coral reefs: present problem ands future concerns resulting from

AMER. ZOOL., 33:524-536 (1993)

Coral Reefs: Present Problems and Future ConcernsResulting from Anthropogenic Disturbance'

ROBERT H. RICHMOND

Marine Laboratory, University of Guam, UOG Station, Mangilao, Guam 96923

SYNOPSIS. Coral reefs, with their vast diversity of invertebrate, verte-brate and algal species, have undoubtedly been subjected to natural dis-turbance since their appearance millions of years ago. Anthropogenicdisturbance has been a factor affecting reefs for a fraction of that time,yet in terms of overall impact, may be of greater concern. Data on habitatdestruction, pesticide and heavy metal accumulation, nutrient loading,sedimentation, runoff and related impacts of man's activities indicate thatmany coastal reefs are endangered by these processes through alterationsin animal-algal symbioses, shifts in competitive interactions, direct mor-tality, reproductive failure, and insufficient recruitment. The death ofcorals critically affects reef communities, as corals provide an importanttrophic link as well as the main habitat structure. While natural distur-bance is an important factor affecting reef interactions, species diversityand evolution, chronic anthropogenic disturbances combined with unsuit-able environments for recovery, are of great concern. Physiological stresscan be measured in corals in addition to outright mortality, allowing theimpacts of specific disturbances to be assessed. Sufficient data for distin-guishing real problems from temporal variability are becoming available,allowing scientists to focus on practical solutions to problems in coralreef management and preservation.

INTRODUCTION

Coral reefs are diverse and productivebiological communities which thrive inshallow and coastal tropical marine envi-ronments. While scleractinian reef-buildingcorals are not necessarily the most abundantor diverse faunal component of coral reefs,they provide the initial trophic link throughtheir symbiosis with algae and produce themajority of the habitat structure for otherreef organisms. Many invertebrate, fish andalgal species are integral members of ahealthy reef community. By understandingtrophic relationships and species interac-tions, we can gain an appreciation for howcoral reefs work, and of critical importance,how changes including anthropogenic dis-turbance, can affect community structureand function.

1 From the Symposium The Crisis in InvertebrateConservation presented at the Annual Meeting of theAmerican Society of Zoologists and the Canadian Soci-ety of Zoologists, 27-30 December 1992, at Vancou-ver, British Columbia.

Coral reefs are unique among high-diver-sity and high-productivity marine com-munities, distinguished by their ability tothrive in clear, oligotrophic waters devoidof high levels of nutrients. As in temperatemarine communities, unicellular algae areresponsible for the initial photosyntheticfixation of carbon on coral reefs; however,here they reside intracellularly within theendodermal cells of the coral host in a mutu-alistic symbiotic association rather than asphytoplankton. The relationship betweenreef-building corals and their symbiotic uni-cellular dinoflagellates ("zooxanthellae") iscentral to the existence of coral reef com-munities. Metabolites are exchangedbetween the algae and the host, and nutri-ents are conserved in an otherwise nutrientlimited environment. Changes in the envi-ronment which affect the symbiotic asso-ciation {e.g., nutrient levels and light) willaffect coral nutrition, metabolism and cal-cification, and hence, the entire reef com-munity.

Coral reefs are biogenic structures which524

ANTHROPOGENIC DISTURBANCE ON CORAL REEFS 525

may form banks, atolls, islands, and sub-stantial masses like the Great Barrier Reefof Australia. Living reefs not only form land,but provide the sand that lines tropicalbeaches, and the structures which bufferwaves that would otherwise cause extensivecoastal erosion. The protection that reefsprovide islands is especially evident duringtyphoons and tropical storms where reefcrests, reef flats and windward spur-and-groove formations dissipate wave energyquickly and efficiently. The balance betweenreef accretion and erosion depends on theliving veneer of corals and coralline algae.The death of key organisms or the shift froman autotrophic to a heterotrophic (suspen-sion/detritus feeding) community shifts thedynamics from carbonate deposition to reeferosion.

