Water Quality in the Negril Area Watershed,
Maya Goreau & Thomas J. Goreau
Global Coral Reef Alliance
NOTE: This report presents detailed water quality measurements during 1996 from every coral reef, every mangrove, every wetland, every river, and every spring in the Negril Area Watershed at the western end of Jamaica. It may be the most thorough set of water quality measurements made throughout the fresh and salt waters of an entire tropical watershed and its coastal zone. By use of the same methods in both land and sea habitats, it has been possible to identify the sources of nutrients throughout the area and make specific proposals regarding the measures need to reduce them and restore the health of aquatic ecosystems. Sharp changes in the management of sewage and in land use will be needed to halt or reverse the severe environmental deterioration underway. An abstract of this report has been published in the Proceedings of the Association of Marine Laboratories of the Caribbean, San Jose, Costa Rica, in 1997. It will be published together with the results of a similar study made at the opposite end of the island.
Anthropogenic and natural impacts are causing deterioration of Caribbean coral reefs. Health of coral reefs and water quality were assessed in the Negril watershed of western Jamaica, and compared to reefs in Antigua, Barbuda, and Panama. The watershed's entire coastal zone, and all major rivers and groundwater springs were analyzed for eight water quality parameters: nitrate-N, ammonia-N, phosphate-P, chlorophyll, temperature, pH, salinity, and dissolved oxygen. Negril's rivers and springs contained low salinity and high nutrients, except where healthy wetlands and tidal mangroves absorbed nutrients. All coastal waters had nutrient and chlorophyll concentrations exceeding thresholds for healthy coral reefs. Negril area reefs had low coral cover and were smothered by eutrophic algae. Other Caribbean reefs were healthier, due to lower human population stresses. To preserve Negril's remaining reefs, water quality must be improved by preventing nutrient discharge into the coastal zone, treating all sewage to the tertiary level, and protecting mangroves and forest nutrient sinks from development.
INTRODUCTION AND REVIEW OF LITERATURE
Coral reefs throughout the Caribbean are undergoing increasing deterioration. A wide variety of anthropogenic and natural causes has been blamed, but verification of the major causes has been limited by a lack of sufficient data. One of the major problems, eutrophication or overgrowth of corals by fleshy algae, has been attributed to excessive seawater nutrients by some scientists (Bell, 1992; Lapointe, Littler, and Littler, 1993; Goreau and Thacker, 1994; Bell & Elmetri, 1995). Massive nutrient loading in coastal waters is thought to stimulate the large blooms of eutrophic algae which smother corals (Dubinsky & Stambler, 1996). Studies establish maximum nutrient threshold concentrations for healthy reefs at .003 ppm phosphorous, and .014 ppm nitrogen (Bell, 1992; Lapointe, 1992)- trace amounts far below the standards for drinking water. Other scientists attribute eutrophication to a decline in herbivory due to overfishing and an increased prevalence of marine pathogens (Hughes, 1995). Uncertainty over the major factors causing reef deterioration has precluded vital steps in reef management and conservation which must be implemented quickly if healthy coral reef ecosystems are to survive.
Within the past few decades, Jamaican reefs have undergone a sharp decline in the percentage live benthic coral cover coupled with eutrophication- a dramatic increase in the percentage algae cover (Goreau, 1992a; Hughes 1995). Algal overgrowth became a major problem in the Negril area watershed in 1990, long after it had devastated more densely populated or earlier developed resort areas in other parts of Jamaica (Goreau, 1991, 1992b,c). In Long Bay, next to the Negril tourist resorts, high concentrations of nitrogen and phosphorous were measured in both algal tissues (Lapointe, 1992; Goreau, 1992b) and in ambient seawater (Wade, 1992).
These high nutrient concentrations could be due to several factors. The Negril area lacks complete sewage treatment to the tertiary level. Though a sewage system is being installed in the developed resort areas of the city of Negril, current plans lack tertiary nutrient removal mechanisms and will release secondarily treated effluents into the Negril River, which empties into Long Bay. It is uncertain whether this sewage treatment plant will release smaller total amounts of nutrients into the bay than already enter the sea via the groundwater and runoff, but it could focus the nutrient discharges into a smaller area. Furthermore, excessive concentrations will remain unreduced in the rest of the Negril area watershed, which will not be connected to the new sewage system. The location of Negril at the extreme downcurrent western end of the island could result in nutrient inputs from upcurrent shorelines. The restoration of water quality in Negril's reefs is highly dependent on the identification and reduction of all of the other sources of nutrients as well as those from the resort zone.
