Abstract
David F. Harnsberger. URBAN CHANNEL INCISION, WATER TABLE
DECLINE, AND NITRATE ATTENUATION IN FLOODPLAIN AQUIFERS,
GREENVILLE, NC (Under the direction of Dr. Michael O’Driscoll) Department of
Geological Sciences, March 2009
Water table decline in riparian areas has been linked to the incision of urban stream
channels by increased stormwater runoff. The contact of riparian groundwater with
shallow, organic-rich surface soils in riparian zones of both urban and agricultural
watersheds may be affected by channel incision. Therefore, channel incision has notable
implications for nitrate (N03‘) attenuation via denitrification and plant uptake in
groundwater of riparian, floodplain areas of urban watersheds. In this study, three
sediment cores (nine cores total; 3.5 to 6.9 m depth) were collected across transects in the
floodplains of three progressively incised, low-order Coastal Plain streams. Standard loss
on ignition (LOI) method was used to assess changes in percent organic matter (% OM)
with depth for the full length of each core. Three well transects were installed in line
with floodplain transects to determine if water table decline adjacent to incised Coastal
Plain streams notably impacts the contact of floodplain groundwater and shallow organic
matter (OM). The impact of water table decline on NOa'-attenuation in floodplain
aquifers was also assessed. Water samples were collected monthly from eighteen
monitoring wells (six per site) from October, 2007 to September, 2008 and were analyzed
for nitrate-N (NOa'-N), ammonium-N (NH4'^-N), dissolved organic carbon (DOC), and
dissolved oxygen (DO).
Results of this investigation suggest organic matter in riparian areas is not confined to
the upper few centimeters of riparian sediment profiles in low-order Coastal Plain
watersheds. Buried soil horizons and isolated hotspots of OM were found in 8 of 9 cores
at depths of 54 to 230 cm. Despite the occurrence of deep and laterally extensive peat
layers at the urban and suburban sites, water table decline caused a notable reduction in
contact between floodplain groundwater and shallow organic matter at the urban and
suburban sites. Though groundwater-organic matter contact was reduced adjacent to
incised channels, buried peats contributed high (> 6 mg/L) concentrations of DOC to
floodplain groundwater. High DOC concentrations in association with low DO
concentrations (< 1 mg/L) provided conditions suitable for denitrification across all three
floodplain aquifers.
Median N03'-attenuation (decrease, loss and/or retention) values (N = 12/site) were
calculated as a percent difference in monthly NOs'-N concentrations between up-gradient
and near-stream locations for each floodplain study site. NOs'-N concentrations were
attenuated 99%, 58%, and 46% at the rural, suburban, and urban floodplain sites. This
decrease in percentage NOs'-attenuation with increasing near-stream water table depth
(0.82 m, 0.99 m, and 1.47 m for the rural, suburban, and urban sites) suggests that water
table decline and impaired NOa'-attenuation across the urban and suburban floodplains
may be related. Though near-stream N03'-N concentrations were significantly higher at
the incised study sites than at the rural site, they were low in comparison to near-stream
concentrations in unbuffered agricultural settings of the US and Canada. Marked
variability in groundwater discharge from each floodplain, caused by variable hydraulic
conductivity of floodplain sediments at each site, prevented a clear relationship between
water table decline and the loading of NO3' to surface water at each site.
URBAN CHANNEL INCISION, WATER TABLE DECLINE, AND NITRATE
ATTENUATION IN FLOODPLAIN AQUIFERS, GREENVILLE, NC
A Thesis
presented to
the Faculty of the Department of Geological Sciences
East Carolina University
In Partial Fulfillment
of the requirements for the Degree
Master of Science in Geology
By
David F. Hamsberger
2009
URBAN CHANNEL INCISION, WATER TABLE DECLINE, AND NITRATE
ATTENUATION IN FLOODPLAIN AQUIFERS, GREENVILLE, NC
by
David F. Hamsberger
Approved By:
Director of Thesis
Dr. Michael A. O’Driscoll
Assistant Professor
ECU Dept, of Geological Sciences
Committee Member
Dr. Terri Woods
Associate Professor
ECU Dept, of Geological Sciences
Committee Member
Dr. Richard K. Spruill
Associate Professor
ECU Dept, of Geological Sciences
Committee Member
Dr. Mark M. Brinson
Professor
ECU Dept, of Biology
Chair of the ECU Dept,
of Geological Sciences Dr. Stephen J. Culver
Dean of the
Graduate School Dr. Paul Gr
Acknowledgements
First and foremost, I would like to thank my thesis advisor, Dr. Mike O’Driscoll. His
support in helping to design the project, conduct field work, and provide edits on the
many draft versions of preliminary work and this document has been invaluable.
Next, I would like to thank the NC Water Resources Research Institute (WRRI) and
the NC Ecosystem Enhancement Program (EEP) for providing funding to purchase
equipment and for helping to fund the water sample analyses.
Thirdly, I would like to extend a most formal thanks to my thesis committee. They
provided valuable feedback on the original proposal of this project and helped me avoid
major problems in the design and implementation of the original concept. Their feedback
on the many drafts of this document has also been invaluable. Thanks specifically to Dr.
Brinson for convincing me that I was not going to be able to core straight into the channel
sediments of Pomes Branch, to Dr. Spmill for helping interpret the odd hydrology at the
rural site and for his demand that the document be condensed, and to Dr. Woods for
lending us her filtration equipment.
Also critical to the completion and success of this project were the many faculty and
grad students of the ECU Geological Sciences Department. ECU Geological Sciences
faculty that were instrumental include: John Woods, Jim Watson, Dare Merrit, Dr. Reide
Corbett, Dr. JP Walsh, and Dr. Steve Culver. ECU grad students that deserve special
thanks include: John DeLoatch, Jason Soban, Brion Byers, Kolt Johnson, Laura
McKenna, Reanna Camp, Lauren Metger, Emma Hardison, David Vandervelde, Brad
Elkins, Dani Dominique, Rob Howard, David Lagomasino, and Andrew Dietsche.
Thanks specifically to Pete Parham for providing draft versions of the lateral transects
from his work in the Coastal Plain of Eastern NC. Transects presented herein are based
on his transects. Thanks to Becky Cooper for her help with the N03'-N and NH4'^-N
analyses, and to Jesse Chadwick at GUC for running 12 months of DOC analyses for us.
Thanks also to the homeowners who provided access to their urban, floodplain riparian
zones with little to no good understanding for what exactly we were up to.
Finally special thanks to my parents. Without their moral and financial support, I
probably would not have been able to finish. Thank you parents!
Table of Contents
List of Figures vi
List of Tables viii
Introduction 1
Study Design 9
Study area and geologic setting 9
Coastal Plain Geology 9
Coastal Plain Hydrology 12
Floodplain geology 14
Well transects and transect stratigraphy 16
Water chemistry 19
Data analysis 22
Results 24
Floodplain hydrogeology 24
Floodplain groundwater levels and temperature 28
Hydraulic gradients and groundwater discharge 32
Groundwater and surface water chemistry 35
Dissolved inorganic-N concentrations 35
DOC and DO concentrations 43
Chloride concentrations 47
Statistical relationships 48
Floodplain N03'-attenuation and loading to incised stream channels 52
Discussion 55
Changes in organic matter with depth 55
Effects of decline on OM contact 55
Effects of water table decline on NOs'-attenuation 58
Comparison to previous studies 63
Conclusions and Management Recommendations 68
Floodplain N-removal is affected by declining water tables 69
References 71
Appendix A: Summary Statistics of Measured Variables 80
Appendix B; Electroconductivity Response Logs 96
Appendix C: Sediment Core Logs 103
Appendix D: Percent Organic Matter Data 113
Appendix E: Hydrologic Data 116
Appendix F: Water Quality Data 119
Appendix G: Monthly Water Chemistry Data 121
List of Figures
Figure 1. Schematic of the nitrogen cycle showing plant and microbial processes that
convert N between its various end members 4
Figure 2. Study site location map showing well transect locations and study site
subwatersheds 10
Figure 3. Correlation of sediment electrical conductivity (EC) log with sediment core
taken at the location of the near-stream well nest of the urban study site showing
conductive response of peat, sandy sediments, and packed wet clay 15
Figure 4. Sample layout diagram showing location, orientation, and spacing of well
transect at the urban study site 17
Figure 5. Hydrogeology of study transects at the urban, suburban and rural sites showing
framework geology, seasonal high (marked ‘H’) water table on 4/12/2008, and seasonal
low (marked ‘L’) water table on 10/5/2007 25
Figure 6. Vertical distribution of % OM based LOI processing of samples from
floodplain sediment cores of each study site 27
Figure 7. Near-stream water table depths measured in deep, near-stream wells of each of
the study sites from 9/14/2007 to 9/4/2008 29
Figure 8. Monthly groundwater levels (m AMSL) from the shallow up-gradient (U) and
near-stream (N) wells of the A) urban, B) suburban and C) rural sites 31
Figure 9. Monthly hydraulic gradient across the floodplains of the three study sites from
10/5/2007 to 9/5/2008 33
Figure 10. Monthly discharge of floodplain groundwater from 10/5/2007 to 9/5/2008
based on a 100 m cross-sectional stream reach 34
Figure 11. Study site transects at the urban, suburban, and rural sites showing median
nitrate-N and ammonium-N (mg/L) values measured in wells at the near-stream,
floodplain, and up-gradient locations 37
Figure 12. Shallow, deep, and in-stream nitrate-N concentrations from the A) urban, B)
suburban, and C) rural floodplains for the one year study period 38
Figure 13. Groundwater and surface water ammonium-N concentrations (mg/L) across
the A) urban, B) suburban, and C) rural floodplains for the one year study period 39
Figure 14. Median nitrate-N, median ammonium-N, and DON (mg/L) from 5/30/2008
in A) shallow and B) deep wells of the urban floodplain 40
Figure 15. Median nitrate-N, median ammonium-N, and DON (mg/L) from 5/30/2008
in A) shallow and B) deep wells of the suburban floodplain 41
Figure 16. Median nitrate-N, median ammonium-N, and DON (mg/L) from 5/30/2008
in A) shallow and B) deep wells of the rural floodplain 42
Figure 17. Study site transects of the urban, suburban and rural sites showing median
annual DOC, DO values from shallow and deep wells of the three study sites 44
Figure 18. Plot of screen depth below water table vs. median DOC from groundwater
samples over the duration of the study period 45
Figure 19. Median well DO concentrations vs. screen depth for floodplain wells of each
of the study sites 46
Figure 20. Median well DO (ppm) concentrations vs. dist. (m) of wells from the stream
edge at each site 46
Figure 21. Correlations between nitrate-N and dissolved oxygen (DO) at the urban,
suburban, and rural sites 49
Figure 22. Correlations between nitrate-N and dissolved organic carbon (DOC) at the
urban, suburban, and rural sites 50
Figure 23. Monthly percent difference in NOa'-N concentrations (%) over twelve
months of chemical data at the urban, suburban and rural sites 53
Figure 24. Seasonality graphs showing variability in A) NO3' loading (g/mo), B)
average near-stream NOb'-N concentrations (mg/L), and C) discharge (m /mo) across the
three floodplains 54
List of Tables
Table 1. Annual hydraulic gradients from up-gradient to near-stream wells across
floodplains of each site 33
Table 2. Descriptive statistics for annual discharge from each floodplain 34
Table 3. Percent difference in median groundwater chloride concentrations (N = 6
months of chloride data) from up-gradient to near-stream wells across each floodplain..47
Table 4. Pearson correlation coefficients for bivariate correlation of monthly
groundwater (gw) levels (m AMSL) and monthly groundwater concentrations (mg/L)
from shallow floodplain wells (N1 and N2 locations) of each site 51
Table 5, Median NOs' concentrations and percentage attenuation (- = loss, + = gain) in
floodplain, riparian groundwater estimated from studies conducted in agricultural
catchments of the central United States, eastern United States, and Canada 65
Introduction
Anthropogenic activity in coastal watersheds has greatly altered the nitrogen (N)
cycle and has significantly accelerated the delivery of N in both organic and inorganic
forms to coastal water bodies (Anderson et ah, 2002; Boyer et ah, 2002; and Howarth et
ah, 2002). Eutrophication, and associated harmful algal blooms lead to fish kills,
increased abundance of toxic algae species, and more frequent anoxic conditions of
coastal water bodies (Anderson et ah, 2002; Burkholder, 2000). Eutrophication continues
to be the single largest pollution problem in coastal rivers and bays of the US (Howarth et
ah, 2002; National Research Council, 2000). For most coastal water systems, non-point
source contribution of nutrients in the form of surface runoff and groundwater inputs
from developed watersheds is the single largest contributor ofN to coastal waters
(Howarth et ah, 2002).
Though agriculture has been shown to be a major source of N in many coastal
watersheds (Howarth et ah, 2002), a considerable body of evidence suggests that urban
watersheds may also contribute elevated levels ofN to receiving coastal waters (Boyer et
ah, 2002; Groffman et ah, 2002; Paul and Meyer, 2001). Boyer et ah (2002) have
suggested that even small shifts in land use in forested catchments can cause large
increases in annual nitrogen export. In their study of 16 catchments across a latitudinal
profile from Maine to Virginia, urban land use was shown to increase the mass of N
applied to the surface area of a catchment. They demonstrated a positive correlation
between “disturbed” land use and total N loading, indicating the importance of
2
urbanization, in addition to agriculture, as a large human-derived source of N to coastal
watersheds.
In a review of urban land-use effects on streams, Walsh et al. (2005) recognized
several symptoms of “the urban stream syndrome.” Most important to this study is the
condition of elevated concentrations of in-stream nutrients. The sources of elevated in-
stream nutrient concentrations in urban catchments include: 1) elevated wet and dry
deposition of atmospheric N (Howarth et ah, 2002; Anderson et ah, 2002), 2) N-based
fertilizer application (Boyer et ah, 2002), 3) legacy pollutants (nutrients sourced in
fertilizer from past agricultural land use) in urban groundwater (Walsh et ah, 2005), 4)
biological fixation of elevated atmospheric N (Boyer et ah, 2002), and 5) import ofN in
food for humans and feed for animals (Boyer et ah, 2002). Though considerable efforts
have been made to reduce the amount ofN in both organic and inorganic forms that is
transferred to streams, N inputs due to sources such as atmospheric deposition and legacy
pollutants are difficult and/or expensive to control (Walsh et ah, 2005).
In addition to the sources of elevated nutrient concentrations in urban streams being
numerous and highly variable, several pathways of migration from source to stream exist.
Of primary concern to this study is the likelihood of increased concentrations of
inorganic-N in urban groundwater. Dissolved inorganic-N in urban groundwater may be
elevated due to: 1) N-based fertilizer application to residential yard space, 2) the
persistence of legacy pollutants from past agricultural land uses, and 3) discharge,
leakage, and overflow from sewage infrastructure.
3
NOa' is the species of inorganic-N that has received the most attention for its role in
limiting or exacerbating eutrophic conditions in coastal water bodies (Howarth et ah,
2002). Background NOs'-N concentrations in streams and groundwater systems vary
spatially and temporally, related to regional N-deposition and ecosystem N-retention
(Stoddard 1994). In a review ofN03‘-N concentration data obtained from the U.S.
Geological Survey’s National Water-Data Storage and Retrieval System (WATSTORE),
Madison and Burnett (1985) attempted to obtain a national perspective on the extent to
which nitrate concentrations in groundwater of the United States had been elevated by
anthropogenic activity. Data from 87,000 wells throughout the United States obtained
from WATSTORE, and an additional 36,000 wells obtained from the Texas Natural
Resource Information System (TNRIS) were analyzed. A concentration of more than 3
mg/L was arbitrarily defined as the average, background concentration of N03'-N in
groundwater of the US (Madison and Burnett, 1985). The use of this “background
concentration” as an average N03'-N concentration for unpolluted natural groundwaters
of the US may be conservative, since 50% of samples assessed in Madison and Burnetts’
study had concentrations below detection limits (Spalding and Exner 1993; USEPA
1990). The maximum N03'-N concentration established by the EPA as safe for drinking
water (the Maximum Contaminant Level or MCL) is 10 mg/L (USEPA 2003).
The cycling of N in streams is closely linked to biologic processes in riparian areas
and channel sediments (Figure 1). In riparian areas, N2 fixation, which is mediated by N-
fixing bacteria, may provide N-inputs to urban groundwater. N-fixing bacteria form
symbiotic associations with plant roots and are free living (Duff and Triska, 2000). Most
4
of the N that is fixed by N-fixing bacteria is assimilated by riparian vegetation, and is
later released to streams as throughfall, leachate from litterfall, or by decomposition of
plant tissue (Duff and Triska, 2000; Wetzel, 1975). Organic matter (OM) in decomposed
plant tissues of riparian soils is further decomposed by heterotrophic bacteria that convert
organically bound N to NH4'^ via the bacterially mediated process of ammonification. In
aerobic soils, free NH4^ is converted to NO3' by nitrifying bacteria in a two step process
called nitrification. First, NH4^ is converted to nitrite (NO2'), and then NO2’ is converted
to NO3' (Vepraskas and Faulkner, 2001).
1
1. Nj fixation
2. Decomposition
T 3. Ammonification/Mineralization
Plants 4. Nitrification
5. Denitrification
6. Plant and microbial assimilation
7.
2 Assimilatory and dissimilatory NO3 reduction to
Organic Matter/Soil
Figure 1. Schematic of the nitrogen cycle showing plant and microbial processes that
convert N between its various end members. Only processes that occur specifically in
floodplain, fresh water environments are shown.
Because N03' that is produced by nitrifying bacteria is mobile and easily transported
(Duff and Triska, 2000; Jones et al, 1995; Holmes et al., 1994) it often moves through
floodplain sediments to anoxic sites, locations or zones of the floodplain environment
where it can be denitrified. Denitrification is the bacterially mediated redox reaction
where bacteria reduce NO3' to N2 or N2O gas as a result of electron transfer. In hydric
soils (soils formed under conditions of saturation), DOC compounds leached from
5
organic substrates often provide electrons for denitrification (Vepraskas and Faulkner,
2001; Ponnamperuma, 1972). Denitrification is important in riparian zones because it is
one of the only bacterially mediated processes in the nitrogen cycle that effectively
removes N from groundwater and streams. Because redox conditions control
denitrification, the occurrence of denitrifieation in floodplain groundwater is often
limited to locations in floodplain aquifers where anoxic conditions exist (Vepraskas and
Faulkner, 2001). Studies by Denver et al. (2004) and Bohlke et al. (2002) suggest that
denitrification is inhibited when groundwater dissolved oxygen (DO) concentrations are
> 1.9 mg/L. NOs' in groundwater may also be assimilated back into plant and microbial
biomass (microbial assimilation). In hypoxic environments, NOs' may be reduced to
NH4^ by both assimilatory nitrate reduction to ammonium (ANRA), where NH4^
produced by the process is incorporated into cellular material, and dissimilatory nitrate
reduction to ammonium (DNRA), where NH4^ produced by the process is excreted back
to the hydric environment.
In floodplain aquifers where conditions fluctuate between oxic and anoxic,
denitrification is often effective at removing NOsTSl. NO3’ for denitrification in
floodplain aquifers is often: 1) imported by infiltrating rain water, 2) input by overbank
flow from flooding stream channels, and/or 3) input by groundwater from upland areas of
Coastal Plain watersheds (Denver et al. 2004; Bohlke and Denver, 1995). When NOb'
enters a zone of anoxia and labile organic matter, denitrification is initiated. NOa' may
also be generated by nitrification of NH4^ in oxic surface soils when precipitation
introduces new water to the floodplain groundwater system. NO3' generated by
6
nitrification in shallow, oxic groundwater is denitrified when it is advectively transported
by groundwater to anoxic microsites of the floodplain aquifer (Patrick and Tusneem,
1972) or when dry periods of relatively little precipitation in low-order watersheds allow
floodplain aquifers to become anoxic (Bohlke et al. 2007).
Various studies have shown that high potential denitrification (DNP) rates in organic-
rich soils are often confined to the upper few centimeters of the riparian sediment profile
(Schoonover and Willard, 2003; Groffman et ah, 2002; Burt et ah, 1999; Groffman et ah,
1996). More recent work by Gurwick et al. (2008) and Hill et al. (2004) suggests that
organic-rich soils and associated high DNP can occur at depths of up to 300 cm in
riparian sediment, though there is a question as to whether the carbon produced by buried
soil horizons is available (labile) for microbial metabolic processing. Organic matter in
surface and buried, organic-rich soils plays an important role for N-cycle dynamics in
floodplain groundwater, for two main reasons: 1) it supplies floodplain groundwater with
DOC (Baker et ah, 2000; Hill, 2000; O’Brien et ah, 1994; Dosskey and Bertsch, 1994)
that can act as an electron donor to promote denitrification (Hill et ah, 2004; Groffman et
al., 2002; Burt et al., 1999), and 2) it provides NH/ (via OM mineralization) to soil water
and floodplain groundwater (Hill, 2000; McLain et ah, 1994; McDowell et ah, 1992) that
can be nitrified under aerobic conditions to increase groundwater NOb'concentrations.
