The effect of stocking density and diet on the growth and survival of cultured Florida apple snails, Pomacea paludosa

Aquaculture 311 (2011) 139–145

Contents lists available at ScienceDirect

Aquaculture j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / a q u a - o n l i n e

The effect of stocking density and diet on the growth and survival of cultured Florida apple snails, Pomacea paludosa Amber L. Garr a,⁎, Helen Lopez a, Rachael Pierce b, Megan Davis a a b

Florida Atlantic University, Harbor Branch Oceanographic Institute, 5600 US 1 North, Fort Pierce, FL 34946, USA South Florida Water Management District, Lake Okeechobee Division, 1480 Skees Road, West Palm Beach, FL 33411, USA

a r t i c l e

i n f o

Article history: Received 10 May 2010 Received in revised form 12 November 2010 Accepted 16 November 2010 Available online 23 November 2010 Keywords: Apple snail Aquaculture Nutrition Pomacea paludosa Stocking density Stock enhancement

a b s t r a c t There has been interest in culturing the Florida apple snail, Pomacea paludosa, for stock enhancement purposes in central and south Florida to help promote snail kite (Rostramus sociabilis) recovery. In 2007, Harbor Branch Oceanographic Institute at Florida Atlantic University began to research techniques necessary to culture Florida apple snails at a commercial scale (tens of thousands per year). This article reviews stocking density and diet experiments that have yielded a protocol for large-scale culture of Florida apple snails. The objectives of this research were to determine the stocking density that supports efficient production and to determine whether diet choice affects growth and survival and can improve captive growth rates at higher stocking densities. Juvenile apple snails were stocked at six densities (10, 20, 40, 60, 80, and 100 snails/m2) in recirculating aquaculture systems with a raised substrate. Although growth was faster in the lowest stocking density compared to the highest density during the first month, the difference subsided in the second month, and overall growth rates and final shell lengths were not statistically different. Survival was not affected by density. A second experiment testing higher densities (100, 175, and 250 snails/m2) showed that snails could be stocked as high as 250 snails/m2 and confirmed that the lowest density is optimal for first-month growth. An initial diet study examining six diets (romaine lettuce, two combination diets of plant material and catfish chow, and three ingredient-only diets) showed shell length growth rates of 3 mm/wk for snails fed the macroalgae Ulva Diet and Catfish Diet (catfish chow only) for two months. In a subsequent experiment, snails stocked at 250 snails/m2 and fed the Ulva Diet grew faster than those at the same density fed the Catfish Diet. The greatest growth occurred in snails fed the Ulva Diet and stocked at 100 snails/m2. Based on these results, it is recommended that the Florida apple snail be cultured in recirculating aquaculture systems with a raised substrate at 100 snails/m2 and an artificial diet of Ulva macroalgae mixed with catfish chow. Snails cultured in this manner are suitable for release into the wild after three months when they reach adult size (25 mm) and reproductive maturity. © 2010 Elsevier B.V. All rights reserved.

1. Introduction In recent years, there has been interest in stock enhancement of Florida apple snail populations in areas of central and southern Florida where hydrological alteration, water management practices, and meteorological events (e.g., droughts, hurricanes) have caused declines. This interest stems from the critical role of the snail in freshwater wetland food webs. The snails are preyed upon by turtles, alligators (Alligator missippienisis) (Delaney and Abercombie, 1986), and bird species such as the white ibis (Eudocimus albus) (Kushlan,

Abbreviations: HBOI-FAU, Harbor Branch Oceanographic Institute at Florida Atlantic University; SD, standard deviation of the mean; SL, juvenile queen conch shell length; G, weekly growth rate; Wt, shell length at time of sampling period; Wi, initial shell length; t, number of weeks; ANOVA, analysis of variance. ⁎ Corresponding author. Tel.: +1 772 242 2578; fax: +1 772 466 6590. E-mail address: [email protected] (A.L. Garr). 0044-8486/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.aquaculture.2010.11.017

1974), limpkin (Aramus guarauna), and boat-tailed grackle (Cassidix mexicanus) (Snyder and Snyder, 1969). Most importantly, they are the predominant food source of the federally endangered snail kite (Rostramus sociabilis) (Sykes et al., 1995; Snyder and Snyder, 1969). Apple snail stock enhancement could potentially reduce the recovery time of apple snail populations following extended periods of severe hydrological conditions. Although stock enhancement is a popular method of compensating for declining aquatic species populations (Gutierrez-Gonzalez and Perez-Enriquez, 2005; Perez-Enriquez et al., 1999; Poteaux et al., 1999), there are many challenges concerning Florida apple snail production. The species has never been artificially propagated on a large scale, and its growth and survival requirements under captive conditions are not fully understood. Further, stocking expansive reintroduction areas at natural densities would require thousands to millions of snails although apple snails are found at relatively low densities (0.05 to 1 snails/m2) in the wild (Karunaratne et al., 2006). As such, rearing space could be a key factor limiting the success of a

