Direct versus indirect effects of human activities on dissolved organic matter in highly impacted lakes

Human activities can alter dissolved organic matter (DOM) in lakes through both direct (i.e., exporting DOM of anthropogenic sources) and indirect effects (i.e., enhancing the autochthonous production of DOM via nutrient loading). Distinguishing between the direct and indirect effects is important to better understand human impacts on aquatic systems, but it remains highly challenging due to the interdependence of associated environmental variables. Here, we demonstrated that disentangling the direct and indirect effects can be achieved through combining large-scale environmental monitoring with the Partial Least Squares Path Modeling (PLS-PM). We presented DOM data from 61 lakes within the floodplain of the Yangtze River (Lakes-YR), China, a region that has been subjected to intense anthropogenic disturbances. We analyzed the amount and composition of DOM through dissolved organic carbon (DOC), chromophoric DOM(CDOM), and fluorescent DOM(FDOM). Four fluorescence components were identified, including one tyrosine-like component, one tryptophan-like component, and two humic-like components. Most of the lakes were dominated by freshly produced DOM with small molecular weights and low humification. Results from the PLS-PM models showed that the autochthonous production was more important than anthropogenic inputs in mediating DOC and CDOM. In contrast, FDOM parameters in lakes were more sensitive to the direct, anthropogenic sources, including treated domestic, industrial wastewater, and the effluents of aquaculture. These sources can be identified by elevated FDOM content per DOC (FDOM: DOC ratio) relative to autochthonous DOM, suggesting the potential of using FDOM as a tracer to identify and monitor the contribution of anthropogenic organic matter to inland waters. (C) 2020 Elsevier B.V. All rights reserved.

Automated summary

The study demonstrates that autochthonous production plays a key role in mediating dissolved organic carbon and chromophoric DOM in lakes, while fluorescent DOM parameters are more sensitive to anthropogenic sources. By combining environmental monitoring with the PLS-PM model, we can better unravel the direct and indirect effects.