The economic value of coral reefs is evi-denced by the islands and land they produceand protect, the fisheries which they sup-port, the tourists they draw, the recreationalopportunities they afford and the diversityof natural products which they producewhich have already proven to be of bio-medical importance. While the value of aresource is often measured in dollars, oneshould not ignore the cultural value of coralreefs, which is every bit as important toislanders and tropical coastal populationsas the rain forest is to its inhabitants. McAl-lister (1988) estimated fisheries losses dueto reef degradation at over $80 million peryear, impacting 127,000 jobs and 637,000family members. Finally, coral reefs aresimply beautiful, and that may be reasonenough to warrant their protection.

Coral reef organisms are usually consid-ered stenotypic, exhibiting a relatively nar-row range of tolerances to environmentalconditions, hence small changes in environ-mental quality can affect critical biologicalprocesses. Reproduction and recruitment arethe two processes by which reef populationsare maintained, and both can be quantita-tively assessed, allowing determination ofsublethal effects of environmental changes.While corals and many other reef inverte-brates are capable of asexual reproduction(fragmentation, tissue sloughing and regen-eration), successful production of larvae,either through release of gametes with exter-

nal fertilization or via internal fertilizationand brooding, is essential for maintainingreef populations (see reviews by Richmondand Hunter, 1990 and Harrison and Wal-lace, 1990). Water and substratum qualityaffect reproduction and recruitment success,and hence, should be the focus of studieson long and short term effects of stress anddisturbance.

The terms "stress" and "disturbance"have been applied to coral reefs and manyother biological communities, with a vari-ety of interpretations. In this paper, I willadhere to the concepts suggested by Rosen(1982) and Brown and Howard (1985) whichrecognize that a gradient of conditions exists,from ideal to the absolute limits of survival,and that effects short of mortality need tobe considered. Stress is a physiological con-dition which results from adverse or exces-sive environmental factors and in corals canbe measured by decreased growth rates,metabolic differences, and biochemicalchanges. Disturbance is an ecological phe-nomenon which includes departures from aroutine set of conditions.

There are varying levels of degradationwhich can be observed on coral reefs, fromthe extreme and obvious (mortality) to moresublime changes in characteristics includingcompetitive dominance among organisms,decreased growth rates, breakdown oforganismal associations, reduced fecundity,reproductive failure, and declining recruit-ment of larvae. Essentially, whether a coralreef is killed in a week, due to sedimentburial, or over a ten year period, due toattrition and lack of recruitment, the resultis the same: the loss of the coral reef com-munity and all of the benefits which it offers.The two plenary presentations at the Sev-enth International Coral Reef Symposium(Guam, 1992) focused on world-widedestruction of coral reefs in the face ofincreasing pressure from man's burgeoningpopulations, indicating an awareness amongcoral reef scientists that anthropogenic (man-induced) disturbance is a critical problem(Buddemeier, 1993; Wilkinson, 1993).

As this paper was prepared for a sym-posium entitled "The crisis in invertebrateconservation," the focus will be on the effectsof disturbance on coral reefs, and specifi-

526 ROBERT H. RICHMOND

cally, with concerns for the health of reefsin the face of mounting pressure from man'sactivities. The range of activities which arecause for concern includes runoff and sed-imentation from development projects,eutrophication from sewage and agriculture,physical impact from maritime activities,dredging, collecting and destructive fishingpractices, pollution from industrial sources,golf courses and oil refineries, and the syn-ergistic impacts of anthropogenic distur-bance on top of natural disturbance.

Some coral researchers have argued thatnatural disturbances such as El-Nino eventsand typhoons have historically devastatedvast stretches of coral reefs, and that bycomparison, documented anthropogenicreef loss has been smaller in scale and ofindeterminate long-term consequences. Forexample, Grigg and Dollar (1990) recognizethe seriousness of damage to local reefs fromhuman activities, but they remain uncon-vinced that anthropogenic disturbances areas serious or pervasive as natural distur-bances. I disagree with arguments thatdownplay the importance of anthropogenicdisturbance because: 1) we cannot controlnature but we can control human impacts;2) differences do exist between the effects ofnatural and anthropogenic disturbance; and3) the synergistic effects of anthropogenicdisturbance on top of natural disturbancechange the conditions for recovery. Thispaper addresses the most common types ofanthropogenic impacts on coral reefs,important differences between types of dis-turbance, summarizes the results of someprevious studies, and offers some ideas onmitigation measures as well as future areasof research.

