Seasonal Colonization of Low Profile Artificial Reefs in Mississippi Coastal Waters: Invertebrates
HARRIET M. PERRY, KIRSTEN LARSEN, JAMES WARREN, and CHRISTINE TRIGG The University of SouthernMississippi, Ins¹tute of Marine Sciences Golf CoastResearch Luiboratirry, P.O. 7000 OceanSpri'ngs, Mississippi 39566-7000USA
ABSTRACI' Artificial reefsserve as fish attractanlsand may increaseproduction of some speciesby increasinghabitat. In an effart to enhancealready established ~onal fisheries and to itxzease numbers of and acct to less common structure-associatedfishes, Mississippi began building new low profile artiTicial reefs and auginentingexisting ones. Over twenty inshore low profile oyster shell reefs were devdaped prior to 1995. Subsequentreef developmenthas utilized limestone gravel in conjunction with oyster sheUor limestone alone. Thecteation of limestonegmvel and/or oyster shell reefs provided an opportunity to obtain information on the colonization of these different substratm by benthic invertebrates.Colonization studies were conducted using crates of artiTicial substrateplaced on a graveVoystershell reef locatedapproximately 300 meters offshore in central Mississippi Sound, Crates contained crushed liinestone gravelor oystershells. Crates were pulled every three months and all organisms were removed. Invertebrates were identified to the lowest taxonomic level, measuredto the nearestO. I mm and weighedto the nearest0.00l g. Dominant invertebtatetaxa included representativesof the following groups: Xanthidae Menippeadina, Zurypanopeus depressus, Panopeus simpsoni!, potcellanidae Petroiisthesarrnatus!, Alpheidae Alpheus angtdatus,A. beterochaeiis!, Palaemonidae Palaernonetes vrdgaris!, Amphipoda Meiita, Apocorophium!, Polychaeta Neanthes saccatea!, Bivalvia Ischadiurnrecurvurn!, and Gastiopoda Strarnonitahm~ia!. Observeddiffetenctn in numbersof individualsand sizebetween summer and winter collectionsappeared to be relatedto recruitment. Dataare presented on speciesabundance and size by substratetype and seasan.
KEY WORDS:Estuarine, invertebrate assemblages, low proQe artificial reef
INTRODUCTION AriiTicial reefsserve as fish attractmtsand may inc@meproduction of some speciesby increasinghabitat. In an effort to enhancealready established recrea¹onal fisheries and io incmm numbers of and access to less common struc~'ociated fishes, Mississippi began building new low profile artificial Perr tl.N. et Il. GCFl:52 2001 reefs and augmentingexisting ones. Although over twenty low profile artificial reefs oyster shcH, concreterubble, limestone gmvcI! have been constructedin Mississippiinshore waters, there are uo dataon reef communitystructure or the association af fish populations with these reefs. A study of the faunal assemblagesassociated with reef colonizahan in Mississippi Sound was begun in December1998 as part of a largerprolpam to assessproductivity of these xeefsin relationto recreationalfishing opportunities. The xescexehreported on hexeinis part of a long-termstudy addressing seiLamal colonizationtsuccession of faunaassociiited with hmestaueaxal oyster aheH reefs in estuarinewaters of MississippiSound. Data presentedrepresent initial colonizationstudies conducted in the sununer of 1998 and the winter of 1998t99.
MATHUALS AND MEIHODS Colamzafionwas studied by placinga seriesof cratesfxHed with 0.025m3of limestonegravel ar aystershell an a newly created,neaxshaxe limestanejsheH reef. The areaadjacent the artificial reef is chlxicterizedby smallpatch subtidal oysterbeds. Averagewater depthis approximately1.5 m. Site locationis shownm Rguxe 1. Thxxty-twosamplexs 6 hmestone,16 oystershell! were deployedin thesummer of 1998. The crateswere placed on eightplastic paHels, four trays to a paHet Rguxe 2!. Four palletscontained crates filled with Iimestoxxeand four palletscontained cxales with oystershell. AH cratesand palletswere laMed. A samplmgscheMe was establishedthat removedone crate fxam each pallet four limestonegravel and four oyster sheH!after an initial soak time of three mouths. At six, nine, and twelve month intervals the xexxumihxgcrates were pulled four limestonegravel and four oystercxates per sampling period!. AdditianaHy, a set of crates was xeplaced four limestone gxaveIand four oyster! each sampling period so that a thxee-manthset of samples was obtainedfor each season. Original samplingschedule caUed for removal af a set of thxeemonth samples and a set of time and six monthsamples during summerand faH af 1998. n~4veiy. Sums sampleswere collected; hawever, HuxxicaneGeorges in September1998 destxayedthe paHetsand no fall sampleswere coHectoL Samplers were xe-deployedin December 1998. Data presentedin this studyaxe from the threemonth summersamples original study! and the three month winter saxuplesthat began the new samplingregime. Crateswcxe iemavol fxam the water, inunediatelyplaced in seasater soaked oystersacks and retxxxned ta the GuH'Caut ResxmchI~sy far proceming. Contentsof eachcrate wexe washed over scteening winter samples!and aH arganisinscaHectixL Samples were frozen prior to analysis.Each sainple was sortedto speciesand the total number and weight nmxsxlaL When available, fifty individualsof eachspecies were randomly selected for ineasuxement.Individuals were measuxzdto the ~ 0.1 mm in length or width brachyurancrabs! using Proceedin s ot the 52nd Gulf and Caribbean Fisheries Institute digitalcalipers. A Sarloriusanalytical balance was used to measureweight to the nearest0,001 g. Y~al-Wallace ANOVA was usedto test for significantdifferences in speciesabundance and sizebetween seasons by substrate.Student's r-test was usedto comparespecies abundance and size between substratm within a season, The level of significancefor all testswas set at aW.05 Data were analyzed usingthe statisticalpackages SPSSx and Quatre Pro Version7.
Figure 1. Lacatianot kwvproiile artllchi reef. Perr H.N. et nl. GCFI:52 200$
Fig txre 2. ramps array
RESULTS AND DISCUSSION Invertebratesassociated with hmeslonegravel and oyster shell suhxtratesfor sutmnerand winter saxnples axe listed in Tables1 and2. Limestonegravel is hereafterreferred to asgravel and oyster shell as shelL Becausethe present data areconcerned with a hmitednumber of samplesthat represent a small portion of the uncompletedtotal study,xesults are presentedwith minimalinterpretation and discussion. For many species,observed differences in numbers of individualsand sizebetween stunxner and winter collectionsappeanR to be related to xecruitmeut,Most of the speciesoccurring in sampleshave been identified as oysterreef associatesin MississippiSound Heard 1979!. The threexanthid crab speciesate Itnown oyster predatorsand Neanthessxxccinea is one of the most comnKsxlyencotmtered nexeids in oyster reef communities. Aklita ni¹da, Isdhutiumrecurvxxm, Pahemonetes vrdgxxris, Alphms aegxxlatus,and A. heterochaeCs'ateIntown from oyster reefs in the area Proceedin s of the 52nd Gulf and Caribbean Flsherles lnstltute H.M. et al. GCFl:52 2001
O
bO O00 O 'Ct O OCV 00 O+ O
O
OCI
O O rt a cu O
CV OQ m O O
V IVi D 4l O 'oa I O 4l M U
0 b Proceedln s of the 52nd Gulf and Caribbean Pisheries Institute
Crustaceans Amphipodswere the mast abundant associatedwith the reef materials,Because the size af theseorganisms allowed. for escapement throughthe small holes in thecrates during sampler retrieval and use of netting to retainthese organisms wasn't begununtil the winter samples,numerical estimatesof relative abundanceare available only for the latter collections. Amphipodswctu identified using the keys of Sheridan 979! andBousfield and Hoover 997!. Two generaof free-livinggammaiideans dominated tbe amphipodfauna on both graveland shell substratcs:Melisma and Apocorophium Table 2!. Speciesof both generaare associatedwith oysterax.fs and shell materialin Mississippi water Heaxd1979!. MelitanMda was exbnneIy abumiant; Mehta Iongisetosa andApocorophium louisiansrn = Corophiumlouisiansm! were common in samples.There was no significant difference in speciescomposition between the two substratesin wintersamples. Monocorophium achenrsicum =Corophium adierusinun!was found only in oystershell materialand in limited numbers. Furypmopeusdcprcssus was the mostthe mostabundant biuchyumn crab collected.This speciesis associatedwith oysterreefs throughout much of its rangeand it is oftenpresent in largenumbers Ryan 1956,Tabb and Manning 1961, Rouse 1970!. It is a known predatorof oyster spat McDermott 1960!. Ovigerousfemales were numerous in summersamples in bothsubstrate types. Thirty-onepercent of the femalescollected in graveland 25% af Qmfemales taken in shell were egg-beating. Size of ovigemusfemales ranged from 3.8 to 13Amm in campacewidth CW!, Widthfrequency distributions for summerand winter samplesby substrateare shownin Rgure 3. Recruitmentwas evidentin summersamples and newly recruitedcrabs were abutuhmton both subsbabm. Therewas no sigmficantdifference in abundanceor size of crabsin either substrate Table 1!. As in summer samples, winter densities in the two su~ were not significancyditTerent; however, there was a significant difference in size of crabs occupying giavel and shell during the winter with meancarapace width greater in shell Table2!. Compaxingseasons, there were significantdifferences in ciubdensities and size between summer and winter in both graveland shellwith crabsmore numerous in summersamples on both substrates Tables 3 and4!. Crabswere significantly larger on both substratesin winter samples. Increase in size in winter samples is, in part, related to decreasedrecruitment and growth of summerrecruits. Variables responsible for reductionin crabdensities in winter samplesmay include emigrationin iesponse to environmental conditions ancUor mortalities related to predation and campetitioufor food. Gneatmabund.mce of crabsin oystersubstrate in winter inay be relatedto their increasedsize and larger niche availabilityin shell habitats. Parr H.M. et al. GCFl:52 2001
Table 3. Seas@el comparison of numerical abUMfance and size for E. deCrsssusand N. success N.S.D. = no slgniTicantdeference, LS = Rmestone, OY = oyster!.
Table 4. Seasonalamyarhon of meansize for E depressusand N. succfnea {N.S.D.= no signNcantdifference, LS = limestone,OY = oyster!. Proceedln a of the 52nd Gulf and Caribbean Fisherfea Institute
Ih O h O Ih O Ih O Ih Ih O Ih N N O ih O Ih O Ih [U&3iVQ !UDMUJ
N
O II
Ui O O Ih O Ih O CV hiO Ih O rh Ih N CV !UBiI1BJ ~iL5$
Figure 3. Widthfrequency distributions of 2'ugyanopeusdeprmmm Perr H.N. et al, OCFI152 2001
Stone crabs, Merrqye adlna, were common only in summer samples,a single specimenwas identified from winter collections. Stone crabs were numerousin both substratetypes and there was no significant difference in abundanceor size betweengravel and shell Table I!. The majority of the crabs collected were under 10.0 mm CW Figure 4!. Stone crabs spawn in the suminer and recruits and easy juveniles are common on oyster reefs and stone rubble breakwatersin Mississippi Sound Stuck and Perry 1992!. ~k of stone crabsin winter samplesmay reflec seasonalemigration from shallow reef areas and/Orinrseased predanOn as8~ with decreasedniche avaihbihty for larger juveniles. Size-specific shelter limitahon has been implicated a8 a factor in determining population levels of large juveniles in northern GOM estuaries Beck 1995, 1997!.
35 30
10
0 0 2 6 10141$22263D343842465D545$62 2 6 1014 18 22 26 30 34 38 42 46 50 54 58 62 Carspscewidth mm! Cen8wcewtd3t mm!
