HAWAU-S-98-002 C2

COMMUNITY STRUCTURE OF FISH AND MACROBENTHOS AT SELECTED SITES FRONTING SAND ISLAND, O'AHU, HAWAPI, IN RELATION TO THE SAND ISLAND OCEAN OUTFALL, YEAR 8 1997

Richard E. Brock

PROJECT REPORT PR-98-07

January l998

WATER RESOURCES RESEARCH CENTER UNIVERSITY OF HAWAI'I AT MANOA Honolulu, Hawai'i 96822 AUTHOR: Dr. Richard E. Brock AssociateResearcher and Fisheries Specialist Sea Grant Extension Service MarineScience Building 204 Universityof Hawai'i at Manoa Honolulu, Hawai'i 96822 Tel.: 808/956-2859 FAX: 808/956-2858

$5.00/copy

Please make remittancein U.S. dollars from a U,S, bank or internationalmoney order to: ResearchCorporation of the University of Hawaii

Mail to: Water Resources Research Center University of Hawai'i at Mixe 2540 Dole St., Holmes Hall 283 Honolulu, Hawai'i 96822 ~ U.S.A. Attn: Publications Office

NOTE: Pleaseindicate PR-98-07 on check or money order for our reference. COMMUNITY STRUCTURE OF FISH AND MACROSENTHOS AT SELECTED SITES FRONTING SAND ISLAND, O'AHU, HAWAI'I, IN RELATION TO THE SAND ISLAND OCEAN OUTFALL, YEAR 8 1997

Richard E. Brock

Project Report PR-98-07

January 1998

PREPARED FOR City and County of Honolulu Departmentof WastewaterManagement Project Report for "A Five-Year Biological and SedimentMonitoring Program on the Marine CommunitiesNear the City's OceanSewer Outfalls" Contract No.: C54997 Project Period: 1 January1997 30 September2002 Principal Investigator: JamesE,T. Moncur

WATER RKSGURCKS RKSKARCH CKNTKR University of Hawai'i at Manoa Honolulu, Hawaii 96822 Any opinions,findings, and conclusionsor recommendationsexpressed in this publicationare those of the authorand do notnecessarily reflect the view of theWater Resources Research Center. ABSTRACT

This reportprovides the resultsof theeighth year of an annualquantitative monitoring carried out in August and September1997! of shallowmarine communities inshore of the SandIsland Ocean Outfall, 0'ahu, Hawai'i. This monitoringeffort focuseson benthicand fish community structure and is designed to detect changesin these communities. Marine communitiesoffshore of Honolulu have receivedconsiderable perturbation over the last 100 years, Dumping of raw sewagein shallow water, which occurred from 1955 to 1977, was halted in 1978;however, point and nonpoint sourcesof pollution from both urban activities andindustry continue. All of thesedisturbances may serve to obscureany impacts that may be causedby treatedeffluent dischargedfrom the deep-oceanoutfall. The marine communities show a considerablerange in developmentthat is probably related to historical impacts. Stationshave been located to take advantageof thesegradients. Analysis of the eight yearsof data showedthat there hasbeen no statistically significantchange in the following biological measures:percent coral cover, numberof coral ,number of invertebratespecies, total numberof invertebratescounted, number of fish species,total number of fishescounted, and the biomassof fishes presentat each station. HurricaneIniki, which occurred in September 1992, impacted marine communities along the south shore of 0'ahu. Coral communities receivedconsiderable damage, especially at the westernmoststudy station.Recovery in these communities is evident from the five years of data collected since the storm. Thus far, this study has not detecteda quantifiablenegative impact from the operationof the Sand Island Ocean Outfall.

CONTENTS

INTRODUCTION Purpose Strategy

MATERIALS AND METHODS . RESULTS 6 7 Station A Kewalo Landfill .

Station B Kalihi Entrance Channel .. 14

Station C Reef Runway 20

Physical Measurementsand Biological Parameters. 24

Green Sea Turtle Observations . 27

DISCUSSION . 27

REFERENCES CITED . 33

APPENDIX 37

Figures

1, General locations of the three permanentstations sampled in this study .... 2. Sketch depicting locations of the five photographicquadrats each on Transects T-1 and T-2, located offshore of the Kewalo Landfill adjacent to an abandoned sewer line...... 3. Sketch depicting locations of the five photographic quadrats each on Transects T-3 and T-4, located adjacent to the Kalihi Entrance Channel .. 4. Sketch depicting locations of the five photographic quadrats each on Transects T-5 and T-6, located offshore of the Honolulu International Airport Reef Runway . 10

Tables

I, Summary of biological observationsmade at TransectT-l, offshore of Kewalo Landfill on 14 August 1997. 11

2. Summary of results for the photographic quadrat survey for 1997. 12 3. Summary of biological observationsmade at TransectT-2, offshore of Kewalo Landfill on 14 August 1997. 15 4. Summary of physical measurementsmade at each of three locations in the vicinity of the transectpairs on 6 December1991, 22 December 1992, 9 September 1993, 14 October 1994, 26 September1995, 21 November 1996, and 15 August 1997 17 5. Summary of biological observations made at Transect T-3, east of the Kalihi Entrance Channel on 14 August 1997 18 6. Summary of biological observationsmade at TransectT-4, east of the Kalihi EntranceChannel on 14 August 1997 .. 7. Summary of biological observationsmade at TransectT-5, approximately760 m offshoreof the HonoluluInternational Airport Reef Runway on 18 August 1997 .. 8. Summary of biological observations made at Transect T-6, approximately840 m offshoreof the HonoluluInternational Airport Reef Runway on 18 August 1997, . 9. Summary of biological parametersmeasured at the six transectlocations in the eight annual surveys .. INTRODUCTION Purpose

In recentyears controversy has arisen regarding the impact that sewageeffluent from the SandIsland WastewaterTreatment Plant may haveon inshorecoral reef species,Much of the geographicalarea of concernin this studywas impacted by the releaseof 62 mgd m /s! of raw sewagein 10 m of wateroff SandIsland from 1955to 1977.Starting in 1978sewage received advanced primary treatment and was released farther offshore of Sand Island from a deepocean outfall 7 to 73 m depth!.Despite studies that demonstratedthe recoveryof inshorebenthic communities once the shallow sewage stress was removed e.g., Dollar 1979!, concerncontinues over the possible impact that the release of sewageeffluent from thedeep- oceanoutfall may be havingon the shallow <20 m depth!marine communities fronting Honolulu and SandIsland. Accordingly, beginningin 1990,this study was undertakenin an attemptto quantitativelyascertain the impactsthat may be occurring.This documentpresents the resultsof the eighthannual survey carried out in Augustand September1997.

Strategy Marine environmentalsurveys are usually performedto evaluatethe feasibility of and ecosystemresponse to specific proposedactivities. Appropriate survey methodologies reflect the natureof the proposedaction s!.An actionthat may havean acuteimpact such aschannel dredging!requires a surveydesigned to determinethe routeof leastharm and the projectedrate and degree of ecosystemrecovery. Impacts that are more chronic or progressiverequire different strategiesfor measurement.Management of chronic stressto a marine ecosystem requiresidentification of systemperturbations that exceedboundaries of natural fluctuations. Thus a thorough understandingof normal ecosystemvariability is required to separatethe impact signal from background"noise." Infrequentnatural events may addconsiderably to the variability or background noise measuredin a marine community. In September 1992 Hurricane Iniki struck the Hawaiian islands and impactedsome marine communitiesalong 0'ahu's south shore. This rare event has provided this study with information on the magnitudeof suchnatural impacts. Rare storm events not withstanding, the potential impacts occurring to the marine ecosystemoffshore of Sand Island and Honolulu are most probably those associatedwith chronic or progressive stresses.Because of the proximity of the population center and industry, marine communitiesfronting Honolulu are subjectedto a wide array of impactsnot usually occurring in other Hawai'i coral communities.Thus a samplingstrategy must attempt to separateimpacts due to wastewatertreatment plant effluent on coral reef communities located at somedistance shoreward from a hostof other perturbationsoccurring in the watersfronting Honolulu. Honolulu Harbor has been the primary commercialport for the Stateof Hawai'i since before the turn of the century Scott 1968!. The harbor is the result of dredging what was originally the drainagebasin of Nu'uanu Stream.Dredging beganbefore 1900,and periodic maintenancedredging still occurs. Until about 1960spoils were droppedjust outside of the harbor,generally to the eastof the SandIsland OceanOutfall. Besidesshipping, the harbor is ringed with industry; pineapplecanneries, gas andoil storage,and numerousother businesses have operatedor are still operating here. Storm drainageinto the harbor and nearby Ke'ehi Lagoon carries runoff from Honolulu's streetsand suburbsinto the ocean.Pollution is well known in the harbor;conditions are describedas early as 1920in referencescited by Cox and Gordon 970!. Sewage has been pumped into the ocean offshore of Kewalo and Sand Island sincethe 1930s.The early inputs were all raw sewagereleased in waternot exceeding20 m in depth.The actualpoint of releasevaried throughtime as different pipes were constructedand used. The multitude of perturbations that occurred in shallow water <20 m! until the constructionof the presentdeep-water outfall in 1978may serveto obscurethe impactsfrom the present discharge. The watersfronting SandIsland, into which the deep-oceanoutfall discharges,may be consideredin terms of gradients.There are numerous"gradients" owing to point-sourceand nonpoint-source such as storm drains and streams!inputs into Honolulu Harbor and the surroundingarea from the above-mentionedactivities. Because many of theseinputs have been occurring for a considerable period of time, the species composition and functional relationships of the benthic and fish communities at any given location in the waters offshore of Honolulu and the harbor are those that have evolved under the influence of theseongoing perturbations. As noted above,if impacts are occurring in the shallow marine communities fronting Honolulu owing to sewageeffluent dischargedfrom the deep-oceanoutfall, they are probably chronic in nature, thus causing a slow decline in the communities so affected. Gradients of "stress" or "impact" should be evident with distance from impact source s!. Thus, to quantitatively define these impacts, one should monitor these communities through time in areassuspected of being impactedas well as in similar communitiesat varying distancesaway from the suspectedsource s!.This rationalehas been used in developingthe samplingstrategy for this study. MATERIALS AND METHODS

