Are global warming and ocean acidification conspiring against marine ectotherms? A meta-analysis of the respiratory effects of elevated temperature, high CO2 and their interaction

Abstract

With the occurrence of global change, research aimed at estimating the performance of marine ectotherms in a warmer and acidified future has intensified. The concept of oxygen- and capacity-limited thermal tolerance, which is inspired by the Fry paradigm of a bell-shaped increase–optimum–decrease-type response of aerobic scope to increasing temperature, but also includes proposed negative and synergistic effects of elevated CO₂ levels, has been suggested as a unifying framework. The objectives of this meta-analysis were to assess the following: (i) the generality of a bell-shaped relationship between absolute aerobic scope (AAS) and temperature; (ii) to what extent elevated CO₂ affects resting oxygen uptake MO2rest and AAS; and (iii) whether there is an interaction between elevated temperature and CO₂. The behavioural effects of CO₂ are also briefly discussed. In 31 out of 73 data sets (both acutely exposed and acclimated), AAS increased and remained above 90% of the maximum, whereas a clear thermal optimum was observed in the remaining 42 data sets. Carbon dioxide caused a significant rise in MO2rest in only 18 out of 125 data sets, and a decrease in 25, whereas it caused a decrease in AAS in four out of 18 data sets and an increase in two. The analysis did not reveal clear evidence for an overall correlation with temperature, CO₂ regime or duration of CO₂ treatment. When CO₂ had an effect, additive rather than synergistic interactions with temperature were most common and, interestingly, they even interacted antagonistically on MO2rest and AAS. The behavioural effects of CO₂ could complicate experimental determination of respiratory performance. Overall, this meta-analysis reveals heterogeneity in the responses to elevated temperature and CO₂ that is not in accordance with the idea of a single unifying principle and which cannot be ignored in attempts to model and predict the impacts of global warming and ocean acidification on marine ectotherms.

Introduction

The influence of environmental temperature on the physiology of aquatic ectothermic animals has been extensively studied and, in most cases, the results are consistent with predictable effects of temperature on biological and chemical processes (e.g. Bělehrádek, 1930; Dell et al., 2011). With the realization that Earth’s climate is changing—that global warming is happening—the focus on temperature in animal physiology research has obviously not subsided. One aspect that has received particular attention in this context is the capacity for aerobic metabolism, also referred to as scope for activity or simply aerobic scope. The absolute aerobic scope (AAS) is defined as the difference between the minimal and maximal rates of aerobic metabolism (Fry, 1971). Given that most processes occurring in an animal require ATP and that the most efficient ATP-producing pathway requires oxygen, AAS is regarded as an indicator of whole-animal performance that links directly to the capacity for activity, growth and reproduction and, thereby, ultimately fitness (Pörtner and Knust, 2007; Wang and Overgaard, 2007). Furthermore, the aerobic capacity is hypothesized to have an optimal temperature, below and above which AAS is reduced (Fry and Hart, 1948; Fry, 1971), and it follows then that the optimal temperature for AAS (T_optAAS) will coincide with the optimum for other measures of performance and thereby overall fitness (T_optFIT). According to this view, the decline in AAS that occurs above T_optAAS (Fig. 1A) is the result of an inability of the cardiorespiratory system to increase the maximal oxygen supply to tissues, $\dot{M} O_{2 \max}$ (Pörtner and Farrell, 2008; Steinhausen et al., 2008; Munday et al., 2012), at a rate that keeps pace with the expected exponential increase in basic oxygen demands (Clarke and Johnston, 1999) reflected by the resting oxygen uptake, $\dot{M} O_{2 rest}$ (Fig. 1A). This is a perfectly valid hypothesis that intuitively makes sense for many physiologists, while also being attractive from a modelling perspective (e.g. Farrell et al., 2008, Del Raye and Weng, 2015).

Proposed effects of temperature and CO2 on aerobic performance. (A) The Fry paradigm (continuous lines), where resting oxygen uptake (M.O2rest) is predicted to increase exponentially with temperature, whereas maximal oxygen uptake (M.O2max) reaches an optimum and then declines. As a result, the absolute aerobic scope (ASS=MO2max−MO2rest) is represented by a bell-shaped curve, and the optimum for AAS (ToptAAS) is assumed to reflect the overall optimum for fitness (ToptFIT). Alternatively (dashed lines), it can be proposed that the increase in M.O2rest may be less pronounced, because of acclimation, and that M.O2max is not limited to the same degree at higher temperature (Clark et al., 2013; Schulte, 2015), resulting in a scope that continues to increase until the critical temperature CTmax (that is, where the animal eventually will die). In this case, it is unlikely that ToptAAS coincides with ToptFIT. (B) Elevated CO2 in the water (PCO2out) has been proposed by some authors to cause a reduction in aerobic scope through ‘reduced tissue functional capacity’ (Pörtner, 2012) or by causing an elevation in M.O2rest (loading stress) and/or a decrease in M.O2max (limiting stress; Heuer and Grosell, 2014). (C) Change in the aerobic scope curve if simultaneous exposure to elevated temperature and hypercapnia has a synergistic effect, where the isolated effect of CO2 (ΔAASCO2) at the thermal optimum is small, or zero, but exacerbated when temperature rises, causing further reduction in aerobic scope (e.g. Pörtner et al., 2005). The combined effect (ΔAAST+CO2) is thus larger than expected from the sum of the two isolated effects (ΔAAST and ΔAASCO2). (D) Alternatively, hypercapnia can cause a reduction in AAS (ΔAASCO2) independent of temperature, and the combined effect of the two stressors is then additive.

Figure 1:

Proposed effects of temperature and CO₂ on aerobic performance. (A) The Fry paradigm (continuous lines), where resting oxygen uptake (⁠ $\dot{M} O_{2 rest}$ ⁠) is predicted to increase exponentially with temperature, whereas maximal oxygen uptake (⁠ $\dot{M} O_{2 \max}$ ⁠) reaches an optimum and then declines. As a result, the absolute aerobic scope (⁠ $ASS = M O_{2 \max} - M O_{2 rest}$ ⁠) is represented by a bell-shaped curve, and the optimum for AAS (T_optAAS) is assumed to reflect the overall optimum for fitness (T_optFIT). Alternatively (dashed lines), it can be proposed that the increase in $\dot{M} O_{2 rest}$ may be less pronounced, because of acclimation, and that $\dot{M} O_{2 \max}$ is not limited to the same degree at higher temperature (Clark et al., 2013; Schulte, 2015), resulting in a scope that continues to increase until the critical temperature CT_max (that is, where the animal eventually will die). In this case, it is unlikely that T_optAAS coincides with T_optFIT. (B) Elevated CO₂ in the water (PCO_2out) has been proposed by some authors to cause a reduction in aerobic scope through ‘reduced tissue functional capacity’ (Pörtner, 2012) or by causing an elevation in $\dot{M} O_{2 rest}$ (loading stress) and/or a decrease in $\dot{M} O_{2 \max}$ (limiting stress; Heuer and Grosell, 2014). (C) Change in the aerobic scope curve if simultaneous exposure to elevated temperature and hypercapnia has a synergistic effect, where the isolated effect of CO₂ (ΔAAS_CO2) at the thermal optimum is small, or zero, but exacerbated when temperature rises, causing further reduction in aerobic scope (e.g. Pörtner et al., 2005). The combined effect (ΔAAS_T_+CO2) is thus larger than expected from the sum of the two isolated effects (ΔAAS_T and ΔAAS_CO2). (D) Alternatively, hypercapnia can cause a reduction in AAS (ΔAAS_CO2) independent of temperature, and the combined effect of the two stressors is then additive.

Open in new tab Download slide

Although it is unlikely that Fry was concerned about the potential implications for global-warming research, his model did form the basis for a more recently developed framework that presents the Fry paradigm and aerobic-scope curve in a perspective more applicable to climate change, namely the concept of ‘oxygen- and capacity-limited thermal tolerance’ (OCLTT; e.g. Pörtner, 2001, 2010, 2014; Pörtner et al., 2004a; Pörtner and Knust, 2007; Pörtner and Lannig, 2009; Bozinovic and Pörtner, 2015). This framework has, for example, formed the basis for conclusions drawn in the most recent report from the Intergovernmental Panel on Climate Change (IPCC), particularly regarding future biodiversity (Pörtner et al., 2014). This framework has been presented as a ‘unifying concept’ (e.g. Farrell, 2016) and is used as an argument to correlate temperature-related shifts in distributions of animals with one specific variable, namely aerobic scope (Pörtner and Farrell, 2008). Indeed, AAS has, for example, been used to predict spawning success or to explain failure in migrating salmon populations (Farrell et al., 2008; Eliason et al., 2011), to compare habitat suitability between native and invasive fish species (Marras et al., 2015), to compare seasons (Cucco et al., 2012) and to predict the impact of climate change on the distribution and abundance of yellowfin tuna (Del Raye and Weng, 2015).

The generality of a compromised oxygen delivery that limits $\dot{M} O_{2 \max}$ at higher temperatures has been questioned, because in some species $\dot{M} O_{2 \max}$ and AAS continue to increase until temperatures close to the critical or incipient lethal limits, CT_max (Fig. 1A; Clark et al., 2013; Ern et al., 2014; Schulte, 2015). To be fair, Farrell (2009) did present a broader variety of AAS curves, and Farrell (2013) also argued that the ‘alternatively’ shaped curve was indeed recognized by Fry. This is true, as Fry (1947) in fact presented unpublished data from thermally acclimated brown bullhead catfish (Ameiurus nebulosus), where AAS clearly did not decline, even at 35°C, which is close to CT_max of that species. This recognition, however, is not apparent from the papers describing the OCLTT hypothesis. Furthermore, it can be questioned whether the underlying assumption that $\dot{M} O_{2 rest}$ increases exponentially is true for all species, when they are given sufficient time to acclimate (e.g. Sandblom et al., 2014). In any event, the result is an AAS curve that continues to increase until just before CT_max. For OCLTT, CT_max is assumed to coincide with the point where AAS is zero, anaerobic metabolism is assumed to be invoked at this point (e.g. Frederich and Pörtner, 2000; Pörtner, 2001; Sokolova and Pörtner, 2003), and CT_max is thus determined entirely by the capacity for oxygen delivery (van Dijk et al., 1999). However, this view has been challenged (e.g. Ern et al., 2014, 2015; Wang et al., 2014; Verberk et al., 2016), and it has been suggested that CT_max may be limited by other processes, such as the loss of neurological function (Prosser and Nelson, 1981; Prosser, 1991; Ern et al., 2015). Nevertheless, an animal can survive at CT_max for only a limited amount of time. Therefore, if AAS continues to increase until very close to CT_max, T_optAAS will be at the very boundary of the thermal window, at temperatures that might not even be within the ecologically relevant range of the animal, and it is likely that the performance of other processes (e.g. growth and reproduction), and thus fitness, have reached their optima at lower temperatures (Clark et al., 2013). Thus, T_optAAS is unlikely to coincide with T_optFIT and is therefore unlikely to guide the distribution of that species.

The Fry paradigm (Fry, 1971) considered temperature a controlling factor, hypoxia a limiting factor, and other factors, such as salinity, masking. The role of temperature and hypoxia in the OCLTT model is based on the Fry paradigm. Fry also considered hypercapnia a limiting factor, but only in special conditions and predominantly in combination with hypoxia. The OCLTT hypothesis, in contrast, incorporates CO₂ more directly, as could be considered necessary from a future perspective involving both continued global warming and ocean acidification (IPCC, 2013). It has been predicted that elevated CO₂ (and thereby reduced pH) will act as a limiting factor much in the same manner as hypoxia (e.g. Pörtner et al., 2005; Pörtner and Farrell, 2008). Briefly, it is proposed that elevated external CO₂ leads to internal accumulation of CO₂, which in turn leads to a reduction in internal pH and reduced ‘tissue functional capacity’ (Pörtner, et al., 2005; Pörtner, 2012; neither provides a more specific definition of the expression) and, lastly, a reduction in whole-animal AAS (Fig. 1B). Physiologically, it is difficult to understand why the relatively modest increase in the partial pressure of CO₂ (PCO₂) that comes with ocean acidification should have an impact on AAS of marine ectotherms. Many of these animals, particularly fish, have evolved mechanisms to maintain internal pH, despite an increase in external PCO₂ and decrease in pH (e.g. Heisler, 1984; Claiborne et al., 2002; Heuer and Grosell, 2014), at levels many times higher (e.g. Brauner and Baker, 2009) than what is projected for the future. It is not obvious why these modest changes should not be compensated for fully, even after prolonged exposure, because studies have shown acclimatory changes in, for example, mRNA expression of many of the channels and transporters involved in acid–base regulation (e.g. Deigweiher et al., 2008; Heuer and Grosell, 2014). In contrast, it has been argued that the new steady-state levels of pH, PCO₂ and HCO₃⁻ resulting from the compensatory regulation may interfere with aerobic capacity (Pörtner et al., 2004b), possibly by reducing $\dot{M} O_{2 \max}$ (limiting stress; Fry, 1971; Farrell, 2016). Although it can be argued that any type of regulation will have an energetic cost (loading stress; Fry, 1971; Farrell, 2016), and hence potentially impair AAS, the question is whether the cost of compensating for the modest changes in external PCO₂ are great enough to cause an energetic deficit. That is to say, why would it cost more to regulate internal pH in tomorrow’s pH 7.8 ocean compared with today’s pH 8.2 ocean? Nonetheless, in the early descriptions of the hypothesis (Pörtner et al., 2005), it was proposed that CO₂ would cause reductions in AAS at the thermal extremes, thereby narrowing the thermal tolerance window (Pörtner, et al., 2005; Metzger et al., 2007; Wittmann and Pörtner, 2013). In other words, it was suggested that elevated temperature and elevated CO₂ would act synergistically (Fig. 1C). The possibility that a reduction in AAS is induced by CO₂ itself has also been included (Pörtner and Farrell, 2008; Pörtner, et al., 2014), in which case the effect of temperature and CO₂ would simply be additive (Fig. 1D).

Several reviews have been written on the subject of ocean acidification and its effect on animal physiology and/or behaviour, either alone (e.g. Heuer and Grosell, 2014; Clements and Hunt, 2015; Nikinmaa and Anttila, 2015) or in combination with elevated temperature (e.g. Hofmann and Todgham, 2010; Harvey et al., 2013; Parker et al., 2013; Wittmann and Pörtner, 2013; Przeslawski et al., 2015), but none of these included, in detail, $\dot{M} O_{2 rest}$ or AAS. With regard to temperature and the shape of the aerobic performance curve, Clark et al. (2013), Schulte (2015) and Farrell (2016) presented excellent discussions of the subject, but with focus on a few examples, and not a quantitative assessment of the available data. Likewise, a recent review by Verberk et al. (2016) limited their discussion of OCLTT to arthropods.

The objective of the present review is therefore to evaluate quantitatively the current knowledge on the effects of temperature, CO₂ and their possible interaction on $\dot{M} O_{2 rest}$ and AAS, to see how well the data fit the predictions described above. Specifically, the following questions are asked. Does AAS in general follow a bell-shaped curve, i.e. is there always an optimal temperature? Does CO₂ in general cause an increase in $\dot{M} O_{2 rest}$ ⁠, and thereby, reduce AAS? And is the combined effect of CO₂ and temperature on $\dot{M} O_{2 rest}$ and AAS generally larger than expected from their sum, i.e. is the interaction synergistic? In addition, the behavioural alterations caused by CO₂, and implications thereof, are briefly discussed. It is important to clarify that this review is not about proving or disproving the OCLTT hypothesis per se, because that would require more studies with a mechanistic approach than what is currently available, but rather put it into the perspective of the data that exist and discuss the implications for future research on how global change may affect animal physiology.

General approach

To answer the questions outlined above, a meta-analysis approach was adopted, similar to analyses by Harvey et al. (2013), Przeslawski et al. (2015) and Steckbauer et al. (2015). The log response ratio (lnRR) was chosen because it is intuitive, while also ensuring that effect sizes for different data sets are spread more evenly along a scale, making it easier to visualize graphically. A meta-analysis in the strictest sense has as output a single mean effect size for a given variable (Harvey et al., 2013), or even combining several variables (Przeslawski et al., 2015). This mean effect size can then be tested statistically against a hypothetical value (commonly zero), or between different animal groups and life stages. Although this approach is attractive because it gives straightforward ‘yes or no’ answers, it also introduces the risk that potentially interesting patterns and range of responses are overlooked. Given that many species are studied (unlike the situation in medical meta-analyses), the breadth of responses may reflect biological diversity. I therefore chose to examine the diversity of responses by presenting data graphically, rather than focusing only on mean effect sizes.

Three categories of studies were considered for the analyses. To be included, studies had to present data on one or more of the following three overall categories.

Category A. The AAS, calculated as the difference between $\dot{M} O_{2 rest}$ (fasted animals showing minimal activity) and $\dot{M} O_{2 \max}$ (during swimming or another form of maximal activity, or after being chased or stressed to exhaustion), measured using respirometry, at three or more temperatures. In total, 87 data sets (from 48 papers on 53 species) were included in this category.
Category B. The $\dot{M} O_{2 rest}$ or AAS at two or more CO₂ levels (one control and a minimum of one elevated). In total, 206 data sets on $\dot{M} O_{2 rest}$ (from 78 papers on 83 species) and 20 data sets on AAS (from 14 papers on 16 species) were included in this category.
Category C. The $\dot{M} O_{2 rest}$ or AAS at two or more CO₂ levels (one control and a minimum of one elevated) in combination with two or more temperatures. In total, 70 data sets on $\dot{M} O_{2 rest}$ (from 32 papers on 43 species) and 12 data sets on AAS (from seven papers on eight species) were included in this category.

For studies reporting effects of three (or more) temperatures in combination with CO₂, or vice versa, lnRR was calculated for each control–treatment contrast. As the primary goal of the analysis was not to calculate a mean effect size across studies, but also to explore the diversity and potential causes of variation, this method was chosen over calculation of an aggregate within-study lnRR (Lajeunesse, 2011). The goal was to include studies on both fish and invertebrates, irrespective of time frame, temperature ranges and CO₂ regimes, but only if sample size and some form of variance estimate for each experimental group were reported or could be extracted from figures. The literature on fish, however, turned out to be dominated by studies on teleosts, whereas the literature on invertebrates was dominated by molluscs (gastropods, cephalopods and bivalves), echinoderms, cnidarians and crustaceans. These marine invertebrates are all calcifying to some extent. Data were extracted from papers using WebPlotDigitizer 3.8 (http://arohatgi.info/WebPlotDigitizer/citation.html). Details of each study, raw values (⁠ $\dot{M} O_{2 rest}$ ⁠, $\dot{M} O_{2 \max}$ and AAS for category A, or $\dot{M} O_{2 rest}$ and AAS for categories B and C), calculated lnRR and reference details, are collected in a Microsoft Excel spread sheet made available as online Supplementary Material. Data were plotted and analysed in GraphPad Prism 6.07 (GraphPad Software Inc., www.graphpad.com).

Calculations

Comparison of Fry aerobic-scope curves (category A)

Species differ in their AAS and hence the magnitude of change with temperature, and it was therefore necessary to express the changes with temperature in relative terms to be able to compare the shape of the curve between different species. The relative aerobic scope (RAS), relative resting oxygen uptake (⁠

R \dot{M} O_{2 rest}

⁠) and relative maximal oxygen uptake (⁠

R \dot{M} O_{2 \max}

⁠), at different measurement temperatures (T_i) within each data set, were therefore calculated as a percentage of the maximum:

R X = \frac{X_{T_{i}}}{X_{max}} \times 100,

(1)

where

X_{T_{i}}

is AAS,

\dot{M} O_{2 rest}

\dot{M} O_{2 \max}

at T_i, and X_max is the highest value of AAS,

\dot{M} O_{2 rest}

\dot{M} O_{2 \max}

within each data set. A similar approach has previously been adopted by other authors (Brett, 1971; Tirsgaard et al., 2015a). Furthermore, the temperature coefficient (Q₁₀) for

\dot{M} O_{2 rest}

was calculated for each data set according to equation 2:

Q_{10} = {(\frac{\dot{M} O_{2 rest, 2}}{\dot{M} O_{2 rest, 1}})}^{10 / (T_{2} - T_{1})},

(2)

where

\dot{M} O_{2 rest, 1}

is the resting oxygen uptake at the lowest temperature, T₁, whereas

\dot{M} O_{2 rest, 2}

is the resting oxygen uptake at the temperature where it was highest, T₂, thus excluding the highest temperatures if

\dot{M} O_{2}

was reduced.

In all cases, the raw values from each study were used for calculations, although data were also converted to a standardized unit (milligrams of O₂ per kilogram per hour) and body mass (100 g). The raw as well as standardized values (⁠ $\dot{M} O_{2 rest}$ and $\dot{M} O_{2 \max}$ ⁠) and calculated values (AAS, RAS, $R \dot{M} O_{2 rest}$ and $R \dot{M} O_{2 \max}$ ⁠) are shown in Supplementary Table ST1.

