Long-term variability in spat collections of the blacklip pearl oyster (Pinctada margaritifera) in Solomon Islands.ABSTRACT Temporal variation in abundance of spat of the blacklip pearl oyster (Pinctada margaritifera) was determined over an eight-year period at two sites, Gizo and Noro, in the Western Province of Solomon Islands Solomon Islands, independent Commonwealth nation (2005 est. pop. 538,000), c.15,500 sq mi (40,150 sq km), SW Pacific, E of New Guinea. The islands that constitute the nation of the Solomon Islands—Guadalcanal, Malaita, New Georgia, the Santa Cruz Islands, . Spat were collected by deploying shade mesh substrata at a depth of 3 m for near-sequential two-month period. Overall, spatfall at the two sites was similar, with summer maximum of 4.9 (Gizo) and 4.7 [collector.sup.-1] (Noro). At both sites, significantly fewer spat were collected in late winter than in summer. Abundance of spat varied among years, although there was no consistency between the two sites regarding years of highest recruitment. Maximum settlement of spat at Gizo was 10.2 [collector.sup.-1] in January 2002, whereas at Noro it was 19.6 [collector.sup.-1] in March 2000. Mean size of spat did not vary over time, but spat at Gizo were significantly smaller (geometric mean (mathematics) geometric mean - The Nth root of the product of N numbers. If each number in a list of numbers was replaced with their geometric mean, then multiplying them all together would still give the same result. 7.6 mm) than those at Noro (mean 8.4 mm). Predators of spat settled on the collectors throughout the year; there were significantly more gastropod gastropod, member of the class Gastropoda, the largest and most successful class of mollusks (phylum Mollusca), containing over 35,000 living species and 15,000 fossil forms. predators at Noro than at Gizo. Gizo and Noro in the Western Province of Solomon Islands are reliable places for collection of blacklip pearl oyster spat. Spat can be caught all year round, although collections will be most effective between November and March. KEY WORDS: blacklip pearl oyster, spat collection, Solomon Islands, temporal variability. INTRODUCTION Blacklip pearl oysters (Pinctada margaritifera) are an economically important species in the tropical Indo-Pacific, primarily through the use of mature oysters for the culture of "black" pearls. At present, culture of black pearls The Black Pearl, originally HEIC Wicked Wench, is a fictional ship in , , and . The Black Pearl is easily recognised by her distinctive black hull and sails. This turns out to be an advantage in more than one way. in the Pacific is confined mostly to the lagoons of eastern Polynesia (Sarver & Sims 1996). In these locations, planktonic plank·ton n. The collection of small or microscopic organisms, including algae and protozoans, that float or drift in great numbers in fresh or salt water, especially at or near the surface, and serve as food for fish and other larger organisms. larval stages larval stage - Describes a period of monomaniacal concentration on coding apparently passed through by all fledgling hackers. Common symptoms include the perpetration of more than one 36-hour hacking run in a given week; neglect of all other activities including usual basics like are entrained by currents within atoll atoll: see coral reefs. atoll Coral reef enclosing a lagoon. Atolls consist of ribbons of reef that may not be circular but that are closed shapes, sometimes miles across, around a lagoon that may be 160 ft (50 m) deep or more. lagoon systems. Expansion of the pearl industry to other parts of the region, where such lagoons are not present, requires identification of an alternative strategy for spat collection, and/or hatchery hatchery a commercial establishment dedicated to the hatching of bird eggs to provide day old chicks and poults to the poultry industry. hatchery liquid the contents of unfertilized eggs. Used in petfood manufacture. production. Friedman and Bell (1996) and Friedman et al. (1998) found that spat could be collected cheaply and easily using simple mesh bag collectors deployed on the seaward slopes of coastal coral reefs coral reefs, limestone formations produced by living organisms, found in shallow, tropical marine waters. In most reefs, the predominant organisms are stony corals, colonial cnidarians that secrete an exoskeleton of calcium carbonate (limestone). in Solomon Islands. Friedman and Bell (1999, 2000) then identified the optimal locations, site characteristics and deployment times for spat collection in Solomon Islands, and Friedman and Southgate (1999) demonstrated that the spat could be grown successfully