Stormwater runoff in urban watersheds has led to significant incision of stream
channels in various geographic settings (Soban, 2007; Paul and Meyer, 2001; Henshaw
and Booth, 2000). Stream channel incision, in turn, has caused significant decline of
7
water tables adjacent to incised stream channels (Hardison, 2008; Groffman et al., 2003;
Groffman et al., 2002). The decline of water tables across riparian zones in urban
catchments has been coined “riparian hydrologic drought” by Groffman et al. (2003), and
has important implications for nitrate attenuation (the net percentage decrease in NO3'
concentrations across a riparian zone that typically results from mechanisms that both
increase and decrease NO3'concentrations) in urban, floodplain groundwater (Kaushal et
al., 2008; Groffman et al., 2003; Groffman et al., 2002; Gold et al. 2001). The
established effects of water table decline on hydrogeochemical conditions in floodplain
sediments and aquifers are three-fold: 1) considerable alteration in the hydrologic and
geochemical condition of unsaturated, surface soils (Groffman, 2003), 2) decreased
contact of floodplain groundwater with organic-rich surface soils that have high
denitrification potential (Bohlke et al., 2007; Groffman et al., 2003; Groffman et al.,
2002; Gold et al. 2001), and 3) the preferential forcing of groundwater flow to deeper
hydrologic flowpaths across floodplain aquifers, where nitrate derived from upland
aquifers is less likely to be denitrified or taken up by riparian vegetation (Bohlke et al.,
2007; Groffman et al., 2003; Gold et al., 2001).
In a study of the effects of water table decline on shallow soil chemistry, Groffman et
al. (2002) suggested that urban hydrologic factors can increase the production and
decrease the consumption of N03' in urban settings. Schilling et al. (2006) have
suggested that N03' concentration increases in groundwater samples from wells across
the floodplain of their study may have been caused by the leaching and transport of NO3'
8
from unsaturated soil during a single large storm event that occurred during the three
months of their investigation.
The primary objective of this study was to determine the effects of water table decline
on NOs'-attenuation in floodplain aquifers of incised urban streams of the Greenville
area. Specifically, the following hypotheses were tested: 1) organic matter in riparian
sediment profiles will be concentrated at shallow depths (0-30 cm), 2) water table decline
adjacent to incised urban streams will reduce contact between floodplain groundwater
and surface organic-rich soil, and 3) water table decline across floodplain aquifers
adjacent to incised urban streams will cause less NOs' to be attenuated from floodplain
aquifers.
Study Design
Study area and geologic setting
This study focused on floodplains at selected locations in the catchments of three
low-order streams (small streams with very few tributaries) of the Tar River Basin, whieh
have undergone various degrees of urban development. Tomes Branch (FB), Meeting
House Branch (MHB), and Phillippi Branch (PB) (Figure 2) were classified as urban,
suburban, and rural respectively based on the percent of surface area classified as
impervious area (Total Impervious Area or TIA) in each catchment. TIA percentages
were calculated for each catchment during a previous study using impervious eover data
from the City of Greenville and the National Land Cover Database (Hardison, 2008) to
be 37% (urban), 22.1% (suburban), and 4% (rural), respectively.
Coastal Plain Geology
The study area (Figure 2) resides within the Inner Coastal Plain of eastern North
Carolina (NC). At the surface, the Coastal Plain is characterized by a series of relict
coastal terraces that slope gently towards the sea separated by scarps that run parallel to
the coast (Maddry, 1979). In the Greenville area, surface terraees are underlain by a vast,
thick, and laterally extensive wedge of unconsolidated sediment that has been roughly
eategorized by Maddry (1979) into 5 hydrogeologic units: the Black Creek, Pee Dee,
Beaufort, Castle Hayne, and Yorktown formations. Overlying the Yorktown is a
relatively thin seetion (~ 10 m; Winner and Coble, 1996) of Quaternary sediment that
includes the floodplain unit examined in this work.
10
North Carolina Pitt County
City of Greenville
GKE F.NVILLE
Fornes Branch
Phillippi Branch
Meeting House Branch
kilometers
Legend
o Study Site Locations
USGS Hydrography of Tar River Tributaries
Added Hydrography (Hardison, 2008)
Subwatersheds of Study Sites
Tar River
Figure 2. Study site location map showing well transect locations and study site
subwatersheds. Hydrography added by Hardison (2008) is based on field observations.
Background image is a cropped portion of the Greenville SE, 7.5” USGS Topographic
Quadrangle, obtained using Terrain Navigator Pro 7.5.
11
The uppermost section of the Yorktown frequently occurs as a thick, bluish gray,
low-permeability clay (Lautier, 2001). The continuity and lateral extent of the confining
clay of the Yorktown is important to geochemical processing of nitrate in unconfmed,
floodplain aquifers. In eastern Pitt County, the area of this study, the Yorktown confining
unit is composed of alternating silt and clay layers and is roughly 1 to 22 m thick, with an
average thickness of 7 m (Winner and Coble, 1996). Maddry (1979) has suggested that
the Yorktown was likely deposited during the transgressive portion of a single
cyclostratigraphic cycle. Relict channel deposits offshore of Virginia and North Carolina
that cut into and through the Yorktown (Shideler and Swift, 1972), suggest that during
the regressive portion of the cyclostratigraphic cycle responsible for the deposition of the
Yorktown, the silts and clays of the Yorktown were left exposed to erosion and fluvial
seaward transport. Finally, Maddry suggests that during a more recent marine
transgression, a thin section of terrigenous, clastic sediments was deposited by both
marine and fluvial processes.
Maddry separates these Quaternary clastic sediments into 6 distinct stratigraphic units
which he labels Qi-Qó- Qi and Q2 are described as a transgressive sequence of
sedimentary units that occur in interstream areas south of the Tar River and presumably
away from the floodplain aquifers of this study. Q3 is classified into four units based on
stratigraphic descriptions that closely resemble the stratigraphy of sediment encountered
during this study. Qsa is described as a fine to coarse, sand unit with basal pebble-sized
sediment and isolated clay-rich lenses. Dark to light green glauconite grains with cracks
and spaces between grains filled with a dark-reddish-brown iron oxide occurred in the
12
sand of Qsa. Qsb and Qsc are described as organic-rich clays and sands that contain
abundant plant debris and occur in select locations of Eastern Pitt County. Q3d is
described as organic-rich sand that occurs at or near the surface with a maximum depth of
burial of about two meters. Maddry suggests that the shallow depth of burial and fresh
appearance of the plant remains in Qsd provide evidence of it being a recent floodplain
deposit less than 8000 years old. Maddry explains that Q3 as a sequence was deposited in
stream valleys tributary to the Tar River that are likely the same valleys filled by
floodplain sediments encountered during this study.
Coastal Plain Hydrology
The hydrogeologic units of the Coastal Plain sediment wedge discussed above consist
of permeable sand, gravel, and limestone aquifers separated by confining units composed
of lower permeability clays (Winner and Coble, 1996). Recharge to these aquifers begins
as rainfall that infiltrates to the surficial aquifer at interstream areas, moves down through
underlying confining units, and then permeates back through the same confining units to
discharge to floodplains and streams of the NC Coastal Plain (Winner and Coble, 1996).
Annual rainfall estimates for the Greenville area suggest that approximately 137 cm of
rain fall in Pitt County each year, with a mean annual air temperature of about 17.4 C
(Southeast Regional Climate Center, 2007). Rainfall for the year of this study (110 cm;
October, 2007 to October, 2008) was below average due to extreme drought conditions
into the late summer and fall of 2007. Mean annual air temperature was 17.1 C.
The combined annual discharge of groundwater from deeper confined aquifers to
surficial aquifers and large, high order streams and rivers is approximately 1.3 cm per
13
year (Heath, 1975). For a watershed in Pitt, Beaufort and Craven counties, estimated
average annual discharge of groundwater from the confined Castle Hayne Limestone unit
was approximately 2.0 cm per year (Winner and Coble, 1996; Winner and Simmons,
1977). Discharge from the surficial aquifer occurs as direct discharge and/or seepage into
streams, lakes and drainage ditches, évapotranspiration from shallow surficial soils and
sediment, and upwelling through basal sediments of large rivers and estuaries (Winner
and Coble, 1996).
Coastal Plain streams and rivers can be described as low-gradient, meandering
systems that often develop broad floodplains that are subjected to frequent and prolonged
inundation (Hupp, 2000). Using a partial-duration series of average daily discharge.
Sweet and Geratz (2003) estimated that the average recurrence interval of bankfull
discharge flows for streams in the Southeastern Plain ecoregion of the NC Coastal Plain
is about 0.19 year, or about 5 bankfull events per year. According to Hupp (2000),
medium and large streams of the south-eastern Coastal Plain have a strong seasonal
variation in discharge driven by seasonal variation in évapotranspiration and rainfall.
Hupp (2000) identifies two distinctly different hydrological seasons for Coastal Plain
streams in their natural state: 1) a low-flow season from about June through to October,
which is often interrupted by tropical storms or hurricanes (Brinson, 2009, pers. comm.)
and 2) a high-flow season from about November through May, when large parts of
adjacent floodplains are often inundated.
14
Generally speaking, at least 40% of the average annual streamflow in the NC Coastal
Plain is derived from groundwater (Harden and Spruill, 2008; Harden et ah, 2003; Spruill
et ah, 1998; Winner and Simmons, 1977). Some studies of streams in the Coastal Plain
streams have found the groundwater component of surface water discharge to be on the
order of 90% (Hutchinson, 2007; Williams and Pinder, 1990). Several studies in
agricultural settings of the Coastal Plain have shown that surface water quality
(concentrations of nutrients and contaminants in surface water) are notably influenced by
groundwater quality (Harden and Spruill, 2008; Spruill, 2004; Tesonero et al. 2004;
Denver et al. 2004).
Floodplain geology
Framework geology for floodplain transects at the study sites was established via a
combination of sediment electrical conductivity logging and sediment core comparison.
A Geoprobe ™ electro-conductivity sensor was used in conjunction with a Geoprobe
direct push sediment sampling rig to conduct conductivity surveys based on a 4-6 point
grid layout that typically extended 20 m by 30 m (Appendix B). Sediment electrical
conductivity logs and sediment cores were correlated to identify peat, sand, and clay
layers in the floodplains of the study sites (Figure 3). The sandy surflcial aquifer was
targeted for installation of three wells per floodplain. Groundwater levels measured in
wells were used to calculate direction of groundwater flow at each proposed study site via
standard three-point solution (Heath, 2004).
15
Fornes Branch Core N1 Fornes Branch Conductivity Log N1
Greenville, NC Greenville, NC
Lat.35°35'41.8"N Lat.35°41.9"N
Long. 77°2r23.8" W Long.77°2r23.85"W
sediment conductivity (mS/m)
60 80 100 120 140 160 180 200 220
description
rooted surficial soil
rooted, gray peat
saturated black peat
coarse-grained sand
coarse-grained sand
- NR
coarse-grained sand
clayey orange coarse-grained sand
clayey orange medium grained sand
Figure 3. Correlation of sediment electrical conductivity (EC) log with sediment core
taken at the location of the near-stream well nest of the urban study site showing
conductive response of peat, sandy sediments, and packed wet clay. EC logs were used
in conjunction with sediment cores to position well screens just below the water table and
just above the confining unit for shallow and deep wells at each nest. Remaining EC logs
are provided in Appendix B.
16
Well transects and transect stratigraphy
After calculating direction of groundwater flow according to standard three point
procedure (Heath, 2004), nested well transects were established at each of the proposed
study sites along distinct groundwater flowpaths. Transects extended from the floodplain
edge or a short distance into the upland-floodplain interface to within 6 (+/- 1) meters of
the stream edge at each of the study sites. Three nests of two wells each were installed
across each of the floodplains. Nests were located 6 (+/- 1), 20 (+/- 1), and 34 (-f/- 1) m
from the banks at the urban (Figure 4) and rural sites and 5, 15, and 25 meters from the
banks at the suburban site. Each nest consisted of wells constructed from PVC pipe (3.2
cm inner diameter) with 50 cm screens installed at variable depths between 1.7 and 6.5 m
below ground, depending on the thickness of the unconfmed aquifer. Wells at the N1
nest locations (Figure 4) closest to each stream are referred to hereafter as “near-stream”
wells; wells at the floodplain, N2 locations are referred to as “floodplain” wells; and
wells at the floodplain edge/upland, N3 locations are “up-gradient”.
17
Figure 4. Sample layout diagram showing location, orientation, and spacing of well
transect at the urban study site. Well nests consist of shallow and deep wells screened
just below floodplain water tables and just above confining unit in sandy sediments of the
unconfined aquifer. Wells at the N1 location are “near-stream” wells, wells at the N2
location are “floodplain” wells, and wells at the N3 location are “up-gradient” for the
remainder of this manuscript. Spacing of wells for the calculation of groundwater flow
direction is also shown.
17
Figure 4. Sample layout diagram showing location, orientation, and spacing of well
transect at the urban study site. Well nests consist of shallow and deep wells screened
just below floodplain water tables and just above confining unit in sandy sediments of the
unconfmed aquifer. Wells at the N1 location are “near-stream” wells, wells at the N2
location are “floodplain” wells, and wells at the N3 location are “up-gradient” for the
remainder of this manuscript. Spacing of wells for the calculation of groundwater flow
direction is also shown.
18
Groundwater levels in transect wells were monitored monthly to estimate hydraulic
gradients and to quantify groundwater contact with surface soil at each site. One
groundwater logger in the shallow, near-stream well at each site recorded changes in
floodplain groundwater levels at 0.5 hour increments. Monthly fluctuations in water
table levels in shallow wells of each site were compared to depth of surface soils to
quantify loss of contact between groundwater and organic-rich soil for the duration of the
study period.
Ground-water discharge (Q) from floodplain sediments was estimated from monthly
head measurements collected from October 5, 2007 to September 5, 2008. Darcy’s law
(Q=KA dh/dl) was used to quantify groundwater discharge, where K was the median site
hydraulic conductivity, A was the cross-sectional area of the surflcial aquifer at a site,
and dh/dl was the hydraulic gradient across each floodplain. Cross-sectional area was
determined for a 100 m length of floodplain sediment and an aquifer thickness based on
monthly calculation of the height of the water table above the impermeable confining unit
at the floodplain, N2 well locations of each transect (monthly aquifer thickness = depth to
confining unit - water table depth in the shallow N2 well for a given month). Discharge
estimates presented in this study represent an estimate of groundwater discharge to a 100
m length of each stream from floodplains adjacent to each of the study sites. Notable
variation in K within floodplains of this study and throughout the catchments of each
stream may cause discharge estimates presented in this manuscript to vary measurably
from actual discharge at both the floodplains of this study and other discharge locations
within each watershed. Slug tests were used to estimate floodplain K in all 6 wells at
19
each site (18 total) (Fetter, 1994; Bouwer and Rice, 1976). Monthly nitrate loading from
the edge of each floodplain was calculated by multiplying the average near-stream nitrate
concentration (average of concentrations from the shallow and deep near-stream wells at
each site) by the monthly groundwater discharge calculated for a 100 m reach of each
floodplain study site.
A single sediment core was taken at each well nest location (3 cores per site) for
detailed stratigraphic analysis. The average core length was 4.5 m with a total of 60.2 m
of sediment collected for this study. Organic profiles of stream cores were established by
sampling cores at 25 cm increments from the floodplain surface to the core base and then
processing discrete sediment samples according to standard loss on ignition technique
(Soil and Plant Analysis Council, 1999). Percent Organic Matter (% OM) results were
paired with simple lithologic descriptions of distinct sediment layers to establish detailed
transects of the riparian stratigraphy at each study site.
Water chemistry
Groundwater and stream samples were collected monthly from October, 2007
through September, 2008. Three well volumes were purged from wells prior to sample
collection. Surface water samples were collected normal to well transects during each
monthly site visit. Surface water and groundwater samples were collected using small
bailers and 200 mL opaque plastic bottles (NOa'-N and NH4'^-N), or amber glass bottles
(DOC). Samples were stored in a cooler, and transported back to the Ragsdale geology
lab where they were filtered (0.45 pm) and stored overnight at 4 C prior to analysis. In
20
rare cases where collected samples were heavily sediment-laden, samples were run first
through a 1.25 pm 934-AH glass fiber filter, then through a 0.7 pm APFF glass fiber
filter, and finally through the 0.45 pm membrane filter. Pre-filters were rinsed twice with
ultra-pure D1 water to prepare them for the pre-filtration process.
Monthly N03'-N and NH4^-N concentrations, and Kjeldahl-N concentrations for the
sampling date of 5/30/2008 were measured at the Central Environmental Laboratory of
East Carolina University (ECU) by Dr. Rebecca Cooper. NOs'-N concentrations were
analyzed using a Westco Scientific SmartChem 200 analyzer according to SmartChem
Method 375-100E-2-Rev.A-02-0906: “Nitrate-Nitrite in water, wastewater, soil extracts
and other aqueous samples” (Westco Scientific, 2006a) with a lower limit of detection of
0.02 mg/L N. NH4^-N concentrations were analyzed according to Method 4500-NH3-F:
“Phenate Method” (APHA, 1998) with a lower limit of detection of 0.0014 mg/L N.
Samples for Kjeldahl-N collected on 5/30/2008 were digested using a Westco Scientific
AD4020 block digestor and were then analyzed by a Westco Scientific SmartChem 200
analyzer according to SmartChem Method 390-200E-Rev.A-0906: “Total Kjeldahl
Nitrogen (TKN) in Water, Waste Water, Soil Extracts and other aqueous samples”
modified from Westco Scientific (2006b) for use with the AD4020 colorimetric, block
digestor. The lower limit of detection for this method was 0.10 mg/L TKN. DON values
(mg/L) presented in the results section, were calculated by subtracting corresponding
NH4'^-N concentrations from the assessed TKN values since TKN = DON + NH4^-N
(Westco Scientific, 2006b). In order to confirm the precision of the analytical techniques,
10 replicate sample pairs collected on 4/14/2008 were analyzed for NOs'-N and NH4'^-N
21
concentrations. Mean difference of the paired analyses indicated that, NOs'-N and NH4^-
N concentrations reported in this manuscript are within 0.002 mg/L and 0.001 mg/L N of
each other respectively.
DOC concentrations were measured by Mr. Jesse Chadwick using a Teckmar
Dorhmann Phoenix 8000 TOC analyzer at the main drinking water treatment plant of
Greenville Utilities Commission (GUC). DOC was analyzed according to the method:
UV Persulfate procedure Method 53IOC (APHA, 2005) with a lower limit of detection of
0.02 mg/L. Samples analyzed forNOs'-N, NH/-N, Kjeldahl-N, and DOC with
concentrations lower than the limit of detection are reported as 0.00 mg/L. Replicate
analysis of groundwater samples (N = 8 pairs) collected on 11/14/2007 indicated that
DOC concentrations reported in this manuscript are within 0.1 mg/L of each other on
average.
Chloride (CP), DO, and specific conductivity (SC) were measured using field meters.
Chloride concentrations were obtained using a Nexsens WQ-Cl Smart USB ion-specific
chloride electrode with a manufacturer’s lower detection limit of 1.8 ppm. Dissolved
oxygen (DO), specific conductivity (SC), and groundwater temperature (GT) were
measured in-situ during monthly collection trips using YSI EcoSense DO200 and EC300
field meters with detection ranges of 0 to 20.00 ppm and 0.0 to 499.9 pS/cm,
respectively. Analyzed samples with concentrations lower than the limit of detection are
reported as 0.00 ppm for Cf and DO and 0.00 pS/cm for SC.
22
Data analysis
Monthly difference in NOs' and Cf concentrations (estimated as a percent) at each
site were estimated for groundwater flowpaths ranging in length from 20 to 28 m.
Monthly percent differences were estimated for each floodplain according to the
following equation:
(near - stream NOj' concentration - up - gradient NO3' concentration ) * jqqo/
up - gradient NO 3' concentration
where:
near-stream NOa' concentration = NOa' or Cf concentration averaged from the near-
stream (N1 ) shallow and deep wells at a site for a given month,
up-gradient NO3' concentration = NO3' or Cf concentration averaged from the up-
gradient (N3) shallow and deep wells at a site for a given month, and
Median concentrations for the year were used to calculate armual percent difference in
NO3' concentrations, hereafter referred to as N03'-attenuation. Positive percent
difference in constituent concentrations represents an increase in concentrations across a
particular floodplain, where negative values represent a concentration decrease.
The influence of dilution on observed decreases in NO3' concentrations across each
floodplain aquifer was determined by measuring percent difference in Cf concentrations
from up-gradient to near-stream wells, since Cf is generally considered non-reactive in
most fresh groundwater systems (Hill 2000; Altman and Parizek, 1995).
23
Bivariate correlation and Mann-Whitney-U comparisons were completed using
Minitab v. 15.1. Linear, and power regressions were completed using Microsoft Excel
2003. Twenty-two data sets grouped on specific hydrologic and chemical variables such
as near-stream groundwater depth, and near-stream nitrate concentrations at each site
were tested for normal distribution via a standard Anderson-Darling normality test. Of
the 22 sets tested, only five data sets were determined to be normally distributed, none of
which repeated for a single variable across all three sites. Thus, non-parametric Mann-
Whitney-U comparison was used to determine if the medians of the hydrologic and
chemical data sets were significantly different. Where bivariate correlations are
presented, both Pearson correlation coefficient (r) and p-values are reported to distinguish
bivariate results from Mann-Whitney-U results, which are represented with a single p-
value. For both analyses, only p values < 0.05 were considered statistically significant.
Results
Floodplain hydrogeology
The floodplain sediment distribution and groundwater flowpaths for the three study
site transects based on floodplain sediment cores (Appendix C) are shown in Figure 5.
