Abstract

In the southeastern Pacific, jack mackerel ( Trachurus murphyi , Carangidae) is a heavily exploited pelagic species, and its presence in Chilean waters in autumn and winter is assumed to be mainly due to an inshore feeding migration. Predator–prey relationships are known to depend on the spatial and temporal scale of observation, but they can also be strongly affected by factors such as diel vertical migration. In studying the case of jack mackerel in detail, we used data from three acoustic surveys carried out in central Chile in 1997, 1998, and 1999. In terms of spatial occupation, jack mackerel behaviour is “atypical” behaviour, i.e. more aggregated during the night than during the day. The patterns we observed can be related to their nocturnal active foraging behaviour. Diel feeding behaviour is therefore a key factor in the aggregating behaviour of jack mackerel and its vulnerability to the purse-seine fishery that targets these nocturnal aggregations. This particular fish diel feeding behaviour also affected predator–prey relationships in relation to the spatial scale. Positive correlations at a “small” spatial scale (<7–25 km) were observed during the night when jack mackerel foraged, but not during the day. Finally, we show that prey biomass was lower where jack mackerel were abundant, which could indicate a jack mackerel top–down control on prey communities.

Introduction

Spatial fish behaviour is a key factor for resource management. It affects catchability and availability of animals to fishing gears and sampling tools ( Fréon and Misund, 1999 ). Occupation of space by fish is a complex issue that has to do with vertical and horizontal position, but also with spatial structuring – whether scattered fish or dense schools. These spatial characteristics are assumed to be linked to local and regional fish abundance (i.e. cluster and stock scales), the exploitation rate, and, more generally, to the structure of the ecosystem: presence of competitors, predators, prey, and oceanographic conditions (e.g. Petitgas and Levenez, 1996 ; Fréon and Misund, 1999 ; Muiño et al ., 2003 ).

Jack mackerel ( Trachurus murphyi , Carangidae), along with anchovy ( Engraulis ringens , Engraulidae) and sardine ( Sardinops sagax , Clupeidae), constitute one of the most important fish resources in the Southeastern Pacific. It has a very large distribution range, from the equator to the austral region of Chile and from the coast of South America to New Zealand and Tasmania ( Serra, 1991 ; Grechina, 1998 ). In the main Chilean fishing ground, the central part of Chile (33°S–40°S), catches ranged from around 1 million to 4 million tonnes during the 1990s ( Arcos et al ., 2001 ). Quiñones et al . (1997) reviewed the main oceanographic, hydrological, and trophic factors that may determine jack mackerel distribution. Jack mackerel is an opportunistic consumer, foraging mainly on macro-zooplankton and micronekton ( Konchina, 1981 ). Off Chile, the main horizontal migration pattern of jack mackerel is an offshore spawning migration in spring, and an inshore feeding migration in autumn and winter ( Serra, 1991 ). Many authors (e.g. Quiñones et al ., 1997 ; Grechina, 1998 ) have therefore assumed that feeding behaviour is one of the main factors determining jack mackerel distribution in Chilean waters.

Predator–prey relationships have been the subject of many studies, many focusing on the importance of the spatial scale of observation (e.g. Rose and Leggett, 1990 ; Horne and Schneider, 1994 ; Bertrand et al ., 2002b ; Davoren et al ., 2002 ; Fauchald and Erikstad, 2002 ; Swartzman et al ., 2002 ), while less is known of the importance of diel vertical behaviour on predator–prey relationships ( Swartzman et al ., 2002 ). Predator avoidance is considered to be one of the main factors influencing diel vertical migration ( Longhurst, 1976 ; Aksnes and Giske, 1990 ). Therefore, the diel vertical behaviour of jack mackerel and its prey may be a key element needed to understand the spatial interactions.

The objective of this work is to study the horizontal and vertical distribution, spatial organization and occupation of jack mackerel, its prey and the spatial relationships. Finally, we examine the hypothesis of Quiñones et al . (1997) , according to which jack mackerel predation can lead to local depletion of prey biomass. For this purpose, we use data from three acoustic surveys from central Chile, where Córdova et al . (1998 , 1999 , 2000) estimated jack mackerel abundance to be 3.5 million, 3.2 million, and 4.1 million tonnes in 1997, 1998, and 1999, respectively.

Material and methods

Data came from three cruises of acoustic evaluation of jack mackerel biomass performed on board the IFOP RV “Abate Molina” in austral autumn–winter: 5 May to 17 June 1997, 3 June to 20 July 1998, and 15 May to 30 June 1999 ( Córdova et al ., 1998 , 1999 , 2000 ). The study area is the main jack mackerel fishing ground ( Figure 1 ) in central Chile (32°S–40°S).

