia (Chen and Chen 1992) and F. zelandiae is found in inter-tidal and sub-tidal marine soft-bottom communities around New Zealand from 0-70 m water depth (McKnight 1969). These species were chosen because both are common and have been shown to be tolerant to moderate changes in pH and temperature; Arachnoides placenta has a wide thermal development window of 15-31oC (Hardy et al. 2014; Karelitz et al. 2016), and even temperatures >31oC do not appear to affect larval development

(Gonzalez-Bernat et al. 2013; Rivera 2015). There are few published works examining the effect of ocean acidification on embryo development in A. placenta; however, Gonzalez-

Bernat et al. (2013) show that decreased pH (10 % at pH 7.70 and >5 % at pH 7.50 across a range of environmentally-relevant seawater temperatures, although later developmental stages are robust to pH change (Karelitz et al. 2016). Fellaster zelandiae has a thermal development

79 optimum temperature of 18-20 oC which will not be negatively impacted by moderate ocean

o warming (+2.0 C) and acidification (pH 7.80) (Rivera 2015). Exposure to high pCO2 had no effect on the thermal development window of early pluteus larvae of Fellaster, as well as confirming that the thermal windows of both sand dollar species are broad (range of +/- 15

°C variation around mean for both) at the pluteus stage (Karelitz et al. 2016).

In eastern Australia the intertidal habitat of A. placenta is in a fast-warming region where mean SST is projected to increase by >2.5 °C in the near future (Cai et al. 2005) and for the intertidal population investigated here will be exacerbated by strong warming of the continent

(Hobday and Lough 2011). The pH mean in coastal seawater around eastern Australia is 7.9-

8.15 and can fluctuate as much as 0.61 pH units diurnally (Pelejero et al. 2005; Wei et al.

2009; Duarte et al. 2013). New Zealand coastal seawater pH has been found to vary by up to

0.9 units over several hours (Hurd et al. 2011, 2017). The variation in both of these environments represent significant physiological and adaptive challenges to the biological communities in coastal environments.

We estimated the adaptive capacity of coastal sand dollars to forecasted elevated temperatures and pCO2 levels, and investigate the contributions of genetic and environmental components for each species. We hypothesised that both species would show strong heritability of tolerance to elevated temperature and pCO2 due to the standing genetic variation within their respective coastal environments, but that Arachnoides would possess more genetic variation in response to ocean warming and acidification than Fellaster because of its broader thermal development window and the larger variation in daily pH and temperature levels in its natural habitat.

3.3 Methods

a) Animal Collection

Fellaster zelandiae were collected by benthic dredge from sub-tidal sandy habitat between

20-70 m depth in Timaru Harbour, New Zealand (44 23' S, 171 15' E) in January 2015 and transferred to Portobello Marine Laboratory, Dunedin, New Zealand. They were held in flow- through ambient conditions for two weeks prior to spawning. Sea temperature at the time of collection was 15.45 °C. The seawater temperature range in the harbour is ~5 – 16 °C with a

Fellaster spawning season (Jan – Feb) average of 16.0 – 17.7 °C (Skerman 1958, Greig et al.

1988). Sea surface temperature varies daily as well as monthly, and can change by up to 6 °C over one tidal cycle in Timaru Harbour (Greig et al. 1988). Seawater pH(T) at the time of collection was 8.00. There is a paucity of detailed pH measurements within Timaru Harbour, but the nearby Munida transect records show that Otago coastal seawater pH varies monthly between 8.20 and 7.90 (data publically available from CARIM and GOA-ON/ NZOA-ON resources).

Arachnoides placenta were collected by hand from inter-tidal sandy coastal areas of

Pallarenda Beach, Queensland, Australia (19 20' S, 146 77'E) in April 2014 and held in flow-through aquaria at the Australian Institute of Marine Science (AIMS), Townsville,

QLD, for two weeks prior to spawning. Water temperature at the time of the collection was

26.0 °C +/- 0.5. Mean sea surface temperature for this area in April is within a range of 25.0 –

27.0 °C (Stuck 2014). Annual variations in sea surface temperature in the area around

Pallarenda Beach (Queensland, Australia) can range from ~10 – 30 °C, with a spawning season average (March – April) of 26 – 28 °C (Haycock 2004). Seawater pH at the time of

81 collection was 8.15 at low tide, determined with a handheld pH probe (model and details given in chapter 2).

b) Experimental temperature and pCO2 manipulations

Arachnoides placenta experiments carried out at the national Sea Simulator facility at AIMS,

Townsville, using the experimental design and set-up as in Sparks et al. (2016) (figure given in chapter 2 Fig. 2.1), consisting of multiple 20 L recirculating ‘water jacket’ seawater tanks set to either ‘ambient’ or ‘elevated’ temperature levels (controlled to within ±0.5 °C of target), within which were the experimental tanks that housed the larvae. Temperature treatments were; ambient temperature mean = 28.10 °C (+/- 0.04 S.E.); elevated temperature mean = 31.80 °C (+/- 0.06 S.E.). Temperature was logged every 5 min throughout the experiment using a HOBO temperature loggers (Onset HOBO Pendant data/alarm logger 64

K UA-001-64, Massachusetts). Pure pCO2 was continually injected and digitally monitored in order to maintain the pCO2 levels at the set treatment levels. The pCO2 treatments were; present-day mean = 519 µatm +/- 35.44 (+/- 0.67 S.E.); and ‘future’ mean = 965 µatm (+/-

0.41 S.E.). Injection pCO2 was logged using in-house logging software at AIMS, and seawater pH was checked three times daily using a benchtop JAZ spectrometer (JAZ

Handheld Spectrometer, Ocean Optics, Florida) and m-cresol purple (Sigma-Aldrich batch

2303-01-7) using methods described in Dickson et al. (2007). Detailed carbonate chemistry during experiments is given in Appendix 10.2.

F. zelandiae experiments were carried out at Portobello Marine Laboratory. For this, an ambient temperature treatment mean = 16.10 °C (+/- 0.03 S.E.) and an elevated temperature treatment mean = 18.65 (+/- 0.40 S.E.) was established. The ‘current’ pCO2 treatment mean =

457.23 ppm (+/- 0.33 S.E.) and the ‘future’ pCO2 level mean was targeted at 900 ppm.

Carbonate chemistry at those pCO2 levels are given in Appendix 10.2. Temperatures were logged using HOBO loggers at 5 min intervals, with temperature controlled using Tunze temperature controllers (model 7028/3) set to either the ambient or elevated temperature levels. Aquarium pH was monitored and controlled via injection of pure CO2 into the header tank using Tunze pH/CO2 controllers (model 7074/2) (Tunze Aquarientechnik GMBH,

Germany). pH was checked spectrally three times daily using a Hitachi-U1100

Spectrophotometer and the methods described above.

For all experiments in both locations, a daily 250 mL water sample was fixed with 100 µl saturated HgCl2 and analysed by potentiometric titration to determine total alkalinity and dissolved inorganic carbon (DIC) content, calibrated against certified reference material

CRM126 AIMS, or certified reference material CRM118 at the University of Otago.

Dissolved pCO2 levels and carbon saturation states were calculated from the total alkalinity, temperature, salinity and pH levels using widely-accepted dissociation constants (Mehrbach et al. 1973; Dickson and Millero 1987).

c) Spawning and fertilization

Each experimental run consisted of gametes from 2 dams crossed with 4 sires in all combinations to create 8 offspring families of known, unique parentage. Mature adults were injected with approx. 1.0 ml of 10-5 M KCl to induce release of gametes. Gametes from each male and female were kept separate until fertilization. Sperm were kept cold on ice until needed, but for no more than 60 min after release. Eggs were acclimated to the seawater temperature and pH in which they would be fertilized and reared for 30 min before being

83 rinsed with fresh filtered seawater of the same temperature and pH composition and placed into jars for fertilization. At this point, egg density was determined from 5x 100 µl subsamples of seawater from the container into which each female was spawning. Mean number of eggs produced by A. placenta and F. zelandiae females was on average 46,000 and 54,000, respectively, as determined by the subsample counts. For both species, sperm concentration was determined by haemocytometer count for each individual male.

Sperm from each male was then diluted with fresh filtered seawater at the appropriate temperature and pH before fertilization to achieve an average sperm concentration of 104 sperm/ml. Fertilization was carried out in situ in the treatments in which the offspring would be raised. The correct volume of sperm was added to each jar of eggs to achieve a final sperm: egg density of 250: 1 for A. placenta and 380:1 for F. zelandiae. The final ratios were different in order to ensure that fertilization kinetics in both species was not limited by sperm availability, as preliminary data indicated that mean fertilization success in F. zelandiae was lower than for A. placenta in ambient seawater conditions (K. Sparks unpubl. data).

Fertilization was confirmed after 20 min for A. placenta and 45 min for F. zelandiae by the presence of a raised fertilisation envelope. At this point, 80 % of the water in each fertilisation jar was removed and the volume topped up to 250 ml with fresh filtered seawater from the same treatment. Fertilised embryos from each jar were divided between three falcon tubes with mesh windows (as described for Sparks et al. 2016) with 63 µm pore size to retain eggs while allowing seawater movement. The dimensions of the window equated to a total area open for water exchange of 10.9 cm2 (~ 5% of total surface area), and water flow through the tubes was ~4.4 mL min-1 (n = 15, σ2 = 0.29 mL min-1).

d) Developmental Scoring

Subsamples of 50 embryos from each falcon tube were taken when embryos reached the 2- arm pluteus stage (18 hr post-fertilization in A. placenta and 36 hr post-fertilization in F. zelandiae), fixed with 2 % buffered formalin and examined for development. Embryos of both species were scored as ‘normal’ if they followed the developmental morphology set out in the literature (Chen and Chen 1992; Nipper et al. 1997; Karelitz et al. 2016). ‘Abnormal’ development was recorded if the embryos violated the ‘normal’ morphology in a significant manner.

3.4 Data Analysis

Data analysis for both species followed the protocol laid out in Sparks et al. (2016). Data were arc-sine transformed and the assumptions of normality checked using a Shapiro-Wilk test. The contributions of genetic and environmental factors to normal development at the 2- arm pluteus stage were analysed using a permutational multivariate ANOVA

(PERMANOVA) performed in the Permanova+ extension of Primer 6 (Clark and Gorley

2006). P-values were derived from Monte Carlo simulations for all analyses. For the design of the linear mixed model, temperature and pCO2 level were fixed factors, experimental block, sire and dam identity were random factors, and sire and dam were nested within experimental block. Pseudo-F statistics were calculated using an average of 999 permutations. Variance contributions for the random factors only were estimated using a reduced error maximum likelihood (REML) analysis in the lmerTest package of R. Fixed factors were excluded from the REML analysis, therefore fixed factors were not used to explain any portion of variation.

Reaction norms were created for each genotypes among the temperature and pCO2 levels.

Log-response ratios (lnRR) for the 2-arm pluteus stage only were calculated for each dam x sire family across each of the six experimental treatments to show the changes in normal development for each family across all treatments.

Broad-sense heritability (h2) was defined as the ratio of the additive genetic portion of a trait

(VA), in this case normal development, to the phenotypic portion of that trait (VB) present within a population (Lynch and Walsh 1998; Malvezzi et al. 2015) thus:

푉퐴 ℎ2 = 푉퐵

Genetic correlations were calculated using variance components generated from the REML analysis. The ‘trait’ in this study was normal development at 2-arm pluteus stage, for which overall correlations r*G and correlations across the pH environment r*G(pH) were calculated

(Eisen and Saxton 1983). Correlations across the thermal environment were not calculated as there was no significant additive genetic variance present for either species. R2 values derived from linear regressions were used to show the association between the two factors.

3.5 Results

a) Effects of environmental factors on population tolerance

Neither Arachnoides placenta nor Fellaster zelandiae showed any significant changes to population mean normal development in response to elevated temperature. In contrast, exposure to elevated pCO2 decreased normal development for A. placenta from 60.0 % (450 ppm) to 39.7 % (900 ppm) (Fig. 3.1 p=0.02 F=9.68) and for F. zelandiae reduced normal development from 84.5 % to 59.9 % under elevated pCO2 (Fig. 3.2 p<0.001 F=24.45). There were no interactions between the two factors.

Figure 3.1: The effect of elevated pCO2 on normal development of sand dollar Arachnoides

placenta 2-arm plutei. Error bars represent 1 S.E. Bars with different letters are significantly different from one another (p< 0.05); n = 245

Figure 3.2: Effect of elevated pCO2 on mean normal development of sand dollar Fellaster

zelandiae 2-arm plutei. Error bars represent 1 S.E. and n = 470 for all. Bars with different

letters above are significantly different (p> 0.005).

b) Heritability and genetic contributions to tolerance

In both species, the interaction between sire identity and pCO2 level explained significant portions of variation among genotypes; 2.14 % of variation in A. placenta (pseudo-F = 0.87 d.f. = 5 p = 0.04 Fig. 3.3) and 4.27 % of variation in F. zelandiae (pseudo-F = 1.93 d.f. = 24 p

= 0.028 Fig. 3.4). Dam identity explained 4.73 % of variation among offspring genotypes in

A. placenta (pseudo-F = 1.62 d.f. = 5 p = 0.05 Fig. 3.3), while [dam x temperature] explained

6.35 % of variation (pseudo-F = 1.06 d.f. = 6 p = 0.05 Fig. 3.3). In both species, the majority of variation among genotypes was explained by the [sire x dam x temperature x pCO2] interaction, which explained 11.91 % of variation in A. placenta (pseudo-F = 4.17 d.f. = 18 p>0.001 Fig. 3.3) and 24.21 % of variation in F. zelandiae (pseudo-F = 2.88 d.f. = 36 p>0.001 Fig. 3.4). Full statistical results are given in Appendix 10.2.

Heritability for the trait ‘normal development’ was h2 = 0.03 for A. placenta and h2 = 0.06 for

2 F. zelandiae at low CO2. At high CO2 heritability estimates were h = 0.06 for A. placenta and h2 = 0.05 for F. zelandiae. There were also significant genetic correlations for the same trait.

2 The genetic correlations were; across the CO2 environment r*G(pH) = 0.14 (R = 0.25) in A.

2 placenta and in F. zelandiae r*G(pH) = 0.24 (R = 0.10).

Figure 3.3: Reaction norms for (A) sire x CO2 and (B) dam x temperature for sand dollar

Arachnoides placenta at 2-arm pluteus stage. Each line represents the response of one

genotype, either male (A) or female (B) across an environmental gradient from n=3 counts.

There are 18 male and 9 female genotypes shown.

Figure 3.4: Reaction norms for 18 male genotypes of sand dollar Fellaster zelandiae across a

CO2 gradient at 2-arm pluteus stage. Each line represents the performance of one male genotype from n=3 counts.

3.6 Discussion

a) Parent identity and embryo performance

The present study examined the range of genotypes within coastal populations of

Arachnoides placenta and Fellaster zelandiae in response to ocean acidification and warming, and demonstrated a strong negative influence of elevated pCO2 on the percentage of normal developmental in the early embryonic stages of both species. The observed negative effect of elevated pCO2 on normal development at the 2-arm pluteus stage corroborates the findings of Karelitz et al. (2016) that changes to ‘normal’ morphology exist among plutei of sand dollars when exposed to reduced pH, including a reduction in body length and changes in the allometry of stomach components. Work by Gonzalez-Bernat et al.

(2013) concluded that morphology, but not survival, of embryonic and larval stages of sand dollars would be impacted by ocean acidification. For example, morphological abnormality in early stomach development (not tested in this study) may lead to increased mortality or loss of feeding competency in later ontogenetic stages (Caballes et al. 2017; Manriquez et al.

2017).

The effect of elevated pCO2 on the percentage of normal development appears to be driven by a subset of individuals from both species; there are eight female Fellaster and three male

Arachnoides genotypes that show relatively high (>10% change in normal development) phenotypic plasticity in comparison to other individuals. Significant sire and [dam x temperature] effects at the pluteus stage for Arachnoides placenta indicate that tolerance to increasing temperature is somewhat dependent on parental pair identity, but also on environmental conditions. Previous exposure to a wide temperature range, as expected in coastal populations, allowed both Arachnoides placenta and Fellaster zealandie plutei to

91 tolerate a 2.5 °C temperature increase. The elevated temperature treatments used in the present study were still within the established range of temperatures within which pluteus larvae of both species would encounter during the spawnng season and known to be tolerated based on thermal tolerance data (A. placenta ~5 -35 °C; F. zelandiae. 13 – 27 °C, Karelitz et al. 2016). Thus, it was not surprising that the 3 °C warming, as well as the lack of interactions between the two stressors, did not have a negative affect on larval development.

Some previous studies have demonstrated that some larvae display a reduction in length with exposure to reduced pH (pH 7.80 Karelitz et al. 2016) and high temperature (>34 °C, Chen and Chen 1992) even within their thermal development range. Increasing pCO2 had a population-level effect on genotypes of plutei of both species, showing that even where species have wide thermal tolerances, there are still subtle selective forcing factors acting on these populations. This result has also been observed in echinoderms such as Sterechinus neumayeri (Suckling et al. 2015) and Strongylocentrotus purpuratus, where body size was shown to be under strong natural selection at high pCO2 even though this species is exposed to high pH variations in its natural environment (Kelly et al. 2013). Sand dollar genotypes which appear to cope well with the experimental conditions may still be subject to natural selection, as evidenced by a change in genotype rank between the CO2 environments for both

Arachnoides and Fellaster.

The heritability of plastic traits will strongly influence the adaptive potential of future generations (Schlegel et al. 2012; Wrange et al. 2014). The heritability estimates in both sand dollar species studied here are lower than heritability etimates for many other echinoderm species, particularly within the Echinoidea (urchins and sand dollars). For example

92 heritability of normal development was h2 = 0.22 in the urchin Pseudoboletia indiana in response to combined ocean acidification and warming, and this was found to be mainly due to contributions from the mother, as there was no heritability associated with sire contributions (Foo et al. 2014). Heritability for the ‘body length’ trait reduced from h2 = 0.14

2 at low pCO2 to h = 0.09 for 2-arm plutei of the urchin Strongylocentrotus franciscanus and again, strong maternal effects were detected (Sunday et al. 2011). Other heritability estimates for urchin species show that a variety of traits from body size, length and normal development are heritable across a CO2 gradient, often but not always due to strong maternal effects (Kelly et al. 2013; Delorme and Sewell 2016). There do not appear to be any other published estimates of trait heritability for sand dollars. The lack of change in heritability estimates of normal development at high CO2 for both sand dollar species in the present study may indicate weak selection potential in the future compared to other subclasses within the

Echinoidea.

In conjunction with overall high percent normal development and other larval studies, stable heritability estimates indicate that temperature is not a limiting factor in sand dollar development at high pCO2 (Gonzalez-Bernat et al. 2013). Exposure to high pCO2 reduced genetic diversity among the purple urchin Strongylocentotus purpuratus (Lloyd et al. 2016) and processes of this sort could eventually produce a range of genotyes with high tolerance to temperature change. In neither Arachnoides nor Fellaster, however, were any genetic correlations (r*G) strong enough to conclude that genotypes that are successful in current populations would also be successful in future populations. There is a relatively small amount of standing genetic variation in these species, but end-of century ocean warming conditions do not appear to challenge the survival prospects of sand dollars, therefore strong adaptive capacity is not of high importance.

b) Environmental selection

Species tolerance to the predicted changing ocean conditions will depend heavily on population exposure history and flexibility, In both F. zelandiae and A. placenta, the decrease in normal pluteus development in response to high CO2 is driven by only a small amount of phenotypic plasticity across genotypes. For Arachnoides there is a significant [female x temperature] interaction which contributes to variation in performance, while for Fellaster there is a small [male x CO2] interaction, indicating some degree of variation in genetic contribution to offspring. Both populations in this study displayed relatively small amounts of standing genetic variation compared to other sub-tropical and temperate echinoderms studied, such as the urchin Heliocidaris erythrogramma from Australia, where genotypes are highly plastic in response to a 3.0 °C increase in temperature at the hatched larval stage (Foo et al.

2016). Genotypes within these two sand dollar populations are not challenged by end-of- century predicted temperature increases, reflective of their wide thermal development window. The evolutionary implications of wide thermal tolerance and little plasticity in response to temperature increases throughout both sand dollar populations is likely to be that long-term survival will not result in genetic bottlenecks during exposure to extreme high- temperature events.

Range shifts are one possible consequence of rapid changes in an environment (Peck 2005), and could affect marine populations in several ways. Australian populations of the sea star

Acanthaster cf. solaris have been shown to have both the adaptive capacity (Sparks et al.

2016) and the broad thermal development window necessary for them to expand their range south around the Australian coastline if that area continues to warm (Lamare et al. 2014).

Both species used in the present study have important ecosystem functions as bioturbators in sandy habitats, and therefore any alterations to their distribution may affect ecosystem function. Based on the lack of adaptive capacity seen in the present study, both species will need to rely on their wide thermal and acidification tolerance ranges in order to colonise new habitats.

c) Conclusions

We estimated the contributions of genetic identity and environmental exposure to adaptive capacity in two sand dollars and concluded that the effect of parent identity is significant but small compared to the effects of environmental exposure history. This result may be influenced by the wide developmental thermal windows of the two species and the natural

CO2 flux in their intertidal (A. placenta) and shallow subtidal (F. zelandiae) habitats. Our hypothesis that these sand dollar populations would show strong heritability of tolerance to ocean warming and acidification was not supported, neither was the hypothesis that

Arachnoides would have a larger heritable genetic component to tolerance than Fellaster.

Although the adaptive capacity of both sand dollar species is relatively low, their already- existing wide tolerance range indicate that predictions of future ocean warming and acidification are unlikely to result in significant evolutionary selection pressures.

Chapter Four: Antagonistic effects of ocean acidification and warming on the New Zealand common cushion sea star Patiriella regularis metabolic energy budget and reproductive capacity over

a 12-month breeding cycle

Photo: Adult Patiriella regularis with reference scale bar. Photo credit: K. Sparks

This chapter is currently in preparation as : Sparks KM and Lamare MD ‘Antagonistic effects of ocean acidification and warming on the New Zealand common cushion sea star Patiriella regularis metabolic energy budget and reproductive capacity over a 12-month breeding cycle.’ Target publication: Journal of Experimental Biology (2016 impact factor 3.32).

This chapter pertains to:

Hypothesis 2: Metabolic energy budget

Adult echinoderms possess the physiological capability to acclimatize to ocean acidification

and warming over long time-scales without a significant cost to metabolic energy budget.

Significance: Metabolic energy budget is a reliable indicator of physiological health.

Changes to metabolism in response to external stressors will indicate acclimation, or lack thereof, to ocean acidification and warming. This has been assessed in only a few species in

response to ocean acidification, so more data is needed for species from a variety of environments in order to understand how species are ‘winners’ or ‘losers’ in climate change

scenarios.

Contributions to this chapter: Kate designed the experiment and built the aquarium system with Miles Lamare. Kate collected and analysed all the data, and wrote this chapter. Miles and Abby Smith provided feedback and edits on an early draft of the manuscript. Peter

Dillingham provided advice about statistical analyses. Kate was funded by the University of

Otago and the University of Otago Ocean Acidification Research Theme.

4.1 Abstract

Energy-related physiological indices are important metric responses of marine populations to combined ocean acidification and warming. I tested the hypothesis that adults of the New

Zealand common cushion stars, Patiriella regularis, would be capable of physiological acclimation to combined ocean acidification and warming over decadal time-scales. Adult P. regularis were exposed to all combinations of three pH(T) levels (ambient 8.10, 7.90 and

7.70) and two temperature levels (ambient and +2.50 °C above ambient) for 12 months, during which time physiological indices of gonad and pyloric cecum growth rate, righting time and tissue inorganic content were measured every three months. At the same time respiration, energy absorption efficiency, and nitrogenous excretion rates were measured to estimate Scope for Growth (SfG) in all treatments. SfG in adults was reduced by ~48 % through long-term exposure to pH 7.70, but elevated temperature at pH 7.70 enabled some physiological compensation for lowered SfG i.e. an increase in SFG of ~45 % at pH 7.70.

Reduced gonad size, pyloric indices and body organic content indicated that P. regularis can re-allocate energy away from gamete production at pH 7.70 in order to maintain homeostasis.

Reduced parental investment in offspring may carry over into subsequent generations despite the ability of the adult P. regularis to acclimate to long-term ocean acidification and warming.

4.2 Introduction

Ocean acidification and warming, driven by increasing atmospheric CO2, is expected to decrease the mean ocean pH by around 0.3 to 0.5 units by the year 2100 (Caldeira and

Wickett 2005; Law et al. 2017) and increase global mean sea surface temperature by around

0.1 °C per decade, with maximum warming occurring at latitudes >30 °S (IPCC 2013). This combination of ocean acidification and warming has been identified as a threat to marine calcifying species (Orr et al. 2006) with groups such as echinoderms identified as a particularly vulnerable phyla (Ries et al. 2009; McClintock et al. 2011). The mean pH decrease in the Southern Ocean over the past 35 years has been at a rate of 0.0011 – 0.0013 pH yr-1 and parts of the Southern Ocean are predicted to be seasonally under-saturated with

98 respect to aragonite by as soon as summer 2030 (McNeil and Matear 2008; Midorikawa et al.

2013). This reduction in ocean pH may have direct effects on marine calcifiers by increasing the energy needed to maintain basic physiological functions required for survival, such as calcification, acid-base regulation, feeding and respiration, therefore impacting whole- organism physiology (Wood et al. 2008; Melzner et al. 2009; Ries 2011; Stumpp et al. 2011).

Nevertheless, populations inhabiting shallow coastal waters, environments characterised by spatial and temporal variability in pH and sea temperature, may be better adapted to ocean warming and acidification than related species from more stable habitats such as the open- ocean and polar seas (Byrne et al. 2010).

One commonly-observed consequence of pH and temperature changes on the physiology of marine organisms is changes in the energy budget of adults (see review by Gaylord et al.

2015). Examining physiological functions such as respiration, energy storage and growth, many of which have been identified as ‘universal physiological principles’ (Pörtner 2008), can give important information not only about whether a populations’ energetic demands and expenditure are being met, but also about the stability of that species’ ecosystem function.

Populations of the copepod Calanus finmarchicus in the northern Atlantic, for example, have higher maintenance costs at high pCO2, resulting in lower productivity, which is theorised to lower the amount of energy transferred to higher trophic levels in the food web, and therefore potentially reduce the productivity of the entire pelagic Atlantic ecosystem (Pedersen et al.

2014). Studying energy-related physiological indices is therefore an important part of understanding both organism-level and ecosystem-level responses to ocean acidification and warming.

Using energy-related physiological indices as a way to measure response to environmental stressors also provides a method of comparing among species with differing habitats or modes of life, and some, such as nitrogenous excretion, have even been suggested as useful biomarkers of stress in marine ecosystems (Sokolova et al. 2012). Physiological parameters such as respiration, energy absorption and excretion can be integrated using metrics such as

Scope for Growth (SfG) which estimates the available energy for gametogenesis and growth after basic metabolic functions have been completed (Bayne and Widdows 1978; Widdows and Hawkins 1989), providing further insight into the population’s overall condition and survival capabilities (Albentosa et al. 2012). For example, exposure to low pH can impact the energy absorption efficiency (a measure of metabolic efficacy) of invertebrates such as the mussel Mytilus coruscus and sea cucumber Apostichopus japonicas (Yuan et al. 2016), and therefore reduce the overall SfG of the animal (Sui et al. 2016). Low pH has also been shown to negatively impact SfG via disruption of internal acid-base regulation resulting in an increase in excreted ammonia (Mytilus edulis, Zittier et al. 2015). Exposure to high temperature can either enhance (jewelled dog whelk Nassarius conoidalis, Zhang et al. 2015) or counteract (Nacella concinna, Schram et al. 2014) the negative effects of ocean acidification on SfG, although the factors that determine how the interaction between these two stressors affects an animal are not very clear.

The storage and allocation of energy within the body of echinoderm individuals reflects the overall condition of the population, and may reveal ecological strategies for dealing with long-term environmental stress. Energy is stored within the bodies of echinoderms, primarily in the form of lipids, and those lipids are usually found within the gonads and body walls, and for sea stars, within the pyloric caecum. The contribution of pyloric caeca to overall body weight typically varies inversely with the gonad production during a reproductive cycle

100

(Lawrence 1973; Jangoux and van Impe 1977). The pyloric index, therefore, reflects energy available for growth except for the months immediately preceding the spawning period

(Caballes et al. 2016). The inverse relationship that Asteroids typically show between pyloric and gonadal indices, as observed in Asterias rubens (Jangoux and Van Impe 1977),

Sclerasterias mollis (Barker and Xu 1991) and Patiriella pseudoexigua (Chen and Chen

1992), indicates that energy transference, in the form of yolk proteins (Alqaisi et al. 2016), from the pyloric caeca into the gonads, is the mechanism that enables a gonad to be built.

Measuring nitrogenous excretion is important in determining the sensitivity of an animal to ocean acidification. There is a well-established link between an increase in hypercapnic stress and an increase in nitrogenous excretion in response to ocean acidification (Calosi et al.

2013; Appelhans et al. 2014; Hu et al. 2014; Zhang et al. 2015; review by Stumpp and Hu

2017), the source of which has been identified as a disruption of the intra-cellular acid-base reaction. A linear relationship between increasing aqueous pCO2 and increasing nitrogenous excretion has been found for some species e.g. Mytilus edulis (Thomsen and Melzner 2010), and this makes excretion in echinoderms a useful tool in ecotoxicological studies (Sugni et al.

2007).

Long-term studies of adult physiology and energy budget in response to either ocean warming or acidification are relatively rare, but have the potential to provide insights into the long-term effects of climate change. Exposure of the sea urchin Hemicentrotus pulcherrimus to 1000 µatm pCO2 for nine months indicated that available energy was preferentially allocated to somatic growth and adult survival over gametogenesis at low pH (Kurihara et al.

2013), a strategy which is effective in the short-term but potentially reduces genotypic

101 diversity in the long-term as reduced energetic input to offspring may result in high mortality and the loss of some genetic lines. After six months’ exposure to increased (+ 200 ppm above ambient) pCO2, both gastropods (Strombus luhuanus) and sea urchins (Hemicentrotus pulcherrimus and Echinometra mathaei) had reduced growth and calcification rates

(Shirayama and Thornton 2005). Metabolic depression was observed as a survival strategy in the marine worm Sipunculus nudus after eight months exposure to seawater equilibrated with

1 % CO2 (pH equivalent 7.05) (Langenbuch and Pörtner 2004). Other studies can use pre- acclimated populations from areas naturally exposed to high temperature or low pH to study the effects of long-term multigenerational acclimation (Falfushynska et al. 2016). These studies, despite using different methodologies, pH levels and incubation times, were able to identify some common changes in animal physiology which may not be apparent in shorter- term experiments. For example H. pulcherrimus growth rates at high CO2 appeared similar to the control groups for the first 20 weeks of a 26-week experiment, before becoming significantly lower (Shirayama and Thornton 2005) and Echinometra urchin long-term growth rates were comparable in low- and ambient-pH conditions (Lamare et al. 2016) .

Metabolic suppression coupled with a reallocation of energy away from gametogenesis, the commonly-seen long-term effects of exposure to low pH, are capable of driving community- level changes over multiple generations, an area of research which needs further study in order to integrate our knowledge of how ontogeny, physiology and genotype interact to influence population-level responses to environmental change.

Long-term studies allow for estimation of the acclimation potential of a species to environmental stressors by not incorporating an acclimation period, which may cause short- term (<10 weeks) studies to either over- or under-estimate population response. Dupont et al.

(2013) found that offspring of urchin (Strongylocentotus droebachiensis) adults acclimated to

102 elevated (1200 µatm) pCO2 for nine months had significantly reduced survival chances, indicating that the reproductive capacity of the parent generation was compromised. This trans-generational effect disappeared when the parents were exposed for 16 months, however, implying a delayed but effective acclimation potential. Similarly, there was no observed change in the metabolism of Mytilus edulis exposed to low pH (7.5 and 7.1) for 36 hours, but over a longer (21 d) exposure, metabolic rates declined (Jakubowska and Normant 2015). In an extreme case, the urchin Echinometra mathaeii shows reduced growth in short term larval experiments, yet adult E. mathaeii can live their whole lives on naturally occurring CO2 vents and can exhibit elevated growth (Uthicke et al. 2016). These results indicate that the effects of ocean warming and acidification can impact reproductive capacity and larval development as well as adult physiology, but the mechanisms of acclimation as well as long-term effects of combined low pH and high temperatures are largely unexamined in a number of marine taxa.

The New Zealand common cushion star Patiriella regularis (Asteroidea: Valvatida, Verrill

1867) is a sub-tidal species with a wide latitudinal distribution throughout New Zealand and its offshore islands (from around 46° S 168°E up to 35° S 183°E) (Fenwick and Horning Jr

1980) and east to Tasmania, Australia (42° S 147° E) where it was unintentionally introduced by human transport (Waters and Roy 2004). P. regularis is a common opportunistic scavenger, and important in rocky-shore coastal food ecosystems by feeding on dead and moribund animals (O’Loughlin et al. 2002). It is especially well-suited to a long-term study of acclimation because it is able to tolerate the short-term changes to physical and chemical seawater parameters which arise from living in coastal environments, and therefore should make a robust study species.

103

Patiriella regularis has a gametogenic cycle that lasts 12 months and culminates in a broad reproductive season spanning December-March (Crump 1971). Patiriella regularis maternal provisioning to eggs is lower than other Patiriella species, reflected in the relatively small oocyte diameter (~150 mm) and lower energetic content (Byrne and Barker 1991; Moreno and Hoegh-Guldberg 1999). Patiriella regularis have planktrophic larvae, and it has been found that the fertilised embryos and blastulae are robust to changes in seawater pH and temperature (Byrne et al. 2013; Karelitz et al. 2017), probably reflecting the wide range of different capnic and thermal environments encountered by drifting larvae. Research into adult thermal and hypercapnic developmental tolerance and acclimation is needed to address the implications of those results published by Byrne et al. (2013), and to understand the continued range expansion of this species north into Australia in the context of ocean warming and acidification.

This study quantifies physiological acclimation responses in adults to elevated temperature and pCO2 over a 12-month period, and identifies which components of the energy budget of adult Patiriella regularis from New Zealand are affected by long-term exposure to combined ocean acidification and warming. I hypothesised that if the overall energy budget of the animal is reduced, P. regularis will prioritise survival of the adults over energy investment to gametes. I also investigated the potential for physiological acclimation over a long exposure time, which may differ from the responses seen in the short-term, including the potential for maintenance of long-term homeostasis.

104

4.3 Methods

a) Animal Collection and Acclimation

One hundred and eighty adult Patiriella regularis (Verrill 1867) were collected by hand and by benthic dredge from sub-tidal benthic sites from around Latham Bay (45.83° S 170.64° E) on the Otago Peninsula, New Zealand in January 2014 and transported immediately to the nearby Portobello Marine Laboratory, where these experiments were conducted.

Animals were kept in 65 L flow-through seawater aquaria for the duration of the experiment, which was 12 months. After collection, thirty animals were randomly assigned to each of six aquaria and held for three weeks in ambient conditions with seawater from Otago Harbour.

After this initial holding period, temperature and pH were adjusted gradually over a period of

12 weeks to meet the target conditions. Temperature was increased at a rate of 0.2 °C every three days until the eventual temperature was 2.5 °C above ambient in the warm-water treatments. pH was adjusted by 0.3 units every three days, on alternating days to temperature adjustment, until the target pH levels were reached. After the acclimation period, the temperature and pH were held steady at the target levels although allowed to vary within set limits with the diurnal cycles ( +/-0.8 °C of mean; 8.25 to 7.98 pH units) of Otago Harbour

(Fig. 4.1)

b) Experiment design and manipulation of temperature and pH

Two temperature treatments were used for this experiment; the control, which was allowed to vary with the incoming water from Otago Harbour, and a temperature which was always 2.5

°C above the harbour temperature. Temperature was adjusted using Tunze Temperature

105

Controllers (model 7028/3; Tunze Aquarientechnik, Penzberg, Germany) attached to Jäger

Aquarium Heaters within a tolerance range of +/- 0.5 °C of the target. Target temperature

(2.5 °C above ambient) was set and adjusted twice daily using the temperature patterns over the past three days Temperature was recorded and logged every 30 minutes using a HOBO

Diamond Pendant Logger (Onset Computer Corporation, Bourne, Massachusetts, USA).

Aquarium pH(T) was altered using bubbling of pure gaseous CO2 and controlled to within 0.05 units of target by a Tunze pH/CO2 Controller Set (model 7074/2; Tunze Aquarientechnik,

Penzberg, Germany). Electrodes were typically calibrated daily and pH readings were checked every second day spectrophotometrically using m-cresol purple and a Hitachi U-

1100 Spectrophotometer (Metrohm AG, Herisau, Switzerland). Every 3 months, water samples were taken, fixed with 250 µL HgCL2 and analysed at the University of Otago’s

Waterworld Chemistry Lab for chemical parameters including total alkalinity and dissolved inorganic carbon. These parameters are given in Appendix 10.3

The control treatment pH varied with the Otago Harbour water but was on average 8.10 throughout the year. There were two low-pH treatment targets; 7.90 and 7.70. These treatments were selected based on the end-of-century predicted conditions for the Southern

Ocean (McNeil and Matear 2008; IPCC 2013) but were also allowed to vary within a range set by the pH variations in the control treatments, adjusted daily based on the previous three days’ natural pH change in Otago Harbour. This was generally a pH change of +/-0.06 over a twelve-hour tidal cycle. Temperature and pH treatments were crossed in a fully orthogonal experimental design to give six treatments in all combinations of three pH levels and two temperature levels. Temperature and pH in the treatments are shown in Figure 4.1. There were no controls on photoperiod in this experiment, however as the location of the experiment was on the edge of Latham Bay, the collection site for wild individuals, the

106

photoperiod inside the aquaria was allowed to vary in natural cycle matching those of Otago

harbour.

Figure 4.1: Temperature (left) and pH(T) (right) in the aquaria used to acclimate adult Patiriella regularis to all combinations of two temperature (ambient; elevated 2.5 °C above ambient) and three pH (ambient; 7.90; 7.70) treatments. Aquaria used were 65 L flow though tanks. Temperature data are weekly averages from n = 336 measurements. pH data are means from n = 20 measurements.

Error bars represent S.D.

At the site of the collection, Patiriella regularis has a peak reproductive season in the Austral

midsummer, late December – January (Crump 1971; Byrne and Barker 1991). During the 12-

month incubation in the experimental treatments, physiological measurements were taken at

3-monthly intervals starting immediately after the reproductive period of the previous year

and covering the period from then until the end of the next years’ reproductive season. The

107 measurements taken were body size, inorganic content, respiration rate, feeding rate, food absorption efficiency and nitrogenous excretion rate, turning time, and gonad and pyloric weights. These data were used to calculate metabolic energy budget and an approximate

Scope for Growth estimate for each animal in each treatment, as well as to assess whether the animals were capable of physiologically acclimating to each combination of pH and temperature. To compare with changes in wild populations, at each time point (0 weeks = Feb

2015; 20 weeks = June 2015; 36 weeks = October 2015; 50 weeks = January 2016) measurements were repeated with 15 fresh-collected wild individuals from Latham Bay.

c) Body Size and Tissue Weight Relationships

Animals were wet-weighed every month throughout the 12-month period. Every three months, a subset (n = 8) of animals from each treatment was dissected, and the body components including pyloric caeca, gonad and body wall were wet-weighed separately.

These components were then dried at 60 °C overnight and dry-weighed once cool. Gonad and pyloric indices were calculated as % dry body weight for animals from each treatment at each time point. Arm length was also tracked monthly in each treatment to assess growth rate.

Body wall and skeletal components were ashed in a muffle furnace (450 °C for 6 hours) to determine inorganic content.

d) Respiration Rate and Ammonia Excretion

Respiration rate was measured in a sub-set (n = 8) of animals from each treatment at 3-month intervals throughout the acclimation process, including pre-acclimation and after 12 months.

Animals were placed in sealed 300 ml chambers of seawater from their respective treatments and the oxygen depletion rate was measured fibre-optically using a PreSens FibOx 3 sensor

108 system, with temperature-compensated oxygen measurements taken using fluorescence from a PreSens Oxygen Sensor Spot (SP-PSt3-NAU) (FibOx sensor, oxygen meter and fluorescent spots PreSens Precision Sensing GmbH, Regensberg, Germany). Each sealed chamber was placed in a water bath at the appropriate experimental temperature for each treatment and oxygen depletion measured over a 6-hour period, or the time taken for oxygen saturation level to drop to a minimum of 80 %, whichever was the shorter time period. Vertical stratification and changes to microbial respiration rate were minimised by the use of mechanical stirrers whenever the chambers were sealed and the experiments in progress.

Animals were wet-weighed before and after the experiment. After the experiment, animals were killed and dry-weighed to normalise respiration rate to body dry weight. During both respiration and excretion experiments, background oxygen depletion was measured using a blank chamber for each run filled with seawater at the same carbon dioxide and temperature conditions as the animal chambers.

Water samples from each respiration chamber were analysed for ammonia (NH3) content using a continuous-flow injection analysis Auto-Analyzer (LaChat QuikChem 8500) at

Portobello Marine Laboratory. Analytical protocol followed the Berthelot reaction to measure ammonia concentration in seawater relative to indophenol blue absorbance at 630 nm

(QuikChem Method 31-107-06-1-B). The system was calibrated using deionized water and spectral analysis was performed using in-house software.

e) Food Absorption Efficiency

Animals were fed on a mixed diet throughout their acclimation time, mimicking the food available in their natural environment. The diet included spotted topshells (Diloma spp.), blue cod (Parapercis colias) muscle tissue, cockles (Austrovenus stutchburyi) and other small

109 invertebrates. During the three-monthly feeding measurements, however, the food source was kept the same; blue cod muscle. Pre-starved (10 d) animals were given a pre-weighed sample of fresh blue cod and allowed to eat for three hours or until satiation, whichever was the shorter time period. The food was then weighed and the eaten mass calculated. Energetic content of blue cod muscle tissue (8.2 J g-1) was determined experimentally by combustion and checked against commercial estimates. Feeding experiments took place in the animals’ acclimation treatments. Animals were wet-weighed before the feeding experiment to normalise feeding rate to body size. After each feeding experiment, faeces were collected over a 36-hour period, weighed and dried to determine organic content using the following equation:

퐴 = 퐼 − 퐹

Where A is the absorption efficiency, I is the energy ingested in food and F is the energy loss as solid faeces.

f) Righting Time

‘Righting time,’ the time taken for an animal inverted on its’ aboral (dorsal) surface to right itself completely and replace all five arms onto a flat surface, was measured in a subset of fed animals from each treatment every three months. Righting time was recorded by video and analysed. Each animal was inverted a maximum of ten times with a break of no more than one minute in between each turn. Righting time was normalised to arm length for analysis.

