© 2004 by ICES/CIEM International Council for the Exploration of the Sea/Conseil International pour l'Exploration de la Mer
Simulating trawl herding in flatfish: the role of fish length in behaviour and swimming characteristics
a Centre for Sustainable Aquatic Resources, Marine Institute, Memorial University of Newfoundland PO Box 4920, St. John's, NL, Canada A1C 5R3
b Northwest Atlantic Fisheries Centre, Department of Fisheries and Oceans PO Box 5667, St. John's, NL, Canada A1C 5X1
c University of New Hampshire 137 Morse Hall, Durham, NH 03824, USA
d Ocean Sciences Centre, Memorial University of Newfoundland St. John's, NL, Canada A1C 5S7
*Correspondence to P. D. Winger: tel: +1 709 778 0678. e-mail: paul.winger{at}mi.mun.ca.
Theoretical modeling indicates that the herding efficiency of flatfish by bottom-trawl sweeps is highly sensitive to subtle changes in fish behaviour. Yet the degree of variation in herding behaviour within a species, population, or individual remains poorly understood because of the difficulty observing and measuring fish behaviour in this region of the gear. The purpose of this study was to simulate herding under laboratory conditions in order to examine the behaviour and swimming characteristics of flatfish. Using a large flume equipped with a moving floor, we examined the effect of fish length on gait use, behaviour, and swimming kinematics in American plaice (Hippoglossoides platessoides). While swimming at a speed comparable to the herding speed of trawl sweeps (0.3 m s–1), smaller plaice (<30 cm) spent a large percentage of time using the kick-swim gait, while larger fish (
30 cm) preferred cruising. In total, 65% of plaice exhibited settling behaviour, analogous to the swim-and-settle behaviour observed in response to trawl sweeps. The frequency of this behaviour and the distance swum between settles were independent of fish length. Only the frequency of gliding changed with the duration of swimming. Like other teleost species, tailbeat frequency decreased with increasing fish length. The results from this study indicate that fish length affects gait use and swimming kinematics in flatfish, but not the frequency of gliding and settling behaviours. These observations support the hypothesis of size-selective herding and provide further insight into the herding efficiency of trawl sweeps.
Keywords: behaviour, bottom-trawl, fish length, flatfish, gait use, herding efficiency, kinematics
Received 13 March 2003; accepted 23 May 2004.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
Bottom-trawl surveys are conducted worldwide for the assessment of fishery resources. Accurate knowledge of the area sampled by a survey trawl is a necessary parameter to derive estimates of population abundance. It is most often defined as the product of trawl width and distance towed (Gunderson, 1993). However, a proportion of fish that initially lie between the trawl doors but not in the trawl path may be herded into the path and caught. This increase in the effective pathwidth of a survey trawl has been the subject of considerable research (e.g. Dickson, 1993; Ramm and Xiao, 1995; Somerton and Munro, 2001; Somerton, 2003). For benthic species such as flatfish, herding is typically induced after direct or near contact with the doors, sand clouds, and sweeps. Underwater observations of trawl sweeps have revealed that flatfish, once disturbed from the seafloor, tend to swim toward the trawl path in a direction perpendicular to the advancing sweeps (Figure 1; Main and Sangster, 1981). To avoid being overtaken, the fish must swim at a speed equal to or greater than the herding speed of the sweeps. This herding speed is determined by the forward towing speed of the trawl and the sweep angle (i.e. angle of attack of the sweep to the direction of tow). Given that all fish are stimulated to swim at the same speed, fish of different sizes will operate at different levels within their performance range. In order to effectively swim at these different levels, flatfish like other teleost fish, are expected to exhibit more than one gait (Alexander, 1989). Gaits represent a multigear system (synonymous with walking, trotting, galloping) for the fractionation of the swimming performance range (Webb, 1994a). Similar to terrestrial animals, individual gaits in fish work over a part of the performance range, and a series of gaits are recruited to cover the entire range. Each gait is defined on the basis of its muscle use, propulsor type, and propulsor kinematics (Webb, 1994b). Gait use and swimming kinematics are not well documented for flatfish species and are predicted to be size-dependent for fish swimming at a fixed speed, such as in response to the sweeps of a bottom-trawl.
