ebook img

Historical and present status of the pearl oyster, Pinctada margaritifera, at Pearl and Hermes Atoll, Northwestern Hawaiian Islands PDF

2006·5.5 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Historical and present status of the pearl oyster, Pinctada margaritifera, at Pearl and Hermes Atoll, Northwestern Hawaiian Islands

HISTORICALAND PRESENT STATUS OFTHE PEARLOYSI EK, PI\CT\DA MARGARITIFERA,AT PEARLAND HERMESATOLL, NORIHW ESlERN HAWAIIAN ISLANDS BY ELIZABETH E. KEENAN', RUSSELL E. BRA1NARD^ and LARRY V. BASCH' ABSTRACT Populations ofthe black-lipped pearl oyster, Pinctada margaritifera, at Pearl and HennesAtoll in the Northwestern Hawaiian Islands were first reported in 1928 and heavily harvested overthe next 2 years.Approximately 150,000 pearl oysters were eitherexported orkilled during the exploitation. An expedition in 1930 to assess post- hai"vestpopulation status found 480 P. margaritifera and determined the population to be severely depleted. Limited surveys in 1994 and 2000 found only a few pearl oysters and led to the conclusion thatthe population was still depleted. In 2003, the National Oceanic andAtmospheric Administration (NOAA)-led multi-agency marine debris removal team spent several months conducting surveys at Pearl and HermesAtoll that includedquantitative observations ofPinctadamargaritifera. Data were collected on location, size, depth, habitat, and orientation ofindividual pearl oysters on the reef Analyses ofthe 1930 and 2003 data sets revealed similar size-frequency distributions of theP. margaritifera population. The population has a spatial distribution within theAtoll similarto the 1930 post-harvest distribution, and some sustained level ofreproduction. Density and depth distribution comparisons from the two survey periods suggest that pearl oysters are significantly more abundant in the shallow waters where they were harvested during the fisherybut at a similardensity overall as they were during the 1930 survey. Although no estimates ofabsolute population size are available for any time period, the large numberofoysters harvested priorto the 1930 survey, togetherwith estimates ofoysterdensity in 1930 and 2003. suggest that the population may neverhave recovered to its pre-exploitation level. INTRODUCTION Thepearl oyster, like other shellfish and many othermarine animals (e.g., abalone; Tegneret al., 1996), has a long history ofexploitation throughout the world. 'JointInstituteforMarineandAtmospheric Research,UniversityofHawaiiandNOAAPacific Islands FisheriesScienceCenter, 1125-BAlaMoanaBlvd.Honolulu.HI 96814USA. E-mail: ekeenan(amail.nmfs. hawaii.edu ^NOAAPacific IslandsFisheries ScienceCenter,2570DoleStreet, Honolulu. HI 96822 USA 'U.S. National ParkService, PacificIslandsCoralReefProgramandUniversityofHawaiiManoa,Hawai'i- PacificIslandsCooperativeEcosystem StudiesUnit.Honolulu, HI USA 334 Records from the pearl industries in India and Venezuela document the discovery, hai'vest, and eventual over-exploitation ofthese populations (Arunachlam, 1952; Romero et al., 1999). Pearl oysters have been prone to exploitation due to the considerable value ofthe pearls and the nacre, or"motherofpearl", ofthe shell, andbecause ofthe animal's sessile nature and tendency to occur in sufficient densities at shallow depths forrelatively easy collection. The first documented discovery ofPinctada margaritifera atPearl and Hermes Atoll in the Northwestern Hawaiian Islands (NWHl) (Fig. 1) was in May 1928 by Captain William B. Anderson ofthe Lanikai Fishing Company(Amerson et al., 1974). Forthe next 2 years, the pearl oysters were heavily harvested fortheirnacre. This shiny portion ofthe shell w/as exported to the U.S. mainland where it was used primarily to make buttons. Although documents concerning the harvest are wanting, conservative estimates are that the shells ofapproximately 100,000 oysters were exported (Galtsoff, 1933). It is estimated that about 50,000 more oysters were killed and discarded, some due to theirpoor shell quality and others in the search forpearls (Galtsoff, 1933).Afterthe extent ofthe hai'vest was realizedby the Hawaii Temtorial government, an expedition was undertaken to assess the population and atemporaryban onharvesting wasput in place. This six-weekexpedition, ledby P. Galtsoffinthe summerof1930, utilized several Filipino divers and produced a lengthy report including data on pearl oyster size, weight, location (Fig. 2, modified from Galtsoff(1933)) and survey effort. Galtsoff(1933) found 480 P. margaritifera andpronounced the population too depleted to sustain further harvesting. At this time theTerritory ofHawaii made the taking ofpearl oysters illegal without pemiission, and a resurvey was suggested in five years to assess the recovery of the population. Subsequently the industry collapsed, coinciding with replacement ofpearl shell withplastic forbutton making and the advent ofcommercial pearl oysterfarms. Due to the lack ofinterest in furtherfishing of/! margaritifera in Hawaii, the suggested 5- yearresurvey at Pearl and Hermes was not conducted; however, the species has remained under state protection since that time. 175°W 170°W 165"W 160°W 155°W ^N ,1 ^-^' Haw a ''^n ^ ^c ^^' ^ ^''^. .