USVI Public Records

Document text

GEOCHEMISTRY OF ST. JOHN AND INFLUENCE ON MARINE SYSTEMS 1988 BIOSPHERE RESERVE REPORT NO. 28 CARLOS R. RAMOS-PEREZ and CINDY GINES-SANCHEZ TERRESTRIAL ECOLOGY DIVISION CENTER FOR ENERGY AND ENVIRONMENT RESEARCH UNIVERSITY OF PUERTO RICO SAN JUAN, PUERTO RICO U.S. DEPARTMENT OF THE INTERIOR NATIONAL PARK SERVICE AND VIRGIN ISLANDS RESOURCE MANAGEMENT COOPERATIVE VIRGIN ISLANDS NATIONAL PARK P.O. BOX 7789, ST. THOMAS U.S. VIRGIN ISLANDS 00801 LOCAL CONTRACTING AGENT ISLAND RESOURCES FOUNDATION RED HOOK BOX 33, ST. THOMAS U.S. VIRGIN ISLANDS 00802 (NPS CONTRACT NO. CX-0001-3-0048) ABSTRACT This study was conducted to determine if three geochemical anomalies in St. John, USVI contribute significant amounts of heavy metals to near shore marine ecosystems. Stream water, sea water, marine sediments, and marine organisms from five watersheds were analyzed for Fe, Mg, Mn, Cu, Cr, Ni, and 4n in order to determine the transport of the metals. Metal concentrations in stream water were typical of small, unpolluted streams, and only iron exceeded the U.S. Environmental Protection Agency (EPA) water quality standard. Magnesium was transported primarily in the dissolved phase, and iron and manganese were transported in the particulate phase. The streams did not appear to have consistently lower or higher concentrations of the metals, but Coral Bay had markedly higher iron and copper than the other streams. Annual metal export was estimated for Fish Bay Gut. Mg and Fe were exported in quantities greater than 100 kg; the other metals only rarely exceeded this quantity. There was great variability in annual export rates depending primarily on the magnitude and frequency of rain events. Sea water samples also had low metal concentrations, but iron and manganese exceeded U.S. EPA water quality standards. There was considerable variability in metal concentrations among the sampling sites. Marine sediments also had low concentrations and high variability among the samples. Intensive sampling conducted in Fish Bay showed that Mn, Ni, and Fe were horizontally zonated with respect to the discharge point of Fish Bay Gut. The high concentrations found in sediments make this sample type the easiest compartment for heavy metal monitoring. Snails, crabs, and sea urchins from Fish Bay were analyzed. Snails and crabs appear to be good indicator organisms because the levels of heavy metal present in their tissue is well above the detection limit of the standard analytical techniques. The geochemical anomalies do not pose a serious threat to the marine ecosystems of St. John island. ACKNOWLEDGEMENTS We would like to express our gratitude to Mr. Gary Owen of the Caribbean Research Institute for his help during the critical moments. Additional field Support was provided by Mrs. Kristen Canoy and her assistants, Vincent and George, of the Virgin Island Environmental Research Station at St. John. We would also like to thank our friends and colleagues at the Center for Energy and Environment Research (CEER) , Lourdes Prieto, IvAn Rosa, and Ileana Mejia for their valuable help during this project. Partial funding for this project was provided by the University of Puerto Rico. li TABLE OF CONTENTS ADSCLACt. cc ccc ccc c ce eee were eee e ee ee sere esr seresssareresee i Acknowledgements......- cece ccc eee eee ee eee eee ee eee neces 1i Table of Contents.....-.... ccc eee eee nace rr iii List of Tables...... Cece eee ee ee eee eee eee ee ee eee eee eee iv List Of FIQUFES... ce we eee eee ee ener e eee reer ereeences Vv INCroduction. . cc ec cee cee eee eee eer e rer eerens Cee w ree 1 SOULCE. . cee cece cc ewer eens eee see ees rear eae a 1 TLANSPOLE. .. ccc eee eee eee eee reese reneeee see ee eee ceee 4 BLOCOKICILY. cc cece cece ec eee eee weer ener reenter eeeeees 5 Objectives... cc. cee cece eserves eee ee eeee ec cee eee ee 6 Materials and Methods...... eee eeeee eee nee ew ee cere eee eee 6 Analytical..........cc eee e nee cees ccc me eee ee eee reece 7 Stream Water Samples.........- ar ara 8 Sea Water SampleS...... cece eee ee cece ec cenc ewe wees 8 Marine Sediments... . cece newer cc nce cece ent e ener ceneees 10 Marine OrganiSMS..... cece cere c cere cccrerecscreeccece 10 Results. ... eee e eee ce cee a ara 11 Export from the Terrestrial Environment............. 11 Heavy Metals in Marine Near-Shore Environments...... 18 CONCLUSIONS... cc eee wee eee ern ene eee cceecccececeeceevees 28 Research and Management Recommendations..........44- 29 Literature Cited iii la. lb. LIST OF TABLES Heavy metal export estimate for Fish Bay, St. John... ee ee ec ccc eee ee ecm e ee eee eee eee 18 Heavy metal concentrations in the coastal waters of St. John... ... ec ccc eee cc cee eee 19 Average heavy metal content of surface coastal waters...... wee ee cece nn sene wee ween ener eves 20 Leachable metals in sediment core samples from St. John island................ eee eee cee ee wens 21 Leachable metals in sediment core samples from Fish Bay, St. John island...... ecw eee cee cece 22 Heavy metal content of surface near-shore Sediments............c2 cc eae em een eee wee rece ee eae 23 Heavy metal content of some marine . invertebrates in Fish Bay, St. John.............30.. 27 iv 6a. 6b. 6c. LIST OF FIGURES Geology of St. John. ...-- ee eceer eee ecereercrcerccecs Metal-rich geological anomalies of St. John........- Sampling stations....... were ec eee teeter eee eee eeee Heavy metal concentrations in St. John streams...... Relationship between precipitation and streamflow for Guinea Gut, St. Jonn......-. eee e eens Leachable Fish Bay, Leachable Fish Bay, Leachable Fish Bay, metals in sediment cores from St. John....... rr ar aera ar ar metals in sediment cores from metals in sediment cores from St. Jonn. . ccc cnc cece cece were csc err eenccece INTRODUCTION On a geological time scale, bedrock weathering and tectonic activities have been the largest sources of trace metals entering surface waters (Williams, et al. 1974). These natural inputs may lead to environmental concentrations of trace metals comparable to those produced by anthropogenic pollution. Miller et al. (1982) found dissolved copper concentrations of 4100 and 1300 ug/l in streams overlying or peripheral to large copper deposits, whereas concentrations ranged from 0.8 to 2.6 ug/l in nearby control areas. A recent study by the US Geological Survey identified several metal deposits in St. John, US Virgin Islands (Tucker, et al. 1985). Researchers discovered anomalously high concentrations of copper, lead, iron, barium, bismuth, and tin. These metals were not restricted to the bedrock, but were also detected in high concentrations in the nearby soils and stream sediments. This geochemical data suggests that bedrock-bound metals are entering the aquatic environment, where they may pose a potential threat to the biota. This study was conducted to provide additional information concerning the potential impact of these heavy metals on marine ecosystems. Source St. John’s geology was described by Donelly (1957). The Water Island formation, which crops out along the southern slopes, is composed chiefly of keratophyre flows and tuffs (Figure 1). Also of volcanic origin, the Louisenhoj formation overlies the Water