|
For herding to be effective, flatfish must also have sufficient endurance necessary to reach the trawl path. The distance required to swim is determined by the sweep angle and the position along the sweep where the fish initially encounter the gear. The probability of successfully swimming this distance is predicted to be size- and temperature-dependent (Winger et al., 1999). In situ observations of flatfish in response to the sweeps of a Danish seine (Hemmings, 1969, 1973) and bottom-trawl (High, 1969; Harden Jones et al., 1977; Main and Sangster, 1981) have revealed that flatfish are equipped with a repertoire of swimming behaviours. In general, once flatfish are disturbed from the seafloor, they tend to swim a short distance (0.5 to 5.0 m) ahead of the advancing sweep and settle onto the seafloor. Each time the sweep approaches, the fish may repeat the behaviour and progressively slip along the sweep toward the mouth of the trawl (Figure 1).
This incremental process of swimming and settling has been described using probabilistic models (Foster, 1969; Foster et al., 1981; Fuwa et al., 1988; Fuwa, 1989; Tanaka et al., 1991). These models demonstrate that herding efficiency is highly sensitive to subtle changes in the swim-and-settle behaviour. Parameters of importance include (i) the number of times a fish encounters the sweep, (ii) the probability of escape upon initial and subsequent encounters with the sweep, (iii) the distance swum following each encounter, and (iv) the total distance the fish is capable of swimming. Unfortunately, empirical data for these parameters are almost entirely lacking because of the difficulty observing and measuring fish behaviour in this region of the gear.
In this study, we undertook an analysis of video recordings collected during an earlier experiment on the swimming endurance of American plaice (Hippoglossoides platessoides) (Winger et al., 1999). We examine the effect of fish length on gait use, behaviour, and swimming kinematics for fish swimming at a speed comparable to the herding speeds of trawl sweeps.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Video recordings collected during an earlier experiment on the swimming endurance of American plaice (Winger et al., 1999) were selected for detailed analysis of gait use, behaviour, and swimming kinematics. In this experiment, individual plaice of varying length were induced to swim at a fixed speed of 0.3 m s–1. Depending on the type of trawl, its operation, and performance, this falls within the range of speeds (
0.2 to 0.7 m s–1) typically expected during herding (see Winger et al., 1999; Glass, 2000). The flume was operated on a flow-through basis with ambient temperature seawater. Internal chamber dimensions were 1.80 x 0.50 x 0.26 m (length x width x depth). Swimming activity was recorded using a Sony CCD-TR500 video camera at a rate of 30 frames s–1. Pairs of electrodes in the downstream end of the chamber emitted a pulsing electrical stimulus (2 Hz) with a peak voltage of 8 V. In the event that fish settled, a continuously moving belt in the floor of the chamber would carry the fish to the downstream end where they would be stimulated to resume swimming. The belt moved in the same direction and at the same speed as the water current and simulated the seafloor beneath a swimming fish. This process is considered to be analogous to the herding behaviour shown in Figure 1, but rather than a moving sweep and stationary water, we have presented a stationary stimulus (electrodes) with moving water and seafloor. This experiment complied with animal care guidelines and was designed to achieve the humane use and care of animals (see Winger et al. (1999) for details). We first examined video recordings for the purpose of investigating gait use and swimming behaviour. Video of fish ranging in length from 20 to 41 cm total length (mean = 31.6 cm; s.d. = 5.6, n = 23) swimming at temperatures of –0.2 to 0.9°C (mean = 0.4°C, s.d. = 0.4) were selected. Using event recording software (Observer 2.0, Noldus Information Technology), we quantified: (i) duration spent in different gaits, (ii) frequency of gliding and settling behaviours, (iii) distance swum between settles, (iv) distance settled ahead of the downstream electrodes, and (v) duration spent in different locations of the swimming chamber. Assuming that gait use or behaviour might change with the onset of fatigue (Webb, 1994a), we only examined video collected at the beginning of an endurance challenge (see Winger et al., 1999), thereby attempting to simulate the experience of fish initially aroused from the seafloor and herded toward the trawl path. Only the first 3 min were examined, as this was expected to represent the maximum period of time that flatfish would be under the influence of a trawl sweep (see Hemmings, 1969; Foster et al., 1981; Winger et al., 1999).
