Clark and colleagues (Clark et al., 2013) have set people straight on how to make high quality measurements of maximum and minimum oxygen uptake rates in fishes. They correctly state that great care and understanding are needed to properly measure a fish's aerobic scope (=maximum metabolic rate–standard metabolic rate). Indeed, such challenges probably contributed to the slow acceptance of the concept of aerobic scope and Topt (the temperature for maximum aerobic scope) after their introduction in the 1940s by a Canadian, Fred Fry. However, while doing so, Clark et al. take some misguided sideswipes at the oxygen- and capacity-limited thermal tolerance (OCLTT) hypothesis.
Clark et al. have an obvious error in their fig. 1B (see Clark et al., 2013). They mislabel the peak of the Fry aerobic scope curve as ‘onset of loss of performance’ such that Topt (they use ToptAS) is not the temperature when aerobic scope is maximal. As a result, Topt is labelled at a lower temperature and reduced aerobic scope. Here, I have assumed the independent variable in their fig. 1 (‘scope for aerobic performance’) is aerobic scope as the text correctly states ‘…aerobic scope continues to increase until temperature approaches lethal levels…’. This mistake apparently created a somewhat slippery slope. They then say, ‘…recent data suggest that ToptAS provides little insight into the preferred temperature or performance of aquatic ectotherms…’, without appropriate literature citations. They even ignore a previous study (Clark et al., 2011) where the authors posit aerobic scope at high temperatures may be important for pink salmon in a global warming situation.
Instead, perhaps their fig. 1B (see Clark et al., 2013) was primarily intended to point out that a Fry aerobic scope curve isn't always bell-shaped (see their fig. 1A). But surely such a point is unnecessary given the various curves illustrated by Fry's original works, many of which were recently reproduced alongside newer data for Pacific salmon (Farrell, 2009). [I trust they were not disrespectfully targeting the single cartoon allowed in a Science Perspective (Pörtner and Farrell, 2008).] Equally curious is why Clark and colleagues do not also consider a Fry aerobic scope curve for a representative eurythermic fish. Killifish, for example, maintain aerobic scope over an acute temperature range of almost 30°C (Healey and Schulte, 2012). As a result, rather than having a specific Topt, killifish have a broad Topt window where absolute aerobic scope varies very little over most of the fish's thermal limit.
To be absolutely clear, for a given temperature, aerobic scope is the maximum amount of oxygen available for any aerobic activity above routine. A Fry aerobic scope curve defines the temperature dependence of aerobic scope, from which one can derive Topt and the temperature when aerobic scope disappears (Tcrit or Tc). The concept cannot be simpler. What may be less clear, and a point of ideological difference, is that aerobic scope is a fundamental capacity; an animal must be at Topt (or within a Topt window) to maximize this potential. It says nothing about when and how it is realized, which involves competing factors such as food availability and species competition. Clearly, Topt maximizes the potential to deliver oxygen to tissues, but growth cannot be maximized without high quality food, for example.
In the context of life history, Clark and colleagues should not suggest ‘…ToptAS has little relevance…’. Topt and Tcrit have been given ecological relevance for a life history bottleneck – a once-in-a lifetime spawning migration up a high-flowing river at peak summer temperatures for adult sockeye salmon (Farrell et al., 2008). It is true, juvenile and adult sockeye salmon typically inhabit lakes, rivers and oceans colder than Topt (Lee et al., 2003; Eliason et al., 2011; Chen et al., 2013). Therefore, while El Nino oceanic warming provides sockeye salmon with a potential for better growth, this potential is not necessarily realized for the population because predatory tuna invade northward as they are no longer biogeographically constrained by the cold northeast Pacific (e.g. Block et al., 1997). Conversely, lake-rearing juvenile sockeye salmon behaviourally exploit their Fry aerobic scope curve to manage oxygen allocation and maximize growth by using a diurnal vertical migration. At dawn and dusk, they feed in warm surface water, which maximizes foraging (swimming) activity, but they then relocate to deep, cooler water to maximize food conversion by lowering routine metabolic rate (Brett, 1971). Moreover, by suggesting that maximum metabolic rate for ambush predators should be measured post-exercise, Clark et al. clearly acknowledge that maximum oxygen delivery is required to recover as quickly as possible after escaping a predator. Thus, they imply that Topt has ecological relevance for recovery. The ecological and evolutionary relevance of the OCLTT is further discussed in the accompanying commentary by Hans-Otto Pörtner and Folco Giomi (Pörtner and Giomi, 2013).
An important benefit of the OCLTT hypothesis is that the Fry aerobic scope curve offers a framework for a mechanistic understanding of thermal responses. All fishes studied to date increase heart rate (fH) with warming, which is the primary driver to increase cardiac output and arterial oxygen transport (e.g. Cech et al., 1976; Steinhausen et al., 2008; Farrell, 2009; Mendonça and Gamperl, 2010; Eliason et al., 2013). However, warming eventually triggers an arrhythmic heartbeat, a cardiac collapse that occurs at a temperature lower than Tcrit and CTmax (the upper critical temperature when a fish can no longer maintain a righting reflex) (Casselman et al., 2012; Anttila et al., 2013). The Arrhenius breakpoint temperature (ABT) for fH was recognized nearly a century ago (Crozier, 1926) and is now informing biogeographic distributions of intertidal ectotherms (Somero, 2010). Thus, mechanisms underlying thermal responses can be related back to a Fry aerobic scope curve.
To close, Clark and colleagues provide guidance for measuring metabolic rate in fishes, a field where techniques vary widely (sometimes unavoidably). Thus, readers must be circumspect because these technicalities may mean that all values for aerobic scope are not necessarily exactly comparable among studies. Unfortunately, this important guidance is marred by some poorly supported statements, which mislead rather than advance the field. Hopefully, they will better explain (and defend) at some time their new proposition: multiple performance–multiple optima [see their fig. 7B (Clark et al., 2013)]. I suspect that if a common dimension can be found for the proposed ‘physiological performances’, summing all these curves may produce a single curve like that in their fig. 1A. The OCLTT remains a valuable conceptual framework around which the various competing factors for oxygen allocation can be examined, with the Fry aerobic scope curve defining Topt. Nowhere do Pörtner and Farrell (Pörtner and Farrell, 2008) suggest that a fish must operate at Topt. Indeed, suboptimal temperatures may be exploited to obtain food and avoid predation. Such are the choices in life.