Sediments of the unconflned aquifers at the three sites can be divided into five main
groups: 1) rooted surflcial soil and gravel, 2) clayey sand, 3) sandy loam, 4) peat, 5)
coarse- to fine-grained sand, 6) densely packed clay. The regional clay that serves as a
confining unit was encountered at all sites and ranged from 2.8 m depth at the near-
stream nest of the urban site to 5.9 m depth at up-gradient nest of the rural site, with a
median depth of 3.3 m (Figure 5). Saturated sandy aquifers at the three sites ranged in
thickness from 1.3-2 m at the near-stream wells to upwards of 4.3 m at the up-gradient
wells of the suburban site (Figure 5). The sand of these floodplain aquifers was typically
overlain by 5-170 cm of organic-rich sediment at the surface, depending on the
occurrence of buried peat across each floodplain. A subsurface peat layer occurred at
both the urban and suburban sites. The peat at the urban site ranged in thickness from
about 1 m at the N1 core to 1.8 m at the N3 core, was light gray to brownish black,
ranged in % OM from about 4-49%, and contained abundant plant debris including roots
and stems to 0.5 cm diameter. The peat at the suburban site was thinner, and less
extensive being 1 meter thick at the N1 core and pinching out between cores N2 and N3.
Peat at the suburban site was similar in color to peat at the urban site from light gray to
black. It had a range of % OM from 2-26%, and plant roots and stems to 0.6 cm
diameter.
25
Legend
Head Measurement
CZD rooted surficial soil and gravel ?i^ high baseflow peatwater table ? shallow sand
^ low baseflow water table CZ3 clayey sand aquifer
A O CDgroundwater flowpath sandy loam clay confining unit» >
— equipotential line FB-Nl FB-N2 FB-N3
I '
ENE WSW
5 13
<
|12
urban
C
4 10
8
-8 -4 0 4 8 12 16 20 24 28 32 36
Distance from Stream(m)
PB-Nl PB-N2 PB-N3
C 1
ENE WSW
rural
3
2
-4 0 4 8 12 16 20 24 28 32
Distance from Stream(m)
Figure 5. Hydrogeology of study transects at the urban, suburban and rural sites showing
framework geology, seasonal high (marked ‘H’) water table on 4/12/2008, and seasonal
low (marked ‘L’) water table on 10/5/2007. Also shown are well screen locations (black
dots), and hydraulic head contours (m) for 1/8/2008. Labels above each transect show
well nest locations. Arrows indicate inferred groundwater flowpaths.
26
Peats at both floodplains contained woody debris including fossilized bark, and thin
sticks to 1 cm diameter that provide evidence of a forested floodplain environment. Peats
were also interbedded from base to surface with thin lenses of fine-grained sand that were
likely deposited by floods. An AMS radiocarbon date (1000 ± 40 Before Present (BP))
for the top of the peat layer at the suburban transect (MHB-N2, 0.8 m depth, Figure 5)
yielded a 2a calendar year calibration of 980 to 1060 Cal AD (“Anno Domini” - in the
year of the lord). The 2a calendar year calibration was calculated using the fNTCALOd
calibration database (2004) as described by Talma and Vogel (1993).
Across each floodplain the greatest organic matter content (% OM) occurred at depths
of 1 m or greater (Figure 6). Because of the absence of a thick peat layer at the rural site,
floodplain sediments at the rural site showed less % OM at than the floodplain sediments
at the urban and suburban sites. Very shallow soil layers (0-20 cm) showed slightly
elevated % OM at all three sites (Figure 6). This pattern is less pronounced at the urban
site, where shallow organic-rich soil grades directly into the underlying peat. A slight
increase in % OM was observed in the samples taken from the confining unit of all three
sites (Figure 6). The % OM peaks seen from 2-2.5 m at the suburban site (Fig. 5B,
MHB-Nl) and from 1.7 to 2 m at the rural site (Fig. 5C, PB-Nl) represent isolated “hot-
spots” of organic matter where woody debris and organic litter were sampled directly.
27
< To Streams
A FB-N3
FB-N2
FB-N1 %OM
0 0 20 0 40 0 60.0
B
%OM
0 0 20 0 40 0 60 0 800
suburban
0 0 SO 10.0 15.0 20.0
C
rural
Figure 6. Vertical distribution of % OM based LOI processing of samples from
floodplain sediment cores of each study site. Cores are spaced 14 m apart at the urban
and rural sites and 10 m apart at the suburban site. X-axis at the rural site is scaled down
relative to the axes of the other sites to show depths where % OM was high at the rural
site despite generally low % OM relative to the urban and suburban sites.
28
The urban site had the highest median % OM of 1.2 % (Appendix Al). The
maximum % OM measured for a single sample was 73.3 % measured at the suburban site
at 222 cm depth, where woody debris was encountered. % OM increased with distance
from the stream channel at the urban site and decreased with distance from the stream
channel at the suburban and rural sites (Appendix Al).
Mean and median hydraulic conductivity (K) for the floodplain sediments at each of
the wells (n = 18) were within an order of magnitude of each other (Appendix A2).
Hydraulic conductivities ranged from 1m/s at the deep near-stream well of the
suburban site to 10' m/s at the deep near-stream well of the rural site. The maximum K
is characteristic of medium-grained sand and the minimum K is characteristic of silt
(Heath, 2004). Though K was highly variable within and across sites, the median K at
the urban site (10'^^ m/s) was significantly lower (p = 0.0328) than the median K at the
suburban site (10'^^ m/s). Median K values at the urban and suburban sites were not
significantly different from the median K at the rural site (10'^ ° m/s).
Floodplain groundwater levels and temperature
The median, near-stream groundwater depth (9/14/2007 to 9/4/2008) as measured
by near-stream loggers at the urban (U) site (1.47 m) was 0.48 m deeper than the water
table at the suburban (S) site (0.99 m; p = 0.000), and 0.65 m deeper than the near-stream
water table at the rural site (0.82 m; p = 0.000) (Appendix A4; Figure 7). Over the one
year study, the near-stream water table showed a measurably smaller range and standard
29
deviation at the urban site (0.74 m, 0.08 m) than at the suburban site (1.02 m, 0.12 m) and
the rural site (1.31 m, 0.25 m) (Appendix A4).
Land Surface
Sep-07 Nov-07 Jan-08 Mar-08 May-08 Jul-08
0 10
Surface soil r 7"1 ^0.2 i ! 9
0.4 if -?lI i— CO: I 8 p
-i- Peat at the U and S ;? 1 1
t
0.6 7
2 0.8
% h|l hi 6
g.
Ü
1 5 £
I 1 1 Ü*
Q.
4
'sj o
Q. 1.4 iWj 3 >
0) tjj3 c
Q 1.6 2 0)
£
1.8 1s' * o
*
2 Í. ... •
-Urban -Suburban Rural Gneenvile Precip.
Figure 7. Near-stream water table depths measured in deep, near-stream wells of each of
the study sites from 9/14/2007 to 9/4/2008. Urban data from 7/15/2008 to 9/4/2008 is
missing due to a logger failure. Both seasonal and storm-induced changes in depth are
consistent across the three sites.
Near-stream groundwater at the rural site saturates shallow surface soils completely
on a near monthly basis, whereas near-stream groundwater at the suburban site and the
urban site never fully saturates shallow surface soil (Figure 7). The Pitt County Soil
Survey suggests that combined 0-horizon and A-horizon soils, where organic matter is
generally concentrated, typically range from about 10-30 cm in thickness for the region
where Pitt County resides (Kamowski et al., 1974). The % OM data from surface soil of
this study is consistent with this range. Thus, with a conservative estimate of 30 cm as
the base of O- and A-horizon soils, near-stream groundwater at the urban site made no
contact with shallow organic-rich soil, one half-hour of contact at the suburban site, and
30
234 hours of contact at the rural site, during the one year period of the study. In addition
to loss of contact with surface organic soils, comparison of water table extremes of this
study (seasonal high and low groundwater levels) to cross-sections of floodplain geology
suggests significant loss of contact between near-stream groundwater and shallow peat at
the incised study sites (Figures 5A and 5B).
Monthly groundwater levels (m Above Mean Sea Level (AMSL)) in shallow up-
gradient wells at the three study sites showed a smaller armual range at the urban site (0.5
m) than at the suburban (0.8 m) and rural (0.6 m) sites respectively (Appendix A7, Figure
8). The annual groundwater level range was smaller at near-stream wells where monthly
measurements at the urban site showed less than 0.1m change for the total duration of
the study period (Appendix A7, Figure 8). Seasonally, floodplain groundwater levels
were relatively constant through the fall of 2007, increased into the late winter of 2007
and spring of 2008, and decreased into the summer of 2008 (Figure 7). Less seasonal
variation was observed in near-stream wells of the urban and suburban sites than at the
rural site (Figure 8).
Median groundwater temperature (N = 72 measurements/site) for the twelve months
of data collected at each site was significantly higher at the urban site (18.6 C) than it was
at the suburban (17.3 C, p = 0.00) and rural sites (17.2 C, p < 0.01) respectively
(Appendix A8). Temperature did not vary significantly with depth or position across the
three floodplain study sites (Appendix A9).
31
A
>
(O
5
d
ü
0)
w
Û. urban
«
Q
Date
- up-gradient (U) ? near-stream (N) Greenville precip.
B
>
<0
d
‘ü
0)
suburban
(5
O
Date
- up-gradient (U) near-stream (N) Greenville Precip.
c
>.
(0
5
d
‘ü
0)
Q.
rural
<5
O
Date
up-gradient (U) - near-stream (N) Greenville precip.
Figure 8. Monthly groundwater levels (m AMSL) from the shallow up-gradient (U) and
near-stream (N) wells of the A) urban, B) suburban and C) rural sites.
32
Hydraulic gradients and groundwater discharge
Hydraulic gradients were steeper at the urban and suburban sites than at the rural site
(Table 1, Figure 9). Annual groundwater discharge, based on median hydraulic
conductivity and monthly hydraulic gradient at each transect, was significantly lower at
the urban site (5.0 m /month) than at the suburban site (12.3 m /month; p = 0.000) and
the rural site (14.4 m^/month; p = 0.000; Table 2). Median annual groundwater discharge
was not significantly different between the suburban and rural sites (p > 0.05). The
annual range in total discharge at the urban site (3 m /month) was considerably lower
than the range at the other sites (Table 2). Seasonal variation in hydraulic gradient and
groundwater levels controlled variation in discharge (Figures 9 and 10).
Seasonally, peak groundwater discharge at the suburban and rural sites lagged behind
peak discharge at the urban site. Discharge peaked in late February at the urban site,
early April at the suburban site, and early May at the rural site (Figure 10). The urban
and the suburban sites showed little change in discharge through the fall of 2007, a time
when discharge at the rural site showed significant and steep decline. All three sites
showed a significant increase in discharge as a response to heavy rains in February of
2008, but the urban site showed no discharge response to frequent late winter and spring
showers. All three sites experienced a measurable decrease in discharge from the spring
of 2008 to the summer of 2008 (Figure 10).
33
Table 1. Annual hydraulic gradients from up-gradient to near-stream wells across
floodplains of each site.
dh/dl (m/m)
Site
urban suburban rural
N 12 12 12
Median 0.031 0.032 0.024
Maximum 0.038 0.051 0.028
annual
Minimum 0.023 0.028 0.013
Range 0.015 0.023 0.014
Standard Deviation 0.004 0.008 0.005
Aug-07 Oct-Oy Dec-07 Feb-08 Apr-08 Jun-08 Jul-08 Sep-08
Date
Figure 9. Monthly hydraulic gradient across the floodplains of the three study sites from
10/5/2007 to 9/5/2008.
34
Table 2. Descriptive statistics for annual discharge from each floodplain.
Discharge (m^/mo)
Site
urban suburban rural
annual N 12 12 12
Median 5.0 12.3 14.4
Minimum 3.5 10.2 8.1
Maximum 6.5 22.8 18.6
Range 3.0 12.6 10.5
— urban
A—— suburban
?—— rural
Greenville
Precip.
Date
Figure 10. Monthly discharge of floodplain groundwater from 10/5/2007 to 9/5/2008
based on a 100 m cross-sectional stream reach. Dashed line represents daily precipitation
(in/day) in Greenville during the course of the study period.
35
Groundwater and surface water chemistry
Dissolved inorganic-N concentrations
Generally, groundwater NOa'-N concentrations were < 1 mg/L, in all wells but the
up-gradient shallow and deep wells of the rural site (Figure 11). A septic system buried
approximately 20 meters west-southwest (WSW) of the up-gradient wells (landowner
Christopher Graves, pers. comm., 2008) at the rural site presumably acted as the
dominant source of nitrate concentrations greater than 8 mg/L measured in the shallow
floodplain and up-gradient wells at the rural site. Median groundwater nitrate
concentrations (N = 12) were significantly lower in shallow wells at the urban (p =
0.0000) and suburban (p = 0.0021) sites than in deeper wells (Appendix A10; Figures
12A and 12B). Concentrations were significantly higher (p = 0.0005) in the shallow
wells relative to deep wells at the rural site (Appendix A10; Figure 12C). Median, near-
stream N03'-N concentrations in groundwater (N = 24/site) were generally less than 0.2
mg/L, and were significantly less than median surface water concentrations (N = 12/site)
at each site (Appendix All; Figure 11).
Groundwater NH4'^-N concentrations were generally < 0.2 mg/L (Figure 11;
Appendices A12 and A13). Higher concentrations were found in the shallow wells of
the urban floodplain and the near-stream wells of the rural floodplain (Figures 11 and
13). NH4'^-N concentrations decreased significantly from the shallow floodplain well of
the urban site to the shallow near-stream well (p < 0.001) and increased significantly
from the floodplain to the near-stream wells at the rural site (p < 0.001). DON
36
concentrations were generally less than 1 mg/L at all three study sites and remained
stable or declined along lateral hydrologic flowpaths for the single date they were
collected and assessed (Appendix A14, Figures 14-16). In general, DON and NH4'^-N
remained stable or declined across each floodplain as N03'-N concentrations declined
(Figures 14-16).
37
Legend
I Installed monitoring well showing EZI] rooted surficial soil and gravel peat
A I l~~l I I shallow sandmedian NO, -N and NH/-N clayey sand aquifer
I concentrations (mg/L). I i sandy loam T'--XA clay confining unit
I 0.8, 0.8 FB-Nl FB-N2 FB-N3
deep-rooted ' I I
ENE trees tended residential lawn wsw
urban
Stream;
0.2,0.1
-8 12 16 20 24 28 32 36
Distance from Stream{m)
SW
MHB-Nl MHB-N2 MHB-N3
I I I
12 Sparse vegetation cleared for sewer access Deep rooted trees and shrubs
11
Stream: suburban
10
9
0.1,0.1
8
AMSL) 7S( 6Vecraticleal -4 0 8 12Distance from Stream (m)PB-N1 PB-N2 PB-N3I Iene Small shrubs Tended residential lawn Small shrubs WSW98(AMSL) 7Smcale 6 ruralVertical 5
4
-4 ' 0 4 8 ' 12 16 ' 20 24 28 32
Distance from Stream(m)
Figure 11. Study site transects at the urban, suburban, and rural sites showing median
nitrate-N and ammonium-N (mg/L) values measured in wells at the near-stream,
floodplain, and up-gradient locations.
38
Depth
Bshallow
Ddeep
?stream
urban
Depth
I shallow
? deep
?stream
suburban
Depth
Bshallow
?deep
?stream
rural
Position
Figure 12. Shallow, deep, and in-stream nitrate-N concentrations from the A) urban, B)
suburban, and C) rural floodplains for the one year study period.
39
Depth
H shallow
?deep
Dstream
urban
Depth
B shallow
Odeep
?stream
suburban
Depth
Hshallow
Gdeep
?stream
rural
Figure 13. Groundwater and surface water ammonium-N concentrations (mg/L) across
the A) urban, B) suburban, and C) rural floodplains for the one year study period.
40
? Nitrate-N
?Ammonium-N
0DON
urban shallow
?Nitrate-N
? Ammonium-N
Soon
urban deep
up-grad. floodplain near-st, stream
Position
Figure 14. Median nitrate-N, median ammonium-N, and DON (mg/L) from 5/30/2008
in A) shallow and B) deep wells of the urban floodplain.
41
?Nitrate-N
? Ammonium-N
B3DON
suburban shallow
Position
B ? Nitrate-N20 ?Ammonium-N
0DON
suburban deep
up-grad, floodplain near-st. stream
Position
Figure 15. Median nitrate-N, median ammonium-N, and DON (mg/L) from 5/30/2008
in A) shallow and B) deep wells of the suburban floodplain.
42
?Nitrate-N
?Ammonium-N
Hdon
rural shallow
Position
?Nitrate-N
?Ammonium-N
0DON
rural deep
Figure 16. Median nitrate-N, median ammonium-N, and DON (mg/L) from 5/30/2008
in A) shallow and B) deep wells of the rural floodplain.
43
DOC and DO concentrations
Median dissolved organic carbon (DOC) concentrations were greater than 1.1 mg/L
throughout the unconfmed aquifers of all three study sites (Figure 17). DOC
concentrations were variable across the floodplain, but generally decreased with depth
below organic-rich sediment layers (Figure 17). Examples of this distribution are the
high DOC concentrations of 6.2, 14.2, and 4.2 mg/L at the shallow wells of the urban site
and the 2.1, 5.1 mg/L median concentrations determined at the shallow, floodplain wells
of the suburban site. DOC concentrations showed a significant inverse relationship with
well-screen depth below the water table across the three sites (Figure 18, R = 0.46).
Concentrations were significantly higher in the shallow wells of the urban and suburban
sites than in the deeper wells (p = 0.000 and p = 0.0018 respectively). Finally, DOC
concentrations were significantly higher at the urban site than at the suburban (p < 0.005)
and the rural (p = 0.056) sites, respectively.
DOC concentrations also varied systematically with lateral groundwater flow (Figure
17). Concentrations decreased significantly along: 1) shallow flow from the shallow N2
to N1 wells at the urban site (p = 0.0016), 2) deep flow from the deep N3 to the deep N1
wells at the urban site (p = 0.0244), 3) and shallow flow from the shallow N2 to N1 wells
of the suburban site (p = 0.0131). DOC increased significantly between the shallow N2
and N1 wells of the rural site (p = 0.0302) likely due to contact with “hot-spots” of
woody plant debris buried in floodplain sands of the rural site.
44
Legend
Installed monitoring wells showing median rooted surftcial soil and gravel peat
dissolved organic carbon and dissolved L \ clayey sand I I shallow sand aquifer
oxygen concentrations in mg/L I I sandy loam clay confining unit
FB-N2 FB-N3
urban
SA(V suburbanS(AV?AS(eMcVeMrcaeMtcmSriartcSlatiLecaiSlcLeal)aeLl)l ) rural
1.5,0.8
1.5,0.6
-4 0 4 8 12 16 20 24 28 32
Distance from Stream(m)
Figure 17. Study site transects of the urban, suburban and rural sites showing median
annual DOC, DO values from shallow and deep wells of the three study sites.
45
Figure 18. Plot of screen depth below water table vs. median DOC from groundwater
samples over the duration of the study period. The value marked * is an outlier from the
deep near-stream wells of the rural site, where high DOC concentrations may represent
contribution of dissolved organic carbon from underlying marine sediments.
DO concentrations inereased significantly with depth at the urban and suburban sites
respectively (Figure 19, R^= 0.58 and 0.63 respectively). Across the floodplain,
groundwater DO concentrations also decreased with proximity to the edge of eaeh stream
channel (Figure 20). Median annual groundwater DO coneentration was signifleantly
higher at the urban site (1.0 ppm) than at both the suburban site (0.7 ppm, p = 0.000) and
the rural site (0.5 ppm, p = 0.000).
46
Figure 19. Median well DO concentrations vs. screen depth for floodplain wells of each
of the study sites. values were 0.58, 0,63, and 0.41 for the urban, suburban and rural
sites respectively.
Figure 20. Median well DO (ppm) coneentrations vs. dist. (m) of wells from the stream
edge at each site. values of 0.42, 0.39, and 0.79 for the urban, suburban and rural sites
respectively suggest median well DO concentrations increase with distance from the
stream edge at each site.
47
Chloride concentrations
Groundwater chloride (Cl') concentrations ranged from 6.8-40 mg/L in the floodplain
aquifers of this study (Appendix A18). Monthly percent difference in Cf concentrations
from the upgradient wells to the down-gradient wells across each floodplain were slight
and inconsistent in comparison to percent difference in NO3' concentrations (Table 3). A
slight decrease in median Cf concentrations was observed at the urban site (-17%), with a
notable increase in concentrations at the suburban site (30%) and a notable decrease in
concentrations at the rural site (-30%).
Table 3. Percent difference in median groundwater chloride concentrations (N = 6
months of chloride data) from up-gradient to near-stream wells across each floodplain.
Chloride cone.
(mg/L)
Position
Sample up-grad. near-St.
Depth Median Median % difference
urban average 11.3 9.4 -16.6
Site suburban average 15.0 19.4 29.8
rural average 28.5 19.9 -30.4
48
Statistical relationships
Correlation of NOa'-N concentrations with the concentrations of the other measured
constituents revealed positive correlations between NOa'-N and DO and inverse
correlations between NOa'-N and DOC. Strong positive relationships were observed
between NOa"-N and DO at the urban site (r = 0.48, p < 0.01) and the suburban site (r =
0.58, p < 0.01) (Figures 21A and 21B)), with no significant relationship between the two
constituents at the rural site (r = 0.15, p = 0.28) (Figure 21C). Significant inverse
relationships were observed between NOa'-N and DOC at the urban site (r = -0.38, p =
0.01), suburban site (r = -0.24, p = 0.040), and rural sites (r = -0.37, p = 0.01) (Figure 22).
