Previous Page Table of Contents Next Page


Poster papers (continue)

THEME 2: POPULATION BIOLOGY AND RESOURCE ASSESSMENT

Biological and genetic characteristics of redfish Sebastes mentella of the Irminger sea and adjacent waters

G.G. Novikov1, A.N. Stroganov1, V.N. Shibanov2 and S.P. Melnikov2
1 Department of Biology, Moscow State University
Vorobjovy Gory, 119992, Moscow, Russia
<andrei_str@mail.ru>

2 Knipovich Polar Research Institute of Marine Fisheries and Oceanography (PINRO)
6, Knipovich Street, Murmansk, 183763, Russia

1. INTRODUCTION

Sebastes mentella is the most important species taken by the fishery of three species of redfish dwelling in the North-East Atlantic. Bottom fishery for deep-sea redfish (S. mentella) (shelf redfish) is carried out on the shelf and slope of Iceland, Faroe Islands and East Greenland (ICES Subareas V, VI, XIV). The pelagic fishery for redfish is prosecuted in the Irminger Sea and adjacent areas of the Labrador Sea (ICES Divisions Va, XII, XIVb, NAFO Divisions 1F and 2HJ). In the early 1990s Magnusson (Magnusson 1991, Magnusson and Magnusson 1995) hypothesized the existence in the pelagial of the Irminger Sea of two types of redfish -“oceanic S.mentella”and “pelagic deep-sea S. mentella”. In his opinion, redfish of the first type are distributed at depths less than 500 m, whereas those of the second type live deeper than 500 m. In the last two decades, Russian scientists carried out investigations on the populational structure of redfish of the pelagic Irminger Sea (Pavlov and Shibanov 1984, Pavlov 1992, Shibanov, Melnikov and Pedchenko 1996, Melnikov 1998, Bakay and Melnikov 2001, 2002, Melnikov, Pedchenko and Shibanov 2001, Melnikov and Bakay 2002, Novikov et al. 2002). The identity of the origin and uniformity of the pelagic deep-sea S. mentella stock in the Irminger Sea and adjacent area of the Labrador Sea over the entire range of the vertical distribution of concentrations was revealed.

In recent years, the intensive fishery for pelagic redfish was carried out at depths greater than 500 m north of the Reykjanes Ridge close to areas of the bottom fishery for redfish. Thus, the relation between redfish from the slope of Iceland and those of the adjacent area of the pelagial of the Irminger Sea is important for developing conservation measures and rational exploitation of this stock.

2. MATERIAL AND METHODS

Results of investigations of redfish caught in two areas are presented. The first area is part of the southwestern slope of Iceland (the Reykjanes Ridge, 63° 00'-63° 10'N and 25° 10'-25° 30'W). Samples were selected from catches taken by a bottom trawl at depths between 430 and 760 m. The second area is adjacent to the Reykjanes Ridge and its western slope over the area bounded by 60°30'-63°00'N and 27°00'-33°00'W. Samples were obtained from catches taken by a pelagic trawl at depths between 500 and 900 m.

Ichthyological material on redfish was collected using the methods of the PINRO. The data were obtained from one scientific-research and four scientific-fishing cruises in June–August 2001. The length-age composition, growth, maturation rate and range of feeding of redfish from the two surveyed areas were analyzed.

Differences in rates of maturation of redfish from different areas were tested using a X2test. Genetic investigations were carried out by starch-gel electrophoresis. A unified enzyme systems population and genetic investigations of redfish in Russia and abroad used NADH-dependent malate dehydrogenase malic enzyme) (MEP-2), NAD-dependent malate dehydrogenase (MDH), glucose 6-phosphate isomerase (PGI), isocitrate dehydrogenase (IDH), lactate dehydrogenase (LDG), phosphoglucomutase (PGM-muscle), superoxide dismutase (SOD) and fluorescent esterase (ESTD). This work followed the methods of e.g. Dushchenko (1986) and Johansen et al. (1996, 1997). For comparison of our results all names of loci and alleles follow (Johansen et al. (1997, pp. 9–11). Distributions were tested for a Hardy-Weinberg distribution. Specimens were not attributed to “oceanic”or ”deep-water”types by morphological analysis.

TABLE 1
Volume of the analyzed ichthyological material

Volume of measurements Number  
Individuals measured23 358 
Maturation analyses13 895 
Feeding analyses9 054 
Individuals aged1 184 
Enzymatic systems analyzed640 

3. RESULTS

3.1 Length-age composition

Length of redfish males and females in the pelagial of the Irminger Sea in 2001 measured from samples at depths of more than 500 m varied from 27 to 49 cm with males between 39–44 cm and females between 42–45 cm. Males represented 53.1 percent of fish measured. Length of redfish on the slope of Iceland fluctuated within the same range as in the pelagial, however, smaller fish predominated in catches: males were 36–41 cm long; females were 35–43 cm long. The sex ratio was approximately equal. The portion of redfish less than 35 cm on the slope was 2.5–3 times higher than for fish in the pelagic sea (Figure 1). Redfish age in the pelagial of the Irminger Sea and on the slope of Iceland varied from 8 to 24 years. Individuals in the pelagial ranged in the age from 17 to 20 years and on the slope, from 14 to 19 years. The age range of redfish in both areas was 16–24 years, however, the portion of fish of age 16 on the slope was larger than in the pelagial (Figure 2).

The result of analysis of length-age composition of redfish showed an absence of young fish in both areas in the investigated range of depths. Thus, redfish stocks in the pelagial of the Irminger Sea and on the slope of Iceland must recruit from other areas.

3.2 Rate of maturation

In the pelagial the mature males were found from 28 cm length, and females from 29 cm. The 50 percent level of fish maturity in males was at length 32 cm and age 10; and for females 33 cm and age 10. Full maturation (100 percent) of males happened at length 38 cm and age 16; and in females, at 41 cm and age 17 (Figures 3A, 4A).On the slope, mature males were first registered at length 34 cm and females at 38 cm. The length of 50 percent maturity in males was reached at length 37 cm and age 15; for females at 41 cm and age 17. All males became mature at length 43 cm and age 20; and females at 47 cm and age 22 (Figures 3B, 4B). Highly significant differences were revealed between rates of maturation of redfish in the pelagial of the Irminger Sea and on the slope of Iceland during the statistical analysis.

A comparative analysis of maturity corroborated the lower rate of maturation of redfish on the slope compared with that of the pelagial of the Irminger Sea (Zakharov 1969, Melnikov 1998). Maturation of fish dwelling on the slope takes place at a greater length and later age. In the pelagial the portion of mature redfish constituted 59 percent for males under 33 cm and 82 percent for females under 37 cm. No mature fish of these length groups were registered on the slope. At the same time in the pelagial, immature males longer than 38 cm and females longer than 41 cm were completely absent. On the slope, immature fish of these lengths constituted 15 percent of males and 27 percent of females. Results of analysis of maturation rate by age corroborated the results of the ratio between mature and immature individuals by length groups (Figure 4). Consequently, the analysis of maturation rate of redfish did not reveal an interrelation, and exchange, between fish stocks in the pelagial of the Irminger Sea and on the slope of Iceland.

FIGURE 1
Length composition of redfish in pelagial of the Irminger Sea (A) and on the southwestern slope of Iceland (B)

FIGURE 1
FIGURE 1

FIGURE 2
Age composition of redfish in pelagial of the Irminger Sea (A) and on the southwestern slope of Iceland (B)

FIGURE 2
FIGURE 2

FIGURE 3
Maturation of males and females of redfish of different length groups in pelagial of the Irminger Sea (A) and on the southwestern slope of Iceland (B)

FIGURE 3
FIGURE 3

FIGURE 4
Maturation of males and females of redfish of different age groups in pelagial of the Irminger Sea (A) and on the southwestern slope of Iceland (B)

FIGURE 4
FIGURE 4

3.3 Growth in length and weight

The comparative analysis of fish of the same age and mean length revealed similar linear growth of redfish males and females in the pelagial of the Irminger Sea and on the slope of Iceland (Table 3). However, the mean weights of males and females on the slope were 1.8–19.4 percent higher than for fish dwelling in the pelagial. The largest deviation in weight (7.0–19.4 percent) was found in redfish at age 15. An analysis of maturation rate showed (Figure 4) that all redfish at this age were mature in the pelagial, whereas on the slope the proportion of immature fish exceeded 50 percent. The immature fish of many species have higher rates of weight growth than mature ones as big part of assimilated food is used for somatic growth rather than for reproductive products (Shatunovsky et al. 1975, Shatunovsky 1980). Therefore, the higher rate of weight growth of redfish on the slope compared to the pelagial can be explained by the greater fraction of immature fish at age 15 or less.

3.4 Feeding

The main food of redfish in the pelagial of the Irminger Sea was mesopelagic fish (Myctophidae, Paralepidae), shrimp and cephalopods (Gonatus fabricii). On the slope off Iceland fish and young squids were eaten 2–3 times less than in the pelagic sea. The portion of Euphausiacea in the food range reached 50 percent (Table 2). Mean stomach fullness of fish in the pelagial was 0.2 and on the slope, 0.8.

TABLE 2
Diet (%) of Sebastes mentella in the pelagial of the Irminger Sea and on the southwestern slope of Iceland in 2001

Food Item Pelagial of the Irminger Sea (%) S.-W. slope of Iceland (%)
Calanus0.23.5
Euphausiids5.150.0
Shrimp22.023.3
Squid18.39.3
Fish38.08.1
Other species16.45.8

In summer, there were peculiarities in the distribution of food plankton in the two areas. In the pelagial of the Irminger Sea, variation in the distribution of food organisms by depth was clearly observed. Planktonic crustaceans were concentrated predominantly in the upper 500 m layer. In the lower 500 m layer of the pelagial, the main fishery concentrations of redfish coincided with the predominantly mesopelagic fish and macroplankton food base (Pavlov 1992). In the area of the Icelandic continental slope the dominant food organisms were euphausids and shrimp (Zakharov 1969). One of the reasons for the differences in the distribution of plankton prey is the characteristics of the hydrological regime in the Irminger Sea and on the slope of Iceland.

3.5 Distribution of redfish in the pelagic Irminger Sea and on the south-western slope of Iceland as indicated by protein type

The distribution of frequencies of alleles of locus MEP-2* in the samples studied reflects the mosaic of distribution of samples with different frequencies at different depths and in different areas (Tables 4, 5 and 6). Frequency of allele “100”of locus MEP-2* in redfish from samples taken at depths of 292 m and 490 m is 0.570 and 0.750 respectively. In samples from 500 to 800 m depths it varied from 0.357 to 1.000. We propose that such amplitudes prove the integrity of the redfish stock from the pelagic Irminger Sea throughout its entire vertical distribution (Bakay and Melnikov 2002).

TABLE 3
Mean length (cm) of the Sebastes mentella by age groups in the pelagial of the Irminger Sea and on the southwestern slope of Iceland in 2001

Age (years) Pelagial of the Irminger Sea S.-W. slope of Iceland
Males Females Males Females
6-22.0--
724.825.5--
827.227.5--
928.028.828.0-
1029.029.431.030.0
1130.530.531.231.2
1233.032.333.332.7
1334.034.034.234.2
1435.035.335.735.5
1536.837.136.936.9
1639.238.338.438.5
1740.239.840.040.4
1841.842.041.842.2
1943.04.3043.243.0
2044.444.444.245.0
2145.745.745.346.0
2246.646.646.546.6
2347.247.8-48.0
2448.048.5--
2549049.4--
26-50.5--

The frequency distribution of the allele MEP-2*100 was analysed for samples from two areas (a) pelagic trawls in the Irminger Sea (Tables 5 and 6) and (b) bottom trawls on the southwestern slope of Iceland. The results (Table 6) showed a similar distribution of frequencies, e.g. the frequency of allele MEP-2*100 varied over a wide range in samples from each area (Tables 5 and 6). A similar distribution of frequencies of allele MEP-2*100 was described by Johansen, Danielsdottir, Meland and Nevdal (1997). Frequencies changed from 0.60 to 0.83 and from 0.42 to 0.61 for the different redfish types.