It was then determined for each data set whether a thermal optimum for aerobic scope (T_optAAS) could be identified clearly or not. To use an objective criterion, T_optAAS was considered to be absent in a data set if RAS did not fall below 90%, to take variation and measurement error into account. To examine the effect of acclimation on the outcome, data were first divided into acute studies (0–5.5 days) and acclimation studies (7–365 days), and a two-by-two contingency table of the outcome (‘T_optAAS yes’ vs. ‘T_optAAS no’) and duration (acute vs. acclimation) was analysed by Fisher’s exact test. Additionally, to examine whether the outcome was significantly influenced by the methodology used to estimate $\dot{M} O_{2 \max}$ ⁠, a two-by-two contingency table of outcome (‘T_opt yes’ vs. ‘T_opt no’) and methodology (⁠ $\dot{M} O_{2 \max}$ measured after chasing vs. $\dot{M} O_{2 \max}$ measured during maximal activity) was analysed by Fisher’s exact test. To avoid bias, more than one data set on the same species from the same research group was included only if it differed in outcome or methodology, and the number of data sets used in these analyses (73) was therefore lower than the total number of data sets (87).

To examine whether the rate at which $\dot{M} O_{2 rest}$ increased with temperature affected the presence or absence of T_optAAS, a Mann–Whitney U-test was used to determine whether the mean Q₁₀ of data sets with different outcome (‘T_opt yes’ vs. ‘T_opt no’) and group (acute vs. acclimation) was significantly different. Given that raw data for $\dot{M} O_{2 rest}$ and $\dot{M} O_{2 \max}$ were not presented for three species [hapuku wreckfish (Polyprion oxygeneios), salema (Sarpa salpa) and marbled spinefoot (Signaus rivulatus)], 84 data sets out of 87 were included in this analysis. Additionally, a Mann–Whitney U-test was used to determine whether the mean duration of studies with different outcome (‘T_opt yes’ vs. ‘T_opt no’) and group (acute vs. acclimation) was significantly different. Given that acclimation time was not specified in one case (brown bullhead), 86 data sets out of 87 were included in this analysis.

Isolated effects of elevated CO₂ on resting oxygen uptake and absolute aerobic scope (category B)

For each dataset, the effect size was calculated as the log response ratio, lnRR, between the measured value (Y) of a given variable (⁠

\dot{M} O_{2 rest}

or AAS) in a control condition, Y_ctrl, and an experimental condition, Y_exp, as described by equation 3 (Hedges et al., 1999; Lajeunesse, 2011):

\ln RR = \ln (\frac{Y_{\exp}}{Y_{ctrl}}) .

(3)

For each effect size, where SD is the standard deviation and n is the sample size from each group, the variation of lnRR, v(lnRR), was calculated according to Hedges et al. (1999):

v (l n R R) = \frac{{(S D_{\exp})}^{2}}{n_{\exp} \times {(Y_{\exp})}^{2}} + \frac{{(S D_{ctrl})}^{2}}{n_{ctrl} \times {(Y_{ctrl})}^{2}} .

(4)

The 95% confidence interval (CI) was estimated based on v(lnRR) as suggested by Hedges et al. (1999):

\begin{array}{l} \ln RR - 1.96 \times \sqrt{v (l n R R)} \leq \ln RR \geq \ln RR \\ + 1.96 \times \sqrt{v (\ln RR)} . \end{array}

(5)

Given that there were both positive and negative values for lnRR, it was decided to present all values graphically, as opposed to presenting only a (weighted) mean effect size. Data were divided into different groups of invertebrates (bivalves, cephalopods, cnidarians, crustaceans, echinoderms and gastropods) and fish (teleosts and elasmobranchs) and plotted separately as a function of life stage (adult and non-adult), temperature, PCO₂ in the experimental treatment and length of CO₂ treatment, to investigate the degree of correlation (assessed by linear regression and Pearson’s r) between lnRR and these variables. Data points were also given different symbols to indicate taxonomic class, and different colours to indicate life stages (adult and non-adult). The data sets are available as Supplementary Tables ST2_MO2rest and ST2_AAS.

To examine whether methodology had a significant impact on the outcome, studies on $\dot{M} O_{2 rest}$ were divided into those where $\dot{M} O_{2 rest}$ was measured during a truly resting state (‘resting’) and those where some routine activity could not be ruled out (‘routine’), whereas studies on AAS were divided into those where $\dot{M} O_{2 \max}$ was estimated immediately after exhaustive exercise (‘post-chase’) and those where it was measured during maximal activity (‘during exercise’). The lnRR values in different data sets were categorized as ‘no effect’ if the 95% CI overlapped zero and as ‘effect’ if the 95% CI did not overlap zero. Data sets in the latter group were then categorized further as ‘decrease’ if the lnRR was negative and ‘increase’ if lnRR was positive. More than one data set from the same research group on the same species was included only if it differed in outcome or methodology; therefore, 125 out of 206 data sets on $\dot{M} O_{2 rest}$ and 18 out of 20 data sets on AAS were included in these analyses. The resulting two-by-two contingency tables of the outcomes (effect vs. no effect, and decrease vs. increase) were then analysed using Fisher’s exact test.

Combined effects of elevated temperature and CO₂ (category C)

For both

\dot{M} O_{2 rest}

and AAS in each data set, lnRR was calculated for each of the isolated effects of temperature and CO₂ (lnRR_T and lnRR_CO2, respectively), as well as the combined effect (lnRR_T_+CO2), as described in equations 3, 4 and 5. To examine the nature of the interaction between temperature and CO₂, the expected combined effect size for each data set, assuming a simple additive effect (lnRR_add), was calculated as described by equation 6:

\begin{array}{l} \ln R R_{add} = \ln (\frac{Y_{add}}{Y_{ctrl}}) = \ln (\frac{Y_{ctrl} + (Y_{temp} - Y_{ctrl}) + (Y_{CO 2} - Y_{ctrl})}{Y_{ctrl}}) \\ = \ln (\frac{Y_{temp} + Y_{CO 2} - Y_{ctrl}}{Y_{ctrl}}) . \end{array}

(6)

More specifically, the additive effect was interpreted as the sum of the control value (Y_ctrl), the change caused by temperature (Y_temp − Y_ctrl) and the change caused by elevated CO₂ (Y_CO2 − Y_ctrl). The observed lnRR_T_+CO2 for $\dot{M} O_{2 rest}$ and AAS in each data set could then be plotted as a function of the expected lnRR to visualize the range and direction of effects. From these plots, the presence of a synergistic, additive or antagonistic interaction could be determined for each point (i.e. study or data set), and also from the slope of the line fitted by linear regression (a similar approach was used by Steckbauer et al., 2015). A slope of 1.0 would indicate perfect additivity; a slope <1 would indicate an antagonistic effect, and a slope >1 would indicate synergism. Data from invertebrates and fish were plotted separately, and data points were given different symbols to indicate taxonomic group and different colours to indicate life stages (adult and non-adult). The data sets are available as Supplementary Tables ST3_MO2rest and ST3_AAS.

Results

Aerobic-scope curves

The changes in relative aerobic scope (RAS) with temperature across different gastropod, bivalve and crustacean invertebrates are shown in Fig. 2A and B, whereas the changes in RAS with temperature across different teleost fish species are shown in Fig. 2C–H.

Figure 2:

Relative aerobic scope as a function of temperature in different invertebrates and teleost fish. Aerobic scope was set to 100% at the temperature where it was highest. Only studies where measurements of absolute aerobic scope (AAS, measured as the difference between maximal and minimal oxygen uptake) were conducted at three or more temperatures are included. Data were not available for marine invertebrate groups other than gastropods (upright triangles), crustaceans (inverted triangles), bivalves (hexagons) and cephalopods (circled crosses) and for fish other than teleosts (circles). Continuous lines represent acclimated individuals (minimum of 7 days at each temperature), whereas dashed lines represents acute measurements (from 30 min to 5.5 days). Multiple lines of the same colours represent different studies on the same species or studies on different populations of the same species. Symbol size reflects sample size at each temperature (<5 = small, 5–10 = medium and >10 = large). For most species, the temperature range over which aerobic scope was measured is representative for the temperatures at which the species occurs, or even up to critical temperatures. Species for which the relevance of the tested temperature range is unclear are indicated by ‘•’. Species, where routine rather than resting oxygen uptake was measured are indicated by ‘©’. (A) Invertebrate species that appear to have an optimal temperature for AAS (T_optAAS). (B) Invertebrate species where AAS increases over the entire investigated temperature range. However, in only two cases (Atlantic blue crab and common cockle) was the highest temperature also used close to the upper limit of the species. (C) Studies on cold-water or temperate fish species acutely exposed to elevated temperature. All have a clear T_optAAS. (D) Studies on primarily temperate fishes after acclimation at different temperatures, where T_optAAS is evident. (E) Coral reef fishes that have a much narrower thermal range than the fishes in (D), but also showing declines in AAS with temperature (one exception being the maroon clownfish) and thus clear T_optAAS. Also, there was an intraspecific effect of latitude, with warmer populations (continuous lines; Lizard Island and Papua New Guinea) showing a rightward shift in their thermal range and a greater decline in aerobic scope with temperature compared with the less warm populations (dotted lines; Heron Island). (F) Studies on temperate fish species after acclimation where T_optAAS is not clearly evident. (G) Studies on fish species with a slightly wider temperature range than species in (F), where T_optAAS is not clearly evident. (H) Freshwater fishes where a T_optAAS is not evident, except for mummichog (relative aerobic scope only declines from 100 to 75% at the highest temperature, which is the highest at which fish could be kept without mortality). The complete data set and reference details are available in Supplementary Table ST1. Note that several independent (but very similar) data sets exist for the coral reef fishes, but to enhance visual clarity of the graph only data from Rummer et al. (2014) and Gardiner et al. (2010) are shown.

Open in new tab Download slide

The first group of invertebrates (Fig. 2A) all showed a clear T_optAAS, even when only three temperatures were investigated. It is noteworthy that only one of these studies used thermally acclimated individuals (Shiba shrimp, Metapenaeus joyneri). The second group (Fig. 2B), in contrast, consists of invertebrates where a clear T_optAAS was not identified within the temperature range measured. In these cases, all but two [Atlantic blue crab (Callinectes sapidus) and common periwinkle (Littorina littorea)] were acute studies, and most studies [except common cockle (Cerastoderma edule) and Atlantic blue crab] used a temperature range that might not have reflected the tolerance of the species.

The teleost fish were divided into six panels according to similarities in duration, temperature range or response curves. Two main patterns emerged: species where a T_optAAS was clearly evident (Fig. 2C–E) and species where RAS did not decrease or only slightly so (remaining above 90% in most cases) at the highest temperature (Fig. 2F–H). These two patterns were present across different types of fishes (cold-water, Fig. 2C, D and F; warm-water, Fig. 2E, G and H; and freshwater, Fig. 2H) and different time frames (acute, Fig. 2C, G and H; and acclimated, Fig. 2D–H). Some species showed a clear T_optAAS over a broad temperature range (ΔT = 10–20°C; Fig. 2C and D), whereas others, particularly the coral reef fish and the bald notothen (Pagothenia borchgrevinki), did so over a narrower temperature range (Fig. 2C and E; ΔT = 4–7°C). For the coral reef fish, it was evident that the temperature window of a given species was not fixed, because it was right shifted in populations from warmer areas. Most of the studies where only minor decreases in RAS were observed were carried out on thermally acclimated individuals and covered the relevant temperature range for each species, such that the highest temperature used was slightly below the temperature where mortality occurred or the maximal temperature found in the habitat.

As expected, the relative resting oxygen uptake (⁠ $R \dot{M} O_{2 rest}$ ⁠) increased over the entire temperature range in most species, both invertebrates (Fig. 3A and B) and fish (Fig. 3C–H). In robust shell (Littoraria undulata; Fig. 3A), common cockle (Fig. 3B) and mummichog (Fundulus heteroclitus; Fig. 3D and H), $R \dot{M} O_{2 rest}$ declined at the highest temperature. The Q₁₀ was between 1 and 3 in most studies and was slightly, but significantly, lower in data sets on acclimated individuals in which a T_optAAS was absent (i.e. RAS never decreased below 90%; Fig. 4A; Mann–Whitney U-test, P = 0.0246). The highest Q₁₀ values, above 4 and as high as 7, were found in the warmest coral reef fish populations (Papa New Guinea and Lizard Island), such as yellowstriped cardinalfish (Ostorhinchus cyanosoma), fourline cardinalfish (Ostorhinchus doederleini), lemon damsel (Pomacentrus moluccensis) and black-axil chromis (Chromis atripectoralis), regardless of acclimation time. The cardinalfishes were also the ones showing the largest decrease in RAS (Fig. 2E). Overall, there was no difference in the outcome of the studies (presence vs. absence of a thermal optimum for aerobic scope, T_optAAS) between acute and acclimated (Fisher’s exact test, P = 0.345). However, the mean duration in studies on acclimated individuals where T_optAAS was absent was significantly longer (Fig. 4B; Mann–Whitney U-test, P = 0.0004) than in the studies in which a T_optAAS was identified.

Figure 3:

Relative resting oxygen uptake as a function of temperature in different invertebrates and teleost fish. Resting oxygen uptake was set to 100% at the temperature where it was highest. Data are from the same data sets and grouped as in the panels of Fig. 2. Different symbols represent gastropods (upright triangles), crustaceans (inverted triangles), bivalves (hexagons), cephalopods (circled crosses) and teleost fish (circles), because these were the only taxonomic groups for which data were available. Continuous lines represent acclimated individuals (minimum of 7 days at each temperature), whereas dashed lines represent acute measurements (from 30 min to 5.5 days). Multiple lines of the same colours represent different studies on the same species or studies on different populations of the same species. Symbol size reflects sample size at each temperature (<5 = small, 5–10 = medium and >10 = large). For most species, the temperature range over which aerobic scope was measured is representative for the temperatures at which the species occurs, or even up to critical temperatures. Species for which the relevance of the tested temperature range is unclear are indicated by ‘•’. Species for which routine rather than resting oxygen uptake was measured are indicated by ‘©’. The complete data set and reference details are available in Supplementary Table ST1.

Open in new tab Download slide

Effect of temperature coefficient (Q10) and acclimation time on the presence (yes) or absence (no) of an optimal temperature (ToptAAS) for absolute aerobic scope. (A) Presence or absence of ToptAAS as a function of Q10 for resting oxygen uptake. (B) Presence or absence of ToptAAS as a function of the duration of temperature treatment in studies using acute treatment and acclimation. Continous lines indicate median ± interquartile range. Note the logarithmic scale in (B). Asterisk indicates significant difference in Q10 and acclimation duration, in studies with different outcome for acclimated species (*P < 0.05, ***P < 0.001; see ‘Results’ for details).

Figure 4:

Effect of temperature coefficient (Q₁₀) and acclimation time on the presence (yes) or absence (no) of an optimal temperature (T_optAAS) for absolute aerobic scope. (A) Presence or absence of T_optAAS as a function of Q₁₀ for resting oxygen uptake. (B) Presence or absence of T_optAAS as a function of the duration of temperature treatment in studies using acute treatment and acclimation. Continous lines indicate median ± interquartile range. Note the logarithmic scale in (B). Asterisk indicates significant difference in Q₁₀ and acclimation duration, in studies with different outcome for acclimated species (*P < 0.05, ***P < 0.001; see ‘Results’ for details).

Open in new tab Download slide

Unlike $R \dot{M} O_{2 rest}$ ⁠, but similar to RAS, there were different responses of the relative maximal oxygen uptake (⁠ $R \dot{M} O_{2 \max}$ ⁠) with increased temperature. In most species of both invertebrates (Fig. 5A and B) and fish (Fig. 5C–H), $R \dot{M} O_{2 \max}$ increased over most of the temperature range measured, showing no or only a minor decline at the upper temperature. Other species showed declines in $R \dot{M} O_{2 \max}$ above mid-range temperatures (Fig. 5D). The coral reef fishes (Fig. 5E) again showed the most pronounced declines, despite having a much narrower thermal range, and this was also the group that showed the largest declines in RAS (Fig. 2E). When studies were divided into ones where $\dot{M} O_{2 \max}$ was measured immediately after chasing or other exhaustive exercise (‘post-chase’) and ones where $\dot{M} O_{2 \max}$ was measured during swimming or other maximal activity (‘during exercise’; Fig. 6), there was no difference in the outcome (presence vs. absence of T_optAAS), in studies using either acute exposure (Fisher’s exact test, P = 0.383) or acclimation (Fisher’s exact test, P = 0.336).

Figure 5:

Relative maximal oxygen uptake as a function of temperature in different invertebrates and teleost fish. Maximal oxygen uptake was set to 100% at the temperature where it was highest. Data are from the same data sets and grouped as in the panels of Fig. 2. Different symbols represent gastropods (upright triangles), crustaceans (inverted triangles), bivalves (hexagons) and cephalopods (circled crosses) and teleost fish (circles). Continuous lines represent acclimated individuals (minimum of 7 days at each temperature), whereas dashed lines represent acute measurements (from 30 min up to 5.5 days). Multiple lines of the same colours represent different studies on the same species or studies on different populations of the same species. Symbol size reflects the sample size at each temperature (<5 = small, 5–10 = medium and >10 = large). For most species, the temperature range over which aerobic scope was measured is representative for the temperatures at which the species occurs, or even up to critical temperatures. Species for which the relevance of the tested temperature range is unclear are indicated by ‘•’. Species for which routine rather than resting oxygen uptake was measured are indicated by ‘©’. The complete data set and reference details are available in Supplementary Table ST1.

Open in new tab Download slide

Figure 6:

Effect of methodology on the presence or absence of an optimal temperature for absolute aerobic scope (T_optAAS). Data are the number of data sets with absence or presence of a T_optAAS (‘no’ or ‘yes’) grouped by methodology (‘acute’ or ‘acclimated’, and ‘post-chase’ or ‘exercise’). There was no significant difference in the proportion of ‘yes’ and ‘no’ outcomes between the two methods in either acute or acclimated studies (see Results for details).

Open in new tab Download slide

Isolated and combined effects ’of temperature and CO₂

The mean log response ratio for the isolated effect of temperature (lnRR_T) was significantly larger than zero (as judged by the 95% CI) in both calcifying invertebrates (Fig. 7A; except cephalopods and cnidarians, the groups with fewest studies) and fish (Fig. 7B), although the magnitude of the effect varied (mostly in invertebrates), and there were cases where $\dot{M} O_{2 rest}$ was unchanged. Only in invertebrates were reductions in $\dot{M} O_{2 rest}$ observed with elevated temperature. There was no clear distinction in responses between different taxonomic groups, except that response magnitude and variation seemed smaller within the crustaceans. Generally, the effects of CO₂ on $\dot{M} O_{2 rest}$ were smaller than the effects of temperature, and more evenly distributed between positive and negative effects, generally resulting in a mean lnRR_CO2 that was not different from zero. Exceptions to this included cephalopods and elasmobranchs, where $\dot{M} O_{2 rest}$ was mostly reduced (note that all were non-adults in these two groups), and crustaceans and echinoderms, where $\dot{M} O_{2 rest}$ increased in most cases. In both invertebrates (except cephalopods) and teleosts, the combined effect of temperature and CO₂ on $\dot{M} O_{2 rest}$ resembled the isolated effect of temperature.

Isolated and combined effects of elevated temperature and partial pressure of CO2 (PCO2) on resting oxygen uptake (MO2rest) and absolute aerobic scope (AAS). Data are log response ratios (lnRR) of MO2rest (A and B) and AAS (C and D) in marine invertebrates (A and C) and fish (B and D) to elevated temperature alone (red symbols), elevated PCO2 alone (blue symbols) and the combined treatment (purple symbols). Different shapes indicate different taxonomic groups (upright triangle, gastropod; circled cross, cephalopod; hexagon, bivalve; inverted triangle, crustacean; diamond, echinoderm; cross, cnidarian; square, elasmobranch; and circle, teleost). Note that the invertebrate species are all calcifying because there were no comparable data available for non-calcifying species. Continous lines indicate mean ± 95% confidence intervals. In (D), lnRR from studies investigating CO2 only (indicated by plus signs) have been added to illustrate the variation in responses, because this is not reflected in the studies having investigated both temperature and CO2. The complete data set and reference details are available in Supplementary Table ST3_MO2rest and ST3_AAS.

Figure 7:

Isolated and combined effects of elevated temperature and partial pressure of CO₂ (PCO₂) on resting oxygen uptake (MO2rest) and absolute aerobic scope (AAS). Data are log response ratios (lnRR) of $M O_{2 rest}$ (A and B) and AAS (C and D) in marine invertebrates (A and C) and fish (B and D) to elevated temperature alone (red symbols), elevated PCO₂ alone (blue symbols) and the combined treatment (purple symbols). Different shapes indicate different taxonomic groups (upright triangle, gastropod; circled cross, cephalopod; hexagon, bivalve; inverted triangle, crustacean; diamond, echinoderm; cross, cnidarian; square, elasmobranch; and circle, teleost). Note that the invertebrate species are all calcifying because there were no comparable data available for non-calcifying species. Continous lines indicate mean ± 95% confidence intervals. In (D), lnRR from studies investigating CO₂ only (indicated by plus signs) have been added to illustrate the variation in responses, because this is not reflected in the studies having investigated both temperature and CO₂. The complete data set and reference details are available in Supplementary Table ST3_MO2rest and ST3_AAS.

Open in new tab Download slide

The lnRR for the isolated effect of temperature on AAS in both calcifying invertebrates (Fig. 7C) and fish (Fig. 7D) was variable. In contrast, CO₂ seemed to affect the AAS of invertebrates negatively in some cases, or not at all, whereas the response was more variable for fish (at least if the studies that investigated only CO₂ are also included). For invertebrates, the combined effect of temperature and CO₂ again resembled that of temperature, whereas it was more variable for fish.

