ICES Journal of Marine Science: Journal du Conseil Advance Access originally published online on June 11, 2007
ICES Journal of Marine Science: Journal du Conseil 2007 64(5):963-972; doi:10.1093/icesjms/fsm075
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Diel variation in the vertical distribution and schooling behaviour of sardine (Sardina pilchardus) off Portugal
1 Instituto de Investigação das Pescas e do MAR (INIAP-IPIMAR), Av. Brasília s/n, 1449–006 Lisboa, Portugal
2 FRS Marine Laboratory Aberdeen, PO Box 101, Victoria Road, Aberdeen AB11 9DB, UK
Correspondence to J. Zwolinski: tel: +351 213027000; fax: +351 213025948; e-mail: juan{at}ipimar.pt
Zwolinski, J., Morais, A., Marques, V., Stratoudakis, Y., and Fernandes, P. G. 2007. Diel variation in the vertical distribution and schooling behaviour of sardine (Sardina pilchardus) off Portugal. – ICES Journal of Marine Science, 64: 963–972.Diel patterns in the schooling behaviour and vertical distribution of pelagic fish schools were studied by examining their echotraces from repeated acoustic survey transects at three inshore sites off the Portuguese coast. At two sites, sardine was the dominant pelagic species, and echotrace characteristics of fish schools were similar to those reported in the literature. At the third site, where there was a multispecies pelagic assemblage that included sardine, there was more variability in several of the school descriptors. At all sites, fish schools expanded after sunset, enlarging their cross-sectional area along the horizontal plane and reducing their mean internal acoustic density, while maintaining their overall mean abundance. Downward migration was rapid (within 1 h) after sunset and simultaneous with school expansion. School-like aggregations with total backscattering similar to daytime schools were present throughout the night, although the proportion of small schools and scattered fish appeared to increase at that time. At dawn, sardine rose back up the water column and rapidly reformed into the typical daytime schools. This pattern of diel vertical migration is opposite to that described for most clupeoids worldwide. The implications of this behaviour on abundance estimation by acoustic monitoring surveys for small pelagic fish are discussed.
Keywords: acoustic surveys, clupeoids, diel cycles, schools, vertical migrations
Received 21 August 2006; accepted 9 April 2007; advance access publication 11 June 2007.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
Schools of clupeoid fish expand at night when light drops to a level at which individual fish cannot perceive conspecific movements, and their ability to behave as a school diminishes (Blaxter and Hunter, 1982). However, the expansion is highly variable and, at night, the fish adopt a range of formations, from homogeneous layers of scattered fish to school-like aggregations that are almost as dense as daytime schools (Blaxter and Holliday, 1963; Blaxter and Hunter, 1982; Azzalli et al., 1985; Fréon and Misund, 1999). Diel cycles of light intensity are also associated with vertical migrations of schooling fish that are usually related to feeding (Mackinson et al., 1999; Mowbray, 2002; Cardinale et al., 2003). The direction and the intensity of these migrations may vary owing to changes in the food availability and variety, life stage, and the presence of competitors or predators (Neilson and Perry, 1990). Despite the large number of variables that may control vertical migrations, most adult clupeoids and other small pelagic fish species undergo vertical migrations towards the surface at night. This has been shown to be the case for herring (Clupea harengus) and sprat (Sprattus sprattus) in the Baltic (Nilsson et al., 2003; Orlowski, 2005); Pacific sardine (Sardinops sagax) and northern anchovy (Engraulis mordax) off the Washington and Oregon coast (Krutzikowsky and Emmett, 2005); and capelin (Mallotus villosus) off Newfoundland (Davoren et al., 2006).
In the particular case of sardine (Sardina pilchardus), studies are scarce, and existing information is contradictory with respect to the diel patterns of aggregation and vertical migration. For example, in a directed small-scale study off the Mediterranean coast of Spain, Fréon et al. (1996) detected discernible aggregations of sardine by day and night. This result contrasts with the findings of Iglesias et al. (2003), which were based on four extensive monitoring surveys in roughly the same area, where schools were absent from the acoustic records at night. However, in the latter work, the definition of what the authors considered a school was not given, so perhaps the effect observed was the result of the use of different criteria in school definition. In terms of vertical migration, sardine appear to follow the typical pattern of clupeoids, i.e. migrating towards the surface at night. This is the case in the English Channel (Cushing, 1957) and in the Mediterranean Sea, where it has been proposed that sardine adjust their position to a preferred range of ambient light intensities (Giannoulaki et al., 1999). Contrasting evidence has arisen from a series of acoustic surveys performed throughout a 24-h cycle off the Portuguese coast prior to 1997: sardine aggregations were deeper at night than by day (Dias et al., 1989). More recently, corroborating this evidence, Zwolinski et al. (2006) detected large aggregations of sardine near an intense spawning ground in proximity to the seabed at night.