Thequantitative sampling of macrofauna of marine communities presents a number of problems;many of theseare related to thescale on which one wishes to quantitatively enumerateorganism abundance. Marine communities inthe waters fronting Sand Island may bespatially defined in a rangeon the order of a fewhundred square centimeters such as the communityliving in aPoci Ilopora meandrina coral head! to many hectares such as areas which arecovered by major biotopes!. Because considerable interest focuses onvisually dominant corals,diurnally exposed macroinvertebrates, andfishes, we designed a sampling program to delineatechanges that may be occurring in communitiesat thisscale. Threestations were selected forthe monitoring ofbenthic and fish community response topossible sewage impacts. Their approximate locations are shown in Figure1, The stations areclose to some stations previously used by Dollar 979!. Thestations and the rationale for their selectionare given below: Station A- Utilized as a controlarea. This stationlies eastof the Kewalo Landfill! presentdeep-ocean outfall in 17.0 to 18,2 m of water. Prevailingcurrents create a westerlymovement of sewageeffluent Dollar 1979!,thus the shallow Kewalo Landfillarea is probablynot directly impacted. At this location,corals occur in areasof emergentlimestone. Localcoverage over short linear distances may exceed 30%.This station is in thevicinity of Dollar's979! station 2. StationB- Locatedabout 120 m eastof theKalihi Entrance Channel KalihiChannel! in approximately15.0 m ofwater. This station is about 900m westof thebypass old! outfall in anarea heavily impactedby the old 955 to 1977! shallow-water dischargeand is veryclose to Dollar's979! station14. Thereis emergentlimestone at this station,but coral coverage is low <1%!. Station C- Locatedin anarea of complexlimestone substratum in Reef Runway! waterranging from 7.5 to 12.0m in depthfronting HonoluluInternational Airport's Reef Runway. This stationis locatedclose to Brock's986! stationthat was monitoredquarterly in 1977-78 AECOS, Inc, 1979! andagain in 1986.It is closeto Dollar's979! station 19.This station was moderately impacted by theold shallow-watersewage outfall Dollar 1979!. At eachstation two transect lines were permanently established using metal stakes and plastic-coatedno.14 copper wire. The transects are 20 m inlength and have an orientation that is perpendiculartoshore. Two transects were established ateach location to provide some replication.Both sample approximately the same benthic communities. On each transect are fivepermanently marked locations m, 5 m,10 m, 15 m, and 20 m! for the taking of e e c D c co m AM m ce> ~ c C m z each eD e 0 <~~c E e g OcDcern Z0 0 m o~c9 m e e m m q!eOI ODdj C c o ecosoc EO m«m N eeeDm C Ve m~m>E e .0 Cb RD C 0 e m mCt5 y >c

m,eD e c y! e e C o W cn e e e 0 sEm iDc e E K» m» e ~CDD ee' e ~cl-c c Icing O Ee=ua c= e m m e « C C e I- omme .5? CL

0 e m e eD c c C 48e c m5Cl e mc, m Lo 5 eg» e «CO ~ C oC8 K o g «0 P C mcm LL COUJ l- photographsof the benthic communities.Cover estimateswerc also madein the field with a 1 mx 1 m quadratplaced at the -1 to 0 m,4 to 5 m,9 to 10 m, 14 to 15 m, and 19 to 20 m marks on the transect line in each survey, Fish abundanceand diversity are often relatedto small-scaletopographical relief over short linear distances.A long transectmay bisect a numberof topographicalfeatures e.g., coral mounds, sand flats, and algal beds!, thus sampling more than one community and obscuring distinctive featuresof individual communities.To alleviate this problem, a short transect0 m in length!, which hasproved to be adequatefor samplingmany Hawai'i benthic communities see Brock 1982; Brock and Norris 1989! is used. Information is collected at each transect location using methods including a visual assessmentof fishes,benthic quadrats for cover estimatesof sessileforms e.g., algae,corals, and colonial invertebrates!,and countsalong the transectline for diurnally exposedmotile macroinvertebrates.Fish censusesare conductedover a 20 m x 4 m corridor the permanent transect line!. All fishes within this areato the water's surfaceare counted. A single diver equippedwith scuba,slate, and pencil entersthe water, thencounts and notesall fishes in the prescribed area method modified from Brock 1954!. Besides counting the numbers of individuals of all fishes seen, the length of each is estimated for later use in the estimation of fish standingcrop by linear regressiontechniques Ricker 1975!. Species-specificregression coefficients have been developedover the last 30 years by the author and others at the University of Hawai'i, Naval UnderseaCenter see Evans 1974!, and the Hawai'i Division of Aquatic Resourcesusing weight andbody measurementsof capturedfishes; for many species the coefficients have beendeveloped using samplesizes in excessof a hundredindividuals, For the 1990 survey two weeks were allowed to elapse from the time of station selection and marking to the time of the first fish censusto reducethe bias causedby wary fishes.The same individual the author!performs all fish censusesto reducebias. Besidesfrightening wary fishes,other problemswith the visual censustechnique include the underestimationof cryptic speciessuch as morayeels family Muraenidae!and nocturnal species such as squirrelfishes family Holocentridae! and bigeyes or 'aweoweo family Priacanthidae!.This problem is compoundedin areasof high relief and coral coveragethat affords numerousshelter sites. Specieslists and abundanceestimates are more accuratefor areasof low relief, althoughsome fishes with cryptic habitsor protectivecoloration, such as scorpionfishesor nohu family Scorpaenidae!and flatfishes family Bothidae!, might still be missed.Another problem is the reducedeffectiveness of the visual censustechnique in turbid water. This is compoundedby the difficulty of countingfishes that move quickly or are very numerous.Additionally, bias relatedto the experienceof the censustaker should be considered in making comparisons between surveys. Despite these problems, the visual census technique is probablythe most accurate,nondestructive assessment method currently availablefor countingdiurnally active fishes Brock 1982!. A number of methodsare utilized to quantitativelyassess benthic communitiesat each station,including the taking of photographsat locationsmarked for repeatedsampling through time each covering 0.67 m~! and the placing of 1 m x 1 m quadratsat markedlocations for repeatedmeasurements. Both the photographsand quadratsare usedto estimatecoverage of coralsand other sessileforms. Photographs,which providea permanentrecord from which to estimatecoverage, were used in the sevenmost recentsurveys 991 through 1997!;the 1 mx 1 mquadrats were usedfor an in-the-fieldappraisal of coveragein all surveys.Cover estimates from photographsand quadratsare all recordedas percentcover. Diurnally exposedmotile macroinvertebratesgreater than 2 cm in somedimension are censused in the same4 m x 20 rn corridor used for the fish counts. Macrothalloid algae encounteredin the 1 m x 1 m quadrats or photographs were quantitatively recordedas percentcover. Emphasiswas placed on those speciesthat were visually dominant,and no attemptwas made to quantitativelyassess the multitudeof microalgal speciesthat constitutethe "algal turf ' socharacteristic of manycoral reef habitats. As requestedby permit agencies,divers madesimple physical measurements at the three stationswhile in the field. Measurementsof percentoxygen concentrationand temperature were made with a YSI Model 57 Oxygen meter, salinity was taken with a hand-held refractometer,and waterclarity wasdetermined using a 12-inchsecchi disk. Data were subjectedto simple nonparametricstatistical procedures provided in the SAS Institute statistical package SAS Institute Inc. 1985!.Nonparametric methods were used to avoid meeting requirementsof normal distribution and homogeneityof variancein the data. Data were analyzed using the Kruskal-Wallis one-way analysis of variance to discern statistically significant differencesamong ranked meansfor each transectsite and sample period; this procedure is outlined by Siegel 956! and Sokal and Rohlf 981!. The a posterioriStudent-Newman Keuls multiple-rangetest SAS InstituteInc. 1985!was also used to elucidate differences between locations. During fieldwork, an effort was made to note the presenceof any green seaturtles a threatenedspecies! within or nearthe study sites.

RESULTS

Field sampling was first undertakenon 27-29 December1990. Station locations were selected and marked in November 1990. The permanent pins were deployed about a week later.Figure 1 sho~sthe approximate locations of thethree stations, each with a pairof transects.Figures 2, 3, and4 aresketches showing the orientationof thepermanent photographicquadrats on each transect line. Data were collected from the same locations as follows: 1991data on 5 6 December1991, 1992 data on 21-22 December1992 and 25January 1993, 1993 data on 7 8 September1993, 1994 data on 20 September and 13 14 October1994, 1995 data on 28 August and 7 September1995, 1996 data on 26 and 29 August aswell as 20-21 November 1996, and 1997 data on 14,15, and 18 August as well as11 and 23 September 1997, Malfunctionof a newNikonos V cameracaused the loss of all photographicquadrat data forall stationsin thefirst 990! fieldeffort. Subsequently, theannual photography effort has beencarried out by membersof the OceanographicTeam, Department of Wastewater Management,City and County of Honolulu.However, the 1990 visually assessed square- meter-quadratsurvey provided information on benthic coverage. Subsequent surveys have usedboth photographic and quadrat methods to assessthe benthiccommunities. It shouldbe notedthat the numbering of photoquadrats has changed since the 1991survey, but the locations remain the same. The resultsare presented below by station.-

Station A Kewalo Landfill StationA islocated 600 m offshoreof the old Kewalo Landfill in waterranging from 17.0 to 18,2 m in depthon a substratumdominated by limestonewith moderatecoral communitydevelopment, Thetwo transects are 35 m apart,out of visualrange of oneanother seeFigure 2!. Waterclarity at thisstation usually ranges from 15to 20m. A summaryof thedata collected at TransectT-l on 14August 1997 is presentedin Tablel. In thequadrat survey, five coral species having a meanestimated coverage of 27.6% wereencountered; the dominant species were Pori tes lobata and Pocillopora meandrina. Four algalspecies Amansia glomerata, Botryocladia skottsbergii, Desmia hornemannii, and Porolithononkodes! having a meancoverage of 0.7%were noted in thequadrats. The macroinvertebratecensus noted the Christmas tree worm Spirobranchus giganteus corni culatus; therock oyster Spondylus tenebrosus; the nudibranch PhyIlida varicosa; and three species,including the long-spined black or wana diadema!, the black seaurchin Tripneustes gratilla, andthe green sea urchin Echinometra mathaei. The results of thefish census carried out at Transect T-1 are summarized in Table 1 andgiven in detailin the Appendix.Table 2 presentsthe results of thephotographic survey carried out on 11and 23 September1997. Mean coral coverage in thephotographic survey was estimated at 16.2%, with Poriteslobata being the dominantcoral. Old abarIdon Sewer Pipe

T-2 T-1 SHORE SHORE

AAC2 AAB1 20.3 m 20.1 m 57 ft deep 56 ft deep

AAC1 AAA4 15.1 m 15.0 m

AAB4 AAA3 10.0 m 10.0 m

AAB3 AAA2 4.95 m 5,05 m

AAB2 AAA1 0,0m 0.0 m 60 ft deep 60 ft deep I Cover End of I Pipe I I I 20.6 m I I12.0 m I I OFFSHORE-EWA OFFSHORE-DIALIOIIDHEAD

FIGURE2. Sketchdepicting locations of five photographicquadrats each on TransectsT-1 and T-2, locatedoffshore of the Kewalo Landfill StationA! adjacentto an abandonedsewer line T-4BAC2 SHORE 20.2 m 45 ft deep

BAC1 15.0 m

BA84 10.0 m

BA83 5.05 m

BA82 460,0 ft mdeep BA81

20.1 m 48 ft deep

BAA4 15.1 m

BAA3 10.1 m

BAA2 5,0 m

BAA1 0.0 m 50 ft deep OFFSHORE

FiGURE3. Sketchdepicting locations of five photographicquadrats on Transects T-3 and T-4, located adjacent to the Kalihi Entrance Channel Station B! T-5 SHORE

CAA1 20.3 m 30 ft deep

CAA2 15,1 m

CAA3 10.3 m

CAA4 5.0 m

CAB1 0.0 m 36 ft deep 0'0mge 150. ea TW CA82 0.0 m 35 fl deep

CA83 5.3 m

CAB4 10,75 m

CAC1 15,1 m

CAC2 20.0 m 38 ft deep OFFSHORE

FIGuRE4. Sketchdepicting locations of fivephotographic quadrats each on TransectsT-5 and T-6, located offshore of the HonoluluInternational Airport Reef Runway StationC!