IMPACTS ON CORAL REEFS

SedimentationSedimentation, which is the most well-

studied impact, may affect corals three dif-ferent ways: photosynthetically, physically,and chemically. As most reef-building cor-als obtain the majority of their nutritionalrequirements via translocation of metabo-lites from their photosynthetic partners(Muscatine et ai, 1981), any reduction inthe available quality and/or quantity of lightwill affect coral nutrition, growth, repro-duction and depth distribution. Bak (1978)

reported decreased coral growth ratesresulting from decreased light levels avail-able to corals due to sedimentation fromdredging. Rinkevich (1989) found that plan-ula production in Stylophora pistillata isenergetically supported by metabolitestranslocated from symbiotic zooxanthellae,linking photosynthesis to reproduction incorals. Other studies (see reviews by Brownand Howard, 1985 and Rogers, 1990) havealso shown that decreased light levels havea detrimental effect on corals, and can limitthe depth range over which corals can exist.

Physically, sediments also interfere withcoral nutrition by coating the feeding sur-faces responsible for catching prey itemsneeded to supplement the energy providedby zooxanthellae. While corals do have theability to cleanse themselves using a com-bination of mucus secretion and ciliaryaction, chronic sedimentation may exact ahigh energetic cost, adding to the overallimpact on the colony. Sedimentation canalter species composition of reefs throughphotosynthetic and physical effects. Changesin relative abundance of morphological typesas well as individual species are an impor-tant reflection of how sedimentation as adisturbance affects community structure.

Sediments can also physically interferewith recruitment of coral larvae, whichrequire a solid substratum upon which tosettle and metamorphose. Te (1992a) foundtissue from newly settled and calcifying col-onies ofPodllopora damicornis "bailed out"of their benthic exoskeleton in response toincreased sedimentation, which is a previ-ously described stress response (Sammarco,1982; Richmond, 1985). Dredging projectshave been particularly damaging to reefs,primarily through the initial physical dis-turbance, habitat alteration and the subse-quent problems associated with sedimen-tation.

Few studies have focused on the chemicaleffects of sediment on corals, which can beimportant. Brown and Holley (1982) andHoward and Brown (1984) studied theeffects of heavy metals (copper, tin and zinccompounds) on adult coral colonies, withmixed results. Goh (1991) determined lowlevels (9 ppm) of nickel caused mortality incoral planulae and significantly reduced lar-val settlement rates at concentrations of 1


ppm. Several studies in Okinawa, Japan,have found that lateritic soils (red clay, highin iron), have been particularly detrimentalto reef corals compared to carbonate sedi-ments (Nishihira, 1987; Yamazato, 1987;Sakai et al, 1989).

An aspect of the chemical effects of sed-iment on coral reefs has not yet beenaddressed: How do chemically treated soilsdeposited on coral reefs affect residentorganisms? Golf course construction ontropical islands has increased at an alarmingrate in the past few years, considering thesize of the land masses involved, the quan-tities of pesticides and fertilizers used, andthe fresh water requirements for maintain-ing greens and fairways. If the pesticidesused bind to the soil, and the soil ends upon the reef due to erosion, the question ofchemical impacts needs to be addressed.Clay particles, which by their small size willoften be carried to the ocean by runoff, havea charge which may change when enteringthe marine environment. Chemicals boundto particles on land may be released in sea-water. Experiments with the organophos-phate pesticide Dursban (chlorpyrifos) foundseawater passed through a column of soiltreated with a quantity of chemical equal tothat applied to golf courses, was toxic to thecoral Pocillopora damicornis (Te, 19926).Acevedo (1991) determined chlorpyrifos atlevels of 1 ppm resulted in up to 50% mor-tality in assays with coral planulae. Bioas-says like these indicate the need for moresuch studies.

RUNOFF/CHEMICAL POLLUTION/WATERQUALITY

A general rule for islands: Whatever isused on land today ends up in the aquiferor coastal zone tomorrow. While sedimentcarried by runoff has been a major focus ofenvironmental studies, little attention hasbeen paid to the chemistry of runoff water.Salinity changes alone have proven to affectcorals, especially on shallow water reef flatswhich are most likely to be impacted byfreshwater runoff (Kato, 1987; Jokiel et al.,1993). What are the specific concerns thatneed to be addressed?

