Suitability of cage culture for Pacific abalone Haliotis discus hannai Ino production in China

Authors


Correspondence: G Zhang, Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, China. E-mail: gfzhang@qdio.ac.cn

Abstract

Pacific abalone (Haliotis discus hannai Ino) aquaculture is a thriving industry in China. This study describes a novel submerged cage culture system for abalone rearing in Fujian, South China. The cage consisted of five vertical slots that were oriented perpendicular to the flow of water. The slots were separated by six vertically connected plastic plates for abalone attachment and shelter at the bottom of the cage. Experiment 1 was designed to determine the appropriate stocking density at the start of the abalone sea-based production cycle. Eight-month-old hatchery reared and size-graded juveniles were transferred to the sea-based culture system. For different stocking densities, shell length of juveniles obtained in this novel culture system on 2, 3.5 and 5 months, respectively, was compared with shell lengths obtained in a traditionally multi-tier basket culture system. In Experiment 2, daily growth rates (DGRs) in shell length and biomass in terms of wet weight of 2-year-old abalones reared in cage and tiered basket culture systems were compared over a 6-month period. Results of Experiment 1 showed that growth of abalone in the cage culture system is density-dependent; the mean final shell length of juveniles obtained was 6.7–15.9% higher than in tiered baskets system even at the same initial stocking density. In Experiment 2, DGRs in shell length of 53.83–78.38 μm day−1 obtained in cage system were significantly higher than that in tiered baskets (< 0.01). And in terms of wet weight biomass, it was 1.48–3.01 times higher in the cage system compared with the traditional system. Abalone survival was more than 87.5% in both culture systems in both experiments. Advantages of the newly established cage culture system included better growth performance of the animals reared and potential improvement of rearing conditions, such as improved water flow velocity and dissolved oxygen.

Introduction

Pacific abalone (Haliotis discus hannai Ino) aquaculture is a thriving and prosperous industry in China, producing more than 20 000 metric tons annually since 2007 (China Bureau of Fisheries 2008). In recent years, the industry has performed exceptionally well in Fujian, South China, owing to the positive heterosis from cross-breeding of stocks from Japan and China (Zhang, Que, Liu & Xu 2004; Deng, Liu, Wu & Zhang 2008) and improvements in long-distance transport techniques. As a result, the production of abalone in Fujian accounts for about two-thirds of the total Chinese output (China Bureau of Fisheries 2008). The majority of the production is reared in suspended multi-tier baskets (Wu, Liu, Zhang & Wang 2009), a method that was developed from a land-based farming system for H. diversicolor supertexta (Chen & Lee 1999).

The commercial production of Haliotidae worldwide is generally carried out in land- and sea-based culture systems. As a result of the high cost of water pumping, aeration, filtration and temperature control, land-based production systems are less energy efficient, although land-based water recirculation technology for abalone production is being developed in New Zealand and Korea (Davis & Carrington 2005; Park, Kim & Jo 2008). In sea-based production systems, substrates that afford shelter for abalone (e.g., net cages, barrels or plastic multi-tier baskets) are suspended below the water surface from floating rafts (Aviles & Shepherd 1996; Capinpin, Toledo, Encena & Doi 1999; Fermin & Buen 2002; Wu et al. 2009). These culture systems are popular in China, Thailand and the Philippines because of their simple design, ease of construction and low investment (Jarayabhand & Paphavasist 1996; Capinpin et al. 1999; Wu et al. 2009). However, husbandry and maintenance tasks, such as cleaning, feeding, measuring and observing the abalones must be performed by pulling these culture vessels out of the water. Abalone production in these sea-based systems is therefore highly labour-intensive, which becomes a significant problem where labour is in short supply or expensive.

In addition to labour shortages, the industry faces a number of other challenges. Rapid expansion of rearing facilities in the inner coastal bays in Fujian has led to the overcrowding of floating rafts, which reduces water flow, thereby affecting the growth and survival of abalone (Fleming et al. 1997; Searcy-Bernal & Gorrostieta-Hurtado 2007; Wassnig, Roberts, Krsinich & Day 2010). In addition, there are significant issues with the biofouling of the rearing baskets. Oysters, sea squirts, sea mosses and other organisms settle on the mesh of the plastic tiers and rapidly block water flow through the baskets (Wu & Zhang 2010). The lines on which the baskets are suspended tend to tangle in areas with high water velocity, such as the mouth of a bay, thereby limiting the areas in which abalone can be reared.