NOa’ concentrations were high along deep flowpaths at the urban and suburban sites with
corresponding high DO concentrations and low DOC concentrations relative to
concentrations of all three constituents in shallow wells (Figures 11 and 17). Such
vertical stratification of concentrations was reversed at the rural site, where up-gradient
septic discharge maintained markedly high nitrate concentrations in shallow up-gradient
wells, and both DO and DOC values were notably low relative to the urban floodplains.
49
urban
suburban
rural
Figure 21. Correlations between nitrate-N and dissolved oxygen (DO) at the urban,
suburban, and rural sites. Correlation between constituents at the rural site was
complicated by the relatively constant contribution of up-gradient nitrate-N by a nearby
septic tank. Outliers were excluded to highlight the general relationship.
50
A 30.0
urban: r = -0.38, p < 0.01
24.0
X
D> 18.0 outliers
urban
12.0
6.0 Í--ifVf 'fl ? ?? Dq ? ? ?
O oo ° tEt3 %
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Nitrate-N (mg/L)
B 30.0
suburban: r = -0.24, p = 0.04
24.0
18.0
suburban
O 12.0 I
S ' *
6.0 I ^
*a1
**
0.0 a. a * *a/*
0.0 0.2 0.4 0.6 0.8
Nitrate-N (mg/L)
rural
Figure 22. Correlations between nitrate-N and dissolved organic carbon (DOC) at the
urban, suburban, and rural sites. Where DOC concentrations were greater than 6 mg/L
(dashed line) nitrate-N concentrations were nearly always less than 1 mg/L. Outliers
were excluded to highlight the general relationship.
51
Bivariate correlation of groundwater levels (m AMSL) and constituent concentrations
in shallow floodplain wells (N1 and N2 locations) showed significant variation across the
three study sites (Table 4). Groundwater levels showed significant positive correlation
with ammonium-N (NH4^-N) and dissolved oxygen (DO) at the urban site (when the
water table was closest to the surface NH4'^-N and DO were elevated). Significant
inverse correlation was observed between groundwater levels and both NOa'-N and
NH4^-N at the suburban site (when the water table was shallow, NOs’-N and NH4^-N
were diminished). Finally, at the rural site a signiflcant inverse correlation was observed
between NH4^-N and groundwater levels (when the water table was shallow, NH4’^-N was
diminished) as well as a significant positive correlation between DO and groundwater
levels (when the water table was shallow, DO was elevated) (Table 4).
Table 4. Pearson correlation coefficients for bivariate correlation of monthly
groundwater (gw) levels (m AMSL) and monthly groundwater concentrations (mg/L)
from shallow floodplain wells (N1 and N2 locations) of each site. Bold-faced values
represent significant relationships.
Site Variable NO3-N DOC NH4^-N DO
urban gw level -0.21 (23) 0.41 (20)^ 0.53 (22)* 0.70 (20)**
suburban gw level -0.65 (24)** 0.28 (24) -0.57 (24)* -0.16(22)
rural gw level 0.36 (24)^ -0.23 (23) -0.52 (24)* 0.49 (20)*
** P <0.005
* P < 0.05
*P<0.1
52
Floodplain NOs'-attenuation and loading to incised stream channels
Monthly differences in NOs'-N concentrations (calculated as a percent) from up-
gradient wells to near-stream wells at each study site were less across the urban and
suburban floodplains than the differences that occurred across the rural floodplain (Figure
23). Relatively high monthly concentration differences at the rural site occurred in
coordination with a shallow near-stream groundwater level (median depth = 0.82 m; N =
16680 measurements). Monthly percent differences at the urban site were significantly
lower than differences at the suburban (p = 0.02) and rural (p <0.01) sites, respectively.
Percent differences at the suburban site were also significantly less than differences at the
rural site (p < 0.01). On an annual basis, NO3' concentrations decreased laterally from
up-gradient to near-stream wells (Figure 11) with significant attenuation along: 1) lateral
groundwater flow at the urban site, where annual median NO3' concentrations were
attenuated 46.4% (p < 0.01), 2) lateral groundwater flow at the suburban site, where
median NO3' concentrations were attenuated 58.3% (p < 0.01), and 3) lateral groundwater
flow at the rural site where median NO3' concentrations were attenuated approximately
99.2% (p< 0.01).
53
Figure 23. Monthly percent difference in NOs’-N concentrations (%) over twelve
months of chemical data at the urban, suburban and rural sites. Median groundwater
(gw) depths at each site are median of 14631, 18269, 16680 measurements recorded at V2
hour intervals by water level loggers in the near-stream shallow wells (N1 locations) of
the urban, suburban and rural sites.
Median shallow and deep near-stream nitrate concentrations of the urban, suburban
and rural floodplains were < 1 mg/L (Figure 11). Average near-stream nitrate
concentrations at the urban and suburban sites were significantly greater than near-stream
nitrate concentrations at the rural site (p < 0.005 in both cases). Nitrate loading (average
near-stream groundwater concentration * groundwater discharge from each floodplain) to
streams at the three sites showed considerable variation across sites (Figure 24-A). The
combined effect of discharge and average near-stream nitrate concentration on loading
caused monthly loading from the suburban floodplain to be significantly greater than
loading from both the urban floodplain (p < 0.0005) and the rural floodplain (p < 0.005)
respectively (Figure 24).
54
Figure 24. Seasonality graphs showing variability in A) NOb' loading (g/mo), B)
average near-stream N03'-N concentrations (mg/L), and C) discharge (m^/mo) across the
three floodplains. Dashed black line (C) shows daily precipitation for the Greenville area
(in/day).
55
Discussion
Changes in organic matter with depth
The notable variation in % OM with depth and proximity to streams disproved the
hypothesis that OM would be concentrated in the shallowest (0-30 cm) portion of the
riparian sediment profiles of Greenville’s low-order streams. The deep and laterally
extensive peats found at the urban and suburban sites of this study were similar to peats
and organic-rich sediments found at depths greater than 50 cm in a recent study
conducted near Ontario, Canada by Hill et al. (2004). Peats of their study had % OM of
30-62 % which is consistent with the 6-49 % OM determined for peats of this study.
They demonstrated that buried, laterally extensive, organic-rich soils and “hotspots” of
buried OM in floodplain sediments have considerable potential to denitrify discharging
groundwater. Advancements in the prediction and quantification of the spatial
distribution of subsurface organic matter are needed to improve the accuracy at which
riparian zone denitrification can be estimated (Groffman et al. 2009).
Effects of decline on OM contact
Deeper groundwater tables adjacent to heavily incised streams have been reported in
both agricultural (Schilling et al. 2004, 2006; Burt et al. 2002; and Hupp, 2000) and
urban settings (Groffman et al. 2002; Groffman et al. 2003; Hardison, 2008). The decline
of water tables adjacent to incised stream channels was also observed during this study.
Deeper water tables adjacent to incised channels are likely due to a combination of: 1)
decreased recharge to the surficial aquifer in urban settings caused by impervious surface
56
cover and 2) rapid drainage of near-stream groundwater to enlarged urban channels.
Floodplain groundwater recharge may also be limited by reduced overbank flow to the
riparian areas of the incised channels, which was documented at two of three incised
study sites of the Greenville area (Hardison et ah, 2009, in press). Undisturbed Coastal
Plain streams have frequent overbank events and interact with adjacent floodplains an
average of five times per year (Sweet and Geratz, 2003). When channels are heavily
incised, floodplains may become inactive (relict).
Incision and water table decline at the urban and suburban sites caused a significant
reduction in contact between floodplain groundwater and shallow O- and A-horizon soils.
Though groundwater levels at all sites showed no contact with surface soil during
baseflow conditions, stormflow groundwater tables at the least incised, rural site made
frequent and prolonged contact with surface soil (Figure 7). Aside from the 0.5 hours of
contact observed at the suburban site following a storm on 4/5/2008 (Figure 7), no
contact of floodplain groundwater with surface O- and A-horizon soils was observed at
the heavily incised urban and suburban sites. Because of the occurrence of thick and
extensive peat layers at the heavily incised sites, stormflow contact of floodplain
groundwater and peat at the two sites occurred regularly (Figure 7). Assuming that near-
saturated conditions observed in floodplains of the headwaters of the rural watershed are
representative of the pre-developed state of the streams of this study, groundwater had
reduced contact with buried peat at the urban and suburban sites (Figure 5). When
groundwater levels were seasonally low, the greatest loss of contact with peat occurred
(Figure 5).
57
DOC concentrations along shallow flowpaths at the urban and suburban sites were
elevated relative to concentrations along deep flowpaths (Figure 17). At these sites, the
shallow well screens were close to overlying peat. In the NC Coastal Plain, Boettger
(2002) observed elevated riparian groundwater DOC concentrations associated with
surface and buried OM (Baldwin Swamp, NC). Devito et al. (2000) and Hill et al. (2000)
report DOC concentrations of 10 to 18 mg C/L in the pore water of saturated peats of
their studies. Shallow median DOC concentrations of 4.2 to 14.2 mg/L at the urban site
and 2.1 to 5.1 mg/L in wells near peat at the suburban site suggest that peat at both sites
contributed considerable DOC to shallow groundwater.
In addition to elevated DOC concentrations, median groundwater temperature (N =
72 measurements) was higher at the urban site was than it was at the suburban site (p =
0.00) as well as the rural site (p <0.01) respectively. This pattern was significant across
seasons (Appendix A8) with median groundwater temperature at the urban site being >1
C greater than median groundwater temperatures at the rural and suburban sites. The
suite of factors that could increase groundwater temperature in urban watersheds include:
leakage from water supply or sewage infrastructure, increased air temperature due to the
“urban heat island effect” (Pickett et ah, 2001), changes in soil evaporation, increased
direct solar radiation due to the removal of riparian vegetation, and advection from paved
asphalt radiated by the sun.
Despite high spatial variability in the hydraulic conductivity of floodplain sediments,
localized highs in hydraulic conductivities at depth in the urban and suburban floodplains
58
suggest the presence of fast groundwater flowpaths. Stratigraphic analyses revealed the
occurrence of coarse- to pebble-sized gravel and sand along the base of the unconfined
aquifer where high sediment hydraulic conductivities were found. This coarse basal layer
is overlain by a distinct horizon where sediment becomes finer and less permeable. Hill et
al. (2004), Boettger (2002), and Hill et al. (2000) have observed that sediments with high
hydraulic conductivity promote fast groundwater velocities and short groundwater
residence times in floodplain aquifers that often reduce the capacity of plant uptake and
denitrification to remove NO3' from the floodplain groundwater system. The general
conclusion by Hill et al. (2000; 2004) was that coarse- to gravel-sized sediment at depth
may cause nitrate-laden groundwater to flow preferentially along deep groundwater
flowpaths. Evidence that this may occur in the current study is provided by elevated
nitrate concentrations in deep wells relative to shallow wells at the urban and suburban
sites.
Effects of water table decline on NOs’-attenuation
Results indicate that urban channel incision and corresponding water table decline
reduce the ability of the floodplain to process NO3'. However, geological variability of
floodplain hydraulic properties also influences groundwater discharge and floodplain N-
processing. On an annual basis nitrate concentrations decreased across each floodplain,
suggesting a net attenuation of NO3' at all three sites. The greatest attenuation of nitrate
occurred across the rural floodplain (99 %) where the median, near-stream water table
was the shallowest (0.82 m; Figure 23). These data suggest that deeper water tables
adjacent to incised urban streams may cause a net inhibition of processes that attenuate
59
NO3' in floodplain aquifers. For the single May 30, 2008 sampling date that DON
concentrations were collected and assessed, five of six groundwater samples from near-
stream wells had DON concentrations that were greater than corresponding NOb'
concentrations (Figures 14-16). This suggests that DON may also play an important role
in N transport to Coastal Plain streams.
Water table decline may act to inhibit NOs'-attenuation in the floodplain aquifers of
this study by: 1) shifting groundwater out of contact with the root zones of riparian
vegetation where nitrate is often assimilated by plants, 2) shifting groundwater out of
contact with organic matter that can promote denitrification, and 3) accelerating the
decomposition of subaerially exposed organic matter, and thus accelerating the leaching
of inorganic-N to floodplain groundwater. Variation in the effects of other NO3’-
attenuation processes and mechanisms, such as dilution, and DNRA across the sites of
this study may also affect net N03'-attenuation.
The concentrations of measured nutrients in shallow groundwater of the respective
floodplains represent the net effect of N-cycle processes in shallow soil and groundwater.
The data of this study provides limited insight to the priority, timing, or importance of
each N-cycle process. The remaining discussion will focus on mechanisms of NO3’
attenuation, how incision and water table decline affect specific N-cycle processes, and
how results compare to other studies throughout the midwestem and eastern U.S. and
Canada.
60
Plant uptake via assimilation of nitrate by the roots of riparian vegetation likely has a
limited influence on groundwater nitrate decline observed during this study. Analyses of
sediment cores showed surficial sand aquifers to be completely free of any root, stem, or
other evidence of deep root systems at each site (Figure 11, Appendix C). Sod root
systems were based at about 6 cm below the land surface at the urban and rural sites.
Evidence for the active root systems of trees and sparse shrubs in cores from the
suburban site suggests that active roots did not extend beyond 30 cm depth (Appendix C).
Boettger (2002) has suggested that the ability of riparian buffers to denitrify shallow
groundwater moving from agricultural fields to channelized streams of Eastern North
Carolina is strongly associated with the saturation of surface, organic-rich soil horizons.
Because thick, laterally extensive peat layers occurred at the heavily incised urban and
suburban sites, high DOC concentrations persisted in shallow wells despite significant
water table decline. Patterns of elevated DOC concentrations along shallow flowpaths
where NO3'concentrations were relatively low (Figures 11 and 17) suggest that NOs' was
depleted by denitrification along shallow hydrologic flowpaths. These patterns are
consistent with studies conducted by Devito et al. (2000) in Ontario who reported
elevated DOC in the presence of decreasing NO3' and DO concentrations along zones
bordering organic matter in two floodplain aquifers where ô'^N-N03' isotopes in
groundwater became enriched. Devito et al. (2000) goes on to suggest that the concurrent
occurrence of decreasing N03'with enrichment of ô'^N-N03' provides strong evidence
for the depletion of NO3' via denitrification for the floodplain aquifers of his study. The
combination of strong positive correlations between DO and N03' (Figure 21) at all three
61
sites in the presence of significant inverse relationships between N03'and DOC (Figure
22) provide additional evidence that NO3’ was depleted by denitrification in the three
floodplain aquifers of this study (Harden and Spruill, 2008; Korom, 1992). Hill et al.
(2000) and Devito et al. (2000) have shown that denitrification is limited by organic
carbon in zones where DOC concentrations are low (< 6 mg/L) and may become nitrate
limited where DOC concentrations rise above 6 mg/L. Thus, denitrification may be NO3'
limited along the shallow flowpath at the urban site and DOC limited along shallow and
deep flowpaths at the suburban and rural sites.
Bivariate correlation of riparian groundwater levels with constituent concentrations in
the shallow wells of the urban and suburban sites are presumably influenced by leaching
of constituents from exposed peats during recharge of the floodplain aquifer. Correlation
of groundwater levels with NH4’^-N concentrations showed a positive relationship at the
urban site which contrasted directly with inverse correlations at the suburban and rural
sites (Table 4).
The occurrence of relatively high NH4^-N concentrations in the shallow wells of the
urban site (Figures 13 and 14) is likely the result of mineralization and leaching of NH4’^
from the shallow peat layer. Storage and release ofN03' from subaerially exposed,
organic-rich layers has been observed in other studies (Angier and McCarthy, 2008;
Schilling, 2006; Groffman, 2002). Because the mineralization of N from actively
decomposing, organic-rich soil typically begins with the release of NH4^ by heterotrophic
microbes (Schlesinger, 1997), elevated NH4^-N and DON concentrations in shallow wells
62
at the urban study site may also be caused by storage and release from shallow organic
matter. The marked decrease in NH4’^-N from the shallow N2 well to the shallow N1
well of the urban site is likely affected by a variety of N-cycle processes and mechanisms
including: assimilation by microbes, adsorption to fine-grained sediments of the
unconfined aquifer, and nitrification ofNH/to form NOs’ that may be subsequently
denitrified as groundwater travels across the urban floodplain.
Dilution of NOb'along lateral groundwater flowpaths can occur due to the mixing of
groundwater with other water with relatively low NO3' concentrations. The notable
decrease in median chloride concentrations (-17%; Table 3) across the urban floodplain
suggests that floodplain groundwater may be diluted by groundwater upwelling from the
clay confining unit, since a median (N = 12), upward vertical hydraulic head gradient (-
0.01 m/m) was measured at the near-stream wells of the urban floodplain. Chloride
concentrations at the suburban site were consistently higher at the near-stream wells than
at the up-gradient wells (Table 3). This may be due to evaporation of groundwater from
the floodplain aquifer, since the residence time of groundwater in floodplain aquifers of
the NC Coastal Plain may range from months to years (Spruill et al., 2005). The notable
decrease in median chloride concentration from the up-gradient wells at the rural site
(28.5 mg/L) to the shallow near-stream wells (19.9 mg/L) suggests dilution may decrease
nitrate concentrations across the rural floodplain by a factor of 30% (Table 3). The
average dilution effect at all three sites was a 24% decline in NO3' concentration across
the floodplain. The average floodplain attenuation or decline in NO3' concentration was
63
68% (Figure 23). Overall, the results suggest that dilution contributed to NO3'
concentration declines across the floodplains but was not the primary control.
As with plant uptake, the reduction of NOa' to NH4'^ (DNRA) also likely had a limited
influence on the observed water table decline, N03'-attenuation relationship. Because the
occurrence of notable DNRA along short groundwater flowpaths should cause concurrent
NH4^ production and NO3’ consumption, constant and decreasing NH4^ concentrations
observed in parallel with significant N03'-attenuation along lateral flowpaths between
wells (Figures 14-16; Appendix A12 and A13) suggest that DNRA had little influence on
observed values of N03'-attenuation. Significant increase in NH4^ concentrations
observed between the floodplain and near-stream wells of the rural site (p < 0.001; Figure
16B) may be caused by DNRA, since exceptionally high nitrate concentrations measured
at floodplain wells of the rural site were more than sufficient to sustain microbial DNRA
(An and Gardner, 2002). NH4^ concentration increases may also be related to
mineralization of organic matter.
Comparison to previous studies
Near-stream groundwater NOs'-N concentrations at the incised sites of this study were
below average with respect to near-stream (2-45 m depending on scale) N03'
concentrations measured in agrieultural settings of the central and eastern US, and
Canada (Table 5). Floodplain groundwater N03’ data of this study is compared to data
measured in various agricultural settings because very little work has been conducted
investigating N-dynamics in groundwater discharging to streams in predominantly urban
64
watersheds. Comparison of near-stream groundwater nitrate concentrations at sites listed
in Table 5 to percentage NOs'-attenuation across each respective floodplain suggests that
near-stream groundwater NOa'concentrations > 0.5 mg/L are indicative of buffer settings
where N03'-attenuation is often impaired (< 85%). In all cases where near-stream
groundwater NO3'concentrations were < 0.5 mg/L, percentage N03'-attenuation was >
95%. The pattern of high percentage N03'-attenuation at riparian study sites with high
concentrations of N03' entering the floodplain groundwater system (incoming N03'
concentrations) at each site (e.g. sites 5B, 1, A-A’ (ENC), B-B’ (ENC) and A-A’
Transect; Table 5) suggests that high incoming NO3'concentrations may enhance N03'-
attenuation. Low percentage N03'-attenuation observed at the urban and suburban sites
of this study may be at least partially due to low NO3'concentrations in groundwater
entering floodplains of the incised study sites.
Table 5, Median NO3' concentrations and percentage attenuation (- = loss, + = gain) in floodplain, riparian groundwater
estimated from studies conducted in agricultural catchments of the central United States, eastern United States, and Canada.
NOs'-N concentrations (mg/L)
Study Location Land use Site label Flowpath Up-gradient Near-stream % NOj'N Reference
context Length(m) Attenuated
Eastern North Harden &agriculture Site 5A“ 65 6.4 <0.05 >-99
Carolina 41.3 0.1 >-99 Spruill,Site 5B 380 2008
SiteSPTU 21 20.2 4.2 -79.2
Eastern North agriculture Site 1 80 12.5 0.3 -97.6 Spruill,
Carolina Site 2 350 8.0 4.0 -50.0 2004
Site 3 90 5.3 0.9 -84.0
Site 4 100 5.6 0.0 -99.6
Eastern North agriculture A-A' 200 12.0 0.1 -99.2 Tesoriero et
Carolina (ENC) B-B' 80 15.0 0.1 -99.3 al. 2005
Eastern North agriculture Baldwin Swamp 12-14 3.1 2.6 -16.1 Boettger,
Carolina w/ vegetated Fork Swamp 12-14 3.3 4.1 +24.2 2002
riparian zone Pea Branch 12-14 0.7 0.3 -57.1
Baldwin Swamp 12-14 2.9 2.8 -3.4agriculture
w/ cultivated Fork Swamp 12-14 6.2 8.2 +32.3
riparian zone Pea Branch 12-14 1.7 2.4 +41.2
Lower agriculture A-A’ Transect 800 12.0 0.1 -99.2 Pfeiffer et
Wisconsin al. 2006
River
Union County, Schoonoveragriculture forest'’ 6.6 6.3 1.2 -81.7
Illinois & Williard,
cane'’ 10 7.5 0.1 -99.3 2003
NE Connecticut agriculture control 20 7.2 1.6 -78.5
Clausen et
al. 2000
treatment'’ 20 4.8 1.4 -69.8
NE Virginia agriculture main x-section” 120 7.1 3.3 -52.9 Snyder et al.