-<••* O* ^-' «5' 125 250 500 NauticalMiles 1 1 1 1 1 f 1 1 1 175°W IZO-W Figure 1. MapoftheHawaiianArchipelago. 335 Figure2. (FromGaltsoff, 1933). Pearl oystersur\'eysitesat PearlandHermesAtoll fromthe 1930sur\'ey. Thesingleblackcirclerepresentsthehighestrelativeabundancefound(33 oysters/diverhour)andis representedas 100percent. Othercircles indicatetheproportionalrelativeabundancefoundatothersitesin the 1930survey.Whitecirclesaresitesatwhichnooysterswerefound. There have been three recent surveys at Pearl and Hermes which included documentation ofpearl oysterpresence: (1) by the U.S. National Marine Fisheries Service (NMFS) in 1993 (Moffitt, 1994); (2) the 2000 NWHI ReefAssessment and Monitoring Program (NOWRAMP) expedition (Maragos and Gulko, 2002); and (3) the NOWRAMP 2002 expedition (Basch, unpublished data). Each ofthese surveys reported onlyafewpearl oyster sightings, suggesting that the population did not rebound from the harvesting event andremained severely depleted. However, researchers had insufficient datato determine an accurate status ofthe pearl oysterpopulation. The 1993 NMFS effort was a two-day survey ofthe general areas which had the highest pearl oysterdensities in Galtsoff's (1933) survey. The densities at the three sites assessed in 1993 were found to be lowerthan in 1930 (Moffitt, 1994). One problem with the 1993 survey is that the methods usedto detemiine the locations ofsurvey sites in 1930 were not sufficiently accurate. Theywere comprised ofcalculations using triangulation ofdistant markers 336 and dead reckoning; moreover, these methods were made less accurate bythe scientists' inability to navigate straight lines through the shallow, reticulated reef. Consequently, it is notpossible to locate the 1930 sites with enough accuracy to make site-by-site comparisons overtime, particularly considering the patchy distribution characteristic of pearl oysters. NOWRAMP The 2000 and 2002 ciiiises were not specifically focused on surveying forpearl oysters. The relatively small areas sui-veyed were selected to record detailed infonnation on the fish, algae, corals, and otherinvertebrate species present. Pearl oysters were also recorded on some ofthese transects. Areport documentingthe results ofthe 2000 cruise states that only a few oysters were found, and they were smaller than those taken in 1930 (Maragos and Gulko, 2002). The transects were purposely located on vaiyinghabitattypes and many were not inpreferredpearl oysterhabitat. The few observations made on the status ofthe pearl oysterand the constraints ofthe surveys limitthe usefulness ofthese surveys fordetennining the status ofthe population. The purposes ofthis study were to: (1) accurately documentthe recent status of the pearl oysterpopulation at Pearl and Hermes by means ofa systematic, quantitative, and broad-scale survey oftheP. margaritiferapopulation at theAtoll, and (2) make initial comparisons betweenhistorical andrecent surveyresuhs. MATERIALSAND METHODS Data Collection As part ofan ongoingNOAA-ledmulti-agency effortto remove derelictfishing gear and othermarine debris from the coral reefecosystems ofthe NWHI, divers from NOAAs Coral ReefEcosystem Division (CRED) have methodically and systematically surveyed large areas ofthe shallow waterreefhabitats at Pearl and HermesAtoll (Donohue et al., 2001). Since 2003, the survey protocols have included extensive pearl oysterobservations. Divers surveyed reefs using snorkel gearwhile swimming orbeing towed along patch orreticulated reefs.Areas surveyed were recorded using a Garmin Geographic Positioning System (GPS) 12 (NAD84) in a small boat closely following divertracks. For each pearl oysterobserved, latitude, longitude, size, and depth were recorded. Maximum shell length and width were measured. Lengthwas measured as the maximum dorsal ventral measurement (DVM), andwidth was recorded as the measurement ofthe shell