Krill demography and large-scale distribution in the Western Indian Ocean sector of the Southern Ocean (CCAMLR Division 58.4.2) in Austral summer of 2006

ARTICLE IN PRESS Deep-Sea Research II 57 (2010) 934–947

Contents lists available at ScienceDirect

Deep-Sea Research II journal homepage: www.elsevier.com/locate/dsr2

Krill demography and large-scale distribution in the Western Indian Ocean sector of the Southern Ocean (CCAMLR Division 58.4.2) in Austral summer of 2006 So Kawaguchi a,n, Stephen Nicol a, Patti Virtue b, Stevie R. Davenport c, Ruth Casper b, Kerrie M. Swadling d, Graham W. Hosie e a

Australian Antarctic Division, and Antarctic Climate and Ecosystems Cooperative Research Centre, Channel Highway, Kingston, Tasmania 7050, Australia IASOS, University of Tasmania, Private Bag 77, HOBART, TAS 7001, Australia Beachcomber Research, PO Box 76, Swansea, Tasmania 7190, Australia d Marine Research Laboratories and School of Zoology, Tasmanian Aquaculture and Fisheries Institute, University of Tasmania, Private Bag 49, HOBART, TAS 7001, Australia e Australian Antarctic Division, Channel Highway, Kingston, Tasmania 7050, Australia b c

a r t i c l e in f o

a b s t r a c t

Article history: Received 30 June 2008 Accepted 30 June 2008 Available online 1 December 2009

Krill demography was studied during a large-scale survey of the South Western Indian sector of the Southern Ocean conducted in late January to late February 2006 (BROKE-West). The survey progressed from 30oE in late January to 80oE in late February and was bounded in the North by 621S. The average krill density calculated from the catches of the RMT 8 net for four geographical strata ranged from 0.8468 individuals per 1000 m-3, with overall mean of 6.7 individuals per 1000 m-3. Krill distribution and its population structure were analysed using cluster and mixture analysis of the length frequency distribution, showing small sized krill (modal size of 25-30 mm) broadly distributed in the centre of the western area whereas middle and large sized krill (modal size of 35-45 mm and 45-55 mm, respectively) mainly distributed in the middle and eastern half of the survey area. These findings are discussed in the context of the observed geographic and oceanographic structures; the continental slope area, fronts, currents and gyres. Proportional recruitment indices ranged 0.089-0.226 for R1 (age 1 + recruitment) and 0.204-0.440 for R2 (age 2+ recruitment) in the four strata. Recruitment analysis revealed differences between western and eastern halves of the surveyed area, suggesting the possible existence of separate self-sustaining populations linked to the Weddell Gyre and Prydz Bay Gyre systems. Krill larvae were mainly observed along the shelf break, implying the early onset of spawning in mid-November. The series of coastal polynyas developing in the early spring may be one of the important habitats allowing use of early season phytoplankton growth for early maturation and spawning, and those larvae would be transported westwards by the coastal current along the shelf slope. Crown Copyright & 2009 Published by Elsevier Ltd. All rights reserved.

Keywords: Antarctic krill Recruitment Density Distribution Sexual maturity Southern Ocean, Indian Ocean sector (301E to 801E)

1. Introduction Many studies on the ecology of Antarctic krill (Euphausia superba) have been carried out since the Discovery expeditions (Marr, 1962); however, most of the survey effort has been in the Southwest Atlantic sector (e.g., Watkins et al., 2003). Although mostly sporadic, there have, however, been studies reported from the Indian Ocean sector in recent years (Hosie, 1994; Pakhomov, 2000; Nicol et al., 2000a, 2000b). This paper describes the population structure of krill in the 30-801E region of the Southern Ocean from data collected during the BROKE-West (Baseline

n

Corresponding author. Tel.: + 61 3 62 323216; fax: + 61 3 62 323351. E-mail address: [email protected] (S. Kawaguchi).

Research on Oceanography, Krill and the Environment – West) survey in January-March 2006. The Discovery Investigations during the 1930s to 1960s provided a general outline of the distribution and abundance of Antarctic krill in the Indian Ocean sector, suggesting an uneven distribution of populations both latitudinally and meridionally (Marr, 1962; Mackintosh, 1972, 1973). The size structure and lifespan of krill in this sector (30-1001E) were investigated by Aseev (1984), which indicated onshore-offshore segregation of krill size classes, with the younger age classes found closest to the coast and the oldest year class generally north of 621S. The BIOMASS (Biological Investigations of Marine Antarctic Systems and Stocks) program of the early 1980s conducted surveys in the Indian Ocean sector that resulted in an estimate of acoustic krill biomass being produced which was subsequently

0967-0645/$ - see front matter Crown Copyright & 2009 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dsr2.2008.06.014

ARTICLE IN PRESS S. Kawaguchi et al. / Deep-Sea Research II 57 (2010) 934–947

used to set a catch limit in statistical Division 58.4.2 (Trathan et al., 1992). A number of ancillary studies were also undertaken on these surveys including demographics in the 52-641E region (Miller, 1985, 1986), 70-801E region (Prydz Bay) (Miquel, 1991), krill-environmental linkages (Hampton, 1985; Miller and Monteiro, 1988), krill-phytoplankton associations (El-Sayed and Hampton, 1980; Weber and El-Sayed, 1985) and physiological studies (Ikeda, 1984). Considerable research on krill was also carried out in the Southwest Indian Ocean sector, the subject of the current study, in the 1980s by scientists from the USSR. These studies included examination of the distribution, inter-annual variability, demography (Pakhomov, 2000) and mortality of krill (Pakhomov, 1995). The overall conclusions from this body of work were that there existed a self-sustaining grouping of krill in the Cooperation Sea (60-801E) south of the Antarctic Divergence. There have also been a number of studies of krill and the pelagic ecosystem in the Prydz Bay Region (Hosie et al., 1988; Hosie, 1991) which have described community structure and demographics of the krill population. Other studies have focussed on krill in restricted ¨ et al., 1990; portions of the Weddell Sea (Siegel, 1982; Bergstrom Daly and Macaulay, 1988; Siegel et al., 1990). Analysis of krill populations in the current survey area is complicated by the large scale circulation pattern that is thought to link stocks in the Southwest Atlantic and the Southwest Indian Oceans through the Lazarev Sea (Siegel, 2005; Siegel, 2006). The eastern sector of the Indian Ocean, CCAMLR Division 58.4.1 (80-1501E), was surveyed in 1996 (the BROKE – Baseline Research on Oceanography, Krill and the Environment - survey) (Nicol et al., 2000a, 2000b). Krill in the Southeast Indian Ocean sector were found associated with the cold waters of the coastal current, which corresponded to the findings of Mackintosh (1973), and the northward extension of the krill-based ecosystem in this area (Nicol et al., 2000a) corresponded to the KerguelenGaussberg stock identified by Mackintosh (1973). This paper reports on krill demography from the BROKE-West survey of CCAMLR Division 58.4.2 (30-80oE) which was carried out in the Austral summer of 2005/06 (Nicol et al., 2008, 2010). This survey was designed to produce results that are compatible with those from the BROKE survey 10 years previously, and with the results of the CCAMLR2000 survey of the South Atlantic

935

(Hewitt et al., 2004). The paper describes a number of features of the population structure of Antarctic krill including seasonal and geographical linkages between the krill population structure and physical and biological parameters measured on the voyage. This study especially focuses on the demography of krill. Analysis of other aspects of krill biology can be found in companion papers: krill growth and condition in relation to environments (Virtue et al., 2010) and acoustic estimates of distribution and abundance (Jarvis et al., 2010).