Figure 4. Width frequency distributions of htsrtippe adina ln sunmar samples
Other crab speciesoccumng in samplesincluded Simpson's mud crab, PanopeL8simpsoni, the porcellanidcrab, Perrolisdiesarxrrarxir and the striped hermit crab, C 312844xriusvirrrxtxis. Poxcelaincrabs and the striped hermit were collected only in summer samplesand thine was no significant difference in density between the two substrate types g able 1!. Perrolisrhes arrntxtx6ris a filter feederbut this specieswill also ~ algae froin rock and shell Fotheiingham and Brunenmeister1989!. The striped hermit is an omnivorous scavenger common in tidal mesh habitats,on rock jetties, and around mud flats Heard 1979!. Adults often utilize shellsof theoyster drill. Simpson'smud crabswere Proceedtn e of the 52nd Gulf arid Caribbean Fleheriee Inetltute collectedin bothseasons in small numbers. ln summersamples, they were significantlymore abundantin gravel,but were significantlylarger in shell Table 1!. In wintercollections, there was no significantdifference in numbers or size between the substratetypes, but larger individuals were associatedwith shell Table 2!. Paeopeussinqrsom is an oysterassociate that occurs subtidally and inteitidallyin MississippiSound and it oftenco-occurs with E, depressions Heard 1979!, Callineciessapid', theblue crab, and C. simiiis, thelesser blue crab, were found onIy in winter samplesin sinall numbers. There was no significant difference in the nmnbersof blue crabsbetween the two substrates,but crabsin shellsamples were significantlylarger Table 2!. Blue crabsin wintergravel sampleswere summer/fail recruits mean size was 0.0 mm CW! thatmove to deeper waters with de~ng water temperatures Steele and Perry 1990!. Peak settlementof bluecrab megalopae occurs in thelate summer/earlyfaII Perryet aL 1995,Johnson snd Perry 1999!. Bothmegalopae and early crab stages are associated with shallow neaxshore waters. The snappingshrimp, A/pheus angpkers, was foundonly in the summer. Densitieswere significantly higher in gravel,but sizewas significantly larger in shell Table 1!. A single specimenof A/pheur hererochcvlis,occurred in winter samplesin gravel Table2!. Bothspecies of Alp/ieusaie associatedwith oyster reef communitiesin the northernGOM McClure 1995!, The grassshrimp, Pi/uern0nerer vuigaris, was common in summer samples. There were significantdifferences in summercatch of grassshrimp between substrates, but no significantdifference in size Table I!. Grassshrimp were more abundant in shellthan in gravel, Heard 979! notedthat this speciesof grassshrimp was often associatedwith foulingcommunities and shell bottomsand may occur with P. piigioaround oyster reefs.
Poiyehaetes/pfatyhelmtsiths The polychaete,N~ suceima,was the only other speciesabundant enoughduring both sampling periods to allow for compiuativetesting between seasonalityand abtus9anceand size and to test for seasonaIdifferences in abundanceand size between substrates. There were no significantdifferences in polychaetedensities or size betweenwinter and summersamples far either substratetype Tables 3 and 4!. In comparing gravel and shell substiutesin summersamples, there was a sigmficantdifference in densityand size of polychaetes.Polychaetes were mcse abundant in graveland were hrger gable 1!. In winter samplesthere was no significantdifferenc in densityor size Table 2!. Hatwotms turbemarians!occurred in summersamples with no significantdifferenc in numbersbetween the two substratetypes. Perr H.M. et al. GCFli52 2001
Molluscs Molluscs associatedwith samplesincluded the ~ mussel, Isr~ renrrvirIri, the oyster drill, Rramonita haemaslomn, and the southern nassa, Nassariusvibcx. Only hooked musselsoomrted in both summer and winter samples Musseisin summersamples occurred in both substrateswith larger individuals found in shell samples Table 1!. Small size of specimens in sumtner samples and much teduced nmnbeis of individuals in winter suggests recruitment and settlementoccuis in the warmer months. Oyster drills were significantly more abundantand larger in shell samplesthan in gravel Table 1!,
CONCLUSIONS High abundanceindices for many of the invertebiatesappearing in summer samples were related to tecnitment Reduced numbers of individuals in winter samples~mured as organismsmigrated from the reef area with falling water temperaturesin the winter or were subjectto increasedpredanon rates associated with a larger suite of predators in warmer months. Early juvenile stages of brachyuiancrabs suffer high mortality rates with both intra- and inter-specific predation influential in contmlling population size. Presentdata representa small portion of the overall study. Completion of the seasonid sampling regime will piovide a nore comprehensivedata set froin which to evaluatethe effect of season and substiate on invertebrate assembhges astxciated with low profile reefs in estuarine waters. This study also provides the baseline data on species compositionand abundancenecessary to evaluateutilization of thesereefs by rccteationalfish species.Concurtent studies indude a fisheryindependent finfis sampling prograin near reef areasand an analysis of the stomachcontents of selectedfish speciesin thosecollections. Creel surveysare also ongoing.
ACKNOWLEDGMENTS The MississippiDepartment of Miuine Resourcesprovided funding for this study through their Tidelands Trust Pmgram. We ate indebted to the many talentedinvertebrate taxonomists at the Gulf Coast ReseiuchLaboratory. Sara LeCroy, Richard Hau4 and Jerry McLeihnd provided patient and invaluable assi~ in teachingus to differentiat the amphipod and polychaetespecies taken in samples. Bradley Randall, Wes Devers, Jude LeDoux, and John Anderson assistedwith field retrieval of collectors. Lisa Engel GCRL! and Michael Buchanan along with additional personnel of the Mississippi I~ertment of Marine Resourcesassisted in sample wash down. Jamie McFemn, Damelle Slade,and Virginia Shervettehelped separate samples, We acknowledge Joanne Lyczkowski-Schulz of the National Marine Fisheries Service for sharing her photolpaphic equipment and expertise with us and Windsor Aguirre for photographs, Proceedin e of the 52nd Gttlf and Caribbean Fisheries Institute
LlTEPATURE CITED Beck, M.W. 1995. Size-specificshelter-Limitation in stone crabs: a test of the demographicbottleneck hypothesis. Ecology 7 6 !:968-980. Beck, M.W. 1997. A test of generalityof the effect of shelterbottlenecks in four stonecrab populations. Ecology 7 g 8!:2487-2503. Bousfield. EL and P.M. Hoover. 19%. The amphipod superfamily Corophioideaon the Pacific Coast of North America. Part V. Family Coiophiidae. Corophiinae, new subfamily, Systematics and distributional ecology. Amphipacifica 2 !.67-139. Fotheriugham,N. and S. Brunenmeister, 1989. Gidde to Gulf Coast Marine Life. Gulf Publishing Company,Houston, Texas. 142 pp. ~ R. 1979. Guide to common tidal marsh invertebrates of the northmtern Gulf of Mexico. Mississippi/Alabama Sea Grant Consortium, MASGP-79-004: 1-82. Johnson,D. R. and H.M. Perry. 1999. Blue crab larval dispersionand retention in the Mississippi Bight Bull. Mar. Sci. 65 !:129-149. McClure, M.R. 1995. Alpheusangulatus, a new speciesof snappingshrimp from the Gulf of Mexico and northwestern Atlantic, with a redemption of Alptieitsheterochaelis Say, 1818 Decapoda:Caridea: Alpheidae!. Proc. Biol. Soc. %'ash. 108 !:84-97. McDermott,J.J. 1960. The predationof oystersand barnaclesby crabsof the family Xanthidae.Proc. Penn.Acad. Sci. 34: 199-211. Perry, H.M. and K.C. Stuck. 1982. The life history of the blue crab in Mississippi with notes on larval distribution. Pages 17-22 in: H.M. Petty and W.A. Vau Engel eds.!, Proceedingsof the Blue Crab Colloquium,Publication 7, Gulf StatesMarine Fisheries Commission, Ocean Springs, Mississippi. Perry, H.M., C.K, Eleuteiius, C.B. Trigg, aud J.R. Warren. 1995. Settlement patternsof Callinectessalndus megslopae in Mississippi Sound: 1991, 1992. Bull. Mar. Sci. 57 !:821-KL3. Perry, H.M., J. Warren.C. Trigg and T. Van Devender. 1998. The blue crab fishery of Mississippi. L Shellfis Res. 17 !:4~33. Rouse, W.I 1970. Littoral CrtLstaceafrom southwest Horida. Quart. Jour. Fla. Acad. Sci. 32 !:127-152. Ryan, EP. 1956. Observationson the life historiesand the distribution of the Xanthidae mud crabs! of ChesapeakeBay. Amer. lNdL Natur. 5 6:138-162, Shendan, P.F. 1979. Three new species of Melita Crustacea: Amphipoda!, with notes on the amphipod fauna of the Apalachicola estuary of NorthwestRorida. NortheastGulf Sci. 3 !:60-73.
510 Perr H.M. et al. GCFI:52 2001
Steele,P. and H. M. Perry editors! 1990. The blue crab fishery of the Gulf of Mexico United States. A regional managementplan. Publication Number 21. Gulf States Marine Fisheries Commission, Ocean Springs, Mississippi. Stuck, K. and H.M. Perry. 1992. Life history characteristicsof Menippea&3a in Mississippi coastal waters. Pages 82-98 in: T.M. Bert ecL!, Proceedingsof a Syrnposiunt on StoneCrab GenusMenippe! Btotagy and Fishers. Fla. Mar. Res. Inst. Publ. 50. Tabb, D.C. and R.B. Manmng, 1961. A checklist of the flora and fauna of norlhern Horida Bay and adjacent brackish waters of the Florida nuunlaud collected during the periodJuly, 1957 through September, 1960. Bul. Mar. Sci. Gslf Caribb. 1 1 !:552-649. The Role of Nearshore Habitats as Nursery Gronnds for Juvenile Fishes on the Northeast Coast of St. Croix, USVI
IVAN MATEO and VGLUAM J, TOBIAS USVI Division of Fish and wildlife Room 203 Lagoon Srreet Complex St. Croix, US Vrgin Islands 00841
ABSTRACT Three protectedbadoeef embaymentson St. Croix's northeastcoast were sampledquantitatively to determinespecies composition and juvenile fish abundancefrom October1998 to September1999. The studysites consisted of Yellowciiff Bay, Teague Bay and Cottongardeu Bay. Juvenile reef fish assemblageswere monitored using thee complementarysampling methods;a visual strip transectcensus, fish txaps and beach seine net By comparing juvenilefish communitiesfrom distincthabitats patchreef, seagrass, rubble, algal plains.and sand! within the threeembayments, significant differences in speciesrichness and abundance of juvenileswere identified. Patch reef habitats had more speciesdiversity than any other habitatamong the thxeeembayments. Most juvenile fish observedamong the siteswere scarids,labrids and hacmulids. The slippery dick Halichoeresbiviualm, the bucktooth parxotfish Spariroma mdkmr, the spottedgoatfish Psemdxxpwneasnxacaiatas and Haemalonspp, were axnong the mast abundant species. Preference of neaxshorehabitats by economically important juvemle reef fishes, suggestsa serious concern for habitat conservation
KEY NORDS: Juvenile,nursery, recruitment
INTRODUCTION The role of coastalhabitats as potential nurserygrouxids for coral reef fishes and many invertebrateshas been widely acceptedby many authors Springerand McErlean 1962, Austin 1971, Macnae 1974, Beumer 1978!. Nearshore habitats such as mangrovesystems, seagrass ineadows and backreefareas constitute an array of multispecies communities and their ecological relationships and connectivity is often wide-rangingand complex. Acting together,these coastal habitatspmvide food, habitat,and shelterfunctions for many organismsin their early stages Ogdenand Zieman 1977,Shulman 1984, Shulman 1985!. Several investigatorshave identified essentialfactors possessedby neaxshoxehabitats within backreefaxeas such as food availability and potentialrefuge sites in these systems,which havebeen recalled as determmingfactors of a nurseryfunction. The intexaction between the structural and functional characteristics of these componentsresults in a heterogenicenvironment that providesrefuge and shelter
512 Mateo I. and W.J. Tobias GCFI:52 2001 as well as foodfor a greatnumber of organisms Phillips and Menez 1988!.The ecologicalimportance of variousnearshore hahitats in backieefareas as nursery attxts for fishes of ~onal and commercial fisheries have been documented Ogdenand Zieman 1977, Zieman et al. 1982,Robblee and Zieinaa 1984, Rookerand Deaais1991!. Bconotnicatlyimportant species have been reported to usethese habitats for nurserygrounds including grunts, parrotfishes, several speciesof snappe-rsand groupers, the spinylobster, the queen conch and others Boulon 1986, Stonerand Waite 1990, Appeldoornet al. 1997!. Major issuesof concernaffecting these nearshore environs include the status of the habitatssupporting the reaeationatand cx>mmercialfisheries, the fishes associatedto a parncularhabitat and to what extent degradationaf coastal habitats could adverselyaffect thesenursery grounds. The presentstudy presents iaforinationoa the communitystructure of the fish assemblagesia aearshoie habitatswithin backreef lagoon areas on the Northeastead of St. Croix Figure 1!. This information is essential in the ~emeat of these areas to allow enhancementof theirecological and fisheries value as nursery gruuads.