10 TABLE 1. Summaryof Biological ObservationsMade at TransectT-l, Offshore of Kewalo Landfill StationA! on 14 August 1997

QuadratDistance Along Transect I. Quadrat Survey Om 5m 10 tn 15 m 20 m

Algae Amansia glomerata 2.0 Botryocladia skottsbergii 0.3 Desmi a hornemannii 0.3 Poroli thon onkodes 1.1 Sponges Spirastrella coccinea 1.0 Corals Porites lobata 31.2 13.5 10.5 24.5 4,2 Pori res compressa 4.8 1.0 Pocillopora meandrina 9.3 7.8 7,0 7.5 15.0 Montipora verrucosa 0,1 0.3 0,3 O.l Montipor a patula 0.2 0.1 0.1 0.5 Sand 5.0 2.5 4.0 1.5 4,0 Rubble 12,0 3.5 2.0 12.0 Hard Substratum 41.0 76.0 74.6 55.5 63.7

No. of II. Macroinvertebrate Census m x 20 m! bdl vidual s

Phylum Annelida Spirobranchus giganteus corni culatus 39 Phylum Mollusca Spondylus tenebrosus Phyllida varicosa Phylum Echinodermata Tripneustes grati lla Echinometra mathaei

111,Fish Census m x 20 m!

37 Species 356 Individuals EstimatedStanding Crop = 602g/m2

XoTE:Results of the5-m~ quadrat sampling of the benthic community arepresented in Part I aspercent cover, counts of diurnally exposedmacroinvertebrates are given in Part II, and a summaryof the fish censusis given in Part lil. Water depth rangesfrom 17.4 to 18.2 m; meancoral coverage is 27.6% quadrat method!.

11 TABLE2. Summaryof Resultsfor the PhotographicQuadrat Survey for 1997

PhotographicQuadrat Station A: Transect T-1 Sampled 11 September1997! AAA1 AAA2 AAA3 AAA4 AAB1 m! m! 0 rn! 5 m! 0 m!

Corals Pori res lobata 30.2 10.9 0.6 7.6 Porites compressa 0.8 Pocillopora meandrina 0.8 5.9 4.2 9.2 10.6 Montipora sp. 0,1 Sand 0.2 0,6 2,2 3.9 Rubble 0.2 5.0 2.8 Hard Substratum 64.7 82.6 88.0 89,6 75.1 Mean Coral Coverage = 16.2%

Photographic Quadrat Station A: Transect T-2 Sampled 11 September1997! AAB2 AAB3 AAB4 AAC1 AAC2 m! m! 0 m! 5 m! 0 m!

Unidentifiedred sponge probablySpi rastrella cocci nea! 0,4 0.1 0.1 Unidentified grey sponge probably Plakortis simplex! 0.1 Corals Porites lobata 26.0 12.9 37.5 21,8 16.2 Porites compressa 2.5 Pocilloporameandrina 0,4 3.6 4,2 4.2 4.8 Sand 0.3 Rubble 1.1 2.8 Hard Substratum 73,0 82,3 54.9 70.7 79.0 Mean Coral Coverage = 26.8%

Photographic Quadrat Station B: Transect T-3 Sampled23 September1997! BAA1 BAA2 BAA3 BAA4 BA81 m! m! 0 m! 5 m! 0 m!

Unidentified grey sponge probably Plakortis simplex! 2.8 Unidentifiedred sponge probablySpirastrella coccinea! 0,1 0.1 0,3 Corals Porites lobata 1.4 0.4 0.1 0.3 Pocilloporameandnna 0,3 Monti para sp. 0.3 Sand 5.3 2.2 2.2 2.2 1.4 Rubble 10,6 6.4 1.4 7.3 Hard Substratum 84.0 90.0 95.5 90.3 95.2 Mean Coral Coverage = 0.6%

12 TABLE 2 Continued

Photo hic ad rat Station B: Transect T-4 Sampled23 September1997! BAB2 BAB3 BAB4 BAC1 BAC2 m! m! l0 m! 5 tn! 0 m!

Unidentified red sponge probably Spirastrella coccinea! 0,3 0.4 0.4 0.6 Microciona sp. orange sponge! 0.1 Cords Porites lobata 10.9 0.8 0.6 Pocillopora meandrina 3.4 4.2 Montt'po ra sp, 0.1 0.3 Sand 2.5 1.1 2.2 5.0 1.1 Rubble 8.4 20,7 1.4 9.5 4,2 Hard Substratum 88.8 66.0 94.5 80.7 89.4 Mean Coral Coverage= 4,3%

Photo hic adrat Station C: Transect T-5 Sampled 11 September1997! CAB l CAA4 CAA3 CAA2 CAA 1 m! m! 0 m! l5 m! 0 m!

Algae Porolithon onkodes 15.7 10.9 7.0 Soft Corals Palythoatuberculosa 0.8 2.8 Anthelia edmondsoni 0.3 Corals Porites lobata 0.6 Pocilloporameandrina 1.1 1,1 3.1 Pavona duerdeni 2.5 Montipora sp. 0.8 1.4 Sand 0,8 Rubble 99.2 10.9 1 1.8 20.4 Hard Substratum 71.4 77.3 63.3 Mean Coral Coverage = 2,3%

Photo hic adrat Station C: Transect T-6 Sampled 11 September1997! CAB2 CAB3 CAB4 CAC I CAC2 rn! m! l0 m! 5 m! 0 m!

Algae Poroli thon onkodes 1 1.2 1.7 13.7 Sof Corals Anthelia edmondsoni 0.3 0.1 Corals Porites lobata 11.2 7.6 PocilloPorameandrina 2.1 0.4 0.3 1.4 Montipora sp. 0,7 3,9 1.4 Sand 0.8 Rubble 80.4 98.9 5.6 75.9 Hard Substratum 74.5 13.4 70.3 20.2 Mean Coral Coverage= 6.1%

Nora: Presented in the body of the table are the percent cover of species and snbstrate types for each transect.

13 In total 37 speciesof fishesrepresenting 356 individuals were encountered on Transect T-1.The most common species included the yellowstripe goatfish or weke Mulloidichthys flavolineatus!,the damselfishes Dascyllus albisella and Chromis ovalis, and the palenose parrotfishor uhu Scaruspsittacus!. The standingcrop of fisheson this transectwas estimated at602 g/m~. The species contributing most heavily to this biomass were M. flavolineatus 1% of the total! and S. psittacus % of the total!. TransectT-2 was established 35 m westof TransectT-1 in waterranging from 17.0to 18.2 m in depth.A summaryof the biological informationcollected on this transectis presentedin Table3. Fouralgal species Amansia glomerata, Tolypiocladia sp., Botryocladia skottsbergii,and Porolithon onkodes! having a meancoverage of 1.7%were noted in the quadratsurvey. Also seen were the red sponge Spirastrella coccinea and the grey sponge Plakortissimplex with a meancoverage of 0.4%,and five coralspecies Porites lobata, P.compressa, Pocillopora meandrina, Montipora verrucosa, and M. patula! having an average coverage of 40.0%. The largest contributor to this coverage was P. lobata. The macroinvertebratecensus noted three mollusk species the rock oyster Spondylus tenebrosus, thecone shell Conus lividus, and the pearl oyster Pinctado marginifera!, one polychaete species Spirobranchus giganteus corniculatus!, and three sea urchin species Echinothrix diadema,E. calamaris, and Echinometra mathaei!. The photographic quadrat survey carried out on 11September 1997 noted both red and grey sponge species probably Spirastrella coccinea andPlakortis simplex! and three coral species with a meancoverage of 26.8% Table 2!. Theresults of thefish census are presented in theAppendix. Twenty-eight fish species representing213 individuals were censused on this transect the most abundant species included Mulloidichthys flavolineatus and the saddleback wrasse or hinalea lauwili Thalassomadupeney!. The standing crop of fisheswas estimated at720 g/m>, and the species that contributedthe mostwere the schoolof M. flavolineatus 81%of the total! and Scaruspsittacus % of the total!.

Station B Kalihi Entrance Channel

Station8 is locatedabout 2.2 km seawardof MokaueaIsland, which is situatedin Ke'ehi Lagoon,and about 900 m westof theo/d outfall, which is nowused as an emergency bypass. The two transectsat this stationwere established on a limestonesubstratum about 120m east of theKalihi EntranceChannel in waterranging from 13.7to 15.0m indepth. Much of the substratumin the vicinity of this stationis composedof sandand rubble. An areaof low emergentlimestone approximately 60 m widex 110rn long,with thelong axis oriented perpendicularto shore,is present.Transect T-3 is locatedon the deeperend of this hard substratumarea. Transect T-4 is parallelto but shorewardand approximately 8 m tothe west of

14 TABLE3. Summaryof BiologicalObservations Made at TransectT-2, Offshoreof Kewalo Landfill StationA! on 14 August 1997

QuadratDistance Along Transect I. QuadratSurvey 5m 10m 15m 20m

Algae Amansia glomerata Poroli thon onkodes 2,0 Tolypiocladiasp. 0.5 Botryocladiaskottsbergii 0.2 Sponges Spirastrelkt coccinea 1.0 G.7 0.2 Plakortissimplex 0,1 Corals Porites lobata 27,0 12.2 74.0 37.0 29.0 Porites compressa 2.5 Pocillopora meandrina 6.1 2.0 4.0 4.0 Montipora verrucosa 0,1 0,4 0.1 Montiporapatula G.7 0.9 Sand 1.5 2,0 Rubble 3.0 3.0 Hard Substratum 71.1 73.8 17.5 50.6 63.5

II. MacroinvertebrateCensus m x 20 m! No. of Individuals

Phylum Mollusca Spondylustenebrosus Conus lividus Pinctadomarginifera Phylum Annelida Spirobranchus gi ganteus corni culatus 53 Phylum Echinodermata Echi nothrix calamaris 1 Echinoth rix diadema 25 Echi nometra mathaei 2

III, Fish Census m x 20 m!