Examination of the variability ’in responses to CO₂

The effect of CO₂ (lnRR_CO2) on $\dot{M} O_{2 rest}$ tended to be more variable in adults than non-adults of both calcifying invertebrates (Fig. 8A) and fish (Fig. 8B), although data were lacking for adult cephalopods and elasmobranchs. Overall, differences in life stage explained only 1.75% of the total variation (two-way ANOVA, F_1,86 = 3.58, P = 0.060). The lnRR_CO2 was dominantly negative in non-adult compared with adult bivalves (Sidak’s multiple comparisons test, P = 0.050) but did not differ between life stages of the other taxonomic groups (P > 0.8 for all). The difference between the taxonomic groups explained only 6.92% of the total variation (two-way ANOVA, F_5,186 = 2.84, P = 0.017), as lnRR_CO2 of non-adult bivalves was more negative than that of non-adult gastropods (Sidak’s multiple comparisons test, P = 0.047) but did not differ between other groups of either non-adults (P = 0.223 for teleosts, P > 0.5 for all others) or adults (P > 0.5 for all).

Effect of elevated CO2 on resting oxygen uptake (M.O2rest). Data are log response ratios to CO2 (lnRRCO2) of marine invertebrates (A, C, E and G) and teleosts and elasmobranchs (B, D, F and H) in non-adults (green) and adults (blue) of different taxonomic groups (upright triangle, gastropod; circled cross, cephalopod; hexagon, bivalve; inverted triangle, crustacean; diamond, echinoderm; cross, cnidarian; square, elasmobranchs; and circle, teleost). Data are also depicted as a function of the temperature at which the experiment was conducted (C and D), as a function of the PCO2 used as the experimental treatment (E and F) and as a function of the duration of the CO2 exposure (G and H). Note that the invertebrate species are all calcifying because there were no comparable data available for non-calcifying species. Continous lines indicate mean ± 95% confidence interval (A and B) or fitted lines from linear regression, with dotted lines representing the 95% confidence interval (C–H) (see Results for details). The complete data set and reference details are available in Supplementary Table ST2_MO2rest.

Figure 8:

Effect of elevated CO₂ on resting oxygen uptake (⁠ $\dot{M} O_{2 rest}$ ⁠). Data are log response ratios to CO₂ (lnRR_CO2) of marine invertebrates (A, C, E and G) and teleosts and elasmobranchs (B, D, F and H) in non-adults (green) and adults (blue) of different taxonomic groups (upright triangle, gastropod; circled cross, cephalopod; hexagon, bivalve; inverted triangle, crustacean; diamond, echinoderm; cross, cnidarian; square, elasmobranchs; and circle, teleost). Data are also depicted as a function of the temperature at which the experiment was conducted (C and D), as a function of the PCO₂ used as the experimental treatment (E and F) and as a function of the duration of the CO₂ exposure (G and H). Note that the invertebrate species are all calcifying because there were no comparable data available for non-calcifying species. Continous lines indicate mean ± 95% confidence interval (A and B) or fitted lines from linear regression, with dotted lines representing the 95% confidence interval (C–H) (see Results for details). The complete data set and reference details are available in Supplementary Table ST2_MO2rest.

Open in new tab Download slide

Overall, temperature did not explain much of the variation in lnRR_CO2 of either invertebrates (Fig. 8C) or fish (Fig. 8D). There was a weak positive relationship between lnRR_CO2 and the temperature at which the experiment was conducted in cnidarians (Pearson correlation, r = 0.456; linear regression, R² = 0.208, P = 0.011) and a weak negative relationship in echinoderms (Pearson correlation, r = −0.361; linear regression, R² = 0.131, P = 0.039). When data were pooled across groups and life stages, there was a weak negative relationship between lnRR_CO2 and temperature in invertebrates (Pearson correlation, r = −0.170; linear regression, R² = 0.025, P = 0.029) and a slightly stronger negative relationship in fish (Pearson correlation, r = −0.455; linear regression, R² = 0.207, P = 0.010).

The treatment PCO₂ did not explain much of the variation in lnRR_CO2 of either invertebrates (Fig. 8E) or fish (Fig. 8F), because there was only a weak negative relationship in crustaceans (Pearson correlation, r = −0.494; linear regression, R² = 0.244, P = 0.008) and in invertebrates altogether (Pearson correlation, r = −0.202; linear regression, R² = 0.041, P = 0.008), but not in fish (Pearson correlation, r = −0.040; linear regression, R² = 0.002, P = 0.830). Likewise, lnRR_CO2 seemed to be unrelated to the duration of the CO₂ exposure in both invertebrates (Fig. 8G; Pearson correlation, r = 0.089; linear regression, R² = 0.008, P = 0.242) and fish (Fig. 8H; r = −0.086; R² = 0.007, P = 0.644).

In the studies on the effect of CO₂ on AAS invertebrates, only one was conducted on a non-adult life stage (jumbo squid, Dosidicus gigas), and in this study elevated CO₂ affected AAS negatively, whereas the effect on adults of other groups varied (Fig. 9A). In fish, the non-adult life stages appeared mostly unaffected by elevated CO₂, whereas the responses in adults were both positive and negative (Fig. 9B), although data for adult elasmobranchs were not available. The difference between non-adult and adult teleosts was not significant (Student’s unpaired t-test, P = 0.610). There was no overall relationship between the effect of CO₂ on AAS and the temperature at which the experiment was conducted in either invertebrates (Fig. 9C; Pearson correlation, r = 0.316; linear regression, R² = 0.100, P = 0.542) or fish (Fig. 9D; r = −0.083; R² = 0.007, P = 0.778). Likewise, was there no relationship with the treatment PCO₂ in either invertebrates (Fig. 9E; Pearson correlation, r = −0.151; linear regression, R² = 0.023, P = 0.776) or fish (Fig. 9F; r = −0.165; R² = 0.027, P = 0.572). In invertebrates, there was a small tendency for the effect to diminish with increased duration of the exposure (Fig. 9G; Pearson correlation, r = 0.699; linear regression, R² = 0.489, P = 0.122), but this was not the case in fish (Fig. 9H; r = 0.200; R² = 0.040, P = 0.493).

Effect of elevated CO2 on absolute aerobic scope (AAS). Data are log response ratios to CO2 (lnRRCO2) of calcifying invertebrates (A, C, E and G) and teleosts and elasmobranchs (B, D, F and H) in non-adults (green) and adults (blue) of different taxonomic groups (inverted triangle, crustacean; upright triangle, gastropod; hexagon, bivalve; circled cross, cephalopod; square, elasmobranch; and circle, teleost). Data are also depicted as a function of the temperature at which the experiment was conducted (C and D), as a function of the PCO2 used as the experimental treatment (E and F) and as a function of the duration of the CO2 exposure (G and H). Note that the invertebrate species are all calcifying because there were no comparable data available for non-calcifying species. Continous lines indicate mean ± 95% confidence interval (A and B) or fitted lines from linear regression, with dotted lines representing the 95% confidence interval (C–H) (see Results for details). The complete data set and reference details are available in Supplementary Table ST2_AAS.

Figure 9:

Effect of elevated CO₂ on absolute aerobic scope (AAS). Data are log response ratios to CO₂ (lnRR_CO2) of calcifying invertebrates (A, C, E and G) and teleosts and elasmobranchs (B, D, F and H) in non-adults (green) and adults (blue) of different taxonomic groups (inverted triangle, crustacean; upright triangle, gastropod; hexagon, bivalve; circled cross, cephalopod; square, elasmobranch; and circle, teleost). Data are also depicted as a function of the temperature at which the experiment was conducted (C and D), as a function of the PCO₂ used as the experimental treatment (E and F) and as a function of the duration of the CO₂ exposure (G and H). Note that the invertebrate species are all calcifying because there were no comparable data available for non-calcifying species. Continous lines indicate mean ± 95% confidence interval (A and B) or fitted lines from linear regression, with dotted lines representing the 95% confidence interval (C–H) (see Results for details). The complete data set and reference details are available in Supplementary Table ST2_AAS.

Open in new tab Download slide

In invertebrates (Fig. 10A), the relationship between the effect of CO₂ on $\dot{M} O_{2 rest}$ (lnRR_MO2rest) and the effect of CO₂ on AAS (lnRR_AAS) was positive and marginally significant (linear regression, R² = 0.665, P = 0.0479). The relationship appeared to be opposite for fish (Fig. 10B; linear regression, R² = 0.242, P = 0.0741), although this pattern was driven by one case (yellowstriped cardinalfish) where $\dot{M} O_{2 rest}$ was strongly elevated by CO₂ exposure, and the relationship was not significant when this point was excluded (linear regression, R² = 0.011, P = 0.736).

The effect of elevated PCO2 on absolute aerobic scope (AAS) as a function of the effect on resting oxygen uptake (MO2rest). Data are log response ratios to CO2 (lnRRCO2) with 95% confidence intervals in calcifying invertebrates (A) and teleosts and elasmobranchs (B) of both non-adult (green) and adult (blue) life stages. Different symbols indicate different taxonomic groups (upright triangle, gastropod; inverted triangle, crustacean; circle, teleosts; and square, elasmobranch). Note that the invertebrate species are all calcifying because there were no comparable data available for non-calcifying species. The continous lines are fitted by linear regression, and dotted lines represent the 95% confidence interval (see Results for details).

Figure 10:

The effect of elevated PCO₂ on absolute aerobic scope (AAS) as a function of the effect on resting oxygen uptake (⁠ $M O_{2 rest}$ ⁠). Data are log response ratios to CO₂ (lnRR_CO2) with 95% confidence intervals in calcifying invertebrates (A) and teleosts and elasmobranchs (B) of both non-adult (green) and adult (blue) life stages. Different symbols indicate different taxonomic groups (upright triangle, gastropod; inverted triangle, crustacean; circle, teleosts; and square, elasmobranch). Note that the invertebrate species are all calcifying because there were no comparable data available for non-calcifying species. The continous lines are fitted by linear regression, and dotted lines represent the 95% confidence interval (see Results for details).

Open in new tab Download slide

When dividing studies according to the methodology used to estimate $\dot{M} O_{2 rest}$ (‘resting’ vs. ‘routine’) and AAS (‘post-chase’ vs. ‘during exercise’) and comparing the outcome (effect vs. no effect) of elevated-CO₂ treatment (Fig. 11A), there was no difference in the proportions for either $M O_{2 \max}$ (Fisher’s exact test, P = 0.454) or AAS (P = 0.620). Likewise, in studies were an effect was found was there no differences in the proportion finding a decrease vs. an increase (Fig. 11B) when comparing either ‘resting’ with ‘routine’ (Fisher’s exact test, P = 0.543) or ‘post-chase’ with ‘during exercise’ (P = 0.467).

Effect of methodology on the outcome of CO2 treatment on resting oxygen uptake (M.O2rest) and absolute aerobic scope (AAS). Data are number of data sets grouped according to whether resting or routine oxygen uptake was measured (for M.O2rest), and whether maximal oxygen uptake (for AAS) was measured after chasing (post-chase) or during exercise. All taxonomic groups are included. (A) The number of data sets showing an effect (95% confidence interval of lnRR not overlapping zero) vs. no effect (95% confidence interval of lnRR overlapping zero). (B) The number of data sets (of the ones with an effect) showing a decrease vs. an increase. There were no significant differences in the proportions (see Results for details).

Figure 11:

Effect of methodology on the outcome of CO₂ treatment on resting oxygen uptake (⁠ $\dot{M} O_{2 rest}$ ⁠) and absolute aerobic scope (AAS). Data are number of data sets grouped according to whether resting or routine oxygen uptake was measured (for $\dot{M} O_{2 rest}$ ⁠), and whether maximal oxygen uptake (for AAS) was measured after chasing (post-chase) or during exercise. All taxonomic groups are included. (A) The number of data sets showing an effect (95% confidence interval of lnRR not overlapping zero) vs. no effect (95% confidence interval of lnRR overlapping zero). (B) The number of data sets (of the ones with an effect) showing a decrease vs. an increase. There were no significant differences in the proportions (see Results for details).

Open in new tab Download slide

Nature of the interaction between ’temperature and CO₂

For $\dot{M} O_{2 rest}$ ⁠, there was a positive significant relationship between measured and expected additive lnRR for both invertebrates (Fig. 12A; linear regression, R² = 0.838, P < 0.0001, measured-lnRR_T_+CO2 = 0.87 × expected-lnRR_T_+CO2 + 0.007) and fish (Fig. 12B; R² = 0.657, P < 0.0001, measured-lnRR_T_+CO2 = 0.66 × expected-lnRR_T_+CO2 + 0.104). In both invertebrates and fish, the slope was significantly smaller than 1 (non-linear regression fit of straight line, rejecting the null hypothesis that slope = 1.0 with P = 0.0142 and P = 0.004, respectively), indicating—if anything—antagonistic rather than synergistic interactions between temperature and CO₂. This was also the case when different taxonomical groups were analysed individually, so there did not appear to be a particular correlation of taxonomic group or life stage with the nature of the interaction.

Figure 12:

Measured combined log response ratio as a function of the expected additive log response ratio. Data are log response ratios (lnRR_T_+CO2 with 95% confidence intervals) for resting oxygen uptake (⁠ $\dot{M} O_{2 rest}$ ⁠; A and B) and absolute aerobic scope (AAS; C and D) of both non-adult (orange) and adult (purple) life stages in calcifying invertebrates (A and C) and fish (B and D). Different symbols indicate different taxonomic groups (upright triangle, gastropod; inverted triangle, crustacean; circled cross, cephalopod; hexagon, bivalve; diamond, echinoderm; cross, cnidarians; circle, teleost; and square, elasmobranch). Note that the invertebrate species are all calcifying because there were no comparable data available for non-calcifying species. The continous lines are fitted by linear regression, and the dotted lines represent the 95% confidence interval. The dashed 1/1 line is where the measured lnRR_T_+CO2 equals the expected additive lnRR.

Open in new tab Download slide

For AAS, there was no significant relationship between measured and expected additive lnRR for invertebrates (Fig. 12C; R² = 0.581, P = 0.134, measured-lnRR_T_+CO2 = 0.415 × expected-lnRR_T_+CO2 + 0.034), but there was a positive significant relationship for vertebrates (Fig. 12D; R² = 0.870, P = 0.002, measured-lnRR_T_+CO2 = 0.776 × expected-lnRR_T_+CO2 + 0.076). In both invertebrates and vertebrates, the slope was smaller but not significantly different from 1 (non-linear regression fit of straight line, not rejecting the null hypothesis that slope = 1.0 with P = 0.155 and P = 0.064, respectively), indicating—if anything—an additive rather than synergistic interaction. As for $\dot{M} O_{2 rest}$ ⁠, there did not appear to be a particular correlation between taxonomic group and the nature of the interaction, although the smaller sample size for AAS of invertebrates and the lack of data on elasmobranchs should be kept in mind. In contrast, it is noteworthy that for the teleosts investigated, the juvenile life stages seemed to display additive effects, whereas the adults showed more antagonistic effects, although this could also be a result of all the four adults being coral reef fish, whereas the juveniles were temperate species.

Discussion

Aerobic-scope curves across species and time frames

The diversity in the shape of the aerobic-scope curves, and particularly, the apparent absence of T_optAAS in many species, is difficult to reconcile with a single unifying model for how temperature affects respiratory variables in marine ectothermic animals. The main impression is that studies on the effect of temperature on AAS yield divergent results, and that this may be related to the diversity in physiology and habitats of the species, and to the time frame of the study. The present analysis therefore supports the responses outlined in the reviews by Clark et al. (2013) and Schulte (2015); that is, some species show an increase–optimum–decrease-type response, as predicted by the Fry paradigm and the OCLTT hypothesis, whereas others seem able to maintain or continue to increase AAS with rising temperature, and a clear T_optAAS is not evident. These patterns were observed in the data on both teleosts and invertebrates (of which only gastropods, crustaceans, cephalopods and bivalves have been investigated). For the species that show bell-shaped curves, the data support the widely accepted notion that thermal reaction norms reflect the variability of habitat temperatures (e.g. Tewksbury et al., 2008), so that animals from variable temperate climates have broad thermal windows, whereas polar or equatorial species, e.g. the coral reef fish, display narrow windows.

As outlined in the Introduction, the underlying cause of the increase in AAS with temperature at the lower end of the thermal window must be that $\dot{M} O_{2 \max}$ increases faster than $\dot{M} O_{2 rest}$ (Fig. 1A). Presumably, this difference arises because $\dot{M} O_{2 rest}$ increases owing to the temperature dependence of underlying chemical processes (following the Arrhenius equation), whereas $\dot{M} O_{2 \max}$ can be enhanced actively by adjustments in the circulatory and respiratory systems, which benefits from increased temperature (e.g. Gräns et al., 2014). At the other side of T_optAAS, $\dot{M} O_{2 rest}$ is assumed to increase inexorably, virtually until the animal dies, while the cardiorespiratory system is unable to support further increases in $\dot{M} O_{2 \max}$ ⁠, resulting in a decrease in scope. The predicted increase in $\dot{M} O_{2 rest}$ is generally found in all species, as is the predicted initial increase in $\dot{M} O_{2 \max}$ ⁠, although the suggested eventual decline is not found in all species. Species with a temperature effect on $\dot{M} O_{2 rest}$ (high Q₁₀), combined with an inability to increase $\dot{M} O_{2 \max}$ ⁠, may therefore be more likely to suffer from reductions in aerobic scope, whereas species with a more moderate Q₁₀, and where the $\dot{M} O_{2 \max}$ does not seem to be limited at high temperatures, will also show a continuous increase in AAS. Although the aerobic scope curve is obviously shaped by a combination of the temperature effect on $\dot{M} O_{2 rest}$ and $\dot{M} O_{2 \max}$ ⁠, the ultimate outcome seems to be most closely reflected by $\dot{M} O_{2 \max}$ ⁠. Importantly, the method used to estimate $\dot{M} O_{2 \max}$ does not seem to influence the shape of the curve. Another methodological aspect that is important to keep in mind is the possibility that some species may increase spontaneous activity as temperature rises (Reynolds and Casterlin, 1982; Forstner and Wieser, 1990; Castonguay and Cyr, 1998; Stoner et al., 2006), whereas others may not (Peterson and Anderson, 1969; Stevens and Fry, 1972; Schurmann and Steffensen, 1994; Crocker and Cech, 1997; Johansen et al., 2014). If routine rather than resting $\dot{M} O_{2}$ levels are measured and used as $\dot{M} O_{2 rest}$ to calculate AAS, one may observe an increase in $\dot{M} O_{2}$ caused by activity rather than an increase in basal oxygen demand, and hence a decrease in aerobic scope that reflects differences in activity rather than aerobic capacity. The majority of studies included in the present analysis measured $\dot{M} O_{2 rest}$ in conditions that can be considered resting, but there was no obvious bias in the outcome of the few studies using routine rates, because two studies found AAS to increase continually with temperature (Clark et al., 2005; Vagner et al., 2015), whereas three studies found a clear T_optAAS (Lee et al., 2003; Dissanayake and Ishimatsu, 2011; Healy and Schulte, 2012a).

There is no doubt that an increase in $\dot{M} O_{2 rest}$ and limitation of $\dot{M} O_{2 \max}$ are likely within an acute scenario (minutes, hours and maybe days, at a new temperature), where the animal is not given time to adjust its metabolic processes and thereby their costs. This is also supported by the data, as several of the acutely performed measurements of AAS reveal a typical bell-shaped increase–optimum–decrease-type response [robust shell (Littoraria undulata; Patnaik et al., 1985); bald notothen (Lowe and Davison, 2006); sockeye salmon (Oncorhynchus nerka; Eliason et al., 2011); shorthorn sculpin (Myoxocephalus scorpius), Arctic sculpin (Myoxocephalus scorpioides) and Arctic staghorn sculpin (Gymnocanthus tricuspis; Seth et al., 2013); goldfish (Carassius auratus; Ferreira et al., 2014); rainbow trout (Oncorhynchus mykiss; Chen et al., 2015); European crayfish (Astacus astacus) and giant tiger prawn (Penaeus monodon; Ern et al., 2015); humpback conch (Gibberulus gibberulus gibbosus; Lefevre et al., 2015); and bonefish (Albula vulpes; Nowell et al., 2015)]. Intriguingly, a study on a high-Arctic population of blue mussel (Mytilus edulis) acclimated to 1°C (Thyrring et al., 2015) found that AAS during an acute temperature challenge was highest at 7°C, even though this population probably never experiences a temperature higher than 5°C during the warmest month of the year. Likewise, there are species that under acute temperature challenges continue to increase AAS up to CT_max [barramundi (Lates calcarifer; Norin et al., 2014); and pink salmon (Oncorhynchus gorbuscha; Clark et al., 2011)] or up to the highest temperatures they may experience in their habitats [jumbo squid (Rosa and Seibel, 2008); Murray cod (Maccullochella peelii; Clark et al., 2005); and golden grey mullet (Liza aurata; Vagner et al., 2015)]. These cases are difficult to reconcile with the view that a limitation of the capacity for oxygen delivery is the main driver of thermal tolerance.

