Eric Traub Thesis Defense!

You are invited to attend Eric Traub’s  public MS thesis defense in Geology.

“The Effects of Biogeochemical Sinks on the Mobility of Contaminants in an Area Affected By Acid Mine Drainage, Huff Run, Ohio.”

(Co-Advisors: David Singer and Anne Jefferson)

Monday, Feb. 22, 12:30 pm in McGilvrey Hall, room 339, Kent State University

Anne’s reads in 2016

For the second year in a row, I’ll be keeping track of the academic literature I read. This storify will serve as a roughly reverse chronological listing of that literature, with occasional color commentary.

Scroll to the bottom to find out how it went and what I read in 2015.

2015 in review (with pictures)

2015 was an incredible year of scientific adventures for the Watershed Hydrology lab. Here are some of our highlights:

Figure 3 from Reilly, D., Singer, D., Jefferson, A., and Eckstein, Y., 2015. Identification of Local Groundwater Pollution in Northeastern Pennsylvania: Marcellus Flow-back or Not?, Environmental Earth Sciences, 73(12): 8097-8109. doi:10.1007/s12665-014-3968-0.


These Piper diagrams show the geochemistry of flowback water from fracking operations in the Marcellus shale of Pennsylvania, relative to literature values of groundwater contaminated with road salt, septic waste, and animal manure. The information on these Piper diagrams, along with other geochemical analyses, were used to assess whether alleged contamination of rural, residential well water in northeastern Pennslyvania was likely to be due to the extensive shale gas development occurring in the area. Based on major and trace ion water chemistry, Anne and her coauthors found no evidence of flowback fluid in the tested residential wells. While this work was not definitive proof that flowback contamination is entirely absent, it was a good reminder that contamination of rural, residential wellwater is much more likely to come from leaky septic systems and other less exotic sources.

Figure 3 from Reilly, D., Singer, D., Jefferson, A., and Eckstein, Y., 2015. Identification of Local Groundwater Pollution in Northeastern Pennsylvania: Marcellus Flow-back or Not?, Environmental Earth Sciences, 73(12): 8097-8109. doi:10.1007/s12665-014-3968-0

Figure 7a from Jefferson, A., Bell, C., Clinton, S., and McMillan, S. 2015. Application of isotope hydrograph separation to understand urban stormwater dynamics, Hydrological Processes, 29(25): 5290-5306. doi: 10.1002/hyp.10680.


This year I was pleased to find a way to merge her interests in urban hydrology and using water stable isotopes as tools in hydrology. We published a paper that shows that stormwater detention ponds produce distinctive isotopic signatures that can be used to trace the influence of stormwater on the receiving urban streams. Using a two end-member mixing model, my team showed that a stormwater pond in suburban Charlotte, North Carolina had an outsized contribution to streamflow during hydrograph recession periods, thanks to the temporary storage and slow release of water from the pond. For more on this project, you can read the paper or take a look at slides from the talk I gave at the Geological Society of America meeting in November.

Figure 4 from Jarden, K.M., Jefferson, A., and Grieser, J.M. 2015. Assessing the effects of catchment-scale green infrastructure retrofits on hydrograph characteristics. Hydrological Processes, online ahead of print. doi: 10.1002/hyp.10736.


Rain gardens and their fancier cousins, bioretention cells, are popping up all over the urban landscape. But how well do they work? Engineers have done an excellent job designing and testing them at the site scale, but few studies have been able to test their effectiveness at larger scales. What happens when a bunch of rain gardens and bioretention cells are added to a neighborhood? Do they make a significant difference in the stormwater runoff entering the local stream? Thanks to a brilliant study design by Cleveland Metroparks, Anne and graduate student Kimm Jarden were able to answer this question with a resounding “Yes, but…” Their paired watershed study showed that the green infrastructure retrofits can very effective at reducing peak and total stormflow, but that seemingly small differences in design and construction can have big ramifications for performance at the neighborhood scale.

This research was highlighted in a press release issued by Kent State in mid-November and was featured in several local news articles.

A bioretention cell from our study. Photo by A. Jefferson, August 2015.

A bioretention cell from our study. Photo by A. Jefferson, August 2015.


Bioretention cells, like the one pictured, are designed to capture stormwater runoff and infiltrate it into the ground. Plants use and evapotranspire some of the infiltrated water. In some designs, the infiltrated water moves deeper and recharges groundwater, while in other designs, a perforated underdrain at the bottom of the bioretention cell eventually delivers the water to the storm sewer system – ideally, after the storm has passed.