Thousands of tourists are large contributors to the problem of excessive nutrients, as most hotels do not treat their sewage adequately before discharging it into the groundwater, the sea, rivers, or wetlands. The majority of the watershed's population lives in the hills, isolated from Negril's under-construction sewage treatment plant. Those who do have toilets invariably use "soak-aways", in which sewage leaches into the surrounding limestone ground, and subsequently finds its way to sea via subsurface flow through caves and cracks. Many people lack toilets, and dispose of their sewage into rivers and sinkholes. It then enters the groundwater, which carries the dissolved nutrients from decomposed sewage downhill and out to sea. People without access to running water take baths and do laundry in rivers and sinkholes. Consequently, large quantities of phosphate-rich detergent and nitrogen-rich sewage find their way into the sea.
Humans also cause extensive physical damage to coral reefs through other mechanisms, such as dropping anchors on fragile corals. During the construction of hotels, nutrient-absorbing mangroves and wetlands are razed, drained, and covered with limestone marl; reef material is dredged and dumped on shore to create new beaches; sediments from construction wash into the sea, blanketing live corals, and making it impossible for young corals to settle. Global warming in the past decade has brought about a new phenomenon known as coral bleaching. In the warmest month of the year, a temperature increase of as little as 1 degree Celsius above average causes the coral animals to expel their zooxanthellae- unicellular algae which symbiotically provide corals with food (Goreau et al., 1993; Goreau and Hayes, 1994). After losing their algal pigments, corals turn pale, becoming "bleached" white. Bleached corals can recover slowly if temperatures do not remain elevated for prolonged periods of time. However, while they are bleached, corals starve and are less resistant to damage or mortality from weedy algal overgrowth or excess sediments.
The preservation of coral reefs is economically vital to Caribbean countries like Jamaica whose foreign exchange earnings are driven by tourism, which is largely based on the beauty of the reef and beaches which are entirely built up by the remains of reef organisms. Sick reef ecosystems are incapable of supporting large marine fish and shellfish populations or humans who depend on reefs for food. Many beaches are being eroded because natural reef breakwaters cannot grow quickly enough to compensate for hurricane damage or rising sea level.
STATEMENT OF PURPOSE
Water quality monitoring to date has largely been confined to parameters measured periodically by the National Water Commission and the Urban Development Corporation at approximately 15 sites directly off the main resort strip of Negril (Wade, 1992), and 4 sites in the Negril Morass (Enell, 1984). Little quantitative information is available on nutrients entering from other coastal or land areas throughout the watershed (Goreau, 1993, 1994). Such information is essential for successful whole watershed environmental management (Goreau et al., 1996). Without identification and regulation of these pollution sources, the health of coral reefs in marine parks and protected areas could be affected by sources which lie outside their boundaries (Goreau et al., 1995). This study was a detailed survey in all major marine and freshwater habitats within the watershed in order to document the conditions of water quality prior to the establishment of the sewage system. Its goals were to establish a baseline for the identification of future changes, to ascertain whether water quality throughout the Negril watershed was adequate for healthy reef growth, and to identify major sources of water quality deterioration. The results should provide critical data for the environmental management steps which must be taken to improve the water quality, environmental health, and economic value of Negril's waters.
MATERIALS AND METHODS
A water quality profile for the entire Negril area watershed was constructed from five separate sampling runs during July 1996. Twenty seven half-liter seawater samples were collected by boat along the northern half of the coastline and twenty seven along the southern coastline. Rivers and groundwater springs were sampled from land in a group of six samples from the southern quarter, and twenty seven samples from the upper three quarters of the watershed. The coastline around the town and tourist areas of Negril was represented separately by fifteen seawater and brackish water samples, at sites which have been monitored by previous surveys. The water samples were taken in coral reefs, mangroves, seagrass beds, wetlands, forest aquifers, springs, caves, and other water supplies used for human consumption- both treated and untreated. These sites are shown in the maps.
In order to prevent contamination, bottles and lids were rinsed three times with the water to be sampled. The bottles were then inverted, plunged vertically into the water, and filled approximately six inches below the surface. This minimized collection of surface water which might be uncharacteristic because of surface films. The bottles were capped underwater, leaving no air spaces which could distort dissolved oxygen readings, and brought to land to analyze. Water temperatures were measured in situ. Dissolved oxygen (DO), salinity, pH, nitrate-N, phosphate-P, Ammonia-N, and chlorophyll were measured on land within a few hours of collection.