to adult size. A shortcoming short·com·ing n. A deficiency; a flaw. shortcoming Noun a fault or weakness Noun 1. of the studies (see earlier) on the availability of spat in Solomon Islands, and other locations in the region (Beer & Southgate 2000), is that they are short term. This poses a potential problem for investors who might consider establishing black pearl farms in the western Pacific, based on collection of wild spat. Ideally, they need a long time-series of data on abundance of spat to gauge whether sufficient oysters can be collected each year to support commercial operations. In this study, we address this information gap for two locations in the Western Province of Solomon Islands by documenting temporal variability in abundance of blacklip pearl oyster spat on collectors deployed over an eight-year period. METHODS The two sites used to assess long-term variation in abundance of blacklip pearl oyster spat in this study, Noro and Gizo (Fig. 1), were selected because they previously yielded above average rates of spat collection in Solomon Islands (Friedman et al. 1998, Friedman & Bell 2000). [FIGURE 1 OMITTED] The methods used to collect spat were based on those developed for Solomon Islands by Friedman et al. (1998) and Friedman and Bell (1999, 2000). At each site, 10-14 spat collectors were attached at intervals coming or happening with intervals between; now and then. See also: Interval along a 100-m longline long·line n. A heavy fishing line usually several miles long and having a series of baited hooks. long made of 12 mm polypropylene rope. Each collector comprised a 0.4 [m.sup.2] panel of 55% shade mesh, folded concertina-fashion, threaded and tied in a bundle with monofilament monofilament, n a single strand of untwisted synthetic material such as nylon; used to create surgical sutures. monofilament line. Anchor lines and subsurface sub·sur·face adj. Of, relating to, or situated in an area beneath a surface, especially the surface of the earth or of a body of water. Adj. 1. buoys, at 20-m intervals, were used to hold the longline at a depth of 3 m. The shade mesh spat collectors were deployed every two months, for duration of two months, each year between 1997 and 2005. Deployment and retrieval of collectors at the two sites was staggered over a few days, for logistic reasons. At both sites, spat were harvested from collectors as close as possible to the last days of December, February, April, June, August and October. These dates are hereafter In the future. The term hereafter is always used to indicate a future time—to the exclusion of both the past and present—in legal documents, statutes, and other similar papers. referred to as periods 1-6 respectively. Exceptions to this, when collections were not made, were at Noro in December 1998, and at both sites in August 2001 and December 2004. At harvest, the size, as dorso-ventral measurement (DVM DVM Doctor of Veterinary Medicine. DVM abbr. Doctor of Veterinary Medicine DVM Doctor of Veterinary Medicine. ), of each live and dead spat on each collector was recorded to the nearest mm. The numbers of the two most common predators of blacklip pearl oyster spat, crabs and gastropods (Cymatium spp), were also recorded each time a collector was removed from the water. Data Analysis The spat collector catch data were examined for variation within and between years' effects using a series of targeted analyses. For most analyses, we reduced the data from each period to average catch per collector; this accommodated the four samples where number of collectors per longline was other than 10. Firstly, we used a 2-way ANOVA anova see analysis of variance. ANOVA Analysis of variance, see there to compare differences in spat abundance, grouping by collection period and by site. ANOVA was carried out using Statistica for Windows 5.1 (Statsoft Inc., Tulsa, Oklahoma Tulsa is the second-largest city in the state of Oklahoma and 45th-largest in the United States. With an estimated population of 382,872 in 2006,[1] it is the principal municipality of the Tulsa Metropolitan Statistical Area, a region of 897,752 residents projected to , USA). All data were transformed using ln (N + 1) but, because of the large number of zeros, the assumption of homogeneity Homogeneity The degree to which items are similar. of variance was not met. Although ANOVA is reasonably robust to deviation from this assumption (Zar 1996), we considered differences to be significant only if P [less than or equal to] 0.01. This procedure reduces the risk of Type I errors and is particularly appropriate when sample sizes are small (Underwood 1981). Significant differences between means were