28 Species 213 Individuals EstimatedStanding Crop = 720g/m2

NoTE.Results ofthe 5-m2 quadrat sampling ofthe benthic community arepresented inPart I aspercent cover, counts of diurnallyexposed macroinvmtebrates aregiven in Part II, anda summaryofthe fish census isgiven in Part III. Waterdepth rangesfrom 17,0to 18.2m; meancoral coverage is 40.0% quadratmethod!.

15 TransectT-3 seeFigure 3!. Thelack of appropriatehard substratum at thisstation necessitated establishingthe two transects in anend-to-end fashion relatively close to oneanother 8 m apart!. Becauseof this proximity, the fish censuseson both transectsat this station are carried outprior to anyother data collection. During our 1997survey water clarity at thisstation was greater than 15.1 m Table 4!. TransectT-3 hasan orientation that is perpendicularto shoreon thelimestone substratum in waterranging from 14,6 to 15.0m indepth, A summaryof thebiological observations made atTransect T-3 is presentedin Table 5, The quadrat survey noted one algal species Lyngbya majuscula! with a meancoverage of 1.7%,two spongespecies Spirastrella coccinea and Plakortissimplex! with a meancoverage of 1.1%,and six coralspecies Porites lobata, Pocilloporameandrina, Montipora verrucosa, M. patula,and M. verrilli!having a mean estimatedcoverage of 2.5%.The macroinvertebrate survey noted the pinna Streptopinna saccata,the Christmastree worm Spirobranchus giganteus corniculatus, and three sea urchin species Echinostrephus aciculatum, Fchinothrix calamaris, and E diadema!.The photographic quadratsurvey found two unidentifiedsponge species probably Plakortis simplex and Spirastrella coccinea! with a meancoverage of 0.7%and three coral species Porites lobata, Pocilloporameandrina, and Montipora sp.! having a meancoverage of 0.6% up0.2% from the last survey!. Thefish census found 24 species representing 136 individuals and an estimated standing cropof 35 g/m>,The most abundant fishes at TransectT-3 includedthe black damselfish or 'alo'ilo'i Dascyllusalbisella!, the brown surgeonfish or ma'i'i'i Acanthurusnigrofuscus!, andthe manybar goatfish or moano Parupeneus rnultifasciatus!. Thefish species contributing heavily to the biomasson TransectT-3 includedtwo wanderingleatherbacks or lai Scomberoideslaysan 18% of the total!,41 manybargoatfish or moano Parupeneus multifasciatus 22% of thetotal!, and two tablebossor 'a'awa Bodianusbilunulatus 28% of the total!. TransectT-4 sampledthe benthic and fish communitiespresent in thevicinity of the KalihiEntrance Channel. This transect sampled the limestone substratum ata depthranging from 13.7to 14.0m. A summaryof thebiological data collected on 14 August1997 is presentedin Table6. Thequadrat survey noted one algal species Lyngbya maj uscula! with a meancoverage of 1.5%,four spongespecies Spirastrella coccinea, Chondrosia chucalla, Plakortissimplex, and Microciona rnaunaloa! having a meancoverage of 1.4%,and four coral species Porites lobata, Pocillopora meandrina, Montipora verrucosa, and M. patula!. Coral coveragewas estimatedat 5,8% up 1.2' from the last survey!,with Porites lobata and Pocilloporameandrina being the major contributors. The macroinvertebrate census noted the boring bivalve Lithophaga sp., the cone sheH Conus miles, the Christmas tree worm

16 TABLE 4, Summaryof PhysicalMeasurements Made at Eachof ThreeLocations in the Vicinity of theTransect Pairs on 6 December1991, 22 December 1992, 9 September1993, 14October 1994, 26 September 1995, 21 November 1996, and 15 August 1997

Oxygen Temperature Depth to Secchi Location and Time % Saturation! Salinity 'C! Extinction %o! Top Bottom Top Bottom m!

6 DECEMBER 1991 Kewalo Landfill 1035 hr 102 101 34 25.1 25.1 >18,3 Kalihi Entrance Channel 1100 hr 101 101 34 25.0 25.1 >15.1 Reef Runway 1150 hr 102 102 34 25.1 24,9 >12.0 22 DECEMBER 1992 Kewalo Landfill 0900 hr 105 34 22.8 22.8 >18.3 Kalihi Entrance Channel !000 hr 34 22,8 22.7 >15,1 Reef Runway 1035 hr 102 34 22.6 22,8 >12.0 9 SEPTEMBER 1993 Kewalo Landfill 0830 hr 103 104 34 25.1 25.0 >18.3 Kalihi Entrance Channel 0910 hr 103 102 34 24,9 25.2 >15.1 Reef Runway 1020 hr 34 25.1 25.1 >12.0 14 OCH3BER 1994 Kewalo Landfill 0835 hr 103 103 34 25.4 25.1 >18.3 Kalihi Entrance Channel 0920 hr 103 34 25,3 25.2 >15.1 Reef Runway 1030 hr 103 103 25.3 25.4 >12.0 26 SEPTEMBER 1995 Kewalo Landfill 0900 hr 104 103 26.2 26.2 >18.3 Kalihi Entrance Channel 0950 hr 102 102 34 26.0 26.2 >15.1 Reef Runway 1115 hr 103 34 V,9 26.1 >12.0 21 NOVEMBER 1996 Kewalo Landfill 0815 hr 102 34 26.1 26.0 >18.3 Kalihi Entrance Channel 0950 hr 103 26.0 26,2 >15,1 Reef Runway 1145 hr 102 103 26.2 26,1 >12,0 15 AUGUST 1997 Kewalo Landfr II 0820 hr 102 26.3 25.7 >1 8.3 Kalihi Entrance Channel 1000 hr 101 26,0 25.8 >15.1 Reef Runway 1200 hr 102 101 25,9 25.7 >12.0

Narx:Oxygen and temperature measurements were made approximately I m belowthe surface and I m abovethe bottom; waterclarity at all stationswas greater than the depth, thus extinction could not be directly measured.

17 TABLE 5. Sumlnaryof Biological ObservationsMade at TransectT-3, East of the Kalihi EntranceChannel Station B, about2.2 km Offshoreof MokaueaIsland in KeehiLagoon! on 14 August 1997

QuadratDistance Along Transect I. Quadrat Survey Om 10m 15m 20m

Algae Lyngbya maj uscula 4.5 3.0 0.7 0.3 Sponge Spirastrella coccinea 0.7 0.2 1.3 Plakorti s simplex 2.3 Corals Pontes lobata 2,6 0.2 1.0 Poci llopora meandrina 0.2 0.9 1.0 Montipora verrucosa 0.1 Montipora patula 0.1 1,1 1.7 Monti pora verrilli 0.1 Sand 6.0 8.0 2.0 6.0 4,0 Rubble 2.0 2.0 5.0 15.0 3.0 Hard Substratum 84,6 84.1 90.5 75.2 86.2

No. of II. Macroinvertebrate Census m x 20 m! Individuals

Phylum Mollusca Streptopinna saccara Phylum Annelida Spirobranchus giganteus corniculatus 19 Phylum Echinodermata Echinothrix diadema Echinostrephus aci culatum

III. Fish Census m x 20m!

24 Species 136 Individuals EstimatedStanding Crop = 35g/m~

NoTE:Results of the5-m2 quadrat sampling of thebenthic community are presented in Part I aspercent cover, counts of diurnallyexposed macroinvertebrates aregiven in PartII, anda summaryof thefish census is givenin PartIII. Waterdepth rangesfrom 14.6to 15.0m; meancoral coverage is 2.5% quadratmethod!,

18 TABLE6. Summaryof BiologicalObservations Made at TransectT-4, Eastof the Kalihi EntranceChannel Station B, about 2,2 km Offshore ofMokauea Island in KeehiLagoon! on 14 August 1997

!. QuadratSurvey QuadratDistance Along Transect 0 ra 10m 20m

Algae Lyngbya maj uscula 0.3 0.1 3.0 2.5 Sponges Spi rastrella coccinea 2.5 2,0 2.1 Chondrosi a chucalla O.l Plakortis simplex 0.1 Micr ociona maunaloa O.l Corals Porites lobata 0.1 16.0 1,7 1.0 2,4 Poci llopora meandrina 0.1 O.l 2.5 2.0 Montipora verrucosa 0,1 0.1 Monti pora patula 0.2 2.3 O,l O. I Sand 7.0 6.0 6.0 . 9.0 4.0 Rubble 9.0 4,0 2.0 3,0 Hard Substratum 80.6 69.5 89.0 82.9 84.9

II. MacroinvertebrateCensus m x 20 m! No. of Individuals

Phylum Mollusca Li thophagasp, Conus miles Phylum Annehda Spirobranchus giganteus corniculatus 29 Sabellastartesanctij osephi Phylum Echinodermata Echinostrephus aciculatum Echi nothrix calamaris Echinothrix diadema

ID. Fish Census m x 20 m!

13 Species 40 Individuals EstimatedStanding Crop = 9 g/m2

NoTE:Results ofthe 5-m2 quadrat sampling ofthe benthic community arepresented inPart I aspercent cover, counts of diurnallyexposed macroinvertebrates aregiven in PartII, anda summaryofthe fish census isgiven in PartIII. Waterdepth rangesfrom 13.7to l4.0 m; meancoral coverage is 5.S% quadratmethod!,

19 Spirobranchusgiganteus corniculatus, thefeatherduster worm Sabellastarte sanctjiosephi, and threesea urchin species Echinostrephus aciculatum, Echinothrix calamaris, and E. diadema!. In thephotographic quadrat survey two sponge species with a meancoverage of 0,6%and threecorals Porites lobata, Pocillopora meandrina, and Montipora sp.! having a mean coverage of 4.3% were seen. Thefish censusnoted 40 individualfishes among 13 species Appendix!. The most commonfishes present on this transect included the manybar goatfish or moano Parupeneus multifasciatus!,thesmalltail wrasse Pseudojuloides cerasinus, and the brown surgeonfish or ma'i'i'i Acanthurusni grofuscus!. The standing crop of fisheson TransectT-4 wasestimated at9 g/m2,with the important contributors including the lagoon triggerfish or humuhumu- nukunuku-a-pua'a Rhinecanthus aculeatus 21% of thetotal!, the squaretail filefish or 'o'ili lepa Cantherhinessandwichiensis 28% of thetotal!, and the lei triggerfishor humuhumulei Suglamenbursa 24% of the total!.