The picture above is one of the bioretention cells Anne and her collaborators studied in the project above. This bioretention cell has an underdrain, so it probably wasn’t super-helpful at reducing total runoff volumes, but did delay water entering the storm sewer system and receiving stream.


Freeze core skillfully collected by graduate student Stuart Baker from a restored stream. Photo by S. Baker.


On Thursday at the American Geophysical Union meeting,  graduate student Stuart Baker gave a talk presenting results of his in-progress M.S. thesis research. Stuart’s research examines how subsurface properties of streambeds change in the months and years following stream restoration. Stuart employed a high density array of piezometers to make repeated hydraulic conductivity and head measurements in two streams. At the end of his field work, he extracted freeze cores (“gravel popsicles”) to see if areas near the head of constructed riffles had measurable differences in sediment size and geochemistry relative to riffle tails or pool positions.

The experimental set up from: Griffith, E., Ortiz, J. and Jefferson, A. 2015. HANDS-ON OCEANOGRAPHY. Mimicking the Rayleigh Isotope Effect in the Ocean. Oceanography. 28 (4).


As part of an NSF-funded project aiming to devise and test methods for improving stable isotope content understanding amongst geoscience undergraduates, my collaborators (with a little help from me) developed a lab module on Rayleigh distillation. This module can be used in an environmental geochemistry class or oceanography class, and is adaptable into a variety of formats (with or without isotope instrumentation, or even just as a data analysis exercise). We’ve written it up so that others can adopt or adapt the technique, and our paper is appeared in the December 2015 issue of Oceanography. The paper and affiliated files (which are open access) have everything you need to use the exercise in your classroom.

West Creek, October 2015. Photo by A. Jefferson


The West Creek watershed has been a major focus of our research for the past 3 years, because it is (a) emblematic of the problems facing urban streams; (b) a place where a lot of effort is being put into managing stormwater; and (c) one of the most beautiful urban streams I have ever seen.  This shot of West Creek in Parma, Ohio, was taken on October 24th, when the Watershed Hydrology class field trip visited the stream and nearby Cleveland Metroparks Watershed Stewardship Center.

An innovative cleanup effort in Baltimore’s urban waterways. Photo by A. Jefferson, November 2015.


At the Geological Society of America annual meeting in Baltimore, I sought out a fantastic example of an innovative and successful approach to improving urban water quality. The Inner Harbor Water Wheel, affectionately known as Mr. Trash Wheel, is moored at the mouth of Jones Falls, an ultra-urban stream that drains much of the City of Baltimore. Floating rubbish is diverted to Mr. Trash Wheel by booms that span the stream, and the stream’s current powers rakes and a conveyor belt that lift the trash out of the water and deposit it into a dumpster barge. Solar panels power the operation when the stream’s velocity is insufficient to power the water wheel. Thousands of pounds of trash can be removed from the stream per day. As an added bonus, Mr. Trash Wheel has a charismatic Twitter account. Read more about the water wheel at Southern Fried Science and help fund a second one.

Anne on a roof, covered with brown vegetation, dramatic sky

On the (dormant) green roof at Cleveland Metroparks’ Watershed Stewardship Center, April 2015


I can’t resist the terrible pun about my work taking me to new heights in 2015. I have a new project monitoring five pieces of green infrastructure at Cleveland Metroparks’ Watershed Stewardship Center, and I’m having a blast learning the ins and outs of this particular site’s green roof, bioretention cells, enhanced swale, and constructed wetland. I have two new graduate students and two fantastic collaborators working with me on the project, and I’m anticipating the arrival of a postdoctoral scholar soon. I think it’s safe to say that my urban hydrology research looks set to soar in 2016.

Research on street-scale bioretention featured in the news

A bioretention cell from our study. Photo by A. Jefferson, August 2015.

A bioretention cell from our study. Photo by A. Jefferson, August 2015.

Our work with Cleveland Metroparks on assessing the effects of retrofitting bioretention cells, rain gardens, and rain barrels into residential neighborhoods in Parma, Ohio has been been featured in several news stories thanks to a nice press release issued by Kent State.

On November 20th, ran the story: “Kent State University, Cleveland Metroparks project reduces storm water, pollution in Parma neighborhoods

The study was also featured in’s “best of the beat” roundup on November 22.