Having reviewed numerous environmen-tal impact statements (EISs), water qualityreports on runoff usually read "all analyses

performed found chemicals assayed for werein quantities below detectable limits." Thissays nothing about the effects of the chem-ical component of runoff on the local floraand fauna. Bioassays are the appropriatetests to be performed, and should be requiredbefore permits are approved. Recentadvances in our understanding of inverte-brate reproduction in general and settle-ment and metamorphosis of benthic inver-tebrate larvae in particular demonstrate thereasons for concern.

Most scleractinian corals are simulta-neous hermaphrodites (containing both maleand female gonads at the same time) whichparticipate in discrete annual multispeciesspawning events (Richmond and Hunter,1990; Harrison and Wallace, 1990). ForGuam and Okinawa, the timing of masscoral spawning events coincides with theheight of the rainy season, when coastalmarine surface waters are most likely to becontaminated from terrigenous runoff.

During the summers of 1989, 1990, and1991 coral fertilization bioassays were per-formed comparing fertilization and devel-opment success of gametes and embryos,respectively, among waters of differingsalinity and sediment content. In one exper-iment performed using water samples col-lected on the night of coral spawning abovea reef adjacent to a stream mouth in Oki-nawa, runoff was found to cause an initial53% drop in fertilization rate compared toa control, and an additional 51% drop inthe number of embryos developing to theplanula larva stage (Richmond, unpub-lished). The experimental treatment sea-water was determined to have a salinity of28.5%o, with suspended solids (red clay) of1.28 g/liter, while the control water was Mil-lipore filtered (0.45 Mm) and had a salinityof 34.4%o.

Since no additional chemical analyseswere performed on the runoff-affected sea-water, it was not possible to determine ifother substances were responsible for the77% drop in larval production compared tothe control. A subsequent experiment todetermine the effects of decreased salinityalone showed an 86% reduction in fertiliza-tion rate accompanying a 20% dilution ofseawater with distilled water. These exper-iments demonstrated that actual coastal


FIG. 1. Sediment plume off Southern Guam. Corals up to several hundred years old were killed by sedimentburial. Water samples from plumes like this were found to cause up to an 86% drop in fertilization rates inspawning corals. The plume is ca. 1 km in diameter.

surface water quality above reefs during coralspawning events was sufficiently reduced tocause reproductive failure. Considering mostcoral species spawn once a year, during therainy season when coastal pollution wouldbe expected to reach its peak, and that mostcoral eggs are buoyant, floating in the sur-face water layer for up to several hours beforefertilization occurs, it is easy to see the linkbetween terrigenous runoff and reproduc-tive failure of spawning reef species. Fur-thermore, chemical cues have been foundto allow synchronization of spawning incorals (Atkinson and Atkinson, 1992; Rich-mond, unpublished). Decreased water qual-ity could also affect these critical cues, pre-venting synchronous release of gametes andresulting in lowered reproductive success.

Suppose spawning occurs in a pristineenvironment, allowing fertilization andembryological development to occur at nat-ural rates. Can fully developed planula lar-vae settle in areas of reduced water quality?Recent discussions of larval recruitment inbenthic invertebrates have suggested thatsettlement and metamorphosis are different

events which should be addressed sepa-rately (Hadfield and Pennington, 1990;Pawlik and Hadfield, 1990). While larvaeof benthic organisms may settle out of theplankton and come in contact with the sub-stratum, metamorphosis may not occurwithout chemosensory recognition of spe-cific inducing molecules (Morse, 1990). Theconcentrations of metamorphic inducers innature are far below detectable limits basedon present technology, yet are obviously insufficient quantities in the marine environ-ment to affect metamorphosis (10~10 M forthe nudibranch Phestilla sibogae, Hadfieldand Pennington, 1990). Pollutants below thedetectable limits of high performance liquidchromatography (HPLC) are not necessarilybelow the limits of interfering with criticalchemical cues in marine invertebrates.