Figure 1

Upper left, study area; graphic view of prey (3-D colour plot) and jack mackerel (surface grey plot) distribution in 1997 (a), 1998 (c), and 1999 (e). Representation of higher densities of jack mackerel (in blue) and index of prey biomass (in red) in 1997 (b), 1998 (d), and 1999 (f).

Figure 1

Upper left, study area; graphic view of prey (3-D colour plot) and jack mackerel (surface grey plot) distribution in 1997 (a), 1998 (c), and 1999 (e). Representation of higher densities of jack mackerel (in blue) and index of prey biomass (in red) in 1997 (b), 1998 (d), and 1999 (f).

Acoustic data were collected with a SIMRAD EK500 echosounder with a 38-kHz split-beam hull-mounted transducer (ES38B). The water column was sampled to a depth of 500 m. The on-axis and off-axis calibration was performed using a 60-mm copper sphere and in accordance with a standard procedure ( Foote et al ., 1987 ).

Survey design consisted of parallel transects running from 5 nautical miles (9.3 km) to 200 nautical miles (370 km) offshore for even transects and to 100 nautical miles (185 km) for odd transects. Inter-transect distance was 20 nautical miles (37 km) up to 100 nautical miles from the coast, and 40 nautical miles (74 km) to between 101 nautical miles and 200 nautical miles from the coast. Nautical area scattering coefficient, s A ( MacLennan et al ., 2002 ), was integrated in 0.5 nautical miles (926 m) elementary sampling distance units (ESDU). A −65 dB integration threshold was applied to s A . In each ESDU, acoustic energy was available in four layers in 1997 (3–25 m; 25–100 m; 100–200 m; 200–500 m), and in seven layers (3–25 m; 25–50 m; 50–100 m; 100–200 m; 200–300 m; 300–400 m; 400–500 m) in 1998 and 1999. For the main statistical analyses, nautical area scattering coefficient (s A ) was normalized using a log(x+1) transformation.

Following previous work ( Bertrand et al ., 1999 , 2002a ), we assume that acoustic energy not assigned to a fish resource is an indicator of the abundance of jack mackerel prey. For each layer, in each ESDU the acoustic energy was partitioned in “resource” and “index of prey biomass”. In central Chile, the main pelagic fish resources are: jack mackerel, common sardine ( Clupea bentinki ), and anchovy. To improve species determination, acoustic echo traces were sampled by pelagic trawling. A total of 43, 29, and 36 pelagic trawl samples were taken in 1997, 1998, and 1999, respectively. Acoustic energy corresponding to resource species other than jack mackerel was not considered in this study.

Since diel periodicity is a structural parameter in the pelagic ecosystem (e.g. Longhurst, 1976 ), ESDUs were classified according to time. According to the season and the latitude, day was defined from 09:00 to 17:00, dusk from 17:00 to 19:00, night from 19:00 to 07:00, and dawn from 07:00 to 09:00 (local time). ESDUs were also classified into three ecological domains according to the topography of each transect: (i) shelf (seabed depth down to 180 m), (ii) shelf break (seabed depth between 180 m and 800 m), and (iii) offshore (seabed depth deeper than 800 m).

A morphological coding of resource and prey echograms was undertaken using the method proposed by Petitgas and Levenez (1996) . Four echo types close to those defined by Reid (2000) were defined for fish: (i) school, (ii) layer, (iii) mixed layer, i.e. no continuous layer, and (iv) scattered. Vertical position of each fish echo trace was measured. In the case of the index of prey biomass, five echo types close to those defined by Bertrand et al . (2002a) were used: (i) large dense patch, (ii) “classic” scattering layer, (iii) small dense patch, (iv) dispersed, and (v) thin scattering layer. “Prey” echo types were characterized for each depth layer of each ESDU. ANOVAs were used to compare vertical distribution of fish, mean s A of jack mackerel, and mean s A of the index of prey biomass according to diel periods, ecological domains, and type of echo traces.

We defined an index of surface occupation (ISO) for jack mackerel as the percentage of ESDU with fish (s A > 0) in relation to the total number of ESDUs. We also calculated the mean nautical area scattering coefficient, s A , calculated on ESDU where fish were present (s A +), which is an index of fish density.