identified using posthoc testing (Tukey HSD HSD Human Services Department HSD High Speed Data HSD Hillsboro School District (Hillsboro, OR) HSD Hybrid Synergy Drive (Toyota/Lexus) HSD High School Diploma HSD Historical Society of Delaware test for unequal N). We analyzed for a monotonic monotonic - In domain theory, a function f : D -> C is monotonic (or monotone) if for all x,y in D, x <= y => f(x) <= f(y). ("<=" is written in LaTeX as \sqsubseteq). , long-term trend over the entire eight-year period using simple linear regression Simple linear regression A regression analysis between only two variables, one dependent and the other explanatory. of time versus catch (Esterby 1996). We did not consider that the dataset was long enough to examine for cyclic periodicity periodicity /pe·ri·o·dic·i·ty/ (per?e-ah-dis´i-te) recurrence at regular intervals of time. pe·ri·o·dic·i·ty n. 1. . Where a consistent collection period effect was identified, see earlier, data were deseasonalized before this long-term regression analysis In statistics, a mathematical method of modeling the relationships among three or more variables. It is used to predict the value of one variable given the values of the others. For example, a model might estimate sales based on age and gender. . In practice, this was only the case for live spat numbers, and this was accomplished by calculating deviation from the overall mean value for each sampling period. Year-to-year variation in live spat abundance was examined by comparing catches using 1-way ANOVA. For each year where all six collection periods were available, we selected the annual maximum to overcome any effects of variation in timing of peak abundance. For this analysis, spat numbers for each of the 10 replicate collectors on each line were used after transformation, using ln(N + 1), to improve homogeneity of variances. Insufficient numbers and multiple zero values precluded such analysis of abundances of dead spat and predators. Differences in spat sizes among collection periods were examined using a 2-way ANOVA (site x period). Data were skewed skewed curve of a usually unimodal distribution with one tail drawn out more than the other and the median will lie above or below the mean. skewed Epidemiology adjective Referring to an asymmetrical distribution of a population or of data towards small size, which was corrected by log-transformation for all tests. RESULTS AND DISCUSSION The long-term patterns of spat collection at the two sites showed variability in the abundance of spat and predators over time, and by site (Fig. 2). The way in which this variation was partitioned within years, between sites and between years, is outlined (see later). [FIGURE 2 OMITTED] Within-year and Between Site Variation When the datasets were grouped by site and collection period, no differences were found between sites, but an effect of period was evident (Table 1). The lack of an interaction effect suggests that the two sites were behaving similarly. Posthoc testing showed that the significant variation in numbers in numbered parts; as, a book published in numbers. See also: Number of spat collected between periods was caused by higher abundances in periods 1, 2 and 3 compared with 5 and 6, with period 4 having intermediate values. This translates to higher spatfall on collectors submerged between November and April than on those in the water from July through October. Grouped by harvest time Noun 1. harvest time - the season for gathering crops harvest farming, husbandry, agriculture - the practice of cultivating the land or raising stock , the lowest mean spat abundance per collector over the eight-year period for Gizo was 0.7 (June-August deployments) and the highest was 4.9 (November-December deployments). For the same approximate periods at Noro, winter yielded 1.4 spat per collector, whereas summer yielded 4.7 spat per collector. Abundances of dead spat did not differ significantly between sites but there was a significant effect of deployment period (Table 2). Once again, posthoc testing showed that collection rates were significantly lower in winter compared with summer. The pattern of a summer maximum in abundance of spat over the eight-year period was similar to that reported for the shorter study by Friedman et al. (1998). It is interesting that the shorter soak time of spat collectors in our study (2 months) yielded the same peak numbers of spat as the six-month soak time used by Friedman et al. (1998). This is consistent with the view that there is an optimal soak time before predators begin to overwhelm ongoing recruitment and perhaps explains why we collected relatively few (23%) dead spat compared with Friedman et al. (1996) (46%). The timing of peak spat settlement in Solomon Islands differed to that found on the Great Barrier Reef Great Barrier Reef, largest complex of coral reef in the world, c.1,250 mi (2,000 km) long, in the Coral Sea, forming a natural breakwater for the coast of Queensland, NE Australia. , where maximum numbers were collected during May to June (Beer & Southgate 2000). Reasons for this difference are unclear. Crabs and Cymatium spp. showed no consistent variation in abundance among collection periods, although Cymatium spp. were more abundant at Noro (average of 0.53 [collector.sup.