SFS Annual Meeting

Kelly Aho (Primary Presenter/Author)
Yale University, kelly.aho@yale.edu;

Jennifer Fair (Co-Presenter/Co-Author)
Yale University, jennifer.fair@yale.edu;

Jacob Hosen (Co-Presenter/Co-Author)
Purdue University, jhosen@purdue.edu;

Ethan Kyzivat (Co-Presenter/Co-Author)
Brown University, ethan_kyzivat@brown.edu;

Laura Logozzo (Co-Presenter/Co-Author)
Yale University, laura.logozzo@yale.edu;

Lisa Weber (Co-Presenter/Co-Author)
Yale University, lisa.weber@yale.edu;

Bryan Yoon (Co-Presenter/Co-Author)
Northeastern University, b.yoon@northeastern.edu;

Jay Zarnetske (Co-Presenter/Co-Author)
Department of Earth and Environmental Sciences, Michigan State University, jpz@msu.edu;

Peter Raymond (Co-Presenter/Co-Author)
Yale University, peter.raymond@yale.edu;

Abstract: Nitrous oxide (N2O) emissions from streams and rivers are a significant, yet highly uncertain, flux in nitrogen cycle models. Current lotic N2O emission estimates assume that N2O production scales with nitrogen loading to streams or rivers. However, field-scale studies question this assumption, highlighting the need for process-based understanding of N2O dynamics. This study reports eight time series of pN2O in nested streams and rivers in Connecticut, USA. During the four-year study period, a multiyear dry period was punctuated by a large precipitation event, followed by a wet year. During the dry period, streams and rivers were consistently oversaturated in N2O and, therefore, N2O sources to the atmosphere. The intense rainstorm that ended the dry period was associated with a sharp decline in pN2O, followed by a prolonged period (6 – 12 months) of N2O undersaturation. This particular hydrologic event triggered a threshold response that smaller hydrologic events did not, and annual climatology likely amplified the effect. This non-linear response is attributed to disturbance of denitrification in the streambed and surrounding watershed. These findings are relevant to aquatic nitrogen cycling and N2O budgets under a changing climate.

Anne Kellerman (Primary Presenter/Author)
Florida State University, akellerman@fsu.edu;

Sarah Ellen Johnston (Co-Presenter/Co-Author)
University of Lethbridge, sarah.johnston3@uleth.ca ;

Peter Hernes (Co-Presenter/Co-Author)
UC Davis, pjhernes@ucdavis.edu;

Robert Spencer (Co-Presenter/Co-Author)
Florida State University, rgspencer@fsu.edu;

Abstract: Storm events and spring freshet are both characterized in rivers by marked increases in discharge and dissolved organic carbon (DOC) concentration, and therefore flux. Approximately 50% of annual DOC flux in Arctic rivers can occur during freshet alone. Enhanced transport during freshet and storm events likely access compositionally distinct dissolved organic matter (DOM), but how this will shift with climate-driven changes in the timing and dynamism of freshet is poorly understood. Here, we use spring freshet as an analogue for a storm event to investigate how changes in hydrologic flow impact DOC concentration and DOM composition. Ultrahigh resolution mass spectrometry and lignin phenol biomarker analysis were used to assess changes in DOM composition and terrestrial contributions to DOM flux over the hydrograph. Aromatic compounds, lignin concentration and lignin yield were elevated during freshet. In tributary rivers, aliphatic compounds were also elevated at the onset of freshet but did not necessarily covary with discharge, reflecting changing hydrological flowpaths over freshet. How this compositional succession may change with perturbations in winter precipitation and timing of snowmelt needs to be considered when assessing the downstream fate of exported DOM.

Grace Neumiller (Primary Presenter/Author)
Colby College, gcneum21@colby.edu;

Abstract: Intense nutrient loading of nitrogen (N) and phosphorus (P) causes sudden regime shifts in freshwater ecosystems from clearwater to turbid conditions with abundant cyanobacterial blooms. The influence of lake habitat (ie. benthic littoral versus pelagic zones) on nutrient limitation of primary production in mesotrophic lakes is largely unknown, particularly in contrast to research on pelagic nutrient limitation in eutrophic systems. Using paired nutrient diffusing substrata and mesocosm experiments, we measured chlorophyll-a concentrations in response to 4 nutrient treatments (N, P, N+P, Control) to assess nutrient limitation patterns in littoral and pelagic zones of two temperate, mesotrophic lakes in late summer, prior to the fall mixing period (October 2020). China Lake has approximately 2.18 times higher overall average [chl-a] in the water column than Great Pond in this late stratification period. China Lake phytoplankton appear to be co-limited by N and P (x? = 10.665 ?gL-1 chl-a), while Great Pond phytoplankton may be P-limited (x? = 2.4167 ?gL-1 chl-a). These data will improve our understanding of nutrient limitation patterns in mesotrophic systems in danger of eutrophication and allow us to incorporate littoral zones in our understanding of whole lake ecosystem productivity.

Maria Lisboa (Primary Presenter/Author)
Cornell University, msl282@cornell.edu;

Abstract: In the Northeastern (NE) US, nutrient pollution is receiving increasing attention, particularly phosphorus (P), in response to the recent increasing frequency of toxic blue-green algal blooms. This study assesses how land use, seasonal weather patterns, and climate extremes interact to control runoff and nutrient loading from tributaries into Owasco Lake. In 2016, NE US experienced a prolonged drought, which resulted in unusual, extreme low flows between August and October. The drought was ended by an intense rain event at the end of October. Results show that during the drought period the expected influence of agricultural and mixed land use on P loads was masked. However, high P loads were observed during the first rain event after the drought, exacerbating the agricultural impact on water quality. This study aids in understanding how climate extremes might affect nutrient runoff dynamics in the NE US. These findings are an important contribution from a management perspective, as projections for the NE US suggest that total precipitation will slightly increase, but large changes are expected in the extremes, with summer rains becoming concentrated in fewer events of higher intensities, interspaced with prolonged dry periods.