The development of floating cages submerged in offshore waters has made cage culture possible for finfish (e.g., salmonids) in Europe and Canada (Korsoen, Dempster, Fjelldal, Oppedal & Kristiansen 2009; Johnston, Keir, M. & Power 2010) and large yellow croaker Pseudosciaena crocear in China (Lu, Xu & Haegen 2008). This culture technique combines high yield, easy husbandry and low labour and energy costs, and is appealing to fish farmers (Mortensen, Toften & Aas 2007). Our objective was to evaluate the utility of a novel attachment substrate for the culture of Pacific abalone in floating cages. We compared the growth of Pacific abalone H. discus hannai that were reared in the new culturing system and in traditional multi-tier baskets.

Materials and methods

A prototype design for an abalone cage culture system

We used a prototype cage that was previously built for large yellow croaker culture in Ningde Bay, Fujian, China (Lu et al. 2008). The cage consisted of a 330 × 330 × 280 cm (L × W × D) rectangular frame made of cypress, with styrofoam floats attached to all four sides. We modified the basic design slightly based on the farming and biological and behavioural needs of this species. The fixed top of the cage was removed so that abalones could be easily loaded into it and subsequently fed. The topless cage was held rigid and then partitioned into 12 smaller netted cages 80 × 80 × 280 cm (Fig. 1a), each suspended from four wooden boards crossed over the original cage frame. In contrast to the single netting used around cages for large yellow croaker, we placed two nets around each small cage, one with a mesh size of 1 cm for holding juvenile abalones, and the other with a 2 cm mesh for holding adults. This design was intended to minimize harassment by fish.

Figure 1.

Configuration of the deployed novel cage and traditional basket culture systems. (a) represents a whole frame consisting of 12 net cages, (b) represents a typically used shelf-like device for abalone sheltering, (c) represents a six-tiered basket traditionally used for abalone farming in south China and (d) represents a tier in the basket culture system.

Special substrates were designed for abalone attachment and shelter in the new culture system. Six plates with several round holes were arranged into a vertical shelf-like device separated by connecting tube-shaped rods, made of PVC with nuts and bolts at each end (Fig. 1b). Each shelf plate was ladder-shaped, with an area of 0.645 m2. The total shelf surface area available for abalone attachment in each cage was 7.622 m2. Abalones attached themselves to the vertical surfaces of the shelf-like device, oriented perpendicular to the water flow. One shelf-like device was placed on the bottom of each small cage, yielding 12 devices in each large 330 cm × 330 cm × 280 cm frame.

Abalones were also reared in the traditional multi-tier baskets (Fig. 1c) for comparison. Each basket consisted of six vertically connected tiers (Fig. 1d), the inside surface of a tier being 0.486 m2 (Chen & Lee 1999; : Wu et al. 2009). The six-tier basket was not set in floating netting cages, but instead hung by lines directly below the seawater surface. Following the locally accepted guidelines for suspending multi-tier baskets, and avoiding the possible tangling of the lines in basket culture, a 330 × 330 × 280 cm frame can house 24 abalone farming baskets. The number of baskets per frame was reduced to 6–12 for frames placed at the mouth of Ningde bay, which is characterized by higher water velocity.

The experiments

The experiments were conducted in Ningde Bay at the Ningde Aquaculture Experimental Station of the Modern Agro-industry Technology Research System, between November 2009 and May 2010 to evaluate the suitability of cage culture for Pacific abalone (Fig. 2).

Figure 2.

Map showing the locations of the experimental sites mentioned in the text.

Experiment 1 was designed to determine the appropriate stocking density at the start of the abalone sea-based production cycle, using 8-month-old juveniles. The experiment was performed at Changyao, in Ningde Bay over a 5-month period between 1 January and 31 May 2010. In Experiment 2, we assessed the growth of animals reared in the prototype cages compared with those in traditional multi-tier baskets, during a 6-month culture period from 1 December 2009 to 1 June 2010. This experiment was performed at two sites, Changyao and Dongchong in Ningde Bay. Changyao is located in the inner part of Ningde Bay, where water velocities are believed to be relatively low, whereas Dongchong is at the mouth of Ningde Bay where water velocities are higher and traditional culture of abalone is rare.