140

A.L. Garr et al. / Aquaculture 311 (2011) 139–145

large-scale stock enhancement program. Prior to this, however, a culture protocol for the Florida apple snail is needed. Developing a program that produces growth and survival rates for cultivated animals that are equal or greater than their wild counterparts at a minimal cost is one of the most critical factors for the success of any stock enhancement aquaculture operation. It has been estimated that the growth rate of juvenile apple snails in the wild is approximately 2 to 3 mm/wk, which allows them to reach adult size (25 mm) two to three months after hatching (Darby et al., 1997; Hanning, 1979). Two variables that may influence growth and survival of the apple snail in captivity are stocking density and diet composition. It has been suggested that the density-dependent growth of apple snails in captivity is due to metabolites or chemicals they release into the water (Carter and Ashdown, 1984; Hanning, 1979; Perry and Arthur, 1991; Thomas et al., 1975; Williamson et al., 1976). Successful apple snail culture has occurred at stocking densities of up to 20 snails/m2 (Darby et al., 1997; Hanning, 1979), although such low densities are neither economical in an aquaculture setting nor viable for large-scale production. It may be that the small aquarium systems used in previous studies inhibited growth by magnifying chemical cues in the water, a limitation that could be mitigated or eliminated by using larger aquaculture systems that are specially designed for the needs of the snail. Although apple snails traditionally have been cultured on a diet consisting mainly of romaine lettuce (Aufderheide et al., 2006; Conner et al., 2008; Corrao et al., 2006; Garr et al., 2008), the cost and quantity requirements of such a diet make it impractical for commercial-scale operations. Additionally, this diet may not yield the fastest growth and quickest development to sexual maturity. Several laboratory studies have shown that the Florida apple snail is capable of growth and survival while consuming a variety of resources including Utricularia sp. (Darby et al., 1997; Hanning, 1979; Sharfstein and Steinman, 2001), Eleocharis sp. (Sharfstein and Steinman, 2001), and Hydrilla sp. (Darby et al., 1997). Other studies have shown that artificial diets also can sustain growth and survival for many cultured species. One experimental mixture of gelatin and fish feed provided sufficient nutrients to culture Pomacea patula (Espinosa-Chávez and Martínex-Jerónimo, 2005), a species closely related to the Florida apple snail. Other artificial diets, such as cyanobacteria, have been used for different Pilid species (Ruiz-Ramírez et al., 2005). Artificial diets often have the additional benefits of being inexpensive and commercially available; advantages that could reduce production costs. The objectives of this study were to determine the stocking density that supports the maximum production of healthy snails in recirculating aquaculture systems in a similar time frame as those in the wild, and to establish whether diet can improve captive growth rate at higher densities. To determine the stocking density that results in the maximum production of snails, we conducted laboratory experiments that 1) examined the effect of stocking density on growth and survival over time, and 2) determined the maximum stocking density at which growth rate begins to markedly decline. We also conducted experiments using varied diets to ascertain 3) the diet that optimizes growth and survival and 4) whether an improved diet affects growth and survival at different population densities.

window screen approximately 5 cm off the bottom of the tank. The window screen was covered with 3 cm of crushed coral aragonite sand (grain size from 2 to 5.5 mm). The aragonite substrate introduced additional calcium to the system, and also served to capture solids and provide biofiltration (Davis, 2005). Water entered each tank through a PVC pipe submerged just below the surface of the water. Each system consisted of a tank with raised aragonite sand bottom, a sump, and a 1/3 to 1/2 HP pump (Fig. 1). For each experiment, plexiglass pieces were used to divide the troughs into the corresponding experimental units. The systems were located inside an aluminum greenhouse, and the daily lighting cycle was 14 h light:10 h dark. Temperature was maintained in the systems with either titanium heaters controlled by a dial thermostat, or with small submersible heaters. All tanks were drained, with the snails inside, and the substrate sprayed down with fresh water biweekly or monthly, depending on the number of snails being used in the experiment. 2.2. Experiment 1: Stocking Density Study This experiment was conducted for eight weeks from October to December 2007 in a recirculating system consisting of two rectangular troughs (3.3 m× 0.6 m ×0.4 m), each divided into nine identical sections of 0.22 m2 separated by plexiglass dividers. Water entered each section of the troughs through a 1/2 inch PVC pipe at an equal rate (3 l/h). Stems of vegetation with attached egg clutches were collected from Lake Kissimmee, Florida and allowed to hatch naturally in the laboratory. Newly hatched snails (3 to 4 mm) were randomly selected and stocked into each section at one of six stocking densities with three replicates per treatment: 10 snails/m2 (2 snails per replicate), 20 snails/m2 (4 snails per replicate), 40 snails/m2 (9 snails per replicate), 60 snails/m2 (13 snails per replicate), 80 snails/m2 (18 snails per replicate), and 100 snails/m2 (22 snails per replicate). The treatments were allocated to the location within the troughs using a randomized numbering system. The snails were fed one leaf of romaine lettuce attached to the side of the tank at the substrate level and remnants were replaced every other day. The shell length (SL) of all snails was measured at the beginning of the experiment, and at four (Month 1) and eight weeks (Month 2). Shell length, as opposed to shell weight, was monitored because it is thought to be a more important factor for reaching sexual maturity (Hanning, 1979). Shell length was determined by measuring the tip of the shell to the longest diagonal point on the lip at the aperture. Average weekly growth rates were determined using the formula: G = ðWt –Wi Þ = t where G equals weekly growth rate, Wt is the shell length of the snail at the sampling period, Wi is the initial snail shell length, and t is time in weeks. Percent survival was calculated at the end of the experiment when all remaining snails were counted. Water quality parameters were recorded throughout the experiment and the system was

2. Materials and methods 2.1. System design All experiments were conducted in recirculating aquaculture systems constructed at Harbor Branch Oceanographic Institute at Florida Atlantic University (HBOI-FAU). The freshwater came from an underground well, and was pretreated with a degassing tower, sand filter, and biofiltration media prior to entering the aquaculture facility. Each tank had a false bottom composed of plastic grating and

Fig. 1. Schematic diagram of an apple snail culture trough: (a) undergravel support, (b) sand substrate, (c) water column, (d) sump, (e) pump, and (f) water return. Plexiglass partitions were attached to the undergravel support. Drawing by Jackie Aronsan (Davis, 2005).