110

4.4 Data Analysis

Tissue weights (wet, dry and ash-free) were analysed by linear regression to describe the relationship between the indices. Gonad weight and pyloric weight were expressed as a percentage of dry body weight for their respective indices. Respiration rate was calculated as

-1 -1 oxygen consumption O2 µg h g dry weight . Ingestion was calculated as grams food eaten per gram body weight per hour, and absorption efficiency was the % of ingested energy that remained in the body. Excretion was calculated as mg NH3 per litre per hour per gram body weight. Measurements of oxygen consumption and nitrogen excretion were used to calculate the O:N ratio at each time point (Conover and Corner 1968). Respiration, feeding and excretion data were converted into Joules and used to calculate Scope for Growth following a modified Winberg equation (Winberg 1960), which states that:

푃 = 퐴 − (푅 + 푈)

P is the Production, otherwise referred to as Scope for Growth, and represents the difference between the energy absorbed from food and energy expenditure for respiration and feeding. A is the product of consumption, or the Absorption Efficiency of ingested food, R is the

-1 Respiration converted to Joules using the oxyenthalpic ratio (527 kJ mol O2 ), and U is the energy loss as excretion.

Data were tested for significant differences using ANOVA and Holm-Sidak post-hoc tests.

111 captive population and an open source population. The ‘wild’ population was included in analyses only to verify the validity of the laboratory control as a simulation of the condition of the wild population at that time point. This analysis assumed that the effects of captivity on overall health would be negligible compared to the differences between temperature and pH treatments.

Scope for Growth calculations were plotted against mean body wet weight for each treatment after 12 months incubation to normalise for the slight differences in mean animal weight and size between treatments.

4.5 Results

a) Body weight indices

After 12 months acclimation, there were no significant differences in the body wet, dry or ash-free weights of animals among any treatments. In all treatments apart from pH7.70, there was no change in body wet weight at the start and end of the experiment. In the pH 7.70 treatment at ambient temperature, body wet weight declined by on average 4.20 % throughout the experiment. There was also no difference between the weights of animals from the wild population versus the laboratory control population. See Appendix 10.3 for mean body weight data.

Pyloric index was lowest overall in December and January, on average 0.95 % body dry weight in all lab-held populations, and highest (3.32 %) in the wild population (Fig. 2).

Pyloric index increased from January to September (Austral spring) when it peaked in all treatments. Mean pyloric index in the individuals acclimated to pH 7.70 at both ambient and

112 elevated temperatures (+ 2.50 °C above ambient) were significantly lower than in individuals acclimated to all other treatments after 6 months acclimation (from June onwards) (ANOVA

F = 2.45 df=41 p = 0.043). Between September and December (Austral spring to early summer) the mean pyloric index decreased in all treatments (Fig. 4.2).

Average gonad index was lowest in January, ranging from 0.40 – 1.65 % dry weight for all laboratory treatments and the wild population (Fig. 4.2a). Gonad index remained low until

June (range = 0.80 – 2.04 % dry weight), before increasing from September to December, which spans the months leading up to the reproductive season in this species. Gonads in ambient pH/ + 2.50 °C treatment in September and December were significantly (2.35 %) larger than in other treatments (ANOVA F = 17.16 df=41 p = 0.035; Holm Sidak t = 5.397 for September and t = 2.619 for December). The only treatment in which the gonad index did not increase from June to December was the pH 7.70 treatment at ambient temperature.

Gonad index in that treatment remained constant at 2.15 % dry weight, compared with in all other treatments where the gonad index increased to over 4.0 % dry weight in December (Fig.

4.2b).

b) Respiration Rate

-1 -1 Respiration rate in the wild population was fastest at 6.54 µg O2 g dry weight h in February

-1 -1 and slowest at 2.00 µg O2 g dry weight h in June, and was comparable to that of the laboratory control population at all times measured (Fig. 4.3). There were no significant differences between respiration rate at either the start or the end of the acclimation period for the control treatment. Individuals acclimated to elevated temperature at all pH levels had

113

significantly faster respiration rates (ANOVA F = 12.90 df=41 p = 0.05) than all adults in the

ambient temperature treatments at the start of the acclimation period. Respiration rate was

slowest for all acclimated populations in winter (June) with an overall mean of 2.02 µg O2 g

dry weight-1 h-1 +/- 0.29 S.E., and stayed low through October before rising to a peak in

summer for all treatments. After 12 months acclimation (December), the only treatment in

which respiration rate was significantly lower than that of the laboratory control treatment

-1 -1 (6.03 µg O2 g dry weight hr ) was in individuals acclimated to pH 7.70 at ambient

-1 -1 temperature (3.36 µg O2 g dry weight hr ) (ANOVA F = 5.23 df=41 p = 0.001), which was

16.05 °C at that time (Fig. 4.3).

Figure 4.2: Pyloric (A) and gonad-somatic (B) indices for Patiriella regularis adults acclimated for 12 months to combinations of two temperature (ambient; elevated 2.5 °C above ambient) and three pH

(ambient; 7.90; 7.70) treatments, as well as the wild population sampled from Otago Harbour. Bars of different shades represent different months throughout the acclimation period (see legend). Data represent mean values (n = 24 for all sampling points) and error bars represent 1 S.E. Both indices are expressed as tissue dry weight as a percentage of total body dry weight. Bars with different letters above are significantly different (p<0.05)

114

Figure 4.3: Respiration rate of adult Patiriella regularis acclimated for 12 months to

combinations two temperature (ambient; elevated 2.5 °C above ambient) and three pH

(ambient; 7.90; 7.70) treatments as well as respiration rate in the wild population from

Otago Harbour. Data are mean respiration rates for each time point (n = 8 for all). Error

bars represent 1 S.D. Bars with different letters above are significantly different (p

<0.05).

c) Absorption and excretion of energy from food

The solid faeces of the wild population were energy-rich (December mean 21.29 J mg-1) compared to the captive control population (December mean 12.03 J mg-1) at the end of the experimental period (12 months). In turn, the energy content in the faeces of individuals

115 exposed to pH 7.70 after 12 months (December mean = 6.53 J mg-1) was significantly lower than the control (December mean = 12.03 J mg-1) (ANOVA F= 8.49df=27 p = 0.01; Holm-

Sidak t = 1.86 p = 0.01 for [Ctrl vs. pH 7.70] interaction within December time-point) despite comparable amounts of food ingested. Absorption efficiency of food was also reduced (by around 5.50 %) after 6 months exposure to pH 7.90 (ANOVA F= 3.67 df=27 p=0.017) and

7.70 (ANOVA F= 5.37 df=27 p=0.04) at ambient temperature versus the control treatment and at pH 7.70 at elevated temperature (ANOVA F = 7.54 df=27 p=0.005) and were still significantly lower than in the control population after 12 months acclimation (Fig. 4.4).

Lowest absorption efficiency was observed in the pH 7.70 treatment at ambient temperature, with a yearly mean of 54.11 %, and the highest absorption efficiency was observed in the control +2.5 °C treatment at ambient pH, with a yearly mean of 68.33 %.

Nitrogenous excretion between the wild population and the laboratory control population were comparable at every time point sampled, staying within a range of 4.67 – 7.69 µg NH3 g dry weight-1 h -1 throughout the year. The excretion of nitrogenous waste is significantly increased by exposure to pH 7.70 versus the control population (Holm-Sidak t = 3.89 p=0.001) (Fig. 4.5). The treatment within which there was the largest difference in excretion between the months was the control +2.5 °C treatment in January, when excreted ammonia levels were almost 4x as high as the control treatment (Holm-Sidak t = 5.42 p=0.001) (Fig.

4.5).

116

Figure 4.4: Energy absorbed by adult Patiriella regularis exposed to combinations of two temperature (ambient; elevated 2.5 °C above ambient) and three pH (ambient; 7.90; 7.70) treatments for 12 months as well as the wild population from Otago

Harbour. Data are means from n=8 individuals calculated as the difference between the energy ingested and excreted. See Appendix 10.4b for values and energy absorption efficiencies. Error bars are S.E. Bars with different letters above are significantly different (p<0.05).

117

Figure 4.5: Nitrogenous excretion (NH3) of adult Patiriella regularis exposed to two

temperature (ambient; elevated 2.5 °C above ambient) and three pH (ambient; 7.90;

7.70) treatments for 12 months as well as data from a nearby wild population from

Otago Harbour. Data are means from n=8 individuals. Error bars represent S.E. Bars

with different letters above are significantly different (p<0.05).

The oxygen: nitrogen (O:N) ratio for each population varied over the acclimation period. The ratios of the wild and laboratory control populations were not comparable for the first six months of the acclimation period. The O:N ratio was significantly lower in the pH 7.90

(Holm-Sidak t = 5.22 p<0.001) and pH 7.70 (Holm-Sidak t = 5.34 p<0.001) treatments versus the control at all time points but there were no interactive effects of temperature and pH on these populations (Fig. 4.6).

118

Figure 4.6: Oxygen:Nitrogen ratios of adult Patiriella regularis over 12 months exposure to two temperature (ambient; elevated 2.5 °C above ambient) and three pH (ambient; 7.90; 7.70) treatments as well as for a nearby wild population from Otago Harbour. Data are means from n = 8 individuals.

Error bars represent S.E. Bars with different letters above are significantly different (p>0.05).

d) Scope for Growth

Scope for Growth (SfG) in the wild population and the laboratory control were not

significantly different, indicating that the control treatment in this experiment accurately

modelled the energetic budget of the wild population in their natural environment. Increasing

temperature did not significantly affect SFG regardless of pH level. However, the SFG was

119 significantly decreased by exposure to pH 7.90 and pH 7.70 in the absence of elevated temperature after 12 months acclimation (ANOVA F = 4.78 df=125 p = 0.048) (Fig. 4.7).

Figure 4.7: Scope for Growth of adult Patiriella regularis after 12 months exposure to

two temperature (ambient; elevated 2.5 °C above ambient) and three pH (ambient; 7.90;

7.70) treatments as well as for a nearby wild population from Otago Harbour. Data are

means from n = 15 individuals. Error bars represent S.E. Bars with different letters above

are significantly different (p<0.05).

120

e) Righting time

Elevated temperature decreased average righting time by 15 sec after 3 months (June)

(ANOVA F = 2.10, d.f. = 144, p = 0.045). Low pH (both 7.90 and 7.70) increased average righting time by 32 sec (ANOVA F = 7.18 d.f. = 144 p = 0.038) in the first 6 months of the acclimation period, but after that time, there were no significant differences in times among the pH or temperature treatments. (Fig. 4.8).

Figure 4.8: Turning time of adult Patiriella regularis over 12 months exposure to

two temperature (ambient; elevated 2.5 °C above ambient) and three pH (ambient; 7.90; 7.70) treatments as well as for a nearby wild population from Otago Harbour.

Data are mean from n = 10 individuals each inverted 8 times. Error bars are S.D.

Bars with different letters above are significantly different (p<0.05).

121

4.5 Discussion

I hypothesised that adult sea stars Patiriella regularis will be capable of long-term acclimation to combined ocean acidification and warming forecasted for the end of the 21st century. Over a 12-month period, the physiological and energetic effects of exposure to combinations of +2.5 °C temperature increase and -0.2 and -0.4 pH units were measured.

Overall, exposure to pH 7.90 and pH 7.70 altered the metabolic energy budget of P. regularis, resulting in low Scope for Growth values, however, exposure to elevated temperature at both of these low pH levels allowed the animals to maintain an energy budget comparable to that of a current-day wild population. The key finding of this study is that long-term physiological acclimation to end-of-century projected temperature and pH conditions is possible in this species, albeit with potential negative consequences. In this respect, Patiriella regularis populations around the south coast of New Zealand should be able to persist as adults, but the reallocation of energy away from reproductive capacity at low pH may result in reduced recruitment in subsequent generations.

a) Mechanisms of maintaining Scope for Growth at high temperature and low pH

Mechanisms of physiological maintenance at higher temperatures compromised aerobic metabolism and excretory function in this study. An increase in water temperature of 2.5 °C at pH 7.90 and 7.70 reduced aerobic respiration by on average 15 % after 12 months relative to the control population, and disrupted other metabolic functions such as righting time

(~2.15 % faster at high temperature/ pH 7.70) and nitrogenous excretion (~200 % increase at high temperature/ pH 7.70). For invertebrates, metabolic responses to low pH appear to be highly species-specific, as both metabolic suppression (Mytilus galloprovincialis Michaelidis et al. 2005; Porites asteroides Albright and Langdon 2010) and metabolic increase

122

(Crassostrea gigas Lannig et al. 2010) have been observed. In general, though, metabolic suppression has been identified as one of the most important responses to ocean acidification

(Kelly and Hofmann 2012) and this study suggests that metabolism and energy budget remain tightly coupled during exposure to low pH and high temperature, so that metabolism is still a useful proxy for physiological state during experimental pH and temperature alteration. .

Increased energy loss to nitrogenous excretion is indicative of internal acid-base disturbance and is a known consequence of exposure to low pH (Pörtner et al. 2004). Sea urchins

(Paracentrorus lividus) are able to compensate for short-term energy imbalances by up- regulating metabolism and intra-coelomic pH buffering to maintain stable excretion levels

(Catarino et al. 2012) while exposed to pH 7.70 and 7.40 for up to 19 days. Some species such as the mussel Mytilus edulis can experience metabolic rates up to 60 % higher than normal when exposed to high temperature and low pH due to an increase in nitrogenous excretion (Lesser 2016). Other studies, however, have found that despite increasing nitrogenous excretion with increasing pCO2 there can be very small effects on overall energy balance (Thomsen and Melzner 2010). Long-term inability to regulate excretion has a high energetic cost, potentially inhibiting the ability of P. regularis to return to pre-acidification homeostatic norms, despite appearing acclimatized.

Within the Echinodermata, echinoids such as urchins are able to compensate for extracellular pH changes (Stumpp et al. 2011; Dupont and Thorndyke 2012; Taylor et al. 2014; Leite

Figueirdo et al. 2016). This phenomenon does not, however, occur within the Asteroidea;

Leptasterias mollis, for example, was not able to compensate for extracellular pH changes

123 over five days (Dupont and Thorndyke 2012). It is possible that the results observed in L. mollis are due to a regulatory mechanism which is not conserved across Echinoderm taxa, possibly due to more efficient buffer capacity in the coelomic fluid of urchins compared to sea stars and holothurians (Collard et al. 2013).

The results of this study suggest that long-term acclimation to pH 7.70 induced individuals to transition into the unfavourable pessimum physiological range (rather than optimum or pejus)

(Frederich and Pörtner 2000) which is characterised by metabolic depression and increased maintenance costs (Sokolova 2013). It is also associated with a reduction in movement and activity that is demonstrated in this experiment by the increase in righting time at pH 7.70 for the first 9 months of the exposure period. The metabolic depression in P. regularis exposed to pH 7.70 was accompanied by a reduction of ~45 % in energy absorbed in the first six months of the acclimation period and ~15 % thereafter versus the laboratory control (>50 % reduction in absorbed energy versus wild population) was observed. Physiological transition with the pessimum range is regarded as a method of ensuring short-term survival by reducing reproductive and somatic growth demands at the expense of long-term persistence (as shown in the molluscs Laternula elliptica and Austrocochlea constricta; Peck et al. 2004; Leung et al. 2017, respectively) and the ecological implications of existing in this compromised state over long time-scales without any acclimation potential could have negative effects over multiple generations.

Evidence for energy resource re-allocation in individual sea stars within the pessimum range is shown by the low Pyloric Index (PI) and Gonad-Somatic Index (GSI) of animals acclimated to pH 7.70. At the whole-animal level, reductions in these indices indicate that

124 during a period of energy deficit, maintaining internal homeostasis is prioritised over somatic growth and reproductive capacity, as has been demonstrated for other echinoderm species in long-term experiments (e.g. Delorme and Sewell 2016). The urchin Echinometra spp. was able to maintain metabolic normality (measured using coelomic fluid pH) at combinations of

+3 °C and ~900 µatm CO2 (equivalent to pH 7.90) when exposed for 10 weeks, but reduced gonad weight indicated compromised reproductive function (Uthicke et al. 2014). Similarly, the urchin Strongylocentrotus droebachiensis was able to compensate for long-term (45 day) exposure to low pH by re-allocating energy away from somatic and gonadal growth (Stumpp et al. 2011). Just as for P. regularis here, S. droebachiensis experienced decreased O:N ratios, a net decrease in energy budget and a small decrease in respiration rate at low pH. The same pattern was observed for Echinometra spp. from natural CO2 vents (Uthicke et al. 2016).

Despite their natural environment being characterised by wide pH fluctuations, these echinoderm populations do not appear to have the physiological capability to acclimatize to long-term pH depression without increased temperature.

The reduction in GSI seen at pH 7.70 is often accompanied by a change in primary metabolite, from protein to carbohydrate, which may indicate that energetic input to gonads is also reduced as a compensatory mechanism during metabolic depression. The low O:N ratios in the pH 7.70 treatment throughout the exposure period in this experiment indicate that this population was metabolising carbohydrates and lipids despite having a high-protein diet, as in general an O:N ratio lower than 8.0 is interpreted as a change in primary metabolite from protein to carbohydrate (Mayzaud 1973; Langenbuch and Pörtner 2002). The likely source of the carbohydrates for this population is the stored energy associated with the pyloric caeca, which is generally higher in carbohydrates in females than it is in males, and is directly relatable to reproductive capacity (Xu and Barker 1990). The observed reduction in both

125

Pyloric and Gonad-Somatic indices in the pH 7.70 population compared to both the wild and captive control populations indicates that individuals at pH 7.70 utilised their energy stores to maintain metabolism, potentially at the cost of reproductive capacity. As enhanced protein metabolism contributes to the active pumping of excess hydrogen ions out of the body

(Thomsen and Melzner 2010), the switch away from protein metabolism seen in this study could further stress individuals at pH 7.70 by not allowing them to efficiently regulate their internal pH.

b) Ecological implications of acclimation to forecasted end-of-century conditions

Based on the majority of bio-energetic indices in this study, physiological acclimation to end- of-century conditions is possible for adult Patiriella, although importantly, potentially at a cost to future generations. Patiriella regularis was able to maintain SFG comparable to present-day populations during exposure to a combination of +2.5 °C and -0.4 pH units, but only at the expense of internal energy reserves. Other studies on SFG within marine populations in response to ocean acidification and/or warming have often concluded that despite short-term disruption, overall fitness is robust to predicted ocean carbonate chemistry changes after long-term exposures. For example, copepod (Calanus finmarchicus) short-term metabolic imbalance was corrected with multi-generational exposure to 2080 µatm pCO2, leading to their ability to withstand forecasted ocean conditions. Short-term disruption to fitness, as observed in this study however, may not disappear after a few generations as in C. finmarchicus as each generation of P. regularis may experience depletion of energy reserves resulting in a cumulative reduction of fitness within a few generations.

126

Increased temperature and low pH therefore may have antagonistic effects on metabolism in this population. Acclimatization to combined temperature increase and pH decrease for 12 months in this study enabled individual sea stars to maintain a respiration rate comparable to both the wild population and the laboratory control. In the +2.5 °C/ pH 7.70 treatment after

12 months, the overall energy budget of the population was not affected. Maintaining SFG in the face of pH stress alone can decrease metabolic rate (Gianguzza et al. 2014), however as for S. droebachiensis (Stumpp et al. 2014), the antagonistic effect of temperature and pH on metabolism should allow this population of P. regularis to persist in the short term. Adult acclimation to end-of-century conditions appears to be dependent to a large extent on metabolic responses in Patiriella, and the evidence for complete metabolic acclimation within one year depends greatly on the interaction between multiple stressors.

Photoperiod, often linked to temperature, can exert some control over gametogenesis, feeding and phenology of echinoderms (Pearse and Eernisse 1982, Pearse and Beauchamp 1985). Sea star growth rates are not under strict photoperiodic control, but there is evidence suggesting that the rate of gametogenesis is faster when urchins are exposed to a shorter photoperiod

(Shpigel et al. 2004) although sea stars may not follow the same pattern (Pearse et al. 1986).

In this study, photoperiod was not controlled for, however as the aquaria were situated immediately next to the collection site for wild individuals, it was assumed that there was no effect of laboratory incubation on the phenology of the animals. Although no measurements of oocyte size or ripeness were made, the captive population is assumed to have undergone a normal gametogenic cycle. Ripe gonads were observed in the captive and wild populations at the same time. The only significant difference in the gonad-somatic index between the wild population and the laboratory control was at the September timepoint, but as the two

127 populations had comparable gonad sizes again by the peak reproductive period, this earlier difference is assumed not to affect gametogenesis.

As an important and widespread species in New Zealand and Tasmania, P. regularis currently exerts a strong influence on marine benthic communities by scavenging dead organic matter and recycling energy through the marine food web. In the future, the decreased aerobic metabolism and decreased food energy absorption shown in this study could have implications for changes in trophic processes in the wider marine community.

Even though the energy budget of the animal is sustainable in the short term, this inability to maintain food energy uptake may alter community dynamics and energy flow through the marine food web. Although marine populations have been observed to migrate polewards in response to ocean warming (Sunday et al. 2015), the projected rapid decrease in pH of the

Southern Ocean and under-saturation with respect to aragonite (McNeil and Matear 2008) may influence latitudinal changes in P. regularis distributions.

c) Conclusions

Adult acclimation of Patiriella regularis to forecasted end-of-century temperature and pH conditions is possible, although maintaining a stable energy budget in those conditions may come at a significant cost to reproduction, energy storage capacity and future generations, as we hypothesised. I was able to determine which components of the metabolic energy budget were affected by exposure to +2.5 °C/ pH 7.70 for 12 months, as aerobic metabolism and food energy absorption were suppressed but able to be compensated for by metabolism of stored carbohydrate energy reserves, while increased nitrogenous excretion showed no acclimation potential. This physiological state, in which multiple bio-energetic components

128 are compromised, is not sustainable for individual P. regularis, however, unless food energy intake can be significantly increased, something that seems unlikely given the reduced activity rates and food absorption efficiency in a high temperature/ low pH environment.

Short-term survival prospects are increased at the cost of long-term survival, and the persistence of this species may rely on its ability to overcome the barriers of reduced metabolism and energy intake.

129

Chapter Five: Physiological acclimation of the Antarctic cushion

star Odontaster validus to elevated temperature, but not low pH,

over 9 months significantly alters metabolic energy budget

Photo: adult Odontaster validus. Photo credit K. Sparks

This chapter is currently in preparation as: Sparks KM, Aguera AA and Lamare MD ‘Long- term physiological acclimation of the Antarctic cushion star Odontaster validus to elevated temperature, but not low pH, significantly alters metabolic energy budget.’

Target publication for this chapter: Polar Biology (2016 impact factor 1.949)

130

This chapter pertains to:

Hypothesis 2: Metabolic energy budget

Adult echinoderms possess the physiological capability to acclimatize to ocean acidification

and warming over long time-scales without a significant cost to metabolic energy budget.

Significance: metabolic energy budget is a reliable indicator of physiological health.

Changes to metabolism in response to external stressors will indicate acclimation, or lack

thereof, to ocean acidification and warming. It has been assessed in only a few species in

response to ocean acidification, so more data is needed for species from a variety of environments in order to understand how species are ‘winners’ or ‘losers’ in climate change

scenarios.

Contributions to this chapter:

Kate Sparks designed the concept. Kate and Miles Lamare built the aquaria. Kate collected all data with assistance from Antonio Agüera, Miles Lamare and Michelle Liddy. Kate analysed the data and wrote this chapter with edits from Miles. Steve Wing collected the animals and assisted with transportation and importing into New Zealand. Antonio Agüera gave feedback on the experimental design and aquarium set-up. Steve and Kate received assistance and funding from Antarctica New Zealand. Antonio was funded by the Université

Libre de Bruxelles.

The authors acknowledge the assistance of Dave Wilson, Doug Mackie and Reuben Pooley at

PML for their help in building the aquaria.

131

5.1 Abstract

In the Southern Ocean, anthropogenic warming is expected to increase the mean water temperature by up to 2.0 °C by the end of the 21st century. At the same time, rapid acidification may decrease the mean pH by up to 0.5 units. This is thought to have negative consequences for the life histories of native cold-adapted marine animals. The ubiquitous

Antarctic cushion sea star, Odontaster validus, may be one of the most thermally-tolerant polar species and therefore may be one of the ‘winners’ of future climate change. In order to test the hypothesis that long-term acclimation to low pH and elevated temperature is possible, adult O. validus from the Ross Sea, Antarctica were exposed to combinations of a +3.0 °C temperature and -0.4 pH unit decrease for 9 months, following a 3-month period of gradual temperature and pH change. During this time, physiological and metabolic parameters were measured and the data used to estimate Scope for Growth (SfG) for this population. Our hypothesis was supported, as there were no reductions in overall SfG of animals subjected to combined warm and acidified seawater. Exposure to elevated temperature in the absence of acidification significantly increased SfG by ~41 % relative to the control treatment.

Maintenance of metabolic energy budget was sustained by an increase in food intake to compensate for increased respiration and excretion at elevated temperature and low pH.

Individuals had the same reproductive capacity in every treatment, indicating that stored energy reserves are sufficient to maintain base metabolism as well as somatic growth and reproductive output.

132

5.2 Introduction

The Southern Ocean (south of 40 °S) has experienced some of the most rapid warming and acidification on the planet throughout the end of the 20th and beginning of the 21st centuries, and the ecological effects of these changes are already being seen including sea ice break-up, declines in krill productivity and changes to polar upwelling and circulation (Böning et al.

2008; IPCC 2013; Constable et al. 2016; Meredith 2016; Piñones and Federov 2016). There are large uncertainties around the extent and variability of Antarctic ocean warming in the future, and the contributions of sea ice melt, changes to surface albedo and anthropogenic forcing are still being debated (Mengel et al. 2016). Mean Southern Ocean temperature has already increased by 0.17 °C since the 1950s and is predicted to increase further by up to 2.5

°C by the end of the 21st century (RCP8.5; IPCC 2007, 2013; Gille 2002).

The Southern Ocean is also a carbon sink, responsible for absorbing up to 40 % of the global ocean CO2 uptake since the 1850s and resulting in rapid acidification (Sabine et al. 2004;

Mikaloff-Fletcher et al. 2006). Southern Ocean acidification and seasonal aragonite undersaturation may be some of the most pressing problems facing Antarctic marine biota

(Gutt et al. 2015), especially as the variability and duration of acidification and aragonite undersaturation events may make it difficult for marine organisms to acclimatise or adapt

(Sunday et al. 2014; Hauri et al. 2015).

The ecophysiological effects of ocean acidification and warming on Antarctic marine communities are expected to be abrupt, rapid and severe (Peck 2002; Morley et al. 2016).

Marked seasonality of food availability to the benthos can limit the growth rates of marine ectotherms during certain times of year (Clarke 1988). Standard metabolic rates are generally

133

2-10 x slower in Antarctic species than in related temperate and tropical species (Ulbricht and

Pritchard 1972; Peck 2002), for example metabolic rate in the Antarctic sea urchin

-1 -1 Sterechinus neumayeri was 0.46-0.65 µmol O2 g h at -1.0 °C whereas the metabolic rate of

-1 -1 the New Zealand urchin Evechinus chloroticus can be 1.5 - 2.0 µmol O2 g h at 13 °C

(Brockington and Clarke 2001; Brockington and Peck 2001). The energy available to polar animals for movement, growth and reproduction is therefore also limited (Johnston et al.

1991) and this could potentially make it harder for Antarctic biota to acclimatise to the increased metabolic demands of high temperature and low pH.

Antarctic marine invertebrates are suggested to be cold-adapted and therefore temperature- sensitive compared to temperate and tropical species (Morley et al. 2016; Peck 2016). A recent meta-analysis has shown that although local extinctions among Antarctic fauna are unlikely, up to 79 % of benthic species may experience negative effects including range constriction and habitat loss (Griffiths et al. 2017). Species will therefore have to be able to acclimatise to temperatures up to 3.0 °C warmer than current averages if they are to remain in their current habitats. More thermally-tolerant species may also need to shift their distribution, for example to follow changes in distributions of prey, find conspecifics for reproduction or to remain in areas where their historically associated assemblages persist

(Gutt 2013).

Odontaster validus, the Antarctic cushion sea star, is a dominant scavenger/predator in

Antarctic benthic communities, often associated with prey species such as Sterechinus neumayeri, Doris kerguelensis and Laternula elliptica (McClintock et al. 1988; Angulo-

Preckler et al. 2017; Moles et al. 2017). It has been modelled to have a trophic position of 2.0

134

– 2.5 within benthic food webs, and may one of the most important omnivores in coastal

Antarctic ecosystems, with a trophic level of 2.91 in McMurdo Sound (McClintock 1994;

Gillies et al. 2012; Wing et al. 2012). The suggestion that Antarctic marine species are comparatively vulnerable to rapid environmental change due to their adaptations to a cold, stable climate (e.g. by Clarke 1980; McClintock et al. 2009; Melzner et al. 2009; Somero

2010; Gonzalez-Bernat et al. 2013) is not applicable to all polar species, and is being challenged in terms of species’ temperature-compensated metabolism, wide thermal development windows and robust larval development (Marsh et al. 2001; Gonzalez-Bernat et al. 2013, Karelitz et al. 2017). The majority of Antarctic benthic animals can tolerate temperature ranges from -1.8 – 4.8 °C (Griffiths et al. 2017), but O. validus may be one of the more thermo-tolerant benthic species, as long-term adult acclimation up to 6.0 °C is possible and CTmax for adults has been estimated at 23.7 °C (Peck and Prothero-Thomas 2002; Peck et al. 2008; Morley et al. 2016).

O.validus has been called a ‘keystone’ species in Antarctic marine environments (McClintock

1988) for its role in regulating populations of other predators such as Acodontaster conspicuous, by feeding on its’ benthic larvae (Dayton et al. 1974). It has been suggested that

O. validus can potentially live for up to 100 years (Pearse 1965), although this value is an estimate and should not be taken as fact without criticism. The behavioural interactions of O. validus with sympatric sea star predators such as Labidiaster annulatus, Diptasterias brandti and Perknaster aurorae revealed that this species possesses a complex behavioural ‘alarm’ system enabling community flight responses to nearby predators, but that no equivalent chemo-sensory system exists (McClintock et al. 2008). This indicates that O. validus has some ‘community’-level behaviours which fosters survival of populations as well as

135 individuals, even though they can be solitary animals which seek to distance themselves from other individuals of the same species (Palumbi and Freed 1988).

O.validus populations reproduce once per year, generally in the Austral winter although individual gametogenic cycle in this species lasts around two years (Pearse et al. 1986). They are broadcast spawners, releasing gametes into the ocean to be fertilised externally, and have very high levels of storage energy in oocytes in order to help them cope with cold water and long development (Pearse 1965; Shilling and Manahan 1994). Once fertilised, eggs take a relatively long time to develop, needing almost 5 days to develop to blastula stage, and not hatching until 8-9 days post-fertilisation (Pearse 1969). Although originally thought to be demersal, it is now known that larvae are pelagic, spending around 23 weeks in the plankton before metamorphosis and settlement (Pearse et al. 1986). During their time in the plankton, it is thought that O.validus larvae are planktotrophic or bacteriotrophic (Chiantore et al.

2002) but also potentially utilise dissolved organic matter (DOM) as a food source (Shillings and Manahan 1994). The winter spawning period, coupled with the long pelagic development stage, has been posited as an adaptation allowing feeding larvae to take advantage of the spring phytoplankton bloom (Pearse 1965).

For Odontaster, survival at elevated temperature is probably linked to a combination of utilisation of stored energy, low maintenance costs and the ability to compensate for physiological stress (Agüera et al. 2015). Elevated temperature is known to increase metabolic rate in polar marine invertebrates, and metabolic compensation for survival at high temperature has been observed in O. validus by an increase in respiration and righting time with temperature increase (Peck et al. 2008). Due to the patchy availability of food in the

136

Antarctic, long-term metabolic elevation in order to cope with ocean warming and acidification may therefore present a physiological challenge to this species. Odontaster validus appears to possess comparatively low levels of stored energy within pyloric caeca compared to other sea stars. For example, the pyloric index in O. validus has been estimated as 41 % body dry weight and containing 15-30 % lipids (24 mg g-1 dry weight-1)

(McClintock et al. 1988). Other sea stars for example Stichaster striatus have pyloric lipid content estimated 20-30 mg per arm (Diaz-Guisado et al. 2006) and Asteria rubens has pyloric indices of 20-48 % body dry weight and pyloric lipid content up to 75 mg g-1 dry weight-1 (Oudejans et al. 1979). The lower stored energy is consistent with theories of low food availability, energetic investment to gametes and slow metabolic rates in Antarctic marine animals (Clarke 1980; Raymond et al. 2007). Even if O. validus is able to alter its metabolic rate to cope with ocean warming, a change in feeding behaviour must also happen as O. validus may not have substantial energetic reserves to cope with short-term metabolic disruption and enable long-term survival.

This study aimed to quantify the energetic costs of exposure to ocean acidification and warming to O. validus. I hypothesised that adults would possess the capacity to acclimate to combinations of elevated temperature and low pH in terms of metabolic energy budget. I further hypothesised that the underlying mechanism for maintenance of homeostasis would be increased food intake to compensate for a reduction in stored energy at high temperature and low pH.

137

5.3 Materials and Methods

a) Animal collection and holding

Adult Odontaster validus were collected from benthic sites around McMurdo Sound,

Antarctica by SCUBA in November 2014. After collection, adults were air transported to

New Zealand in 60 L refrigerated containers with fresh filtered seawater. Mean seawater pH(T) in McMurdo Sound at the time of collection was 8.04 (± 0.06) and mean water temperature was -1.2 °C (±0.1 °C). Animals were imported to New Zealand under MPI Permit

#2014054106 and held in PC2 conditions at Portobello Marine Laboratory for the duration of the experiment.

b) Experimental design and manipulations of temperature and pH

Once in New Zealand, 200 adult sea stars were held in ambient conditions for a period of nine months. During this time temperature in seawater aquaria was at 0.5 °C ±0.04 and pH(T) was kept within a range of 8.08 – 8.19. The animals were fed a diet of blue cod (Parapercis colias) muscle tissue once every 3 weeks. In August 2015, 22 animals were induced to spawn for the experiment described in Chapter 7. The remaining 178 animals were randomly assigned to one of six experimental treatments, each a specific combination of pH and temperature chosen based on IPCC emission scenario RCP8.5 (IPCC 2013). Temperature treatment levels were either kept ‘ambient’ at 0.5 °C ±0.02 or elevated to 3.5 °C ±0.2. The pH treatment levels were ambient at 8.15 ±0.04 or lowered to 7.70 ±0.03. Introduction to elevated temperature and low pH was incremental as per the recommendations of Suckling et al.

(2014); aquarium temperature was raised at a rate of 0.3 °C per week, and pH was lowered at the rate of 0.03 units every three days after the aquaria had reached experimental temperature, giving a total acclimation time of twelve weeks.

138

Aquarium design is shown in Figure 5.1. Fresh filtered seawater at the appropriate temperature was pumped into the sump tank, where pure CO2 was bubbled in. Water at the appropriate temperature and CO2 levels was then pumped out of the sump tank and into a header tank that controlled the water supply to three aquaria in which adult O. validus were kept at a density of 0.8 individuals/ L3 with a total of 8 individuals per 12 L tank. Each system comprised a CO2 mixing tank (sump), a header tank and three treatment tanks and held a total of 120 L of seawater at any time. Water flowed into the animal tanks at a maximum rate of 3

L per minute. A 50% was made weekly (i.e. 60 L week-1) with fresh filtered seawater at the appropriate treatment temperature. Seawater was removed and replaced in the mixing (sump) tank so that experimental pH conditions were equalised before water was allowed to flow into the header or animal tanks (Figure 5.2). This system did not have the ability to mimic the light/ dark cycles in the Ross Sea, as the light levels in the aquaria were left to vary seasonally with those of Dunedin. The consequences of this are addressed in the discussion section 5.6b.

This system design is in category B-4 of the classifications suggested by Cornwall and Hurd

(2016) as one header tank supplied three pseudo-replicated tanks for each treatment, but true replication in this experimental design resulted from each system being replicated more than once in an independent environment (different fridges). In order to make the experimental design ‘appropriate’ for ocean acidification research, a nested approach to statistical analysis was adopted (see Data Analysis section below) which therefore allows the experimental design to be reclassified as an appropriate A-type design. Temperature and pH over the acclimation period are given in Appendix 10.4.

139

Figure 5.1: Aquarium design for long-term acclimation of adult O. validus to elevated temperature and low pH. (a) experimental design; (b) photo of working aquarium, photo by Miles Lamare. This system was replicated twice for each of four treatments giving a total of eight aquaria. This system was used for adult and F1 larval O. validus experiments detailed here and in chapter 7.

140

Figure 5.2: Mean weekly temperature (a) and pH (b) in aquaria over the 47-week acclimation

period. Error bars represent S.E. Red dashed rectangle indicates spawning period of Odontaster

validus. Data points are averages of (a) 1344 data points or (b) 84 data points.

After the acclimation period the animals were kept at the treatments conditions for a further nine months giving a total exposure time of twelve months. During this time, a subset of eight animals were destructively sampled every 3 months and data collected on a wide range of physiological parameters. At the end of the acclimation period in August 2015 the remaining animals were induced to spawn and the embryos used in the experiment detailed in Chapter 7.

After spawning, the remaining living adults, both reproductively mature and spent, were sacrificed, the same physiological parameters measured, and the experiment stopped. Details of the data collected on adult sea stars every 3 months follow the same protocol as the experiment detailed in chapter 4 and are as follows;

141

c) Body size and tissue weight relationships

These components were then dried in a furnace at 60 °C overnight and dry-weighed once cool. Gonad and pyloric indices were calculated as % dry body weight for animals from each treatment at each time point. Arm length was also tracked monthly in each treatment to assess growth rate. Body wall and skeletal components were ashed in a muffle furnace (450 °C for 6 hours) to determine inorganic content.

d) Respiration rate and ammonia excretion

Respiration rate was measured in a sub-set (n = 8) of animals from each treatment at 3-month intervals throughout the acclimation process, including pre-acclimation and after 9 months.

Animals were placed in sealed 300 ml chambers of seawater from their respective treatments and the oxygen depletion rate was measured fibre-optically using a PreSens FibOx 3 sensor system, with temperature-compensated oxygen measurements taken using fluorescence from a PreSens Oxygen Sensor Spot (SP-PSt3-NAU) (FibOx sensor, oxygen meter and fluorescent spots PreSens Precision Sensing GmbH, Regensberg, Germany). Each sealed chamber was placed in a water bath at the appropriate experimental temperature for each treatment and oxygen depletion measured over a 6-hour period, or the time taken for oxygen saturation level to drop to a minimum of 80 %, whichever was the shorter time period. Animals were wet-weighed before and after the experiment. After the experiment, animals were killed and dry-weighed to normalise respiration rate to body dry weight. Vertical stratification and changes to microbial respiration rate were minimised by the use of mechanical stirrers

142 whenever the chambers were sealed and the experiments in progress. During both respiration and excretion experiments, background oxygen depletion was measured using a blank chamber for each run filled with seawater at the same carbon dioxide and temperature conditions as the animal chambers.

Water samples from each respiration chamber were analysed for ammonia (NH3) content using a continuous-flow injection analysis Auto-Analyzer (LaChat QuikChem 8500) at

Portobello Marine Laboratory. Analytical protocol followed the Berthelot reaction to measure ammonia concentration in seawater relative to indophenol blue absorbance at 630 nm

(QuikChem Method 31-107-06-1-B). The system was calibrated using deionized water and spectral analysis was performed using in-house software.

e) Food absorption efficiency

Animals were fed on a mixed diet throughout their acclimation time. The diet included blue cod (Parapercis colias) muscle tissue, urchins (Pseudechinus novaezelandiae), red cod

(Pseudophycis bachus) muscle tissue and other small invertebrates. During the three-monthly feeding measurements, however, the food source was kept the same; blue cod muscle. Pre- starved (21 d) animals were given a pre-weighed sample of fresh blue cod and allowed to eat for three hours or until satiation, whichever was the shorter time period. The food was then weighed and the eaten mass calculated. Energetic content of blue cod muscle tissue was determined experimentally by combustion and checked against commercial estimates.

Feeding experiments took place in the animals’ acclimation treatments. Animals were wet- weighed before the feeding experiment to normalise feeding rate to body size. After each

143 feeding experiment, faeces were collected over a 36-hour period, weighed and dried to determine organic content using the following equation:

퐴 = 퐼 − 퐹

Where A is the absorption efficiency, I is the energy ingested in food and F is the energy loss as solid faeces.

f) Righting time

‘Righting time,’ the time taken for an animal inverted on its’ aboral (dorsal) surface to right itself completely and replace all arms onto a flat surface, was measured in a subset of fed animals from each treatment every three months. Righting time was recorded by video and analysed. Each animal was inverted a maximum of ten times with a break of no more than one minute in between each turn. Righting time was normalised to arm length for analysis.

5.4 Data analysis

Analysis of the data followed the same protocol as described in Chapter 4. Tissue weights

(wet, dry and ash-free) were analysed by linear regression to describe the relationship between the indices. Gonad weight and pyloric weight were expressed as a percentage of dry body weight for their respective indices. Respiration rate was calculated as oxygen

-1 -1 -1 consumption µg O2 h g dry weight . Ingestion was calculated as grams food eaten per gram body weight per hour, and absorption efficiency was the % of ingested energy that remained in the body. Food absorption efficiency estimates were derived from Dayton et al.

(1974). Excretion was calculated as µg NH3 per litre per hour per gram body weight.

Measurements of oxygen consumption and nitrogen excretion were used to calculate the O:N

144 ratio at each time point (Conover and Corner 1968). Respiration, feeding and excretion data were converted into Joules and used to calculate Scope for Growth following a modified

Winberg equation (Winberg 1960), which states that

푃 = 퐴 − (푅 + 푈)

where P is the Production, otherwise referred to as Scope for Growth, and represents the difference between the energy absorbed from food and energy expenditure for respiration and feeding. A is the product of consumption, or the Absorption Efficiency of ingested food, R is the Respiration converted to Joules using the oxyenthalpic ratio, and U is the energy loss as excretion.