1-tptire $. Locatke of the three 'ernbayrnents studied at the Northeast Coast of St. Croix, USVL
MEIHODS The nearshorenursery habitat in threeprotected badmef embaymentsoa St. Croix's is~east coast were sampled from October 1998 to September 1999 to
513 Praceedin s ot the 52nd Gtiif end Caribbean Fisheries Institute quantitatively determinespecies composition and fish abundance. The study sites consistedof CottongardeaBay, Teague Bay aad Yellowchff Bay. The distancefrom the shoie to the roef within thesebays varies from approximately 500 to 700 m. These areashave a very high propensity for habitat degruMon dueto theimpact of humanactivities. Development activities include: a public beachand residentialshoreline development ia CottongardenBay, a yacht club, residentialhcsaes, restaurant and condominiumsin TeagueBay; aad residential developmentin Yellowcliff Bay. Rsh traps, visual traasects and seine nets were used as assessment techniques. Ten 50 m traasects were randomly establishedfor location aud direction ia each of the three backreef embaymeatsto conduct visual fish censuses.The total numberof transectsrequired to sampleaI habitattypes was basedon preliminaryfish censmcumulative species/~ counts. Two divers simultaneouslyconducted monthly fish censusesalong a 2-m wide strip either side of the transectline. Habitat type, fish species,fish numberand fish size rangewere recorded for eachtransect. Size categories were characterited by c5 cm, 5 - 10 cin, > I cm. For most speciesjuveniles less than or equal to 5 cm wererecorded as recruits. A total of 10 rectangular,92 x 57 x 19 cm, baitedfish traps made from vinyl-coated 1.3 cm wire mesh, weie set for 24 hours at randomly selectedlocations in each embayment. The traps were rebaitedaad relocated to the next embayinent. A total of two beach seine hauls were conductedin eacharea on a montMybasis during the trapping period. The beach seinenet measured308 m x 122cm with weightsand floats attached;mesh size was 13 cm. Two PVC pipes bent at the bottom were fixed vertically at the endsof thenet and pushed by two personsfor a sweptarea of 20 - 25m adjacent to the shoreline. All fish caught in the traps and seine net were identified, enum~ measured fork lengthaad total length! to the nearestmillimeter, and releasedat the point of capture. Total numberof fishesand speciesper censuswere examiaedby site and by habitatiadividually with a Kiuskall-Wal}isone-way ANOVA on ranks Sokal and Rohlf 1981! after log x+1! and squareroot traasformaaonsfailed to pass Kolmogorov-Lilhefors normality test. Differences in the size distribution of individuals among habitats and within sites were tested with Chi-square contingencytables. Rsh speciesdiversity was calculated from the Shannon- WeaverDiversity Index H'! Shannonand Weaver,1949! and the evennessindex J'! was calculated as welf for each census Pielou 1978!. Number of fish and speciesabuadance caught on fish traps and seinenets wereexainiaed by areawith a Kruskall-Wailisone-way ANOVAon ranks Sokal andRohlf 1981!after log x+1! aad squareroot tiunsformationsfailed to pass Kolmogorov-Liihefors normality test. Species were ranked in oilier of aksindance.
514 hlateo I. and W.J. Tobiaa GCFI:52 2001
RESULTS
Visual Strip Transect Census A totalof 4471 individualswere observed in visualcensuses representing 54 speciesand 21 Families in CottongardenBay. The most abundantfish was the shpperydick Halichoeresbivitattiis 8%! followedby the Bucktoothpaixotfish Saunaradhes 9%! Table1!, A totalof 7829fishes represeriting 66 species and 25 Familieswere observedin TeagueBay. Speciescoinposition were dominatedby Haemuhdrecruits Haerwulon happ, 8%! followedby the slippery dick Halichoeresbiviraous 3%! of the total fish abundance Table 1!. The speciescomposition at YeffowcliffBay consistedof a totalof 7388 individuals nqmesenting74 speciesand 27 Families,Once again grunt recruits Haemulidae! dominatedthe speciescomposition accoimtiug for 33% of the total fish abundancefollowed by the slipperydick Halichoeres bivirarrus 7%! Table 1!. Fwqpuss was the most abundant habitat found in all Cottongarden Bay ~ts accountingalmost 94% of the habitatsfound. Seagrass and algal plain weie the mostabundant habitats found in all TeagueBay trunsectsaccounting 84% and 11% of the habitat, respectively. Most of the habitat found was seagrass 0%! andalgal plain 6%! in YellowcliffBay Figure2!.
Flp ure 2. Proptwtional habitatdestrhv5on N = 72,000 iran!
Monthly variaticaisin meanfish abundaiiceand speciesrichness densities ainongsites showed different peaks throughout the year of studyin all three einbayments.A very distinctdecline in numberof individualsand species densitieswas observed from January to Match, 1999 Figure3a, 3b!. Speciesheteiogenity and evenness mean values followed the samemonthly trendsas for abundanceof fishesand numberof speciesdensities. Species diversityand evennesswere higherin October1998 and throughthe summer monthsand chxmasedduring Ihe monthsof Januarythough March 999!. There
515 tsroceeditt s of the 52itd Gulf and Caribbean Fisheries institute were no significantdifferences ~.05! in numberof fishes,number of species, diversity and evennessindexes among sites. Teague Bay had the highestnumber of fish density per site while Yellowcliff Bay had the highestnumber of species densitiesper site. Yellowcliff Bay had higher H' valueswhile J' valueswhere higher in CottongardenBay Figure 4a, 4b!. When numlxs of fish and number of speciesdensities per habitat were examined,patch ieef habitatshad the highestnumber of fish and speciesdensities exceptin CottangardenBay whererubble habitatshad the highestdensity Rgure 5a, Sb!. Kruskall-Waliis ANOvA demonstrated significant differences p 0.05! in abundanceand speciesdiversity amonghabitats. Patch reef and rubble habitats had sigmficantly higher mean numberof fish per transectthan algal plains and sand habitats. As for species richness, patch reef and rubble habitats had significantly higher meannuinber of speciesper IranscMthan all other habitats. Diversity indexes H'! were also significantly higher pc0,05! in patch reefs than the rest of the other habitats.Rubble and seagpeshabitats had diversity indexeshigher than sandand algal plain Figure 6a!. There were no significant diKaeem p=0.05! in evennessvalues among habitats Figure 6b!.
Table 1, Summary af visual strip traftsect resuts and the four most abundant speciesfound.
S Totat Cotlangarden Bay Helichoeres lvitratus 1,687 37.73 Sperisomerertierts 1,289 28.83 Thetassome bffesotatum 218 4.88 Scerus teeniopferus 138 3.09 Total Number of Fish 4,471 Total Number of Speciee 54 Teegue Bay Heemu!on spp. 2,983 38.10 Hatiohoeres bivittstus 1,058 13.51 Heemuton eu~m 921 11.76 Sperisomarediarts 496 6.34 Total Number of Fish 7.829 Total Number of Species 66 YelkwvcttffSay Heemuhw spp. 2,465 33.36 Hetiohoeres biwttatue 1,967 Sperisomaredierrs 443 6,00 Acenthuruschiriuryus 281 3.80 Tate! Number of Fish 7,388 Total Number of S 74
5I6 Mateo I. and W.J. Tobias GCFI:52 2001
100 90 o 80 70 60 pP50 p 40 S 30 C 20 Z 10 C 0 co Q W m 0> a> 0 4r0 Wk 4r2l 8 Z 0 z ~ ~ Z A'
Figure 3. A! Monthlyvartaten in meannumber of fish per site; B! Monthly variation oi mean number of speck+ per sNe Proceedin s of the 52nd Gulf and Caribbean Fisheries institute
Cg CO 07 g ITI CD CD CD CD CD CD Cl 0 h 0 C h a h a h 0 Z O ~ ll. CD Legend:QConOngarden Bay league Bay +Yellocccc-1itTBay
Figure 4. A! Monthlyvariations in ShannonWeaver Oiversityper site; B! Monthlyvariations in Pielou Evennessper site
518 Mateo I. end W.J. Tabtas GCFI:52 2004
180 E 160 > i.0
O 80 60 E 40 Z 20 0
Figure S. A! Mean numberof fish per habitat;8! Mean numberof speciesper habitat
519 Proceedin s of the 52nd Qulf and Caribbean Fisheries Institute
Legend; g Seagraaaei Patch Reef +Algal Plain ~ Sand p Rubble
13 0$ 0$ D.7 Db e 0$ 04 c pg 02 0.1 DO Co'Roegapten Teague B 3y YettovegffBay Bay
Figure 8. A! MeanDiversity Indices H'! psr habitat;8! INsanEvenness J'! per habitat Mateo I. arid W.J. Toblaa GCFI:52 2001
Analysis of habitat and site utilization by size groups demons-trated significantdifferences p 0.05! in sizedistribution proportions among habitats and sites. Seagrasshabitat had the highest proportions on all size groups among habitats followed by algal plain, except in the size group 5 - 10 cm where patch reef habitats had the highest proportion in this size group. All habitats contained recruits 5 cm and the vast majority of recruits were distributed on seqpa Table 2. Size Class Distribution r Habitat Habitat Area Habitat Species'. Fish Abundance, AbsoluteDensity m2! Cover Richness.:<5cm 5-lOom 0 cm Total l A B Seagrsss 57326 79.62 81 8492 3907 1334 13733! 0.24 0.001 Algal Plain 9052 12.57 58 1662 409 1019 3090'.,0.34 0.006 Patch reef 2836 3.94 1019 832 349 2200! 0,77 0.020 Sand 2462 3.42 12 207 47 25 279i 0.11 0,005 Rubble 324 0.45 31 324 44 18 386! 1.19 0,090 Total 72000 :. 11704 5239 2745 19&88l A= Fish Abundance Absolute Density Num. Fish/m ! B=Species Richnese Absolute Density Num. Species/mQ Fish Trap Results A totalof 696individuals were caught by trapsrepresentiag I6 speciesand 12 Families in CottongardenBay Table 3!. The inost abundantfish was the spotted goatfish Pseadupeneusrnacufarus 8%! followed by the bucktooth partotfish Sparisamaradians 8%! at CottongardenBay. There was a total of 760 fishes representing26 speciesand 14 Families around TeagueBay, The speciescomposition around Teague Bay was dominated by the french grunt Haenmbn jhrvoSteaium 1%! followed by the spottedgoatfish Pseudupeneas mandatus 9%!. A total of 438 individuals representing19 speciesand I I Famihes was found at Yellowcliff Bay. The spotted goatfish Pseadlpeneus mand'aiusdominated the speciescomposition 7%! followed by the yeliow4ul snapper Qcyurirs ciirysurus 9%! Table 3!. 