Island formation and is composed of augite andesite breccias and tuffs with intercalated conglomerates. The Outer Brass Limestone overlies the Louisenhoj formation and is thin-bedded, silicified limestone. The Tutu formation is coarse volcanic wacke, composed almost entirely of weathered debris from the Louisenhoj formation. Tucker, et al. (1985) concluded that these formations were not the principal source of the metals. Instead they identified three geological anomalies which they believe to be the source of the metals (Figure 2). They postulate that these anomalies were formed by the emplacement of intrusive bodies of molten lava which flowed through the existing bedrock. The anomalies are metal rich and are characterized by the presence of iron enrichment, copper minerals, alunite, and gossan. MILES eann rr Z . . axxaJ Louisenhoj Fm. iiss) Tutu Pm. GQ Diurite coe; Water Island Fm. Alluvium Outer Brass Limestone Figure 1. Geology of St. John. Adapted from Donelly (1957). "(¢g6T) ‘Te qa ‘zeyon, worz peadepy ‘uyor “3S JO SeTTeuoue TeoThotoeb yoTs-Teqwow “Z erznbyTd Ajowouy j021way0a5 Ajowouy jOJ way I0ad yaad USI The largest anomaly is centered around Bordeaux Mountain and extends from the crest of the mountain to the sea. The authors suggest that a large body of magma flowed through the already deposited bedrock. This anomaly contains intensely argillized areas and nearly vertical iron oxide veins. Many of the streams draining the area contain sediments high in Ag, Bi, Sb, Ba, Cu, and Pb. The second largest anomaly is centered around Fish Bay Creek and is also characterized by high concentrations of Ag, Bi, Sb, Ba, Cu, Pb, and Se in a small area. This anomaly may have been caused by a small metalization zone, possibly related to a geological fault. The third anomaly, the Shore anomaly, is characterized by gossan and intrusive rock containing pyrite. Transport Trace metals are affected by various factors as they are transported through streams, lakes, and estuaries on their way to the sea. Metals in stream or sea water may be either dissolved or associated with suspended solids (generally by adsorption). Equilibrium conditions exist between the two phases depending on parameters such as PH, oxidation reduction potential, salinity, and concentration of suspended solids (Williams, et al. 1974, Salomons 1985, Santschi 1984). Sediments are by far the largest depository of metals ‘transported by streams and sea water. Santschi, et al. (1984) calculated that 50 to 100% of the cadmium and lead entering Narragansett Bay became incorporated, at least temporarily, into the sediments. This retention of metals makes sediments an ideal indicator of pollution. Of the trace metals associated with sediments, only a small fraction is released to the water column (Hunt 1983). These metals have been incorporated into the sediments by adsorption or complexation and do not form a part of the chemical lattice of the sediment particles (Chester and Voutsinou 1981, Gupta and Chen 1975). These so-called non- residual metals represent a small fraction of the total metal content of the sediments (Salomons 1985). Transport in St. John: Of the factors that could potentially affect the transport of the bedrock-bound metals through the environment, St. John’s climate and hydrology are most important. St. John receives very little rainfall, and drought conditions exist 66% of the time (BC&E 1979). The highest rainfall occurs from August to November (Cosner 1972), but seasonal cycles of precipitation are hard to establish due to the annual variability (Bowden, et al. 1970). Although St. John receives an average of 44 inches of rain annually, only 1 inch (2.3%) becomes streamflow (Cosner 1972). The net result is that the island’s streams are intermittent, with long periods between flows. Bowden, et al. (1970) reported that in Charlotte Amalie, stream flow occurred in only 14 occasions in a ten-year period (1958 to 1968). These few periods of runoff were generally short lived and characterized by high flows. The antecedent moisture conditions are very important in determining the amount of stream flow produced by a storm event. Cosner (1972) estimated that only 0.1% of rainfall became runoff when preceded by drought conditions, and 15.1% when preceded by heavy rain. Several other factors could affect the transport of metals through the environment. Soils are clayey and range from shallow to moderately steep (Rivera, et al. 1970). More than 84% of the island has slopes greater than 30% (BCSE 1979). The combination of easily transported clayey soils and steep topography tends to accelerate soil loss, which also accelerates the loss of soil-bound metals. Fortunately, a land-use survey conducted in 1975 concluded that nearly 88% of the island was forested (BC&E 1979), which diminishes soil loss. ; B; a Trace metals are toxic to the marine organism above an availability threshold, but many are essential to metabolism at lower concentrations. The toxicity of heavy metals to marine biota will depend on chemical characteris- tics and on availability (Rainbow 1985, Louma 1983). Results from several studies suggest that, for many species, there is a direct relationship between exposure to heavy metals and uptake (Bryan 1976). Uptake from solution is the major source of metals for most organisms since the bulk of the metals ingested passed through unassimilated (Pentreath 1973, Sick and Baptist 1979). The availability of free ions (a very small fraction of the total dissolved concentration) appears to control metal uptake from solution (Sunda and Guillard 1976, Anderson and Morel 1982). However, studies with deposit feeding organisms such as clams (Fowler and Unlu 1978) have found significant bioaccumulation from the ingestion of sediments. These metals are either maintained in a metabolically available form which may have toxic effects, or they may be detoxified. The negative effects of elevated concentrations of heavy metals to marine organisms have been well documented by the use of "in vitro" tests (Louma 1983). However, the result of these laboratory toxicity tests may lead to false or misleading conclusions because the conditions of the assay do not necessarily reflect the physiochemical conditions of the natural ("in vivo") environment. Field studies such as the one conducted by Rygg (1986) have shown that elevated metal concentrations can adversely affect marine communities. Rygg found a significant negative relationship between species diversity in benthic communities and sediment concentrations of heavy metals, especially copper. The consequences of gross heavy metal pollution in estuarine and marine environments have been frequently observed, but the detection of subtle or gradual ecological effects is difficult. Therefore, the environmental consequences of chronic metal contamination have probably gone largely unnoticed. This study was conducted to provide additional information concerning the potential heavy metal contamination in St. John by accomplishing the following tasks: 1. Measuring heavy metals in stream water, sea water, and marine sediments; 2. Estimating heavy metal export to marine ecosystems; and 3. Measuring heavy metal concentrations in marine organisms and estimating