1998
Table 5. (Continued)
Ontario, Canada agriculture transect A 180 26.0 1.1 -95.9 Hill et al.
transect B 150 35.7 0.1 -99.7 2000
NE Maryland agriculture transect A 420 14.0 0.1 -99.2 Bohlke and
transect B 780 16.0 13.0 -18.8 Denver,
1995
Fairmount Co., agriculture B-B' 4300 20.0 9.4 -53.0 Shedloek et
Delaware al. 1993
Caroline Co., agriculture A-A' (rt. side) 125 3.5 0.1 -97.2 Speiran,
Virginia A-A' (left side) 120 7.4 2.5 -66.8 2003a
Rockingham agriculture A-A' 180 6.2 2.5 -59.7 Speiran,
Co., Virginia 2003b
Average: 374 m 15mg/L 4 mg/L -62%
^NOa' cone, values are mean ± standard error
‘’NOa’ cone, values are reported in-text by the referred author(s).
ON
ON
67
Sites listed in Table 5 that had near-stream groundwater NOb'-N concentrations
comparable to or higher than concentrations observed in this study possessed buffer
conditions and/or hydrogeologic characteristics that inhibited NO3' processing across
each riparian zone. The deep surficial aquifer at Site 2 (Spruill, 2004) for instance, likely
allowed nitrate-laden groundwater to bypass the roots of buffer vegetation at the site. In
the absence of a buried, organic-rich soil horizon, percentage NOs'-attenuation across the
deep hydrologic flowpath at Site 2 was approximately 50%. Boettger (2002) suggested
that an environment of markedly high hydraulic conductivity and corresponding fast
groundwater flow velocity at his Baldwin Swamp study site may have prevented
adequate groundwater residence time in the riparian zone of the site to sustain vigorous
N03' removal via denitrification and plant uptake. Though riparian areas of the sites of
this study had almost no deep-rooted riparian vegetation, sites listed in Table 5 with deep
and laterally extensive peats similar to the peats found at the urban and suburban sites of
this study such as Site 1 (Spruill, 2004) and transects A and B (Hill et al. 2000) recorded
nearly complete N03'-attenuation (> -95%) across riparian, floodplain aquifers. Inhibited
N03'-attenuation values observed at the urban and suburban sites of this study relative to
values observed at those sites are likely related to increased production and decreased
consumption of NO3' provoked by declining water tables across the urban and suburban
floodplain study sites.
Conclusions and Management Recommendations
Subsurface organic matter in riparian areas adjacent to low-order Coastal Plain
streams was not always confined to the upper few centimeters of the riparian sediment
profile. Subsurface organic matter occurred as shallow O- and A-horizon soils, buried
peat layers, and as “hotspots” of concentrated plant debris in sand of the unconfmed
aquifers at each site.
Groundwater levels declined adjacent to incised streams up to 22 m from the edge of
the stream channel. This decline had a notable effect in reducing contact between
floodplain groundwater and buried organic matter at the urban and suburban sites.
Despite this reduction in contact, groundwater DOC concentrations were often elevated,
especially along shallow flowpaths near peat layers. Groundwater DOC concentrations
greater than 6 mg/L and DO concentrations < 1 mg/L in shallow wells were common,
suggesting conditions were suitable for denitrification in aquifers adjacent to all three
low-order streams.
Monthly NO3' concentrations declined across all three floodplains on an annual basis.
Floodplain groundwater N03'-attenuation across the floodplain was greatest at the rural
site where the shallowest water table occurred. Less attenuation occurred at the urban
and suburban sites where the water table was deeper. This suggests that channel incision
and associated water table decline may cause a net decrease in N03'-attenuation across
floodplain aquifers of low-gradient Coastal Plain streams. Though the contribution of
DOC from buried peat may help to maintain conditions suitable for denitrification, fast
69
groundwater flowpaths along coarse-grained, basal sediments at the urban and suburban
sites may allow nitrate-laden groundwater to bypass shallow reducing zones. Elevated
concentrations of NH4'*^ in shallow wells at the urban site in combination with a positive
correlation between floodplain groundwater levels and NH4'^ concentrations at the urban
site suggest that the subaerial exposure of peat by declining water tables may contribute
inorganic-N (mineralization) to unconfined groundwater. Although near-stream NO3'
concentrations at the urban and suburban sites were elevated relative to the rural site, the
loading of NOs' to low-order stream channels is also influenced by groundwater
discharge. Because groundwater discharge is controlled by the hydraulic properties of
floodplain sediments and hydraulic gradients, discharge of groundwater from the
floodplains of the sites of this study, and thus the loading NO3' to streams of this study,
was variable across sites. These results suggest that for locations in the catchments of
low-order streams where elevated groundwater discharge occurs in tandem with impaired
attenuation of NO3' motivated by declining water tables, focused “hotspots” of NO3'
loading may occur.
Floodplain N-removal is affected by declining water tables
Restoration of water tables adjacent to incised streams could improve N03'-
attenuation in floodplain aquifers and could be achieved by; 1) reconstructing and
stabilizing the stream channels in a manner that elevates in-stream water levels to
promote contact between groundwater and organic-rich soil horizons, and 2) capturing
and re-directing stormwater from urban stormwater management networks to detention
ponds up-gradient of floodplain aquifers. The effect would be to both recharge
70
floodplain groundwater levels and to cause nitrate in captured stormwater to be
attenuated in floodplain aquifers prior to being discharged to impacted low-order streams.
In a study of the quantitative effects of two different types of geomorphic stream
restoration techniques on rates of in situ N-removal via denitrification, Kaushal et al.
(2008) concluded that geomorphic restoration whereby stream banks are armored and
lowered to promote hydrologic “connection” of surface-water with shallow soils of
riparian areas, promotes high rates of denitrification in riparian groundwater. Because
the low-order tributaries of this study are often bordered directly by residential property,
restoration that promotes hydrologic connectivity with floodplain areas may not always
be practical. Based on this current study, it is recommended that geomorphic restoration
be considered where hydrologic reconnection with riparian areas can occur without
flooding residential property.
References
Altman, S. J., and Parizek, R.R. 1995. Dilution of nonpoint-source nitrate in
groundwater. Journal ofEnvironmental Quality 24: 707- 718.
American Public Health Association (APHA). 2005. Standard Methods for the
Examination of Water and Wastewater, 2E* edition. APHA- American Water
Works Association (AWWA)-Water Pollution Control Federation (WEF):
Washington, D.C.
American Public Health Association. 1998. Standard Methods for the Examination of
Water and Wastewater, 20*'’ edition. APHA-AWWA-WEF: Washington, D.C.
An, S., and Gardner, W. S. 2002. Dissimilatory nitrate reduction to ammonium (DNRA)
as a nitrogen link, versus denitrification as a sink in a shallow estuary (Laguna
Madre/Baffm Bay, Texas). Marine Ecology - Progress Series. 237: 41-50.
Anderson, D.M., Gilbert, P.M. and Burkholder, J.M. 2002. Harmful algal blooms and
eutrophication: nutrient sources, composition, and consequences. Estuaries. 25(4):
704-726.
Angler, J. T., McCarty, G.W. 2008. Variations in base-flow nitrate flux in a first-order
stream and riparian zone. Journal of the American Water Resources Association.
44(2): 367-380.
Baker, M.A., Valett, H.M., and Dahm, C.N. 2000. Organic carbon supply and
metabolism in a shallow groundwater ecosystem. Ecology. 81 (11): 3133.
Boettger, P. C. 2002. A comparison of shallow groundwater nitrate concentrations in
vegetated vs. cultivated riparian zones adjacent to streams channelized for
agricultural drainage. M.S. Thesis, East Carolina University, Greenville, NC.
Bohlke, J.K., O'Connell, M.E. and Prestegaard, K.L. 2007. Ground water stratification
and delivery of nitrate to an incised stream under varying flow conditions. J. of
Environ. Qual, 36(3): 664-680.
Bohlke, J.K., Wanty, R., Tuttle, M., Delin, G., and Tandon, M. 2002. Denitrification in
the recharge area and discharge area of a transient agricultural nitrate plume in a
glacial outwash sand aquifer, Minnesota. Water Resources Research. 38, 7: 10.1-
10.26.
Bohlke, J.K., and Denver, J.M. 1995. Combined use of groundwater dating, chemical,
and isotopic analyses to resolve the history and fate of nitrate contamination in two
72
agricultural watersheds, Atlantic coastal plain, Maryland (Paper 95WR01584).
Water Resources Research. 31 (9): 2319.
Bouwer, H., and Rice, R.C. 1976. A slug test for determining hydraulic conductivity of
unconfmed aquifers with completely or partially penetrating wells. Water
Resources Research, 12:423-428.
Boyer, E.W., Goodale, C.L., Jaworski, N.A. and Howarth, R.W. 2002. Anthropogenic
nitrogen sources and relationships to riverine nitrogen export in the northeastern
U.S.A. Biogeochemistry, 57-58(1): 137-169.
Burkholder, J.M. 2000. Eutrophication and Oligotrophication. In: Encyclopedia of
Biodiversity. Academic Press: New York, NY, pp. 649-70.
Burt, T.P., Matchett, L.S., Goulding, K.W.T., Webster, C.P., and Haycock, N.E. 1999.
Denitrification in riparian buffer zones: the role of floodplain hydrology.
Hydrological Processes, 13 (10): 1451-1463.
Burt, T.P., Pinay, G., Matheson, F.E., Haycock, N.E., Butturini, A., Clement, J.C.,
Danielescu, S., Hefting, M.M., Hillbricht-Ilkowska, A., Maitre, V. 2002. Water
table fluctuations in the riparian zone: comparative results from a pan-European
experiment. Journal ofHydrology, 265 (1-4): 129-148.
Clausen, J.C., Guillard, K., Sigmund, C.M., and Dors, K.M. 2000. Teehnical Reports -
Ecosystem Restoration - Water Quality Changes from Riparian Buffer Restoration
in Connecticut. Journal ofEnvironmental Quality, 29 (6): 1751.
Denver, J.M., Ator, S.W., Debrewer, L.M., Ferrari, M.J., Barbaro, J.R., Hancock, T.C.,
Brayton, M.J., and Nardi, M.R. 2004. Water quality in the Delmarva Peninsula.
Delaware, Maryland, and Virginia, 1999-2001. Reston, Va., U.S. Geological
Survey Circular 1228, 32 pp.
Devito, K.J., D. Fitzgerald, A.R. Hill, and R. Aravena. 2000. Nitrate dynamics in relation
to lithology and hydrologic flow path in a river riparian zone. Journal of
Environmental Quality, 29: 1075-1084.
Dosskey, M.G., and Bertsch, P.M. 1994. Forest sources and pathways of organic matter
transport to a blackwater stream: a hydrologic approach. Biogeochemistry. 24 (1): 1.
Duff, J.H. and Triska, F.J. 2000. Nitrogen Biogeochemistry and Surface-Subsurface
Exchange in Streams. In: Jones, J.B. and Mulholland, P.J. Streams and Ground
Waters. San Diego, CA: Academic Press. 200.
Fetter, C. W. 1994. Applied Hydrogeology. Prentice-Hall, Inc.: Englewood Cliffs, NJ.
73
251-256.
Gold, A.J., Groffman, P.M., Addy, K., Kellogg, D. Q., Stolt, M., Rosenblatt, A.E. 2001.
Landscape attributes as controls on ground water nitrate removal capacity of
riparian zones. Journal of the American Water Resources Association, 37(6): 1457-
1464.
Groffman, P.M., Butterbach-Bahl, K., Fulweiler, R.W., Gold, A.J., Morse, J.L., Stander,
E.K., Tague, C., Tonitto, C., Vidon, P. 2009. Challenges to incorporating spatially
and temporally explicit phenomena (hotspots and hot moments) in denitrification
models. Biogeochemistry. 93 (1-2): 49-77.
Groffman, P.M., Bain, D.J., Band, L.E., Belt, K.T., Brush, G.S., Grove, J.M., Pouyat,
R.V., Yesilonis, I.C., and Zipperer, W.C. 2003. Down by the Riverside: Urban
Riparian Ecology. Frontiers in Ecology and the Environment, 1(6): 315-321.
Groffman, P.M., Boulware, N.J., Zipperer, W.C., Pouyat, R.V., Band, L.E., and
Colosimo, M.F. 2002. Soil nitrogen cycle processes in urban riparian zones.
Environ, and Sci. Technol, 36: 4547^552.
Groffman, P.M., Howard, G., Gold, A.J. and Nelson, W.M. 1996. Microbial Nitrate
Processing in Shallow Groundwater in a Riparian Forest. Journal ofEnvironmental
Quality, 25 (6): 1309-1316.
Gurwick, N.P., Yavitt, J.B., Groffman, P.M., Gold, A.J., Blazejewski, G., and Stolt, M.
2008. Microbially available carbon in buried riparian soils in a glaciated landscape.
Soil Biology and Biochemistry. 40 (1): 85-96.
Harden, S.L., and Spruill, T.B. 2008. Factors affecting nitrate delivery to streams from
shallow ground water in the North Carolina Coastal Plain. U.S. Geological Survey
Scientific Investigations Report 2008-5021. 39 pp. (online only at
http.V/pubs. water, usgs.gov/sir2008-5021)
Harden, S.L., Fine, J.M., and Spmill, T.B. 2003. Hydrogeology and ground-water quality
of Brunswick County, North Carolina. U.S. Geological Survey Water-Resources
Investigations Report 03-4051. 90 pp.
Hardison, E.C. 2008. Effects of Urbanization on the Ecological Condition of Small
Streams in the Inner Coastal Plain ofNorth Carolina. M.S. Thesis, East Carolina
University, Greenville, NC.
Hardison, E.C., O’Driscoll, M.A., DeLoatch, J.P., Howard, R.J., and Brinson, M.M.
2009. Urban land use, channel incision, and riparian water table decline along low-
74
order Coastal Plain streams, North Carolina. In Press: Journal of American Water
Resources.
Heath,(3R):. 2004. Basic Ground-Water Hydrology: U.S. Geological Survey Water SupplyP(4a)p:er 2220, 86 pp.Heath, R. 1975. Hydrology of the Albemarle-Pamlico Region of North Carolina: USGSWater Resources Investigation 9-75, 98 pp.Hefting, M., Clement, J.C., Dowrick, D., Cosandey, A.C., Bernal, S., Cimpian, C., Tatur,A., Burt, T.P., Pinay, G., 2004. Water table elevation controls on soil nitrogencycling in riparian wetlands along a European climatic gradient. Biogeochemistry,67(1): 113-134.Henshaw, P. C., and Booth, D. B. 2000. Re-equilibration of stream channels in urbanwatersheds. Journal ofthe American Water Resources Association, 36:1219-1236.Hill, A.R. 2000. Stream Chemistry and Riparian Zones. In: Jones, J.B. and Mulholland,P.J. Streams and Ground Waters. San Diego, CA: Academic Press. 87-94.Hill, A.R., Devito K.J., Campagnolo, S. and Sanmugadas, K. 2000. Subsurfacedenitrification in a forest riparian zone: Interactions between hydrology andsupplies of nitrate and organic carbon. Biogeochemistry 51: 193-223Hill, A.R., Vidon, P.G.F., and Langat, J. 2004. Denitrification potential in relation tolithology in five headwater riparian zones. Journal ofEnvironmental Quality, 33911-919.Holmes, R.M., Fisher, S.G. and Grimm, N.B. 1994. "Parafluvial nitrogen dynamics in adesert stream ecosystem". Journal ofthe North American Benthological Society. 13468.
Howarth, R., Sharpley, A., and Walker, D. 2002. Sources of nutrient pollution to coastal
waters in the United States: Implications for achieving coastal water quality goals.
Estuaries and Coasts, 25(4): 656-676.
Hupp, C.R. 2000. Hydrology, geomorphology and vegetation of Coastal Plain rivers in
the south-eastern USA. Hydrological Processes, 14(16-17): 2991-3010.
Hutchinson, H.W. 2007. Hydrologic and isotopic response of low-order coastal plain
streams to urban land use. M.S. Thesis, East Carolina University, Greenville, NC:
126 pp.
rNTCAL04 calibration database. 2004. rNTCAL04 Radiocarbon Age Calibration.
75
Intcal04: Calibration Issue of Radiocarbon. 46 (3).
Jones, J.B., Fisher, S.G., and Grimm, N.B. 1995. Nitrification in the hyporheic zone of a
desert stream ecosystem. Journal ofthe North American Benthological Society. 14
(2): 249.
Kamowski, E.H. and the United States. 1974. Soil survey of Pitt County, North Carolina:
Washington, NC, Soil Conservation Service: for sale by the Supt. of Docs., U.S.
Govt. Print. Off.
Kaushal, S.S., Groffman, P.M., Mayer, P.M., Striz, E. and Gold, A.J., 2008. Effects of
stream restoration on denitrification in an urbanizing watershed. Ecological
Applications. 18(3): 789-804.
Korom, S.F. 1992. Natural denitrification in the saturated zone: a review. Water
Resources Research. 41: 1657-1668.
Lautier, J.C. 2001. Hydrogeologic framework and groundwater conditions in the North
Carolina Central Coastal Plain. NC Dept, of Environment and Natural Resources,
Div. of Water Resources, 38 pp.
Madison, R.J. and Burnett, J.O. 1985. Overview of the occurrence of nitrate in ground
water of the United States. U.S. Geological Survey Water Supply Paper 2275.
Maddry, J. 1979. Geologic history of Coastal Plain streams eastern Pitt County, NC. M.S.
Thesis, East Carolina University, Greenville, NC. 102 pp.
McClain, M.E., Richey, J.E., and Pimentel, T.P. 1994. Groundwater nitrogen dynamics at
the terrestrial-lotic interface of a small catchment in the Central Amazon Basin.
Biogeochemistry. 27 (2): 113.
McDowell, W.H., Bowden, W.B., and Asbury, C.E. 1992. Riparian nitrogen dynamics in
two geomorphologically distinct tropical rain forest watersheds: subsurface solute
patterns. Biogeochemistry. 18 (2): 53.
National Research Council. 2000. Clean coastal waters: Understanding and reducing the
effects of nutrient pollution. National Academy Press: Washington, D.C. 300 pp.
O'Brien, A.K., Eshleman, K.N., and Pollard, J.S. 1994. Interactions Between
Groundwater and Surface Water in a Virginia Coastal Plain Watershed. 2. Acid-
Base Chemistry. Hydrological Processes. 8 (5): 411.
Patrick Jr., W.H. and Tusneem, M.E. 1972. Nitrogen Loss from Flooded Soil. Ecology.
53 (4): 735-737.
76
Paul, M.J. and Meyer, J.L., 2001. Streams in the urban landscape. Annual Review of
Ecology andSystematics, 32(1): 333.
Pfeiffer, S.M., Bahr, J.M., and Beilfuss, R.D. 2006. Identification of groundwater
flowpaths and denitrification zones in a dynamic floodplain aquifer. Journal of
Hydrology, 325 (1), 262.
Pickett, S.T.A., Cadenasso, M.L., Grove, J.M., Nilón, C.H., Pouyat, R.V., Zipperer,
W.C., and Costanza, R. 2001. Urban Ecological Systems: Linking Terrestrial
Ecological, Physical, and Socioeconomic Components of Metropolitan Areas.
Annual Review ofEcology and Systematics, 32: 127-57.
Ponnamperuma, F.N. The Chemistry of Submerged Soils. New York: Academic Press,
1972.
Schilling, K.E., Li, Z. and Zhang, Y.-K., 2006. Groundwater-surface water interaction in
the riparian zone of an incised channel. Walnut Creek, Iowa. Journal ofHydrology,
327(1-2): 140-150.
Schilling, K.E., Zhang, Y.K. and Drobney, P. 2004. Water table fluctuations near an
incised stream. Walnut Creek, Iowa. Journal ofHydrology, 286(1-4): 236-248.
Schlesinger, W.H. 1997. Biogeochemistry. Academic Press: San Diego, CA. 588 pp.
Schoonover, J.E., and Williard, K.W.J. 2003. Ground water nitrate reduction in giant
cane and forest riparian buffer zones. Journal ofthe American Water Resources
Association, 39 (2): 347-354.
Shedlock, R.J, Hamilton, P.A., Denver, J.M, Phillips, P.J. 1993. Multiscale approach to
regional ground-water quality assessment of the Delmarva Peninsula. In: Alley,
W.M. (ed.) Regional Ground-Water Quality. Van Nostrand Reinhold, New York,
NY, USA.
Shideler, G.L., and Swift, D.J.P. 1972. Seismic reconnaissance of post-Miocene
deposits. Middle Atlantic continental shelf - Cape Henry, Virginia, to Cape
Hatteras, North Carolina. Marine Geology, 12(3): 165-185.
Snyder, N. J., Mostaghimi, S., Berry, D. F., Reneau, R. B., Hong, S., McClellan, P. W.,
and Smith, E.P. 1998. Impact of Riparian Forest Buffers on Agricultural Nonpoint
Source Pollution. Water Resources Bulletin: 34 (2), 385.