2. Material and Methods 2.1. Net sampling Krill were sampled using a RMT 1 +8 net with mesh sizes of 4.5 mm (RMT 8) and 315 mm (RMT 1) (Baker et al., 1973). Hard cod-ends were used at all times to keep the catch in good condition. An electro-mechanical net release and real time depth sensor was mounted above the net. The nets were equipped with flowmeters to calculate the volume of water filtered. The entire trawling operation was conducted at a ship’s speed of approximately 2.0 knots. Regular trawls were conducted at 50 pre-determined stations along every second transect in the survey area with closer spacing for sites southwards of the continental slope where changes in the biological and physical environment are most evident (Hosie et al., 2000). These trawls consisted of standard double oblique tows from the surface to a depth of 200 m, or to within approximately 15 m of the sea-floor in shallower water (Fig. 1). Wire speed was held at 0.7 to 0.8 m s 1 during paying out and 0.3 m sec 1 during hauling (approximately 0.5 m sec 1 and 0.2 m sec 1 respectively at vertical depth change rate). The net mouth angle is assumed constant during hauling within the speed ranges given above (Pommeranz et al., 1982). When the net reached maximum depth, the winch was stopped for about 30 seconds to allow the net to stabilise before retrieving. When hauling, the propeller thrust was turned off when the net reached depth of 15 to 20 m; this was to minimise the effects of the propeller action on the net operation and avoid damage to the samples.

Fig. 1. Location of net samples taken during the BROKE-West survey. Sampling commenced in the northwest corner of the survey area (301E, 621S) and progressed eastwards. The positions of regular net hauls and target trawls directed at acoustically-detected aggregations are indicated. The historical positions of the major frontal systems (from Orsi et al.,1995) are overlaid on the map. SBACC: Southern Boundary of Antarctic Circumpolar Current, SACCF: Southern Antarctic Circumpolar Current Front.

ARTICLE IN PRESS 936

S. Kawaguchi et al. / Deep-Sea Research II 57 (2010) 934–947

Target trawls were aimed at acoustically detected targets for determining the composition of these targets (69 tows in total). The position of identified targets was recorded, the ship was turned around and the ship’s track was set so that the required target could be trawled. The ship’s speed was kept below 2.0 knots before passing over the target, so that the net could be lowered to the desired depth. Fine adjustments were made throughout the trawl by monitoring the echo-sounder and the net was opened once it was estimated that the net was inside the acoustically detected target. 2.2. Sample processing 2.2.1. RMT 8 The total volume of the catch from every regular trawl was measured using the water replacement method to obtain quantitative data on the abundance of the catch (Harris et al., 2000). Whenever the volume of catch was 41l, the samples was randomly sub-sampled and measured. Since a portion of the krill caught by target trawling was also used for live krill experiments, only estimates of the number of individuals caught was made, and accurate volume measurements were not performed to minimize damage to krill through handling. All Antarctic krill in the catch (or a sub-sample if the volume of the catch was 41l and individuals mainly consisted of krill) were sorted out and the standard length (SL1) and carapace length (SL4) (Morris et al., 1988) of between 50 and 150 individuals were measured. Additionally, the length of the digestive gland (the longest axis) was measured using digital callipers for up to 50 individuals to obtain an index of condition (Nicol et al., 2004). The maturity stages of measured individuals were identified following Makarov and Denys (1981). Zooplankton, fish and squid were immediately sorted out from the catches, species identified, and their numbers were recorded (Van de Putte et al., 2010). 2.2.2. RMT 1 For the regular trawls, the whole sample, or if the sample volume was too large, a random sub-sample of a known proportion of catch was immediately fixed with 10% formalin seawater. In the laboratory, samples were split with a Folsom plankton splitter so at least 1000 individuals were counted per sample. Animals were identified to species where possible, and, in

the case of larval euphausiids to developmental stage (Kirkwood, 1982).

2.3. Data recording and analysis of population data The survey area was stratified into four areas; north and south of the 2500 m isobath (to delineate deep ocean and coastal areas including the shelf edge), and east and west at the 50oE line (Fig. 2). Krill numerical densities were calculated for each of the four strata by using simple arithmetic means (both including and excluding zero catches) using the software TRAWLCI (de la Mare, 1994a). Mean krill densities for the total area were calculated as sums of krill abundance in each stratum divided by the total area of that stratum. TRAWLCI software is the standard method used by CCAMLR to estimate densities from trawl samples (de la Mare, 1994a). This software was developed for calculating asymptotic confidence intervals for estimates of abundance obtained from standardised net surveys using likelihood ratios from Aitchinson’s delta distribution. The statistical distribution of net haul densities has to allow for an often substantial probability that a given haul will produce a zero estimate of density, especially for organisms forming discrete aggregation such as krill. The delta distribution consists of a discrete probability at the origin and a lognormal distribution for the non-zero observations (de la Mare, 1994a). Analyses by de la Mare (1994a) indicate that the abundance estimates appear to be unbiased. The spatial distribution of the krill population was analysed from the trawl data using cluster analysis to compare betweenstation similarities in size and maturity stage composition. To avoid any bias caused by stations with small sample sizes only the stations with 20 or more measured specimens were used in the cluster analysis. Both regular and target trawl samples were used for these analyses (Table 1). The hierarchical fusion of clusters was performed using Ward’s method to link homogeneous clusters, and the Euclidean distance coefficient was used for the similarity index (Postel et al., 2000). Krill recruitment was estimated by applying length-density data from routine trawls to the software CMIX (de la Mare, 1994b). The usual methods such as MacDonald and Pitcher’s (1979) method assume that length frequency data are representative of a population, with the frequencies in each length class

Fig. 2. Density distribution of post-larval krill from the regular net hauls. Positions of the fronts are determined during the survey. SBACC: Southern Boundary of Antarctic Circumpolar Current, SACCF: Southern Antarctic Circumpolar Current Front (see Williams et al., 2010).

ARTICLE IN PRESS S. Kawaguchi et al. / Deep-Sea Research II 57 (2010) 934–947

937

Table 1 The mean density of the entire survey area from the regular trawls calculated using the simple arithmetic mean, including and excluding zero-catches, and by the TRAWLCI method. Average density

TRAWLCI method

All samples N 1000 m Indian Ocean Sector BROKE-W (This study) Stratum 1 Stratum 2 Stratum 3 Stratum 4 All Strata BROKE (Nicol et al., 2000a, 2000b) Survey area West Survey area East Survey area All

3.46 2.37 0.84 67.44 6.69

Excluding zero catches SE

3

1.81 1.54 0.70 45.60 4.34

No. of Stations

N 1000 m

15 10 15 10 50

4.33 2.96 1.31 84.31 8.48

SE 3

4.33 1.34 0.99 56.02 6.28

Density within distribution range No. of Stations

15 10 15 10 50

N 1000 m

SE 3

4.61 1.97 2.23 81.67 8.71

3.15 1.41 1.57 76.23 5.45

Area of the stratum (x103 km2)

No. of Stations

15 10 15 10 50

627.0 109.9 566.7 95.8 1399.5

4.73 1.07 2.65

Cooporation Sea Pakhomov (2000) Hosie et al. (1988)

86.6 5.94

Cosmonaut Sea Pakhomov (2000)

58.8

0.84

Southwest Atlantic Sector CCAMLR 2000 (Siegel et al., 2000) Survey strata Antarctic Peninsula Scotia Sea South Sandwich Islands All strata

44.8 53.8 268.9 124.2

25.4 35.2 106.8 35.9

58.8 77.9 308.8 158.9

32.8 50.6 121 45.4

90.3 38.3 754.3 247.5

68.2 23.4 615.1 136.4

Bathymetric zone Oceanic Shelf

89.5 236.2

37 102.7

120.7 256.7

49.4 110.8

153.3 487.1

91 379.1

Net derived krill densities from comparable published surveys are also listed.

having Poisson distributions. As krill form discrete aggregations, Aitchinson’s delta distribution is assumed to take this statistical property into account; the CMIX Software was developed to do this (de la Mare, 1994b). Proportions of recruits were calculated for age 1 (R1) and age 2 (R2). Absolute recruitment was calculated by multiplying R1 by the krill density derived by TRAWLCI method.

3.1.2. Comparison between areas Densities (calculated by the three methods, above) within each area were divided into four strata by longitude and bathymetry (Fig. 3); they are also summarised in Table 1. The highest numerical densities were observed in Stratum 4 (eastern inshore area) recording a mean of 82 individuals 1000 m-3 using the TRAWLCI method. Mean densities in all the other three strata were less than 5 individuals 1000 m-3.

