The Effect of Hypoxia on Fish Swimming Performance and Behaviour

Chapter 6

The Effect of Hypoxia on Fish Swimming Performance and Behaviour P. Domenici, N. A. Herbert, C. Lefrançois, J. F. Steffensen and D. J. McKenzie

Abstract Oxygen depletion, hypoxia, can be a common stressor in aquatic habitats, including aquaculture. Hypoxia limits aerobic swimming performance in fish, by limiting their aerobic metabolic scope. Hypoxia also elicits changes in spontaneous swimming activity, typically causing a decrease in swimming speed in sedentary species and an increase in active species. However, fish do have the capacity to avoid hypoxia and actively choose well-oxygenated areas. Hypoxia causes differences in fish behaviour in schools, it may reduce school density and size and influence activities such as shuffling within schools. Hypoxia also influences predator–prey interactions, in particular by reducing fast-start performance. Thus, through effects on swimming, hypoxia can have profound effects on species distributions in the field. In aquaculture, effects of hypoxia may be particularly significant in sea cages. It is therefore important to understand the nature and thresholds of effects of hypoxia on swimming activity to extrapolate to potential impacts on fish in aquaculture. P. Domenici (&) CNR-IAMC Loc. Sa Mardini, Torregrande, Oristano, Italy e-mail: [email protected] N. A. Herbert Leigh Marine Laboratory, University of Auckland, PO Box 349, Warkworth 0941, New Zealand C. Lefrançois UMR 6250 LIENSS (CNRS-University of La Rochelle), 2 rue Olympe de Gouges, 17000 La Rochelle, France J. F. Steffensen Marine Biological Laboratory, Biological Institute, University of Copenhagen, Strandpromenaden 5, 3000 Helsingør, Denmark D. J. McKenzie UMR 5119 Ecologie des Systèmes Marins Côtiers, Université Montpellier II, Place Eugène Bataillon cc 093, Montpellier cedex 5, 34095 Montpellier, France

A. P. Palstra and J. V. Planas (eds.), Swimming Physiology of Fish, DOI: 10.1007/978-3-642-31049-2_6, Ó Springer-Verlag Berlin Heidelberg 2013

129

130

P. Domenici et al.

6.1 Introduction Water has a low capacitance for oxygen, air-saturated water only contains a few milligrams per litre of oxygen at normal atmospheric pressures. As a result, oxygen levels can be depleted quite easily by respiring organisms, especially in nutrientrich environments with a large microbial biomass, or stagnant areas with poor vertical mixing. Oxygen depletion, hypoxia, is therefore a natural phenomenon. As such, fish have evolved to cope with hypoxia, although their relative tolerance depends on the species, its habitat and lifestyle. Hypoxia may occur over a variety of timescales, defined by Kemp et al. (2009) as (1) permanent, (2) persistent seasonal, both stratified and vertically mixed, (3) episodic and (4) diel. Hypoxia is also a common symptom of degraded water quality caused by nutrient pollution and eutrophication. Over the last few decades, eutrophication of coastal waters has been linked to increases in the frequency, duration and geographical extent of hypoxic events, which are recognised as important environmental problems globally (Diaz and Rosenberg 2008). Hypoxic events have the potential to significantly impact coastal fisheries (Diaz 2001). Shifts in spatial distribution and the structure of benthic and nekton assemblages can occur through direct mortality during extreme local events, especially of sluggish species, and through sublethal effects such as increased emigration of vagile species. Additionally, fish exposed to hypoxic conditions grow slower and produce fewer viable offspring (Petersen 1987; Plante et al. 1998; Dean and Richardson 1999; Smith and Able 2003; Taylor and Miller 2001). Changes in assemblage structure and loss of habitat can have bottom-up effects on food web structure such as losses in key prey species resulting in further ecological effects (Diaz 2001). Hypoxia may have profound effects on production efficiency of fish in culture, through its depressive effects on growth. Species differ greatly in their relative tolerance of hypoxia, as a function of the environment in which they have evolved. Thus, cyprinids that have evolved in slow-moving or static and warm waters, where hypoxia can develop, are known to be more tolerant of oxygen depletion than salmonids, which rarely encounter any hypoxia in their cool fast-flowing habitats. Oxygen levels are generally carefully monitored and controlled in modern commercial finfish aquaculture, except for some air-breathing species in south-east Asia (Lefevre et al. 2011). Hypoxia can, however, develop in sea cage aquaculture through oceanographic and eutrophic forces. Little is known about how it influences the behaviour of cultured fish (Oppedal et al. 2011). The current review therefore provides an opportunity to extrapolate what we know about how hypoxia affects fish swimming performance and behaviour to culture scenarios. The impact of hypoxia on aquatic ecosystems is modulated by the physiology and behaviour of the organisms (Kramer et al. 1997; Domenici et al. 2007a, b; Chapman and McKenzie 2009). Knowledge of the processes that regulate the interactions between hypoxia and ecologically relevant variables, such as growth and survival, is fundamental for understanding and predicting the effects of such interactions, particularly on fish in aquaculture. A number of physiological and behavioural effects are

6 The Effect of Hypoxia

131

observed at sublethal levels of hypoxia, which presumably mediate subsequent effects on fish activity and distribution (Domenici et al. 2007a, b; Chapman and McKenzie 2009). Fish can sense oxygen levels in the ventilatory water stream and in their blood and, when oxygen levels in these milieux decline, they engage a suite of physiological responses which, together, aim to improve oxygen uptake at the gills, transport in the blood and release at the tissues (Randall 1982; Burleson et al. 1992; Richards 2009). Nonetheless, the reduced oxygen availability in hypoxia limits the ability of fish to provide oxygen for metabolic activities, their aerobic metabolic scope (MS; Fry 1947, 1971; Claireaux et al. 2007). The effects of hypoxia on aerobic metabolism, and MS, can be modelled as shown in Fig. 6.1. The standard metabolic rate (SMR) is the minimal rate of oxygen uptake required to support essential maintenance functions in ectotherms. In hypoxia, fish can typically use physiological responses to maintain oxygen uptake at or above SMR down until a critical dissolved oxygen (DO) threshold, termed critical O2 partial pressure (Pcrit) or saturation (Scrit). Below this critical threshold, the fish is no longer able to support maintenance metabolism and its metabolic rate is dependent on the oxygen level in its external O2 environment (Fig. 6.1, Schurmann and Steffensen 1997). Whilst SMR provides for essential core function, all other activities such as growth, reproduction and swimming activities (the focus of this review), depend on an ability to increase oxygen uptake and delivery above and beyond SMR, and within the limits of aerobic metabolic scope (MS). MS was first defined by Fry (1947) as the difference between the maximum metabolic rate (MMR) and SMR (Chabot and Claireaux 2008; Claireaux et al. 2000). Maximum metabolic rate measured in fish swimming at the maximum sustained speed is also called active metabolic rate (AMR) by some authors (e.g. Schurmann and Steffensen 1997; Claireaux et al. 2000). Fish MMR is increasingly limited as hypoxia becomes progressively more severe (Fig. 6.1, Claireaux et al. 2000; Cook et al. 2011). Fish must balance multiple metabolic demands, and swimming is an energetically demanding process (Schurmann and Steffensen 1997). Thus, hypoxic limitations to MS may limit a fish’s ability to perform activities such as swimming. Such limitations to MS will also force the fish to make decisions about how to use the available oxygen. The manner by which fish use the available oxygen will become increasingly important as they approach Pcrit, because MS will eventually be exhausted (SMR = MMR at Pcrit, Fig. 6.1). Behavioural responses are, therefore, likely to be driven by the physiological effects of hypoxia (Lefrançois and Claireaux 2003; Fritsche and Nilsson 1989). It is essential that the fish exhibit a suitable behavioural response to an O2 restricted environment because this could have a profound effect on their survival. Hypoxia may therefore elicit changes in spontaneous activity, schooling and predator–prey interactions. These may differ depending upon the species and the severity of hypoxia. This chapter reviews the effects that hypoxia can have on fish swimming performance and swimming behaviours. We discuss aerobic swimming performance in which work under controlled (i.e. forced) speed tested the effect of hypoxia on sustained (aerobic) swimming using swim tunnel observations. The effect of hypoxia on

132

P. Domenici et al.

Fig. 6.1 Conceptual overview of how hypoxia (low partial pressure of oxygen, PO2) affects the various rates of mass-specific O2 consumption (MO2) at set temperatures. SMR = standard metabolic rate; MMR = maximum metabolic rate; MS = metabolic scope; Pcrit = critical oxygen pressure where SMR = MMR. Concepts taken from Fry (1947), Claireaux et al. (2000), Schurmann and Steffensen (1997) and Cook et al. (2011)

spontaneous activity, i.e. unforced aerobic activity, is discussed next. Effects on schooling are included because hypoxia may have effects on this behaviour, given that schooling and swimming performance are tightly linked from an energetic point of view (Weihs 1973; Herskin and Steffensen 1998; Johansen et al. 2010). Schooling behaviour implies swimming, but it is commonly assessed by measures of interindividual relations (such as interindividual distances) rather than by swimming performance per se. Schooling (or at least organised circular swimming. Føre et al. 2009) also occurs in aquaculture and hypoxia has the potential to influence production efficiency by modulating this behaviour in sea cage systems (Oppedal et al. 2011). Whilst effects on aerobic swimming performance can easily be explained by the direct limitation imposed by hypoxia on MS, other effects such as changes in spontaneous activity and in anaerobic swimming during predator–prey interactions (fast start escape responses) may instead reflect behavioural strategies adopted by a given species. For example, certain fish species may reduce spontaneous swimming activity to reduce oxygen requirements (Metcalfe and Butler 1984; Fisher et al. 1992; Schurmann and Steffensen 1994). On the other hand, certain species, particularly highly active marine species, increase their swimming activity when exposed to hypoxia (Dizon 1977; Bejda et al. 1987; Domenici et al. 2000a). This apparent paradox has been interpreted as a behavioural strategy to increase the chances of escaping the hypoxic zone in a heterogeneous environment. How fish react to inescapable and escapable low O2, will therefore influence the survival of fish in the wild and under culture.