We fitted linear regressions between predator and prey densities to obtain indices of spatial correlation. Computation was performed within transect by one ESDU step for a measurement scale of 1–30 ESDU (i.e. 0.9–27.8 km), then by five ESDU steps to a maximum of 110 ESDUs (i.e. 102 km). Regressions were computed for the presence of jack mackerel for ESDU continuous in space and belonging to the same diel period. To take into account the vertical diel migration of prey communities, the biomass was measured from the surface to 500 m during the day and to 200 m during the night. For graphical analysis, the nautical area scattering coefficient of fish and the index of prey biomass were interpolated to the whole study area using a surface mapping system (SURFER Golden Software, 1995) with the natural neighbour method.

Results

Jack mackerel space occupation

Jack mackerel was significantly more abundant in the oceanic domain than in the shelf or the shelf break (ANOVA, p < 0.01). Its vertical distribution was significantly influenced by the diel period (ANOVA, p < 0.01): mean vertical depth was about 100 m during the day and 20 m during the night ( Table 1 ). Horizontal spatial distribution varied according to the year ( Figure 1 ). In 1997, jack mackerel was distributed throughout the whole survey area, but was less abundant in the southern part. In 1998, it was almost absent north of 36°S and led to our cutting the study area into two zones, north and south of 36°S. In 1999, jack mackerel was mainly observed south of 35°S, although a concentration of this species was also observed in the northern part.

Table 1

Mean vertical depth of jack mackerel echo traces in 1997, 1998, and 1999 according to diel period. The standard deviation of the mean is given in parentheses.

Mean depth (m) Dawn Day Dusk Night 
1997 50 (24) 88 (33) 57 (36) 17 (11) 
1998 71 (33) 112 (27) 99 (30) 22 (17) 
1999 66 (38) 96 (28) 67 (35) 24 (22) 
Mean depth (m) Dawn Day Dusk Night 
1997 50 (24) 88 (33) 57 (36) 17 (11) 
1998 71 (33) 112 (27) 99 (30) 22 (17) 
1999 66 (38) 96 (28) 67 (35) 24 (22) 

Jack mackerel index of surface occupation (ISO) strongly varied according to year, diel period, and ecological domain ( Table 2 ). ISO increased from 1997 to 1999, and was higher offshore than in the shelf and shelf break domains. It was also higher during the day than during the night (with the exception of 1998 when the whole survey area is taken into account). ISO was as high as 60% in the offshore domain of the southern part of the study area in 1998. Mean s A calculated on ESDU where fish were present (s A +) was significantly higher during the night than during the day ( Table 2 , ANOVA, p < 0.01). The pattern was less clear when comparing s A + according to the ecological domains and results were very different from year to year ( Table 2 ).

Table 2

Index of jack mackerel surface occupation (ISO) and mean s A calculated on ESDU where fish were present (s A +) according to the diel period and the ecological domain. In 1998, data are presented for the whole study area and north and south of 36°S. The total number of ESDUs in each case is given in parentheses.

 1997 1998 1998 North 1998 South 1999 
 
 

 

 

 

 
 ISO (%)  s A +  ISO (%)  s A +  ISO (%)  s A +  ISO (%)  s A +  ISO (%)  s A +  
Total 29.6 (5 858) 678 33.1 (6 481) 841 2.9 (2 679) 91 54.4 (3 802) 869 42.7 (6 227) 684 
Day 33.4 (3 089) 516 33.9 (2 950) 599 3.0 (1 299) 102 58.2 (1 651) 619 47.2 (2 991) 626 
Night 29.7 (1 076) 1 183 34.7 (1 933) 1 357 0.3 (660) 52.5 (1 273) 1 361 36.7 (1 386) 1 017 
Dawn 21.1 (877) 842 19.3 (908) 915 0.9 (443) 36.8 (465) 936 37.8 (1 072) 490 
Dusk 24.5 (816) 555 43.3 (690) 455 11.6 (277) 94 64.6 (413) 498 43.1 (778) 653 
Shelf 7.0 (402) 306 16.0 (437) 342 0.0 (153) 24.6 (284) 342 14.0 (321) 958 
Shelf break 11.6 (336) 121 10.7 (336) 1 036 0.0 (106) 15.7 (230) 1 036 29.7 (313) 1 419 
Offshore 32.6 (5 120) 687 35.7 (5 708) 871 3.2 (2 420) 91 59.6 (3 288) 886 45.1 (5 593) 652 
 1997 1998 1998 North 1998 South 1999 
 
 

 

 

 