3. Results

3.1.3. Larvae Most of the krill larvae were caught along the shelf-slope (Fig. 4). All three calyptopis stages (CI, CII, and CIII), and a single furcilia stage (FII) were caught in the net survey, but no nauplii or metanauplii (the youngest stages) were caught (Table 2). Calyptopis larvae were all caught at stations in Strata 3 and 4. The stage most frequently caught was CII (6 stations), followed by CI (5 stations). CIII was only caught at one station with low density. Furcilia stage (FII) was caught at two stations, one in Stratum 1 and the other in Stratum 2. It takes krill embryos roughly 30-50 days to develop to CI-II and 75-85 days for them to reach FII (Ikeda, 1984). This means those calyptopis that were caught in mid- to late January were laid as embryos during midDecember to mid-January, whereas furcilia (caught in late January) were laid early to mid-November.

3.1. Distribution and abundance 3.1.1. Overall The numerical density of post larval krill calculated from the 50 net tows at predetermined trawl stations ranged from 0 to 401 individuals 1000 m-3. The mean density of the entire survey area was calculated by three methods: simple arithmetic mean including zero-catches, simple arithmetic mean excluding zerocatches, and TRAWLCI, yielding densities of 6.7, 8.5, and 8.7 individuals 1000 m-3 respectively (Table 1). Zero krill densities were recorded from 12 stations. 22 stations exhibited densities of less than 1 individual 1000 m-3, 12 stations had 1 to 10 individuals 1000 m-3, 2 stations had 10 to 100 individuals 1000 m-3 and 2 stations had densities over 100 individuals 1000 m-3. Trawls which recorded densities of 4 10 individuals 1000 m-3 were all from stations along the shelf break/slope. The two highest mean densities (401 & 268 individuals 1000 m-3) were recorded along 701E and 801E lines respectively (Fig. 2).

3.2. Age structure and recruitment Table 3 summarises the mean sizes of the age groups derived through CMIX analysis using length-frequency-distributions (LFD)

ARTICLE IN PRESS 938

S. Kawaguchi et al. / Deep-Sea Research II 57 (2010) 934–947

Fig. 3. The areas of four strata divided by bathymetry (2500 m) and the 55oE line.

Fig. 4. Distribution of larval krill from the RMT 1 net.

Table 2 Summary information of stations locations and densities where larval krill were caught in the RMT 1 net. Larval stages

Number of stations caught

Range of densities (inds. 1000 m-3)

Strata caught

CI CII CIII FII

5 6 1 2

53.7-9795.9 184.6-763.0 22.6 10.4-52.8

3&4 3&4 4 1&2

Table 3 Mean sizes and S.D. of age classes identified in each of the survey strata. Age class 1+

Stratum Stratum Stratum Stratum *

1 2 3 4*

2+

3+

4+

mean (mm)

(sd)

mean (mm)

(sd)

mean (mm)

(sd)

mean (mm)

(sd)

28.6 33.6 30.5 –

3.14 6.97 4.99

38.5 39.1 40.0 –

2.25 1.40 2.36

44.2 45.0 44.9 -

3.89 2.26 0.90

49.8 47.4 49.2 -

4.92 6.90 2.10

CMIX analysis unsuccessful.

ARTICLE IN PRESS S. Kawaguchi et al. / Deep-Sea Research II 57 (2010) 934–947

939

Table 4 Proportional and absolute recruitment indices in each of the four strata. Strata

Stratum1 Stratum2 Stratum3 Stratum4 a

Proportional recruitment

Density (TrawlCI)

R1

R2

(inds 1000 m

0.20 0.23 0.09 0.122a

0.31 0.20 0.44 1a

4.61 1.97 2.23 81.67

3

)

Absolute recruitment (inds 1000 m 0.922 0.453 0.201 9.96a

3

)

Area of the stratum (x103km2) 570.5 109.9 509.3 95.8

Derived as a simple proportion of o 33 mm to 433 mm krill due to an unsuccessful analysis using CMIX.

from regular trawls in each stratum. Analyses for Strata 1, 2 and 3 successfully converged. Since the output for Stratum 4 did not successfully converge, proportions of populations were not available for Stratum 4 through CMIX analysis. The krill population mainly consisted of four age groups (1+ , 2+ , 3+, and 4+ ), except for Stratum 4 which mostly contained size ranges which correspond to age 1+ and 2 +. Proportional recruitment indices (R1 and R2) and absolute recruitments (AbsR) in each stratum are summarized in Table 4. In both oceanic (Strata 1 and 3) and coastal strata (Strata 2 and 4), R1s were generally higher in the western strata (Stratum1 and 2) compared to the eastern strata (3 and 4). The highest R1 was observed in Stratum 2 (0.226). AbsR was highest in Stratum 1 (0.922) resulting from the higher numerical density in Stratum 1 compared to Stratum 2. However, R1 in one of the eastern strata (strata 3) was low (0.089) with an AbsR value of 0.201. R1 and AbsR for Stratum 4 were 0.122 and 9.96 respectively, but this value must be viewed with caution since the values were derived as a simple proportion of o33 to 4 33 mm krill due to an unsuccessful analysis using CMIX. Although this high value of AbsR for Stratum 4 was mostly driven by a single station out of 10 stations that showed an extremely high numerical density, the result generally suggest that stratum 4 had the highest absolute recruitment amongst all strata. 3.3. Maturity Fig. 5 shows composite length frequency distribution (LFD) for the developmental stages for each stratum. The smallest size group consisted almost exclusively of juvenile krill. The medium size group and the largest size group consisted of immature (sub-adult) and mature (adult) krill respectively, although their size ranges overlapped. Populations in the two western strata had relatively high proportions of mature krill compared to the two eastern strata. Stratum 4 especially seemed to lack mature krill in the larger size range. The detailed maturity structure (Fig. 6) indicates that the most abundant stage in each of the strata was either immature (subadult) males or females. In stratum 1, the population seems to be approaching the peak of their reproductive period, showing relatively even proportions of mature females in various ovarian developmental stages (3FB-3FD) with a smaller proportion of spent females (3FE). Both strata 2 and 3 were at the height of their maturity showing an increasing proportion of gravid (3FB), and spent (3FE) female krill stages. The population in stratum 4 had finished their reproductive season since only spent females and no gravid females were observed.

the size of the catch and the filtered water volume before the data were combined to provide composite LFDs (Fig. 9). Since the numerical abundance per filtered volume was not calculated for target trawls, the composite length frequency distribution was generated by simply averaging each of the 1mm size bins (Fig. 10). The cluster analysis for both regular and target trawls revealed the following three length frequency groups. Small-medium size clusters: C-AR and AC; medium size clusters: C-BR and BT; and large size clusters C-CR and CT for regular and target trawls, respectively. Fig. 11 describes the three size cluster distributions along transects derived only using data from regular trawls. This information was used to calculate the factors for converting integrated backscattering area to areal krill biomass density (Jarvis et al., 2010). Although the clusters’ LFDs are not directly comparable between regular and target trawls due to the difference in the method used to generate the composite LFDs, there are certain similarities between the results, especially the positions of the observed peaks and their size ranges. In the smallest clusters (C-AR and C-AT), C-AT had peaks between 27-29 mm range that matched with the main peak of C-AR. This size component did not appear in both other larger clusters from target trawls (C-BT and C-CT). The medium size clusters (C-BR and C-BT) both consisted of the main size component of 35-42 mm, although C-BT appeared to have larger components compared to C-BR. The largest size cluster from target trawls (C-CT) consisted of a population with a single peak at 48 mm, which also appeared in the regular trawls (C-CR). The size range of the two largest groups was similar (37-53 mm for C-CR and 42-55 mm for C-CT). Due to these similarities between the clusters resulting from regular and target trawls we classified the clusters as ‘‘Small’’ (C-AR and C-AT), ‘‘Medium’’ (C-BR and C-BT), and ‘‘Large’’ (C-CR and C-CT), and generated an integrated cluster distribution map (Fig. 12). The ‘‘Small’’ population distributed centrally in the area south of Southern Boundary of Antarctic Circumpolar Current (SBACC) where the Weddell Gyre had a strong influence. The northern margin of this group was neighboring the ‘‘Medium’’ population which extended further southeastwards. The ‘‘Medium’’ and ‘‘Large’’ population were also distributed closer to the continent. The ‘‘Large’’ populations distributed along the continental shelf, offshore of the eastern half of the survey area, where the Prydz Bay gyre has its influence. The ‘‘medium’’ population was mainly encountered in the eastern survey area, with some ‘‘medium’’ clusters distributed in between the SBACC and Southern Antarctic Circumpolar Current Front (SACCf).