Fernando Carvallo (Co-Presenter/Co-Author)
Texas A&M University–Corpus Christi, fcarvallo@islander.tamucc.edu ;

Christopher Frazier (Co-Presenter/Co-Author)
Texas A&M Corpus Christi, christopher.frazier@tamucc.edu;

Christopher Groff (Co-Presenter/Co-Author)
Texas A&M Corpus Christi, cgroff@islander.tamucc.edu;

Victoria Jenkins (Co-Presenter/Co-Author)
Texas A&M Corpus Christi, vjenkins@islander.tamucc.edu;

Sean Kinard (Co-Presenter/Co-Author)
Virginia Institute of Marine Science, s2kinard@gmail.com;

Alexander Solis (Co-Presenter/Co-Author)
VIMS, alexander.tr.solis@gmail.com;

Bradley Strickland (Co-Presenter/Co-Author)
Virginia Institute of Marine Science, bastrickland273@gmail.com;

James Hogan (Co-Presenter/Co-Author)
Texas A&M University – Corpus Christi, james.hogan@tamucc.edu;

Christopher Patrick (Co-Presenter/Co-Author)
Virginia Institute of Marine Science (VIMS), cpatrick@vims.edu;

Matt Whiles (Co-Presenter/Co-Author)
University of Florida, mwhiles@ufl.edu;

Amber Ulseth (Co-Presenter/Co-Author)
Sam Houston State University, amber.ulseth@epfl.ch;

Connor Brown (Primary Presenter/Author)
Sam Houston State University, clb150@SHSU.EDU;

Abstract: Anthropogenic pressures of land use and climate change have the potential to impact chemical and biological factors that can affect stream ecosystem function. Ecosystem metabolism, which is gross primary production (GPP) and ecosystem respiration (ER), is a metric of stream ecosystem function as it integrates nutrient cycling and energy flow. We estimated daily GPP and ER using high frequency oxygen data from nine Texas coastal streams located along a precipitation and land use gradient. The most arid stream watershed consisted of predominantly shrub/grassland cover (55%), whereas the most mesic watershed consisted of predominantly agricultural land cover (90%). Unlike more temperate systems, these coastal streams did not show strong seasonal variation of GPP or ER. Average GPP ranged from 0.2 to 1.2 g O2 m-2 d-1, slightly peaking in the middle of the precipitation and land use gradients. Average ER ranged from -2.5 to -9.4 g O2 m-2 d-1 with no apparent trend along either gradient. These results suggest local factors, such as light and nutrients, are likely driving ecosystem metabolism in these systems, rather than broad-scale processes.

Erin Seybold (Primary Presenter/Author)
Kansas Geological Survey, University of Kansas, erinseybold@ku.edu;

Dustin Kincaid (Co-Presenter/Co-Author)
University of Vermont, dustin.kincaid@uvm.edu;

Brittany Lancellotti (Co-Presenter/Co-Author)
University of Vermont, blancell@uvm.edu;

E. Carol Adair (Co-Presenter/Co-Author)
University of Vermont, carol.adair@uvm.edu;

Julia Perdrial (Co-Presenter/Co-Author)
University of Vermont, julia.perdrial@uvm.edu;

Andrew Schroth (Co-Presenter/Co-Author)
University of Vermont, aschroth@uvm.edu;

Abstract: Winter in the northeastern US is changing in response to climate change. These changes include reduced snowpack and increased frequency of rain-on-snow (ROS) events – a type of extreme event that has been linked with flooding and infrastructure damage. Changes in the form of winter precipitation have the potential to impact the timing and magnitude of water and nutrient fluxes to downstream ecosystems. We investigated the impact of winter storm events on nutrient export in two watersheds in northern Vermont (agricultural vs. forested) using high frequency sensors located in the stream and riparian zone. We found that a significant portion of annual DOC, NO3-, and SRP loads were mobilized during ROS events (>20%) and that disproportionately more nitrate was mobilized (relative to water export) during ROS events than during spring snowmelt, indicating that these events are hot moments of nutrient export from catchments. Nitrate hysteresis patterns suggest elevated terrestrial-aquatic connectivity during ROS events, which may represent a shift in the degree of hydrologic connectivity between uplands and the stream during winter months. These results demonstrate that changing winter precipitation will have significant hydrobiogeochemical impacts on water quality in northern catchments.

Scott Ensign (Co-Presenter/Co-Author)
Stroud Water Research Center, ensign@stroudcenter.org;

Marc Peipoch (Primary Presenter/Author)
Stroud Water Research Center, mpeipoch@stroudcenter.org;

Abstract: Phytoplankton growth in rivers is limited by transport time, yet longitudinal increases in suspended algal biomass sometimes suggest unrealistically high growth rates. Hydraulic storage in backwaters and benthic retention are two hypotheses for this phenomenon that we address here by examining storm hysteresis behavior of chlorophyll concentrations in order to infer the location and type (benthic/suspended) of phytoplankton sources. We monitored suspended chlorophyll concentration and turbidity in-situ during 54 storm events at two locations of a 6th-order river with similar discharge but contrasting hydraulic storage caused by nine milldams between the sites. Upstream from the dams, storm chlorophyll hysteresis showed concentration effects (Δ[Chl]=5.4±0.9) and counterclockwise rotation, suggesting stormflow concentrated algae from far upstream. When autotrophic conditions (P/R>1) preceded storms, chlorophyll hysteresis switched to clockwise rotation indicating a more proximal benthic source of algae. During each individual storm, hysteresis downstream of the milldams indicated a more proximal source of suspended algae than at the upstream site. Unlike chlorophyll, turbidity hysteresis was similar between sites, supporting our inference that benthic sources account for water-column chlorophyll transport at the upstream site but hydraulic storage promotes planktonic growth and export at the downstream site.