An attempt was made to reveal differences observed in the distribution of a rare allele 20 at the locus MDH*. The distribution in some areas of alleles of the locus MDH* was characterized by peculiarity connected with a low level of polymorphism (Tables 7 and 8). It was shown that the presence of the rare allele MDH*20 could change in samples over different years and differed in some measure in the fish from the pelagic zone of the Irminger Sea and on the southwestern slope of Iceland (percentage of individuals-carriers of the rare allele in the united samples). Allele MDH*20 was found in 5 percent of the individuals from the pelagial of the Irminger Sea in the year of 2000 and in 2001. Only 2.8 percent of such fish were detected. In samples taken by bottom trawl on the southwestern slope of Iceland the portion of individuals of redfish with allele MDH*20 was 1.7 percent.

The close ratio between individuals with the rare allele 20, locus MDH*, probably indicates a genetic unity of redfish stock in the two investigated areas, but one can also assume some differentiation of redfish from the southwestern slope of Iceland caused by the specific hydrological conditions of the area. In connection with the higher percentage of individuals with rare allele in the pelagial of the Irminger Sea, one can expect the migration of genetic material from the pelagial of the Irminger Sea to the southwestern slope of Iceland. The other loci investigated were monomorphic (PGM, IDH, LDH, ESTD, SOD). These demonstrated a low measure of polymorphism (PGI) in the studied areas.

TABLE 4
Mean weight of the Sebastes mentella by age groups in the pelagial of the Irminger Sea and on the southwestern slope of Iceland in 2001 (grams)

Age (years) Pelagial of the Irminger Sea S.-W. slope of Iceland
Males Females Males Females
6-136.0--
7180.0211.5--
8235.2249.0--
9257.7307.5300.0-
10298.5360.6319.0386.0
11360.0385.0429.9433.9
12433.3422.3497.6499.8
13476.6473.4542.7556.1
14520.0539.0593.2627.4
15616.0673.6704.3670.4
16785.7762.7799.0750.0
17819.9828.6862.2863.1
18928.7944.7961.4987.4
191030.71031.71046.51043.9
201075.71101.81076.71203.1
211150.91132.51120.81159.2
221203.11233.21262.51308.3
231292.91285.9-1230.0
241304.71348.2--
251585.01494.8--
26-1585.0--

TABLE 5
Frequencies of alleles of MEP-2* locus in samples of Sebastes mentella in the pelagial of the Irminger Sea in 2000

Sample No. Latitude (N) Longitude (W) Trawling depth range (m) Volume of sample Alleles of the MEP-2* locus
Min max 60 100
161.3028.10740835100.2000.800
261.2028.10705770100.3500.650
361.3028.20640740100.3000.700
461.2028.10760780100.1000.900
561.2028.10740760100.3000.700
661.3028.10760770100.2000.800
761.2028.10760760100.3000.700
861.4029.10780800100.0001.000
961.2028.10650720100.1500.850
1061.3028.20720750100.1500.850
1161.1428.10580600100.3000.700
1261.1028.10540700100.1000.900

TABLE 6
Frequencies of alleles of MEP-2* locus in samples of Sebastes mentella in 2001

Trawl no. Latitude (N) Longitude (W) Trawling depth range (m) Volume of sample Alleles of the MER-2* locus
Min max mean 60 100
Pelagial of the Irminger Sea (R.V. AtlantNIRO)
6260.6531.53640691666170.4120.588
6360.6031.53277307292630.4210.579
6461.8728.92660740700180.5000.500
6562.0527.28788853821350.2000.800
6662.5527.20710790750180.3330.667
6863.1225.73695915805250.1200.880
Southwestern slope of Iceland (R.V. AtlantNIRO)
6963.0025.15639639639500.3800.620
7063.0325.42755755755280.2500.750
7163.1725.57824824824110.4090.591
7263.1525.17430430430370.6220.378
7363.1225.18509509509500.4500.550
7463.1025.20597597597500.4900.510
7662.9724.12597597597290.2410.759
7762.8524.02688688688500.3000.700
7863.0723.80525525525500.3200.680
Pelagial of the Irminger Sea (R.V. A.Fridriksson)
27562.2528.10  560500.3600.640
28761.3021.02  68070.6430.357
29061.3026.54  49040.2500.750
29161.3028.37  670130.3460.654
29360.3130.25  650230.3260.674

TABLE 7
Frequencies of alleles of MDH* locus in samples of Sebastes mentella in the pelagial of the Irminger Sea in 2000

Sample no. Latitude (N) Longitude (W) Trawling depth range (m) Volume of sample Alleles of the MDH* locus
min max 20 100
161.3028.1074083510 1.000
261.2028.10705770100.0500.950
361.3028.2064074010 1.000
461.2028.1076078010 1.000
561.2028.10740760100.0500.950
661.3028.1076077010 1.000
761.2028.10760760100.0500.950
861.4029.1078080010 1.000
961.2028.10650720100.0500.950
1061.308.20720750100.1000.900
1161.1428.1058060010 1.000
1261.1028.1054070010 1.000

TABLE 8
Frequencies of alleles of MDH* locus in samples of Sebastes mentella in 2001

Sample no. Latitude (N) Longitude (W) Trawling Depth range (m) Volume of sample Alleles of the MDH* locus
min Max 20 100
Pelagial of the Irminger Sea (R.V. AtlantNIRO)
6260.6531.53640691170.0590.941
6360.6031.5327730763 1.000
6461.8728.92660740180.0560.944
6562.0527.28788853350.0570.943
6662.5527.2071079018 1.000
6863.1225.7369591525 1.000
Southwestern slope of Iceland (R.V. AtlantNIRO)
6963.0025.15639639500.0400.960
7063.0325.4275575528 1.000
7163.1725.5782482411 1.000
7263.1525.1743043037 1.000
7363.1225.1850950950 1.000
7463.1025.20597597500.0400.960
7662.9724.1259759729 1.000
7762.8524.02688688500.0200.980
7863.0723.80525525500.0200.980
Pelagial of the Irminger Sea (R.V. A. Fridriksson)
27562.2528.10560 500.0200.980
28761.3021.02680 7 1.000
29061.3026.54490 4 1.000
29161.3028.37670 130.0760.924
29360.3130.25650 23 1.000

4. CONCLUSIONS

These investigations established the following.

5. LITERATURE CITED

Bakay, Y.I. & S.P. Melnikov 2001. Peculiarities of aggregations of redfish S. mentella in the Irminger pelagial. NWWG 2001, Working Document. 5: 1–20.

Bakay, Y.I. & S.P. Melnikov 2002. Vertical structure of Sebastes mentella concentrations in the pelagic open part of the Irminger Sea. NAFO SCR Doc. 02/10. Serial No. 4611: 1–21.

Dushchenko, V.V. 1986. Polymorphism of NADH-dependent malate dehydrogenase in Sebastes mentella Travin (Scorpaenidae) from the Irminger Sea. Voprosy ikhtiologii. 26(3): 522–524 (in Russian).

Johansen, T., A.K. Danielsdottir, K. Kristinsson, P.H. Petersen & G. Nevdal 1996. Studies on the relationship between deep-sea and oceanic Sebastes mentella in the Irminger Sea by the use of haemoglobin and allozyme analyse. ICES C.M.1996/G: 1–27.

Johansen, T., A.K. Danielsdottir, K. Meland & G. Nevdal 1997. Studies on the relationship between deep-sea and oceanic Sebastes mentella in the Irminger Sea by the use of haemoglobin, allozyme analyses and RAPD. ICES C.M.1997/HH: 1–13.

Magnusson, J. 1991. Eitt og annao um uthafskarfa. AEgir. 84: 3–7.

Magnusson, J. & J.V. Magnusson 1995. Oceanic redfish (Sebastes mentella) in the Irminger Sea and adjacent waters. SCI. MAR. 59 (3–4): 241–254.

Melnikov, S. P. 1998. Peculiarities of deepwater redfish, Sebastes mentella, distribution by depths in the Irminger Sea. NAFO SCR Doc. 98/16. Ser. No. N2995: 1–9.

Melnikov, S.P., A.P. Pedchenko & V.N. Shibanov 2001. On status of the pelagic redfish aggregations in NAFO Div. 1F. NAFO SCR Doc. 01/115. Ser. No. N 4503: 1–20.

Melnikov, S.P. & Y.I. Bakay 2002. Spatial structure of pelagic concentrations of Sebastes mentella of the Irminger Sea and adjacent waters. NAFO 2002, SCR Doc. 02/15. Serial No. 4616: 1–22.

Novikov, G.G., A.N. Stroganov, K.I. Afanasjev, V.N. Shibanov & S.P. Melnikov 2002. Analysis of population and genetic characteristics of redfish of the Irminger Sea. NWWG 2002. Working Document N19: 1–16.

Pavlov, A.I. 1992. Feeding and peculiarities of Sebastes mentella distribution in the pelagic waters of the Irminger Sea. In: Investigations of bioresources of the North Atlantic. Selected papers. PINRO. Murmansk.: 96–116 (in Russian).

Pavlov, A.I. & V.N. Shibanov 1984. Investigations of biological resources in pelagial and thalassobathyal in the open North Atlantic. Complex fishery investigations of PINRO in the North Basin: results and prospects. Sbornik nauchnykh trudov PINRO. Murmansk.:104–117 (in Russian).

Shatunovsky, M.I. 1980. Ecological regularities of metabolism of marine fishes. Moscow, Nauka. 288pp. (in Russian).

Shatunovsky, M.I., M.P. Bogoyavlenskaya, I.F. Veltishcheva & N.V. Maslennikova 1975. Investigations of generative exchange of the Baltic cod. Trudy VNIRO. 96: 57–62 (In Russian).

Shibanov, V.N., S.P. Melnikov & A.P. Pedchenko 1996. Dynamics of commercial stock of oceanic-type redfish Sebastes mentella in the Irminger Sea in 1989–1995 from results of Russian summer trawl-acoustic survey. ICES C.M. 1996/G:46.: 1–19.

Zakharov, G.P. 1969. Ecology and fishery for marine redfishes (Sebastes marinus and Sebastes mentella Travin) in the area of Iceland and Greenland. Abstract of Doctor Thesis. Moscow: 1–25 (in Russian).

Inter-and intraspecific variation of eyespot formation in the secondary deep-sea fish genus Neobythites (Ophidiidae, Ophidiiformes)

F. Uiblein1 and J.G. Nielsen2
1 Institute of Marine Research
Nordnesgaten 50, P.b. 1870 Nordnes, N-5817 Bergen, Norway
<franz.uiblein@imr.no>

2 Zoological Museum, University of Copenhagen
Copenhagen, Denmark

1. INTRODUCTION

Among the 50 hitherto known species of Neobythites, a genus of sedentary deep-sea fishes that occurs in subtropical and tropical waters at moderate depths (cf. Nielsen 1999, 2002), 23 show well-developed eyespots or ocelli, i.e. dark spots surrounded by a contrasting white ring, on the dorsal fin. To obtain further insights into the ecological function and evolutionary origin of these conspicuous structures we compared co-occurring species and populations from different areas with respect to several characters including number, size and position of ocelli.

