Relations between river flow, primary production and fatty acid composition of particulate organic matter in San Francisco and Chesapeake Bays: a multivariate approach

Estuarine environments influence particulate organic matter (POM) composition in the coastal zone as regions of high primary production and through their roles in the modification and exchange of organic materials. In this study, fatty acids (FA) associated with surface water particulate organic matter (POM) were collected seasonally and used to identify spatial and temporal variations in organic matter sources within two important estuaries in the USA–San Francisco Bay and Chesapeake Bay. Factor analysis was used to identify the dominant source signatures and the environmental processes important in controlling FA distributions within each estuary. In the San Francisco Bay dataset, Factor 1 explained 29% of the variance and had positive loadings for odd–numbered branched FA (C13, C15, C17 and C19), 16:1ω7 and long-chain FA (LCFA). Factor 1 was also negatively correlated with salinity and temperature,‘ suggesting that concentrations of these FA were controlled by seasonal fluctuations in freshwater inflow and the delivery of allochthonous materials. In contrast, loadings on Factor 2 (16% variance) in SFB were most positive for 16:0 and 18:0 and most negative for 20:5ω3, 14:0, 16:1ω7 and C16 polyunsaturated FA. These compounds are generally attributed to plankton sources. Factor 2 was inversely correlated with chlorophyll concentrations suggesting it was useful in resolving between FA derived from the spring (diatom) vs. non-spring plankton community. In Chesapeake Bay, Factor 1 explained 24% of the variance and was characterized by high positive loadings for 12:0 and C22 polyunsaturated FA and negative loadings for 16:1ω7 and n- and branched C13, C15 and C17 acids. Factor 1 was significantly correlated with salinity, suggesting it was useful for identifying regional differences in FA composition. Factor 2 (18% of the variance) was weighted positively for 18:1ω9 and negatively for 14:0 and 16:1ω7. This factor was inversely related to samples collected under bloom conditions (chl a>10 μg l−1) and chl a concentrations (r2=0.543; P <0.0001), indicating its usefulness for identifying temporal and regional differences in phytoplankton production and species composition. POM quality was, on average, higher in Chesapeake Bay vs. San Francisco Bay. Concentrations of total FA and polyunsaturated FA were higher in Chesapeake Bay, suggesting that POM is generally more labile and potentially more useful to heterotrophic organisms than in San Francisco Bay. Moreover, POM quality increased along the Chesapeake Bay salinity gradient. Overall, this study demonstrates that biomarker methods can provide insights useful in unraveling the complex factors that control the quality of POM in estuaries, particularly when coupled with environmental data (salinity, temperature and phytoplankton distributions) and multivariate statistical methods.