perpendicularto the length. Fora small number (approximately 10 percent) ofthe oysters no measurements, depths, or locations were recorded. Since identification ofjuvenile recmits to species requires more time and greatertaxonomic skills than were available, and usually requires obsei-vation inthe laboratory,juvenile oysters (<1.5 cm)were not included inthe data analysis. For a subset ofobservations, additional datawere collected, when time allowed, on habitat (substratum, dominantbiotic covercategory), and orientation ofindividual oysters on the reef. For40% percentofobservations substrate was documented, and for 59% percent orientation was recorded. Habitatwas characterizedbypercent coverofthe 337 substratum types in the 1 m- area centered on an oyster. The substrate categories were recorded as algae, sand, coral, and coral cement. The algae category consisted ofmacro- algae only, which were not identified to species. The coral cement category encompassed coral rubble and dead coral either exposed or with associated turfor coralline algae. Orientation, the angle between the plane ofthe oyster's shell and the substrate it was attached to, was classified as horizontal, vertical ordiagonal. DataAnalysis Field data were transcribed daily to an Excel worksheet containing all parameters foreach oyster. The GPS tracklines and waypoints were imported into ESRl Arcview 3.2 Geographic hiformation System (GIS) software, where they were used to map both the reefs surveyed and the point location ofoysters on those reefs (Fig. 3). The total area ofthe reefsurveyed during 2003 was detenninedby creating polygons inArcView which delineated the reefcontours ofareas where divers swam. These polygons were created using an Ikonis satellite image ofthe atoll (Fig. 4). The areas ofall polygons were added to obtain total reefarea suneyed. Several manipulations ofGaltsoff's (1933) observations were performed to enable comparison ofhis and oursurveyresults. Galtsoff(1933) reported survey effort in diverminutes. He reported thatthe divers covered a reefat the speed of42.7 ft/min (0.01 km/min), and that in orderto coverthe entire breadth ofthe reefthe divers swam a zigzagpattern with a width of60 to 100 ft. Galtsoffreported survey effort only for areas where oysters were found. In orderto compensate forthe rest ofthe survey area effort, we assumed that the average effort fora site with no oysters reported was approximately the same as forthe sites where oysters were observed. There were 32 siteswith oysters and 32 without, so the total minutes were doubled for a best approximation ofsuney effort. To estimate the distance sui"veyed from effort we multiplied the survey rate (0.01 km/min) by total minutes (4,562 min) fora result of58.9 km. We multipliedthis distance bythe associated width (60 to 100 ft. or0.01 to 0.03 km) to estimate the survey area covered, with a result of 1.1 km- to 1.8 km-. RESULTS Atotal of 1,057 pearl oysters were found at Pearl and HermesAtoll during the 2003 summer survey. The pearl oysters were distributed primarily throughout the inner lagoonarea (Fig. 2) with the exception often observations where individuals were found on the sandflats orouter fringing reefs. The lagoon habitat was surveyed by swimming only; we did not factor in the towed-diver survey areas in ourdensity estimates as they were largely performed over sand and on habitat unsuitable forpearl oysters. This facilitated comparisons with Gahsoff's (1933) results as observers in the 1930 surv'ey intentionally avoidedthe sandflats. Area computations using GIS resulted in a total lagoon survey area of5.9 km-. With an observed total of1,047 pearl oysters in this area, we calculated an average density of 177 pearl oysters/km- in the lagoon area surveyed. 338 Figure3. Distribution mapof2003 pearl oystersightings. PearlandHennesAtoll.Areassurveyedby swimmingaredisplayedasblack lines,andoystersarerepresentedaswhitesquares. Onlythesurveyareas andoysters inthe innerlagoonwereusedinthedensitycalculations.Theblackboxrepresentsthearea portrayedinFigure4. This density estimate assumes that all oysters present were observed bythe divers, when in reality some oysters were missed. Therefore, this is not an estimate ofabsolute oyster density, but a density estimate that can be compared to the 1930 survey, assumingthat in each study there was the same probability that an oysterpresent in the surveyed areawas m m observedby the divers. Pearl oysters were found at depths ranging from 0.31 to 6.1 with amean of1.36 m and standarddeviation (sd) of0.87 m (Fig. 