A.L. Garr et al. / Aquaculture 311 (2011) 139–145

cleaned once each month. Separate ANOVAs were used to detect differences in average weekly growth rates between treatments for Month 1, Month 2, and Overall growth, followed by Tukey's Studentized Range Test if significant differences were found (SAS v. 9.1, SAS Institute Inc., 2005). Data were tested for homogeneity of variances using Shapiro–Wilk tests and were found to be normally distributed. 2.3. Experiment 2: Upper Limit Stocking Density Study Based on the results of the previous experiment, we attempted to determine the upper threshold limit at which density begins to negatively affect the growth and survival of newly hatched snails. We conducted this experiment using two recirculating systems identical to each other that were designed for juvenile snail growout. Each system consisted of eight toughs (2.4 m × 0.6 m × 0.3 m) with a raised aragonite sand bed, a sump and pump. This experiment was conducted for twelve weeks from August to November 2008. Egg clutches were collected as described above from Lake Kissimmee, Florida in August 2008 and were allowed to hatch naturally in the laboratory. Newly hatched snails (3 to 4 mm) were stocked into the troughs at three densities: 100 snails/m2 (120 snails per replicate), 175 snails/m2 (210 snails per replicate), and 250 snails/m2 (300 snails per replicate). Nine troughs were stocked to provide three replicates per treatment. Two replicates of each treatment were randomly placed in the first system, and one replicate was placed in the second system. Snails were fed romaine lettuce to satiation for the first month, and this was supplemented with an artificial diet (Ulva Diet from Experiment 3) for the rest of the experiment, based on preliminary results from Experiment 3. Shell length was recorded each month for a 20% subsample of snails haphazardly chosen from each tank. Average weekly growth rates (for Month 1, Month 2, Month 3, and Overall growth) were determined using the formula described above. Survival was calculated at the end of the experiment when all remaining snails were counted. Water quality parameters were recorded and the troughs were cleaned biweekly. An ANOVA was used to determine among-treatment differences in average weekly growth rates and percent survival, followed by Tukey's Studentized Range Test for comparative differences between treatments (SAS v. 9.1). Data were tested for homogeneity of variances using a Shapiro–Wilk test and were found to be normally distributed. 2.4. Experiment 3: Diet Study This experiment was conducted for eight weeks from August to October 2008. Egg clutches were collected from the Fort Drum Wildlife Management Area, Florida in August 2008 and hatched naturally in the laboratory. Juvenile snails (4 to 6 mm) were stocked at a density of 100 snails/m2 in a recirculating system consisting of four troughs (3 m × 0.6 m × 0.4 m) divided by plexiglass walls into six identical compartments (0.3 m2 each). Water entered each compartment though 1/2 inch PVC at a rate of 3 l/h. Six experimental diets were examined with four replicates per treatment. Each treatment was randomly allocated within the system using a randomized numbering system. Diet 1 was romaine lettuce. Diet 2 (Ulva Diet) consisted of 25% Ulva sp., a marine macroalgae, and 75% commercial catfish chow (Purina™). This artificial diet has produced high growth rates for marine gastropods (Strombus sp.) (Davis and Shawl, 2005), and was hypothesized to be beneficial for freshwater apple snails as well. Diet 3 (Utric Only) consisted of 100% dried Utricularia sp. collected from Lake Kissimmee and mixed into a gel pellet. This diet was chosen because previous studies have shown success culturing apple snails using Utricularia sp. as a feed (Darby et al., 1997; Hanning, 1979; Sharfstein and Steinman, 2001). Diet 4 (Ulva Only) consisted of 100% dried Ulva sp. cultured at HBOI-FAU and mixed into a gel pellet. Diet 5 (Utric Diet) consisted of 25% Utricularia