 
 ISO (%)  s A +  ISO (%)  s A +  ISO (%)  s A +  ISO (%)  s A +  ISO (%)  s A +  
Total 29.6 (5 858) 678 33.1 (6 481) 841 2.9 (2 679) 91 54.4 (3 802) 869 42.7 (6 227) 684 
Day 33.4 (3 089) 516 33.9 (2 950) 599 3.0 (1 299) 102 58.2 (1 651) 619 47.2 (2 991) 626 
Night 29.7 (1 076) 1 183 34.7 (1 933) 1 357 0.3 (660) 52.5 (1 273) 1 361 36.7 (1 386) 1 017 
Dawn 21.1 (877) 842 19.3 (908) 915 0.9 (443) 36.8 (465) 936 37.8 (1 072) 490 
Dusk 24.5 (816) 555 43.3 (690) 455 11.6 (277) 94 64.6 (413) 498 43.1 (778) 653 
Shelf 7.0 (402) 306 16.0 (437) 342 0.0 (153) 24.6 (284) 342 14.0 (321) 958 
Shelf break 11.6 (336) 121 10.7 (336) 1 036 0.0 (106) 15.7 (230) 1 036 29.7 (313) 1 419 
Offshore 32.6 (5 120) 687 35.7 (5 708) 871 3.2 (2 420) 91 59.6 (3 288) 886 45.1 (5 593) 652 

Aggregating behaviour of jack mackerel varied substantially between years and diel periods ( Table 3 ). School proportion increased from 1997 to 1999. More layers were observed in 1997 and 1998 than in 1999. Proportions of scattered echo traces remained more stable. During the day, scattered echo traces always dominated in number, but they contributed little to the total fish acoustic energy as schools contributed to most of the acoustic returns.

Table 3

Jack mackerel echo trace frequency in number (n) and in acoustic energy (s A ) by year during the day and night.

 1997 1998 1999 
 
 

 

 
 Day Night Day Night Day Night 
 
 

 

 

 

 

 
Echo trace % in n  % of s A % in n  % of s A % in n  % of s A % in n  % of s A % in n  % of s A % in n  % of s A 
School 26.8 44.2 31.7 30.9 23.1 49.8 44.3 70.4 35.3 79.6 48.6 63.2 
Layer 0.7 18.9 3.8 7.0 4.4 26.3 3.3 22.1 0.9 9.8 2.7 10.9 
Mixed layer 5.9 28.8 38.7 57.6 6.6 18.9 1.7 1.3 0.9 3.8 8.2 16.2 
Scattered 66.6 8.6 25.9 4.5 65.9 4.9 50.7 6.2 62.9 6.7 40.4 9.7 
 1997 1998 1999 
 
 

 

 
 Day Night Day Night Day Night 
 
 

 

 

 

 

 
Echo trace % in n  % of s A % in n  % of s A % in n  % of s A % in n  % of s A % in n  % of s A % in n  % of s A 
School 26.8 44.2 31.7 30.9 23.1 49.8 44.3 70.4 35.3 79.6 48.6 63.2 
Layer 0.7 18.9 3.8 7.0 4.4 26.3 3.3 22.1 0.9 9.8 2.7 10.9 
Mixed layer 5.9 28.8 38.7 57.6 6.6 18.9 1.7 1.3 0.9 3.8 8.2 16.2 
Scattered 66.6 8.6 25.9 4.5 65.9 4.9 50.7 6.2 62.9 6.7 40.4 9.7 

Index of prey biomass distribution

The index of prey biomass was always significantly higher in the shelf break than in any other ecological domain ( Table 4 , Figure 1 , ANOVA, p < 0.01). However, it is important to note that on the shelf, and particularly in the shelf break domain, it was difficult to discriminate acoustic energy attributed to fish and “prey” because of the presence of numerous scattered hake or hake patches. Therefore, the index of prey biomass was probably overestimated at the shelf break. Offshore, where the main jack mackerel biomass was distributed, the index of prey biomass was higher in 1999 than in 1997 and 1998. In 1998, we observed that the index of prey biomass was significantly higher (ANOVA, p < 0.01) in the northern part of the study area, where very few jack mackerel were present. In terms of aggregating behaviour, it is important to note that we observed ten times more (3% vs. 0.3%) dense patchy structures in 1999 than in other years. According to the aggregating behaviour, the shape of echo traces, and the result of experimental trawling, we can assume that these patchy structures were composed of mesopelagic fish, which also strongly dominated (in weight) jack mackerel stomach contents ( Córdova et al ., 2000 ). When large prey patches were present they contributed to the major part of the prey density (see the peaks of high prey biomass in Figure 1e ).