Soban, J. R. 2007. Stream channel response to urban land use in the inner coastal plain of
North Carolina. M.S. Thesis, East Carolina University, Greenville, NC: 154 pp.
77
Soil and Plant Analysis Council, Inc. 1999. Soil Analysis: Handbook of Reference
Methods. CRC Press LLC: Boca Raton, FL. 245 pp.
Southeast Regional Climate Center. 2007. Historical Climate Summaries for North
Carolina, Dept, of Natural Resources. South Carolina, http://radar.meas.ncsu.edu/
Shedlock, R.J, Hamilton, P.A., Denver, J.M, Phillips, P.J. 1993. Multiscale approach to
regional ground-water quality assessment of the Delmarva Peninsula. In: Alley,
W.M. (ed.) Regional Ground-Water Quality. Van Nostrand Reinhold, New York,
NY, USA.
Spalding, R.F., and Exner, M.E. 1993. Occurrence of Nitrate in Groundwater. Journal of
Environmental Quality. 22 (3): 392.
Speiran, G.K. 2003a. Polecat Creek Watershed, Virginia. In: Eindsey, B. D. Residence
times and nitrate transport in ground water discharging to streams in the
Chesapeake Bay Watershed. Water-resources investigations report, 03-4035. New
Cumberland, Pa: U.S. Dept, of the Interior, U.S. Geological Survey, pp. 71-86.
Speiran, G.K. 2003b. Muddy Creek Watershed, Virginia. In: Eindsey, B. D. Residence
times and nitrate transport in ground water discharging to streams in the
Chesapeake Bay Watershed. Water-resources investigations report, 03-4035. New
Cumberland, Pa: U.S. Dept, of the Interior, U.S. Geological Survey, pp. 87-107.
Spruill, T.B. 2005. Geochemistry and characteristics of nitrogen transport at a confined
animal feeding operation in a coastal plain agricultural watershed, and implications
for nutrient loading in the Neuse River Basin, North Carolina, 1999-2002. Scientific
investigations report, 2004-5283. Reston, VA: U.S. Dept, of the Interior, U.S.
Geological Survey. http://purl.access.gpo.gov/GPO/LPS102347.
Spruill, T. B. 2004. Effectiveness of riparian buffers in controlling ground-water
discharge of nitrate to streams in selected hydrogeologic settings of the North
Carolina Coastal Plain. Water Science and Technology. 49: 63-70.
Spruill, T.B., Hamed, D.A., Ruhl, P.M., Elmers, J.L., McMahon, G., Smith, K.E.,
Galeone, D.R., Woodside, M.D. 1998. Water Quality in the Albemarle-Pamlico
Drainage Basin, North Carolina and Virginia, 1992-95: U.S. Geological Survey
Circular 1157. 41 pp.
Stoddard, J. L. 1994. Long-term changes in watershed retention of nitrogen. In L.A.
Baker (ed.) Environmental chemistry of lakes and reservoirs. Advances in
Chemistry Series. Washington, DC: American Chemical Society. Volume 237. pp.
223-284.
78
Sweet, W.V., and Geratz, J.W. 2003. Bankfull Hydraulic Geometry Relationships and
Recurrence Intervals for North Carolina's Coastal Plain. Journal ofthe American
Water Resources Association. 39: 861-872.
Talma, A.S., and Vogel, J.C. 1993. A Simplified Approach to Calibrating ''’C Dates.
Radiocarbon. 35 (2): 317-322.
Tesoriero, A.J., Spruill, T.B., and Eimers, J.L. 2004. Geochemistry of shallow ground
water in coastal plain environments in the southeastern United States: implications
for aquifer susceptibility. Applied Geochemistry, 19(9): 1471-1482.
Tesoriero, A.J., Spruill, T.B., Mew, H.E., Farrell, K.M., and Harden, S.L. 2005. Nitrogen
transport and transformations in a coastal plain watershed: Influence of
geomorphology on flow paths and residence times. Water Resources Research, 41
(2): W02008.
U.S. Environmental Protection Agency (USEPA). 2003. List of Contaminants & their
Maximum Contaminant Levels (MCLs), EPA 816-F-03-016; accessed online in
March, 2009 at http://www.epa.gov/safewater/contaminants/index.html.
U.S. Environmental Protection Agency. 1990. National survey of pesticides in drinking
water wells phase 1 report. USEPA Rep. 570/9-90-015. Office of Water, Office of
Pesticides and Toxic Substances. Washington, D.C. 98 pp.
Vepraskas, M.J. and Faulkner, S.P. 2001. Redox Chemistry of Hydric Soils. In:
Richardson, J.L., and Vepraskas, M.J. Wetland Soils: Genesis. Hydrology,
Landscapes, and Classification. Boca Raton, Fla. [u.a.]: Lewis Publ. 417 pp.
Walsh, C. J., Roy, A. H., Feminella, J.W., Cottingham, P.D., Groffman, P.M., and
Morgan, R.P. 2005. The urban stream syndrome: current knowledge and the search
for a cure. Journal of the North American Benthological Society. 24: 706-723.
Westco Scientific, 2006a. SmartChem Method 375-100E-2-Rev.A-02-0906: Nitrate-
Nitrite in water, wastewater, soil extracts and other aqueous samples. Details
available at www.WestcoScientfic.com.
Westco Scientific, 2006b. Modified SmartChem Method 390-200E-Rev.A-0906: Total
Kjeldahl Nitrogen (TKN) in Water, Waste Water, Soil Extracts and other aqueous
samples for use with a Colorimetric, Block Digestor. Modified by Cooper, B.C.
2006. Copies available on request. Details available at www.WestcoScientfic.com.
Wetzel, R.G. Limnology. Philadelphia: Saunders, 1975.
79
Williams, J.B., and Pinder, J.E. III. 1990. Ground Water Flow and Runoff in a Coastal
Plain Stream. Journal of the American Water Resources Association. 26(2): 343-
352.
Winner, M.D. JR., and Coble, R.W. 1996. Hydrogeologic framework of the North
Carolina Coastal Plain. U.S. Geol. Surv. Prof Pap. 1404-1. 106 pp.
Winner, M.D. JR., and Simmons, C.E. 1977. Hydrology of the Creeping Swamp
Watershed, North Carolina: with reference to potential effects of stream
channelization. USGS Water-resources investigation 77-26. Raleigh, NC. 54 pp.
Worrall, F., Burt, T., and Adamson, J. 2004. Can climate change explain increases in
DOC flux from upland peat catchments? The Science ofthe Total Environment. 326
(1): 95.
Appendix A: Summary Statistics of Measured Variables
Table Al. Descriptive statistics for the vertical distribution of organic matter
(processed from samples as a percent) at the three study sites. Median values for
cores from well nest locations reflect thickness and extent of buried organic matter
across each floodplain.
Organic Matter (%)
Position
Site
near-st. floodplain up-grad. Total
urban N 24 23 32 79
Median .7 1.3 2.6 1.2
Min .0 .3 .0 .0
Max 10.1 27.7 48.6 48.6
suburban N 29 32 33 94
Median 1.9 .9 .6 .9
Min .0 .0 .1 .0
Max 73.3 26.2 6.3 73.3
rural N 21 23 37 81
Median .6 .4 .5 .4
Min .2 .0 .0 .0
Max 12.4 1.6 4.5 12.4
Table A2. Descriptive statistics for hydraulic conductivity across the three
floodplains.
log K (m/s)
Site
urban suburban rural
N 6 6 6
Mean -6.5 -5.3 -6.0
Median -6.6 -6.2 -6.0
Min -7.1 -6.8 -7.8
Max -6.1 -4.7 -5.7
Range -6.1 -4.7 -5.7
81
Table A3. Logarithm of hydraulic conductivity (m/s) in wells of the urban, suburban,
and rural sites. Hydraulic conductivities were generally lower in the shallow wells of
the urban and suburban sites and the deep wells of the rural site. Higher
conductivities in deep wells of the urban and suburban sites suggest faster
groundwater flow velocity along the deeper hydrologic flowpaths.
log K (m/s)
position
near-st. floodplain up-grad.
urban shallow -6.8 -6.5 -7.0
deep -7.1 -6.5 -6.1
suburban depth -6.2 -6.1 -6.4
shallow -4.7 -5.2 -6.8
rural deep -6.7 -5.7 -6.3
depth -7.8 -5.8 -5.8
Table A4. Descriptive data summary for continuous near-stream water table depth
measured in near-stream wells of the three study sites.
Depth to water table (m)
Site
urban suburban rural
N 14236 14236 14236
Mean 1.44 .96 .81
Median 1.47 .99 .82
annual Minimum .82 .20 .00
Maximum 1.55 1.22 1.31
Range .74 1.02 1.31
Standard
.08 .12 .25
Deviation
82
Table A5. Descriptive statistics for monthly depth to groundwater data at the three
study sites.
Site: Urban
Nest location
Site
Statistic Near-St. Floodplain Up-grad. Total
N (# of samples) 12 12 12 36
annual median 1.60 1.53 1.56 1.58
max 1.67 1.69 1.84 1.84
min 1.52 1.27 1.36 1.27
range 0.15 0.42 0.48 0.57
Site: Suburban
Nest location
Site
Statistic Near-St. Floodplain Up-grad. Total
N (# of samples) 12 12 12 36
annual median 1.18 1.06 1.58 1.17
max 1.25 1.17 1.73 1.73
min 0.94 0.71 0.96 0.71
range 0.32 0.47 0.77 1.02
Site: Rural
Nest location
Site
Statistic Near-St. Floodplain Up-grad. Total
N (# of samples) 12 12 12 36
annual median 1.05 0.88 3.90 1.04
max 1.68 1.01 4.02 4.02
min 0.71 0.42 3.45 0.42
range 0.97 0.59 0.58 3.60
83
Table A6. Seasonal groundwater discharge from the floodplain of each site.
Discharge (m3/mo)
Site
urban suburban rural
Season fall Count 3 3 3
Median 4.6 11.2 12.1
Minimum 3.5 11.1 9.9
Maximum 5.0 11.3 15.9
Range 1.5 .2 6.0
winter Count 3 3 3
Median 5.3 13.6 8.9
Minimum 5.1 13.3 8.1
Maximum 6.5 19.0 14.2
Range 1.5 5.8 6.1
spring Count 3 3 3
Median 6.0 18.3 17.3
Minimum 5.6 14.9 17.0
Maximum 6.1 22.8 18.6
Range .5 8.0 1.6
summer Count 3 3 3
Median 4.5 10.2 14.4
Minimum 4.5 10.2 14.4
Maximum 4.7 10.6 15.0
Range .3 .3 .7
84
Table A7. Data summary for groundwater levels (m AMSL) in shallow, up-gradient
and near-stream wells at the three study sites. Shallow groundwater levels at both the
up-gradient and near-stream wells had the smallest annual range at the urban site.
Groundwater level (m AMSL)
Site
urban suburban rural
Position Up-gradient N 12 12 12
Median 11.6 10.0 5.1
Minimum 11.3 9.9 4.9
Maximum 11.8 10.6 5.5
Range .5 .8 .6
Near-stream N 12 12 12
Median 10.7 9.4 4.5
Minimum 10.7 9.3 4.3
Maximum 10.8 9.6 4.9
Range .1 .3 .6
85
Table A8. Descriptive Statistics for seasonal groundwater temperatures across the
range of urban land use based on hand-dipped monthly measurements in all 6 wells of
each floodplain.
Temp (C)
Site
urban suburban rural
Season annual N 72 72 72
Med. 18.6 17.3 17.2
Min. 15.3 13.5 11.7
Max. 22.6 20.8 21.9
Range 7,3 7.3 10,2
fall N 18 18 18
Med. 20.0 18.5 19.0
Min, 19,6 17.3 16.5
Max. 22.6 20.5 20.9
Range 3.0 3.2 4.4
winter N 18 18 18
Med. 17.0 14.8 14.2
Min. 15.3 13.5 11,7
Max. 18.5 16.5 17.2
Range 3.2 3.0 5.5
spring N 18 18 18
Med. 17.1 15.3 15.5
Min. 16.1 14.2 14.1
Max. 18,6 17.1 17.7
Range 2.5 2.9 3.6
summer N 18 18 18
Med. 20.8 19.0 19.5
Min. 18,1 18.0 17.1
Max, 22.0 20.8 21.9
Range 3.9 2.8 4.8
86
Table A9. Descriptive statistics based on hand-dipped monthly measurements for
annual temperature (C) by position in shallow and deep wells of the three study sites.
Temp (C)
Position
up-gradient floodplain near-st.
Site urban deep N 12 12 12
Median 19.0 19.1 19.3
Minimum 15.8 15.4 16.0
Maximum 21.0 21.9 22.6
shallow N 12 12 12
Median 18.2 19.1 19.1
Minimum 16.3 15.3 15.6
Maximum 19.9 22.2 21.9
suburban deep N 12 12 12
Median 16.7 16.9 18.3
Minimum 13.7 13.8 14.5
Maximum 19.0 20.4 20.8
shallow N 12 12 12
Median 16.6 16.7 17.3
Minimum 13.7 13.5 14.1
Maximum 18.9 19.6 20.3
rural deep N 12 12 12
Median 17.2 17.4 17.4
Minimum 14.9 12.1 14.1
Maximum 19.5 21.9 20.0
shallow N 12 12 12
Median 17.3 16.8 17.1
Minimum 15.0 11.7 12.5
Maximum 19.7 21.9 20.9
Table AlO. Descriptive data for NOa'-N concentrations (mg/L) from floodplain wells
of the three sites grouped by depth.
Nitrate-N (mg/L)
position
up-grad. floodplain near-st. total
Site urban shallow Count 12 12 12 36
Median .1 .1 .1 .1
Minimum ,0 .0 .0 .0
Maximum .2 ,2 .3 .3
Range .2 ,2 .2 .3
deep Count 12 12 12 36
Median .4 .3 .2 .3
Minimum .1 .0 .1 .0
Maximum .5 ,5 .3 .5
Range .4 .5 ,2 .5
suburban shallow Count 12 12 12 36
Median .1 .0 .1 .1
Minimum .0 .0 .1 .0
Maximum .3 .2 .2 .3
Range .2 .2 .1 .3
deep Count 12 12 12 36
Median .5 .1 .1 .1
Minimum .4 .0 .1 ,0
Maximum .6 .3 .2 .6
Range .2 .3 .2 .5
rural shallow Count 12 12 12 36
Median 9.3 9.6 .1 8.3
Minimum 7.0 7.5 .0 ,0
Maximum 12,3 11,2 .2 12.3
Range 5.3 3.7 .2 12.3
deep Count 12 12 12 36
Median 3.9 1.0 .1 1.0
Minimum 2,0 .3 ,0 .0
Maximum 5.3 3.7 .1 5.3
Range 3.4 3.4 .1 5.3
88
Table All. Descriptive data for in-stream nitrate concentrations compared to
average near-stream nitrate concentrations from shallow and deep near-stream wells.
Nitrate-N (mg/L)
Position
near-st. avg. stream
Site urban N 12 12
Median .1 .2
Minimum .1 .0
Maximum .2 .5
Range .1 .5
suburban N 12 12
Median .1 1.6
Minimum .0 .9
Maximum .2 2.3
Range ,1 1,4
rural N 12 12
Median .1 .3
Minimum .0 .0
Maximum .1 ,8
Range .1 .8
Table A12. Comparative statistics for nitrate-N and ammonium-N (mg/L) from shallow wells along the three study
transects.
Nitrate-N (mg/L) Ammonium-N (mg/L)
Position Position
up-qrad. floodplain near-st. stream up-qrad. floodplain near-st. stream
urban Count 12 12 12 12 12 12 12 12
Median .1 .1 .1 .2 .4 2.6 1.1 .1
Minimum .0 .0 .0 .0 .0 .0 .8 .0
Maximum .2 .2 .3 .5 .6 5.3 1.3 .3
Range .2 .2 .2 .5 .6 5.3 .5 .3
suburban Count 12 12 12 12 12 12 12 12
Median .1 .0 .1 1.6 .1 .1 .1 .1
Minimum .0 .0 .1 .9 .0 .0 .0 .0
Maximum .3 .2 .2 2.3 .1 .2 .1 .3
Range .2 .2 .1 1.4 .1 .1 .1 .2
rural Count 12 12 12 12 12 12 12 12
Median 9.3 9.6 .1 .3 .1 .1 .7 .0
Minimum 7.0 7.5 .0 .0 .0 .0 .5 .0
Maximum 12.3 11.2 .2 .8 .1 .2 1.1 .1
Range 5.3 3.7 .2 .8 .1 .2 .6 .1
00
'O
Table A13. Comparative statistics for nitrate-N and ammonium-N (mg/L) from deep wells along the three study transects.
Nitrate-N (mg/L) Ammonium-N (mg/L)
Position Position
up-grad. floodplain near-st. stream up-grad. floodplain near-st. stream
Site urban Count 12 12 12 12 12 12 12 12
Median .4 .3 .2 .2 .2 .1 .2 .1
Minimum .1 .0 .1 .0 .1 .0 .1 .0
Maximum .5 .5 .3 .5 .4 .3 .2 .3
Range .4 .5 .2 .5 .3 .3 .1 .3
suburban Count 12 12 12 12 12 12 12 12
Median .5 .1 .1 1.6 .1 .1 .1 .1
Minimum .4 .0 .1 .9 .0 .0 .0 .0
Maximum .6 .3 .2 2.3 .1 .1 .1 .3
Range .2 .3 .2 1.4 .0 .1 .1 .2
rural Count 12 12 12 12 12 12 12 12
Median 3.9 1.0 .1 .3 .1 .1 .5 .0
Minimum 2.0 .3 .0 .0 .1 .0 .1 .0
Maximum 5.3 3.7 .1 .8 .4 .2 1.8 .1
Range 3.4 3.4 .1 .8 .3 .2 1.7 .1
o
Table Al4. Comparative statistics for nitrate-N and DON (mg/L) by position and depth across the three study transects.
Nitrate-N (mg/L) DON from 5/30/2008 (mg/L)
position position
up- up- near-
qradient floodplain near-st. stream gradient floodplain st. stream
urban shallow N 12 12 12 12 1 1 1 1
Median .1 .1 .1 .9 .2 .7 .3 .3
deep N 12 12 12 12 1 1 1 1
Median .4 .3 .2 .9 .1 .5 .6 .3
suburban shallow N 12 12 12 12 1 1 1 1
Median .1 .0 .1 7.0 .2 .6 .3 .2
deep N 12 12 12 12 1 1 1 1
Median .5 .1 .1 7.0 .2 .8 ,4 .2
rural shallow N 12 12 12 12 1 1 1 1
Median 9.3 9.6 .1 1.5 .3 .7 .5 .7
deep N 12 12 12 12 1 1 1 1
Median 3.9 1.0 .1 1.5 .7 .3 .7 .7
Table A15. Seasonal statistics for nitrate-N and ammonium-N (mg/L) from shallow and deep wells along the three
transects.
Nitrate-N (mg/L) Ammonium-N (mg/L)
Season Season
fall winter spring summer fall winter spring summer
urban shallow N 9 9 9 9 9 9 9 9
Median .1 .0 .0 .1 1.1 .3 1.0 .9
Range .2 .2 .1 .2 2.2 2.2 5.1 4.0
deep N 9 9 9 9 9 9 9 9
Median .2 .3 .4 .3 .2 .1 .2 .3
Range .3 .4 .4 .3 .2 .1 .3 .2
suburban shallow N 9 9 9 9 9 9 9 9
Median .2 .1 .0 .1 .1 .1 .1 .1
Range .2 .1 .1 .1 .1 .1 .1 .1
deep N 9 9 9 9 9 9 9 9
Median .1 .1 .1 .2 .1 .0 .1 .1
Range .5 .5 .4 .5 .1 .1 .0 .1
rural shallow N 9 9 9 9 9 9 9 9
Median 11.1 8.2 7.5 8.4 .1 .1 .1 .1
Range 12.3 10.4 9.5 10.2 1.1 .7 .7 .7
deep N 9 9 9 9 9 9 9 9
Median 1.0 .5 .5 1.6 .1 .1 .2 .2
Range 2.8 5.3 4.4 4.6 .6 .2 .9 1.7
o
to
93
Table A16. Descriptive statistics for DOC concentrations across study sites by
floodplain (f.p.) position. Concentrations are grouped by depth to show variation
with depth and distance from organic rich sediment layers.
DOC (mg/L)
Position
up-
gradient floodplain near-st f.p. total
Site urban shallow N 12 12 12 36
Median 6.2 14.2 4.2 5.7
Minimum 1.3 .0 3.1 .0
Maximum 7.7 27.2 6.1 27.2
Range 6.4 27.2 2.9 27.2
deep N 12 12 12 36
Median 2.6 1.9 1.3 2.0
Minimum .0 .4 1.0 .0
Maximum 20.5 3.6 6.5 20.5
Range 20.5 3.2 5.5 20.5
suburban shallow N 12 12 12 36
Median 1.4 5.1 2.1 2.8
Minimum .5 1.0 .6 .5
Maximum 4.5 13.8 8.2 13.8
Range 4.0 12.9 7.5 13.3
deep N 12 12 12 36
Median 1.5 1.2 1.1 1.3
Minimum .4 .2 .1 .1
Maximum 15.4 3.8 3.7 15.4
Range 14.9 3.6 3.6 15.2
rural shallow N 12 12 12 36
Median 1.5 2.2 3.6 2.3
Minimum .3 .5 1.6 .3
Maximum 5.5 5.5 7.5 7.5
Range 5.2 5.0 5.9 7.2
deep N 12 12 12 36
Median 1.5 1.6 9.0 2.2
Minimum .5 .9 4.9 .5
Maximum 3.7 5.2 33.3 33.3
Range 3.2 4.4 28.5 32.8
94
Table Al7. Descriptive statistics for dissolved oxygen (DO) concentrations across
the floodplains of the three study sites. Data has been grouped by depth and
floodplain position for consideration of spatial relationships.