Ambient water temperature at the two sites was measured daily, whereas salinity and pH were measured weekly during the experimental period. In addition, we measured water velocity and dissolved oxygen at the Changyao site during Experiment 2. Water flow velocity (cm s−1) was recorded in both culture systems using two continuous data loggers (ALEC Compact-EM; ALEC, Kobe, Japan). The data were combined into 1-h intervals for analysis of a complete tidal cycle (25 h). Similarly, a dissolved oxygen (DO) data loggers (YSI 556; YSI, Yellow Springs, OH, USA) was placed into each of the culture systems and recorded dissolved oxygen levels (mg L−1) every 2 h. Limited to the experimental conditions of our laboratory and experimental station, the DO and water velocity data loggers were borrowed from the Key Laboratory of Ocean Circulation and Wave, Chinese Academy of Sciences and a local working station in Fujian and were deployed for one complete tidal cycle in mid-January, 2010. We used a paired sample t-tests to determine whether there were significant differences in the two environmental variables between the cage and basket cultures (< 0.05).

Experiment 1: The effect of stocking density

The animals used in Experiment 1 were obtained from a cohort produced at a hatchery in Rongcheng, Shandong Province (N37°10′06″, E122°35′44″). After transport to Changyao (N26°40′29.71′′, E119°48′27.45′′) in December 2009, the 8-month-old juveniles were graded for size by shell length. After allowing 20 days for acclimation to the new water conditions, abalones with mean shell length of 28.15 ± 1.56 mm were selected for the experiment, which began on 1 January 2010.

Density was defined as the percentage of unit surface area of the substrate occupied by animals. The total surface area of a shelf-like device in a culture cage was 7.622 m2, which compared with 0.486 m2 for a single tier in the tiered baskets. The cages were stocked at one of five density: 6.53%, 9.80%, 13.07%, 16.33% and 19.60%, corresponding to about 1200, 1800, 2400, 3000 and 3600 individuals per shelf-like device in each cage respectively. Three replicates were assembled for each stocking density in the culture cages, and a total of 15 net cages were used in the experiment. To minimize the effects of cage location (within the farm) on abalone growth, the cages were distributed randomly within three neighbouring frames (330 × 330 × 280 cm) in the mid-region of the abalone farm. In the tiered baskets (controls), 75 animals were stocked per tier, equal to a density of 6.53% (the rearing density typically used in the traditional culture system). We stocked three replicate baskets consisting of 18 tiers and these were placed adjacent to the experimental cages.

Before transport, juvenile abalones were fed a formulated diet and dry Porphyra spp. powder in the hatchery. After transport and size-grading, the animals were fed fresh Gracilaria lemaneiformis prior to February 2010, followed by fresh Laminaria japonica in excess every 4 days during the remainder of the experiments. The change in diet, as is typically practised by the abalone farming industry in Fujian, was due to changes in the availability of different seaweeds at various times. The seaweed feed was cut into pieces for abalone foraging in the traditional baskets, whereas whole strings of the seaweed were placed in the slots between the vertical plates in the new prototype cages.

Shell length was recorded at 2, 3.5 and 5 months by measuring 75 animals from each of two cage replicates, selected at random. Animals in the control groups were measured in the same way and at the same time. Measurements were taken using digital vernier callipers to an accuracy of 0.01 mm. The measurements of each replicate were accomplished within 10 min to minimize disturbance. We compared growth in terms of mean shell length in Experiment 1, rather than daily growth rate in shell length (μm day−1), between the two culture systems as the abalone were not individually marked. Data from the replicates were pooled as we found no evidence of a significant difference among the replicates (two-way ANOVA,< 0.05). We used a one-way ANOVA and least square means post hoc comparison to evaluate the effect of stocking densities using SPSS 10.0. Data were transformed to a natural logarithm before analysis to increase the homogeneity of variances and the normality of the residual variation (Sokal & Rohlf 1981).