6.2 The Effects of Hypoxia on Aerobic Swimming Performance It is well established that aquatic hypoxia limits the aerobic swimming performance of fish. Various studies using swim tunnels and an incremental ‘‘critical swimming speed’’ (Ucrit) protocol have demonstrated that fish have a lower Ucrit in hypoxia as


133

compared to normoxia (e.g. Dahlberg et al. 1968; Jones 1971; Bushnell et al. 1984; Jourdan-Pineau et al. 2010; Petersen and Gamperl 2010, Zhang et al. 2010). Table 6.1 shows that there is considerable interspecific variation in sensitivity of Ucrit to hypoxia. Whenever it has been investigated, however, reduced swimming performance has been shown to be a direct consequence of limitations to MMR and MS (Jones 1971; Bushnell et al. 1984; Jourdan-Pineau et al. 2010; Petersen and Gamperl 2010; Zhang et al. 2010; Fu et al. 2011). Presumably, fish are all limited in their ability to provide oxygen to the slow-twitch oxidative ‘‘red’’ muscles which power steady aerobic swimming. In addition, hypoxia reduced the stamina of golden grey mullet Liza aurata swimming at their optimal swimming speed (the speed with the lowest cost of transport) (Vagner et al. 2008). This effect was probably due to the use of anaerobic metabolism to supplement swimming at such low speeds, which poses a limit to the amount of time a fish can engage in activities such as habitat exploration and food searching under hypoxic conditions.

6.2.1 Effects of Prior Acclimation A number of studies have investigated whether prior acclimation to low levels of oxygen can improve the ability to perform Ucrit exercise in hypoxia (Kutty 1968; Bushnell et al. 1984; Petersen and Gamperl 2010; Fu et al. 2011). Fu et al. (2011) found that prior exposure of goldfish Carassius auratus to severe hypoxia (3 % saturation for 48 h at 12 °C) improved their exercise performance by 18 % when they were then tested in hypoxia at 10 % of the prevailing air saturation (Table 6.1), and raised their MMR by 35 %. This was linked to the ability of cyprinids to modify their gill structure and blood oxygen carrying capacity in response to severe hypoxia (e.g. Sollid and Nilsson 2006). The acclimated goldfish had a 71 % increase in lamellar surface area and 25 % increase in blood haemoglobin concentration compared to normoxic controls (Fu et al. 2011). Indeed, the acclimated goldfish actually exhibited improved Ucrit performance in normoxia when compared to control animals (Table 6.1, Fu et al. 2011). Other studies, on various species with various protocols (Table 6.1), have failed to show any effect of hypoxia acclimation on subsequent ability to perform aerobic exercise in hypoxia (Kutty 1968; Petersen and Gamperl 2010). Furthermore, in Atlantic cod Gadus morhua, chronic exposure to moderate hypoxia (40–45 % saturation for at least six weeks at 10 °C) appeared to exert some negative effects, causing a decline in cardiac stroke volume under resting conditions and during exercise (Petersen and Gamperl 2010). These species differences may be linked to the acclimation protocol, in particular the severity of hypoxia employed and whether the animals were returned to normoxia prior to measuring their exercise performance (Fu et al. 2011). It may also reflect an inability of some species such as the rainbow trout Oncorhynchus mykiss or the Atlantic cod to produce the same plastic phenotypic modifications of gill structure and blood carrying capacity as seen in

12 23 15 15 20 10 10 25 25 12 12 30 30

6.0 ± 6.3 ± *1.8 *1.8 *4.2 1.5 ± 1.5 ± 2.8 ± 2.8 ± 5.3 ± 5.8 ± 2.0 ± 2.1 ±

*66 *79 *55 *55 114 ± 3 (6) *61 *63 *33 *33 *29 *31 *46 *48 0.1 0.1 0.1 0.1 0.1 0.2 0.2 0.1

(10) (10) (6) (6) (8) (8) (7) (7)

0.2 (15) 0.2 (14)

Length s-1

cm s-1 *44 *52 41 ± 6 (8) 41 ± 2 (8) 105 ± 2 (6) *41 *43 *24 *17 *18 *21 *28 *48

cm s-1 3.5 ± 4.4 ± *1.4 *1.4 *3.9 1.0 ± 1.0 ± 2.0 ± 1.5 ± 3.3 ± 3.9 ± 1.2 ± 2.0 ± 0.3 0.1 0.1 0.1 0.1 0.2 0.1 0.1

(10) (10) (6) (8) (8) (8) (7) (7)

0.1 (18) 0.1 (15)

Length s-1

a

Length, body lengths, numbers in brackets are sample size acclimated for 2 weeks to an hypoxic PO2 of 40 mmHg (*25 % air saturation) b acclimated for at least 6 weeks to an hypoxic PO2 of *8–9 kPa (40–45 % air saturation) c exposure to an hypoxic O2 level at of 50 % air saturation had no significant effect on Ucrit d acclimated for 48 h to an hypoxic O2 concentration of 0.3 mg l-1 mmHg (*3 % air saturation) e when individual fish permitted access to surface to breathe air

O. mykiss O. mykiss O. mykiss O. mykissa D. labrax G. morhua G. morhuab S. meridionalisc S. meridionalis C. auratus C. auratusd G. carapo G. carapoe

(°C) *50 *50 *25 *25 50 40-45 40-45 24 12 9 9 19 19

(%) 42 34 25 25 8 32 32 27 47 38 33 40 *0

(%) Jones (1971) Jones (1971) Bushnell et al. (1984) Bushnell et al. (1984) Jourdan-Pineau et al. (2010) Petersen and Gamperl (2010) Petersen and Gamperl (2010) Zhang et al. (2010) Zhang et al. (2010) Fu et al. (2011) Fu et al. (2011) McKenzie et al. (2012) McKenzie et al. (2012)

Table 1 Effects of various different levels of hypoxia (reported as % of air saturation) on critical swimming speed (Ucrit) in various finfish species at various temperatures Hypoxic Ucrit Hypoxia Decline Reference Species T Normoxic Ucrit

134 P. Domenici et al.


135

cyprinids (Kutty 1968; Bushnell et al. 1984; Petersen and Gamperl 2010; Fu et al. 2011).

6.2.2 Air-Breathing Fish A number of actinopterygian fish species, especially in tropical habitats, have evolved the ability to gulp air at the water surface and store this in a variety of vascularised air-breathing organs, to then extract the O2 for metabolism (Graham 1997). Such species are bimodal breathers, they retain gills with some function in gas-exchange. In freshwater fish, air breathing is believed to have evolved as a response to aquatic hypoxia (Randall et al. 1981; Graham 1997) and hypoxia stimulates air breathing in all species studied to date (Graham 1997; Chapman and McKenzie 2009). Increased activity and exercise also, however, stimulate air breathing (Graham 2006) in all species that have been studied (Grigg 1965; Farmer and Jackson 1998; Seymour et al. 2004, 2007; McKenzie et al. 2012). When submitted to controlled increases in swimming speed, bimodal species show pronounced increases in rates of oxygen uptake from air (Farmer and Jackson 1998; Seymour et al. 2004, 2007; McKenzie et al. 2012). Thus, when air-breathing fish swim, they make continual visits to the surface to gulp air. A stepwise Ucrit protocol caused an exponential increase in air-breathing frequency in the two species that have been studied to date, the bowfin Amia calva (Farmer and Jackson 1998) and the banded knifefish Gymnotus carapo (McKenzie et al. 2012). In the knifefish, air breathing contributed about 35 % of MS during a Ucrit protocol in aquatic normoxia (Fig. 6.2). If denied access to the surface, however, the knifefish were able to achieve the same Ucrit and MS by gill ventilation alone (Fig. 6.2), indicating that air breathing was not necessary to sustain aerobic performance in aquatic normoxia (McKenzie et al. 2012). In aquatic hypoxia, though, denial of access to the surface caused a profound decline in Ucrit performance and MS (Table 6.1; Fig. 6.2), In aquatic hypoxia with access to the surface to breath air, however, the knifefish were able to increase the proportion of their MS that was met by air breathing, and avoid any hypoxic limitation of Ucrit performance (Table 6.1; Fig. 6.2). They were able to do this by a very marked increase in the frequency of air breathing, whereby they took bouts of breaths in quick succession, whilst swimming just below the surface (McKenzie et al. 2012). Thus, at least in G. carapo, air breathing allows the animal to avoid hypoxic limitations to aerobic exercise performance and MS. Therefore, air breathing presumably allows this, and probably other bimodal species, to colonise hypoxic aquatic habitats without any limitations to their ability to perform activities such as exercise and digestion. This might seem like a potential advantage for any fish species, given that even in normoxia water has a low capacitance for oxygen. The fact, however, that only about 450 species of bony fish breathe air (Graham 1997), a very small percentage of the *25,000 species of actinopterygian, indicates that there must be ecological costs. These are probably, at least in part, the increased risk of predation, in

136

P. Domenici et al.

50

(a)

Ucrit (cm s-1)

40 30 20 10

max. O2 uptake rate (mg kg-1 h-1)

0 400

(b) 300

200

100

0 norm + air normoxia hypo + air hypoxia

Fig. 6.2 Mean (± S.E.) critical swimming speed (Ucrit, Panel A) and maximum rates of oxygen uptake (Panel B) from water (dark grey) and air (light grey), in banded knifefish Gymnotus carapo exercised in either aquatic normoxia, or aquatic hypoxia (25 % air saturation), with or without access to the surface to breathe air. N = 7 in all cases. The combination of aquatic and aerial oxygen uptake is their maximum oxygen uptake. Critical swimming speed and maximum oxygen uptake are limited in aquatic hypoxia without access to air; when access is allowed to the surface in aquatic hypoxia, the knifefish compensated for limitations to aquatic uptake by breathing air. Data from McKenzie et al. (2012)

particular from above the surface (Smith and Kramer 1986). The key challenge in the farming of air-breathing species under low O2 conditions might, therefore, be to reduce the risk of aerial predation (e.g. birds).

6.2.3 Effects of Hypoxia on Metabolic Prioritisation There has been some interest in understanding how water-breathing fish prioritise between an activity like swimming, and other metabolic activities when aquatic hypoxia limits MS. The effects of hypoxia on prioritisation between swimming and specific dynamic action (SDA) have been investigated in two species, the European sea bass Dicentrarchus labrax and the southern catfish Silurus meridionalis (Jourdan-Pineau et al. 2010; Zhang et al. 2010). The SDA response is the transient increase in metabolic rate and oxygen demand that follows consumption of a meal,


137

and is the cost of processing the food (Jobling 1994). SDA and swimming are considered to be the two major metabolic activities in the lives of most fish and are both very relevant to cultured fish as they exert a strong influence over growth and productivity (Dupont-Prinet et al. 2010; Brown et al. 2011). When D. labrax and S. meridionalis are submitted to a Ucrit protocol when digesting a meal, they both achieve a higher MMR than when exercised in a fasted state, indicating that they can meet the costs of both activities simultaneously. Jourdan-Pineau et al. (2010) found that, when exposed to hypoxia at 50 % of air saturation, D. labrax suffered a significant decline in Ucrit and MMR relative to their normoxic performance. Furthermore, there was no longer any difference in MMR between the fasted versus fed animals in hypoxia, indicating that hypoxia had limited their ability to meet the costs of both swimming and the SDA. In fact, they appeared to prioritise swimming performance because hypoxic Ucrit did not differ between fasted and fed seabass, indicating that they were allocating all of their limited MS to sustain aerobic exercise (Jourdan-Pineau et al. 2010). Zhang et al. (2010) found that hypoxia equivalent to 24 and 12 % of air saturation both severely limited Ucrit and MMR in S. meridionalis. Similar to the seabass, there was no difference in MMR in fasted versus fed catfish in hypoxia. Fed catfish, however, had significantly lower Ucrit than fasted fish in hypoxia, indicating that some of the limited MS was not being allocated towards exercise but, presumably, towards continued digestion (Zhang et al. 2010). These differences in metabolic prioritisation between species may be of particular significance to fish under culture, if hypoxia is a common occurrence. For example, species that prioritise digestion over swimming activity may exhibit less hypoxic depression of growth rates.

