Dissolved organic matter transformations in a freshwater rivermouth

Abstract

River-to-lake transitional areas are biogeochemically active ecosystems that can alter the amount and composition of dissolved organic matter (DOM) as it moves through the aquatic continuum. However, few studies have directly measured carbon processing and assessed the carbon budget of freshwater rivermouths. We compiled measurements of dissolved organic carbon (DOC) and DOM in several water column (light and dark) and sediment incubation experiments conducted in the mouth of the Fox river (Fox rivermouth) upstream from Green Bay, Lake Michigan. Despite variation in the direction of DOC fluxes from sediments, we found that the Fox rivermouth was a net sink of DOC where water column DOC mineralization outweighed the release of DOC from sediments at the rivermouth scale. Although we found DOM composition also changed during our experiments, alterations in DOM optical properties were largely independent of the direction of sediment DOC fluxes. We found a consistent decrease in humic-like and fulvic-like terrestrial DOM and a consistent increase in the overall microbial composition of rivermouth DOM during our incubations. Moreover, greater ambient total dissolved phosphorus concentrations were positively associated with the consumption of terrestrial humic-like, microbial protein-like, and more recently derived DOM but had no effect on bulk DOC in the water column. Unexplained variation indicates that other environmental controls and water column processes affect the processing of DOM in this rivermouth. Nonetheless, the Fox rivermouth appears capable of substantial DOM transformation with implications for the composition of DOM entering Lake Michigan.

Supplementary information: The online version contains supplementary material available at 10.1007/s10533-022-01000-z.

Keywords: Carbon; Nutrients; PARAFAC; Sediments; Water column.

Fig. 1

Location of study sites on the mouth of the Fox River downstream from the De Pere dam (black lines) in northeastern Wisconsin, USA. The number and type of incubation experiments (Sed = sediment; Wat = water column) are annotated beside site locations. Bathymetry (depth, meters) obtained from the National Oceanic and Atmospheric Administration, National Centers for Environmental Information. Service layer credits: Esri, HERE, Garmin, GeoTechnologies, Inc., USGS, EPA.

Fig. 2

Summary of DOC fluxes from all sediment and water column (left) incubations, and net change in DOC in the Fox rivermouth (right). Negative values indicate decreases in DOC, and positive values indicate increases in DOC. Box plots show the median (line), quartiles (box extents), and 1.5 times the interquartile range (whiskers), and outliers (points) of the entire dataset. Red squares and error bars represent the mean and 80% credible interval of the posterior probability distribution with the percentage of the distribution outside of zero reported

Fig. 3

Contingency table between the change in DOM optical properties (increase, +; decrease, –) and DOC (increase, +; decrease, –) in sediment and water column incubation experiments. The size of each bubble indicates the proportion of observations for each condition. Bayes factors (BF) are reported whereby larger numbers indicate greater evidence against the null hypothesis of independence

Fig. 4

Contingency table between the change in PARAFAC components (increase, +; decrease, –) and DOC (increase, +; decrease, –) in sediment and water column incubation experiments. The size of each bubble indicates the proportion of observations for each condition. Bayes factors (BF) are reported whereby larger numbers indicate greater evidence against the null hypothesis of independence

Fig. 5

Summary of the change in DOM composition over all sediment (a) and light (b) and dark (c) water column incubation experiments at the rivermouth scale. DOM properties (left) and PARAFAC components (right) were centered to zero and scaled by the global standard deviation for visual comparison. Box plots show the median (line), quartiles (box extents), and 1.5 times the interquartile range (whiskers), and outliers (points) of the entire dataset. Negative values indicate a decrease and positive values indicate an increase in optical properties over the incubation. Red squares and error bars represent the mean and 80% credible interval of the posterior probability distribution with the percentage of the distribution outside of zero reported. Asterisk indicates strong unidirectionality in mean values signified by greater than 90% of the posterior probability distribution outside of zero

Fig. 6

Summary of net changes in DOM composition at the rivermouth scale. DOM properties (left) and PARAFAC components are centered to zero and scaled by standard deviation for visualization. Positive and negative values indicate an increase or a decrease in DOM properties, respectively. Red squares and error bars represent the mean and 80% credible interval of the posterior probability distribution with the percentage of the distribution outside of zero reported. Asterisk indicates strong unidirectionality in mean values signified by greater than 90% of the posterior probability distribution outside of zero

Fig. 7

Summary of dissolved organic carbon (DOC) and dissolved organic matter (DOM) processing in the Fox rivermouth. Illustration not to scale (vertical exaggeration x100). ^‡ Note, credible intervals overlap zero for DOC processing