Experiment 2: Comparison of abalone growth between cage and basket culture

Animals used in Experiment 2 were from a cohort obtained locally in Fujian, China. All the animals came from a single hatchery reared cohort produced in November 2007. After being reared in a hatchery for 8 months, the abalone seeds were transferred to the sea-based multi-tier basket culture system and reared for an additional period prior to the experiment. These 2-year-old animals were graded into three size groups (A–C) by wet weight in November 2009 (Table 1). After grading, the animals were transferred to culture cages at Changyao. Twelve replicates of each size group were reared in separate cages, randomly assigned to three 330 × 330 × 280 cm frames. The initial stocking densities in the cages were the same as that recommended for commercial production in multi-tier baskets: 10.15%, 8.42% and 12.82% for abalones of sizes A, B and C respectively (Table 1). Cage culture was also conducted at Dongchong using animals from the same cohort, at the same initial stocking density as at Changyao.

Table 1. Initial body sizes and stocking density of Pacific abalone reared in Experiment 2
Size groupInitial body sizeDensity (%)a Cage or tier
Shell length (mm)Wet weight (g)
  1. a

    Density (%) was defined as the percentage of unit surface area of the substrate occupied by animals.

A60.77 ± 2.4845.20 ± 4.2810.15
B55.34 ± 3.0137.54 ± 4.018.42
C48.29 ± 2.1331.76 ± 3.6512.82

Following local traditional multi-tier basket culture practices, 24 multi-tier baskets were held in each frame at Changyao. At Dongchong, where water flow rates are believed to be higher, only 12 baskets were held in each frame to prevent tangling of the cage lines. In Experiment 2, animals were set up in multi-tier baskets at the same initial stocking density at both sites, to serve as controls. Twenty-four baskets containing animals from each size group (A to C) were placed at random into the three frames as controls at the Changyao site. Similarly, 12 baskets containing animals from each size group (A to C) were placed at random into three frames as controls at the Dongchong site. For each size group, the initial stocking density at the two sites was the same as in the culture cages (Table 1). Experimental cages and multi-tier baskets in this experiment were also deposited in adjacent frames in the mid-region of the abalone farms.

As diet affects shell colour in abalones, the animals were fed different types of algae to generate shell colour markers for identification of initial shell length but not initial wet weight. According to local protocols, dry L. japonica is fed to abalone on summer and autumn days when fresh G. lemaneiformis is of low yield and high price. Animals that are fed dry L. japonica develop a greenish colour in the recently deposited sections of their shell. On 1 December 2009 when the experiment was initiated, the abalone were switched to a diet of fresh G. lemaneiformis, which results in a brownish colour in the shell that is deposited thereafter. The change in shell colour from green to brown provides a marker of initial length. Experimental animals were then fed with fresh G. lemaneiformis before February 2010 and fresh L. japonica in excess every 4 days during the rest of Experiment 2. The seaweed feeding regime in the two culture systems was the same as in Experiment 1. Daily growth rate (DGR, μm day−1) was calculated using the following formula

display math

where LT and Lt are final and initial shell lengths of 100 animals from each of two cages and two baskets selected at random. We tested for differences in DGR between experimental groups using an analysis of multi-factor variance using a General Linear Model (GLM; SAS Institute 1999). Sources of variance in DGR were partitioned into replicate, culture site (Changyao or Dongchong), size group (A, B, or C), culture system (cage or tiered basket) and the interactions among these factors. Differences were considered significant if < 0.05. If no significant differences existed among replicates, replicates were combined and data were retested using GLM in variance analysis.

As shell colour marker cannot be used to determine initial wet weight, daily growth rate in term of individual wet weight cannot be calculated. Instead, wet weight biomass was calculated for each specific size group in each frame at both sites, Changyao and Dongchong. The wet weight of animals in one cage or one basket was multiplied by the number of cages or baskets in a frame, to estimate the biomass in each frame. We used a paired sample t-tests to compare the biomass of each frame between the cage and basket cultures (< 0.05).