Jay Zarnetske (Co-Presenter/Co-Author)
Department of Earth and Environmental Sciences, Michigan State University, jpz@msu.edu;

Arial Shogren (Co-Presenter/Co-Author)
University of Alabama, ashogren@ua.edu;

Chao Song (Co-Presenter/Co-Author)
Taizhou University, songchaonk@163.com ;

Benjamin Abbott (Co-Presenter/Co-Author)
Brigham Young University, Department of Plant and Wildlife Sciences, benabbott@byu.edu;

Frances Iannucci (Co-Presenter/Co-Author)
University of Alaska Fairbanks, fiannucci@alaska.edu;

Alexander Medvedeff (Co-Presenter/Co-Author)
University of Vermont, Alexander.Medvedeff@uvm.edu;

Caroline Weidner (Primary Presenter/Author)
Michigan State University, weidne11@msu.edu;

William Breck Bowden (Co-Presenter/Co-Author)
University of Vermont, breck.bowden@uvm.edu;

Abstract: As climate change causes shifting hydrologic behavior, it is critical to understand how changing flow impacts transport processes in vulnerable ecosystems, including the Arctic. Effective discharge (Qeff) has been proposed as an ecosystem-level metric that integrates biogeochemical and hydrological controls to describe the most “effective” flow that produces the greatest quantity of material export in a watershed relative to its frequency. Here, we estimated Qeff in two Arctic tundra watersheds, the Kuparuk River (little lake influence) and Oksrukuyik Creek (high lake influence), and made comparisons across monitoring years and watersheds. We used high-frequency measurements of dissolved organic carbon (DOC) and discharge (Q) across four thaw seasons to assess flow frequency distribution and quantify Qeff. We hypothesize that lakes present in the landscape retain DOC, reducing the overall watershed DOC flux and resulting in lower Qeff in Oksrukuyik. Preliminary results show that Qeff is greater for Kuparuk than Oksrukuyik across study years, potentially indicating the role of lakes. This study provides an initial assessment of Qeff across space and time, which will improve our understanding of transport processes in the Arctic as more high-frequency datasets become available.

Margaret Zimmer (Co-Presenter/Co-Author)
University of California, Santa Cruz, margaret.zimmer@ucsc.edu;

Galen Gorski (Primary Presenter/Author)
University of California, Berkeley, galengorski@berkeley.edu;

Abstract: Water quality sensor data can help disentangle interactions between anthropogenic, hydrologic, and biogeochemical factors that drive nutrient export across a range of spatial and temporal scales in agricultural watersheds. We analyzed daily nitrate concentration (c) and discharge (Q) data for a four-year period (2016-2019) from five nested, agricultural watersheds in the midwestern United States that contribute nutrient loads to the Gulf of Mexico. The watersheds span two distinct landforms shaped by differences in glacial history resulting in natural soil properties that necessitated different drainage infrastructure across the study area. To investigate nutrient export patterns under different hydrologic conditions, we partitioned the hydrograph into stormflow and baseflow periods and examined the c-Q patterns of these periods separately on annual, seasonal, and event time scales. Stormflow c-Q behavior was consistent across watersheds, but baseflow c-Q behavior was seasonally dynamic and linked to intensity of agriculture and density of built drainage infrastructure. This suggests that how humans ‘replumb’ the subsurface in response to geologic conditions has implications for hydrologic connectivity, homogenization of source areas, and subsequently nutrient export during both baseflow and stormflow.

Abagael Pruitt (Primary Presenter/Author)
University of Notre Dame, abagaelpruitt@gmail.com;

Jennifer Tank (Co-Presenter/Co-Author)
University of Notre Dame, jtank@nd.edu;

Shannon Speir (Co-Presenter/Co-Author)
University of Arkansas, slspeir@uark.edu;

Ursula Mahl (Co-Presenter/Co-Author)
University of Notre Dame, Ursula.H.Mahl.1@nd.edu;

Anna Vincent (Co-Presenter/Co-Author)
University of Notre Dame, avincen5@nd.edu;

Todd Royer (Co-Presenter/Co-Author)
O’Neill School of Public and Environmental Affairs, Indiana University, Bloomington, tvroyer@indiana.edu;

Abstract: Agricultural land use can increase erosion thereby increasing sediment export in streams. Land cover change, by planting cover crops during the winter fallow period, may protect soils and reduce storm-driven erosion and soil loss to waterways. To quantify cover crop impacts on sediment export, we measured continuous turbidity at the outlets of two agricultural watersheds (Indiana, USA) for four water years. We related turbidity to total suspended solid (TSS) using grab samples, generating a multiyear record of daily TSS loads. In the Kirkpatrick Ditch Watershed (KDW), 12-32% of croppable acres were cover cropped, while coverage was ~68% in the Shatto Ditch Watershed (SDW). Despite 25% lower runoff, average annual TSS export from KDW was ~7x higher (1140 kg/ha/yr) than from SDW (170 kg/ha/yr) and was 10x higher when we partitioned TSS losses for the top 10% of flows. TSS export was positively correlated with runoff in KDW (R= 0.88). In contrast, we did not see this relationship in SDW, likely due to infrequent overland flows combined with higher cover crop coverages. Overall, we show that cover crops may limit sediment loss during vulnerable periods when fields are bare.