5). The average shell length ofpearl oysters measured was 20.2 cm (sd=4.76 cm, n = 963). Shell length ranged from 1.5-33.0 cm. Pearl oysters smallerthan 1.5 cm were excluded from analysis since oysters ofthat size could notbe accurately identified 339 Miles Figure4.Anenlargementofasectionofthe lagoonatPearland HenriesAtoll (Areaportrayed isoutlined bytheblackboxinFigure3).Theblackareasarethepolygonscreated inArcview3.2todelineatethe surveyedreefareaatPearlandHermesduring2003. Pearloystersarerepresentedbywhitesquares. to Species in the field. TheP. margaritifera size frequency distribution (Fig. 6) has a single mode. However, immature oyster recruits, or spat (shell length <5 cm), were excluded from the size-frequency data set. The mean shell length of20.2 cm found in 2003 is remarkably similarto the mean shell length of20.23 cm forthe 164 adult pearl oysters measuredby Galtsoff(1933), although the distributions have different shapes (Fig. 6). For419 (40% oftotal surveyed) pearl oysters observed, data were recorded on substratum type. Within the lagoon, the typical substratum composition consisted of: sand 11% (sd=24.8), coral 13% (sd= 17.3), algae 28% (sd= 32.1), and coral cement 48% (sd = 34.6). The oysters were found in various orientations. In the subset ofoysters forwhich orientation datawere collected (n = 624), mostwere horizontal (53%)). Ofthe remaining oysters, 32%) were vertical, and 15% were diagonal. 340 Percent of Population (%) 2 4 6 8 10 12 14 16 18 0.305 1.22 2.135 3.05 -^4.575 Q T 5.49 7.32 8.845 10.37 12.2 14.03 Figure5. Depthrange frequencydistribution forpearl oystersurveys in 1930and2003. Percentvaluesfor the 1930datasetareestimatesbasedonthegiven meandepth rangeandminimum andmaximumdepths fromGaltsoff(1933). 25 2- 20 c o iS 15 3 Q. O Q. 10 0O) <u Q. 9 11 13 15 17 19 21 23 25 27 29 31 33 Shell Length (cm) Figure6. Size-frequencydistribution ofthepearl oysterpopulationatPearl andHennesAtoll. NorthwesternHawaiianIslandsin 1930(N=197)and2003 (N=963). 341 DISCUSSION In distribution maps (Figs. 3 and 4) \\c indicate that pearl oyslers are wides|-)icad throughout theAtoll lagoon. When comparing our results with GaltsolTs 1933) post- ( harvest data, we considered whetherchanges in the population occurred since that time. By visually comparing the maps from the 1930 and 2003 surveys (Figs. 2 and 3), a general idea ofthe difference in spatial distribution can be obtained. Although the locations from the 1930 map (Galtsoff, 1933) are only roughly estimated, no oysters were recorded in 1930 forthe reefs in the southeast and south ends ofthe islands. When those areas were surveyed in 2003, relatively high levels ofpearl oysters were observed on almost all reefs. It is likely there has been new recruitment and population expansion into this region since the 1930 survey. Moreover, reefs in the south central and north central lagoon, where some oysters were seen in 1930, were not surveyed in 2003; some oysters maybe present in these areas. Determining whetherPiiictada margaiitifera populations have recovered to pre-harvest levels is complicated by the factthatthere are no estimates ofpre-harvest population density, and no estimates ofabsolute oyster abundance at Pearl and Hermes forany time period. Comparisons between early and recent post-harvest data sets are facilitatedbythe fact that Galtsoff(1933) did report numbers ofoysters found and the survey effort, which can be used to estimate oysterdensity in the 1930 survey. After converting Galtsoff's (1933) reported effort into survey area, we determined an average densityof209 to 349 pearl oysters/km- during his surveys. In our2003 survey, we estimated an average density within the lagoon areas of 77 pearl oysters/km-, lower but 1 ofthe same order ofmagnitude as the density found in 1930, and presumably lower than the densityjustpriorto exploitation. Given the lack ofdatabetween the two surveys, changes in pearl oysterabundance during the intervening 73 years cannot be detennined. However, ifabundance has not reached pre-e.xploitation levels, it is useful to ask why. One explanation forthis wouldbe that adultpearl oysterdensities were reduced by exploitation below a threshold whereAUee effects (or inverse density dependence) came into play (Levitan, 1995). Pinctadamargaiitifera is abroadcast spawnerwith planktonic larvae (Pouvreauet al. 2000); consequently, reduced adult densities could have imposed a directbottleneck on fertilization success, and subsequent embryonic, larval, and recruitment success (Pouvreau et al., 2000). With a lowered adult