StationC ReefRunway StationC liesbetween 76G and 840 m seawardof the runway in waterranging from 7.5 to 12.0m indepth. The substratum of this area is a mosaicof emergent limestone spur and grooveformations grading seaward into a seriesof low limestonemounds. The general orientationof thespur and groove formations is perpendicularto theshoreline and direction of usualwave impact. The spurs, which are 5 to4G m inwidth and 30 to 80m inlength, are spacedfrom 10 to 100m apart.Sand is thedominant substratum in the intervening areas. The maximumtopographical relief formedby thesespurs is about3.5 m. Justseaward of this is a zoneof low emergentlimestone where "patches" of hardbottom 5 m x 10 m to severalhundred squaremeters in sizeare present. Spacing between these limestone areas is 10to 50m; again, sand is found in the intervening areas.Corals are restrictedto the areasof hard substratum. Waterclarity at this station was about 12 m duringour 1997 visit; usually clarity here does not exceed12 m. On 15August 1997, the depth to secchidisk extinction was greater than the water depth i.e., more than 12 m, seeTable 4!. HurricaneIniki, which occurred in September 1992, caused considerable damage to the benthiccommunities atStation C. A large approximately 60m indiameter! sand patch located betweenTransects T-5 andT-6 hadlargely been replaced by coral rubble. Much of thehard substratumon bothtransects was broken and the underlying limestone rock exposed, and crevicesand holes were filled in with coralrubble. These physical changes noted in the September1992 survey were much the same in theAugust 1997 survey. TransectsT-5 andT-6 wereestablished on spursor ridgesof limestone see Figure 4!. TransectT-5 was established ona limestoneridge at a depthof9.1 to 11.Gm. Table 7 presents

20 TABLE7. Summaryof Biological Observations Made at Transect T-5, Approximately 760 m Offshoreof theHonolulu International Airport Reef Runway Station C! on l8 August1997

QuadratDistance Along Transect I, Quadrat Survey 5 tn 10m 15 m 20m

Algae Poroli thon onkodes 6.0 7,0 4.0 Desmia hornemannii 0.3 Amansia glomerata 0,4 Soft Corals Anthelia edmondsoni 0,1 0.4 0,7 Palythoa tuberculosa 0.5 1.0 Corals Porites lobata 0.2 0.2 Pori tes compressa 0,2 Pocillopora meandrina 0.1 0.4 0.7 1.0 3.5 Montipora patula 1,0 0.8 2.9 1.3 Pavona varians 0.2 0,6 0.4 Pavona duerdeni 4.2 0,5 Sand 1.5 Rubble 95.3 3'7.0 19.0 9.0 Hard Substratum 3.0 53.8 72.1 80.6 88.6

II. Macroinvertebrate Census m x 20 m! No, of Individuals

Phy! um Mollusca Drupa morum Spondvlus tenebrosus Phy1um Echinodermata Echi nothrix diadema Tri pneustes grati lla

III. Fish Census m x 20 m!

27 Species 156 Individuals EstimatedStanding Crop = 71 g/m~

NoTE:Results of the5-m quadratsamphng of thebenthic community are presented in Part I aspercent cover, counts of diurnallyexposed macroinvehtebrates aregiven in PartII, anda summaryof the fish census isgiven in PartIII, Waterdepth rangesfroin 9.1 to 11.0m; meancoral coverageis 3.6% quadratmethod!, theresults of thebiological survey carried out at Transect T-5. The quadrat survey noted three algalspecies Porolithon onkodes, Desmia hornemannii, and Amansia glomerata! having a mean coverage of 3.8%, two soft coral species Palythoa tuberculosa and Anthelia edmondsoni! with a meancoverage of 0.5%, and six coral species Porites Iobata, P. compressa,Pocillopora meandrina, Montipora patula, Pavona varians, and P. duerdeni! havinga meancoverage of 3.6%.This coverage is downfrom 3,7% in theprevious survey, The invertebratecensus found the drupe shell Drupa morum, the rock oyster Spondylus tenebrosus,and two sea urchin species Echinothrix diadema and Tripneustes gratilla! in the transectarea. The photographicquadrat survey completed on 11 September1997 Table2! notedone algal species the crustose coralline alga Porolithon onkodes! with a meancoverage of 8.4%, two soft coral species Palythoa tuberculosaand Anthelia edmondsoni!with a mean coverageof 0.8%;and four coral species Porites Iobata, Pocillopora meandrina, Pavona duerdeni,and Montipora sp.! having a meanestimated coverage of 2,3%,which is up 0.9% from last year. Thefish census Appendix! noted 156 individuals among 27 species.The most common speciesincluded Acanthurus nigrofuscus, and Ctenochaetus strigosus. The standing crop of fisheson TransectT-5 was estimatedat 71 g/m>,with the most importantcontributors includingCtenochaetus strigosus 8% of the total! andThalassoma duperrey 7% of the total!. Transect T-6 was established approximately 80 m seaward of Transect T-5. The substratumat TransectT-6 was similar to that at TransectT-5 and is situatedon a limestone spurthat is about40 m in width and 80 m in length.Water depth at this locationvaries between 10.7and 11,6 m, A summaryof thebiological observations made on Transect T-6 is givenin Table 8. The quadrat survey found two algal species Porolithon onkodesand Desmia hornemannii!having a meancoverage of 0,4%,one sponge species Spirastrella coccinea! with a meancoverage of 0.02%,one soft coral species Anthelia edmondsoni! with a meancoverage of 0.2%,and five coral species Porites lobata, P, compressa,Pocillopora meandrina, Montiporapatula, and Pavona varians! having a meancoverage of 6.9%.This coral coverage estimateis up from the 1996estimate of 6.6%. The censusof macroinvertebratesnoted four species:the rock oyster Spondylus tenebrosus and three sea urchin species Tripneustes gratilla, Echinometramathaei, and Echinothrix diadema!. The photo quadrat survey Table 2! notedone algal species Porolithon onkodes! with a meancoverage of 5.8%,one soft coral species Anthelia edmondsoni! with a meancoverage of 0.08%,and three coral species PoritesIobata, Pocillopora meandrina, and Montipora sp.! with a meancoverage of 6.1%. Thefish censusnoted 182 individuals belonging to 28 speciesin the4 m x 20 m area, The most abundantfishes on TransectT-6 includedThalassoma duperrey, Acanthurus

22 TABLE8, Summaryof BiologicalObservations Made at TransectT-6, Approximately840 m Offshoreof theHonolulu international Airport ReefRunway Station C! on 18August 1997

QuadratDistance Along Transect I. Quadrat Survey Om 10m 15 m 20m

Algae Porolithon onkodes 1.0 Desmia hornemannii 0.4 0.5 Sponges Spirastrella coccinea 0.1 Soft Corals Antheli a edmondsoni 0,2 0.1 O.l 0.5 Corals Pori res lobata 9.5 0.2 0.1 6,3 4.1 Porites compressa 1.0 Pocillopora meandrina 1.2 1.3 0.3 0,7 0.1 Monti pora patula 4,0 2.4 2.0 Pavona varians 1,1 Sand 2.0 Rubble 11.0 59,5 97.6 12.0 80,3 Hard Substratum 72,0 35.0 78.4 15.0

No. of II. Macroinvertebrate Census m x 20 m! Indi viduals

Phylum Mollusca Spondylus tenebrosus Phylum Echinodertnata Fchi nometra mathaei Echinothrix diadema Tripneustes gratilla

III, Fish Census m x 20 m!

28 Species 182 Individuals EstimatedStanding Crop = 71g/m2

NoTE:Results of the5-m2 quadrat sampling of thebenthic community are presented in Part I aspercent cover, counts of diurnally exposedmacroinvertebrates are given in Part II, and a summaryof the fish censusis given in Part III. Water depth rangesfrom 10.7to 11.6m; meancoral coverageis 6.9% quadratmethod!.

23 nigrofuscus,and Ctenochaetusstrigosus. The standingcrop of fisheson this transectwas estimatedat 71g/m2, with the largest contributors including Acanthurus nigrotu scus 2% of the total!, Thalassomaduperrey 4% of the total!,and Ctenochaetus strigosus 0% of the total!.

PhysicalMeasorements arid Biological Parameters Physicalmeasurements were made in the morningon 15August 1997. These data are presentedin Table4. Little variationwas noted in temperature5.7' to 26.3'C!,percent oxygen saturation01% to 102%!,or salinity all 34%a!,despite the fact that measurements for oxygen and temperaturewere madeboth at the surfaceand about 1 mabove the bottom. In all casesthe secchi disk measurements did notyield an extinction value; water clarity was such thatfrom the surface the disk was still plainlyvisible on the bottom. As hasbeen suggested previously,a bettermethod of determiningwater clarity would be to collectwater samples and measureturbidity with a nephalometerin the laboratory, The biological datafor all eight surveys990 through 1997!are summarizedas means for eachtransect in Table 9. The previous annualdata are from Brock 992a, 1992b, 1993, 1994,1995, 1996, 1997!. Differences are apparent for someof theparameters among the eight years.Some change is evidentin the benthic measures such ascoral cover! betweenthe 1991 and1992 pre- and post-hurricane! surveys, and this is to beexpected. Despite these changes theKruskal-Wallis ANOVA shows that there has been no statistically significant change over theeight-year period. Specifically, there have been no statisticallysignificant changes from the 1990survey through the 1997field effort for meancoral cover on a transect p > 0.92,df = 7, notsignificant!, mean number of coralspecies on a transect p > 0.12,df = 7,not significant!, meannumber of invertebratespecies on a transect p > 0.16,df = 7, not significant!,mean numberof individual invertebrateson a transect p > O.S3,df = 7, not significant!,mean number of fish specieson a transect p > 0.51, df = 7, not significant!, mean number of individualfish on a transect p > 0.56,df = 7, notsignificant!, and mean standing crop of fisheson a transect p > 0.93,df = 7, not significant!.Similarly, the Student Newman Keuls multiplerange test demonstrated no statisticallysignificant differences among any of these parametersbetween the six transectsand eight samplingperiods. Thebiological parameters measured in theeight surveys i.e., number of coralspecies, percent coral cover, number of macroinvertebratespecies, number of macroinvertebrate individuals,number of fishspecies, number of fishindividuals, and biomass of fishes!point to thefact that the Kewalo Landfill station has the most diverse communities, followed by the Reef Runwaystation. The leastdiverse communities appear to be at the Kalihi Entrance Channel station. This hierarchy has not changed over the eight survey years. The low TABLE9. Summaryof Biological ParametersMeasured at the Six TransectLocations in the Eight Annual Surveys 1990 1991 Parameter Transect T-! Transect T-! I 2 3 4 5 6 I 2 3 4 5 6

% Coral Cover 18 30 4 3 2 10 18 29 4 3 2 7 No. of Coral 4 5 4 3 5 5 5 4 3 3 5 4 Species No. of Invertebrate 4 5 6 6 2 2 4 4 4 6 2 2 Species No. of Invertebrate 12 15 25 25 3 5 13 18 17 44 10 10 Individuals No. of Fish 38 37 24 16 29 31 31 26 22 12 28 29 Species No. of Fish 455 481 310 126 197 267 260 240 138 68 176 202 Individuals Fish Biomass 763 824 91 30 129 293 148 221 72 20 101 183 g/m~!

