Comparisons were conducted between estimates of photosynthetic capacity (Pmax) and photosynthetic efficiency (α) calculated with different models of the photosynthesis vs. irradiance curve. Values computed on the same data sets are different according to the models used. Estimates for Pmax with the exponential and hyperbolic tangent models (without a term for photoinhibition) are in good agreement (4% difference). The same comparison for α shows poor agreement (24% difference between the two models). When a parameter for the intercept is added to the two models, the lack of agreement increases to 8% for Pmax and 46% for α. When the mean photosynthetic parameters calculated with the two models are introduced into various published models for calculating primary production, differences in the resulting estimates range between 20 and 133%. Comparing the exponential model with a term for photoinhibition to the hyperbolic tangent model (without a term for photoinhibition) shows a 24% difference in the estimate of α. Equations are given for transforming values calculated with the various models.

The freshwater-saltwater transition zones of upper estuaries often contain high concentrations of particulate organic matter (POM) that could potentially support their productive food webs. Our objectives were to define the carbon and nitrogen isotopic characteristics of POM across the estuarine transition zone (ETZ) of the St. Lawrence River, Canada, and to determine the availability of this material to primary consumers. The ␦ 13 C of seston from upstream freshwater samples (Ϫ26.3 Ϯ 0.9‰) indicated a large contribution of terrestrially derived organic material, whereas downstream sites had higher values (Ϫ24.8 Ϯ 0.5‰) attributable to dilution with marine phytoplankton. The ETZ turbidity maximum had more negative values (Ϫ27.0 Ϯ 1.6‰), especially the Ͻ5-m fraction (Ϫ30.0 Ϯ 1.3‰) that contributed 56-86% of the POM. The ␦ 13 C of the dominant consumer species in the ETZ (Bosmina longirostris, Keratella sp., Eurytemora affinis, and adult Dreissena polymorpha) averaged Ϫ21.0 Ϯ 0.9‰, well above the seston values. The ␦ 15 N of the three animals of lowest trophic position averaged 7.3 Ϯ 0.8‰, less than 2‰ above the seston ␦ 15 N (5.7 Ϯ 0.2‰ for all sites). Our results imply that the bulk POM is largely detrital and unavailable to the ETZ food web and that the primary consumers feed selectively on phytoplankton despite its low contribution (Ͻ10%) to total POM. A cross-system comparison of isotopic data shows that the large enrichment of 13 C in consumers relative to POM is unusual, underscoring the distinctive character of turbid, upper estuarine ecosystems.

Large rivers are generally heterogeneous and productive systems that receive important inputs of dissolved organic matter (DOM) from terrestrial and in situ sources. Thus, they are likely to play a significant role in the biogeochemical cycling of the DOM flowing to the oceans. The asymmetric spatial gradient driven by directional flow and environmental heterogeneity contributes to the fate of DOM flowing downstream. Yet, the relative effects of spatial connectivity and environmental heterogeneity on DOM dynamics are poorly understood. For example, since environmental variables show spatial heterogeneity, the variation explained by environmental and spatial variables may be redundant. We used the St. Lawrence River (SLR) as a representative large river to resolve the unique influences of environmental heterogeneity and spatial connectivity on DOM dynamics. We used three-dimensional fluorescence matrices combined with parallel factor analysis (PARAFAC) to characterize the DOM pool in the SLR. Seven fluorophores were modeled, of which two were identified to be of terrestrial origin and three from algal exudates. We measured a set of environmental variables that are known to drive the fate of DOM in aquatic systems. Additionally, we used asymmetric eigenvector map (AEM) modeling to take spatial connectivity into account. The combination of spatial and environmental models explained 85% of the DOM variation. We show that spatial connectivity is an important driver of DOM dynamics, as a large fraction of environmental heterogeneity was attributable to the asymmetric spatial gradient. Along the longitudinal axis, we noted a rapid increase in dissolved organic carbon (DOC), mostly controlled by terrestrial input of DOM originating from the tributaries. Variance partitioning demonstrated that freshly produced protein-like DOM was found to be the preferential substrate for heterotrophic bacteria undergoing rapid proliferation, while humic-like DOM was more correlated to the diffuse attenuation coefficient of UVA radiation.

We studied the contribution of aquatic macrophytes and allochthonous sources to the pool of fluorescent dissolved organic matter (FDOM) in a large river system composed of several distinct water masses that flow alongside one another in the same riverbed. Using three dimensional fluorescence combined with parallel factor analysis (PARAFAC), we characterized FDOM found in the St. Lawrence River (Lake Saint-Pierre, QuØbec, Canada), and from macrophyte leaching experiments. Eight fluorescent components were identified, three of which were dominant in macrophyte experiments and were similar to protein-like, autochthonous fluorophores identified in previous studies. The remaining components corresponded to humic and fulvic acids, and a principal component analysis revealed that their distribution in Lake Saint-Pierre was different than that of protein-like fluorophores, suggesting a different origin. Concentrations of dissolved organic carbon were strongly associated with the distribution of the allochthonous components. The distribution of protein-like FDOM in Lake Saint-Pierre matched that of macrophytes in the lake and the abundance of allochthonous FDOM was explained by the connectivity with the terrestrial ecosystem. Nearshore water masses carrying large loads of newly imported organic matter from proximal tributaries showed the maximum abundances and the older water masses, from the center of the lake, carried smaller quantities of terrestrial organic matter, thus originated mainly from Lake Ontario, several hundred kilometers upstream of Lake Saint-Pierre. This study demonstrates that macrophytes are a net source of protein-like FDOM and could represent an important supply of autochthonous DOM in shallow, productive environments.

We measured characteristics of the underwater light spectra (e.g., attenuation of ultraviolet [UV] radiation, photosynthetically active radiation) and select dissolved and particulate physicochemical properties (e.g., chromophoric dissolved organic carbon [CDOM], dissolved and particulate organic carbon, inorganic dry weights, beam attenuation coefficients, particulate absorption coefficients, and nutrients) in different water masses of fluvial Lac Saint-Pierre (Canada). We used these variables as tracers to reveal the extent and magnitude of spatial and temporal heterogeneity in this large, shallow, fluvial lake of the St. Lawrence River. We superimposed these tracer variables over radiance data obtained from satellite images to identify spatial and temporal changes in the distribution of different water masses and their bio-optical components. The underwater light environment showed strong horizontal (longitudinal and lateral) variability because of the strong connectivity between the terrestrial and aquatic environments in the lake's tributaries and adjoining wetlands. Analyzing the downstream distribution of optical and chemical variables as a function of transport time rather than distance from source tributaries allowed us to demonstrate large differences in the age of the different water masses depending on the characteristics of the source tributary, in-stream processes, and distance from its source. CDOM explained most of the UV attenuation and allowed the greatest discrimination between water masses.