The concept of OCLTT has, however, not only been applied to acute circumstances, but is also suggested to explain species distribution limits and hence predict changes in distributions under global warming and ocean acidification (e.g. Bozinovic and Pörtner, 2015), processes occurring at a much slower pace than what can be mimicked in the laboratory. This automatically introduces a contradiction with the well-established paradigm that animals, given enough time, will down-regulate their basal oxygen demand and may even compensate fully for the elevated temperature (Bullock, 1955; Segal, 1961; Hazel and Prosser, 1974; Johnston and Dunn, 1987; Sandblom et al., 2014; Seebacher et al., 2015), presumably to minimize energy demand. Although this makes sense, at least if it is assumed that animals are generally energy restricted, it does conflict with the assumption that $\dot{M} O_{2 rest}$ increases exponentially, which contributes to the shape of the aerobic performance curve under the Fry paradigm and hence the OCLTT hypothesis. The ability to acclimate could thus explain why several species appear to show increases [common cockle (Newell, 1966); cutthroat trout (Oncorhynchus clarkii; Dwyer and Kramer, 1975); smooth plough shell (Bullia rhodostoma; Brown and da Silva, 1984); Atlantic blue crab (Booth and McMahon, 1992); European sea bass (Dicentrarchus labrax; Claireaux et al., 2006); southern blue catfish (Ictalurus meridionalis; Pang et al., 2010); largemouth bronze gudgeon (Coreius guichenoti; Tu et al., 2012); and nine-spined stickleback (Pungitius sinensis; Bruneaux et al., 2014)] or only minor decreases [sockeye salmon (Brett, 1964); common periwinkle (Newell and Pye, 1970, 1971); rainbow trout (Dickson and Kramer, 1971); finger plough shell (Bullia digitalis; Brown and da Silva, 1983); turbot (Scophthalmus maximus; Mallekh and Lagardere, 2002); mummichog (Healy and Schulte, 2012a); common triplefin (Forsterygion lapillum; Khan and Herbert, 2012; Khan et al., 2014a); Atlantic halibut (Hippoglossus hippoglossus; Gräns et al., 2014); salema and marbled spinefoot (Marras et al., 2015)] in AAS after acclimation to an increased temperature.

It is clear that acute studies can yield temperature response curves that are different from those obtained in the same species after acclimation to the different temperatures. This is, for example, illustrated by different studies on Atlantic cod (Gadus morhua). Sylvestre et al. (2007) found that RAS was already reduced to 78% at 13°C, whereas Tirsgaard et al. (2015a) found that RAS had not yet decreased at 15°C, in individuals of a similar size, after 2–3 weeks of acclimation. Perhaps even more extreme is the case of rainbow trout, where acute studies by Chen et al. (2015) showed severe decreases in RAS to 45% at 25°C, whereas Dickson and Kramer (1971) found RAS to be maintained at 94% after 2 weeks of acclimation at 25°C. The maximal AAS was similar in the two studies (550 and 498 mg O₂ kg⁻¹ h⁻¹, respectively). In both Atlantic cod and rainbow trout, the T_optAAS with acute temperature changes coincided with the temperature at which the fish were acclimated. Another illustrative example of the importance of acclimation, and even transgenerational acclimation for $\dot{M} O_{2 rest}$ (and presumably also AAS), comes from rearing studies. In their first study, Donelson et al. (2011) showed that spiny chromis damselfish (Acanthochromis polyacanthus), reared from the larval stage (30 days post-hatch) at a 3°C higher temperature than the natural temperature in their habitat, partly compensated for the rise in $\dot{M} O_{2 rest}$ that was displayed by conspecifics acutely exposed to the same temperature. Subsequently, it was shown (Donelson et al., 2012) that offspring of spiny chromis damselfish kept at the +3°C regime also displayed $\dot{M} O_{2 rest}$ values that were fully compensated, revealing transgenerational acclimation. Overall, these examples and the present analysis show that studies of shorter duration (i.e. no or little acclimation) may be more likely to identify a T_optAAS, indicating a difference in the physiological effects between acute and long-term exposures. Such a difference is obviously important to keep in mind when interpreting aerobic scope data and using it for modelling.

One can, of course, argue that studies failed to reveal a decline in AAS because temperatures were not high enough, i.e. if one had continued to increase the temperature one would eventually see a decline in AAS. But the majority of the studies finding that AAS does not decline have used a range of temperatures matching the distributional temperature range [except lemon sole (Microstomus kitt), common dab (Limanda limanda; Duthie, 1982); Atlantic cod (Claireaux et al., 2000); qingbo (Spinibarbus sinensis; Pang et al., 2013); and goldfish (Ferreira et al., 2014)] and still show that AAS within these more ecologically relevant temperatures is not compromised. Thus, in these cases it is unlikely that AAS is the factor restricting the distribution of that species, and alternative mechanisms must be investigated to identify factors that can be used in the prediction of distributional changes with climate change. That is, even if AAS is not limited at the upper temperature range of a species, there may be other important performance measures that have a T_opt, such as reproduction and growth. One example is the barramundi, where AAS is the same at 38°C as at 29°C (after acclimation), but where the optimal temperature for growth and even the preferred temperature is 31°C (Norin et al., 2014). In Atlantic cod, evolutionary bioenergetics modelling suggested AAS to be a poor predictor of optimal temperature for fitness as a whole (Holt and Jørgensen, 2015). Furthermore, a recent meta-analysis revealed that the acclimatory plasticity characteristic of aerobic metabolism is not observed to the same extent for lethal temperature limits (Gunderson and Stillman, 2015). For species living in environments with increased likelihood of short but extreme temperature peaks (e.g. tidepools and streams), a limited plasticity of critical temperature may be more important than AAS in determining future success, because extreme climatic events are expected to be more common in the future (Rahmstorf and Coumou, 2011; Fey et al., 2015).

The number of examples that do not follow the predictions by the Fry paradigm, and where the OCLTT hypothesis is therefore not supported, has grown to a proportion that can hardly be classified as exceptions to a rule. But nonetheless, these cases cannot be considered as arguments for completely abandoning the concept either (or ‘throw out the baby with the bathwater’, as it was put by Farrell, 2016). It cannot be ruled out that OCLTT guides the performance of such animals that, even when given time to acclimate, still have a T_optAAS, or at least a decline in AAS at elevated temperatures, also within the ecologically relevant temperature range [sockeye salmon (Lee et al., 2003); common sole (Solea solea; Lefrancois and Claireaux, 2003); bald notothen (Lowe and Davison, 2006); hapuku wreckfish (Khan et al., 2014b); and various coral reef fish (Nilsson et al., 2009; Gardiner et al., 2010; Johansen and Jones, 2011; Rummer et al., 2014)]. In these cases, the acclimated response resembles the acute response, and this may be a direct result of an elevation in $\dot{M} O_{2 rest}$ and/or limitation of $\dot{M} O_{2 \max}$ that cannot be alleviated through acclimation. This may even be the expected outcome for species that are adapted to a narrow temperature range and are lacking or have lost the genes necessary for acclimation outside this range. Many coral reef fish and polar species appear to be good examples of animals with a thermal history that has led to a narrow thermal window for physiological performance. A recent study found a good correlation between performance data from the field (activity and growth) and from the laboratory (AAS; Payne et al., 2016), but notably, this correlation was obtained after excluding all the species where AAS did not decline at high temperature. In any case, finding that T_optAAS and T_optFIT correlate does not prove causal relationship between the two, and as such, a more mechanistic experimental approach is necessary to demonstrate OCLTT in a species (e.g. Healy and Schulte, 2012b; Overgaard et al., 2012; Ellis et al., 2013; Verberk et al., 2013; Wang et al., 2014; Brijs et al., 2015; Ern et al., 2015), even if evidence both for and against the predictions from the hypothesis exists.

Does elevated CO₂ in general cause resting oxygen uptake to increase?

Hypercapnia has for long been used in fish physiology research as a tool for studying the regulation of respiration and control of acid–base balance, and much of our basic understanding of pH regulation in fish is based on such experiments. These studies involved exposure to levels of CO₂ many times higher than those relevant from an ocean acidification perspective. In some cases, these high CO₂ levels interfered directly with ventilation (e.g. Kinkead and Perry, 1991; Crocker and Cech, 2002; Vulesevic et al., 2006; Perry and Abdallah, 2012), and thereby, possibly $\dot{M} O_{2}$ and AAS. To evaluate the effects of future ocean acidification, it is much lower PCO₂ levels that matter, because the current predictions are in the 1000 µatm range (Meinshausen et al., 2011; Doney et al., 2012).

It can be argued that CO₂ exposure could cause an increase in $\dot{M} O_{2 rest}$ ⁠, either through direct increases in costs associated with adjusting to elevated internal CO₂, or alternatively, by inducing a general stress reaction. Indeed, for some animals in some conditions, $\dot{M} O_{2 rest}$ has been shown to be elevated significantly during prolonged CO₂ exposure [yellowstriped cardinalfish (Munday et al., 2009); Atlantic oyster (Crassostrea virginica; Beniash et al., 2010); Pacific oyster (Crassostrea gigas; Lannig et al., 2010); serpent star (Ophiura ophiura; Wood et al., 2010); Schayer’s brittlestar (Ophionereis schayeri; Christensen et al., 2011); purple sea urchin (Paracentrotus lividus; Catarino et al., 2012); dwarf cushion star (Parvulastra exigua; McElroy et al., 2012); green sea urchin (Strongylocentrotus droebachiensis; Dorey et al., 2013); bald notothen, emerald rockcod (Trematomus bernacchii) and striped rockcod (Trematomus hansoni; Enzor et al., 2013)]. Nevertheless, in most of the published experiments, elevated CO₂ did not significantly affect $\dot{M} O_{2 rest}$ [e.g. Atlantic cod (Melzner et al., 2009; Tirsgaard et al., 2015b); an Arctic pteropod (Limacina helicina; Comeau et al., 2010); Shiba shrimp (Metapenaeus joyneri; Dissanayake and Ishimatsu, 2011); Acesta excavata (Hammer et al., 2011); Zhikong scallop (Chlamys farreri; Mingliang et al., 2011); burrowing shrimp (Upogebia deltaura; Donohue et al., 2012); various copepods (Li and Gao, 2012; Hildebrandt et al., 2014; Zervoudaki et al., 2014; Thor and Dupont, 2015); the bivalves Chlamys nobilis, Perna viridis and Pinctada fucata (Liu and He, 2012); marbled rockcod (Notothenia rossii; Strobel et al., 2012); Northern shrimp (Pandalus borealis; Arnberg et al., 2013); porcelain crab (Petrolisthes cinctipes; Carter et al., 2013); common starfish (Asterias rubens; Collard et al., 2013); hard-shelled clam (Mercenaria mercenaria; Dickinson et al., 2013; Matoo et al., 2013); the brittlestars Ophiothrix fragilis and Amphiura filiformis (Carey et al., 2014); the sea cucumbers Holothuria parva and Holothuria scabra (Collard et al., 2014); small-spotted catshark (Scyliorhinus canicula; Green and Jutfelt, 2014); Atlantic halibut (Gräns et al., 2014); European sea bass (Pope et al., 2014); white-spotted bamboo shark (Chiloscyllium punctatum; Rosa et al., 2014a); Pacific sea urchins (Echinometra sp. A; Uthicke et al., 2014); rainbow abalone (Haliotis iris; Cunningham et al., 2015); red drum (Sciaenops ocellatus; Esbaugh et al., 2016); various coral reef fish (Couturier et al., 2013; Ferrari et al., 2015); Antarctic dragonfish (Gymnodraco acuticeps; Flynn et al., 2015); humpbacked conch (Lefevre et al., 2015); common slipper shell (Crepidula fornicate; Noisette et al., 2015, 2016); European lobster (Homarus gammarus; Small et al., 2015); Norway lobster (Nephrops norwegicus; Wood et al., 2015); and cone-shaped Nassa (Nassarius conoidalis; Zhang et al., 2015)] or even caused a reduction [e.g. peanut worm (Sipunculus nudus; Pörtner et al., 1998); velvet swimming crab (Necora puber; Small et al., 2010); grooved carpet shell (Ruditapes decussatus; Fernández-Reiriz et al., 2011); Chilean blue mussel (Mytilus chilensis; Navarro et al., 2013); spiny chromis damselfish (Rummer et al., 2013); common dolphinfish (Coryphaena hippurus; Pimentel et al., 2014); European squid (Loligo vulgaris; Rosa et al., 2014b); and hard-shelled mussel (Mytilus coruscus; Wang et al., 2015)].

A decreased $\dot{M} O_{2 rest}$ could be interpreted as something positive because, all else being equal, it would imply reduced maintenance costs and potentially higher AAS. The mechanism behind a CO₂-induced reduction in $\dot{M} O_{2 rest}$ is rarely discussed, but has nonetheless often been interpreted as something negative. It has been suggested that ‘uncompensated extracellular pH might be the trigger for these reductions’ [quote from Stumpp et al. (2011); based on Langenbuch and Pörtner (2002) and Pörtner et al. (2005)]. Melatunan et al. (2011) observed a reduction in $\dot{M} O_{2 rest}$ in common periwinkle that appeared to be compensated for by an increase in anaerobic metabolism, which is not a sustainable strategy and therefore likely to be negative for fitness or even survival, but the mechanism behind a CO₂-induced reduction in respiratory capacity was not discussed. In embryos of common cuttlefish (Sepia officinalis), the observed reduction in $\dot{M} O_{2 rest}$ was interpreted as a form of adaptive short-term metabolic depression invoked to conserve energy (Rosa et al., 2013). Carbon dioxide has also previously been linked to metabolic depression, although this may apply to very high CO₂ levels, where it has anaesthetic effects (e.g. Guppy and Withers, 1999).

The absolute value of $\dot{M} O_{2 rest}$ ⁠, and the way it is influenced by one stressor in a given situation, is not necessarily straightforward to interpret. Factors that could contribute to variation in the response are the duration of the CO₂ exposure, PCO₂ level and temperature. Firstly, one can expect that the longer the animals are exposed to a certain condition, the likelier they are to have compensated for the challenge through acid–base regulation; general stress responses should have subsided, and the harder it might be to detect a remaining effect. In contrast to this reasoning, the few studies using an exposure time of less than a day found no effect of elevated CO₂ (Rosa and Seibel, 2008; Rivest and Hofmann, 2014), whereas the outcome of long-term experiments (more than a months to a year) were reductions (Fernández-Reiriz et al., 2011; Rosa et al., 2014a), no effect (Melzner et al., 2009; Dickinson et al., 2013; Gräns et al., 2014; Hildebrandt et al., 2014; Uthicke et al., 2014; Cunningham et al., 2015; Noisette et al., 2016; Strahl et al., 2015; Thor and Dupont, 2015) or increases (Beniash et al., 2010; Enzor et al., 2013; Matoo et al., 2013) in $\dot{M} O_{2 rest}$ ⁠. Overall, it was not possible to detect a direct effect of the exposure times used, of which most were 4 days or more.

Secondly, it could be expected that higher CO₂ levels led to larger increases in $\dot{M} O_{2 rest}$ ⁠, because the loading stress would be higher. Curiously, of the few studies that used a high CO₂ level (5000–33 000 µatm), most found no change in $\dot{M} O_{2 rest}$ (Deigweiher et al., 2008; Melzner et al., 2009; Dissanayake and Ishimatsu, 2011; Hammer et al., 2011; Tirsgaard et al., 2015b), whereas others found a decrease (Christensen and Colacino, 2000; Small et al., 2010; Mingliang et al., 2011; Hu et al., 2014; Sun et al., 2016). Other than that, most of the recent studies have been concerned with climate change, and a relatively narrow range of CO₂ levels have been used, around 800–1200 µatm, because this is the current RCP8.5 predicted level for the year 2100 (Meinshausen, et al., 2011; Doney, et al., 2012). As these levels are low in a classic physiological sense and cover a narrow range, it is not surprising that they do not explain the wide variation observed in $\dot{M} O_{2 rest}$ response ratios.

Lastly, temperature obviously has a strong influence on physiological performance, and it could be argued that animals at higher temperatures with higher metabolic demands may be more susceptible to possible stressors, such as CO₂. Unlike the uniform range of CO₂ treatment levels, animals from a wide range of temperatures have been studied, making it more likely to detect a relationship between temperature and the response to elevated CO₂, if one existed. But again, negative, neutral and positive effects were spread evenly across the temperature range. Notably, only one of the experiments conducted at the highest temperatures (25–30°C) showed a significant increase in $\dot{M} O_{2 rest}$ (Munday et al., 2009), whereas three studies showed a decrease with CO₂ treatment (Rummer et al., 2013; Rosa et al., 2014b; Wang et al., 2015).

But still, it cannot be ruled out that some of the studies, which measured routine rather than resting $\dot{M} O_{2}$ ⁠, failed to detect a difference because a truly resting state was not obtained. A truly resting state may, however, be difficult to confirm in some types of animals, such as corals and bivalves. Importantly, the direction (decrease vs. increase) of responses was not affected by overall methodology, indicating that the divergence in the CO₂ effects, when these are found, may in fact be a result of different physiological mechanisms. A more obvious difference between studies is, of course, that they have been conducted on different species, and identifying the mechanistic background for each of the outcomes will require further and more detailed studies on those particular species. Only a few species have been investigated multiple times but in slightly different experimental conditions. In Atlantic oysters, Beniash et al. (2010) found an increase in $\dot{M} O_{2 rest}$ at elevated CO₂ in juveniles, but no effect in adults, as did Matoo et al. (2013). In the purple sea urchin (Strongylocentrotus purpuratus), Stumpp et al. (2011) found an increase after 21 days, whereas both Stumpp et al. (2011) and Padilla-Gamiño et al. (2013) found no effect after 4 days. In the Chilean abalone (Concholepas concholepas), Manríquez et al. (2013) found no effect in adults, whereas Lardies et al. (2014) found an increase in $\dot{M} O_{2 rest}$ in juveniles, although with a higher PCO₂ and temperature and shorter treatment duration. In adult blue mussels, Zittier et al. (2015) and Sun et al. (2016) found no effect of elevated CO₂ at 15°C, whereas Thomsen and Melzner (2010) found an increase in $\dot{M} O_{2 rest}$ at 8°C. Lastly, in the king scallop (Pecten maximus) an increase in $\dot{M} O_{2 rest}$ was found after 50 days (Schalkhausser et al., 2014), but not after 30 days, of high-CO₂ treatment (Schalkhausser et al., 2013). There may even be differences in the response between different populations of a species, because, for example, Thor and Oliva (2015) found $\dot{M} O_{2 rest}$ to be elevated after high-CO₂ exposure in the copepod Pseudocalanus acuspes from Skagerrak, whereas the same treatment did not have an effect on individuals from Svalbard.

Does elevated CO₂ in general reduce ’aerobic scope?

The variability detected in the response in $\dot{M} O_{2 rest}$ to elevated CO₂ is also reflected in AAS, which decreases in some studies [jumbo squid (Rosa and Seibel, 2008); yellowstriped and fourline cardinalfish (Munday et al., 2009); and Shiba shrimp (Dissanayake and Ishimatsu, 2011)], but not in others [ambon damsel (Pomacentrus amboinensis), lemon damsel, brown dottyback (Pseudochromis fuscus; Couturier et al., 2013); European sea bass (Pope et al. 2014); red drum (Esbaugh et al., 2016); small-spotted catshark (Green and Jutfelt, 2014); and humpbacked conch (Lefevre et al., 2015)], and in fact, even increases in a third, albeit small, group [spiny chromis damselfish (Rummer et al., 2013); and Atlantic halibut (Gräns et al., 2014)]. Overall, these outcomes were not influenced by methodology, because the proportion of studies finding an effect vs. no effect, and an increase vs. a decrease, was the same for studies measuring $\dot{M} O_{2 \max}$ during exercise or after chasing. The relationship between the effect of elevated CO₂ on $\dot{M} O_{2 rest}$ and the effect of elevated CO₂ on AAS was weak, which does not support the hypothesis that it is an increase in $\dot{M} O_{2 rest}$ that causes AAS to decrease. Why, therefore, is AAS sometimes affected by CO₂? Obviously, if it is not because $\dot{M} O_{2 rest}$ changes, it must be $\dot{M} O_{2 \max}$ that changes. It has been hypothesized that CO₂ acts as a limiting stressor, reducing $\dot{M} O_{2 \max}$ by, for example, interfering with respiratory pigments [Heuer and Grosell (2014) cites Pörtner et al. (2004b), who cites Tamburrini et al. (1998)], but no further studies appear to have examined this hypothesis. For the cases of CO₂-induced reductions of $\dot{M} O_{2 \max}$ ⁠, it has been suggested that it is changes in metabolic pathways, specifically a shift in the balance between anaerobic and aerobic pathways (as reported by Michaelidis et al., 2007), that supress AAS and/or $\dot{M} O_{2 \max}$ (Munday, et al., 2009). Schalkhausser et al. (2013) found that $\dot{M} O_{2 \max}$ and AAS were reduced in the king scallop, but these findings were not replicated in a later study (Schalkhausser et al., 2014), and the authors did not attempt to explain either the decreased $\dot{M} O_{2 \max}$ or the discrepancy between their experiments. In the cases where elevated CO₂ had no effect, this was attributed to a sufficient capacity for acid–base regulation (Melzner, et al., 2009), also with long-term exposure. Maybe for good reasons, i.e. to avoid unfounded speculation, some studies reflect little or only vaguely upon the mechanistic cause for the observed response, whether it was a reduction (e.g. Rosa and Seibel, 2008; Dissanayake and Ishimatsu, 2011; Tirsgaard et al., 2015b) or no change (Schalkhausser et al., 2013; Lefevre et al., 2015). Given the hypothesis that CO₂ acts as a limiting factor, it is counterintuitive to find that $\dot{M} O_{2 \max}$ increases in some animals exposed to elevated CO₂ . In the Atlantic halibut, increased $\dot{M} O_{2 \max}$ appeared to be associated with increased maximal pumping ability of the heart (Gräns, et al., 2014), although it is unclear why CO₂ would have that effect. Rummer et al. (2013) suggested that increased respiratory surface area contributed to an increased $\dot{M} O_{2 \max}$ ⁠, and even hypothesized that an interaction between acidosis and stress could have led to catecholamine release, which increased oxygen delivery to the muscles by reducing the affinity of haemoglobin for oxygen. Couturier et al. (2013) suggested increased swimming speed as a possible explanation for increased $\dot{M} O_{2 \max}$ when measured in a swim respirometer, which in turn was hypothesized to be associated with behavioural alterations caused by high CO₂. In other words, it could be that neural effects of exposure to elevated CO₂ either increase the drive to swim fast or take away any behavioural inhibition that may suppress maximal exercise efforts. In addition, Couturier et al. (2013) pointed towards the interesting possibility that elevated CO₂ does indeed incur energetic costs, which are not measurable in resting conditions, but become evident when the fish are pushed to the limit of their capacity, forcing them to increase $\dot{M} O_{2 \max}$ ⁠. If this suggestion is correct, then an increased $\dot{M} O_{2 \max}$ or AAS should not necessarily be interpreted as something promoting survival and fitness, a notion also put forward by Di Santo (2015), although the latter study did not measure $\dot{M} O_{2 \max}$ per se.