Benjamin Abbott (Co-Presenter/Co-Author)
Brigham Young University, Department of Plant and Wildlife Sciences, benabbott@byu.edu;

Adam Norris (Primary Presenter/Author)
Brigham Young University, 96adamn@gmail.com;

Abstract: Increases in wildfire extent and severity are affecting terrestrial and aquatic ecosystems across the western US. Changing wildfire regime could have serious consequences for already fractured aquatic habitat and stressed biogeochemical cycles. We investigated the effects of a megafire (610 km2) on stream metabolism and nutrient uptake in river reaches along elevational and land use gradients in central Utah. We used a paired watershed experimental design in which we compared burned and unburned watersheds with week-long deployments of high-frequency sensors that measured dissolved organic carbon, nitrate and oxygen. We found that stream reaches were extremely diverse in their metabolisms and nutrient dynamics, with net uptake and net production observed in different parts of the surface water network. In addition to wildfire effects, we investigated the interactive influences of light, topography, riparian vegetation, and hydrological conditions on metabolism and nutrient balance. The high level of spatial and temporal variability we observed in nutrient uptake complicates efforts to estimate network-wide effects of wildfire, but hybrid approaches linking permanent stations with short-term deployments could provide a fuller picture of how changing disturbance regimes will affect water and nutrient budgets for ecosystems and society.

Sasha Wagner (Primary Presenter/Author)
Rensselaer Polytechnic Institute, wagnes3@rpi.edu;

Riley Barton (Co-Presenter/Co-Author)
Rensselaer Polytechnic Institute, bartor2@rpi.edu];

Lauren Giggy (Co-Presenter/Co-Author)
University of California, Santa Cruz, lgiggy@ucsc.edu;

Peter Willits (Co-Presenter/Co-Author)
University of California, Santa Cruz, pwillits@ucsc.edu;

Margaret Zimmer (Co-Presenter/Co-Author)
University of California, Santa Cruz, margaret.zimmer@ucsc.edu;

Abstract: Wildfires alter the lateral transfer and in-stream export of carbon, thus impacting river biogeochemistry at the watershed scale. The frequency and intensity of both wildfire and storm events are increasing as a result of our changing climate, but the biogeochemical links between precipitation, runoff, and material export is still poorly understood in fire-affected systems. Given the logistical challenges inherent to sampling coincident wildfire and rainfall events, capturing the “first flush” of carbon after a burn has remained particularly elusive. Our experimental watershed (Diablo Range, CA, USA) was burned in the summer of 2020 and we collected hydrologic data and stream samples during the first post-fire storm events that winter. Our preliminary data suggest the export of dissolved PyC may be decoupled from DOC during the first flush – A novel finding since these two carbon forms are typically correlated in river systems. We expect DOC and PyC export to be controlled by the hydrologic routing of water through the system (e.g., surface runoff versus groundwater inputs) to produce an integrated view of how watershed biogeochemistry is altered by wildfire and rainfall on short timescales.

Robert Hensley (Primary Presenter/Author)
Battelle, National Ecological Observatory Network , hensley@battelleecology.org;

Joel Singley (Co-Presenter/Co-Author)
University of Colorado, joel.singley@colorado.edu;

Michael Gooseff (Co-Presenter/Co-Author)
University of Colorado, michael.gooseff@colorado.edu;

Abstract: Como Creek and West St. Louis Creek are two sites in the NEON Rocky Mountain Domain. Located around 3,000m elevation, they among the highest instrumented streams in the U.S. In these snowmelt dominated streams, nitrate (NO3-N) and dissolved organic carbon (DOC) peak during the initial onset of the spring melt pulse, producing clockwise C-Q hysteresis over seasonal time-scales. This “first flush” suggests rapid mobilization of nutrient and carbon stores. However, over shorter time-scales (e.g. daily pulses within the pulse), NO3-N and DOC lag discharge, producing smaller counterclockwise hysteresis within the larger clockwise loops. Following the melt pulse, we observed increased rates of stream metabolism, potentially a response to warmer temperatures. This raises the possibility that the lower nutrient and carbon concentrations observed during this period may also partly be due to increased biological processing (in the stream and also the contributing area). Given that approximately ¾ of streamflow in the western U.S. derives from snowmelt, understanding the biogeochemical dynamics of melt pulses is of critical importance, particularly in the face of warming winter temperatures and reduced snowpack.

Patrick Kelly (Primary Presenter/Author)
Rhodes College, kellyp@rhodes.edu;

William Renwick (Co-Presenter/Co-Author)
Miami University, wrenwick@miamioh.edu;

Michael Vanni (Co-Presenter/Co-Author)
Miami University, vannimj@miamioh.edu;

Abstract: Large storms drive a significant proportion of a lake’s annual phosphorus (P) load, particularly in agricultural watersheds. However, nitrogen (N) dynamics during these events have received less attention. This has led to an incomplete understanding of how load N:P ratios vary during periods of high magnitude nutrient delivery to the lake. We assessed how load N:P varies with discharge in a 21-year time series of nutrient loading to Acton Lake, a hypereutrophic reservoir in southwestern Ohio, USA. In general, high discharge events disproportionately increased P load relative to N. This increase is particularly noteworthy when comparing high discharge events to base flow periods with the same cumulative discharge, as high discharge events deliver up to 400% more SRP compared to base flow while NO3 remains proportional. This suggests high discharge events provide periods of greater overall nutrient load and significantly reduced N:P compared to base flow conditions. Comparing this relationship across the time series also demonstrates increased conservation tillage in the watershed has steepened the relationship between discharge and P loads, but not N. As a result, high discharge events contribute even lower N:P than under conventional till.

