Community oxygen and nutrient fluxes in seagrass beds of Florida Bay, USA

We used clear, acrylic chambers to measure in situ community oxygen and nutrient fluxes under day and night conditions in seagrass beds at five sites across Florida Bay five times between September 1997 and March 1999. Underlying sediments are biogenic carbonate with porosities of 0.7-0.9 and with low organic content (< 1.6%). The seagrass communities always removed oxygen from the water column during the night and produced oxygen during daylight, and sampling date and site significantly affected both night and daytime oxygen fluxes. Net daily average fluxes of oxygen (-4.9 to 49 mmol m(-2) day(-1)) ranged from net autotrophy to heterotrophy across the bay and during the 18-month sampling period. However, the Rabbit Key Basin site, located in the west-central bay and covered with a dense Thalassia testudinum bed, was always autotrophic with net average oxygen production ranging from 4.8 to 49 mmol m(-2) day(-1). In November 1998, three of the five sites were strongly heterotrophic and oxygen production was least at Rabbit, suggesting the possibility of hypoxic conditions in fall. Average ammonium (NH4) concentrations in the water column varied widely across the bay, ranging from a mean of 6.9 mu mol l-1supercript stop at Calusa in the eastern bay to a mean of 0.6 mu mol l-1supercript stop at Rabbit Key for the period of study. However, average NH4 fluxes by site and date (-240 to 110 mu mol m(-2) h(-1)) were not correlated with water column concentrations and did not vary in a consistent diel, seasonal, or spatial pattern. Concentrations of dissolved organic nitrogen (DON) in the water column, averaged by site (15-25 mu mol l(-1)), were greater than mean NH4 concentrations, and the range of day and night DON fluxes (-920 to 1,300 mu mol m(-2) h(-1)), averaged by site and date, was greater than the range of mean NH4 fluxes. Average DON fluxes did not vary consistently from day to night, seasonally or spatially. Mean silicate fluxes ranged from -590 to 860 mu mol m(-2) h(-1) across all sites and dates, but mean net daily fluxes were less variable and most of the time contributed small amounts of silicate to the water column. Mean concentrations of filterable reactive phosphorus (FRP) in the water column across the bay were very low (0.021-0.075 mu mol l(-1)); but site average concentrations of dissolved organic phosphorus (DOP) were higher (0.04-0.15 mu mol l(-1)) and showed a gradient of increasing concentration from east to west in the bay. A pronounced gradient in average surficial sediment total phosphorus (1.1-12 mu mol g DW-1) along an east-to-west gradient was not reflected in fluxes of phosphorus. FRP fluxes, averaged by site and date, were low (-5.2 to 52 mu mol m(-2) h(-1)), highly variable, and did not vary consistently from day to night or across season or location. Mean DOP fluxes varied over a smaller range (-8.7 to 7.4 mu mol m(-2) h(-1)), but also showed no consistent spatial or temporal patterns. These small DOP fluxes were in sharp contrast to the predominately organic phosphorus pool in surficial sediments (site means = 0.66-7.4 mu mol g DW-1). Significant correlations of nutrient fluxes with parameters related to seagrass abundance suggest that the seagrass community may play a major role in nutrient recycling. Integrated means of net daily fluxes over the area of Florida Bay, though highy variable, suggest that seagrass communities might be a source of DOP and NH4 to Florida Bay and might remove small amounts of FRP and potentially large amounts of DON from the waters of the bay.