We also examined a selected number of video recordings from Winger et al. (1999) to determine the relationship between fish length and tailbeat frequency (f) during steady cruising. This was conducted for fish ranging in length from 23 to 44 cm total length (mean = 33.9 cm; s.d. = 5.3, n = 52) swimming at temperatures of –0.2 to 3.0°C (mean = 1.4°C, s.d. = 1.0). Tailbeat frequency was defined as the number of tailbeat cycles completed per second, where one tailbeat represented one complete oscillation of the tail. The total distance travelled per tailbeat, referred to as the stride length (S), was determined by S = 0.3 m s–1/f.
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
American plaice swam in the horizontal orientation typical of pleuronectiform species. Swimming was powered by undulations of the body and caudal fin, referred to as BCF propulsion (Webb, 1994a). Two gaits were observed and easily recognizable, with plaice often swimming for a variable time period in one before switching to the other. The cruising gait was characterized as steady beating of the caudal fin without deviation in tailbeat amplitude. The kick-swim gait was similar to cruising, but was punctuated with single large amplitude tailbeats every 3–5 s to increase thrust. Smaller plaice (<30 cm), swimming at higher relative speeds, spent a large percentage of time using the kick-swim gait (Figure 2). As fish size increased (>30 cm), there was a noticeable shift toward cruising. The burst-and-coast gait was not observed, but would have been expected at higher performance levels (Webb, 1994a).
|
Plaice swimming in either gait frequently exhibited gliding and settling. These were considered to be "unsteady" swimming behaviours (Webb, 1994a) and not representative of different gaits. Gliding was characterized as a brief unpowered phase in which the fish glided toward the chamber floor, typically with the body rigid and tilted slightly downward. The frequency of gliding ranged from 0.3 to 4.9 times per minute (mean = 2.5, s.d. = 1.3) and was independent of fish length (F[1,21] = 2.68, p = 0.12). The frequency of gliding tended to decrease with the duration of swimming, most notably in the small (20–27 cm) and medium (28–34 cm) sized fish (Figure 3a).
|
Settling was characterized by leaving the water column and resting on the chamber belt. It was always preceded by a glide and typically took place in the midstream or upstream locations of the swimming chamber. This resulted in a brief rest, while riding backward, until stimulated by the downstream electrodes. In total, 65% (15 of 23) of plaice, ranging in length from 23 to 41 cm, demonstrated this behaviour. Among these, the frequency of settling ranged from 0.3 to 1.4 times per minute (mean = 0.7, s.d. = 0.4) and was independent of fish length (F[1,14] = 1.73, p = 0.21). The frequency of settling did not change in any consistent manner with the duration of swimming (Figure 3b).
The distance swum between settles (i.e. product of the elapsed time and speed) was highly variable between and within individuals, with distances ranging from 1.2 to 50.3 m. Individual mean distances ranged from 10.5 to 50.3 m and were independent of fish length (Figure 4a, F[1,14] = 1.79, p = 0.21). The distance settled ahead of the downstream electrodes was also highly variable (Figure 4b). Mean distances ranged from 0.2 to 1.7 m and were similarly independent of fish length (F[1,14] = 0.01, p = 0.92). Of the plaice which employed settling, 53% (8 of 15) did so on more than one occasion. For these fish, neither the distance swum nor the distance settled changed with the duration of swimming (p > 0.05).
|
The proportion of time spent horizontally in different locations of the swimming chamber was size-dependent and did not show any consistent shift over the duration of swimming (Figure 5). Small plaice (20–27 cm) spent the majority of their time (87.8 ± 7.9%) struggling to swim in the downstream end and in close proximity to the electrodes. Medium (28–34 cm) and large (35–41 cm) plaice tended to utilize the full length of the chamber, with medium fish spending the majority of their time in the midstream (47.1 ± 6.9%) and large fish in the upstream (53.3 ± 21.5%) areas, respectively. Most of the fish (20 of 23) made contact with the downstream electrodes at some point during the swimming test. The frequency of contact ranged from 0.3 to 5.0 times per minute (mean = 1.6, s.d. = 1.3) and significantly decreased with increasing fish length (F[1,19] = 6.51, p = 0.02).