The combined acoustic surveys of Portugal and Spain in spring have been used systematically in the assessment of the Iberian sardine stock by ICES, providing a relative index of fish abundance at age to tune an assessment model (e.g. ICES, 2006a). After several years of methodological and technical improvements since they started in 1984, acoustic surveys off Portugal have adopted the current format, formulated on the basis of a decision of the ICES Planning Group for Acoustic Surveys in ICES Subareas VIII and IX to limit acoustic surveying to daylight (ICES, 1998). The main reasons given for this were the increased difficulty in attributing echotraces to individual species in the absence of clear school formations at night, and logistic limitations in crew size that prevented routine fishing taking place at night. This decision to change the sampling strategy would have benefited from a thorough study of fish behaviour at night using the precise digital acoustic processing tools that are now available. The aim of the current work is to confirm and describe the atypical diel behaviour of sardine off Portugal, which may be a regional adaptation, and to document how changes in behaviour take place within the transition periods (dawn and dusk). The behaviour is described in relation to the properties of sardine schools derived from analyses of echosounder data obtained across the diel cycle at three sites sampled during two recent acoustic surveys off Portugal. The implications of the behaviour for acoustic survey estimates of sardine and other pelagic fish are discussed.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Data sampling
Data were collected during the spring and autumn 2005 acoustic surveys carried out by the Portuguese Fisheries Research Institute (IPIMAR) off the Portuguese coast and Gulf of Cadiz (Figure 1a). Three study sites were selected, based on results from previous acoustic surveys which indicated that they were likely to be suitable locations for studying the diel cycle and vertical migration of small pelagic fish: two in spring (to the south and west of the survey area) and one in autumn (to the north of the area). Biological samples were taken using a midwater trawl with a vertical opening of
12 m towed at speeds of 3.5–4 knots for an average duration of 30 min. Acoustic samples were taken using a Simrad EK500 vertical echosounder operating at 38 kHz with a hull-mounted 7° x 8° beam width transducer. The echosounder pulse rate was set at 1 s–1, and the pulse duration was 1 ms. Vessel speed was 10 knots (±1 knot). Acoustic data were stored on a PC and post-processed using the echo integration and image analysis software Movies+ (Weill et al., 1993). This delivered digitized acoustic data with a vertical and horizontal resolution of 0.1 and 4.9 m, respectively.
|
The northern and western sites (Figure 1b and c) were located in regions where sardine was the dominant pelagic species, as indicated by the species composition of adjacent trawls (Table 1). Data from these sites were used to describe day/night differences in schooling and vertical position of sardine in a single-species environment in which interaction with other species was minimal. The northern site, consisting of a single transect of 8 nautical miles (hereafter referred to as miles) with depths of 20–100 m, was sampled once during the night of 30 November (from 22:30 to 23:15) followed by a daytime passage the following morning, from 07:45 to 8:30. The western site, consisting of a transect roughly 5 miles long with depths of 15–90 m, was sampled from 19:45 to 6:40 on the night of 29/30 April, with an interruption between 23:30 and 5:30. The relatively short length of the western transect allowed data to be collected along eight repeated passes, providing intense temporal coverage over the daylight transition periods (dusk and dawn). At both sites, the average target strength (TS) of sardine was derived according to the formula recommended by the ICES Planning Group for Acoustic Surveys in ICES Subareas VIII and IX (ICES, 1998): TS = 20 log (L) – 72.6 (dB), where length L (cm) was derived from length compositions taken from the nearest trawl.
|
The southern site (Figure 1d) was sampled during a 24 h period (8/9 May) and consisted of three transects
12 miles long, set 4 miles apart. The seabed consisted of a gentle slope with depth ranging from 20 to 100 m. Sampling there was aimed at describing the diel change in the organization and depth distribution of the pelagic fish assemblage. Two trawls were used to determine the species composition of the pelagic assemblage in water shallower than 55 m. The assemblage comprised a mixture of sardine (Sardina pilchardus), bogue (Boops boops), chub mackerel (Scomber japonicus), and some horse mackerel (Trachurus trachurus) (Table 1). This is a common situation off the southern Portuguese coast (Algarve), which represents a transition region between the sardine-dominated west coast and the multispecies pelagic assemblage of the Gulf of Cadiz, where sardine are often found mixed with many other small pelagic fish species (ICES, 2006b).