Guam's southern reefs provide an exam-ple of how water quality impacts coral reefsthrough sublethal effects. During the periodfrom 1988-90, a major road constructionproject was undertaken on southern Guam,which is geologically volcanic with steeplysloping, highly erodible lateritic soils. Large


FIG. 2. Coastal reef off Southern Guam which was killed by sedimentation. A fairly dry typhoon removedsediment accumulations from the reef.

quantities of sediment ladened freshwaterrunoff impinged on coastal reefs, causinghigh levels of coral mortality, rapid growthof fleshy algal species, and large parcels ofreduced salinity/quality seawater (Figs. 1-4). Local fishermen have complained ofdecreased fisheries and reef vitality not onlyon these coastal reefs, but also on offshoreislands and reefs not directly impacted bycontact with the sediment. Inspection ofthese reefs revealed live adult coral colonies,but no signs of larval recruits for the periodcoinciding with and following constructionactivities and increased levels of sedimen-tation and runoff.

In contrast, surveys of Kossol Reef,Republic of Palau, performed in September,1992, revealed up to 12 coral larval recruitsper square meter in an area devastated bya typhoon in 1989. Coral recruits could beidentified from each of the year classes since1989, including small colonies resultingfrom the 1992 summer spawning. The pointis, while levels of stress may be sublethal toadult coral colonies, they may be sufficientto cause reproductive and recruitment fail-

ure on nearby and distant reefs. Consideringcoral planulae remain competent (able tosuccessfully settle and metamorphose) forperiods from days to months (Richmond,1987, 1988), a regional view of coral reefpopulation dynamics is needed. Reproduc-tive failure in one area may affect recruit-ment elsewhere. Numerous genetic studieshave shown gene flow among populationsof marine invertebrates. The significance ofenvironmental degradation in one area onrecruitment in another should not beignored.

Oil pollution is an extreme example ofhow chemicals, in this case hydrocarbons,can impact reefs. Research performed in theGulf of Eilat has documented coral mor-tality, decreased fecundity and recruitmentfailure in response to chronic oil pollution(Fishelson, 1973; Loya, 1975, 1976; Loyaand Rinkevich, 1979; 1980). Chemical fish-ing techniques including the use of cyanidefor collecting aquarium fish and chlorinebleach for consumptive fishing have alsohad a negative impact on reefs in the Phil-ippines and Micronesia.


FIG. 3. Porites colony being overgrown by algae.

Water quality is a critical considerationin understanding anthropogenic impacts oncoral reef communities, especially in howpollutants affect egg-sperm interactions andchemosensory cues critical to reproductivesynchrony, fertilization success, larval set-tlement, metamorphosis and recruitment.

SEWAGE

The effects of sewage on coral reefs,reviewed by Pastorok and Bilyard (1985),result from several factors including nutri-ents, sediment (suspended solids) and toxicsubstances. The overall impact of sewageon a coral reef community depends on site-specific conditions including volume ofsewage, level of treatment, presence of toxicmaterials, and receiving water characteris-tics.

The effects of sewage-related nutrientenrichment on coral reef communities havebeen documented, and include alteration ofcompetitive interactions, reduction of coralcalcification rates from decreased light lev-els and increased phosphate concentrations,

and increased mortality from bacterialinfection (Smith et al, 1981; Pastorok andBilyard, 1985; Tomascik and Sander, 1987).Corals are adapted to live in nutrient poorenvironments (Muscatine and Porter, 1977),and are relatively slow-growing comparedto algae, sponges, tunicates and other groupsof sessile, benthic organisms. Nutrients notonly increase the biomass of phytoplankton,affecting light transmission and increasingthe biochemical oxygen demand (B.O.D.)to the point that corals may be impacted(Guzman et al., 1990), but also give a com-petitive advantage to faster growing benthicspecies. The green algae Dictyosphaeria cav-ernosa, formed large mats, covering andkilling corals in Kaneohe Bay, Hawaii, dueto sewage pollution (Smith et al., 1981;Evans et al., 1986). Additionally, portionsof the Kaneohe Bay community shifted froma coral-dominated autotrophic communityto an algae-dominated suspension feedingcommunity because of eutrophication. Spe-cies of sabellid and serpulid worms and otherboring organisms thrived during the period


FIG. 4. Nutrient input caused an increase in fleshy algae biomass, which is now binding the sediment, preventingnormal turbulence from cleansing the reef of accumulated sediment.

of sewage input, eroding the carbonatestructure of the reefs and undermining thebase upon which coral recolonization tookplace following sewage diversion.