2. RESULTS

Ocellus-bearing Neobythites species occur at shallower mid-depths than those without. Optimal signal transmission may be restricted to the lower shelf and upper slope. Species with single ocelli reach mostly into shallower waters than species with two or more ocelli. N. stefanovi occurs to a particularly great depth in the Red Sea, most probably due to the high temperature isothermy (Uiblein 1996).

In populations of N. stefanovi (northern Indian Ocean) and N. ocellatus (northern Carribbean) that partly overlap in distribution with the morphologically similar species N. steatiticus and N. gilli, respectively, a significant deviation in ocellus position was found when compared to allopatric populations. This character displacement supports the hypothesis that ocelli may have an important signaling function related to intraspecific communication. Earlier evidence on intraspecific variation in ocellus size in N. stefanovi and published data on other ocellus-bearing fishes, however, do suggest a primary anti-predator function (Uiblein, Nielson and Klausewitz 1994, Uiblein 1996).

3. DISCUSSION

In fishes that live under extended dawn conditions of the deep shelf and upper slope signal transmission by ocelli may be particularly efficient and hence these structures may have evolved to support several ecological functions. Two hypotheses shall be further examined. First, ocelli may serve at least two different ecological functions as an anipredator device and a means for species recognition. Second, ocelli are still subject to an ongoing evolutionary process and have evolved repeatedly to serve various functions (i.e. they are homoplastic characters). At present a morphological phylogeny of Neobythites is planned.

4. LITERATURE CITED

Nielsen, J.G. 1999. Atlantic occurrence of the genus Neobythites (Pisces: Ophidiidae) with three new species. Bull.Mar. Science 64:335–372.

Nielsen, J.G. 2002. Revision of the Indo-Pacific species of the Neobythites (Teleostei, Ophidiidae) with 15 new species. Galathea Report 19: 1–104.

Uiblein, F. 1995. Morphological variability between populations of Neobythites stefanovi (Pisces: Ophidiidae) from the deep Red Sea and the Gulf of Aden. Mar.Ecol.Prog.Ser. 124: 23–29.

Uiblein, F. 1996. Constraints and exploratory windows in light-reduced marine habitats. In: Uiblein, F., Ott, J. & M. Stachowitsch (Eds.): Deep-sea and extreme shallow-water habitats: affinities and adaptations. Biosystematics and Ecology Series, 11: 165–182.

Uiblein, F., J.G. Nielsen & W. Klausewitz 1994. Depth dependent morphological variation in two ophidiiform fishes from the deep Red Sea: evidence for species-specific structure in vertical distribution. Cybium 18: 15–23.

The yellow squat lobster (Cervimunida johni) fishery off Northern Chile, 1994–2001

E. Acuña, J.C. Villarroel, A. Bodini and M. Andrade
Departamento de Biología Marina, Facultad de Ciencias del Mar
Universidad Católica del Norte
Sede Coquimbo, Casilla 117, Coquimbo, Chile
<eacuna@ucn.cl, jcvillar@ucn.cl>

1. INTRODUCTION

The yellow squat lobster (Cervimunida jonhi) along with the red squat lobster (Pleuroncodes monodon) are two species of the family Galatheidae that occupy mainly the lower continental shelf and upper slope between 150 and 300 m depth off the Chilean coast. This particular habitat has a low oxygen content and is thought to provide a refuge against predators (Villarroel, Acuña and Andrade 2001). Both species play an important role in the trophic foodweb since they are the prey of many benthodemersal fishes and so transfer a great amount of energy towards upper levels (Arancibia and Meléndez 1987, Villarroel, Acuña and Andrade 2001).

The yellow squat lobster is an important fishing resource off central-northern Chile where it is caught by bottom trawls. According to Acuña, González and González (2003), data on yellow squat lobster was scarce until the beginning of the 1990s in this zone, when the first direct assessments using the swept-area method begun. These authors did an analysis of the biological data obtained between 1995–2000, which indicated a decrease in size and size at first sexual maturity in the yellow squat lobster, which they attributed as probably owing to fishing mortality and, or, environmental variability. This work incorporates new information about the demographic structure of the resource and catch and effort data from the central-northern Chilean fishery. The objective is to describe inter-annual differences in these parameters that may be related to biological variability.

2. METHODS

Data of commercial fishing hauls performed between 26° 04'S and 32° 10'S and between 120 and 420 m depth were analyzed. The study area was divided into Region III (26°04'–29° 12'S) and Region IV (29°12'–32°10'S). Biological data were obtained between 1994 and 2001 and fishing data since 1996.

The median size (cephalothorax length (CL), mm), sex proportion and the size at first sexual maturity were analyzed from 1995 to 2001. The size at first sexual maturity was determined using the method described by Udupa (1986).

Fishing data time series were analyzed based on fishing effort, catch and catch per unit of effort (CPUE, kg/hour of trawling). Analysis of spatial-temporal distribution used data from the fishing hauls and latitude and longitude records. Thus, a series of year-to-year maps were prepared with CPUE values obtained between 1996 and 2001.

3. RESULTS

The spatial distribution of CPUE shows that most of the hauls and the highest relative biomass was located between 28 °S and 32 °S during 1996 and 1997. Two main concentration zones were observed south and north of 30 °40'S. A contraction in the distribution of fishing effort was observed between 28 °S and 30 °S during 1998, with the most important concentrations located between 29°S and 28°S. The spatial distribution showed an expansion again during 1999, but as several restricted areas, isolated from each other but spread through the study area that showed lower CPUE values. A new concentration of the fishing area towards the central area was observed during 2000, similar to 1998, although CPUE values were lower than in 1998. Higher CPUE values were found mainly between 28°S and 30°S in the central area and between 31°S and 32°S in a smaller area during 2001 (Figure 1).

FIGURE 1
Spatial-temporal distribution of the relative biomass of the yellow squat lobster of central-northern Chile

FIGURE 1

This species shows sexual dimorphism in size; males are larger than females. The median size of the males increased significantly in Region III between 1994 and 1998, when a significant decrease was observed. A similar trend was observed for females, but the median size decreased slightly from 28.6 mm CL to 27 mm CL from 1997 to 2001 (Figure 2). Median size in Region IV increased for both sexes between 1994 and 1997, when a sharp decrease was observed. However, both sexes showed a slight increase during 2001, suggesting a recovery of this biological characteristic.

The size at first sexual maturity showed an increase of 15 mm CL between 1994 and 1998. Then, it decreased from 35.4 to 23.9 mm CL between 1998 and 1999 and increased again between 1999 and 2001 (Figure 3).

There was an increase in the fraction of males in Region III between 1996 and 1998 and a decrease between 1998 and 2000, being around 50 percent in 1994 and 2001. The proportion of males increased between 1996 and 1999 in Region IV and decreased between 1999 and 2001, reaching around 50 percent (Figure 4).

FIGURE 2
Annual time series by region showing the yellow squat lobster median size, for both sexes

FIGURE 2

The median effort by haul increased between 1996 and 1999 in Region III and between 1996 and 2000 in Region IV. The highest fishing effort (hours of trawling) was observed during 1999 and 2000 in RegionsIII and IV respectively. The median catch by haul decreased significantly between 1998 and 1999 in Region III, and remained at these low levels between 1999 and 2001. This decrease occurred between 1997 and 1998 in Region IV with the lowest values observed in 2000. The CPUE decreased between 1996 and 1999 in Region III, and between 1997 and 1999 in Region IV. In both regions, the relative biomass remained stable at low levels during the last years, but showed a slight increase during 2001 (Figure 5).

FIGURE 3
Annual time series showing the size at first sexual maturity of females yellow squat lobster

FIGURE 3

4. DISCUSSION

Fishing removes larger and older individuals, changing the size and age structure of exploited populations. Reducing the abundance of larger specimens leads to a decrease of the stock spawning potential (Goñi 1998). Our results, show that an increase of the median size and size at first sexual maturity was observed along with a decrease of the biomass during the first fishing period (1994–1998), suggesting that the effect of this decrease in the demography of the species lagged a couple of years.

FIGURE 4
Annual time series by region showing the yellow squat lobster sex ratio

FIGURE 4
FIGURE 4

Inter-annual fluctuations in the availability of fishing resources are due to multiple factors. Hannah (1993) and Ye (2000) have demonstrated the influence of the environmental variability and the level of the spawning stock on crustacean recruitment. Population changes are also influenced by natural and fishing mortality as it has been found in other fisheries (Sinclair 2001, Myers, Barrowman, Hoenig and Qu 1996).

According to Wolff and Aroca (1995) and Acuña et al. (1995) during the beginning of the study period the population of the yellow squat lobster showed no signs of overexploitation. The population was relatively abundant and there was a large supply of recruits resulting in a decrease in median size (Villarroel et al. 2001, Acuña et al. 1995, Acuña, González and González 2003). However, low recruitment occurred during 1998 and the increase of fishing effort caused a decrease in biomass. This decrease mainly affected the spawning stock and gravid females carrying eggs, whose proportion in the population was high in the catches. Campodónico (2002) pointed out this feature indicating that the decrease of the biomass was caused by a high level of fishing mortality and a decrease in the parental stock biomass.

FIGURE 5
Annual time series by region showing median effort, catch and CPUE of the yellow squat lobster fishery of central-northern Chile

FIGURE 7
FIGURE 7
FIGURE 7

It is difficult to determine if fishing could be the only factor causing the decrease in resource biomass, but it might be the most important. In the yellow squat lobster fishery, increases in fishing effort owing to an increase in number of vessels occurred between 1988 and 1998 (Wolff and Aroca 1995, Acuña et al. 1998) and was concentrated in the southern area where greater biomasses and larger specimens were found (Acuña et al. 1998). The decrease in CPUE coincides with an increase of fishing time and the spatial distribution of the fishing grounds contracted to several small, isolated areas.

An improvement in the biological and demographic parameters was observed during 2001, indicating a slight recovery of the resource. This was also observed during 2002 (Guerrero et al. 2003) coinciding with a sharp decrease in landings and catch quotas during the last two years (Figure 6). Campodónico (2002) suggested that better recruitment was the important factor in the recovery of the stock.

FIGURE 6
Annual time series showing landings and catch quota for the yellow squat lobster fishery from regions III and IV

Note the annual quota system began in 1995.

FIGURE 6

5. LITERATURE CITED

Acuña, E., H. Arancibia, A. Mujica, K. Brokordt & C. Gaymer 1995. Estudio biológico-pesquero del langostino amarillo (Cervimunida johni) en la III y IV Región, mediante el uso de la flota arrastrera con base en Coquimbo. Informe final Proyecto U.C. del Norte -Sede Coquimbo/ Instituto de Investigación Pesquera de la VIII Región/ Empresas Pesqueras Coquimbo, 107pp. + 2 Anexos.

Acuña, E., H. Arancibia, A. Mujica, L. Cid & R.Roa 1998. Análisis de la pesquería y evaluación indirecta del stock de langostino amarillo en la III y IV Regiones. Informes Técnicos FIP FIT/IT 96-08, 153 pp.

Acuña, E., M.T. González & M. González 2003. La pesquería de langostinos y camarón nailon en la zona norte de Chile: 1995–2000. In Yañez, E. (Ed.). Actividad pesquera y de acuicultura en Chile. Ediciones Universitarias de Valparaíso, Chile, pp. 263–288.

Arancibia, H. & R. Meléndez 1987. Alimentación de peces concurrentes en la pesquería de Pleuroncodes monodon Milne Edwards. Investigación Pesquera (Chile) 34: 113–28.