|
Tailbeat frequency (f) decreased linearly with increasing fish length (Figure 6). Values ranged from 0.86 to 1.97 Hz and can be expressed by: f = –0.034 L + 2.512 using least squares regression (F[1,50] = 83.468, p < 0.001), where L is fish length (cm). The corresponding stride lengths (S) ranged from 15.22 to 34.85 cm and specific stride length (S/L) from 0.52 to 0.85 L.
|
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Hemmings (1969) was the first to provide qualitative descriptions for the different behaviour patterns that flatfish may employ in response to the sweep of a bottom-trawl. Once disturbed from the seafloor, they may choose to (i) swim slower than the speed of the advancing sweep, in which case they would be overtaken and escape from the gear, (ii) swim continuously at the same speed as the sweep, keeping station a certain distance ahead of the stimulus, or (iii) swim at a speed greater than the sweep for a period of time and then slow down or settle onto the seafloor. In situ observations of flatfish in response to the sweeps of a Danish seine (Hemmings, 1969, 1973; Wardle, 1983) and bottom-trawl (High, 1969; Harden Jones et al., 1977; Main and Sangster, 1981) have confirmed the use of all three responses. However, functional explanations for these different behaviours are completely lacking or unknown. The results from the present study indicate that fish length plays an important role. Smaller plaice were clearly unable to swim using cruising alone and were forced to use periodic "kicking" in order to generate sufficient thrust. They also spent the greatest percentage of time in the downstream end of the swimming chamber, made contact with the electrodes often, and exhibited higher tailbeat frequencies compared to their larger conspecifics. These observations are consistent with the length-dependent swimming capability of flatfish (Hashimoto et al., 1996; Winger et al., 1999) and teleost fish in general (see review by Webb, 1994a). If they are representative of field conditions we would predict the herding of flatfish by trawl sweeps to be highly size-selective. Smaller fish would be expected to struggle while swimming, utilize the kick-swim gait, exhibit high tailbeat frequencies, maintain a position in close proximity to the herding stimuli, and make frequent contact with the sweep or sand cloud. As fish length increases, swimming capability improves. Larger fish would be expected to use the cruising gait more frequently, exhibit lower tailbeat frequencies, and have the ability to match or exceed the speed of an advancing sweep.
Those flatfish that swim faster than the speed of an advancing sweep will increase the distance between them and the sweep up to a point at which they are likely to slow down or settle onto the seafloor. The speed and distance swum will determine the future point along the sweep that the fish are likely to encounter (Figure 1). As Hemmings (1969) pointed out, regardless of whether a fish swims fast or slow, settles or swims continuously, the distance that it would have to swim in order to reach the trawl path remains the same, provided it does not change its swimming trajectory. However, because each physical encounter with the sweep entails a certain probability of escape (Foster et al., 1981; Fuwa et al., 1988), the type of behaviour chosen will have a significant effect on the probability of capture. In this study, we found that the decision to settle, frequency of settling, and distance swum between settles, were all independent of fish length for the size range tested. This does not support the hypothesis put forward by Tanaka et al. (1991) that settling behaviour may be size-dependent. Functional explanations for this behaviour are therefore unclear. Given that only 65% of the plaice demonstrated settling and that the distances swum were highly variable (1.2–50.3 m), it appears likely that some additional factor(s) are responsible for this unexplained variation. Possible factors may be fish condition, visual acuity, and experience. Factors not investigated in this study but that are suspected to affect the frequency of settling include water temperature (He, 2003) and the cost of lift-off (Joaquim et al., 2004).