Dustin Kincaid (Primary Presenter/Author)
University of Vermont, dustin.kincaid@uvm.edu;

Scott Hamshaw (Co-Presenter/Co-Author)
University of Vermont, shamshaw@uvm.edu;

Kristen Underwood (Co-Presenter/Co-Author)
University of Vermont, kristen.underwood@uvm.edu;

Erin Seybold (Co-Presenter/Co-Author)
Kansas Geological Survey, University of Kansas, erinseybold@ku.edu;

E. Carol Adair (Co-Presenter/Co-Author)
University of Vermont, carol.adair@uvm.edu;

Julia Perdrial (Co-Presenter/Co-Author)
University of Vermont, julia.perdrial@uvm.edu;

Donna Rizzo (Co-Presenter/Co-Author)
University of Vermont, drizzo@uvm.edu;

Beverley Wemple (Co-Presenter/Co-Author)
University of Vermont, beverley.wemple@uvm.edu;

Andrew Schroth (Co-Presenter/Co-Author)
University of Vermont, aschroth@uvm.edu;

Abstract: Hydrological events, such as storms, transport large quantities of nitrogen (N) and phosphorus (P) from watersheds. Understanding and predicting event N and P export is challenging because complex drivers of solute transport and event regimes interact with watershed characteristics to produce large variability in N and P export. However, novel statistical tools can extract information from a large number of interacting variables to provide insight into the processes that drive event N and P export. Here we use discharge, high-frequency sensor-derived dissolved N and P estimates, precipitation, and riparian soil physicochemical data from ~100 hydrological events to determine how land use, seasonality, hydrometeorological event characteristics, and antecedent conditions influence dissolved N and P export. We clustered the event-scale data using self-organizing maps to identify distinct event regimes and determine their impacts on dissolved N and P export. The events, which spanned varied land uses, event characteristics, and antecedent conditions clustered into distinct event regimes that featured unique export dynamics, including yields and concentration-discharge relationships. The resulting event regimes inform our understanding of how watershed export may change in the future as a result of climate or land use change.

Wilfred M. Wollheim (Primary Presenter/Author)
University of New Hampshire, wil.wollheim@unh.edu;

Sarah Bower (Co-Presenter/Co-Author)
University of New Hampshire, Sarah.bower@unh.edu;

Andrew Robison (Co-Presenter/Co-Author)
University of New Hampshire, andrew.robison@unh.edu;

Christopher Whitney (Co-Presenter/Co-Author)
University of New Hampshire, chris.whitney@unh.edu;

Abstract: Stream networks can denitrifiy large amounts of anthropogenic nitrate, particularly when they have abundant fluvial wetlands. Retention is expected to decline when flows increase due to reduced residence times and biogeochemical saturation. Further, fluvial wetlands are also potential sources of greenhouse gases (GHG: N2O, CH4, CO2). We quantified N retention and GHG production in two fluvial wetland dominated stream flow paths (~10km) of contrasting nitrate loads and across storm flows. We hypothesized that storms would lead to a downstream breakthrough of nitrate, increased N2O evasion, and reduced CH4 evasion. We also hypothesized that these changes would be greater in the high compared to low N flow path. Three storms events were sampled in each flow path, at baseflow and during two points in the storm hydrograph. We found here was no nitrate breakthrough during storms, indicating little evidence of saturation. During storms, GHG evasion increased, but there was a lower response in high than low nutrient flow path, suggesting little evidence of tradeoff between N removal and GHG emissions. Findings suggest nutrient management relying on N removal by fluvial wetland dominated streams will not result in increased GHG evasion.

Natalie Murphy (Co-Presenter/Co-Author)
Virginia Tech, nataliem14@vt.edu;

Erin Hotchkiss (Co-Presenter/Co-Author)
Virginia Polytechnic Institute and State University (Virginia Tech), ehotchkiss@vt.edu;

Kristen Bretz (Primary Presenter/Author)
Virginia Tech, kabretz@vt.edu;

Abstract: Headwater streams expand, contract, or fragment seasonally with changing precipitation and evapotranspiration. Stream fragmentation and reconnection likely disrupt carbon fluxes and fate, but uncertainty surrounds quantifying the effects of intermittency on stream carbon. To understand how fragmentation affects carbon dynamics in a temperate mountain stream, we tracked CO2 and dissolved organic matter (DOM) concentrations with stream drying. We measured high-frequency CO2 at the stream’s base and sampled dissolved CO2 and CH4 at key locations along the valley up to the stream’s source every few weeks. High-frequency CO2 was variable and inversely tracked local precipitation. Spatial CO2 variability mirrored temporal variability: the range of site medians across 7 months was 243 mmol m-3 while the range of monthly medians was 219 mmol m-3. Similarly the range in monthly median CH4 was 0.222 mmol m-3 while the range in site medians was 0.195 mmol m-3. Ongoing work integrates surface water intermittency mapping, water residence time, DOM fluxes, and carbon metabolism with gas fluxes. As the prevalence of stream intermittency is expected to increase with climate change, we must better quantify the effects of flow impermanence on carbon cycling and fate in streams.

Tyler Hampton (Primary Presenter/Author)
University of Waterloo, tyler.hampton@uwaterloo.ca;

Nandita Basu (Co-Presenter/Co-Author)
University of Waterloo, nandita.basu@uwaterloo.ca;

Simon Lin (Co-Presenter/Co-Author)
University of Waterloo, simon.lin@uwaterloo.ca;

Abstract: Wildfires have been documented to cause drastic deterioration in water quality in streams and rivers. We compiled a comprehensive meta-analyses of water quality data from burned watersheds, using criteria for paper title abstract and keywords, and examining the response of water quality parameters such as suspended sediments, nutrients, and organic carbon. For 105 watersheds across the globe, we show consistent increases (~75% of sites) in suspended sediments, and nitrogen and phosphorus species (inorganic and organic); along with strong correlation between species, such as total P and sediment. While findings of increased mean concentrations of various nutrients are not new, many studies also document that water quality worsens during extreme events (e.g. storms, spring freshet). We examined studies’ concentration-duration curves and we found a consistent “hockey stick” effect where increases in the highest concentrations are much greater in magnitude than increases at lower concentrations. This shows the importance of sampling and measuring water quality and flow during these events. Our study documents strong heterogeneity in responses of water quality to wildfire that have been unreported so far in the literature.

