Major inter-annual variations in microbial dynamics in the Gulf of Trieste (northern Adriatic Sea) and their ecosystem implications

Comprehensive, multi-year studies of microbial processes in an ecosystem context are important in understanding microbial regulation of ecosystem structure, function and variability. A 3 yr study on microbial community was carried out in a coastal area in the northern Adriatic Sea (Gulf of Trieste). It is a semi-enclosed shallow system, mostly controlled by pulsed external inputs, which determine a high variability of plankton communities. Samples were collected monthly at 2 stations from January 1999 to December 2001. Data illustrate remarkable inter- and intra-annual variability in parameters relevant to carbon biogeochemistry and ecosystem energy flow patterns. Integrated primary production (PP; 135 g C m–2 yr–1) in 1999 tripled (414 g C m–2 yr–1) in 2000, returning to a low level (150 g C m–2 yr–1) in 2001. Bacterial production (by 3H-leucine incorporation) accounted, on average, for 35.3% of net PP in 1999, 9.5% in 2000 and 29.4% in 2001. Flux into the microbial loop via bacterial C demand, assuming 30% growth efficiency, was the dominant C flow pathway in 1999 and 2001, accounting for almost 100% of the net PP but a lesser fraction (40%) of net PP in 2000. Total microbial respiration (<200 μm fraction, R< 200μm) exceeded PP for almost the whole period, indicating that the system was potentially net heterotrophic at the expense of external input of organic matter. Our results suggest that in 1999 and 2001, when PP was low, it mostly fuelled, and was respired within, the microbial loop. During 2000 there was a major shift in C flux partitioning. The coupling between PP and bacteria was loose, and larger fractions of PP were presumably channelled into other paths of C flow such as grazing food chain, sedimentation and ‘storage’ in the water column as dissolved and colloidal organic phases. Organic C in excess produced over the 2 mo preceding visible accumulation of gelatinous aggregates (mucilage) was sufficient in magnitude to account for the mucilage C pool. Our results are consistent with the hypothesis that mucilage is derived from accumulated slow-to-degrade dissolved organic matter.