Eos, Vol. 70, No. 33, August 15, 1989 Jellyfish Lake, Palau A Model Anoxic Environment for Geochemical Studies PAGES 777-778, 783 William C. Burnett, William M. Landing Department of Oceanography, Florida State University, Tallahassee W. Berry Lyons Institute for the Study of Earth, Oceans, and Space, University of New Hampshire, Durham William Orem U.S. Geological Survey, Reston, Va. Elemental cycling in suboxic to anoxic ma- rine environments has received considerable attention in the past few years. The recent expeditions of the RIV Knorr to the Black Sea, for example, illustrate the interest in deci- phering the geochemical processes active in such environments [Murray, 1988]. This interest has been stimulated by the recognition that the type of elemental cycling that occurs in restrictive basins such as the Black Sea also occurs in many other environ- ments. Examples include hydrothermal sys- tems, estuarine and marine sediments, and other water columns with restrictive circula- tion. In addition, evidence that widespread anoxia has occurred in oceans of the geologic past [Fischer and Arthur, 1977; Jenkyns, 1980] requires us to obtain a better understanding of geochemical processes under such condi- tions. We report here on preliminary geo- chemical measurements made in a stratified marine lake—an unusual, yet useful environ- ment as an analog of an anoxic ocean. A group of islands in the southern part of Palau (Figure 1), in the western Caroline Is- lands, consists almost exclusively of fossilized carbonate reefs uplifted by tectonic forces. These emergent islands contain areas that were once part of a much larger lagoon sur- rounded by coral reefs. These areas are now sea-level salt water lakes separated from the ocean by elevated limestone cliffs. There are about 70 such stranded lagoons in Palau; jel- lyfish Lake is surely the best known [Hamner et al, 1982; Muscatine and Marian, 1982; Hamner, 1982]. Jellyfish Lake is on an uninhabited carbon- ate island called Eil Malk, about 20 km south of Koror, capital of Palau. The lake is about 30 m deep, stratified with anoxic hypolimni- on, characterized by damped and delayed tides, and displaying no heat accumulation. Roughly rectangular, about 100 m x 40 m, it has steep sides and is separated into western and eastern basins, the western being deeper. The upper 15 m of the lake have oxygenat- ed, nutrient-poor and turbid waters. The sur- face layer has extremely high productivities and large populations of a restricted number of metazoans. The two types of jellyfish in- habiting the lake, Aurelia and Mastigias, are Cover. Research raft on Jellyfish Lake, Palau. The scientists are sampling lake wa- ter and sediments. See "Jellyfish Lake, Pa- relatively free of predators and have an un- usual migratory behavior that optimizes their life cycle in this unique environment. The combined population is an extraordinary 1.6 million jellyfish! Stratification in the lake is maintained by dilution of surface waters from the persistent Fig. 1 Map showing the location of the Palau Islands and Eil Malk, the island where Jellyfish Lake is. lau: A Model Anoxic Environment for Geochemical Studies," page 777. rainfall of the area. Several parts per thou- sand dilution is common at the surface; near- marine salinities occur in the deeper waters. The chemocline occurs at the density inter- face between oxygenated surface waters and anoxic deeper waters: it is a dense purple bacterial plate, probably a photosynthetic bac- teria such as Chromatium. This bacterial layer is 1-2 m thick and apparently degrades al- most all falling particulates and absorbs all sunlight. Divers who have gone beneath the plate report that the deeper water, while very clear, is completely dark. The underlying waters contain extremely high levels of hydrogen sulfide, ammonia, phosphate and silica. Sulfide levels in the lake, approaching 2.5 mM near the bottom, are among the highest reported for any natu- ral marine water column in the world, much higher than maximum values reported for such well-known examples as the Black Sea (0.29 mM; Dyrssen et al. [1986]), Saanich Inlet (0.006 mM; Jacobs [1984]), and the Cariaco Trench (0.050 mM; Jacobs [1984]), but lower than deep water from Framvaren Fjord (7.0 mM; Jacobs et al. [1985]). Continued study over several years by Hamner and his group has shown that these physical and chemical characteristics are very stable in the tropical climate of Jellyfish Lake, with no perceptible seasonal change. The combination of stability, limited inputs and outputs, a simple food chain, and a high degree of anoxia seemed to us to be an ideal environment for study of geochemical pro- cesses in an anoxic marine environment. Of particular interest at the outset was a study of the diagenetic reactions occurring in the or- ganic rich sediments at the lake bottom. We originally hypothesized that stratified marine lakes may be modern analogs of the environ- ment that produced the significant quantities of authigenic dolomite and phosphorite that occur on carbonate islands throughout the tropical Pacific Ocean. Sampling Strategy and Methodology In February and March 1987 we spent more than 6 weeks studying the geochemical character of Jellyfish Lake [Burnett et al, 1987a]. We analyzed the lake water and sedi- ment pore water for parameters that includ- ed phosphate, ammonium, silicate, alkalinity, pH, chloride, fluoride, sulfate, sulfide, iron and manganese. All processing and chemical analyses were done in the laboratory of the Micronesian Mariculture Demonstration Cen- ter (MMCD) in Koror. Our team had a graduate student and div- ing coordinator from Florida State Universi- ty, two professional divers from the National Oceanic and Atmospheric Administration, and several volunteers from Earthwatch Ex- peditions, Inc. MMCD was laboratory, base of operations and dormitory during our entire stay. At MMCD we prefabricated a 3.6 x 3.6 m raft, using standard-sized lumber and ply- wood available in Koror. The raft had a 1.2 x 1.2 m center opening for deploying equip- ment. For buoyancy, we used 12 empty 55- gallon oil drums donated by a local power company. The drums were cleaned and coat- ed with epoxy paint to minimize contamina- tion of the lake. After construction, we took the raft apart, transported the pieces to Eil