As a result of the difficulty (or reluctance) of publishing negative results, it may even be that counting the number of studies showing positive or negative effects of elevated CO₂ on aerobic performance variables gives an overestimation of the general effects that expected future CO₂ may have on marine animal respiration and metabolism. One may make the general reflection that the degree of ocean acidification caused by a projected future PCO₂ of ∼1000 µatm does not make the ocean acidic (i.e. pH < 7.0) but will reduce pH from today’s pH ∼8.2 to ∼7.8. At least with regard to teleost fish, this will bring the water pH closer to the blood pH, and if there is any energetic cost involved in maintaining a pH gradient between blood and water, then this cost would be reduced. One may also reflect upon the fact that a pH of 7.8 would be readily tolerated by virtually all freshwater fish, because this is a very ‘average’ freshwater pH. Thus, effective mechanisms needed to handle variations in water pH do exist in teleost fish, although it could be argued that they may have been lost in many marine teleosts living in stable pH conditions. Consequently, there are no obvious reasons to expect that the acidification as such would pose a serious problem, at least for fish, and this may be part of the explanation for the divergent results.

Calcifying invertebrates could be another matter, although results vary even within this group. Although one might have expected $\dot{M} O_{2 rest}$ to be increased more generally by elevated CO₂ in calcifying organisms, the few existing studies on adult tropical coral do not show an effect of CO₂ on $\dot{M} O_{2}$ (dark respiration; Takahashi and Kurihara, 2013; Strahl et al., 2015; Comeau et al., 2016), whereas $\dot{M} O_{2}$ was reduced in a cold-water coral (Hennige et al., 2014). Likewise, studies on coral larvae have not found an effect of CO₂ on $\dot{M} O_{2}$ (Nakamura et al., 2011; Putnam et al., 2012; Cumbo et al., 2013; Edmunds et al., 2013; Rivest and Hofmann, 2014). However, the elevation of CO₂ in itself can still be expected to be problematic for growth and calcification of calcifying organisms (e.g. Fabry, 2008; Ries et al., 2009; Gutowska et al., 2010; Mingliang et al., 2011; Bramanti et al., 2013; Kaplan et al., 2013; Wolfe et al., 2013; Byrne et al., 2014; Watson, 2015) and also for fish (e.g. Baumann et al., 2012; Frommel et al., 2012; Chambers et al., 2014) on the background of the increasing number of studies revealing neural effects of relatively modest rises in PCO₂ (see Beyond aerobic scope: behavioural effects of CO₂).

Combined effects of temperature and CO₂ on resting aerobic metabolism

As discussed in Aerobic scope curves across species and time frames, the expectation that elevated temperature causes $\dot{M} O_{2 rest}$ to increase, the magnitude of which will be dependent on the acclimatory ability, is well founded from decades of research. It is much less so for CO₂, and data seem to reflect this. Putting the two together, which is inevitable from a global change perspective, does not make matters clearer. A prediction could be that CO₂ is a stressor, which causes an increase in $\dot{M} O_{2 rest}$ that simply adds on top of the increase caused by temperature. Indeed, the combined effect of elevated temperature and CO₂ on respiration seems to be additive, when looking at it quantitatively. Notably, in several cases the combined outcome is additive simply because there was no measurable effect of CO₂ on $\dot{M} O_{2 rest}$ (Comeau et al., 2010; Dissanayake and Ishimatsu, 2011; Strobel et al., 2012; Arnberg et al., 2013; Padilla-Gamiño et al., 2013; Carey et al., 2014; Rivest and Hofmann, 2014; Schalkhausser et al., 2014; Cunningham et al., 2015; Lefevre et al., 2015; Noisette et al., 2016; Small et al., 2015; Zhang et al., 2015), whereas in other cases either temperature or CO₂ alone had an effect (Rosa and Seibel, 2008; Munday et al., 2009; McElroy et al., 2012; Matoo et al., 2013; Hildebrandt et al., 2014; Flynn et al., 2015; Kreiss et al., 2015). There are only a few cases suggesting synergistic effects, where the change during combined exposure is significantly higher than expected from the sum of the isolated effects, one example being a coral reef cardinalfish (fourline cardinalfish; Munday et al., 2009), the other being an Antarctic fish (bald notothen; Enzor, et al., 2013). The mechanism behind these synergistic effects, however, is not clear. It is also worth noting that several of the studies failing to show a significant interaction fall below the additive line (Fig. 10A and C). That is, in many cases temperature causes an increase in $\dot{M} O_{2 rest}$ as expected, but when the two environmental stressors are applied together, the increase is significantly lower than expected from an additive response [purple sea urchin (Catarino et al., 2012); dwarf cushion star (McElroy et al., 2012); cauliflower coral (Cumbo et al., 2013); striped rockcod, dusky rockcod (Trematomus newnesi; Enzor et al., 2013); Amphiura filiformis (Carey et al., 2014); and white-spotted bamboo shark (Rosa et al., 2014a)].

Obviously, there can be arguments both for and against an increase in $\dot{M} O_{2 rest}$ being negative for survival or fitness (depending, for example, on nutritional status and food availability), but if one assumes that saving energy is an adaptive strategy for the average animal, an antagonistic interaction can be interpreted such that CO₂ alleviates some of the costs associated with increased temperature. In contrast, in the few instances where temperature causes a decrease in $\dot{M} O_{2 rest}$ ⁠, CO₂ seems to have the opposite effect and causes an increase in $\dot{M} O_{2 rest}$ despite the expected decrease from temperature alone [sea urcin (Echinometra sp.; Uthicke et al., 2014); and Atlantic cod (Kreiss et al., 2015)] or at least causing the decrease to be smaller than expected [common periwinkle (Melatunan et al., 2011); and hard-shelled mussel (Wang et al., 2015)], which is probably not beneficial (again, if it is assumed that low $\dot{M} O_{2 rest}$ is beneficial). A decrease in $\dot{M} O_{2 rest}$ with increased temperature is clearly not the expected outcome, and it is noteworthy that it is observed only in invertebrates, whereas the majority of the teleosts investigated show no change or the expected increase in $\dot{M} O_{2 rest}$ ⁠.

Combined effects of temperature and CO₂ on absolute aerobic scope

The predictions from the OCLTT hypothesis for how elevated temperature and CO₂ in combination affect AAS are relatively straightforward (Fig. 1C and D). Two scenarios are commonly depicted; one where interaction between CO₂ and temperature is additive (CO₂ reduces AAS over the entire temperature range), and one where it is synergistic (CO₂ narrows the thermal windows, by causing a further reduction in AAS with increased or decreased temperature). Synergistic effects will only be worse than additive effects in a global warming scenario if it is assumed that the animal is presently living near T_optAAS (which most animals will strive to do, according to the OCLTT). Although there is currently a limited number of studies that have looked at the effect of both temperature and CO₂ on AAS, the data seem to support an additive outcome [jumbo squid (Rosa and Seibel, 2008); and European sea bass (Pope et al., 2014)], of which some are simply a result of a very small or absent effect of CO₂ [Atlantic halibut (Gräns et al., 2014); and humpbacked conch (Lefevre et al., 2015)]. In a few instances, a positive antagonistic effect is evident, where temperature seems to alleviate the negative effect of CO₂ [Shiba shrimp (Dissanayake and Ishimatsu, 2011); and king scallop (Schalkhausser, et al., 2014)]. Interestingly, the nature of the interaction between CO₂ and temperature can vary within a species, as for some coral reef fish, where it can be antagonistic with a moderate temperature increase and turn additive at higher temperature (yellowstriped and fourline cardinalfish; Munday et al., 2009). Furthermore, there seems to be a correlation between the response and either life stage or type of fish, because the cases with negative combined effects (Fig. 10C) were adult coral reef fish, whereas the unaffected fish were juvenile temperate fish, making it difficult to speculate on the underlying factors. Obviously, more studies of this kind on a wider range of fish species are needed to draw firm conclusions, but at least it is clear that synergistic effects have not been observed. For the animal, it is perhaps less important if it is succumbing to an additive or synergistic effect. However, for scientists trying to predict ecological effects of global change, an important implication of an additive rather than synergistic effect (when CO₂ has an effect at all) is that it is inherently easier to incorporate additive effects in mathematical models.

Beyond aerobic scope: behavioural ’effects of CO₂

Aerobic scope is a conceptually attractive measure of performance, because in theory it represents the overall capacity, that is, the amount of energy that can be devoted to different activities at any given point. This capacity is nonetheless still theoretical, because it is difficult to know how and when an animal uses this capacity in nature, where many other factors are at play (e.g. food availability and conspecifics; Farrell, 2013, 2016). A recent meta-analysis on reproduction and survival concluded synergistic effects to be dominant (Harvey et al., 2013), which contrasts with the dominantly additive effect on respiratory performance found in the present analysis, indicating that the underlying mechanisms may differ.

When it comes to the effect of climate change on variables beyond AAS, there is an additional twist to the story, namely behaviour. So far, behavioural studies on the interaction between warming and acidification are limited to a recent study by Ferrari et al. (2015) on coral reef fish [ambon damsel and Nagasaki damsel (Pomacentrus nagasakiensis)]. Strikingly, they found the interactive effect on predation rate to be synergistic (i.e. there was no effect of CO₂ or temperature by themselves, but there was an effect when the two were combined), whereas prey selectivity was antagonistically affected (i.e. both temperature and CO₂ by themselves reversed prey preference, and any preference was completely abolished when both stressors were combined). Behavioural effects of CO₂ or low pH were reviewed in detail by Briffa et al. (2012) and more recently by Clements and Hunt (2015) and Nagelkerken and Munday (2016), who found that the majority of published studies find an effect of elevated CO₂. However, these direct effects of CO₂ on behaviour are not part of the OCLTT framework and were judged to have ‘low confidence’ in the most recent edition of the IPCC report (Pörtner et al., 2014).

For some species, it is not surprising that CO₂ does not have an effect. The epaulette shark (Hemiscyllium ocellatum), which is a shallow-reef species, is exposed to rather severe hypoxia and hypercapnia on a diurnal basis, and in addition to being hypoxia tolerant (e.g. Speers-Roesch et al., 2012a,b) it is therefore likely to be tolerant of elevated CO₂, and accordingly, it is not behaviourally affected (Heinrich et al., 2014, 2016). This may also be the case for Atlantic cod, which is known to feed in hypoxic–hypercapnic waters (Strand and Huse, 2007; Neuenfeldt et al., 2009) and which is also not behaviourally affected by elevated CO₂ (Jutfelt and Hedgärde, 2015). The behavioural changes seen in other cases can easily be interpreted as negative, such as being attracted to predators, the wrong habitat or the wrong food (e.g. Vargas et al., 2013; Munday et al., 2014; Sundin and Jutfelt, 2016), failing to learn (Chivers et al., 2014), altered prey handling (Dodd et al., 2015) and loss of the ability to respond to important cues (e.g. Dixson et al., 2010; Simpson et al., 2011; Lönnstedt et al., 2013; Munday et al., 2013; Allan et al., 2014; Manríquez et al., 2013, 2014; Watson et al., 2014; Sui et al., 2015). The interpretation can be more difficult when it comes to something like lateralization (e.g. Jutfelt et al., 2013) or activity (e.g. Pimentel et al., 2014; Green and Jutfelt, 2014; Spady et al., 2014). Although a ‘hard-wired’ preference for left or right might decrease response time, it can also be argued that it limits the options and makes an individual predictable. The benefit of being more or less active depends on the situation; if an animal is more active it might be more likely to find food or conspecifics, but it might also be more visible and thereby vulnerable to predation. Importantly, in the present context, behavioural changes related to elevated PCO₂ could affect measurements of $\dot{M} O_{2 rest}$ and $\dot{M} O_{2 \max}$ and could explain the elevation of these variables seen in some studies.

In contrast to the limited understanding of whether and how elevated PCO₂ affects $\dot{M} O_{2 rest}$ and AAS, there is now an increasing understanding of the mechanism that underlies the behavioural alterations. Since first suggested from experiments showing that the effects of CO₂ on olfaction and lateralization can be reversed by treatment with the GABA_A receptor antagonist gabazine (Nilsson et al., 2012), several studies have linked these clearly neurological changes to an altered function of this major inhibitory neurotransmitter receptor. Thus, Chivers et al. (2014) showed that high-CO₂-induced learning deficiency in ambon damselfish could be reversed by gabazine treatment, and the direct neuronal effect of continuous CO₂ exposure on retinal function in spiny chromis damselfish was also reversed by gabazine (Chung et al., 2014). Subsequently, similar results have been seen in temperate marine teleosts [three-spined stickleback (Gasterosteus aculeatus; Lai et al., 2015); and split-nose rockfish (Sebastes diploproa; Hamilton et al., 2014)], and maybe more surprisingly, in the early freshwater life stage of pink salmon (Ou et al., 2015) and the facultative air-breathing striped catfish (Pangasianodon hypophthalmus; Regan et al., 2016). Indeed, the effects of CO₂ on the GABA_A receptor are apparently not limited to fish, because they have also been found in a gastropod, the humpbacked conch (Watson et al., 2014). The link between elevated CO₂ and the GABA_A receptor probably lies in the fact that this receptor is an ion channel with conductivity for Cl⁻ and HCO₃⁻, and disruption of the neuronal gradients for these ions could readily alter its function (Nilsson et al., 2012; Heuer and Grosell, 2014). For fish, acid–base regulation is very much dependent on regulation of the levels of Cl⁻ and HCO₃⁻, and it is perhaps not so surprising that neural functions depending on these ions could be affected by hypercapnia.

Over the last few years, it has thus become clear that neural functions of many marine species are vulnerable to ocean acidification. Although reduced AAS may reduce fitness in the long run, it can be acutely detrimental for an animal not to respond with an appropriate behaviour to cues in its environment. Likewise, it does not matter how large and unaffected AAS is by ocean acidification, if behaviour becomes severely maladaptive. As an example, a study on humpbacked conch showed that AAS per se was not impaired by elevated CO₂, whereas it had a negative impact on behaviour, so that a larger proportion of snails failed to elicit an escape response when exposed to odour from a predator (Watson et al., 2014). In this case, one would arrive at two completely different predictions for the future of these snails, had only one of the variables been investigated. It is therefore crucial to implement both physiology and animal behaviour, and the consequences of alterations caused by warming and acidification, in attempts to predict the impact of climate change (Nagelkerken and Munday, 2016).

From an experimental point of view, care has also to be taken to assure that measurements of $\dot{M} O_{2 rest}$ and $\dot{M} O_{2 \max}$ ⁠, in ocean acidification conditions, are not confounded by behavioural alterations, including changes in activity or drive to exercise. Even more worryingly, although studies have shown negative effects of temperature and CO₂ on oxygen uptake to be alleviated through developmental or transgeneration acclimation (Donelson et al., 2011, 2012; Donelson and Munday, 2012; Donelson, 2015), this may not be the case for behavioural abnormalities induced by elevated CO₂ (Allan et al., 2014; Munday, et al., 2014; Welch et al., 2014). Interestingly, a recent study on three-spined stickleback showed that offspring from parents exposed to high CO₂ had lower survival when reared in control conditions compared with offspring reared in high-CO₂ water (Schade et al., 2014), but the effect appeared to be opposite for growth. Obviously, more studies are warranted to characterize these seemingly complex relationships.

Conclusions

Bozinovic and Pörtner (2015) argue that ‘Development of unifying concepts such as OCLTT is relevant for interpreting existing and future findings in a coherent way and for projecting the future ecological and evolutionary effects of climate change on whole-organism and ecosystem functioning’. In one sense, it is easy to agree with this statement, because the creation of predictive and testable mechanistic models can be hugely successful in inspiring researchers and moving science forward. Indeed, the idealized thermal performance curves proposed under the concept of OCLTT for AAS and the hypothesized influence of CO₂ on these curves are widely used as a framework for research. It is also a healthy sign that models are questioned, tested, rejected or improved, in light of available and new data. The patterns emerging from the present analysis indicate that such performance curves may not fit the majority of marine animals, particularly for the combined effects of high temperature and high CO₂. Even within one species, the effect of elevated CO₂ and temperature, and thereby the predicted outcome of climate change, may vary, depending on which variable is examined, and the effect on one variable may not predict the effect on another (e.g. Gräns et al., 2014; Watson et al., 2014).

The idea that ‘Available knowledge suggests unifying physiological principles of CO₂ effects, across animal groups and phyla’ (Pörtner, 2010) is not supported by the data reviewed here. Synergistic effects of ocean acidification on variables such as $\dot{M} O_{2 rest}$ and AAS appear to be rare; instead, additive effects seem to prevail, with many examples also of antagonistic effects. This variability indicates intricate underlying mechanisms. Although it is clear that climate change can have severe effects and that AAS might be a useful parameter for modelling the outcomes in some cases, species-by-species research and attempts to uncover the cause of different responses are still essential, and generalizations should be made with caution. The recent discovery of neurological and thereby behavioural effects of CO₂ exposure is currently treated as a parallel threat to marine ectotherms and a parallel area of research, because it is not linked to the OCLTT framework. However, the neurological alterations, although equally important on their own, may also prove to be confounding factors in experimental studies of other aspects of physiological performance, including resting and maximal rates of oxygen uptake.

Supplementary material

Supplementary material is available at Conservation Physiology online.

Funding

This work was supported by the Research Council of Norway, the University of Oslo, and the European Cooperation in Science and Technology (COST) action ‘Conservation Physiology of Marine Fishes’ [FA1004].

Acknowledgements

The author is grateful for insightful and constructive comments given by Göran E. Nilsson on the drafts of the manuscript. Furthermore, the author wishes to thank the three referees for providing useful suggestions that significantly improved the paper.