6.2.4 Effects of Hypoxia on Recovery from Exercise Although this section is dedicated to aerobic swimming, the Ucrit test that is typically used as a measure of aerobic performance actually has a significant element of anaerobic swimming, with recruitment of fast-twitch glycolytic ‘‘white’’ muscle at the highest speeds in the incremental protocol (Burgetz et al. 1998; discussed in Webb 1998; McKenzie and Claireaux 2010). As a result, fish often show elevated post-exercise oxygen consumption (EPOC, often described as an ‘‘oxygen debt’’) after Ucrit swimming (Beamish 1978; Farrell et al. 1998), which reflects the costs of recovering homoeostasis and replenishing substrates in the white muscle (Wood 1991; Richards et al. 2002). The effects of hypoxia on the ability to respond to EPOC have not been investigated. Farrell et al. (1998) showed, however, that sockeye salmon Oncorhynchus nerka could repeat a Ucrit protocol to the same level of performance with only a 45 min recovery period in normoxia, but could not do so in a degree of moderate hypoxia (*66 % saturation) that did not significantly impair their initial Ucrit performance. Svendsen et al. (2012) have also shown that recovery from a hypoxic anaerobiosis, which also engendered an ‘‘oxygen debt’’, is prolonged by hypoxia (30 % saturation) in

138

P. Domenici et al.

rainbow trout, so the same would presumably be true for recovery from EPOC. This may be significant in aquaculture if fish are subjected to anaerobic exercise when, for example, they attempt to evade capture. It would then be very important to ensure adequate oxygen supply for recovery. Interestingly, moderate sustained aerobic exercise appears to improve recovery from such anaerobiosis in rainbow trout (Milligan et al. 2000).

6.3 The Effects of Hypoxia on Spontaneous Activity Hypoxia can elicit a number of effects on spontaneous activity in fish. They may increase, decrease or show no change in their swimming activity, and these reactions have each been linked to lifestyles (see review by Chapman and McKenzie 2009; Fig. 6.3). For example, sluggish species are commonly thought to down-regulate their speed whilst active ecotypes more often up-regulate their activity during hypoxia. However, because some species show a tendency to both increase and decrease their swimming speed, depending on the severity of hypoxia (Brady et al. 2009; Herbert and Steffensen 2005), broad generalizations should be made with caution. A more informative approach might therefore be to consider the swimming response of fish to hypoxia as a species-specific (or even individual-based) trade-off between (1) aerobic capacity, (2) the hypoxic environment (severity of hypoxia, duration of exposure and likelihood of escape), (3) the routine energy-demands of the fish (i.e. high or low)and (4) the ecological benefit that is conferred by a change in swimming speed (e.g. survival, avoidance etc.). Indeed, there is evidence that the effects of severe hypoxia (*20 % DO) on the spontaneous activity of individual seabass (Dicentrarchus labrax) depends on their routine metabolic rate, hence level of energy demand; individuals with high metabolic rate showed a greater change in activity (Killen et al. 2012a). However, in that hypoxia study of D. labrax the increased activity of high metabolic rate fish was linked to a behaviour known as aquatic surface respiration (Killen et al. 2012a), where fish swim up to ventilate the surface layer of the water which is in contact with air, and therefore has a higher oxygen saturation (see Chapman and McKenzie 2009 for a review). Swimming speed responses may also differ with perceived risk thresholds that exist when fish are either solitary or held in shoaling groups (Lefrançois et al. 2009). Therefore, to gain a true understanding of why fish adopt a particular change in swimming speed during hypoxia we should probably consider the trade-offs and contexts that apply to different species and individuals under relevant low O2 conditions. The review of Chapman and McKenzie (2009) provides a thorough account of which species increase or decrease their swimming speed when exposed to hypoxia so the current discussion focuses more on the expected trade-offs facing a few species for which there is a good range of information regarding hypoxic behaviour, ecology and physiology. An attempt is also made to emphasise the few new research findings that have appeared since the review of Chapman and McKenzie (2009), particularly with respect to the avoidance behaviour of water-breathing fish towards the end of this


139

Fig. 6.3 The effects of hypoxia on swimming speed in fish. Quantitative data are available for nine species, this figure presents mean percentage changes in speed relative to the normoxic control, note also the inverted abscissa. For each species, the number in brackets is the mean normoxic swimming speed in length s-1. Figure taken from Chapman and McKenzie (2009) with permission from Elsevier

section. This understanding will hopefully help to predict the response of fish in culture, under escapable or inescapable conditions.

6.3.1 Down-Regulation of Activity The Atlantic cod Gadus morhua and weakfish Cynoscion regalis commonly inhabit large areas of hypoxia, possibly to forage on zoobenthos (Neuenfeldt et al. 2009; Chabot and Claireaux 2008; Stierhoff et al. 2009) and show a similar behavioural response to hypoxia, characterised by an initial increase in swimming speed at moderate O2 levels (65 % DO) followed by a pronounced 21–41 % drop in swimming speed during more severe hypoxia (20–40 % DO) (Fig. 6.4a; Herbert and Steffensen 2005; Brady et al. 2009). G. morhua do not have exceptional metabolic scope at any level of hypoxia (Claireaux et al. 2000) nor do they have an exceptionally high tolerance of low O2 (Schurmann and Steffensen 1997) but their two-faceted behavioural response to hypoxia does appear to hold adaptive value

140

P. Domenici et al.

for this species (Herbert and Steffensen 2005). It is commonly believed that decreased swimming speed conserves energy and allows fish to operate within the limits of their available scope, thus offsetting stress to the last possible moment (Chapman and McKenzie 2009; Claireaux and Chabot 2005; Nilsson et al. 1993). Also, since spontaneous activity is far more costly than straight-line swimming (Boisclair and Tang 1993), the absolute 0.18 length s-1 drop in spontaneous swimming speed would undoubtedly reduce metabolic expenditure in cod during hypoxia. This situation certainly appears to fit Atlantic cod that do not show signs of anaerobic stress until encountering O2 levels close to their Pcrit (Fig. 6.4a–b), a point at which anaerobic metabolism would be unavoidable anyway due to zero aerobic scope (Fig. 6.1; Herbert and Steffensen 2005). Heightened swimming speed on the other hand increases energy demand which could result in stress, or even death, if safe O2 areas are not found quickly. Assuming that increased activity improves the chances of escape (but see discussion below), the initial increase in swimming speed seen by cod (Fig. 6.4a) might represent a safe avoidance reaction because it is performed at sufficiently high O2 levels where MS is not overly constrained (Claireaux et al. 2000) and major stress is not observed (Fig. 6.4b, Herbert and Steffensen 2005). Therefore, by first attempting to escape a sudden decline in O2 but then quickly resigning to inescapable conditions, cod appear to avoid anaerobic thresholds with strategic shifts in swimming behaviour. The final ‘‘sit-and-wait’’ strategy adopted by cod, and other species such as carp Cyprinus carpio and weakfish (Brady et al. 2009; Herbert and Steffensen 2005; Nilsson et al. 1993; Schurmann and Steffensen 1994), might enhance survival under inescapable low O2 conditions but it does limit any ability to seek higher O2 levels and probably also places a limit on other non-essential processes such as feeding, growth and reproduction. On that basis, activity down-regulation as a low O2 strategy response is potentially more applicable to species that inhabit extensive areas of hypoxia and whose life histories can accommodate a degree of flexibility in growth and reproduction. In terms of hypoxia in seacage aquaculture, production efficiency might not be optimised but species that down-regulate (vs. increase) their swimming speed will probably have a greater chance of surviving low O2 episodes at sites prone to extensive and/or long-term hypoxia.

6.3.2 Increased Activity Active pelagic species, such as herring Clupea harengus and tuna (e.g. Katsuwonus pelamis), appear to increase their swimming speed during hypoxia by at least 10 % (Figs. 6.3 and 6.4), which is commonly viewed as an avoidance reaction (Dizon 1977; Domenici et al. 2000a, b; Fitzgibbon et al. 2010; Herbert and Steffensen 2006). On this basis, the response might have adaptive potential but it does also carry high risks if areas with suitable O2 are not found quickly. The marked swimming speed response does not, however, indicate low O2 tolerance or a tendency to live in nonhypoxic areas. For example, tuna are highly active pelagic carnivores, and skipjack


141

Fig. 6.4 The behavioural and physiological response of Atlantic cod and Atlantic herring (Gadus morhua and Clupea harengus) to progressive stepwise hypoxia at 10 °C. Figure a and c show the differential swimming speed of the two species in response to a progressive stepwise decline in PO2 (kPa). Differential speed is calculated as the difference between expected and observed swimming speeds, and therefore allows for diurnal shifts in background activity over time. A positive differential speed indicates faster swimming whilst a negative differential indicates slower speeds. Filled symbols indicate that water PO2 is in a steady state (with corresponding kPa values given above the horizontal bar). Open symbols denote that water PO2 is unsteady and declining to the next level. Figure c and d show the levels of lactate detected in the blood of the two species at various steady state PO2 levels during the progressive hypoxia experiment (closed symbols). Open symbols with a dashed line indicate the maximum level observed in the blood following a bout of strenuous chasing. Values with an asterisk are significantly different from control (high PO2 level) responses (* P \ 0.05; ** P \ 0.01). All data are from Herbert and Steffensen (2005, 2006). Reproduced with permission from Marine Biology, Springer

tuna Katsuwonus pelamis show dramatic increases in swimming speed during hypoxia (Dizon 1977) but other studies suggest that some species (e.g. the bigeye tuna Thunnus obesus and the southern Bluefin Thunnus maccoyii) are surprisingly hypoxia tolerant (Pcrit = 1.6–2.5 mg O2 l-1) and routinely exploit low O2 layers (*1 ml O2 l-1) (Lowe et al. 2000); (Fitzgibbon et al. 2010). Being obligate ram ventilators, tuna are perhaps less flexible in their response to low O2 with an increase in swimming speed representing a means of ensuring adequate ventilation during hypoxia (Fitzgibbon et al. 2010). Whilst increased swimming speed might feasibly enable tuna to escape low O2 regions more quickly, the rigidity and high-risk nature of this response does possibly present a major cause-for-concern if oxygen minimum