Data were tested for significant differences using ANOVA and Holm-Sidak post-hoc tests.

The ‘wild’ population as well as each temperature x pH treatment at each time point were analysed against the laboratory control population; treatments which were not the control were not tested against the wild population due to the expected differences between a closed captive population and an open source population. The ‘wild’ population was included in analyses only to verify the validity of the laboratory control as a simulation of the condition of the wild population at that time point. This analysis assumed that the effects of captivity on overall health would be negligible compared to the differences between temperature and pH treatments.

145

5.5 Results

a) Body weight indices

After 9 months there were no significant differences in the wet, dry or ash-free weights of animals acclimated among treatments (ANOVA F = 12.25 df= 66 p<0.05). On average, animal dry weight was 2 – 5 % of wet weight, and ash-free weight was 20 – 80 % of dry weight. Mean animal wet weight in the control treatments was 10.45 g (+/- 0.44 S.E.) at the start of the acclimation period (Dec) and 9.81 g (+/- 0.38 S.E.) in peak reproductive season

(Aug) pre-spawning (see Appendix 10.4a for body weight values).

After 9 months, acclimation to elevated temperature significantly reduced gonad-somatic index (ANOVA F = 10.66 df = 25 p = 0.047) by 1.1 % compared to the control treatment at pH 8.1 (Holm-Sidak t = 3.26 p = 0.047) and 1.9 % at pH 7.7 (Holm-Sidak t = 2.39 p =

0.025). Acclimation to low pH at ambient temperature for 9 months reduced gonad-somatic index by 2 % at ambient temperature (Holm-Sidak t = 2.366 p=0.019) versus the control treatment (Fig. 5.3a).

146

Acclimation to elevated temperature significantly increased pyloric index after 3 and 6 months (ANOVA F= 3.88 df = 25 and F = 3.78 df= 25 respectively, p<0.05) by on average

13 % at pH 8.1 versus the control treatment. At ambient temperature, pyloric index was highest in June, but at elevated temperature it was highest in March. In all treatments, pyloric index was lowest in December and August. After 9 months acclimation, however, there were no differences in pyloric index between any of the groups. There was no effect of acclimation to low pH on pyloric index at either temperature level sampled and no interaction between the two factors (Fig. 5.3b).

Figure 5.3: Gonad-somatic (A) and pyloric (B) indices for adult Odontaster validus acclimated

for 9 months to combinations of two temperature (ambient = 0.5 °C; elevated = 3.5 °C) and two

pH (ambient = 8.15; low = pH 7.7) levels. Bars of different shades represent sampling time

points. Bars are mean values (n=8) and error bars represent 1 S.E. Both indices are expressed as

a percentage of body dry weight. Bars with different letters above are significantly different (p<0.05).

147

b) Respiration rate

Respiration rate in the control population at the beginning of the acclimation period was 0.29

-1 -1 µg O2 g dry weight hr ±0.10 (SE) and increased to 0.51 µg O2 g dry weight hr ±0.006 (SE) after 9 months (Fig. 5.4). Exposure to elevated temperature for 9 months significantly increased mean respiration rate (ANOVA F = 25.71 df= 27 p <0.001) to 1.11 µg O2 g dry weight hr-1 at pH 8.1 (Holm-Sidak t = 8.47 p <0.001) but exposure to low pH at elevated

-1 temperature decreased mean respiration rate to 0.10 µg O2 g dry weight hr (Holm-Sidak t =

8.18 p = 0.01). At low temperature, regardless of pH level there were no significant differences between respiration rate at the start and end of the exposure period (p<0.05). In all treatments apart from the [elevated temperature + low pH] environment, respiration rate was highest in August (Fig. 5.4). In the [elevated temperature + low pH] treatment, respiration rate was lowest in August.

c) Absorption and excretion of energy

Mean feeding rate of animals was 0.15 mg eaten g wet weight-1 hr-1 at the start of the acclimation period. Low pH exposure did not affect feeding rate, however exposure to elevated temperature significantly (ANOA F = 9.62 df= 24 p =0.44) increased feeding rate to

0.47 mg eaten g wet weight-1 hr-1. Feeding rates are given in Appendix 10.4.

Mean energy absorbed at the beginning of the acclimation period (Dec) was 0.38 mJ g dry weight-1 hr-1 and in all treatments, there were no significant differences between energy absorption at the beginning and end of the experiment (Fig. 5.5). However, after 3 months exposure, energy absorption in the elevated temperature treatments, regardless of pH level, was on average 0.35 mJ g dry weight-1 hr-1 higher than in the ambient temperature treatments

148

(ANOVA F= 22.25 df= 27 p = 0.041). Mean absorption efficiency was 65.5 % (+/- 0.44 SE) throughout the experiment.

Figure 5.4: Respiration rate of adult Odontaster validus acclimated for 9 months to

combinations of two temperature (ambient = 0.5 °C; elevated = 3.5 °C) and two pH

(ambient = 8.15; low = pH 7.7) levels. Bars of different shades represent sampling time

points. Bars are mean values (n=8) and error bars represent 1 S.E. Both indices are

expressed as a percentage of body dry weight. Bars with different letters above are

significantly different (p<0.05).

149

Energy content in solid faeces was 5.50 J g-1 in the control treatment at the start of the experiment and was not affected by acclimation time, temperature or pH treatment. Mean absorption efficiency in this experiment was 74.4 % and was not affected by either temperature (ANOVA F = 12.26 df= 27 p = 0.03) or pH (ANOVA F= 4.78 df = 27 p = 0.04).

Exposure to low pH significantly increased nitrogenous excretion from 6.09 µg NH3 g dry

-1 -1 -1 -1 weight h at the start on the exposure period to 16.11 µg NH3 g dry weight h after 9

-1 -1 months at ambient temperature and 18.65 µg NH3 g dry weight h at elevated temperature

(Fig. 5.6). O:N ratio was significantly decreased by exposure to low pH at ambient temperature for the first 6 months of the acclimation period, but by the end point of the experiment (August) there were no differences between ambient and low pH. Elevated temperature decreased O:N ratio vs. the control treatment (ANOVA F = 12.58 p = 0.05) but there were no interactive effects between the two stressors (Fig. 5.7).

150

Figure 5.5: Absorption of energy from food in adult Odontaster validus acclimated over 9 months to combinations of two temperature (ambient = 0.5 °C; elevated = 3.5 °C) and two pH (ambient =

8.15; low = pH 7.7) levels. Bars of different shades represent sampling time points. Bars are mean values (n=8) and error bars represent 1 S.E. Both indices are expressed as a percentage of body dry weight. Bars with different letters above are significantly different (p<0.05).

151

Figure 5.6: Nitrogenous excretion in adult Odontaster validus acclimated over 9 months to combinations of two temperature (ambient = 0.5 °C; elevated = 3.5 °C) and two pH (ambient =

8.15; low = pH 7.7) levels. Bars of different colour represent sampling time points. Bars are mean values (n=8) and error bars represent 1 S.E. Both indices are expressed as a percentage of body dry weight. Bars with different letters above are significantly different (p<0.05).

152

Figure 5.7: Oxygen: Nitrogen (O:N) ratio in adult Odontaster validus acclimated over 9 months to

combinations of two temperature (ambient = 0.5 °C; elevated = 3.5 °C) and two pH (ambient =

8.15; low = pH 7.7) levels. Bars of different colour represent sampling time points. Bars are mean

values (n=8) and error bars represent 1 S.E. Bars with different letters above are significantly

different (p<0.05).

d) Scope for Growth

Scope for Growth (SFG) was 19.60 J g-1 h-1 in the control treatment at the end of the experiment and was significantly increased to 27.81 J g-1 h-1 by exposure to elevated temperature at pH 8.1 (ANOVA F = 7.314 df = 99 p <0.001) (Fig. 5.8). There was no

153

significant (ANOVA F = 3.24 df= 99 p =0.55) effect of exposure to low pH on SfG nor was

there any interaction between the two factors.

Figure 5.8 Scope For Growth (SFG) in adult Odontaster validus after 9 months acclimation to combinations of two temperature (ambient = 0.5 °C; elevated = 3.5 °C) and two pH (ambient =

8.15; low = pH 7.7) levels. Bars are mean values (n=8) and error bars represent 1 S.E. Bars with different letters above are significantly different (p<0.05).

154

e) Righting behaviour

Righting time at the start of the acclimation period was 437.25 s and was significantly affected by exposure time and temperature (Fig. 5.9). After 3 months acclimation (March), righting time was significantly (ANOA F = 7.42 df = 27 p =0.01) lowered by exposure to elevated temperature (Holm-Sidak t = 4.94 p = 0.017) and low pH (Holm-Sidak t = 2.86 p =

0.010). From 6 months exposure (Jun) onward to the end of the experiment (August), elevated temperature significantly (ANOVA F = 4.04 decreased righting time in all treatments including the control (Holm-Sidak t = 4.58 p = 0.025) but there was no effect (ANOVA F =

0.32 df = 14 p =0.81) of low pH.

Figure 5.9 Righting time of adult Odontaster validus acclimated over 9 months to combinations of

two temperature (ambient = 0.5 °C; elevated = 3.5 °C) and two pH (ambient = 8.15; low = pH 7.7)

levels. Bars of different shades represent sampling time points. Bars are mean values (n=8) and

error bars represent 1 S.E. Bars with different letters above are significantly different (p<0.05).

155

5.6 Discussion

In this experiment, I hypothesised that adult Odontaster validus would be capable of long- term physiological acclimation to combinations of elevated temperature and low pH after 9 months. In order to investigate this hypothesis, I exposed O. validus to combinations of +3.0

°C temperature increase and a -0.4 pH unit reduction. Exposure to a combination of low pH and elevated temperature did not affect metabolic energy budget, and this was possibly due to an increase in feeding rate as well as reduced gonad size.

a) Physiological acclimation to ocean acidification and warming

In this study, Odontaster validus appears capable of physiological acclimation to ocean acidification and warming. This conclusion is supported by other literature suggesting that this species is thermotolerant up to temperatures far higher than would usually be experienced in present-day conditions (i.e. Peck 2002, 2004, 2008). The scaling of metabolism with temperature increase is seen in the increased respiration rates at elevated temperature in this study. Our respiration rates at 0.5 °C are directly comparable to other published studies showing the relationship between temperature and respiration rate in this species (Belman and Giese 1974; Agüera et al. 2014) and other polar marine echinoderms

(Marsh and Manahan 1999; Brockington and Peck 2001; Watson et al. 2014).

The steady values of energy absorption throughout the experiment at high temperature reflect the increase in feeding rate at higher temperature, as there was no difference in the assimilation efficiency values between the treatments. The consistently high (>8.0) O:N ratios also indicate protein as the primary metabolite, reflecting the high protein diet in this

156 study and indicating that stored energy reserves were not being used as a supplementary source of dietary energy. It has also been shown that O. validus can select food items of higher calorific value during times of starvation (Kidawa 2009), meaning that feeding behaviour modification at high temperature could favour high-energy food consumed in greater quantities in order to maintain homeostasis. An increase in feeding rate with temperature, is seen in some, but not all, polar marine invertebrates. The urchin Sterechinus neumayeri fed at a faster rate at lower temperatures during a long-term exposure (Morley et al. 2016) and the amphipod Gondogeneiea antarctica lost both appetite and feeding preference with a 2.0 °C water temperature increase (Schram et al. 2015). The opposite trend was observed for the Antarctic limpet Nacella concinna which increased feeding rate with temperature from -0.6 – 12.3 °C (Morley et al. 2014). As both S. neumayeri and G. antarctica are grazers, and O. validus and N. concinna occupy higher trophic positions, there could be a correlation with food source or digestive function and feeding performance at warmer temperatures.

Behavioural modification, a measure of overall physiological performance, at elevated temperature in this study was also seen in the righting time, which is often acknowledged as a suitable parameter to measure animal physiological state (Kidawa et al. 2010). The exact upper thermal limits to righting behaviour are uncertain as it has been reported that O. validus loses the ability to right itself around either 4-5 °C (Kidawa et al. 2010 or >8 °C (Peck et al.

2008). The results of this experiment suggest that a temperature increase of +3.0 °C is within the temperature tolerance range of this population, and agrees with both of the above studies that animals move faster at high temperatures. Scope for Growth indicates the ‘leftover’ energy available for growth and movement, and therefore the faster turning times were as

157 expected, given the higher SfG values seen at elevated temperature/ ambient pH in this experiment.

The values for nitrogenous excretion in this study are within the range reported for other

+ h-1 Antarctic marine invertebrates such as the sea urchin S. neumayeri (162-415 nmol NH4

Brockington and Clarke 2001; Brockington and Peck 2001), the limpet Nacella concinna (66

+ -1 - - 170 nmol NH4 h Fraser et al. 2002) and the amphipod Waldeckia obesa (5 – 45 µg NH3 h

1 Chapelle et al. 1994). Excretory responses to low pH in marine invertebrates can vary from decreasing, as in the bivalve Limopsis marionensis (Pörtner et al. 1999); increasing as in S. neumayeri (Suckling et al. 2014) or no change as in the gastropod Crepidula fornicata and the coral Desmophyllum dianthus (Noisette et al. 2015; Gori et al. 2016). High excretion is sometimes explained as a compensatory mechanism for high intracellular bicarbonate ion accumulation at low pH (Stumpp and Yu 2017), whereby proton excretion at low extracellular pH is linked to bicarbonate foundation, although the exact mechanism of this process is unclear.

Another study of long-term acclimation of the polar urchin S. neumayeri estimated that physiological acclimation to low pH took 6 months (Suckling et al. 2015). The results of this study suggest that O. validus is capable of physiological acclimation within 3 months. After 3 months exposure to low pH at the ambient temperature, O. validus did not show any significant effect of low pH on gonad-somatic index, pyloric index, respiration rate, food absorption rate or efficiency, righting time or O:N ratio, and this trend continued throughout the exposure period. The only parameters measured which differed significantly from the control in the ambient temperature/ low pH treatment after 3 months were nitrogenous

158 excretion, which had increased, and gonad-somatic index, which had decreased. Acclimation to low pH at high temperature also seemed likely after 3 months, but many parameters showed some significant differences at the longer exposure time-points. Such parameters include pyloric index, which was comparable to the control treatment for the first 6 months but then became significantly lower, and the gonad-somatic index, which was lower than the control in all other treatments after 9 months despite being comparable for the first 6 months.

This result suggests that time taken to acclimate in long-term experiments depends on the parameter being measured as well as the treatment levels.

b) Energy allocation to reproductive output at high temperature and low pH

There is little evidence for energy resource re-allocation at elevated temperature and low pH in this study. Resource re-allocation has been suggested as a way for animals to maintain homeostasis during physiological stress, and often results in reduced gonad size due to lack of stored energy reserves. This has been observed in several temperate echinoderm species such as Strongylocentrotus droebachiensis (Stumpp et al. 2011). However, Antarctic echinoderms do not appear to do this, as both O. validus and S. neumayeri are able to acclimate their reproductive physiology to combined high temperature and low pH, possibly due to their longer gametogenic cycles (Suckling et al. 2015). Sterechinus neumayeri was able to maintain egg size and energetic investment to offspring under long-term experimental acidification, therefore keeping offspring developmental performance and survival high

(Llodra 2002; Suckling 2014). Odontaster validus can allocate a higher proportion of energetic reserves to growth and maintenance than other echinoderms, as well as having a constant, high energetic input to gametogenesis (Aguera et al. 2015) which negates the need for drastic resource reallocation.

159

The initiation and timing of gametogenesis in Odontaster validus are under photoperiodic control, and gametogenesis is stimulated when a photoperiod over 12 hours is sustained over a number of days (Pearse and Bosch 2002). In this study, photoperiod was not able to be controlled for, due to limitations of the laboratory set-up. The differences in daylight hours between Antarctica and Dunedin are evident throughout the summer and winter when

Antarctica experiences 24-hour light and dark cycles, respectively. Gametogenesis in the captive population may therefore have become out-of-phase with the natural population during those seasons. The impact of this on the findings of this study may have been to make the population develop ripe gonads earlier or later in the captive population than they would in the wild. The captive population did, however, undergo a normal gametogenic cycle as defined by McClintock (1988) and therefore the findings of this study are still relatable to wind populations. Bearing in mind that some differences between the captive and wild population were not able to be quantified as much in this experiment as they were in the

Patiriella regularis acclimation experiment in chapter five, the results of this experiment may be less comparable to those results because of the lack of photoperiod control.

Sustained reproductive capacity at elevated temperature and low pH indicates that this species may thrive in future conditions. Temperature, but not pH, appears to be the dominant environmental influence on reproductive capacity in this study. It has been estimated that

Odontaster can live for up to 100 years (Pearse 1969; Kidawa and Janecki 2011) and therefore will produce new cohorts of offspring over a period encompassing major change but within one life-cycle.

160

Offspring survival is partly dependant on egg lipid content, which in turn is dependent on parent nutrition (Bayne et al. 1978; Bertram and Strathmann 1998; Moore and Manahan

2007). As both gonad-somatic index and energy absorption were not affected by acidification or warming in this experiment, maternal lipid investment to offspring and therefore pre- feeding embryo survival prospects should remain stable into the future. One major caveat to this statement is that this study could not attempt to match the feeding behaviour of wild O. validus, as it is largely unknown how often they feed. Generally thought of as an opportunistic scavenger and predator, O. validus may experience seasonal depressions in food availability and quality. The feeding regime in this study was intended to enable as many O. validus as possible to survive the duration of the experiment, which is clearly necessary for practical reasons. It is not, however, an accurate representation of their diet composition or feeding schedule in the wild, and this fact may lessen the strength of any conlusions reached in this chapter. The offspring of female O. validus over successive years should not differ in their provisioning due to reduced parent SfG and reproductive investment. Odontaster validus embryos and larvae are tolerant to warming up to +3 °C

(Stanwell-Smith and Peck 1998) and reduced pH 7.8 (Gonzalez et al. 2012). If anything, ocean warming may contribute to increased egg provisioning by females and therefore improve the survival prospects of future cohorts.

c) Conclusions

The hypothesis that adult Odontaster validus would be capable of long-term physiological acclimation to combined ocean acidification and warming was supported. End-of-century seawater conditions in the Southern Ocean may not represent a significant challenge to the metabolic energy budget or reproductive capacity of this species. The higher maintenance costs associated with increased aerobic metabolism at elevated temperature are able to be

161 compensated for by an increase in feeding and therefore the ingested energy. Although this study has limitations, particularly around the control of photoperiod and experimental design,

I do not believe that they reduce the accuracy of the data analysis nor the strength of the conclusion. The results of this study support others which have theorised that long-term acclimation to elevated temperature is possible for O. validus. Antarctic species with long generation times will need this type of acclimation potential in order to survive, and O. validus may be one of the ‘winners’ of climate change.

162

Chapter Six: Maladaptive trans-generational effects of exposure

to low pH are mitigated by both genetic and non-genetic factors

for sea stars Patiriella regularis at elevated temperature.

20 µm

Photo: Patiriella regularis gastrula with scale bar. Photo credit: K. Sparks

163

This chapter is currently in preparation as: Sparks KM and Lamare MD ‘Maladaptive trans- generational effects of exposure to low pH are mitigated by both genetic and non-genetic factors for sea stars Patiriella regularis at elevated temperature.’ Target publication for this chapter: Evolution (2016 impact factor 4.20)

This chapter pertains to:

Hypothesis 3: Trans-generational responses to ocean acidification and warming

The phenotype of F1 offspring will depend on the temperature and pH environment to which

the parents are acclimated, thereby increasing fitness in the same environment.

Significance: This is a novel, understudied and increasingly important area of research into adaptive

capacity of marine populations. Trans-generational plasticity is an important survival and adaptive mechanism for marine populations and this area of research is becoming more and more important in

the face of abrupt, rapid climate change.

Contributions to this chapter:

Kate Sparks designed the experiment, partly based off an earlier QG protocol designed by

Shawna Foo. Kate and Miles Lamare built the aquariua. Kate carried out the experiments, collected and analysed all the data. Kate wrote this chapter. Miles provided feedback and edits on an early draft of the manuscript. Kate was funded by the University of Otago and the

University of Otago Ocean Acidification Research Theme.

164

6.1 Abstract

Trans-generational plasticity (TGP) is one way in which marine biological populations could adapt to rapid ocean acidification and warming. I investigated the hypothesis that populations of the New Zealand cushion sea star Patiriella regularis would possess sufficient adaptive capacity towards combined ocean acidification and warming, and that this would have a heritable genetic component. Populations of wild P. regularis were exposed to combinations of two temperature (ambient; ambient +2.5 °C) and three pH(T) (ambient =8.15; 7.90; 7.70) levels for 12 months. F1 offspring of both wild and acclimated parents were raised to gastrulation in all combinations of the same treatments as their parents. The effect of parent exposure to elevated temperature on offspring development was negligable. Parent exposure to pH 7.70 for 12 months reduced offspring % normal development by ~ 13 % regardless of the offspring temperature or pH environment. Offspring exposure to pH 7.70 reduced % normal development by ~25 %. Among F1 offspring of parents at the end of the exposure period, 17.62 % of variation in % normal development between families was explained by dam (=maternal) identity, and strong maternal effects (m2 = 0.41) were detected. Broad-sense heritability was h2 = 0.06 at ambient pH and h2 = 0.16 at pH 7.70. The hypothesis was not supported, and maladaptive TGP in response to low pH and strong maternal effects indicate intense natural selection for this population in the near future.

6.2 Introduction

Both phenotypic plasticity (of an individual) and trans-generational plasticity (TGP) are important in the context of global climate change because in a rapidly changing environment, mean population phenotype often lags behind the ‘optimum’ phenotype for a specific niche

(Bürger and Lynch 1997, Lande 2009). In response to ocean acidification and warming, TGP

165 may aid marine populations in responding to rapid environmental change. The Southern

Ocean, for example, is predicted to experience an increase in mean surface temperature of up to 2.0 °C and mean surface pH reduction of up to 0.4 units units by the year 2100 (IPCC 2013 scenario RCP8.5). The adaptive potential of marine populations to ocean acidification and warming may depend on TGP (Evans et al. 2015), especially in the face of widespread biodiversity losses caused by the interaction between these two stressors (Byrne 2011).

Transgenerational plasticity (TGP) may help a population adapt to a rapidly changing environment when mutational evolution cannot keep up with the pace of natural selection, and is often seen as a precursor to adaptation (Pfennig et al. 2010, Crozier and Hutchings

2014, Jensen et al. 2014). Transgenerational plasticity (sometimes called Anticipatory

Parental Effects or APEs) is a form of non-genetic inheritance whereby transmission of information between generations occurs through non-DNA based trait alteration (Foo and

Byrne 2016) originating from either the environment or phenotype of the parent (Agrawal

2001; Gunderson et al. 2017), and is usually adaptive or, less commonly, maladaptive

(Ghalambor et al. 2007; Guillaume et al. 2015).

Commercial aquaculture programs which use wild broodstock to selectively breed populations with desirable characteristics indicate that wild populations of marine invertebrates such as Pacific oysters Crassostrea gigas (Kong et al. 2015), urchins

Strongylocentrotus intermedis (Liu et al. 2005) and tiger prawns Penaeus monodon (Sun et al. 2015) are able to modify their mean population phenotype within one or two generations in response to strong environmental selection pressure. This conclusion, when applied to ocean acidification and warming research, can help to identify vulnerable and resilient species. For example, the gastropod Nucella lapillus is able to counteract the negative effects

166 of long-term exposure to acidification and temperature to different degrees for each stressor

(Ruhl et al. 2017). This is not true of all species, as offspring of adult oysters Saccostrea glomerata exposed to combinations of a +4 °C temperature rise and a doubling of seawater pCO2 from 392 – 856 µatm performed worse in early developmental stages than offspring of parents exposed to one stressor alone (Parker et al. 2017). In some marine populations, warming and acidification can be considered ‘opposing evolutionary vectors’ across multiple generations (Gibbin et al. 2017) due to the complex, but often antagonistic, effect of exposure to multiple stressors.

Changes in fitness-correlated trait heritability are often not clearly directional under stressful conditions such as acidification and warming (Reusch 2014). However, TGP is often difficult to pinpoint in lab studies on marine organisms outside of commercial aquaculture programs, as it requires an experimental design in which the parent environment is stable on a time- scale long enough that reproductive effects, if any, can be identified (such as a long-term lab acclimation period or a stable vent/ non-vent mesocosm experiment). One such reciprocal transplant study in which embryos from adult Echinometra sp. from naturally-occurring high-

CO2 vent sites were transplanted to non-vent sites, and vice versa, concluded that the place of origin of the parents (i.e. vent or non-vent sites) did not influence the development of the offspring at 48 hr, leading the study to conclude that transgenerational plasticity in response to CO2 levels in these populations could not be proven (Lamare et al. 2016). This in situ experiment supported the results of laboratory experiments on the same species, which found no trans-generational plasticity in response to ocean acidification among offspring of parents from vent and non-vent sites (Uthicke et al. 2016). The gastropod mollusc Hexaplex trunculus, however, which naturally occurs at both high-CO2 and low-CO2 sites around Italy’s

Ischia vents, has shown evidence of altered population demography and increased variance in

167 reproductive success between individuals within a site, often seen as the precursor to trans- generational plasticity and evolution (Harvey et al. 2016). Genetic variance in fitness- correlated traits increases as environmental stress increases, indicating that pre-adaptive evolutionary responses to environmental change (e.g. TGP) occur in conjunction with increased adaptive capacity in wild populations (Rowiński and Rogell 2017).

The sea star Patiriella regularis is found all over New Zealand and is an invasive species in parts of Tasmania, Australia (Byrne et al. 2013). Fertilization and early embryonic life- history stages, often considered to be the most vulnerable to environmental changes, are robust to temperatures of up to 26 °C (Byrne et al. 2010), around 5 °C higher than mean upper coastal New Zealand water temperature limit over the period from 1967 – 2011

(Bowen et al. 2017) and pH changes of up to 0.4 units (Byrne et al. 2010, Karelitz et al.

2017). The wide thermal and capnic tolerance of this species has contributed to its success in colonising new areas, and forecasted ocean warming and acidification conditions may open up more possible new areas of colonisation. Range shifts in marine populations due to effect of ocean warming have already been observed to influence biodiversity, community composition and animal behaviour (Bates et al. 2014a, Vergés et al. 2014, Sunday et al. 2015,

Gaylord et al. 2015, Nagelkerken and Munday 2016). The potential for Patiriella to expand its range into new areas could be an important consideration for biodiversity and conservation.

I hypothesised that adult Patiriella regularis exposed to elevated temperature and low pH would have the adaptive capacity to modify the phenotype of their offspring in order to

168 increase survival and fitness in their own environment. I also hypothesised that a significant component of this trait variation would be heritable.

6.3 Methods

a) Animal collection and acclimation to increased temperature and low pH

In order to test the hypothesis, adult Patiriella regularis were collected from Otago Harbour in the south of New Zealand at the height of their reproductive maturity in January 2015. The source population for this study is near the southern limit of this species distribution, and is characterised by wide temperature and pH fluctuations. Otago Harbour has an annual temperature range of 5 – 20 °C (Greig et al. 1988) but temperature can vary by as much as 6

°C over one tidal range and a mean annual pCO2 range of 280 – 430 ppm (equivalent to pH

8.0 – 8.25) with daily pH variations of up to 0.15 units over a tidal cycle (Nelson 2016, Bates et al. 2014b).

Once collected, adult P. regularis were bred from using a quantitative genetic design in order to characterise the range of genotypes among the wild population. The adults were then randomly assigned to one of six temperature/ pH combination treatments and allowed to acclimate for one reproductive cycle (12 months) culminating in December 2015 when the adult population was sexually ripe. At this point, the remaining adults were bred from again using the same quantitative genetic breeding design as for the wild population, and the offspring (referred to as the F1 generation from here) were raised to the gastrula stage in

169 either the same treatment as their parents or a different treatment. Offspring of all adults were raised in all treatments.

b) Spawning and fertilization

The basic method for the spawning, fertilization and development of embryos to gastrula stage (48 hr post-fertilization) was kept the same for the wild P. regularis population in

January 2015 was the acclimated population 12 months later.

Adults were dissected and the gonads removed and soaked in 10-5 M 1-methyladenine to induce maturation for 60 min. Gonads from each individual were kept separate. At this point, egg density from each female was determined by removing 5x 100µl subsamples and counting egg number. Eggs from females were placed in 250 ml filtered seawater (FSW) from the treatment in which they would be raised and allowed to acclimate for a further 30 min before fertilization.

Egg density between females was relatively consistent, with each female producing on average 80,000 eggs. Undiluted sperm concentration individual males varied between 107 and

1010 sperm ml-1 but did not differ significantly between males acclimated to different treatments. Mean sperm: egg ratio at fertilisation was 300: 1 across all treatments. This ratio was decided upon based on data from the wild population fertilisation carried out in 2015.

The speed of development of P. regularis embryos in the F1 generation did not differ significantly from that of the wild population, therefore the duration of each experimental run or ‘block’ was consistent with previous experiments, allowing the embryos to be exposed to their treatments for the same amount of time.

170

Fertilization occurred in the treatments in which the embryos would be raised, following the

North Carolina II breeding design and protocols set out in Sparks et al. (2016) so that eggs from all females were fertilized by all males, and the offspring of each M x F pair were kept separate throughout the experiment so that parentage could be tracked. Fertilization was >80

% after 45 min and identified by the presence of a fertilization envelope. At this point embryos were rinsed and divided between three falcon tubes with cut-out mesh windows, details of which are given in Sparks et al. (2016). Once the embryos reached the gastrula stage, embryos were removed, fixed and scored for normal development following the developmental schedule set out in Byrne et al. 2013 for this species.

c) Experimental treatments and manipulations of temperature and pH

The aquarium temperature/ pH treatments were as follows; the control treatment, which was allowed to vary with the water temperature and pH in Otago Harbour; and then all combinations of a temperature treatment always 2.5 °C above the ambient water temperature, pH 7.90 and pH 7.70. All treatments were allowed to vary within the natural ranges of Otago

Harbour water over a tidal cycle. Temperature and pH variations over the 12-month adult acclimation period are given in chapter 4 Figure 4.1. Temperature and pH were adjusted gradually as per the recommendations of Suckling et al. (2014). These treatment levels were chosen in line with IPCC 2013 scenario RCP8.5 to represent a ‘worst-case’ scenario for forecasted end-of-century temperature and pH conditions.

The pH in the aquaria was recorded in the morning, noon and evening every day during the experimental period using the Tunze pH controllers and checked using m-cresol purple. Mean pH(T) in the control treatment was 8.17 + 0.028 (S.D.) and varied tidally and daily within a

171 range from 8.10 – 8.22. The adult P. regularis population had been acclimated for 12 months to their relative pH treatments, targeted at pH 7.9 and 7.7. Overall, mean pH(T) in the targeted

7.9 treatment (across both temperature levels) was 7.87 + 0.037 (S.D.). In the targeted 7.7 treatment (across both temperature levels), mean pH(T) was 7.69 + 0.022 (S.D.). Water chemistry parameters (DIC, Alkalinity, pCO2) for each treatment during the experimental period are given in Appendix 10.3. Temperature was recorded every 30 min using HOBO loggers. Within the +2.5 °C treatments, temperature was raised using Jager aquarium heaters and controlled to within +/- 0.3 °C of a target using Tunze Temperature controllers.

6.4 Data Analysis

Data collected were number of normal and abnormal individuals in a sample. These data were arcsine transformed then analysed using a nested multivariate ANOVA design in the

PERMANOVA+ extension of Primer 6 software (Clark and Gorley 2006) to determine the contribution of fixed and random factors to variation among genotypes. Temperature and pH were fixed factors, sire and dam identity were random and nested within each experimental run (“block”) and within the aquarium tank from which the parents were drawn. For analysis of the F1 generation, parent acclimation treatment was added as a random factor nested within tank. Holm-Sidak post-hoc test identified differences between treatment levels. Variance components for random factors were generated using the lmerTest package in R (which used a REML method to estimate variances) using a variation on the protocol described by

Kielland et al. (2017), and those variances used to calculate broad-sense heritability (h2) and maternal effects (m2).

Heritability and maternal effects were calculated following an application of the animal protocol set out in Kruuk (2004) and as used by Munday et al. (2017). The additive genetic

172 variance, nongenetic variance and residual error variances generated by the REML analysis were used in calculations derived from Freeman and Herron (2004). For each male and female genotype within the wild and acclimated populations, reaction norms were generated to examine the performance of each genotype across either the temperature or pH environmental gradient. A genotype was determined to be plastic if the slope of the reaction norm describing that genotype was not equal to 0. Transgenerational plasticity was identified as the presence of a heritable ‘anticipatory parental effect’ (APE) wherein the F0 generation altered the phenotype of their F1 offspring to alter fitness in the same environment to which they were acclimated.

Interpretation of reaction norms such as those in figures 6.2 – 6.4 should focus on the spread of data, not an ‘average.’ As each line represents the response of one animal, the spread of lines and changes to the rank order of genotypes in certain environments indicate both individual performance and, taken as a group, selection potential. Tightly-grouped lines on a graph indicate that each genotype is responding to environmental change in a similar manner i.e. very little genotypic variation within the studied population (e.g. Fig. 6.4e). Widely- spread lines indicate genotypic variation within a population e.g. Fig. 6.4m, and therefore the potential for the ‘best performing’ genotypes in each environment to be selected for.

Reaction norm panels (6.2 – 6.4) are presented in full in this thesis, to help understanding of the estimations of variance between genotype and environment which are presented in the chapter appendices e.g. Appendix 10.5a. Reaction norms are presented here without estimations of error, a choice reflecting the method of data collection and analysis. As discussed in section 2.3c, during scoring, embryos from all three falcon tubes were counted

173 and data pooled to represent a replicate of the treatment. If any one tube failed to contain embryos, data from the remaining two samples from that tube were pooled for the replicate.

This method of data collection means that the count data for each individual are pseudo- replicates of each other. It may be misleading, therefore, to present error bars on the reaction norms, as the three embryo counts are not independent of each other. The implications of this decision are discussed further in the general discussion section 8.1d.

6.5 Results

a) Transgenerational plasticity

Both parent (F0) and offspring (F1) environment contributed to developmental success (Fig.

6.1). Parent exposure to pH 7.70 for 12 months significantly reduced mean F1 normal development at pH 7.70 by on average 13 % at both ambient (Fig. 6.1d) and elevated (Fig.

6.1e) temperature (Holm-Sidak t = 2.85 p<0.05). Regardless of parent pH exposure, when F1 embryos were raised at pH 7.70 they experienced significantly lower (mean 25 %) developmental normality (Holm-Sidak t = 3.25 p<0.05) (Fig. 6.1c-e).

F1 exposure to pH 7.70 also reduced normal development by 37 % among offspring of adults acclimated to ambient temperature but did not have any effect on F1 offspring of adults acclimated to elevated temperature (Fig. 6.1c). Among the F1 offspring of adults kept in both the control treatment and the pH 7.90 treatment, there was no effect of temperature or pH on normal development (Fig. 6.1a-b).

174

b) Heritability and genetic contributions to tolerance

Among the wild population, dam identity explained 3.98 % of variation (pseudo-F = 4.47 p =

0.04), [dam x temperature] explained 1.84 % of variation (pseudo-F = 2.96 p = 0.01) and

[dam x pH] explained 1.60 % of variation among genetic lines (pseudo-F= 2.63 p = 0.05)

(Fig. 6.2). Dam identity among F1 families of acclimated parents explained 17.62 % of variation between genetic lines and the interaction between [dam x temperature] (Fig. 6.3) and [dam x pH] (Fig. 6.3) explained 4.51 % and 3.71 % of variation respectively. Strong

2 2 heritable maternal effects (m ) were detected among F1 (m =0.41) genotypes.

Among the wild population there was no influence of sire identity or [sire x environment] interactions. Sire identity explained 7.67 % of variation between genetic lines in the acclimated F1 population and the interaction between [sire x temperature] (Fig. 6.4) and [sire x pH] (Fig. 6.4) explained 2.11 % and 5.04 % of variation respectively. The interaction between sire and dam identity explained 2.34 % of variation among F1 genotypes (Fig. 6.5).

Broad-sense heritability for the ‘normal development’ trait was h2 = 0.10 at ambient pH and h2=0.24 at low pH for genotypes at the end of the acclimation period, and h2 = 0.06 at ambient pH and h2=0.16 at low pH for the wild population.

175

H H

on on

Patiriella

) environment is shown by the shaded by environment ) shown is

1 F

gastrulae is dependent on both parent and offspring pH environment pH in sea star both on and parent offspring dependent gastrulaeis

F <0.05) Error bars denote SE. n=75 for all.denote bars Error for n=75 SE. <0.05) p

) environment is given along the horizontal axis and offspring ( axisthe and offspring along horizontal given is environment )

0 F

Normaldevelopment of

Parent(

Figure 6.1 Figure regularis. effect 8.10; (B) exposure raised temperature at Effect pH offspring on parent of exposure temperature parent of (A) bars. effect p 7.70; parent of (D) raised 7.90; at effect (C) raised exposure pH offspring on at temperature pH offspring parent of raised case at letters exposure °C. 18.5 offspring on raised at Lower effect pH exposureparent of °C; 16.0 offspring on (E) significance of denote( levels

176

genotypes from a wild population from Otago Harbour, NZ NZ Harbour, Otago a from wild population from genotypes

Patiriella regularis Patiriellaregularis

Reaction female for norms

: :

Figure 6.2 Figure across a temperature gradient (A) and a pH gradient (B). Each line represents the response of one female genotype from one of female Each genotype response line gradient(B). the represents and a pH a across gradient temperature (A) reaction of norms. counts. n=3 explanation for 6.4 See section

177

M) are M)

e(G

4 for for 4

genotypes acclimated for 12 months to the treatment described along the top the of acclimated along described treatment months 12 to the genotypes for

F) are reaction norms across a temperature gradient and graphs along the linalong bottom andgraphs a are gradient temperature reaction across norms F)

Patiriella regularis Patiriellaregularis

eactionfemale for norms

: :

Figure 6.3 Figure (A along the columns. top line Graphs graph from counts. n=3 Seesection genotype 6. one of response female line gradient.Each a the pH reactionrepresents across norms explanation norms, reaction of

178

M) are M)

the top the of acclimatedalong described treatment months to the 12 for

genotypes genotypes

F) are reaction norms across a temperature gradient and graphs along the linealong bottom (G andgraphs a are gradient temperature reaction across norms F) - from counts. n=3 for See section 6.4 one of response male genotype the nerepresents

regularis Patiriella

Reactionmale for norms : :

graph columns. Graphs along the top line (A along the columns. top line Graphs graph li gradient.Each a pH reactionacross norms explanationnorms. reaction of 6.4 Figure

179

Figure 6.5 Normal development at gastrulation for Patiriella regularis gastrulae is dependent on

the interaction between sire and dam identity across 144 families. Sire x dam pair is given across

the horizontal axis and each data point is the mean from n=18 counts. Error bars represent S.E.

6.6 Discussion

This study used a quantitative genetic breeding design to examine the range of genotypes within the population of Patiriella regularis both from a wild population and populations acclimated for 12 months to combinations of elevated temperature (+2.5 °C) and low pH ( -

0.2 and -0.4 units). Parent exposure to pH 7.70 for 12 months reduced offspring normal

180 development across all treatments by 13 – 37 % depending on the environment in which the offspring were raised, but offspring exposure to elevated temperature erased some of these differences. For females and males acclimated to pH 7.70, there were no genotypes which performed better at pH 7.70 than in the control treatment (pH 8.10) despite the acclimation period. There is a heritable genetic component to the ‘normal development’ trait in both ambient and simulated near-future conditions. Maternal identity was a strong factor in determining which genotypes responded better to elevated temperature and low pH and strong anticipatory maternal effects were detected among F1 genetic lines. Both genetic and non-genetic factors contribute to the success of this population in adapting to rapid ocean warming and acidification, and that this population possesses the capacity to adapt to end-of- century conditions.

a) Maladaptive trans-generational plasticity

The maladaptive APEs of the type seen in this study are often seen when parents are exposed to a rapidly-changing, multi-stressor environment (Burgess and Marshall 2014; Grether 2014;

Bonduriansky and Crean 2017; Kazerouni et al. 2017; Miller et al. 2017; Parker et al. 2017).

Temperature and pH are antagonistic stressors in this study to the extent that larval exposure to elevated temperature compensates for the negative trans-generational effects of parent exposure to low pH. [Dam x pH] was a significant interaction among both the wild and acclimated populations, however [sire x pH] only explained a significant portion of variation among genotypes after the parent acclimation period. The conclusions of this study, that TGP is maladaptive at low pH for both male and female genotypes and therefore likely to induce intense selection pressure, are supported by other studies indicating that intense trait selection and environmental unpredictability can induce maladaptive TGP (Moran et al. 1994;

181

Bonduriansky et al. 2012; Crean et al. 2012). The larger h2 values estimated for the low pH environment at the end of the acclimation period also supports this conclusion.

The evolutionary consequences of maladaptive plasticity may be uncertain for this population, as maladaptive plastic responses are not the ‘evolutionary dead ends’ that maladaptive genetic responses can be (Langerhans and DeWitt 2002; Grether 2014). By definition, maladaptive TGP creates increased divergence between the expressed trait and the phenotypic optimum, leading to a large lag and further intense selection (Lande and Shannon

1996; Burger 1999; Rogers and Morris 2013; Kopp and Matuszewski 2014). Plastic compensation may eventually overcome any maladaptive TGP (Rogers and Morris 2013) and therefore minimise the phenotypic lag, possibly acting as the precursor to cryptic evolution

(Grether 2005). Processes such as allele transfer and canalization may influence the anticipatory capacity of future generations, as has been reported for F2 guppies (Poecilia reticulata) which showed some negative carry-over effects of parent thermal environment in the F1 generation (Morris and Rogers 2013; Le Roy et al. 2017).

b) Maternal effects

The results of this study do not support others which indicate that maternal effects such as provisioning and condition-transfer effects are not influenced by environmental unpredictability (Barbosa et al. 2015; Kuijper and Johnstone 2017), as dam identity explained a higher proportion of variation among genotypes at the end of the acclimation period than it did in the wild population. Female reaction norms in response to both temperature and pH gradients in this study do not show the same degree of narrowing as male genotypes, with the trans-generational consequence that strong maternal effects (m2) were detected. The ‘bet

182 hedging’ strategy, in which mothers experiencing environmental unpredictability produce a wider range of phenotypes among their offspring (Burgess and Marshall 2014; Garcia-

Gonzalez et al. 2014), may apply in this case. It has also been reported in female sticklebacks

(Gasterosteus aculeatus) in response to maternal exposure to elevated temperature (Shama

2015) and is highly relevant in multi-stressor environments as it can maximise the fitness of a subset of offspring, although at the expense of siblings from the same spawning event.