References

Allan

BJM

Miller

McCormick

Domenici

Munday

(

2014

)

Parental effects improve escape performance of juvenile reef fish in a high-CO₂ world

Proc Biol Sci

281

20132179

Arnberg

Calosi

Spicer

Tandberg

Nilsen

Westerlund

Bechmann

(

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

Google Scholar

Crossref

WorldCat

Baumann

Talmage

Gobler

(

2012

)

Reduced early life growth and survival in a fish in direct response to increased carbon dioxide

Nature Clim Change

–

Google Scholar

Crossref

WorldCat

Bělehrádek

(

1930

)

Temperature coefficients in biology

Biol Rev

–

Google Scholar

Crossref

WorldCat

Beniash

Ivanina

Lieb

Kurochkin

Sokolova

(

2010

)

Elevated level of carbon dioxide affects metabolism and shell formation in oysters Crassostrea virginica

Mar Ecol Prog Ser

419

–

108

Google Scholar

Crossref

WorldCat

Booth

McMahon

(

1992

)

Aerobic capacity of the blue crab, Callinectes sapidus

Physiol Zool

1074

–

1091

Google Scholar

Crossref

WorldCat

Bozinovic

Pörtner

H-O

(

2015

)

Physiological ecology meets climate change

Ecol Evol

1025

–

1030

Bramanti

Movilla

Guron

Calvo

Gori

Dominguez-Carrió

Grinyó

Lopez-Sanz

Martinez-Quintana

Pelejero

et al. . (

2013

)

Detrimental effects of ocean acidification on the economically important Mediterranean red coral (Corallium rubrum)

Glob Change Biol

1897

–

1908

Google Scholar

Crossref

WorldCat

Brauner

Baker

(

2009

)

Patterns of acid-base regulation in fish

. In

Glass

Wood

, eds,

Cardio-respiratory Control in Vertebrates: Comparative and Evolutionary Aspects

Springer-Verlag

Berlin

, pp

–

Google Scholar

Google Preview

OpenURL Placeholder Text

WorldCat

Brett

(

1964

)

The respiratory metabolism and swimming performance of young sockeye salmon

J Fish Res Board Can

1183

–

1226

Google Scholar

Crossref

WorldCat

Brett

(

1971

)

Energetic responses of salmon to temperature. A study of some thermal relations in the physiology and freshwater ecology of sockeye salmon (Oncorhynchus nerka)

Am Zool

–

113

Google Scholar

Crossref

WorldCat

Briffa

de la Haye

Munday

(

2012

)

High CO₂ and marine animal behaviour: potential mechanisms and ecological consequences

Mar Pollut Bull

1519

–

1528

Brijs

Jutfelt

Clark

Gräns

Ekström

Sandblom

(

2015

)

Experimental manipulations of tissue oxygen supply do not affect warming tolerance of European perch

J Exp Biol

218

2448

–

2454

Brown

da Silva

(

1983

)

Acute metabolic-rate: temperature relationships of intact and homogenized Bullia digitalis (Gastropoda, Nassariidae)

T Roy Soc S Afr

–

Google Scholar

Crossref

WorldCat

Brown

da Silva

(

1984

)

Effects of temperature on oxygen-consumption in two closely-related whelks from different temperature regimes

J Exp Mar Biol Ecol

145

–

153

Google Scholar

Crossref

WorldCat

Bruneaux

Nikinmaa

Laine

Lindström

Primmer

Vasemägi

(

2014

)

Differences in the metabolic response to temperature acclimation in nine-spined stickleback (Pungitius pungitius) populations from contrasting thermal environments

J Exp Zool A

321

550

–

565

Google Scholar

Crossref

WorldCat

Bullock

(

1955

)

Compensation for temperature in the metabolism and activity of poikilotherms

Biol Rev

311

–

342

Google Scholar

Crossref

WorldCat

Byrne

Smith

Wes

Collard

Dubois

Graba-landry

Dworjanyn

(

2014

)

Warming influences Mg²⁺ content, while warming and acidification influence calcification and test strength of a sea urchin

Environ Sci Technol

12620

–

12627

Carey

Dupont

Lundve

Sigwart

(

2014

)

One size fits all: stability of metabolic scaling under warming and ocean acidification in echinoderms

Mar Biol

161

2131

–

2142

Google Scholar

Crossref

WorldCat

Carter

Ceballos-Osuna

Miller

Stillman

(

2013

)

Impact of ocean acidification on metabolism and energetics during early life stages of the intertidal porcelain crab Petrolisthes cinctipes

J Exp Biol

216

1412

–

1422

Castonguay

Cyr

(

1998

)

Effects on temperature on spontaneous and thyroxine-stimulated locomotor activity of Atlantic cod

J Fish Biol

303

–

313

Google Scholar

Crossref

WorldCat

Catarino

Bauwens

Dubois

(

2012

)

Acid–base balance and metabolic response of the sea urchin Paracentrotus lividus to different seawater pH and temperatures

Environ Sci Pollut Res Int

2344

–

2353

Chambers

Candelmo

Habeck

Poach

Wieczorek

Cooper

Greenfield

Phelan

(

2014

)

Effects of elevated CO₂ in the early life stages of summer flounder, Paralichthys dentatus, and potential consequences of ocean acidification

Biogeosciences

1613

–

1626

Google Scholar

Crossref

WorldCat

Chen

Snow

Lawrence

Church

Narum

Devlin

Farrell

(

2015

)

Selection for upper thermal tolerance in rainbow trout (Oncorhynchus mykiss Walbaum)

J Exp Biol

218

803

–

812

Chivers

McCormick

Nilsson

Munday

Watson

S-A

Meekan

Mitchell

Corkill

Ferrari

MCO

(

2014

)

Impaired learning of predators and lower prey survival under elevated CO₂: a consequence of neurotransmitter interference

Glob Change Biol

515

–

522

Google Scholar

Crossref

WorldCat

Christensen

Colacino

(

2000

)

Respiration in the burrowing brittlestar, Hemipholis elongata Say (Echinodermata, Ophiuroidea): a study of the effects of environmental variables on oxygen uptake

Comp Biochem Physiol A Mol Integr Physiol

127

201

–

213

Christensen

Nguyen

Byrne

(

2011

)

Thermotolerance and the effects of hypercapnia on the metabolic rate of the Ophiuroid ophionereis Schayeri: inferences for survivorship in a changing ocean

J Exp Mar Biol Ecol

403

–

Google Scholar

Crossref

WorldCat

Chung

W-S

Marshall

Watson

S-A

Munday

Nilsson

(

2014

)

Ocean acidification slows retinal function in a damselfish through interference with GABA_A receptors

J Exp Biol

217

323

–

326

Claiborne

Edwards

Morrison-Shetlar

(

2002

)

Acid-base regulation in fishes: Cellular and molecular mechanisms

J Exp Zool

293

302

–

319

Claireaux

Webber

Lagardère

J-P

Kerr

(

2000

)

Influence of water temperature and oxygenation on the aerobic metabolic scope of Atlantic cod (Gadus morhua)

J Sea Res

257

–

265

Google Scholar

Crossref

WorldCat

Claireaux

Couturier

Groison

(

2006

)

Effect of temperature on maximum swimming speed and cost of transport in juvenile European sea bass (Dicentrarchus labrax)

J Exp Biol

209

3420

–

3428

Clark

Ryan

Ingram

Woakes

Butler

Frappell

(

2005

)

Factorial aerobic scope is independent of temperature and primarily modulated by heart rate in exercising murray cod (Maccullochella peelii peelii)

Physiol Biochem Zool

347

–

355

Clark

Jeffries

Hinch

Farrell

(

2011

)

Exceptional aerobic scope and cardiovascular performance of pink salmon (Oncorhynchus gorbuscha) may underlie resilience in a warming climate

J Exp Biol

214

3074

–

3081

Clark

Sandblom

Jutfelt

(

2013

)

Aerobic scope measurements of fishes in an era of climate change: respirometry, relevance and recommendations

J Exp Biol

216

2771

–

2782

Clarke

Johnston

(

1999

)

Scaling of metabolic rate with body mass and temperature in teleost fish

J Anim Ecol

893

–

905

Google Scholar

Crossref

WorldCat

Clements

Hunt

(

2015

)

Marine animal behaviour in a high CO₂ ocean

Mar Ecol Proc Ser

536

259

–

279

Google Scholar

Crossref

WorldCat

Collard

Catarino

Bonnet

Flammang

Dubois

(

2013

)

Effects of CO₂-induced ocean acidification on physiological and mechanical properties of the starfish Asterias rubens

J Exp Mar Biol Ecol

446

355

–

362

Google Scholar

Crossref

WorldCat

Collard

Eeckhaut

Dehairs

Dubois

(

2014

)

Acid–base physiology response to ocean acidification of two ecologically and economically important holothuroids from contrasting habitats, Holothuria scabra and Holothuria parva

Environ Sci Pollut Res Int

13602

–

13614

Comeau

Jeffree

Teyssié

J-L

Gattuso

J-P

(

2010

)

Response of the Arctic pteropod Limacina helicina to projected future environmental conditions

PLoS ONE

e11362

Comeau

Carpenter

Edmunds

(

2016

)

Effects of pCO₂ on photosynthesis and respiration of tropical scleractinian corals and calcified algae

ICES J Mar Sci

fsv267

10.1093/icesjms/fsv267

Google Scholar

OpenURL Placeholder Text

WorldCat

Crossref

Couturier

Stecyk

JAW

Rummer

Munday

Nilsson

(

2013

)

Species-specific effects of near-future CO₂ on the respiratory performance of two tropical prey fish and their predator

Comp Biochem Physiol A Mol Integr Physiol

166

482

–

489

Crocker

Cech

(

1997

)

Effects of environmental hypoxia on oxygen consumption rate and swimming activity in juvenile white sturgeon, Acipenser transmontanus, in relation to temperature and life intervals

Environ Biol Fishes

383

–

389

Google Scholar

Crossref

WorldCat

Crocker

Cech

(

2002

)

The effects of dissolved gases on oxygen consumption rate and ventilation frequency in white sturgeon, Acipenser transmontanus

J Appl Ichthyol

338

–

340

Google Scholar

Crossref

WorldCat

Cucco

Sinerchia

Lefrançois

Magni

Ghezzo

Umgiesser

Perilli

Domenici

(

2012

)

A metabolic scope based model of fish response to environmental changes

Ecol Model

237&238

132

–

141

Google Scholar

Crossref

WorldCat

Cumbo

Fan

Edmunds

(

2013

)

Effects of exposure duration on the response of Pocillopora damicornis larvae to elevated temperature and high pCO₂

J Exp Mar Biol Ecol

439

100

–

107

Google Scholar

Crossref

WorldCat

Cunningham

Smith

Lamare

(

2015

)

The effects of elevated pCO₂ on growth, shell production and metabolism of cultured juvenile abalone, Haliotis iris

Aquac Res

12684.

10.1111/are.12684

Google Scholar

OpenURL Placeholder Text

WorldCat

Crossref

Deigweiher

Koschnick

Pörtner

H-O

Lucassen

(

2008

)

Acclimation of ion regulatory capacities in gills of marine fish under environmental hypercapnia

Am J Physiol Regul Integr Comp Physiol

295

R1660

–

R1670

Dell

Pawar

Savage

(

2011

)

Systematic variation in the temperature dependence of physiological and ecological traits

Proc Natl Acad Sci USA

108

10591

–

10596

Del Raye

Weng

(

2015

)

An aerobic scope-based habitat suitability index for predicting the effects of multi-dimensional climate change stressors on marine teleosts

Deep-Sea Res Pt II

113

280

–

290

Google Scholar

Crossref

WorldCat

Dickinson

Matoo

Tourek

Sokolova

Beniash

(

2013

)

Environmental salinity modulates the effects of elevated CO₂ levels on juvenile hard-shell clams, Mercenaria mercenaria

J Exp Biol

216

2607

–

2618

Dickson

Kramer

(

1971

)

Factors influencing scope for activity and active and standard metabolism of rainbow trout (Salmo gairdneri)

J Fish Res Board Can

587

–

596

Google Scholar

Crossref

WorldCat

Di Santo

(

2015

)

Ocean acidification exacerbates the impacts of global warming on embryonic little skate, Leucoraja erinacea (Mitchill)

J Exp Mar Biol Ecol

463

–

Google Scholar

Crossref

WorldCat

Dissanayake

Ishimatsu

(

2011

)

Synergistic effects of elevated CO₂ and temperature on the metabolic scope and activity in a shallow-water coastal decapod (Metapenaeus joyneri; Crustacea: Penaeidae)

ICES J Mar Sci

1147

–

1154

Google Scholar

Crossref

WorldCat

Dixson

Munday

Jones

(

2010

)

Ocean acidification disrupts the innate ability of fish to detect predator olfactory cues

Ecol Lett

–

Dodd

Grabowski

Piehler

Westfield

Ries

(

2015

)

Ocean acidification impairs crab foraging behaviour

Proc Biol Sci

282

20150333

Donelson

(

2015

)

Development in a warm future ocean may enhance performance in some species

J Exp Mar Biol Ecol

472

119

–

125

Google Scholar

Crossref

WorldCat

Donelson

Munday

(

2012

)

Thermal sensitivity does not determine acclimation capacity for a tropical reef fish

J Anim Ecol

1126

–

1131

Donelson

Munday

McCormick

Nilsson

(

2011

)

Acclimation to predicted ocean warming through developmental plasticity in a tropical reef fish

Glob Change Biol

1712

–

1719

Google Scholar

Crossref

WorldCat

Donelson

Munday

McCormick

Pitcher

(

2012

)

Rapid transgenerational acclimation of a tropical reef fish to climate change

Nature Clim Change

–

Google Scholar

Crossref

WorldCat

Doney

Ruckelshaus

Emmett Duffy

Barry

Chan

English

Galindo

Grebmeier

Hollowed

Knowlton

et al. . (

2012

)

Climate change impacts on marine ecosystems

Ann Rev Mar Sci

–

Donohue

Calosi

Bates

Laverock

Rastrick

Mark

Strobel

Widdicombe

(

2012

)

Impact of exposure to elevated pCO₂ on the physiology and behaviour of an important ecosystem engineer, the burrowing shrimp Upogebia deltaura

Aquat Biol

–

Google Scholar

Crossref

WorldCat

Dorey

Lançon

Thorndyke

Dupont

(

2013

)

Assessing physiological tipping point of sea urchin larvae exposed to a broad range of pH

Glob Change Biol

3355

–

3367

Google Scholar

OpenURL Placeholder Text

WorldCat

Duthie

(

1982

)

The respiratory metabolism of temperature-adapted flatfish at rest and during swimming activity and the use of anaerobic metabolism at moderate swimming speeds

J Exp Biol

359

–

373

Google Scholar

PubMed

OpenURL Placeholder Text

WorldCat

Dwyer

Kramer

(

1975

)

The influence of temperature on scope for activity in cutthroat trout, Salmo clarkii

T Am Fish Soc

104

552

–

554

Google Scholar

Crossref

WorldCat

Edmunds

Cumbo

Fan

T-Y

(

2013

)

Metabolic costs of larval settlement and metamorphosis in the coral Seriatopora caliendrum under ambient and elevated pCO₂

J Exp Mar Biol Ecol

443

–

Google Scholar

Crossref

WorldCat

Eliason

Clark

Hague

Hanson

Gallagher

Jeffries

Gale

Patterson

Hinch

Farrell

(

2011

)

Differences in thermal tolerance among sockeye salmon populations

Science

332

109

–

112

Ellis

Sacobie

CFD

Kieffer

Benfey

(

2013

)

The effects of dissolved oxygen and triploidy on critical thermal maximum in brook charr (Salvelinus fontinalis)

Comp Biochem Physiol A Mol Integr Physiol

166

426

–

433

Enzor

Zippay

Place

(

2013

)

High latitude fish in a high CO₂ world: synergistic effects of elevated temperature and carbon dioxide on the metabolic rates of Antarctic notothenioids

Comp Biochem Physiol A Mol Integr Physiol

164

154

–

161

Ern

Huong

DTT

Phuong

Wang

Bayley

(

2014

)

Oxygen delivery does not limit thermal tolerance in a tropical eurythermal crustacean

J Exp Biol

217

809

–

814

Ern

Huong

DTT

Phuong

Madsen

Wang

Bayley

(

2015

)

Some like it hot: thermal tolerance and oxygen supply capacity in two eurythermal crustaceans

Sci Rep

10743

Esbaugh

Ern

Nordi

Johnson

(

2016

)

Respiratory plasticity is insufficient to alleviate blood acid–base disturbances after acclimation to ocean acidification in the estuarine red drum, Sciaenops ocellatus

J Comp Physiol B

186

–

109

Fabry

(

2008

)

Marine calcifiers in a high-CO₂ ocean

Science

320

1020

–

1022

Farrell

(

2009

)

Environment, antecedents and climate change: lessons from the study of temperature physiology and river migration of salmonids

J Exp Biol

212

3771

–

3780

Farrell

(

2013

)

Aerobic scope and its optimum temperature: clarifying their usefulness and limitations – correspondence on J. Exp. Biol. 216: 2771-2782

J Exp Biol

216

4493

–

4494

Farrell

(

2016

)

Pragmatic perspective on aerobic scope: peaking, plummeting, pejus and apportioning

J Fish Biol

322

–

343

Farrell

Hinch

Cooke

Patterson

Crossin

Lapointe

Mathes

(

2008

)

Pacific salmon in hot water: applying aerobic scope models and biotelemetry to predict the success of spawning migrations

Physiol Biochem Zool

697

–

709

Fernández-Reiriz

Range

Álvarez-Salgado

Labarta

(

2011

)

Physiological energetics of juvenile clams Ruditapes decussatus in a high CO₂ coastal ocean

Mar Ecol Prog Ser

433

–

105

Google Scholar

Crossref

WorldCat

Ferrari

MCO

Munday

Rummer

McCormick

Corkill

Watson

Allan

BJM

Meekan

Chivers

(

2015

)

Interactive effects of ocean acidification and rising sea temperatures alter predation rate and predator selectivity in reef fish communities

Glob Change Biol

1848

–

1855

Google Scholar

Crossref

WorldCat

Ferreira

Anttila

Farrell

(

2014

)

Thermal optima and tolerance in the eurythermic goldfish (Carassius auratus): relationships between whole-animal aerobic capacity and maximum heart rate

Physiol Biochem Zool

599

–

611

Fey

Siepielski

Nusslé

Cervantes-Yoshida

Hwan

Huber

Fey

Catenazzi

Carlson

(

2015

)

Recent shifts in the occurrence, cause, and magnitude of animal mass mortality events

Proc Natl Acad Sci USA

112

1083

–

1088

Flynn

Bjelde

Miller

Todgham

(

2015

)

Ocean acidification exerts negative effects during warming conditions in a developing Antarctic fish

Conserv Physiol

10.1093/conphys/cov033

Google Scholar

OpenURL Placeholder Text

WorldCat

Crossref

Forstner

Wieser

(

1990

)

Patterns of routine swimming and metabolic rate in juvenile cyprinids at three temperatures: analysis with a respirometer-activity-monitoring system

J Comp Physiol B

160

–

Google Scholar

Crossref

WorldCat

Frederich

Pörtner

H-O

(

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

–

R1538

Frommel

Maneja

Lowe

Malzahn

Geffen

Folkvord

Piatkowski

Reusch

TBH

Clemmesen

(

2012

)

Severe tissue damage in Atlantic cod larvae under increasing ocean acidification

Nature Clim Change

–

Google Scholar

Crossref

WorldCat

Fry

FEJ

(

1947

)

Effects of the environment on animal activity

Publications of the Ontario Fisheries Research Laboratory

–

Google Scholar

OpenURL Placeholder Text

WorldCat

Fry

FEJ

(

1971

)

The effect of environmental factors on the physiology of fish

. In

Hoar

Randall

, eds,

Fish Physiology, Vol 6, Environmental Relations and Behavior

Academic Press

New York, London

, pp

–

Fry

FEJ

Hart

(

1948

)

The relation of temperature to oxygen consumption in the goldfish

Biol Bull

–

Gardiner

Munday

Nilsson

(

2010

)

Counter-gradient variation in respiratory performance of coral reef fishes at elevated temperatures

PLoS ONE

e13299

Gräns

Jutfelt

Sandblom

Jönsson

Wiklander

Seth

Olsson

Dupont

Ortega-Martinez

Einarsdottir

et al. . (

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

Green

Jutfelt

(

2014

)

Elevated carbon dioxide alters the plasma composition and behaviour of a shark

Biol Lett

2014.0538

Google Scholar

Crossref

WorldCat

Gunderson

Stillman

(

2015

)

Plasticity in thermal tolerance has limited potential to buffer ectotherms from global warming

Proc Biol Sci

282

20150401

Guppy

Withers

(

1999

)

Metabolic depression in animals: physiological perspectives and biochemical generalizations

Biol Rev

–

Gutowska

Melzner

Pörtner

Meier

(

2010

)

Cuttlebone calcification increases during exposure to elevated seawater pCO₂ in the cephalopod Sepia officinalis

Mar Biol

157

1653

–

1663

Google Scholar

Crossref

WorldCat

Hamilton

Holcombe

Tresguerres

(

2014

)

CO₂-induced ocean acidification increases anxiety in rockfish via alteration of GABA_A receptor functioning

Proc Biol Sci

281

2013.2509

Google Scholar

Crossref

WorldCat

Hammer

Kristiansen

Zachariassen

(

2011

)

Physiological effects of hypercapnia in the deep-sea bivalve Acesta excavata (Fabricius, 1779) (Bivalvia; Limidae)

Mar Environ Res

135

–

142

Harvey

Gwynn-Jones

Moore

(

2013

)

Meta-analysis reveals complex marine biological responses to the interactive effects of ocean acidification and warming

Ecol Evol

1016

–

1030

Hazel

Prosser

(

1974

)

Molecular mechanisms of temperature compensation in poikilotherms

Physiol Rev

620

–

677

Healy

Schulte

(

2012

Thermal acclimation is not necessary to maintain a wide thermal breadth of aerobic scope in the common killifish (Fundulus heteroclitus)

Physiol Biochem Zool

107

–

119

Healy

Schulte

(

2012

Factors affecting plasticity in whole-organism thermal tolerance in common killifish (Fundulus heteroclitus)

J Comp Physiol B

182

–

Hedges

Gurevitch

Curtis

(

1999

)

The meta-analysis of response ratios in experimental ecology

Ecology

1150

–

1156

Google Scholar

Crossref

WorldCat

Heinrich

DDU

Rummer

Morash

Watson

S-A

Simpfendorfer

Heupel

Munday

(

2014

)

A product of its environment: the epaulette shark (Hemiscyllium ocellatum) exhibits physiological tolerance to elevated environmental CO₂

Conserv Physiol

10.1093/conphys/cou047

Google Scholar

OpenURL Placeholder Text

WorldCat

Crossref

Heinrich

DDU

Watson

S-A

Rummer

Brandl

Simpfendorfer

Heupel

Munday

(

2016

)

Foraging behaviour of the epaulette shark Hemiscyllium ocellatum is not affected by elevated CO₂

ICES J Mar Sci

633

–

640

Google Scholar

Crossref

WorldCat

Heisler

(

1984

)