141

sp. and 75% commercial catfish chow mixed into a gel pellet. Diet 6 (Catfish Only) was 100% ground catfish chow pellets mixed into a gel pellet. Diets 3–6 were prepared at HBOI-FAU following the recipe in Davis and Shawl (2005). The Ulva Diet was prepared by the commercial company Bonney, Laramore, and Hopkins, Inc. Juvenile snails were fed the experimental diets for three days prior to the initiation of the experiment and the first SL measurement to allow time for the snails to adjust to the diets. Snails were fed the diets at a rate of 0.07 g/snail/day for the first month and 0.14 g/day/snail for the second month as the snails grew larger. This amount was based on previous observations in the laboratory and allowed the snails to be fed to satiation each day. All of the snails were measured for SL at the beginning of the experiment and each month thereafter, and average weekly growth rates were determined. Due to large differences in growth and survival among diets during the first month, only the diets supporting the highest growth were continued during the second month. Water quality parameters were recorded and the system was cleaned each month. Among-treatment differences in average weekly growth rates and survival were analyzed using an ANOVA followed by Tukey's Studentized Range Test for comparative differences between treatments (SAS v. 9.1). Data were tested for homogeneity of variances using a Shapiro–Wilk test and were found to be normally distributed. 2.5. Experiment 4: Stocking Density and Diet Study Based on the growth and survival results of the initial diet study (Study 3), a second experiment was designed, post hoc, to compare the growth and survival of snails fed the two best experimental diets at varying stocking densities. This experiment was conducted for eight weeks from September to November 2008 in the same systems as Experiment 3. Preliminary density experiments with the apple snails at the culture facility had shown that snails could be stocked as high as 250 snails/m2 without affecting survival rates, but with an effect on growth rates during the first month (Garr et al., 2008). Therefore, this experiment examined whether an improved diet would enable snails to be stocked at higher densities without impacting growth or survival rates. Egg clutches were collected from Lake Kissimmee, Florida in September 2008 and hatched in the laboratory. Juvenile snails (3–4 mm) were stocked into one of two density treatments: 100 snails/m2 (22 snails) and 250 snails/m2 (55 snails), and fed one of two diet treatments: Ulva Diet with 25% Ulva and 75% commercial catfish chow or Catfish Only Diet with 0% macroalgae. Each treatment was replicated four times using the same aquaculture system as the first diet experiment. All treatments were randomly allocated within the system using a randomized numbering system. Snails were fed each day at a rate of 0.07 g/snail/day for the first month and 0.14 g/snail/day for the second month. Shell length and average weekly growth rates were calculated using the same methods as the previous experiments. Among-treatment differences in mean weekly growth rates and survival and interaction effects of diet and density were analyzed using a two-way ANOVA followed by Tukey's Studentized Range Test for comparative differences between treatments if necessary (SAS v. 9.1). Data were tested for homogeneity of variances using a Shapiro–Wilk test and were found to be randomly distributed. 3. Results 3.1. Laboratory vs. natural water quality Basic water quality parameters were similar in Lake Kissimmee (2008) and Lake Okeechobee (1976–1977) with the exception of alkalinity and hardness (Table 1). Based on field data obtained from local freshwater lakes where snails are common, laboratory water quality parameters remained within the natural range of apple snail tolerance

142

A.L. Garr et al. / Aquaculture 311 (2011) 139–145

Table 1 Water quality data collected from natural apple snail habitats in the field compared to water quality data collected for each experiment in the laboratory. The results are expressed as mean ± SD and the number of samples is in parentheses. Field

Temperature (°C) Dissolved oxygen (mg/l) pH Ammonia (mg/l) Nitrate (mg/l) Alkalinity (mg/l CaCO3) Hardness (mg/l CaCO3) b

Lake Okeechobee (1976–1977)b

Stocking Density Study System 1

System 2

28 (1) 6.7 (1) 7.9 ± 0.1 (3) 0.01 ± 0.01 (3) 0.9 ± 0.6 (3) 548 ± 5.7 (2) 331 ± 12.7 (2)

22.0 ± 6.0 (16) n/a 8.0 ± 0.41 (16) 0.61 ± 0.31 (16) 1.3 ± 0.49 (16) 111.0 ± 32.0 (16) 151.0 ± 40.0 (16)

24.9 ± 3.1 (12) 13.0 ± 5.5 (8) 8.1 ± 0.2 (8) 0.06 ± 0.14 (9) 0.5 ± 0.3 (8) 166.4 ± 16.8 (9) 158.0 ± 16.9 (9)

24.9 ± 3.0 (12) 12.4 ± 5.3 (7) 8.1 ± 0.1 (7) 0.03 ± 0.03 (9) 0.9 ± 0.8 (8) 193.0 ± 37.7 (9) 170.4 ± 12.6 (9)

(Table 1). However, due to the way in which the water was circulated throughout the experimental units, it was impossible to attribute potential water quality effects (or impacts) to differences in stocking densities. In other words, any metabolites dissolved in the water as the result of stocking density were distributed evenly among all treatments, potentially dampening the effects of density on growth. 3.2. Experiment 1: Stocking Density Study The mean SL at stocking was 3.4 ± 0.1 mm (SD), and there were no significant differences in initial SL among densities (p = 0.435). Average weekly growth rates tended to decline with increased stocking density during the first month, but were indistinguishable among densities during the second month (Fig. 2). In the first month, snails stocked at 10 snails/m2 had significantly faster growth rates than those stocked at 100 snails/m2 (p = 0.029), and snails stocked at 20 snails/m2 grew faster than those stocked at the 40, 60, and 100 snails/m2 (p = 0.041, 0.024, and 0.005, respectively). No significant density-dependent growth responses were detected during the second month, and overall growth rates for the experiment were also similar among all densities, averaging 2.5 mm/wk and yielding an average final SL of 20.4 ± 0.9 mm. Survival averaged 97.5% and ranged from 86 to 100% among individual replicates with no significant differences among treatments (p = 0.736). 3.3. Experiment 2: Upper Limit Stocking Density Study The mean SL was 3.7 ± 0.8 mm (SD) at stocking with no significant differences between densities (p = 0.667). Stocking densities of 100 3.5

3.0

Mean Growth Rate (mm/wk)

Upper Limit Stocking Density Study

Experimental Diet Study

Stocking Density and Diet Study

28.5 ± 1.0 (17) 7.4 (1) 8.6 (1) 0.01 (1) 4.5 ± 1.2 (2) 168 (1) 246 (1)

27.1 ± 0.6 (3) 13.6 ± 0.6 (2) 8.2 ± 0.2 (4) 0.02 ± 0.01 (6) 2.7 ± 1.9 (6) 184.3 ± 46.4 (6) 167.3 ± 22.2 (6)

27.2 ± 1.1 (5) 12.8 ± 5.9 (4) 8.2 ± 0.1 (6) 0.02 ± 0.03 (9) 2.2 ± 1.7 (9) 178.0 ± 45.8 (9) 160.0 ± 23.8 (9)

Garr et al., 2008. Hanning, 1979.