Table 4

Mean acoustic energy (s A ) of the index of prey biomass by year according to ecological domain. The standard error of the mean is given in parentheses. In 1998, data are presented for the whole study area and north and south of 36°S.

 1997 1998 1998 North 1998 South 1999 
Shelf 876 (48) 249 (36) 194 (27) 278 (30) 843 (71) 
Shelf break 2 202 (142) 2 331 (41) 1 879 (193) 2 539 (225) 1 276 (118) 
Offshore 346 (6) 239 (4) 290 (5) 202 (6) 493 (9) 
 1997 1998 1998 North 1998 South 1999 
Shelf 876 (48) 249 (36) 194 (27) 278 (30) 843 (71) 
Shelf break 2 202 (142) 2 331 (41) 1 879 (193) 2 539 (225) 1 276 (118) 
Offshore 346 (6) 239 (4) 290 (5) 202 (6) 493 (9) 

Predator–prey relationships

Jack mackerel and prey correlation according to the spatial scale showed a similar pattern in 1997 and 1998 ( Figure 2 ). During the day, the correlation was negative in 1998 or null in 1997 up to a scale of 25 km, and then nil up to a scale of about 40 km, after which positive correlations were observed at larger scales. In 1998, the correlation was positive only up to a scale of 65 km. During the night, jack mackerel and prey were positively correlated at a “small” spatial scale (up to 7 km in 1997 and 32 km in 1998). At medium spatial scale, no correlation was observed. Positive correlation was observed from a scale of 45 km in 1997 and a scale of 74 km in 1998. In 1999, the results were quite different from results during the day, with nil correlations observed at small scale. At a scale of 7 km up to almost 90 km, jack mackerel and prey were negatively correlated, after which the correlations were positive at larger scales. A similar pattern was observed during the night, but positive correlations began at 70 km.

Figure 2

Slope of diurnal and nocturnal predator–prey relationships according to the spatial scale (in km) for each year and according to the diel period. (a) 1997 day; (b) 1997 night; (c) 1998 day; (d) 1998 night; (e) 1999 day; and (f) 1999 night. Large red circle = significant correlation (p < 0.05); small black diamond = non-significant correlation (p > 0.05).

Figure 2

Slope of diurnal and nocturnal predator–prey relationships according to the spatial scale (in km) for each year and according to the diel period. (a) 1997 day; (b) 1997 night; (c) 1998 day; (d) 1998 night; (e) 1999 day; and (f) 1999 night. Large red circle = significant correlation (p < 0.05); small black diamond = non-significant correlation (p > 0.05).

The graphical comparison of predator and prey distribution ( Figure 1 ) illustrates that prey were globally less abundant in the main jack mackerel distribution area. This pattern was particularly evident in 1998. The index of prey biomass ( Table 4 ) was significantly lower (ANOVA, p < 0.01) in the offshore southern part of the study area, where most jack mackerel were present, than in the offshore northern part, where almost no jack mackerel were observed. On the contrary, inside the main jack mackerel stock range, peaks of prey biomass were associated with high fish densities most of the time ( Figure 1 ).

Discussion

Jack mackerel is an opportunistic feeder foraging on a large range of prey – from copepods to mesopelagic fish ( Konchina, 1981 ). Its diet consisted mainly of euphausids in 1997 and of mesopelagic fish in 1998 and 1999 ( Córdova et al ., 1998 , 1999 , 2000 ). The presence of jack mackerel in Chilean waters in autumn and winter is assumed to be mainly due to a feeding migration towards the coast. In such contexts, our results illustrated that predator–prey relationships are of major concern when studying jack mackerel distribution and behaviour.

Describing a biological system or functional relationships requires an appropriate choice of scales of observation ( Levin, 1992 ). Rose and Leggett (1990) showed that in the predator–prey relationships (cod–capelin) at large scales (>4–10 km), predator and prey densities were positively correlated. As the scale decreased and approached the aggregation scale (3–5 km), the strength of the correlation decreased and became non-significant. Finally, at scales smaller than this dimension, predator and prey densities were negatively correlated. This theoretical scheme fits well (at another range of scale) with our day results in 1998 and to a lesser extent in 1997 with a higher scale range ( Figure 2 ). During the night we observed a very different pattern. In 1997 and 1998 we observed a positive correlation at smaller spatial scale (<7–25 km according to the case), no correlation at medium spatial scale (8–70 km), then again a positive correlation at a larger spatial scale (>45–70 km).