4. Discussion 3.4. Length cluster analysis Three major clusters based on size at a dissimilarity level of 50% were identified for the regular trawls (Fig. 7) and for target trawls (Fig. 8). Each LFD from the regular trawls was weighted by

The populations of krill in the Southwest Indian Ocean in 2006 exhibited traits in common with other studies in this area and with the populations in adjacent areas of the Southern Ocean. The patterns of distribution of life history stages, however,

ARTICLE IN PRESS 940

S. Kawaguchi et al. / Deep-Sea Research II 57 (2010) 934–947

0.4

0.3 MAT

Stratum 1

0.35

IMM

JJ

0.3

J

02

0.25

Frequency

Frequency

MAT

Stratum 3

0.25

IMM

02 0.15

0 15 01

01 0.05

0.05

Length (mm)

65

62

59

56

53

50

47

44

41

38

Length (mm)

0.35

16 MAT

MAT

0.3

Strat m 2 Stratum

14

IMM J

Stratum 4

IMM J

12 Frequency

0 25 0.2 0.15 0.1

10 8 6

Length (mm)

65

62

59

56

53

50

47

44

41

38

35

32

29

26

23

65

61

57

53

49

45

41

37

33

29

0 25

0 21

2

17

0.05

20

4

17

Frequency

35

32

29

26

23

20

17

65

62

59

56

53

50

47

44

41

38

35

32

29

26

23

20

0 17

0

Length (mm)

Fig. 5. Composite length-frequency-distributions from the regular trawls in the four geographical strata. The population is broken down by maturity. MAT: mature; IMM: immature; J: juveniles.

were distinctly different from other regions that have been studied.

4.1. Overall Krill density A number of studies on net-derived krill density have been published in the past (Table 1). Reported densities from the Southwest Indian Ocean Sector are highly variable, but are usually at the lower end of the scale of reported densities. Reports from the Southwest Atlantic Sector, also indicate variability but densities are consistently higher (Siegel et al., 2004). Average densities for the four geographical strata in our study ranged from 0.84-67.4 (overall 6.7 individuals 1000 m-3), and these were mostly within, but towards the lower end of the previously reported range of densities reported from the same region (Cosmonaut Sea and Cooporation Sea) (Pakhomov, 2000). In fact the overall average of 6.7 for the current survey is mainly due to the high density in the shelf break area (Stratum 4) and if we exclude this value, the density range is 0.8-3.5 individuals 1000 m-3. The density range within the Prydz Bay area in our survey ranged from 0.01-401, with an average of 67 individuals 1000 m-3. It should also be noted that values reported in

Pakhomov (2000) result from a range of sampling gear and depths; net mouth area varies from 0.25 up to 30 m2 and tows sampled depths that sampled the surface (0-52 m) as well as deeper ones with 0-200 m, and the results included aimed trawls (see Table 1 of Pakhomov, 2000 for details). Most of the high densities reported in Pakhomov (2000) are generally those from net tows in the upper shallower layer, where krill densities are expected to be relatively high, whereas our net tows encompassed the entire top 200 m of the surface layer. An overall density of 3.2 individuals 1000 m-3 was observed during a krill survey in the Lazarev Sea (6oW-3oE) (Siegel, 2006), which was conducted just prior to the commencement of our survey (5 Dec 2005 - 2 Jan 2006). This area showed higher densities in the south-eastern part of the survey (around 67oS) (Siegel, 2006); similar to the pattern we observed in our Strata 1 and 2. The original BROKE survey in 1996 (Nicol et al., 2000a, 2000b), which was conducted immediately to the east of our study area, showed densities towards the lower end of our current range. There was one transect in common between the two surveys, although the timing differed. Transect-1 (801E line) of the BROKE survey which showed a density of 4.2 individuals 1000 m-3 (Nicol et al., 2000a, 2000b) was repeated during BROKE-West survey as

ARTICLE IN PRESS S. Kawaguchi et al. / Deep-Sea Research II 57 (2010) 934–947

0.4

Stratum1

Stratum3 ProPorption

ProPorption

0.3

941

0.2

0.1

0

0.3 0.2 0.1 0

Maturity Stage

Maturity Stage 0.4

Stratum4

Stratum2 ProPorption

ProPorption

0.2

0.1

0.1

03 0.2 01 0

0

Maturity Stage

Maturity Stage

Fig. 6. The detailed composition of maturity stages in each of the four strata from the regular trawls. Abbreviations in X axis correspond to definitions in Makarov and Denys (1981). J: juveniles; 2M: sub-adult males; 3MA: adult males IIIA; 3MB: adult males IIIB; 2F: sub-adult females ; 3FA: adult females IIIA; 3FB: adult females IIIB; 3FC: adult females IIIC; 3FD: adult females IIID; 3FE: adult females IIIE.

60

C CR C-CR

C AR C-AR

a-12 20 Sta

40

S 38 Sta-3 a-37 7 Sta

a-34 4 Sta

Sta a-72

3 Sta a-93

StaS -69

StaS -50

Sta a-12 22

Sta a-10 01

117 S Sta-1

Sta-5 S 52

19 Sta a-11

Sta-4 S 42

66 S Sta-6

a-79 9 Sta

a-116 Sta

S Sta--18

S Sta--17

11 S Sta-1

20

Stta-104

Stta-19

Height

C-BR

Fig. 7. Tree diagram resulting from the cluster analysis of length-frequency-distributions from regular trawls.

Transect 11 which recorded 32 individuals 1000 m-3. Transect 9 along 70oE line in BROKE-W was even higher with the average of 50 individuals 1000 m-3. Although these values may not be directly comparable due to temporal differences between surveys (late January in BROKE and late February in BROKE-West), the net-derived krill density along the 801E line was noticeably higher during BROKE-West. This difference is also reflected in the acoustic results, in which mean densities south of 63oS were

58 m2n mile-2 during BROKE (Pauly et al., 2000) and 110 m2n mile-2 during BROKE-West using Method 1 (Jarvis personal communication), which is broadly consistent with Pauly et al. (2000) and Hewitt et al. (2004), (Jarvis et al., 2010). Krill occurred throughout the survey area. The overall density in our study was an order of magnitude lower than that of the Southwest Atlantic sector. It is worth noting that even our highest average density, which was observed at one of the shelf regions

ARTICLE IN PRESS 942

S. Kawaguchi et al. / Deep-Sea Research II 57 (2010) 934–947

100

80

Height

60

20

0

Stta-6 Sta-84 71 Sta-7 47 Stta-4 74 Stta-7 a-46 Sta a-96 Sta Sta a-11 13 a-16 Sta a-28 Sta Sta-5 58 78 Sta-7 Sta-10 03 Sta-9 Sta-2 24 Sta--22 St a-70 Sta-27 Sta-40 S -118 StaSta-3 36 Sta a-12 25 Sta--26 Stta-100 Sta-60 Stta-68 Stta-85 Sta-64 a-67 Sta Stta-45 Stta-49 Sta-99 Sta-7 76

40

Fig. 8. Tree diagram resulting from the cluster analysis of length-frequency-distributions from target trawls.

0.15 C-AR

0.1

Proportion

Proportion

0.15

0.05 0

C-AT

0.1 0.05 0

15

20

25

30

35

40

45

50

55

60

65

15

25

Length (mm)

55

65

0.2 C-BR

0.15

Proportion

Proportion

45

Length (mm)

0.2

0.1 0.05

C-BT

0.15 0.1 0.05 0

0 15

20

25

30

35 40 45 Length (mm)

50

55

60

65

15

25

35 45 Length (mm)

55

65

0.25

0.15 C-CR

0.1

Proportion

Proportion

35

0.05

0.2

C-CT

0.15 1 0.05 0

0 15

20

25

30

35

40

45

50

55

60

65

Length (mm)

15

25

35

45

55

65

Length (mm)

Fig. 9. Composite length-frequency-distributions of the three clusters based on regular trawls. C-AR, C-BR, and C-CR.

Fig. 10. Composite length-frequency-distributions of the three clusters based on target trawls. C-AT, C-BT, and C-CT.