Acid-base regulation in fishes

. In

Hoar

Randall

, eds,

Fish Physiology, Vol 10, Part A, Gills – Anatomy, Gas Transfer, and Acid-Base Regulation

Academic Press

New York, London

, pp

315

–

401

Hennige

Wicks

Kamenos

Bakker

DCE

Findlay

Dumousseaud

Roberts

(

2014

)

Short-term metabolic and growth responses of the cold-water coral Lophelia pertusa to ocean acidification

Deep-Sea Res Pt II

–

Google Scholar

Crossref

WorldCat

Heuer

Grosell

(

2014

)

Physiological impacts of elevated carbon dioxide and ocean acidification on fish

Am J Physiol Regul Integr Comp Physiol

307

R1061

–

R1084

Hildebrandt

Niehoff

Sartoris

(

2014

)

Long-term effects of elevated CO₂ and temperature on the Arctic calanoid copepods Calanus glacialis and C. hyperoreus

Mar Pollut Bull

–

Hofmann

Todgham

(

2010

)

Living in the now: physiological mechanisms to tolerate a rapidly changing environment

Ann Rev Physiol

127

–

145

Google Scholar

Crossref

WorldCat

Holt

Jørgensen

(

2015

)

Climate change in fish: effects of respiratory constraints on optimal life history and behaviour

Proc Biol Sci

2014.1032

Google Scholar

OpenURL Placeholder Text

WorldCat

Casties

Stumpp

Ortega-Martinez

Dupont

(

2014

)

Energy metabolism and regeneration are impaired by seawater acidification in the infaunal brittlestar Amphiura filiformis

J Exp Biol

217

2411

–

2421

Intergovernmental Panel on Climate Change (IPCC)

(

2013

)

Climate Change 2013: The Physical Science Basis

Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change

Cambridge University Press

Cambridge, UK

Johansen

Jones

(

2011

)

Increasing ocean temperature reduces the metabolic performance and swimming ability of coral reef damselfishes

Glob Change Biol

2971

–

2979

Google Scholar

Crossref

WorldCat

Johansen

Messmer

Coker

Hoey

Pratchett

(

2014

)

Increasing ocean temperatures reduce activity patterns of a large commercially important coral reef fish

Glob Change Biol

1067

–

1074

Google Scholar

Crossref

WorldCat

Johnston

Dunn

(

1987

)

Temperature acclimation and metabolism in ectotherms with particular reference to teleost fish

Symp Soc Exp Biol

–

Google Scholar

PubMed

OpenURL Placeholder Text

WorldCat

Jutfelt

Hedgärde

(

2015

)

Juvenile Atlantic cod behavior appears robust to near-future CO₂ levels

Front Zool

Jutfelt

Bresolin de Souza

Vuylsteke

Sturve

(

2013

)

Behavioural disturbances in a temperate fish exposed to sustained high-CO₂ levels

PLoS ONE

e65825

Kaplan

Mooney

McCorkle

Cohen

(

2013

)

Adverse effects of ocean acidification on early development of squid (Doryteuthis pealeii)

PLoS ONE

e63714

Khan

Herbert

(

2012

)

The behavioural thermal preference of the common triplefin (Forsterygion lapillum) tracks aerobic scope optima at the upper thermal limit of its distribution

J Therm Biol

118

–

124

Google Scholar

Crossref

WorldCat

Khan

Iftikar

Herbert

Gnaiger

Hickey

AJR

(

2014

Thermal plasticity of skeletal muscle mitochondrial activity and whole animal respiration in a common intertidal triplefin fish, Forsterygion lapillum (family: Tripterygiidae)

J Comp Physiol B

184

991

–

1001

Khan

Pether

Bruce

Walker

Herbert

(

2014

Optimum temperatures for growth and feed conversion in cultured hapuku (Polyprion oxygeneios)—is there a link to aerobic metabolic scope and final temperature preference?

Aquaculture

430

107

–

113

Google Scholar

Crossref

WorldCat

Kinkead

Perry

(

1991

)

The effects of catecholamines on ventilation in rainbow trout during hypoxia or hypercapnia

Respir Physiol

–

Kreiss

Michael

Lucassen

Jutfelt

Motyka

Dupont

Pörtner

(

2015

)

Ocean warming and acidification modulate energy budget and gill ion regulatory mechanisms in Atlantic cod (Gadus morhua)

J Comp Physiol B

185

767

–

781

Lai

Jutfelt

Nilsson

(

2015

)

Altered neurotransmitter function in CO₂-exposed stickleback (Gasterosteus aculeatus): a temperate model species for ocean acidification research

Conserv Physiol

10.1093/conphys/cov018

Google Scholar

OpenURL Placeholder Text

WorldCat

Crossref

Lajeunesse

(

2011

)

On the meta-analysis of response ratios for studies with correlated and multi-group designs

Ecology

2049

–

2055

Langenbuch

Pörtner

(

2002

)

Changes in metabolic rate and N excretion in the marine invertebrate Sipunculus nudus under conditions of environmental hypercapnia: identifying effective acid–base variables

J Exp Biol

205

1153

–

1160

Google Scholar

PubMed

OpenURL Placeholder Text

WorldCat

Lannig

Eilers

Pörtner

Sokolova

Bock

(

2010

)

Impact of ocean acidification on energy metabolism of oyster, Crassostrea gigas—changes in metabolic pathways and thermal response

Mar Drugs

2318

–

2339

Lardies

Arias

Poupin

Manríquez

Torres

Vargas

Navarro

Lagos

(

2014

)

Differential response to ocean acidification in physiological traits of Concholepas concholepas populations

J Sea Res

127

–

134

Google Scholar

Crossref

WorldCat

Lee

Farrell

Lotto

MacNutt

Hinch

Healey

(

2003

)

The effect of temperature on swimming performance and oxygen consumption in adult sockeye (Oncorhynchus nerka) and coho (O. kisutch) salmon stocks

J Exp Biol

206

3239

–

3251

Lefevre

Watson

S-A

Munday

Nilsson

(

2015

)

Will jumping snails prevail? Influence of near-future CO₂, temperature and hypoxia on respiratory performance in the tropical conch Gibberulus gibberulus gibbosus

J Exp Biol

218

2991

–

3001

Lefrancois

Claireaux

(

2003

)

Influence of ambient oxygenation and temperature on metabolic scope and scope for heart rate in the common sole Solea solea

Mar Ecol Prog Ser

259

273

–

284

Google Scholar

Crossref

WorldCat

Gao

(

2012

)

A marine secondary producer respires and feeds more in a high CO₂ ocean

Mar Pollut Bull

699

–

703

Liu

(

2012

)

Effects of ocean acidification on the metabolic rates of three species of bivalve from Southern coast of China

Chin J Oceanol Limn

206

–

211

Google Scholar

Crossref

WorldCat

Lowe

Davison

(

2006

)

Thermal sensitivity of scope for activity in Pagothenia borchgrevinki, a cryopelagic antarctic nototheniid fish

Polar Biol

971

–

977

Google Scholar

Crossref

WorldCat

Lönnstedt

Munday

McCormick

Ferrari

MCO

Chivers

(

2013

)

Ocean acidification and responses to predators: can sensory redundancy reduce the apparent impacts of elevated CO₂ on fish?

Ecol Evol

3565

–

3575

McElroy

Nguyen

Byrne

(

2012

)

Respiratory response of the intertidal seastar Parvulastra exigua to contemporary and near-future pulses of warming and hypercapnia

J Exp Mar Biol Ecol

416–417

–

Google Scholar

Crossref

WorldCat

Mallekh

Lagardere

(

2002

)

Effect of temperature and dissolved oxygen concentration on the metabolic rate of the turbot and the relationship between metabolic scope and feeding demand

J Fish Biol

1105

–

1115

Google Scholar

Crossref

WorldCat

Manríquez

Jara

Mardones

Navarro

Torres

Lardies

Vargas

Duarte

Widdicombe

Salisbury

et al. . (

2013

)

Ocean acidification disrupts prey responses to predator cues but not net prey shell growth in Concholepas concholepas (Loco)

PLoS ONE

e68643

Manríquez

Jara

Mardones

Torres

Navarro

Lardies

Vargas

Duarte

Lagos

(

2014

)

Ocean acidification affects predator avoidance behaviour but not prey detection in the early ontogeny of a keystone species

Mar Ecol Prog Ser

502

157

–

167

Google Scholar

Crossref

WorldCat

Marras

Cucco

Antognarelli

Azzurro

Milazzo

Bariche

Butenschön

Kay

Di Bitetto

Quattrocchi

et al. . (

2015

)

Predicting future thermal habitat suitability of competing native and invasive fish species: From metabolic scope to oceanographic modelling

Conserv Physiol

10.1093/conphys/cou059

Google Scholar

OpenURL Placeholder Text

WorldCat

Crossref

Matoo

Ivanina

Ullstad

Beniash

Sokolova

(

2013

)

Interactive effects of elevated temperature and CO₂ levels on metabolism and oxidative stress in two common marine bivalves (Crassostrea virginica and Mercenaria mercenaria)

Comp Biochem Physiol A Mol Integr Physiol

164

545

–

553

Meinshausen

Smith

Calvin

Daniel

Kainuma

MLT

Lamarque

Matsumoto

Montzka

Raper

SCB

Riahi

et al. . (

2011

)

The RCP greenhouse gas concentrations and their extensions from 1765 to 2300

Climatic Change

109

213

–

241

Google Scholar

Crossref

WorldCat

Melatunan

Calosi

Simon

John

Widdicombe

(

2011

)

Exposure to elevated temperature and PCo₂ reduces respiration rate and energy status in the periwinkle Littorina littorea

Physiol Biochem Zool

583

–

594

Melzner

Göbel

Langenbuch

Gutowska

Pörtner

H-O

Lucassen

(

2009

)

Swimming performance in Atlantic cod (Gadus morhua) following long-term (4–12 months) acclimation to elevated seawater

Aquat Toxicol

–

Metzger

Sartoris

Langenbuch

Pörtner

(

2007

)

Influence of elevated CO₂ concentrations on thermal tolerance of the edible crab Cancer pagurus

J Therm Biol

144

–

151

Google Scholar

Crossref

WorldCat

Michaelidis

Spring

Pörtner

H-O

(

2007

)

Effects of long-term acclimation to environmental hypercapnia on extracellular acid−base status and metabolic capacity in Mediterranean fish Sparus aurata

Mar Biol

150

1417

–

1429

Google Scholar

Crossref

WorldCat

Mingliang

Jianguang

Jihong

Bin

Shengmin

Yuze

Yaping

(

2011

)

Effect of marine acidification on calcification and respiration of Chlamys farreri

J Shellfish Res

267

–

271

Google Scholar

Crossref

WorldCat

Munday

Crawley

Nilsson

(

2009

)

Interacting effects of elevated temperature and ocean acidification on the aerobic performance of coral reef fishes

Mar Ecol Prog Ser

388

235

–

242

Google Scholar

Crossref

WorldCat

Munday

McCormick

Nilsson

(

2012

)

Impact of global warming and rising CO₂ levels on coral reef fishes: what hope for the future?

J Exp Biol

215

3865

–

3873

Munday

Pratchett

Dixson

Donelson

Endo

Reynolds

Knuckey

(

2013

)

Elevated CO₂ affects the behavior of an ecologically and economically important coral reef fish

Mar Biol

160

2137

–

2144

Google Scholar

Crossref

WorldCat

Munday

Cheal

Dixson

Rummer

Fabricius

(

2014

)

Behavioural impairment in reef fishes caused by ocean acidification at CO₂ seeps

Nature Clim Change

487

–

492

Google Scholar

Crossref

WorldCat

Nagelkerken

Munday

(

2016

)

Animal behaviour shapes the ecological effects of ocean acidification and warming: moving from individual to community-level responses

Glob Change Biol

974

–

989

Google Scholar

Crossref

WorldCat

Nakamura

Ohki

Suzuki

Sakai

(

2011

)

Coral larvae under ocean acidification: survival, metabolism, and metamorphosis

PLoS ONE

e14521

Navarro

Torres

Acuña

Duarte

Manriquez

Lardies

Lagos

Vargas

Aguilera

(

2013

)

Impact of medium-term exposure to elevated pCO₂ levels on the physiological energetics of the mussel Mytilus chilensis

Chemosphere

1242

–

1248

Neuenfeldt

Andersen

Hinrichsen

(

2009

)

Some Atlantic cod Gadus morhua in the Baltic sea visit hypoxic water briefly but often

J Fish Biol

290

–

294

Newell

(

1966

)

Effect of temperature on the metabolism of poikilotherms

Nature

212

426

–

428

Google Scholar

Crossref

WorldCat

Newell

Pye

(

1970

)

The influence of thermal acclimation on the relation between oxygen consumption and temperature in Littorina littorea (L.) and Mytilus edulis L

Comp Biochem Physiol

385

–

397

Google Scholar

Crossref

WorldCat

Newell

Pye

(

1971

)

Quantitative aspects of the relationship between metabolism and temperature in the winkle Littorina littorea (L.)

Comp Biochem Physiol B

635

–

650

Google Scholar

Crossref

WorldCat

Nikinmaa

Anttila

(

2015

)

Responses of marine animals to ocean acidification

. In

Botana

Louzao

Vilariño

, eds,

Climate Change and Marine and Freshwater Toxins

Walter de Gruyter GmbH & Co.

Berlin/Boston, pp

–

124

Nilsson

Crawley

Lunde

Munday

(

2009

)

Elevated temperature reduced the respiratory scope of coral reef fishes

Glob Change Biol

1405

–

1412

Google Scholar

Crossref

WorldCat

Nilsson

Dixson

Domenici

McCormick

Sorensen

Watson

Munday

(

2012

)

Near-future carbon dioxide levels alter fish behaviour by interfering with neurotransmitter function

Nature Clim Change

201

–

204

Google Scholar

Crossref

WorldCat

Noisette

Richard

Le Fur

Peck

Davoult

Martin

(

2015

)

Metabolic responses to temperature stress under elevated pCO₂ in Crepidula fornicata

J Mollus Stud

238

–

246

Google Scholar

Crossref

WorldCat

Noisette

Bordeyne

Davoult

Martin

(

2016

)

Assessing the physiological responses of the gastropod Crepidula fornicata to predicted ocean acidification and warming

Limnol Oceanogr

430

–

444

Norin

Malte

Clark

(

2014

)

Aerobic scope does not predict the performance of a tropical eurythermal fish at elevated temperatures

J Exp Biol

217

244

–

251

Nowell

Brownscombe

Gutowsky

Murchie

Suski

Danylchuk

Shultz

Cooke

(

2015

)

Swimming energetics and thermal ecology of adult bonefish (Albula vulpes): a combined laboratory and field study in Eleuthera, the Bahamas

Environ Biol Fish

2133

–

2146

Google Scholar

Crossref

WorldCat

Hamilton

Eom

Lyall

Gallup

Jiang

Lee

Yun

S-S

Brauner

(

2015

)

Responses of pink salmon to CO₂-induced aquatic acidification

Nature Clim Change

950

–

955

Google Scholar

Crossref

WorldCat

Overgaard

Andersen

Findsen

Pedersen

PBM

Hansen

Ozolina

Wang

(

2012

)

Aerobic scope and cardiovascular oxygen transport is not compromised at high temperatures in the toad Rhinella marina

J Exp Biol

215

3519

–

3526

Padilla-Gamiño

Kelly

Evans

Hofmann

(

2013

)

Temperature and CO₂ additively regulate physiology, morphology and genomic responses of larval sea urchins, Strongylocentrotus purpuratus

Proc Biol Sci

280

2013.0155

Google Scholar

Crossref

WorldCat

Pang

Cao

Peng

(

2010

)

The effects of feeding on the swimming performance and metabolic response of juvenile southern catfish, Silurus meridionalis, acclimated at different temperatures

Comp Biochem Physiol A Mol Integr Physiol

155

253

–

258

Pang

Yuan

X-Z

Cao

Z-D

S-J

(

2013

)

The effects of temperature and exercise training on swimming performance in juvenile qingbo (Spinibarbus sinensis)

J Comp Physiol B

183

–

108

Parker

Ross

Connor

Pörtner

Scanes

Wright

(

2013

)

Predicting the response of molluscs to the impact of ocean acidification

Biology

651

–

692

Patnaik

Rao

DGVP

(

1985

)

Studies on the standard and active rates of metabolism of a tropical intertidal gastropod Littorina (Littoraria) undulata (Gray): effect of body size and temperature

Ital J Zool

101

–

109

Google Scholar

OpenURL Placeholder Text

WorldCat

Payne

Smith

van der Meulen

Taylor

Watanabe

Takahashi

Marzullo

Gray

Cadiou

Suthers

(

2016

)

Temperature dependence of fish performance in the wild: links with species biogeography and physiological thermal tolerance

Funct Ecol

12618

10.1111/1365-2435.12618

Google Scholar

OpenURL Placeholder Text

WorldCat

Crossref

Perry

Abdallah

(

2012

)

Mechanisms and consequences of carbon dioxide sensing in fish

Respir Physiol Neurobiol

184

309

–

315

Peterson

Anderson

(

1969

)

Influence of temperature change on spontaneous locomotor activity and oxygen consumption of Atlantic salmon, Salmo salar, acclimated to two temperatures

J Fish Res Board Can

–

109

Google Scholar

Crossref

WorldCat

Pimentel

Pegado

Repolho

Rosa

(

2014

)

Impact of ocean acidification in the metabolism and swimming behavior of the dolphinfish (Coryphaena hippurus) early larvae

Mar Biol

161

725

–

729

Google Scholar

Crossref

WorldCat

Pope

Ellis

Scolamacchia

Scolding

JWS

Keay

Chingombe

Shields

Wilcox

Speirs

Wilson

et al. . (

2014

)

European sea bass, Dicentrarchus labrax, in a changing ocean

Biogeosciences

2519

–

2530

Google Scholar

Crossref

WorldCat

Pörtner

H-O

(

2001

)

Climate change and temperature-dependent biogeography: oxygen limitation of thermal tolerance in animals

Naturwissenschaften

137

–

146

Pörtner

H-O

(

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

Pörtner

H-O

(

2012

)

Integrating climate-related stressor effects on marine organisms: unifying principles linking molecule to ecosystem-level changes

Mar Ecol Prog Ser

470

273

–

290

Google Scholar

Crossref

WorldCat

Pörtner

H-O

(

2014

)

How and how not to investigate the oxygen and capacity limitation of thermal tolerance (OCLTT) and aerobic scope – remarks on the article by Gräns et al

J Exp Biol

217

4432

–

4433

Pörtner

H-O

Farrell

(

2008

)

Physiology and climate change

Science

322

690

–

692

Pörtner

H-O

Knust

(

2007

)

Climate change affects marine fishes through the oxygen limitation of thermal tolerance

Science

315

–

Pörtner

H-O

Lannig

(

2009

)

Oxygen and capacity limited thermal tolerance

. In

Jeffrey

, eds.

Fish Physiology, Vol 27, Hypoxia

Academic Press

New York, London

, pp

143

–

191

Google Scholar

Google Preview

OpenURL Placeholder Text

WorldCat

Pörtner

H-O

Reipschläger

Heisler

(

1998

)

Acid–base regulation, metabolism and energetics in Sipunculus nudus as a function of ambient carbon dioxide level

J Exp Biol

201

–

Google Scholar

PubMed

OpenURL Placeholder Text

WorldCat

Pörtner

H-O

Mark

Bock

(

2004

Oxygen limited thermal tolerance in fish?: Answers obtained by nuclear magnetic resonance techniques

Respir Physiol Neurobiol

141

243

–

260

Pörtner

H-O

Langenbuch

Reipschläger

(

2004

Biological impact of elevated ocean CO₂ concentrations: lessons from animal physiology and earth history

J Oceanogr

705

–

718

Google Scholar

Crossref

WorldCat

Pörtner

H-O

Langenbuch

Michaelidis

(

2005

)

Synergistic effects of temperature extremes, hypoxia, and increases in CO₂ on marine animals: from earth history to global change

J Geophys Res

110

C09S10

Google Scholar

Crossref

WorldCat

Pörtner

H-O

Karl

Boyd

Cheung

WWL

Lluch-Cota

Nojiri

Schmidt

Zavialov

(

2014

)

Ocean systems

. In

Field

Barros

Dokken

Mach

Mastrandrea

Bilir

Chatterjee

Ebi

Estrada

Genova

et al. . eds.

Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change

Cambridge University Press

Cambridge, UK

and

New York, NY, USA

, pp

411

–

484

Google Scholar

Google Preview

OpenURL Placeholder Text

WorldCat

Prosser

(

1991

)

Temperature. In Prosser CL ed. Environmental and Metabolic Animal Physiology, Ed 4. Wiley-Liss, Inc., New York, NY, USA, pp 109–165

Prosser

Nelson

(

1981

)

The role of nervous systems in temperature adaptation of poikilotherms

Annu Rev Physiol

281

–

300

Przeslawski

Byrne

Mellin

(

2015

)

A review and meta-analysis of the effects of multiple abiotic stressors on marine embryos and larvae

Glob Change Biol

2122

–

2140

Google Scholar

Crossref

WorldCat

Putnam

Mayfield

Fan

Chen

Gates

(

2012

)

The physiological and molecular responses of larvae from the reef-building coral Pocillopora damicornis exposed to near-future increases in temperature and pCO₂

Mar Biol

160

2157

–

2173

Google Scholar

Crossref

WorldCat

Rahmstorf

Coumou

(

2011

)

Increase of extreme events in a warming world

Proc Natl Acad Sci USA

108

17905

–

17909

Regan

Turko

Heras

Andersen

Lefevre

Wang

Bayley

Brauner

Huong

DTT

Phuong

et al. . (

2016

)

Ambient CO₂, fish behaviour and altered GABAergic neurotransmission: exploring the mechanism of CO₂-altered behaviour by taking a hypercapnia dweller down to low CO₂ levels

J Exp Biol

219

109

–

118

Reynolds

Casterlin

(

1982

)

Thermoregulatory behaviour, diel activity and the relationship of spontaneous locomotor activity to temperature in the sea raven, Hemitripterus americanus (Gmelin)

J Fish Biol

–

Google Scholar

Crossref

WorldCat

Ries

Cohen

McCorkle

(

2009

)

Marine calcifiers exhibit mixed responses to CO₂-induced ocean acidification

Geology

1131

–

1134

Google Scholar

Crossref

WorldCat

Rivest

Hofmann

(

2014

)

Responses of the metabolism of the larvae of Pocillopora damicornis to ocean acidification and warming

PLoS ONE

e96172

Rosa

Seibel

(

2008

)

Synergistic effects of climate-related variables suggest future physiological impairment in a top oceanic predator

Proc Natl Acad Sci USA

105

20776

–

20780

Rosa

Trübenbach

Repolho

Pimentel

Faleiro

Boavida-Portugal

Baptista

Lopes

Dionísio

Leal

et al. . (

2013

)

Lower hypoxia thresholds of cuttlefish early life stages living in a warm acidified ocean

Proc Biol Sci

280

20131695

Rosa

Baptista

Lopes

Pegado

Paula

Trübenbach

Leal

Calado

Repolho

(

2014

Early-life exposure to climate change impairs tropical shark survival

Proc Biol Sci

281

20141738

Rosa

Trübenbach

Pimentel

Boavida-Portugal

Faleiro

Baptista

Dionísio

Calado

Pörtner

H-O

Repolho

(

2014

Differential impacts of ocean acidification and warming on winter and summer progeny of a coastal squid (Loligo vulgaris)

J Exp Biol

217

518

–

525

Rummer

Stecyk

JAW

Couturier

Watson

S-A

Nilsson

Munday

(

2013

)

Elevated CO₂ enhances aerobic scope of a coral reef fish

Conserv Physiol

10.1093/conphys/cot023

Google Scholar

OpenURL Placeholder Text

WorldCat

Crossref

Rummer

Couturier

Stecyk

JAW

Gardiner

Kinch

Nilsson

Munday

(

2014

)

Life on the edge: thermal optima for aerobic scope of Equatorial reef fishes are close to current day temperatures

Glob Change Biol

1055

–

1066

Google Scholar

Crossref

WorldCat

Sandblom

Gräns

Axelsson

Seth

(

2014

)

Temperature acclimation rate of aerobic scope and feeding metabolism in fishes: implications in a thermally extreme future

Proc Biol Sci

281

20141490

Schade

Clemmesen

Wegner

(

2014

)

Within- and transgenerational effects of ocean acidification on life history of marine three-spined stickleback (Gasterosteus aculeatus)

Mar Biol

161

1667

–

1676

Google Scholar

Crossref

WorldCat

Schalkhausser

Bock

Stemmer

Brey

Pörtner

H-O

Lannig

(

2013

)

Impact of ocean acidification on escape performance of the king scallop, Pecten maximus, from Norway

Mar Biol

160

1995

–

2006

Google Scholar

Crossref

WorldCat

Schalkhausser

Bock

Pörtner

H-O

Lannig

(

2014

)

Escape performance of temperate king scallop, Pecten maximus under ocean warming and acidification

Mar Biol

161

2819

–

2829

Google Scholar

Crossref

WorldCat

Schulte

(

2015

)

The effects of temperature on aerobic metabolism: towards a mechanistic understanding of the responses of ectotherms to a changing environment

J Exp Biol

218

1856

–

1866

Schurmann

Steffensen

(

1994

)

Spontaneous swimming activity of Atlantic cod Gadus morhua exposed to graded hypoxia at three temperatures

J Exp Biol

197

129

–

142

Google Scholar

PubMed

OpenURL Placeholder Text

WorldCat

Seebacher

White

Franklin

(

2015

)

Physiological plasticity increases resilience of ectothermic animals to climate change

Nature Clim Change

–

Google Scholar

Crossref

WorldCat

Segal

(

1961

)

Acclimation in molluscs

Am Zool

235

–

244

Google Scholar

Crossref

WorldCat

Seth

Gräns

Sandblom

Olsson

Wiklander

Johnsson

Axelsson

(

2013

)

Metabolic scope and interspecific competition in sculpins of Greenland are influenced by increased temperatures due to climate change

PLoS ONE

e62859

Simpson

Munday

Wittenrich

Manassa

Dixson

Gagliano

Yan

(

2011

)

Ocean acidification erodes crucial auditory behaviour in a marine fish

Biol Lett

917

–

920

Small

Calosi

White

Spicer

Widdicombe

(

2010

)

Impact of medium-term exposure to CO₂ enriched seawater on the physiological functions of the velvet swimming crab Necora puber

Aquat Biol

–

Google Scholar

Crossref

WorldCat

Small

Calosi

Boothroyd

Widdicombe

Spicer

(

2015

)

Stage-specific changes in physiological and life-history responses to elevated temperature and pCO₂ during the larval development of the European lobster Homarus gammarus (L.)

Physiol Biochem Zool

494

–

507

Sokolova

Pörtner

(

2003

)

Metabolic plasticity and critical temperatures for aerobic scope in a eurythermal marine invertebrate (Littorina saxatilis, Gastropoda: Littorinidae) from different latitudes

J Exp Biol

206

195

–

207

Spady

Watson

S-A

Chase

Munday

(

2014

)

Projected near-future CO₂ levels increase activity and alter defensive behaviours in the tropical squid Idiosepius pygmaeus

Biol Open

1063

–

1070

Speers-Roesch

Brauner

Farrell

Hickey

AJR

Renshaw

GMC

Wang

Richards

(

2012

Hypoxia tolerance in elasmobranchs. II. Cardiovascular function and tissue metabolic responses during progressive and relative hypoxia exposures

J Exp Biol

215

103

–

114

Speers-Roesch

Richards

Brauner

Farrell

Hickey

AJR

Wang

Renshaw

GMC

(

2012

Hypoxia tolerance in elasmobranchs. I. Critical oxygen tension as a measure of blood oxygen transport during hypoxia exposure

J Exp Biol

215

–

102

Steckbauer

Ramajo

Hendriks

Fernandez

Lagos

Prado

Duarte

(

2015

)

Synergistic effects of hypoxia and increasing CO₂ on benthic invertebrates of the central chilean coast

Front Mar Sci

10.3389/fmars.2015.00049

Google Scholar

Crossref

WorldCat

Steinhausen

Sandblom

Eliason

Verhille

Farrell

(

2008

)

The effect of acute temperature increases on the cardiorespiratory performance of resting and swimming sockeye salmon (Oncorhynchus nerka)

J Exp Biol

211

3915

–

3926

Stevens

Fry

FEJ

(

1972

)

The effect of changes in ambient temperature on spontaneous activity in skipjack tuna

Comp Biochem Physiol A Comp Physiol

803

–

805

Stoner

Ottmar

Hurst

(

2006

)

Temperature affects activity and feeding motivation in pacific halibut: implications for bait-dependent fishing

Fish Res

202

–

209

Google Scholar

Crossref

WorldCat

Strahl

Stolz

Uthicke

Vogel

Noonan

SHC

Fabricius

(

2015

)

Physiological and ecological performance differs in four coral taxa at a volcanic carbon dioxide seep

Comp Biochem Physiol A Mol Integr Physiol

184

179

–

186

Strand

Huse

(

2007

)

Vertical migration in adult Atlantic cod (Gadus morhua)

Can J Fish Aquat Sci

1747

–

1760

Google Scholar

Crossref

WorldCat

Strobel

Bennecke

Leo

Mintenbeck

Pörtner

Mark

(

2012

)

Metabolic shifts in the antarctic fish Notothenia rossii in response to rising temperature and PCO₂

Front Zool

Stumpp

Wren

Melzner

Thorndyke

Dupont

(

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 A Mol Integr Physiol

160

331

–

340

Sui

Huang

Wang

(

2015

)

Anti-predatory responses of the thick shell mussel Mytilus coruscus exposed to seawater acidification and hypoxia

Mar Environ Res

109

159

–

167

Sun

Tang

Zhou

Wang

(

2016

)

Comparative studies on the effects of seawater acidification caused by CO₂ and HCl enrichment on physiological changes in Mytilus edulis

Chemosphere

144

2368

–

2376

Sundin

Jutfelt

(

2016

)

9–28 d of exposure to elevated pCO₂ reduces avoidance of predator odour but had no effect on behavioural lateralization or swimming activity in a temperate wrasse (Ctenolabrus rupestris)

ICES J Mar Sci

620

–

632

Google Scholar

Crossref

WorldCat

Sylvestre

E-L

Lapointe

Dutil

J-D

Guderley

(

2007

)

Thermal sensitivity of metabolic rates and swimming performance in two latitudinally separated populations of cod, Gadus morhua L

J Comp Physiol B

177

447

–

460

Takahashi

Kurihara

(

2013

)

Ocean acidification does not affect the physiology of the tropical coral Acropora digitifera during a 5-week experiment

Coral Reefs

305

–

314

Google Scholar

Crossref

WorldCat

Tamburrini

Romano

Carratore

Kunzmann

Coletta

di Prisco

(

1998

)

The hemoglobins of the Antarctic fishes Artedidraco orianae and Pogonophryne scotti: amino acid sequence, lack of cooperativity, and ligand binding properties

J Biol Chem

273

32452

–

32459

Tewksbury

Huey

Deutsch

(

2008

)

Putting the heat on tropical animals

Science

320

1296

Thomsen

Melzner

(

2010

)

Moderate seawater acidification does not elicit long-term metabolic depression in the blue mussel Mytilus edulis

Mar Biol

157

2667

–

2676

Google Scholar

Crossref

WorldCat

Thor

Dupont

(

2015

)

Transgenerational effects alleviate severe fecundity loss during ocean acidification in a ubiquitous planktonic copepod

Glob Change Biol

2261

–

2271

Google Scholar

Crossref

WorldCat

Thor

Oliva

(

2015

)

Ocean acidification elicits different energetic responses in an Arctic and a boreal population of the copepod Pseudocalanus acuspes

Mar Biol

162

799

–

807

Google Scholar

Crossref

WorldCat

Thyrring

Rysgaard

Blicher

Sejr

(

2015

)

Metabolic cold adaptation and aerobic performance of blue mussels (Mytilus edulis) along a temperature gradient into the high Arctic region

Mar Biol

162

235

–

243

Google Scholar

Crossref

WorldCat

Tirsgaard

Behrens

Steffensen

(

2015

The effect of temperature and body size on metabolic scope of activity in juvenile Atlantic cod Gadus morhua L

Comp Biochem Physiol A Mol Integr Physiol

179

–

Tirsgaard

Moran

Steffensen

(

2015

Prolonged SDA and reduced digestive efficiency under elevated CO₂ may explain reduced growth in Atlantic cod (Gadus morhua)

Aquat Toxicol

158

171

–

180

Yuan

Huang

Johnson

(

2012

)

Aerobic swimming performance of juvenile largemouth bronze gudgeon (Coreius guichenoti) in the Yangtze river

J Exp Zool A

317

294

–

302

Google Scholar

Crossref

WorldCat

Uthicke

Liddy

Nguyen

Byrne

(

2014

)

Interactive effects of near-future temperature increase and ocean acidification on physiology and gonad development in adult Pacific sea urchin, Echinometra sp. A

Coral Reefs

831

–

845

Google Scholar

Crossref

WorldCat

Vagner

Lacoue-Labarthe

Zambonino Infante

J-L

Mazurais

Dubillot

Le Delliou

Quazuguel

Lefrançois

(

2015

)

Depletion of essential fatty acids in the food source affects aerobic capacities of the golden grey mullet Liza aurata in a warming seawater context

PLoS ONE

e0126489

van Dijk

Tesch

Hardewig

Pörtner

H-O

(

1999

)

Physiological disturbances at critically high temperatures: a comparison between stenothermal Antarctic and eurythermal temperate eelpouts (Zoarcidae)

J Exp Biol

202

3611

–

3621

Google Scholar

PubMed

OpenURL Placeholder Text

WorldCat

Vargas

de la Hoz

Aguilera

Martín

Manríquez

Navarro

Torres

Lardies

Lagos

(

2013

)

CO₂-driven ocean acidification reduces larval feeding efficiency and changes food selectivity in the mollusk Concholepas concholepas

J Plankton Res

1059

–

1068

Google Scholar

Crossref

WorldCat

Verberk

WCEP

Sommer

Davidson

Viant

(

2013

)

Anaerobic metabolism at thermal extremes: a metabolomic test of the oxygen limitation hypothesis in an aquatic insect

Integr Comp Biol

609

–

619

Verberk

WCEP

Overgaard

Ern

Bayley

Wang

Boardman

Terblanche

(

2016

)

Does oxygen limit thermal tolerance in arthropods? A critical review of current evidence

Comp Biochem Physiol A Mol Integr Physiol

192

–

Vulesevic

McNeill

Perry

(

2006

)

Chemoreceptor plasticity and respiratory acclimation in the zebrafish Danio rerio

J Exp Biol

209

1261

–

1273

Wang

Overgaard

(

2007

)

The heartbreak of adapting to global warming

Science

315

–

Wang

Lefevre

Iversen

Findorf

Buchanan

McKenzie

(

2014

)

Anaemia only causes a small reduction in the upper critical temperature of sea bass: is oxygen delivery the limiting factor for tolerance of acute warming in fishes?

J Exp Biol

217

4275

–

4278

Wang

(

2015

)

Physiological energetics of the thick shell mussel Mytilus coruscus exposed to seawater acidification and thermal stress

Sci Total Environ

514

261

–

272

Watson

S-A

(

2015

)

Giant clams and rising CO₂: light may ameliorate effects of ocean acidification on a solar-powered animal

PLoS ONE

e0128405

Watson

S-A

Lefevre

McCormick

Domenici

Nilsson

Munday

(

2014

)

Marine mollusc predator-escape behaviour altered by near-future carbon dioxide levels

Proc Biol Sci

281

20132377

Welch

Watson

S-A

Welsh

McCormick

Munday

(

2014

)

Effects of elevated CO₂ on fish behaviour undiminished by transgenerational acclimation

Nature Clim Change

1086

–

1089

Google Scholar

Crossref

WorldCat

Wittmann

Pörtner

H-O

(

2013

)

Sensitivities of extant animal taxa to ocean acidification

Nature Clim Change

995

–

1001

Google Scholar

Crossref

WorldCat

Wolfe

Dworjanyn

Byrne

(

2013

)

Effects of ocean warming and acidification on survival, growth and skeletal development in the early benthic juvenile sea urchin (Heliocidaris erythrogramma)

Glob Change Biol

2698

–

2707

Google Scholar

Crossref

WorldCat

Wood

Spicer

Lowe

Widdicombe

(

2010

)

Interaction of ocean acidification and temperature; the high cost of survival in the brittlestar Ophiura ophiura

Mar Biol

157

2001

–

2013

Google Scholar

Crossref

WorldCat

Wood

Eriksson

Nordborg

Styf

(

2015

)

The effect of environmental stressors on the early development of the Norway lobster Nephrops norvegicus (L.)

J Exp Mar Biol Ecol

473

–

Google Scholar

Crossref

WorldCat

Zervoudaki

Frangoulis

Giannoudi

Krasakopoulou

(

2014

)

Effects of low pH and raised temperature on egg production, hatching and metabolic rates of a Mediterranean copepod species (Acartia clausi) under oligotrophic conditions

Mediterranean Mar Sci

–

Google Scholar

Crossref

WorldCat

Zhang

Shin

PKS

Cheung

(

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

–

Zittier

ZMC

Bock

Lannig

Pörtner

H-O

(

2015

)

Impact of ocean acidification on thermal tolerance and acid–base regulation of Mytilus edulis (L.) from the North Sea

J Exp Mar Biol Ecol

473

–

Google Scholar

Crossref

WorldCat

Author notes

Editor: Steven Cooke

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.

Download all slides

Month:	Total Views:
February 2017	55
March 2017	72
April 2017	27
May 2017	14
June 2017	18
July 2017	25
August 2017	13
September 2017	22
October 2017	40
November 2017	54
December 2017	81
January 2018	135
February 2018	107
March 2018	139
April 2018	114
May 2018	97
June 2018	74
July 2018	81
August 2018	71
September 2018	95
October 2018	83
November 2018	122
December 2018	87
January 2019	80
February 2019	96
March 2019	100
April 2019	105
May 2019	89
June 2019	86
July 2019	86
August 2019	63
September 2019	103
October 2019	140
November 2019	108
December 2019	68
January 2020	82
February 2020	121
March 2020	150
April 2020	106
May 2020	114
June 2020	88
July 2020	82
August 2020	78
September 2020	137
October 2020	149
November 2020	158
December 2020	248
January 2021	101
February 2021	118
March 2021	140
April 2021	124
May 2021	113
June 2021	77
July 2021	67
August 2021	67
September 2021	115
October 2021	114
November 2021	115
December 2021	65
January 2022	77
February 2022	65
March 2022	103
April 2022	109
May 2022	147
June 2022	96
July 2022	52
August 2022	77
September 2022	193
October 2022	219
November 2022	140
December 2022	102
January 2023	108
February 2023	105
March 2023	138
April 2023	124
May 2023	143
June 2023	72
July 2023	65
August 2023	82
September 2023	143
October 2023	148
November 2023	112
December 2023	53
January 2024	83
February 2024	91
March 2024	152
April 2024	124
May 2024	136
June 2024	66
July 2024	82
August 2024	68
September 2024	95
October 2024	108
November 2024	98
December 2024	50
January 2025	72
February 2025	98
March 2025	99
April 2025	81
May 2025	17

Article Contents

Are global warming and ocean acidification conspiring against marine ectotherms? A meta-analysis of the respiratory effects of elevated temperature, high CO₂ and their interaction

Abstract

Introduction

General approach

Calculations

Comparison of Fry aerobic-scope curves (category A)

Isolated effects of elevated CO₂ on resting oxygen uptake and absolute aerobic scope (category B)

Combined effects of elevated temperature and CO₂ (category C)

Results

Aerobic-scope curves

Isolated and combined effects ’of temperature and CO₂

Examination of the variability ’in responses to CO₂

Nature of the interaction between ’temperature and CO₂

Discussion

Aerobic-scope curves across species and time frames

Does elevated CO₂ in general cause resting oxygen uptake to increase?

Does elevated CO₂ in general reduce ’aerobic scope?

Combined effects of temperature and CO₂ on resting aerobic metabolism

Combined effects of temperature and CO₂ on absolute aerobic scope

Beyond aerobic scope: behavioural ’effects of CO₂

Conclusions

Supplementary material

Funding

Acknowledgements

References

Author notes

Supplementary data

Citations

Views

Altmetric

Email alerts

Citing articles via

Latest

Most Read

Most Cited

Article Contents

Are global warming and ocean acidification conspiring against marine ectotherms? A meta-analysis of the respiratory effects of elevated temperature, high CO2 and their interaction

Abstract

Introduction

General approach

Calculations

Comparison of Fry aerobic-scope curves (category A)

Isolated effects of elevated CO2 on resting oxygen uptake and absolute aerobic scope (category B)

Combined effects of elevated temperature and CO2 (category C)

Results

Aerobic-scope curves

Isolated and combined effects ’of temperature and CO2

Examination of the variability ’in responses to CO2

Nature of the interaction between ’temperature and CO2

Discussion

Aerobic-scope curves across species and time frames

Does elevated CO2 in general cause resting oxygen uptake to increase?

Does elevated CO2 in general reduce ’aerobic scope?

Combined effects of temperature and CO2 on resting aerobic metabolism

Combined effects of temperature and CO2 on absolute aerobic scope

Beyond aerobic scope: behavioural ’effects of CO2

Conclusions

Supplementary material

Funding

Acknowledgements

References

Author notes

Supplementary data

Citations

Views

Altmetric

Email alerts

Citing articles via

Latest

Most Read

Most Cited

This Feature Is Available To Subscribers Only

Are global warming and ocean acidification conspiring against marine ectotherms? A meta-analysis of the respiratory effects of elevated temperature, high CO₂ and their interaction

Isolated effects of elevated CO₂ on resting oxygen uptake and absolute aerobic scope (category B)

Combined effects of elevated temperature and CO₂ (category C)

Isolated and combined effects ’of temperature and CO₂

Examination of the variability ’in responses to CO₂

Nature of the interaction between ’temperature and CO₂

Does elevated CO₂ in general cause resting oxygen uptake to increase?

Does elevated CO₂ in general reduce ’aerobic scope?

Combined effects of temperature and CO₂ on resting aerobic metabolism

Combined effects of temperature and CO₂ on absolute aerobic scope

Beyond aerobic scope: behavioural ’effects of CO₂