Dubinsky et al. (1990) have also dem-onstrated that nutrient enrichment via sew-age reduces the photosynthetic efficiency ofcorals, as algal cells increase in density tothe point of becoming self-shading. Sincethe coral-zooxanthellae symbiosis evolvedunder nutrient limited conditions, it is rea-sonable to assume that the relationship willbecome altered in response to changes inthe level of nutrients available. Furtherstudies of the physiological effects of suchchanges are needed to determine the sub-lethal or long-term effects of sewage andnutrient enrichment on coral reefs.

While the effects of suspended solids fromsewer outfalls have been compared to thosefrom terrigenous runoff and sedimentation,the two types of sediment differ in physical,chemical and toxicological characteristics,which must be considered when assessingimpacts (Pastorok and Bilyard, 1985). Sew-

age suspended solids are primarily organic,can contain adsorbed toxins, and increaseB.O.D. more than inorganic sediment asso-ciated with runoff. The toxic component ofsewage depends on the sources of input, andis primarily a concern in industrial or agri-cultural areas where industrial wastes andpesticides are included in the effluent.

TEMPERATURE STRESS

The negative impacts of increased tem-perature on corals have been documentedfrom both anthropogenic and naturalsources. Jokiel and Coles (1974) found coralmortality associated with the heated ther-mal discharge from a cooling system for apower plant in Hawaii. Glynn (1990)reported widespread coral mortality in theeastern Pacific associated with increasedtemperatures accompanying the 1982-83 ElNino event. In both cases the cause of mor-tality appeared to be the breakdown of thesymbiotic association between the zooxan-thellae and the coral host (bleaching). Animportant distinction between the two


sources of stress is duration. Studies follow-ing the two examples presented here foundcorals recovering after the temperature stresswas removed, indicating corals can reboundfrom acute temperature disturbances. The1982-83 El Nino ended naturally, but it tookredesigning and rebuilding the power plantoutfall before recovery occurred at theHawaii site.

CORAL BLEACHING

Widespread coral bleaching (loss of zoo-xanthellae) has been observed in both theAtlantic and Pacific oceans, and has beenlinked to unusually high temperatures andirradiance (Williams and Williams, 1988;Williams et al., 1987). This topic is reviewedin a special issue of the journal Coral Reefs(vol. 8, no. 4, 1990), which presents dis-cussions of both natural and anthropogenicsources of temperature increase. The rela-tionship between bleaching events and ozonedepletion/global warming is presently beingstudied by several groups of researchers(Buddemeier, 1993). If the connection canbe proven, it will be an example of globalrather than local anthropogenic impacts oncoral reefs.

CORAL DISEASES

Four types of coral diseases have been"identified": white band disease, black banddisease, bacterial infection, and shut-downreaction (Antonius, 1981). While there is adegree of uncertainty as to the causesresponsible for each disease, they all appearto be stress-related. This is one area wheresynergisms are believed to play an impor-tant role, as stressed corals seem to be themost susceptible. Tumors, bacterial attackand parasitic worms have been observed inareas where corals have been stressed bysediment, sewage, pesticides, heavy metalsand other human impacts (Mitchell andChet, 1975; Brown and Howard, 1985;Glynn et al., 1989; C. Hunter, personalcommunication).

Toxic WASTE

Within the last few years, a new threathas emerged, endangering the health of cor-al reefs of the Pacific Islands. Several com-panies have been targeting cash-poor devel-

oping Pacific Islands as potential sites fordisposing of toxic wastes. In 1990, a pro-posal from an Australian company was sub-mitted to the Government of Palau for theconstruction of a toxic waste disposal incin-erator/power plant. Among the materialslisted as suitable to be used as fuel to pro-duce electricity were coal tailings, sewagesludge, dioxins, lead, sulphur, and cyanidecontaminated wastes, PCBs, and heavymetal liquors (Graves, 1992; LohningBrothers, 1990). In 1992, the U.S. Securitiesand Exchange Commission suspended trad-ing of stock in a company called PacificWaste Management Inc., which was pro-moting the development of the toxic wasteincinerator in Palau, noting concernsincluding several name changes of the com-pany, and that the Palau constitution has aprovision banning the importation of toxicmatter (Graves, 1992; North, 1992).