To interpret such results, it is necessary to analyse the diel behaviour of both jack mackerel and prey communities. In the southeastern Pacific, during the day, jack mackerel is distributed at mid-depth ( Table 1 ) in a layer where very little prey are present. Actually, most of the mesopelagic communities migrate in deeper waters in daytime, out of the reach of jack mackerel, which are assumed to “rest”. During this period, jack mackerel were mainly distributed in the form of small schools and scattered fish with a fairly low level of density (s A +) ( Tables 2 and 3 ). During the night, patterns differed strongly. Jack mackerel distribution, in surface waters, overlapped with one of the mesopelagic communities which had migrated towards the surface. Jack mackerel were then in a phase of active foraging ( Barbieri et al ., 1998 ). Their spatial organization changed with a slight decrease of the index of space occupation, but a strong increase in density within the ESDU ( Tables 2 and 3 ). Such results do not comply with the standard pattern for pelagic fish, which are assumed to disperse at night and to aggregate in schools during the day (e.g. Fréon et al ., 1996 ; Cardinale et al ., 2003 ).

The atypical aggregating behaviour we observed could be related to the nocturnal foraging habits of jack mackerel in the southeastern Pacific. Cardinale et al . (2003) showed that a high degree of aggregation could be related to periods of active feeding behaviour. Most of the small or medium-sized pelagic fish feed at dawn, dusk, or daytime and do not actively forage at night. This is so in the case of most Trachurus species throughout the world (e.g. Pillar and Barange, 1998 ), but not of jack mackerel in the Humboldt Current system – at least not during its feeding migration. Fishermen take advantage of jack mackerel behaviour as they target these winter nocturnal dense aggregations ( Hancock et al ., 1995 ). Diel feeding behaviour is therefore a key factor in jack mackerel aggregative behaviour and its vulnerability to a purse-seine fishery. This has important implications for interpretation of the nocturnal predator–prey relationships we observed ( Figure 2 ). When predators are actively foraging, positive correlation with their prey can be observed even at small scales ( Rose and Leggett, 1990 ). It is what we observed in 1997 and 1998. Therefore, the form of predator–prey relationship is strongly influenced by diel fish behaviour. We have demonstrated that such a relationship should be studied according to the spatial scale, but also in relation to the fish vertical migration and feeding behaviour.

The form of the scale-dependent predator–prey relationships was different in 1999, and could be related to the higher prey availability in the oceanic domain in 1999 ( Table 4 ). The 1999 survey was performed during a La Niña period, while in 1997 it was performed at the beginning of the strong 1997–1998 El Niño, and at the end of this event in 1998. The higher prey availability and number of dense prey patches probably affected the form of predator–prey relationship ( Swartzman et al ., 1999 ; Bertrand et al . 2002b ). When dense patches of prey were present in the main zone of jack mackerel distribution, high densities of fish were located close to these patches ( Figure 1 ). However, with the simple statistical approach we used, it was difficult to study the local variability of predator–prey relationships related to the presence or absence of prey patches.

The last part of this work concerns the potential impact of jack mackerel foraging on the prey community. Our results are in accordance with the predator depletion hypothesis of Quiñones et al . (1997), as prey biomass was lower where jack mackerel were abundant ( Figure 1 ). The most evident case was observed in 1998. In the southern part of the study area, where some 3 million tonnes of foraging jack mackerel were present in approximately 120 000 km 2 (i.e. about 25 tonnes km −2 ), the index of prey biomass was significantly lower than in the northern part. However, it is difficult to prove such a hypothesis because we only have a snapshot of the distribution of fish and prey in a specific area. We need more data on a relevant spatio-temporal scale to confirm that jack mackerel exert a local top–down control on the prey community. Nevertheless, data from the Chilean purse-seine fishery targeting jack mackerel ( Barría et al ., 1999 ) gave us some additional information. In 1998, for example, the fishing fleet was concentrated in the southern part of the study area in June (the acoustic survey track began in the south at the beginning of June), before slowly migrating north (where more prey were present) in July and August. The impact of the jack mackerel population on the mesopelagic communities was less evident in 1997 than in 1998 and 1999. In 1997, fewer mesopelagic fish were present in the study area and jack mackerel fed mainly on euphausids. With the acoustic setting used for the present study, euphausids were probably underestimated and our index of prey biomass may have been biased in 1997.