Future temperature increases of the kind predicted for the next century are not likely to drive natural selection in P. regularis. Several female and male reaction norms show phenotypic buffering (reaction norm slope = 0) in response to temperature increase. In the coral

Acropora pulchra the maintenance of key processes such as growth rate in acidified conditions is due to phenotypic buffering (Shaw et al. 2016), and it has been shown that a history of exposure to fluctuating CO2 enabled the phytoplankton species Emiliania huxleyi,

Gephyrocapsa oceanica and Chaetoceros affinis to buffer long-term capnic stress (Hattich et al. 2017). Buffering is therefore an adaptive strategy for a trait (in this case normal development) which is not under intense selection pressure (in this case from temperature increase).

c) Ecological implications of maladaptive plasticity

Larval dispersal patterns are heavily influenced by larval phenotypic plasticity, which is often assumed to increase local adaptation (Hedgecock 1986; Warner 1997; Hollander et al. 2006).

The maladaptive phenotypic plasticity seen in this study, however, may constrain larval dispersal patterns in a southward direction. The Southern Ocean is acidifying relatively quickly (McNeil and Matear 2008) and the source population for P. regularis in this study is

183 one of the southernmost populations of this species (Byrne et al. 2013). The Otago population of P. regularis may therefore be among the first P. regularis populations to experience an acidified Southern Ocean. The evolutionary ‘cost’ (sensu DeWitt et al. 1998; Uller and Pen

2011) of trait expression when plasticity is maladaptive may be a major constraint on the future expansion of this species (Langerhans and DeWitt 2002; Marshall et al. 2010).

d) Conclusions

There is clear evidence that parent exposure to low pH is maladaptive for P. regularis, but that the antagonistic effects of increased temperature and low pH may reduce these negative trans-generational effects. Maternal identity and maternal pH environment have a strong effect on non-genetic inheritance. [Sire x environment] interactions and large heritability estimates indicate that this population is capable of adapting to combined ocean acidification and warming, but the negative effects of low pH in the absence of elevated temperature may influence present-day larval dispersal patterns in edge-of-range populations.

184

Chapter Seven: Trans-generational plasticity in response to ocean

acidification and warming is strongly dependent on maternal

identity in the Antarctic sea star Odontaster validus

Photo: Adult Odontaster validus predating on Sterechinus neumayeri. Photo credit: K. Sparks

This chapter is currently in preparation as: Sparks KM, Byrne MA, Wing SR and Lamare MD ‘Trans-

generational plasticity in response to ocean acidification and warming is strongly dependent on maternal

identity in the Antarctic sea star Odontaster validus.’ Target publication for this chapter: Proceedings of

the Royal Society Part B (2016 impact factor 4.94) 185

This chapter pertains to:

Hypothesis 3: Trans-generational responses to ocean acidification and warming

The phenotype of F1 offspring will depend on the temperature and pH environment to which the parents are acclimated, thereby increasing fitness in the same environment.

Significance: This is a novel, understudied and increasingly important area of research into adaptive capacity of marine populations. Trans-generational plasticity is an important survival and adaptive mechanism for marine populations and this area of research is becoming more and more important

in the face of abrupt, rapid climate change. Contributions to this chapter:

Contributions to this chapter:

Kate Sparks designed this experiment using an original protocol designed by Shawna Foo.

Kate collected and analysed all data with help from Michelle Liddy. Miles Lamare, Steve

Wing and Maria Byrne provided feedback on an earlier draft of this chapter. Kate was funded

by the University of Otago and the University of Otago Ocean Acidification Research

Theme.

7.1 Abstract

Anticipatory parental effects (APEs), the alteration of offspring phenotype by parents in

response to a changing environment, is a form of trans-generational plasticity that may allow

marine populations to pre-adapt to a rapidly warming and acidifying ocean. This study

exposed the keystone Antarctic cushion star Odontaster validus for 12 months to

environmentally relevant ambient (0.5°C) and future warming (3.5°C) temperatures and

ocean acidification scenarios (pH(T) 8.10 and 7.70). F1 offspring were then generated using a

186 quantitative genetic breeding design and raised to the gastrula stage in all combinations of

2 elevated temperature and pCO2. Broad-sense heritability (h ) was 0.34, indicating that trait plasticity is partly attributable to genetic variation. There was little variation among offspring phenotype attributable to sire identity, but dam identity accounted for 5.79 % of variation among genotypes in the wild population and 23.27 % of variation among genotypes in the F1 population. Maternal identity accounted for 23.27 % of variation among genotypes in the F1 population. Strong maternal effects (m2) were observed at the gastrula stage (m2 = 0.51), indicating that maternal environmental history is a strong determinant factor in non-genetic inheritance. Maternal effects increased trans-generational fitness in response to elevated temperature. These results suggests that O. validus has a pre-adaptive capacity to respond to ocean warming and acidification that may facilitate population persistence over the next century.

7.2 Introduction

An individual altering the phenotype of its offspring relative to its own phenotype in response to an environmental change, termed an ‘anticipatory parental effect’ (APE) (Burgess and

Marshall 2014), is one of a number of documented forms of trans-generational phenotypic plasticity (TGP). Transgenerational phenotypic plasticity is a form of non-genetic inheritance whereby transmission of information between generations occurs through non-DNA based trait alteration (Foo and Byrne 2016) originating from either the environment or phenotype of the parent (Agrawal 2001), such as maternal provisioning or epigenetics, and is usually adaptive or, less commonly, maladaptive (Guillaume et al. 2015). Different phenotypes can be produced from the same genotype in response to environmental variability (Pfennig et al.

2010). Anticipatory maternal effects are generally hard to isolate, as it is thought that

187 maternal contribution to offspring success comes both in the form of genetic and non-genetic material. As maternal environmental history can influence offspring fitness, maternal effects provide a mechanism for an adaptive plastic response in response to the same stressors the mother has experienced (Mousseau and Fox 1998).

Broad-sense heritability (h2) estimates the portion of trait phenotypic variation due to genetic factors (Freeman and Herron 2004; Wray and Visscher 2008). Heritability is influenced by genotypic diversity and environmental variation. Heritability estimates in recent studies have been used to gauge the trans-generational responses of marine populations in response to warming in fish (Munday et al. 2017) and the combination of warming and acidification in urchins (Foo et al. 2014). Heritability is being used to estimate the potential for adaptation of marine populations in the face of a changing climate, as it can increase our understanding of the way in which population responses will interact with evolutionary change, and the potential “winners and losers” (Somero 2010). Evidence that parent environment can be a good predictor of offspring response has been shown, e.g. in oysters (Parker et al. 2015), and indicates that some marine invertebrates can acclimate to future ocean acidification and warming to various degrees (Suckling et al. 2015; Kelly et al. 2016; Uthicke et al. 2016). In some populations long-term acclimation to ocean acidification drives TGP towards evolutionary change at both the individual and population level (Ross et al. 2016).

It has been theorised that long-term adaptation to a thermally stable environment has made

Antarctic marine species extremely sensitive to changes in temperature (e.g. by Peck et al.

2004). There are still uncertainties around the mechanisms whereby polar species have adapted, for example in terms of protein folding and genetic duplication, but some physiological functions of Antarctic marine species, including acclimation to temperature change, are significantly slower than would be expected were the acclimation period determined completely by thermal environment (Peck 2016). This knowledge is the first step

188 towards research that can link acclimation to rapid change into the physiological state of the adult population. In addition, the slower-than-expected thermal acclimation rates among polar marine species opens the question of whether APEs are able to compensate for such slow acclimation, and whether trans-generational plasticity is as viable and important in Antarctic species compared with temperate species and marine vertebrates such as fish (i.e. Shama et al. 2014).

Odontaster validus is a sea star with a circum-polar Antarctic distribution. In Antarctic benthic communities, it can be one of the most numerous species present, and is a keystone predator (Peck et al. 2008). It has a broad thermal range (adult competency 0 - 9.0 °C; pre- settlement larval competency -0.5 to +2.0 °C; Peck and Prothero-Thomas 2002; Peck et al.

2008; Karelitz et al. 2016), and fertilization and early embryo development are robust to reduced pH as low as pH 7.50; Gonzalez-Bernat et al. 2013). The wide temperature and pH tolerance are unusual for a polar species (Peck et al. 2008; Peck 2016), but enable individuals to withstand the seasonal temperature variation of up to 2 °C and seasonal pH changes of up to 0.4 units found in the Ross Sea (IPCC 2013). Odontaster validus has been identified as one of the most stress tolerant Antarctic species, especially with respect to thermal tolerance

(Peck and Prothero-Thomas 2002; Peck et al. 2008; Peck 2016). As stress tolerance often conveys an overall resilience to perturbation, this sea star may possess the genetic variation to withstand end of century forecasted ocean warming and acidification (IPCC 2013). In echinoderms and many other invertebrates it is well known that the thermal history of the parents determines the thermal tolerance of the offspring, an important phenotypic trait that contributes to the ability of embryos and larvae to cope with environmental warming (Byrne

2011).

189

I addressed the hypothesis that acclimation of adult O. validus to increased temperature and low pH over 12 months would result in modification of offspring phenotype with respect to tolerance to these stressors, thereby increasing fitness in the environment to which their parents were acclimated. If this is the case, I ask the question as to what portion of this trait is heritable.

7.3 Methods

a) Animal collection

Adult Odontaster validus were collected from benthic sites around Ross Island, McMurdo

Sound, Antarctica by SCUBA in November 2014. After collection, adult O. validus were held in flow-through aquaria for a period of ten days at Scott Base, Antarctica, before being air transported back to New Zealand in 60 L refrigerated containers filled with fresh filtered seawater at the ambient conditions from the Scott Base aquarium tanks.

Mean water temperature in the Scott Base aquarium was -1.2 °C (±0.1 °C) within the mean monthly temperature range for McMurdo Sound (-1.80 to -0.90 °C; McNeil and Matear 2008;

Kapsenberg et al. 2015). Mean seawater pH(T) in McMurdo Sound over the collection period was pH 8.04 and varied daily within a range of ±0.06 units, measured spectrally (method set out in Foo et al. 2016). Detailed seawater chemistry (DIC, Alkalinity, pCO2) at the time of collection is given in Appendix 10.6. Animals were imported to New Zealand under MPI

Permit #2014054106 to New Zealand and held in PC2 containment conditions at the

University of Otago’s Portobello Marine Laboratory (PML) for the duration of the study.

190

b) Adult acclimation

In New Zealand, 200 adult O. validus were held for 8 months (November 2014 – August

2015) in ‘ambient’ conditions (-1.2 °C) to standardise physiology and feeding among the population. Aquarium conditions are given in Appendix 10.6 In August 2015 a sub-set (22 individuals; 8 females and 14 males) of the animals kept in the ‘ambient’ conditions were spawned and mated using a quantitative genetics breeding design (see Spawning and

Fertilisation section below for details of method) and the offspring raised to gastrulation to examine the range of genotypes within the wild population.

The adults were then exposed to combinations of two temperature (0.5 °C ± 0.2, 3.5 °C ± 0.2

SE) and two seawater pH treatments (pH(T) = 8.10 ± 0.15, pH(T) 7.70 ± 0.03 SE) in a fully- crossed, orthogonal, and replicated factorial design for a further 9 months. This is longer than the suggested minimum time period (6-8 months) needed for other adult Antarctic species

(Sterechinus neumayeri) to acclimate to elevated temperature and low pH (Suckling et al.

2015). The first 3 months of this was an acclimation period, where temperature and pH were adjusted gradually as recommended by Suckling et al. (2015). For this, aquarium temperature was raised at a rate of 0.3 °C per week, and pH was lowered at the rate of 0.03 units every three days. Aquarium design is given in chapter 5 figure 5.1.

After the 12 months the acclimated adults were bred, and their offspring (hereafter referred to as the F1 generation) were reared to gastrulation in either the same treatment as their parents or at a different pH or temperate to that of their parent. Offspring of all parents were reared in all treatment combinations. The experimental design was fully crossed so that offspring of each male-female pair were kept separate.

191

c) Spawning and Fertilisation

Ovaries were removed by dissection and placed in ice-cold 10-5 M 1-methyladenine to induce maturation of oocytes and release. Eggs from each individual were placed in 250 ml containers with filtered experimental seawater (FSW) in all combinations of temperature and pH and allowed to acclimate for one hour to the treatment in which the embryos would be reared. Sperm was then added to achieve a final sperm: egg ratio of 2500:1. Offspring of all pairs were reared in all four treatment levels (0.5 °C & pH 8.10, 0.5 °C & pH 7.70, 3.5 °C & pH 8.10, 3.5 °C and pH 7.70), including in the same treatment as their parents were acclimated, and in the 3 other treatment combinations.

After 60 minutes the presence of a fertilisation envelope indicated >60 % fertilisation. Eggs were rinsed to remove excess sperm and the containers were replenished with fresh FSW from the same treatment. Offspring from each family (M x F pair) were kept separate by crossing individual male and female gametes so that parentage of each offspring population could be tracked, following the North Carolina II breeding design and fertilisation protocols

(Delorme and Sewell 2016; Foo et al. 2014, 2016; Sparks et al. 2017), in which two females were crossed with four males in all combinations in each experimental run (“block”).

Embryos from each fertilisation container were then divided among three falcon tubes with mesh-insert windows (details of flow and dimensions given in Sparks et al. 2017) and placed into the recirculating aquaria described in Chapter 5 Fig. 5.1. At 6 d post-fertilisation, subsamples of ~50 gastrulae were removed from each falcon tube and scored for normal development according to the developmental schedule for this species (Pearse 1969).

192

d) Manipulations of experimental conditions

Adults were acclimated in recirculating aquaria in a temperature-controlled environment in which pH was manipulated by the addition of pure CO2 into a mixing tank at the correct temperature. Temperature in the aquaria was monitored constantly and logged every 30 minutes using HOBO loggers (Onset HOBO Pendant data/alarm logger 64K model UA-001-

64, Massachussetts). CO2 injection and pH in the aquaria were controlled using a Tunze pH/CO2 controller 7074/2 which was set to the relevant pH(T) (eventually 7.70). Carbonate chemistry was measured every 3 months throughout the parent acclimation period as well as daily while larvae were being raised by removing a 250 ml subsample of water from each treatment, fixing with 100 µl HgCl2 and analysing for DIC, TA and pCO2 content (see

Appendix 10.6).

Adults were kept in recirculating aquaria in a temperature-controlled environment in which pH was manipulated by the addition of pure CO2 into a mixing tank at the correct temperature. Water from this tank then flowed into a header tank that controlled the water supply to three aquaria in which adult O. validus were kept at a density of 0.8 individuals/ L3 with a total of 8 individuals per 12 L tank. Each system comprised a CO2 mixing tank

(sump), a header tank and three treatment tanks and held 120 L of seawater at any time.

Water flowed into the animal tanks at a maximum rate of 3 L per minute. Seawater was replaced at a rate of 50 % per week (60 L per week), with fresh filtered seawater at the appropriate temperature. Seawater was removed and replaced in the mixing (sump) tank so that experimental pH conditions were equalised before water was allowed to flow into the header or animal tanks. Temperature in the aquaria was monitored constantly and logged every 30 minutes using HOBO loggers (Onset HOBO Pendant data/alarm logger 64K model

193

UA-001-64, Massachussetts). CO2 injection and pH in the aquaria were controlled using a

Tunze pH/CO2 controller 7074/2 which was set to the relevant pH(T) (eventually 7.70).

7.4 Data analysis

Data on percentage of normal development were arcsine transformed prior to analysis to satisfy assumptions of normality, then analysed using a nested multivariate ANOVA design in the Permanova+ extension of Primer 6 software (Clark and Gorley 2006) to determine contributions of fixed and random factors to variation among offspring genotypes. Holm-

Sidak post-hoc testing separated out interactions between levels of factors. Temperature and pH were fixed factors. Sire and dam identity were random factors nested within experimental block and within experimental tank to account for the aquarium system design used (outlined above). Block and tank were both random factors. Variance components for random factors were estimated in the lmerTest package of R, which uses a reduced-error-maximum- likelihood (REML) analysis. Data tables in Appendix 10.7.

This design meets the minimum requirements described by Burgess and Marshall (2014) to identify APEs in that (1) the manipulations of both parent and offspring environment were orthogonal and fully-crossed, (2) the F1 generation were reared in a quantitative genetic design which allowed the separation of embryos into families with traceable parentage, (3) the mortality among the parent generation was low, (4) all individuals in the captive population were bred from, (5) the offspring of each M x F pairing were split between either the same environment as their parents or another environment (in this study, F1 offspring of all pairings were assigned to all environments) and (6) the trait measured in the F1 generation

(in this study, normal development) is correlated to survival and fitness.

194

Heritability (h2) and maternal effects (m2) were estimated using variance components generated by the REML analysis across each environment (either temperature, pH or temperature x pH) following an adaptation (based on Foo et al. 2014) of the animal model

(Kruuk 2004). The variance components used in the two calculations were the additive genetic variance (sire), the non-genetic variance (dam) and the residual error variance.

For each male and female genotype from both the control and acclimated populations, reaction norms were generated to examine the performance of each genotype across the temperature and pH environments in which the offspring were raised. If the slope of a reaction norm describing the response of a phenotype to an environment changed depending on environment, that genotype was concluded to be plastic (Freeman and Herron 2004;

Suckling et al. 2015).

7.5 Results

a) Trans-generational plasticity

Mean percentage of normal development of F1 larvae was significantly affected by both the pH and temperature treatment in which they were raised (pH treatment: pseudo-F(DF = 1) =

4.245, p = 0.045; temperature treatment pseudo-F(DF = 1) = 7.265 p = 0.049). There was a significant interaction between temperature and pH 7.70 (Holm-Sidak t = 4.488 p = 0.033), with the response to each variable dependent on adult exposure history (Fig. 7.1). For example, offspring of adults acclimated to pH 7.70 and 3.50 °C had on average 10 % lower normal development even when the offspring were raised in ambient pH (Fig. 7.1d).

195

Parental environment also significantly affected larval development at gastrulation and

explained 12.15 % of variation among F1 genotypes (pseudo-F = 8.1941 p = 0.050).

Offspring developed in the 3.50 °C treatment had on average 15 % higher normality when

their parents had been acclimated to warm conditions at either ambient (Holm-Sidak t =

2.868 p = 0.006) or low pH (Holm-Sidak t = 3.877 p = 0.012) (Fig. 7.1).

Figure 7.1: Normal development of F1 Odontaster validus gastrulae is dependent on both parent and offspring environment in response to elevated temperature and low pH. Parent environment is given along the horizontal axis and offspring environment is shown by the shaded bars. (A) Effect of parent temperature exposure on offspring raised at pH 8.1; (B) effect of parent temperature exposure on offspring raised at pH 7.7; (C) effect of parent pH exposure on offspring raised at 0.5 °C; (D) effect of parent pH exposure on offspring raised at 3.5 °C. Letters denote levels of significance. Error bars are S.E. n=75 for all.

196

b) Heritability and maternal effects

Dam identity was a significant factor in determining offspring response. At the gastrula stage, dam identity explained 5.79 % of variation among genotypes in the wild population (pseudo-

F = 4.89 p = 0.001), and 23.27 % of variation among genotypes in the acclimated population

(pseudo-F = 12.51 p=0.014). Three female genotypes indicated the presence of phenotypic buffering (reaction norm slope = 0) in the offspring of one female that had been acclimated to

0.5 °C/ pH 7.7 and the offspring two females that were acclimated to 3.5 °C/ pH 7.70 (Fig.

7.2). The interaction between [dam x temperature] explained 6.22 % of variation among genotypes (Fig. 7.2).

Broad-sense heritability (h2) from dam identity was estimated to be h2 = 0.34 (± 0.08 SE) among offspring of acclimated genotypes. For the wild population, h2 = 0.28 at low pH.

Strong maternal effects (m2) were estimated; m2 = 0.51 (± 0.11 SE) across both the temperature and pH treatments.

197

Figure 7.2: Dam identity interacts with temperature to influence offspring performance among populations of Odontaster validus. Data are reaction norms where each line represents the response of one female genotype to increasing temperature. (A) female genotypes among wild population; (B) female genotypes kept in control conditions in lab for 12 months; (C) female genotypes kept at 0.5 °C and pH 7.70 for 12 months; (D) female genotypes kept at 3.5 °C and pH 8.10 for 12 months, and; (E) female genotypes kept at 3.5 °C and pH 7.7 for 12 months. N=3 for all.

198

7.6 Discussion

This study exposed adult Antarctic sea star Odontaster validus for 12 months to environmentally relevant ambient and future scenarios of temperature and pH to explore the potential of trans-generational plasticity and adaptation in their offspring. I tested the hypothesis that parent O. validus will be able to modify the phenotype of their offspring to ensure fitness in response to their own environment.

a) Parent acclimation to elevated temperature

I found that acclimation of female O. validus in near future warming and acidification conditions resulted in a modification of offspring phenotype, improving their fitness in response to increased temperature and, to a lesser extent, acidification. Parent temperature history was a reliable predictor of offspring performance in the same environment, a common phenomenon in some marine invertebrates such as urchins (Suckling et al. 2015). In this respect, F1 generation embryos from mothers acclimated to warmer temperatures had higher developmental normality than offspring of parents acclimated to colder temperatures. Within each environment, the rank order of individual female genotypes changed, indicating that natural selection favours certain genotypes in each environment.

Results from previous studies of TGP in marine populations have demonstrated that invertebrates are capable of some degree of heritable plasticity in response to ocean acidification and warming (Reusch 2014). Response to reduced pH in larvae of the oyster

Saccostrea glomerata indicated that maternal provisioning was relevant in determining offspring developmental success in the next generation (F1) but not any subsequent

199 generations (F2) suggesting that APEs give way to genetic change over longer timescales

(Parker et al. 2015; Ross et al. 2015). This conclusion is borne out by studies which examine differences in gene transcription as well as phenotypic changes in long-term exposures, for example gene translation in adult Saccostrea glomerata differed after exposure to high CO2 for three generations dependent on both parent exposure history and trait variation among genetic lines (Goncalves et al. 2016). It is worth noting, however, that long-term exposure to stress does not increase phenotypic plasticity in all cases, as for the copepod Pseudocalanus acuspes in which a reduction in phenotypic diversity accompanied exposure to high CO2 among the F1 generation (Kelly et al. 2016), leaving many questions still to be answered about the interaction between plasticity and evolutionary adaptation.

For O. validus, offspring of parents acclimated to increased temperature showed a significant increase in normal development when exposed to increased temperature compared to offspring of parents from the colder treatments, indicating that parent exposure to increased temperature confers a level of resilience to their offspring. This result is documented for several marine invertebrates (Suckling et al. 2015; Ross et al. 2016; Parker et al. 2017) and is a tool used to condition parents to improve outcomes in aquaculture hatcheries (Saillant et al.

2001; Parker et al. 2017). This trait will be important as O. validus will not be able to benefit from any poleward range shifts suggested for other marine species as oceans continue to warm (Cheung et al. 2009; Byrne et al. 2011).

In terms of pH responses, F1 gastrulae were only resilient to reduced pH 7.70 when their parents were exposed to warm temperatures. Often, in populations subjected to multiple stressors, each stressor has a different effect magnitude on offspring resilience, and this can

200 increase phenotypic diversity, perhaps leading to an increase in resilience to ocean acidification and warming (Byrne et al. 2011), and successful pre-adaptation to climate change (Chakravarti et al. 2016; Gibbin et al. 2016). Offspring resilience to oxidative damage in polar echinoderms has been demonstrated, as for S. neumayeri embryos, maternal provisioning mitigated oxidative damage (Lister et al. 2017).

A subset of females in the acclimated population demonstrated the ability for phenotypic buffering, which maintains a stable phenotype despite environmental temperature change.

Again, this type of TGP has already been demonstrated for S. neumayeri females in response to oxidative stress from UV exposure (Lister et al. 2017). This is most noticeable among females acclimated to +3.5 °C/ pH 7.70 in which the reaction norms of three female genotypes remained consistent across all thermal environments. F1 generation gastrulae from females acclimated to elevated temperature developed ‘best’ in elevated temperature, although this effect was not consistent across developmental stages. As O. validus embryos are resilient to low pH anyway (Gonzalez-Bernat et al. 2012), the effect of parent exposure may not be easily visible. This study indicates that complete acclimation to increased pCO2 may not be possible within the time scale of one reproductive cycle of this species. Although reaction norms for wild population adults indicate that the majority of genotypes perform better in warmer temperatures, the same genotypes show evidence of phenotypic buffering

(reaction norm slope = 0) when exposed to 1400 ppm CO2. The benefits of phenotypic buffering as a life history strategy are that fitness and survival are not necessarily correlated with environment, specifically, that a fitness-correlated trait (such as normal development in this study) is not a function of a changing environment. In the coral Acropora pulchra the maintenance of key processes such as growth rate in acidified conditions is due to phenotypic buffering, which allows the individual polyps to survive for longer (Shaw et al. 2016), and it

201 has been shown that a history of exposure to fluctuating CO2 enabled the phytoplankton species Emiliania huxleyi, Gephyrocapsa oceanica and Chaetoceros affinis to buffer long- term capnic stress (Hattich et al. 2017). This strategy is beneficial to an individual in a scenario where a population exists at the edge of its tolerance range, and therefore the ability to maintain homeostasis under stress is an adaptive advantage (Reusch 2014). Although it has been stated that Antarctic fauna are likely to have narrow phenotypic variance due to the

‘stable Antarctic’ theory, evidence that other Antarctic species, such as the sponge

Homaxinella balfourensis, possess greater than expected phenotypic plasticity in response to rapid environmental change (Morley et al. 2016) indicates that phenotypic buffering may not be a viable long-term solution.

Defining the benefits of plasticity in O. validus may not be possible without knowing the timescales on which the adult population completely acclimates to ocean warming and acidification, but there are some important conclusions which can be drawn. Strong heritable maternal effects and the ability to pass on thermal tolerance will afford those more adaptable individuals some fitness advantage in the near-future over their non-plastic con-specifics. The population of O. validus in this study contains a sub-set of individual females who are able to pass on their acclimated tolerance to elevated temperature to the next generation, and therefore possess the capacity to adapt to rapid warming. However, the lack of genetic correlations between trait responses to the two stressors in this study indicate that the conditions predicted for the end of this century (RCP 8.5) may not drive beneficial natural selection in this particular trait. Antarctic marine populations are not able to migrate poleward in order to mitigate warming temperatures as has been predicted for other higher latitude species (Mieszkowska and Sugden 2016) such as the sea star Asterias amurensis (Byrne et al.

2016) and many commercially-important species (Woodsworth-Jefcoats et al. 2017). This

202 means that Antarctic populations must adapt to the predicted warming conditions for this century (+2.5°C this century; RCP 8.5; IPCC 2013) while remaining in the same geographical location.

The results of this study, which show significant differences between larvae from different pH treatments after 6 days, are slightly different to those of some other published studies, but the same general trend can be seen. Work by Gonzalez-Bernat et al. (2013b) concluded that low pH affected survival only from 10 days post-fertilisation onwards. In this study, it was not possible to take the embryos to 10 days old, for a number of reasons. Firstly, maintaining survival of the embryos from each treatment was tricky and I felt that attempting to keep them alive for 10 d would have been very difficult. A late-stage crash in one treatment out of many would render that entire experimental run unusable, leading to delays. Secondly, as the development time of this species is so much slower than the others used in this thesis, exposure time may have become a confounding variable, for example had this species been exposed to low pH for 10 d where P. regularis had only been exposed for 3 days. As the embryos successfully reached gastrulation, it was felt that 6 d was an appropriate length of time for this experiment, both to collect meaningful data in the form of significant differences between treatments and to avoid unneccesary artifacts of sampling. If circumstances had allowed a longer experiment, I have confidence that the observed trends would have been the same, but future researchers undertaking this type of experiment may wish to plan their sampling time-points to avoid time constraints.

203

b) Maternal investment in offspring

I found that heritable maternal effects are a strong deterministic factor in non-genetic inheritance. Maternal effects are highly context-dependent as they represent an anticipatory response to a particular environment or variable. The presence of maternal effects alone, therefore, is not enough to conclude that O. validus is able to adapt to ocean warming and acidification, although taken in conjunction with the strong heritability estimates in this study, does indicate that a large component of normal development at elevated temperature is genetic.

Heritability and maternal effects in combination with the contribution of [dam x environment] interactions to offspring developmental success in this study mean that this population of Odontaster validus may be capable of rapid adaptation to ocean warming, but that tolerance of ocean acidification may need to be handled by each successive generation in terms of adult physiological acclimatization. Ocean warming is predicted to influence the

Southern Ocean by the end of the 21st century (IPCC 2013) and the results of this study suggest that this species may be one of the evolutionary ‘winners’ in this scenario. However, in combination with other environmental stressors such as acidification, O. validus will need a combination of phenotypic plasticity and physiological tolerance in order to respond to combined stressors on timescales too short for genetic mutations to become functionally active in altering genotype at the population level.

Although this study does not contain any examination of the mechanism by which phenotypic plasticity is inherited, one possible vehicle is maternal investment in egg quality. Eggs, which contain high-energy compounds such as lipoproteins and phospholipids, are the sole energy source for pre-feeding embryos. Therefore, the higher the energetic investment in the egg, the

204 better chance that the offspring has of normal larval development. Maternal effects such as these have been demonstrated within one generation before; in Evechinus chloroticus maternal antioxidant provisioning is able to alleviate oxidative stress in the next generation

(Marshall and Uller 2007). Ocean warming and acidification have been suggested, both in this study and others (Portner and Farrell; 2008; Kelly et al. 2016; Morley et al. 2016), to increase the energetic demands of marine species, meaning that adults may reallocate their energetic reserves to themselves at the expense of the next generation.

c) Conclusions

This study provides clear evidence that exposure of female O. validus to warming over a relevant period (i.e. over an annual reproductive cycle) modifies the phenotype of their offspring thereby enhancing their fitness in a warming environment. Maternal identity and maternal environment strongly influence offspring developmental success, meaning that this population of Odontaster validus may be capable of pre-adaptation to ocean warming, predicted to influence the Southern Ocean by the end of the 21st century.

205

Chapter Eight: Discussion

Standing genetic variation and trans-generational plasticity

among populations of echinoderms in response to ocean

acidification and warming

Figure 8.1 Thesis workflow as it pertains to hypotheses.

Note on species names: Acanthaster spp. will be used to refer to both of the species commonly referred to as either Acanthaster planci or Acanthaster cf. solaris due to recent reclassification.

206

8.1 Summary of main findings relating to hypotheses

a) Hypothesis 1: Standing genetic variation within wild populations

Hypothesis 1 stated that ‘wild populations of echinoderms will possess the adaptive capacity to combinations of high temperature and high pCO2 equivalent to those forecasted for the end of the 21st century in polar, subtropical and temperate locations.’ This hypothesis was investigated by characterising the standing genetic variation within wild populations of the sea stars Acanthaster spp., Patiriella regularis and Odontaster validus, and the sand dollars

Arachnoides placenta and Fellaster zelandiae in response to combinations of elevated temperature and pCO2 (measured experimentally as low pH).

‘Adaptive capacity’ in this hypothesis was indicated by a significant sire x environment interaction and significant trait heritability for the ‘normal development’ trait at gastrulation among offspring of wild populations. Our hypothesis was supported both subtropical species, and the temperate F. zelandiae, as significant sire x environment interactions and broad-sense heritability values (h2; Table 1) were found for those species. This hypothesis was not supported for P. regularis or O. validus as there were no sire x environment interactions within the wild population for these species, and therefore standing genetic variation in response to ocean acidification (the dominant stressor in both species) was deemed not to confer any adaptive capacity.

I found that a subset of individuals within the Acanthaster spp. population from eastern

Australia possessed the capacity to adapt to the combined ocean acidification and warming

207 conditions forecasted for the year 2100 (Hobday and Lough 2011). Variation among individual genotypes within this population is dependent on genetic factors conferred by sire identity and sire x environment interactions with little maternal influence. In both sand dollar species examined, A. placenta and F. zelandiae, tolerance to elevated temperature and low pH was not strongly heritable in any combination of stressors, but some effects of sire and dam identity were detected indicating population resilience across all genotypes. The only species in which a strong influence of dam identity was observed were P. regularis and O. validus, and no significant effect of sire identity on normal development were detected in those species. These results are summarised in Table 1.

Table 1: Summary of results from Hypothesis 1. Data are mean values from wild

populations of echinoderms only. Significant values only reported; ‘NS’ indicates that no

significant effect was found (p>0.05). ‘D.E.S.’ = Dominant environmental stressor on

population mean normal development. VS = variation among families due to sire identity.

h2 = broad sense heritability in response to D.E.S.

2 Species Population source D.E.S. Temp x pH V(S) h environment interaction Acanthaster spp. Subtropical (Aus) Temp. Antagonistic 0.87 0.44 Arachnoides placenta Subtropical (Aus) pCO2 None 0.14 0.06 Fellaster zelandiae Temperate (NZ) pCO2 None 0.24 0.05 Patiriella regularis Temperate (NZ) Temp. None NS 0.16 Odontaster validus Polar (Antarctica) Temp. Antagonistic NS 0.28

208

b) Hypothesis 2: Physiological acclimation to ocean acidification and warming.

Hypothesis 2 stated that ‘Echinoderms possess the physiological capability to acclimatize to ocean acidification and warming over long time-scales without a significant cost to metabolic energy budget.’ This hypothesis was tested using adult sea stars P. regularis and O. validus.

Wild populations were exposed to combinations of elevated temperature for 12 months culminating in the peak reproductive season of each species. Before, during and after the exposure period, bioenergetic and physiological measurements were taken allowing the estimation of Scope for Growth (SFG).

This hypothesis was supported for both species. Both P. regularis and O. validus were able to acclimate to a combination of elevated temperature and low pH over 12 months, indicated by no change in their SFG values after 12 months in the combined stressor treatment (Table 2).

Both species, however, were affected by a reduction in seawater pH in the absence of elevated temperature suggesting that the effect of these two stressors on sea star metabolism is antagonistic. Odontaster validus was able to alter its feeding behaviour in order to compensate for the increased metabolic demands of combined warming and acidification, but

P. regularis was only able to do so by reallocating stored energy away from reproductive capacity towards maintenance of homeostatic balance.

209

Table 2: Summary of results from Hypothesis 2. ‘DEF’ = Dominant environmental factor. (+/-)ive

= positive or negative effect of the DEF on SfG. * = ‘at end of exposure period in response to

combined low pH and high temperature.’

Species Population D.E.F. Temp. x pH Reduction Decreased Main source (+/-)ive interaction? in SfG* reproductive mechanism of environment capacity?* homeostatic maintenance? Patiriella Temperate pCO2 Antagonistic N Y Energy regularis (NZ) (-) reallocation Odontaster Polar Temp. Antagonistic N N Increase in food validus (Antarctica) (+) ingestion

c) Hypothesis 3: Trans-generational plasticity in response to ocean acidification and

warming

Hypothesis 3 stated that ‘the phenotype of F1 offspring will depend on the temperature and

pH environment to which the parents are acclimated.’

This hypothesis was tested using P. regularis and O. validus offspring of the individuals

acclimated to combinations of elevated temperature and low pH used for hypothesis 2. After

the exposure period, adult sea stars were bred and their offspring raised to gastrulation in

either the same or a different environment to their parents. Trans-generational plasticity was

mainly identified as Anticipatory Parental Effects (APEs).

Our hypothesis was supported for both species examined, but with different evolutionary

implications for each species. Trans-generational plasticity in P. regularis was maladaptive in

response to reduced pH so that fitness was decreased among offspring of parents acclimated

to low pH regardless of the environment in which the offspring were raised. Among O.

210 validus F1 gastrulae, trans-generational plasticity was also maladaptive in response to ocean acidification, but the antagonistic effect of elevated temperature in the offspring environment was able to counteract the maladaptive plasticity in order to maintain high % normal development. In both species examined, APEs were mainly attributable to maternal identity, and strong maternal effects were detected among the F1 generation (Table 3). Strong heritability values in both species indicated that trait plasticity is partly attributable to genetic variation, and therefore that the APEs detected here may be the basis for pre-adaptation in O. validus.

Table 3: Summary of results from Hypothesis 3. Data are mean values from F offspring of parents

exposed to combinations of elevated temperature and low pH prior to reproduction. ‘DES’ =

Dominant environmental stressor on population mean normal development. h2 = broad sense

heritability in response to D.E.S. M2 = maternal effects in response to D.E.S. * = ‘in response to

† combined stressors.’ = ‘among F1 offspring of the acclimated population.’

Species Population D.E.S Temp x pH (h2)† (m2) Direction of source interaction? plasticity* environment Patiriella Temperate pCO2 None 0.24 0.41 Maladaptive regularis (NZ) Odontaster Polar pCO2 Antagonistic 0.34 0.51 Adaptive validus (Antarctica)

211

d) Experimental limitations and recommendations for further work

Laboratory- based experiments face challenges replicating the complex multi-stressor nature of the real-world environment. In particular, this study encountered problems due to some necessary concessions made for experimental design and robustness balanced with space and time considerations. Aquarium designs for multi-stressor studies are prone to pseudo- replication problems (Cornwall and Hurd 2016) which this study was able to work around with a statistics-based approach. A key consideration for future experimental design should be statistical validity and the avoidance of pseudo-replication with considered experimental design.

One aspect of this thesis which entailed working with pseudo-replicated data was the creation of the reaction norm graphs. Due to the pseudo-replicated count data from each falcon tube

(section 2.3c), the data points making up each reaction norm could not be used to generate estimations of error as they were not independent counts. Presenting these graphs without estimations of error seemed like an appropriate way to treat these data, but it may hamper understanding of the validity of these types of graph, especially in published manuscripts.

Readers may feel they are unable to trust the data presented in reaction norms without estimations of error. Each individual reaction norm panel displays a relationship which was estimated to explain a significant portion of variance from the PERMANOVA model, and the calculations of variance were careful to omit repeated count data which were pseudo- replicates of each other. I feel that the data presented in the reaction norms are useful to explain the relationships which were indicated as ‘significant,’ but must be presented alongside the variance component estimations in order to be useful. I hope that presenting reaction norms alongside the estimations of variance will enable readers to better understand

212 the complex interactions between individual genotypes and small changes in each environmental variable, even without error bars.

Issues surrounding photoperiod could have the potential to confound the results of this study, especially those from chapters 4 – 7. As discussed in sections 4.5b and 5.5b, the lack of photoperiod control stemmed both from the assumption that this would not be detrimental to the animals as well as from practical limitations on the equipment and aquaria used. The truth of these assumptions are somewhat borne out by the experiment, as both species examined produced ripe gametes at the expected times of year. Phenological control of gametogenic cycles appeared to be normal, however as the light and dark exposures were not explicitly controlled in this study as they have been for others (e.g. McClintock et al. 1988), these assumptions were not quantified. If this study were to be repeated I would urge future researchers using Odontaster validus to consider mimicking the light/ dark cycles of

Antarctica, especially for long term experiments or those looking at environmental controls on gametogenesis.

Another major challenge of this work arose from working across many locations and measuring pH using different techniques and equipment across the different chapters. The data for chapter 2 and the Arachnoides part of chapter 3 were collected in Australia; the data for chapters 4 and 6 were collected in Dunedin and the data for chapters 5 and 7 were collected between Antarctica and Dunedin. The different standards to which equipment in each place measures and controls the environmental variables made for a greater-than- expected variation among treatments. For example, ‘ambient’ pCO2 in the SeaSim experiments with Arachnoides rose over 500 ppm while in Dunedin it could be below 450

213 ppm. In that case, the ‘ambient’ carbon dioxide levels in the SeaSim were constructed to reflect the surrounding seawater and atmospheric conditions, which were higher in carbon dioxide than either Dunedin or Scott Base. The carbon dioxide delivery system in Scott Base was a modified version of the set-up in Dunedin, but both those systems relied on researchers to manually calibrate and adjust flow of carbon dioxide into the aquaria, whereas the SeaSim has an advanced feedback system with both internal and external calibration systems built-in.

This also meant that the response times of the carbon dioxide injection systems varied between experimental set-ups, with the result that the treatment levels used were not as consistent as they should have been. In turn, this could have altered our perception of the

‘normal’ development of embryos in control conditions. We checked the look of our embryos against other published photos of the same species wherever possible, and sought advice from experienced larval biologists including Maria Byrne, Mike Barker and Sven Uthicke whenever possible to ensure that what I identified as ‘normal’ really was the norm for that species, and I am confident that even though some of the carbon dioxide treatments were a little higher than others, it did not affect the appearance of normal embryos in the control treatments.

Internal controls on carbon dioxide feedback cycles were tight, so that carbon dioxide levels within the treatments held steady in accordance with research suggesting that treatment variability, as well as treatment level, in important in ocean acidification research. As well as this, in each case the treatments were matched as well as possible to the local seawater conditions that each species would have been familiar with. Even though, for example,

Arachnoides embryos were held at a higher pCO2 than Patiriella, their conditions were not outside the range that either species would have experienced in the wild. Although some allowances can be made in these kind of experiments for variation in local conditions, tighter

214 control on carbon dioxide treatment levels would allow for more direct comparison between experiments, and between my results and other published studies.

8.2 Adaptive potential in echinoderms

a) Adaptive capacity in response to multiple stressors

In this study, the dominant stressor at the gastrula stage varied between temperature and pH depending on species. Others have found that to be the case, but in most studies temperature has often been found to have the larger effect size (e.g. by Nguyen et al. 2012). The main finding of this study pertaining to adaptive capacity in wild populations of echinoderms is that population resilience to climate change will partly depend on adaptive capacity to both elevated temperature and low pH. Regardless of which environmental stressor has the larger effect size, both elevated temperature and low pH pose potential mortality bottlenecks which may impact local larval recruitment regimes. Our findings agree with other studies which have found that the potential for adaptation is not uniform across environments, phylogenetic groups or developmental stages (Byrne and Przeslawski 2013; Kroeker et al. 2013; Hofmann et al. 2014).