521 Proceedtn a of the 52nd Gutf and Caribbean Flaherlea institute Table 3. Size Class Distribution r Site Ste Area Species. ,'Fish Abundance : 'AbsoluteDensity m' Richness'. Table4. Summaryof trapsresults and the four most abundant species found. N= 120 total tra s for each site, Total Fish Cottongarden Bay Psevdvpeneusmecufafus 269 38.65 Sperisorna radians 267 38.36 Hoiocenfrus edscensionis 54 7.76 ~urus ch+surus 32 4,60 Total Number of Fish 696 Totai Number of Species 16 Teague Bay Haemuion fievotineeturn 239 31.45 Pseudvpeneusmecviefvs 146 19.21 Ocyvrue chrysvrus 79 10,39 Heernulon auroiineaium 43 5.66 Total Number of Fish 760 Total Number of Species 26 Yellowcliff Bay Pseudupeneus mecuiatvs 209 47.72 Ocyuruschrysurus 40 9.13 Spansoma radians 30 6.65 Heemuion ffevoiinestum 24 5.48 Total Number of Rsh 438 Total Number of Species 19 Mateo I. arid W.J. Tohltte GCFI:52 2001 Monthly mean variations in fish abundaacecaught per trap among sites mnged from 1.8 to 21.8 fish per trap Figure 7a!. Mcaa Catch Per Unit Effort CPUE! at CottoagardenBay was highestia February CPUE = 13.7 fishes/trap! and lowest in January CPUE = 1.9!. In Teague Bay CPUE was highest in April 1.8 fishes per trap! and lowest in March CPUE = 2.1!. Yellowcliff Bay monthly CPUE was highestin Septeraber CPUE 7.7 fishes/trap!and lowest on April CPUE=1.8!. In Cot!ongaidenBay, monthly meannumber of speciesper trap rangedfroin 1.1 to 2.3, while meanspecies diversity in TeagueBay rangedfrom 0.9 to 4.9 CottoagardenBay CPUE valuesranged from I. 1 1 9 numberof species!trap. Although TeagueBay had more numberof fishes and speciesrichness per trap, thee were no significant differences p > 0.05! in the number of fishes, and aumberof speciesper site Fiigure7b!. Beach Seiue Results A total of 61 individuals were caught by seine net representing 9 species aad 7 Families in CottongardeaBay gable 5!. The most abundantfish was the permit Trac/iinoarsfakcuus 6%! followed by the slendermojarra Kucinoslomas jonensis 9%! at Cottongardea Bay. There was a total of 263 fishes representing29 species and 19 Families around Teague Bay. The species compositionwas dominated by the horseyejacir Caranr laais 8%! followed by the slendermojaita, Eacinosromusjoncnsis 9%!. The speciescomposition at Yellowcliff Bay contributedto a total of 10 individuals representing4 species aad 4 Families. The slender mojarta Eueinosromusj oncnsis and the hound fish FylosorLr crocodrllris dominatedthe speciescomposition accountingfor 50%, and30%, respectively,of total fish abundance, Monthly mean variations in fish abundance caught per seine net haul among sites mnged from 0 to 29 fish per haul Figure 8a!. Mean CPUE at CottongardenBay was highestia May 0 fishes per haul! aad lowest during the months from December1998 to February 1999,aad froin August to September 1999. Ia Teague Bay CPUE was highest in May 9 fishes/haul! and lowest from Januaryto February1999. Yellowcliff Bay monthly CPUE remainedlow throughoutthe study. The highestCPUE was in Decemberwith 4 fishes/haul, Monthly incan number of species per haul ranged from 0 to 2.5 in CottongardeaBay while meanspecies diversity in TeagucBay rangedfrom 0.5 to 4.67. Cottongardea Bay CPUE values ranged from 0 to 2 speci~. There were significantdifferences pc0.05! in the numberof fishesaad number of speci~ among sites. Teague Bay had significantly more fishes aad species/haulthan any of the other sites Figure 8b! Proceedin s of the 52nd Guif and Caribbean Fisheries institute Legend:QCottongarden Bay league Bay +Yel!OwcliffBay Figure 7, A! Monthlymean variationin numberof fishescaught on trap per site;1! Monthlymean venation in numberof speciescaught on trapper site, Mateo I. and W.J. Toblaa GCFI.52 2001 Legend. QCottozgardenBay league Bay +Yellowcliff Bay Figure 8. A! Monthlymean amh5on ln numberof fishes caughton beach seine net; B! Monthlymean variationin numberof speciescaught on beach seine net Proceeds s af the 52nd Gulf and Caribbean Fieheriee Irtatitttte Table 5. Summary of beach seine net results and the four most abundant s 'es found N= 24 total net hauls/site . Total Fish Cottongarden Bay Trachinofus fa/cetus 16 26.23 Eucinostomusjones/a 12 19.67 Cerarur /atua 10 16.39 Trschinotus gooder' 7 1'l.48 Total Number of Fish 61 Total Number of Species 9 Teague Bay Caranr /atua 74 28.14 Eucinostomus j onesis 51 19,39 Ocyurus chrysurus 27 10,27 Sphoeroides spengieri 16 6.06 Total Number of Fish 263 Total Number of Species 29 YellowciN Bay Eucinostornusj onesis 50 Tylosorrrs croaxfn/us 51 3 30 Chaeforfon oaprstratus 10 Sparisoma radians 10 Total Number of Rsh 10 Total Number of S ecies 4 DISCUSSION The role of nearshorehabitats for juvenile fishes in the North~ coastof St Croix is demonstratedto be quantifiable in terms of juvenile fish diversity and abundanceby area, The data suggeststhat some of the function and attributesof thesecoastal systems such as their high productivity and structural coin-plexity are conducivefor a nursery function. The presenceof nearshore seagrassand patch reefs habitats is likely to influence the juvenile fish assemblagesin adjacent coastedareas by enhancing the numbers of micto- habitats,ptmriding refuge from predationand foragingarea. Many of the species utilizing ncarshorehabitats as nursery areaswere found throughoutthe study regionand in somecases in relativelyhigh densities.These species included Haerurdouspp., Haenudouauroir'nearurn,Haemulon Jfavo tineaturn, and Scarm taeruopterus. Mateo I. xxrxxI W,J. Toblas GCFI:52 2001 The juvenile fish assemblagesin tbe backxeef lagoonal areas of Couonyuden Bay, Teague Bay, and Yellowcliff Bay have a shared trait with those found in many other reef fish populationsin that these habitats support few speciesin large numbers Ogden and Zieman 1977,Brothers and McFaxland 1981!. Three species, Sparisonxa radians, Halichoeres biviNaixxs, Haenadon spp., accotmtedfor >65% of aU fishes observedin the visual census. The more abundaat speciesof the fish community in this study belonged to the lower ixopbic level species TaMe 1!. Some dominant species are ~cnally and commerciaUy baked to the local nearshoie fishery. The fish iababitiag the backxeefareas ia CottcegardxsxBay, TeagueBay aud Yellowchff Bay were of smaller sizes 10 cm! and very few reached over 10 cm. This data indicates that backieef areas apparently plays aa imporuuxt role as a nursery grounds for economicallyimportant species at somepoint during their life cycle. At the sites studied, fish species formed distiact communities within seagxas.x,sand, patch reefs, algal plains and rubble habitais. The seal ass aad algal plain babitalsat all ihxee embaymentswere dominatedby small xesident species HaIicIxoeres spp., Qarisoma radians, and juveniles of nonresident species that are of economic importancelike Haemrdon~. and Oc'tnxraschrysxxxxxs. Rubble aad patch reef habitats harbored more speciesper tmnsect and were mostly dominated by small juvenile damselfishes, parrotfishes, grunts, and docioxfishes. Seasonalityof speciesrichness and abunxhmceaf fishes was observedin most cases, including the number of individuals of some of the most abundant species. This seiLxonalitymay be due to severalphenomena which axeprobably not mutually exclusive. Previous studies done at Teague Bay state that during the wanner months from July to November there is a significant increasein algal cover and biomass which probably induces an incxeasein invertebrate populations Rogersand Salesky 1981!. This increasein food availability may favour some hexbivoies such as Spvxrisrxmaradians, Scarus croicensis, Scanxs raeniopiernsaad someinveriebraie omnivores HaIichoeres bivixiaixxs, Haenndon spp. and small carnivorousflsbes as Ocyxxrxxsc!xrysxxrus. Nevertheless, ismaexl variatioas may have anoNux complementary origin. Seveml authors have pxcduced strong evidence for the inAuence of settlement aad recruitment pnxeaies of larvae and juvenile fishes on seasonalfluctuation of both species richness and abundance af individuals of fish communities which showed higher values in summer Williams and Sale 1981, Dohexty and Williams 1982!. Dming this study smnmerpulses of recruitmentwere evident for some species. This fact affected mainly schooling speciessuch as Haexxudonspp,, although varying in magnitudebetween areas. The lower abuxs9anceof fishes particularly recruits in winter which may be a causeof the decxeasein nmnberof fhh individuals aad speciesdiversity in these 5Z7 Proceedlri a of the 52nd Golf attd Caribbean Fiaheriee lnetitute periodsmay be due to severalmechanisms. The most ptobable include natural mortality,predation and or migrationto otherreef areas. The natumImortality and the mortalitydue to predationprobably affect all recruitsdespite the existenceof preferentialprey/predator relationship Hixon 1991!. Although somejuveniles settle directly into adult habitatsuch as the pomacentrids, chaetodontidsand labrids Fowler 1990, Eckert 1984!. Mostjuvenile reef fishes occurin shallowerareas than their conspecific adults. Thus migrations of some specieswhen the adult or subsdultstage is reachedfrom shallow to deeperwaters mayprovide a partialexplanation for theseasonal variatian observed at leastfor fish speciessuch as gruntsand snappers. However, differences in timing and intensityof recruitmentof Haernllonspp, Halichoeres bivittanis, Sparisoma radiansand Ocyarus cIiryslrns between sites suggests different factors affecting larvalsupply such as spawning cycle, currents, etc. Loweroverall recruitment to CottongardenBay may be due to lower larval supplybut also shortage suitable habitat. Shortageof habitat is a possibility since habitat cover was diffeient amongsites. Yellowcliff Bay had more habitatcomplexity than CottongardenBay. Almost95% of thehabitat found iu CottongardenBay was seagiass,and occurrence of suitablehabitats for fish recruitssuch as patch reefs algal Pain and rubbleareas were rare. Rnally, it is demonstratedthat the backieeflagoon areas of Cottongarden Bay, TeagueBay and Yellowcliff Bay are iinportantnurseries for inany economicallyimportant fishes. While variability amongsites and months occuin+ trendswere similar and uMhcated the importanceof nearshorehabitats for juvenilefishes. Sincethe majority of juvenilesin nearshorehabitats are of economicallyimportant species and they appear to preferentiallyinhabit backreef lagoonalshallow areas, these habitats must be conservedto ensurethe continued viability of fisheriesresources in St. Croix. ACKNOWLEL'KxMEÃ7S We thank W, Ventura, H. Rivers, G. Santana, C. Cortez, and D. Durant for theirlogistical support provided during the field survey.This studywas funded by U.S.Rsh andWildlife ServiceFederal Aid Gmnt;F714 under the Dingell- Johnson Sport Fish Restoration Act. UTEPATURE CITED Appeldoorn,R.S. C.W. Recksieck,R L Hill, F,E Paganand G.D. Dennis, 1997. Marine protectedareas and reef fish movements;the role of habitat in controllingontogenetic moveinents, Proc ge Int. Coral Reef. Symp. 2: 1917-1922. Mateo I. and W.J. Toblaa GCFli52 2001 Austin, KM. 1971. A survey of the ichthyofauaa of the mangroves of western PuertoRico during December,1967 - August, 1968. Caribb. J. Sci. 11 -2!:27-39. Beumer, J.P. 1978. Feeding ecology of four fishes from a mangrove creek in north Queenslaad,Ausnalia. J. Fish. Biol. 1 2 !:475490. Boulon, R.H. 1986. Fisheries habitats of the Virgin Islands region of ecological importanceto the fisheries resourcesof the Virgin islarxh Biosphere Reserve. St. Thomas, USVI: island Resources Foundation, Brothers, E.R, and W.N. McFarland. 1981. Correlations between otolith microstructure,growth and life history transitions in newly recruited French Grunts Haernulon jlavolineaturn! Rapp. P.v. reun. Cons. Penn. Int. Explor. Mer. 178:369-374. Doherty, P3. and D.McB. Wi]liarns. 1988. The replenishmentof coral reef fish populations. Oceanogr. Mar. Biol. Annu. Rev. 26:487-551. Eckert, GL 1984. Anaual aad spatial variation in recruitment of Iabroid fishes amongsever reefs in the CapricorittBunkerGroups, Great Barrier Reef. Mar. Biol. 7$:123-127. Fowler, A.J, 1990. Spatial and temporalpatterns of distribution and abundance of chaetodontid fishes at One Tree Reef, southern GBR. Mar. EcoL Prog. Ser. 64:39-53. Hixon, MA. 1991. Predation as a process structuring coral reef fish communities. Pages475-508 ia: P.F. Sale ed.! Theecology of fishes on coral reefs. Academic Press, London. Macnae, W. 1974. Mangrove forest and fisheries. Indian Ocean Programme Publ, No. 34. Rome. Indian OceanFisheries Commission. 35 p. Ogden,J.C. and J.C. Zieman. 1977. Ecological aspectsof coral reef, seagrass bedsin the Caribbean, Pages377-389 ia: Roc. of Third International Cora/Reef Syrup. R.S.M.A.S. Univ, of Miami: 656 pp, Phillips, R.C. and E.G. Menez. 