Mixing Downstream of Stream Confluences Alters Carbon and Nutrient
Cycling in Freshwater Networks
Abstract
Stream confluences are ubiquitous features in freshwater networks, have
distinct hydrogeomorphic characteristics relative to upstream
tributaries and downstream reaches, and serve as junctions of previously
independent streams. Confluences may enhance or disrupt biological
processes. How ecosystem functions (e.g., carbon metabolism, nutrient
removal) change at confluences remains a knowledge gap in our
understanding of the processes controlling water quality at the
network-scale. To test how carbon and nutrient cycling may differ
between confluences and their tributaries, we estimated dissolved
organic carbon (DOC) and PO43- uptake in October 2018 and July 2019 in
two tributary reaches as well as downstream of their confluence mixing
zone using pulse injections of roasted barley leachate (a standardized,
colored DOC source), K2HPO4, and NaCl (a non-bioreactive tracer). We
hypothesized that biological processes would be enhanced at confluences
due to the delivery and mixing of different microbial communities and/or
carbon and nutrient sources. We calculated PO43- and DOC uptake
velocities (vf-PO4, vf-DOC) and compared them across sites and season.
In October 2018, vf-PO4 in each tributary was 10.2 and 4.9 mm/min while
vf-DOC was 0.84 and 0.38 mm/min. vf-PO4 downstream (6.6 mm/min) was
lower than vf-PO4 predicted from a mixing model of upstream vf-PO4 and
proportional flow contributions of tributaries (10.1 mm/min), suggesting
in-stream PO43- uptake was suppressed as a result of confluence mixing.
Conversely, vf-DOC downstream (0.94 mm/min) was higher than vf-DOC
predicted from a mixing model (0.75 mm/min). This difference in measured
and predicted vf-DOC was supported by bioassay experiments, which found
enhanced DOC uptake downstream of the mixing zone. DOC uptake within the
confluence mixing zone was spatially heterogeneous (0.00 to 0.19 day-1)
and varied more within mixing zone transects than among the two
tributary reaches. Ongoing analyses are comparing uptake estimates among
seasons. Our results suggest that DOC and PO43- uptake at confluences
cannot be estimated from tributary DOC and PO43- uptake alone. A
critical next step in this work is to identify the mechanisms behind
confluence-derived changes in carbon metabolism and nutrient removal
across freshwater networks.