DO (mg/L)
Position
up-
gradient floodplain near-st fp. total
N 11 11 11 33
Median 1.0 1.0 .8 .8
shallow Minimum .5 .4 .3 .3
Maximum 3.8 5.6 1.3 5.6
Range 3.3 5.2 .9 5.3
urban
N 11 11 11 33
Median 1.9 1.5 .8 1.5
deep Minimum .4 .3 .0 .0
Maximum 3.7 2.7 1.9 3.7
Range 3.3 2.4 1.9 3.7
N 11 11 11 33
Median .7 .3 .4 .5
shallow Minimum .5 .2 .2 .2
Maximum 3.7 1.4 1.2 3.7
Range 3.2 1.2 1.0 3.5
Site suburban
N 11 11 11 33
Median 1.2 .4 .6 .7
deep Minimum .6 .2 .1 .1
Maximum 1.7 .9 1.5 1.7
Range 1.1 .7 1.3 1.6
N 11 11 11 33
Median .8 .5 .4 .5
shallow Minimum .3 .1 .2 .1
Maximum 1.7 1.3 1.7 1.7
Range 1.4 1.2 1.5 1.6
rural
N 11 11 11 33
Median .6 .4 .3 .4
deep Minimum .2 .2 .1 .1
Maximum 1.9 2.4 1.1 2.4
Range 1.7 2.2 1.0 2.3
95
Table A18. Chloride concentrations from floodplain well samples at the three sites
grouped by depth. Chloride concentrations were determined for samples from five
monthly sampling events from 2/29/2008 to 8/13/2008.
Chloride cone. (mg/L)
Position
up-grad floodplain near-st in-stream
Site urban shallow N 5 5 5 5
Median 12.6 7.2 10.9 21.8
Minimum 10.1 5.1 7.1 6.7
Maximum 15.1 8.0 11.8 22.6
Range 5.0 2.9 4.7 15.9
deep N 5 5 5
Median 10.2 7.8 9.4
Minimum 9.6 5.7 7.4
Maximum 13.2 10.5 10.9
Range 3.7 4.9 3.5
suburban shallow N 5 5 5 5
Median 15.1 16.8 17.3 24.2
Minimum 12.0 11.8 15.6 15.1
Maximum 16.9 30.1 28.0 43.0
Range 4.9 18.3 12.4 27.9
deep N 5 5 5
Median 20.0 21.0 23.5
Minimum 14.2 14.0 17.8
Maximum 38.4 31.1 29.3
Range 24.2 17.1 11.5
rural shallow N 5 5 5 5
Median 28.1 40.0 25.6 15.7
Minimum 23.0 30.5 17.3 12.2
Maximum 31.5 44.4 44.0 18.7
Range 8.4 13.9 26.7 6.5
deep N 5 5 5
Median 28.4 31.1 19.0
Minimum 21.1 23.7 10.2
Maximum 32.0 35.5 48.2
Range 10.9 11.8 38.0
Appendix B: Electroconductivity Response Logs
Figure Bl. Generalized location map showing approximate layout of the 5-6 point
electroconductivity grids used to estimate floodplain stratigraphy in preparation for well
transect installation at each study site.
Appendix B1
Urban Electroconductivity Logs
 
99
Appendix B2
Suburban Electroconductivity Logs
Suburban 01 Suburban 02
Sediment EC (mS/m) Sediment EC (mS/m)
60 110 160 210 60 100 140 180 220
Suburban 03 Suburban 04
Sediment EC (mS/m) Sediment EC (mS/m)
60 100 140 180 220 0 50 100 150 200
 
101
Appendix B3
Rural Electroconductivity Logs
102
* Electroconductivity log Rural 04 likely experienced considerable interference from an
overlying electric dog fence.
Appendix C: Sediment Core Logs
Figure Cl. Symbol key for lithologic symbols used on the core logs to delineate distinct
stratigraphic units and their characteristics.
Lithologic symbols used for core logs
rooted surface soil PPl loamy sand
n~l sand with clay lenses clay
plant debris and wood chips FPl shell material lenses
^ peat gravel driveway
r~n fine-grained sand [v^ very coarse sand and gravel
1 : 1 medium-grained sand WM roots and plant debris
1 ? , '\ coarse-grained sand major facies change
1 ~ 1 sandy loam
Note i: Core labels correspond with well nest locations shown on
floodplain transects throughout this manuscript, i.e. Urban
Floodplain Core N1 corresponds with the near-stream N1
wells shown in Figures 4,12, and 15.
Urban Floodplain Core NI
Lat. 35°35'41.8"N
Long. 77°21'23.77" W
rooted surficial soil
rooted, gray peat
saturated black peat
coarse-grained sand
NR
coarse-grained sand
NR
coarse-grained sand
clayey orange coarse-grained sand
clayey orange medium grained sand
clay
Urban Floodplain Core N2
Lat. 35°35'41.9"N
Long. 77°21'24.r'W
£
-C
a
<v %OM
•D 0 10 20 30
0
surficial soil with gravel 0.5-3 cm diam.
surficial soil with gravel 0.5-3 cm diam.
light gray peat
1 dark, organic rich peat
organic rich peat with thin lenses of
coarse-grained sand
clayey fine-grained sand
2
clayey, orange, coarse-grained sand
3
orange, coarse-grained sand
4
5
Urban Floodplain Core N3
Lat. 35°35’41.9"N
Long. 77°2r24.1"W
E
4-»
Q.
OJ %OM
~o 0 25 50
0
rooted surficial soil bearing traces of plant debris
and woody plant material
gravel (1-4 cm grains)
organic rich, gravelly surficial soil
gravel (1-4 cm grains)
light brown peat bearing lenses of coarse-
and fine-grained sand
1
light brown peat
dark brown to black peat
2
rooted, dark brown peat
dark brown peat
3
fine-grained sand
medium-grained sand
4 coarse-grained sand
orange-colored, coarse-grained sand
deep orange-stained clay
5
Suburban Floodplain Core N1
Lat. 35°34'47.8" N
Long. 77°19'56.7"W
Û. %OM
<D
?D 0 40 80
108
Suburban Floodplain Core N2
Lat. 35°34'48.1" N
Long. 77°19’57.2" W
Suburban Floodplain Core N3
Lat. 35°34'48.6" N
Long. 77°19'57.6" W
E
f. %OM
Rural Floodplain Core N1
Lat. 35°35'23.0"N
Long. 77° 15'45.5"W
Ê
S % OM
(U 0 5 10 15
Rural Floodplain Core N2
Lat. 35°35'22.9"N
Long. 77°15'46.0"W
g
Q. % OM
Rural Floodplain Core N3
Lat. 35°35'22.8"N
Long. 77°15’46.5"W
sandy surficial soil
fine-grained sand and dirt bearing trace
woody debris
fine-grained sand and dirt bearing trace
driveway gravel
fining upward light gray, fine- to
medium-grained sand
sandy loam
light gray, fine-grained sand
loamy sand
coarse-grained sand with lag pebble deposits
fine-grained sand
medium-grained, poorly sorted sand
NR
coarse-grained sand with thin lenses of finer
sediment
NR
coarse-grained sand with lenses of
fine-grained sand and mud
NR
orange, loamy, fine-grained sand
densely packed clay
densely packed clay with -1% trace shell frags
Appendix D: Percent Organic Matter Data
Table Dl. Sediment sample depths and LOI results for the urban site.
Core
Urban N1 Urban N2 Urban N3
Depth (m) % OM Depth (m) % OM Depth (m) % OM
14 1.9 4 0.6 16 2.5
28 2.7 9 1.8 25 1.7
42 5.5 31 1.7 36 4.2
92 10.1 39 2.0 44 3.1
100 9.4 49 1.3 78 2.8
122 4.1 60 4.2 89 2.7
130 6.5 71 9.8 101 7.2
139 2.7 82 3.9 114 10.7
147 0.4 90 11.2 122 30.2
157 0.0 100 1.3 148 31.3
164 0.3 110 14.6 160 40.5
172 0.1 117 27.7 171 48.6
218 0.0 131 1.2 185 38.3
226 0.1 141 1.5 198 19.7
256 0.2 149 0.3 210 35.1
261 1.0 160 0.4 218 20.3
273 0.4 174 0.3 225 21.0
289 0.0 234 0.3 246 9.5
308 0.3 259 0.5 319 0.5
321 0.4 289 0.4 329 0.4
328 0.5 330 0.4 344 0.1
339 1.1 337 0.5 348 0.2
344 1.2 344 0.7 402 1.0
348 1.0 411 0.2
432 0.1
440 0.1
455 0.9
465 0.3
Median 0.7 1.3 3.0
Min 0.0 0.3 0.1
Max 10.1 27.7 48.6
Range 10.1 27.4 48.5
114
Table D2. Sediment sample depths and LOI results for the suburban site.
Core
Suburban N1 Suburban N2 Suburban N3
Depth (m) % OM Depth (m) % OM Depth (m) % OM
2 4.2 48 2.9 6 6.3
13 1.9 59 26.2 18 0.9
22 1.7 69 12.4 27 0.2
55 4.1 77 6.6 75 0.8
63 4.7 86 7.4 86 0.8
74 7.5 93 20.2 98 0.4
83 25.5 100 3.0 108 0.3
94 21.0 110 2.4 126 0.2
106 3.9 123 5.1 140 0.6
122 9.6 134 9.4 156 0.3
130 21.3 189 0.1 172 0.3
138 10.0 226 0.2 213 0.7
146 1.8 259 2.7 225 0.5
154 4.9 269 0.2 237 0.8
204 0.1 278 1.2 248 0.3
212 2.5 289 0.1 260 0.1
220 61.8 298 0.4 310 0.3
228 73.3 307 0.2 323 0.5
233 0.9 316 1.7 337 0.5
282 0.1 325 0.4 389 0.4
291 0.1 334 0.9 401 0.4
302 1.6 349 0.6 415 1.0
312 0.1 361 0.9 425 0.2
324 0.2 373 0.5 440 1.1
342 1.5 385 0.5 451 1.0
372 0.3 397 0.8 468 0.7
398 0.7 409 2.5 478 0.3
425 0.5 421 0.9 490 0.8
433 0.9 509 0.9
445 0.7 520 1.2
533 0.9
544 1.1
560 0.9
Median 2.2 0.9 0.6
Min 0.1 0.1 0.1
Max 73.3 26.2 6.3
Range 73.2 26.1 6.2
115
Table D3. Sediment sample depths and LOI results for the rural site.
Core
Rural N1 Rural N2 Rural N3
Depth (m) % OM Depth (m) %OM Depth (m) % OM
4 6.8 14 1.6 9 1.5
16 0.2 66 0.3 20 1.5
29 0.3 75 0.3 61 1.8
43 0.2 85 0.2 74 4.5
99 0.8 96 0.4 85 0.8
107 0.4 107 0.9 97 0.8
115 2.7 147 0.3 108 0.9
121 1.4 158 0.1 118 1.7
130 2.6 168 0.1 157 1.0
180 12.4 181 0.5 182 0.6
188 10.9 204 0.4 209 1.3
195 0.6 224 0.1 219 0.8
201 0.8 288 1.2 236 1.0
212 0.3 299 0.3 253 0.1
221 0.3 310 0.2 306 0.2
267 0.3 320 0.4 315 0.5
275 0.2 333 1.3 325 0.5
287 0.2 341 0.4 334 0.4
298 0.3 359 1.3 342 0.5
306 0.7 370 0.5 395 0.4
320 1.6 382 1.2 409 0.2
420 0.6
433 0.1
445 0.5
458 0.1
502 0.7
543 0.2
557 0.3
570 0.2
601 2.1
612 0.9
Median 0.6 204.0 0.4 315.0 0.6
Min 0.2 14.0 0.1 9.0 0.1
Max 12.4 382.0 1.6 612.0 4.5
Range 12.3 368.0 1.5 603.0 4.4
Appendix E: Hydrologic Data
Table El. Depth to water (m) values for shallow and deep wells at each site.
Well 10/5/2007 11/8/2007 11/29/2007 1/8/2008 2/1/2008 2/29/2008 4/12/2008 5/10/2008 5/30/2008 7/15/2008 8/13/2008 9/4/2008
Urban N1-S 1.66 1.61 1.60 1.59 1.57 1.58 1.52 1.58 1.62 1.67 1.61 1.66
Urban N1-D 1.64 1.66 1.62 1.59 1.57 1.55 1.58 1.57 1.61 1.65 1.66 1.64
Urban N2-S 1.69 1.55 1.58 1.51 1.50 1.37 1.27 1.41 1.50 1.64 1.55 1.62
Urban N2-D 1.58 1.55 1.56 1.50 1.50 1.40 1.33 1.43 1.49 1.59 1.58 1.61
Urban N3-S 1.84 1.58 1.63 1.52 1.54 1.36 1.39 1.42 1.50 1.65 1.67 1.69
Urban N3-D 1.69 1.66 1.67 1.63 1.62 1.52 1.42 1.51 1.57 1.69 1.68 1.72
Suburban N1-S 1.25 1.19 1.19 1.15 1.15 0.98 0.94 1.08 1.16 1.25 1.23 1.23
Suburban N1-D 1.22 1.16 1.16 1.11 1.12 0.94 0.90 1.04 1.25 1.23 1.22
Suburban N2-S 1.17 1.09 1.10 1.01 1.03 0.78 0.71 0.91 1.03 1.17 1.15 1.13
Suburban N2-D 1.10 1.01 1.01 0.94 0.96 0.72 0.63 0.80 0.95 1.11 1.09 1.06
Suburban N3-S 1.69 1.65 1.65 1.50 1.52 1.15 0.96 1.24 1.45 1.72 1.73 1.69
Suburban N3-D 1.76 1.63 1.73 1.54 1.57 1.20 1.01 1.30 1.51 1.78 1.77 1.73
Rural N1-S 1.30 1.07 1.10 0.92 0.90 0.90 0.71 0.90 1.00 1.18 1.26 1.16
Rural N1-D 1.41 1.10 1.05 0.94 1 05 0.84 0.81 0.92 1.05 1.25 1.33 1.21
Rural N2-S 1.01 0.88 0.95 0.88 0.89 0.66 0.42 0.53 0.65 0.87 0.96 1.01
Rural N2-D 0.91 0.82 0.94 0.87 0.90 0.66 0.41 0.53 0.64 0.84 0.94 1.00
Rural N3-S 3.92 3.90 4.02 3.90 3.92 3.69 3.45 3.55 3.67 3.89 3.97 4.02
Rural N3-D 3.96 3.92 4.02 3.81 3.92 3.69 3.45 3.54 3.65 3.88 3.97 4.01
Table E2. Calculated elevation head (m) values for shallow and deep wells at each site..
Well 10/5/2007 11/8/2007 11/29/2007 1/8/2008 2/1/2008 2/29/2008 4/12/2008 5/10/2008 5/30/2008 7/15/2008 8/13/2008 9/4/2008
Urban N1-S 10.7 10.7 10.7 10.8 10.8 10.8 10.8 10.8 10.7 10.7 10 7 10.7
Urban N1-D 10.7 10.7 10.7 10.8 10.8 10.8 10.8 10.8 10.7 10.7 10.7 10.7
Urban N2-S 11.2 11.3 11.3 11.3 11.4 11.5 11.6 11.4 11.4 11.2 11.3 11.2
Urban N2-D 11.3 11.3 11.3 11.3 11.4 11.5 11.5 11.4 11.4 11.3 11.3 11.2
Urban N3-S 11.3 11.6 11.6 11.7 11.7 11.8 11.8 11.8 11.7 11.5 11.5 11.5
Urban N3-D 11.5 11.5 11.5 11.6 11.6 11.7 11.8 11.7 11.6 11.5 11.5 11.5
Suburban N1-S 9.3 9.3 9.3 9.4 9.4 9.6 9.6 9.5 9.4 9.3 9.3 9.3
Suburban N1-D 9.3 9.4 9.4 9.4 9.4 9.6 9.6 9.5 10.5 9.3 9.3 9.3
Suburban N2-S 9.4 9.5 9.4 9.5 9.5 9.8 9.8 9.6 9.5 9.4 9.4 9.4
Suburban N2-D 9.5 9.5 9.5 9.6 9.6 9.8 9.9 9.7 9.6 9.4 9.5 9.5
Suburban N3-S 9.9 9.9 9.9 10.1 10.1 10.4 10.6 10.4 10.1 9.9 9.9 9.9
Suburban N3-0 9 8 10.0 9.9 10.1 10.0 10.4 10.6 10.3 10.1 9.8 9.8 9.9
Rural N1-S 4 3 4.5 4.5 4.7 4.7 4.7 4.9 4.7 46 4.4 4.3 4.4
Rural N1-D 4.2 4.5 4.5 4.6 4.5 4 7 4.8 4.7 4.5 4.3 4.2 4 4
Rural N2-S 4 9 5.1 5.0 5.1 5.0 5.3 5.5 5.4 5.3 5.1 5.0 4 9
Rural N2-D 5.0 5.1 5.0 5.1 5.0 5.3 5.5 5.4 5.3 5.1 5.0 4.9
Rural N3-S 5.0 5.1 4.9 5.1 5.1 5.3 5.5 5.4 5.3 5.1 5.0 4.9
Rural N3-D 5.0 5.0 4.9 5.2 5.0 5.3 5.5 5.4 5.3 5.1 5.0 5.0
Table E3. Monthly hydraulic gradient (m/m) and discharge (m^/month) data for the three floodplain aquifers.
Site
Urban Suburban Rural
dh/dl Q dh/dl Q dh/dl Q
Date (m/m) (mVmo) (m/m) (m^/mo) (m/m) (m^/mo)
10/5/2007 0.023 3.5 0.031 11.2 0.028 15.9
11/8/2007 0.031 5.0 0.030 11.3 0.020 12.1
11/29/2007 0.029 4.6 0.029 11.1 0.017 9.9
1/8/2008 0.033 5.3 0.035 13.6 0.015 8.9
2/1/2008 0.031 5.1 0.034 13.3 0.013 8.1
2/29/2008 0.038 6.5 0.044 19.0 0.021 14.2
4/12/2008 0.034 6.1 0.051 22.8 0.023 17.0
5/10/2008 0.036 6.0 0.044 18,3 0.027 18.6
5/30/2008 0.034 5.6 0.038 14.9 0.026 17.3
7/16/2008 0,031 4.7 0.029 10.6 0.025 15.0
8/13/2008 0.028 4.5 0.028 10.2 0.024 14.4
9/5/2008 0.028 4.5 0.028 10.2 0,024 14.4
Appendix F: Water Quality Data
Table FI. Monthly groundwater specific conductivity data (uS/m) measured in wells of the three study sites.
s pecific Conductivity (uS/m)
Well 10/5/07 11/8/07 11/29/07 1/8/08 2/1/08 2/29/08 4/12/08 5/10/08 5/30/08 7/15/08 8/30/08 9/5/08
Urban N1-S 72.0 61.8 59.3 77.7 61.0 122.5 84.9 68.9 64.1 68.8 64.4 68.5
Urban N1-D 244.3 123.2 152.0 83.0 107.2 54.7 59.9 59 60.5 58.8 61.5 67.1
Urban N2-S 215.2 178.5 141.1 52.8 95.1 100.1 76 6 86 108.4 122.4 104.8 101.1
Urban N2-D 59.3 53.0 55.1 87.7 49.5 48.8 51.4 53.2 56.8 62.9 59.7 59.5
Urban N3-S 151.5 112.5 113.0 78.7 82.4 92.9 70.6 94.5 119.2 99.5 98.5 105.8
Urban N3-D 57.5 57.8 56 6 52.6 51.0 50 0 534 71.4 72.4 71 5 65.6 58.0
Suburban N1-S 73.6 65.6 111.6 71.5 80.9 75.6 73.4 76.8 88.4 81.2 75.2 72.6
Suburban N1-D 151.2 138.7 145.6 144.1 142.8 136.2 133.0 145.6 157.2 151.5 149 6
Suburban N2-S 76.9 144.3 101.0 150.8 127.4 111.6 73.0 78.2 150.2 82.4 87.7 155.4
Suburban N2-D 126.2 144.7 132.9 134.9 135.4 125.1 143.1 125.8 97.4 142.2 130.6 138.5
Suburban N3-S 75.8 73.1 84.0 74.5 74.8 76.6 786 80.3 68.6 81.8 79.5 81.1
Suburban N3-D 59.7 57.9 58.2 54.1 52.3 51.5 52.5 54.4 97.1 61.6 58.4 60.5
Rural N1-S 156.3 59.2 136.6 142.9 150.0 82.0 74.5 108.4 213.7 193.4 178.0 146.0
Rural N1-D 275.0 169.9 250.1 204.7 216.4 193.8 53.9 132.3 227.0 223.0 150.1 248.2
Rural N2-S 227.3 219.7 213.4 185.0 169.5 168.6 167.7 177.6 230.0 216.1 230.1 238.9
Rural N2-D 192.8 179.8 162.3 152.5 161.3 151.6 169.1 169.8 203.5 190.0 193.6 178.7
Rural N3-S 233.5 219.2 223.5 221.9 221.3 176.4 152.7 174.1 234.0 186.2 197.6 182.4
Rural N3-D 180.7 173.4 164.0 172.3 179.3 166.5 172.1 168.1 222.8 170.7 174.0 155.4
Table F2. Monthly groundwater temperature data (°C) measured in shallow and deep wells of the three study sites.