Data analysis
School detection and echo integration were carried out on data from all three sites to describe the fish schooling behaviour and patterns in schooling based on individual school characteristics. Schools were defined according to Reid et al. (2000), as images of acoustically unresolved, multiple fish aggregations, with horizontal and vertical continuity between acoustic samples. Schools were detected by image analysis using the shoal-integration tool supplied in Movies+ (Diner et al., 2002), with the following input parameters: minimum school length = 5 m; minimum height = 1 m; and minimum mean school Sv=–55 dB; all set over a minimum acquisition Sv threshold of –60 dB (the value used in routine acoustic estimation; ICES, 2006b). Throughout the text, the term "school" refers to fish aggregations extracted by the Movies+ shoal-integration tool with these input parameters. Once a school had been detected by this image analysis software, the following school descriptors were calculated for each: morphometric, e.g. length, height, area, perimeter, fractal dimension [2 log (perimeter/4)/log (area)]; energetic, e.g. volume backscattering strength, Sv, and school Nautical Area Scattering Coefficient (NASC) relative to 1 mile; and positional, e.g. altitude and depth. These descriptors were calculated as corrected values, to account for the effect of beam width and pulse duration, according to Diner (2001). Because of the large number of small schools identified with these input parameters, an additional minimum NASC threshold was applied at each site to retain those schools contributing 95% of the total school scattering. The threshold used for each site is given in Table 2. The use of this additional threshold ensured that most (95%) of the acoustic energy attributed to pelagic fish was included, while excluding a large number of unrepresentative schools (in terms of biomass) that would otherwise add noise to the analysis of school characteristics. Day/night differences in the descriptors of schools were tested using a Welch t-test for unequal variances, using a confidence level (
) of 0.05. At the western site, where a temporal trend could be followed, especially through the repeated coverage over twilight periods, a trend analysis was also carried out by fitting a locally weighted regression (loess) to the descriptors considered over time of day. Additionally, echo integration was performed by layers for equivalent distance sampling units (EDSU) 1 mile long to evaluate the contribution (percentage) of the backscatter from fish schools to the total backscattered acoustic energy from the whole water column.
|
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
The contribution of school echo energy to the total echo energy from the whole water column was expressed as a simple percentage for each 1-mile EDSU. At the northern site, no significant differences were found between the day (52%) and night (57%) contribution of schools to the total backscatter (t = 0.45, p = 0.66). In contrast, at the western site, 97% of the total scattering was contained in schools during the day, but this dropped to 77% at night (a significant difference, t = 3.3, p = 0.018). At the southern site, daytime schools contributed to 82% of the total NASC, whereas by night their contribution was 64% (marginally not significantly different, t = 2.0, p = 0.054). Moreover, there were no indications of relative diel bias in the acoustic integration of the whole water column at any site: the daytime mean values were 3647, 4279, and 1438 m2 mile–2 for the northern, western, and southern sites, respectively, and the night-time averages for the same sites were 3108, 2792, and 1314 m2 mile–2. No differences were statistically significant. The data suggest that the schools extracted from each 24-h period are representative, in terms of fish biomass, of each site and, as such, can be used for comparing school descriptors.
A statistical summary of school descriptors at each site is given in Table 2. A preliminary analysis of the daytime school descriptors showed that the schools from the northern and western sites were similar in every parameter studied except elongation, and that the joint set of northern and western sites were statistically different from those of the southern site in all descriptors except mean school mean volume backscattering strength (MVBS). This is most likely because of the different species composition between the southern and the other two sites, and adds some support to the conclusion from biological sampling and echogram scrutiny to the definition of the northern and western fish assemblages as being mainly sardine. In light of these differences, diel differences in school descriptors were examined for two populations of schools: one population from the northern and western sites combined and another for the schools from the southern site.
In all, 137 schools (51 night and 86 day) were retained for analysis from the northern and western sites, where sardine was the only, or clearly dominant, pelagic fish species. The contrast between day and night for the relevant school descriptors of this population of sardines is shown in Figure 2. The MVBSs of the schools were significantly higher (t = 11.3, p < 0.01) by day (average of –38.9 dB) than by night (–49.3 dB). This reduction in the internal density of schools at night was accompanied by a significant increase (t = 9.6, p < 0.01) in their cross-sectional area, suggesting that the expansion of schools promoted the lowering of the internal school acoustic density (MVBS) by night. These schools did not show significant differences in mean log(NASC) with time of day, suggesting that the larger and less dense night schools contained, on average, the same quantity of fish as the smaller, denser day schools. School expansion was more intense along a horizontal plane than in a vertical plane, leading to a significant increase in a school's mean elongation (school length/school height) at night (t = 5.2, p < 0.01). The longer, larger, and less dense night schools also showed evidence of a more irregular shape, displaying a significantly higher fractal dimension by night than by day (t = 7.4, p < 0.01). In some cases, this pattern of school expansion produced extremely long fish aggregations, which were very different from the smaller, more regular, and more compact fish schools identified by day (Figure 3a and b). Unusually, for clupeoid fish, the schools migrated vertically downwards at night, where they were constrained to a layer <15 m from the seabed.