(Stratum 4), was in fact only at the same level of the mean density observed during the CCAMLR 2000 survey of the Southwest Atlantic (Siegel et al., 2004). These observations agree with the general impression that the average krill density (by what ever technique is used) in the Indian Ocean Sector is lower than that

observed in the Southwest Atlantic Sector (Nicol et al., 2000a, 2000b; Atkinson et al., 2004; Siegel, 2005). The key distribution area for krill is generally thought to be along the shelf break/slope area and this is reflected in higher mean krill densities, and a concentration of research activities and

ARTICLE IN PRESS S. Kawaguchi et al. / Deep-Sea Research II 57 (2010) 934–947

943

C-BR

C-BR

C-CR

C-AR

C-CR

C-CR C-BR

C-BR C-CR

C-AR

C-BR

Fig. 11. Distribution map of the clusters resulting from regular trawls. Amalgamated length frequency distributions for each cluster (C-AR, C-BR, and C-CR) are presented in Fig. 8.

M L

M

L

S M S

M

M

L

L M S

M

Fig. 12. Integrated krill cluster distribution. S: Small population (C-AR&AT), M: Medium population (C-BR&BT), L: Large population (C-CR&CT).

fishery operations (Kawaguchi et al., 1997; Kawaguchi and Nicol, 2007). Recent habitat analysis utilising net based data has indicated that krill are confined to the inner shelf at the southern Antarctic Peninsula, and further into the Atlantic sector they are abundant in open ocean. By contrast in the Prydz Bay region, concentrations occur just offshore of the shelf break (Atkinson et al., 2008). In our study, three of the sampling transects showed extremely high net-derived densities around the shelf break; however, this pattern was not observed on the other three transects and krill were also caught at the northern boundary of the survey. For the acoustic survey, which measured krill density on all 11 transects, the shelf-break zone did support a higher mean density than the rest of the survey area (62 gm-2 for shelfbreak zone and 10 gm-2 for the whole survey), but this was due mainly to noticeably higher densities around the shelf-break on transects 1, 9 and 11. Krill appear to be responding to certain bathymetric features through an interaction between bathymetry, surface currents and sea-ice dynamics. The shelf break provided the location for the highest densities of krill, but only in certain areas and it is likely that the factors determining areas of peak krill abundance are complex and multi-factorial (Jarvis et al., 2010). The suggested northern limit to the distribution of krill in the 80-1501E region was the SBACC (Nicol et al., 2000a), however in the 30-801E region, in this study, the krill population extended to the north of the SBACC, at least up to the SACCF. Although limited

sampling occurred north of the SACCF during BROKE-West, few krill were caught in this area raising the possibility that this frontal system may be the northern boundary of the krill population in the Southwest Indian Ocean sector. In the Southeast Indian Ocean, the SBACC and SACCF are closely aligned geographically (Bindoff et al., 2000) and it is possible that the krill population there also extends to the SACCF but that the sampling resolution during the BROKE survey was insufficient to discriminate between the two distributions. 4.2. Recruitment Proportional recruitment indices, R1 (0.089-0.226) and R2 (0.204-0.440) (Table 4), were both within the range of previous studies from the same area as well as those reported from other areas (Siegel and Nicol, 2000). It is difficult to discuss recruitment in detail with only a single year’s survey, however, it is worthwhile noting that recruitment indices showed stable values for at least two years (2004 and 2005 cohorts) in Strata 1 & 2 (the area mostly dominated by influence of the Weddell Gyre system). On the other hand Strata 3 and 4 seemed to have had low recruitment from the 2005 cohort. Two recent small-scale (100 x 100 km2) ecosystem studies conducted off east Antarctica (Nicol et al., 2008) indicated very low or no recruitment from year 2000 and 2002 cohorts (0.027 and 0.000, respectively, Kawaguchi, unpublished data).

ARTICLE IN PRESS 944

S. Kawaguchi et al. / Deep-Sea Research II 57 (2010) 934–947

Proportional recruitment from the 2004 and 2005 cohorts in the Lazarev Sea were reported to be 0.406 and 0.658 (Siegel, 2006), which were higher than we calculated for our Strata 1 and 2 but the results were in general agreement that both areas showed good recruitment from these two year classes. These observations strongly suggest that recruitment in Strata 1 and 2 is likely to be operating in the same system with the population in the Lazarev Sea, but is largely independent from Strata 3 and 4. Therefore, recruitment in the two systems (the Weddell Gyre and the Prydz Bay Gyre) seems to lack strong linkages and may be indicative of separate populations of krill. Proportional recruitment is calculated from the relative abundance of smaller krill in the net hauls (de la Mare 1994b). The assumption in this calculation is that the entire post-larval population is being adequately sampled. Inspection of the length frequency histograms produced from both regular and target tows indicates that across the entire survey area very few juveniles were sampled by the nets, but that more juveniles were caught in the regular trawls (Fig. 13). The surface layer is not sampled by the target trawls and is undersampled by double oblique trawls. Juvenile krill are generally known to distribute mainly within the top 10m of the water column (Pakhomov, 2000), and if this layer is systematically under-sampled, then this will result in erroneously low estimates of recruitment. Future studies on krill recruitment in the Indian Ocean sectors, and elsewhere, will need to investigate whether the net sampling regimes used to determine recruitment are producing unbiased samples of the entire population.

4.3. Larvae Pakhomov (2000) produced a conceptual diagram for the Cosmonaut and Cooperation Seas explaining the distribution pattern of larval and spawning stock accumulation. This diagram showed the spawning stock being accumulated along both sides of the Antarctic Divergence with larvae being distributed along both sides of the Divergence, but more distant from it. In our study the stations with highest larval density generally matched with the area of highest post-larval density. However, this does not necessary disagree with Pakhomov’s model since there is a obvious time lag (1-2 months) between the timing of spawning events and the development of the products of spawning to the calyptopis stage which we encountered Calyptopis larvae in our study were mostly caught at stations around the slope (2000-2500 m) (Fig. 4). Also along the slope there is a strong westwards coastal current and it is likely that larvae are transported from the Prydz Bay region to the west by this current (Hosie, 1994). These larvae may eventually be entrained into the Prydz Bay Gyral system and recruit back into the local population, however they might also be transported further west to recruit into the population in the downstream area. Further surveys later in the season would be required to clarify this. The observed distribution of larvae in 2006 was generally in agreement with the earlier findings of Hosie (1991), who indicated calyptopis larvae distribution being closely correlated with the surface currents, being transported westwards along the Mawson coast from Prydz Bay.

BROKE-West 160

Target

BROKE 500

Target

400

120

300 80 200 40

100 0

0 15 19 23 27 31 35 39 43 47 51 55 59 63 (mm)

15 19 23 27 31 35 39 43 47 51 55 59 63 (mm)

0

0

20

20

40

40

60

60

80 80 100

Regular

Regular

Fig. 13. Cummulative length frequency distribution of krill caught by target trawls and Regular trawls for the entire survey. Left panel: BROKE-West (current study), and right panel: BROKE survey (1996).