-1]) than at Gizo (0.33 [collector.sup.-1]) (Fig. 3, Table 2). [FIGURE 3 OMITTED] Inter-annual Variation and Multiyear Trends Live spat showed no multiyear trends in abundance after the data were de-seasonalized. Nevertheless, there were clearly differences in abundance of spat in some years, which appear to be caused by differences in timing of the recruitment peak and in size of the peaks (Fig. 2). Out of 13 site-years when a full year of data were collected, the summer peak was in period 1 in 46% of cases, period 2 in 30%, period 3 in 15% and period 6 in 8%. At Gizo, the greatest annual spatfall peak occurred in period 6 in 2001 (10.2 [collector.sup.-1]) and at Noro in period 2 of 2000 (19.6 [collector.sup.-1]). We examined the significance of the magnitudes of annual peaks using 1-way ANOVA for each site (Table 3). Only years when data were obtained for all sampling periods could be used, and this removed 2005 from both sites. Note that for this analysis, because of one occasion when the summer peak spat collection was in period 6 of the previous calendar year, we considered that each spat year began with period 6 of the calendar year before and ran through to period 5 of the year in question. Significantly different peak spatfalls were evident between years at both sites. Posthoc testing showed that at Noro the 2000 peak yielded more spat than any other year except 1999, whereas at Gizo, abundance in 2002 was significantly higher than in 1998, 2001, 2003 and 2004, with no other significant differences. The number of crabs at Gizo was the only temporal dataset that showed evidence of a linear trend over time, in this case one of declining abundance (Table 4). The absence of any reciprocal effect on spat numbers shows that, under the conditions of deployment, the small changes in crab numbers did not affect spat survival. Spat Size There were significant differences in the mean sizes of spat harvested during each collection period and between sites (Table 1). Maximum size tended to be in period 2, towards the latter part of summer, when water temperatures can be expected to be highest. This effect was more pronounced at Gizo than at Noro. The median sizes of live spat on collectors at Noro and Gizo were 9 and 7 mm, and the geometric means were significantly different at 7.6 and 8.1 mm, respectively (Table 1). Caution should be taken in interpreting this size difference, because this result was obtained from a large dataset, and the differences were within the likely measurement error of spat. Thus, the ecological significance of such a small difference is questionable. The arithmetic mean (mathematics) arithmetic mean - The mean of a list of N numbers calculated by dividing their sum by N. The arithmetic mean is appropriate for sets of numbers that are added together or that form an arithmetic series. of spat for both sites was 10 mm. This is consistent with the earlier data of Friedman and Bell (2000), who reported that spat attained an arithmetic mean size of ~10-14 mm within three months of deployment of collectors. CONCLUSION Gizo and Noro in the Western Province of Solomon Islands are reliable places for the collection of blacklip pearl oyster spat. However, collection of spat is expected to be most cost-effective between November and March, when average yields of >4 spat per collector for two-month deployments can be expected in most years. ACKNOWLEDGMENTS The authors thank the staff of the WorldFish Center's Western Pacific Research Station, who worked to collect this dataset through difficult times. Mason Tauku supervised much of the fieldwork and was responsible for data entry. The authors also thank Dr Steve Purcell of WorldFish for helpful discussions on curve fitting Curve fitting is finding a curve which matches a series of data points and possibly other constraints. This section is an introduction to both interpolation (where an exact fit to constraints is expected) and regression analysis. Both are sometimes used for extrapolation. . Funding for this work was provided partially by the Australian Centre for International Agricultural Research Two key international agricultural research organizations are:
LITERATURE CITED Beer, A. C. & P. C. Southgate. 2000. Collection of pearl oyster (family Pteriidae Noun 1. family Pteriidae - pearl oysters Pteriidae mollusk family - a family of mollusks Bivalvia, class Bivalvia, class Lamellibranchia, class Pelecypoda, Lamellibranchia - oysters; clams; scallops; mussels ) spat at Orpheus Island, Great Barrier Reef (Australia). J. Shellfish shellfish, popular name for certain edible mollusks (see Mollusca), e.g., oysters, clams, and scallops, and for certain edible crustaceans, e.g., crabs, lobsters, and shrimps. All are aquatic invertebrates with shells; they are not fish. Res. 19:821-826. Esterby, S. R. 1996. Review of methods for the detection and estimation of trends with emphasis on water quality applications. Hydrol. Process. 10:127-149. Friedman, K. J. & J. D. Bell. 1996. Effects of different susbtrata and protective mesh bags on collection of spat of the pearl oysters Pinctada margaritifera (Linnaeus 1758) and Pinctada maculata (Gould 1850). J. Shellfish Res. 15:525-541. Friedman, K. J. & J. D. Bell. 1999. Variations in abundance of black pearl oyster (Pinctada margaritifera (Linne.)) spat from inshore in·shore adv. & adj. 1. Close to a shore. 2. Toward or coming toward a shore. inshore Adjective in or on the water, but close to the shore: and offshore reefs in Solomon Islands. Aquaculture aquaculture, the raising and harvesting of fresh- and saltwater plants and animals. The most economically important form of aquaculture is fish farming, an industry that accounts for an ever increasing share of world fisheries production. 178:273-291. Friedman, K. J. & J. D. Bell. 2000. Shorter immersion times increase yields of the blacklip pearl oyster, Pinctada margaritifera (Linne.), from spat collectors in Solomon Islands. Aquaculture 187:299-313. Friedman, K. J., J. D. Bell & G. Tiroba. 1998. Availability of wild spat of the blacklip pearl oyster, Pinctada margaritifera, from "open" reef systems in Solomon Islands. Aquaculture 167:283-299. Friedman, K. J. & P. C. Southgate. 1999. Growout of blacklip pearl oysters, Pinctada margaritifera collected as wild spat in Solomon Islands. J. Shellfish Res. 18:159-167. Sarver, D. J. & N. A. Sims. 1996. Overcoming the scarcity of pearl oysters (Pintuda margaritifera) in Micronesia and Hawaii: new areas and opportunities for pearl farming. Aquaculture '95. World Aquaculture Society, Baton Rouge Baton Rouge (băt`ən r  zh) [Fr.,=red stick], city (1990 pop. 219,531), state capital and seat of East Baton Rouge parish, SE La. , LA, USA. pp 406-407.
Underwood, A. J. 1981. Techniques for analysis of variance in experimental marine biology marine biology, study of ocean plants and animals and their ecological relationships. Marine organisms may be classified (according to their mode of life) as nektonic, planktonic, or benthic. Nektonic animals are those that swim and migrate freely, e.g. and ecology. Oceanogr. Mar. Biol. Ann. Rev. 19: 513-605. Zar, J. H. 1996. Biostatistical analysis, 3rd ed. New York New York, state, United States New York, Middle Atlantic state of the United States. It is bordered by Vermont, Massachusetts, Connecticut, and the Atlantic Ocean (E), New Jersey and Pennsylvania (S), Lakes Erie and Ontario and the Canadian province of : Prentice Hall Prentice Hall is a leading educational publisher. It is an imprint of Pearson Education, Inc., based in Upper Saddle River, New Jersey, USA. Prentice Hall publishes print and digital content for the 6-12 and higher education market. History In 1913, law professor Dr. . C. OENGPEPA, (1) * I. HAWES, (1) I. LANE, (1,2) K. FRIEDMAN (3) AND J. BELL (4) (1) The WorldFish Center, PO Box 77, Gizo, Solomon Islands; (2) Currently at 18, Hibiscus hibiscus: see mallow. hibiscus Any of about 250 species of shrubs, trees, and herbaceous plants that make up the genus Hibiscus, in the mallow family, native to warm temperate and tropical regions. Drive, Valla Beach, NSW NSW New South Wales Noun 1. NSW - the agency that provides units to conduct unconventional and counter-guerilla warfare Naval Special Warfare 2448, Australia; (3) Secretariat of the Pacific Community, B.P. DS, 98848 Noumea, New Caledonia New Caledonia, Fr. Nouvelle Calédonie, internally self-governing territory of France (2005 est. pop. 216,000), land area 7,241 sq mi (18,760 sq km), South Pacific, c.700 mi (1,130 km) E of Australia. ; (4) The WorldFish Center, c/o Secretariat of the Pacific Community, B.P. DS, 98848 Noumea, New Caledonia * Corresponding author. E-mail: c.oengpepa@cgiar.org
TABLE 1.