Finally, we can assume that when a very high biomass of foraging fish is present, the classic concept of higher fish abundance in rich prey areas ( Bertrand et al ., 2002b ) could be brought into question because of the potential impact of predator foraging on prey communities. We can therefore expect a negative relationship between predator and prey when working at the scale of the fish stock range.

We thank the council of the “Fondo de Investigación Pesquera” of Chile, who authorized the use of data for this work. We are grateful to François Gerlotto and Sophie Bertrand for their useful comments on the manuscript. Stephane Gauthier is warmly thanked for revising the English of the paper. The comments of three anonymous reviewers helped to improve the paper. The work is based on the results of Franco-Chilean ECOS project C97B06 and is a contribution of the Research Unit “ACTIVE” UR061 from IRD.

References

Aksnes
D.L.
Giske
J.
Habitat profitability in pelagic environments
Marine Ecology Progress Series
 , 
1990
, vol. 
64
 (pg. 
209
-
215
)
Arcos
D.F.
Cubillos
L.A.
Núñez
S.P.
The jack mackerel fishery and El Niño 1997–98 effects off Chile
Progress in Oceanography
 , 
2001
, vol. 
49
 (pg. 
597
-
617
)
Barbieri
M. A.
Córdova
J.
Lillo
S.
Evaluación hidroacústica del recurso jurel en la zona centro–sur (fase crucero metodológico)
1998
 
Informe final Fondo de Investigación Pesquera n° 95–07. 113 pp
Barría
P.
Aranis
A.
Mora
S.
Böhm
G.
Catasti
V.
Peña
H.
González
C.
Vejar
F.
Monitoreo de las capturas de jurel en la zona centro–sur. 1999
1999
 
Informe final Fondo de Investigación Pesquera n° 98–01. 102 pp
Bertrand
A.
Bard
F-X.
Josse
E.
Tuna food habits related to the micronekton distribution in French Polynesia
Marine Biology
 , 
2002
, vol. 
149
 (pg. 
1023
-
1037
)
Bertrand
A.
Josse
E.
Bach
P.
Gros
P.
Dagorn
L.
Hydrological and trophic characteristics of tuna habitat: consequences on tuna distribution and longline catchability
Canadian Journal of Fisheries and Aquatic Sciences
 , 
2002
, vol. 
59
 (pg. 
1002
-
1013
)
Bertrand
A.
Le Borgne
R.
Josse
E.
Acoustic characterisation of micronekton distribution in French Polynesia
Marine Ecology Progress Series
 , 
1999
, vol. 
191
 (pg. 
127
-
140
)
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
, vol. 
16
 (pg. 
283
-
292
)
Córdova
J.
Barbieri
M. A.
Lillo
S.
Evaluación hidroacústica de jurel
Evaluación hidroacústica de jurel en la ZEE de Chile
 , 
1998
Santiago, Chile
Instituto de Fomento Pesquero
 
161 pp
Córdova
J.
Barbieri
M. A.
Lillo
S.
Evaluación hidroacústica de jurel
Evaluación hidroacústica de jurel en la ZEE de Chile
 , 
1999
Santiago, Chile
Instituto de Fomento Pesquero
 
200 pp
Córdova
J.
Barbieri
M. A.
Lillo
S.
Evaluación acústica de jurel
Evaluación hidroacústica de jurel en la ZEE de Chile
 , 
2000
Santiago, Chile
Instituto de Fomento Pesquero
 
200 pp
Davoren
G.K.
Montevecchi
W.A.
Anderson
J.T.
Scale-dependent associations of predators and prey: constraints imposed by flightlessness of common murres
Marine Ecology Progress Series
 , 
2002
, vol. 
245
 (pg. 
259
-
272
)
Fauchald
P.
Erikstad
K.E.
Scale-dependent predator–prey interactions: the aggregative response of seabirds to prey under variable prey abundance and patchiness
Marine Ecology Progress Series
 , 
2002
, vol. 
231
 (pg. 
279
-
291
)
Foote
K. G.
Knudsen
H. P.
Vestnes
D. N.
MacLennan
D. N.
Simmonds
E. J.
Calibration of acoustic instruments for fish density estimation: a practical guide
1987
 
ICES Cooperative Research Report, No. 144. 69 pp
Fréon
P.
Gerlotto
F.
Soria
M.
Diel variability of school structure with special reference to transition periods
ICES Journal of Marine Science
 , 
1996
, vol. 
53
 (pg. 
459
-
464
)
Fréon
P.
Misund
O.A.
Dynamics of pelagic fish distribution and behaviour: effects on fisheries and stock assessment
 , 
1999
London
Blackwell Science
 