bioaccumulation. MATERIALS AND METHODS Samples were collected from several locations in St. John. Stream water, sea water, and marine sediments were taken in Fish Bay, Reef Bay, Little Lameshur Bay, Great Lameshur Bay, and Coral Bay. All sampling sites were located in watersheds draining the geochemical anomalies, and stream sediments had moderate to high concentrations of trace metals (Tucker, et al. 1985). Most streams were located in undisturbed watersheds, but the Fish Bay and Coral Bay Guts had significant residential development upstream from the sampling sites. More intensive sampling was conducted in Fish Bay because of strong development pressures in the area. At the time the samplings were conducted, several houses and roads were being constructed. These activities resulted in considerable soil movement to the stream channel. The location of the stream channel had recently been altered by the developers, thereby changing water movement through the mangroves. Fish Bay’s coastal zone has remained largely unaltered. A fringe of red mangrove borders the eastern half of the bay, and a rock outcrop borders the western half. Until recently, Fish Bay Gut drained into the bay at a small channel cut into the mangrove; the current discharge point is difficult to locate because the stream had not yet formed a well defined channel. The old drainage channel was used as the reference point for sample collection. Bottom sediments were uncompacted and deep. There were few rooted algae covering the bay floor, which made sediments highly susceptible to resuspension. Turbidity in the inner bay was very high, and visibility was often less than 2 feet. Aside from several sharks, there was low observable quantity and diversity of fish in the Bay. Analytical Trace metal concentrations were determined using an atomic absorption spectrophotometer (Perkin Elmer 2380) equipped with a deuterium lamp background corrector. Iron and magnesium were determined by direct aspiration; total chromium, copper, manganese, nickel and zinc were determined using a Perkin Elmer graphite furnace model HGA 400. Since metal concentrations were expected to be very low, extreme care was taken to avoid contaminating the samples during sampling and processing. All samples for metal analysis were taken in 1 liter polyethylene bottles soaked for 24 hours in 50% nitric acid and rinsed with distilled, deionized water. Furthermore, samples were processed in CEER laboratories in laminar flow hoods. Acids used to digest the samples or to acidify them were either UltrexR nitric acid or Intra-analyzed® hydrochloric acid. Blanks were analyzed for the different digestion techniques. A rigorous quality control program in accordance with US EPA protocol was run with the samples to ensure the reliability of the results (EPA 1979, EPA 1983). This program included the use of blanks, duplicate and replicate analyses, standard additions, and US EPA unknowns. W. caail Samples from the Fish Bay (2 locations), Little Lameshur Bay, Great Lameshur Bay, Bordeaux, and Coral Bay Guts (Figure 3) were analyzed for dissolved and total metals. All samples were taken 3 meters upstream from the point where the stream intersects a road except for Little Lameshur Bay Gut, where the sample was taken at the intersection with Reef Bay trail. The Great Lameshur Bay Gut samples were taken 50 meters upstream from the Virgin Islands Environmental Research Station field station. Samples were stored at 4 °C and processed within 48 hours. For the dissolved metal analysis, an aliquot was filtered through a prewashed 0.45 um membrane (Millipore HAWP) and acidified with concentrated nitric acid. The unfiltered portion of the sample was also acidified with concentrated nitric acid for the determination of total metals. Samples for total metals were digested using the hot concentrated nitric acid technique (EPA 1983). Particulate metals were defined as the difference between dissolved and total metals. Samples for the determination of total suspended solids (TSS) were collected at the same time that Samples for heavy metals were taken. TSS were determined gravimetrically using the glass fiber filtration technique (APHA 1980). Sea Water Samples Sea water samples were collected just off shore and 15 meters off shore from the gut outlets in Reef Bay, Little Lameshur Bay, Great Lameshur Bay, and Coral Bay (Figure 3). More intensive sampling was conducted in Fish Bay. When the samples were taken, there was no stream inflow. In order to minimize contamination, samples were collected by hand from a slow-moving fiberglass boat or by diving. Concentrated nitric acid and storage at 4 °C were used to preserve the samples prior to analysis. ZL oF | Sazequnn Ld 2 S FS a ¢ / eAle Ze YY fog | *“SaQTs QuUusUTpesS pue AaAeM edS s}eOTPUT “SORTS weszqS SReOTpPUT JF OF W $79930e7T *suotqeqgs but tdues kog uowouuty *¢ sanbty Samples were digested for the determination of total metals with the addition of 5 ml of 6N hydrochloric acid per 50 ml and heating (Brooks, et al. 1967). Iron, copper, manganese, and zinc were determined by direct aspiration (Sturgeon, et al. 1980). Marine sed; Surface sediments were collected wherever sea water samples were taken (Figure 3), except in Fish Bay (Figure 6a). Sediments were obtained while diving using acid- washed plastic cores (15.2 cm long x 5.1 cm diameter). The sediment samples were digested using the hydrochloric acid stripping method described in Wood and Acosta Cintrén (1976). Sediment cores were dried at 120 °C for 7 days. The samples were then sieved and weighed in duplicates. 175 ml of 50% hydrochloric acid was added to each sample, which was allowed to stand for 24 hours. The supernatant was decanted, and another 125 ml of acid was added. After an hour, the supernatant was decanted, and the sample was rinsed 3 times with distilled, deionized water. The digestate was taken to a final volume of 500 ml, and an aliquot was centrifuged at 3000 rpm for 5 minutes. The digestates were diluted for the determination of metals. Mari ; ; Invertebrates (sea urchins, crabs and snails) were caught in Fish Bay by hand and stored frozen. The crabs and snails were associated with mangroves, where they live in the roots and trunks of the trees. The sea urchins were found on a flat at the mouth of the gut. Fish traps using lettuce as bait were set up within the bay, but no fish were captured. Soft tissues (e.g. liver, kidneys, and heart) were removed by dissection from the crabs and sea urchins; total body tissues were used for the snails. The dry ashing technique described in AOAC (1980) was used to digest the tissues. Tissue samples were oven dried at 120 °C for 24 hours and placed in a muffle furnace for 6 hours at 500 °C. 50% hydrochloric acid was added to the cool, carbon-free residue, which was then warmed for 4 hours or until the sample dissolved. The solution was later transferred to a 50 ml volumetric flask and an aliquot was centrifuged and filtered through a pre-washed 0.45 um membrane. 