Campodónico, I. 2002. Cuota global anual de captura 2003 para la pesquería del langostino amarillo de la III y IV Región. Informe Técnico (R. Pesq.) № 98. 15 pp.

Goñi, R. 1998. Ecosystem effects of marine fisheries: an overview. Ocean & Coastal Management 40: 37–64.

Guerrero, A., R. Bustos, M. Ahumada & P. Arana 2003. Monitoreo de las pesquerías de crustáceos bentónicos desarrolladas entre la III y IV Región, año 2002. Informe Final Convenio Marco, №9/2003. 236 pp.

Hannah, R.W. 1993. Influence of environmental variation and spawning stock levels on recruitment of ocean shrimp (Pandalus jordani). Canadian Journal of Fisheries and Aquatic Sciences 50: 612–622.

Myers, R.A., N. J. Barrowman, J. M. Hoenig & Z. Qu 1996. The collapse of cod in Eastern Canada: the evidence from tagging data. ICES Journal of Marine Science 53: 629–640.

Sinclair, A F. 2001. Natural mortality of cod (Gadus morhua) in the Southern Gulf of St Lawrence. ICES Journal of Marine Science 58: 1–10.

Udupa, K.S. 1986. Statistical method of estimating the size at first maturity in fishes. Fishbyte, August: 8–10.

Villarroel, J.C., E. Acuña & M. Andrade 2001. Feeding and distribution of the big eye flounder Hippoglossina macrops off northern Chile. Marine and Freshwater Research 52:833–841.

Wolff, M. & T. Aroca 1995. Population dynamics and fishery of the Chilean squat lobster Cervimunida johni Porter, (Decapoda, Galatheidae) off the coast of Coquimbo, northern Chile. Revista de Biología Marina, Valparaíso 30: 57–70.

Ye, Y. 2000. Is recruitment related to spawning stock in penaeid shrimp fisheries? ICES Journal of Marine Science 57: 1103–1109.

A note on the reproductive biology of the longnose velvet dogfish (Centroscymnus crepidater) from the Northeast Atlantic1

1This note is a summary of the poster entitled “The reproductive biology of the longnose velvet dogfish, Centroscymnus crepidater, in the northeast Atlantic”, presented at Deep-Sea 2003. The main body of this research is currently in press.

C.P. Nolan1,2 and F. Hogan1
1 Zoology Department, Trinity College
Dublin 2, Ireland

2 Irish Sea Fisheries Board (BIM)
Crofton Road, Dun Laoghaire, Co. Dublin, Ireland
<cnolan@bim.ie> <spacemarv@yahoo.com>

The reproductive biology of the little known and generally discarded deep-water shark Centroscymnus crepidater was examined from a total sample of 97 females and 44 males. Specimens were collected by observers from trawls between 710 m and 1317 m aboard commercial fishing vessels over a three-month period from October to December 2002 in the north-east Atlantic. Individuals ranged from 350 mm to 868 mm total length and mature males and females dominated the sample. Females were generally larger than males, with male and female modal sizes identified at 625 cm and 800 cm total length respectively. Counts of ova greater than 1 mm in diameter ranged from 26 to 117. Ova diameter ranged from 1 mm to 58 mm with the most common diameter between 1 mm and 3 mm. Of the 17.5 percent of the females found carrying embryos, the uterine fecundity ranged from 1 to 9 embryos. A total of 48 embryos were recorded in various stages of development, ranging from 0.2 g to 98.0 g in mass and 32 mm to 261 mm in total length with a male:female ratio of 1:1.125. Liver mass ranged from 200 g to 735 g and contributed 24 percent and 25 percent to the total body mass in males and females, respectively. The hepatosomatic index (HSI) was relatively constant in males but showed a significant decrease in females following ovulation, consistent with the translocation of resources from the liver to developing ova. The diversion of resources to ova production was also evident in the gonadosomatic index (GSI) where a maximum, of 18.16 percent was recorded in a mature female. Consistent with other Centroscymnus species, C. crepidater does not appear to have a defined breeding season. Ovarian and uterine fecundity are low indicating that the reproductive capability of this species is low. Results have been interpretted relative to the limited data on reproduction available for this species.

A note on the fecundity and reproductive capability of orange roughy (Hoplostethus atlanticus Collett 1889) on the Porcupine Bank, Northeast Atlantic1

1 This note is a summary of the poster entitled “The fecundity and reproductive capability of orange roughy off the Porcupine Bank, north-east Atlantic”, presented at Deep-Sea 2003. The main body of this research is currently in preparation.

C. Minto1 and C.P. Nolan1, 2
1 Zoology Department, Trinity College
Dublin 2, Ireland

2 Irish Sea Fisheries Board (BIM)
Crofton Road, Dun Laoghaire, Co. Dublin, Ireland.
<mintoc@eircom.net> <cnolan@bim.ie>

The first comprehensive analysis of the fecundity and reproductive capability of orange roughy (Hoplostethus atlanticus) from a specific area in the Northeast Atlantic is reported. Specimens were collected from commercial vessels targeting this species on the Porcupine Bank (ICES sub-area VII), in waters of between 1400 m and 1650 m depth. Between September and December 2002, a non-random, stratified, sampling protocol was implemented by fisheries biologists to collect mature female fish between 300 mm and 540 mm SL. Ovaries from 70 individuals, representing the majority of 10 mm SL size classes sampled, formed the analytical sample. A novel and semi-automated method of oocyte counting was developed, allowing digital images of oocytes to be annotated, counted and stored.

Total fecundity ranged between 20352 and 244578 oocytes per female and mean total fecundity was estimated to be 97368 oocytes per female (S.D. = 48322). Mean relative fecundity was estimated to be 33376 oocytes per kg (S.D. = 11407). Total fecundity showed an increase with both length and weight; however, the relationship between fecundity and age appears to decline for extremely old fish (>120 years). Macroscopic analyses showed that 50 percent of females were not mature until they reached 28.7 years and 387 mm SL.

The relationships between fecundity and the variables: length, weight, and age are characterised by high variability. The variability is proposed to be related to heterogeneity in habitat productivity, as fish from different regions aggregate, and the genetic structure of the gravid population.

A comparison with stocks from the southern hemisphere indicates that orange roughy from the northeast Atlantic generally mature at a larger size and have a higher mean fecundity than those found in the southern hemisphere. This may reflect differences in growth rates influenced by environmental variables and fishing pressure. Lower productivity in the northeast Atlantic is suggested to be the cause of an increase in the length and age at maturity.

It is emphasized that the populations currently being fished are the product of spawning events that occurred prior to the start of the fishery and that the effects of fishing on the overall population fecundity will only be evident when the age of the fishery is approximately equivalent to the age of the fish at maturity.

Deepwater trawl and longline sampling levels

L. Borges1,2, M. Clarke3, E. Rogan2 and R. Officer3

1Corresponding address: Marine Institute, Snugboro Road, Abbotstown, Dublin 15, Ireland

2 Aquaculture and Fisheries Development Centre, Department of Zoology, Ecology and Plant Science
University College Cork, Lee Maltings, Prospect Row Cork, Ireland

3 Marine Institute, Galway Technology Park
Parkmore, Galway, Ireland
<e.rogan@ucc.ie>

Ireland, as a member state of the European Union, must implement observer schemes to improve the quality of the scientific data used for assessment of deep-water species (Council Regulation No 2347/2002). Optimal sampling levels for deep-water trawl and longline sampling are presented here in relation to precision limitations. The analysis is based on surveys carried out between 1993 and 2000 in the North East Atlantic using two of the main commercial gears: trawl (four surveys) and longline (three surveys). Due to the hierarchical structure of the data -hauls/sets nested within surveys-amultistage analysis was performed, according to the approach detailed in Allen et al. (2002), to determine the levels of precision achieved in the past.

The analysis is based on mixed-models, where the unexplained variability of catch per unit of effort (CPUE) is partitioned over the two sampling levels of trip and haul (random effects), and in the process, the major parameters likely to affect catches (for example: gear, year, depth, area, catch richness haul position, temperature and salinity) are also determined (fixed effects). CPUE was chosen as the dependent variable due to the multipurpose nature of the surveys studied. The results show that trawl catches are highly variable -36 percent -coefficient of variation (CV) while longline catches shows the opposite, a 9 percent CV. Longline catches are also highly dependent on depth, particularly when depth strata are considered in 300 meters intervals (200, 400 and 500 m depth intervals were also tested). Another depth-related parameter, temperature, was also significant in explaining catch variability, but could not be considered in the final model due to low sampling levels.

None of the parameters studied appear to explain catch variability in trawl fisheries. Only random variability is present, in both sampling levels of trip and haul, thus future sampling schemes should reflect this fact. In both gears, area and year were not significant factors in the analysis. When recommended sampling levels are stratified by depth, sampling should be grouped in 300 m depth strata to improve precision.

Trawl sampling levels should be doubled to achieve at least a 25 percent CV annually, while longline levels can be reduced by half and still achieve a 14 percent CV. The recommended sampling levels are a product of the sampling plan objectives, i.e. catch qualification and quantification and also collection of biological information. If the resulting data are to be used specifically in discard studies, then geographical area and the corresponding increase in sampling should also be taken into account in future sampling schemes.

ACKNOWLEDGMENTS

L. Borges was funded by the Marine Institute and the Marine RTDI Measure, Productive Sector Operational Programme, National Development Plan 2000–2006 (Grant-aid Agreement No.PHD/01/002). We wish to thank the crews and the scientific personnel involved in all surveys.

LITERATURE CITED

Allen, M., Kilpatrick, D., Armstrong, M., Briggs, R., G. Course, G. & N. Perez 2002. Multistage Cluster Sampling Design and Optimal Sample Sizes for Estimation of Fish Discards from Commercial Trawlers. Fisheries Research. 55: 11–24.

The estimation of catch levels for new orange roughy fisheries onseamounts: a meta-analysis of seamount data

M. Clark, B. Bull and Dianne Tracey
National Institute of Water and Atmospheric Research
P. O. Box 14–901, Wellington, New Zealand
<m.clark@niwa.co.nz>

1. INTRODUCTION

Seamounts are widely regarded as being fragile habitats that are susceptible to overfishing. This implies that careful management is required in the initial stages of any fishery development to reduce the risks from rapid expansion in fishing effort and possible overexploitation of the fish resources.

The task of designing and carrying out appropriate fish abundance surveys on seamounts can be a lengthy, expensive and complicated task. And, the stocks to be assessed may be small and localized, which raises the question of whether a research programme is warranted or cost-effective.

2. METHODOLOGY

Physical attributes and catch data of deepwater fisheries were compiled for 77 seamounts in the New Zealand region (Figure 1). Characteristics of location, depth, size, elevation above the seafloor, age, continental association, geological origin, distance offshore and from surrounding seamounts and degree of spawning were defined. These were then analysed as independent variables against the minimum orange roughy population size estimated from the historical amount of catch taken from the seamounts to investigate whether they could be useful predictors of likely safe catch from newly found seamounts.

Multiple regression procedures were used to model the effects of the physical variables on orange roughy stock size. There were two stages in the analysis. First, biomass was modelled on individual seamounts grouped in regions (as a categorical variable) and included predictors specific to individual seamounts. This analysis showed region, depth of the peak, and slope of the seamount to be significant. The relative region effects are shown in Figure 2. A second analysis was carried out where the region effects were modelled, using predictors related to entire regions. This showed latitude (Figure 3) and association (continental and oceanic) to be important. The predictive power of the models was tested by cross validation and compared with simpler models to assess their informative value.