Marine nutrients were measured colorimetrically, using Hach reagents and a DR2000 Spectrophotometer (Hach, 1992). Nutrients were measured in parts per million (ppm) of nitrogen or phosphorous. Nitrate-N was measured by the Cadmium Reduction method. Ammonia-N in was measured with the Salicylate method. The Ascorbic Acid method was used to measure Phosphate-P. Chlorophyll concentration was measured using a WETStar fluorometer, which excites the pigments with blue light and measures the amount of fluorescent red light emitted by chlorophyll, as the voltage output of a photomultiplier, converted to equivalent concentrations in micrograms per liter, or parts per billion (ppb), using the manufacturer's suggested calibration (WET Labs, 1996). Temperature in degrees Celsius was read off a calibrated mercury thermometer to the nearest tenth of a degree. Salinity as conductivity converted to parts per thousand (ppt), pH (in pH units), and dissolved oxygen (in percent of saturation with respect to the atmosphere) were measured with an Aqua Check water analyzing probe (Aqua Check, 1994).
Replicability to estimate sampling and analytical precision of measurements was determined by analysis of three separate samples taken at one site. This site was the most remote ofshore site analyzed, and had the lowest concentration of nutrients. The mean and standard deviation of each parameter are shown in the table below:
Parameter Mean Standard deviation
temperature: 29.0 degrees C + 0.000
dissolved oxygen: 98 % saturation + 0.000
pH: 8.24 + 0.010
phosphate: .00767 ppm + 0.005
nitrate: .03 ppm + 0.000
ammonia: .0167 ppm + 0.0153
salinity: 36.08 ppt + 0.000
chlorophyll: 1.553 ppb + 0.092
Underwater ecology was observed in as broad an area as possible around sites by rapidly swimming along stress gradients and across transitions between habitats. The traditional approach of recording the species Iying along transect lines was not used because it results in the detailed observation of a too small and frequently unrepresentative area for the time invested. The composition of organisms and substrates covering the bottom was approximated visually to the nearest 10 percent, and expressed in percentage live coral cover, percentage bleached coral, percentage dead coral or rubble, percentage fleshy algae cover, and percentage calcareous algae. Species of algae and corals present were identified, noting which were dominant.
Previous water quality monitoring was largely confined to resort areas off of the city of Negril. As fewer than 20 of the 101 sites studied here had been previously examined for water quality, this data represents a considerable increase in quantitative information for the entire watershed. Spearman Rank Order Correlation cefficients indicate a statistically significant direct relationship between pH and salinity, and statistically significant inverse relationships between all four nutrients measured and both salinity and pH. This correlation indicates that excess nutrients enter the sea via fresh water. All relationships between parameters are shown in figures 1a and 1b. The data obtained are shown in maps 2 through 9, and in Figures 2 through 9. Distinct patterns seen in each environmental variable are discussed below.
1) Salinity (Figure 2)
Salinity remained fairly constant in marine sampling sites, at the average concentrations for seawater. Lower salinity values were observed in fresh and brackish waters, with increasing salinity occurring with decreasing distance from the sea. Salt mixing was observed in the Great Morass along the length of the watershed's two major rivers, the North and South Negril rivers, suggesting significant tidal exchange of river and sea waters. 53 of 101 sites had salinities above 36 parts per thousand, representing marine coastal waters. 24 samples, with salininties less than 2 ppt were freshwater samples, The remaining 24 brackish water samples with salinities between 2 and 36 ppt fairly evenly covered the entire range between fully fresh and fully marine conditions (Figure 2).
2) Temperature (Figure 3)
Sea surface temperatures measured throughout the watershed hovered at, or above, the Jamaican coral bleaching temperature threshold of 29.6 degrees Celsius (Goreau et al., 1993). Temperatures were highest in inshore areas due to higher turbidity and poor water circulation. Rivers and groundwater springs and sinkholes tended to be cooler. Fresh and brackish water temperatures increased as distance from the sea decreased, due to mixing with the warmer seawater. 41 samples had temperatures below 29.5 degrees celsius. 60 samples were warmer than this coral bleaching threshold temperature, with 26 of them between 29.5 and 30.0. These temperatures were measured very close to the warmest time of year, but they did not stay long at this level and soon fell. Satellite sea surface temperature records for the Negril area show that temperatures briefly reached the threshold but fell quickly (Goreau, Hayes, & Strong, 1997). A mild bleaching event was observed to take place in Negril at this time, with only the most sensitive species being affected. New bleaching, affecting only 10 to 20% of the corals, and hardly noticeable to casual observers, was seen to develop in highly sensitive but rapidly recovering species such as Millepora complanata, the rapidly affected and slowly recovering species like Palythoa caribbeorum, and patchy new paling was seen in more slowly affected species like Diploria strigosa, Montastrea annularis, and Meandrina meandrites. Montastrea cavernosa and Siderastrea siderea also showed new bleaching but this was masked in the many coral heads of those species which had not yet recovered from bleaching the previous year's severe Caribbean-wide bleaching event.