142

P. Domenici et al.

zones expand further (Keeling et al. 2010; Stramma et al. 2012) or tuna encounter hypoxia within the confines of sea cage aquaculture. The Atlantic herring Clupea harengus is also an active schooling species that shows a consistent increase in swimming speed when O2 is low (\40 % DO) and declining (Fig. 6.4c; Domenici et al. 2000a, b; Herbert and Steffensen 2006). Whilst not thought to be a common situation for active schooling fish, herring are surprisingly hypoxia tolerant by virtue of high haemoglobin-oxygen binding affinities (Herbert et al. 2006) and their ability to preserve aerobic function under extremely low O2 levels (i.e. down to 6.4 kPa or 30 % DO) (Fig. 6.4d; Herbert and Steffensen 2006). They also choose to reside in hypoxia on a frequent basis, either through en masse schooling (Dommasnes et al. 1994) or the selection of low O2 waters that exclude less tolerant predators (Domenici et al. 2002; Herbert and Steffensen 2006). The ability of an active pelagic species to partner physiological low O2 tolerance with elevated speed during hypoxia might therefore assist low O2 survival at a time of high routine expenditure. Whether the collective increase in swimming speed during schooling constitutes a coordinated avoidance response has not yet been demonstrated however. Why herring increase their speed in response to declining low O2 levels but not at low steady levels is also unclear (Fig. 6.4c; Herbert and Steffensen 2006). Domenici et al. (2002) show that hypoxia does not affect the reshuffling rate of leaders so it could be argued that increased swimming in trailing positions (where hypoxia is most intense in large schools. Moss and McFarland (1970)) do not allow rearward fish to escape. More work is certainly required to ascertain the functional basis (if any) of swimming speed changes under inescapable conditions (see below). These examples are used to show that the resultant swimming response of fish to hypoxia is not necessarily dictated by ecotype and/or routine swimming performance. Rather they reflect a species-specific trade-off between intrinsic physiological characteristics, the nature of the hypoxic encounter and the need to maintain routine activities (e.g. schooling) under environmentally challenging conditions.

6.3.3 Behavioural Avoidance of Hypoxia All hypoxia studies to date provide interesting insights into the swimming speed reaction of different species, but they have almost exclusively employed progressive inescapable hypoxia that never allow fish the opportunity to escape deleterious O2 conditions. This is not entirely an ecologically relevant condition as fish can often locate O2 refuges within extensive low O2 areas (Herbert et al. 2011). It is also not entirely relevant to high volume sea cage culture where low O2 may be vertically graded (Oppedal et al. 2011). Studies using inescapable hypoxia are not therefore able to resolve whether the hypoxia-induced increase in speed seen by a number of species does indeed associate with hypoxia avoidance or whether it is simply a reaction to an inescapable situation (Dizon 1977; Domenici et al. 2000a, b; Herbert and Steffensen 2005, 2006). Hypoxia avoidance studies employing avoidance test chambers and behavioural tracking techniques provide a valuable contribution to the


143

Fig. 6.5 The swimming speed response of Atlantic cod (G. morhua) presented with a range of water PO2 choices in a laminar flow choice chamber at 11.4 °C. The choice chamber received two streams of water which were varied in terms of PO2. a Water PO2 presentations over time. Within this experiment, both sides of the choice chamber were held at a high level of PO2, after which they deoxygenated to a critically low level of *4.3 kPa. The PO2 of one side was then raised progressively and cod were thus presented with a choice of different O2 pressures which they could choose between. b The swimming speed of cod (closed square symbols indicating mean speed ± 95 % CI) subjected to the different O2 levels in the choice chamber. Arrows indicate the level at which cod avoided the lowest PO2 level. Swimming speed values with an asterisk are significantly different from the initial control (high PO2 level) response. Data from Herbert et al. (2011). Reproduced with permission from Marine Biology, Springer

field. Whilst increased swimming speed has been widely upheld as an avoidance reaction, Herbert et al. (2011) showed that the avoidance behaviour of cod was not associated with increased swimming speed (Fig. 6.5). Perhaps surprisingly, when cod avoid low O2, they actually slow down with a reduction in speed. Poulsen et al. (2011) showed that increased swimming speed was not functionally linked with hypoxia avoidance in rainbow trout at high or moderate levels of low O2. Cook et al. (2011) showed no modulation of swimming speed by the New Zealand snapper Pagrus auratus at any level of O2 prior to avoidance. The yellowtail kingfish Seriola lalandi also showed a burst and rest mode of swimming during progressive inescapable hypoxia, which disappeared when escapable low O2 conditions were presented (Cook and Herbert 2012). There is therefore accumulating evidence that heightened swimming speed, does not necessarily constitute part of a functional avoidance response across a range of different ecotypes. The increased swimming speed of some fish species in response progressive hypoxia may therefore simply be a

144

P. Domenici et al.

Fig. 6.6 Oxygen level behind schools composed of different numbers of individuals of blacksmith (Chromis punctipinnis). Y = -0.256 Log(X) ? 8.172; R2 = 0.96; p \ 0.0001; Data from Green and McFarland (1994)

panic reaction with no adaptive value and should therefore be minimised in aquaculture at all costs. Examining the swimming speed reaction of other active pelagic fish such as tuna and herring whilst also allowing the opportunity to escape will hopefully provide future insights.

6.4 The Effects of Hypoxia on Schooling Many gregarious fish species that live in coastal areas, lagoons and estuaries may be subject to hypoxic conditions seasonally. Pelagic species that live in large schools such as herring may also face hypoxia as reported in some areas like the Kattegat and some Norwegian Fjords (Domenici et al. 2002; Dommasnes et al. 1994; Hognestad 1994). Furthermore, field work has demonstrated that the oxygen level within a school tends to decrease along its axis of motion as a result of the oxygen consumption by the fish in the front of the school (McFarland and Moss 1967; Green and McFarland 1994; Fig. 6.6). The probability that fish in the centre/ back of a school experience low oxygen conditions may be particularly great for large schools. As a result, school size itself may be limited by oxygen levels (Steffensen 1995). Understanding the effect of hypoxia on schooling is important because many of the aspects involved (e.g. school resizing) may be heavily constrained under culture conditions.

6.4.1 School Volume and Spacing Hypoxia can affect schooling behaviour by inducing changes in both their structure and dynamics (Moss and McFarland 1970; Israeli and Kimmel 1996; Domenici et al. 2000a, 2002). One of the main changes is an increase in school volume


145

Fig. 6.7 a Example of the effect of hypoxia on school volume (Vt in L3) and speed (in Length s-1) in a school of herring subject to progressive hypoxia. b The effect of oxygen saturation on the average herring school volume. c The effect of hypoxia on the frequency of O-turn manoeuvres in schools of herring. From Domenici et al. (2007a). Asterisks represent significant differences from normoxia

(Figs. 6.7a and b; Domenici et al. 2002), which may provide individuals with more oxygen available, thereby counteracting the limiting effect of hypoxia on schools. Changes in volume may result in an increase of the oxygen available for rear fish in two ways: (1) by increasing cross-sectional area (i.e. width and depth) of the school and (2) by increasing school length, and therefore oxygen availability, provided that sufficient mixing of water mass occurs within the school. Such an increase in school volume may work for schools in the wild but not schools in the laboratory or aquaculture when fish are kept in confined set-ups. It is possible that fish in the laboratory or under culture may be compelled to show those behaviours that would minimise oxygen distress in natural situations (Domenici et al. 2007a, b). Experimental data on schooling herring show an increase from a specific volume (volume of water per fish) of 1.5 L3 (1.5 Length 3) in normoxia to [5 L3 at 20 % oxygen saturation (Domenici et al. 2002). This increase in volume may be explained in two ways: (1) As a behaviour response to hypoxia evolved to minimise the oxygen distress and (2) As related to a decrease in sensory performance of each single fish (Domenici et al. 2007a, b). Although the increase in volume was reflected in an increase in all dimensions of the school (X Y and Z), the school’s horizontal spread (school area) increased significantly starting from 30 % oxygen saturation, whilst school depth was significantly different from that in normoxia only at B20 % oxygen saturation. Similarly, Israeli and Kimmel (1996) found that only the horizontal dimensions increased in hypoxia in Carassius auratus. It is therefore possible that the

146

P. Domenici et al.

first effect of hypoxia may be to induce an increase in the horizontal spacing between fish, thereby allowing fish to keep some hydrodynamic advantages from following their neighbours (Herskin and Steffensen 1998; Johansen et al. 2010). A relatively flat school is known to maximise such energetic advantages, whilst the displacement of fish in different vertical planes would not confer such a great benefit (Weihs 1973; Abrahams and Colgan 1985). The rate at which oxygen declines may modulate the effect of hypoxia on schooling. Progressive hypoxia induced within 1.5 h did not cause any changes in schooling behaviour (measured using a density index, a parallel orientation index and swimming speed), whilst acute changes altered swimming speed significantly (Moss and McFarland 1970). Such fast responses may be due to peripheral O2 sensitive receptors located in the gills (Smatresk 1990). It is important to determine the time course of oxygen level changes in the wild, which may change from species to species and depend on environmental characteristics of the region. Whilst temporal changes such as those related to plant and algal oxygen consumption may be of the order of hours (Domenici et al. 2007a, b), fish that swim across oxygen gradient may experience a change from normoxia to hypoxia within seconds (Domenici et al. 2007a, b).

6.4.2 School Integrity and Spontaneous Activity Hypoxia is known to have an effect on the spontaneous activity of various fish species (see previous sections). As schooling structure and dynamics are largely affected by activity level and swimming speed (Pitcher and Partridge 1979), it is theoretically possible that the change in school structure observed in hypoxia may be an indirect result of a change in activity. However, work on herring shows that this is not the case, because the increase in school volume is relatively decoupled from the increase in swimming speed. Speed peaks at 30 % whilst school volume peaks at 20 % oxygen saturation (Fig. 6.7; Domenici et al. 2000a, 2002) Work on herring shows that the activity level shown by fish prior to hypoxia exposure modulated their response to hypoxia. School disruption occurred at higher oxygen levels in schools that had a higher spontaneous activity prior to the severe hypoxia (Domenici et al. 2000a, b). This may be because fish that had higher activity prior to hypoxia may experience higher overall exhaustion and distress and their schools may therefore disperse at a higher oxygen level than schools which have low spontaneous activity prior to hypoxia. School break up may ultimately be caused by the effect of respiratory distress on the performance of the sensory channel that help to maintain school cohesion, such as lateral line and vision (Partridge and Pitcher 1980). Field work by Mcfarland and Moss (1967) shows that school break up can occur at the back end of the school at oxygen levels near 5 mg l-1. They suggest that this is a way for fish to decrease school size by forming smaller schools at the back end, thereby avoiding massive oxygen


147

decrease throughout the school. However, this is not really a response available to fish schooling under intensive rearing conditions.