Results

Experiment 1

More than 98% of the abalones survived the 5-month experimental period at all density levels. At the end of the culture period, the mean shell length of abalones in the various cage treatments ranged between 39.45 ± 4.32 and 45.57 ± 4.21 mm, and compared with an average of 39.30 ± 4.36 mm for abalones cultured in multi-tier baskets (Table 2).

Table 2. Mean shell lengthsa (mm ± SD) of juvenile abalones stocked at varying initial densities at each sampling time in Experiment 1
Culture systemDensity (%)bSL (mm ± SD)
Initial2 months3.5 months5 months
  1. a

    Within each row, means with the same letter are not statistically different (> 0.05).

  2. b

    Definition of density (%) is the same as in Table 1.

Cage6.5328.15 ± 1.56a33.41 ± 3.32a38.82 ± 3.52a45.56 ± 4.21a
9.8028.15 ± 1.56a31.82 ± 2.75b36.33 ± 2.92b42.69 ± 3.92b
13.0728.15 ± 1.56a31.54 ± 2.90b34.35 ± 3.46c40.94 ± 4.11c
16.3328.15 ± 1.56a31.20 ± 2.68c33.85 ± 3.69c39.52 ± 4.03d
19.6028.15 ± 1.56a30.56 ± 3.27d33.14 ± 3.66d39.44 ± 4.33d
Tier (Control)6.5328.15 ± 1.56a31.31 ± 2.37c34.11 ± 3.06c39.30 ± 4.36d

We observed a significant decrease in growth as rearing density increased (Table 2; F5, 1994 = 14.665, 22.197 and 19.855, < 0.001). In cage cultured animals, the highest shell length (45.57 ± 4.21 mm) occurred in animals that were held at a density of 6.53%, compared with that of 39.45 ± 4.32 mm in animals held at a density of 19.60%. For animals held at the same stocking density (6.53%), the mean shell length achieved was 6.7–15.9% higher in cage cultured animals than for those reared in the baskets at the end of the experiment (Table 2).

Experiment 2

Daily growth rate in shell length

More than 87.5% of abalones from the three size groups (A, B and C) survived throughout the experimental period in Experiment 2 at both sites. The culture system had a significant effect on growth rate (Fig. 3), but there was no significant interaction between culture system and size group (< 0.001, Table 3). The animals grew more rapidly in cage culture (53.83 ± 23.34 to 78.38 ± 19.30 μm day−1) than in the tiered baskets (52.48 ± 18.35 to 63.72 ± 21.42 μm day−1) at both sites. Culture site and size group also affected growth rate (three-way analysis of variance < 0.001). DGR was higher at Dongchong than at Changyao and was also positively correlated with size. The interactions between culture sites and size group, and among sites, size group and culture systems were significant (< 0.05), whereas the remaining interaction terms were not significant (> 0.05).

Figure 3.

Daily growth rates (DGRs, μm day−1 ± SD) of the three size groups (A, B and C) under various conditions.

Table 3. Variance analysis of daily growth rates (DGRs) based on a general linear model in Experiment 2
Sourced.f.M.S.F-value P
System166 157.84136.93<0.0001
Group226 002.7953.82<0.0001
Group × system2472.270.980.3766
Site138 013.6678.68<0.0001
Site × group21823.173.770.0233
Site × system11175.662.430.1191
Site × group × system21725.713.570.0285

Biomass in terms of wet weight

The biomass of cage cultured animals was 1.48–1.52 higher than that of basket cultured animals at both sites. The difference was highly significant at the termination of the experiment for all size groups (Fig. 4; < 0.001). The range of biomass was 319.77 ± 11.91 kg to 385.63 ± 14.66 kg at Changyao and 329.25 ± 12.03 kg to 390.40 ± 19.12 kg at Dongchong for cage cultured animals, and between 220.32 ± 12.96 kg to 257.4 ± 15.38 kg and 114.12 ± 6.48 kg to 129.66 ± 8.93 kg, at Changyao and Dongchong, respectively, for basket cultured animals.

Figure 4.

Pacific abalone biomass in wet weight (kg ± SD) obtained after 6-month culture in cage (Cage) and multi-tier basket (Tier) culture systems. A, B and C represents the size group A, B and C.