|
|
At the southern site, 408 schools (258 day and 150 night) were retained for day/night comparison of the relevant school descriptors (Figure 4). The general pattern of school descriptor dynamics in relation to time of day as described above was maintained despite differences in absolute values of the school descriptors between the southern and the two other sites. There was a significant reduction in school MVBS (t = 8.7, p < 0.01) at night, accompanied by an increase in cross-sectional area (t = 8.4, p < 0.01) similar to the sardine schools described above. School expansion was again stronger on a horizontal plane, leading to a significant change in a school's mean elongation values (t = 2.96, p < 0.01). Mean fractal dimension was also greater at night (t = 6.1, p < 0.01), indicating more irregularity of the schools' contours, and mean log(NASC) values were not statistically different between day and night schools (t = 0.88, p = 0.38), as observed for sardine at the northern and western sites. Finally, in terms of their vertical position, night schools were exclusively constrained to a layer <15 m from the seabed.
|
At the western site, repeated coverage of the relatively short transect every 45 min also allowed for monitoring of sardine school properties during the light transition periods of dusk and dawn. This provided additional information on the diel differences described above. The temporal variation of school MVBS, area, and minimum school altitude is presented in Figure 5. The transect sampled around sunset shows high MVBS, mostly small school areas, and schools clearly present throughout the water column, revealing typical daytime behaviour. Within the next 30 min, MVBS started to drop, school area to increase, and schools to descend, a process that can be inferred from the sequence of Figure 3c (before descent and expansion of the schools) and 3d (at the time of descent). The negative trends of MVBS and altitude continued until the nautical twilight (
1 h after sunset), then bottomed out. In the early hours of the next morning, MVBS and altitude were similar to the low values measured the previous night, implying some form of stability over the night-time period.
|
During the dawn's nautical twilight, i.e. the period when the sun is between 6° and 12° below the horizon, the values of MVBS and altitude were not significantly different from the mean night-time values. However, within the civil twilight, i.e. when the sun is between 6° and 0° below the horizon, MVBS rose rapidly to the values of the previous day, and schools became progressively more pelagic and their cross-sectional area reduced. An analogous process is shown in Figure 3e and f within two acoustic transects during the early morning at the southern site. These observations suggest that the day/night differences in school descriptors and vertical positioning in the water column shown in Figures 2 and 4 are the result of the relatively rapid re-organization of fish during twilight, followed by periods of relative stability by day and by night. This view was further corroborated by the data from the 24-h cycle study at the southern site (Figure 6).
|
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
In recent years, the development of high-precision acoustic devices (e.g. digital echosounders, multibeam sonar), acquisition software (e.g. MOVIES+, Echoview), and the refinement of related acoustic theory (e.g. correction of individual school descriptors) have facilitated the study of schooling behaviour in pelagic fish (see Simmonds and MacLennan, 2005, for a review). However, most existing studies focus on herring, for which descriptions of schooling behaviour during spawning (Skaret et al., 2003), feeding (Cardinale et al., 2003), and overwintering (Huse and Korneliussen, 2000) are used to improve understanding of spatial organization and habitat utilization (Mackinson et al., 1999). Most of these studies also showed a pronounced diel pattern in herring schooling behaviour and vertical positioning that may affect the estimation of species abundance using acoustic methods (e.g. Fréon and Misund, 1999; Huse and Korneliussen, 2000; Orlowski, 2005). For sardine (genera Sardina and Sardinops), information on typical schooling behaviour is provided either in single-species studies (Scalabrin et al., 1996; Coetzee, 2000; Muiño et al., 2003a; Castillo and Robotham, 2004), or within the description of multispecies pelagic communities (Petitgas and Levenez, 1996; Barange and Hampton, 1997; Muiño et al., 2003b). Very few studies have described the diel behaviour of schooling in Sardina pilchardus (Fréon et al., 1996; Giannoulaki et al., 1999), and some contradictory reports of aggregation behaviour and vertical distribution patterns have been reported recently (Iglesias et al., 2003; Zwolinski et al., 2006). For this study, repeated acoustic observations off Portugal throughout a diel cycle were used to clarify local day/night differences in sardine schooling and to describe school dynamics during the light transition periods (dusk and dawn).