Matthew Chen (Primary Presenter/Author)
University of Nebraska-Lincoln, mchenl511@gmail.com;

Abstract: The purpose of this study was to examine nitrogen and phosphorus dynamics in Niobrara River following a major flood event. In the early spring of 2019, a major rain-on-snow event occurred across the US Midwest led to historic levels of flooding across many states. We took this opportunity to examine how the Niobrara River responded in the two years following the flood by sampling the river for two years post-flood and comparing our results with historic values. Based on historic water quality data we found a positive relationship between phosphorus and discharge, leading us to predict elevated phosphorus levels in our post-flood samples. As expected, our results show that phosphorus increased in 2019 following the flood. In contrast, historic nitrogen levels were found to have no such relationship with discharge. Diverging from this prediction, nitrogen was shown to not only to be diluted, but the daily load of inorganic nitrogen was significantly reduced as well. In agriculturally dominated regions it is often expected for nutrients levels to increase following runoff events. Our results suggest that nitrogen is not being sourced from the surrounding environment.

Linh Nguyen (Primary Presenter/Author,Co-Presenter/Co-Author)
University of Vermont, lnguye11@uvm.edu;

Ali Javed (Co-Presenter/Co-Author)
University of Vermont, Ali.Javed@uvm.edu;

Donna Rizzo (Co-Presenter/Co-Author)
University of Vermont, drizzo@uvm.edu;

Scott Hamshaw (Co-Presenter/Co-Author)
University of Vermont, shamshaw@uvm.edu;

Abstract: To ensure high quality drinking water, a common issue that needs addressing is the presence and excessive amount of turbidity in drinking water sources. Recent years have seen a rise in popularity of turbidity sensors deployed at water monitoring stations that allow for continuous suspended sediment monitoring. Such monitoring networks collect massive volumes of high-frequency, real-time streamflow (Q) and turbidity (C) data, offering opportunities for the development of new data-driven analysis methods. This research presents an event-based analysis tool to 1) process high-frequency C, Q data, 2) automate detection and delineation of storm events using C, Q data simultaneously, 3) visualize C-Q hysteresis, and 4) relate storm events across multiple stations. We demonstrate the application of the tool using sensor data collected at several USGS water stations in the Upper Esopus Creek watershed in New York that experience significant erosion resulting in high turbidity. The results will assist watershed managers in designing and assessing the effectiveness of erosion mitigation strategies. This tool can be adapted to other water quality constituent investigations such as solutes and nutrients at the event scale.

Qipei Shangguan (Primary Presenter/Author)
University of Montana, qipei.shangguan@umontana.edu;

Chun-Ze Lai (Co-Presenter/Co-Author)
University of Montana, czlai1981@hotmail.com;

Cory Beatty (Co-Presenter/Co-Author)
University of Montana, Department of Chemistry and Biochemistry, cory.beatty@mso.umt.edu;

Fischer Young (Co-Presenter/Co-Author)
University of Montana, Department of Chemistry and Biochemistry, fischer.young@umconnect.umt.edu;

Reggie Spaulding (Co-Presenter/Co-Author)
Sunburst Sensor, LLC, reggie@sunburstsensors.com;

Michael DeGrandpre (Co-Presenter/Co-Author)
University of Montana, michael.degrandpre@umontana.edu;

Abstract: Total alkalinity (AT) is an important parameter in the study of aquatic biogeochemical cycles, chemical speciation modelling and many other important fundamental and anthropogenic (e.g. industrial) processes. We know little about its short term variability, however, because studies are based on traditional bottle sampling typically with coarse temporal resolution. In this work, an autonomous AT sensor, named the Submersible Autonomous Moored Instrument for Alkalinity (SAMI-alk), was tested for freshwater applications. A comprehensive evaluation was conducted in the laboratory using freshwater standards, demonstrating good stability, precision and accuracy (±0.1-0.4%) over the AT range from 800 to 3000 µmol/L. Three SAMI-alks were deployed from19 September to 11 October 2019 in the Clark Fork River, MT, USA, with a suite of other sensors. These data reveal the complex AT dynamics that are typically missed by coarse sampling. We observed AT diel cycles as large as 60-80 µmol/L, as well as a rapid change caused by a runoff event. Significant errors in inorganic carbon system modelling result if these short-term variations are not considered. This study demonstrates both the feasibility of the technology and importance of high resolution AT measurements.

Emily Elliott (Primary Presenter/Author)
University of Pittsburgh, eelliott@pitt.edu;

Anusha Balangoda (Co-Presenter/Co-Author)
University of Pittsburgh, ANB287@pitt.edu;

Rebecca Forgrave (Co-Presenter/Co-Author)
University of Pittsburgh, ref61@pitt.edu;

Catherine Zidar (Co-Presenter/Co-Author)
University of Pittsburgh, caz47@pitt.edu;

Richard Dabundo (Co-Presenter/Co-Author)
University of Pittsburgh, RCD35@pitt.edu;

Elijah Hall (Co-Presenter/Co-Author)
University of Pittsburgh, edh47@pitt.edu;

Abstract: In 2015 and 2019, the Ohio River experienced record-setting Microcystis blooms, where toxins released from these blooms stretched 600 and 300 miles, respectively. We explored long-term trends (1976-2020) in the concentration and flux of nutrients in the Ohio River using the Weighted Regressions on Time, Discharge, and Season method (WRTDS). USGS discharge records at four monitoring locations were coupled with nutrient chemistry measured by the Ohio River Valley Water Sanitation Commission (ORSANCO). We evaluated hydrological shifts in Ohio River discharge using long-term daily average flow during bloom years. Our results document decreases in flow-normalized total nitrogen and total phosphorus concentrations and fluxes in the Ohio River at three of four monitoring locations. Our analyses indicate that both bloom years were characterized by extremely wet springs followed by late summer dry conditions. Our findings suggest that abnormally long residence times, coupled with warm water and low-flow drought periods, fueled excessive Microcystis growth in 2015 and 2019. Our results highlight the need to develop 1) best management practices that decouple extreme precipitation events from nutrient delivery, and 2) reservoir management approaches that modulate discharge during drought conditions.