Environmental variables

Ambient water temperature at the two sites ranged from a low of 10.8°C in February to a peak of 21.5°C in June. Salinity and pH ranged from 29.43 to 31.35 and 7.83 to 8.13 respectively. There was a significant difference in water velocity and dissolved oxygen between the two culture systems at the Changyao site (< 0.0001, Figs 5 and 6).

Figure 5.

Water flow velocity (cm s−1 ± SD) in the cage (Cage) and tiered baskets (Tier) culture systems throughout a complete tidal cycle.

Figure 6.

Dissolved oxygen (mg L−1) in the cage (Cage) and tiered baskets (Tier) culture systems throughout a complete tidal cycle.

Discussion

This study evaluated a new containment system for holding and culturing large quantities of Pacific abalone H. discus hannai. Key to the success of a cage culture system is the provision of a suitable substrate for abalone attachment. The design of this prototype takes the behavioural patterns of H. discus hannai into consideration and tried to simulate their position and feeding habits in nature. Abalones prefer areas where the current is high and often move about to find such locations (Hayakawa, Watanabe & Kittaka 1987; Tissot 1992). Water quality on the flat bottom of suspended plastic baskets or boxes was often impacted by high levels of metabolic wastes in abalone culture (Jarayabhand & Paphavasist 1996). The structure of the shelf-like device, with its vertical slots and arranged plates, allow for higher levels of water flow than the traditional closed baskets. The better growth and high survival performance of the animals tested in the newly established culture system demonstrated the suitability of cage culture for H. discus hannai, compared with that in tiered baskets.

Aviles and Shepherd (1996) summarized daily growth rates (DGRs) in shell length of Haliotidae animals reared in sea-based culture systems. Without exception, substrates for abalone sheltering, such as net cages, barrels or plastic multi-tier baskets reported previously are suspended below the water surface from floating rafts. Compared with the small (30–50 mm length, width and height) and relative closed system used before, our cage system showed a more open and larger culture space for H.discus hannai. The DGRs of 53.83–78.38 μm day−1 obtained in the new cage culture system were higher than the 27–107 μm day−1 reported by Aviles and Shepherd (1996), Nie, Ji and Yan (1996) and Alcantara and Noro (2006), but lower than the 67.08–135.75 μm day−1 reported previously by our research team (Wu et al. 2009). Maybe the animals from the varying ageing groups contributed to the lower growth rates recorded in the present experiments. Experimental evidence showed most of the energy budget for the young goes to somatic growth, and to respiration for adults in H. tuberculata (Peck, Culley & Helm 1987) or gonad development in H. asinina (Capinpin & Corre 1996). Results from Experiment 1 in this study showed a relative higher growth rate (about 116.97 μm day−1, from 28.15 ± 1.56 mm to 45.56 ± 4.21 mm, at a density of 6.53%) during a 5-month experimental time than that in a previous study (3-month period, 105.19 μm day−1, from 28.11 ± 1.11 mm to 39.26 ± 3.47 mm, at a similar density of 7.04%, Wu et al. 2009). Considering the ambient temperature in this study was 10.8–21.5°C compared with 12.6–22.8°C during the previous study, the newly tested cage culture system is promising.

As individual tagging of abalone is difficult (Kube, Appleyare & Elliott 2007), we changed shell colour to delimit growth periods of abalone. The validity of using a change in shell colour by changing the diet is subject to discussion. Under all conditions, the change should be instantaneous, distinct and occur uniformly in all abalone. In Experiment 2, with a relative low stocking density of 8.42–12.82%, a distinct colour change occurred in all the size-graded animals after 2 days on the new diet. Furthermore, our previous study revealed no significant difference (> 0.05) between the initial shell lengths measured at the beginning of experiment and those remeasured based on the colour change on termination of the experiment (Wu et al. 2009), demonstrating the validity and accuracy of the colour change to tag initial shell length. Nevertheless, an individual abalone tagging method, e.g. glued numbered tags onto shells (Kube et al. 2007), should be developed to allow for repeated measurement and samplings in further experiments.