10 RESULTS Export from the Terrestrial Environment Heavy._Metals in Stream Water: Dissolved, particulate, and total metal concentrations in stream water appear in Tables la and ib. Only iron and magnesium were found in concentrations higher than 100 ug/l; manganese, Copper, zinc, nickel, and total chromium content was low in all samples. Concentrations of dissolved and particulate iron varied considerably among the samples. Particulate iron concentrations, which ranged from <120 to 8,200 ug/l, were higher than dissolved concentrations, which ranged from <120 to 804 ug/l. The highest metal concentrations found in this study were dissolved magnesium, which ranged from 3,130 to 11,320 ug/l. In some cases, all of the magnesium was in the dissolved phase, and particulate concentrations were below the detection limit of 400 ug/l. None of the samples contained measurable dissolved manganese, but most contained detectable particulate manganese, which ranged from <5 to 61 ug/l. Dissolved copper concentrations were lower than 15 ug/l, and only one sample exceeded that value for particulate copper. Concentrations of zinc, nickel, and chromium were generally below the limit of detection for this study. These values are typical of small, uncontaminated streams. They also agree with previous samplings conducted by the US Geological Survey in St. John (Robison, et al. 1973). From 1962 to 1966, the dissolved metal concentrations for Guinea Gut were: iron <10-100 ug/I1, manganese <10 ug/l, and magnesium 48,000-85,000 ug/l. The large discrepancy in the magnesium values might be due to differences in land use or metal content of the bedrock. Most of the metals were below the water quality standards set by US EPA for stream water. Total copper, zinc, and chromium concentrations were much lower than the respective standards of 40, 50, and 50 ug/l. On the other hand, 17 of the 19 samples exceeded the total iron standard of 300 ug/l. Magnesium was the only metal transported primarily in the dissolved phase (Figure 4). Iron and manganese concentrations were much higher in the particulate phase than in the dissolved phase, suggesting that transport occurred primarily associated with suspended particles. In the case of manganese, all of the detectable metal was exported in the particulate phase. Copper transport was not dominated by one phase. 41 le r4 ¢ | tt 9 S> | OST‘6 OOF> o006‘8 | 9t a z> | OF le s> | oOpe’‘et o€a’s oto’s vA zZ S be G> S> | OSL‘TT O€F OZE'IT I | | 1 2t Ot z> 1 as Gb S> | 0£8°S OFL‘2 OET’E a) 9 z> 1 Ot $> S> | 08¢‘6 Oz20’E o92'9 fe S @> 1 Sb OF S> | 0996 o8e8’z o82'9 1 2 St l 19 S> S> | O6F‘OT OOF> O6F ‘OT a € rT | 6% rE S> | O41‘6 0€9 Obs ‘8 a: € S 1 62 bz S> | OZE*IL OFF 089 ‘OT | --+---- + + ee ee Ig z> S | es 8p s> 1 0468 oOOF> oaLe’s | -----~--+-- ~~ ee wee eee me ee me f oT g a> | 6€ be S> | 096‘b OTS’T oOGb‘E lo lt | 009%L Pe @> 9 i) ¢> S> | OSL‘L OOF> OG4'L {--- wee te a ot ee 0 ee es ee oe ta ee ee ee ee ee ee PL € b <4 9T S> | 06T‘S oO0F> O16’? a) € S | €t 8 S> | o92’L oOOFP> o9z2'4 oe) z b 1 g> G> S> | OB8€‘2 OOF> o8E‘L |---- ween wee eee ee ee ee ce toe ae ee se se ee ee ee | be €2 tT 199 19 S> | 026‘b OLb’T OSP’e It € 8 1 2t L s> | 009°S OOb> oOTP‘’S 1 61 It 8 | Te 92 S> | OoLb‘S o8s 068 ‘Fr | ---------+--- meee ene ee aac ee ee meee ew enn Eos d a [2 d a | L d a | zeddop I esourbhuep | untsoubew |------~----- ---- ene nn won-5--- eee *sTeqjow (1) [Teq309 pue /(g) aqetnotjred UT pesseidxs ere suoTjerquaou0D TIw a a we ee ee ne a a ae | | OZI=GIWIT NOILOSEZaI | | mt re ee nee eee ae a a a oe te | | ose 099 OgI> | €€°9 | MoT 91-S puog Aeg ustal I ot6’2 984%2 bet | | MOT bI-S puog Aeg ystal { 28€ 292 OcT> | org | MOT 8-S puog Aeg ystal | | | I | o66‘e€ oL8’e€ ozt> | ogee 1 gz'z BI-S ebptig Aeg ystal | pos 909 set | bree | 29°0 91-S ebptig Aeg ystal 1 09L‘T OF9‘T ozt> | | pL*o PI-G ebptig Aeg ystal | 2g¢ OzT> €os | | gg°9 8-S abpyig Aeq ystal | o€€‘9 9029 pet | etre f p9°O L~S ebptiza Aeg ystal 1 OLe'T tettt 6’T 1 Te°Le | 2970 L-S abptig Aeg ustal ibatiashataieiantereaahetataietetatatetatatetated ieheleieieieinte tates aaiaaleniaaiaeeaaietatateietetebetated oteateleteatetetetatetatetetad | | 9€2 O2T> b2T | | HOIH 61-S InD Aeg anysowey rajear9| eee ee eee eee er ee ae ee ee ee oe we ee ee ee ee ee ow ee ee ~| | o6€’2 zét‘z 86T | SE°L | HOIH 81-S Ino Aeg aznysouey aT33TT| | pos | oo'e 1 MOT bI-G any Aeq rnyseweq o143TT} 1 oLt‘2 296°T g0z | LO°T | MOT 8-S ano Aeg anyseuey eT33TT| = ee | | OPO‘T 604 Tee | 9e°e2 | HDIH BI-S 3n5 ‘uan xneepsog| | oze‘’t 9s bos | €4°¢ 1. MOT bI-S 3nD "UAW xneapiog| | S9T O@T> b2t | 26°S | MOT 8-S 2N5 *uqW xneapiog| wee en eee a ee a em | | o9€’s etz’s Zbt t OoO°P2T | HOTH 8I-s any Aeg Tez09] | o€6’t €69°T Lez | ps"e | MOT PI-S ang keg Tez09] | o19‘€ o6P‘e O@T> | oLcet ft mor 8-S any Keg Tez05] wee ee eee a a ee | ! L d a | (t/6u) jqaouwHoOsia Le6t aqabis| I uoqy | ss. | WWauLs aLva *“SPTTOS pepuedsns [e403 03 szaejer soy ‘(d) PeATOSSTp OjUT UMOp usayorq eae pue T/bn *sweozqs uyor “4S uT sTeqjeuw Aaeoy ‘eT ataqes 12 i) > e> | S> 1 8 S e> 18 lop €> €> | G> I i) | 9 €> €> 19 ,e€ €> €> | G> let 0 €> | g> lt €> €> | ¢> t ot L €> | G> 1g €> €> | S> I €> €> e> | S> 9 €> e> 19 | €> 1 ¢> 19 €> €> |. G> ly €> e> | G> 1b €> e> | S> | €> €> €> | ¢> '9 > €> | TT 1 €> €> €> {| G> | ST ZT e> | G> la d fej IL | un ;WwoOTYD I TOYOTN G> S> g> | MOT 9T-S puog Aeg ustal ¢> g> g> | MOT bI-s puog Aeg ustal G> G> s> {| MOT 8-S puog Aeg ustal | | > > g> | Bz°Z 8I-S abptza keg ustal S> g> g> | 29°70 9T-S ebpyig Aeg ystal G> > G> | ¥L°O PI-S abptia Aeg ystal G> G> s> | 6S°0 8-S afptia Aeg ystal l G> s> | p9°0 L-S ebptizq Area ystal g> g> g> $ 2970 L-S ebptaa Aeg ustal ee ee ee ee ce ee ee i ee a ee ee er eee { S> g> G> | HOIH 61-G any Aeq anyseuey ze3¢e75| ew ew ee ee ee ee ee ete = | eI 8 9 { HDIH 8t-s anp Aeg anysowey 93T33F7! G> | MOT pI-S any Aeq anyseweyT eT339F7] ¢> G> S> | MOT 8-S ano Aeg azanysaweqT ata3til S> G> ¢> | HOTH 8I-S an5 ‘uqW xneepizog| S> G> s> | mot bI-S 4n5 "UIW xNeveprog| “G> S> G> | MOT 8-S 3n5 *‘uaqW xneepi0g| ee rns ee ee a a a ea ee er ee ere { QT 8 8 | HOI 8I-s ang keg Tez0D{ G> G> g> [ Mot vI-s any Keg Tezr05| 6 G> S> | MoT 8-S anp Aeg Ter0D} a al ee a a a ee ee ee eee | L d a [SOUVHOSIG LB6T aLIs! DUyZ | WwauLs qdivwd | *sTeqjow (1) TeqOR pue ’(q) eqeTnoTIIed 4(q) PeATOSSTp OUT UMOp USyOIq eaze pue T/bn uy pesseidxe ere suoT ,erzqueou0D TV *sweezqs uyor °3$ UT sTeqeu AaeoH “qT sTQeL 13 © 3 Y Vs Ee The small number of samples taken in each gut makes it difficult to determine relationships between stream flow, suspended solids, and metal transport, but some generalizations can be made. Magnesium concentrations decreased with increasing discharge, and manganese content increased. The other metals showed no marked relationship between flow rate and metal concentration. Total suspended solids (TSS) were also higher at high flows than at low flows. TSS were positively correlated with particulate copper, manganese, and, to a lesser extent, iron (r“= 0.854, 0.713, and 0.678, respectively). Particulate manganese and copper were also correlated with particulate dron (r“=0.843 and 0.592, respectively). In order to compare the heavy metal content of the guts, averages were calculated for the three sampling dates on which all guts were sampled (Figure 4). These averages showed considerable variation, which could be due to differences in metal content of the bedrock and in land use within the watershed. None of the guts appeared to have consistently lower or higher