3.4. Experiment 3: diet study After the first month, average weekly growth rates for snails fed the Ulva Diet (3.2 mm/wk) and Catfish Only (3.0 mm/wk) were significantly higher than the other four diets (p = 0.001; Fig. 4). Juvenile snails fed these diets grew up to three times faster than those fed other diets. Growth rates were similar for the first four weeks among Lettuce (1.4 mm/wk), Utric Only (0.9 mm/wk), and Utric Diet (1.3 mm/wk) groups. All of the snails fed Ulva Only diet died before the end of the first month due to apparent starvation since they were never observed feeding on these pellets. Snails fed Catfish Only and Ulva Diet averaged 19.7 ± 0.7 mm and 20.3 ± 1.5 mm, respectively, which was significantly different than snails fed lettuce (13.3 ± 0.3 mm, p = 0.0001), Utric Only (10.4 ± 0.4 mm, p = 0.0001), and Utric Diet (13.1 ± 1.4 mm, p = 0.0001). There was no measurable growth in Ulva Only snails prior to their death. First-month survival for all other treatments was 100%. A second month of growth was examined for the Ulva Diet and Catfish Only snails (Fig. 4) because these diets showed the most

Month 1 (Oct-Nov) Month 2 (Nov-Dec) Overall

ab b ac

2.5

snails/m2 yielded significantly higher growth rates (1.3 mm/wk) than densities of 250 snails/m2 (0.7 mm/wk) during the first month (p = 0.013), but no difference in growth rate was detected during the second (p = 0.597) or third (p = 0.434) months (Fig. 3). There were no differences in the overall growth rate among the densities (p = 0.579). Average weekly growth rates (1.3 to 1.4 mm/wk) in this experiment were lower than in the previous stocking density experiment. There was no density-dependent difference in survival rates (p = 0.875), which averaged 79.1% across both densities and ranged from 75.1 to 81.5%.

abc ac

c

2.0

1.5

1.0

0.5

3.5

Mean Growth Rate (mm/wk)

a

Laboratory

Lake Kissimmee (Summer 2008)a

Month 1 (Aug - Sept) Month 2 (Sept - Oct) Month 3 (Oct-Nov) Overall

3.0 2.5 2.0 A 1.5 AB 1.0

B

0.5 0.0 100 snails/m2

0.0 10 snails/m2

20 snails/m2 40 snails/m2 60 snails/m2 80 snails/m2 100 snails/m2

Fig. 2. Mean weekly growth rate (mm/wk, ± SD, n = 3) of juvenile apple snails stocked at six densities over a two month period. Significant pairwise differences between densities were found only during month one and are denoted by letters. Solid bars represent mean growth rate over the entire experiment.

175 snails/m2

250 snails/m2

Fig. 3. Mean weekly growth rate (mm/wk, ± SD, n = 4) of juvenile apple snails stocked at three different densities over a three month period. Significant pairwise differences between densities were found only during month one and are denoted by letters. Solid bars represent mean growth rate over the entire experiment. Snails were fed romaine lettuce during the first month of growth and were switched to an artificial diet for the remainder of the experiment.

A.L. Garr et al. / Aquaculture 311 (2011) 139–145 3.5

a

a

Month 1 Month 2

Mean Growth Rate (mm/wk)

3.0 A

2.5

A

2.0 b

b 1.5

A difference did appear, however, within the Catfish Only diet between the high and low densities during the second month of growth (p= 0.004), with high-density snails experiencing a lower growth rate. At the end of the experiment, snails fed the Ulva Diet at both densities were significantly larger than the snails fed the Catfish Only diet (pb 0.001 for Catfish Only at low density, and p= 0.027 for Catfish Only at high density). The mean final SL for the snails fed Ulva Diet was 25.4± 1.1 in low density (100 snails/m2) and 23.8± 0.7 mm in high density (250 snails/m2). Snails fed the Catfish Only diet averaged 23.0±0.7 mm in low density and 22.1± 0.8 mm in high density.

b 1.0

4. Discussion

0.5

0.0 Lettuce

Catfish

Utric Only

Utric Diet

Ulva Only

Ulva Diet

Fig. 4. Mean weekly growth rates (mm/wk) (± SD, n = 4) of juvenile apple snails fed six experimental diets for one month. Letters above each column represent significant pairwise differences among treatments (p b 0.05). None of the snails in the Ulva Only Diet survived to the one month interval. The black bars represent a second month of analysis for the Ulva and Catfish diets only.

promise. Although growth rates decreased during the second month (1.6 mm/wk for Ulva Diet and 2.0 mm/wk for Catfish Only), there was no significant difference between the two treatments (p = 0.565). Final SL averaged 27.1 ± 2.8 and 27.0 ± 2.1 mm for snails fed the Catfish Only and Ulva Diet, respectively. The survival rate was 70.5% (±14.2 SD) for snails fed the Catfish Only diet and 86.5% (±5.2 SD) for the snails fed the Ulva Diet, which was not significantly different (p = 0.078). 3.5. Experiment 4: Stocking Density and Diet Study The average initial SL for all snails was 3.6 ± 0.22 mm. There were significant differences in average weekly growth rates during the first month (p = 0.015), the second month (p = 0.001), and in the overall growth rates (p = 0.002). Although a significant diet and density interaction effect did appear during the first month of growth (p = 0.007), there was no effect when considering the experiment overall (p = 0.659). Snails fed the Ulva Diet and stocked at 100 snails/m2 had significantly faster growth during the first month than snails stocked at 100 snails/m2 and fed the Catfish Only diet (p= 0.011) (Fig. 5). Within the diet treatments, there was no difference in growth rates during the first month despite differences in stocking density (p = 0.050). 3.5 3.0

