Oecologia (1991) 88:305-310 Oecologia 9 Springer-Verlag 1991 Original papers Water relations of coastal plant communities near the ocean/freshwater boundary Leonel da Silveira Lobo Sternberg 1, Naomi Ish-Shaiom-Gordon 1, Michael Ross z, and Joseph O'Brien 2 1 Department of Biology,Universityof Miami, Coral Gables, FL 33124, USA 2 National Audubon Society, 115 Indian Mound Trail, Tavernier, FL 33070, USA Received March 11, 1991 / Accepted in revised form June 18, 1991 Summary. Salinity and isotope ratios were determined in water from several wells in the Florida Keys, and tidal inlets. Both D/H and lsO/160 ratios of water from wells and tidal inlets were highly correlated to their salinity. Water from standing pools was enriched in deuterium and oxygen-18 relative to their salinity because of evap- oration processes, l sO/160 and D/H ratios of stem water from plants of several different communities at Sugar Loaf Key, ranging from hardwood hammocks to man- groves, were highly correlated to their predawn water potential. The correlation was consistent with the presence of high salinity in waters with high 180 and D content. Most individuals from each community were either utilizing water with isotopic characteristics typical of freshwater or of ocean water, while only a few in- dividuals had stem water with isotopic ratios inter- mediate to these two water sources. Key words: Water potential - Mangrove - Hardwood hammock - Salinity - Stable isotopes South Florida and many other areas in the Caribbean basin have a flat topography which allows for the interac- tion between saline ocean water and rain-derived fresh- water. This interaction has produced a mosaic of several vegetation types exhibiting different salinity tolerances, ranging from saline tolerant mangrove vegetation to relatively non-tolerant hardwood hammock and cypress stands (Hoffmeister 1974). Previous research has shown that freshwater and ocean water in South Florida have different isotopic compositions, with freshwater showing less deuterium and oxygen-18 enrichment than ocean water (Sternberg and Swart 1987). It was also shown that by analyzing plant water, many interesting questions regarding the placement of various coastal communities in the interface between freshwater and ocean water can be answered. Three major communities were analyzed; Qffprint requests to . L. Sternberg hardwood hammock, mangrove margin, and mangrove. The mangrove margin community was loosely identified as a community occurring in sandy berms in the midst of mangrove. Plants in this community were charac- terized by succulent stem or leaves, with such species as Borrichia frutescens L.(DC.) and Sesuvium por~ tulacastrum (L.)L. The findings of this previous study showed that hardwood hammocks were located in the freshwater end of the salinity spectrum, mangrove mar- gin species occupied the middle of the spectrum, and mangroves, although mostly in the ocean water end of the spectrum, had several individuals utilizing fresh- water. Implicit in these conclusions is the assumption that salinity and isotopic ratios follow a linear relation caused by the various proportions of freshwater and ocean water during mixing. There are at least two ways in which this relationship does not hold. Firstly, rain water can mix with salts deposited by high tides, thus giving that water elevated salinities but isotopic ratios typical of freshwater. Secondly, freshwater bodies can undergo evaporation causing isotopic enrichment thus giving them high isotopic ratios typical of ocean water but low salinities typical of freshwater. In order to detect deviations from a simple ocean-freshwater mixing model we determined the isotopic composition and salinity from water of 28 wells through out the Florida Keys (Fig. 1). Salinity of water from these wells ranged from 2%, to 40%0. In addition we sampled water from several bays and tidal flows through out the Florida Keys (Fig. 1). We further hypothesized that if the salinity of water is related to its isotopic composition, plant predawn water potential should be related to the isotopic com- position of stem water. This relationship may be possible in southern Florida because its high rainfall minimizes the effect of soil matric potential as a determinant of plant predawn water potential. Plant predawn water potential should, therefore be mostly related to the soil water osmotic potential. Thus to determine whether the overall water status of different plant communities is dependent on where they are placed along a gradient of