concentrations of magnesium, iron, copper, total chromium, or manganese. However, Coral Bay Gut had markedly higher copper and iron content than the other guts. ; Export Estimate: In order to estimate heavy metal export from the Fish Bay watershed, the stream water heavy metal results were combined with estimated stream flow values. Since the Fish Bay watershed is ungaged, data from Guinea Gut were used to test the validity of hydrological models. Stream flow data for the Guinea Gut station were obtained from Cosner (1972), Curtis, et al. (1983 and 1984), and US Geological Survey preliminary data. Precipitation data were obtained from two sources: from 1963 to 1966, from the Guinea Gut watershed (Cosner 1972), and from 1983 to 1985, from the Coral Bay watershed (NOAA 1983, 1984, and 1985). The Guinea Gut watershed data showed that export calculations should be based on annual stream runoff rather than individual storm events. The relationship between precipitation and stream runoff for individual storms was very poor (Figure 5). This great variability in the amount of runoff produced by storms of similar magnitude was probably due to the influence of antecedent weather conditions in the watershed (Cosner 1972). If a storm was preceded by drought conditions, evapotranspiration could account for almost 100% of storm waters, and stream flow would be nearly 0. Therefore, the export estimates were based on annual streamflow and precipitation, which are significantly correlated (Figure 5). 15 Sorry miDswan Kae) ZQ ABAOMPMEPMAHM 0.7 + 0.4 ~ INDIVIDUAL b STORMS 10 $2 35 ~ 25 « 20 ~1. 13 ~ 10 - Figure 5. te ' 100. PRECIPITATION 140 (CM) Relationship between precipitation and streamflow for Guinea Gut, St. John. 16 The next step in the calculation was to extrapolate the precipitation/runoff relationship from Guinea Gut to the Fish Bay watershed by adjusting the runoff data for the drainage area of the two watersheds (0.37 and 1.77 miZ, respectively; Cosner 1972). Due to the sparsity of hydrological information for St. John, this manipulation represents the best estimate of runoff in Fish Bay. However, Jordan (1972) hypothesized that runoff from the major watersheds of St. John was not directly proportional to the drainage area. The annual runoff data can be subdivided into two groups (Table 2). In some years (1964, 1966, 1983, and 1985), storm events greater than 2.6 cm (approximately 1 inch) accounted for less than 20% of yearly runoff. Most metal export in these years probably occurred over a long period of time and during base flow conditions. However, in other years (1963, 1965, and 1984), storm events greater than 2.6 cm accounted for more than 65% of yearly runoff. Most metal export in these years probably occurred at high stream flows and over a relatively short period of time. Therefore, the runoff data were divided into "base flow" and “storm flow" years, and the estimated annual runoff values for Fish Bay were multiplied by the average total metal concentrations in Fish Bay Gut during low flow conditions (5-7 to 5-16), and high flow conditions (5-18), respectively (Table la and lb). Generally, there was little difference between the high flow and low flow metal concentrations, probably due to the small number of stream water samples analyzed. There was great variability in the annual metal export (Table 2). Each year, only iron and magnesium were exported in quantities greater than 100 kg and exceeded 1,000 kg several years. The other metals only rarely exceeded 100 kg. Since zinc and nickel concentrations were generally below the detection limit, the values presented in Table 2 represent the greatest possible export (based on the results of this study). These values are comparable to export calculations made by Hart, et al. (1982) for an Australian stream. It is very probable that only a fraction of the metals exported by a stream find their way to the near shore marine ecosystems. Many of the island’s streams do not drain directly into the sea. Mangroves and salt ponds may act as heavy metal sinks by accumulating metal-rich sediments. 17 Table 2. Heavy metal export estimate for Fish Bay, St. John. A. Hydrological data for Guinea Gut: Number of Rain % Yearly Runoff Year Rainfall Runoff Storms <2.6 cm Accounted for by (cm) (cm) Per Year Storms <2.6 cm 1964 68.1 0.23 1 1 1966 77.2 0.43 5 17 1983 134.3 36.40 3 12 1985 94.6 3.43 3 2 1963 90.2 1.47 5 66 1965 96.3 3.56 7 93 1984 128.4 20.35 6 99 B. Metal Export Estimates for Fish Bay Gut: Runoff Fe Mg Mn cu wi cr Year (L X 10°6) (kg) (kq) (kq) (kq) (kg) (kg) 1964 72 154 715 2.00 0.89 <0.36 0.42 1966 136 291 1,351 3.78 1.68 <0.68 0.79 1983 11,456 24,552 113,918 318.91 141.21 <57.28 66.44 1985 1,079 2,313 10,732 30.04 13.30 <5.406 6.26 1963 464 1,850 2,722 23.25 §.52 2.78 2.78 1965 1,119 4,466 . 6,570 56.11 13.32 6.72 6.72 1984 6,404 25,550 37,589 321.04 76.20 38.42 38.42 The quantity and chemical partitioning of the trace metals probably varied considerably in the "base flow" and "storm flow" years. Bradley (1984) concluded that the mass flow of metals is greatest during flood peaks. He also concluded that dissolved metal concentrations decrease at high flows, and the quantity of metals bound to particles increases. Therefore, export probably would be greatest during the "storm flow" years, but metals would be relatively unavailable to the environment because they would be tightly bound to suspended particles. During the “base flow" years, there would be less export, but the metals would be in the more bioavailable dissolved state. { i N - re Envi Sea Water: Concentrations of total metals in sea water followed a pattern similar to that observed in stream water (Table 3). Of the metais analyzed, iron had the highest concentration (range: 915 to 71 ug/l), followed by 18 manganese (4 to 387), zine (10 to 20), and copper (all samples <10). These values were considerably higher than the values obtained in Bermuda, and lower than those obtained in Jobos Bay, Puerto Rico (Table 4). Bermuda, like St. John, is a small island relatively free of anthropogenic contamination, whereas Jobos Bay receives the treated effluents of several industries. The difference in concentrations between Bermuda and St. John is especially interesting since it may be an indication that the metal- rich bedrock in St. John is indeed causing an increase in trace metals in coastal waters. However, the different digestion techniques utilized in the studies should be taken into consideration. The Bermuda samples were not digested, and the samples in this study were digested using 10% HCl. Several of the samples exceeded the EPA water quality standards for coastal waters. Seven of the fifteen samples had more than 200 ug/l of iron, and four samples had more than 100 ug/l of manganese. All of the zinc and copper samples were below the 50 ug/l standards. There was considerable variability among the sampling sites. Fish Bay apparently had higher concentrations of iron and manganese, and lower concentrations of zinc than the other bays. However, this fact is probably due to the greater number of samples taken in Fish Bay. Table 3. Heavy metal concentrations in the coastal waters of St. John. Distance refers to the approximate distance from shore in the