6.4.3 Schooling Dynamics Schooling fish can have positional preference in relation to a number of physiological factors such as hunger level (Krause and Ruxton 2002) and their MS (Killen et al. 2012b). Nevertheless, these positions are not completely fixed, and there is evidence that individuals in a school perform a certain level of positional turn-over (Pitcher et al. 1982; Krause and Ruxton 2002). Positional reshuffling may allow individual fish to be exposed to the various advantages and disadvantages related to each position. For instance, front positions are known to confer feeding advantages (Krause 1993) but higher predation risk (Bumann et al. 1997), whilst being behind leaders can confer hydrodynamic advantages (Herskin and Steffensen 1998; Johansen et al. 2010; Killen et al. 2012b). Shuffling behaviour can be affected by hypoxia. Domenici et al. (2002) found that the individual in the lead tended to perform a turning manoeuvre that repositioned them at the back of the school. This manoeuvre (termed ‘‘O-turn’’) was observed on average about 0.8 per minute per individual in normoxia. In hypoxia, O-turn frequency decreased dramatically to \0.1 O-turn manouvre per minute per fish (Fig. 6.7c). Interestingly, however, the decrease in shuffling behaviour did not affect the time fish spent in the leading positions. Hence, whilst hypoxia appeared to inhibit ‘‘active’’ reshuffling through the O-turn manoeuvres, the overall internal mixing in the school was not affected. This is perhaps because internal mixing was maintained by other mechanisms such as an increased rate of overtaking or falling back of individual fish. The observed decrease in O-turn frequency may have been caused due to their cost, which may not be supported during respiratory distress such as that caused by hypoxia.

6.4.4 Trade-Offs in Schooling: The Effect of Biotic and Abiotic Factors School structure and behaviour may be the result of a number of trade-offs. Various studies suggest that horizontal dimensions may be the first component of school volume to be affected (Israeli and Kimmel 1996; Domenici et al. 2002). In normoxia, relatively small spacing (about 1 Length between individuals) is desirable because it allows fast within-school communication via hydrodynamic signals (Gray and Denton 1991). By affecting spacing, possibly because of decreased sensory performance, hypoxia is likely to have a negative effect on the effectiveness of anti-predator manoeuvres (Domenici et al. 2007a, b). Furthermore,

148

P. Domenici et al.

small spacing can confer hydrodynamic advantages (Weihs 1973). Whilst larger interindividual distances may affect the hydrodynamic advantages of schooling, it is likely that increasing horizontal spacing will decrease the hydrodynamic benefit of schooling to a lesser extent than large vertical spacing. Abrahams and Colgan (1985) tested the hypothesis that fish school shape may be the result of two conflicting forces, i.e. (1) schools should be relatively flat, to maximise the hydrodynamic advantages of following the vorticity produced by fish in front (2) schools should be spread in depth, to maximise the visual fields of each individual, allowing for predator perception otherwise blocked by neighbours in the same plane. Abrahams and Colgan (1985) found that when no predator was present, characin fish tended to form relatively flat schools, whilst a larger depth was observed in the presence of predators. In hypoxia, the horizontal school dimension is the first variable to be affected (Israeli and Kimmel 1996; Domenici et al. 2007a), implying schools become more flat than deep, maximising energetic advantages over of visual fields for predator detection. Since these studies were based on laboratory observations, it would be interesting to test in the field if a similar trade-off occurs in the presence and absence of predators. School shapes commonly found in the wild may be due to a compromise between maximising oxygen availability and antipredator advantages (Brierly and Cox 2010). Being in the centre of the school reduces the risk of predation, but also decreases oxygen availability. School shapes were measured as B = Surface/ Volume and yield a value of approximately 3.3 m-1. According to Briely and Cox (Brierly and Cox 2010), this shape corresponds to the one that optimises shelter from predation and oxygen availability. Brierly and Cox (2010) however do not distinguish between moving and steady schools. It is possible that motion may introduce a further factor affecting school shape, i.e. hydrodynamic advantages. Further field studies could help establishing how fish deal with these trade-offs. In fact, many hypoxia-induced behaviours (increase of interindividual distances, and increase in horizontal dimensions) are in contraposition with those induced by predator presence. Another trade-off is represented by shuffling rates. Being in the front can be advantageous for feeding, but it is associated with higher predation risk and higher energetic costs (Bumann et al. 1997; Herskin and Steffensen 1998). In hypoxia, being in the front converts the advantage of avoiding a further reduction in oxygen level due to the oxygen consumption of fish in the front (Fig. 6.6). However, although hypoxia affects ‘‘active’’ reshuffling (through its affect on the frequency of O-turn manoeuvres), no effect on the overall shuffling rates were found (Domenici et al. 2002); Hence, it is possible that other factors may be more important in regulating shuffling rates, such as nutritional state energetics and perceived risk of predation (Krause 1993; Krause et al. 1998). Hypoxia clearly affects the schooling behaviour of fish in the wild but caution should be exercised in extrapolating the conclusions to common culture species like Atlantic salmon Salmo salar that are assumed to ‘‘school’’ in sea cage aquaculture. Salmon do show organised circular ‘‘school-like’’ swimming in sea cages (Oppedal et al. 2011) but this behaviour is not true schooling involving a


149

highly synchronous change in swimming direction and speed. Rather it is believed to occur as a result of cumulative avoidance reactions where fish avoid collisions with conspecifics and/or enclosure netting and gradually settle into the same circular pattern of swimming. The effect of hypoxia on the individual and school group dynamics of individual herring, for example, may not therefore extend to salmon in culture. More work is clearly required to resolve the schooling response of true schooling species and species like S. salar in aquaculture.

6.5 The Effects of Hypoxia on Predator–Prey Interactions In addition to the effect of hypoxia discussed in the sections above, previous work indicates that sublethal levels of oxygen may also influence survival by affecting predator–prey interactions (e.g. Breitburg et al. 1994; Robb and Abrahams 2002; Shoji et al. 2005; Domenici et al. 2007a; Rosal and Seibel 2008). Indeed, differential responses to hypoxia by predators and prey is likely to be a major factor in shaping the future abundance and distribution of fish and other marine organisms in increasingly hypoxic coastal areas around the world (Domenici et al. 2007a, b). Predation on fish in culture is also problematic and requires us to understand the likely impact of environmental forces such as low O2. Indeed, hypoxia may provide excellent opportunities for predation by birds at fish farm sites. During predator–prey interactions, both predators and prey swim anaerobically. Previous authors have therefore hypothesised that hypoxia may have no effect on the brief, anaerobic activity associated with escape (Beamish 1978). On the other hand, systemic hypoxia could impair brain and sensory function, which are fundamental for the execution of fast start motions (Domenici et al. 2007a). Accordingly, as we shall see, the potential consequences of hypoxia for predator attacks and prey responses are not confined to the effect of hypoxia on locomotion, but also on other variables such as sensory performance, hunger levels in the predators, aquatic surface respiration, and therefore these are discussed below along with the effect of hypoxia on locomotion per se.

6.5.1 Effects on Prey Escape Performance Predator–prey relationships in fish are largely based on sensory-motor performance (Domenici and Blake 1997). Hypoxia impaired locomotor performance of the escape response in golden grey mullet Liza aurata (Lefrançois et al. 2005), which can be expected to increase their vulnerability to predation (Walker et al. 2005). Golden grey mullet showed an increase in the proportion of single bend responses and a consequent decrease in proportion of double bend responses in hypoxia (at 10 % of air saturation) compared to normoxia. Single bend (SB)

150

P. Domenici et al.

(a)

SB

0.20

DB

(b)

a

0.15 D (m)

b 0.10 0.05 0.00 Double Bend 3.5

(c)

250

a

200

2.5

b

2.0 1.5 1.0

Amax(m s-2)

Vmax(m s-1)

3.0

Single Bend

(d) a b

150 100 50

0.5

0

0.0 Double Bend

Single Bend

Double Bend

Single Bend

Fig. 6.8 a Examples of tracings of two types of escape response: single bend (SB, performed in hypoxia) and double bend (DB, performed in normoxia). Single bend response only occurred in hypoxia (50 % of the hypoxic trials) whilst in normoxia only double bend responses occurred. Midline and centre of mass (dot) of the fish are shown at 10 ms intervals from the frame preceding the onset of the response. Arrow indicated the head. Whilst DB response shows a reversal in the direction of turning of the head (at frame six) and the fish return flip is complete, in SB the fish goes into a glide after frame six. b, c and d show the locomotor performance associated to SB and DB escape responses pooled independently of the oxygen treatment (D Distance covered, Vmax maximum swimming speed, Amax maximum acceleration). Modified from Lefrançois et al. (2005) with permission from the Journal of Fish Biology

responses, in which only one muscle contraction occurs during fast start, are known to result in lower performance than double bend (DB) responses with two muscle contractions (Domenici and Blake 1997). Whilst in normoxia all responses were of DB type, in hypoxia SB responses occurred in *50 % of the fast start responses and were associated with a significant reduction in both cumulative distance covered and maximum swimming speed (Fig. 6.8; Lefrançois et al. 2005). The effect of hypoxia on performance was significant only when fish were not allowed to perform aquatic surface respiration (ASR), which highlights the importance of the trade-off between (1) swimming to the surface to breath the oxygenated layer of water and becoming more visible to aerial predators versus (2) staying in the water column in order to avoid visibility by aerial predators, but experiencing low oxygen levels (Kramer et al. 1983; Kramer 1987; Domenici et al. 2007a). This trade off is a common feature of fish that perform ASR in coastal areas and lagoons, and laboratory work has shown that ASR in hypoxia can be modulated by turbidity and the presence of a model predator (Shingles et al. 2005). When exposed to a predator bird model, ASR in mullet was observed at lower oxygen levels than without the presence of the model predator. However, this delay in ASR was abolished when the experiments were carried out in turbid