In December, 1992, a barge containing5,200 tons of petroleum contaminated soiltransited Guam en route to the MarshallIslands, where the material was to be usedin construction of a causeway between twoislands in Kwajalein Atoll (Brooks, 1992a,b; Glauberman, 1993). Concerns about theenvironmental acceptability of the pro-posed activity resulted in the fully-loadedbarge returning to its point of origin inHonolulu, Hawaii, where alternate meansof disposal are being considered.

The costs of handling and disposing oftoxic materials are high, and islands in needof income, and without adequate techno-logical expertise, are attractive opportuni-ties for getting rid of such wastes at minimalcosts. I believe this situation remains animportant concern for Pacific coral reefs. Aswith pesticide runoff, the problem may bewith long-term effects, bioaccumulation ofsubstances by organisms and the results ofchronic exposure.

DESTRUCTIVE FISHING PRACTICES

The use of dynamite and poisons, includ-ing chlorox and cyanide, have been respon-sible for the destruction of coral reefsthroughout the world. Because of the sizeof the areas concerned, and the general lackof resources for enforcement, educationappears to be more successful than legisla-


tion in controlling these practices. Povertyreduces the alternatives for fishermen whomust feed their families and rely on fishingas a source of protein and income. This sameproblem has lead to another anthropogenicdisturbance on reefs: overfishing. The useof fish traps made of long-lasting materialswith small mesh sizes results in the captureof pre-reproductive juveniles, affectingfuture populations, and the death of fishwhen traps become dislodged during storms,yet continue to capture fish which eventu-ally starve. Several types of net fishing havealso been responsible for over-exploitationof reef resources as well as impact damageto coral reefs. As with all biological com-munities, each species plays an importantrole in the dynamics of balance. The deple-tion of grazers, for example, may eventuallylead to an overgrowth of algae. While it issimple to prove how damaging destructivefishing practices are to the productivity offisheries, the economic realities of day-to-day life on some tropical islands makes thesolution difficult to obtain.

CROWN-OF-THORNS STARFISH

The coral-eating starfish, Acanthasterplancii, has been the focus of a debate onthe fate of coral reefs since major outbreakswere observed in the late 1960s and early1970s (reviewed in a special issue of thejournal Coral Reefs, vol. 9, no. 3, 1990).While it has been documented that hun-dreds of km2 of coral reefs have been dev-astated by population blooms of the starfish,the debate centers on whether the outbreaksare natural events, having occurred repeat-edly over geologic time, or if the situationhas arisen as a result of man's activities.Although sediment core data have indicatedAcanthaster outbreaks occurred 10,000 yearsago, recent studies have shown a relation-ship between nutrient input and recruit-ment success of the larvae (Birkeland, 1982).Studies of echinoderm reproduction havedemonstrated that the success of recruit-ment of their planktotrophic larvae dependson phytoplankton availability followingspawning. Events that increase nutrientavailability on coral reefs can affect repro-duction and recruitment in Acanthaster.While outbreaks may be considered natural,

an increasing number and/or the persistenceof these events may be linked to anthro-pogenic nutrient input.

DISCUSSION

Anthropogenic versus natural disturbanceThis is certainly not the first discussion

or comparison of anthropogenic versus nat-ural disturbance, nor will it be the last. Asscientists, we often look for trends, gener-alizations and rules, perhaps even wheresuch things are not appropriate. It is easyto pick specific examples which supporteither side of the debate: 1) that there is littlequalitative difference between anthropo-genic and natural disturbance to coral reefs(Grigg and Dollar, 1990), and 2) that impor-tant differences do exist, which affect recov-ery, mitigation and management decisions(Johannes, 1975; Loya, 1976). For example,Grigg and Dollar (1990) report the impactof a kaolin spill on French Frigate Shoal,Hawaii, was trivial, yet tropical storms causecatastrophic coral mortality. The stress onFrench Frigate Shoal was acute, kaolin isinert, non-toxic with no B.O.D., and the siteis an area with high water motion, farremoved from any other chronic sources ofpollution. Had the same spill occurred in aharbor, or on a coastal reef adjacent to apopulated area, the long-term impact wouldhave been much greater, especially due tosynergisms and continued interference withrecovery and recruitment. As far as effectsof tropical storms and typhoons, while theoverall appearance could be described ascatastrophic, such events usually crop thereef rather than completely kill the corals.One of my primary collecting sites for stud-ies of coral reproduction is the windward,exposed reef behind the University of GuamMarine Laboratory. During the past sevenyears, there have been at least eight majortyphoons, during which wave wash hasreached as high as 33 feet above sea level.There has not been a single year that I havenot been able to collect a variety of gravidcorals from this site.