Echograms taken by day from the western and northern sites showed sardine schools to be distributed over the inner shelf (up to 80 m deep), and containing a wide range of school sizes. These are typical of the bathymetric distribution patterns of sardine off the coast of Portugal (Marques et al., 2005; ICES, 2006b) and of pelagic school size distribution (Fréon et al., 1993; Marchal and Petitgas, 1993). Despite the observed variation in sardine school size, daytime schools maintained a horizontally elongated shape (median length to height ratio of 5:1), and a relatively smooth perimeter (mean fractal dimension of 1.4). These values are consistent with similar measurements made by Scalabrin et al. (1996) for sardine in the Bay of Biscay. This consistency becomes more evident when sardine school shape descriptors are compared with those of the mixed species assemblage at the southern site, where the presence of other species (chub mackerel, bogue, and horse mackerel) resulted in fish schools with greater variability in shape (elongation, fractal dimension) and size (area), in keeping with similar observations by Massé et al. (1996). Although sardine school shape varies significantly over time and in relation to local abundance and demographic structure (Coetzee, 2000; Iglesias et al., 2003; Castillo and Robotham, 2004), intraspecific variations are usually less pronounced than interspecific ones. This may allow for general characterization of school descriptors that are typical for, and sometimes unique to, the species (Haralabous and Georgakarakos, 1996; Scalabrin et al., 1996). Such intraspecific homogeneity may result from species-specific patterns in the spatial organization of fish within schools (Soria et al., 2003), which is a function of fish size (Misund, 1993). Our data support this hypothesis, because the lack of significant difference between the MVBS of the daytime sardine schools from the northern and western sites arose from the same pattern of space occupation by individual fish within the schools: these sardine occupied a volume of 7.4 and 10.4 cubic body lengths at the northern and western sites, respectively, which was not significantly different (t = 1.12, p = 0.26).
During the early hours of darkness, sardine and pelagic fish schools off Portugal expanded, but identifiable aggregations were still clearly discernible on the echograms. The pattern of night schooling was similar to that described by Fréon et al. (1996) for sardine, McMahon and Tash (1979) for shad (Dorosoma petense), and Azzali et al. (1985) for small pelagic fish in the Adriatic Sea, with larger and less compact aggregations at night than by day. School expansion was not omnidirectional, occurring largely along the horizontal plane, as reported for most clupeoids (Blaxter and Hunter, 1982; Soria et al., 2003). This resulted in a significant increase in school elongation that, in extreme cases, was similar to the large night formations reported by Zwolinski et al. (2006). However, for small schools, night expansion may also have promoted disaggregation to a level that excluded the schools from image-analysis detection, explaining the apparent lesser contribution of schools to the total NASC by night. The transition between typical day and night formations was quite rapid, generally within the first hour of darkness (from the onset of civil twilight), similar to the results of Azzali et al. (1985). The expansion ceased thereafter and school descriptors were relatively constant until dawn (Figures 5 and 6). This night stability contradicts the hypothesis that schools expand and disaggregate passively while light levels remain below a certain threshold (Blaxter and Hunter, 1982; Fréon et al., 1996; Nilsson et al., 2003). If that was the case, school MVBS should have shown a gradual decrease from the civil twilight until dawn, instead of the significant drop only during the first hour of darkness. Nevertheless, the different degree of cohesion in night-time aggregations between northern and western sites (inferred from the schools MVBS, not shown), suggest that school expansion is highly variable, and that the attractive forces that maintain fish aggregated at night are of a different nature from those that keep fish aggregated within daytime schools. In the morning, the loose night aggregations persisted until dawn, after which they rapidly resumed the typical daytime school shape, similar to that reported by Fréon et al. (1996).
The diel cycle of school aggregation we observed was concurrent with changes in the vertical position of schools. Similar to previous observations off Portugal (Dias et al., 1989; Zwolinski et al., 2006), schools descended towards the seabed during their expansion at dusk, and remained in close contact with the seabed throughout the night. Although the descent may be linked to spawning behaviour of the species—reproduction occurs at dusk, close to the seabed (Ganias et al., 2003; Zwolinski et al., 2006)—similar patterns have been observed off Portugal outside the spawning season (Dias et al., 1989). This pattern of vertical migration is opposite to that reported for sardine in the English Channel (Cushing, 1957) and in the eastern Mediterranean (Giannoulaki et al., 1999). There, as in most clupeoids, feeding migrations are towards the surface at night (Neilson and Perry, 1990; Mackinson et al., 1999; Cardinale et al., 2003). For pelagic fish that feed mainly on large zooplankton by particulate feeding, light above a certain threshold is needed for visual predation (Blaxter and Hunter, 1982). For such species, upward migration at dusk has been identified as a strategy for increasing feeding opportunities at twilight (Blaxter and Holliday, 1963; Cardinale et al., 2003). On the contrary, filter-feeding clupeoids can feed on abundant or small prey even in darkness (Batty et al., 1986). For sardine off the Portuguese Atlantic coast (already proven to be efficient filter-feeders; Garrido et al., 2007), an inversion of the general diel pattern of vertical migration may either be due to the presence of sufficient food close to the seabed or to processes unrelated to feeding. An indication that processes other than feeding may shape diel vertical migrations of pelagic fish is provided by overwintering herring in Norwegian fjords, for which dusk ascent has been reported during a period when fish do not eat (Huse and Korneliussen, 2000).
