Rivermouth Alteration of Agricultural Impacts on Consumer Tissue δ15N

Abstract

Terrestrial agricultural activities strongly influence riverine nitrogen (N) dynamics, which is reflected in the δ15N of riverine consumer tissues. However, processes within aquatic ecosystems also influence consumer tissue δ15N. As aquatic processes become more important terrestrial inputs may become a weaker predictor of consumer tissue δ15N. In a previous study, this terrestrial-consumer tissue δ15N connection was very strong at river sites, but was disrupted by processes occurring in rivermouths (the ‘rivermouth effect’). This suggested that watershed indicators of N loading might be accurate in riverine settings, but could be inaccurate when considering N loading to the nearshore of large lakes and oceans. In this study, the rivermouth effect was examined on twenty-five sites spread across the Laurentian Great Lakes. Relationships between agriculture and consumer tissue δ15N occurred in both upstream rivers and at the outlets where rivermouths connect to the nearshore zone, but agriculture explained less variation and had a weaker effect at the outlet. These results suggest that rivermouths may sometimes be significant sources or sinks of N, which would cause N loading estimates to the nearshore zone that are typically made at discharge gages further upstream to be inaccurate. Identifying definitively the controls over the rivermouth effect on N loading (and other nutrients) will require integration of biogeochemical and hydrologic models.

Do rivermouths alter nutrient and seston delivery to the nearshore?

“…Many of these are sites that were previously reported in Larson et al . (, ). Site selection was constrained by logistical issues including that we could not sample in Canada (due to travel restrictions) and many rivers and rivermouths were inaccessible with our field equipment.…”

“…Some sites sampled in Larson et al . () are not included here because we did not collect stable water isotopes at those sites. We sampled most locations once and sites in the Ford, Manitowoc and Pere Marquette tributaries were sampled twice.…”

“…Two of the filters were stored in cryovials in liquid nitrogen and returned to the UMESC, where the filters were lyophilised and then shipped to Colorado Plateau Stable Isotope Laboratory (http://www.mpcer.nau.edu/isotopelab/isotope.html) for measurement of total carbon and nitrogen on the filters [nitrogen isotopes on many of these samples are reported in Larson et al . ()]. The final two filters were sent to Davy Environmental Laboratories, La Crosse, WI, where samples were digested in a sulphuric acid/persulphate solution [following EPA Method 365.1, Rev 2.0 (Dell, )] and then analysed for TP using a discrete analyser (Seal AQ2; Seal Analytical, Inc., Mequon, Wisconsin, USA).…”

“…At each site, we sampled three locations: (i) at an upstream location outside the influence of seiche-driven lake water inputs (River; R), (ii) at the outlet where the rivermouth entered the adjacent lakes (RM) and (iii) at a nearshore location in the lake that was outside the river plume at the time of sampling (L). Many of these are sites that were previously reported in Larson et al (2013bLarson et al ( , 2014. Site selection was constrained by logistical issues including that we could not sample in Canada (due to travel restrictions) and many rivers and rivermouths were inaccessible with our field equipment.…”

“…Of the remaining locations, all sites with long-term discharge monitoring were sampled and additional sites were randomly sampled to ensure more complete representation of each Great Lake. Some sites sampled in Larson et al (2013b) are not included here because we did not collect stable water isotopes at those sites. We sampled most locations once and sites in the Ford, Manitowoc and Pere Marquette tributaries were sampled twice.…”