3. RESULTS

For the individual seamount model, the approximating formula is

Predicted biomass = exp (intercept + region effect + depth of top effect + slope effect)

where the intercept is 6.89 and the region, depth of top, and slope effects are given by Clark, Bull and Tracy (2001). This formula can be used where a new seamount is found within an existing region. For example, the approximate prediction for Mt. Ghost (if it were a new feature on the southern Louisville Ridge) would be:

Biomass = exp (6.89 + 0.83 (Louisville South) + 0.76 (top depth is 620 m) + 0.08 (slope is 0.14))

= 5 200 t.

FIGURE 1
Regions and seamount features included in the study

The Chatham Rise regions are indicated by “CR”.

FIGURE 1

FIGURE 2
Effect of region on estimated seamount biomass

The length of each bar indicates the average relative abundance of orange roughy on seamounts in the region once the effects of the ‘slope’ and ‘depth of top’ predictors have been removed.

FIGURE 2

FIGURE 3
Effect of latitude on orange roughy abundance by region

The height of the curve indicates the relative abundance of orange roughy at that latitude, all else being equal. The x-axis is restricted to the range of latitudes in which most of the data set lies.

FIGURE 3

For a new region, using the region effect predicted by the region regression model, then

Predicted region effect = exp (intercept + latitude effect + association effect)

where the intercept is 0.73 and the region, depth of top, and slope effects are given in Clark, Bull and Tracey (2001).

4. CONCLUSIONS

Data on the physical features of a seamount can be informative in predicting possible stock size. Simple formulae can be used to obtain estimates of biomass, although predictions are approximate only, as the data show a wide scatter. However, it appears they can be useful in helping guide initial management of a developing fishery on a seamount.

5. ACKNOWLEDGEMENTS

This work was funded by the Ministry of Fisheries (project ORH200103) and was based on seamount data provided by research under The New Zealand Foundation for Research, Science, and Technology contract CO1X0224

6. LITERATURE CITED

Clark, M.R., B. Bull & D.M Tracey 2001. The estimation of catch levels for new orange roughy fisheries on seamounts: a meta-analysis of seamount data. New Zealand Fisheries Assessment Report 2001/75. 40 pp.

How old are these deep-sea species? Applying improved techniques to validate longevity

L. Paul1, Dianne Tracey1, H. Neil1, P. Horn1, A. Andrews2 and R. Sparks3
1 National Institute of Water and Atmospheric Research Ltd (NIWA)
P O Box 14901, Wellington, New Zealand
<l.paul@niwa.co.nz> <d.tracey@niwa.co.nz> <h.neil@niwa.co.nz> <p.horn@niwa.co.nz>

1 Moss Landing Marine Laboratories
8272 Moss Landing Road, Moss Landing, California 95039
<andrews@mlml.calstate.edu>

1 Rafter Radiocarbon Laboratory, Institute of Geological and Nuclear Sciences (GNS)
PO Box 31–312, Lower Hutt, New Zealand
<R.Sparks@gns.cri.nz>

Careful resource management is required for deep-sea fish due to their great longevity (and thus low productivity) and late maturation. Age estimation, required for such management, in the deep-sea environment has historically been difficult. There is growing confidence in the current age interpretations of deep-sea fish species. Their characteristic of high longevity is likely to be a resilient adaptation to the deep-sea environment, with its assumed low productivity and, or, episodic recruitment. Here we describe two advances in deep-sea age estimation -radiometric age determination and bomb carbon age validation.

The otolith growth rings of many deepwater fish species are often closely spaced and difficult to count. Although ring-counting is likely to remain the most common procedure, other techniques should be used to validate the ages of these species. The techniques used for shallow-water species (e.g. size frequency analysis, tagging, otolith-edge growth patterns and otolith weights) are not suitable for deepwater species. The two most promising validation techniques, radiometric and bomb radiocarbon analyses, are difficult to carry out, but have the advantage of being independent of the otolith's appearance.

Ageing studies of New Zealand's deepwater species have been based on otolith growth ring counts. The assumption is that these zones, although narrow and somewhat different from those of more near-shore species, and often are particularly difficult to read, represent annual growth. As found elsewhere, recent age readings are giving higher, rather than lower, ages.

The following are examples of deepwater species that have been investigated: black cardinalfish (Epigonus telescopus), black oreo (Allocyttus niger), smooth oreo (Pseudocyttus maculatus), orange roughy (Hoplostethus atlanticus), ridge-scaled rattail (Macrourus carinatus), rubyfish (Plagiogeneion rubiginosum), sea perch (Helicolenus spp.), and bluenose (Hyperoglyphe antarctica). Most of these species have high ring counts, and three have maximum ages of at least 100 years. Ages at maturity can be as high as 36 years. The slow growth rates, high longevity, low levels of sustainable yields, and slow recovery rates make them particularly vulnerable to overfishing. Figures 1a and 1b provide examples of growth zones present on orange roughy and rubyfish otolith thin sections.

FIGURES 1A AND 1B
Images of whole orange roughy (top) and rubyfish (bottom) otoliths and their respective longitudinal and transverse thin sections showing annual growth zones

FIGURES 1A AND 1B

Currently, more emphasis is being placed on validation studies that include radioisotope ratios (226Ra/210Pb) and the time-specific, bomb-produced, radiocarbon (14C) marker. We are carrying out further applications of these techniques, which we are optimistic will resolve issues with previous age validation studies of our deepwater species.

Radiometric age determination is based on measuring the activity of pairs of radionuclides present in the otolith, which are known to decay from one into the other at a known rate over time; for example 226Ra (radon) into 210Pb (lead). They are incorporated at equilibrium into the otolith core when it is formed and their ratio then changes over subsequent years. Earlier radiometric studies of orange roughy and oreos used pooled samples of whole otoliths, which caused a problem of between-individual variability in otolith mass growth rates. In an attempt to overcome this problem, otolith cores and a refined radiometric technique were used. Very small core samples were analysed using thermal ionization mass spectrometry at Moss Landing Marine Laboratories.

Results indicate that the application of the refined radiometric technique is feasible to varying degrees for oreos, cardinalfish, and orange roughy, but most successful for the latter two. For the orange roughy pilot study, calculated radiometric age was in close agreement with zone count age for three of the four samples. Estimated and radiometric age data for orange roughy are shown in Figure 2. There was strong agreement for the oldest sample (77 y), indicating that this is a long-lived species. For cardinalfish, a juvenile group and an old aged group of otoliths were analysed; the calculated radiometric age agreed closely with zone count age.

The bomb-radiocarbon age validation approach uses the sudden pulse of radiocarbon (14C) released into the atmosphere from nuclear testing in the 1950s. In one procedure, the radiocarbon level in otolith cores demonstrates whether the fish was born pre-bomb or post-bomb, and if born during the period of radiocarbon increase (1950–1970) an age can be assigned. In this development, radiocarbon levels can be measured across an otolith, and the position of the bomb pulse compared with the counted age.

FIGURE 2
Plot of estimated and radiometric age data for orange roughy
FIGURE 2
FIGURE 3
Results of the bomb radiocarbon study for bluenose and rubyfish
FIGURE 3

Bluenose and rubyfish have otoliths with complex growth zones. From zone counts, maximum age estimates for bluenose have ranged from 15 to 50 years. Rubyfish have a maximum age from zone counts of 80+ years. The bomb chronometer procedure was used to determine whether these ages were realistic. The 14C level in carbonate samples drilled from otolith cores confirmed that the selected fish (five of each species) were born ‘pre-bomb’ era. Samples were then drilled in a sequence across otolith cross-sections to locate the position of the 1950–60 rise in 14C. This position defined a known time interval that was compared with same time interval as determined by zone counts. Results of the bomb radiocarbon study for bluenose and rubyfish are shown in Figure 3. We consider that this procedure has advantages over the core-only method. Consideration was given to the time-lag issue, i.e., the time required for 14C to move from surface waters down to the depth (500–1000 m) at which the adults of both these species reside. Zone counts for rubyfish matched well with the bomb carbon curve, confirming the reliability of the zone count method. However, while the maximum age for bluenose has been shown to be in excess of 40 years, the poor match of zone counts to the bomb carbon curve indicate that problems with ageing of this species have yet to be fully resolved.

This research was funded by the New Zealand Ministry of Fisheries under projects DEE200202, INS200002.

An overview of the orange roughy (Hoplostethus sp.) fishery off Chile

I. Payá, M. Montecinos, V. Ojeda and L. Cid
Instituto de Fomento Pesquero
Blanco 839
Valparaíso, Chile
<ipaya@ifop.cl>

1. INTRODUCTION

There are several orange roughy fisheries in different parts of the world: New Zealand, Australia, Namibia, South Indian Ocean and the North East Atlantic. Several reviews have been published on these fisheries: Clark (1996) summarizes the methods used for biomass estimation; Tracey and Horn (1999) review the age determination and Branch (2001) summarizes fisheries, estimation methods, biology and stock structure in Namibia and worldwide. However, Chilean orange roughy information is only available in technical reports and no comprehensive review exists. Therefore, the purpose here is to provide an overview of this rather new orange roughy fishery in Chilean waters, including catch, effort, catch rates, biology, stock assessment and management information.

The first commercial aggregations of orange roughy in Chilean waters were discovered in 1997–1998 by a fishing industry assisted by a New Zealand captain. Because of this finding the Chilean government through its Fishery Development Institute (IFOP) made a scientific exploration for this resource in 1998. This survey used both the R.V. Abate Molina and a commercial vessel (Lillo et al. 1999). Orange roughy were found in three main areas of seamounts: Juan Fernandez archipelago (JF), Bajo O'Higgins (BO) and Punta Sierra (PS) (Figure 1). Hence, in 1998 according to Chilean fishery law, the orange roughy fishery was declared as a “Fishery Development Regime”(FDR), which means that the annual quota is fixed and allocated between fishery industries by means of an auction system.

From its beginning the fishery has been monitored by the IFOP using scientific observers onboard the commercial vessels. This monitoring system has been collecting logbooks, which have permitted analysis of spatial and temporal fishing activities. The observers also collected biological information concerning the catches.

The fishery is based on good catch rates taken during spawning aggregations that occur in June–July above seamounts. Spawning behaviour has been analysed using gonad indices and histological studies. Age and growth studies were carried out in 2001 using NIWA (National Institute of Water and Atmospheric Research Ltd, New Zealand) ageing techniques (Gili et al. 2002). However, no juvenile otoliths were available, so growth models were fitted integrating Chilean data with juvenile data from New Zealand (Gili et al. 2002). In this review we include updated growth models fitted with Bayesian inference incorporating only Chilean data.

The Chilean Undersecretary for Fisheries contracted NIWA in 1999 to advise on the development of a fishery monitoring, stock assessment and management of orange roughy fishery (Clark 2000). More recently the IFOP received cooperation from the NatMIRC (National Marine Information and Research Centre; Namibia) to learn from Namibian orange roughy experiences (Montecinos, Payá and Canales 2003). However, there have been several difficulties in developing an Chilean orange roughy stock assessment model, the most important being the short time series, the lack of knowledge of spatial and temporal stock dynamics, the unreliability of abundance indices based on catch rates and the lack of acoustic surveys. There is also concern about the results of stock assessments in others countries because Chilean orange roughy displays the same general pattern that has been observed in most of the orange roughy fisheries in the world, i.e. after a few years of exploitation the catch rate and total catch decrease strongly but the length structure of fish in the catch does not change too much. Therefore, hypotheses of fishing-driven disruptions; habitat perturbation and intermittent spawning as possible causes of local abundance depletion that have been postulated for this species by McAllister and Kirchner (2002) and should be tested in Chile.