Physostomous sardine experiencing large vertical excursions might demonstrate changes in TS because of the compression of the swimbladder. At the time of writing, we know of no studies that have considered depth-dependence in sardine TS. Ona's (2003) depth-dependent function for the related physostomous species, herring, gives very much higher values of TS in absolute form than the depth-invariant TS function used on Portuguese surveys (some +6 dB, which would result in a reduction to 25% of abundance). The function used in those surveys is the TS function for herring provided by Degnbol et al. (1985), derived from in situ TS measurements on a range of herring sizes (8–32 cm) at depths of 7–50 m. If one considers the depths at which these in situ measurements were made, and weights the mean depth according to the number of echoes taken at depth, the mean depth on which the relation is based is 29.7 m. Ona's (2003) b20 parameter (–65.4) can then be set to a value (–71.2) whereby fish at a depth of 29.7 m have the same TS as fish at 29.7 m using the TS/length function of Degnbol et al. (1985). One can then begin to appreciate the depth-dependence in TS (without knowing its exact nature, or absolute magnitude, because of the slightly different physiology of the species). The difference in TS using the latter approach between fish at 29 m and fish at 41 m (the average sardine depth by day and by night, respectively) is 0.3 dB. Fish at 41 m, having the lower TS, would, therefore, be only 6% more abundant than fish at 21 m, for the same acoustic density. This is probably not significant enough a change to implement given the uncertainty of the implementation associated with its derivation from another species and the very large difference in TS that Ona's (2003) function delivers. These results suggest that the depth-invariant TS function of Degnbol et al. (1985), derived from fish at 30 m, is a reasonable function to use to average out depth-dependent effects, and that its use is appropriate to calculate numerical quantities (e.g. school packing densities, fish density by surface area) that are comparable between day and night schools.
The diel patterns in sardine aggregation off Portugal described here indicate significant changes in school shape and internal density between day and night. Typical daytime schools were more compact and conspicuous, sharing properties with daytime descriptions of sardine schools from other areas and facilitating scrutiny of echograms collected by day. Such properties become more variable by night owing to an expansion of the schools. This diel contrast was one of the reasons invoked by the ICES Planning Group (ICES, 1998) to abandon acoustic sampling by night, leading in recent years to a considerable increase in the duration of these monitoring surveys off the Iberian Peninsula. The results of the current study, in particular the lack of a difference in school NASC values obtained by day and by night, indicate that detection of schools is possible by night, so further consideration should be given to surveying for the whole 24-h period. Significant gains in precision of abundance estimates could be expected with day and night sampling, because of the increased number of acoustic samples and the more even distribution of acoustic energy across the EDSUs (Fréon et al., 1993; Axenrot et al., 2004). Nevertheless, studies aiming to determine the net result between possible biases should be performed before the use of night-time acoustic samples in sardine abundance estimation. Negative biases would include the systematic disaggregation of small day schools and the increase of sardine within the acoustic dead zone (Ona and Misund, 1996). Positive biases could possibly include less avoidance through a diminished school response (Fréon and Misund, 1999). Notwithstanding these technical issues, the logistical concerns of operating a research vessel for 24 h every day would need to be considered from a cost/benefit perspective.
|
Acknowledgements |
|---|
The work of JZ was funded by the Portuguese Ministry of Science through a PhD grant. AM was funded by the "Programa Nacional de Amostragem Biológica'' PNAB–DCR. The authors are grateful to "Programa PELAGICOS" funded by the Portuguese Ministry of Science and to PNAB–DCR for funding IPIMAR's acoustic surveys. Finally, we thank the two anonymous referees who provided detailed comments that helped to improve the final manuscript.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Axenrot T., Didrikas T., Danielsson C., Hanson S. Diel patterns in pelagic fish behaviour and distribution observed from a stationary, bottom-mounted, and upward-facing transducer. ICES Journal of Marine Science (2004) 61:1100–1104.
Azzali M., Buracchi G., Conti S., Gambetti S., Luna M. Relationships between the forms of pelagic fish distribution and nycthemeral periods. A tentative model of behaviour. Oebalia (1985) 11:471–488.