1988. Seagrassess. Smithsoraan Contributions to the Marine Sciences. 34. IG4 pp. Pielou, E.C. 1978. Pages289 315 ia: Population and communityecology. Principlesand Methods. Robblee, M.B. and J.C. Zieman. 1984. Diel variation in the fish fauna of a tropical seagrmsfeeding grouacL Bull Mar. Sci. 34 !:335-345. Rogers, C,S. and N.H. Salesky. 1981. Productivity of Acropora Prdrnata Lamatck!, MacroscopicAlgae. and Algal Turf from Taguebay Reef, St. Croix, US Virgin lslaads. L Exp, Mar. Biol. FcoL 49:179-187 Rooker, J.R and G.D. Dennis. 1991. Diel lunar aad seasceal changes in a mangrovefish assemblagesoff southwesternPuerto Rico. Bull. Mar. Sci. 49 !:684-698 Proceedln s of the 52nd Gulf and Caribbean Fisheries Institute Shannon, C,E. and W. Weaver. 1949. The Mathematical Theory of communication. Urbana IIL University of Illinois Press. Shulman, M J. 1984. Resource iiinitation and recruitment patterns in a coral reef fish assemblages.L Exp. Mar. Biol. EcoL 74:85-109. Shuiman, M J. 1985. Recruitment of coral reef fishes: effects of distribution of ~rs and shelter. EcoL 6 6:1056-1066. Soksl, R.R., and F.J. Rohlf. 1981, Biometry. W H Freeman and Company, New York. Pp. 959. Springer, V.G. and AJ. Mcerlean. 1962. Seasonalityof fishes on a south Florida shore. Bull. Mar. Sci. 1 2 !:39-60 Stoner, A.W. and J.M. Waite. 1990. Distribution and behavior of queen conch, Strorabm' gigas, relative to seagrassstanding cd. Fish. Bul!. $8:573-585. Wilhams, D.McB. and P.F. Sale. 1981. Spatial and temporal patterns of recruitmentof juvenile coral reef fishes to coral hahitatswithin One Tice Lagoon,Great Barrier Reef. Mar. Biol. 6 4:245-253. Zieman, J, C., S.A. Mack and A.L. Mills. 1982. The ecology of the seagiassessof south Florida: a community protile. U.S. Fish. Wildlife. Serv. Biol. Serv. Prog. FWS/OBS-&2-&5, 530 BniM It, But%'ill They Come? Preliminary Findingsof RefugeLimitation Bottienecking in Jnveniie Menippe aditta in the Mississippi Sound VIRGINIA SHERVEITE1, HARRIET PERRY2,PATRICIA BIESIOI'1, KIRSZEN LARSEhP, AND JAMES WARREN'. ~Departmentof Biological Sciences Universityof SouthernMississippi Hattiesbnrg, Mississippi 99406-5018USA 2GtrlfCoast Research Laboratory, Institute of Marine Sciences University of SonthernMssissippi OceanSprings, Mississippi 39566-7000 USA ABSTRACT Many marine organismsare restricted to habilats which provide essential refuge, Menageadina, the westernGulf stonecrab, depends on patchyhard substratain the otherwisesoft-bottomed Mississippi Sound for individualand populationsurvival. Menippeadina supports small, developingfisheries in Louisianaand Texas, and occurs as an incidentalcatch in theblue in' fisheryin Mississippiand Alabama Stonecrab zoeae, megalopae, and small juveniles 0 24 mm carapacewidth! are relativelyabundant in MississippiSound and the speciesdoes not apped to be recruitmentlimited. Largmjuveniles are less conunonand their nmnbersmay be relatedto quantityand quality of suitable habitat. There is strong evidencethat refuge limitation exerts control on both populationsize structureand density of stonecrabs, The establishmentof low propre reefs in the MississippiSound has provided an opportunityfor preliminaryinvestigations of refugelimitation in local stonecrab populations Preliminaryresults indicate that suitablehabitat is lacking for juvenile stone crabsand competitionfor availablehabitat may be acutebetween stone crabs and otherxantbids &ttypanopensdepressus, Panopens simpsoni! as well as the toadfish,Opsanus beta KEY WORDS.Low profileartificial reef, Menippe adina, shelter limitation INTRODUCTION The westernGulf stonecrab, k&~e adina, currently supportssmall, developing fisheries in Louisiana and Texas, and occurs as an incidental catch in theblue crab, Callinectes sapidls, fisheryin Mississippiand Alabama Guillory et aL 1995!. Akntppeadieu can completeits life cycle within Mississippi waters;larvae ate not exportedoffshore, but areretained in the estuary.Stuck andPerry 992! reportedthat zoeae,megalopae, and smalljuveniles 0 - 24 mm CW! wererelatively aterskant in MississippiSound and that the species did Proceedin e ot the 52nd Gulf end Caribbean Fisheriee Institute notappear to berecruitment limited. Larger juveniles were less abuiidant and theirnumbers may be related ta quantityand quality of suitablehabitat. Thus, survivorshipof juvenilecrabs may be the primedeterminant of year-class suength. Thereis evidencethat shelterlimitation exeru controlon both popuiatiansize structure and density in stonecrabs. Caddy and Stamatopoulos 990! usedfractal theary to examinethe relationshipbetween topographic complexity niche availability! and the capacity of naturidsubstrates to support individuals of differentsizes. They suggested that the availabilityof niches,not food, determinedboth individualand populationsize and influenced timing and size at onsetof migrationfor some organisms.The shelterbottleneck hypothesis proposes that refuge limitation exertscontrol on populationsize structure and density and that its effectsare size-specificand not consistentthroughout an organism's life history Caddyand Stamatopoulos1990!, Beck 99l! notedthat as individualsnear the critical size,growth slo~s and the populationis diminishedby increasedmortahty andIoremigiation from the area. Shelterbottlenecks have been identified as contributingto size-structurein several species of nephtopdand palinurid lobsters Punish and Polovina1994, Wahleaiid Steneck1991, Howard 1980!. Beck 995, 1997! examinedsize-specific shelter limitation in Afenippeand suggestedthat shelter bottlenecks may be importantin theregulation of stone crabpopulations in the northern Gulf af Mexico GOM! because this atea lacks the extensivebard bottom habitats favored by thesecrabs, In examiningshelter linutation at four sites in narthwcst Florida, he found three of the sites shelter liinitedand the size classes affected differed amaiig the areas. In additionto affectingsize structureof stonecrab populations, availability of suitablehabitat may exert control on stonecrab densities by providing"escape space".Jeffxies aud Lawtou 984! exanunedthe concept af "enemy-freespace" and identified it as an impartant component of inany species' ecologies. Availability of structurally complex habitat offering refuge from predationwas linked to blue crab productionby Heck and Caen 995!, Orth and van Montfrans 990!, Wilsonet al. 990!, Heckand Thaman 984!, andHeck and Wilson 987!. Heck and Caen 995! notedthat althoughblue crab recruitment in the GOM greatlyexceeded recruitment along the Atlanticcoast, numbers of surviving juveniles were similar. They attributed the geographicalsimilarities in juvenileabundance to higherpredawn rates in the GOM andnoted that the greaterdiversity of low latitudepredators may limit piedahou-freerefuges. The establishmentof low profile artiTicialreefs in the Mississippi Sound by the MississippiDepartment of MarineResources provided an opportunityto investigaterefuge limitation in juvenilestone crabs aud to examinethe rolesof predationand inter-specific competitian in conuollingpopulation levels. These reefsprovide habitat for an anay of marineinvertebmtes, including Menippe Shervette V. et al. GCFI:52 2001 adina,Panopeussimpsoni, and Buypanopeus depressus and known predators of thesexanthid crabs including toadfish, Opsunus beta Larsenet al. 1999,Perry et al. 1999!. MATERIALS AND METHODS The site location is shown in Figure I. The area is characterizedby smaH natural patch oysterreefs and a newly establishedlow profile artificial reef. Fitlure 1. Locatke of low profile aNficlal reef To testfor refugelimitation in local stonecrab populations, experimental nichesamplers were constructed. The samplers,composed of sectionsof PVC pipe imbeddedin a cementmatrix Figure 2!, were designedto measurestone craboa~qxmcy at size. Samplerswere located on a portionof theartificial reef site at a distanceof appmximately200 m from shore. Pipe diameterswere selectedto cover the size iange of stone crab juveniles collectediu the MississippiSound Stuck and Perry 1992!. Four sizes of pipe were used in constructing the samplers: 13 mm .5 in!, 26 mm ,0 in!, 39 mm .5 in!, and52 mm .0 in!. Sectionsof pipewere iinbedded at a 45 Figure 2. Shelter samphrs RESULTS Stone crabs, Menippeadinrr, consistently began to occupy the sheltersby late June Figure 3!, with frequencyof occurrenceranging from 3 - 30% in the four different shelter size classes. The two smaHest diameter shelter size classes 3 and 26 mm! were occupied most frequently throughout the study, but stone crabsalso occurredin oue or both of the two largestshelter size classes 9 and 52 mm! between 28 June and October 1. The xanthid crab, I'anoperrs simpsoni, occupied shelters from the beginning of the study Frgure4! but had the greatestfrequency of occrmencein the two Shervette V. et ei. GCFI:52 2005 smaHest shelter size classes during the first three sampling periods, when M. adinaoccupation was low or nonexistent Between2 Septemberand I October, P. simpsoni occurredonly sporadicallyin the two largestshelter size classes. The Xanthid Crab, Z~7PanopeuSNprerSuS, WaS preSent during eaCh sampIing period Figure 5!, but its percentoccurrence was lowest among the three crab species, usually below l0%. This crab occurred with greatest frequency in the smallest shelter size class, but was found in all four sizes of shelter. Toadiish, Opsanusbeta, dominant the 52 nun diametershelters and occurred occasionally in the 39 mm diameter shelters Figure 6!. Toadlish showed a maximum occupation rate of 70% on 28 June and a Iow of 7% in July and on 2 September.No toadfishoccurred on the last two samplingdays. 70 60 + 50 40 4X 20 10 0 ++4 ~ >~ >4 ~ Qq Qq D4 'F ~o 1990 Figure 3. Percent occupation by shelter diameter for M etgntr ~60 sg! 10 0 ~q ~q 1090 Figure 4. Percent occupation by shatter diameter for P. simp:~ Proceedin s of the 52nd Gulf and Caribbean Fisheries institute 70 00 D s 50 s ~40 S20 20 ]00 !E ~!+ ~.!+ p+ q~ ~0 o" u y + >' p 2999 Figure 5. Percent occupation by shelter diameter for E. depresses 70 60 C~50 0 S40 P sS30 20 0 Figure 6. Percent occupation by shelter diameter for O. beta In Ripne 7, carapacewidths of the three crab speciesare comparedin relation to their occunence in the different sizes of shelter. Crabs with carapex widths between 6 - 17 mm, telardless of species,occurred in the 13 mm diametershelters Figere 7!. In termsof absolutenumbers, more Menippeadina occupiedthis slnaBestdiameter shelter than either of the two other xanthids. However, small crabs ~ 20 mm CW! of each speciesalso occurredin eachof Shervette V. et et. GCFf:52 2001 the three larger sized shelters, sometimes in combination with toadfislL The size rangeof crabsin the 26 mm diametershelters was 5 29 inm for M. adrna,5- 26 mm for Panopeussimpsoni, and 5 - 2l mm for EurypanopeNsd'epre~ums. Only two P. siIrtpsoni with caral'eccwidths gteaterthan 30 mm were found in the 39 mm diameter shelteis and none in the 52 mm diameter shelters whereas SiXM. adieu wilh Cara~m widlhs greaterthan 30 mm were fOundin the 39 mm diameter shelters and six in the 52 mm diameter shelters. Shelteroccupation rates for crabswith caripem widths less than25 mm are given in Figure8. During the first two monthsof the study,only PaIropms simpsoni and Zurypmtopems depressus occurred in the shelters. From June onward,JHenqpe adit had a greaterpercent occupation rate than either of the other two xanthids. 