When defending projects or activitiespotentially harmful to coral reefs, paid con-sultants often argue that if there are no datathat prove the proposed activity is detri-


mental, the project or activity should beallowed. This sets the stage for classical typeII statistical error: accepting a false hypoth-esis. Simply put: the absence of data show-ing harm often indicates a lack of data ratherthan no effect. When adequate and accuratedata demonstrate no detrimental effects,then and only then should projects beapproved.

Data summarized in this paper supportthe hypothesis that most chronic distur-bances are more damaging than acute dis-turbances, especially when considering coralreef recovery. Anthropogenic disturbances,like runoff, sedimentation, sewage outfalls,and oil pollution are characteristicallychronic perturbances. They generally causeproblems not only by inducing coral mor-tality, but by affecting reproduction andrecruitment, and hence, recovery. If suchproblems can be controlled, limiting themto more episodic and acute disturbances, orremoving the stress altogether, reefs can, andwill recover. Perhaps that is the message ofvalue we can extract from impact studies.Corals in Kaneohe Bay, Oahu, Hawaii, didrecover from both fresh water kills andeutrophication when sewage input wasdiverted elsewhere (Evans et al., 1986; Hol-thus et al., 1986; Jokiel et al., 1993).

Devastating events normally consideredas natural, like Acanthaster outbreaks(Endean, 1973) and red tides (Guzman etal., 1990), have been found to have links tohuman activities increasing runoff andeutrophication (Birkeland, 1982). Recoveryof Jamaican reefs following Hurricane Allen(Woodley et al., 1981) appears to have beenimpaired by anthropogenic impacts of over-fishing herbivorous fishes and terrigenousrunoff coincident with the loss of grazingurchins by disease. This further emphasizesthe types of synergistic interactions that canoccur, affecting the extent of mortality aswell as the possibilities for recovery (Pear-son, 1981).

One solution which would drasticallyreduce the amount of anthropogenic dis-turbance on coastal coral reefs is to changethe common engineering practice of usingthe coastal zone as a dumping ground forstorm drainage, runoff and sewage.Whiletemperate marine environments and fish-

eries thrive on nutrient input, coral reefssuffer whenever water clarity goes down and/or nutrient levels go up. Better erosion con-trol standards, increased retention of fresh-water on land, and diversion of sewage toareas which carry the material away fromreefs are all solutions which can readily beapplied.

Globally, coral reef preserves are criti-cally needed to serve as refuges for coralsand other reef organisms (Buddemeier,1993). Wilkinson (1993) estimates 70% ofthe world's coral reefs are already seriouslydegraded (10%), in a critical state of beinglost within the next 10-20 years (30%), orthreatened to disappear within the next 20-40 years (30%), leaving an estimated 30%as stable, and capable of surviving fromhundreds to thousands of years. More dataand studies, especially in forms and forumsavailable and comprehensible to the generalpublic and decision makers are needed ifpresent trends are to be reversed.

In conclusion, acute, natural disturbancesare critical to maintenance of diversity onreefs (Connell, 1978), and in the case oftropical storms and typhoons, may actuallyserve to reduce anthropogenic disturbanceby removing accumulated sediments depos-ited by erosion and sedimentation. Char-acteristically chronic, anthropogenic distur-bance, while often sublethal, and hence,more difficult to assess on the short term,can cause more serious damage by pre-venting recovery while acting to weakencorals and other reef organisms to the pointof eventual mortality. We cannot controlnature, but with adequate and accurate data,can make decisions which control theimpacts of man's activities on coral reefcommunities.

ACKNOWLEDGMENTS

The author gratefully acknowledges Dr.M. G. Hadfield for the invitation to partic-ipate in this symposium, and for his manyhelpful comments and Drs. C. E. Birkeland,P. W. Glynn, P. L. Jokiel, R. A. Kinzie, III,D.A. Krupp, and K. Yamazato for numer-ous discussions on the subject matter of thispaper and critical comments. Support forresearch included in this paper was pro-vided by NIH/MBRS Grant #53-LA-61158-

coral reefs: present problem ands future concerns resulting from

Documents