Barange M., Hampton I. Spatial structure of co-occurring anchovy and sardine populations from acoustic data: implications for survey design. Fisheries Oceanography (1997) 6:94–108.[CrossRef][Web of Science]
Batty R., Blaxter J., Libby D. Herring (Clupea harengus) filter-feeding in the dark. Marine Biology (1986) 91:371–375.[CrossRef]
Blaxter J., Holliday F. The behaviour and physiology of herring and other clupeids. Advances in Marine Biology (1963) 1:261–393.
Blaxter J., Hunter J. The biology of the clupeoid fishes. Advances in Marine Biology (1982) 20:1–223.
Cardinale M., Casini M., Arrhenius F., Håkansson N. Diel spatial distribution and feeding activity of herring Clupea harengus and sprat Sprattus sprattus in the Baltic Sea. Aquatic Living Resources (2003) 16:283–292.[CrossRef][Web of Science]
Castillo J., Robotham H. Spatial structure and geometry of schools of sardine Sardinops sagax in relation to abundance, fishing effort, and catch in northern Chile. ICES Journal of Marine Science (2004) 61:1113–1119.
Coetzee J. Use of a shoal analysis patch system (SHAPES) to characterize sardine schools. Aquatic Living Resources (2000) 13:1–10.[CrossRef][Web of Science]
Cushing D. H. The number of pilchards in the Channel. Fishery Investigations, Ministry of Agriculture Fisheries and Food. Series II (1957) 21:27.
Davoren G., Anderson T., Montevecchi A. Shoal behaviour and maturity relations of spawning capelin Mallotus villosus off Newfoundland: demersal spawning and diel vertical movement patterns. Canadian Journal of Fisheries and Aquatic Sciences (2006) 63:268–284.
Degnbol P., Lassen H., Staehr. In-situ determination of target strength of herring and sprat at 38 and 120 kHz. Dana (1985) 5:45–54.
Dias C., Soares E., Marques V. Acoustic abundance estimations of sardine (Sardina pilchardus, Walb.) off the Portuguese coast. ICES Document CM 1989/H (1989) 52:24.
Diner N. Correction on school geometry and density: approach based on acoustic image simulation. Aquatic Living Resources (2001) 14:211–222.[CrossRef][Web of Science]
Diner N., Marchalot C., Berger L. Echo integration by shoal using Movies+ software. (2002) France: IFREMER. 35.
Fréon P., Gerlotto F., Soria M. Diel variability of school structure with special reference to transition periods. ICES Journal of Marine Science (1996) 53:459–464.
Fréon P., Misund O. Dynamics of Pelagic Fish Distribution and Behaviour: Effects on Fisheries and Stock Assessment. (1999) Oxford: Blackwell Science.
Fréon P., Soria M., Mullon C., Gerlotto F. Diurnal variation in fish density estimate during acoustic surveys in relation to spatial distribution and avoidance reaction. Aquatic Living Resources (1993) 6:221–234.[CrossRef]
Ganias K., Somarakis S., Machias A., Theodorou A. Evaluation of spawning frequency in a Mediterranean sardine population (Sardina pilchardus sardina). Marine Biology (2003) 142:1169–1179.
Garrido S., Marçalo A., Zwolinski J., van der Lingen C. D. Laboratory investigations on the effect of prey size and concentration on the feeding behaviour of Sardina pilchardus. Marine Ecology Progress Series (2007) 330:189–199.[Web of Science]
Giannoulaki M., Machias A., Tsimenides N. Ambient luminance and vertical migration of the sardine Sardina pilchardus. Marine Ecology Progress Series (1999) 178:29–38.[Web of Science]
Haralabous J., Georgakarakos S. Artificial neural networks as a tool for species identification of fish schools. ICES Journal of Marine Science (1996) 53:173–180.
Huse I., Korneliussen R. Diel variation in acoustic density measurements of overwintering herring (Clupea harengus L.). ICES Journal of Marine Science (2000) 57:903–910.
ICES. Report of the Planning Group for Acoustic Surveys in ICES Subareas VIII and IX. (1998) 2:17. ICES Document CM 1998/G.
ICES. Report of the Working Group on the Assessment of Mackerel, Horse Mackerel, Sardine, and Anchovy. (2006a) 36:601. ICES Document CM 2006/ACFM.
ICES. Report of the Working Group on Acoustic and Egg Surveys for Sardine and Anchovy in ICES Areas VIII and IX. (2006b) 01:126. ICES Document CM 2006/LRC.
Iglesias M., Carrera P., Muiño R. Spatio-temporal patterns and morphological characterisation of multispecies pelagic fish schools in the North-Western Mediterranean Sea. Aquatic Living Resources (2003) 16:541–548.[CrossRef][Web of Science]
Krutzikowsky G., Emmett R. Diel differences in surface trawl fish catches off Oregon and Washington. Fisheries Research (2005) 71:365–371.[CrossRef][Web of Science]
Mackinson S., Nøtestad L., Guénette S., Pitcher T., Misund O., Fernö A. Cross-scale observations on distribution and behavioural dynamics of ocean feeding Norwegian spring-spawning herring (Clupea harengus L.). ICES Journal of Marine Science (1999) 56:613–626.