Across the board, it is difficult to pinpoint the contributions of phylogeny, evolutionary history and environmental exposure to multiple stressor tolerance. There is no clear pattern in the literature across multi-stressor studies of this kind using temperature and pH as to which environmental factor has the larger effect size. For many studies, interactive effects of ocean warming and acidification are of greater ecological relevance, given that multiple stressors

215 are assumed to always co-exist in nature (Przeslawski et al. 2008; Byrne et al. 2009; Byrne

2011; Garzke et al. 2016).

Stressor interactions can be broadly categorised as additive (Talmage and Gobler 2011;

Armstrong et al. 2017), synergistic (Reynaud et al. 2003; Lischka et al. 2011) or antagonistic

(Gooding et al. 2009). Antagonistic effects have been found from environments including sub-tropical coral reefs, intertidal banks, and coastal zones (Edmunds et al. 2012, Duarte et al. 2014). Antagonistic temperature x pH interactions are common, accounting for 38 % of all multiple-stressor interactions, and can be considered ‘opposing evolutionary vectors’ due to the frequency with which antagonistic effects are found (Crain et al. 2008; Byrne and

Przeslawski 2013; Gibbin et al. 2017). However, other meta-analytical studies found that the most common type of interaction when studying parameters such as calcification, reproduction and survival was synergistic, accounting for 65 % of interactions (Harvey et al.

2013; Przeslawski et al. 2015), indicating that the complexity of multi-stressor interactions is not yet fully understood.

Of the temperature x pH interactions found in this study, the majority were antagonistic.

Combinations of sub-lethal stressors have been suggested to lead to antagonistic, not synergistic, effects on survival-correlated traits (Przeslawski et al. 2015), and antagonistic stressor interactions have also been implicated in limiting the potential for adaptation to multiple stressors (McBryan et al. 2013; Lange and Marshall 2017). This study presents a middle ground: Acanthaster spp. possessed the capacity to adapt to combined ocean acidification and warming, demonstrated by positive genetic correlations and sire x environment interactions, despite elevated temperature and low pH having an antagonistic

216 interactive effect on population mean normal development. Likewise, Odontaster validus showed that TGP can be adaptive despite antagonistic interactions over multiple generations, although in that case there were no significant sire x environment interactions among the population. In both these cases, although the two stressors were antagonistic, one stressor had a larger effect size than the other. This led to the effect of the ‘smaller’ stressor being not only cancelled out, but actively reversed so that the overall effect appears positive in the direction of the stressor with the larger effect size.

The lack of interactive effects detected in this study among F. zelandiae, P. regularis and A. placenta wild populations despite significant genotype x environment interactions for all species indicates that these types of interactions are more complex than is sometimes possible to identify. The North Carolina II breeding design used in this study is able to tease out subtle interactions, but in no case can it explain every contribution to variation among genotypes.

Integrating genetic barcoding or transcriptomic data into these types of experiments would complement those datasets, and allow the genetic basis for adaptive capacity to be checked.

One interesting aspect of the findings pertaining to hypothesis 1 in this study is the large

2 difference in VS and h between Acanthaster spp. and others. Phylogeny interacts with environmental exposure history and local niche adaptation to influence this result. The influence of phylogeny can be seen from the results of the two sand dollar species in this study, Arachnoides placenta and Fellaster zelandiae. Although from different environments, their adaptive responses are most similar out of all species studied here and they are phylogenetically the two closest-related species studied, both being representatives of the only two genera (Arachnoides Leske 1778 and Fellaster Durham 195) within the subfamily

217

Arachnoidinae (Echinodermata: Echinoidea: Clypeasteroidea: Clypeasteridae Duncan 1889).

Phylogenetic considerations may be important when looking at the results of animals from different environments, as environment is not the only predictor of tolerance and there needs to be more incorporation of phylogenetic relationships into modelling of species responses.

The influence of environmental exposure history is not uniform despite broad latitudinal trends. Although the Acanthaster and Arachnoides populations in this study were taken from geographically-similar locations in northern Queensland and from bodies of water with similar ambient temperature and pH regimes, their adaptive responses are very different.

While Arachnoides has low heritability estimates and genetic correlations, Acanthaster has the opposite. As discussed above, these differences may be partly due to the phylogenetic distance between the two subtropical species, but also the particular dynamics of Acanthaster spp. life history modes. As climate change has been suggested to be a co-factor in

Acanthaster outbreaks on the Great Barrier Reef (Uthicke et al. 2015), the dominant influence of temperature on larval development is not surprising. In temperate and polar areas, species in this study are more likely to have a proportionally larger influence of maternal identity on larval development (Patiriella and Odontaster) if that species is classified within the order Valvatida (Echinodermata: Asteroidea: Valvatacea Perrier 1884), however, once again the responses of sand dollars bear more similarities to each other than to species from similar latitudes, such as F. zelandiae and P. regularis. Neither phylogeny, nor latitudinal origin is the clearly dominant factor in determining species response, showing the importance of determining adaptation to distinct local niche parameters, as emphasised by many other studies (Ettison and Shaw 2001; Sanford and Kelly 2011; Popova et al. 2016).

218

b) The deterministic influence of parental environment and physiological state on

offspring development in response to ocean acidification and warming

This study presents evidence of both adaptive and maladaptive trans-generational plasticity in response to ocean acidification and warming. In both O. validus and P. regularis, this plasticity is strongly dependent on maternal identity. The main finding of this study pertaining to trans-generational plasticity is that it exists alongside strongly heritable genetic components of trait variability in marine echinoderm populations, primarily through female genetic lines which can be key drivers of adaptation. Plasticity, in conjunction with ecology and evolution, should be an important consideration when identifying the ‘winners and losers’ of climate change (Webster et al. 2013). Others have shown that trans-generational plasticity is an important mechanism of adaptation to climate change, but it can only be adaptive when it is not costly to the parents (DeWitt et al. 1998; Auld et al. 2010; Scheiner et al. 2017). Our findings do not support the ‘bellwether’ theory (Fabry et al. 2008) that the responses of polar species can be used to predict the responses of related species from lower latitudes to climate change, based on the adaptive and acclamatory responses seen in this study.

It appears that Odontaster validus would be a good choice of model system for teasing out the effects of APEs (genetic) versus provisioning (non-genetic), as there was no evidence of compromised reproductive capacity among females in the F0 generation in response to combined low pH and high temperature. Among offspring of both the wild population and the population kept in the long-term experiment for 12 months, maternal identity explained large portions of variation among F1 genotypes, but a larger portion among the acclimated population. The difference in the size of maternal effect between the two offspring

219 populations is therefore due to an Antipicatory Maternal Effect and not differences in maternal provisioning to eggs. The increase in the variation explained by dam identity as well as the strong heritability and maternal effects (m2) seen among the offspring of the acclimated population indicate that a larger portion of the trait ‘normal development’ is heritable when the dam is exposed to environmental stressors. Natural selection, therefore, is capable of influencing population tolerance between two generations, and this is more strongly heritable through the maternal genetic line when the parents are fully acclimated to an environment which selects strongly for pH and temperature tolerance.

The adaptive plasticity seen in O. validus contrasts with the overall negative trans- generational effect seen in P. regularis. In Patiriella, it is harder to tease out what portion of offspring performance is due to maternal provisioning or condition-transfer effects and what portion is due to true APEs, as the reproductive capacity of the parent generation was strongly compromised by long-term exposure to ocean acidification and warming. Changes to parent condition which affect the next generation (a.k.a. condition transfer effects

Bonduriansky and Crean 2017) often have both genetic and epigenetic components and in

Patiriella, it seems as though the negative trans-generational carryover effects of parent exposure to ocean acidification are also a combination of multiple interacting epigenetic, environmental and genetic factors. Parent loss of homeostatic equilibrium is also known to be a source of maladaptive TGP (Ghalambor et al. 2007; Chevin et al. 2013). All of these factors can act as selective cues in the right environment, and the conditions which favour genetic or non-genetic inheritance may change depending on the strength of environmental selection

(English et al. 2015). The strong heritability estimates for the F1 offspring of acclimated parents indicates that a large portion of their poor performance is genetic, which indicates an

APE. The low normal development among F1 offspring of adults acclimated to pH 7.70 at

220 either ambient or elevated temperature, however, is probably due to a combination of poor parent physiological condition, reduction in offspring fitness and lack of provisioning.

This type of situation, in which parents reduce offspring fitness to maximise their own survival chances, has been called a ‘selfish’ APE (Marshall and Uller 2007) but that term may not fully apply in this case. The terms ‘fitness drag’ or ‘fitness lag,’ (as used by Chevin et al. 2010; Nunney et al. 2016; Scheiner et al. 2017) indicate that the current population mean genotype does not match the optimum population genotype for a particular niche, usually when environmental change is rapid. ‘Fitness drag’ may therefore be a better term to describe the mechanism underlying the maladaptive plasticity seen in P. regularis as it implies that the eventual conclusion of this situation will be that plasticity inhibits adaptive change, as seen in this study.

Both Odontaster and Patiriella share strong heritability of normal development and adaptive maternal effects, but have a lack of sire x environment interactions among genotypes from either wild of acclimated populations. Maternal effects are indeed potential sources of adaptation in response to climate change (Runcie et al. 2012; Jensen et al. 2014; Shama et al.

2014; Barbosa et al. 2015). On the other hand, sire contribution of genetic material to the next generation does not seem to confer any developmental advantage in either present-day conditions or the near-future conditions forecasted for the year 2100. Instead, parent environment, not parent genetic identity, appears to be the stronger selection cue at elevated temperature and low pH. Others have shown that the influence of parent environment on offspring performance increases when parents are from highly stressful environments

(Chirgwin et al. 2017). The effect of the offspring environment only appears to be a strong

221

‘selective context’ (sensu Uller et al. 2013) when the mismatch between parent and offspring environment releases the offspring from the stressors within the maternal environment. An example from this study is the increase in normal development among P. regularis embryos from parents acclimated to pH 7.70 when offspring were raised in the control conditions versus offspring also raised at pH 7.70. In the context of future ocean acidification and warming, this situation could occur if populations were able to take advantage of ‘climate change refugia’, which have been proposed to reduce biodiversity losses in the future

(Taberlet and Cheddadi 2002; Stewart et al. 2010; Tittensor et al. 2010).

c) Wider context of these findings and recommendations for further research

Our overarching research question was: do echinoderms adapt or acclimate to climate change

(ocean acidification and warming) physiologically and across generations? And the main answer is that some echinoderm populations do possess the capacity to adapt to climate change, both individually and across multiple generations. The studies presented here add to the landscape of literature which advances our knowledge of how echinoderm-dominated ecosystems respond to change.

Echinoderm-dominated habitats (generally defined as <10,000 individuals m-2), e.g. ‘urchin barrens’, are found in a wide range of environments, and are generally associated with the aftermath of mechanical destruction such as storm events, leading to reductions in diversity, changes to habitat usage, ecosystem collapse and destructive grazing, for example around temperate kelp forests (Ebeling et al. 1985; Gagnon et al. 2004; Filbee-Dexter and Scheibling

2014). Increasing global seawater temperatures may be aiding the spread of these ecosystems

(Ling et al. 2008; Hoegh-Guldberg and Bruno 2010), and the associated changes to habitat

222 usage (Johnson et al. 2011) may fundamentally change these communities (e.g. Harris and

Tyrell 2001). Many echinoderm-dominated ecosystems are model systems for climate-change research, for example, the California Current Large Marine Ecosystem (LME) or the Ischia undersea vent ecosystems (Uthicke et al. 2009, 2016; Klinger et al. 2017). This kind of underwater CO2 vent site is important as a ‘natural lab’ for studying the future effects of climate change on marine populations. Researchers in this area can now be more confident in assuming local adaptive capacity in echinoderm populations, which in the past has not been so clear (e.g. Calosi et al. 2013; Lamare et al. 2016).

In New Zealand, echinoderm-dominated assemblages are often seen in benthic communities disturbed by deep-sea trawling, such as the Chatham Islands orange roughy fishery (e.g.

Clark and Rowden 2009). Within the New Zealand Exclusive Economic Zone (EEZ) there are currently (as of December 2017) several proposed areas of deep-sea oil exploration, in areas such as Canterbury, Taranaki and Otago, and deep-sea mining prospects such as

Seafloor Massive Sulfides in the Kermadec Volcanic Arc. These activities are of course associated with benthic disturbance to sensitive habitats (Boschen et al. 2015). Companies interested in removing undersea resources will need to characterise resulting environmental disturbance, which often includes changes to echinoderm abundance. The findings of this study may be relevant to future environmental impact surveys, as echinoderms are sensitive to mechanical and pollution disturbances (Warwick and Clark 1993; Tuck et al. 1998) as well as ocean acidification and warming (Wittmann and Pörtner 2013), and therefore it will be necessary to separate out what changes to community composition are due to climate change, and what changes are due to anthropogenic disturbance.

223

The ‘bellwether’ hypothesis (Fabry et al. 2008) posits that as the effects of ocean acidification may be seen sooner at higher latitudes, so that polar seas may act as the bellwether for acidification-induced changes. This theory is only applicable to biological communities if the responses of polar species and their temperate and tropical counterparts are comparable in direction and magnitude. Some other authors have found that habitat-type is a stronger influence on echinoderm larval resilience to ocean acidification than phylogenetic relatedness or temperature despite very slight latitudinal trends (Byrne et al.

2013), and that in general, the responses of polar echinoderm larvae to ocean acidification and warming are not consistent with the responses of related temperate species over a range of both genetic and physiological parameters (Clark et al. 2009; Ericson et al. 2010, 2012;

Catarino et al. 2012). Here too we found that Antarctic marine echinoderms may be more thermally-resilient than their temperate counterparts, and that their responses are not sufficient to predict the responses of related non-polar species. This study is the first to investigate this issue using multi-generational exposures and a quantitative genetic approach together with heritability estimates. We therefore provide the first robust data undermining the bellwether hypothesis over multiple generations and long timescales. Researchers, policy- makers and conservationists working on Southern Ocean species vulnerabilities should be careful not to assume that the bellwether hypothesis can be applied indiscriminately.

Because of the intrinsic environment-specific nature of TGP, it is difficult to draw broad conclusions about the role of non-genetic inheritance in marine populations. TGP research in the field of ocean acidification and warming is still in its infancy, and there are many wide knowledge gaps including the mechanisms of TGP in marine populations, effects of parent environment stability, whether TGP can give rise to cryptic speciation in novel environments and how TGP can either accumulate or dissipate across more than two generations (Donelson et al. 2017). This study provides useful information to improve predictive models of smaller-

224 scale variability and resilience to future climate change. The next steps in this field of research could include examination of a broader suite of TGP drivers and traits, including how ontogeny may alter the effects of TGP. In this way, questions of adaptation versus acclimation across multiple generations may become a little clearer.

Transcriptomic and metabolomic data would complement the type of data collected in this study, but time and budget considerations prevented this. My recommendation is that future studies in this area should marry genetic and phenotypic approaches in order to get a more complete profile of the adaptive responses of key species. Studies in the past have tended to approach the problem of adaptive capacity from either a DNA-based or functional trait-based approach, leaving the rare meta-analytic studies to identify areas where the two studies overlap and where knowledge gaps exist.

8.3 Conclusions

 This is the first study to combine data about parent physiology and environment over

long time-scales with estimates of heritability and trans-generational plasticity in the

next generation in response to ocean acidification and warming.

 Adaptive capacity, defined by sire x environment interactions, in response to elevated

temperature and low pH is present in a subset of modern echinoderm populations

from subtropical and temperate latitudes.

 The polar sea star Odontaster validus displayed high physiological plasticity and pH-

tolerance, enabling both within- and trans-generational adaptive capacity to ocean

acidification and warming.

225

 The physiological and adaptive responses of the Antarctic sea star Odontaster validus

are not sufficient to predict the responses of the related temperate sea star Patiriella

regularis to ocean acidification and warming, suggesting that polar species are not

always useful as a “bellwether” for temperate ecosystems.

 The mechanisms of physiological acclimation to ocean acidification and warming

differ depending on the species examined. It is important to look at metabolic, bio-

energetic and behavioural indices when assessing whole-animal physiology in

response to temperature and pH stress. In this study, we applied this approach to

characterise within-generation metabolic acclimation potential in the sea stars

Patiriella regularis and Odontaster validus.

 Species-specific developmental mode and plasticity within regional environments are

important considerations to identify ‘winners and losers’ of climate change as broad

latitudinal and phylogenetic trends in adaptive capacity are weak, and interact with

adaptations to local environmental variation in ways which are difficult to separate

out.

 Trans-generational plasticity research in marine populations in response to climate

change is a developing field with some large knowledge gaps; more work is needed to

understand the role and importance of TGP across life stages and environments.

226

References

Agrawal AA, Ackerly DD, Adler F, Arnold AE, Cáceres C, Doak DF, Post E, Hudson PJ, Maron J, Mooney KA, Power M, Schemske D, Stachowicz J, Strauss S, Turner MG, Werner E (2007) Filling Key Gaps in Population and Community Ecology. Front. Ecol. Environ. 5:145–152

Agrawal AA (2001) Transgenerational Consequences of Plant Responses to Herbivory: An Adaptive Maternal Effect? Am Nat 157:555–569 doi: 10.1086/319932

Agüera A, Ahn I-Y, Guillaumot C, Danis B (2017) A Dynamic Energy Budget (DEB) model to describe Laternula elliptica (King, 1832) seasonal feeding and metabolism. PLoS One 12:e0183848 doi: 10.1371/journal.pone.0183848

Agüera A, Collard M, Jossart Q, Moreau C, Danis B (2015) Parameter Estimations of Dynamic Energy Budget (DEB) Model over the Life History of a Key Antarctic Species: The Antarctic Sea Star Odontaster validus (Koehler, 1906). PLoS One 10:e0140078 doi: 10.1371/journal.pone.0140078

Albentosa M, Vinas L, Besada V, Franco A, Gonzalez-Quijano A (2012) First measurements of the scope for growth (SFG) in mussels from a large-scale survey in the North Atlantic Spanish coast. Sci Total Environ 435:430–445

Albright R, Langdon C (2011) Ocean acidification impacts multiple early life history processes of the Caribbean coral Porites asteroides. Glob Chang Biol 17:2478–2487

Allan BJM, Miller GM, McCormick MI, Domenici P, Munday PL (2014) Parental effects improve escape performance of juvenile reef fish in a high-CO2 world. Proc R Soc Part B Biol Sci 281:20132179 doi: 10.1098/rspb.2013.2179

Alqaisi K, Lamare MD, Grattan D, Damsteegt E, Schneider W, Lokman P (2016) A comparative study of vitellogenesis in Echinodermata: Lessons from the sea star. Comp Biochem Physiol - A Mol Integr Physiol 198:72–86

Angulo-Preckler C, Tuya F, Avila C (2017) Abundance and size patterns of echinoderms in coastal soft-bottoms at Deception Island (South Shetland Islands, Antarctica). Cont Shelf Res 137:131–141 doi: 10.1016/J.CSR.2016.12.010

Anthony KRN, Maynard JA, Diaz-Pulido G, Mumby PJ, Marshall PA, Cao L, Hoegh- Guldberg O (2011) Ocean acidification and warming will lower coral reef resilience. Glob Chang Biol 17:1798–1808 doi: 10.1111/j.1365-2486.2010.02364.x

Appelhans YS, Thomsen J, Opitz S, Pansch C, Melzner F, Wahl M (2014) Juvenile sea stars exposed to acidification decrease feeding and growth with no acclimation potential. Mar Ecol Prog Ser 509:227–239 doi: 10.3354/meps10884

Armstrong EJ, Allen TR, Beltrand M, Dubousquet V, Stillman JH, Mills SC (2017) High pCO2 and elevated temperature reduce survival and alter development in early life

227

stages of the tropical sea hare Stylocheilus striatus. Mar Biol 164:107 doi: 10.1007/s00227-017-3133-x

Arnberg M, Calosi P, Spicer JI, Tandberg A, Nilsen M, Westerlund S, Bechmann RK (2013) Elevated temperature elicits greater effects than decreased pH on the development, feeding and metabolism of northern shrimp (Pandalus borealis) larvae. Mar Biol 160:2037–2048 doi: 10.1007/s00227-012-2072-9

Auld JR, Agrawal AA, Relyea RA (2010) Re-evaluating the costs and limits of adaptive phenotypic plasticity. Proceedings Biol Sci 277:503–11 doi: 10.1098/rspb.2009.1355

Babcock R, Mundy C (1992) Reproductive biology, spawning and field fertilization rates of Acanthaster planci. Mar Freshw Res 43:525 doi: 10.1071/MF9920525

Bacher C, Gangnery A (2006) Use of dynamic energy budget and individual based models to simulate the dynamics of cultivated oyster populations. J Sea Res 56:140–155 doi: 10.1016/j.seares.2006.03.004

Bakun A, Black BA, Bograd SJ, García-Reyes M, Miller AJ, Rykaczewski RR, Sydeman WJ (2015) Anticipated Effects of Climate Change on Coastal Upwelling Ecosystems. Curr Clim Chang Reports 1:85–93 doi: 10.1007/s40641-015-0008-4

Barbosa M, Lopes I, Venâncio C, Janeiro MJ, Morrisey MB, Soares AMVM (2015) Maternal response to environmental unpredictability. Ecol Evol 5:4567–4577 doi: 10.1002/ece3.1723

Barker MF, Xu RA (1991) Population differences in gonad and pyloric caeca cycles of the New Zealand seastar Sclerasterias mollis (Echinodermata: Asteroidea). Mar Biol 108:97–103 doi: 10.1007/BF01313476

Barker M, Russell M (2008) The distribution and behaviour of Patiriella mortenseni and P. regularis in the extreme hyposaline conditions of the Southern New Zealand Fiords. J Exp Mar Bio Ecol 355:76–84

Bates A, Pecl G, Frusher S, Hobday A, Wernberg T, Smale D, Sunday J, Hill N, Dulvy N, Colwell R, Holbrook N, Fulton E, Slawinkski D, Feng M, Edgar G, Radford B, Thompson P, Watson R (2014) Defining and observing stages of climate-mediated range shifts in marine systems. Glob Environ Chang 26:27–38 doi: 10.1016/J.GLOENVCHA.2014.03.009

Bates NR, Astor YM, Church MJ, Currie K, Dore JE, Gonzalez-Davila M, Lorenzoni L, Muller-Karger F, Olafsson J, Santana-Casiano JM (2014) A Time-Series View of Changing Surface Ocean Chemistry Due to Ocean Uptake of Anthropogenic CO₂ and Ocean Acidification. Oceanography 27:126–141 doi: 10.2307/24862128

Bayne B, Holland D, Moore M, Lowe D, Widdows J (1978) Further studies on the effects of stress in the adult on the eggs of Mytilus edulis. J Mar Biol Assoc United Kingdom 58:825–841

228

Bayne B, Widdows J (1978) The physiological ecology of two populations of Mytilus edulis (L.). Oecologia 37:137–162

Bellwood DR, Hughes TP, Folke C, Nyström M (2004) Confronting the coral reef crisis. Nature 429:827–833 doi: 10.1038/nature02691

Belman BW, Giese AC (1974) Oxygen consumption of an Asteroid and an Echinoid from the Antarctic. Biol Bull 146:157–164 doi: 10.2307/1540614

Bernhardt J, Leslie H (2013) Resilience to climate change in coastal marine ecosystems. Ann Rev Mar Sci 5:371–392

Bertram D, Strathmann R (1998) Effects of maternal and larval nutrition on growth and form of planktotrophic larvae. Ecology 79:315–327

Bingham BL, Bacigalupi M, Johnson LG (1997) Temperature Adaptations of Embryos from Intertidal and Subtidal Sand Dollars (Dendraster excentricus, Eschscholtz). Northwest Sci 71:108–114

Bonduriansky R, Crean AJ (2017) What are parental condition-transfer effects and how can they be detected? Methods Ecol Evol. doi: 10.1111/2041-210X.12848

Bonduriansky R, Crean AJ, Day T (2012) The implications of nongenetic inheritance for evolution in changing environments. Evol Appl 5:192–201 doi: 10.1111/j.1752- 4571.2011.00213.x

Bonduriansky R, Day T (2009) Nongenetic Inheritance and Its Evolutionary Implications. Annu Rev Ecol Evol Syst 40:103–25 doi: 10.1146/annurev.ecolsys.39.110707.173441

Böning CW, Dispert A, Visbeck M, Rintoul SR, Schwarzkopf FU (2008) The response of the Antarctic Circumpolar Current to recent climate change. Nat Geosci 1:864–869 doi: 10.1038/ngeo362

Boschen R, Rowden A, Clark M, Barton S, Pallentin A, Gardner J (2015) Megabenthic assemblage structure on three New Zealand seamounts: implications for seafloor massive sulfide mining. Mar Ecol Prog Ser 523:1–14

Bowen M, Markham J, Sutton P, Zhang X, Wu Q, Shears NT, Fernandez D, Bowen M, Markham J, Sutton P, Zhang X, Wu Q, Shears NT, Fernandez D (2017) Interannual Variability of Sea Surface Temperature in the Southwest Pacific and the Role of Ocean Dynamics. J Clim 30:7481–7492 doi: 10.1175/JCLI-D-16-0852.1

Brierley AnS, Kingsford MJ (2009) Impacts of Climate Change on Marine Organisms and Ecosystems. Curr Biol 19:R602–R614 doi: 10.1016/J.CUB.2009.05.046

Brockington S, Clarke A (2001) The relative influence of temperature and food on the metabolism of a marine invertebrate. J Exp Mar Bio Ecol 258:87–99 doi: 10.1016/S0022-0981(00)00347-6

229

Brockington S, Peck LS (2001) Seasonality of respiration and ammonium excretion in the Antarctic echinoid Sterechinus neumayeri. Mar Ecol Prog Ser Mar Ecol Prog Ser Mar Ecol Prog Ser 219:159–168

Brodie J, Fabricius K, De’ath G, Okaji K (2005) Are increased nutrient inputs responsible for more outbreaks of crown-of-thorns starfish? An appraisal of the evidence. Mar Pollut Bull 51:266–278 doi: 10.1016/J.MARPOLBUL.2004.10.035

Brothers C, Mcclintock JB (2015) The effects of climate-induced elevated seawater temperature on the covering behaviour, righting response, and Aristotle’s lantern reflex of the sea urchin Lytechinus variegatus. J Exp Mar Bio Ecol 467:33–38

Burger R, Lynch M (1997) Adaptation and extinction in changing environments. In: Bijlsma K, Loeschcke V (eds.) Environmental Stress, Adaptation and Evolution. Birkhauser, Basel, Switzerland, pp 209–239

Bürger R (1999) Evolution of Genetic Variability and the Advantage of Sex and Recombination in Changing Environments. Genetics 153:1055–1069

Burgess SC, Marshall DJ (2014) Adaptive parental effects: The importance of estimating environmental predictability and offspring fitness appropriately. Oikos. doi: 10.1111/oik.01235

Burton T, Metcalfe NB (2014) Can environmental conditions experienced in early life influence future generations? Proc R Soc Part B Biol Sci 281:20140311 doi: 10.1098/rspb.2014.0311

Busby PE, Newcombe G, Dirzo R, Whitham TG (2014) Differentiating genetic and environmental drivers of plant-pathogen community interactions. J Ecol 102:1300–1309 doi: 10.1111/1365-2745.12270

Bylenga CH, Cummings VJ, Ryan KG (2017) High resolution microscopy reveals significant impacts of ocean acidification and warming on larval shell development in Laternula elliptica. PLoS One 12:e0175706 doi: 10.1371/journal.pone.0175706

Byrne M (2010) Impact of climate change stressors on marine invertebrate life histories with a focus on the Mollusca and Echinodermata. In: You Y, Henderson-Sellers A (eds.) Climate alert: climate change monitoring and strategy. Sydney University Press, Sydney, pp 142–185

Byrne M (2011) Impact of ocean warming and ocean acidification on marine invertebrate life history stages: vulnerabilities and potential for persistence in a changing ocean. Oceanogr Mar Biol An Annu Rev 49:1–42 doi: 10.1016/j.marenvres.2011.10.00

Byrne M (2012) Global change ecotoxicology: Identification of early life history bottlenecks in marine invertebrates, variable species responses and variable experimental approaches. Mar Environ Res 76:3–15 doi: 10.1016/j.marenvres.2011.10.004

Byrne M, Prowse TAA, Sewell MA, Dworjanyn S, Williamson JE, Vaïtilingon D (2008) Maternal provisioning for larvae and larval provisioning for juveniles in the

230

toxopneustid sea urchin Tripneustes gratilla. Mar Biol 155:473–482 doi: 10.1007/s00227-008-1045-5

Byrne M, Przeslawski R (2013) Multistressor Impacts of Warming and Acidification of the Ocean on Marine Invertebrates’ Life Histories. Integr Comp Biol 53:582–596 doi: 10.1093/icb/ict049

Byrne M, Barker M (1991) Embryogenesis and larval development of the Asteroid Patiriella regularis viewed by light and Scanning Electron Microscopy. Biol Bull 180:332–345

Byrne M, Gonzalez-Bernat MJ, Foo S, Soars N, Lamare M (2013) Effects of ocean warming and acidification on embryos and non-calcifying larvae of the invasive sea star Patiriella regularis. Mar Ecol Prog Ser 473:235–246

Byrne M, Selvakumaraswamy P, Ho M, Woolsey E, Nguyen H (2011) Sea urchin development in a global change hotspot, potential for southerly migration of thermotolerant propagules. Deep Sea Res Part II Top Stud Oceanogr 58:712–719 doi: 10.1016/J.DSR2.2010.06.010

Byrne M, Villinski J, Cisternas P, Siegel R, Popdi E, Raff R (1999) Maternal factors and the evolution of developmental mode: Evolution of oogenesis in Heliocidaris erythrogramma. Dev Genes Evol 209:275–283

Byrne M, Gall M, Wolfe K, Agüera A (2016) From pole to pole: the potential for the Arctic seastar Asterias amurensis to invade a warming Southern Ocean. Glob Chang Biol 22:3874–3887 doi: 10.1111/gcb.13304

Byrne M, Ho M, Selvakumaraswamy P, Nguyen HD, Dworjanyn SA, Davis AR (2009) Temperature, but not pH, compromises sea urchin fertilization and early development under near-future climate change scenarios. Proc R Soc Part B Biol Sci 276: doi: 10.1098/rspb.2008.1935

Byrne M, Ho M, Wong E, Soars NA, Selvakumaraswamy P, Shepard-Brennand H, Dworjanyn SA, Davis AR (2011) Unshelled abalone and corrupted urchins: development of marine calcifiers in a changing ocean. Proc R Soc B Biol Sci 278:2376– 83 doi: 10.1098/rspb.2010.2404

Byrne M, Lamare M, Winter D, Dworjanyn S, Uthicke S (2013) The stunting effect of a high CO2 ocean on calcification and development in sea urchin larvae, a synthesis from the tropics to the poles. Philos Trans R Soc London B 368

Byrne M, Soars NA, Ho MA, Wong E, McElroy D, Selvakumaraswamy P, Dworjanyn SA, Davis AR (2010) Fertilization in a suite of coastal marine invertebrates from SE Australia is robust to near-future ocean warming and acidification. Mar Biol 157:2061– 2069

Caballes C, Pratchett M, Buck A (2017) Interactive effects of endogenous and exogenous nutrition on larval development for crown-of-thorns starfish. Diversity 9 doi: 10.3390/d9010015

231

Caballes CF, Pratchett MS, Kerr AM, Rivera-Posada JA (2016) The role of maternal nutrition on oocyte size and quality, with respect to early larval development in the coral-eating starfish, Acanthaster planci. PLoS One 11:1–21 doi: 10.1371/journal.pone.0158007

Cai W, Shi G, Cowan T, Bi D, Ribbe J (2005) The response of the Southern Annular Mode, the East Australian Current, and the southern mid-latitude ocean circulation to global warming. Geophys Res Lett 32:L23706

Caldeira K, Wickett M (2005) Ocean model predictions of chemistry changes from carbon dioxide emissions to the atmosphere and ocean. J Geophys Res 110:C09S04

Calosi P, Rastrick SPS, Graziano M, Thomas S, Baggini C, Carter C, Hall-Spencer J, Milazzo M, Spicer J (2013) Distribution of sea urchins living near shallow water CO2 vents is dependent upon species acid-base and ion-regulatory abilities. Mar Pollut Bull 30:470–484

Calosi P, De Wit P, Thor P, Dupont S (2016) Will life find a way? Evolution of marine species under global change. Evol Appl 9:1035–1042 doi: 10.1111/eva.12418

Calosi P, Rastrick SPS, Lombardi C, De Guzman HJ, Davidson L, Jahnke M, Giangrande A, Hardege JD, Schulze A, Spicer JI, Gambi M-C (2013) Adaptation and acclimatization to ocean acidification in marine ectotherms: an in situ transplant experiment with polychaetes at a shallow CO2 vent system. Philos Trans R Soc Lond B Biol Sci 368:20120444 doi: 10.1098/rstb.2012.0444

Campbell A, Levitan D, Hosken D, Lewis C (2016) Ocean acidification changes the male fitness landscape. Sci Rep 6:31250

Carey N, Harianto J, Byrne M (2016) Sea urchins in a high-CO2 world: partitioned effects of body size, ocean warming and acidification on metabolic rate. J Exp Biol 219:1178–86 doi: 10.1242/jeb.136101

Catarino A, Bauwens M, Dubois P (2012) Acid-base and metabolic response of the sea urchin Paracentrorus lividus to different seawater pH and temperature. Environ Sci Pollut 19:2344–2353

Catarino A, De Ridder C, Gonzalez M, Gallardo P, Dubois P (2012) Sea urchin Arbacia dufresnei (Blainvill 1825) larvae response to ocean acidification. Polar Biol 35:455–461

Chakravarti LJ, Jarrold MD, Gibbin EM, Christen F, Massamba-N’Siala G, Blier PU, Calosi P (2016) Can trans-generational experiments be used to enhance species resilience to ocean warming and acidification? Evol Appl 9:1133–1146 doi: 10.1111/eva.12391

Challener RC, McClintock JB (2013) Exposure to extreme hypercapnia under laboratory conditions does not impact righting and covering behavior of juveniles of the common sea urchin Lytechinus variegatus. Mar Freshw Behav Physiol 46:191–199 doi: 10.1080/10236244.2013.800759

232

Chan K, Grunbaun D, O’Donnell M, Grünbaum D, O’Donnell M (2011) Effects of ocean- acidification-induced morphological changes on larval swimming and feeding. J Exp Biol 214:3857–3867 doi: 10.1242/jeb.054809

Chapelle G, Peck L, Clarke A (1994) Effects of feeding and starvation on the metabolic rate of the necrophagous Antarctic amphipod Waldeckia obesa (Chevreux, 1905). J Exp Mar Bio Ecol 183:63–76 doi: 10.1016/0022-0981(94)90157-0

Charmantier A, McCleery RH, Cole LR, Perrins C, Kruuk LEB, Sheldon BC (2008) Adaptive phenotypic plasticity in response to climate change in a wild bird population. Science 320:800–3 doi: 10.1126/science.1157174

Chen B, Chen C (1992) Reproductive cycle, larval development, juvenile growth and population dynamics of Patiriella pseudoexigua (Echinodermata: Asteroidea) in Taiwan. Mar Biol 113:271–280

Chen C, Chen B (1992) Effects of high temperature on larval development and metamorphosis of Arachnoides placenta (Echinodermata: Echinoidea). Mar Biol 112:445–449

Cheung WWL, Lam VWY, Sarmiento JL, Kearney K, Watson R, Pauly D (2009) Projecting global marine biodiversity impacts under climate change scenarios. Fish Fish 10:235– 251 doi: 10.1111/j.1467-2979.2008.00315.x

Chevin LM, Collins S, Lefèvre F (2013) Phenotypic plasticity and evolutionary demographic responses to climate change: Taking theory out to the field. Funct Ecol. doi: 10.1111/j.1365-2435.2012.02043.x

Chevin LM, Lande R, Mace GM (2010) Adaptation, plasticity, and extinction in a changing environment: Towards a predictive theory. PLoS Biol. doi: 10.1371/journal.pbio.1000357

Chevin L-M, Lande R, Mace GM (2010) Adaptation, Plasticity, and Extinction in a Changing Environment: Towards a Predictive Theory. PLos ONE Biol e1000357 doi: 10.1371/journal.pbio.1000357

Chia F (1968) Some observations on the developmental and cyclic changes of the oocytes in a brooding starfish, Leptasterias hexactis. J Zool Soc London 154:453–461

Chirgwin E, Marshall DJ, Sgrò CM, Monro K (2017) The other 96%: Can neglected sources of fitness variation offer new insights into adaptation to global change? Evol Appl 10:267–275 doi: 10.1111/eva.12447

Chirgwin E, Monro K, Sgrò CM, Marshall DJ (2015) Revealing hidden evolutionary capacity to cope with global change. Glob Chang Biol. doi: 10.1111/gcb.12929

Christensen J (2007) Regional climate projections. In: Solomon S (ed.) Climate Change 2007: The Physical Science Basis. Cambridge University Press, Cambridge, UK, pp 849–940

233

Clark D, Lamare M, Barker M (2009) Response of sea urchin pluteus larvae (Echinodermata: Echinoidea) to reduced seawater pH: a comparison among a tropical, temperate and a polar species. Mar Biol 156:1125–1137

Clark JS, Poore AGB, Ralph PJ, Doblin MA (2013) Potential for adaptation in response to thermal stress in an intertidal macroalga. J Phycol 49:630–639 doi: 10.1111/jpy.12067

Clark K, Gorley R (2006) PRIMER v6: user manual/tutorial

Clark M, Rowden A (2009) Effect of deepwater trawling on the macro-invertebrate assemblages of seamounts on the Chatham Rise, New Zealand. Deep Sea Res Part I Oceanogr Res Pap 56:1540–1554

Clarke A (1991) What is cold adaptation and how should we measure it? Am Zool 31:81–92

Clarke A (1990) Temperature and Evolution: Southern Ocean cooling and the Antarctic marine fauna. In: Kerry KR, Hempel G (eds.) Antarctic Ecosystems: Ecological Change and Conservation. Springer Berlin Heidelberg, Berlin, Heidelberg, pp 9–22

Clarke A (1988) Seasonality in the antarctic marine environment. Comp Biochem Physiol Part B Comp Biochem 90:461–473 doi: 10.1016/0305-0491(88)90285-4

Clarke A (1983) Life in cold water: the physiological ecology of polar marine ectotherms. In: Oceanography and Marine Biology: An Annual Review. Aberdeen University Press, Aberdeen, pp 341–453

Clarke A (1980) A reappraisal of the concept of metabolic cold adaptation in polar marine invertebrates. Biol J Linn Soc 14:77–92

Clarke A, Murphy EJ, Meredith MP, King JC, Peck LS, Barnes DKA, Smith RC (2007) Climate change and the marine ecosystem of the western Antarctic Peninsula. Philos Trans R Soc Lond B Biol Sci 362:149–66 doi: 10.1098/RSTB.2006.1958

Collard M, Laitat K, Moulin L, Catarino A, Grosjean P, Dubois P (2013) Buffer capacity of the coelomic fluid in echinoderms. Comp Biochem Physiol Part A Mol Integr Physiol 166:199–206

Collins S, Rost B, Rynearson TA (2014) Evolutionary potential of marine phytoplankton under ocean acidification. Evol Appl 7:140–155 doi: 10.1111/eva.12120

Comeau S, Gorsky G, Alliouane S, Gattuso J-P (2010) Larvae of the pteropod Cavolinia inflexa exposed to aragonite undersaturation are viable but shell-less. Mar Biol 157:2341–2345 doi: 10.1007/s00227-010-1493-6

Conover R, Corner E (1968) Respiration and nitrogen excretion by some marine zooplankton in relation to their life cycles. J Mar Biol Assoc United Kingdom 48:49–75

Constable A, Melbourne-Thomas J, Corney S, Arrigo K, Barbraud C, Barnes D, Bindoff N, Boyd P, Brandt A, Coasta D, Davidson A, Ducklow H, Emmerson L, Fukuchi M, Gutt J, Hindell MA, Hofmann E, Hosie G, Iida T, Jacob S, Johnston N, Kawaguchi S, Kokubun

234

N, Koubbi P, Lea M, Makhado A, Massom R, Meiners K, Meredith M, Murphy E, Nicol S, Reid S, Richerson K, Riddle M, Rintoul S, Smith Jr W, Southwell C, Stark J, Sumner M, Swadling K, Takahashi K, Trathan P, Welsford D, Weimerskirch H, Westwood K, Weinecke B, Wolf-Gladrow D, Wright S, Xavier J, Ziegler P (2014) Climate change and Southern Ocean ecosystems I: how changes in physical habitats directly affect marine biota. Glob Chang Biol 6:2–23 doi: 10.1111/gcb.12623

Constable A, Meredith M, Ducklow H, Murphy E, Linse K, Kawaguchi S (2016) Impacts and effects of ocean warming on Antarctic ecosystems and species. In: Laffoley D, Baxter JM (eds,) IUCN Global marine and polar programme: Explaining ocean warming: Causes, scale, effects and consequences. IUCN, Switzerland, pp 239–255

Convey P, Scott D, Fraser WR (2003) Biophysical and habitat changes in response to climate alteration in the Arctic and Antarctic. Adv Appl Biodivers Sci 77–82

Cornwall CE, Hurd CL (2016) Experimental design in ocean acidification research: problems and solutions. ICES J Mar Sci 73:572–581

Cox PM, Betts RA, Jones CD, Spall SA, Totterdell IJ (2000) Acceleration of global warming due to carbon-cycle feedbacks in a coupled climate model. Nature 408:184–187

Crain CM, Kroeker K, Halpern BS (2008) Interactive and cumulative effects of multiple human stressors in marine systems. Ecol Lett 11:1304–1315 doi: 10.1111/j.1461- 0248.2008.01253.x

Crain W, Bushman F (1983) Transcripts of paternal and maternal actin gene alleles are present in interspecific sea urchin embyro hybrids. Dev Biol 100:190–196 doi: 10.1016/0012-1606(83)90210-5

Crean A, Bonduriansky R (2014) What is a paternal effect? Trends Ecol Evol 29:554–559 doi: 10.1016/J.TREE.2014.07.009

Crean AJ, Dwyer JM, Marshall DJ (2013) Adaptive paternal effects? Experimental evidence that the paternal environment affects offspring performance. Ecology 94:2575–2582 doi: 10.1890/13-0184.1

Crean AJ, Dwyer JM, Marshall DJ (2012) Fertilization is not a new beginning: The relationship between sperm longevity and offspring performance. PLoS One 7 doi: 10.1371/journal.pone.0049167

Crozier LG, Hutchings JA (2014) Plastic and evolutionary responses to climate change in fish. Evol. Appl.