FIGURE 1
Orange roughy fishing grounds by areas

(JF=Juan Fernandez Archipelago; BO= Bajo O'Higgins and; PS: Punta Sierra)
Frames A and B enclosed the areas showed in Figure 2.

FIGURE 1

The precautionary approach suggests that a fishing effects hypothesis that is risk averse should be used until other hypotheses have been tested. However, Namibian experience on the Frankies ground showed that local abundance could recover in only two years after banning the fishery (Boyer, Hampton and Kirchner pers.com.). Therefore we have suggested that a fishery exclusion experiment should be done in Chile.

During 2003 the first acoustic surveys was made by industry and the results will soon be released, enough information will then be available to make a full stock assessment. Nevertheless, the research focus should be on testing the perturbation hypothesis and producing reliable abundance indices from commercial catches and acoustic surveys.

2. MATERIAL AND METHODS

2.1 Monitoring system

Fishing activities have been monitored from the beginning of the fishery by IFOP observers onboard commercial vessels. The observers carry out three types of sampling: logbooks; length structure and biological samples. Logbook records include: target species; haul number; geographic satellite position (GPS) at the beginning and the end of haul; depth of net at the beginning and at the end of trawling; time at the beginning and at end of trawling; catch by species, sea surface temperature; sea bottom temperature; and net types. Length frequencies are collected by random sampling and recorded data includes: haul number; fork length; sexes and maturity stages in females. Biological sampling is size stratified and includes recording haul number, fork length, total fish weight, gutted fish weight, gonad weight, stomach contents, identification of sex ratios, maturity stages, and collection otoliths and bycatch species.

2.2 Fishing grounds

The fishing grounds were represented on a contour map of sea bottom based on data downloaded from GEBCO and GEODAT. The haul locations from 1999 to 2003 were included in a 3D representation of Juan Fernandez archipelago, Bajo O'Higgins and Punta sierra bottom topography. This topography was raised in situ during the 1999 scientific exploration survey (Lillo et al. 1999, Young et al. 2000).

2.3 Environmental information

The fishing ground environment was characterized by comparing water masses with fishing haul frequency by 50 m depth intervals. The water mass classification proposed by Silva (1985) was used. Haul frequency by 0.5°C bottom temperature recorded by net sensor was calculated. Both frequencies were computed using all fishing hauls recorded from 1999 to 2003.

2.4 Fishing activities

The evolution of fishing activities was analysed in terms of catch, fishing effort (trawling hours) and catch rates (tonnes/trawling hour) by year and fishing ground. Catch rates (R) were calculated using a ratio estimator

where C is catch, E the fishing effort and i the index of hauls number.

In order to check that orange roughy were caught during spawning concentration, fishing season was characterized in terms of historical (1999–2003 mean) patterns of catch, catch rates and gonadic index by month. The evolution of length of fishing season was analyzed computing catch per 10-day intervals during different years.

2.5 Spawning time

To characterize the relationship between spawning and fishing season, historical means of female gonadic indices (GI) were calculated by

To investigate a possible time sequence of spawning throughout fishing grounds the GI were analysed by 10-day intervals for each fishing ground.

2.6 Size structure

Fork length frequency distributions by sex were calculated from samples weighted by catch. Distributions were analyzed by years and fishing areas for the Juan Fernandez archipelago, Bajo O'Higgins and Punta Sierra. Because of similarity of the distributions over years, the distributions were characterized by their modes and proportion of individuals >45 cm for males and >50 cm for females. These proportions do not have any special biological meaning, but represent the larger fish of stocks that could be removed first under heavy fishing pressure.

2.7 Age, growth and natural mortality

Gili et al. (2002) undertook age determinations from sectioned otolith rings applying the same technique developed by New Zealand scientists. Data analyses were based on samples taken in 2001, which did not have any juvenile fishes. Gili et al. (2002) fitted a preliminary growth model complementing Chilean orange roughy data with juvenile data from New Zealand orange roughy. This procedure assumed that Chilean orange roughy and New Zealand orange roughy have the same growth parameters. This is unlikely because Chilean orange roughy are larger than those from New Zealand (Branch 2001).

In this work we determined the ages of otoliths from 315 males and 364 females sampled in 2002. Combining these samples with samples read in 2001, the total number of otoliths available was 611 males and 675 females. Consistency of the age determination technique and criteria were assured because the otolith reader, Mr Luis Cid, has been member of the reading team from 2001 (Gili et al. 2002).

Neither 2002 data nor 2001 data included juvenile fish, therefore we applied a Bayesian inference for estimating the parameters in the von Bertalanffy (1938) growth model:

= L (1- e (-K(t0-t))

where t is the standard length estimate at age t and L, K and t0 are model parameters. The log of data likelihood to maximize is:

LD = -0.5 n ln (res / n)

where n is the total number of observations and Li is the standard length of fish in the samples which were transformed from fork length (FL) using the Young et al. (2000) equations

Lm = 0.416 + 0.9FLm for males

Lf = 0.467 + 0.902FLf for females

A normal prior distribution was used for the to parameter, therefore the log of prior likelihood is

where u is the t0 estimated for New Zealand orange roughy by Doonan (1994) and currently used in stock assessment (Annala et al. 2002). The coefficient of variation (c.v.) was assumed to be 0.9. Male u is -0.44 and female u is -0.6. For L and K, prior distribution uniform distributions from - to + were used. The maximum total likelihood, LT= LD+LP, was found by minimizing -LT by a quasi-Newton algorithm.

Natural mortality was estimated by Gili et al. (2002) using different methods based on the growth parameters. In the current work we estimated natural mortality using updated growth parameters and applying the same methods used by these authors. Alajarata (1984), for whom natural mortality is function of the longevity Tmax, used the relation

M = - ln (0.01) /Tmax

where the longevity is estimated by Taylor (1958) as

Alverson and Carney (1975) based M on the time of cohort that maximized the weights and longevity.

Taylor (1958) based M on the parameters K and to

A Chapman and Robson (1960) catch curve method was also applied. This method was used because it is more precise and less biased than lineal regression methods (Dunn, Francis and Doonan 2002). Natural mortality was supposed to be similar to total mortality (Z), because of the unfished conditions of the stock in the early years of the fishery and the low level of catches, i.e.

where a is mean age (over the recruited ages) and n is the sample size. The variance is given by

This method was applied to catch-at-age data, which was calculated using length-frequency data and 5-year age-length key.

3. RESULTS

3.1 Distribution of fishing grounds and environment

Two exploratory surveys were made in August–September 1998, which covered the area from Nazca (25° 00'S) to Mocha Island (37° 00'S) (Lillo et al. 1999). A total of 23 seamounts were explored but orange roughy were found on only five, located in the Juan Fernandez archipelago, Bajo O'Higgins and Mocha Island. A third exploratory cruise was made in October 1999 from 24° 00'S to 28° 00'S where eight seamounts were explored but orange roughy were not found. The fourth exploratory survey was conducted in September 2000 from 31°01' to 31°30'S where one seamount was explored, which later was exploited by a fishery (Young et al. 2000). The last exploratory survey was made in July 2003 and covered from 37°48'S to 41°09'S, where two seamounts were found but orange roughy was practically absent; it composed only 0.1 percent of the research catches (Young et al. 2004). All the orange roughy detected have been adult fish and the juvenile distribution areas remain unknown.

The known distribution of orange roughy is limited to spawning areas where fishes form dense aggregations and are easily caught by trawlers. No juveniles have been available to the fishery. Fishing grounds are distributed along three main areas (a) Juan Fernandez archipelago area, which has five fishing grounds (JF1 to JF5) in four seamounts, JF2 and JF3 are located in two crests of the same seamount; (b) Bajo O'Higgins area that has 2 fishing grounds (BO1 and BO2) in one seamount and (c), Punta Sierra, which is a fishing ground on only one seamount (Figure 2). During these years the fishing hauls have been located to the north or north-east of the seamount, suggesting that some specific conditions could occur in those places that concentrate the spawning fishes.

FIGURE 2
Fishing hauls recorded from 1999 to 2003 over the seamounts located in Juan Fernandez and Bajo O'Higgins areas (upper image) and Punta Sierra (lower image)
Both images have a 30x vertical exaggeration.
FIGURE 2
FIGURE 2

Oceanographic knowledge of these three areas is scarce. However, the Juan Fernandez archipelago, which is the main fishing area, is located in a transition zone between warm equatorial waters and cool Sub-Antarctic waters (Silva 1985). Below this zone there is the intermediate Antarctic water mass characterized by 4–7°C temperatures, 34.3–34.5 ‰ salinity, 3–4 ml/l O2 and low nutrient contents. Of all hauls that have been made since 1999 97 percent have been located below 450 m in intermediate Antarctic waters (Figure 3).

FIGURE 3
Haul frequency and water masses by depth

FIGURE 3

Bottom temperature recorded from fishing hauls were made in 1999–2003 in the Juan Fernandez archipelago, Bajo O'Higgins and Punta Sierra and have fluctuated between 4 and 7.5°C and the most frequent values has been 5–6°C (Figure 4).

FIGURE 4
Haul frequency by bottom temperature recorded in the net

FIGURE 4

3.2 Fishing activities

The fishery fleet consisted of 5–8 'ice' trawlers but most orange roughy has been caught by only four vessels. The length of the vessels is 44 m. Total annual catches increased from 779 t in 1999 to 1 868 t in 2001 and then decreased to 1 514 t in 2002. For 2003, the preliminary catch is 1 246 t (Table 1). Total annual catches have always been lower than the TAC (Figure 5). In the first year, 48 percent of TAC remained uncaught. This uncaught TAC decreased to 6 percent in 2001 but increased to 50 percent in 2003.

TABLE 1
Total allowable catch (TAC) and annual catch officially reported by SERNAPESCA

Year TAC (t) Catch (t) TAC remains (t)
19991 500779721
20001 5801 48298
20012 1401 868272
20022 5001 514986
2003*2 5001 2461 254

* = Preliminary data.

FIGURE 5
Annual catch by fishing ground and TAC remaining at the end of the fishing season

Right ordinate -fishing effort

FIGURE 5

From 1999 to 2003, orange roughy have been caught mainly (84%) in the JF area, followed by PS (10%) and BO (6%). Relative importance of fishing grounds has changed through years. The most important fishing grounds have been JF4 (40%) in 1999–2000; JF1 (29%) and JF2 (25%) in 2001; JF1(37%), JF4 (20%) and PS (20%) in 2002; and JF3 (35%) in 2003.

The evolution of fishing effort through the years is the result of several interacting variables (Figure 5). In the first year there was obviously a learning component, some vessels searched for the resource in February, June and October, but the fishery season was concentrated between July and September and most fish were caught in August (65%). In 2000 trawling hours decreased and the fishing season started in May and finished in August, however most of fish (72%) were caught during June–July (Young et al. 2000). The maximum fishing effort was in 2002 and then decreased in 2003 because one vessel had operational problems and could not fish.

Catch rates increased from 5 tonnes/h in 1999 to 15 tonnes/h in 2002 and then fell to 6–4 tonnes/h in the last two years. This decline was observed for all the fishing grounds, except JF2 (Figure 6). Historically, the seasonal pattern of catch rates is strongly related with spawning behaviour. The best catch rates are achieved during June–July when the gonad index reached a maximum and the resource is most concentrated (Figure 7). Consequently, greater catches are made during these months. The length of the fishing season has fluctuated between 3 to 4 months in different years (Figure 8). The shortest season occurred in 1999 and the largest in 2002. The peak fishing season moved from 21–30 August in 1999 to 1–10 June in 2002 and then returned to 1–10 August in 2003.