Size-related differences in swimming kinematics, particularly tailbeat frequency, are easily observed among fish swimming in the trawl mouth (Wardle, 1993). Large fish are seen cruising at low tailbeat frequencies while smaller fish struggle at higher frequencies in order to maintain position ahead of the gear. This same scale-effect is expected to occur at lower swimming speeds during herding by trawl sweeps. Swimming at a fixed speed of 0.3 m s–1, tailbeat frequency (f) in this study decreased from an average of 1.7 Hz for a 23 cm plaice to 1.0 Hz for a 44 cm plaice. This drop in f with increasing fish length corresponds to a near doubling of the stride length (S) over this size range. Specific stride lengths (S/L) for American plaice are larger than those previously reported for North Sea plaice (Pleuronectes platessa) (Webb, 2002), but are consistent with the range of most teleost species (Videler and Wardle, 1991).
In summary, it is generally accepted that the herding of fish by trawl sweeps is size-dependent (Wardle, 1983; Engås and West, 1987; Engås and Godø, 1989; Isaksen and Valdemarsen, 1994; Winger et al., 1999; Glass, 2000; Somerton and Munro, 2001), despite the lack of direct empirical observations. The results from this study indicate that fish length affects gait use and swimming kinematics in flatfish. These observations, together with the finding that endurance is size-dependent (Winger et al., 1999), support the hypothesis that trawl herding is size-selective. Quantitative examination of in situ herding behaviour under different environmental conditions and for different trawl riggings is necessary to test this prediction.
|
Acknowledgements |
|---|
We thank Kurt Gamperl, Bill Montevecchi, Joanne Morgan, Dave Scruton, and Pierce Bailey for helpful discussion and comments on the manuscript. This project was funded through the Department of Fisheries and Oceans ADM Strategic Science Project No. 9040 and by the Marine Institute of Memorial University of Newfoundland. Financial support for P.D.W. was provided by the Natural Sciences and Engineering Research Council of Canada.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Alexander R.M. (1989) Optimization and gaits in the locomotion of vertebrates. Physiological Reviews 69:1199–1227.
Dickson W. (1993) Estimation of the capture efficiency of trawl gear. I. Development of a theoretical model. Fisheries Research 16:239–253.[Medline]
Engås A. and Godø O.R. (1989) The effect of different sweep lengths on the length composition of bottom-sampling trawl catches. Journal du Conseil International pour l'Exploration de la Mer 45:263–268.
Engås A. and West C.W. (1987) Trawl performance during the Barents Sea cod and haddock survey: potential sources of gear-related sampling bias. Fisheries Research 5:279–286.[CrossRef][Web of Science]
Fuwa S. (1989) Fish herding model by ground ropes considering reaction of fish. Nippon Suisan Gakkaishi 55:1767–1771.
Fuwa S., Sato O., Nashimoto K., Higo N. (1988) Fish herding model by ground rope. Nippon Suisan Gakkaishi 54:1155–1159.
Foster J. J. (1969) The influence of fish behaviour on trawl design with special reference to mathematical interpretations of observations on the swimming speeds of fish and results of C.F. experiments. Proceedings of the FAO Conference on Fish Behaviour in Relation to Fishing Techniques and Tactics19–27 October 1967Bergen FAO Fisheries Reports, 62: 731–773.
Foster J.J., Campbell C.M., Sabin G.C.W. (1981) The fish catching process relevant to trawls. Canadian Special Publication of Fisheries and Aquatic Sciences 58:229–246.
Glass C.W. (2000) Conservation of fish stocks through bycatch reduction: a review. Northeastern Naturalist 7:395–410.[CrossRef]
Gunderson D.R. (1993) Surveys of Fisheries Resources(John Wiley, New York) 248 pp.
Harden Jones F.R., Margetts A.R., Greer Walker G., Arnold G.P. (1977) The efficiency of the Granton otter trawl determined by sector-scanning sonar and acoustic transponding tags. Rapports et Procès-Verbaux des Réunions du Conseil International pour l'Exploration de la Mer 170:45–51.
Hashimoto S., Hiraishi T., Suzuki K., Yamamoto K., Nashimoto K. (1996) Swimming ability of bastard halibut (Paralichthys olivaceus) at the bottom of net cage. Nippon Suisan Gakkaishi 62:12–16.