density there would be less likelihood that female gametes wouldbecome fertilized in the watercolumn, as has been shown foroctocorals, sea urchins, abalone, and other sessile or sedentaiy benthic marine invertebrates (Levitan, 1995; Tegneret al., 1996; Coma and Lasker, 1997). Subsequently, ifa larva was produced and dispersed proximate to a suitable settlement site, the likelihoodthat it would encountera settlement cue associated with an adult shell alsowould be more remote. In otherwords,Allee effects would be furtherenforced given that pearl oyster larvae tend to settle gregariously on the shells ofadult oysters (Pascal and Zampatti, 1995; Zhao et al.. 2002). Comparison ofthe pearl oysterpopulation depth distribution between the 1930 and2003 surveys shows some intriguing differences (Fig. 5). In 1930, oysters were reported as ranging from 2.5 to 15.0 m, and were most abundant from 4.4 to 8.3 m. 342 Galtsoff(1933) also reports that, according to CaptainAnderson, when the oysters were first discovered they were very abundant in waterdepths of 1 to 3 m. In our 2003 survey, we found oysters from 0.3 to 6.0 m depth, but animals were most abundant in the 0.5 to 2.2 m range (detennined using the mean of 1.36 m + 0.87 m sd). In Figure 5, we illustrate the difference in the depth ranges between the two studies. The absence ofany oysters in waters shallowerthan 2.5 m in 1930 is evidence ofthe heavy harvesting effort at these shallow depths in the immediately preceding years. Seventy-three years later, we found oysters to be very abundant at these shallow depths, suggesting that the remaining population contributedto areseeding ofshallow areas ofthe reef What remains elusive at this time is an explanation forthe apparent scarcity ofoysters at deeperdepths in our survey. The most likely explanation is that oysters still occur in higherabundance at these greaterdepths but that our sampling did not detect them. Since the lagoon surveys were perfoiTnedby snorkeling, the divers spent most oftheirtime at ornearthe surface while surveying. Oysters which may have been on the deeperreefslopes mayhave been missed because oftheirsmaller size in relationto othersearch images (since the primarymission was to locate generally largermarine debris) and the greaterdistance with depth fromthe divers. We have no reason to believe that the lagoonal reefs have changed in a way that would impose biological limits to the depth range ofthe oysters. Alternatively, though less likely, pearl oysters may recruit preferentially to shallower depths and may be less abundant in deeperareas due to this preference in combination with (1) reduced adult densities at depth sustained overthe post-harvest period and (2)Allee effects. Adirected survey forpearl oysters at Pearl and Hermes conducted along multiple-depth contours would help determine the distribution and other population parameters ofoysters at deeperdepths. The present study indicates that thePinctada margaritiferapopulation at Pearl and HenriesAtoll is reproducing at some level, as indicated by individuals ofabroad range ofsize classes, including recniits. The mean shell length ofthe 963 pearl oysters measured in the 2003 survey was 20.2 cm. The oysters were foundpredominantly on coral cement and macro-algae dominated habitat. This observation contrasts with Galtsoff's (1933) report that most oysters were found "confined exclusivelyto those sections where the bottom is covered with corals." Initially, we thoughtthatthe difference between surveys in composition ofoyster-occupied habitat might be attributed to differences in the depth range, but examination ofthe data showed similar coral percent coveratall depths. The shell orientation ofthe oysters was measured in our surveybecause ourinitial observations oforientation were inconsistent with a comment in Galtsoff's (1933) report. Galtsoff(1933) noted that oysters were found in a vertical orslightly inclinedposition, while we commonly observed oysters in a horizontal position. Our results indicate that only about 1/3 ofoysters were oriented vertically, and> 1/2 were horizontal. These differences in orientation maybe aresidual artifactofharvesting, ormayreflect depth- related differences in the nature ofnear-boundary layerwatermovements which the animals mayrespondto by orienting themselves, eitherto minimize drag dueto sheer forces, orto optimize filter-feeding efficiency in differentflowregimes. We report the first systematic, quantitative survey forpearl oysters throughout

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.