ARTICLE IN PRESS S. Kawaguchi et al. / Deep-Sea Research II 57 (2010) 934–947

4.4. Maturity and reproduction Krill in the eastern strata were more advanced in maturity compared to those in the west, and krill in the coastal strata were generally more advanced than those in the oceanic strata (Fig. 6). One of the most obvious explanations for the east-west maturity difference is the time lag in survey (since the survey progressed from west to east), but this observed difference may also have other causes, described later. The general concept of onshore-offshore krill distribution during the reproductive season is that krill in advanced maturity stages are distributed further offshore when compared to less mature ones (Siegel, 1988; Nicol et al., 2000a, 2000b). Interestingly, our observations in the Southwest Indian Ocean sector did not exhibit this pattern, especially in the western half of our survey area; along the coast populations were more advanced in maturity compared to oceanic ones. The same pattern was observed in the same season in the Lazarev Sea (Siegel, 2006). The main driving force for boosting population maturity is the phytoplankton bloom that occurs following the seasonal sea-ice retreat (Cuzin-Roudy and Labat, 1992; Kawaguchi et al., 2006a). Since seasonal sea-ice in the area under influence of the Weddell Gyre retreats later than that in the Prydz Bay Gyre (Williams et al., 2010), krill maturation could start earlier in the eastern area compared to the west, and in general progress from north to south. Interestingly, Antarctic krill females in the Prydz Bay region are also capable of spawning as early as the end of November as indicated by the presence of a few advanced developmental stages in January samples (Hosie, 1991). Siegel (2006) also observed calyptopis I in late December along the shelf slope in Lazarev Sea, which indicates onset of spawning as early as late November far towards south from the seasonal ice edge. Since krill are not thought to spawn under the ice (Cuzin-Roudy and Labat, 1992; Spiridonov, 1995), it is possible that the extensive polynya observed in October 2005 along the coast around 20oE (Williams et al., 2010) may have served as the habitat for this early spawning event, being the source of larvae transported by the coastal current along the shelf slope. Although levels could be moderate, earliest possible opening of ocean surface due to polynya formation allows early start of phytoplankton growth compared to ice covered regions (Arrigo and van Dijken, 2003; Schwarz et al., 2010), which may positively contribute to early krill maturity in those regions. Alternatively, possibilities of spawning events in icecovered regions, or other transport mechanisms of these larvae can not be completely denied. Clearly, more studies on krill condition during late winter to early spring is essential to further understanding of the process of early onset of their reproduction events. Krill growth, condition and biological parameters in relation to oceanographic environments during the survey are further discussed in Virtue et al. (2010). Furcilia larvae were caught in late January in the Cosmonaut Sea (Fig. 4b). This might also be explained by the influence of the local polynya (the Cosmonaut Polynya) providing an environment for early spawning. Alternatively, the furcilia, which is a slow developing stage, may represent age 1 + larvae spawned in the summer of the previous season as suggested by Melnikov and Spiridonov (1996) for the Western Weddell Sea. However, the Weddell Sea furcilia were at an advanced stage, and it is difficult to envisage that early stage furcilia (FII) in our study were individuals which had overwintered.

4.5. Population structure 4.5.1. Age composition The analysis of the length frequency density using the CMIX program in our study implied existence of four age groups (Age 1+, 2+, 3+, and 4+) (Table 3). Aseev (1984), from his study in the

945

Indian Ocean sector (30-1001E), was the first researcher to indicate that krill might have a life span of greater than five years. Pakhomov (1995) examined age structure of krill in the Indian Ocean sector using an extensive LFD dataset from the former Soviet Union. Through his analysis he consistently found four post larval age groups, and sometimes up to five to the north of the Antarctic Divergence in the Cooperation Sea, of which age 5+ individuals were later suggested to be from the previous years’ age 4+ populations in Cosmonaut Sea drifting into the area (Pakhomov, 2000). The mean size of four of the age groups in our study agreed with his analysis, although age 5+ individuals were not identified. There are some drawbacks to the method of determining age structure through mixture analysis of LFD (Pakhomov, 1995) due to possible variation in growth rate within a range of locations due to environmental differences (e.g. food concentration and temperature; Quetin et al., 1994; Pakhomov, 1995) Also, there is the potential for the mixture of different age classes in a same size range due to differences in the rates of growth and shrinkage depending on local food conditions (amount and types) during winter, and reduced growth rates towards the end of their lifespan (Nicol, 2000; Pakhomov, 1995). LFD analysis also assumes that the same population is being repeatedly sampled, but this is obviously difficult to verify for the open ocean, especially the Southern Ocean (Nicol, 2000). Moreover, there is now compelling evidence that male krill grow faster than females (Bargmann, 1945; Marr, 1962; Virtue et al., 1996; Kawaguchi et al., 2007), which means LFD analysis using the same growth functions for males and females is problematic (Kawaguchi et al., 2007). Recent attempts to measure krill growth using the instantaneous growth rate technique (a method to measure a direct growth rate of individual animals freshly caught at sea; Quetin and Ross, 1991; Nicol et al., 2000a, 2000b) is starting to provide data on the growth rate of individual krill of various size, which can be related to environmental conditions (e.g., Kawaguchi et al., 2006b; Atkinson et al., 2006). An initial attempt to generate a growth trajectory (Candy and Kawaguchi, 2006) using IGR growth model was successful, and further data collection through field experiments will help refine the model in relation to food environment and difference between sexes. Extensive IGR experiments were conducted on board during the BROKE-West survey, and are discussed in detail in Virtue et al. (2010). Measured growth rates of adult krill showed both seasonal and regional differences.

4.5.2. Population structure and oceanography Length frequency distributions derived through cluster analysis showed substantial variation across the survey area. The integrated cluster map is overlaid on the conceptual current field derived by the shipboard ADCP during the survey (for details see Williams et al., 2010) (Fig. 14). The area where the ‘‘Small’’ population, which was dominated by juvenile krill, occurred coincided almost exactly with the area where the influence of Weddell Gyre dominated. On the other hand the eastern half of the area was dominated by ‘‘medium’’ and ‘‘large’’ krill, whereas, closer to the continent where the coastal current dominates only ‘‘medium’’ krill were found. The life cycle of krill in the Prydz Bay Gyre is tightly linked with the circulation pattern with small to medium, less mature krill residing in the centre and the mature spawning stock distributed in the outer regions. This suggests that a stationary circulation pattern south of the Antarctic Divergence (SBACC) in the Cooperation Sea may act as a nursery ground for the Prydz Bay region krill stock (Pakhomov, 2000). This distribution pattern was also clearly observed in this study showing a large population generally

ARTICLE IN PRESS 946

S. Kawaguchi et al. / Deep-Sea Research II 57 (2010) 934–947

Small

Large Large

Medium

Medium

Large Medium

Medium

Cosmonaut Small Sea

Cooperation Sea L Large

Medium

Large Small

Medium

Prydz Bay Gyre (Williams et al ., 2008) Outer eastern limb of the Weddell Gyre (Williams et al ., 2008) S th Southern ACC ffrontt (SACCf) off ACC (Williams et al ., 2008) Southern Boundary of ACC (SB) (Williams et al., 2008) Antarctic Slope Current, associated with the Antarctic Slope Front (Williams et al., 2008)

Fig. 14. Integrated cluster map is overlaid on the conceptual current field derived by the shipboard ADCP (Williams et al., 2010).

distributed in the oceanic area, with a ‘‘medium’’ population towards the center of Prydz Bay Gyre. The difference in the distribution of the different sized krill together with the likely disconnection of recruitment between the two gyral systems (Hunt et al., 2007) as observed in our study strongly supports the idea that the Prydz Bay Gyre is acting as a means of providing a stable habitat for the regional population (Pakhomov, 2000).

5. Conclusions The current study outlined krill density, distribution, recruitment, reproduction, and population structure in the Southwest Indian Ocean in relation to some of the oceanographic features. A strong relationship between krill population structure and oceanography is suggested, which supports the existence of separate, possibly self sustaining populations in the Weddell Gyre and Prydz Bay Gyre systems as suggested in earlier studies. The series of coastal polynyas developing in the early spring could be one of the important mechanisms for providing habitats for early maturation and spawning.

Acknowledgments The authors acknowledge the efforts and support of the colleagues on board, in particular J. Kitchener, L. Finley, J. Foster, M. Brown, T. Yoshiki, A. Van de Putte, T. Yoshida, and the officers and crew of the Aurora Australis. Thanks are extended to K. Meiners and B. Pasquer for their constructive comments on the manuscript. D. Smith and A. Bender are appreciated for generating maps. This work forms part of Australian Antarctic Science Project 2655 and is also a contribution towards the output of the Antarctic Marine Ecosystems Program of the Antarctic Climate and Ecosystems Cooperative Research Centre funded by the Australian government’s Cooperative Research Centres Programme. References Arrigo, K.R., van Dijken, G.L., 2003. Phytoplankton dynamics within 37 Antarctic coastal polynya systems. Journal of Geophysical Research 108 (C8), 27, doi:10.1029/2002JC001739.