Mean Growth Rate (mm/wk)

143

Month 1 (Sept-Oct) Month 2 (Oct-Nov) Overall

a c

A

ab c

AB

b

ab c

B

B

2.5 d 2.0 1.5 1.0 0.5 0.0 Ulva Low Density (100 snails/m2)

Ulva High Density (250 snails/m2)

Catfish Low Density (100 snails/m2)

Catfish High Density (250 snails/m2)

Fig. 5. Mean weekly growth rates (mm/wk) (± SD, n = 4) of juvenile apple snails at two different stocking densities being fed two different experimental diets for two months. Letters above each column represent significant pairwise differences among treatments (p b 0.05). Black bars represent mean growth rate over the entire experiment.

Previous research has suggested that snail growth at high densities in laboratory conditions may be inhibited by metabolites or other chemicals released by the snails, most likely in their slime trials (Carter and Ashdown, 1984; Hanning, 1979; Thomas et al., 1975; Perry and Arthur, 1991; Williamson et al., 1976). Although we did not measure chemical metabolites released by the snails in these studies, a recirculating system that utilizes a raised, multi-surface substrate that can be easily cleaned may assist in mitigating the potential effects of growth-inhibiting compounds. When these production systems are drained, scrubbed, and sprayed down on a monthly basis, snails can be cultured at 100 to 250 snails/m2 and reach adulthood (sexual maturity) within three months of hatching. Future experiments that examine the influence of metabolites over a longer period of time may be useful in developing large-scale production facilities with recirculating aquaculture systems. Apple snails can be stocked in recirculating aquaculture systems with an artificial raised substrate at densities as high as 100 snails/m2 and maintain an average growth rate of 2.5 mm/week for the first month. Although densities higher than 100 snails/m2 could be used, this may slow the fast growth usually seen in the first month after hatch (Hanning, 1979). Natural densities for adult Florida apple snails are typically 1 snail/m2 or less (Darby et al., 2003; Karunaratne et al., 2003) and are estimated to be 30–120 snails/m2 for newly hatched juveniles (Darby et al., 2008). Previous laboratory experiments have noted negative growth impacts to apple snails stocked at densities higher than 8 snails/m2 (Conner et al., 2008) or 20 snails/m2 (Darby et al., 1997) for Pomacea paludosa and 2 snails/m2 for P. canaliculata (Tanaka et al., 1999). Research on the husbandry techniques for the apple snail Marisa cornuarietis also found decreased growth rates in stocking densities greater than 12 snails/m2 (0.2 snails/l) (Aufderheide et al., 2006). Similar patterns have also been seen for land snails (Baur and Baur, 1992; Cameron and Carter, 1979). Conversely, Alves et al. (2006) showed that two apple snail species, P. lineate and P. bridgesi, could be cultured at densities from 91 to 272 snails/m2 (or 0.5 to 1.5 snails/l). Likewise, the edible land snail, Helix aspersa, can be raised in culture conditions with frequent tank cleaning at densities from 100 snails/m2 (Jess and Marks, 1995) up to several hundred snails/m2 (Dupont-Nivet et al., 2000). When we tested the upper limits of stocking density, there was an increase in growth rates during the second month instead of the decline typically seen during subsequent months as snails grow larger. This could be due, in part, to the addition of the artificial Ulva Diet as a supplemental food source during the second month. Our experiments have shown growth rates of 1.5–2 mm/wk in apple snails fed romaine lettuce, and this rate has been improved upon using an artificial diet. The increased growth could also be attributed to the haphazard method of sub-sampling 20% of the snails, where a potentially biased selection of larger snails could have skewed the data. Research on diet development for the queen conch, Strombus gigas, has shown that Ulva macroalgae has a protein content of 5.3% compared to plain Catfish chow (34.2%) and the Ulva Diet (31.4%) on a dry weight basis (Garr et al., unpublished data). Although analytical