The local Record-Courier ran a fantastic feature on the work (unfortunately behind a paywall) on November 30th. They called the article “Kent State professor studies rain gardens’ effect on storm water runoff.

To read the full scientific story of the work, check out our recent publication:
Jarden, K.M., Jefferson, A., and Grieser, J.M. 2015. Assessing the effects of catchment-scale green infrastructure retrofits on hydrograph characteristics. Hydrological Processes, online ahead of print. doi: 10.1002/hyp.10736.

Green infrastructure research featured on Kent Wired

Kent Wired, the electronic version of Kent State University’s student media, ran a story on Saturday about the work Kimm Jarden and I have been doing on the effectiveness of green infrastructure retrofits in a neighborhood in Parma, Ohio.  Hopefully I’ll have more to say about this in the next few days. In the meantime, if you want a glimpse of what we’ve been up to, you can check out the news article here.

How low will they go? The response of headwater streams in the Oregon Cascades to the 2015 drought

From a distance, Anne has been watching an incredibly unusual summer play out in the Pacific Northwest, following a winter with far less snow (but more rain) than usual. Folks on the ground in Oregon have been collecting data on the response of the Oregon Cascades streams to “no snow, low flow” conditions. Anne is making minor contributions to the following poster, to be presented in Session No. 291, Geomorphology and Quaternary Geology (Posters) at Booth# 101 on Wednesday, 4 November 2015: 9:00 AM-6:30 PM.


LEWIS, Sarah L.1, GRANT, Gordon E.2, NOLIN, Anne W.1, HEMPEL, Laura A.1, JEFFERSON, Anne J.3 and SELKER, John S.4, (1)College of Earth Ocean and Atmospheric Sciences, Oregon State University, Corvallis, OR 97331, (2)Pacific Northwest Research Station, USDA Forest Service, 3200 SW Jefferson Way, Corvallis, OR 97331-8550, (3)Department of Geology, Kent State University, Kent, OH 44242, (4)Biological & Ecological Engineering, Oregon State University, Corvallis, OR 97331,

Larger rivers draining the Oregon Cascades are sourced from headwater systems with two distinct runoff regimes: surface-flow dominated watersheds with flashy hydrographs, rapid baseflow recession, and very low summer flows; and spring-fed systems, with slow-responding hydrographs, long baseflow recession, and summer flow sustained by deep groundwater fed coldwater springs. Our previous research has explored these differences on both the wet west-side and dry east-side of the Cascade crest, as expressed in contrasting discharge and temperature regimes, drainage efficiency, low and peak flow dynamics, and sensitivity to snowpack and climate change scenarios. In 2015, record low winter snowpack combined with an anomalously dry spring resulted in historically low flows across our research sites and throughout Oregon. These extreme meteorological conditions, equivalent to a 4°C warming scenario, offer an exceptional opportunity to witness how these contrasting stream networks might respond to anticipated changes in amount and timing of recharge.
Conceptually, channel network response to decreasing discharge may involve both lateral and longitudinal contraction. Lateral contraction, the decrease of wetted channel width and depth, occurs in both surface-flow and spring-fed streams as flows diminish. Longitudinal contraction may be expressed as (a) a gradual drying of the stream channel and downstream retreat of the channel head, (b) a “jump” of the channel head downstream to the next spring when an upper spring goes dry, or (c) no change in channel head despite diminishing flows. We hypothesize that while individual stream channels may display a combination of these dynamics, surface-flow and spring-fed watersheds will have distinctive and different behaviors. We field test our hypothesis by monitoring channel head locations in 6 watersheds during the low flow recession of 2015, and repeatedly measuring discharge, water quality and hydraulic geometry at a longitudinal array of sites along each surface-flow or spring-fed channel. The resulting data set can be used to explore the fundamental processes by which drainage networks accommodate decreasing flows.

Hydrologic response to watershed metrics describing urban development and mitigation with stormwater control measures

Watershed Hydrology lab collaborator and Ph.D. candidate Colin Bell will be giving a talk in T106. From Green Roofs and Gutters to Urban Streams: Advancing Urban Watershed Hydrology through Innovative Field and Modeling Approaches on Monday, 2 November 2015 at 2:25 pm in Room 342 (Baltimore Convention Center).