60 50 340 0 30 0~ 2010 13 26 39 52 Shelter dian»ter rrsu! Figure 7. Size frequency distribution of crabs by shelter diatneter Proceedin s of the 52nd Gulf end Ceribbeen Fisheries Institute t2 ee1p 0 8 X 6 CCl 0 o0 2 Co p Month Figure 8. Monthlypercent occupation of sheitersby crabs under25 mm CW DISCUSSION Shelter, competition, and predation are important factors in shaping community struck. For shelter-dwellingmatine organisms,quantity and qualityof suitablehabitat may be moreimportant in regtdatingpopulation size than food availability. The role of refuge limitation in determining juvenile stone crab population levels has been addressedin several studies with size- specificshelter bottlenecks identified as beinginAuential in settingpopuIation abundance Gibbs 1994, Beck 1995, 1997!. Although shelter occupation data from the current study are limited to the initial seven months of sampler deployment,there was no evidenceof shelter limitation for the size classesof crabsobserved. Occupancyrale in sheltersdid not exceed30 % for any of the cmb speciescollected to date. With one exception,the two smallestdiameter sheltetswete the mostfrequently occupied throughout the study. Selectionof sheltersappeared to be telated to size frequencyof animals occurring on the reef. Size class distributions of crabs available to occupy sheltersare a function of overall speciessize and recruitment. During the time period coveted in this study, recruiunentoccurred for alI three xanthid species. Temporal periodicity of recruitment,however, was variable. Menippeadimr did not occur on samplersuntil June. Of the three species,Eurypmopeus depresses was found less f'requently overall. This is the smallest of the xanthids collected and it occurredprimarily in the smallestshelter. Perry et al. 999! found this crab to be highly abundantin the area of study and its low occupancyrate suggests that it is not habitat limited. After the amval of stone crabs in June, occupancy rates for the other species declined. suggesting that some form of Shervette V. et ai. GCFI:52 2001 competition may be occurriag even though the majority of available shelters wereunoccupieL Shelterswere also occupiedby the taadfish, Opsmwsbeta. Many of the 52 mm diameter shelters were occupied by two toadfish, one male and oae female, during the first two sampling periods but after that only males, apparently guardingtaadfish eggs, were found. Toadfishoccupaacy rates dropped in June aad remamedlow thereafter,possibly becausepeak reproductiveseason had p88scL Stomach content analysesof Opsanusbete associatedwith the samplers demonstratedthat toadftshpreyed on all threespecies of xaathids Predationmay be an important factor ia controlling populationlevels of thesecrabs. ACKNOWLEDGlvKNTS This work is a resultof researchsponsored in part by the National Oceanic aad Atmospheric Administration, U.S. Depntment of Commerce, the Mississippi-AlabamaSea Grant Consortium, aad the University of Southern Mississippi, Institute of Marine Sciences,Gulf Coast ResearchLaboratory. The U.S. Govermaeataad the Mississippi-AlabamaSea Grant Consortiumare authorized to produce and distribute reprints notwithstanding any copyright notation that may appearhereaa. The views expressedherein are thoseof the authorsand do not necessarilyreflect views of NOAA or any of its subagencies. We are indebt@i to the following individuals who helped with field sarnpliag:Robin McCall, Bradley Randali,Jude LeDoux, Eric Pedeison,Brett Blackbura. Chuck Flowers, Todd RandaII,Roanie Bond, Glenn Hendiix, Wes Devers, John Anderson, aad Monty Simmons. David Chiluiza, Lesber Salazar, Windsor Aguirte, Run Schofield, JasonLemas, Andres Soto also assistedwith variousproject activities. We thank Lisa Engel for her adnanistrativeexpertise. REFERENCES CITED Beck, MW. 1995. Size-specific shelter Iimitaaon in stone crabs: a test of the demographicbottleneck hypothesis. Ecology 7 6 !:968-980. Beck, M.W; 1997. A test of generality of the effects of shelter bottlenecksin four stonecrab populations. Ecology 7 $ 8!:2487-2503. Caddy,J.F. and C. Shunatopoulos.1990. Mapping growth and mortality rates of crevice-dwellingorgaaisnis onto a perforatedsurface: the relevance of 'cover' to the carryingcapadty of naturaland artificial habitats. Eruuu; Coast. $heff Sci. 3 1:87-106. Gibbs,B.R 1994. Experimentson shelteravailability and interaclioason juvenile stone crabs, hkaippe ad&ra Williams and Felder!. Master's Thesis, University of SouthAlabama, Mobile, Alabama. 62 pp. Guillory, V., KM. Perry, aad R.L Learn 1995. A profile of the westernGulf Proceedin a of the 52nd Gulf and Caribbean Fisheries Institute stonecrab, Menippe adina. Gulf StatesMarine FisheriesCommission, TechnicalReport 3. 60 pp. Heck, K.I, Jr. and LD. Coen. 1995. Predadon and the abundanceof juvenile blue crabs:a comparisonof selectedeast and Gulf coast USA! studies. Bull, Mar. Sci. 57 !:877483. Heck, K.L, Jr. and T.A. Thoman. 1984. The nurseryrole of seagrmsmeadows in the upperand lower reachesof the ChesapeakeBay. Es~ 7 !:70-92. Heck,K.L, Jr. andK.A. Wilson. 1987. Predationrates on decapodcriLstaceaits in latitudinallyserrated seagnmcoinmunities: a studyof spatialand temporalvariation using tethering techniques. J. Exp. Mar. Biol. Ecol. 1 07:87-100. Howard,A.E. 1980. Substratecontrols on the size compositionof lobster Homarusgammants! populations. J, Cons.Int, Explor.Mer 39: 130- 133. Jeffries,MJ. and J.H. Lawton. 1984. Enemy-freespam and the structureof ecological communities. Biol. J. Linn. Soc. 23:269-286. Larsen, K, H, Perry, J. Warren, and C. Trigg. 2001 Seasotial colonization of artificial low profile reefsin Mississippicoastal waters. vertebrates. Proc. Gulf Carib. Fish. Inst. 52:488--'{97. Orth, RJ. and J. van Montfrans. 1990. Utilization of marsh and seagrnss habitatsby earlystages of CaNnectessapidus: a latitudinalperspective. Bull. Mar. Sci. 46:126-144. Pamsh, F.A. and J.J. Polovina. 1994. Habitat thresholds and bottlenecks in productionof the spiny lobster Panulints marginatus!in the northwestern Hawaiian islands. Bull. Mar. Sci. 5 4: 151-163. Perry,H K. Larsen,J. Warren,and C, Trigg. 2001. Seasonalcolonization of artificiallow profilereefs in Mississippicoastal waters: invertebrates. Proc. Gulf Carib. Fish. Inst. 52:498-51L Stuck, K.C. and H.M. Perry. 1992. Life history characteristicsof Menippe adinain Mississippicoastal waters. Pages82-98 in: T.M. Bert ed.!, Proceedingsof a Symposiumon Stone Crab {'GenusMenippe! Biology and Fisheries. Fl. Mar. Res. Publ. 50. Wahle, R.A. and R.S. Steneck. 1991. Recruitment habitats and nursery groundsof the American lobster llotnarus ameri~tus: a demographic bottleneck? Mar. Ecol.Prog. Ser. 69:231-243. Wilson, K.A., K.W. Able, and K.L Heck, Jr. 1990. Predation rates on juvenile blue crabsin estuarinenursery habitats: evidence for the importanceof macroalgae Viva ~!. Mar. Ecol, Prog. Ser. 5 8:243-251. Dynamics of Low-profile, inshore Arti6cial Reefs in The Mississippi Sonnd JUDE J. ~UX, JAMES R, WARREN, WESLEY DEVERS, LISA ENGH and MICHAEL BUCHANAN ??ieUniversity of SouthernMississippi Institute of Marine Science Gulf CoastResearclt laboratory, P.O. 7000 Ocean Springs, Mississippi 39566-7000 USA ABSTRACT The popularity of low-profile reefsas fishing banksand the needto increase the potential for harvcstablefood from the ocean has prompted many coastal statesto begin artificial reef programs. The temtorial waters of Mississippi contaia severalopen Gulf artificial reef sites and inshore,artificial, low profile reefs within the Mississippi Sound. past studies have provided valuable information on the successof the olrshorereefs and their attractionand possible supportof recieahonally important speaes. Iittle documeateddata is available about inshore, low profile reefs relative to their productivity and abihty to attract fish. Mississippi has establishedover twenty inshore low pmfile reefs using oyster shell, crushed limestone, aad concrete rubble. The in~ number of artificial fishing reefs in Mississippi's coastal waters over the last five years has provided an immediate need in obtaining information on the association of fish populations with these reefs and their subsequent use by the fishing public. An assessmentaad monitoring program for four of thesereefs was implementedin 1998 usiag entanglementgear, a 4.88meter lined otter trawl, aad custom fish traps. Interoefa creel surveys were also initiated to obtain usagedata by the fishing public. Substratesamples from oyster aad liraestone were also taken to provide information oa seasonalityand colonization of benthic fauna on thesereefs. These data may provide information on the trophic relationshipsbetween attracted fiafishes and a particularsubstrate. The attiacdon of finfishes by the artificial reef varied sigmficantiy by the geographic location, substratecompositioa, various abiotic factors, and a variety of aathropogenic factor. Cynoscionarenarius, Sconiberomorusrnaculatus, and Micropogonias tauftdattts were the most numemus, in deneeing abundance, recreationally important speciesobserved from the sampledreefs. These specieswere iaore numeious on the reef site than to the adjacent off reef sites; ho~ever the differenceswere not statistically different becauseof the high variabihty in the catch. KEY WORDS: Estuarine,low profile attificial reef,recreational fishing Proceedin s of the 52rtd Gulf end CarlbbeetI Fiaherlee lrietltute INTRODUCrION Aruficial reefs are generallycc~ed in coastalwaters to enhancelish populationsand improvenear shorefisheries as the result of fish attraction and/or production of new biomass Fabi and Fiorentini 1994!. The primary goalsof artificial reefsin coastalhabitats have been to enhancethe productionof reef-associatedspecies i.e., macroaigae,invertebrates, and fishes!and to increase the convenienceor efficiencyof harvestingreef-associated species Seamanet al. 1989, Seamanand Sprague1991, Pratt 1994!. Inshore artificial reefsprovide additional advaniagesfor open gulf fishermenby providing proximate, economicalaccess to fishingIocations during times when inclimateweather discouragesoffshore travel. The territorialwaters of Mississippicontain several openGulf artificialreef sitesand approximatelytwenty inshore, artificial, low profilereefs within the Mississippi SounL Stroud 965! hasprovided valuable informationon thesuccess of theoffshore reefs and their attraction and possible supportof recreahonallyimportant species. Little documented data is available about inshore, low profile reefs relative to their productivity and ability to amact fish. Mississippi,over the past twenty years, has establishedand maintainedseveral inshore low profile reefs using clam shell, oystershell, crushedlimestone, and concreterubble. With the itszmiseover the pastfive years in new reefsbeing developed in Mississippi's~ waters an iinmediateneed is warrantedin obtaininginformation on the associationof fish populationswith these reefs and their subsequent use by the fishing public. The main objectiveof the study was to evaluatethe speciescomposition