151

waters. The extent to which individuals perform ASR in hypoxia is also directly related to their metabolic rate and oxygen demand (Killen et al. 2012a, b), such that those with elevated metabolic rate are presumably more sensitive to predation at the surface by, for example, birds during hypoxic episodes. Since muscle contraction in fast starts is fuelled anaerobically (Domenici and Blake 1997), the hypoxia-related decline in performance (i.e. increased number of single bends) was suggested to be due to changes in the balance between physiological exhaustion and the need to escape from a predator attack. The threshold at which such changes are expected may be species specific. European sea bass, for example, exposed to the same hypoxic conditions as L. aurata, did not experience any impairment in fast start locomotor performance (Lefrançois and Domenici 2006). Escape responses are typically triggered by one of a pair of giant neurons (the Mauthner cells) as a result of sensory input from mechano-acoustic or visual cues, although alternative neural pathways have been described (Eaton and Hackett 1984). Therefore, in addition to locomotor performance, the outcome of an escape response may depend on the sensory performance of the prey. Sensory functionality can be assessed by measuring variables such as responsiveness (i.e. the proportion of animals that respond to the stimulus) and escape latency (i.e. the time interval between stimulus onset and the first detectable movement of the escape response). Both variables are related to the reactivity of the prey to an external stimulus such as an approaching predator, and their motivation to escape. Hypoxia causes a decrease in responsiveness in golden grey mullet (Lefrançois et al. 2005), European sea bass (Lefrançois and Domenici 2006), as well as in common sole (Cannas et al. 2012), although not to the same extent. At 10 % of air saturation, 69 % of L. aurata but only 37 % of D. labrax responded to an external acoustic stimulus, compared with close to 100 % in normoxia. On the other hand, a reduction in responsiveness might be expected in flatfish, because of their benthic and mimetic lifestyle. When facing a predator, a significant proportion of common sole Solea solea tended to freeze rather than escape, and total responsiveness was only 70 % in normoxic conditions (Ellis et al. 1997). If the same species was startled by a stimulus that mimicked aerial predation, only about 57 % responded in normoxia, and this was reduced to 15 % in hypoxia at 15 % of air saturation (Cannas et al. 2012). Such low proportions of responding individuals suggest a hypoxia-related reduction of acoustic/visual sensitivity and/or motivation to escape. However, when a response is observed, the time course was not affected by hypoxia, which had no significant effect on latency in either L. aurata or D. labrax over a wide range of oxygen conditions (Lefrançois et al. 2005; Lefrançois and Domenici 2006). Alterations in escape performance, such as reduction in speed, may have important ecological consequences on the outcome of predator–prey interactions since it increases predation success (Walker et al. 2005). The increasing proportion of non-responses in prey exposed to hypoxia is also likely to raise their vulnerability to predators, since unless predators make an error, the absence of an escape attempt leads unavoidably to prey capture. In addition, hypoxia influenced the

152

P. Domenici et al.

directionality of the response, namely whether the initial C-bend was oriented away or towards the threat, in L. aurata (Lefrançois et al. 2005) and D. labrax (Lefrançois and Domenici 2006). The proportion of ‘away:towards’ escape response was affected, being not significantly different from 50:50 in at \20 % of air saturation in L. aurata and \50 % in D. labrax. Hypoxia tended therefore to induce random directionality in both of these species, suggesting a significant impairment of the left–right discrimination in the individuals tested. It is worth noting that most individuals which showed a C-bend oriented towards the predator at the initiation of the fast start nonetheless showed a final escape trajectory away from the stimulus. However, ‘correcting’ a tactical error during the progress of the escape response may induce a significant delay in getting away from the predator. Since the initial milliseconds may be crucial for survival, a high proportion of ‘towards’ response may reduce the probability of success.

6.5.2 Effects on Prey Visibility In addition to effects on locomotor and non-locomotor escape performance, oxygen may influence the vulnerability of a prey through effects on behaviour and/or physiology. In addition to causing aquatic surface respiration and aerial breathing in a number of species (Chapman and McKenzie 2009), a primary reflex response to hypoxia is increased ventilatory activity (Randall 1982; Burleson et al. 1992). Whilst this aids in regulating oxygen supply, in some species it may also raise the risk of predation. Elevated ventilation in known to increase the visibility of the prey, especially for cryptic species, as well as the release of chemical cues that may contribute to detection. Fear of predation inhibits gill ventilation in many species (Shingles et al. 2005; Cannas et al. 2012) but, in startled S. solea, this inhibitory effect is less pronounced in hypoxia than in normoxia (Cannas et al. 2012), which may reduce the effectiveness of crypsis.

6.5.3 Effects on Predator Behaviour In the case of oxygen-sensitive predators, such as fish and cephalopods, predator– prey interactions are likely also to be influenced by hypoxic impacts on predator behaviour and/or physiological performance. It has been demonstrated in many species that hypoxia diminishes appetite, reducing thereby the occurrence of strikes by predators (e.g., Breitburg et al. 1994, 1999; Chabot and Dutil 1999; Pichavant et al. 2000; Mallekh and Lagardère 2002; Robb and Abrahams 2002; Shimps et al. 2005). These reductions were observed at various oxygen levels in different species, indicating a species-specific pattern of limiting oxygen thresholds, similar to other effects of hypoxia. For instance, in G. morhua, D. labrax or turbot Scophthalmus maximus fed on dead prey, ingestion rate was significantly reduced between 60 and


153

40 % of air saturation (Chabot and Dutil 1999; Thetmeyer et al. 1999; Mallekh and Lagardère 2002). Jordan and Steffensen (2007) showed the SDA-response in Atlantic cod lasted 212 h in hypoxia compared to only 95 h in normoxia. Predation rate itself decreased in juvenile striped bass Morone saxatilis and adult naked goby Gobiosoma bosc (Breitburg et al. 1994), as well as sea bream larvae (Pagrus major, Shoji et al. 2005) whilst feeding on live prey in severely hypoxic conditions (i.e. between 28 and 13 % air saturation). Predation tends to increase, however, if the predator is more tolerant of hypoxia than the prey, as shown by the increase predation rates of sea nettles on fish larvae (Breitburg et al. 1994) or perhaps in the case of seals preying on cultured fish. The different effects of hypoxia on predators and prey from different taxa are likely to be fundamental factors affecting their relative abundance in the wild or stock biomass in captivity.

6.6 Conclusions Hypoxia has profound effects on the ability of fish to perform aerobic exercise, on their spontaneous activity, including how they behave in schools, and on their performance in predator–prey encounters. Fish will also avoid hypoxia when given the opportunity. These are general conclusions, there is a great deal of diversity among species. Some species, such as cyprinids, appear to be able to acclimate to hypoxia, with plastic changes to gill morphology and blood oxygen carrying capacity that help them ameliorate limiting effects of hypoxia on performance. Air breathing, which is believed to have evolved in response to aquatic hypoxia, may allow such species completely to avoid hypoxic limitations to exercise performance, though increased air breathing may exposed these fish to higher predation risk (Kramer et al. 1983; Kramer 1987). The effects of hypoxia on spontaneous activity, which may differ between active and sedentary species, may be engendered by the fish being in an inescapable situation—they may behave very differently if they can escape to better oxygenated areas. Hypoxia can have major effects on schooling structure and dynamics and this may be particularly detrimental for large schools. The effects of hypoxia on predator–prey interactions will depend on the relative sensitivity of each. In particular, fish may be less able to escape and therefore at greater risk of predation by air breathers, such as aquatic mammals and birds. Thus, in nature, hypoxia can be expected to limit the ability of fish to simultaneously forage, swim against currents, and perform activities such as swimming and digesting. These physiological effects, plus direct effects of hypoxia on avoidance reactions, may have profound effects on fish distributions. By affecting schooling dynamics, hypoxia may have a negative impact on the effectiveness of schooling as an antipredator behaviour and on swimming energetics, in addition to limiting the size of schools. Effects of predator–prey interactions may cause complex changes to foodweb structures and relationships.

154

P. Domenici et al.

In aquaculture, water oxygen levels are generally carefully monitored and regulated, specifically to avoid effects of hypoxia. Hypoxia may however become an issue for sea cage culture in coastal areas. It is important to understand the nature of the cultured species’ tolerance of hypoxia, and its behavioural responses. In particular, if a gentle swimming current is used to promote growth and/or improve welfare, it is then essential to understand thresholds for effects of hypoxia on swimming performance, and the potential implications of behavioural responses to hypoxia. Acknowledgements PD received funding from the European Community’s Seventh Framework Programme (FP7/2007-2013) under Grant Agreement No. 266445 for the project Vectors of Change in Oceans and Seas Marine Life, Impact on Economic Sectors (VECTORS).

References Abrahams MV, Colgan P (1985) Risk of predation, hydrodynamic efficiency and their influence on school structure. Environ Biol Fishes 13:195–202 Beamish FWH (1978) Swimming capacity. In: Hoar WS, Randall DJ (eds) Fish physiology, vol VII. Academic Press, New York, pp 101–187 Bejda AJ, Studholme AL, Olla BL (1987) Behavioural responses of red hake, Urophycis chuss, to decreasing concentrations of dissolved oxygen. Environ Biol Fishes 19:611–621 Boisclair D, Tang M (1993) Empirical analysis of the influence of swimming pattern on the net energetic cost of swimming in fishes. J Fish Biol 42:169–183 Brady DC, Targett TE, Tuzzolino DM (2009) Behavioral responses of juvenile weakfish (Cynoscion regalis) to diel-cycling hypoxia: swimming speed, angular correlation, expected displacement, and effects of hypoxia acclimation. Can J Fish Aquat Sci 66:415–424 Breitburg DL, Steinberg N, DuBeau S, Cooksey C, Houde ED (1994) Effects of low dissolved oxygen on predation on estuarine fish larvae. Mar Ecol Prog Ser 104:235–246 Breitburg DL, Rose KA, Cowan JH (1999) Linking water quality to larval survival: predation mortality of fish larvae in an oxygen-stratified water column. Mar Ecol Prog Ser 178:39–54 Brierly AS, Cox MJ (2010) Shapes of krill swarms and fish schools emerge as aggregation members avoid predators and access oxygen. Cur Biol 20(19):1758–1762 Brown EJ, Bruce M, Pether S, Herbert NA (2011) Do swimming fish always grow fast? Investigating the magnitude and physiological basis of exercise-induced growth in juvenile New Zealand yellowtail kingfish, Seriola lalandi. Fish Physiol Biochem 37:327–336 Bumann D, Krause J, Rubenstein D (1997) Mortality risk of spatial position in animal groups: the danger of being in the front. Behaviour 134:1063–1076 Burgetz IJ, Rojas-Vargas A, Hinch SG, Randall DJ (1998) Initial recruitment of anaerobic metabolism during sub-maximal swimming in rainbow trout (Oncorhynchus mykiss). J Exp Biol 201:2711–2721 Burleson ML, Smatresk NJ, Milsom WK (1992) Afferent inputs associated with cardioventilatory control in fish. In: Hoar WS, Randall DJ, Farrell AP (eds) Fish Physiology, vol XIIB. Academic Press, New York, pp 389–426 Bushnell PG, Steffensen JF, Johansen K (1984) Oxygen consumption and swimming performance in hypoxia-acclimated rainbow trout Salmo gairdneri. J Exp Biol 113:225–235 Cannas M, Domenici P, Lefrançois C (2012) The effect of hypoxia on ventilation frequency in startled common sole Solea solea. J Fish Biol 80:2636–2642 Chabot D, Claireaux G (2008) Environmental hypoxia as a metabolic constraint on fish: the case of Atlantic cod, Gadus morhua. Mar Pollut Bull 57:287–294