north/south (N/S) and east/west (E/W) axis. For ID codes refer to Figure 3. Distance ID N/S E/W Fe Mn zn Cu Code Site (m) (m) (ug/l) (ug/l) (ug/l) (ug/1) 7 Coral Bay 15 0 276 56 10 <10 6 Sanders Bay 15 07 190 9 15 <10 5 Greater Lameshur E 15 0 86 4 17 <10 4 Greater Lameshur W 15 0 162 9 10 <10 3 Little Lameshur 15 0 71 4 18 <10 2 Reef Bay E 15 0 24 12 15 <10 1 Reef Bay W 15 0 24 9 10 <10 Fish Bay 0 0 407 387 10 <10 Fish Bay 20 0 762 340 10 <10 Fish Bay 40 0 915 368 10 <10 Fish Bay 60 0 467 104 10 <10 Fish Bay 80 0 484 73 20 <10 Fish Bay 100 0 446 65 10 <10 19 Marine Sediments: The same concentration pattern that was seen in the stream and sea water samples could also be seen in the marine sediments (Tables 5 and 6). Metals could be arranged in order of decreasing concentration: magnesium (23,800-2,250 ug/g), iron (25,610-1,230 ug/g), manganese (610-4 ug/g), total chromium (388-6 ug/g), copper (33-1 ug/g), nickel (30-1 ug/g), and zinc (<35 ug/g). These values were 2 to 90 times lower than concentrations obtained for Mayaguez Bay, in Puerto Rico, which receives industrial and domestic pollution (Table 7). Nichols and Towle (1977) studied sediment heavy metal concentrations in Benner Bay, St. Thomas. They obtained the following values: copper 140 to 0.2, and zinc 150 to 0.12 mg/g. These values are not comparable to values obtained in this study because the Benner Bay sediments were digested using hot concentrated nitric acid, which liberates matrix- bound metals in addition to the non-matrix metals liberated by leaching with hydrochloric acid. There was considerable variation in the metal content of the sediments, possibly due to differences in factors that control adsorption, such as organic matter (Salomons 1985), Manganese and iron oxides (Williams, et al. 1974), and percent non-calcareous residue (Jickells and Knap 1984). There was considerable variation in the organic matter content, sand and rock content, and vegetation of the sampling places, which could affect metal concentrations. As was the case in the stream water samples, manganese and copper were positively correlated with iron (p<0.05)° Table 4. Average heavy metal content of surface coastal waters. NA means not available. Fe Mn Cu an Ni (ug/1) (ug/1) (ug/1) (ug/1) (ug/1) Bermudal 1.07 0.19 0.2 0.3 0.14 St. John 332 111 <10 13 NA Jobos Bay, 541 252 9.6 103 NA Puerto Rico? 1. Total leachable metals (non-filtered samples without digestion) (Jickells and Knap 1984). 2. Total metals (non-filtered samples with HCl digestion). This study. 3. Dissolved metals (Puerto Rico Environmental Quality Board data for 1984). 20 eT S8 ST ST 8 OT 9 60 09 8S 6 ct (65/6n) 4D ET T OT 9T 9 ST O€ (6/6n) TN GE> ct GE> SE> GE> SE> GE> Se> SE> Se> SE> GE> GE> (6/6n) (6/5n) uz no mTAAMNHDUOAMN NH N Sepod dI 104g 9 v9 TTT BT TP €8 T?2 TET PT SZ 6 v (6/6n) uW 060°L OTT‘L ocr ’6 0€6‘Z OLb‘E 0€9‘L 0€0°S OTL ‘OT 00 L 00L‘F 0€9 ‘OT 099‘ZT (6/6n) BW OZZ‘L O8L‘L 0Z0‘L 090‘¢€ oce7‘'9 OZL’S OST’F Orr’L 06072 OSP‘T ObE‘T 0Ez‘T (6/6n) oq ST 0 ST 0 ST 0 ST 0 ST 0 ST ST (ul) AONV LSI SSH AVG TWHOD AVd TWHOO AVWd SUYAANWS AVG SUAANWS UNHSANWT YNHSANWT ANHSANWT UNS aNWTT UNS anNWT UNS aWNW'T ad AVG JaaY M AVG Aaa LS L9 L2 LD LT LT ANMOTTNMNWOWOME Adoo ALIS aI "€ °bta O} reyer “erTO0Uus WOTF BOUeIASTP oqeuTxordde ou OF SASForT soueqstGd *pueTsStT uyor *3g woaz satdues ez0d JusUTpes uT sTejeW sTqeyuoreT “G STOePL 21 GE G Se> OT €S 099 ‘ZT OTT‘OT 00Z2 OO€ TE S G€> ZT 8h OOF ‘OT 097'6 0OT 00€ o€ v Ge> GT L9 08. ‘OT 089’6 0S OO€ gt Z Se> L LE 0726 ‘L OeP’s 0 OO€ SOE ZT Ge> 8Z 8hZ OOF ‘LT 019 ‘GZ 0S- O0€ Ez € Se> 9T GE 006 ‘6 OOF‘L OST 00Z2 ce G Se> LT €G Ost ‘ot 0L6 ‘6 0S 002 oT Zz GE> S vS Onl ‘6 06T’S 0 00Z LT Zz Se> L vS 06h ‘TT 08z‘9 0S- 002 €z € Ge> L Zv 082 ‘OT 06c ‘9 0 OST 7A OT Se> ST 09T 070 ‘OT OLL‘S 09 OOT 6Z G Ge> €T OL . 08S ‘6 09S‘0T OP Oot SZ L Ge> VI SPT 008’8 O€L‘ZT 0Z 00. 02 8 Se> TZ T09 OZ2‘L O€7 ‘St 0Z- OOT Ze 6 Ge> LZ OTS 0L0/0T 029 ‘9T 0F- 00T €€ Tt Ge> €€ GST 020 ‘TT 06h ‘6T 09- OOT WN Z Ge> 0z Z9E 0999 O9T ‘TIT 0 06 9 T SE> v OF 0Sz‘z Ozb'z 0 08 V6T 9 GE> ST SET 008 ’EZ O6T ‘IT ) 09 88E TT Se> 8z 98T O€L‘OT O8Z’LT 0 0S 82 8 G€> G Zee 0798 OTE ‘6 0 0Z WN 8 SE> €T 08h 0€6 ‘6 06S ‘FT 0 OT (6/6n) (6/6n) (6/6n) (6/6n) (6/5n) (6/6n) (6/6n) (w) (ul) ID TN uz ng uW BW | M/a S/N FONVLSIA ‘eqep oeTqeTtTeae-uou surow WN ‘“STxe (M/a) semM/qAsee pue pue (S/N) YQnos/yqz0U eyy UT AnD Aeg YSTa Jo ehrzeyostp Jo Jutod 9yW woiy seoue {stp oj}eutxoadde ey of szeyear BOUeASTd ‘“pueTST uYyor °I3S ‘Keg uStg wozjz seTdues er09 juewtTpes ut sTeqjow eTqeyoeetT “9 eTqQeL 22 An intensive sampling effort was conducted in Fish Bay to determine if superficial sediment-linked metals were stratified in relation with the gut’s discharge point. Three of the six metals studied showed a slight horizontal zonification. Manganese, nickel, and to a lesser extent, iron had markedly higher concentrations in the samples taken within 100 meters of the gut’s discharge area than in more distant samples (Figure 6a and b). There were no discernible horizontal patterns for magnesium, copper, or chromium (Figure 6b and 6c). The lack of a strong pattern is not surprising since the stream runoff events that bear metals to the bay occur only 1 to 7 times per year. Post- depositional mixing of particles by physical and biological mechanisms could then affect the original heavy metal imprint on the sediments (Santschi, et al. 1984, Polprasert 1982). The water circulation pattern within the bay, which is yet unknown, may also affect the deposition pattern of the metals. Samples taken in other bays had lower concentrations of iron, magnesium, manganese, copper, and total chromium than samples taken in Fish Bay (Tables 5 and 6). This difference is probably due to the unequal sample sizes at each site. The greater number of samples taken in Fish Bay permitted samples with higher concentrations to appear in the study. Marine Invertebrates: Concentrations of manganese, copper, magnesium, and iron for several species of snails, crabs, and sea urchins are given in Table 8. Metal content varied considerably among the different organisms. The concentrations found in snail and crab tissues were generally higher than those found in sea urchins. However, this observation was not tested statistically due to the small sample size and large standard deviations. Table 7. Heavy metal content of surface near-shore sediments. Site Cr Cu Fe Mn Ni an (ug/g) (ug/g) (mg/g) (ug/g) (ug/g) (ug/g) Mayaguez Bay, 690 31 60 827 487 70 Puerto Rico Fish Bay, 68 12 9 9. 7 <35 St. John 1. Adapted from Wood and Acosta Cintrén (1976). 2. This study. 