and the potential attiaction of the ~anally important speciesby the Iow-profile artificial reefs placedin the Mississippi Sound. These speciesinclude, but are not limited to: Red drum Sciaenops ocenahcs!, Spotted Seatrout Cynoscion nebllosus!, Silver Trout Cynoscion nothus!, White Seatrout Cynoscion amrarias!, Atlantic Croaker Mcropogonias umtsdarls!, Southern Kingfish h4enticirrhus americurucs!, Southern Flounder Paraiichlhys leihostignia!, Spanish Mackerel Scornberomonrsmac~!, and Cobia Rachycenuon amarfian!. MATERIALS AND METHODS An assessmentand monitoring program for four of twenty-two available reefs were implemented in March 1998. The four reefs chosen were geographicallywide spreadalong the entire MississippiCoast Figure 1!. Two reefs were composedof crushedlime stoneand two were composedof concrete rubble. Each sample reef was located in highly accessible areas well known and uulized by the fishing public. The sample sites selected were: Hancock County Reef, located just west of the mouth of St. Louis Bay N 30'17.298, W 8%20.831!; Long Beach Reef, located just east of the Long Beach Harbor N 30 20.852, W 89 08,016!; East Deer Island Reef; located just east of Deer Ledoux J.J. et al. GCFI:S2 2001 Island N30'21.170, W 88 48310!;and Round Island Reef, located just westof Round Island N 30'17.970, W 8F35.884!. Figure 1. Low-PraNle,Inshore, ArtNcN Reef Sample Sites in the lNIsissippi Sound Proceedln e of the 52rtd Gulf ertd Caribbean Fleheriee Ittetltute Eachsite was sampledonce a monthusing a 228.6 metermulti-mesh size gill net, a 4.88meter Iined otter trawl, and up to five customfish traps.The gill netwas deployed across the reef and was allowed to fishfor onehour. While the gill netwas in thewater, the five trapswere randomly placed over the reef, and a tenminute trawl sample was taken on thereef. Uponretrieval of theon-reef gill net,a secondgill netset was deployed at anadjacent "control" site approximately 0.8 kilometersfroin the reef. A ten minute trawl samplewas obtainedat approximatelythe same distance froin the reef site,The fish trapswere retrieved afterthe end of thenet sampling on thereef and control site approximatelythree hours!. Total numberand lengthrange were recordedin the field for all non- recreauonalspecies. Total length,fork length, standardlength nun!, weight g!, and sex was recordedfor recreationalspecies after returning to the Gulf Coast R~ Laboratory GCRL!. Bimonthly interceptcreel surveyswere crsiducted. The surveyperiod was divided into weekendsfholiday'sand weekdays.Fium thesetwo categories,dates were randomlyselected: one weekdayand one weekends'holidaywere chosen. Surveyquestions included time spentfishing, and speciesand size rangecaught At the end of eachinterview the captainof the vesselwas given a "trip card" to be filled out at the end cf the fishing day and mailed to GCRL This caid providedadditional information on the numbersand sizeof fish that werecaught, Data was analyzed using Wilcoxon Paiied Signed Rank Test. RESULTS AND DISCUSSION All finfish speciescaptured can be found in Table 1. Forty-seven 7! total gill net sets and tiawls were done on the four reef sites over the twelve-month samplingperiod, producingfoity-six 6! different speciesof finfish. Preliminary statistical analysis was done on five recieahonaI species. Speciesinciuded: C. arenarue, C. nebulosus,M aniericanns,M undnlatas,and S. inacalalus. The total numbercapnaed on the reef and off the reef were paired for each species. A paired sign rank test was used to determine if the number on the reef was significantly different from off the reef. A "p" value was used to determine whether or not to reject the null hypothesis that the means = 0 for an alpha value = 0.05. P Valuesranged froin 1.0 to 0.0119974. Of the five most abundant recreational species, C. nebalosNs, was the only fish that was statisticallymote abundant p = 0.0119%'4!on the reef than off the reef sites. C. arenariuswas the most abuta9antfish reportedby recreanona!fishers on the artificial reefs Rgure 2!. Creel catchesof recreationalfishers contained C. an.nanusthirty-four 4! percentof the time. M. arnericanas,C. nebulosus,and M. raididaruswere also presenttwenty-five 5!, Eighteen 8!, and 4 percentof the time, respectively. The attraction of finfishes to the artificial reef varied significantly by the Ledottx J.J. et al. GCFI:S2 2001 geographiclocation, substratecomposition, various abiotic factors,and a variety of autfuopogenicfactors. The number of thesespecies, generally speaking, was more numero'> on the reef site when compared to adjacent off reef sites. Crushed limestone expresses itself as a suitable artificial habitat. The reefs composedof this substratematerial have attracteda variety of both pelagic and bottom dwelling fishes. o e o e o e o e o R Co P4 Pi 8OUSJJAQQO!U8088J Figure 2. Speciesoccurrence ln creel catch in Mlsshalppicoastal waters Proceedins of the 52ndGulf and CaribbeanFieheriee inetitute 8Ãw~mwnnn Ol Fl t N l At ~ZkzZ5~<~zj40 oeaC» W cv»gnCV~ ~ »»»l4 O O'eeW»»-»»teO N 74 ~ ~ ~ ~ IO C16 l1 V 9RQ~~~ »tern 4! CVW ~~ A P4 XmA~~~»»»m&mne4&'4 C W C4 I » N, Im I 0 ~ 4@4 R 44 4 4u 4~ v4 4 OK3 4~~ ~ %a~ 5 o- 8 8 8 ~0 %vnPv 8 0 c4 8Pent@ t 0»» '»»' A~~~~»t»emery»»» 4»t %' tV 8 I- »»!;8:t.14~'» « ijittj,t; ip'k j k'-jNijl Ledottx J.J. et al. GCFI:52 2001 Round Island Reef Geolpaphic location has played an important ro/e in the abundanceand sustaiuabilityof thesereefs. The RoundIsland reef sitewas highly susceptible to variousabiotic factors including weather,wave action, and oceancurrents. It is the deepestof the four reefs,with an averagedepth of 2,43 meters. This reef is moreopen to the Gulf of Mexicothan tbe others. It basptoduced a larger variety of fishes,including R. camafunt,and variousshark species. This areahas been noted as a possible shark nursery ground, Sampling has shown that during the summermonths, a large abundanceof sharkscau be found on or in very close proximity to this reef area. Since this site seemsto be heavily utilizedby sportand commercial fishermen, it hasbeen noted that during the monthsof shrimpseason, fjsb abundancedecreased. Total captureof finflsh frotn sampling during this high use period reflects this decrease in abundance. East Deer Island Reef The EastDeer Island samplesite is anotherheavily usedflsbing areautilized by ~onal, commercial and charter boat interests. It bas been noted in samplingthat during timesof heavy boat traffic in the Biloxi Ship Channel,fish were sparse in numbers. Through sampling efforts, it has been noted that excessivesedimentation due to HurricaneGeorges in September1998 bas bad a detrimental effect ou this atria. Tbe dredging of the ship channel has also impactedthe site by increasingforeign sedimentdeposition. The ship channel runsapproximately one-half of a kilometeraway from the site. Hancock County Reef and Long Beach Reef The rclnaining two reefs are protectedfrom some of the aforementioned factors. Thesetwo reefs,Hanc M County reef, and Long Beachreef are located in very close proximity to the beach or shoreline areas. Both are located approximately 300 meters from the shore. These reefs, combined, have supportedthe greatestnumber of ~onally important species. ACKNO'O'LEDGBvKNTS We are greatly indebtedto the Mssissippi Departmentof Marine Rescsucte for funding this project LITERATURE CITED Fabi, G., and L Fiorentini. 1994. Comparisonbetween an artificial reef and a control site in the adriatic sea. analysis of four years of monitorin. Bull. Mar. Sci. 55 -3!.538-558. Pratt,J.R 1994.Artificial habitatsand ecosystem restoration: mme'ng for the future. BulL Mar. Sci. 55:268-275 Proceedirx e of the 52nd Gulf arid Caribbenix Fiaheriee irxetittxte Seaman. W., and LM. Sprague. 1991. ArtijMal habitats for nxxarineand freshwaterjhherxes. Academic Press, San Diego, CA. Seaman,W., R.M.Bnckley,and J3. Polovina 1989. Advances in knowledge and priorities for research,technology and xnxxnagementrelated to artificial aquatichabi tats. Bull.Mar Sci. 44:527-532. Strond, R. H. 1965. Ax5Ticiai reefs as a tool of sport fishery maxiagementin coastal marine waters. Proceedings of Intexnational Game Fish Conference. Miami, Fla. November 12, 1965 Ef5caey of Two- and Three-Chamber Light-traps for Presett!etnent Fishes and Invertebrate Plankton frotn Mangrove and Cora! Reef Habitats at Key Largo, Florida DAVID G. UNDQUISTt, FRANK J. HERNANDEZ, Jr.2, ILEANA E CLA VIJOt and MAUP~N E 4%911'AKER> >DeparOnentof J3iological Sciences and Centerfor Marine ScienceResearch University of North Carolina ar Wilrnington Wilrnington, NC 28403 USA 2Departntentof Oceanographyand CoastalSciences LotdsianaState University Baton Rottge, LA 70803 USA ABSTRACT Light-tzapshave been«sed in ichthyo- aad iaverlebrateplankton studiesto samplethe larvalaud juvenile stages of fishesand planktonic invertebrates that are often inadequatelysampled with conventionalnets. Light-traps can be used in shallow areaswhere the «seof towed nets is diflicult, such as in maagroves aad aver coral reefs. Our purposewas to field test aud comple the efficacy of two- and three-chamberlight-ttaps in thesetwo habitats.Tha~hamber traps significantly out-performedtwo-chamber traps in both the maagtovesand over the reef, collecting7 aad 9.5 times the numberof fish larvae per hour, respectively.The thee-chamberuaps also sampleda greasediveisity and abundance of invertebrate plankton. KEY WORDS: Ichthyoplankton, invertebrate pl~ light attractioa, sainpliag gear INTRODUCFION Light-traps havebeen used recently ia ichthyoplanktonstudies to samplethe older. post-flexion pelagic stages and presettlementjuveniles that are not sampledadequately with conventionalnet tows Brogan 1994,Choat et al. 1993, Hernandezand Lindquist 1999!. The traps have demonstratedcertain biases, including beingsi2e aad taxoa selective Doherty 1987!,bin they do offer several advantagesas a samplingdevice. They caa be usedin shallow areaswhere the useof towed netscaa be difficult Broom 1994!. Larvae are collectedlive aad canbe placedm aquariafor furtherstudies and identiTicatioapurposes Doherty 1987!or releasedby diversand ~ for behavioralanalyses Leis et aL 1998, 1999!. Severaldesigns, aH modeledafter Doherty's 987! desiga,have been usedexteasivdy around differen reef systems,particularly the GreatBarner Reef, and have denionstratedthe ability to sample individuals from many different families. However, light-traps have not specifically been used to assessthe Proceedin a of the 52nd Gulf and Cartbbearr Flaherlaa Institute relativeabundance of all capturedmarine invertebrate plankton see Thormid 1992 and Moltschaniwskyland Doherty 1994, 1995, and Kraerner1996 for juvenile squidcatches!. In May, 1994, we werc able to collectzooplankton usingtwo differentlight-trap designs in Key Largo,Florida in a mangrove lagoonand above a coralreef. Our purposewas to field testand compareboth designsin termsof theirefficiency as presettlementlarval fish and invertebrate plankton samplers. METHODS and MATCRIAID Sampleswere collected from a shallow ~ 2 m deep!mangrove lagoon channel located near an inlet on the southwestend of Largo Sound 5'6.5'N, 80'24,2'W! on May 9 - 11, 1994,and from ThreeSisters, a reef < 7 m deep! locatedapproximately 7.5 km from