1992 1993 Parameter Transect T-! Transect T-! I 2 3 4 5 6 I 2 3 4 5 6

% Coral Cover 20 21 2 3 0.4 3 26 26 2 4 2 5 No. of Coral 6 6 4 4 4 6 6 7 5 4 5 6 Species No. of Invertebrate 6 5 6 5 3 3 8 7 6 6 2 4 Species No. of Invertebrate 15 12 21 14 7 6 22 20 12 12 6 6 Individuals No, of Fish 36 19 15 9 23 36 31 30 10 19 23 29 Species No. of Fish 312 153 33 27 136 247 343 150 43 63 152 247 Indi vidual s Fish Biomass 736 247 30 14 69 108 1,039 273 31 96 61 79 g/m~!

1994 1995 Parameter Transect T-! Transect T-! I 2 3 4 5 6 I 2 3 4 5 6

% Coral Cover 26,2 31.4 1.6 3.9 1.9 8,3 23.1 37,8 1.7 3.6 3,5 5.3 No. of Coral 6 6 5 4 6 6 6 7 3 4 5 6 Species No. of Invertebrate 7 7 8 6 4 4 9 7 5 7 4 6 Species No. of Invertebrate 27 32 16 14 6 12 62 99 5 7 4 11 Individuals No. of Fish 27 22 13 9 26 27 37 37 17 19 32 34 Species No. of Fish 289 168 30 20 169 224 256 329 95 128 240 257 Individuals Fish Biomass 726 723 27 23 63 86 301 457 27 25 77 93 g/m~!

25 TABLE 9 Continued

1996 1997 Parameter Transect T-! Transect T-! 1 2 3 4 5 6 1 2 3 4 5 6

% Coral Cover 27.3 33.8 2.6 4.6 3.7 40.0 2.5 5,8 3.6 6.9 No. of Coral 5 8 5 4 7 6 5 5 6 4 6 5 Species No. of Invertebrate 6 7 6 4 4 3 6 7 5 7 4 4 Species No. of Invertebrate 22 43 7 28 12 5 49 85 38 45 6 6 Individuals No. of Fish 33 27 25 11 19 30 37 28 24 13 27 28 Species No. of Fish 554 266 160 20 166 234 356 213 136 40 156 182 Individuals Fish Biomass 444 868 157 15 72 83 602 720 35 9 71 71 g/m2!

NoTE:Each transect samples 80 m~ of substratumfor fishesand invertebrates other than corals. Coral data given in percent cover! are from 5 m of substratumsampled on eachtransect. Data for 1990through 1996surveys are from Brock 992a, 1992b, 1993, 1994, 1995, 1996, 1997!.

26 biologicaldiversity at the Kalihi Entrance Channel station is notsurprising in viewof thefact thatthis station was heavily impacted by theold shallow-wateroutfall until 1978and that there is not muchtopographic relief to provideshelter at this location. Froma commercialfisheries standpoint, a numberof importantspecies have been encounteredconsistently in thevicinity of theKewalo Landfill and Reef Runway stations, includingthe yellowstripe goatfish or weke Mulloidichthys flavolineatus!, the yellowfin goatfishor weke'ula Mulloidichthys vanicolensis!, themackerel scad or 'opelu Decapterus macarellus!,the brick soldierfish or menpachi Myripristis amaenus!, ln someyears the emperorfish or mu Monotaxis grandoculis!,the blue trevally or 'omilu Caranx melampygus!,and the grey snapper or uku Aprionvirescens! were also present.

Creen Sea Turtle Observations Priorto HurricaneIniki, greensea turtles Cheloniamydas! were usually seen in the vicinity of TransectT-6. Theseturtles have been absent in this areasince the hurricane, probablydue to theloss of restinghabitat by infilling i.e.,the resting site was completely coveredwith coral rubble!. The coral rubble continues tomove about with each passing storm event.In 1994coral rubble accounted for 10,2%of thesubstratum covered in thequadrats examinedin thefield. In September1995 coralline rubble in thesesame quadrats covered 46.5%of thesubstratum, in November 1996 it covered42.1%, and in August1997 it covered 52.1%.The presence of thiscoralline rubble further decreases the availability of localshelter for fishesand invertebrates. During the August 1997 field work,green turtles were seen in other areas about200 m eastof TransectsT-5 andT-6!.

0 I SC U SS ION

Sincetheir delineationin December1990, the six transectshave been visited on a number of occasionstoensure that, among other things, the permanent markers are remaining in place. Duringthese visits reconnaissance surveys are carried out in theareas surrounding the selected stations.At a minimum,these qualitative surveys cover about 4 hectaresaround each of the threestations. The resultingqualitative observations suggest that the marinecommunities sampledat thethree stations are representative of those found in thesurrounding areas. Theworking hypothesis is thatall threestations, being situated in relativelyshallow water,are outsideof the zoneof influenceof the presentdeep-ocean outfall. However,if impactsfrom the outfall are occurring to the shallow-watercoral reef areasshoreward of the outfall,our monitoringshould be ableto quantitativelydiscern these impacts. Because of

27 bottomtime constraints, potential dangers with deep diving, and the fact that coral community developmentis usuallygreatest in waterless than 30 m deep,the placement of biological monitoringstations was restricted to watersup to 20 m deepin this study.Monitoring the shallow-waterstations provides additional information regarding the recoveryof these communities from the perturbationof raw sewagereleased from the old shallow-wateroutfall from 1955 to 1977.Dollar's 979! study showedthat the Kewalo Landfill station was not directly impactedby dischargefrom the old outfall, but the Kalihi EntranceChannel station was "acutely" perturbedand the stationoffshore of the Reef Runwayreceived an "intermediate" levelof disturbance.Additionally, in themid-1970s the construction of theReef Runway must havecontributed to thedisturbance of benthiccommunities at thisstation Chapman 1979!. The result of theseimpacts is still evidentin the averagecoral coverestimates made at thesestations: themean coverage offshore of theKalihi Entrance Channel is only3,2%, fronting the Reef Runway station it is 4.5%, and offshore of the Kewalo Landfill it is 27,2%. The shallow marine ecosystem fronting Sand Island and Honolulu has received considerableperturbation from human activity over the last 100 years. Among the perturbations has beenthe disposalof raw sewageeffluent in shallow water from the 1930suntil 1977-78, whenthe deep-oceanoutfall becameoperational. From 1955through 1977 the shallow-water outfall released62 mgd m3/s! of raw sewage.Dollar 979! noted two distinct zonesof impact to marinecommunities: the areaof "acute" perturbationwas an ellipse500 m to the east and 1,000 m to the west of the outfall. Outside this area the impacts were evident in a decreasinggradient with distancefrom the outfall. The maximal extent of impactattributed to this sewageinput was 1.9 km to theeast and 5.8 km to thewest of the outfall.The ellipsoid shapeof the zoneof influencewas attributed to thepredominant westerly direction of current flow. The Kewalo Landfill stationis 4.75 km eastand inshoreof the terminusof the outfall, the Kalihi EntranceChannel station is about2.1 km eastand inshoreof the terminus,and the Reef Runwaystation is about3.25 km inshoreand west of theterminus Figure 1!. Presumably,the presentoutfall releasesthe sewageeffluent below the pycnocline,and little interactionoccurs with the inshore biota. Dollar's 979! findings suggestthat if the material was carried to inshore waters, impactsto shallow marine communities would occur in those communities situatedprimarily to the westof the outfaU. The KewaloLandfill stationserves as a controlstation in this study;although coral coverageand fish communitydevelopment are greaterat this location,the stationhas received perturbations in the past. The two transects T-1 and T-2! that samplethe Kewalo Landfill station are situatedclose to an old, nonoperablesewage discharge pipe. Operationsutilizing this dischargepipe ceasedsometime before 1955;the pipe wasprobably used sometime in the