155

Chabot D, Dutil JD (1999) Reduced growth of Atlantic cod in non-lethal hypoxic conditions. J Fish Biol 55:472–491 Chapman LJ, McKenzie DJ (2009) Behavioral responses and ecological consequences. In: Richards JG, Farrell AP, Brauner CJ (eds) Hypoxia. Elsevier, London, pp 25–77 Claireaux G, Chabot D (2005) A review of the impact of environmental hypoxia on fish: the case of Atlantic cod. Comp Biochem Physiol 141A:S177 Claireaux G, Webber DM, Lagardere JP, Kerr SR (2000) Influence of water temperature and oxygenation on the aerobic metabolic scope of Atlantic cod (Gadus morhua). J Sea Res 44:257–265 Claireaux G, Handelsman C, Standen E, Nelson JA (2007) Thermal and temporal stability of swimming performance in the European sea bass. Physiol Biochem Zool 80(2):186–196 Cook DG, Herbert NA (2012) The physiological and behavioural response of juvenile kingfish (Seriola lalandi) differs between escapable and inescapble progressive hypoxia. J Exp Mar Biol Ecol 413:138–144 Cook DG, Wells RMG, Herbert NA (2011) Anaemia adjusts the aerobic physiology of snapper (Pagrus auratus) and modulates hypoxia avoidance behaviour during oxygen choice presentations. J Exp Biol 214:2927–2934 Dahlberg ML, Shumway DL, Doudoroff O (1968) Influence of dissolved oxygen and carbon dioxide on swimming performance of largemouth bass and coho salmon. J Fish Res Bd Can 25:49–70 Dean TL, Richardson J (1999) Responses of seven species of native freshwater fish and a shrimp to low levels of dissolved oxygen. N Z J Mar Freshw Res 33(1):99–106 Diaz RJ (2001) Overview of hypoxia around the world. J Environ Qual 30:275–281 Diaz RJ, Rosenberg R (2008) Spreading dead zones and consequences for marine ecosystems. Science 321:926–929 Dizon AE (1977) Effect of dissolved oxygen concentration and salinity on swimming speed of two species of tunas. Fish B-NOAA 75:649–653 Domenici P, Blake RW (1997) The kinematics and performance of fish fast start swimming. J Exp Biol 200:1165–1178 Domenici P, Steffensen JF, Batty RS (2000a) The effect of progressive hypoxia on swimming activity and schooling in Atlantic herring. J Fish Biol 57:1526–1538 Domenici P, Batty RS, Simila T (2000b) Spacing of wild schooling herring while encircled by killer whales. J Fish Biol 57:831–836 Domenici P, Ferrari RS, Steffensen JF, Batty RS (2002) The effect of progressive hypoxia on school structure and dynamics in Atlantic herring Clupea harengus. Proc R Soc B 269:2103–2111 Domenici P, Claireaux G, McKenzie DJ (2007a) Environmental constraints upon locomotion and predator–prey interactions in aquatic organisms: an introduction. Phil Trans R Soc 362:1929–1936 Domenici P, Lefrançois C, Shingles A (2007b) The effect of hypoxia on the antipredator behaviours of fishes. Philos Trans R Soc 362:2105–2121 Dommasnes A, Rey F, Røttingen I (1994) Reduced oxygen concentrations in herring wintering areas. ICES J Mar Sci 51:63–69 Dupont-Prinet A, Chatain B, Grima L, Vandeputte M, Claireaux G, McKenzie DJ (2010) Physiological mechanisms underlying a trade-off between growth rate and tolerance of feed deprivation in the European sea bass (Dicentrarchus labrax). J Exp Biol 213:1143–1152 Eaton RC, Hackett JT (1984) The role of Mauthner cells in fast-starts involving escape in teleost fish. In: Eaton RC (ed) Neural mechanisms of startle behavior. Plenum Press, New York, pp 213–266 Ellis T, Howell BR, Hughes RN (1997) The cryptic responses of hatchery-reared sole to a natural sand substratum. J Fish Biol 51:389–401 Farmer CG, Jackson DC (1998) Air-breathing during activity in the fishes Amia calva and Lepisosteus oculatus. J Exp Biol 201(7):943–948

156

P. Domenici et al.

Farrell AP, Gamperl AK, Birtwell IK (1998) Prolonged swimming, recovery and repeat swimming performance of mature sockeye salmon Oncorhynchus nerka exposed to moderate hypoxia and pentachlorophenol. J Exp Biol 201:2183–2193 Fisher P, Rademacher K, Kils U (1992) In situ investigations on the respiration and behaviour of the eelpout Zoarces viviparus under short-terms hypoxia. Mar Ecol Prog Ser 88:181–184 Fitzgibbon QP, Seymour RS, Buchanan J, Musgrove R, Carragher J (2010) Effects of hypoxia on oxygen consumption, swimming velocity and gut evacuation in southern bluefin tuna (Thunnus maccoyii). Environ Biol Fish 89:59–69 Føre M, Dempster T, Alfredsen JA, Johansen V, Johansson D (2009) Modelling of Atlantic salmon (Salmo salar L.) behaviour in sea-cages: A Lagrangian approach. Aquaculture 288:196–204 Fritsche R, Nilsson S (1989) Cardiovascular responses to hypoxia in the Atlantic cod, Gadus morhua. J Exp Biol 48:153–160 Fry FEJ (1947) Effects of the environment on animal activity. University of Toronto studies Biological series, No. 55, pp 1–62 Fry FEJ (1971) The effect of environmental factors on the physiology of fish. In: Hoar WS, Randall DJ (eds) Fish Physiology, vol 6. Academic Press, New York, pp 1–98 Fu SJ, Brauner CJ, Cao ZD, Richards JG, Peng JL, Dhillon R, Wang YX (2011) The effect of acclimation to hypoxia and sustained exercise on subsequent hypoxia tolerance and swimming performance in goldfish (Carassius auratus). J Exp Biol 214:2080–2088 Graham JB (1997) Air breathing fishes: evolution, diversity, and adaptation. Academic Press, San Diego Graham JB (2006) Aquatic and aerial respiration. In: Evans DD, Claiborne JB (eds) The physiology of fishes, 3rd edn. CRC Press, Boca Raton, pp 85–117 Gray JAB, Denton EJ (1991) Fast pressure pulses and communication between fish. J Mar Biol Assoc UK 71:83–106 Green D, McFarland WN (1994) Impact of foraging blacksmiths on constituents in the water column: implications on school behaviour and structure. In Halverson WL, Maender GJ (eds) Proceedings of the 4th California Islands symposium: update on the status of resources.Santa Barbara Museum of Natural History, Santa Barbara, CA, pp 97–102 Grigg GC (1965) Studies on the Queensland lungfish, Neoceratodus forsteri (Krefft) III Aerial respiration in relation to habits. Aust J Zool 13:413–421 Herbert NA, Steffensen JF (2005) The response of Atlantic cod, Gadus morhua, to progressive hypoxia: fish swimming speed and physiological stress. Mar Biol 147:1403–1412 Herbert NA, Steffensen JF (2006) Hypoxia increases the behavioural activity of schooling herring: a response to physiological stress or respiratory distress? Mar Biol 149:1217–1225 Herbert NA, Skov PV, Wells RMG, Steffensen JF (2006) Whole blood-oxygen binding properties of four cold-temperate marine fishes: blood-affinity is independent of pH-dependent binding, routine swimming performance and environmental hypoxia. Physiol Biochem Zool 79:909–918 Herbert NA, Skjæraasen JE, Nilsen T, Salvanes AG, Steffensen JF (2011) The hypoxia avoidance behaviour of juvenile Atlantic cod (Gadus morhua L.) depends on the provision and pressure level of an O2 refuge. Mar Biol 158:737–746 Herskin J, Steffensen JF (1998) Reduced tail beat frequency and oxygen consumption due to hydrodynamic interactions of schooling sea bass, Dicentrarchus labrax L. J Fish Biol 53:366–376 Hognestad PT (1994) The Lake Rossfjord herring (Clupea harengus L.) and its environment. ICES J Mar Sci 51:281–292 Israeli D, Kimmel E (1996) Monitoring the behavior of hypoxia-stressed Carassius auratus using computer vision. Aquacult Eng 15(6):423–440 Jobling M (1994) Fish Bioenergetics. Chapman and Hall, London Johansen JL, Vaknin R, Steffensen JF, Domenici P (2010) Kinematics and energetic benefits of schooling in the labriform fish, striped surfperch Embiotoca lateralis. MEPS 420:221–229 Jones DR (1971) The effect of hypoxia and anemia on the swimming performance of rainbow trout (Salmo gairdneri). J Exp Biol 55:541–551 Jordan AD, Steffensen JF (2007) The effect of ration size and hypoxia on the specific dynamic action in the cod. Physiol Biochem Zool 80:178–185