Crump R (971) Annual reproductive patterns in three geographically separated populations of Patiriella regularis (Verrill), a common New Zealand Asteroid. J Exp Mar Biol Ecol 7:137–162

Cruz J, Garrido S, Pimentel M, Rosa R, Santos A, Re P (2013) Reproduction and respiration of a climate change indicator species: effect of temperature and variable food in the copepod Centropages chierchiae. J Plankton Res 35:1046–1064

235

Dahlhoff EP, Fearnley SL, Bruce DA, Gibbs AG, Stoneking R, McMillan DM, Deiner K, Smiley JT, Rank NE (2008) Effects of temperature on physiology and reproductive success of a montane leaf beetle: implications for persistence of native populations enduring climate change. Physiol Biochem Zool 81:718–32 doi: 10.1086/590165

Dam HG (2013) Evolutionary adaptation of marine zooplankton to global change. Annu Rev Mar Sci 5:349–70 doi: 10.1146/annurev-marine-121211-172229

Davis MB, Shaw RG, Etterson JR (2005) Evolutionary responses to changing climate. Ecology 86:1704–1714 doi: 10.2307/3450613

Dayton P, Robilliard G, Paine R, Dayton L (1974) Biological accommodation in the benthic community at McMurdo Sound, Antarctica. Ecol Monogr 44:105–128

De Wit P, Dupont S, Thor P (2016) Selection on oxidative phosphorylation and ribosomal structure as a multigenerational response to ocean acidification in the common copepod Pseudocalanus acuspes. Evol Appl. doi: 10.1111/eva.12335

De’ath G, Fabricius KE, Sweatman H, Puotinen M (2012) The 27-year decline of coral cover on the Great Barrier Reef and its causes. Proc Natl Acad Sci U S A 109:17995–9 doi: 10.1073/pnas.1208909109

Delorme NJ, Sewell MA (2016) Genotype-by-environment interactions during early development of the sea urchin Evechinus chloroticus. Mar Biol 163:215 doi: 10.1007/s00227-016-2987-7

Deutsch C, Ferrel A, Seibel B, Pörtner H-O, Huey RB (2015) Climate change tightens a metabolic constraint on marine habitats. Science (80- ) 348:1132–5 doi: 10.1126/science.aaa1605

DeWitt T, Siha A, Wilson D (1998) Costs and limits of phenotypic plasticity. Trends Ecol Evol 13:77–81 doi: 10.1016/S0169-5347(97)01274-3

Diaz-Guisado D, Gaymer C, Brokordt K, Lawrence J (2006) Autotomy reduces feeding, energy storage and growth of the sea star Stichaster striatus. J Exp Mar Bio Ecol 338:73–80 doi: 10.1016/J.JEMBE.2006.06.037

Dickson A, Millero FJ (1987) A comparison of the equilibrium constants for the dissociation of carbonic acid in seawater media. Deep Sea Res 34:1733–1743

Dickson A, Sabine C, Christian J (2007) Guide to best practices for ocean CO2 measurements. North Pacific Marine Sciences Organization, Sidney, British Columbia

Donelson JM, Salinas S, Munday PL, Shama LNS (2017) Transgenerational plasticity and climate change experiments: Where do we go from here? Glob Chang Biol. doi: 10.1111/gcb.13903

Draghi JA, Whitlock MC (2012) Phenotypic plasticity facilitates mutational variance, genetic variance, and evolvability along the major axis of environmental variation. Evolution (N Y). doi: 10.1111/j.1558-5646.2012.01649.x

236

Duarte C, Navarro J, Acuna K, Torres R, Manriquez P, Lardies M, Vargas C, Lagos N, Aguilera V (2014) Combined effects of temperature and ocean acidification on the juvenile individuals of the mussel Mytilus chilensis. J Sea Res 85:308–314 doi: 10.1016/J.SEARES.2013.06.002

Duarte CM, Hendriks IE, Moore TS, Olsen YS, Steckbauer A, Ramajo L, Carstensen J, Trotter JA, McCulloch M (2013) Is ocean acidification an open-ocean syndrome? Understanding anthropogenic impacts on seawater pH. Estuaries and Coasts 36:221–236 doi: 10.1007/s12237-013-9594-3

Dulvy NK, Freckleton RP, Polunin NVC (2004) Coral reef cascades and the indirect effects of predator removal by exploitation. Ecol Lett 7:410–416 doi: 10.1111/j.1461- 0248.2004.00593.x

Dunbar MJ (1960) The Evolution of Stability in Marine Environments Natural Selection at the Level of the. Am Nat 94:129–136

Dupont S, Ortega-Martínez O, Thorndyke M (2010) Impact of near-future ocean acidification on echinoderms. Ecotoxicology 19:449–462 doi: 10.1007/s10646-010-0463-6

Dupont S, Dorey N, Stumpp M, Melzner F, Thorndyke M (2013) Long-term and trans-life- cycle effects of exposure to ocean acidification in the green sea urchin Strongylocentrotus droebachiensis. Mar Biol 160:1835–1843

Dupont S, Thorndyke M (2012) Relationship between CO2 – driven changes in extracellular acid-base balance and cellular immune response in two polar echinoderm species. J Exp Mar Bio Ecol 424:32–37

Ebeling AW, Laur DR, Rowley RJ (1985) Severe storm disturbances and reversal of community structure in a southern California kelp forest. Mar Biol 84:287–294 doi: 10.1007/BF00392498

Edmunds PJ, Brown D, Moriarty V (2012) Interactive effects of ocean acidification and temperature on two scleractinian corals from Moorea, French Polynesia. Glob Chang Biol 18:2173–2183 doi: 10.1111/j.1365-2486.2012.02695.x

Eisen EJ, Saxton A (1983) Genotype by environment interactions and genetic correlations involving two environmental factors. Theor Appl Genet 67:75–86 doi: 10.1007/BF00303929

English S, Pen I, Shea N, Uller T (2015) The information value of non-genetic inheritance in plants and animals. PLoS One 10:e0116996 doi: 10.1371/journal.pone.0116996

Engqvist L, Reinhold K (2016) Adaptive trans-generational phenotypic plasticity and the lack of an experimental control in reciprocal match/mismatch experiments. Methods Ecol Evol 7:1482–1488 doi: 10.1111/2041-210X.12618

Ericson J, Ho M, Miskelly A, King C, Virtue P, Tilbrook B, Byrne M (2012) Combined effects of two ocean change stressors, warming and acidification, on fertilization and

237

early development of the Antarctic echinoid Sterechinus neumayeri. Polar Biol 35:1027–1034

Ericson J, Lamare M, Morley S, Barker M (2010) The responses of two ecologically important Antarctic invertebrates (Sterechinus neumayeri and Parborlasia corrogatus) to reduced seawater pH: effects on fertilisation and embryonic development. Mar Biol 157:2689–2702

Etterson JR, Shaw RG (2001) Constraint to adaptive evolution in response to global warming. Science 294:151–4 doi: 10.1126/science.1063656

Evans TG, Padilla-Gamiño JL, Kelly MW, Pespeni MH, Chan F, Menge BA, Gaylord B, Hill TM, Russell AD, Palumbi SR, Sanford E, Hofmann GE (2015) Ocean acidification research in the “post-genomic” era: Roadmaps from the purple sea urchin Strongylocentrotus purpuratus. Comp Biochem Physiol Part A Mol Integr Physiol 185:33–42 doi: 10.1016/j.cbpa.2015.03.007

Fabricius KE, De’ath G, Noonan S, Uthicke S (2014) Ecological effects of ocean acidification and habitat complexity on reef-associated macroinvertebrate communities. Proc R Soc Part B Biol Sci 281:20132479 doi: 10.1098/rspb.2013.2479

Fabricius KE, Okaji K, De’ath G (2010) Three lines of evidence to link outbreaks of the crown-of-thorns seastar Acanthaster planci to the release of larval food limitation. Coral Reefs 29:593–605 doi: 10.1007/s00338-010-0628-z

Fabry VJ, Seibel BA, Feely RA, Orr JC (2008) Impacts of ocean acidification on marine fauna and ecosystem processes. ICES J Mar Sci 65:414–432 doi: 10.1093/icesjms/fsn048

Fabry V, McClintock J, Mathis J, Grebmeier J (2009) Ocean acidification at high latitudes: The Bellwether. Oceanography 22:160–171 doi: 10.5670/oceanog.2009.105

Falfushynska HI, Phan T, Sokolova IM (2016) Long-Term Acclimation to Different Thermal Regimes Affects Molecular Responses to Heat Stress in a Freshwater Clam Corbicula Fluminea. Sci Rep 6:39476 doi: 10.1038/srep39476

Falkner I, Byrne M, Sewell MA (2006) Maternal provisioning in Ophionereis fasciata and O. schayeri: brittle stars with contrasting modes of development. Biol Bull 211:204–7 doi: 10.2307/4134542

Fang J, Schonberg, CHLMello-Athayade M, Hoegh-Guldberg O, Dove S (2014) Effects of ocean warming and acidification on the energy budget of an excavating sponge. Glob Chang Biol 20:1043–1054 doi: 10.1111/gcb.12369

Feely RA, Sabine CL, Lee K, Berelson W, Kleypas J, Fabry VJ, Millero FJ (2004) Impact of anthropogenic CO2 on the CaCO3 system in the oceans. Science (80-81) 305:362

Feely RA, Sabine C, Hernandez-Ayon J, Ianson D, Hales B (2008) Evidence for upwelling of corrosive “acidified” water onto the continental shelf. Science (80-81) 320:

238

Feil R, Fraga MF (2012) Epigenetics and the environment: emerging patterns and implications. Nat Rev Genet 13:97–109 doi: 10.1038/nrg3142

Fenwick GD, Horning DS, Horning Jr DS (1980) Echinodermata of The Snares islands, southern New Zealand. New Zeal J Mar Freshw Res 144:437–445 doi: 10.1080/00288330.1980.9515888

Filbee-Dexter K, Scheibling RE (2014) Sea urchin barrens as alternative stable states of collapsed kelp ecosystems. Mar Ecol Prog Ser 495:1–25 doi: 10.3354/meps10573

Fitzer SC, Caldwell GS, Close AJ, Clare AS, Upstill-Goddard RC, Bentley MG (2012) Ocean acidification induces multi-generational decline in copepod naupliar production with possible conflict for reproductive resource allocation. J Exp Mar Bio Ecol 418–419:30– 36 doi: 10.1016/j.jembe.2012.03.009

Fitzpatrick B (2012) Underappreciated consequences of phenotypic plasticity for ecological speciation. Int J Ecol. doi: 10.1155/2012/256017

Fletcher SEM, Gruber N, Jacobson AR, Doney SC, Dutkiewicz S, Gerber M, Follows M, Joos F, Lindsay K, Menemenlis D, Mouchet A, Müller SA, Sarmiento JL (2006) Inverse estimates of anthropogenic CO2 uptake, transport, and storage by the ocean. Global Biogeochem Cycles 20:GB2002 doi: 10.1029/2005GB002530

Flynn EE, Todgham AE (2017) Thermal windows and metabolic performance curves in a developing Antarctic fish. J Comp Physiol B 1–12 doi: 10.1007/s00360-017-1124-3

Foo SA, Byrne M (2016) Acclimatization and Adaptive Capacity of Marine Species in a Changing Ocean. In: Advances in Marine Biology 74: 69-116

Foo SA (2015) Acclimatisation and adaptive capacity of sea urchins in a changing ocean: Effects of ocean warming and acidification on early development and the potential to persist. Doctoral Thesis, University of Sydney.

Foo SA, Dworjanyn SA, Khatkar MS, Poore AGB, Byrne M (2014) Increased temperature, but not acidification, enhances fertilization and development in a tropical urchin: Potential for adaptation to a tropicalized eastern Australia. Evol Appl 7:1226–1237 doi: 10.1111/eva.12218

Foo SA, Dworjanyn SA, Poore AGB, Byrne M (2012) Adaptive capacity of the habitat modifying sea urchin Centrostephanus rodgersii to ocean warming and ocean acidification: Performance of early embryos. PLoS One 7: doi: 10.1371/journal.pone.0042497

Foo SA, Sparks K, Uthicke S, Karelitz S, Barker M, Byrne M, Lamare MD (2016) Contributions of genetic and environmental variance in early development of the Antarctic sea urchin Sterechinus neumayeri in response to increased ocean temperature and acidification. Mar Biol 161:130–141

Forsman A (2015) Rethinking phenotypic plasticity and its consequences for individuals, populations and species. Heredity (Edinb) 115:276–84 doi: 10.1038/hdy.2014.92

239

Fraser KPP, Clarke A, Peck LS (2002) Feast and famine in Antarctica: seasonal physiology in the limpet Nacella concinna. Mar. Ecol. Prog. Ser. 242:169–177

Frederich M, Pörtner HO (2000) Oxygen limitation of thermal tolerance defined by cardiac and ventilatory performance in spider crab, Maja squinado. Am J Physiol Regul Integr Comp Physiol 279:R1531-8

Freeman S, Herron J (2004) Evolutionary Analysis, 3rd edn. Prentice Hall, New Jersey

Frölicher TL, Sarmiento JL, Paynter DJ, Dunne JP, Krasting JP, Winton M, Frölicher TL, Sarmiento JL, Paynter DJ, Dunne JP, Krasting JP, Winton M (2015) Dominance of the Southern Ocean in anthropogenic carbon and heat uptake in CMIP5 Models. J Clim 28:862–886 doi: 10.1175/JCLI-D-14-00117.1

Fung IY, Doney SC, Lindsay K, John J (2005) Evolution of carbon sinks in a changing climate. Proc Natl Acad Sci USA 102: 11201-11206

Gagnon P, Himmelman JH, Johnson LE (2004) Temporal variation in community interfaces: Kelp-bed boundary dynamics adjacent to persistent urchin barrens. Mar Biol 144:1191– 1203 doi: 10.1007/s00227-003-1270-x

Garcia E, Clemente S, Hernandez J (2015) Ocean warming ameliorates the negative effects of ocean acidification on Paracentrotus lividus larval development and settlement. Mar Environ Res 110:61–68 doi: 10.1016/J.MARENVRES.2015.07.010

Garcia-Gonzalez F, Yasui Y, Evans JP (2014) Mating portfolios: bet-hedging, sexual selection and female multiple mating. Proc R Soc B Biol Sci. doi: 10.1098/rspb.2014.1525

Garzke J, Hansen T, Ismar SMH, Sommer U (2016) Combined effects of ocean warming and acidification on copepod abundance, body size and fatty acid content. PLoS One 11:e0155952 doi: 10.1371/journal.pone.0155952

Gavery MR, Roberts SB (2013) Predominant intragenic methylation is associated with gene expression characteristics in a bivalve mollusc. PeerJ 1:e215 doi: 10.7717/peerj.215

Gaylord B, Kroeker K, Sunday J, Anderson K, Barry J, Brown N, Connell S, Dupont S, Fabricius K, Hall-Spencer J, Klinger T, Milazzo M, Munda P, Russell B, Sanford E, Schreiber S, Thiyagarajan V, Vaughn M, Widdicombe S, Harley C (2015) Ocean acidification through the lens of ecological theory. Ecology 96: 3–15

Ghalambor CK, McKay JK, Carroll SP, Reznick DN (2007) Adaptive versus non-adaptive phenotypic plasticity and the potential for contemporary adaptation in new environments. Funct Ecol. doi: 10.1111/j.1365-2435.2007.01283.x

Ghanizadeh Kazerouni E, Franklin CE, Seebacher F (2017) Parental exposure modulates the effects of UV-B on offspring in guppies. Funct Ecol 31:1082–1090 doi: 10.1111/1365- 2435.12817

240

Gianguzza P, Visconti G, Gianguzza F, Vizzini S, Sara G, Dupont S (2014) Temperature modulates the response of the thermophilous sea urchin Arbacia lixula early life stages to CO2-driven acidification. Mar Environ Res 93:70–77 doi: 10.1016/J.MARENVRES.2013.07.008

Gibbin E, Chakravarti L, Jarrold M, Christen F, Turpin V, Massamba-N’Siala G, Blier P, Calosi P (2016) Can multi-generational exposure to ocean warming and acidification lead to the adaptation of life-history and physiology in a marine metazoan? J Exp Biol 220:551–563

Gienapp P, Teplitsky C, Alho JS, Mills JA, Merilä J (2008) Climate change and evolution: Disentangling environmental and genetic responses. Mol Ecol 17:167–178 doi: 10.1111/j.1365-294X.2007.03413.x

Gille S (2002) Warming of the Southern Ocean since the 1950s. Science (80) 295:1275–1277 doi: 10.1126/science.1058304

Gillies C, Stark J, Johnstone G, Smith S (2012) Carbon flow and trophic structure of an Antarctic coastal benthic community as determined by δ13C and δ15N. Estuar Coast Shelf Sci 97:44–57 doi: 10.1016/J.ECSS.2011.11.003

Gomez-Mestre I, Jovani R (2013) A heuristic model on the role of plasticity in adaptive evolution: plasticity increases adaptation, population viability and genetic variation. Proc R Soc B Biol Sci 280:20131869–20131869 doi: 10.1098/rspb.2013.1869

Goncalves P, Anderson K, Thompson EL, Melwani A, Parker LM, Ross PM, Raftos DA (2016) Rapid transcriptional acclimation following transgenerational exposure of oysters to ocean acidification. Mol Ecol 25:4836–4849 doi: 10.1111/mec.13808

Gonzalez-Bernat MJ, Lamare M, Uthicke S, Byrne M (2013) Fertilisation, embryogenesis and larval development in the tropical intertidal sand dollar Arachnoides placenta in response to reduced seawater pH. Mar Biol 160:1927–1941

Gonzalez-Bernat MJ, Lamare M, Barker M (2013) Effects of reduced seawater pH on fertilisation, embryogenesis and larval development in the Antarctic seastar Odontaster validus. Polar Biol 36:235–247

Gooding RA, Harley CDG, Tang E (2009) Elevated water temperature and carbon dioxide concentration increase the growth of a keystone echinoderm. Proc Natl Acad Sci U S A 106:9316–21 doi: 10.1073/pnas.0811143106

Gori A, Ferrier-Pagès C, Hennige SJ, Murray F, Rottier C, Wicks LC, Roberts JM (2016) Physiological response of the cold-water coral Desmophyllum dianthus to thermal stress and ocean acidification. PeerJ 4:e1606 doi: 10.7717/peerj.1606

Grans A, Jutfelt F, Sandblom E, Jonsson E, Wiklander K, Seth H, Olsson C, Dupont S, Ortega-Martinez O, Einarsdottir I, Bjornsson BT, Sundell K, Axelsson M (2014) Aerobic scope fails to explain the detrimental effects on growth resulting from warming and elevated CO2 in Atlantic halibut. J Exp Biol 217:711–717 doi: 10.1242/jeb.096743

241

Gray M, Langdon C, Waldbusser G, Hales B, Kramer S (2017) Mechanistic understanding of ocean acidification impacts on larval feeding physiology and energy budgets of the mussel Mytilus californianus. Mar Ecol Prog Ser 563:81–94 doi: 10.3354/meps11977

Greig M, Ridgway N, Shakespeare B (1988) Sea surface temperature variations at coastal sites around New Zealand. J New Zeal Mar Freshw Res 22:391–400

Grether GF (2005) Environmental change, phenotypic plasticity, and genetic compensation. Am Nat 166:E115-23 doi: 10.1086/432023

Grether GF (2014) Redesigning the genetic architecture of phenotypically plastic traits in a changing environment. Biol J Linn Soc 112:276–286 doi: 10.1111/bij.12064

Griffiths HJ, Meijers AJS, Bracegirdle TJ (2017) More losers than winners in a century of future Southern Ocean seafloor warming. Nat Clim Chang 7:749–754 doi: 10.1038/nclimate3377

Guillaume AS, Monro K, Marshall DJ (2016) Transgenerational plasticity and environmental stress: do paternal effects act as a conduit or a buffer? Funct Ecol. doi: 10.1111/1365- 2435.12604

Guinand B, Vandeputte M, Dupont-Nivet M, Vergnet A, Haffray P, Chavanne H, Chatain B (2017) Metapopulation patterns of additive and nonadditive genetic variance in the sea bass (Dicentrarchus labrax). Ecol Evol 7:2777–2790 doi: 10.1002/ece3.2832

Gunderson AR, Armstrong EJ, Stillman JH (2016) Multiple stressors in a changing world: The need for an improved perspective on physiological responses to the dynamic marine environment. Ann Rev Mar Sci 8:357–378 doi: 10.1146/annurev-marine-122414- 033953

Gunderson AR, Dillon ME, Stillman JH (2017) Estimating the benefits of plasticity in ectotherm heat tolerance under natural thermal variability. Funct Ecol 31:1529–1539 doi: 10.1111/1365-2435.12874

Gutt J, Bertler N, Bracegirdle TJ, Buschmann A, Hosie G, Isla E, Schloss IR, Smith CR, Tournadre J, Xavier JC, Comiso J, Hosie G, Isla E, Schloss IR, Smith CR, Tournadre J, Xavier JC (2010) The Southern Ocean ecosystem under multiple climate change stresses -an integrated circumpolar assessment. Glob Chang Biol 21:1434–1453 doi: 10.1111/geb.12794

Gutt J, Griffiths HJ, Jones CD (2013) Circumpolar overview and spatial heterogeneity of Antarctic macrobenthic communities. Mar Biodivers 43:481–487 doi: 10.1007/s12526- 013-0152-9

Hamdoun A, Epel D (2007) Embryo stability and vulnerability in an always changing world. Proc Natl Acad Sci U S A 104:1745–50 doi: 10.1073/pnas.0610108104

Hansen M, Olivieri I, Waller D, Nielsen E, GeM Working Group (12AD) Monitoring adaptive genetic responses to environmental change. Mol Ecol 21:1311–1329

242

Hardy N, Byrne M (2014) Early development of congeneric sea urchins (Heliocidaris) with contrasting life history modes in a warming and high CO2 ocean. Mar Environ Res 102:78–87 doi: 10.1016/J.MARENVRES.2014.07.007

Hardy N, Lamare M, Uthicke S, Wolfe K, Doo S, Dworjanyn SA, Byrne M (2014) Thermal tolerance of early development in tropical and temperate sea urchins: inferences for the tropicalization of eastern Australia. Mar Biol 161:395–409

Harley CDG, Randall Hughes A, Hultgren KM, Miner BG, Sorte CJB, Thornber CS, Rodriguez LF, Tomanek L, Williams SL (2006) The impacts of climate change in coastal marine systems. Ecol Lett 9:228–241 doi: 10.1111/j.1461-0248.2005.00871.x

Harris J, Lewis L, Smith J (2015) Quantifying scales of spatial variability in algal turf assemblages on coral reefs. Mar Ecol Prog Ser 532:41–57 doi: 10.3354/meps11344

Harris LG, Tyrrell MC (2001) Changing community states in the Gulf of Maine: Synergism between invaders, overfishing and climate change. Biol Invasions 3:9–21 doi: 10.1023/A:1011487219735

Hart MW, Marko PB (2010) It’s about time: Divergence, demography, and the evolution of developmental modes in marine invertebrates. Integr Comp Biol 50:643–661 doi: 10.1093/icb/icq068

Harvey BP, Gwynn-Jones D, Moore PJ (2013) Meta-analysis reveals complex marine biological responses to the interactive effects of ocean acidification and warming. Ecol Evol 3:1016–1030 doi: 10.1002/ece3.516

Harvey BP, McKeown NJ, Rastrick SPS, Bertolini C, Foggo A, Graham H, Hall-Spencer JM, Milazzo M, Shaw PW, Small DP, Moore PJ (2016) Individual and population-level responses to ocean acidification. Sci Rep 6:20194 doi: 10.1038/srep20194

Harvey B, Al-Janabi B, Broszeit S, Cioffi R, Kumar A, Aranguren-Gassis M, Bailey A, Green L, Gsottbauer C, Hall E, Lechler M, Mancuso F, Pereira C, Ricevuto E, Schram J, Stapp L, Stenberg S, Rosa L (2014) Evolution of marine organisms under climate change at different levels of biological organisation. Water 6:3545–3574 doi: 10.3390/w6113545

Hattich GSI, Listmann L, Raab J, Ozod-Seradj D, Reusch TBH, Matthiessen B (2017) Inter- and intraspecific phenotypic plasticity of three phytoplankton species in response to ocean acidification. Biol Lett 13:20160774 doi: 10.1098/rsbl.2016.0774

Hauck J, Völker C, Wolf-Gladrow DA, Laufkötter C, Vogt M, Aumont O, Bopp L, Buitenhuis ET, Doney SC, Dunne J, Gruber N, Hashioka T, John J, Quéré C Le, Lima ID, Nakano H, Séférian R, Totterdell I (2015) On the Southern Ocean CO2 uptake and the role of the biological carbon pump in the 21st century. Global Biogeochem Cycles 29:1451–1470 doi: 10.1002/2015GB005140

Hauri C, Friedrich T, Timmermann A (2015) Abrupt onset and prolongation of aragonite undersaturation events in the Southern Ocean. doi: 10.1038/NCLIMATE2844

243

Hawkins E, Sutton R (2012) Time of emergence of climate signals. Geophys Res Lett 39 doi: 10.1029/2011GL050087

Haycock L (2004) The reproduction and recruitment of the sand dollar Arachnoides placenta (L.) (Echinoidea: Echinodermata) from differing habitats on the North Queensland coast. James Cook University

Hedgecock D (1986) Is gene flow from pelagic larval dispersal important for the adaptation and evolution of marine invertebrates? Bull Mar Sci 39:550–564

Hendry AP, Farrugia T, Kinnison M (2008) Human influences on rates of phenotypic change in wild animal populations. Mol Ecol 17:20–29

Hettinger A, Sanford E, Hill TM, Lenz EA, Russell AD, Gaylord B (2013) Larval carry-over effects from ocean acidification persist in the natural environment. Glob Chang Biol. doi: 10.1111/gcb.12307

Hobday AJ, Pecl GT (2014) Identification of global marine hotspots: sentinels for change and vanguards for adaptation action. Rev Fish Biol Fish 24:415–425 doi: 10.1007/s11160- 013-9326-6

Hoegh-Guldberg O, Hughes L, McIntyre S, Lindenmayer DB, Parmesan C, Possingham HP, Thomas CD (2008) Assisted colonization and rapid climate change. Science (80- ) 321:345–6 doi: 10.1126/science.1157897

Hoegh-Guldberg O, Mumby P, Hooten AJ, Steneck RS, Greenfield P, Gomez E, Harvell CD, Sale PF, Edwards AJ, Caldeira K, Knowlton N, Eakin CM, Iglesias-Prieto R, Muthiga N, Bradbury RH, Dubi A, Hatziolos ME (2007) Coral reefs under rapid climate change and ocean acidification. Science 315:1811 doi: 10.1126/science.1137094

Hoegh-Guldberg O (1999) Climate change, coral bleaching and the future of the world’s coral reefs. Mar Freshw Res 50:839 doi: 10.1071/MF99078

Hoegh-Guldberg O, Bruno JF (2010) The impact of climate change on the world’s marine ecosystems. Science 328:1523–8 doi: 10.1126/science.1189930

Hoegh-Guldberg O, Pearse JS (1995) Temperature, food availability, and the development of marine invertebrate larvae. Am Zool 35:415–425 doi: 10.1093/icb/35.4.415

Hoffmann AA, Shirriffs J, Scott M (2005) Relative importance of plastic vs genetic factors in adaptive differentiation: geographical variation for stress resistance in Drosophila melanogaster from eastern Australia. Funct Ecol 19:222–227 doi: 10.1111/j.1365- 2435.2005.00959.x

Hoffmann AA, Sgrò CM (2011) Climate change and evolutionary adaptation. doi: 10.1038/nature09670

Hofmann GE, Evans TG, Kelly MW, Padilla-Gamiño JL, Blanchette CA, Washburn L, Chan F, McManus MA, Menge BA, Gaylord B, Hill TM, Sanford E, LaVigne M, Rose JM, Kapsenberg L, Dutton JM (2014) Exploring local adaptation and the ocean acidification

244

seascape: Studies in the California Current Large Marine Ecosystem. Biogeosciences 11:1053–1064 doi: 10.5194/bg-11-1053-2014

Hofmann GE (2017) Ecological Epigenetics in Marine Metazoans. Front Mar Sci 4:4 doi: 10.3389/fmars.2017.00004

Hofmann GE, Barry JP, Edmunds PJ, Gates RD, Hutchins DA, Klinger T, Sewell MA (2010) The effect of ocean acidification on calcifying organisms in marine ecosystems: An organism- to- ecosystem perspective. Annu Rev Ecol Evol Syst 41:127–47 doi: 10.1146/annurev.ecolsys.110308.120227

Hollander J, Collyer ML, Adams DC, Johannesson K (2006) Phenotypic plasticity in two marine snails: constraints superseding life history. J Evol Biol 19:1861–1872 doi: 10.1111/j.1420-9101.2006.01171.x

Holt RD (1990) The microevolutionary consequences of climate change. Trends Ecol Evol 5:311–315 doi: 10.1016/0169-5347(90)90088-U

Hu M, Tseng Y-C, Su Y-H, Lein E, Lee H-G, Lee J-R, Dupont S, Stumpp M (2017) Variability in larval gut pH regulation defines sensitivity to ocean acidification in six species of the Ambulacraria superphylum. Proc R Soc Part B Biol Sci 284:20171066 . doi: 10.1098/rspb.2017.1066

Hu M, Casties I, Stumpp M, Ortega-Martinez O, Dupont S (2014) Energy metabolism and regeneration are impaired by seawater acidification in the infaunal brittlestar Amphiura filiformis. J Exp Biol 217:2411–2421

Hughes AR, Inouye BD, Johnson MTJ, Underwood N, Vellend M (2008) Ecological consequences of genetic diversity. Ecol Lett 11:609–623 doi: 10.1111/j.1461- 0248.2008.01179.x

Hurd CL, Cornwall CE, Currie K, Hepburn C, McGraw C, Hunter K, Boyd P (2011) Metabolically induced pH fluctuations by some coastal calcifiers exceed projected 22nd century ocean acidification: a mechanism for differential susceptibility? Glob Chang Biol 17:3254–3262

Hutchings P, Ahyong S, Byrne M, Przeslawski R, Wörheide G (2007) Vulnerability of benthic invertebrates of the Great Barrier Reef to climate change. In: Waycott M, Coolier C, McMahon K, Ralph P, McKenzie L, Udy J, Grech A (eds.) Climate change and the Great Barrier Reef: A vulnerability assessment, 1st edn. Department of Primary Industries and Fisheries, Queensland, pp 309–356

IPCC (2007) Climate change 2007: synthesis report. Contribution of working groups I, II and III. (eds.) Lenny Bernstein, Peter Bosch, Osvaldo Canziani, Zhenlin Chen, Renate Christ, Ogunlade Davidson, William Hare.

IPCC (2013) Climate Change 2013: The Physical Science Basis. Working Group 1 Contribution to the Fifth Assessment report of the Intergovermental Panel on Climate Change. Cambridge University Press, Cambridge, UK and New York, USA

245

Jakubowska M, Normant M (2015) Metabolic rate and activity of blue mussel Mytilus edulis trossulus under short-term exposure to carbon dioxide-induced water acidification and oxygen deficiency. Mar Freshw Behav Physiol 48:25–39 doi: 10.1080/10236244.2014.986865

Jangoux M, van Impe E (1977) The annual pyloric cycle of Asterias rubens L. (echinodermata: Asteroidea). J Exp Mar Bio Ecol 30:165–184 doi: 10.1016/0022- 0981(77)90010-7

Jangoux M, Lawrence JM (1987) Echinoderm studies, volume 2. CRC Press

Jenkins NL, Sgro CM, Hoffmann AA (1997) Environmental stress and the expression of genetic variation. In: Bijlsma R, Loeschcke V (eds.) Environmental Stress, Adaptation and Evolution, 1st edn. Birkhauser, Basel, Switzerland, pp 79–96

Jensen N, Allen RM, Marshall DJ (2014) Adaptive maternal and paternal effects: Gamete plasticity in response to parental stress. Funct Ecol. doi: 10.1111/1365-2435.12195

Johnson CR, Banks SC, Barrett NS, Cazassus F, Dunstan PK, Edgar GJ, Frusher SD, Gardner C, Haddon M, Helidoniotis F, Hill KL, Holbrook NJ, Hosie GW, Last PR, Ling SD, Melbourne-Thomas J, Miller K, Pecl GT, Richardson AJ, Ridgway KR, Rintoul SR, Ritz DA, Ross DJ, Sanderson JC, Shepherd SA, Slotwinski A, Swadling KM, Taw N (2011) Climate change cascades: Shifts in oceanography, species’ ranges and subtidal marine community dynamics in eastern Tasmania. J Exp Mar Bio Ecol 400:17–32 doi: 10.1016/j.jembe.2011.02.032

Johnson LG, Babcock RC (1994) Temperature and the larval ecology of the crown-of-thorns starfish, Acanthaster planci. Biol Bull 187:304–308 doi: 10.2307/1542287

Johnston IA, Clarke A, Ward P (1991) Temperature and metabolic rate in sedentary fish from the Antarctic, North Sea and Indo-West Pacific Ocean. Mar Biol 109:191–195

Jutterstrom S, Anderson L (2005) The saturation of calcite and aragonite in the Arctic Ocean. Mar Chem 94:101–110 doi: 10.1016/J.MARCHEM.2004.08.010

Kamya PZ, Dworjanyn SA, Hardy N, Mos B, Uthicke S, Byrne M (2014) Larvae of the coral eating crown-of-thorns starfish, Acanthaster planci in a warmer-high CO 2 ocean. Glob Chang Biol 20:3365–3376 doi: 10.1111/gcb.12530

Kandiano ES, Van Der Meer MTJ, Schouten S, Fahl K, Sinninghe Damsté JS, Bauch HA (2017) Response of the North Atlantic surface and intermediate ocean structure to climate warming of MIS 11. doi: 10.1038/srep46192

Kapsenberg L, Kelley AL, Shaw EC, Martz TR, Hofmann GE (2015) Near-shore Antarctic pH variability has implications for the design of ocean acidification experiments. Sci Rep 5:9638 doi: 10.1038/srep09638

Karelitz SE, Uthicke S, Foo SA, Barker MF, Byrne M, Pecorino D, Lamare MD (2017) Ocean acidification has little effect on developmental thermal windows of echinoderms from Antarctica to the tropics. Glob Chang Biol 23:657–67. doi: 10.1111/gcb.13452

246

Keever CC, Puritz JB, Addison JA, Byrne M, Grosberg RK, Toonen RJ, Hart MW (2013) Shallow gene pools in the high intertidal: extreme loss of genetic diversity in viviparous sea stars (Parvulastra). Biol Lett 9:20130551 doi: 10.1098/rsbl.2013.0551

Kelly MW, Hofmann GE (2013) Adaptation and the physiology of ocean acidification. Funct Ecol 27:980–990 doi: 10.1111/j.1365-2435.2012.02061.x

Kelly MW, Padilla-Gamiño JL, Hofmann GE (2013) Natural variation and the capacity to adapt to ocean acidification in the keystone sea urchin Strongylocentrotus purpuratus. Glob Chang Biol. doi: 10.1111/gcb.12251

Kelly MW, Pankey MS, DeBiasse MB, Plachetzki DC (2017) Adaptation to heat stress reduces phenotypic and transcriptional plasticity in a marine copepod. Funct Ecol 31:398–406 doi: 10.1111/1365-2435.12725

Kelly RP, Eernisse DJ (2007) Southern hospitality: A latitudinal gradient in gene flow in the marine environment. Evolution (NY) 61:700–707 . doi: 10.1111/j.1558- 5646.2007.00055.x

Kidawa A (2009) Food selection of the Antarctic sea star Odontaster validus (Koehler): Laboratory experiments with food quality and size. Polish J Ecol 57:139–147

Kidawa A, Janecki T (2011) Antarctic benthic fauna in the global climate change. Pap Glob Chang 18:71–86 doi: 10.2478/v10190-010-0006-5

Kidawa A, Potocka M, Janecki T (2010) The effects of temperature on the behaviour of the Antarctic sea star Odontaster validus. Polish Polar Res 31:273–284 doi: 10.2478/v10183−010−0003−3

Kielland ØN, Bech C, Einum S (2017) No evidence for thermal transgenerational plasticity in metabolism when minimizing the potential for confounding effects. Proc R Soc Part B Biol Sci 284:20162494 doi: 10.1098/rspb.2016.2494

Kleitman N (1941) The effect of temperature on the righting of echinoderms. Biol Bull 80:292–298 doi: 10.2307/1537716

Klinger T, Chornesky EA, Whiteman EA, Chan F, Largier JL, Wakefield WW (2017) Using integrated, ecosystem-level management to address intensifying ocean acidification and hypoxia in the California Current large marine ecosystem. Elem Sci Anthr 5(0):16 doi: 10.1525/elementa.198.t1

Kong N, Li Q, Yu H, Kong L-F (2015) Heritability estimates for growth-related traits in the Pacific oyster ( Crassostrea gigas ) using a molecular pedigree. Aquac Res 46:499–508 doi: 10.1111/are.12205

Koojiman S (1986) Energy budgets can explain body size relations. J Theor Biol 121:269– 282 doi: 10.1016/S0022-5193(86)80107-2

Kopp M, Matuszewski S (2014) Rapid evolution of quantitative traits: Theoretical perspectives. Evol Appl. doi: 10.1111/eva.12127

247

Kroeker KJ, Kordas RL, Crim RN, Singh GG (2010) Meta-analysis reveals negative yet variable effects of ocean acidification on marine organisms. Ecol Lett 13:1419–1434 doi: 10.1111/j.1461-0248.2010.01518.x

Kroeker KJ, Kordas RL, Crim R, Hendriks IE, Ramajo L, Singh GS, Duarte CM, Gattuso J-P (2013) Impacts of ocean acidification on marine organisms: quantifying sensitivities and interaction with warming. Glob Chang Biol 19:1884–1896 doi: 10.1111/gcb.12179

Kroeker KJ, Kordas RL, Harley CDG (2017) Embracing interactions in ocean acidification research: confronting multiple stressor scenarios and context dependence. Biol Lett 13:20160802 doi: 10.1098/rsbl.2016.0802

Kruuk LEB (2004) Estimating genetic parameters in natural populations using the “animal model.” Philos Trans R Soc London B 359:873–890 doi: 10.1098/rstb.2003.1437

Kuhnhold H, Kamyab E, Novais S, Indriana L, Kunzmann A, Slater M, Lemos M (2017) Thermal stress effects on energy resource allocation and oxygen consumption rate in the juvenile sea cucumber, Holothuria scabra (Jaeger, 1833). Aquaculture 467:109–117 doi: 10.1016/J.AQUACULTURE.2016.03.018

Kuijper B, Johnstone RA (2017) Maternal Effects and Parent-Offspring Conflict. Evolution 220 doi: 10.1101/168310

Kurihara H, Yin R, Nishihara G, Soyano K, Ishimatsu A (2013) Effect of ocean acidification on growth, gonad development and physiology of the sea urchin Hemicentrotus pulcherrimus. Aquat Biol 18:281–292 doi: 10.3354/ab00510

Kutzbach JE, Ruddiman WF, Vavrus SJ, Philippon G, Kutzbach JE (2010) Climate model simulation of anthropogenic influence on greenhouse-induced climate change (early agriculture to modern): the role of ocean feedbacks. Clim Change 99:351–381 doi: 10.1007/s10584-009-9684-1

Lamare MD, Pecorino D, Hardy N, Liddy M, Byrne M, Uthicke S (2014) The thermal tolerance of crown-of-thorns (Acanthaster planci) embryos and bipinnaria larvae: implications for spatial and temporal variation in adult populations. Coral Reefs 33:207– 219

Lamare MD, Liddy M, Uthicke S (2016) In situ developmental responses of tropical sea urchin larvae to ocean acidification conditions at naturally elevated pCO2 vent sites. Proc R Soc Part B Biol Sci 283:20161506 doi: 10.1098/rspb.2016.1506

Lamit LJ, Lau MK, Næsborg RR, Wojtowicz T, Whitham TG, Gehring CA (2015) Genotype variation in bark texture drives lichen community assembly across multiple environments. Ecology 96:960–971 doi: 10.1890/14-1007.1

Lande R (2009) Adaptation to an extraordinary environment by evolution of phenotypic plasticity and genetic assimilation. J Evol Biol. doi: 10.1111/j.1420-9101.2009.01754.x

Lande R, Shannon S (1996) The Role of Genetic Variation in Adaptation and Population Persistence in a Changing Environment. Evolution (NY) 50:434 doi: 10.2307/2410812

248

Lange R, Marshall D (2017) Ecologically relevant levels of multiple, common marine stressors suggest antagonistic effects. Sci Rep 7:6281 doi: 10.1038/s41598-017-06373-y

Langenbuch M, Portner H (2002) Changes in metabolic rate and N excretion in the marine invertebrate Sipunculus nudus under conditions of environmental hypoxia. J Exp Biol 205:1152–1160

Langenbuch M, Portner H (2004) High sensitivity to chronically elevated CO2 levels in a eurybathic marine sipunculid. Aquat Toxicol 70:55–61 doi: 10.1016/J.AQUATOX.2004.07.006

Langerhans RB, Dewitt TJ (2002) Plasticity constrained: over-generalized induction cues cause maladaptive phenotypes. Evol Ecol Res 4:857–870

Lannig G, Eilers S, Pörtner HO, Sokolova IM, Bock C (2010) Impact of Ocean Acidification on Energy Metabolism of Oyster, Crassostrea gigas - Changes in Metabolic Pathways and Thermal Response. Mar Drugs 8:2318–2339 doi: 10.3390/md8082318

Law CS, Bell JJ, Bostock HC, Cornwall CE, Cummings VJ, Currie K, Davy SK, Gammon M, Hepburn CD, Hurd CL, Lamare M, Mikaloff-Fletcher SE, Nelson WA, Parsons DM, Ragg NLC, Sewell MA, Smith AM, Tracey DM (2017) Ocean acidification in New Zealand waters: trends and impacts. New Zeal J Mar Freshw Res 8330:1–41 doi: 10.1080/00288330.2017.1374983