Result of a 2-way ANOVA of number and size of P. margaritifera
spat on collectors grouped by collection period (1-6) and site (Noro
or Gizo). Numbers of spat were transformed using ln(N + 1).
Significant effects (P < 0.01) are italicized.
Variable Source of Variation DF MS F P
Number Site 1 0.01 0.01 0.88
Period 5 2.77 8.57 <0.001
Site x Period 5 0.05 0.17 0.97
Residual 82 0.32
Size Site 1 6.97 17.7 <0.001
Period 5 4.43 11.3 <0.001
Site x Period 5 3.34 8.8 <0.001
Residual 1794 0.39
TABLE 2.
Results of the 2-way ANOVA of effects of deployment period and
site on the number of dead P. margaritifera, and numbers of crabs
and Cymatium spp, on spat collectors at Gizo and Noro. All
variables were transformed using ln(N + 1). Effects considered
significant are italicized.
Variable Source of Variation DF MS F P
Dead spat Site 1 0.72 4.81 0.03
Period 5 0.55 3.70 0.004
Site x Period 5 0.29 2.94 0.10
Residual 82 0.15
Crabs Site 1 0.04 0.14 0.70
Period 5 0.64 2.05 0.08
Site x Period 5 0.73 2.33 0.04
Residual 82 0.31
Cymatium Site 1 0.76 8.01 0.006
Period 5 0.06 0.67 0.65
Site x Period 5 0.12 1.25 0.29
Residual 82 0.09
TABLE 3.
Results of the 1-way ANOVA of effects of year on the maximum
number of P. margaritifera spat (ln(N + 1)) on spat collectors
deployed at any period within that year at Noro and Gizo. Only
years when data from 10 individual spat collectors were available
for each of the six sampling periods were used in this analysis.
Site Source of Variation DF MS F P
Noro Year 6 2.84 6.77 <0.001
Residual 61 0.42
Gizo Year 6 1.63 3.49 0.005
Residual 62 0.46
TABLE 4.
Results of the linear regression analysis of the time-series from Gizo
and Noro of mean numbers of live spat (de-seasonalized), crabs and
Cymatium spp. Regression slopes (as change in catch [collector.sup.-1]
[y.sup -1]) are given only for significant regressions.
Site Variable [R.sup.2] F df p
Noro Live Spat 0.06 3.42 1,47 0.07
(deseasonalized)
Crabs 0.06 2.53 1,47 0.12
Cymatium spp 0.06 2.44 1,47 0.12
Gizo Live Spat 0.03 1.17 1,47 0.28
(deseasonalized)
Crabs 0.16 10.3 1,47 <0.01
Cymatium spp 0.00 0.11 1,47 0.74
Site Variable Slope [+ or -] error
Noro Live Spat
(deseasonalized)
Crabs
Cymatium spp
Gizo Live Spat
(deseasonalized)
Crabs -0.21 [+ or -] 0.07
Cymatium spp
|
|
||||||||||||||||||
Printer friendly
Cite/link
Email
Feedback
Reader Opinion