28 1940s A. Muranaka, Oceanographic Team,Department ofWastewater Management, Cityand Countyof Honolulu,personal communication!. Thedevelopment of Kewalo Basin and the entrancechannel inthe mid-1930s would have created considerable turbidity that probably impactedthis station, which is about200 m westof theKewalo Basin entrance channel. From a historicalperspective, human-induced perturbations have probably occurred in all marine communitiessituated in shallowwaters fronting Honolulu during the last 100 years. The KewaloLandfill station was selected asthe control station for thisstudy because of its relativelydiverse coral and fish communities, aswell as its locationwell to theeast of the presentdeep-ocean outfall presumablyout of thezone of influence!. On 11September 1992 the Hawaiian islands were struck by HurricaneIniki. The hurricanepassed directly over Kaua'i, with sustained winds of 144mph and gusts to 172 mph resultingin considerabledamage to improvementsand forests on that island and the west leeward!coast of 0'ahu.To a lesserextent, high surf caused damage tomarine communities alongthe south, east, and west shores of 0'ahu,Kaua'i, Maui, Lana'i, and Hawai'i; this damagewas primarily to coralcommunities. In many areas a largeamount of sandand other loosematerial was moved and/or advected outof theshallow areas i.e., depths of lessthan 27 m! into deeperwaters. On 0'ahu,storm waves emanating from the southeast were estimatedtoexceed 7 m in height and were breaking in water at least 20 m deep personal observations!. Stormdamage tobenthic and fish communities isfrequently patchy, resulting ina mosaic ofdestruction personal observations; Walsh 1983!, and an occasional storm event generating highsurf is one of themost important parameters in determining thestructure ofHawai'i coral communities Dollar 1982!. Numerous studies have shown that storm-generated surfmay keep coralreefs in a nonequilibriurnorsubclimactic state Grigg and Maragos 1974; Connell 1978; Woodleyet al. 1981;Grigg 1983!. The large expanses of near-featureless lavaor limestone substratumpresent around much of theHawaiian islands at depths less than 30 m attestto the forceand frequency of these events Brock and Norris 1989!. These same wave forces also impingeupon and impact fish communities Walsh 1983!. HurricaneIniki caused damage tocoral communities atall three study sites. The greatest impactoccurred tothe benthic communities atStation C ReefRunway!, where portions ofthe Porolirhon-coveredsubstratum upto 1 mx 2m inarea and up to 0.75 m indepth! were completelyremoved. Other areas were entirely covered with coral rubble at scalesfrom 10 m> to over30 m>, In somecases a "blanket"up to 0.5 m ofrubble buried coral colonies or killed thelower portions oflarger colonies. The hurricane broke many coral colonies into pieces; someof thesehave survived where they have been lodged into thesubstratum. These live fragmentsareresponsible forthe increase inthe number ofcoral species seen in some quadrats

29 . betweenthe pre-991! andpost-hurricane 992 through1997! surveys. This phenomenon i.e., live fragments!also servedto lessenthe decreasein coral cover encounteredin someof thequadrats where coverage was low priorto thestorm. Despite these large changes, many of thebenthic components survived, and these communities are recovering well, asevidenced in the increasesin coral cover.However, since Hawai'i corals arerelatively slow growing, it will be yearsbefore the impactof the hurricanewill no longerbe evidentin the benthiccommunities at the study sites. The hurricane also impacted the fish communities at the sample sites. Coral rubble depositedin depressionsserves to lessenthe rugosityof the submarinetopography i.e., shelter!available to fishes.The loss of local sheltercauses fishes to moveand take up residenceelsewhere. At the Reef Runway station, where considerablerubble was present, many of the residentfishes suchas the schoolof emperorfish or mu [Monoraxisgrandoculis]! were no longer found on Transect T-6 after the hurricane. These fish had moved about 100 m eastto an areawhere the coral and benthic communities remained relatively intact. Despitethe impactof HurricaneIniki, the summarydata in Table9, whichspans eight years December 1990to August 1997!,show that there has beenno statistically significant change in the biological parametersmeasured in this study. However, despite the lack of significant change,some parameters show high variability throughtime. Among theseare the number of fish censusedand the estimatedstanding crop of fish. Thesechanges have been greatestat the Kewalo Landfill station. Relative to many other locations in the Hawaiian Islands,the fish communityis well developedat the Kewalo Landfill station.The high standingcrop estimatesin manyyears are much greaterthan for mostcoral reefs;the maximum fish standingcrop encounteredon naturalcoral reefs is about 200 g/mz Goldmanand Talbot 1975; Brock et al, 1979!. Three explanationsfor the high biomassof fishes at the Kewalo Landfill stationare ! the sheltercreated by the old sewagedischarge pipe andgrowth of coral on this pipe locally enhancesthe fish community, ! chanceencounters occur with roving predators or planktivorous and/or other schooling species during censuses,and ! commencingin the summerof 1993a scubadive tour operationbegan feeding the fish in the vicinity of the pipe. The fish feedinghas resulted in an aggregatingeffect of somespecies such as the yellowstripe goatfishor weke Mulloidichthysflavolineatus!, the bluelined snapperor ta'ape Lujtanuskasmira!, and the black triggerfishor humuhumuele'ele Melichthysniger!. Space and cover are important agentsgoverning the distribution of coral reef fishes Risk 1972;Sale 1977;Gladfelter and Gladfelter 1978; Brock et al. 1979;Ogden and Ebersole 1981; Anderson et al, 1981; Shulman et al. 1983; Shulman 1984; Eckert 1985; Walsh 1985; Alevizon et al. 1985!. Similarly, the standingcrop of fishes on a reef is correlatedwith the degreeof vertical relief of the substratum.Thus Brock 954!, using visual techniqueson Hawai'i reefs,

30 estimatedthe standing crop of fishesto rangefrom 4 g/m on sandflats to 186g/m2 in anarea of considerablevertical relief. If structuralcomplexity or topographicalrelief is importantto coral reef fish communities, then the addition of materials to increase this relief in otherwise barrenareas may serve to locally enhancethe biomassof fish.The additionaltopographical relief is usuallyin theform of artificialreefs, but anyunderwater structure such as a deployed sewerline! will havea similareffect. The old sewagedischarge pipe is set abovethe seafloor, creatingconsiderable local topographicalrelief about2 m high! in an areawhere the maximum naturalvertical relief doesnot exceed25 cm. The shelterand high topographical relief must fostergreater development of the fish community seeBrock andNorris 1989!. Chanceencounters with large roving predators such as the grey snapperor uku or [Aprion virescens!,the emperorfish or mu [Monotaxisgrandoculis], the amberjackor kahala [Seriola dumerili!, the blue trevally or papio [Caranx melampygus],the papioCaranx orthogrammus!,schools of planktivorousfishes the mackeralscad or 'opelu[Decapterus macarellus],the sleek unicornfish or kalaholo [Nasohexacanthus], the spotted unicornfish or kala lolo [N, brevirostris], the miHetseedbutterfly fish or lauwiliwili [Chaetodonmi liaris!, the sergeantmajor or mamo [Abudefduf abdominalis]!, or other schoolingspecies the yellowstripegoatfish or weke [Mulloidichthysflavolineatus]! may greatly increasethe counts andbiomass at a particulartransect. The presence of thesewage discharge pipe serves to focus numerouspredators and schooling fishes in the vicinity of the two transectsat the Kewalo Landfill station;hence, an encounterwith thesefishes during a censuswill resultin high biomassestimates. In 1990,at TransectT-6 Reef Runway! chanceencounters with a small schoolof Monotaxisgrandoculis accounted for 51% of the biomassthere, whereas at Transect T-2 Kewalo Landfill! chance encounters with Naso hexacanthus and N. brevirostris accounted for 40% of the biomassand with Seriola dumerili and Caranxorthogrammus, for 21% of the biomass.In 1991N. hexacanthusand N. brevirostrisand some predators were present around TransectsT-1 andT-2 but did not enterthe actualcensus area while the countswere being made,thus they do not appearin the data.In 1992the large schoolof M. flavolineatus that is residentto the old sewagedischarge pipe madeup 78% and93% of the biomasson TransectT- 1 andTransect T-2, respectively;in 1993, 87% and 79%; in 1994,31% arid 44%', in 1995, 68% and 72%; and in 1997, 71% and 81%. In the 1996census an interestingencounter was madewith a roving school of juvenile yellowfin tuna or shibi Thunnus albacares! on Transect T-2. These fishes remained in the general area for the duration of our underwatersampling on 20 November,The estimated weight of that part of the schoolentering the censusarea was 63.9 kg, accountingfor 92% of the standingcrop at this transect.Encounters with tunain shallowwater are rather unusual, but if they do occur, it is usually in areaswhere fish communitiesare relatively well-developed,

31 suchas on artificial reefs.In general,these predaceous fishes are attractedto areaswhere there is a likelihoodof encounteringforage. The encounter with thesetuna supports the point that the fish communities in the vicinity of the Kewalo transectsare relatively diverse and well- developed. Makingbiological measurements underwater can often be a time-consumingprocess; use of thephotographic technique lessens bottom time in measuringcoral and other benthic species coverage.However, as noted by Brock992b!, inspectionof theresults of thecoral coverage datafrom visual assessmentof quadratsin the field relativeto the dataobtained using the photographicmethod points out severalthings. First, meancoral coverageestimates are in reasonableagreement using either method, and the regression of visualversus the photographic coveragedata shows a statisticallysignificant relationship. However, the photographic quadrat techniquedoes not discernsmall coral coloniesor other small colonial benthicspecies such as the soft coral AntheIia edmondsoni;these are easily seenin the field using the visual assessmentmethod. Both methodswork, but the techniqueselected should be done so while keepingthe objectivesof the studyin inind. This studywill continueto useboth methods. The six transectsselected for this study show a considerablerange in community developmentthat is probablyrelated to historicalimpacts. Separating the impact of advanced primary-treatedeffluent released at depthfrom a multitudeof otherongoing and historical impactsthat haveoccurred in and to the shallow marine communitiesfronting SandIsland is difficult at best. The added natural disturbanceof Hurricane Iniki on 11 September1992 provided additional impact to thesecommunities that varied tremendouslywith location. However,the siting of thesepermanent stations to capitalizeon presumedgradient s!of impact createdby the variety of land-derivedsources, as well as the repeatedsampling of these permanentstations, should allow delineationof any changesattributable to the dischargeof sewageeffluent from the Sand Island deep-ocean outfall. The sampling of thesestations during the first two years 990 and 1991!showed that there was little changeto the communities during that time, suggestingthat therewas no quantitativelydefinable impact to shallow-water benthic and fish communitiesdue to the operationof the outfall. Many of the changesseen in the 1992survey appear to berelated to thenatural storm event that occurred in Septemberof that year. The 1993 through 1997 data suggest that recovery from the hurricane is well underway,particularly in the coralcommunities.