He P. (2003) Swimming behaviour of winter flounder (Pleuronectes americanus) on natural fishing grounds as observed by an underwater video camera. Fisheries Research 60:507–514.[CrossRef][Web of Science]
Hemmings C. C. (1969) Observations on the behaviour of fish during capture by the Danish seine net, and their relation to herding by trawl bridles. Proceedings of the FAO Conference on Fish Behaviour in Relation to Fishing Techniques and Tactics19–27 October 1967Bergen FAO Fisheries Reports, 62: 645–655.
Hemmings C.C. (1973) Direct observation of the behaviour of fish in relation to fishing gear. Helgoländer Wissenschaftliche Meeresuntersuchungen 24:348–360.[CrossRef][Web of Science]
High W. L. (1969) SCUBA diving, a valuable tool for investigating the behaviour of fish within the influence of fishing gear. Proceedings of the FAO Conference on Fish Behaviour in Relation to Fishing Techniques and Tactics19–27 October 1967Bergen FAO Fisheries Reports, 62: 253–667.
Isaksen B. and Valdemarsen J.W. (1994) Bycatch reduction in trawls by utilizing behaviour differences. In Fernö A. and Olsen S. (Eds.). Marine Fish Behaviour in Capture and Abundance Estimation(Fishing News Books, Oxford) pp. 69–83 221 pp.
Joaquim N., Wagner G. N., Gamperl A. K. (2004) Cardiac function and critical swimming speed of the winter flounder (Pleuronectes americanus) at two temperatures. Comparative Biochemistry and Physiology Part A (In Press).
Main J. and Sangster G.I. (1981) A study of the fish capture process in a bottom trawl by direct observations from a towed underwater vehicle. Scottish Fisheries Research Report 23:1–23.
Ramm D.C. and Xiao Y. (1995) Herding in groundfish and effective pathwidth of trawls. Fisheries Research 24:243–259.[CrossRef][Web of Science]
Somerton D.A. (2003) Bridle efficiency of a survey trawl for flatfish: measuring the length of the bridles in contact with the bottom. Fisheries Research 60:273–279.[CrossRef][Web of Science]
Somerton D.A. and Munro P.T. (2001) Bridle efficiency of a survey trawl for flatfish. Fishery Bulletin 99:641–652.[Web of Science]
Tanaka E., Matuda K., Hirayama N. (1991) A simulation model of gear efficiencies of trawlers for flatfish. Nippon Suisan Gakkaishi 57:1019–1028.
Videler J.J. and Wardle C.S. (1991) Fish swimming stride by stride: speed limits and endurance. Reviews in Fish Biology and Fisheries 1:23–40.[CrossRef]
Wardle C.S. (1983) Fish reaction to towed fishing gears. In MacDonald A.G. and Priede I.G. (Eds.). Experimental Biology at Sea(Academic Press, London) pp. 167–196 414 pp.
Wardle C.S. (1993) Fish behaviour and fishing gear. In Pitcher T.J. (Ed.). Behaviour of Teleost Fishes(Chapman and Hall, London) pp. 609–643 715 pp.
Webb P.W. (1994) Exercise performance of fish. In Jones J.H. (Ed.). Comparative Vertebrate Exercise Physiology: Phyletic Adaptations(Academic Press, San Diego, CA) pp. 1–49 236 pp.
Webb P.W. (1994) The biology of fish swimming. In Maddock L., Bone Q., Rayner J.M.V. (Eds.). Mechanics and Physiology of Animal Swimming(Cambridge University Press, Cambridge) pp. 45–62 250 pp.
Webb P.W. (2002) Kinematics of plaice, Pleuronectes platessa, and cod, Gadus morhua, swimming near the bottom. Journal of Experimental Biology 205:2125–2134.
Winger P.D., He P., Walsh S.J. (1999) Swimming endurance of American plaice (Hippoglossoides platessoides) and its role in fish capture. ICES Journal of Marine Science 56:252–265.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. M. Fraser, S. P. R. Greenstreet, R. J. Fryer, and G. J. Piet Mapping spatial variation in demersal fish species diversity and composition in the North Sea: accounting for species- and size-related catchability in survey trawls ICES J. Mar. Sci., May 1, 2008; 65(4): 531 - 538. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||