Lawrence J (1973) Level, content, and caloric equivalents of the lipid, carbohydrate, and protein in the body components of Luidia clathrata (Echinodermata: Asteroidea: Platyasterida) in Tampa bay. J Exp Mar Bio Ecol 11:263–274 doi: 10.1016/0022- 0981(73)90026-9

Le Roy A, Loughland I, Seebacher F (2017) Differential effects of developmental thermal plasticity across three generations of guppies (Poecilia reticulata): canalization and anticipatory matching. Sci Rep 7:4313 doi: 10.1038/s41598-017-03300-z

Lebrato M, Andersson AJ, Ries JB, Aronson RB, Lamare MD, Koeve W, Oschlies A, Iglesias-Rodriguez MD, Thatje S, Amsler M, Vos SC, Jones DOB, Ruhl HA, Gates AR, McClintock JB (2016) Benthic marine calcifiers coexist with CaCO3 -undersaturated seawater worldwide. Global Biogeochem Cycles 30:1038–1053 doi: 10.1002/2015GB005260

Lee JW, Applebaum SL, Manahan DT (2016) Metabolic cost of protein synthesis in larvae of the Pacific oyster (Crassostrea gigas) is fixed across genotype, phenotype, and environmental temperature. Biol Bull 230:175–87 doi: 10.1086/BBLv230n3p175

Leite Figueirdo D, Branco P, dos Santos D, Emerenciano A, Iunes R, Borges J, Roberto J, Cunha da Silva M (2016) Ocean acidification affects parameters of immune response and extracellular pH in tropical sea urchins Lytechinus variegatus and Echinometra luccunter. Aquat Toxicol 180:84–94 doi: 10.1016/J.AQUATOX.2016.09.010

Lenton A, Tilbrook B, Law R, Bakker D, Doney SC, Gruber N, Hoppema M, Ishii M, Lovenduski NS, Matear RJ, McNeil BI, Metzl N, Mikaloff Fletcher SE, Monteiro P,

249

Rödenbeck C, Sweeney C, Takahashi T (2013) Sea-air CO2 fluxes in the Southern Ocean: 1990–2009. In: 39th International Carbon Dioxide Conference. Beijing

Lesser MP (2016) Climate change stressors cause metabolic depression in the blue mussel, Mytilus edulis, from the Gulf of Maine. Limnol Oceanogr 61:1705–1717 doi: 10.1002/lno.10326

Leung JYS, Russell BD, Connell SD (2017) Mineralogical plasticity acts as a compensatory mechanism to the impacts of ocean acidification. Environ Sci Technol 51:2652–2659 doi: 10.1021/acs.est.6b04709

Levitus S, Antonov J, Boyer T, Baranova O, Garcia H, Locarnini R, Mishonov A, Reagan J, Seidov D, Yarosh E, Zweng M (2012) World ocean heat content and thermosteric sea level change (0-2000m), 1955-2010. Geophys Res Lett 39: L10603 doi: 10.1029/2012GL051106

Lewis E, Wallace D (1998) Program developed for CO2 system calculations. ORNL/CDIAC- 105. Carbon Dioxide Information Analysis Centre, U.S. Department of Energy, Oak Ridge, Tennessee

Ling S, Johnson C, Ridgway K, Hobday A, Haddon M (2009) Climate-driven range extension of a sea urchin: inferring future trends by analysis of recent population dynamics. Glob Chang Biol 15:719–731

Lischka S, Büdenbender J, Boxhammer T, Riebesell U (2011) Impact of ocean acidification and elevated temperatures on early juveniles of the polar shelled pteropod Limacina helicina: mortality, shell degradation, and shell growth. Biogeosciences 8:919–932 doi: 10.5194/bg-8-919-2011

Lister KN, Lamare MD, Burritt DJ (2017) Maternal antioxidant provisioning mitigates pollutant-induced oxidative damage in embryos of the temperate sea urchin Evechinus chloroticus. Sci Rep 7:1954 doi: 10.1038/s41598-017-02077-5

Liu X, Change Y, Xiang J, Cao X (2005) Estimates of genetic parameters for growth traits of the sea urchin, Strongylocentrotus intermedius. Aquaculture 243:27–32 doi: 10.1016/J.AQUACULTURE.2004.10.014

Llodra E (2002) Fecundity and life-history strategies in marine invertebrates. Adv Mar Biol 43:87–170

Lloyd M, Makuhov A, Pespeni M (2016) Loss of genetic diversity as a consequence of selection in response to high pCO2. Evol Appl 9:1124–1132

Lough JM, Hobday AJ, Lough JM (2011) Projected climate change in Australian marine and freshwater environments. Mar Freshw Res 62:1000–1014 doi: 10.1071/MF10302

Lyman JM, Johnson GC (2014) Estimating global ocean heat content changes in the upper 1800 m since 1950 and the influence of climatology choice. J Clim 27:1945–1957 doi: 10.1175/JCLI-D-12-00752.1

250

Lynch M, Walsh B (1998) Genetics and analysis of quantitative traits. Sinauer, Massachussetts, USA

Maltby L, Naylor C, Calow P (1990) Effect of stress on a freshwater benthic detritivore: Scope for growth in Gammarus pulex. Ecotoxicol Environ Saf 19:285–291

Malvezzi AJ, Murray CS, Feldheim KA, DiBattista JD, Garant D, Gobler CJ, Chapman DD, Baumann H (2015) A quantitative genetic approach to assess the evolutionary potential of a coastal marine fish to ocean acidification. Evol Appl 8:352–362 doi: 10.1111/eva.12248

Manriquez P, Torres R, Matson P, Lee M, Jara M, Seguel M, Sepulvida F, Pereira L (2017) Effects of ocean warming and acidification on the early benthic ontogeny of an ecologically and economically important echinoderm. Mar Ecol Prog Ser 563:169–184

Marsh AG, Maxson RE, Manahan DT (2001) High macromolecular synthesis with low metabolic cost in Antarctic sea urchin embryos. Science 291:1950–2 doi: 10.1126/science.1056341

Marsh AG, Manahan DT (1999) A method for accurate measurements of the respiration rates of marine invertebrate embryos and larvae. Mar Ecol Prog Ser 184:1–10

Marsh AG, Pasqualone AA (2014) DNA methylation and temperature stress in an Antarctic polychaete, Spiophanes tcherniai. Front Physiol 5:173 doi: 10.3389/fphys.2014.00173

Marshall DJ, Monro K, Bode M, Keough MJ, Swearer S (2010) Phenotype–environment mismatches reduce connectivity in the sea. Ecol Lett 13:128–140 doi: 10.1111/j.1461- 0248.2009.01408.x

Marshall DJ (2015) Environmentally induced (co)variance in sperm and offspring phenotypes as a source of epigenetic effects. J Exp Biol 218:107–113 doi: 10.1242/jeb.106427

Marshall DJ (2008) Transgenerational plasticity in the sea: Context-dependent maternal effects across the life history. Source Ecol 89:418–427

Marshall DJ, Heppell SS, Munch SB, Warner RR (2010) The relationship between maternal phenotype and offspring quality: Do older mothers really produce the best offspring? Ecology 91:2862–2873 doi: 10.1890/09-0156.1

Marshall DJ, Uller T (2007) When is a maternal effect adaptive? Oikos 116:1957–1963 doi: 10.1111/j.2007.0030-1299.16203.x

Mayzaud P (1973) Respiration and nitrogen excretion of zooplankton. II. Studies of the metabolic characteristics of starved animals. Mar Biol 21:19–28 doi: 10.1007/BF00351188

McBryan TL, Anttila K, Healy TM, Schulte PM (2013) Responses to temperature and hypoxia as interacting stressors in fish: Implications for adaptation to environmental change. Integr Comp Biol 53:648–659 doi: 10.1093/icb/ict066

251

Mcclintock JB, Pearse JS, Bosch I (1988) Population structure and energetics of the shallow - water antarctic sea star Odontaster validus in contrasting habitats. Mar Biol 99:235–246 doi: 10.1007/BF00391986

Mcclintock JB (1994) Trophic biology of antarctic shallow-water echinoderms. Mar Ecol Prog Ser 111:191–202

McClintock JB (1989) Energetic composition, reproductive output, and resource allocation of Antarctic asteroids. Polar Biol 9:147–153 doi: 10.1007/BF00297169

McClintock JB, Amsler MO, Angus RA, Challener RC, Schram JB, Amsler CD, Mah CL, Cuce J, Baker BJ (2011) The Mg-calcite composition of Antarctic echinoderms: Important implications for predicting the impacts of ocean acidification. J Geol 119:457–466 doi: 10.1086/660890

Mcclintock JB, Angus RA, Mcdonald MR, Amsler CD, Catledge SA, Vohra YK (2009) Rapid dissolution of shells of weakly calcified Antarctic benthic macroorganisms indicates high vulnerability to ocean acidification. Antarct Sci Antarct Sci Ltd 21:449– 456 doi: 10.1017/S0954102009990198

McKnight DG (1969) Infaunal benthic communities of the New Zealand continental shelf. New Zeal J Mar Freshw Res 3:409–444 doi: 10.1080/00288330.1969.9515308

Mcneil BI, Matear RJ (2008) Southern Ocean acidification: A tipping point at 450-ppm atmospheric CO2. Proc Natl Acad Sci 105:18860–18864

Mehrbach C, Culbertson C, Hawley J, Pytkowicz R (1973) Measurement of the apparent dissociation of carbonic acid in seawater at atmospheric pressure. Limnol Oceanogr 18:897–907

Melzner F, Gutowska MA, Langenbuch M, Dupont S, Lucassen M, Thorndyke MC, Bleich M, Pörtner H-O (2009) Physiological basis for high CO2 tolerance in marine ectothermic animals: pre-adaptation through lifestyle and ontogeny? Biogeosciences 6:2313–2331

Melzner F, Thomsen J, Koeve W, Oschlies A, Gutowska M, Bange H, Hansen H, Kortzinger A (2013) Future ocean acidification will be amplified by hypoxia in coastal environments. Mar Biol 160:1875–1888

Mendo T, Semmens JM, Lyle JM, Tracey SR, Moltschaniwskyj N (2016) Reproductive strategies and energy sources fuelling reproductive growth in a protracted spawner. Mar Biol 163 doi: 10.1007/s00227-015-2785-7

Menéndez R, Megías AG, Hill JK, Braschler B, Willis SG, Collingham Y, Fox R, Roy DB, Thomas CD (2006) Species richness changes lag behind climate change. Proceedings Biol Sci 273:1465–70 doi: 10.1098/rspb.2006.3484

Mengel M, Feldmann J, Levermann A (2016) Linear sea-level response to abrupt ocean warming of major West Antarctic ice basin. doi: 10.1038/NCLIMATE2808

252

Meredith MP (2016) Understanding the structure of changes in the Southern Ocean eddy field. Geophys Res Lett 43:5829–5832 doi: 10.1002/2016GL069677

Merilä J, Hendry AP (2014) Climate change, adaptation, and phenotypic plasticity: the problem and the evidence. Evol Appl 7:1–14 doi: 10.1111/eva.12137

Michaelidis B, Ouzounis C, Paleras A, Pörtner HO (2005) Effects of long-term moderate hypercapnia on acid–base balance and growth rate in marine mussels Mytilus galloprovincialis. Mar. Ecol. Prog. Ser. 293:109–118

Michel L, Dubois P, Eleaume M, Fournier J, Gallut C, Jane P, Lepoint G (2017) Increased sea ice cover disrupts food web structure in coastal Antarctica. In: Proceedings of the12th SCAR Biology Symposium. Open Repoository and Bibliography 2017, Bruges

Midorikawa T, Inouye H, Ishii M, Sasano D, Kosugi N, Hashida G, Nakaoka S, Suzuki T (2012) Decreasing pH trend estimated from 35-year time series of carbonate parameters in the Pacific sector of the Southern Ocean in summer. Deep Sea Res Part I Oceanogr Res Pap 61:131–139 doi: 10.1016/J.DSR.2011.12.003

Mieszkowska N, Sugden H (2016) Climate-driven range shifts within benthic habitats across a marine biogeographical transition zone. In: Dumbrell A, Kordas RL, Woodward G (eds.) Advances in Ecological Research, 55th edn. Oxford Academic Press, Oxford, UK, pp 344–350

Miller GM, Watson S-A, Donelson JM, Mccormick MI, Munday PL (2012) Parental environment mediates impacts of increased carbon dioxide on a coral reef fish. Nat Clim Chang 2 doi: 10.1038/NCLIMATE1599

Moles J, Wägele H, Cutignano A, Fontana A, Ballesteros M, Avila C (2017) Giant embryos and hatchlings of Antarctic nudibranchs (Mollusca: Gastropoda: Heterobranchia). Mar Biol 164:114 doi: 10.1007/s00227-017-3143-8

Moore M, Manahan DT (2007) Variation among females in egg lipid content and developmental success of echinoderms from McMurdo Sound, Antarctica. Polar Biol 30:1245–1252 doi: 10.1007/s00300-007-0285-z

Moran PJ, Bradbury RH, Reichelt RE (1988) Distribution of recent outbreaks of the crown- of-thorns starfish (Acanthaster planci) along the Great Barrier Reef: 1985-1986. Coral Reefs 7:125–137 doi: 10.1007/BF00300972

Moran PJ, De’ath G (1992) Estimates of the abundance of the crown-of-thorns starfish Acanthaster planci in outbreaking and non-outbreaking populations on reefs within the Great Barrier Reef. Mar Biol 113:509–515 doi: 10.1007/BF00349178

Moreno G, Hoegh-Guldberg O (1999) The energetics of development of three congeneric seastars (Patiriella Verrill, 1913) with different types of development. J Exp Mar Bio Ecol 235:1–20 doi: 10.1016/S0022-0981(98)00031-8

Morley SA, Bates AE, Lamare M, Richard J, Nguyen KD, Brown J, Peck LS (2017) Rates of warming and the global sensitivity of shallow water marine invertebrates to elevated

253

temperature. J Mar Biol Assoc United Kingdom 96:159–165 doi: 10.1017/S0025315414000307

Morley SA, Berman J, Barnes DKA, de Juan Carbonell C, Downey R V., Peck LS (2016) Extreme phenotypic plasticity in metabolic physiology of Antarctic demosponges. Front Ecol Evol 3:157 doi: 10.3389/fevo.2015.00157

Morley SA, Lai C-H, Clarke A, Tan KS, Thorne MAS, Peck LS (2014) Limpet feeding rate and the consistency of physiological response to temperature. J Comp Physiol B 184:563–570 doi: 10.1007/s00360-014-0814-3

Morley SA, Suckling CC, Clark MS, Cross EL, Peck LS (2016) Long-term effects of altered pH and temperature on the feeding energetics of the Antarctic sea urchin, Sterechinus neumayeri. Biodiversity 17:34–45

Morris MRJ, Rogers SM (2013) Overcoming maladaptive plasticity through plastic compensation. Curr Zool 59:526–536

Mousseau T, Fox C (1998) The adaptive significance of maternal effects. Trends Ecol Evol 13:403–407 doi: 10.1016/S0169-5347(98)01472-4

Mulvaney R, Abram NJ, Hindmarsh RCA, Arrowsmith C, Fleet L, Triest J, Sime LC, Alemany O, Foord S (2012) Recent Antarctic Peninsula warming relative to Holocene climate and ice-shelf history. Nature 489:141–144 doi: 10.1038/nature11391

Munday PL (2014) Transgenerational acclimation of fishes to climate change and ocean acidification. F1000Prime Rep 6:99 doi: 10.12703/P6-99

Munday PL, Donelson JM, Domingos JA (2017) Potential for adaptation to climate change in a coral reef fish. Glob Chang Biol 23:307–317 doi: 10.1111/gcb.13419

Munday PL, Warner RR, Monro K, Pandolfi JM, Marshall DJ (2013) Predicting evolutionary responses to climate change in the sea. Ecol Lett 16:1488–1500 doi: 10.1111/ele.12185

N. Reznick D, K. Ghalambor C (2001) The population ecology of contemporary adaptations: what empirical studies reveal about the conditions that promote adaptive evolution. Genetica 112/113:183–198 doi: 10.1023/A:1013352109042

Nagelkerken I, Munday PL (2016) Animal behaviour shapes the ecological effects of ocean acidification and warming: moving from individual to community-level responses. Glob Chang Biol 22:974–989 doi: 10.1111/gcb.13167

Nanninga GB, Berumen ML (2014) The role of individual variation in marine larval dispersal. Front Mar Sci 1:71 doi: 10.3389/fmars.2014.00071

Näslund J, Lindström E, Lai F, Jutfelt F (2015) Behavioural responses to simulated bird attacks in marine three-spined sticklebacks after exposure to high CO2 levels. Mar Freshw Res 66:877 doi: 10.1071/MF14144

254

Nelson K (2016) Biofilm response to ocean acidification and the effects on serpulid polychaete settlement. Masters Thesis, University of Otago.

Nguyen H, Byrne M (2014) Early benthic juvenile Parvulastra exigua (Asteroidea) are tolerant to extreme acidification and warming in its intertidal habitat. J Exp Mar Bio Ecol 453:36–42 doi: 10.1016/J.JEMBE.2013.12.007

Nguyen HD, Doo SS, Soars NA, Byrne M (2012) Noncalcifying larvae in a changing ocean: warming, not acidification/hypercapnia, is the dominant stressor on development of the sea star Meridiastra calcar. Glob Chang Biol 18:2466–2476 doi: 10.1111/j.1365- 2486.2012.02714.x

Nilsson GE, Dixson DL, Domenici P, Mccormick MI, Sørensen C, Watson S-A, Munday PL (2012) Near-future carbon dioxide levels alter fish behaviour by interfering with neurotransmitter function. Nat Clim Chang 2: doi: 10.1038/NCLIMATE1352

Nipper MG, Martin ML, Williams EK (1997) The optimisation and validation of a marine toxicity test using the New Zealand echinoid, Fellaster zelandiae. Australas J Ecotoxicol 3:109–115

Noisette F, Bordeyne F, Davoult D, Martin S (2015) Assessing the physiological responses of the gastropod Crepidula fornicata. Limnol Oceanogr 61:430–444

Nunney L (2016) c. J Hered 107:15–24 doi: 10.1093/jhered/esv084

Nussey DH, Wilson AJ, Brommer JE (2007) The evolutionary ecology of individual phenotypic plasticity in wild populations. J Evol Biol 20:831–844 doi: 10.1111/j.1420- 9101.2007.01300.x

O’Loughlin PM, Waters JM, Roy MS (2002) Description of a new species of Patiriella from New Zealand, and review of Patiriella regularis (Echinodermata, Asteroidea) based on morphological and molecular data. J R Soc New Zeal 32:697–711 doi: 10.1080/03014223.2002.9517716

Orr JC, Fabry VJ, Aumont O, Bopp L, Doney SC, Feely RA, Gnanadesikan A, Gruber N, Ishida A, Joos F, Key RM, Lindsay K, Maier-Reimer E, Matear R, Monfray P, Mouchet A, Najjar RG, Plattner G-K, Rodgers KB, Sabine CL, Sarmiento JL, Schlitzer R, Slater RD, Totterdell IJ, Weirig M-F, Yamanaka Y, Yool A (2005) Anthropogenic ocean acidification over the twenty-first century and its impact on calcifying organisms. Nature 437: doi: 10.1038/nature04095

Oudejans RCHM, Van Der Sluis H, Van Der Plas AJ (1979) Changes in the biochemical composition of the pyloric caeca of female seastars, Asterias rubens, during their annual reproductive cycle. Mar Biol 53:231–238

Pan T-CF, Applebaum SL, Lentz BA, Manahan DT (2016) Predicting phenotypic variation in growth and metabolism of marine invertebrate larvae. J Exp Mar Bio Ecol 483:64–73 doi: 10.1016/j.jembe.2016.06.006

255

Pan T-CF, Applebaum SL, Manahan DT (2015) Experimental ocean acidification alters the allocation of metabolic energy. Proc Natl Acad Sci U S A 112:4696–701 doi: 10.1073/pnas.1416967112

Pandolfi JM, Connolly S, Marshall D, Cohen A (2011) Climate change, human impacts, and the resilience of coral reefs. Science (80- ) 333:418–422 doi: 10.1126/science.1085046

Parker LM, O’Connor WA, Raftos DA, Pörtner HO, Ross PM (2015) Persistence of positive carryover effects in the oyster, Saccostrea glomerata, following transgenerational exposure to ocean acidification. PLoS One. doi: 10.1371/journal.pone.0132276

Parker LM, O’Connor WA, Byrne M, Coleman RA, Virtue P, Dove M, Gibbs M, Spohr L, Scanes E, Ross PM (2017) Adult exposure to ocean acidification is maladaptive for larvae of the Sydney rock oyster Saccostrea glomerata in the presence of multiple stressors. Biol Lett 13:20160798 doi: 10.1098/rsbl.2016.0798

Parmesan C (2006) Ecological and evolutionary responses to recent climate change. Annu Rev Ecol Evol Syst 37:637–669 doi: 10.1146/annurev.ecolsys.37.091305.110100

Patiño S, Keever CC, Sunday JM, Popovic I, Byrne M, Hart MW (2016) Sperm Bindin divergence under sexual selection and concerted evolution in sea stars. Mol Biol Evol 33:1988–2001 doi: 10.1093/molbev/msw081

Pearse JS (1969) Slow developing demersal embryos and larvae of the antarctic sea star Odontaster validus. Mar Biol 3:110–116 doi: 10.1007/BF00353429

Pearson R, Endean R (1969) A preliminary study of the coral predator Acanthaster planci (L.) (Asteroidea) on the Great Barrier Reef. Fish Notes 3:27–55

Pechenik JA (2006) Larval experience and latent effects--metamorphosis is not a new beginning. Integr Comp Biol 46:323–333 doi: 10.1093/icb/icj028

Pechenik J (1989) Environmental influences on larval survival and development. In: Giese A, Pearse J, Pearse V (eds.) Reproductive biology of invertebrates. Blackwell Science, Ltd, California, p 551

Pechenik JA (1999) On the advantages and disadvantages of larval stages in benthic marine invertebrate life cycles. Mar. Ecol. Prog. Ser. 177:269–297

Peck LS, Prothero-Thomas E (2002) Temperature effects on the metabolism of larvae of the Antarctic starfish Odontaster validus, using a novel micro-respirometry method. Mar Biol 141:271–276 doi: 10.1007/s00227-002-0834-5

Peck LS (2002) Ecophysiology of Antarctic marine ectotherms: limits to life. Polar Biol 25:31–40 doi: 10.1007/s003000100308

Peck LS (2005) Prospects for survival in the Southern Ocean: vulnerability of benthic species to temperature change. Antarct Sci 17:497–507

Peck LS (2016) A Cold Limit to Adaptation in the Sea. Trends Ecol. Evol. 31: 13-26

256

Peck LS, Clark MS, Morley SA, Massey A, Rossetti H (2009) Animal temperature limits and ecological relevance: effects of size, activity and rates of change. Funct Ecol 23:248– 256 doi: 10.1111/j.1365-2435.2008.01537.x

Peck LS, Webb KE, Miller A, Clark MS, Hill T (2008) Temperature limits to activity, feeding and metabolism in the Antarctic starfish Odontaster validus. Mar Ecol Prog Ser 358:181–189 doi: 10.3354/meps07336

Peck LS, Conway LZ (2000) The myth of metabolic cold adaptation: oxygen consumption in stenothermal Antarctic bivalves. J Geol Soc London 177:441–450

Peck LS, Webb KE, Bailey DM (2004) Extreme sensitivity of biological function to temperature in Antarctic marine species. Funct Ecol 18:625–630

Peck L, Prothero-Thomas E (2002) Temperature effects on the metabolism of larvae of the Antarctic starfish Odontaster validus, using a novel micro-respirometry method. Mar Biol 141:271–276 doi: 10.1007/s00227-002-0834-5

Pecorino D, Lamare MD, Barker MF (2013) Reproduction of the Diadematidae sea urchin Centrostephanus rodgersii in a recently colonized area of northern New Zealand. Mar Biol Res. doi: 10.1080/17451000.2012.708046

Pedersen SA, Håkedal OJ, Salaberria I, Tagliati A, Gustavson LM, Jenssen BM, Olsen AJ, Altin D (2014) Multigenerational exposure to ocean acidification during food limitation reveals consequences for copepod Scope for Growth and vital rates. Environ Sci Technol 48:12275–12284 doi: 10.1021/es501581j

Pelejero C, Calvo E, McCulloch M, Marshall J, Gagan M, Lough J, Opdyke B (2005) Preindustrial to modern decadal variability in coral reef pH. Science (80) 309:2204– 2207 doi: 10.1126/science.1113692

Perera A, Davis P, Robinson JJ (2004) Functional role of a high molecular mass protein complex in the sea urchin yolk granule. Dev Growth Differ 46:201–211 doi: 10.1111/j.1440-169X.2004.00737.x

Pespeni MH, Sanford E, Gaylord B, Hill TM, Hosfelt JD, Jaris HK, LaVigne M, Lenz EA, Russell AD, Young MK, Palumbi SR (2013) Evolutionary change during experimental ocean acidification. Proc Natl Acad Sci 110:6937–6942 doi: 10.1073/pnas.1220673110

Pfennig D, Wund M, Snell-Rood E, Cruickshank T, Schlichting C, Moczek A (2010) Phenotypic plasticity’s impacts on diversification and speciation. Trends Ecol Evol 25:459–467 doi: 10.1016/J.TREE.2010.05.006

Pigliucci M (2001) Phenotypic plasticity: beyond nature and nurture. Johns Hopkins University Press.

Piñones A, Fedorov A V (2016) Projected changes of Antarctic krill habitat by the end of the 21st century. Geophys Res Lett 43: doi: 10.1002/2016GL069656

257

Pistevos JCA, Calosi P, Widdicombe S, Bishop JDD, Pistevos JCA, Calosi P (2011) Will variation among genetic individuals infl uence species responses to global climate change? Oikos 120:675–689 doi: 10.1111/j.1600-0706.2010.19470.x

Poloczanska ES, Brown C, Sydeman W, Kiessling W, Schoeman D, Moore P, Brander K, Bruno J, Buckley L, Burrows M, Duarte C, Halpern B, Holding J, Kappel C, O’Connor M, Pandolfi J, Parmesan C, Schwing F, Thompson S, Richardson A (2013) Global imprint of climate change on marine life. Nat Clim Chang 3:1–7 doi: 10.1038/nclimate1958

Popova E, Yool A, Byfield V, Cochrane K, Coward AC, Salim SS, Gasalla MA, Henson SA, Hobday AJ, Pecl GT, Sauer WH, Roberts MJ (2016) From global to regional and back again: common climate stressors of marine ecosystems relevant for adaptation across five ocean warming hotspots. Glob Chang Biol 22:2038–2053 doi: 10.1111/gcb.13247

Pörtner HO, Peck L, Zielinski S, Conway LZ (1999) Intracellular pH and energy metabolism in the highly stenothermal Antarctic bivalve Limopsis marionensis as a function of ambient temperature. Polar Biol 22:17–30 doi: 10.1007/s003000050386

Pörtner H-O, Axelsson M, Altimiras J, Sandblom E, Claireaux G, Seth H, Olsson C, Dupont S, Ortega-Martinez O, Einarsdottir I, Björnsson BT, Sundell K, Axelsson M (2010) Oxygen- and capacity-limitation of thermal tolerance: a matrix for integrating climate- related stressor effects in marine ecosystems. J Exp Biol 213:881–893 doi: 10.1242/jeb.037523

Pörtner H-O (2008) Ecosystem effects of ocean acidification in times of ocean warming: a physiologist’s view. Mar. Ecol. Prog. Ser. 373:203–218

Pörtner H (2002) Climate variations and the physiological basis of temperature dependent biogeography: systemic to molecular hierarchy of thermal tolerance in animals. Comp Biochem Physiol Part A Mol Integr Physiol 132:739–761 doi: 10.1016/S1095- 6433(02)00045-4

Pörtner H, Farrell A (2008) Physiology and climate change. Science (80- ) 322:690–692 doi: 10.1126/science.1121365

Pouvreau S, Bourles Y, Lefebvre S, Gangery A, Alunno-Bruscia M (2006) Application of a dynamic energy budget model to the Pacific oyster, Crassostrea gigas, reared under various environmental conditions. J Sea Res 56:156–167 doi: 10.1016/J.SEARES.2006.03.007

Pratchett MS (2005) Dynamics of an outbreak population of Acanthaster planci at Lizard Island, northern Great Barrier Reef (1995–1999). Coral Reefs 24:453–462 doi: 10.1007/s00338-005-0006-4

Pratchett MS, Caballes C, Rivera-Posada J, Sweatman H (2014) Limits to understanding and managing outbreaks of crown-of-thorns starfish (Acanthaster spp.). In: Hughes R, Hughes D, Smith P (eds.) Oceanography and Marine Biology: An Annual Review 52. Taylor & Francis, pp 133–2000

258

Prowse TAA, Sewell MA, Byrne M (2007) Fuels for development: evolution of maternal provisioning in asterinid sea stars. Mar Biol 153:337–349 doi: 10.1007/s00227-007- 0809-7

Prowse TAA, Byrne M (2012) Evolution of yolk protein genes in the Echinodermata. Evol Dev 14:139–151 doi: 10.1111/j.1525-142X.2012.00531.x

Przeslawski R, Ahyong S, Byrne M, Wörheide G, Hutchings P (2008) Beyond corals and fish: the effects of climate change on noncoral benthic invertebrates of tropical reefs. Glob Chang Biol 14:2773–2795 doi: 10.1111/j.1365-2486.2008.01693.x

Przeslawski R, Byrne M, Mellin C (2015) A review and meta-analysis of the effects of multiple abiotic stressors on marine embryos and larvae. Glob Chang Biol 21:2122– 2140 doi: 10.1111/gcb.12833

Putnam HM, Gates RD (2015) Preconditioning in the reef-building coral Pocillopora damicornis and the potential for trans-generational acclimatization in coral larvae under future climate change conditions. J Exp Biol. doi: 10.1242/jeb.123018

Putnam HM, Davidson JM, Gates RD (2016) Ocean acidification influences host DNA methylation and phenotypic plasticity in environmentally susceptible corals. Evol Appl 9:1165–1178 doi: 10.1111/eva.12408

Räsänen K, Kruuk LEB (2007) Maternal effects and evolution at ecological time-scales. Funct Ecol 21:408–421 doi: 10.1111/j.1365-2435.2007.01246.x

Raven J, Caldeira K, Elderfield H, Hoegh-Guldberg O, Liss P, Riebesell U, Shepherd J, Turley C, Watson A (2005) Ocean acidification due to increasing atmospheric carbon dioxide. London

Raymond J, Himmelman J, Guderley H (2007) Biochemical content, energy composition and reproductive effort in the broadcasting sea star Asterias vulgaris over the spawning period. J Exp Mar Bio Ecol 341:32–44 doi: 10.1016/J.JEMBE.2006.10.030

Ren J, Schiel D (2008) A dynamic energy budget model: parameterisation and application to the Pacific oyster Crassostrea gigas in New Zealand waters. J Exp Mar Bio Ecol 361:42–48 doi: 10.1016/J.JEMBE.2008.04.012

Reusch TBH (2014) Climate change in the oceans: evolutionary versus phenotypically plastic responses of marine animals and plants. Evol Appl 7:104–122 doi: 10.1111/eva.12109

Reynaud S, Leclercq N, Romaine-Lioud S, Ferrier-Pages C, Jaubert J, Gattuso J-P (2003) Interacting effects of CO2 partial pressure and temperature on photosynthesis and calcification in a scleractinian coral. Glob Chang Biol 9:1660–1668 doi: 10.1046/j.1365- 2486.2003.00678.x

Riebesell U, Fabry VJ, Hansson L, Gattuso J (2010) Guide to best practices for ocean acidification research and data reporting. Luxembourg.

259

Ries JB, Cohen AL, McCorkle DC (2009) Marine calcifiers exhibit mixed responses to CO2- induced ocean acidification. Geology 37:1131–1134 doi: 10.1130/G30210A.1

Ries JB (2011) Skeletal mineralogy in a high-CO2 world. J Exp Mar Bio Ecol. doi: 10.1016/j.jembe.2011.04.006

Rivera C (2015) The developmental responses of two closely-related species of sand dollars, the tropical Arachnoides placenta, and the temperate, Fellaster zelandiae, to varied seawater temperature and pH. Masters thesis, University of Otago.

Rivest EB, Hofmann GE (2014) Responses of the metabolism of the larvae of Pocillopora damicornis to ocean acidification and warming. PLoS One 9:e96172 doi: 10.1371/journal.pone.0096172

Roche RC, Pratchett MS, Carr P, Turner JR, Wagner D, Head C, Sheppard CRC (2015) Localized outbreaks of Acanthaster planci at an isolated and unpopulated reef atoll in the Chagos Archipelago. Mar Biol 162:1695–1704 doi: 10.1007/s00227-015-2708-7

Ross PM, Parker L, Byrne M (2016) Transgenerational responses of molluscs and echinoderms to changing ocean conditions. ICES J Mar Sci J du Cons. doi: 10.1093/icesjms/fsv254

Ross P, Parker LM, O’Connor WA, Bailey E (2011) The impact of ocean acidification on reproduction, early development and settlement of marine organisms. Water 3:1005– 1030

Rowiński PK, Rogell B (2017) Environmental stress correlates with increases in both genetic and residual variances: A meta-analysis of animal studies. Evolution (N Y) 71:1339– 1351 doi: 10.1111/evo.13201

Rühl S, Calosi P, Faulwetter S, Keklikoglou K, Widdicombe S, Queirós AM (2017) Long- term exposure to elevated pCO2 more than warming modifies early-life shell growth in a temperate gastropod. ICES J Mar Sci J du Cons 74:1113–1124 doi: 10.1093/icesjms/fsw242

Runcie DE, Garfield DA, Babbitt CC, Wygoda JA, Mukherjee S, Wray GA (2012) Genetics of gene expression responses to temperature stress in a sea urchin gene network. Mol Ecol 21:4547–62 doi: 10.1111/j.1365-294X.2012.05717.x

Sabine C, Feely R, Gruber N, Key R, Lee K, Bullister J, Wanninkhof R, Wong C, Wallace D, Tilbrook B, Millero F, Peng T, Kozyr A, Ono T, Rios A (2004) The oceanic sink for anthropogenic CO2. Science (80) 305:367–371

Saillant E, Chatain B, Fostier A, Przybyla C, Fauvel C (2001) Parental influence on early development in the European sea bass. J Fish Biol 58:1585–1600 doi: 10.1111/j.1095- 8649.2001.tb02314.x

Sancho LG, Green T, Pintado A (2007) Slowest to fastest: Extreme range in lichen growth rates supports their use as an indicator of climate change in Antarctica. Flora - Morphol Distrib Funct Ecol Plants 202:667–673 doi: 10.1016/J.FLORA.2007.05.005

260

Sanford E, Kelly MW (2011) Local adaptation in marine invertebrates. Ann Rev Mar Sci 3:509–535 doi: 10.1146/annurev-marine-120709-142756

Sarmiento JL, Hughes TM, Stouffer RJ, Manakbe S (1998) Simulated responses of the ocean carbon cycle to anthropogenic climate warming. Nature 393:245–248

Sasse TP, Mcneil BI, Matear RJ, Lenton A (2015) Quantifying the influence of CO2 seasonality on future aragonite undersaturation onset. Biogeosciences 12:6017–6031 doi: 10.5194/bgd-12-5907-2015

Schaum CE, Collins S (2014) Plasticity predicts evolution in a marine alga. Proc R Soc B Biol Sci 281:20141486–20141486 doi: 10.1098/rspb.2014.1486

Schaum C-E, Rost B, Collins S (2016) Environmental stability affects phenotypic evolution in a globally distributed marine picoplankton. ISME J. doi: 10.1038/ismej.2015.102

Scheiner SM, Barfield M, Holt RD (2017) The genetics of phenotypic plasticity. XV. Genetic assimilation, the Baldwin effect, and evolutionary rescue. Ecol Evol 7:8788–8803 doi: 10.1002/ece3.3429

Schlegel P, Binet MT, Havenhand JN, Doyle CJ, Williamson JE (2015) Ocean acidification impacts on sperm mitochondrial membrane potential bring sperm swimming behaviour near its tipping point. J Exp Biol. doi: 10.1242/jeb.114900

Schlegel P, Havenhand JN, Gillings MR, Williamson JE (2012) Individual Variability in Reproductive Success Determines Winners and Losers under Ocean Acidification: A Case Study with Sea Urchins. PLoS One 7:e53118 doi: 10.1371/journal.pone.0053118

Schlichting CD, Pigliucci M (1998) Phenotypic evolution: a reaction norm perspective, 1st edn. Sinauer Associates Incorporated, Sunderland, USA

Schram J, Shoenrock K, McClintock JB, Amsler C, Angus R (2014) Multiple stressor effects of near-future elevated seawater temperature and decreased pH on righting and escape behaviors of two common Antarctic gastropods. J Exp Mar Bio Ecol 457:90–96 doi: 10.1016/J.JEMBE.2014.04.005

Schram JB, Mcclintock JB, Amsler CD, Baker BJ (2015) Impacts of acute elevated seawater temperature on the feeding preferences of an Antarctic amphipod toward chemically deterrent macroalgae. Mar Biol 162:425–433 doi: 10.1007/s00227-014-2590-8

Searle S, Casella G, McCulloch C (1992) Variance Components, 1st edn. John Wiley and Sons, New York, USA

Sewell MA, Hofmann GE (2011) Antarctic echinoids and climate change: A major impact on the brooding forms. Glob Chang Biol. doi: 10.1111/j.1365-2486.2010.02288.x

Shama LNS (2015) Bet hedging in a warming ocean: predictability of maternal environment shapes offspring size variation in marine sticklebacks. Glob Chang Biol 21:4387–4400 doi: 10.1111/gcb.13041

261

Shama LNS, Strobel A, Mark FC, Wegner KM (2014) Transgenerational plasticity in marine sticklebacks: maternal effects mediate impacts of a warming ocean. Funct Ecol 28:1482–1493 doi: 10.1111/1365-2435.12280

Shaw E, McNeil B (2014) Seasonal variability in carbonate chemistry and air–sea CO2 fluxes in the southern Great Barrier Reef. Mar Chem 158:49–58 doi: 10.1016/J.MARCHEM.2013.11.007

Shaw EC, Carpenter RC, Lantz CA, Edmunds PJ (2016) Intraspecific variability in the response to ocean warming and acidification in the scleractinian coral Acropora pulchra. Mar Biol 163:210 doi: 10.1007/s00227-016-2986-8

Shea N, Pen I, Uller T (2011) Three epigenetic information channels and their different roles in evolution. J Evol Biol 24:1178–1187 doi: 10.1111/j.1420-9101.2011.02235.x

Shirayama Y, Thornton H (2005) Effect of increased atmospheric CO2 on shallow water marine benthos. J Geophys Res 110:C09S08 doi: 10.1029/2004JC002618

Skerman T (1958) Seasonal variations in sea-water surface temperatures within New Zealand harbours. New Zeal J Geol Geophys 1:197–218

Sokolova IM (2013) Energy-limited tolerance to stress as a conceptual framework to integrate the effects of multiple stressors. Integr Comp Biol 53:597–608 doi: 10.1093/icb/ict028

Sokolova I, Frederich M, Bagwe R, Lannig G, Sukhotin A (2012) Energy homeostasis as an integrative tool for assessing limits of environmental stress tolerance in aquatic invertebrates. Mar Environ Res 79:1–15 doi: 10.1016/J.MARENVRES.2012.04.003

Somero GN (2011) Comparative physiology: a “crystal ball” for predicting consequences of global change. Am J Physiol Regul Integr Comp Physiol 301:R1-14 doi: 10.1152/ajpregu.00719.2010

Somero GN (2012) The physiology of global change: Linking patterns to mechanisms. Ann Rev Mar Sci 4:39–61 doi: 10.1146/annurev-marine-120710-100935

Somero G (2010) The physiology of climate change: how potentials for acclimatization and genetic adaptation will determine “winners” and “losers.” J Exp Biol 213:912–920 doi: 10.1242/jeb.037473

Sparks K, Foo S, Uthicke S, Byrne M, Lamare MD (2016) Paternal identity influences response of Acanthaster planci embryos to ocean acidification and warming. Coral Reefs 36:325–338

Stewart JR, Lister AM, Barnes I, Dalén L (2010) Refugia revisited: individualistic responses of species in space and time. Proc R Soc Part B Biol Sci 277:661–71 doi: 10.1098/rspb.2009.1272

Stillman JH, Paganini AW (2015) Biochemical adaptation to ocean acidification. J Exp Biol. doi: 10.1242/jeb.115584

262

Stuck E (2014) Extracellular pH changes in Echinoderms as a consequence of ocean acidification. Masters thesis, University of Otago

Stumpp M, Dupont S, Thorndyke MC, Melzner F (2011) CO2 induced seawater acidification impacts sea urchin larval development II: Gene expression patterns in pluteus larvae. Comp Biochem Physiol - A Mol Integr Physiol 160:320–330 doi: 10.1016/j.cbpa.2011.06.023

Stumpp M, Wren J, Melzner F, Thorndyke MC, Dupont ST (2011) CO2 induced seawater acidification impacts sea urchin larval development I: Elevated metabolic rates decrease scope for growth and induce developmental delay. Comp Biochem Physiol Part A Mol Integr Physiol 160:331–340 doi: 10.1016/j.cbpa.2011.06.022

Stumpp M, Trubenbach K, Brennecke D, Hu M, Melzner F (2012) Resource allocation and extracellular acid–base status in the sea urchin Strongylocentrotus droebachiensis in response to CO2 induced seawater acidification. Aquat Toxicol 110–111:194–207 doi: 10.1016/J.AQUATOX.2011.12.020

Stumpp M, Hu MY (2017) pH regulation and excretion in echinoderms. In: Weihrauch D, O’Donnell M (eds.) Acid-Base Balance and Nitrogen Excretion in Invertebrates. Springer, Switzerland, pp 261–273

Suckling CC, Clark MS, Peck LS, Cook EJ (2014) Experimental influence of pH on the early life-stages of sea urchins I: different rates of introduction give rise to different responses. Invertebr Reprod Dev 58:148–159 doi: 10.1080/07924259.2013.875950

Suckling CC, Clark MS, Beveridge C, Brunner L, Hughes AD, Harper EM, Cook EJ, Davies AJ, Peck LS (2014) Experimental influence of pH on the early life-stages of sea urchins II: Increasing parental exposure times gives rise to different responses. Invertebr Reprod Dev. doi: 10.1080/07924259.2013.875951

Suckling CC, Clark MS, Richard J, Morley SA, Thorne MAS, Harper EM, Peck LS (2015) Adult acclimation to combined temperature and pH stressors significantly enhances reproductive outcomes compared to short-term exposures. J Anim Ecol 84:773–784 doi: 10.1111/1365-2656.12316

Sugni M, Mozzi D, Barbaglio A, Bonasoro F, Candia Carnevali MD (2007) Endocrine disrupting compounds and echinoderms: new ecotoxicological sentinels for the marine ecosystem. Ecotoxicology 16:95–108 doi: 10.1007/s10646-006-0119-8

Sui Y, Kong H, Huang X, Dupont S, Hu M, Storch D, Pörtner H-O, Lu W, Wang Y (2016) Combined effects of short-term exposure to elevated CO2 and decreased O2 on the physiology and energy budget of the thick shell mussel Mytilus coruscus. Chemosphere 155:207–216 doi: 10.1016/j.chemosphere.2016.04.054

Sun MM, Huang JH, Jiang SG, Yang Q Bin, Zhou FL, Zhu CY, Yang LS, Su TF (2015) Estimates of heritability and genetic correlations for growth-related traits in the tiger prawn Penaeus monodon. Aquac Res 46:1363–1368 doi: 10.1111/are.12290

263

Sunday JM, Crim RN, Harley CDG, Hart MW (2011) Quantifying rates of evolutionary adaptation in response to ocean acidification. PLoS One 6:1–8 doi: 10.1371/journal.pone.0022881

Sunday JM, Pecl GT, Frusher S, Hobday AJ, Hill N, Holbrook NJ, Edgar GJ, Stuart-Smith R, Barrett N, Wernberg T, Watson RA, Smale DA, Fulton EA, Slawinski D, Feng M, Radford BT, Thompson PA, Bates AE (2015) Species traits and climate velocity explain geographic range shifts in an ocean-warming hotspot. Ecol Lett 18:944–953 doi: 10.1111/ele.12474

Sunday J, Calosi P, Dupont S, Munday P, Stillman J, Reusch T (2014) Evolution in an acidifying ocean. Trends Ecol Evol 29:117–125 doi: 10.1016/J.TREE.2013.11.001

Taberlet P, Cheddadi R (2002) Quaternary refugia and persistence of biodiversity. Science (80- ) 297:2009–2010

Tadros W, Lipshitz H (2009) The maternal-to-zygotic transition: a play in two acts. Development 136:3033–3042 doi: 10.1242/dev.033183

Talmage SC, Gobler CJ (2011) Effects of elevated temperature and carbon dioxide on the growth and survival of larvae and juveniles of three species of northwest Atlantic bivalves. PLoS One 6:e26941 doi: 10.1371/journal.pone.0026941

Taylor JR, Lovera C, Whaling PJ, Buck KR, Pane EF, Barry JP (2014) Physiological effects of environmental acidification in the deep-sea urchin Strongylocentrotus fragilis. Biogeosciences 11:1413–1423 doi: 10.5194/bg-11-1413-2014

Terlizzi A, Anderson MJ, Fraschetti S, Benedetti-Cecchi L (2007) Scales of spatial variation in Mediterranean subtidal sessile assemblages at different depths. Mar. Ecol. Prog. Ser. 332:25–39

Thomsen J, Melzner F (2010) Moderate seawater acidification does not elicit long-term metabolic depression in the blue mussel Mytilus edulis. Mar Biol 157:2667–2676 doi: 10.1007/s00227-010-1527-0

Thor P, Dupont S (2015) Transgenerational effects alleviate severe fecundity loss during ocean acidification in a ubiquitous planktonic copepod. Glob Chang Biol. doi: 10.1111/gcb.12815

Tittensor DP, Baco AR, Hall-Spencer JM, Orr JC, Rogers AD (2010) Seamounts as refugia from ocean acidification for cold-water stony corals. Mar Ecol. doi: 10.1111/j.1439- 0485.2010.00393.x

Todgham AE, Hofmann GE (2009) Transcriptomic response of sea urchin larvae Strongylocentrotus purpuratus to CO2-driven seawater acidification. J Exp Biol 212:2579–2594 doi: 10.1242/jeb.032540

Tonin R, Gerdol R, Tomaselli M, Carbognani M, Petraglia A, Wellstein C (2016) Functional traits variation in snowbed plants along a snow cover gradient as a prediction of adaptation to climate warming. EVS.