348 pp
Grechina
A.
Arcos
D.
Historia de investigaciones y aspectos básicos de la ecología del jurel Trachurus symmetricus murphyi (Nichols) en alta mar del pacífico Sur
Biología y ecología del jurel en aguas Chilenas
 , 
1998
Concepción, Chile
Instituto de Investigación Pesquera
(pg. 
11
-
34
212 pp
Hancock
J.
Hart
J.B.
Antezana
T.
Searching behaviour and catch of horse mackerel ( Trachurus murphyi ) by industrial purse-seiners off south-central Chile
ICES Journal of Marine Science
 , 
1995
, vol. 
52
 (pg. 
991
-
1004
)
Horne
J.K.
Schneider
D.C.
Lack of spatial coherence of predators with preys: a bioenergetic explanation for Atlantic cod feeding on caplin
Journal of Fish Biology
 , 
1994
, vol. 
45
 (pg. 
191
-
207
)
Konchina
Y.V.
The Peruvian jack mackerel, Trachurus symmetricus murphyi , a facultative predator in the coastal upwelling ecosystem
Journal of Ichthyology
 , 
1981
, vol. 
21
 (pg. 
46
-
59
)
Levin
S.A.
The problem of pattern and scale in ecology
Ecology
 , 
1992
, vol. 
73
 (pg. 
1943
-
1967
)
Longhurst
A.R.
Cushing
D.H.
Walsh
J.J.
Vertical migration
The Ecology of the Seas
 , 
1976
London
Blackwell
(pg. 
117
-
137
467 pp
MacLennan
D.N.
Fernandes
P.G.
Dalen
J.
A consistent approach to definitions and symbols in fisheries acoustics
ICES Journal of Marine Science
 , 
2002
, vol. 
59
 (pg. 
365
-
369
)
Muiño
R.
Carrera
P.
Petitgas
P.
Beare
D.J.
Georgakarakos
S.
Haralambous
J.
Iglesias
M.
Liorzou
B.
Massé
J.
Reid
D.G.
Consistency in the correlation of school parameters across years and stocks
ICES Journal of Marine Science
 , 
2003
, vol. 
60
 (pg. 
164
-
175
)
Petitgas
P.
Levenez
J.J.
Spatial organisation of pelagic fish: echogram structure, spatio-temporal condition, and biomass in Senegalese waters
ICES Journal of Marine Science
 , 
1996
, vol. 
53
 (pg. 
147
-
153
)
Pillar
S.C.
Barange
M.
Feeding habits, daily ration and vertical migration of the cape horse mackerel off South Africa
South African Journal of Marine Science
 , 
1998
, vol. 
19
 (pg. 
263
-
274
)
Quiñones
R.
Serra
R.
Nuñez
S.
Arancibia
H.
Córdova
J.
Bustos
F.
Tarifeño
E.
Relación espacial entre el jurel ( Trachurus symmetricus murphyi ) y sus presas en la zona centro-sur de Chile
Gestión de sistemas oceanograficos del Pacifico oriental
 , 
1997
(pg. 
187
-
202
Comisión Oceanográfica Intergubernamental de la UNESCO, IOC/INF–1046. 432 pp
Reid
D.
Report on echo trace classification
2000
 
ICES Cooperative Research Report, No. 238. 107 pp
Rose
G.A.
Leggett
W.C.
The importance of scale to predator–prey spatial correlation: an example of Atlantic fishes
Ecology
 , 
1990
, vol. 
71
 (pg. 
33
-
43
)
Serra
R.
Important life history aspects of the Chilean jack mackerel, Trachurus symmetricus murphyi
Investigación Pesquera
 , 
1991
, vol. 
36
 (pg. 
67
-
83
)
Swartzman
G.
Brodeur
R.
Napp
J.
Hunt
G.
Demer
D.
Hewitt
R.
Spatial proximity of age-0 walleye pollock ( Theragra chalcogramma ) to zooplankton near the Pribilof Islands, Bering Sea, Alaska
ICES Journal of Marine Science
 , 
1999
, vol. 
56
 (pg. 
545
-
560
)
Swartzman
G.
Napp
J.
Brodeur
R.
Winter
A.
Ciannelli
L.
Spatial patterns of pollock and zooplankton distribution in the Pribilof Islands, Alaska nursery area and their relationship to pollock recruitment
ICES Journal of Marine Science
 , 
2002
, vol. 
59
 (pg. 
1167
-
1186
)