Jun Xu (Primary Presenter/Author)
LSU, yjxu@lsu.edu;

Xiaoyue Ni (Co-Presenter/Co-Author)
LSU, 1096312319@qq.com;

Zhen Xu (Co-Presenter/Co-Author)
LSU, zhenxu1988@gmail.com;

Lee Potter (Co-Presenter/Co-Author)
Louisiana State University, lpotte6@lsu.edu;

Abstract: Light conditions are known as an important factor affecting photosynthesis and respiration of aquatic ecosystems. However, knowledge is still limited about weather effects on the attenuation of photosynthetically active radiation (PAR) in shallow eutrophic lakes. This study aimed to investigate vertical distribution of PAR under different weather conditions and their influence on dissolved oxygen (DO) and other biotic water parameters. From January to June 2020, a series of in-situ measurements were conducted at University Lake in Baton Rouge, Louisiana during sunny, partly cloudy, cloudy, and rainy days. The measurements included light irradiance and PAR in the air, at the depth of 16 cm and 50-75 cm below water surface. Other parameters including colored dissolved organic matter (CDOM), DO, fluorescence, and turbidity of the lake water were also recorded. Preliminary results showed a strong effect of weather conditions on PAR distribution. Attenuation of light was high during the rainy days, corresponding with increased CDOM and turbidity. DO concentration of the lake water increased with increasing light irradiance. This presentation will discuss how daily weather changes could affect dissolved carbon dioxide in water bodies because of the change in PAR attenuation.

Jacob Fries (Primary Presenter/Author)
University of Notre Dame, jfries3@nd.edu;

Shannon Speir (Co-Presenter/Co-Author)
University of Arkansas, slspeir@uark.edu;

Jennifer Tank (Co-Presenter/Co-Author)
University of Notre Dame, jtank@nd.edu;

Abagael Pruitt (Co-Presenter/Co-Author)
University of Notre Dame, apruitt2@nd.edu;

Ursula Mahl (Co-Presenter/Co-Author)
University of Notre Dame, Ursula.H.Mahl.1@nd.edu;

Todd Royer (Co-Presenter/Co-Author)
O’Neill School of Public and Environmental Affairs, Indiana University, Bloomington, tvroyer@indiana.edu;

Abstract: Agricultural land use has altered sediment transport dynamics in small watersheds across the Midwestern US. During the fallow season, bare fields are vulnerable to erosion, and planting winter cover crops (CC) may reduce sediment loss to the waterways, thereby decreasing downstream transport. We used a multiyear record of high-frequency sonde data from two agricultural watersheds to examine turbidity (as NTU) patterns using hysteresis (HI) and flushing indices (FI) for 140 individual storm events. Patterns in HI and FI were similar between watersheds; we found 74% of storms had positive HI, suggesting sediment was rapidly mobilized with storm onset. Typically, FI was also positive, suggesting there was a “first flush” of sediments as flows increased. Unexpectedly, there were no clear patterns in HI or FI relative to storm size. In the watershed with low CC coverage (<30%), NTU increased rapidly (slope=1) with flow increases, while in the watershed with high CC coverage (>65%), we saw gradual increase during storms (slope=0.4). Cover crops may be effective at mitigating sediment flushes from agricultural fields during storms and could provide resilience in the face of increased storm frequency and intensity under future climate.

Adam Ward (Primary Presenter/Author)
Indiana University, adamward@indiana.edu;

Steven Wondzell (Co-Presenter/Co-Author)
US Forest Service, Steven.Wondzell@usda.gov;

Abstract: In-stream DOC dynamics have been studied by two largely disparate groups of scholars, with distinctly different motivations. Catchment hydrologists have traditionally used in-channel DOC as a basis for inferring runoff generation mechanisms, almost exclusively requiring that sources of water and DOC are coupled. Through this lens, DOC must be generated in the catchment and transported into the stream channel. Separately, stream ecologists take DOC as part of the aquatic carbon cycle and consider its role in sustaining ecosystems. Through this lens, DOC is part of a continuum of size fractions and need not be associated with an inflow of water. Here, we consider how differences in conceptual models have led to diverging but potentially complementarily explanations for in-channel DOC. We present evidence from a headwater mountain stream in the Cascade Mountains (Oregon, USA) critically evaluating the evidence for previously published conceptual models, and conclude with a hypothesized conceptual model for the channel itself as the source of observed DOC hysteresis in response to storm events.