From the results of both Experiment 1 and 2, abalones grew better in cage culture system designed compared with that in traditional tiered basket culture system. In Experiment 1, when compared with current methods used within the industry, culturing abalone at high density in sea cages provides acceptable growth rates. Considering the acceptable stocking density of 6.53% (75 individuals per tier) following the local traditional practices, a higher initial density of 16.33–19.60% can be appropriate in the cage culture system without compromising growth rate. It is noticeable that the mean final shell length achieved in cage culture was 6.7–15.9% higher than in the baskets at the same initial stocking density (6.53%). Higher growth rates obtained in cage culture system were also detected in each sized groups in Experiment 2. In addition to improved shell growth, the wet weight biomass of the animals was increased by 1.48–3.01 times in the new culture cages compared with the traditional baskets. The increase in biomass was primarily due to the structural superiorities of newly established cage culture system for holding more cage units in a frame, thus increasing the number of abalones per frame, while the individual growth was also better.

It's reported that high load of feeds to abalones could lead to dissolved oxygen competition and restricted water movement in sea-based culture system (Takami, Kawamura & Yamashita 1997; Minh, Petpiroon, Jarayabhand, Meksumpun & Tunkijjanukij 2010). This seems to be more significant when high stocking density and relatively smaller and closed culture system were implemented in abalone culture. Abalone growth is inhibited by increased levels of metabolic wastes, disease and reduced dissolved oxygen (Jarayabhand & Paphavasist 1996; Takami et al. 1997). In this study, higher levels of water flow velocities and dissolved oxygen and higher growth performance of abalones were observed in the cage culture system rather than the traditional tiered baskets system (Figs 3, 5 and 6). Water flow is believed to stimulate the feeding behaviour, which enhances the growth of abalone (Shepherd 1973). Dissolved oxygen is classified as a limiting factor for growth by means of limiting the scope for aerobic metabolism (Fry 1971; Brett 1979). Ingerson and Geddes (1995) also reported that the energy cost to oxygen conformers in dealing with declining oxygen consumption can severely limit growth. Harris, Maguire, Edwards and Johns (1999) suggested that abalone may be more sensitive to periods of hypoxia than several other aquatic animals. The present study provided experimental evidence that differences of water flow and dissolved oxygen occurred between varying culture systems. It is possible that more healthy environmental conditions might explain the growth superiorities of the newly cage culture system. Considering that the water flow and dissolved oxygen in the two culture systems was only monitored over one 24-h period, a longer observation period is recommended.

Unlike other floating cage or barrel systems reported previously (Benson 1986; Aviles & Shepherd 1996; Alcantara & Noro 2006), or the traditional tiered basket culture system in China (Chen 1989; Wu et al. 2009), it's not necessary to pull culture vessels out of the seawater surface as no fixed top were placed in the newly established cage system. Whole strings of the fresh algae including G. lemaneiformis and L. japonica could be supplied to excess in the slots between neighbouring plates of the shelf-like devices of the culture cages. Following the locally accepted guidelines for suspending multi-tier baskets, feeding abalones within eight frames is a tough task for one farmer. With the newly established cage culture method, daily feeding is easy to carry out, requiring about 1/3 of the labour provided in basket culture. Considering the biomass increase in a unit frame in this study (Fig. 4), the new method could increase the biomass by 6–7 times with the same labour input, compared with the traditional method.

This study describes a novel submerged cage culture system for H. discus hannai rearing in Fujian, South China. The study examined growth rate and biomass of abalones in newly established cage culture and compared with the results using traditional multi-tier basket culture system. Results showed remarkable increases in growth rate and biomass of abalone in cage culture method compared with the traditional basket system.

Acknowledgments

We are very grateful to the anonymous reviewer for his careful revision and many constructive comments. This study was supported by the Non-profit Project for Molluscan Industry (no. 3-53, or ny-hyzx07-047 formerly), the China Northeast Rejuvenation Project fund by Shenyang Branch, CAS, the earmarked fund for China Agriculture Research System and the Major State Basic Research Development Program of China (973 Program) (no. 2010CB126401). We are grateful to Dengyue Fishery & Seafood for providing facilities and animals for the study. We also thank Mr Liu Lei for assistance in measuring abalones, Dr Wang Fan for his provision of data loggers and Dr Zhang Guangtao for his kind comments on our experiments.

Ancillary