Date
Well 10/5/07 11/8/07 11/29/07 1/8/08 2/1/08 2/29/08 4/12/08 5/10/08 5/30/08 7/15/08 8/13/08 9/4/08
Urban N1-S 21.9 20.9 20.0 18.0 16.9 15.6 16.8 17.6 18.2 20.5 21.1 21.8
Urban N1-D 22.6 21.1 20.0 18.5 17.6 16.0 16.7 17.3 17.7 20.1 20.6 21.2
Urban N2-S 22.2 21.4 19.6 18.0 16.2 15.3 16.1 17.2 18.6 20.7 21.4 22.0
Urban N2-D 21.2 21.1 19.9 17.7 16.9 15.4 16.2 17.2 18.3 20.8 21.4 21.9
Urban N3-S 19.6 19.9 19.7 18.2 17.2 16.3 16.3 16.6 17.1 18.1 18.8 19.6
Urban N3-D 19.8 20.0 19.6 18.4 17.1 15.8 16.2 17 17.1 20.2 21.0 19.6
Suburban N1-S 20.3 19.5 18.0 16.5 14.1 14.2 15.1 16.6 16.3 19.7 19.7 20.0
Suburban N1-D 20.5 19.4 18.3 16.2 14.8 14.5 15.0 17.1 20.8 19.3 20.0
Suburban N2-S 19.6 19.0 17.3 15.5 14.7 13.5 14.2 15.3 16.1 18.4 18.5 19.6
Suburban N2-D 18.8 18.3 17.9 15.8 14.8 13.8 14.3 15.3 15.9 19.0 18.7 20.4
Suburban N3-S 18.9 18.4 17.3 15.4 14.7 13.7 14.6 15.5 15.9 18.3 18.3 18.8
Suburban N3-D 18.6 18.0 17.5 15.7 15.3 13.7 14.5 15.2 15.9 18.0 18.3 19.0
Rural N1-S 20.9 19.1 17.8 13.8 14.0 12.5 15.0 15.6 16.4 20.1 20.1 20.8
Rural N1-D 20.0 18.8 18.4 17.2 14.1 14.2 15.0 16.1 15.9 17.5 18.6 18.9
Rural N2-S 20.3 19.1 16.5 13.3 12.7 11.7 14.4 15.4 17.1 20.5 21.9 21.3
Rural N2-D 20.1 18.6 17.1 14.3 12.5 12 1 14.1 15.2 17.7 20.5 21.9 21.4
Rural N3-S 19.7 19.2 18.8 17.2 16.4 15.0 15.0 15.3 15.9 17.4 18.5 18.5
Rural N3-D 19.5 18.7 18.7 17.2 16.1 14.9 15.5 15.4 16.2 17.1 18.0 18.6
Appendix G: Monthly Water Chemistry Data
Table Gl. Groundwater chemistry data from the sampling date of 10/5/2007.
Well NO3 -N (mg/L) NH2-N (mg/L) DOC (mg/L) DO (mg/L) Cl- (mg/L)
Urban N1-S 0,3 1,3 3.1 6.4
Urban N1-D 0.1 0.2 6,5 6,8
Urban N2-S 0,1 2.4 13.7 6,9
Urban N2-D 0.3 0.1 0.7 5.7
Urban N3-S 0.1 0.6 6,4 7.2
Urban N3-D 0.1 0.2 5.9 6.9
Urban SW 0.1 0.1 2.0
Suburban N1-S 0.2 0.1 3.5 6.4
Suburban N1-D 0,1 0.1 1.5 6.5
Suburban N2-S 0.2 0.1 2.7 5.8
Suburban N2-D 0.1 0.1 1.8 6.5
Suburban N3-S 0.2 0.1 1.8 5,8
Suburban N3-D 0.5 0,1 2.6 6.1
Suburban SW 2.3 0.1 0.5
Rural N1-S 0.1 1,1 3.7
Rural N1-D 0.1 0.4 8.3 3.7
Rural N2-S 10.3 0.1 2,0 5.0
Rural N2-D 1.2 0.1 1.8 4.7
Rural N3-S 11.1 0.1 1.1 4.6
Rural N3-D 2.9 0.1 3.2 3.9
Rural SW 0.1 0.0 3.1
122
Table G2. Groundwater chemistry data from the sampling date of 11/8/2007
Well NO3 -N (mg/L) NHi*-N (mg/L) DOC (mg/L) DO (mg/L) Cl- (mg/L)
Urban N1-S 0.1 1.1 4.0 0.8
Urban N1-D 0.3 0.2 3.3 1.9
Urban N2-S 0.1 2.6 14.2 4.3
Urban N2-D 0.3 0.2 1.8 1.6
Urban N3-S 0.1 0.5 6.5 3.8
Urban N3-D 0.2 0.2 2.9 1.9
Urban SW 0.1 0.1 1.3
Suburban N1-S 0.1 0.1 0.6 0.9
Suburban N1-D 0.2 0.1 0.5 0.7
Suburban N2-S 0.1 0.1 7.6 1.4
Suburban N2-D 0.1 0.1 0.5 0.9
Suburban N3-S 0.2 0.1 1.1 3.7
Suburban N3-D 0.5 0.1 0.5 1.6
Suburban SW 2.2 0.3 0.9
Rural N1-S 0.1 0.7 4.3 1.7
Rural N1-D 0.1 0.7 10.2 0.1
Rural N2-S 11.2 0.1 2.0 0.7
Rural N2-D 1.0 0.1 1.1 2.4
Rural N3-S 11.8 0.1 1.8 1.4
Rural N3-D 2.0 0.1 2.0 0.6
Rural SW 0.3 0.1 2.8
123
Table G3. Groundwater chemistry data from the sampling date of 11/29/2007
Well NOa -N (mg/L) NHi'-N (mg/L) DOC (mg/L) DO (mg/L) Cl- (mg/L)
Urban N1-S 0.0 0.8 4.2 0.5
Urban N1-D 0,2 0.1 3.1 0,5
Urban N2-S 0.2 2.6 8.6 0.5
Urban N2-D 0.0 0.0 1.4 1.8
Urban N3-S 0.0 0,5 6.0 0.7
Urban N3-D 0.3 0.2 3.2 3.7
Urban SW 0.0 0.0 2.2 3.5
Suburban N1-S 0.1 0.0 8.2 0.7
Suburban N1-D 0.1 0.1 1.1 0.7
Suburban N2-S 0.0 0.0 6.7 0.5
Suburban N2-D 0,1 0.0 0.6 0.4
Suburban N3-S 0.3 0.0 2.8 0.7
Suburban N3-D 0.5 0.0 1,1 0.6
Suburban SW 1.6 0.1 1.5 4.2
Rural N1-S 0.0 0.8 2.5 0.4
Rural N1-D 0.1 0.2 4.9 0.3
Rural N2-S 11.1 0.1 1,3 0.5
Rural N2-D 0,9 0.0 1.2 0.3
Rural N3-S 12.3 0.1 1.3 0.3
Rural N3-D 2.7 0.2 0.9 0.6
Rural SW 0.0 0,1 11.3 1.4
124
Table G4. Groundwater chemistry data from the sampling date of 1/8/2008
Well NO3 -N (mg/L) NHGN (mg/L) DOC (mg/L) DO (mg/L) Cl- (mg/L)
Urban N1-S 0.1 1.2 4.9 0.8
Urban N1-D 0.1 0.1 1.1 0.6
Urban N2-S 0.0 0.0 0.5
Urban N2-D 0.1 0.1 0.9 0,8
Urban N3-S 0.0 0.2 3,1 1.1
Urban N3-D 0.5 0.1 1.4 2.5
Urban SW 0.3 0.3 2.1 6.1
Suburban N1-S 0.1 0.0 0.9 1.2
Suburban N1-D 0.1 0,0 0.1 1.0
Suburban N2-S 0.0 0.1 6.0 0.6
Suburban N2-D 0,3 0.0 0.4 0.9
Suburban N3-S 0.1 0,0 0.5 1,0
Suburban N3-D 0.5 0.0 0.6 1.7
Suburban SW 1.6 0.1 1.0 7.0
Rural N1-S 0.2 0.7 2.7 1.0
Rural N1-D 0.0 0.2 8.5 0.8
Rural N2-S 9.9 0.1 1.4 1.3
Rural N2-D 0.6 0.1 1.5 1.1
Rural N3-S 10.4 0.0 1.4 1.7
Rural N3-D 3.4 0.1 1.0 1.2
Rural SW 0.3 0.0 3.7
125
Table G5. Groundwater chemistry data from the sampling date of 2/1/2008
Well NO3 -N (mg/L) NH2-N (mg/L) DOC (mg/L) DO (mg/L) Cl-(mg/L)
Urban N1-S 0.2 1.1 4.2 0.6
Urban N1-D 0.1 0.2 1.0 0.0
Urban N2-S 0.0 0.0 3.5
Urban N2-D 0.3 0.1 0.4 1.5
Urban N3-S 0.0 0.0 5.1 0.5
Urban N3-D 0.5 0.1 0.0 2.2
Urban SW 0.2 0.0 4.7 8.7
Suburban N1-S 0.1 0.1 1.1 0.4
Suburban N1-D 0.1 0.1 0.6 0.4
Suburban N2-S 0.0 0.1 4.1 0.4
Suburban N2-D 0.0 0.1 0.2 0.3
Suburban N3-S 0.1 0.0 1.1 0.7
Suburban N3-D 0.4 0.0 0.4 1.6
Suburban SW 1.5 0.0 1.1 8.1
Rural N1-S 0.0 0.7 1.6 0.3
Rural N1-D 0.1 0.1 13.7 0.3
Rural N2-S 8.5 0.0 0.5 0.5
Rural N2-D 0.3 0.1 1.4 0.4
Rural N3-S 9.7 0.1 0.5 1.1
Rural N3-D 5.3 0.1 0.5 0.4
Rural SW 0.4 0.0 3.3 10.8
126
Table G6. Groundwater chemistry data from the sampling date of 2/29/2008
Well NO3 -N (mg/L) NH«*-N (mg/L) DOC (mg/L) DO (mg/L) Cl- (mg/L)
Urban N1-S 0.1 1.2 5.7 1.2 10.9
Urban N1-D 0.2 0.1 1.1 1.9 7.4
Urban N2-S 0.0 2.2 1.3 5.1
Urban N2-D 0.3 0.1 1.3 1.8 5.7
Urban N3-S 0.0 0.3 4.5 0.7 10.1
Urban N3-D 0.5 0.1 1.3 2.6 13.2
Urban SW 0.3 0.1 2.1 9.9 21.8
Suburban N1-S 0.1 0.1 1.8 0.3 28.0
Suburban N1-D 0.1 0.1 0.6 0.6 29.3
Suburban N2-S 0.0 0.1 9.8 0.2 30.1
Suburban N2-D 0.0 0.1 0.5 0.7 31.1
Suburban N3-S 0.0 0.1 1.2 0.6 16.9
Suburban N3-D 0.4 0.0 0.7 1.1 20.8
Suburban SW 1.8 0.1 1.2 9.3 26.6
Rural N1-S 0.0 0.5 7.5 0.3 17.3
Rural N1-D 0.0 0.3 4.9 0.3 19.0
Rural N2-S 8.2 0.1 2.9 1.0 44.4
Rural N2-D 0.5 0.1 2.3 1.4 31.1
Rural N3-S 7.1 0.1 1.6 0.8 31.5
Rural N3-D 4.6 0.1 1.8 1.9 29.6
Rural SW 0.8 0.0 5.3 10.5 18.6
127
Table G7. Groundwater chemistry data from the sampling date of 4/12/2008
Well NOj -N (mg/L) NH/-N (mg/L) DOC (mg/L) DO (mg/L) Cl- (mg/L)
Urban N1-S 0,0 1.2 6.1 0.8 11.8
Urban N1-D 0.2 0.1 1,3 1,9 10.9
Urban N2-S 0.0 5.3 5.6 7.5
Urban N2-D 0.4 0.3 3.0 2.7 10.5
Urban N3-S 0.0 0.2 1.3 1.0 15.1
Urban N3-D 0,5 0.1 20.5 3.6 11.1
Urban SW 0.2 0.0 5.4 22.3
Suburban N1-S 0.1 0.0 1.4 0.4 21.1
Suburban N1-D 0.1 0,0 3.6 1.1 24.8
Suburban N2-S 0.0 0.1 1.0 0,5 17.1
Suburban N2-D 0.1 0.0 3.6 0.7 22,7
Suburban N3-S 0.0 0.0 0,8 0.7 15.7
Suburban N3-D 0.4 0.0 15,4 1.1 20.0
Suburban SW 1.6 0.0 1.4 7,6 24.2
Rural N1-S 0.1 0.6 3.6 0.8 28.8
Rural N1-D 0.0 0,3 14.5 0,2 10.2
Rural N2-S 7.5 0,1 2.5 0.3 40,0
Rural N2-D 0.5 0.1 1.9 0.7 31.9
Rural N3-S 7,0 0.1 2.7 0,8 29.8
Rural N3-D 4.4 0.1 1.5 0.4 32.0
Rural SW 0,6 0.0 7.9 3.4 18.7
128
Table G8. Groundwater chemistry data from the sampling date of 5/1/2008
Well NO3 -N (mg/L) NHGN (mg/L) DOC (mg/L) DO (mg/L) Cl- (mg/L)
Urban N1-S 0,0 0.9 5.4 0,4 7.1
Urban N1-D 0.2 0.2 2.0 0.5 8,2
Urban N2-S 0.1 3.1 19.6 1.9 6.1
Urban N2-D 0.5 0.2 2.5 0.3 7.5
Urban N3-S 0,1 0.4 7.7 0.8 11.1
Urban N3-D 0.2 0.4 4.1 0,5 9.6
Urban SW 0.3 0,1 8.6 3.8 11,6
Suburban N1-S 0.1 0.0 3.5 0.2 15.6
Suburban N1-D 0.1 0.1 0.9 0.3 17,8
Suburban N2-S 0,0 0.1 13.8 0.2 14.2
Suburban N2-D 0.1 0.0 1.6 0.2 17.8
Suburban N3-S 0.1 0.1 3.8 0.7 12.0
Suburban N3-D 0.4 0.1 4.2 0.9 14.2
Suburban SW 1.5 0.1 2.2 6,3 16.4
Rural N1-S 0.1 0.5 4.8 0.7 21.8
Rural N1-D 0.0 0.9 33.3 1.1 19.1
Rural N2-S 8.2 0.1 2.9 0.4 30.5
Rural N2-D 0.5 0.1 4.0 0.8 23.7
Rural N3-S 9.2 0.0 1.8 0.6 23.9
Rural N3-D 4.4 0.4 1.4 0.2 21.1
Rural SW 0.5 0.1 6.5 6.6 15.7
129
Table G9. Groundwater chemistry data from the sampling date of 5/30/2008
Well NO3 -N (mg/L) NHGN (mg/L) DOC (mg/L) DO (mg/L) Cl- (mg/L)
Urban N1-S 0.0 1.0 3.9 0,6
Urban N1-D 0,2 0.2 1.4 1.2
Urban N2-S 0.0 2.3 18.5 0.4
Urban N2-D 0,5 0.2 3,6 1.4
Urban N3-S 0,2 0,5 6,8 1.9
Urban N3-D 0.4 0.3 2.1 0.4
Urban SW 0.2 0.1 3.2 6.9
Suburban N1-S 0.1 0.1 3.1 0.9
Suburban N1-D 0.0 0.0
Suburban N2-S 0.0 0.2 7.5 0.3
Suburban N2-D 0.1 0.1 3.8 0.2
Suburban N3-S 0.0 0.0 4.5 1,5
Suburban N3-D 0.5 0,1 4.2 0.7
Suburban SW 1.7 0.1 1.8 7.1
Rural N1-S 0.1 0,7 5.3 0.2
Rural N1-D 0.1 0.9 9.5 0.1
Rural N2-S 9.4 0.1 2.4 0.1
Rural N2-D 1.1 0.1 1.5 0.2
Rural N3-S 9.5 0.1 1.2 0.4
Rural N3-D 3.6 0.2 1.2 0.6
Rural SW 0.4 0.1 5.3 8.3
130
Table GIO. Groundwater chemistry data from the sampling date of 7/16/2008
Well NO3 -N (mg/L) NH2-N (mg/L) DOC (mg/L) DO (mg/L) Cl- (mg/L)
Urban N1-S 0.0 1.0 3.8 0.8 11.5
Urban N1-D 0.3 0.2 1.0 0.8 9.4
Urban N2-S 0.0 4.0 25.4 0.7 8.0
Urban N2-D 0.4 0.3 2.0 0.9 7.8
Urban N3-S 0.2 0.6 2.4 12.6
Urban N3-D 0.4 0.3 1.3 10.2
Urban SW 0.2 0.1 4.7 3.9 23.0
Suburban N1-S 0.1 0.1 2.3 0.2 22.6
Suburban N1-D 0.1 0.1 3.7 1.5 17.3
Suburban N2-S 0.1 0.1 2.5 0.2 23.5
Suburban N2-D 0.1 0.1 3.0 0.4 11.8
Suburban N3-S 0.1 0.1 1.4 0.5 14.0
Suburban N3-D 0.6 0.1 1.9 1.5 15.1
Suburban SW 2.0 0.1 1.3 6.4 28.4
Rural N1-S 0.0 0.8 2.0 0.3 38.4
Rural N1-D 0.0 1.8 24.4 0.4 43.0
Rural N2-S 10.0 0.1 4.2 0.3 44.0
Rural N2-D 1.6 0.2 3.8 0.3 48.2
Rural N3-S 8.4 0.1 2.3 1.0 40.0
Rural N3-D 4.1 0.2 2.0 0.5 35.5
Rural SW 0.1 0.0 2.1 5.7 15.0
131
Table Gil. Groundwater chemistry data from the sampling date of 8/13/2008
Well NO3 -N (mg/L) NHGN (mg/L) DOC (mg/L) DO (mg/L) Cl- (mg/L)
Urban N1-S 0.0 0.9 4.3 0.3 9.2
Urban N1-D 0.2 0.2 1.9 0.5 10.2
Urban N2-S 0.1 3.4 27.2 0.4 7.2
Urban N2-D 0.2 0.1 2.6 0.5 8.3
Urban N3-S 0.1 0.5 7.0 0.7 12.7
Urban N3-D 0.4 0.3 2.5 1.0 9.7
Urban SW 0.5 0.1 7.4 6.1 6.7
Suburban N1-S 0.1 0.1 6.4 0.3 16.6
Suburban N1-D 0.2 0.1 3.7 0.1 21.1
Suburban N2-S 0.1 0.1 3.5 0.3 16.8
Suburban N2-D 0.1 0.0 3.2 0.4 21.0
Suburban N3-S 0.1 0.1 3.3 0.7 13.7
Suburban N3-D 0.6 0.0 1.5 1.3 17.0
Suburban SW 0.9 0.1 3.6 6.7 15.1
Rural N1-S 0.1 0.8 2.7 0.5 25.6
Rural N1-D 0.1 0.9 7.4 0.7 14.9
Rural N2-S 9.2 0.2 1.3 0.4 35.7
Rural N2-D 1.3 0.2 0.9 0.4 29.2
Rural N3-S 8.5 0.1 0.3 0.3 28.1
Rural N3-D 4.7 0.1 0.6 0.2 25.0
Rural SW 0.2 0.1 3.7 6.4 12.2
132
Table G12. Groundwater chemistry data from the sampling date of 9/5/2008
Well NO3 -N (mg/L) NHGN (mg/L) DOC (mg/L) DO (mg/L) Cl- (mg/L)
Urban N1-S 0.1 0.9 3.5 1.3 6.4
Urban N1-D 0.2 0.2 1.2 1.5 8.6
Urban N2-S 0.1 3.8 7.1 1.0 3.3
Urban N2-D 0.2 0.3 2.0 1.5 5.8
Urban N3-S 0.0 0.0 2.1
Urban N3-D 0.5 0.3 2.6 1.5 5.2
Urban SW 0.1 0.1 1.9 4.6 8.8
Suburban N1-S 0.1 0.1 1.8 0.3 12.1
Suburban N1-D 0.2 0.1 1.3 0.5 15.1
Suburban N2-S 0.0 0.1 3.8 0.2 20.4
Suburban N2-D 0.0 0.1 0.7 0.3 15.5
Suburban N3-S 0.1 0.1 1.4 0.8 8.3
Suburban N3-D 0.5 0.1 1.5 1.2 11.0
Suburban SW 1.8 0.1 1.1 4.6 13.4
Rural N1-S 0.1 0.7 3.7 0.2 13.4
Rural N1-D 0.1 0.5 6.4 0.3 11.3
Rural N2-S 10.2 0.1 5.5 0.7 28.1
Rural N2-D 3.7 0.1 5.2 0.4 20.5
Rural N3-S 7.7 0.1 5.5 0.8 18.2
Rural N3-D 3.2 0.1 3.7 0.8 11.2
Rural SW 0.1 0.0 3.0 6.3 7.5