3) pH (Figure 4)
Fresher waters had lower pH than seawater at the sites sampled, probably due to high levels of carbon dioxide in groundwater and soils, derived from respiration of roots and decomposition of organic matter. Forty-eight seawater samples had pH levels above 8.4. A further 21 sites had sufficient seawater admixture to be buffered by it to pH values between 8.2 and 8.4. The remaining freshwater and dilute brackish samples were far more acidic, with 31 falling between pH 7.1 and 8.2. The most acidic sample of all, pH 6.6, was found in a shallow flooded area of wetland in the Royal Palm Reserve, overlying a peat soil. Due to stagnant conditions, this site was highly enriched in dissloved humic compounds and other products of organic decomposition, such as high acidity.
4) Dissolved Oxygen (Figure 5)
Percent oxygen saturation was satisfactory at all sampling sites because all samples were taken from surface waters, which were oxygenated due to wave action or mixing. Values would probably decrease with depth in several fresh and brackish water sites with strong temperature or salinity stratification. No samples were found to be supersaturated in oxygen, which would happen in locations where photosynthesis exceeds respiration. This indicates that all the waters measured showed aquatic habitats in which the decomposition of organic matter exceeds its production and its rate of diffusion from the atmosphere. Therefore they are being supplied with excess decomposable organic matter from the land. Thirty-six samples had oxygen saturation between 98% and 100%. Forty had values between 94% and 98%, and 24 fell between 92% and 94%. No anoxic waters were sampled since all sampling was near the surface where oxygen can enter from the atmosphere, but lower values would have been found deeper in the brackish water sites, which were strongly vertically stratified, with stagnant seawater trapped below warm fresh and brackish surface waters. This situation developed after the river was dredged in 1960, which allowed seawater to penetrate up the length of the dredged channels.
5) Phosphate (Figure 6)
Phosphorous concentrations were typically higher in fresh and brackish water samples than in open ocean. Phosphate contributions alone were below the reef eutrophication threshold of .003 ppm for total phosphorous in 5 of 101 sites. Concentrations exceeded the limits at 96 locations. 29 samples had concentrations less than three times the limit, or 0.01 ppm. Another 58 were between threefold and 15 fold excess, 0.05 ppm, and 13 lay between 15 fold excess and 40 fold excess. The highest sites are likely to have been recently contaminated with sewage or detergents. The actual excess of available phosphorous is certain to be even higher than that of phosphate alone, since the dissolved organic phosphorous content was not measured. However the more frequent nitrogen excess measured in samples suggests that phosphorous is the major limiting nutrient, in accord with the results obtained from analyzing the nitrogen, phosphorous, and carbon contents of algae in Negril (Lapointe, 1992), making these coastal ecosystems highly sensitive to small inputs of phosphorous from sewage or detergents.
6) Nitrate (Figure7)
The highest nitrate concentrations measured were in groundwater springs and in fresh and brackish river waters. Marine nitrate concentrations exceeded reef eutrophication threshold values for total nitrogen in virtually all sites. Only 1 out of 101 sites had values of nitrate-N plus ammonia-N below the reef eutrophication limit of .014 ppm. 79 sites had less than 10 fold excess above the eutrophication limits, 20 were between 10 and 50 times too high for corals, and 22 were over 50 times too high. The highest value was found at the Whitehall well, which is thought to contain sewage contaminated groundwaters. Unfortunately this well is widely used as a water supply in the surrounding community.
7) Ammonia (Figure 8)
Ammonia values were generally low compared to nitrate because this form of nitrogen is very quickly used by marine and aquatic plants. Many marine samples along the southern coast had undetectable ammonia concentrations, while most of the northern coast marine samples were near the limit of detetection, geting notably higher neashore and in bays. Approximately a third of the samples had ammonia levels which, in the absence of any nitrate, would be below the critical eutrophication limits for total nitrogen, about a third were between one and and three times the limit, and the rest above this. However these values, added to the nitrate-N values in the same samples exceeded the total nitrogen eutrophication limits in all but one site.
