The impact of surface water-groundwater interactions on water quality in an urban stream

Date of Award

May 2016

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Earth Sciences


Laura K. Lautz


Chloride, Nitrate, Road Salt, Surface water-groundwater interactions, Urban streams

Subject Categories

Physical Sciences and Mathematics


Stream restoration is a billion dollar business in the United States, and one of the processes restoration projects often try to improve is surface water-groundwater interaction. Groundwater is a key component of nutrient cycling, as it is where dissolved organic carbon and zones of anoxia can interact with the nitrate dissolved in the surface water. Groundwater also generally has greater total dissolved ions, as movement through pore space is many times slower than flowing surface water. Restoration projects that enhance surface water-groundwater interaction may help improve urban stream water quality.

In an effort to decrease floods, many cities have physically altered streams to move water downstream as quickly as possible. Engineering projects result in straightened channels, armored banks with boulders or cement, incised channels, and removed riparian vegetation, sometimes even burying streams and making them part of storm sewer systems. This process, plus the large amount of impervious surface cover in cities, results in runoff going directly into streams, to be discharged downstream, instead of infiltrating to groundwater; this causes the streams to have flashy hydrographs and lowers baseflow. Many nonpoint source pollutants are discharged to urban streams, including heavy metals, pathogens, road salt, and nutrients. Urban planning is evolving and city planners now often recognize the importance of streams, even if just for aesthetic purposes, and when it is monetarily feasible, some cities are restoring urbanized streams to more natural designs. The goal of my research is to study how surface water-groundwater interactions, promoted by restoration practices, impact water quality in urban systems.

The study site for this research is a first-order stream located in Syracuse, NY. Meadowbrook Creek emerges from a stormwater retention basin to flow 5.5 km downstream before discharging into the Erie Canal. There are two distinct reaches to the stream: the upper 4 km, which are heavily urbanized and flow along a bouldered, incised channel down the middle of a boulevard with no riparian vegetation. The most downstream 1.5 km, however, are more natural, with a meandering channel, established floodplain, and mature riparian vegetation. This dichotomy was used to investigate how effects of stream restoration, specifically reconnection of surface water to groundwater, impact surface water chemistry.

I made bi-weekly longitudinal surveys of stream water chemistry in the creek from May 2012 until June 2013. Chloride concentrations in the upstream, urbanized reach were influenced by discharge of road salt during snow melt, peaking at 1440 mg/L. However, the natural, downstream reach had less temporal variation, only peaking at 1049 mg/L, due to buffering by lower chloride groundwater discharge. In the summer, there was little-to-no nitrate in the upper reach due to limited sources and high primary productivity, but concentrations reached over 1 mg N/L in the lower reach where riparian vegetation shade the stream. When temperatures fell below freezing, nitrate concentrations in the upper reach increased to 0.58 mg N/L, but were still lower than the natural reach, which averaged 0.88 mg N/L. Urban stream restoration projects that restore floodplain connection may impact water quality by storing high salinity road runoff during winter overbank events and discharging that water year-round, thereby attenuating seasonal fluctuations in chloride. Contrary to prior findings, we observed floodplain connection and riparian vegetation may alter nitrate sources and sinks such that nitrate concentrations increase longitudinally in connected urban streams downstream of disconnected reaches.

I investigated the processes controlling chloride movement and storage in the lower floodplain further by building a 3D groundwater flow and solute transport model. The model was calibrated to hydraulic head across the floodplain and the observed baseflow groundwater discharge to the stream. The model was validated by comparing simulated chloride concentrations in the floodplain groundwater to observed field data. The main mechanism transporting chloride to the floodplain aquifer was winter overbank flood events during snowmelt or rain-on-snow precipitation, and flooding recharges floodplain groundwater directly. The model accurately predicts the maximum winter chloride concentration in the floodplain and the residence time of chloride as it is slowly released to surface water. Restoration projects that promote overbank flooding in winter have the potential to mitigate the negative ecosystem impacts of road salt in areas with high chloride contamination by temporarily storing chloride in groundwater, buffering surface water concentrations, and minimizing first-flush concentrations following periods of road salting.

Finally, I investigated the sources and sinks of nitrate in the stream to evaluate what impact restoration may have on nutrient cycling. Altered channel morphology in urban streams changes the relative importance of nutrient sinks throughout the year, limiting denitrification and promoting assimilation. Nitrate injection tests and isotopic analysis of nitrate were performed to identify sources and sinks throughout the year, along with indirect measures of autotrophic uptake. The main source of nitrate to the system appears to be leaky sewers, as shown by the nitrate isotopes and fluoride concentrations, with a potential soil nitrogen signal as well. The urban reach has little-to-no nitrate in the spring through fall due to autotrophic uptake by filamentous green algae, as indicated by measurements of uptake properties along with measures of canopy density and algae mats. When temperatures decrease in the winter, autotrophic uptake shuts off and nitrate concentrations increase. The downstream, natural reach has high nitrate concentrations throughout the year, as the assimilation sink is minimized due to riparian shading. Nitrate isotopes do not indicate denitrification despite the surface water-groundwater interaction between the stream and floodplain. Uptake will lower nitrate concentrations for a large portion of the year, but restoration efforts that promote denitrification through increased dissolved organic carbon and increased hydrologic residence time may allow for permanent removal.


Surface provides description only. Full text is available to ProQuest subscribers. Ask your Librarian for assistance.

This document is currently not available here.