23 ~ Uys te. 8h 3a PAR SEE AE SAN OM eee OF ies Ae oy Stk! pe bt Boney 3 FFs ae wh fy oe ee Fel secondary Forest Figure 6a. Leachable metals in sediment cores from Fish Bay, St. John. 24 is TATOO CSET EE “ +e Ae)! by ees LIK Mangrove (| BSeeade*¥ Figure 6b. Leachable metals in sediment cores from Fish Bay, St. John. 25 (mg/g) 3 oS ad BA Say, REE ASE? fe SEE WERT ae : cone of we Mangrove pocondary Figure 6c. Leachable metals in sediment cores from Fish Bay, St. John. 26 9°*VOT - €°T6 TST - OTT veSL - 8829 OvT - SZI L°g - G°Z c qeap 0°S - T’T S - @ Los9t - €€0S 8S2 - GZT O°L - GE i uTYyoIn eas L°96T - €°2 90 - LOT PCTOT - T6S 8T8E€ - SeT 8°L - 0°S 8 Treus HONVY + a p6 - # 6'L6 «66Z CF TET) «6B CF 9069 «TT:SOF EET T'0 F 9°Z Z qezo “°T + T’°€ T FOP PLOG + LBE8 Lg + 8ST Q°T + 8° v uTyoan ees 8°8S + 72°88 TZU ¥ 9EZ O06T ¥F OTSL Té¢eT + T9PT 0°T + 0°9 8 TFeUsS (6/6n) (6/6n) (5/6n) (6/6n) oT3zey Azd/JemM ON nod UW BY oa NOTIWIASd GUYVGNVLS * FOWURAY cW *qyUBTOM ensstj Azp Jo swisj uT pesseidxe erze senteaA TITw ‘pezATeue STeENPTATpUT JO Aequnu ay Sjusseidert N ‘*TUUOSTA SNJSIY pue ‘/SSsoTdsp VISTA ‘PyeTUeNIs STSd6Tussy araem sqeazo oui pue ‘SnyUeTNosSe SSejsiseudtz7yL sem uTyoin ess sy ‘BiSFTTNHuS CUTACIITT sem Tteus SUL “‘Uuyor "43g ‘Aeq USTG UT SsaqeAIqeyTSAUT eUuTIeU sWOS JO JusaqUOD Tejew Aaeay °g aTqeL 27 Ze Manganese concentrations ranged from values as high as 436 ug/g in snails to values as low as <5 ug/g in sea urchins. The same pattern was found to be true for copper and iron, with maximum concentrations of 197 and 3,818 ug/g, respectively. The concentrations of nickel and total chromium were below the analytical detection limit. The large variation in metal content of the species is not surprising since it is well known that some organisms have a significantly higher capacity to bioaccumulate metals than others due to physiological and behavioral differences (Louma 1983). The levels of copper in crab tissue found in this study (36.4 and 38.9 ug/g wet weight) were higher than those reported by Sanders (1984) for two South Carolina estuaries (7.5 to 9.8 ug/g wet weight). This difference can be attributed to differences in methodology since Sanders analyzed only edible tissue. The tissue analyzed in this study was soft tissue which usually has higher metal concentrations. Snails and crabs appear to be good indicator organisms since the level of heavy metals present in their tissue is well above the detection limits of the standard analytical techniques. In order to assess their suitability, information about seasonal and intra-sample variability must be generated and studied, and the implications for biomonitoring determined. CONCLUSIONS The geochemical anomalies identified by Tucker, et al. (1985) do not pose a serious threat to the marine ecosystems of St. John island. Heavy metals are transported by the island’s streams to the sea, and can be detected in stream water, sea water, and marine sediments. However, these levels were typical of uncontaminated environments and, with the exception of iron and manganese, were mostly below water quality standards set by US EPA. The concentrations of trace metals in these three sample types followed a similar concentration pattern: Mg > Fe > Mn > Cu > Cr > Ni > Zn. The much higher concentrations found in marine sediments make this sample type the easiest compartment for heavy metal monitoring. Although samples from several watersheds were analyzed, no conclusions can be made concerning the relative impa: of the geochemical anomalies in these watersheds due to the 28 limited number of samples taken. Samples from Fish Bay had higher concentrations of some of the metals, but this fact can be attributed to the greater number of samples taken in Fish Bay. More intensive sampling, together with a better understanding of the role of metal sinks such as salt ponds and mangroves, are needed to address this question. The export estimates suggest that large quantities of some metals are transported by the island’s streams. More than 100 kg of iron and manganese were consistently exported by Fish Bay, and export of the other metals only rarely exceeded 100 kg. However, two factors must be taken into consideration in order to assess the potential impact of these metals. First, not all of these metals reach coastal waters because salt ponds and mangroves retard or stop metal movement. By slowing down water velocity, these ponds facilitate the incorporation of metals into the sediments. Second, the chemical partitioning of the metals determines the bioavailability to marine organisms. Dissolved ions are potentially more toxic than metals adsorbed onto particles. Only magnesium was transported primarily in the dissolved phase; iron and manganese were exported in the particulate phase. The levels of trace metals present in the animal tissues analyzed were not dangerously high and should not pose a threat to the organisms. There was considerable difference in the metal content of the different species collected in Fish Bay, with those associated with mangroves having the highest concentrations. Snails and crabs should be considered as indicator organisms because of their abundance and the relatively high concentration of trace metals found in their tissues. Research and Management Recommendations 1. Monitor iron and manganese concentrations in stream and sea water to determine how often the EPA standards are exceeded. Since iron is much easier to analyze and is highly correlated with manganese and copper, this metal might be used as an indicator of metal contamination. 2. Determine the transport of lead through the ecosystem. Tucker, et al. (1985) found very high concentrations of lead in the geochemical anomalies. 3. Determine the fraction of stream-transported metals that become incorporated into the sediments of salt ponds and mangroves. 29 4, Establish a long-term biomonitoring program. In order to assess the suitability of the indicator organisms, the intra-species and seasonal variability in metal content need to be studied. , 5. Study the effects on the ecosystem of the high sedimentation and turbidity in Fish Bay; determine the relative importance of natural and anthropogenic factors in causing this sediment load. 30 LITERATURE CITED Anderson, M.A. and F.M.M. Morel. 1982. The influence of aqueous iron on the uptake of iron by the coastal diatom Thalassiosira weissflogii. Limnol. Oceanogr. 27:789-857. AOAC. 1980. Official methods of analysis of the Association of Official Analytical Chemists. 13th ed. APHA. 1980. Standard methods for the examination of water and wastewater. American Public Health Association. 1134 PP. BC&E. 1979. A sediment reduction program. B.C. and E. Inc. Gainesville, Florida, USA. Bowden, M.J., N. Fischman, P. Cook, J. Wood, and E. Omasta. 1970. Climates, water balance, and climate change in the north west Virgin Islands. Caribbean Research Institute, College of the Virgin Islands, St. Thomas, VI. Bradley, S.B. 1984. Flood effects on the transport of heavy metals. Inter. J. Environ. Studies 22:225-230. Brooks, R.R., B.J. Presley, and I.R. Kaplan. 1967. Talanta 14:309. Bryan, G.W. 1976. p. 185-302 in R. Johnston (ed.) Marine Pollution, Academic Press, London. Chester, R. and F.G. Voutsinou. 1981. The initial assessment of trace metal pollution in coastal sediments. Mar. Poll. Bull. 12(3):84-91. Cosner, O.J. 1972. Water in Saint John, U.S. Virgin Islands. US Geological Survey Open-file Report. Available from: USGS Water Resources Div., San Juan, PR. Curtis, R.E., S. GuzmAn-Rios, and P.L. Diaz. 1984. Water resources data for Puerto Rico and the U.S. Virgin Islands 1984. US Geological Survey Report USGS-WRD-PR-84-1. Available from: USGS Water Resources Div., San Juan, PR. Curtis, R.E., S. Guzm&an-Rios, and P.L. Diaz. 1983. Water resources data for Puerto Rico and the U.S. Virgin Islands 1983. US Geological Survey Report USGS-WRD-PR-83-1. Donelly, T.W. 1957. Geology of St. Thomas and St. John, Virgin Islands. Ph.D. thesis, Princeton University, Department of Geology. 