8) Chlorophyll (Figure 9)
The chlorophyll concentrations equaled or exceeded the critical limit for coral reefs of .5 parts per billion (ppb) at 99 of the 101 sites. 2 samples were below the critical concentration limit, 82 of the samples had between 1 and 10 times the critical value, and 16 were above this. The highest values were found in stagnant water in wetland ponds. The two sites which were below the limit were sinkholes in caves which received very little light, but had high nutrient concentrations. Chlorophyll levels were highest in fresh and brackish waters, and decreased as distance to the sea decreased in a similar manner as the nutrients.
9) Ecological observations
Benthic composition was almost entirely weedy green and red algae (the species present were indicative of eutrophication), turf algae, or rubble. Coral cover in coastal waters was extremely low, ranging from less than 1 percent to 25 percent. Moderate levels of coral bleaching were observed, which became more pronounced with prolonged elevated seawater temperature during the course of the study. The waters were usually green along the coast, ranging from blue-green to turbid olive. Water color underwent a spatial gradation, becoming pure oceanic blue several miles offshore. Dark river water with suspended peat formed plumes which extended into the sea, resulting in localized higher water temperatures due to sunlight absorption. The impacts of several coral reef pathogens were observed, including CLD (Caribbean Coralline Lethal Disease) a new disease similar to CLOD which was only recently discovered in the Pacific (Littler and Littler, 1995). The disease, which attacks calcified crustose coralline algae, was observed for the first time in the Caribbean (Goreau et al., in press). Fresh and brackish water frequently supported a relatively large biomass of aquatic plants. Waters adjacent to intact mangroves, forests, and wetlands frequently contained reduced nutrient levels compared to their upstream sources, as nutrients were absorbed or filtered out by aquatic and terrestrial vegetation before entering, the sea. This was a natural form of biological treatment of effluent.
A) INTERACTIONS BETWEEN NEGRIL WATERSHED WATER QUALITY PARAMETERS
1 ) POSITIVE INTERACTIONS
All measured water quality parameters were compared with each other, using non-parametric rank order statistics, to determine which variables significantly correlated with one another (Figure 1a and 1b). Only phosphate levels were found not to be significantly positively correlated with any other variable. Very strongly positive correlations between variables, with a probability of over 99.9%, were found between chlorophyll and ammonia, and between temperature and salinity, temperature and pH, and salinity and pH. Strongly positive correlations, with probability greater than 99%, was found between ammonia and nitrate, and ammonia and dissolved oxygen. A positive correlation, with probability greater than 95%, was found between salinity and dissolved oxygen. The lack of positive correlations between chlorophyll and any other variable except ammonia indicates that ammonia, despite its relatively low levels, is critical in driving growth of phytoplankton in the water.
2) NEGATIVE INTERACTIONS
Very strongly negative correlations (above 99.9% probability) were found between salinity and ammonia, between salinity and chlorophyll, between pH and ammonium, between pH and chlorophyll, and betwen nitrate and temperature. Strongly negative correlations (above 99% probability) were found between salinity and phosphate, and between ammonia and temperature. Negative correlations (above 95% probability) were found between dissolved oxygen and phosphate, between phosphate and pH, between nitrate and salinity, nitrate and pH, and nitrate and chlorophyll. These results indicate that low salinity freshwater is highly associated with high ammonium and chlorophyll levels, and less so with regard to nitrate and phosphate. The significant negative relationship between nitrate and chlorophyll is somewhat surprising, and probably indicates that nitrate is present in excess with regard to other nutrients.
B) WATER QUALITY WITH REGARD TO NEGRIL REEFS
Corals grow best in clean coastal water, almost devoid of nutrients. Marine sampling sites contained phosphate concentrations up to 40 times higher than the established nutrient thresholds for coral reefs to survive without being overgrown by algae. Marine nitrate and ammonia concentrations were up to 14.3 times too high, and chlorophyll was up to 10.3 times too high in marine waters. inland rivers, groundwater, sinkhole, and submarine spring measurements were generally much higher than coastal marine measurements. These fresher waters contained up to 174.3 times the reef threshold limit for nitrogen, 78.3 times the threshold for chlorophyll, and 39 times the critical value for phosphorous. These concentrations decreased as samples were collected progressively from upstream rivers and spring sources outwards into the open ocean. The salinity values obtained showed that there was a high amount of mixing and exchange between fresh water and seawater in tidal rivers in the Great Morass. High nutrient concentrations in fresh and brackish water frequently were found accompanied by low salinity and pH. This suggests that fresh water containing land-generated nutrients acts as the major source of nutrients fertilizing the marine environment.