shore 5'1.6'N, 80'23.7'WJ, on May 12, 1994 Figure1!. Figure 1. Charl of studyarea showingthe two studysites. Depthcontours shown are 6 and 16 rn. 1 irtd uist D.G. et al. GCFI:52 2000 Two different light-trap designs were utilized to collect fish larvae and presettlementjuveniles. Onedesign, constructed by ProgressiveTechnologies, Inc. FTI! in Plantation,Florida, and modeledafter Doheity 987!, had three chambersand a timer, which alternatelyturned the bulbs in the two upper chamberson and off, drawingorganisms into the trap. The other, simpler design,constructed at UNCW's Centerfor Marine Scient Research CMSR!, and niodeledafter Brogan 994!, had only two chamberswith tbe two 8 watt florescentbulbs in eachchamber remained on continuously. Traps werc deployedon site beginningat sunsetusing a 7 m outboani {R/V Tomtate! feign UNCW'sNational Undersea ~ch Centerfacihties in Key Largo,Florida. Eachnight in tbe lagoon,three sets of sampleswere completed using 3 three- chambermodels n = 27! and 1 tw~bamber model {n = 9!. Above the reef, three sets wereconipleted using 2 three-chamberlight traps n = 6! and I two- chamberlight trap n = 3!. In the mangrove,traps were moored to tbe bottom with anchors.Above the reef,the trapswere linked togetherat 50 m intervals with line andattached to a mooring.Buoys were attached to eachtrap above the reef,allowirig then to remainin the watercolumn near the surface.Traps were allowed to fish for approximately1 hour before being collected. The duration thateach trap iisbed was recorded, Upon retrieval, the contents of eachtrap were rinsed with seawaterinto a sieve. Organismswere then rinsed into a collection jar andfixed with 95%ethanol. Samples were sorted in thelaboratory for larval andjuvenile fishes and zoopl~ Fish larvae were countedand measuredto thenearest 0.5 mm. Fishidentifications were made to thelowest possible taxon using many resources,primarily Fahay 983!, Leis and Rennis 983!, and referencematerials provided by Drs. A.B. Powell and R.E Robbins at a larval fish workshopconducted at the NationalMarine Fisheries Service laboratory in Beaufort, North Carolina. Invertebrateplankton were identified to the lowest possibletaxon and enumerated, except in eightcases where high volumeswere subsampled,using a. volumetric technique to extrapolatetotal numbersin the sample. RESULTS Tbearty thiee-chambermodel significantly out-performed the Brogan two- chambermodel in boththe mangrove and on thereef, collecting 7 and9.5 times the numberof fish larvae/hr,respectively Table 1!. A total of 11 Doberty samplesand seven Brogan samplescontained no fish larvae. A total of 60 individualsrepresenting at leastfive familieswas sampledusing tbe Doberty modelin the mangrove,while tbe Broganmodel sampled only four individuals reiiresentingtwo families.Above the reef,the Dohertyinodel collected 29 fish from at least nine families, and the Brogandesign collected two individuals from two families. Ovemll. individuals from at least 12 families were collected,and 551 Proceedln s of the 52nd Gulf and Caribbean Fisheries Institute the coinbined capturerate was 1.9 fisb/hr. Clupeids primarily Je rkinskr spp.! were the most abundantlarvae collected by the Doherty trap, dominating both the inangrove 2%! and reef 5%! sainples TaMe2!. Table 1. Number Of fiah larvae and fiSh Nveniles Sampled per hour in ihe ma rove lagoon and above the reef. Location Trap Type Number of Fish Sampled per Hour 0oherty Brogan 0.3 Doherty Brogan 0.6 Table 2. Total number, size range, and percent total of fish larvae and fish juvenites sampled with light-traps in the mangrove and above the reef. =no data. MANGROVE REEF DOHERTY BROGAN DOHERTY BROGAN ane y Atherinidae 7 19 - 49.5 12 1 5.0 25 p 0 Gupeidae 44 14.5-57 72 3 16-17 75 16 4.5-18 55 Cyprinodontidae 1 7.0 2 0 0 Q 0 0 Engraulidae 4 24-38 7 0 ' 0 Q 0Qtk pp Gobiidae 0 0 0 0 4.5 Haemuiidae 0 0 0 92I 0 2 5-55 5.0 50 Labrisomidae 0 0 0 0 1 7.5 Q Ophidiidae p ~ 0 0 0 1 5.D 03D 0730 0 Pomacentridae p ~ p 0 0 1 8.0 p Synodontidae p w p P ~ Q 2 19-2D 1 350 50 Tetraodontidae D 0 0 W Q 3.5 0 p Unidentified 4 25-3 7 0 *P 0 4 5-7.5 14 0 0 The Doherty traps sampled individuals fram a larger size range .5 mm- 57.0 mm! than the Brogan traps .0 mm-35.0 mm!, although the number of individuals sampledwith the latter model was too low for comparisotL The majority of individuals sampledwith both trap designson the reef and in the mangrovelagoon were post-flexion larvae and juveniles Table 3!. t.lnd tiist D.G. et el. GCFI:52 200f Table 3. Summary of the different stages of fish larvae and juveniles samphd with li ht-tr Site/Trap % Preflexlon L Flexion 5 Postflexlon/ juveniles llangrove Doherty Brogan Reef Doherty 17 17 Brogan Both trap types sampled diverse invertebrate plankton, including holoplankters such as copeleds, chaetognaths,and polychaetesas well as meroplanliWtssuch as stomatopads,zaea, and megalopae.The total numberof individuals sampled varied far each trap type, but the three-chamber nap caught an average of four to seventimes mote inve~ plankttm than the two-chambertrap. Differences in total catcheswere observed at the two habitats. The mangrovelays had a mean of 768 individuals per twocsmber trap and 5,842 per threewhamber tmp while these numbersfor the cond tref habitat were 188 and 809, respectively. The composition of invertebrate plankton also differed at the two habitats Rgure 3!. Harpacticaid capepods comprised 68 percent of the mean total at the mangrove lagoon, but this is biased by a single three-chambertrap catch of 37,404 individuals. Cumaceans were the second most abua~t taxon at the mangrove lagoon with nearly 19 percent of the mean totaL Amphipods and calanaid copepods were the abundant taxa caught at the coral reef with 22 and 18 percentof the meantotals, ~vety, for the two- and three-chambertraps. DISCUSSION The Brogantw~ber madeldid not sampleas well as the Doherty three- chambermodel in terms of numberof individual fish sampled,number of fish taxa sampled,and ~range of fish sampled. However,only one Brogantrap was availablefor use in the study, and did not get field testedas extensivelyas the Eioherty model. Both uaps appear to collect the "target" size-range they are designed to sample, the past-flexion larvae and juvenile stages. With the exceptionof family Cyprinodontidae,all families representedin our studieshave beensampled in otherlight trap studies.The capturerate for the Dohertytrap above the reef is slightly higher than that reported in Doherty's original stuily 987!, but lower than subsequentstudies, which range from 11.8 - 313.5 Proceedin a of the 52nd Gulf and Caribbean Fieheriea Institute larvae/hr. It must be noted, however, that our site was a small patch reef and not compuable to the Great Barrier Reef sites in terms of larval abundanceand diversity. 50 al 50 10 /g~XY~ / ~//// Figure 8. Comparisonoi major invertebrate moplankton taxa at the Nvo habNniscaptured in bath two- and threex4an4er light-traps. Charresults indicate that the three-chambertrap is muchmore efficient in the capnue of invertebt3replankton than the two-chambertrap. The tuner light systemof the ~hamber trap helpsattract and retaina greaternumber of individuals comparedto the continuouslight syslem of the tw~hamber trap. Moreover,the th~amber uap has twice the numberof entry ventscompared to the two-chambertrap six vs. three! and thus allows a greater number of individualsto enterthe uppercoHecting chamber. The main disadvantagesof the ttuee-chambertrap ate cost $3,000in 1994! as weU as greaterbulk and weight. Both light-trapdesigns sampled the diversityof invertebrateplankters almost equaIly. Only cyclopoid copepodswere entirely absentin the tw~hamber trap but they were rarein the duce chamberedtrap, as well, Differencesin the compositionand abundance of invertebraleplanktem at the two sites were observedin our study. These different' may be duc to distributional patternsin planktonic taxa that may be conuolled by prevailing current patterns at the two sites. Differences may also be due to biological factor since some organismsmigrate at night from the soft substratesof the mangrovesediments to the water column e.g., ~ticoid copepods!and are thus differeutially available depending upon temporal and spatial activity Lied tiiet D.G. et al. GCFl:52 2001 ACKNO%~>GElvtENTS We gratefully acknowledge NOAA's National Undersea Ri~arch Center at UNCW for providing facilities and logistic support for the field work at Key Largo. Funding was provided by NOAA's CooperativeInstitute for Fisher Oceanography and a UNCW Faculty Development grant. D. Aten at CMSR was instrumental in the designing and manufacturing of the two-chamber hght- trap. M. Boris, F. Chapman,A. Henrik', F, Leyva, B. Spudich, and D. Swearingen helped with the field collections. Dr. A. McCrary kindly helped with the invertebrate zooplankttm identifications. This paper is contribution No. 125/6, Center for Marine ScienceResearch, University of North Carolina at Wi1mington. LJTERKTURE ClTED Biogan, M.W. 1994. Two inethods of sampliug fish larvae over reefs: a comparisonfrom the Gulf of California. Mar. Biol. 11 8:33~ Choat, J.H., Doherty, P.J., Kerrigan, B., Leis, J.M. 19%. A comparison of towed nets, purse seine, and light-aggregationdevices for sampling larvae and pelagic juveniles of coral reef fishes. Fish, BulL U.S. 9 I:195-209 Doherty, PL 1987. Light-traps:selective but useful devicesfor quantifying the distributions and abundsnces of larval fishes. Bull. Mar. Sci. 41:423- 431 Fahay,M.P. 1983, Guide to the early stagesof marine fishes occurring in the WesternNorth Atlantic Ocean,Cape Hatterasto the SouthernScotian Shelf. J. Northw. AtL Fish. Sci. 4: 423 pp. Hernandez,FL, Jr. aud D,G, Lindquist 1999. A comparisonof two light-tiap designsfor samplinglarval and presettlementjuvenile fish abovea reef in southernOnslow Bay, North Carolina.Bull. Mar. Sci. 64:173-184. Kraemer,JL 1996.Size frequencyand light trap catchefliciencies of the squid, Loligo sp., dming the hatchling and juvenile stages in the coastai waters of Onslow Bay, North Carolina. Honors thesis, University of North Carolina at Wilmington, 30 pp. Leis, J.M., Rennis, D.S. 1983. The Larvae of Indo-Pacific Reef Fishes. University of Hawaii Press, Honolulu. 371 pp. Leis, J.M and B.M. Carson-Ewart 1998. Complex behaviourby coral-reeffish larvae in open-waterand near-reefpelagic enviiorunents &iv. BioL Fish. 53:259-266. Leis, J.M. and 8 M Carson-Ewart 1999. In situ swimming and settlement behavior of larvae of an Indo-Pacific coral-reef fish, the coral trout Plecrropornas leolnrrdrrs Pisces:Settanidae!. Mar. BioL 1 34:51-64. 555 Proceedln a of the 52nd Gijlf and Carlbbeari Flaherlea Inst!tute Moltschaniwskyj,N.A. and P.J.Doherty. 1994. Distribution and abundanceof two juvenile tropical Phofololigo species Cephalopoda:Laliginidae! in the central Great Bamer Reef Lagoon. Fish. Bull. U.S. 9 2:302-912, Moltschaniwskyj, NA. and PJ. Doherty. 1995. Cross-shelf distribution patternsof tropical juvenile cepaRopods sampled with light-traps, her. Fresliwursr Res. 46:707-714. Thorrold, S.R. 1992. Evaluating the performance of light-traps for samphng small fish and squid in open waters of the central Great Bamer Reef lagoon.Mar. EcoL Prog. Ser.8 9:277-285