Tyler Hampton (Co-Presenter/Co-Author)
University of Waterloo, tyler.hampton@uwaterloo.ca;

Nandita Basu (Co-Presenter/Co-Author)
University of Waterloo, nandita.basu@uwaterloo.ca;

Puvaanah Arrumugam (Primary Presenter/Author)
University of Waterloo, parrumug@uwaterloo.ca;

Abstract: Dissolved organic carbon (DOC) from riverine export is the largest flux of reduced carbon from land to ocean (0.25 Pg of carbon annually). DOC originates from the decomposition of organic matter, fueling local riverine ecosystem metabolism and the global carbon cycle. DOC’s transport from land to sea is predominantly governed by climate, catchment hydrology, and landscape features. DOC flux in 80% of U.S. catchments are transport-controlled rather than source-controlled, indicating DOC flux is regulated by the movement of water, instead of DOC’s rate of production and abundance. However, changing precipitation and temperature regimes have modified hydrological regimes and may in turn alter DOC flux. We analyzed 28 watersheds in the U.S. with long term data to explore temporal changes (20 years) in (1) DOC flux (2) concentration-discharge relationships, and (3) precipitation and temperature regimes paired with DOC export. We applied the weighted regressions for time, discharge and seasons (WRTDS) model, combined with linear and multiple linear regression, to explore temporal and spatial trends in DOC flux. Mapping changes on a large spatial scale helps to identify areas with significant changes to DOC patterns and understand changes in carbon export.

KathiJo Jankowski (Co-Presenter/Co-Author)
U.S. Geological Survey, Upper Midwest Environmental Sciences Center, kjankowski@usgs.gov ;

Denise Bruesewitz (Co-Presenter/Co-Author)
Colby College, dabruese@colby.edu;

Megan Johnston (Co-Presenter/Co-Author)
Emory University, megan.irene.johnston@emory.edu;

Jeffrey Houser (Co-Presenter/Co-Author)
USGS Upper Midwest Environmental Sciences Center, jhouser@usgs.gov;

Douglas Baumann (Co-Presenter/Co-Author)
The University of Wisconsin - La Crosse, dbaumann@uwlax.edu;

Barbara Bennie (Co-Presenter/Co-Author)
The University of Wisconsin - La Crosse, bbennie@uwlax.edu;

Molly Van Appledorn (Co-Presenter/Co-Author)
U.S. Geological Survey, Upper Midwest Environmental Sciences Center, mvanappledorn@usgs.gov;

Taryn Waite (Primary Presenter/Author)
Colby College, trwait21@colby.edu;

Abstract: Climate-related changes in the frequency and intensity of storms will have important consequences for the hydrology and biogeochemistry of rivers. However, our understanding of storm-related biogeochemical dynamics in large rivers lags that of small streams. To fill this gap, we used high-frequency sensor data collected during four consecutive summers from a main channel and backwater site of the Upper Mississippi River. We identified high discharge events and calculated event concentration-discharge responses for both physical-chemical (nitrate, turbidity and FDOM) and biological (chlorophyll-a and cyanobacteria) parameters using metrics of hysteresis and slope. We found a range of responses across events, particularly for nitrate, which lacked general patterns. Although FDOM and turbidity exhibited more consistent responses across events, contrasting hysteresis metrics indicated that FDOM was flushed to the river from more distant sources than turbidity. Lastly, we found that the event characteristics best explaining concentration responses differed between sites, with event magnitude more frequently related to responses in the main channel, and antecedent wetness conditions associated with variation in responses in the backwater. Our results indicate that understanding event responses in large rivers will require incorporating lentic-lotic diversity across large spatial scales.

Y. Lu (Primary Presenter/Author)
University of Alabama, yuehan.lu@ua.edu;

Abstract: The export of terrestrial dissolved organic matter (DOM) to streams plays a central role in mediating inland water carbon cycles. Here, we examined high-resolution time series (every 15 minutes) of fluorescent DOM over one year in a Coastal Plain forested stream in the southeastern United States. We recorded 25 storm events during the study period that occurred only on 8% of total days but were responsible for 38% of DOC export. Storm export of DOM was mediated primarily by the duration of antecedent dry periods and storms. Specifically, the hysteresis index decreased with the length of the antecedent dry period but increased with the prior-storm discharge. During five recorded pairs of successive storm events, preceding storms lowered the availability of watershed DOM to be flushed in succeeding ones. The flushing index was positively correlated with the duration of a storm's rising limb, and long-lasting storms correlated more with flushing than with diluting. These observations demonstrate that dry periods were important for restoring mobilizable DOM in soils, and rainfalls of extended duration can effectively leach and transport DOM from distal and diffuse sources.

Andrew Robison (Primary Presenter/Author)
University of New Hampshire, andrew.robison@unh.edu;

Wilfred M. Wollheim (Co-Presenter/Co-Author)
University of New Hampshire, wil.wollheim@unh.edu;

Lauren Koenig (Co-Presenter/Co-Author)
University of Connecticut, Lauren.Koenig@uconn.edu;

Jody Potter (Co-Presenter/Co-Author)
University of New Hampshire, jody.potter@unh.edu;

Lisle Snyder (Co-Presenter/Co-Author)
University of New Hampshire, lisle.snyder@unh.edu;

William H McDowell (Co-Presenter/Co-Author)
University of New Hampshire, bill.mcdowell@unh.edu;

Abstract: While streams and rivers have been identified as important locations of carbon dioxide (CO2) emissions, uncertainty remains in the temporal variability of these emissions. Of particular interest are the dynamics of CO2 emissions during storms, which may affect the transport of CO2 from the terrestrial environment, the processing of organic carbon within the stream, and the gas transfer velocity. The development of high frequency pCO2 sensors enables monitoring of CO2 in these ecosystems at temporal scales necessary to analyze CO2 dynamics during storms. We deployed such sensors at ten streams and rivers spanning a watershed size and land use gradient to investigate the impact of storms on CO2 emissions. We find that while CO2 emissions increase during storms, the contribution to annual emissions is not significantly different than the time-weighted average. The magnitude of storm CO2 emissions appears to be affected by watershed size and wetland cover, possibly a result of the relative balance of internal and external sources of CO2. These findings have important implications for monitoring of CO2 emissions from lotic ecosystems broadly, suggesting monitoring of storms may not be necessary to accurately estimate mean emission rates.