Coastal nutrient loading appeared to originate primarily from anthropogenic sources. The lack of nutrient-reducing tertiary sewage treatment coupled with poor garbage collection services leads many people within the Negril watershed to dispose of sewage and garbage in sinkholes, rivers, and the sea. Runoff contributes additional nutrients and sediments from agricultural fertilizers and pesticides, untreated sewage, and construction sites. Ultimately, these practices result in the contamination of groundwater, springs, rivers, wetlands- all of which flow into the sea. All coastal marine sites observed were eutrophic and infested with weedy green algae. Samples taken offshore from the dried up Orange River and the Orange Bay garbage dump, which had been closed for several years, showed high nutrient concentrations. Nutrients which had accumulated in the soil appeared to be leaching out into the sea. However, high nutrient and chlorophyll levels were found even off of undeveloped shores, indicating that the region's position downcurrent from more developed areas to the east (along both the north and south coasts) is contributing to eutrophication of the entire Negril area. This will greatly restrict any benefits from local sewage treatment unless it is extended to upcurrent areas as well as to the resort zone.
Nearly all Negril coastal sampling sites were characterized by poor water visibility, due to resuspension of bottom sediments, coastal erosion, and soil and peat which enter the sea via runoff and rivers. The widespread deforestation on hillsides and along shorelines decreases the amount of rainwater absorbed into the ground, and increases surface runoff and the erosion of soil and nutrients. The sediments quickly wash into the coastal waters, limiting the recruitment of young corals, which can anchor themselves only to clean limestone substrates. Excessive suspended solids in the seawater have blanketed and smothered corals, and prevented light from reaching them at sites downcurrent from river mouths. Under such conditions, it is difficult for coral polyps to gather food, and for their symbiotic algae to photosynthesize, and they become more prone to weedy algal overgrowth. This was also a problem even near the northern and southern limits of the watershed, including the Green Island, Orange Bay, Samuels Bay, Little Bay, and Salmon Point areas. Additionally, the destruction of aquatic vegetation ecosystems by draining and infilling prevents natural terrestrial effluent filtration. Nutrient-loaded sewage and runoff is filtered as it trickles through densely vegetated regions. Suspended solids and nutrients are removed by plants before they reach the sea. Control is needed over the destruction of wetlands. This reduces their capacity to act as nutrient and soil absorbers, and can even turn them into nutrient sources. A release of nutrients occurs when the vegetation is cut or if the peat is allowed to drain, decompose, and erode.
C) COMPARISON WITH OTHER CARIBBEAN REEFS
In the eastern Caribbean islands of Antigua and Barbuda, which also lack adequate sewage treatment and garbage disposal, nutrient-associated problems were far less prevalent. Though preliminary signs of eutrophication were evident in the algal species composition, nutrient concentrations were lower than any measured in the Negril watershed in Jamaica. Calcareous sand-producing algae dominated the population, and the growth of weedy algae was usually very minor. Nutrients in these areas could scarcely be measured by the methods used in Jamaica (Goreau & Goreau, 1996), and even waters off the major harbors, towns, and hotels in Antigua contained undetectable levels of nutrients. The chief differences between Jamaica and Antigua/Barbuda are the latter's lower population density and greatly reduced levels of rain and freshwater runoff. Though the same detrimental human practices occur in both countries, the coastal population of Antigua and Barbuda is small enough that the surrounding reef ecosystems are generally not eutrophic.
Coral reefs along the Caribbean coast of Panama were typically plagued by erosional and agricultural sedimentation, and the resuspension of marine sediments. Turbidity was high and marine visibility was poor by Jamaican standards. However, all the observed Panamanian reefs had higher percentages of live coral cover and less weedy algae than the coastal reefs of the Negril area watershed (Goreau & Goreau, 1996). Though the coral population was dominated by the most sediment-resistant species, the overall health and biodiversity far surpassed that of Jamaican reefs. In recent analyses, nutrient concentrations were scarcely detectable (Goreau 1997). Both fish and sea urchins were abundant in Barbuda and scarce in Antigua. Fish were abundant in Panama, but sea urchins were very rare. In Jamaica, fish were scarce, but sea urchins were abundant at some sites. The coincidence of eutrophic sites with nutrient sources, and lack of inverse association with herbivores supports the theory that the most influential factors for the health of Caribbean coral reefs lie in excessive nutrient loading in coastal waters rather than a decline in herbivory. Sewage disposal practices differ little between these countries. The advanced state of deterioration of Jamaican reefs is probably a consequence of the higher level of stresses resulting from the considerably higher population density and overdevelopment of the coastal zone in Jamaica compared to Antigua, Barbuda, or Panama.