157

Jourdan-Pineau H, Dupont-Prinet A, Claireaux G, McKenzie DJ (2010) An investigation of metabolic prioritization in the European sea bass, Dicentrarchus labrax. Physiol Biochem Zool 83:68–77 Keeling RF, Kortzinger A, Gruber N (2010) Ocean deoxygenation in a warming world. Annu Rev Mar Sci 2:199–229 Kemp PS, Tsuzaki T, Moser ML (2009) Linking behaviour and performance: intermittent locomotion in a climbing fish. J Zool 277(2):171–178 Killen SS, Marras S, Ryan MR, Domenici P, McKenzie DJ (2012a) A relationship between metabolic rate and risk-taking behaviour is revealed during hypoxia in juvenile European sea bass. Funct Ecol. doi:10.1111/j.1365-2435.2011.01920.x Killen SS, Marras S, Steffensen JF, McKenzie DJ (2012b). Aerobic capacity influences the spatial position of individuals within fish schools. Proc R Soc B. doi:10.1098/rspb.2011.1006 Kramer DL (1987) Dissolved oxygen and fish behaviour. Environ Biol Fish 18:81–92 Kramer DL, Manley D, Bourgeois R (1983) The effect of respiratory mode and oxygen concentration on the risk of aerial predation in fishes. Can J Zool 61:653–665 Kramer DL, Rangeley RW, Chapman LJ (1997) Habitat selection: patterns of spatial distribution from behavioural decisions. In Godin JGJ (ed) Behavioural ecology of teleost fishes. Oxford University Press, Oxford, pp 37–80 Krause J (1993) The relationship between foraging and shoal position in a mixed shoal of roach (Rutilus rutilis) and chub (Leuciscus cephalus)—a field study. Oecologia 93(3):356–359 Krause J, Ruxton GD (2002) Living in groups. Oxford University Press, Oxford Krause J, Reeves P, Hoare D (1998) Positioning behaviour in roach shoals: the role of body length and nutritional state. Behaviour 135:1031–1039 Kutty MN (1968) Influence of ambient oxygen on the swimming performance of goldfish and rainbow trout. Can J Zool 46:647–653 Lefevre S, Huong DTT, Kim NT, Wang T, Phuong NT, Bayley M (2011) A telemetry study of swimming depth and oxygen level in a Pangasius pond in the Mekong Delta. Aquaculture 315:410–413 Lefrançois C, Claireaux G (2003) Influence of ambient oxygenation and temperature on metabolic scope and scope for heart rate of the sole (Solea solea). Mar Ecol Prog Ser 259:273–284 Lefrançois C, Domenici P (2006) Locomotor kinematics and responsiveness in the escape behaviour of European sea bass (Dicentrarchus labrax) exposed to hypoxia. Mar Biol 149:969–977 Lefrançois C, Shingles A, Domenici P (2005) The effect of hypoxia on locomotor performance and behaviour during escape in the golden grey mullet (Liza aurata). J Fish Biol 67:1711–1729 Lefrançois C, Ferrari RS, da Silva JM, Domenici P (2009) The effect of progressive hypoxia on spontaneous activity in single and shoaling golden grey mullet Liza aurata. J Fish Biol 75:1615–1625 Lowe TE, Brill RW, Cousins KL (2000) Blood oxygen-binding characteristics of bigeye tuna (Thunnus obesus), a high-energy-demand teleost that is tolerant of low ambient oxygen. Mar Biol 136:1087–1098 Mallekh R, Lagardère JP (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 60(5):1105–1115 McFarland WN, Moss SA (1967) Internal behavior in fish schools. Science 156:260–262 McKenzie DJ, Claireaux G (2010) Effects of environmental factors on the physiology of sustained aerobic exercise. In: Domenici P, Kapoor BG (eds) Fish Locomotion—an ethoecological perspective. Science Publishers, New Hampshire, pp 296–332 McKenzie DJ, Steffensen JF, Taylor EW, Abe AS (2012) The contribution of air-breathing to aerobic scope and exercise performance in the banded knifefish Gymnotus carapo L. J Exp Biol 215:1323–1330 Metcalfe JD, Butler PJ (1984) Changes in activity and ventilation in response to hypoxia in unrestrained, unoperated dogfish (Scyliorhinus canicula L.). J Exp Biol 180:153–162 Milligan CL, Hooke GB, Johnson C (2000) Sustained swimming at low velocity following a bout of exhaustive exercise enhances metabolic recovery in rainbow trout. J Exp Biol 203:921–926

158

P. Domenici et al.

Moss SA, McFarland WN (1970) Influence of dissolved oxygen and carbon dioxide on fish schooling behavior. Mar Biol 5:100–107 Neuenfeldt S, Andersen KH, Hinrichsen H-H (2009) Some Atlantic cod Gadus morhua in the Baltic Sea visit hypoxic water briefly but often. J Fish Biol 75(1):290–294 Nilsson GE, Rosen P, Johansson D (1993) Anoxic depression of spontaneous locomotor activity in crucian carp quantified by a computerized imaging technique. J Exp Biol 180:153–162 Oppedal F, Dempster T, Stien LH (2011) Environmental drivers of Atlantic salmon behaviour in sea-cages: a review. Aquaculture 311:1–18 Partridge BL, Pitcher TJ (1980) The sensory basis of fish schools: relative roles of lateral line and vision. J Comp Physiol 135:315–325 Petersen CL (1987) Energy budgets for juvenile rainbow trout at various oxygen concentrations. Aquaculture 62:289–298 Petersen LH, Gamperl AK (2010) Effect of acute and chronic hypoxia on the swimming performance, metabolic capacity and cardiac function of Atlantic cod (Gadus morhua). J Exp Biol 213:808–819 Pichavant K, Person-Le-Ruyet J, Le Bayon N, Sévère A, Le Roux A, Quémerer L, Maxime V, Nonotte G, Bœuf G (2000) Effects of hypoxia on growth and metabolism of juvenile turbot. Aquaculture 188:103–104 Pitcher TJ, Partridge BL (1979) Fish school density and volume. Mar Biol 54:383–394 Pitcher TJ, Wyche CJ, Magurran AE (1982) Evidence for position preferences in schooling mackerel. Anim Behav 30:932–934 Plante S, Chabot D, Dutil JD (1998) Hypoxia tolerance in Atlantic cod. J Fish Biol 53:1342–1356 Poulsen SB, Jensen LF, Nielsen KS, Malte H, Aarestrup K, Svendsen JC (2011) Behaviour of rainbow trout Oncorhynchus mykiss presented with a choice of normoxia and stepwise progressive hypoxia. J Fish Biol 79(4):969–979 Randall D (1982) The control of respiration and circulation in fish during exercise and hypoxia. J Exp Biol 100:275–288 Randall DJ, Burggren WW, Farrell AP, Haswell MS (1981) The evolution of air-breathing in vertebrates. Cambridge University Press, Cambridge Richards JG (2009) Metabolic and molecular responses of fish to hypoxia. In: Richards JG, Farrell AP, Brauner CJ (eds) Hypoxia. Elsevier, London, pp 443–485 Richards JG, Heigenhauser GJF, Wood CM (2002) Lipid oxidation fuels recovery from exhaustive exercise in white muscle of rainbow trout. Am J Physiol 282:R89–R99 Robb T, Abrahams MV (2002) The influence of hypoxia on risk of predation and habitat choice by the fathead minnow, Pimephales promelas. Behav Ecol Sociobiol 52:25–30 Rosa R, Seibel BA (2008) Synergistic effects of climate-related variables suggest future physiological impairment in a top oceanic predator. PNAS 105(52):20776–20780 Schurmann H, Steffensen JF (1994) Spontaneous swimming activity of Atlantic cod Gadus morhua exposed to graded hypoxia at three temperatures. J Exp Biol 197:129–142 Schurmann H, Steffensen JF (1997) Effects of temperature, hypoxia and activity on the metabolism of juvenile Atlantic cod. J Fish Biol 50:1166–1180 Seymour RS, Christian K, Bennett MB, Baldwin J, Wells RMG, Baudinette RV (2004) Partitioning of respiration between the gills and air-breathing organ in response to aquatic hypoxia and exercise in the Pacific tarpon, Megalops cyprinoides. Physiol Biochem Zool 77:760–767 Seymour RS, Farrell AP, Christian K, Clark TD, Bennett MB, Wells RMG, Baldwin J (2007) Continuous measurement of oxygen tensions in the air-breathing organ of Pacific tarpon (Megalops cyprinoides) in relation to aquatic hypoxia and exercise. J Comp Physiol B 177:579–587 Shimps EL, Rice JA, Osborne JA (2005) Hypoxia tolerance in two juvenile estuary dependent fishes. J Exp Mar Biol Ecol 325:146–162 Shingles A, McKenzie DJ, Claireaux G, Domenici P (2005) Reflex cardioventilatory responses to hypoxia in the flathead grey mullet (Mugil cephalus) and their behavioural modulation by perceived threat of predation and water turbidity. Physiol Biochem Zool 78:744–755 Shoji J, Masuda R, Yamashita Y, Tanaka M (2005) Effect of low dissolved oxygen concentrations on behavior and predation rates on red sea bream Pagrus major larvae by


159

the jellyfish Aurelia aurita and by juvenile Spanish mackerel Scomberomorus niphonius. Mar Biol 147:863–868 Smatresk NJ (1990) Chemoreceptor modulation of the endogenous respiratory rhythm in vertebrates. Am J Physiol 259:887–897 Smith KJ, Able KW (2003) Dissolved oxygen dynamics in salt marsh pools and its potential impacts on fish assemblages. Mar Ecol Prog Ser 258:223–232 Smith RS, Kramer DL (1986) The effect of apparent predation risk on the respiratory behavior of the Florida gar (Lepisosteus platyrhincus). Can J Zool 64:2133–2136 Sollid J, Nilsson GE (2006) Plasticity of respiratory structures-adaptive remodeling of fish gills induced by ambient oxygen and temperature. Respir Physiol Neurobiol 154:241–251 Steffensen JF (1995) Possible limitations of speed and size of swimming fish schools, based on oxygen consumption of herring, Clupea harengus, measured at different swimming speeds. J Physiol 483:192 Stierhoff KL, Targett TE, Power JH (2009) Hypoxia-induced growth limitation of juvenile fishes in an estuarine nursery: assessment of small-scale temporal dynamics using RNA:DNA. Can J Fish Aquat Sci 66:1033–1047 Stramma L, Prince ED, Schmidtko S, Jiangang L, Hoolihan JP, Visbeck M, Wallace DWR, Brandt P, Körtzinger A (2012) Expansion of oxygen minimum zones may reduce available habitat for tropical pelagic fishes Nature. Clim Change 2:33–37 Svendsen JC, Steffensen JF, Aarestrup K, Frisk M, Etzerodt A, Jyde M (2012) Excess posthypoxic oxygen consumption in rainbow trout (Oncorhynchus mykiss): recovery in normoxia and hypoxia. Can J Zool 90:1–11 Taylor JC, Miller JM (2001) Physiological performance of juvenile southern flounder, Paralichthys lethostigma (Jordan and Gilbert 1884), in chronic and episodic hypoxia. J Exp Mar Biol Ecol 258(2):195–214 Thetmeyer H, Waller U, Black KD, Inselmann S, Rosenthal H (1999) Growth of European sea bass Dicentrarchus labrax L. under hypoxic and oscillating oxygen condition. Aquaculture 174:355–367 Vagner M, Lefrancois C, Ferrari RS, Satta A, Domenici P (2008) The effect of acute hypoxia on swimming stamina at optimal swimming speed in flathead grey mullet Mugil cephalus. Mar Biol 155(2):183–190 Walker JA, Ghalambor CK, Griset OL, Kenney DM, Reznick DN (2005) Do faster starts increase the probability of evading predators? Funct Ecol 19:808–815 Webb PW (1998) Swimming. In: Evans DD (ed) The physiology of fishes, 2nd edn. CRC Press, Boca Raton, pp 3–24 Weihs D (1973) Hydromechanics and fish schooling. Nature 241:290–291 Weihs D (1974) Energetic advantages of burst swimming of fish. J Theor Biol 48(1):215–229 Wood CM (1991) Acid-base and ion balance, metabolism, and their interactions, after exhaustive exercise in fish. J Exp Biol 160:285–308 Zhang W, Cao ZD, Peng JL, Chen BJ, Fu SJ (2010) The effects of dissolved oxygen level on the metabolic interaction between digestion and locomotion in juvenile southern catfish (Silurus meridionalis Chen). Comp Biochem Physiol 157A:212–219