31 EPA. 1983. Methods for chemical analysis of water and wastes. EPA-600/4-79-020. EPA. 1979. Handbook for analytical quality control in water and wastewater laboratories. EPA-600/4~-79-019. Fowler, S.W. and M.Y. Unlu. 1978. Factors affecting bioaccumulation and elimination of arsenic in the shrimp Lysmata seticaudata. Chemosphere 9:711-720. Gupta, S.K. and K.Y. Chen. 1975. Partitioning of trace metals in selective chemical fractions of ‘nearshore sediments. Environmen. Letters. 10(2):129-158. Hart, B.T., S.H.R. Davies, and P.A. Thomas. 1982. Transport of iron, manganese, cadmium, copper and zinc by Magella Creek, Northern Territory, Australia. Water Res. 16:605- 612. Hunt, C.D. 1983. Variability in the benthic Mn flux in coastal marine ecosystems resulting from temperature and primary production. Limnol. Oceanogr. 28 (5) :913-923. Jickells, T.D. and A.H. Knap. 1984, The distribution and geochemistry of some trace metals in the Bermuda coastal environment. Estuarine, Coastal and Shelf Sci. 18:245-262. Jordan, D.G. 1972. Land-use effect on the water regimen of the U.S. Virgin Islands. Geol. Survey Prof. Paper 800-D, pp. D211-D216. Louma, S.N. 1983. Bioavailability of trace metals to aquatic organisms: A review. The Science of the Total Environment. 28:1-22. Miller, W.R., W.H. Ficklin, and R.E. Learned. 1982. Hydrochemical prospecting for porphyry copper deposits in the tropical-marine climate of Puerto Rico. J. of Geochem. Exploration. 16:217-233. Nichols, M. and E.L. Towle. 1977. Water sediment and ecology of the Mangrove Lagoon and Benner Bay, St. Thomas. Island Resources Foundation Technical Report 1, St. Thomas, USVI. 159 pp. NOAA. 1985. Climatological data annual summary. Puerto Rico and Virgin Islands. Vol. 30, Num. 13. National Climatic Data Center, Asheville, NC. NOAA. 1984. Climatological data annual summary. Puerto Rico and Virgin Islands. Vol. 29, Num. 13. 32 NOAA. 1983. Climatological data annual summary. Puerto Rico and Virgin Islands. Vol. 28, Num. 13. Pentreath, R.J. 1973. The accumulation and retention of 652n and °4mMn by the plaice, Pleuronectes platessa L. J. Exp. Mar. Biol. Ecol. 12:1-15. Polprasert, C. 1982. Heavy metal pollution in the Chao Phraya River estuary, Thailand. Water Res. 16:775-784. Rainbow, P.S. 1985. The biology of heavy metals in the sea. Intern. J. Environmental Studies. 25:195-211. Rivera, L.H., W.D. Frederick, C. Farris, E.H. Jensen, L. Davis, C.D. Palmer, L.F. Jackson, and W.E. Mckinzie. 1970. Soil Survey of the Virgin Islands of the United States. US Soil Conservation Service, San Juan, PR. Robison, T.M., et al. 1973. Water Records of the U.S. Virgin Islands 1962~1969. USGS Water Resources Div., San Juan, PR. Rygg, B. 1986. Heavy metal pollution and log-~normal distribution of individuals among species in benthic communities. Mar. Poll. Bull. 17:31-36. Salomons, W. 1985. Sediments and water quality. Environ. Technol. Letter. 6:315-376. : Sanders, M. 1984. Metals in crab, oyster and sediment in two South Carolina estuaries. Mar. Poll. Bull. 15(4):159-161. Santschi, P.H., S. Nixon, M. Pilson, and C. Hunt. 1984. Accumulation of sediments trace metals (Pb, Cu) and total hydrocarbons in Narragansett Bay, Rhode Island. Est. Coastal and Shelf Sci. 19:427-449. Santschi, P.H. 1984. Particle flux and trace metal residence time in natural waters. Limnol. Oceanogr. 29(5): 1100-1108. Sick, L.V. and G.J. Baptist. 1979. Cadmium incorporation by the marine copepod Pseudodiaptomus coronatus. Limnol. and Oceanogr. 24:453-462. Sturgeon, R.E., $.S. Berman, J.A.H. Desaulniers, A.P. Mykytiuk, J.W. McLaren, and D.S. Russell. 1980. Comparison of methods for the determination of trace elements in seawater. Anal. Chem. 52:1585-1588. 33 Sunda, W.G. and R.R.L. Guillard. 1976. The relationship between cupric ion activity and the toxicity of copper to phytoplankton. J. Mar. Res. 34:511-529. Tucker, R.E., H.V. Alminas, and R.T. Hopkins. 1985. Geochemical evidence for metalization on St. Thomas and St. John, U.S. Virgin Islands. US Geological Survey Open-File Report 85-297. Williams, S.L., D.B. Aulenbach, and N.C. Clesceri. 1974, Distribution of trace metals in aquatic environment. pp. 77-127 in A.J. Rubin (ed.) Aqueous-environmental chemistry of metals. Ann Arbor Science Publishers, Ann Arbor, Michigan. Wood, E.D. and N. Acosta Cintré6n. 1976. Accuracy in determining trace element concentrations in marine sediments. National Bureau of Standards Special Publication 422. 34 Carthbean Headquarters RED HOOK BOX 33, ST. THOMAS U.S. VIRGIN ISLANDS 00802 é ‘ +3 ISLAND RESOURCES FOUNDATION POSTSCRIPT (809) 775-6225 mae! : (202) 265-9712 Pssst! The butler didn't do it. But who did? Do we have your attention? Wouldn't you Like to know who was behind the scenes putting this series of reports together? WE DID IT! ALL twenty-nine technical reports ... all 2200 pages of research findings generated by more than a dozen institutional and individual subcontractors rr And the attached Biosphere Reserve "Research Report" is only a small part of the much larger technical document series which focuses on the re- source. base of the Virgin Islands Biosphere Reserve. Funded by the U.S. National Park Service and done under the aegis of the Virgin Islands. Resource Management Cooperative, VIRMC (see inside front cover of this report), the project invclved more than one-half of VIRMC's members as subcontractors, drawn from a dozen different disciplines. Island Resources Foundation (IRF) is proud of its behind-the-scene role as Washington, D.C. Office 1718 P STREET, N.W.. SUITE T4 WASHINGTON, D.C. 20036 project Organizer, prime Contractor, program manager and publication series” editor. We are equally proud of our role as an active member of VIRMC which offers great promise as an organizational vehicle for undertaking joint, interdisciplinary planning, research and monitoring projects. Penultimately, a word of appreciation to five persons who have been ex- traordinarily helpful to Island Resources Foundation in carrying out its role as primary contractor for VIRMC's Virgin Islands Biosphere Reserve program. I am taking note of those who assisted with the overall technical and management tasks of coordinating more than twenty-five subcontracts, arranging for text and copy editing, and proof reading the thousands of pages of reports. For their many services I am indebted to Dr. Caroline Rogers, Executive Officer of VIRMC (for technical review}, Phylis Rubin (manuscript copy editor), Jean-Pierre Bacle (for layout and publication co- ordination), Sandra Tate (for assisting with project administration and word processing) and Judith Towle (for fiscal management). Except for further distribution of documents, the project management tasks of Island Resources Foundation are almost completed. We have, however, only laid one course of "baseline" foundation stone for the structure of the Virgin Islands Biosphere Reserve. Other builders of the information base are now needed, both in the Virgin Islands and elsewhere in the East- ern Caribbean. We hope that the spirit of cooperation implicit in VIRMC will continue. 7 eG ‘Edward L. Towle, President Island Resources Foundation March 1988