1 ) CORAL REEF CONSERVATION IMPLICATIONS
Even the better coral reef areas observed could not be considered truly healthy ecosystems, because algal cover greatly predominated over live coral, and the reefs were no longer growing wave-resistant limestone structures which provide suitable habitat for other organisms and safeguard against coastal erosion. Instead, small communities of marine organisms were found centered around isolated coral colonies. The high nutrient and chlorophyll concentrations found indicate that the current land-based nutrient sources to the coastal zone pose a tremendous threat to the livelihood of Negril's coral reef ecosystems, and the humans who depend on them. For the reefs to survive, essentially all current nutrient sources must be prevented from entering the sea.
2) SEWAGE TREATMENT IMPLICATIONS
The greatly excessive levels of nitrogen, phosphorous, and chlorophyll in Negril area coastal waters and their source in freshwater inputs, imply that considerable reduction of nutrients in rivers and groundwaters is needed before coral reefs can recover. The high values measured in all coastal areas indicates that there are many widespread sources of nutrients, besides those from the area to become part of the sewage collection scheme. The data also suggest that all significant nutrient sources of human origin need to be reduced to nearly zero levels because the background levels in the least contaminated freshwater source areas are already excessive, especially with regard to nitrate. The reductions which would result from Negril's resort area sewage plant alone will be incapable of reducing the already excessive levels of nutrients now entering the sea via rivers which have their origin in other parts of the watershed outside the plant's collection area.
If water quality throughout the entire watershed is not improved, beneficial effects of sewage treatment for the resort areas will be very limited. Adequate reduction of nutrients to improve coastal ecosystem health can only be achieved through a combination of improved sewage treatment to the tertiary level and better land management to prevent the loss of nutrients and soil from terrestrial and agricultural ecosystems. The extent of deterioration of reefs in all parts of the Negril area coastline indicates that water quality improvement must focus on areas outside the resort area as well as within it.
3) WATERSHED MANAGEMENT IMPLICATIONS
Erosion rates must be reduced by preventing further deforestation of watershed recharge areas and protecting wetlands from being cleared, drained, and filled with dredged sediment or marl to prepare for construction. Planting the banks of water courses with a tree buffer zone on both sides would also trap excess sediments and nutrients before they can erode into the ocean. Protection of all wetlands and mangroves is needed to ensure that they continue to serve as nutrient absorbers. Once they are damaged, the vegetation and peat decomposes, and they turn into sources of nutrients to the water instead. Healthy, intact aquatic vegetation communities such as mangroves and wetlands can be used as nutrient-filtering biological tertiary sewage treatment. These areas are capable of safely absorbing controlled effluents because the excess nutrients would fertilize terrestrial plant life, rather than induce marine algal blooms. Long term monitoring of the entire watershed beyond the resort area is needed to determine if water quality is being effectively improved through better environmental management.
Costs of travel and equipment were provided from a Pew Fellowship in Environment and Conservation to T. Goreau and a grant from the European Union to the Negril Coral Reef Preservation Society. This research would have been impossible without the support and valuable local observations from Katy Thacker, the Negril Marine Park Rangers, Karen McCarthy, and the Negril Coral Reef Preservation Society; Louis Daley and the Negril Environmental Protection Trust; Foster and Kim Derrick, Ashton Williams, John and Era Birk and the Environmental Awareness Group of Antigua; Gabriel Despaigne, Angel Tribaldos, Richard Peralta and the Asociacion Oceanica de Panama; Angel Gonzalez-Diaz and Pro Mar, and the Comisones Ecoturismo y Medio Ambiente of Playon Chico and Carti Yantupu, and Carlos Alberto Arango and family for generous hospitality.