Marchal E., Petitgas P. Precision of acoustic fish abundance estimates: separating the number of schools from the biomass in the schools. Aquatic Living Resources (1993) 6:211–219.[CrossRef]
Marques V., Chaves C., Morais A., Cardador F., Stratoudakis Y. Snipefish (Macroramphosus spp.) distribution and abundance off Portugal (1998–2003). Scientia Marina (2005) 69:563–576.[Web of Science]
Massé J., Koutsikopolous C., Patty W. The structure and spatial distribution of pelagic fish schools in multispecies clusters: an acoustic study. ICES Journal of Marine Science (1996) 53:155–160.
McMahon T., Tash J. The use of chemosenses by threadfin shad Dorosoma petense to detect conspecific, predators and food. Journal of Fish Biology (1979) 14:289–296.[CrossRef][Web of Science]
Misund O. Dynamics of moving masses: variability in packing density, shape, and size among herring, sprat, and saithe schools. ICES Journal of Marine Science (1993) 50:145–160.
Mowbray F. Changes in the vertical distribution of capelin Mallotus villosus off Newfoundland. ICES Journal of Marine Science (2002) 59:942–949.
Muiño R., Carrera P., Iglesias M. The characterization of sardine (Sardina pilchardus Walbaum) schools off the Spanish–Atlantic coast. ICES Journal of Marine Science (2003a) 60:1361–1372.
Muiño R., Carrera P., Petitgas P., Beare D., Georgakarakos S., Haralambous J., Iglesias M., et al. Consistency in the correlation of school parameters across years and stocks. ICES Journal of Marine Science (2003b) 60:164–175.
Nakken O., Dommasnes A. The application of an echo integration system in investigations of the stock strength of the Barents Sea capelin. (1975) 25:20. ICES Document CM 1975/B.
Neilson J., Perry R. Diel vertical migration of marine fishes: an obligate or facultative process? Advances in Marine Biology (1990) 26:115–168.[Web of Science]
Nilsson L., Thygesen U., Lundgren B., Nielsen B., Nielsen J., Beyer J. Vertical migration and dispersion of sprat (Sprattus sprattus) and herring (Clupea harengus) schools at dusk in the Baltic Sea. Aquatic Living Resources (2003) 16:317–324.[CrossRef][Web of Science]
Ona E. An expanded target-strength relationship for herring. ICES Journal of Marine Science (2003) 60:493–499.
Ona E., Misund R. Acoustic sampling and signal processing near the seabed: the deadzone revisited. ICES Journal of Marine Science (1996) 53:677–690.
Orlowski A. Experimental verification of the acoustic characteristics of the clupeoid diel cycle in the Baltic. ICES Journal of Marine Science (2005) 62:1180–1190.
Petitgas P., Levenez J. J. Spatial organization of pelagic fish: echogram structure, spatio-temporal condition, and biomass in Senegalese waters. ICES Journal of Marine Science (1996) 53:147–153.
Reid D., Scalabrin C., Petitgas P., Massé J., Aukland R., Carrera P., Georgakarakos S. Standard protocols for the analysis of school based data from echo sounder surveys. Fisheries Research (2000) 47:125–136.[CrossRef][Web of Science]
Scalabrin C., Diner N., Weill A., Hillion A., Mouchot M-C. Narrowband acoustic identification of monospecific fish shoals. ICES Journal of Marine Science (1996) 53:181–188.
Simmonds E. J., MacLennan D. N. Fisheries Acoustics: Theory and Practice. (2005) 2nd edn. Oxford: Blackwell Science. 437.
Skaret G., Nøtestad L., Ferno A., Johannessen A., Axelsen B. Spawning of herring: day or night, today or tomorrow. Aquatic Living Resources (2003) 16:299–306.[CrossRef][Web of Science]
Soria M., Bahri T., Gerlotto F. Effect of external factors (environment and survey vessel) on fish school characteristics observed by echosounder and multibeam sonar in the Mediterranean Sea. Aquatic Living Resources (2003) 16:145–157.[CrossRef][Web of Science]
Weill A., Scalabrin C., Diner N. Movies-B: an acoustic detection description software. Application to shoal species classification. Aquatic Living Resources (1993) 6:169–296.[CrossRef]
Zwolinski J., Mason E., Oliveira P., Stratoudakis Y. Fine scale distribution of sardine (Sardina pilchardus) eggs and adults during a spawning event. Journal of Sea Research (2006) 56:294–304.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||