Aseev, Y.P., 1984. Size structure of krill populations and life span in the Indian Ocean sector of the Antarctic. Hydrobiological Journal 20 (60), 89–94. Atkinson, A.A., Siegel, V., Pakhomov, E., Rothery, P., 2004. Long-term decline in krill stock and increase in salps within the Southern Ocean. Nature 432, 100–103. Atkinson, A., Shreeve, R.S., Hirst, A.G., Rothery, P., Tarling, G.A., Pond, D.W., Korb, R.E., Murphy, E.J., Watkins, J.L., 2006. Natural growth rates in Antarctic krill (Euphausia superba): II. Predictive models based on food, temperature, body length, sex, and maturity stage. Limnol Oceanogr 51, 973–987. Atkinson, A., Siegel, V., Pakhomov, E.A., Rothery, P., Loeb, V., Ross, R.M., Quetin, L.M., Schmidt, K., Fretwell, P., Murphy, E.J., Tarling, G.A., Fleming, A.H., 2008. Circumpolar oceanic habitats for Antarctic krill. Marine Ecology Progress Series. Baker, A.C., Clarke, M.R., Harris, M.J., 1973. The N.I.O. combination net (RMT 1+ 8) and further developments of Rectangular Midwater Trawls. Journal of the Marine Biological Association of the United Kingdom 53, 176–184. Bargmann, H.E., 1945. The development and life-history of adolescent and adult krill, Euphausia superba. Discovery Report 23, 103–176. ¨ ¨ Bergstrom, B.I., Hempel, G., Marschall, H.-P., North, A., Siegel, V., Stromberg, J.-O., 1990. Spring distribution, size composition and behaviour of krill Euphausia superba in the western Weddell Sea. Polar Record 26 (157), 85–89. Bindoff, N.L., Warner, M.J., Rosenberg, M.A., 2000. On the circulation and water masses over the Antarctic continental slope and rise between 80 to 1501E. Deep-Sea Research II 47, 2299–2326. Candy, S.G., Kawaguchi, S., 2006. Modelling growth of Antarctic krill. II. Novel approach to describing the growth trajectory. Marine Ecology Progress Series 306, 17–30. Cuzin-Roudy, J., Labat, J.P., 1992. Early summer distribution of Antarctic krill sexual development in the scotia-Weddell region: a multivariate approach. Polar Biology 12, 65–74. Daly, K.L., Macaulay, M.C., 1988. Abundance and distribution of krill in the ice edge zone of the Weddell Sea, austral spring 1983. Deep-Sea Research 35 (1), 21–41. de la Mare, W.K., 1994a. Estimating confidence intervals for fish stock abundance estimates from trawl surveys. CCAMLR Science 1, 203–207. de la Mare, W.K., 1994b. Estimating krill recruitment and its variability. CCAMLR Science 1, 55–69. El-Sayed, S.Z., Hampton, I., 1980. Phytoplankton/krill investigations in the southwester Indian sector of the Southern Ocean. Antarctic Journal USA 1980 Rev, 143–144. Hampton, I., 1985. Abundance, distribution and behaviour of Euphausia superba in the Southern Ocean between 15o and 30oE during FIBEX. In: Siegfried, W.R., Condy, P.R., Laws, R.M. (Eds.), Antarctic nutrient cycles and food begs. Springer-Verlag, Berlin, pp. 294–303. Harris, R.P., Wiebe, P.H., Lenz, J., Skjoldal, H.R., Huntley, M., 2000. In: ICES zooplankton methodology manual. Academic Press, London, pp. 1–684. Hewitt, R.P., Watkins, J., Naganobu, M., Sushin, V., Brierley, A.S., Demer, D., Kasatkina, S., Takao, Y., Goss, C., Malyshko, A., Brandon, M., Kawaguchi, S., Siegel, V., Trathan, P., Emery, J., Everson, I., Miller, D., 2004. Biomass of Antarctic krill in the Scotia Sea in January/February 2000 and its use in revising an estimate of precautionary yield. Deep-Sea Research II 51 (12–13), 1215–1236. Hosie, G.W., 1991. Distribution and abundance of euphausiid larvae in the Prydz Bay region, Antarctica. Antarctic Science 3, 167–180. Hosie, G.W., 1994. Multivariate analyses of the macrozooplankton community and euphausiid larval ecology in the Prydz Bay region, Antarctica. ANARE Reports 137, pp 209.

ARTICLE IN PRESS S. Kawaguchi et al. / Deep-Sea Research II 57 (2010) 934–947

Hosie, G.W., Ikeda, T., Stolp, M., 1988. Distribution, abundance and population structure of the Antarctic krill (Euphausia superba Dana) in the Prydz Bay region, Antarctica. Polar Biology 8, 213–224. Hosie, G.W., Schultz, M.B., Kitchener, J.A., Cochran, T.G., Richards, K., 2000. Macrozooplankton community structure off East Antarctica (80–1501 East) during the Austral summer of 1995/96. Deep-Sea Research II 47, 2437–2463. Hunt, B.P.V., Pakhomov, E.A., Trotsenko, B.G., 2007. The macrozooplankton of the Cosmonaut Sea, east Antarctica (301E–601E), 1987–1990. Deep-Sea Research I 54 (7), 1042–1069. Ikeda, T., 1984. Development of the larvae of the Antarctic krill (Euphausia superba Dana) observed in the laboratory. Journal of Experimental Marine Biology and Ecology 75, 107–117. Jarvis, T., Kelly, N., Kawaguchi, S., van Wijk, E., Nicol, S., 2010. Acoustic characterisation of the broad-scale distribution and abundance of Antarctic krill (Euphausia superba) off East Antarctica (30–801E) in January–March 2006. Deep-Sea Research II 57 (9–10), 916–933. Kawaguchi, S., Nicol, S., 2007. Learning about Antarctic krill from the fishery. Antarctic Science 19 (2), 219–230. Kawaguchi, S., Ichii, T., Naganobu, M., 1997. Catch per unit effort and proportional recruitment indices from Japanese krill fishery data in Subarea 48.1. CCAMLR Science 4, 47–63. Kawaguchi, S., Yoshida, T., Finley, L., Cramp, P., Nicol, S., 2006a. The Krill maturity cycle: a conceptual model of the seasonal cycle in Antarctic krill. Polar Biology 19 (2), 219–230. Kawaguchi, S., Candy, S.G., King, R., Naganobu, M., Nicol, S., 2006b. Modelling growth of Antarctic krill. I. Growth trends with sex, length, season and region. Marine Ecology Progress Series 306, 1–15. Kawaguchi, S., Finley, L.A., Jarman, S., Candy, S.G., Ross, R.M., Quetin, L.B., Siegel, V., Trivelpiece, W., Naganobu, M., Nicol, S., 2007. Male krill grow fast and die young Marine Ecology Progress Series 345, 199–210. Kirkwood, J.M., 1982. A guide to the Euphausiacea of the Southern Ocean. ANARE Research Notes Number 1, 45pp. Marr, J.W.S., 1962. The natural history and geography of the Antarctic krill (Euphausia superba Dana) Discovery Report 32, 33–464. MacDonald, P.D.M., Pitcher, T.J., 1979. Age-groups from size-frequency data: a versatile and efficient method of analysing distribution mixtures. Journal of the Fisheries Research Board of Canada 36, 987–1001. Mackintosh, N.A., 1972. Life cycle of Antarctic krill in relation to ice and water conditions. Discovery Reports 36, 1–94. Mackintosh, N.A., 1973. Distribution of post-larval krill in the Antarctic. Discovery Reports 36, 95–156. Makarov, R.R., Denys, C.J., 1981. Stages of sexual maturity of Euphausia superba Dana. BIOMASS Handbook 11, 1–11. Melnikov, I.A., Spiridonov, V.A., 1996. Antarctic krill under perennial sea ice in the western Weddell Sea. Antarctic Science 8 (4), 323–329. Miller, D.G.M., 1985. The South African SIBEX-I Cruise to Prydz Bay region, 1984 IX: Krill (Euphausia superba Dana). South African Journal of Antarctic Research 15, 33–41. Miller, D.G.M., 1986. Results from biological investigations of krill (Euphausia superba) in the Southern Ocean during SIBEX-I. Memoirs of the National Institute of Polar Research Special Issue 40, 117–139. Miller, D.G.M., Monteiro, P.M.S., 1988. Variability in the physical and biotic environment of the Antarctic krill (Euphausia superba Dana), south of Africa: some results and a conceptual appraisal of important interactions. In: Sahrhage, D. (Ed.), Antarctic Ocean and Resources Variability. Springer-Verlag, Berlin Heidelberg, pp. 245–257. Miquel, J.C., 1991. Distribution and abundance of post-larval krill (Euphausia superba Dana) near Prydz Bay in summer with reference to environmental conditions. Antarctic Science 3 (3), 279–292. Morris, D.J., Watkins, J.L., Ricketts, C., Buchholz, F., Priddle, J., 1988. An assessment of the merits of length and weight measurements of Antarctic krill Euphausia superba. British Antarctic Survey Bulletin 79, 27–50. Nicol, S., 2000. Understanding krill growth and aging: the contribution of experimental studies. Canadian Journal of Fisheries and Aquatic Science 57 (Suppl. 3), 168–177. Nicol, S., Pauly, T., Bindoff, N.L., Wright, S., Thiele, D., Hosie, G.W., Strutton, P.G., Woehler, E., 2000a. Ocean circulation off east Antarctica affects ecosystem structure and sea-ice extent. Nature 406, 504–507. Nicol, S., Kitchener, J., King, R., Hosie, G., de la Mare, W.K., 2000b. Population structure and condition of Antarctic krill (Euphausia superba) off East Antarctica (80–1501E) during the Austral summer of 1995/1996. Deep-Sea Research II 47, 2489–2517. Nicol, S., Virtue, P., King, R., Davenport, S.R., McGaffin, A.F., Nichols, P., 2004. Condition of Euphausia crystallorophias off East Antarctica in winter in comparison to other seasons. Deep-Sea Research II 51, 2215–2224. Nicol, S., Clarke, J., Romaine, S.J., Kawaguchi, S., Williams, G., Hosie, G.W., 2008. Krill (Euphausia superba) abundance and Ade´lie penguin (Pygoscelis adeliae) breeding performance in the waters off the Be´chervaise Island colony, East Antarctica in two years with contrasting ecological conditions. Deep-Sea Research II. SO-GLOBEC Special Issue Part II 55, 540–557.

