Microbial production, enzyme activity, and carbon turnover in surface sediments of the Hudson River estuary.

The detrital food web is a major nexus of energy flow in nearly all aquatic ecosystems. Energy enters this nexus by microbial assimilation of detrital carbon. To link microbiological variables with ecosystem process, it is necessary to understand the regulatory hierarchy that controls the distribution of microbial biomass and activity. Toward that goal, we investigated variability in microbial abundance and activities within the tidal freshwater estuary of the Hudson River. Surface sediments were collected from four contrasting sites: a mid-channel shoal, two types of wetlands, and a tributary confluence. These samples, collected in June to August 1992, were sorted into two to four size fractions, depending on the particle size distribution at each site. Each fraction was analyzed for bacterial biomass (by acridine orange direct counting), bacterial production (by (3)H-thymidine incorporation into DNA), fungal biomass (by ergosterol extraction), fungal production (by biomass accrual), and the potential activities of seven extracellular enzymes involved in the degradation of detrital structural molecules. Decomposition rates for particulate organic carbon (POC) were estimated from a statistical model relating mass loss rates to endocellulase activity. Within samples, bacterial biomass and productivity were negatively correlated with particle size: Standing stocks and rates in the <63-μm class were roughly twofold greater than in the >4-mm class. Conversely, fungal biomass was positively correlated with particle size, with standing stocks in the largest size class more than 1OX greater than in the smallest. Extracellular enzyme activities also differed significantly among size classes, with high carbohydrase activities associated with the largest particles, while oxidative activities predominated in the smallest size classes. Among sites, the mid-channel sediments had the lowest POC standing stock (2% of sediment dry mass) and longest turnover time (approximately 1.7 years), with bacterial productivity approximately equal to fungal (56 vs. 46 μg C per gram POC per day, respectively). In the Typha wetland, POC standing stock was high (10%); turnover time was about 0.3 years; and 90% of the microbial productivity was fungal (670 vs. 84 μg C per gram POC per day). The other two sites, a Trapa wetland and a tributary confluence, showed intermediate values for microbial productivity and POC turnover. Differences among sites were described by regression models that related the distribution of microbial biomass (r (2) = 0.98) and productivity (r (2) = 0.81) to particle size and carbon quality. These factors also determined POC decomposition rates. Net microbial production efficiency (production rate/decomposition rate) averaged 10.6%, suggesting that the sediments were exporting large quantities of unassimilated dissolved organic carbon into the water column. Our results suggest that studies of carbon processing in large systems, like the Hudson River estuary, can be facilitated by regression models that relate microbial dynamics to more readily measured parameters.