32 REFERENCES CITED

AECOS,Inc. 1979. Postconstruction water quality, benthic habitat and epifaunal survey for thereef runway, Honolulu International Airport. Final report, Part B: Benthicbiology. In HonoluluInternational Airport reef runway post-construction environmental impact report,vol. 2, technicalreport, ed. G.A. Chapman unlisted appendix, 69 pp.!. Prepared by ParsonsHawaii for Air TransportationFacilities Division, Departmentof Transportation, State of Hawaii. Alevizon,W., R. Richardson,P.Pitts, and G. Serviss. 1985. Coral zonation and patterns of community structurein Bahamianreef fishes.Bull. Mar, Sci. 36:304 318. Anderson,G.R,V,, A.H. Ehrlich,P.R. Ehrlich, J.D. Roughgarden,B.C. Russell,and F.H. Talbot. 1981.The communitystructure of coral reef fishes.Am. lVat. 117:476 495. Brock,R.E. 1982.A critiqueon the visualcensus method for assessingcoral reef fish populations. Bull. Mar. Sci. 32:269 276. Brock,R.E. 1986. Postconstruction biological survey for theHonolulu International Airport reefrunway: Eight years later. In Surveyof thewater quality, benthic habitat and infaunal populationsfor KeehiLagoon, Hickam Harbor and marine pond, Honolulu International Airport,ed. OI Consultants,Inc., B 1 B28.Prepared for KFCAirport, Inc., Honolulu. Brock,R.E. 1992a. Community structure of fishand macrobenthos at selected sites fronting SandIsland, Oahu, in relationto the SandIsland deep ocean outfall, December1990 survey.Special Report 03.24:92, Water Resources Research Center, University of Hawaii at Manoa, Honolulu.27 pp. Brock,R.E. 1992b. Community structure of fishand macrobenthos at selected sites fronting SandIsland, Oahu, in relationto the SandIsland deep ocean outfall, December1991 survey.Special Report 04.30:92, Water Resources Research Center, University of Hawaii at Manoa, Honolulu.36 pp, Brock,R.E. 1993. Community structure of fish andmacrobenthos at selected sites fronting SandIsland, 0'ahu, in relationto theSand Island deep ocean outfall, year 3 1992. SpecialReport 01.29:93, Water Resources Research Center, University of Hawai'iat Manoa,Honolulu. 37 pp. Brock,R.E. 1994. Community structure of fish andmacrobenthos at selected sites fronting SandIsland, 0'ahu, Hawai'i, in relationto theSand Island Ocean Outfall, year 4 1993. ProjectReport PR-94-14, Water Resources Research Center, University of Hawai'iat Manoa,Honolulu. 36 pp. Brock,R,E. 1995. Cormnunity structure of fish andmacrobenthos at selected sites fronting SandIsland, 0'ahu, Hawai'i, in relationto theSand Island Ocean Outfall, year 5 1994, ProjectReport PR-95-08, Water Resources Research Center, University of Hawai'iat Manoa,Honolulu. 37 pp. Brock,R.E. 1996. Community structure of fish andmacrobenthos at selected sites fronting SandIsland, 0'ahu, Hawai'i, in relationto theSand Island Ocean Outfall, year 6 1995. ProjectReport PR-96-05, Water Resources Research Center, University of Hawai'i at Manoa, Honolulu. 38 pp.

33 Brock,R.E. 1997.Community structure of fish andmacrobenthos at selectedsites fronting SandIsland, 0'ahu, Hawai'i,in relationto theSand Island Ocean Outfall, year 7 1996. Project Report PR-97-04,Water ResourcesResearch Center, University of Hawai'i at Manoa, Honolulu. 39 pp, Brock,R.E., C. Lewis,and R.C. Wass. 1979. Stability and structure of a fishcommunity on a coralpatch reef in Hawaii. Mar. BioL 54:281-292. Brock,R.E., and J.E. Norris. 1989.An analysisof the efficacyof four artificial reefdesigns in tropical waters. BuIL Mar. Sci. 44:934 941, Brock, V,E. 1954.A preliminaryreport on a methodof estimatingreef fish populations. J, Wildlife Mgmr. 18:297 308.

Chapman,G.A. ed.!. 1979.Honolulu International Airport reef runway post-construction environmentalimpact report, vol. 2, technicalreport. Prepared by ParsonsHawaii for Air TransportationFacilities Division, Department of Transportation,State of Hawaii.76 pp. + appendixes,

Connell, J. 1978. Diversity in tropical rain forests and coral reefs. Science 199:1302 1310.

Cox, D,C., and L.C. Gordon,Jr. 1970.Estuarine pollution in the stateof Hawaii Volume 1: Statewidestudy. Tech. Rep. No. 31, Water ResourcesResearch Center, University of Hawaii at Manoa,Honolulu. 151pp.

Dollar, S.J. 1979.Ecological response to relaxationof sewagestress off SandIsland, 0'ahu, Hawai'i. Tech. Rep.No, 124,Water ResourcesResearch Center, University of Hawaii at Manoa, Honolulu. 78 pp.

Dollar, S.J. 1982. Wave stress and coral community structure in Hawaii. Coral Reefs 1:71-81.

Eckert, G.J. 1985. Settlement of coral reef fishes to different natural substrata and at different depths,Proc. 5th Int, Coral ReefCongr. 5:385 390.

Evans,E.C. ed.!. 1974.Pearl Harbor biological survey final report. ReportNo, NUC-TN- 1128,Naval UnderseaCenter, Hawaii Laboratory,

Gladfelter, W.B., and E.H, Gladfelter. 1978. Fish community structure as a function of habitatstructure on West Indian patchreefs. Rev. Biol. Trop. 26 Supplement1!:65-84.

Goldman,B., and F.H. Talbot. 1975.Aspects of the ecology of coral reef fishes. In Biology and Geology of Coral Reefs, Vol. III, Biology 2, ed. O.A. Jones and R. Endean, 124-154. New York: Academic Press.

Grigg, R. 1983.Community structure,succession and developmentof coral reefs in Hawaii Mar. Ecol. Prog. Ser. 11:1 14.

Grigg, R., and J. Maragos. 1974. Recolonizationof hermatypic corals on submergedlava flows in Hawaii. Ecology 55:387 395.

Odgen, J.C., and J.P. Ebersole. 1981. Scale and community structureof coral reef fishes: A long-term study of a large artificial reef. Mar. Ecol. Prog. Ser. 4:97-104.

34 Ricker,W.E. 1975.Computation and interpretation of biologicalstatistics of fish populations. Bull. Fish. Res,Bd. Canada 191.382 pp.

Risk, M.J. 1972.Fish diversity on a coralreef in the Virgin Islands.Atoll Res.Bull. 153:1 6.

Sale, P.J. 1977. Maintenanceof high diversity in coral reef fish communities. Am. Nat, 111: 337-359.

SAS Institute Inc. 1985. SAS user's guide: Basics, version 5 edition. SAS Institute Inc., Cary, N.C. 1,290 pp.

Scott, E.B. 1968. The saga of the Sandwich Islands. Crystal Bay, Lake Tahoe, Nevada: Sierra-TahoePublishing Co. 933 pp. Shulrnan,M.J. 1984.Resource limitation and recruitmentpatterns in a coral reef fish assemblage.J, Erp. Mar. Biol. Ecol. 74:85 109.

Shulman, M.J., J.C. Ogden, J.P. Ebersole, W.N. McFarland, S.L. Miller, and N.G. Wolf, 1983.Priority effects in therecruitment of juvenilecoral reef fishes, Ecology 64:1508 1513.

Siegel,S. 1956.Nonparametric statistics for the behavioralsciences. New York: McGraw-Hill Book Co. 312 pp. Sokal,R,R., and F.J. Rohlf. 1981.Biometry, 2d ed.New York: W.H. Freeman.855 pp. Walsh,W.J. 1983.Stability of a coralreef fish conununityfollowing a catastrophicstorm. Coral Reefs 2:49&3.

Walsh, W.J. 1985.Reef fish community dynamicson small artificial reefs: The influence of isolation, habitatstructure, and biogeography.BulL Mar. Sci. 36:357 376.

Woodley, J,D., and 19 others. 1981. Hurricane Allen's impact on Jamaicancoral reefs. Science 214:749 755.

35

APPENDIX Results of the Quantitative Visual Fish CensusesConducted at Six Locations Offshore of Sand Island, 0'ahu, Hawai'i, 14 and 18 August 1997

TMlsect FAMILY and Species T-1 T-2 T-3 T-4 T-5 T-6

MUI~NIDAE Gymnothorax meleagris Gymnothoraxeurostus Gymnothoraxundulatus

SYNODONTlDAE Sauruja vartegatus

SCORPAENIDAE Scorpaenopsiscacopsis

APOGONIDAE Apogon kallopterus

CARANGIDAE Scomberoideslaysan Caranx melampygus

MULLIDAE ' Muloidichthys flavolineatus 55 75 Parupeneusmultifasciatus 4 12 41 Parupeneuspleurostigma .1 1

CK~ODONTIDAE Forcipiger flavissimus Chaetodon multicinctus Chaetodon ornatissimus Chaetodon unimaculatus Chaetodon miliaris Heniochusdi phreutes

POMACANTHIDAE Centropygepotteri 1 1 Dascyllus albisella 52 15 Plectroglyphidodonj ohnstonianus Plectroglyphidodonimparipennis Chromis vanderbilti 13 Chromi s hanut Chromis agilis Chromis ovalis 54 Stegastesfasci olatus

CIRRHITIDAE Paracirrhites arcatus Ci rrhi topsfasciatus

LAB RIDAE Labroidesphthi rophagus Bodianus bilunuiatus Chei linus bimaculatus Pseudochei li nus octotaenia 1 1 1 3 Thalassomaduperrey 22 35 16 36 Gomphosusvanus 3 3 4 Cons gatmard 2 Coris venusta 2 1

38 Results Continued

Transect FAMILY and Species T-1 T-2 T-3 T4 T-5 T-6

Pseudojuloides cerasinus 1 Stethojulis balteata 2 4 Macropharyngodongeo+roy 6 Halichoeres ornatissimus 1

SCARIDAE Calotomus carolinus 1 Scarus psi ttacus 35 20 13 15 Scans sordidus 14 Scarus perspicillatus 14 Scarus rubroviolaceus

BL.~lIIDAE Exallis brevis Cirripectus variolosus

GOBIIDAE Fsi logobi us mainlandi

ACANTHURIDAE Acanthurus nigrofuscus 19 16 30 32 Acanthurus nigroris 21 9 7 3 Acanthurus dussumieri 3 Acanthurus oli vaceus 3 Ctenochaetus strigosus 4 44 25 Zebrasomaflavescens 16 Naso lituratus 106 2 2 Naso hexacanthus Naso brevirostris Zanclus cornutus

BALISTIDAE Melichthys niger Meli chthys vidua Suglamenbursa Rhinecanthus aculeatus

MONACANTHIDAE Pervagormelanocephalus Pervagor spilosoma Cantherhines sandwichiensis

TETRAODONTIDAE Arothron hispidus Canthigasterj actator Canthi gaster ci nctus Canthigaster rivulata

Total No. of Species 37 28 24 13 27 28 Total No. of Individuals 356 213 136 40 156 182 EstimatedStanding Crop g/m2! 602 720 35 9 71 71

NOTE: Each entry in the body of the table representsthe total number of individuals of each speciesseen; totals are presentedat the foot of the table, along with an estimate of the standing crop of fishes presentat each location, All censuseswere carried out by the author.

39