BELL, Colin D., Agricultural and Biological Engineering, Purdue University, West Lafayette, IN 47907, MCMILLAN, Sara K., Agricultural and Biological Engineering, Purdue University, West Lafayette, IN 47907-2093, JEFFERSON, Anne J., Department of Geology, Kent State University, Kent, OH 44242 and CLINTON, Sandra, Department of Geography and Earth Sciences, University of North Carolina at Charlotte, Charlotte, NC 28223,

Stormwater control measures (SCMs) are designed to mitigate changes in hydrologic response to hydrometeorological forcing caused by urban development. Total imperviousness (TI) is a metric that effectively quantifies this urban development, but does not contain information about the extent of SCM mitigation within the watershed. The hydrologic records of 16 urban watersheds in Charlotte, NC spanning a range of TI (4.1 to 54%) and mitigation with SCMs (1.3% to 89%) were analyzed to identify which of a suite of easily-determined watershed metrics best predict hydrologic behavior. We tested the watershed metrics TI, percent forested coverage, impervious area unmitigated by SCMs, effective impervious area, percent SCM-mitigated area, and a newly-developed metric called the mitigation factor. Linear models proved TI to be the best predictor of the 10th, 30th, 50th, 70th, and 90th percentiles of the distributions of peak unit discharge and rainfall-runoff ratios. In addition, TI was the best predictor of a watershed’s ability to buffer small rain events and the rate at which a stream responds once that buffering capacity is exceeded. Additional variables describing hydrograph record flashiness and water yield were best correlated to unmitigated imperviousness and forest coverage, respectively. For the range of watersheds considered, simple metrics that quantify SCM mitigation of both total watershed area and impervious area were neither the strongest primary control nor a consistent, secondary control on storm event behavior across sites. The dominance of TI as a control on hydrology over metrics of stormwater mitigation could either be attributed to the range of sites considered (14 out of 16 sites had less than 20% SCM mitigated area) or because the watershed metrics were not able to consider the spatial arrangement of impervious surfaces and SCMs. Our results have implications for policy makers designing standards that seek to minimize stream ecosystem degradation due to hydrologic disturbances from urbanization.

Quantifying the influences of stormwater control measures on urban headwater streamflow

The Watershed Hydrology Lab will be at the Geological Society of America meeting in November in Baltimore. Anne will be giving an invited talk in the Urban Geochemistry session (T32) on Sunday, November 1st at 9 am in BCC room 308. Here’s what she’ll be talking about:

Quantifying the influences of stormwater control measures on urban headwater streamflow

Anne Jefferson1, Colin Bell2, Sara McMillan2, and Sandra Clinton3
1. Department of Geology, Kent State University, 221 McGilvrey Hall, Kent, OH 44242 USA. Phone: 1-330-672-2746 Email:
2. Department of Agricultural and Biological Engineering, Purdue University, West Lafayette, Indiana, USA.
3. Department of Geography and Earth Sciences, University of North Carolina at Charlotte, 9201 University City Boulevard, Charlotte, NC 28223, USA.

Stormwater control measures are designed to mitigate the hydrological consequences of urbanization, but their as-built effectiveness in altering patterns of urban streamflow remains poorly quantified. Stream gaging and water stable isotopes were used to understand the effects of stormwater ponds and wetlands on hydrograph characteristics and water sourcing in four urban headwater streams in Charlotte, North Carolina. At the small watershed scale (0.15-1.5 km2), runoff ratio and peak discharge are more strongly related to impervious area than area treated by stormwater controls. For one stream during 10 events, we used stable isotopes to quantify contributions of retention pond discharge to streamflow, taking advantage of the unique isotope signature of pond outflow. The pond, which drains 25% of the watershed’s impervious area, contributed an average of 10% (0-21%) of the streamflow on the rising limb and 12% (0-19%) of discharge at peak flow. During recession, this pond contributed an average of 32% (11-54%) of the stream’s discharge, reflecting the pond’s design goals of temporarily storing and delaying runoff. The isotopic signature of the pond’s discharge also reveals varying water residence times (hours to weeks) within the structure, which may have implications for nutrient and metal fluxes into the stream. Our results suggest that even when individual stormwater control measures are working as designed, they are insufficient to fully mitigate the effects of urbanization on stream hydrology. They also demonstrate the combination of traditional hydrometric and tracer-based techniques can reveal a nuanced view of stormwater influences on urban streams. Such hydrological nuance will be necessary to develop strong mechanistic understanding of biogeochemical processes in urban streams and watersheds.