Aqua Check, 1994, Water Analyzer Technical Manual, 26p., Perstorp Analytical, Wilsonville, Orecon
Bell, P., 1992, Eutrophication and coral reefs - some examples in the Great Barrier Reef Lagoon, Water Research, 26: 553-568
Bell, P., & I. Elmetri, 1995, Ecological indicators of large-scale eutrophication in the Great Barrier Reef Lagoon, Ambio, 24: 208-215
Dubinsky, Z., & N. Stambler, 1996, Marine pollution and coral reefs, Global Change Biology, 2: 511 -526
Enell, M., 1984, Water chemistry of the Negril and Black River Morasses, Jamaica, 166p., Petroleum Corporation of Jamaica, Kingston, Jamaica & Institute of Limnology, University of Lund, Sweden
Goreau, M., T. Goreau, & F. Derrick,1996, Ecological assessment of Antigua and Barbuda Reefs, Global Coral Reef Alliance, Camabridge, MA & Environmental Awareness Group, St. Johns, Antigua
Goreau, T., 1991, Coral reef health in the Negril area: survey and recommendations, p. 32-70 in K. Thacker (ed.), Protecting Jamaica's coral reefs: final report of the Negril reef mooring buoy workshop and installation project, Negril Coral Reef Preservation Society, Negril, Jamaica and Reef Relief, Key West, Florida
Goreau, T. 1992a, Bleaching and reef community change in Jamaica: 1951-1991, American Zoologist 32: 683-695
Goreau, T., 1992b, Negril: environmental threats and recommended actions, p. 9-37 in K. Thacker (ed.), Protecting Jamaica's coral reefs: water quality issues, Negril Coral Reef Preservation Society, Negril, Jamaica
Goreau. T., 1992c, Coral reef protection and coastal development in Western Jamaica, p. 39-65 in K. Thacker (ed.), Protecting Jamaica's coral reefs: water quality issues, Negril Coral Reef Preservation Society, Negril, Jamaica
Goreau, T., 1994, The Negril Environmental Protection Area: habitats, environmental problems, and recommended actions, in K. Thacker (ed.), Protecting Jamaica's coral reefs: the Negril Environmental Protection Area, Negril Coral Reef Preservation Society, Negril, Jamaica
Goreau, T., 1995, Negril Environmental Protected Area, prospects and progress in 1995, in K. Thacker (ed.), Protecting Jamaica's coral reefs: planning the Negril Environmental Protected Area, Negril Coral Reef Preservation Society, Negril, Jamaica
Goreau, T., 1997, Water quality and ecological assessment of Caribbean coral reefs in Panama, Global Coral Reef Alliance, Cambridge, MA
Goreau, T., J. Brown, E. Gordon, G. Gordon, V. Guthrie, N. Ruddock, R. Evans, & K. Thacker, 1995, Negril & Green Island area environmental protection plan, 63 p., Negril Environmental Protection Trust, Negril, Jamaica & Natural Resources Conservation Authority, Kingston, Jamaica
Goreau, T., L. Daley, S. Ciappara, J. Brown, S. Bourke, & K. Thacker, 1996, Community-based whole-watershed and coastal zone management in Jamaica, Proceedings of the 8th International Coral Reef Symposium, Panama
Goreau, T., M. Goreau, & W. Hilbertz, 1996, Ecological status, management, and restoration needs of Caribbean coral reefs in Panama, Global Coral Reef Alliance, Cambridge, MA
Goreau, T. & R. Hayes, Coral bleaching and ocean "hot spots", Ambio, 23: 176-180
Goreau, T., R. Hayes, J. Clark, D. Basta, & C. Robertson, 1993, Elevated sea surface temperatures correlate with Caribbean coral reef bleaching, p. 225-255 in R. Geyer (ed.), A global warming forum: scientific, economic, and legal overview, CRC Press, Boca Raton, Florida
Goreau, T., & K. Thacker, 1994, Coral reefs, sewage, and water quality standards, Proceedings of the Caribbean Water & Wastewater Conference, Kingston, Jamaica
Goreau, T et al, in press, Rapid spread of diseases in Caribbean Coral Reefs, proceedings of the Association of Marine Labs of the Caribbean 28th reunion.
Hach Company, 1992, Hach water analysis handbook, Hach Co., Loveland, Colorado
Hughes, T., 1995, Catastrophes, phase shifts, and large scale degradation of a Caribbean coral reef, Science, 265: 1547-1551
Lapointe, B., 1992, Eutrophication thresholds for macroalgal overgrowth of coral reefs, p. 105-112 in K. Thacker (ed.), Protecting Jamaica's coral reefs: water quality issues, Negril Coral Reef Preservation Society, Negril, Jamaica
Lapointe, B., M. Littler, & D. Littler, 1993, Modification of benthic community structure by natural eutrophication: the Belize barrier reef, p. 317- 326 in Proceedings of the 7th International Coral Reef Symposium, Guam
Littler, M., & D. Littler, 1995, Impact of CLOD pathogen on Pacific coral reefs, Science, 267: 1356-1360
Wade, B., 1992, Nutrient levels in Negril wetland and coastal waters, p. 91-104 in K. Thacker (ed.), Protecting Jamaica's coral reefs: water quality issues, Negril Coral Reef Preservation Society, Negril, Jamaica
Wet Labs, 1996, WETStar miniature fluormeter user's quide, Philomath, Orenon