947

Nicol, S., Raymond, B., Meiners, K., 2010. BROKE-West, a large ecosystem survey of the South West Indian Ocean sector of the Southern Ocean, 30–801E (CCAMLR Division 58.4.2). Deep-Sea Research II 57 (9–10), 693–700. Orsi, A.H., Whitworth III, T., Nowlin, W.D.J., 1995. On the meridional extent and fronts of the Antarctic Circumpolar Current. Deep-Sea Research 42, 641–673. Pakhomov, E.A., 1995. Demographic studies of Antarctic krill Euphausia superba in the Cooperation and Cosmonaut Seaas (Indian sector of the Southern Ocean). Marine Ecology Progress Series 119, 45–61. Pakhomov, E.A., 2000. Demography and life cycle of Antarctic krill, Euphausia superba, in the Indian sector of the Southern Ocean: long-term comparison between coastal and open-ocean regions. Canadian Journal of Fisheries and Aquatic Science 57 (Suppl. 3), 68–90. Pauly, T., Nicol, S., Higginbottom, I., Hosie, G., Kitchener, J., 2000. Distribution and abundance of Antarctic krill (Euphausia superba) off East Antarctica (80–1501E) during the Austral summer of 1995/1996. Deep-Sea Research II 47, 2465-2488. Pommeranz, T., Herrmann, C., Kuhn, A., 1982. Mouth angles of the Rectangular Midwater Trawl (RMT1+ 8) during paying out and hauling. Meeresforschung 29, 267–270. Postel, L., Fock, H., Hagen, W., 2000. Biomass and abundance. In: Harris, R.P., Wiebie, P.H., Lenz, J., Skjoldal, H.R., Huntley, M. (Eds.), ICES Zooplankton Methodology Manual. Academic Press, London, pp. 83–192. Quetin, L.B., Ross, R.M., 1991. Behavioral and physiological characteristics of the Antarctic krill, Euphausia superba. American Zoologist 31, 49–63. Quetin, L.B., Ross, R.M., Clarke, A., 1994. Krill energetics: seasonal and environmental aspects of the physiology of Euphausia superba. In: El-Sayed, SZ (Ed.), Southern Ocean Ecology-the BIOMASS perspective. Cambridge University Press, Cambridge, pp. 165–184. Siegel, V., 1982. Investigations on krill (Euphausia superba) in the Southern Weddell Sea. Meeresforschung 29, 244–252. Siegel, V., 1988. A concept of seasonal variation of krill (Euphausia superba) distribution and abundance west of the Antarcitc Peninsula. In: Sahrhage, D (Ed.), Antarctic Ocean and resources variability. Springer, Berlin Heidelberg New York, pp. 219–230. Siegel, V., 2005. Distribution and population dynamics of Euphausia superba: summary of recent findings. Polar Biology 29, 1–22. Siegel, V., 2006. Demography of Antarctic krill in the Lazarev Sea (Subarea 48.6) in the 2005/06 season. WG-EMM-06/10. Siegel, V., Nicol, S., 2000. Population parameters. In: Everson, I (Ed.), Krill: Biology ecology and fisheries. Blackwell, pp. 103–149. ¨ ¨ Siegel, V., Bergstrom, B., Stromberg, J.O., Schalk, P.O., 1990. Distribution, size frequencies and maturity stages of krill, Euphausia superba, in relation to the sea-ice in the northern Weddell Sea. Polar Biology 10, 549–557. Siegel, V., Kawaguchi, S., Ward, P., Litvinov, F., Sushin, V., Loeb, V., Watkins, J., 2004. Krill demography and large-scale distribution in the southwest Atlantic during January/February 2000. Deep-Sea Research II 51, 1253–1273. Spiridonov, V.A., 1995. Spatial and temporal variability in reproductive timing of Antarctic krill (Euphausia superba Dana). Polar Biology 15, 161–174. Schwarz, J.N., Raymond, B., Williams, G.D., Pasquer, B., Marsland, S.J., Gorton, R.J., 2010. Biophysical coupling in remotely–sensed wind stress, sea surface temperature, sea ice and chlorophyll concentrations in the South Indian Ocean. Deep-Sea Research II 57 (9–10), 701–722. Trathan, P.N., Agnew, D.J., Miller, D.G.M., Watkins, J.L., Everson, I., Thorley, M.T., Murphy, E.J., Murray, A.W.A., Goss, C., 1992. Krill biomass in Area 48 and Area 58: recalculation of FIBEX data. CCAMLR Selected Scientific Papers 1992, 157–181. Van de Putte, A.P., Jackson, G.D., Pakhomov, E, Flores, H., Volckaert, F.A.M., 2010. Distribution of squid and fish in the pelagic zone of the Cosmonaut Sea and Prydz Bay region during the BROKE-West campaign. Deep-Sea Research II 57 (9–10), 956–967. Virtue, P., Nichols, P.D., Nicol, S., Hosie, G., 1996. Reproductive trade-off in male Antarctic krill, Euphausia superba. Marine Biology 126, 521–527. Virtue, P., Kawaguchi, S., McIvor, J., Nicol, S., Wotherspoon, S., Brown, M., Casper, R., Davenport, S., Finlay, L., Foster, J., Yoshida, T., Yoshiki, T., 2010. Krill growth and condition in Western Indian Ocean sector of the Southern Ocean 30–801E in austral summer 2006. Deep-Sea Research II 57 (9–10), 948–955. Watkins, J.L., Hewitt, R., Naganobu, M., Sushin, V., 2003. The CCAMLR 2000 Survey: a multinational, multi-ship biological oceanography survey of the Atlantic sector of the Southern Ocean. Deep-Sea Research II, 1205–1456. Weber, L.H., El-Sayed, S.Z., 1985. Spatial variability of phytoplankton and the distribution and abundance of krill in the Indian Ocean sector of the Southern Ocean. In: Siegfried, W.R., Condy, P.R., Laws, R.M. (Eds.), Antarctic Nutrient Cycles and Food Webs. Springer-Verlag, Berlin, Heidelberg, pp. 284–293. Williams, G.D., Nicol, S., Bindoff, N.L., Aoki, S., Meijers, A., Marsland, S.J., Klocker, A., Iijima, Y., 2010. Surface oceanography of BROKE-West, along the Antarctic Margin of the South-West Indian Ocean (301-801E). Deep-Sea Research II 57 (9–10), 738–757.