Retrofitting stormwater retention on headwater streets: hydrologic effects of catchment-scale green infrastructure

At the Geological Society of America meeting, Anne will be giving an invited talk in (T106) From Green Roofs and Gutters to Urban Streams: Advancing Urban Watershed Hydrology through Innovative Field and Modeling Approaches. On Monday, November 2nd, at 1:35 pm in BCC room 342, Anne will be talking about:

Retrofitting stormwater retention on headwater streets: hydrologic effects of catchment-scale green infrastructure
Anne J. Jefferson1*, Kimberly M. Jarden1, and Jennifer M. Grieser2
1. Department of Geology, Kent State University, 221 McGilvrey Hall, Kent, OH 44242;
2. Cleveland Metroparks, 2277 W Ridgewood Dr, Parma, OH 44134
*corresponding author
The detrimental effects of urban stormwater can be lessened by disconnecting impervious surfaces and redirecting runoff to stormwater control measures, but retrofitting stormwater ponds into fully-developed urban landscapes is challenging. Decentralized green infrastructure, such as rain barrels, rain gardens, and street-connected bioretention cells, may be a more feasible and attractive approach, but the catchment-scale effectiveness of such retrofits is poorly understood. In a residential neighborhood in suburban Cleveland, Ohio, a before-after-control-impact design, in which streets served as subcatchments, was used to quantify hydrologic effectiveness of street-scale investments in green infrastructure. On a residential treatment street, voluntary participation resulted in 13.5% of parcels having green infrastructure installed over a two year period. Storm sewer discharge was measured pre– and post- green infrastructure implementation and peak discharge, total runoff volume, and hydrograph lags were analyzed. Green infrastructure installation succeeded in reducing peak discharge by up to 33% and total storm runoff by up to 40%. Lag times increased following the first year’s installation of green infrastructure, in which street side bioretention cells were built with underdrains. In the second year, bioretention cells were built without underdrains and lag times did not change further. We conclude that voluntary green infrastructure retrofits that include treatment of street runoff can be effective for substantially reducing stormwater, but that small differences in design and construction can be important for determining the level of the benefit.

Woodchips and young plants in foreground, rain garden in middle distance, and houses in the background.

Example of a bioretention cell and rain garden studied in this project.

AGU Abstract: Dynamic Hydraulic Conductivity, Streambed Sediment, and Biogeochemistry Following Stream Restoration

The Watershed Hydrology Lab will be represented at the AGU Fall Meeting in December in the session on “Groundwater-Surface Water Interactions: Identifying and Integrating Physical, Biological, and Chemical Processes.”

Dynamic Hydraulic Conductivity, Streambed Sediment, and Biogeochemistry Following Stream Restoration

Anne Jefferson, Stuart Baker, and Lauren Kinsman-Costello, Kent State University, Kent, OH, United States

Stream restoration projects strive to improve water quality and degraded habitat, yet restoration projects often fall short of achieving their goals. Hyporheic exchange facilitates biogeochemical interaction which can contribute to positive water quality and habitat, but there are limited data on how restoration affects hyporheic processes. Hyporheic flowpaths can be altered by the processes and products of stream restoration, as well as the transport of fine sediment through the stream bed post-restoration. In two northeastern Ohio headwater streams, variations in hydraulic conductivity and pore water chemistry were monitored following restoration, as measures of hyporheic functioning. A second-order stream restored in August 2013, had a slight decrease in average hydraulic conductivity but an increase in heterogeneity from pre-restoration to four months post-restoration. Data collected 10 and 15 months post-restoration show continued declines in hydraulic conductivity throughout large constructed riffles. These piezometers also indicate dominance of downwelling throughout the riffles with only isolated upwelling locations. Grain size analysis of freeze cores collected in streambed sediments show differences suggesting fluvial transport and sorting have occurred since construction was completed. Pore water sampled from piezometers within the riffles had Mn2+ concentrations ten times higher than surface water, suggesting redox transformations are occurring along hyporheic flowpaths. A first-order stream reach, immediately downstream of a dam, restored in April 2014 had no significant change in average hydraulic conductivity between 1 and 2 months post-restoration, but many individual piezometers had increases of over 100% in high gradient positions or decreases of over 50% in low gradient positions. Changes in hydraulic conductivities in both restored streams are thought to be an adjustments from disturbance to a new dynamic equilibrium influenced by the morphology and sediment regime established by restoration, suggesting these are important processes to consider in the design of such projects.

One of the study streams, 3 months post-restoration.

One of the study streams, 3 months post-restoration.