264

Tuck I, Hall S, Robertson M, Armstrong E, Basford D (1998) Effects of physical trawling disturbance in a previously unfished sheltered Scottish sea loch. Mar Ecol Prog Ser 162:227

Tyler P, Young C, Clarke A (2000) Temperature and pressure tolerances of embryos and larvae of the Antarctic sea urchin Sterechinus neumayeri (Echinodermata: Echinoidea): potential for deep-sea invasion from high latitudes. Mar Ecol Prog Ser 192:173–180

Ulbricht R, Pritchard A (1972) Effect of Temperature on the metabolic rate of sea urchins. Biol Bull 142:178–185 doi: 10.2307/1540254

Uller T (2008) Developmental plasticity and the evolution of parental effects. Trends Ecol Evol 23:432–438 doi: 10.1016/J.TREE.2008.04.005

Uller T, Nakagawa S, English S (2013) Weak evidence for anticipatory parental effects in plants and animals. J Evol Biol 26:2161–2170 doi: 10.1111/jeb.12212

Uller T, Pen I (2011) A theoretical model of the evolution of maternal effects under parent- offspring conflict. Evolution (N Y) 65:2075–2084

Uthicke S, Liddy M, Nguyen HD, Byrne M (2014) Interactive effects of near-future temperature increase and ocean acidification on physiology and gonad development in adult Pacific sea urchin, Echinometra spp. A. Coral Reefs 33:831–845 doi: 10.1007/s00338-014-1165-y

Uthicke S, Logan M, Liddy M, Francis D, Hardy N, Lamare M (2015) Climate change as an unexpected co-factor promoting coral eating seastar (Acanthaster planci) outbreaks. Sci Rep 5:8402 doi: 10.1038/srep08402

Uthicke S, Ebert T, Liddy M, Johansson C, Fabricius KE, Lamare M (2016) Echinometra sea urchins acclimatized to elevated pCO2 at volcanic vents outperform those under present- day pCO2 conditions. Glob Chang Biol 22:2451–2461 . doi: 10.1111/gcb.13223

Uthicke S, Pecorino D, Albright R, Negri AP, Cantin N, Liddy M, Dworjanyn S, Kamya P, Byrne M, Lamare M (2013) Impacts of ocean acidification on early life-history stages and settlement of the coral-eating sea star Acanthaster planci. PLoS One 8:e82938 doi: 10.1371/journal.pone.0082938

Uthicke S, Schaffelke B, Byrne M (2009) A boom–bust phylum? Ecological and evolutionary consequences of density variations in echinoderms. Ecol Monogr 79:3–24 doi: 10.1890/07-2136.1 van Oppen MJH, Oliver JK, Putnam HM, Gates RD (2015) Building coral reef resilience through assisted evolution. Proc Natl Acad Sci U S A 112:2307–13 doi: 10.1073/pnas.1422301112

Vergés A, Steinberg PD, Hay ME, Poore AGB, Campbell AH, Ballesteros E, Heck KL, Booth DJ, Coleman MA, Feary DA, Figueira W, Langlois T, Marzinelli EM, Mizerek T, Mumby PJ, Nakamura Y, Roughan M, van Sebille E, Gupta A Sen, Smale DA, Tomas F, Wernberg T, Wilson SK, Wilson SK (2014) The tropicalization of temperate marine

265

ecosystems: climate-mediated changes in herbivory and community phase shifts. Proc R Soc Part B Biol Sci 281:20140846 doi: 10.1098/rspb.2014.0846

Villinski JT, Villinski JC, Byrne M, Raff RA (2009) Convergent maternal provisioning and life-history evolution in echinoderms. Evolution (N Y) 56:1764–1775 doi: 10.1554/0014-3820

Visser ME (2008) Keeping up with a warming world; assessing the rate of adaptation to climate change. Proc R Soc Part B Biol Sci 275:649–59 doi: 10.1098/rspb.2007.0997

Waddington C (1942) Canalization of development and the inheritance of acquired characters. Nature 381:563

Walther G-R, Post E, Convey P, Menzel A, Parmesank C, Beebee TJC, Fromentin J-M, Hoegh-Guldberg O, Bairlein F (2002) Ecological responses to recent climate change. Nature 416:389–395

Warner R (1997) Evolutionary ecology: how to reconcile pelagic dispersal with local adaptation. Coral Reefs 16:S115–S120

Warwick R, Clarke K (1993) Comparing the severity of disturbance: a meta-analysis of marine macrobenthic community data. Mar Ecol Prog Ser 93:221–231

Waters JM, Roy MS (2003) Marine biogeography of southern Australia: phylogeographical structure in a temperate sea-star. J Biogeogr 30:1787–1796 doi: 10.1046/j.0305- 0270.2003.00978.x

Watson S-A, Fabricius KE, Munday PL (2017) Quantifying pCO2 in biological ocean acidification experiments: A comparison of four methods. PLoS One 12:e0185469 doi: 10.1371/journal.pone.0185469

Watson S-A, Morley SA, Bates AE, Clark MS, Day R, Lamare MD, Martin S, Southgate P, Tan K, Tyler P, Peck LS (2014) Low global sensitivity of metabolic rate to temperature in calcified marine invertebrates. Oecologia 174:45–54

Watts SA, Lawrence JM (1990) The effect of temperature and salinity interactions on righting, feeding and growth in the sea star Luidia clathrata (SAY). Mar Behav Physiol 17:159–165 doi: 10.1080/10236249009378765

Webster M, Colton M, Darling E, Armstrong J, Pinsky M, Knowlton N, Schindler D (2017) Who should pick the winners of climate change? Trends Ecol Evol 32:167–173 doi: 10.1016/J.TREE.2016.12.007

Webster N, Pantile R, Botté E, Abdo D, Andreakis N, Whalan S (2013) A complex life cycle in a warming planet: gene expression in thermally stressed sponges. Mol Ecol 22:1854– 1868 doi: 10.1111/mec.12213

Wei G, McCulloch M, Mortimer G, Deng W, Xie L (2009) Evidence for ocean acidification in the Great Barrier Reef of Australia. Geochim Cosmochim Acta 73:2332–2346 doi: 10.1016/J.GCA.2009.02.009

266

Wernberg T, Russell B, Moore P, Ling S, Smale D, Campbell A, Coleman M, Steinberg P, Kendrick G, Connell S (2011) Impacts of climate change in a global hotspot for temperate marine biodiversity and ocean warming. J Exp Mar Bio Ecol 400:7–16 doi: 10.1016/J.JEMBE.2011.02.021

Whitman DW, Agrawal AA (2009) What is phenotypic plasticity and why is it important? In: Whitman DW, Ananthakrishnan T (eds.) Phenotypic Plasticity of Insects: Mechanisms and Consequences, 1st edn. CRC Press, Enfield, USA, pp 1–47

Widdows J, Hawkins AJS (1989) Partitioning of Rate of Heat Dissipation by Mytilus edulis into Maintenance, Feeding, and Growth Components. Physiol Zool 62:764–784 doi: 10.1086/physzool.62.3.30157926

Wilson L, Fulton C, Hogg A, Joyce K, Radford B, Fraser C (2016) Climate-driven changes to ocean circulation and their inferred impacts on marine dispersal patterns. Glob Ecol Biogeochem 25:923–939

Winberg G (1960) Rate of metabolism and food requirements of fishes. J Fish Res Board Canada 194:1–239

Wing ASR, Mcleod RJ, Leichter JJ, Frew RD, Lamare MD, Wing SR, Mcleod RJ, Leichter JJ, Frew RD, Lamare MD (2012) Sea ice microbial production supports Ross Sea benthic communities : influence of a small but stable subsidy. Ecology 93:314–323

Wittmann A, Portner H-O (2013) Sensitivities of extant marine taxa to ocean acidification. Nat Clim Chang 3:995–1001

Wolfe K, Graba-Landry A, Dworjanyn S, Byrne M (2015) Larval phenotypic plasticity in the boom-and-bust crown-of-thorns seastar, Acanthaster planci. Mar Ecol Prog Ser 539:179–189 doi: 10.3354/meps11495

Wolfe K, Dworjanyn SA, Byrne M (2013) Effects of ocean warming and acidification on survival, growth and skeletal development in the early benthic juvenile sea urchin (Heliocidaris erythrogramma). Glob Chang Biol 19:2698–2707 doi: 10.1111/gcb.12249

Wolfe K, Graba-Landry A, Dworjanyn SA, Byrne M (2015) Larval Starvation to Satiation: Influence of Nutrient Regime on the Success of Acanthaster planci. PLoS One 10:e0122010 doi: 10.1371/journal.pone.0122010

Wong BBM, Candolin U (2015) Behavioral responses to changing environments. Behav Ecol 26:665–673 doi: 10.1093/beheco/aru183

Wood HL, Spicer JI, Widdicombe S (2008) Ocean acidification may increase calcification rates, but at a cost. Proceedings Biol Sci 275:1767–73 doi: 10.1098/rspb.2008.0343

Woodworth-Jefcoats P, Polovina J, Drazen J (2016) Climate change is projected to reduce carrying capacity and redistribute species richness in North Pacific pelagic marine ecosystems. Glob Chang Biol. doi: 10.1111/gcb.13471

267

Woodworth-Jefcoats PA, Polovina JJ, Drazen JC (2017) Climate change is projected to reduce carrying capacity and redistribute species richness in North Pacific pelagic marine ecosystems. Glob Chang Biol 23:1000–1008 doi: 10.1111/gcb.13471

Wrange A, Andre C, Torbjörn L, Lind U, Blomberg A, Jonsson P, Havenhand JN (2014) Importance of plasticity and local adaptation for coping with changing salinity in coastal areas: a test case with barnacles in the Baltic Sea. BMC Evol Biol 14:156–161

Wray N, Visscher P (2008) Estimating Trait Heritability. Nat Educ 1:29

Wright S (1931) Evolution in Mendelian populations. Genetics 6:97–159

Xu R, Barker M (1990) Laboratory experiments on the effects of diet on the gonad and pyloric caeca indices and biochemical composition of tissues of the New Zealand starfish Sclerasterias mollis (Hutton) (Echinodermata: Asteroidea). J Exp Mar Bio Ecol 136:23–45 doi: 10.1016/0022-0981(90)90098-W

Yamamoto-Kawai M, McLaughlin F, Marmack E, Nishino S, Shimada K (2009) Aragonite undersaturation in the Arctic ocean: Effects of ocean acidification and sea ice melt. Science (80- ) 326:1098–1100 doi: 10.1126/science.1097403

Yuan X, Shao S, Yang X, Yang D, Xu Q, Zong H, Liu S (2016) Bioenergetic trade-offs in the sea cucumber Apostichopus japonicus (Echinodermata: Holothuroidea) in response to CO2-driven ocean acidification. Environ Sci Pollut Res 23:8453–8461 doi: 10.1007/s11356-016-6071-0

Zhang H, Shin PKS, Cheung SG (2015) Physiological responses and scope for growth upon medium-term exposure to the combined effects of ocean acidification and temperature in a subtidal scavenger Nassarius conoidalis. Mar Environ Res 106:51–60 doi: 10.1016/j.marenvres.2015.03.001

Zittier ZMC, Bock C, Lannig G, Pörtner HO (2015) Impact of ocean acidification on thermal tolerance and acid–base regulation of Mytilus edulis (L.) from the North Sea. J Exp Mar Bio Ecol 473:16–25 doi: 10.1016/j.jembe.2015.08.001

268

Appendices

Appendix 10.1: Supplementary Material for Chapter 2

Appendix 10.1a Water chemical and physical parameters averaged for each of three experimental blocks for Acanthaster spp. quantitative genetic experiments. All data displayed are taken from individual measurements except temperature and pH(T) which are block averages (temperature n = 312; pH n = 3) in which case the given values represent mean and range. All measurements taken at the Australian Institute of Marine Science (AIMS) in December 2013. Water chemistry analysed by AIMS as described by Uthicke et al. (2013). DIC = dissolved inorganic carbon; DO = dissolved oxygen

Temp (°C) 27 29 31

CO2 Ambient High Ambient High Ambient High

Block pCO2 (ppm) 491.73 902.97 518.91 950.46 564.53 915.71 One Alkalinity (µmol kg) 2395 2400 2376 2373 2410 2386 DIC (µmol kg) 2095 2168 2049 2172 2080 2146 7.94 ± 7.77 ± 7.99 ± 7.77 ± 7.96 ± 7.77 ± pH(T) 0.03 0.04 0.03 0.05 0.03 0.05 27.8 ± 27.4 ± 29.7 ± 31.1 ± Temp. (°C) 29.7 ± 0.4 30.3 ± 0.4 0.4 0.4 0.5 0.5 DO (% sat.) 97.9 no data 99.5 98.0 98.1 no data

Block pCO2 (ppm) 427.36 904.92 472.59 902.22 540.78 948.71 Two Alkalinity (µmol kg) 2369 2377 2378 2373 2376 2381 DIC (µmol kg) 2051 2155 2034 2154 2040 2142 8.04 ± 7.80 ± 8.03 ± 7.77 ± 7.97 ± 7.76 ± pH(T) 0.03 0.04 0.03 0.05 0.05 0.05 27.1 ± 26.9 ± 28.8 ± 31.2 ± Temp. (°C) 29.1 ± 0.5 30.8 ± 0.4 0.5 0.5 0.4 0.4 DO (% sat.) 98.2 98.1 no data 97.9 98.6 98.2

Block pCO2 (ppm) 487.07 839.54 502.22 904.73 490.88 900.68 Three Alkalinity (µmol kg) 2382 2379 2387 2380 2392 2383 DIC (µmol kg) 2059 2151 2046 2144 2057 2137 7.80 ± 8.03 ± 7.80 ± 8.02 ± 7.79 ± pH(T) 8.0± 0.02 0.05 0.04 0.08 0.05 0.05 27.30 ± 27.30 ± 29.40 ± 28.90 ± 30.90 ± 31.10 ± Temp. (°C) 0.5 0.5 0.6 0.5 0.5 0.3 DO (% sat.) 98.10 98.10 98.20 98.80 98.30 98.50

269

Appendix 10.1b Daily averaged pH(T) over the course of the experiment measured three times daily (morning, midday, evening) using m-cresol purple and an Ocean Optics Jaz hand-held spectrometer with a SC-PH cuvette kit attachment. Data are means calculated from three replicate pH measurements within 24 h. Error bars represent one standard deviation.

8.2

8.1

8.0 450 ppm/ 27 °C 450 ppm/ 29 °C

(T) 450 ppm/ 31 °C 7.9 900 ppm/ 27 °C

pH 900 ppm/ 29 °C 900 ppm/ 31 °C 7.8

7.7

7.6 Dec 6 Dec 7 Dec 8 Dec 9 Dec 10 Dec 11 Day

270

Appendix 10.1c Water temperatures over the course of the experiment. Data points represent means from 312 measurements recorded at five minute intervals using Onset HOBO Pendant data/alarm logger 64K UA-001-64. Error bars represent one standard deviation

450 ppm/ 27 °C 450 ppm/ 29 °C 450 ppm/ 31 °C 31 900 ppm/ 27 °C 900 ppm/ 29 °C 900 ppm/ 31 °C

Temperature (

Dec 6 Dec 7 Dec 8 Dec 9 Dec 10 Dec 11 Date

271

Appendix 10.1d Reference photographs showing developmental morphology of Acanthaster cf solaris in ambient seawater; a 4-cell embryo; b normal and abnormal development 2 h post- fertilisation; c formation of morula and blastulation; d normal cleaved embryo; e abnormal embryo showing outer wall rupture; f abnormal embryo showing internal rupture and asymmetry; g normal gastrula 26 h post-fertilisation; h abnormal gastrula showing asymmetry and ‘grainy’ internal structures. Scale bar = 50 µm in all photos. An individual cleaved embryo was scored as ‘abnormal’ if it violated one or more of these criteria: rough internal radial symmetry; sphericity; cell wall integrity

(internal and external); clear and defined development of internal structures; uniform ‘dark brown’ appearance of all cells, i.e., cells not appearing grainy, light or ‘mottled’. An individual gastrula was scored as ‘abnormal’ if it violated one or more of these criteria: longitudinal symmetry; cell wall integrity (internal and external); clear development of ‘keyhole-shaped’ internal blind-ended gut; size roughly equal to that of the surrounding individuals and those of other families

272

Appendix 10.1e Mean fertilisation rates of Acanthaster spp. subjected to high temperature and high pCO2. 27 °C represents a ‘control’ temperature. Black bars indicate 450 ppm CO2 and grey bars indicate 900 ppm CO2. Error bars represent one standard deviation. Letters indicate levels of significance. N = 81 for all treatments.

273

Appendix 10.1f Analysis by permutational MANOVA (PERMANOVA) performed on Acanthaster spp. gastrulae subjected to two temperatures and three pCO2 levels. Block, temperature and pCO2 were fixed factors, and dam and sire were random factors nested within block. Variance components were calculated using a REML (Restricted Error Maximum Likelihood) analysis. Bold/ underlined values are significant contributions to offspring variation (p<0.05). Sire and dam are both nested within block (bl). NT= not tested (fixed factors only- see Data Analysis section) “Perms” = Permutations. “SS”= sum of squares. “%var.” = % variance explained by that factor. NT= not tested; the percent of variance explained was not calculated for fixed effects but were part of the overall model

Source of variation Gastrula d.f. SS Pseudo-F P-value %var. Fixed Factors explained pCO2 1 889 0.41 0.622 NT Temperature 2 63170 6.54 0.026 NT Temperature x pCO2 2 11151 5.77 0.023 NT Random Factors Block 2 23451 2.74 0.004 17.54 Block x Temperature x pCO2 4 3868 0.97 0.525 4.13 Block x pCO2 2 4302 1.22 0.031 <0.01 Block x Temperature 4 19335 2.48 0.006 2.68 Sire (bl) 9 27538 23.30 0.007 15.05 Dam (bl) 3 5116 1.29 0.303 3.85 Sire(bl) x pCO2 9 9642 1.16 0.40 <0.01 Dam (bl) x pCO2 3 4328 1.56 0.202 0.27 Sire(bl) x Temperature 18 20140 1.11 0.354 <0.01 Dam(bl) x Temperature 6 7447 1.24 0.31 1.33 Sire(bl) x Dam (bl) 9 11958 3.12 0.001 2.37 Sire(bl) x Temperature x pCO2 18 24663 1.01 0.451 2.87 Dam(bl) x Temperature x pCO2 6 6117 0.76 0.689 <0.01 Sire(bl) x Dam(bl) x pCO2 9 8334 2.18 0.60 <0.01 Sire(bl) x Dam(bl) x Temperature 18 18112 2.36 0.001 1.94 Sire(bl) x Dam(bl) x Temperature x pCO2 18 24391 3.18 0.001 22.81 Residual 277 1102 50.84

274

Appendix 10.2: Supplementary Material for Chapter 3

Appendix 10.2a Carbonate chemistry parameters for seawater taken during Arachnoides placenta and Fellaster zelandiae spawning and larval rearing. Data are treatment means +/-

S.E.

Temp (°C) 29 32

CO2: Ambient High Ambient High

Arachnoides pCO2 (ppm) 519.45 (+/-0.67) 965.50(+/-0.41) 523.61(+/-0.20) 992.99(+/0.89) placenta Alkalinity (µmol kg) 2395.0 (+/-0.85) 2400.4 (+/-0.63) 2394.6 (+/-0.21) 2409.0(+/0.82) DIC (µmol kg) 2110 (+/-1.20) 2234 (+/-1.08) 2085(+/-0.67) 2206(+/-0.97)

pH(T) 8.11(+/-0.01) 7.88(+/-0.01) 8.11(+/-0.02) 7.82 (+/-0.02) Temp. (°C) 28.10 (+/-0.04) 27.90 (+/-0.10) 31.80 (+/-0.06) 31.75 (+/-0.1)

Temp (°C) 16 19

CO2: Ambient High Ambient High

Fellaster pCO2 (ppm) 457.23 (+/-0.33) 822.19 (+/-0.15) 457.23 (+/-0.40) 893.66(+/0.44) zelandiae Alkalinity (µmol kg) 2271.4 (+/-0.78) 2269.1(+/-0.70) 2275.4 (+/-0.36) 2259.9 (+/1.9) DIC (µmol kg) 2117.5 (+/-1.01) 2230.8 (+/-0.22) 2109.6 (+/-0.66) 2178.4 (+/-12)

pH(T) 8.13 (+/- 0.04) 7.81 (+/-0.02) 8.16 (+/-0.03) 7.81 (+/-0.01) Temp. (°C) 16.1 (+/-0.30) 16.5 (+/-0.01) 18.65 (+/- 0.40) 18.4 (+/-0.40)

275

Appendix 10.2b Results of PERMANOVA analysis performed on data from pluteus stage larvae of sand dollars Arachnoides placenta and Fellaster zelandiae subjected to two temperature and two pCO2 levels. Block, temperature and pCO2 were fixed factors, and dam and sire were random factors nested within block (bl). Variance components were calculated using a REML (Restricted Error Maximum Likelihood) analysis. Bold/ underlined values are significant contributions to offspring variation (p<0.05). “%var.” = % variance explained by that factor. NT= not tested; the percent of variance explained was not calculated for fixed effects but were part of the overall model.

Source of Variation Arachnoides placenta 2-arm pluteus %var. Fixed Factors d.f. Pseudo-F P-value explained Temperature 1 0.66997 0.642 NT pCO2 1 2.8131 0.068 NT pCO2 x Temperature 1 0.52601 0.724 NT

Block (Bl) 4 2.062 0.129 6.44 Block x pCO2 4 2.1596 0.042 4.64 Block x Temperature 4 1.041 0.483 Dam (Bl) 5 1.0652 0.047 4.73 Sire (Bl) 5 0.41269 0.044 1.83 pCO2 x Dam (Bl) 5 0.63715 0.721 3.25 pCO2 x Sire (Bl) 5 0.41897 0.865 2.14 Temperature x Dam (Bl) 5 1.0586 0.048 6.35 Temperature x Sire (Bl) 5 0.63206 0.017 3.79 Dam (Bl) x Sire (Bl) 5 3.5386 0.002 4.43 Block x pCO2 x Temperature 4 1.3465 0.032 7.14 pCO2 x Temperature x Dam (Bl) 5 1.4305 0.036 7.47 pCO2 x Temperature x Sire (Bl) 5 0.58324 0.072 3.05 pCO2 x Dam (Bl) x Sire (Bl) 5 4.0794 0.001 5.11 Temperature x Dam (Bl) x Sire (Bl) 5 4.7901 0.001 5.99 pCO2 x Temperature x Dam (Bl) x Sire (Bl) 5 4.1713 0.001 11.91 Residual 131 32.81

276

Fellaster zelandiae Source of Variation 2-arm pluteus %var. Fixed Factors d.f. Pseudo-F P-value explained Temperature 1 3.5879 0.15 NT pCO2 2 3.2236 0.033 NT Temperature x pCO2 2 2.0339 0.22 NT Random Factors Block 3 1.9782 0.071 Sire (bl) 12 1.0746 0.439 0.18 Dam (bl) 6 0.31819 0.944 1.03 Block x Temperature 3 2.6361 0.36 3.93 Block x pCO2 6 2.5068 0.011 6.56 Sire (bl) x Temperature 12 1.4514 0.212 1.74 Dam (bl) x Temperature 6 1.0262 0.422 0.06 Sire (bl) x pCO2 24 1.9363 0.028 4.27 Dam (bl) x pCO2 12 1.86 0.066 2.45 Sire (bl) x Dam (bl) 18 1.3178 0.241 1.45 Block x Temperature x pCO2 6 3.5209 0.002 17.94 Sire (bl) x Temperature x pCO2 24 1.2843 0.237 3.12 Dam (bl) x Temperature x pCO2 12 1.2971 0.264 2.03 Sire (bl) x Dam (bl) x Temperature 18 1.056 0.415 0.51 Sire (bl) x Dam (bl) x pCO2 36 0.8317 0.697 2.30 Sire (bl) x Dam (bl) x Temperature x pCO2 36 2.8854 0.001 24.21 Residual 36 27.39

Appendix 10.3: Supplementary Material for Chapter 4 Appendix 10.3a Detailed carbonate chemistry parameters for aquaria during the 12-month exposure period, taken every 3 months.

JAN pH(T) Temp. (°C) Alkalinity (µmol kg-1) DIC (µmol kg-1) pCO2 (ppm)

Control 8.12 16.10 2271.40 1986.21 473.11

Control +2.5 °C 8.14 16.10 2269.10 1978.43 341.40 pH 7.90 8.10 16.10 2260.30 2063.24 339.04 pH 7.90 +2.5 °C 8.08 16.10 2275.40 2091.18 436.29 pH 7.70 8.11 16.10 2259.90 2076.53 476.43 pH 7.70 +2.5 °C 8.14 16.10 2248.60 2077.85 401.70

277

JUN pH(T) Temp. (°C) Alkalinity (µmol kg-1) DIC (µmol kg-1) pCO2 (ppm)

Control 8.11 8.23 2275.7 1963.07 375.4333

Control +2.5 °C 8.07 10.84 2282.90 1984.12 320.14 pH 7.90 7.81 8.18 2299.10 2102.84 402.39 pH 7.90 +2.5 °C 7.79 10.61 2276.20 2125.54 473.65 pH 7.70 7.61 8.10 2278.20 2177.86 685.93 pH 7.70 +2.5 °C 7.63 10.9337 2282.00 2159.89 726.67

OCT pH(T) Temp. (°C) Alkalinity (µmol kg-1) DIC (µmol kg-1) pCO2 (ppm)

Control 8.15 12.09 2270.1 1937.76 300.47

Control +2.5 °C 8.11 14.3583 2276.00 1967.23 337.92 pH 7.90 7.85 12.06 2283.60 2107.84 690.71 pH 7.90 +2.5 °C 7.83 13.413 2247.60 2082.31 716.25 pH 7.70 7.65 12.0864 2271.90 2175.59 1150.70 pH 7.70 +2.5 °C 7.67 14.6542 2273.10 2169.57 1094.51

DEC pH(T) Temp. (°C) Alkalinity (µmol kg-1) DIC (µmol kg-1) pCO2 (ppm)

Control 8.14 14.37 2234.00 1911.48 304.10

Control +2.5 °C 8.10 16.56 3347.10 2943.23 318.60 pH 7.90 7.84 14.50 2275.50 2104.53 706.56 pH 7.90 +2.5 °C 7.82 16.51 2279.60 2117.02 746.02 pH 7.70 7.64 14.21 2291.00 2197.78 1190.21 pH 7.70 +2.5 °C 7.66 16.59 2287.10 2186.81 1129.62

278

Appendix 10.3b Body weight indices for adult Patiriella regularis after exposure to combinations of high temperature and low pH for 12 months and at the same time for the wild population.

Mean wet S.D. wet Mean dry S.D. dry Mean ash- S.D. ash- weight (g) weight weight (g) weight free weight free weight (g) Wild 14.27 6.65 4.29 1.52 2.88 0.88 Control 13.80 8.91 3.60 2.30 1.83 0.24 Control + 2.5 °C 8.79 3.04 2.36 0.78 1.87 0.05 pH 7.90 10.79 3.53 3.16 0.66 2.16 0.59 pH7.90 + 2.5 °C 9.61 2.70 2.97 0.85 2.15 0.56 pH 7.70 12.10 3.21 3.34 0.10 2.42 0.03 pH7.70 +2.5 °C 11.84 2.96 2.89 0.74 2.09 0.46

Appendix 10.3 c Ingestion, excretion and energy absorption of Parapercis colias muscle tissue by Patiriella regularis over 12 months by populations acclimated to different combinations of high temperature and low pH. Data given are means (n=5 for all laboratory treatments, n=15 for all wild populations at all time-points).

Energy Ingested (J g dw-1) Energy in faeces (J g dw-1) Absorption Efficiency (%) Treatment Jan Jun Oct Dec Jan Jun Oct Dec Jan Jun Oct Dec Wild 20.36 12.48 19.28 26.65 10.60 7.31 14.36 21.29 52.05 58.54 74.48 79.90 Control 12.40 5.14 18.25 17.64 7.96 3.09 11.37 12.03 64.11 60.27 62.32 68.21 Control +2.5 14.59 13.52 18.45 16.67 9.66 5.42 12.60 11.23 68.21 69.33 68.25 67.34 °C pH 7.90 10.69 8.40 12.60 18.75 7.59 5.39 8.60 12.60 71.00 64.10 68.25 67.21 pH 7.90 +2.5 12.69 11.99 9.35 12.35 8.38 7.70 6.44 8.40 66.00 64.17 68.89 68.04 °C pH 7.70 14.32 13.33 12.68 14.56 8.81 7.40 5.98 6.53 61.49 55.50 47.17 44.82 pH 7.70 +2.5 12.65 15.88 12.64 14.65 6.88 8.40 5.96 10.60 54.40 52.93 51.46 49.99 °C

279

Appendix 10.4: Supplementary Material for Chapter 5

Appendix 10.4a Body weight indices for adult Odontaster validus after exposure to combinations of high temperature and low pH for 12 months and at the same time for the wild population.9 months.

Mean wet S.D. wet Mean dry S.D. dry Mean ash- S.D. ash- weight (g) weight weight (g) weight free weight free weight (g) Control 9.81 0.38 3.08 0.26 2.43 0.24 Control + 2.5 °C 12.34 0.28 3.83 0.09 1.17 0.05 pH 7.70 14.02 0.35 4.39 0.20 1.35 0.11 pH7.70 +2.5 °C 13.96 0.32 4.04 0.18 1.24 0.10

Appendix 10.4b Ingestion, excretion and energy absorption of Parapercis colias muscle tissue by Odontaster validus over 12 months by populations acclimated to different combinations of high temperature and low pH. Data given are means (n=5).

Feeding rate (mg g dw-1 h-1) Energy in faeces (J mg dw-1) Absorption Efficiency (%) Treatment Dec Mar Jun Aug Dec Mar Jun Aug Dec Mar Jun Aug Control 0. 0.18 0.27 0.34 0.55 0.56 0.58 0.58 64.84 44.74 66.22 53.45 Control +2.5 0.15 0.16 0.23 0.47 0.58 0.56 0.57 0.54 66.15 35.84 74.32 69.23 °C pH 7.70 0.15 0.22 0.21 0.31 0.55 0.58 0.57 0.55 61.20 25.23 30.70 49.74 pH 7.70 +2.5 0.15 0.28 0.41 0.44 0.55 0.59 0.59 0.59 55.06 21.30 65.62 57.12 °C

280

Appendix 10.4c Carbonate chemistry parameters throughout the O. validus adult acclimation period.

DEC pH(T) Temp. (°C) Alkalinity (µmol kg-1) DIC (µmol kg-1) pCO2 (ppm)

Control 8.20 0.56 2276.65 2276.6 428.31

Control +2.5 °C 8.18 1.12 2268.53 2253.9 402.26 pH 7.70 7.98 3.88 2268.99 2180.5 916.61 pH 7.70 +2.5 °C 7.98 3.41 2769.73 2217.2 961.43

MAR pH(T) Temp. (°C) Alkalinity (µmol kg-1) DIC (µmol kg-1) pCO2 (ppm)

Control 8.22 0.40 2251.56 2038.3 437.09

Control +2.5 °C 8.08 0.51 2273.36 1995.3 416.55 pH 7.70 7.75 2.89 2273.64 2018.2 1645.93 pH 7.70 +2.5 °C 7.78 3.01 2300.22 2079.6 1420.99

JUN pH(T) Temp. (°C) Alkalinity (µmol kg-1) DIC (µmol kg-1) pCO2 (ppm)

Control 8.21 0.50 2289.32 2166.1 403.04

Control +2.5 °C 8.10 1.05 2271.04 1988.6 404.90 pH 7.70 7.69 3.46 2232.07 1973.0 1100.87 pH 7.70 +2.5 °C 7.71 3.47 2249.50 2185.6 1640.49

AUG pH(T) Temp. (°C) Alkalinity (µmol kg-1) DIC (µmol kg-1) pCO2 (ppm)

Control 8.19 1.10 2199.90 2254.4 451.65

Control +2.5 °C 8.12 1.03 2355.51 2165.8 408.33 pH 7.70 7.71 3.06 2273.36 2265.5 1490.66 pH 7.70 +2.5 °C 7.72 3.69 2270.49 1941.3 1341.21

281

Appendix 10.5: Supplementary Material for Chapter 6

Appendix 10.5a Analysis by permutational MANOVA (PerMANOVA) performed on gastrula-stage Patiriella regularis embryos subjected to two temperatures and three pCO2 levels. Block, temperature and pCO2 were fixed factors, and dam and sire were random factors nested within block (bl). Variance components were calculated using a REML (Restricted Error Maximum Likelihood) analysis. Bold/ underlined values are significant contributions to offspring variation (p<0.05). “%var.” = % variance explained by that factor. NT= not tested; the percent of variance explained was not calculated for fixed effects but were part of the overall model.

Patiriella regularis Source of Variation Gastrulae %var. Fixed Factors d.f. Pseudo-F P-value explained pCO2 2 7.90 0.01 NT Temperature 1 10.21 0.04 NT Temperature x pCO2 2 0.34 0.86 NT Random Factors Block 2 11.40 0.001 14.34 Dam (bl) 5 4.47 0.04 3.98 Sire (bl) 11 1.21 0.39 1.13 Block x pCO2 4 0.60 0.88 0.85 Block x Temperature 2 0.87 0.57 1.40 Dam (bl) x pCO2 8 2.63 0.05 1.60 Sire (bl) x pCO2 18 1.33 0.30 3.63 Dam (bl) x Temperature 8 2.96 0.01 1.84 Sire (bl) x Temperature 9 0.66 0.74 1.84 Sire (bl) x Dam (bl) 5 0.78 0.62 0.52 Block x Temperature x pCO2 4 1.75 0.13 4.46 Dam (bl) x Temperature x pCO2 4 2.88 0.05 1.62 Sire (bl) x Temperature x pCO2 18 1.94 0.10 0.93 Sire (bl) x Dam (bl) x pCO2 10 1.13 0.31 1.52 Sire (bl) x Dam (bl) x Temperature 5 2.32 0.02 1.55 Sire (bl) x Dam (bl) x Temperature x pCO2 10 1.05 0.04 24.27 Residual 228 30.52

282

Appendix 10.6: Supplementary Material for Chapter 7

Appendix 10.6a Contributions of fixed and random factors to variation at the gastrula stage for O. validus embryos from (Table A) wild population and (B) lab-acclimated population. Bold/ underlined values are significant contributions to offspring variation (p<0.05). “%var.” = % variance explained by that factor. NT= not tested; the percent of variance explained was not calculated for fixed effects but were part of the overall model. TABLE A: WILD PARENTS Gastrulae Source of Variation (wild parents) Fixed Factors % var. d.f. Pseudo-F P-value explained Temperature 1 4.47 0.04 NT pCO2 2 0.29 0.92 NT Temperature x pCO2 2 1.10 0.38 NT Random Factors Block 3 0.90 0.57 4.28 Sire (bl) 11 0.52 0.91 1.63 Dam (bl) 5 4.89 0.001 5.79 Block x pCO2 6 2.52 0.01 5.23 Block x Temperature 3 2.13 0.16 1.33 Sire (bl) x pCO2 16 1.52 0.11 2.80 Dam (bl) x pCO2 6 2.12 0.04 1.47 Sire (bl) x Temperature 8 1.19 0.36 2.71 Dam (bl) x Temperature 9 7.24 0.001 6.22 Sire (bl) x Dam (bl) 7 3.91 0.001 2.71 Block x Temperature x pCO2 6 0.90 0.62 1.37 Sire (bl) x Temperature x pCO2 16 0.75 0.81 2.05 Dam (bl) x Temperature x pCO2 6 1.87 0.04 1.94 Sire (bl) x Dam (bl) x pCO2 14 1.16 0.25 1.60 Sire (bl) x Dam (bl) x Temperature 7 2.84 0.001 19.67 Sire(bl) x Dam (bl) x Temperature x pCO2 14 1.73 0.01 21.68 Residual 177 17.51

283

TABLE B: ACCLIMATED PARENTS Gastrulae Source of Variation (acclimated parents) Fixed Factors % var. df Pseudo-F P-value explained Temperature 1 4.245 0.045 NT pH 1 7.265 0.049 NT Temperature x pH 3 6.21 0.047 NT Random Factors Block 5 2.403 0.555 2.235 Parent Acclimation Treatment (Par. Acclim. Tr.) 11 8.191 0.050 12.152 Sire (bl) 7 0.902 0.559 2.231 Dam (bl) 6 12.513 0.014 23.273 Block x pCO2 6 1.109 0.423 1.398 Block x Temperature 13 1.341 0.347 1.664 Par. Acclim. Tr. x pH 9 1.746 0.081 1.550 Par. Acclim. Tr. x Temperature 9 2.214 0.021 1.994 Sire (bl) x pH 13 0.935 0.513 3.369 Dam (bl) x pH 9 0.660 0.676 1.784 Sire (bl) x Temperature 20 1.501 0.299 3.146 Dam (bl) x Temperature 12 11.518 0.005 18.100 Sire (bl) x Dam (bl) 18 0.647 0.710 2.480 Temperature x pH x Par. Acclim. Tr. 18 1.803 0.177 3.837 Sire (bl) x Temperature x pH 18 0.805 0.609 3.086 Dam (bl) x Temperature x pH 18 0.866 0.561 0.249 Sire (bl) x Dam (bl) x pH 18 0.939 0.530 3.601 Sire (bl) x Dam (bl) x Temperature 18 0.546 0.785 2.095 Sire (bl) x Dam (bl) x Par. Acclim. Tr. 23 0.881 0.049 3.335 Sire (bl) x Dam (bl) x Temperature x pH x Par. Acclim. Tr. 24 14.252 0.054 8.114 Residual 110 3.790

284