FIGURE 6
Catch rate by fishing ground in Juan Fernandez area and total catch rate over years
FIGURE 6
FIGURE 7
Total catch, catch rate and gonadic index by month
FIGURE 7

FIGURE 8
Length of fishing season based on catch by 10-day periods

FIGURE 8

3.3 Reproductive behaviour

The annual reproductive cycle and maturity stages of fish concentrated on the fishing grounds have been studied by gonad index and macroscopic maturity scales, which were validated by histological analysis (Young et al. 2004). These authors postulated a spawning sequence by fishing grounds in Juan Fernandez archipelago: late May in JF1, late June in JF2, early July in JF4, and first half of August in JF3. There is enough information of gonad indices in JF2, JF3 and JF4 to support this spawning sequence, but there is not enough information for JF1 because there are no samples in July (Figure 9). Because of samples are spatially restricted to fishing operations, spawning timing on the other fishing grounds are unknown.

It has not been possible to estimate the age of 50% maturity because all samples consist of adult fishes. Juveniles were absent in the fishery. Using the relationship between the transition zone of the otoliths and age of first maturity (Francis and Horn 1997), Gili et al. (2002) made a preliminary estimation of age of first of maturity of 30 years for males and 32 years for females, which corresponds to a 30 cm fork length. However these estimations are biased because there are no juveniles in the samples.

Young et al. (2000) estimated an absolute fecundity of 33 669–379216 eggs and a relative fecundity of 16 056–115 944 eggs/kg body weight.

3.4 Length composition of catch

Females are larger than males in all fishing areas and there is an increasing of size from Juan Fernandez to Punta Sierra, indicating an increasing trend toward the coast (Figure 10). The distribution of fork lengths through years seems relatively stable, however some changes in the modal fork length have occurred (Figure 11). The overall male mode decreased 2 cm from 1999 to 2003, except in the Punta Sierra area. This decline is not clear in females because the mode decreased 4 cm from 1999 to 2001 but increased 3 cm in the last year; however the mode decreased 2 cm in Bajo O'Higgins and Punta Sierra.

FIGURE 9
Sequence of spawning by fishing ground based on gonadic index behaviour by 10-day periods

The sequence is JF1, JF2, JF4 and JF3.

FIGURE 9

The fraction of large males (arbitrarily fishes larger than 45 cm) decreased from 25 percent in 1999 to 10 percent in 2003 in Juan Fernandez area and occurred in all areas (Figure 12). The fraction of large females (fish larger than 50 cm) in Juan Fernandez area decreased from 24 percent in 1999 to 11 percent in 2003, but this decrease was smaller over the total area.

3.5 Age, growth and natural mortality

New growth models that included 2001 and 2002 data were fitted by Bayesian inference and produced L and k parameters similar to those estimated by Gili et al. (2002), but gave a smaller to value (Table 2 and Figure 13). Consequently, natural mortality estimated by different methods was similar to those estimated by Gili et al. (2002) (Table 3). In this work we did not use the method of Rikheter and Evanof (1976) and Roff (1988), which require the maturity curve or age of 50 percent maturity because these parameters are biased due to the absence of juveniles in the samples.

TABLE 2
Growth parameter models estimated in this work and by Gili et al. (2002)

Sex Parameter This work Gili et al.(2002) Prior distribution
MalesLoo43.459843.68U~(-inf,+inf)
 K0.0372990.037U~(-inf,+inf)
 to-0.63384-1.639N~(-0.44,0.9*-0.44)
FemalesLoo48.73349.0U~(-inf,+inf)
 K0.0305690.030U~(-inf,+inf)
 to-1.1324-1.97N~(-0.60,0.9*-0.60)
Both sexesLoo47.024447.51U~(-inf,+inf)
 K0.03200.030U~(-inf,+inf)
 to-0.6060-2.0N~(-0.3,0.9*-0.30)

FIGURE 10
Fork length composition of catch by sex, fishing areas and year

Vertical lines are shown only for helping visualization and have no biological meaning.

FIGURE 10

3.6 Stock assessment

Stock assessment models have not been applied to Chilean orange roughy for several scientific and management reasons. In the beginning of the fishery there was little knowledge of the biology of the resource and most of the information available came from other stocks around the world. Therefore, the Chilean Under-Secretary for Fisheries organized a workshop in 1999 in which fisheries entrepreneurs, IFOP researchers, managers and a panel of international experts composed by Andy Smith (Private consultant, New Zealand), Graham Patchell (Sealord Group, New Zealand) and Malcom Clark (NIWA, New Zealand) participated. The main recommmendations (Clark 2000) follow.

  1. Monitoring of the whole fishery season should be done by IFOP to collect biological and fishery information required for characterizing the catches.
  2. In the mid-term an acoustic survey should be designed and implemented. A preliminary hypothesis of number of stocks should be postulated based on spatial and temporal analysis of spawning stock distribution and structure by size/age.
  3. In the long term a stock assessment model should be developed based on at least four annual acoustic surveys and fishery and biological information of catches.

FIGURE 11
Mode of fork length by sex, fishing area and year

FIGURE 11

Following recommendations a series of studies have been made to monitor the fishery (Tascheri et al. 2001, 2002), to estimate fecundity and maturity (Young et al. 2000) and to estimate age and growth and natural mortality (Gili et al. 2002).

A technical Committee was established in 1999 with members of the fishing industry, IFOP researchers and managers, which has been undertaking fishery monitoring and quota discussions. The fishery industries have been cooperative and a medium term Collaborative Research Program (CRP) has been agreed to between government and fishery industries. Researchers contracted by the industry have been developing a low-cost acoustic monitoring system based on acoustic data recorded onboard fishing vessels. This system should be operative in 2004. They also contracted South African scientists (Ian Hampton and Dave Boyer) to conduct the first acoustic survey on Chilean orange roughy, which was done in 2003. The results will be soon available.

FIGURE 12
Proportion of males larger than 45 cm and females greater than 50 cm by fishing area and year

FIGURE 12

3.7 Resource management

The orange roughy fishery has been managed from the beginning. In accordance with Chilean fishery law in 1998 this fishery was declared an “Incipient Development Regime”. Therefore new fishing authorizations were no permitted and the rights for quota allocation by fishery industries were fixed by means of an auction. The fishing law states (<http://www.subpesca.cl/pagina%20juridica/ Leyes.htm>):

“An extraordinary fishing permit shall be granted to the successful bidders that shall give them the right to catch annually, for a 10-year period, a maximum amount equivalent to the result of multiplying the appropriate annual global catch quota by the fixed coefficient awarded in the respective fishery unit, and it shall begin to apply as of the calendar year following the year the award is made.

So that auctions shall be held during all the years this system is in effect, in the first public auction the total of the annual global catch quota shall be awarded for a period of ten years, granting extraordinary fishing permits with a variable coefficient, which shall decrease by ten percent every year. As of the second year the first award is in effect, ten percent cuts shall be auctioned annually, which shall be obtained during the first ten years through the proportional discount from all the successful bidders that possess the original extraordinary fishing permits granted for this purpose. The successful bidders at these auctions shall be granted extraordinary fishing permits with the same characteristics indicated in the preceding clause.”

Nine fishing companies were successful bidders at the first auction and five shared 99 percent of total quota: Pesca Chile, Friosur, El Golfo, Bio-Bio and Emdepes (Figure 14). However, in 2003 95 percent of annual catch was made by only three companies: Pesca Chile, El Golfo and Bio-Bio. El Golfo had a small catch in 2003 because its vessel could not fish due to operational problems. Emdepes never caught any orange roughy because of an agreement with another company to fish their quota. However, in the last year the other company was unable to catch their own quota and therefore did not catch Emdepes's quota either.

The annual quota is fixed each year and does not include any allocation by area. Thus, industries can catch their quota wherever and whenever they want. Annual quotas have not been estimated quantitatively. The first quota in 1999 was based on a qualitative analysis of the relative productivity of stocks in the world based on a precautionary approach (Young et al. 2000). Later quotas were increased to encourage industry cooperation with government in the fishery monitoring and research activities.

TABLE 3
Natural mortalities estimated in this work and by Gili et al. (2002)

Sex Method This work Gili et al. (2002) Observation
MaleRikheter & Evanof-0.66Requires age 50% maturity that is not available
 Alagaraja0.05780.058 
 Alverson & Carney0.10150.103 
 Roff-0.042Requires maturity curve that is not available
 Taylor0.03750.038 
 Chapman & Robson0.041 (0.037–0.046) 95% Confidence intervals in bracket
FemaleRikheter & Evanof-0.057Requires age 50% maturity that is not available
 Alagaraja0.04640.048 
 Alverson & Carney0.08090.084 
 Roff-0.031Requires maturity curve that is not available
 Taylor0.03120.031 
 Chapman & Robson0.046 (0.041–0.051) 95% Confidence intervals in bracket
BothRikheter & Evanof-0.058Requires age 50% maturity that is not available
 Alagaraja0.04890.049 
 Alverson & Carney0.08530.086 
 Roff-0.047Requires maturity curve that is not available
 Taylor0.0322  

FIGURE 13
Growth models fitted with Bayesian inference

(line= model; dotted line= Gili et al. (2002)

FIGURE 13

FIGURE 14
The 2003 quota allocation and annual catch by fishing industry

FIGURE 14

4. DISCUSSION

4.1 Distribution of fishing grounds and environment

The known distribution of Chilean orange roughy is restricted to seamount areas where fishes concentrate to spawn. Although 34 seamounts have been explored by five surveys, orange roughy have been found on only seven of them. So far no juveniles have been found in Chilean waters. This seems to be a worldwide characteristic of orange roughy fisheries, except for three locations where juveniles of 0 and 1 age classes were found: off the west coast of north Island, New Zealand at 250 m of depth; on the north Chatham rise of 50–175 km from the main spawning aggregation and on the Frankies ground in Namibia (Branch 2001). The occurrence of Chilean orange roughy in other habitats described for this species, e.g. canyons and soft flat bottom, have not been explored because this species is normally highly dispersed in such habitats.

Chilean orange roughy aggregations are distributed in waters of mainly at 4.0–6.5°C. This temperature range is similar to the 4.5–6.5 °C range described for worldwide orange roughy distribution (Clark 1999, Branch 2001). This tight range of water temperature is considered more important than depth in determining orange roughy occurrence (Branch 2001). Orange roughy off Chile are distributed mainly from 500 to 800 m, which is similar to the depth range of 600–800 m described for orange roughy off Namibia (Boyer and Hampton 2001), but more shallow than in Australia, New Zealand and the North-East Atlantic where orange roughy are distributed from 800 to 1200 m (Branch 2001). Chilean orange roughy are distributed mainly in Antarctic Intermediate Water as are orange roughy off New Zealand (Clark et al. 2000).

Small scale circulation studies of habitats where orange roughy spawn have not been done in Chile. However, such analysis are needed to understand why fishing hauls, and therefore fishable spawning aggregations, are always located to the North-East of seamounts. Gubbay (2003) recently reviewed this kind of study and summarized “seamounts can interrupt the flow of water and therefore affect the hydrography and current system in the vicinity as well as further afield. Tides can be amplified creating fast currents, and eddies may form and be trapped over seamounts in closed circulations known as Taylor columns. Other effects include locally enhanced turbulent vertical mixing and the creation of jets”.