144

A.L. Garr et al. / Aquaculture 311 (2011) 139–145

tests were not conducted on the romaine lettuce, Utric Only, or Utric Diets in this experiment, it can be hypothesized that these diets and romaine lettuce alone may not provide enough protein to maximize growth for captive snails. Additionally, the marine macroalgae Ulva sp. did not appear to be palatable without the mixture of the catfish chow. In the queen conch experiments, conch fed the same Catfish Only diet (tested in this experiment) had a significantly lower survival than conch fed a combination of catfish chow and the macroalgae Agardheilla sp. (Acosta-Salmón et al., 2010). Although the survival rates of the apple snails in our experiments were not statistically different, we saw a similar trend of slightly lower survival for the snails fed the Catfish Only diet. Data from our snail experiment differs from Sharfstein and Steinman (2001), where Ultricularia provided the best growth in comparison to other natural periphyton food sources tested. In their experiment, growth was recorded over 32 days, and snails only grew an average of 2.15 mm, or roughly 0.07 mm/wk. Although the initial SL was larger, this growth is notably lower than what is reported in our experiments. Our results suggest that the combination of a plant source and a supplementary protein source (catfish chow) may provide the best nutritional diet for captive-raised snails. Additional experimentation on the type of plant material needed and the ratio of supplementary protein necessary to increase growth rates would be beneficial to a snail production facility. The snails fed the Ulva Diet and Catfish Only diet showed exceptional growth rates of greater than 2.5 mm/wk during the first month of the experimental diet study. This is similar to growth rates seen in the wild (Hanning, 1979), and is a rate that could support large-scale culture of the apple snail. It appears that an artificial diet high in protein, particularly the Ulva Diet, will not only sustain high growth rates and high survival for the first month, but also will enable the snails to be stocked at densities as high as 250 snails/m2 after the first month of growth during the remainder of the production cycle (3 months for adult snails) Prior to our experiments, the Florida apple snail had not been successfully cultured in large numbers at densities greater than 20 snails/m2 (Darby et al., 1997; Hanning, 1979). Additionally, the same artificial Ulva Diet also supported optimal growth and breeding of Strombus species raised in a comparable recirculating aquaculture system (Shawl and Davis, 2004; Spring and Davis, 2003). Our research suggests that artificial Ulva Diet is ideal for culture of Florida apple snails. While the romaine lettuce diet supports high growth rates, it is cost-prohibitive compared to diets developed specifically for aquaculture applications. Although the Catfish Only diet also provided high growth rates, it is not commercially available in the gel diet form in bulk, while the Ulva Diet can be produced for approximately $2.50 a pound and in a formulation (dried pellets) can be stored for several months (Garr, unpublished data). Nevertheless, due to the cost differential between commercial non-gelatin-bound catfish chow (approximately $0.20/lb) and the experimental diets, the use of the former should be explored for stock enhancement applications in spite of the potential for lower survival, growth rate, or feed conversion efficiency. With capital costs estimated at $131,000 and operational expenses at $95,000 per year, present production costs for apple snails for stock enhancement using Ulva Diet are estimated to be $2.15 per stocked adult (Garr, unpublished data). This is quite high compared to typical stock enhancement costs of $0.30 to $0.95 per individual for catfish, bream, and bass (Allstate Resources, personal communication). Therefore, significant effort needs to be directed at translating our initial laboratory-based successes into a commercially viable operation. Current research is examining the use of a commercially available aquaculture diet (sinking catfish pellets) as a means to reduce expenses. Because the Florida apple snail is being cultured to enhance depleted natural stocks, it is important to know how readily the snails will switch from a commercial to a natural diet once they are released into the field. Preliminary research shows that adult apple snails released directly from

the hatchery into the field are producing healthy egg clutches within a few weeks of release (Pierce, unpublished data). Understanding the production capacities necessary to produce robust snails brings us one step closer to being able to use stock enhancement of cultured snails a potential method to augment the recovery of natural populations. Acknowledgements This research was supported by Contract 4100000028 given by the South Florida Water Management District's Lake Okeechobee Division to Harbor Branch Oceanographic Institute at Florida Atlantic University. The authors would like to thank T. East, D. Unsell, P. McCormick, and P. Darby for their assistance with the project. The authors are grateful to P. McCormick, B. Sharfstein, R. Shuford, and L. Macke for comments on earlier versions of this manuscript. This is a Harbor Branch Oceanographic Institute at Florida Atlantic University contribution number 1816. References Acosta-Salmón, H, Shawl, A., Davis, M., Capo, T.R., 2010. Addition of macroalgae in artificial diets improves survival of juvenile queen conch Strombus gigas (Linné, 1758). World Aqua. Mag. March, 16–19. Alves, T., Lima, P., Lima, S.F.B., Ferri, A.G., Barros, J.C., Machado, J., 2006. Growth of Pomacea lineata and Pomacea bridgesi in different stock densities. Thalassas 22, 55–64. Aufderheide, J., Warbritton, R., Pounds, N., Fileemperador, S., Staples, C., Caspers, N., Forbes, V., 2006. Effects of husbandry parameters on the life-history traits of the apple snail, Marisa cornuarietis: effects of temperature, photoperiod, and population density. Invertebr. Biol. 125, 9–20. Baur, A., Baur, B., 1992. Responses in growth, reproduction and life span to reduced competition pressure in the land snail Balea perversa. Oikos 63, 298–304. Cameron, R.A.D., Carter, M.A., 1979. Intra- and interspecific effects of population density on growth and activity in some helicid land snails (Gastropoda: Pulmonata). J. Anim. Ecol. 48, 237–246. Carter, M.A., Ashdown, M., 1984. Experimental studies on the effects of density, size, and shell colour and banding phenotypes on the fecundity of Cepaea nemoralis. Malacologia 25, 291–302. Conner, S.L., Pomory, C.M., Darby, P.C., 2008. Density effects of native and exotic snails on growth in juvenile apple snails Pomacea paludosa (Gastropoda: Ampullariidae): a laboratory experiment. J. Mollus. Stud. 74, 355–362. Corrao, N.M., Darby, P.C., Pomory, C.M., 2006. Nitrate impacts on the Florida apple snail, Pomacea paludosa. Hydrobiologia 568, 135–143. Darby, P.C., Valentine-Darby, P.L., Bennetts, R.E., Croop, J.D., Percival, H.F., Kitchens, W.M., 1997. Ecological studies of apple snails (Pomacea paludosa, SAY). Report prepared for the South Florida Water Management District and St. Johns River Water Management District. December. 152 pages. Darby, P.C., Valentine-Darby, P.L., Percival, H.F., 2003. Dry season survival in a Florida apple snail (Pomacea paludosa Say) population. Malacologia 45, 179–184. Darby, P.C., Bennetts, R.E., Percival, H., 2008. Dry down impacts on apple snail (Pomacea paludosa) demography: implications for wetland water management. Wetlands 28, 561–575. Davis, M., 2005. Species Profile: Queen Conch, Strombus gigas. Southern Regional Aquaculture Center Publication 7203. 12 pp. Davis, M., Shawl, A.L., 2005. A guide for culturing queen conch, Strombus gigas. Manual of Fish Culture. American Fisheries Society Symposium 46, 125–142. Delaney, M.F., Abercombie, C.L., 1986. American alligator food habits in north central Florida. J. Wildl. Manage. 50, 348. Dupont-Nivet, M., Coste, V., Coinon, P., Bonnet, J.-C., Blanc, J.-M., 2000. Rearing density effect on the production performance of the edible snail Helix aspera Müller in indoor rearing. Ann. Zootechnol. 49, 447–456. Espinosa-Chávez, F., Martínex-Jerónimo, F., 2005. Growth and fecundity of Pomacea patula (Caenogastropoda:Ampullariidae) when fed on gel diets of Scenedesmus incrassatulus (Chlorophyceae). Veliger 47, 213–217. Garr, A., Lopez, H., Davis, M., 2008. A manual to culture the Florida apple snail. Final report prepared for the South Florida Water Management District. December. 14 pages. Gutierrez-Gonzalez, J.L., Perez-Enriquez, R., 2005. A genetic evaluation of stock enhancement of blue abalone Haliotis fulgens in Baja California, Mexico. Aquaculture 247, 233–242. Hanning, G. W., 1979. Aspects of reproduction in Pomacea paludosa (Mesogastropoda: Pilidae). College of Arts and Sciences. Tallahassee, Florida State University. Master of Science. Jess, S., Marks, R.J., 1995. Population density effects on growth in culture of the edible snail Helix aspersa var. maxima. J. Mollus. Stud. 61, 313–323. Karunaratne, L.B., Darby, P.C., Bennetts, R.R., 2003. The effects of wetland habitat structure on Florida apple snails density. Wetlands 26, 1143–1150. Karunaratne, L.B., Darby, P.C., Bennetts, R.E., 2006. The effects of wetland habitat structure on Florida apple snail density. Wetlands 26, 1143–1150.