4.2 Fishing activities

So far, commercial catch rates of Chilean orange roughy have not been standardized because the understanding of fishing behaviour and fishing tactics is too poorly known to show that catch rates are related to stock abundance. In the first year, fishermen were clearly learning how to fish orange roughy. In the second year with more experience, they increased their catch rates more than twofold and in the third year they increased them again. But, in the fourth and fifth years catch rates decreased to the levels recorded in the first year. Although the same catch rate patterns have been observed in orange roughy off Namibia (McAllister and Kirchner 2001 and 2002), New Zealand (Annala et al. 2002) and Australia (Wayte and Bax 2002), there are doubts that this decline is related to stock abundance. Four hypotheses have been postulated to explain the declining of abundances showed by commercial standardized catch rates and acoustic biomass in orange roughy off Namibia (McAllister and Kirchner 2002):

  1. Catch removal. The observed declines were mainly attributable to catches.
  2. Fishing. The observed declines resulted from the dispersion of orange roughy aggregation by fishing.
  3. Intermittent aggregation. The observed declines were caused by temporary factors unrelated to fishing. Orange roughy may aggregate on an intermittent basis depending on various environment conditions. Fish will re-aggregate on fishing grounds, but when they will do remains unpredictable.
  4. Mass emigration or mortality. The observed declines were caused by either mass mortality or mass emigration, and the original large abundance observed on the fishing grounds is unlikely to be re-established in the near future.

Recent Namibian experience on the Frankies fishing ground supports the ‘Fishing and, or, Intermittent aggregation’ hypotheses. Commercial fishing activities in Frankies started in 1995. Standardized commercial CPUE indices decreased from 4.028 (relative to mean value) in 1995 to 0.441 in 1998. The first acoustic survey in 1987 estimated a biomass of 29 567 tonnes. The second gave an estimate of 478 tonnes in 1998. Therefore, Frankies was closed to commercial fishing activities in April 1999, however only three years later the acoustic biomass estimate was 25 839 tonnes, i.e. a recovery to the level of 1987 (Brandão and Butterworth 2003).

4.3 Length composition of catch

Chilean orange roughy shares several worldwide characteristics in the length compositions of commercial catches. Frequencies are strongly unimodal, juveniles are absent, females more common than males and catch length frequencies do not change over time (Clark et al. 2000, Branch 2001).

Chilean orange roughy are the second largest in the world after those found in the North-East Atlantic. The fork-length range in Chilean orange catches was from 35 to 55 cm, which corresponds to 32–49 cm standard length. The orange roughy of the North-East Atlantic had a standardized size range of 40–58 cm which was greater than for orange roughy off New Zealand and Australia, which had a size range of 20–50 cm. The orange roughy off Namibia had a size range of 20–32 cm (Branch 2001).

4.4 Age, growth and natural mortality

Growth parameters fitted using prior distributions for to based on to values for orange roughy from New Zealand had similar parameters that were estimated by Gili et al. (2002). This means that the results from the approximation used by these authors, which included juvenile data from orange roughy off New Zealand, were comparable to the results obtained using Bayesian inference. It seems that the growth increments in adult fishes from both countries are similar, so the to parameters reflect mainly the juvenile growth patterns. The procedures that have been applied to estimate growth model are interim procedures until Chilean juvenile data become available.

Natural mortality values were estimated using different methods were similar, except to that of Alverson and Carney (1975). Values of M estimated by the Chapman and Robson (1960) method were similar to values estimated by methods based on growth parameters. Values of M for Chilean orange roughy are close to the value of 0.045 used in the New Zealand and 0.055 used for Namibian orange roughy stock assessments (Annala et al. 2002; Brandão and Butterworth 2003).

4.5 Stock assessment and management

So far, a complete stock assessment using a model specific for Chilean orange roughy has not been done. The main difficulties for stock assessment have been the short time series of fishery data, lack of knowledge of spatial and temporal stock dynamics, unreliable abundance indices based on catch rates and lack of acoustic surveys. Now, acoustic survey results will be soon available and sufficient information should be available to provide a stock assessment. However, the research focus should be on testing the perturbation hypothesis and producing reliable abundance indices from commercial catches and acoustic surveys.

5. LITERATURE CITED

Alagaraja, K. 1984. Simple method for estimation of parameter for assessing exploited fish stock. Indian J. Fish. 31:177–208.

Alverson, D. & M. Carney 1975. A graphic review of growth and decay of population cohort. J. Cons. Int. Explor. Mer. 36: 133–143.

Annala, J.H., K.J. Sullivan, C.J. O'Brien, N.W. McL Smith & S.J.A.Varian 2002. Report from the Fishery Assessment Plenary, May 2002: stock assessments and yield estimates. 640pp. (unpublished report held in NIWA library, Wellington).

Branch, T.A. 2001. A review of orange roughy Hoplostethus atlanticus fisheries, estimation methods, biology and stock structure. In A decade of Namibian Fisheries Science. A.Payne, A.I.L., Pillar, S.C. and R.J.M. Crawford (Eds). S. Afr. J .Mar. Sci. 23: 181–203.

Boyer, D.C. & I. Hampton 2001. Development of acoustic techniques for assessment of orange roughy Hoplostethus atlanticus biomass off Namibia, and of methods for correcting for bias. In A decade of Namibian Fisheries Science. A. Payne, A.I.L., Pillar, S.C. and R.J.M. Crawford (Eds). S. Afr. J .Mar. Sci. 23: 223–240.

Brandão, A. & D.S. Butterworth 2003. Stock assessment of Namibian orange roughy populations using an age-structured production model and all available indices of abundance from 1994 to 2002, and making allowance for annually variable aggregation of the stocks. Marine Resource Assessment & Management Group (MARAM). Department of Mathematics and Applied Mathematics. University of Cape Town. DWFWG/WkShop/Mar03/doc. 39pp.

Chapman, D.G. & D.S. Robson 1960. The analysis of a catch curve. Biometrics 16, 354–368.

Clark, M. 1996. Biomass estimation of orange roughy: a summary and evaluation of techniques for measuring stock size of deep-water fish species in New Zealand. J. Fish. Biol. 49 (Suppl. A):114–131.

Clark, M. 2000. Notes on research and management aspects of the developing Chilean Orange Roughy Fishery. Technical report. Undersecretary for Fisheries, Chile.

Clark, M.R., O.F. Anderson, R.I.C.C. Francis & D.M. Tracey 2000. The effects of commercial exploitation on orange roughy (Hoplostethus atlanticus) from the continental slope of the Chatham rise, New Zealand, from 1979 to 1997. Fisheries Research. 45: 217–238.

Doonan, I.J. 1994. Life history parameters of orange roughy; estimates for 1994. New Zealand Fisheries Assessment Research Document 94/19.

Dunn, A., R.I.C.C. Francis & I.J. Doonan 2002. Comparison of the Chapman-Robson and regression estimators of Z from Catch curve data when non-sampling stochastic error is present. Fisheries Research. 59(1–2):149–159.

Francis, C.R.I. & P.L. Horn 1997. Transition zone in otoliths of orange roughy (Hoplostethus atlanticus) and its relationship to the onset of maturity. Marine Biology 129: 681–687.

Gili, R., L.Cid, H. Pool, Z. Young, D. Tracey, P. Horn & P.Marriott 2002. Estudio de edad, crecimiento y mortalidad natural de los recursos Orange roughy y Alfonsino. Fondo de Investigaciones Pesqueras. FIP 2000-12. Instituto de Fomento Pesquero. 107pp + Figures, Tables and Annexes.

Gubbay, S. 2003. Seamounts of the North-east Atlantic. OASIS, Hamburg & WWFGermany, Frankfurt am Main. 38pp.

Lillo, S., R. Bahamonde, B. Leiva, M. Rojas, M.Barbieri, M. Donoso, R Gili., J. Cordova & G. Ojeda 1999. Prospección del recurso Orange Roughy (Hoplostethus sp.) y su fauna acompañante entre la I y X región. Instituto de Fomento Pesquero. FIP 98-05. 45pp + figures and tables.

McAllister, M. & C. Kirchner 2001. Development of Bayesian stock assessment methods for Namibian orange roughy Hoplostethus atlanticus. In A decade of Namibian Fisheries Science. A. Payne, A.I.L., Pillar, S.C. and R.J.M. Crawford (Eds). S. Afr. J .Mar. Sci. 23:241–264.

McAllister, M. & C. Kirchner 2002. Accounting for structural uncertainty to facilitate precautionary fishery management: illustration with Namibian Orange Roughy. Bulletin of Marine Science. 70(2): 499–540.

Montecinos, M., I. Payá & C. Canales 2003. Investigación CTP Orange roughy, 2003. Instituto de Fomento Pesquero. 50pp + tables and figures.

Rikheter, V.A. & V.N. Efanov 1976. On one of the approaches to estimation of natural mortality. ICNAF. Res.Doc. 76/VI/8: 12pp.

Roff, D.A. 1988. The evolution of migration and some life history parameters in marine fishes. Environ. Biol. 22(2): 133–146.

Silva, N. 1985. Algunas características físicas y quimicas de las masas de agua alrededor de las islas Robinson Crusoe y Santa Clara (Archipielago de Juan Fernández). 35–42 pp. In: Investigaciones marinas en el archipiélago de Juan Fernández. P. Arana [Ed]. Escuela de Ciencias del Mar. Facultad de recursos naturales. Universidad Católica de Valparaíso.

Taylor, C. 1958. Temperatura, growth and mortality The Pacific Cockle. J. Cons. CIEM. 26: 117–124.

Tascheri, R., J. Sateler, J. Merino, V. Ojeda, J. Olivares, R. Gili, R. Bravo, H. Miranda, C.Vera, L.Adasme & C. Bravo 2001 Programa de Seguimiento del Estado de Seguimiento de las Principales Pesquerías Nacionales. Investigación Situación Pesquería Demersal Centro Sur, 2000. Instituto de Fomento Pesquero. 153pp + 87 figures and 73 tables + annexes.

Tascheri, R., J. Sateler, J. Merino, O. Carrasco, J. González, E. Díaz, V. Ojeda, J.Olivares, R. Gili, R.Bravo & L. Cid 2003. Programa de Seguimiento del Estado de Seguimiento de las Principales Pesquerías Nacionales. Investigación Situación Pesquería Demersal Centro Sur Informe Final 2002. Instituto de Fomento Pesquero. 309pp.

Tracey, D.M. & P.L. Horn 1999. Background and review of ageing orange roughy (Hoplostethus atlanticus, Trachichthyidae) from New Zealand and elsewhere. N.Z. J.Mar. Freshwat. Res. 33:67–86.

Von Bertalanffy, L. 1938. A quantitative theory of organic growth. Human Biology 10: 181–213.

Wayte, S. & N. Bax (compls) . 2002. Stock assessment report 2002 orange roughy Hoplostethus atlanticus. South East Fishery Stock Assessment Group. 31pp + annex.

Young, Z., F. Balloting, J. Rivera, M. Ortego, R. Tascheri, M. Rojas & S. Lillo 2000. Estudio biológico-pesquero del recurso orange roughy. Instituto de Fomento Pesquero. Fondo de Investigación Pesquera. FIP. №99–05.

Young, Z., A. Zuleta & R. Tascheri 2000. Investigación CTP de orange roughy 2001. Technical report. Instituto de Fomento Pesquero. 30pp + tables + figures.

Young, Z. 2001. Investigación CTP de orange roughy 2002. Technical report. Instituto de Fomento Pesquero. 30pp + Annex.

Young, Z., E. Díaz, R. Bahamonde, R. Tascheri, Y. Muñoz, A. Olivares, M. I. Ortego, J.Rivera & J.Oliva 2004. Monitoreo y Prospección de orange roughy, 2001. Instituto de Fomento Pesquero. FIP № 2001–04. 127pp.


Previous Page Top of Page Next Page