A.L. Garr et al. / Aquaculture 311 (2011) 139–145 Kushlan, J.A., 1974. Ecology of the white ibis in southern Florida. Ph.D. Dissertation. University of Miami, Coral Gables. 129 pp. Perez-Enriquez, R., Takagi, M., Taniguchi, N., 1999. Genetic variability and pedigree tracing of a hatchery-reared stock of red sea bream (Pagrus major) used for stock enhancement, based on microsatellite DNA markers. Aquaculture 173, 413–423. Perry, R., Arthur, W., 1991. Shell size and population density in large helicid land snails. J. Anim. Ecol. 60, 409–421. Poteaux, C., Bonhomme, F., Berrebi, P., 1999. Microsatellite polymorphism and genetic impact of restocking in Mediterranean brown trout (Salmo trutta L.). Heredity 82, 645–653. Ruiz-Ramírez, R., Esinosa-Chávez, F., Martínex-Jerónimo, F., 2005. Growth and reproduction of Pomacea patula catenacensis Baker, 1922 (Gastropoda: Ampullariidae) when fed Calothrix sp. (Cyanobacteria). J. World Aquacult. Soc. 36, 87–95. SAS Institute Inc., 2005. SAS/STAT Users Guide: Version 9.1. SAS Institute, Cary, NC. Sharfstein, B., Steinman, A.D., 2001. Growth and survival of the Florida apple snails (Pomacea paludosa) fed 3 naturally occurring macrophyte assemblages. J. N. Am. Benthol. Soc. 20, 84–95. Shawl, A., Davis, M., 2004. Captive breeding behavior of four Strombidae conch. J. Shellfish Res. 23, 157–164.

145

Snyder, N.F.R., Snyder, H.A., 1969. A comparative study of mollusk predation by limpkins, everglade kites, and boat-tailed grackles. Living Bird 8, 177–223. Spring, A., Davis, M., 2003. Improving the culture conditions of juvenile queen conch (Strombus gigas Linné) for grow out purposes. In: Aldana-Aranda, D. (Ed.), El Caracol Strombus gigas: Conocimiento Integral para su Menejo Sustentable en el Caribe. Programa Iberoamericano de Ciencia y Tecnología para el Desarrollo. Mérida, Yucatán, México, pp. 147–153. Sykes, P.W., Rodgers, J.A., Bennetts, R.E., 1995. Snail kite (Rostrhamus sociabilis), in: Poole, A., Gill, F., (Eds.), The Birds of North America, No. 171. The Academy of Natural Sciences, Philadelphia and the American Ornithologists' Union, Washington, DC, USA. Tanaka, K., Watanabe, T., Higuchi, H., Miyamoto, K., Yusa, Y., Kiyonaga, T., Kiyota, H., Suzuki, Y., Wada, T., 1999. Density-dependent growth and reproduction of the apple snail, Pomacea canaliculata: a density manipulation experiment in a paddy field. Res. Pop. Ecol. 41, 253–262. Thomas, J.D., Goldsworth, G.J., Benjamin, M., 1975. Studies on the chemical ecology on Biomphalaria glabrataI: the effects of chemical conditioning by snails kept at various densities on their growth and metabolism. J. Zool. Lond. 175, 421–437. Williamson, P., Cameron, R.A.D., Carter, M.A., 1976. Population density affecting adult shell size of Cepaea nemoralis L. Nature 263, 496–497.