ArticlePDF Available

Fate of Effluent-Borne Contaminants beneath Septic Tank Drainfields Overlying a Karst Aquifer

Authors:

Abstract and Figures

Groundwater quality effects from septic tanks were investigated in the Woodville Karst Plain, an area that contains numerous sinkholes and a thin veneer of sands and clays overlying the Upper Floridan aquifer (UFA). Concerns have emerged about elevated nitrate concentrations in the UFA, which is the source of water supply in this area of northern Florida. At three sites during dry and wet periods in 2007-2008, water samples were collected from the septic tank, shallow and deep lysimeters, and drainfield and background wells in the UFA and analyzed for multiple chemical indicators including nutrients, nitrate isotopes, organic wastewater compounds (OWCs), pharmaceutical compounds, and microbiological indicators (bacteria and viruses). Median NO3-N concentration in groundwater beneath the septic tank drainfields was 20 mg L(-1) (8.0-26 mg L(-1)). After adjusting for dilution, about 25 to 40% N loss (from denitrification, ammonium sorption, and ammonia volatilization) occurs as septic tank effluent moves through the unsaturated zone to the water table. Nitrogen loading rates to groundwater were highly variable at each site (3.9-12 kg N yr(-1)), as were N and chloride depth profiles in the unsaturated zone. Most OWCs and pharmaceutical compounds were highly attenuated beneath the drainfields; however, five Cs (caffeine, 1,7-dimethylxanthine, phenol, galaxolide, and tris(dichloroisotopropyl)phosphate) and two pharmaceutical compounds (acetaminophen and sulfamethoxazole) were detected in groundwater samples. Indicator bacteria and human enteric viruses were detected in septic tank effluent samples but only intermittently in soil water and groundwater. Contaminant movement to groundwater beneath each septic tank system also was related to water use and differences in lithology at each site.
Content may be subject to copyright.
1181
Groundwater quality eff ects from septic tanks were investigated
in the Woodville Karst Plain, an area that contains numerous
sinkholes and a thin veneer of sands and clays overlying the
Upper Floridan aquifer (UFA). Concerns have emerged
about elevated nitrate concentrations in the UFA, which is
the source of water supply in this area of northern Florida. At
three sites during dry and wet periods in 2007–2008, water
samples were collected from the septic tank, shallow and deep
lysimeters, and drainfi eld and background wells in the UFA and
analyzed for multiple chemical indicators including nutrients,
nitrate isotopes, organic wastewater compounds (OWCs),
pharmaceutical compounds, and microbiological indicators
(bacteria and viruses). Median NO3–N concentration in
groundwater beneath the septic tank drainfi elds was 20 mg
L1 (8.0–26 mg L1). After adjusting for dilution, about 25 to
40% N loss (from denitrifi cation, ammonium sorption, and
ammonia volatilization) occurs as septic tank effl uent moves
through the unsaturated zone to the water table. Nitrogen
loading rates to groundwater were highly variable at each site
(3.9–12 kg N yr1), as were N and chloride depth profi les in the
unsaturated zone. Most OWCs and pharmaceutical compounds
were highly attenuated beneath the drainfi elds; however, fi ve
OWCs (caff eine, 1,7-dimethylxanthine, phenol, galaxolide, and
tris(dichloroisotopropyl)phosphate) and two pharmaceutical
compounds (acetaminophen and sulfamethoxazole) were
detected in groundwater samples. Indicator bacteria and
human enteric viruses were detected in septic tank effl uent
samples but only intermittently in soil water and groundwater.
Contaminant movement to groundwater beneath each septic
tank system also was related to water use and diff erences in
lithology at each site.
Fate of E uent-Borne Contaminants beneath Septic Tank Drain elds
Overlying a Karst Aquifer
Brian G. Katz,* Dale W. Gri n, and Peter B. McMahon U.S. Geological Survey
Harmon S. Harden Florida State University
Edgar Wade and Richard W. Hicks Florida Department of Environmental Protection
Je rey P. Chanton Florida State University
A    households in the United States uses
a septic tank for wastewater disposal (USEPA, 2003).
Conventional septic tanks are the most commonly used sys-
tems and consist of a tank that allows solids to settle out and
septic tank effl uent (STE) to fl ow to a drainfi eld where it per-
colates downward through soil and the unsaturated zone toward
the water table. Various biogeochemical processes occur beneath
the drainfi eld that can modify and attenuate certain chemical
and microbiological constituents in the STE. Numerous studies
have assessed the impacts of STE on groundwater quality, with
particular emphasis on the transformation and fate of nutrients
(Wilhelm et al., 1994; Aravena and Robertson, 1998; Hinkle et
al., 2008), bacteria and viruses (DeBorde et al., 1998; Rose et al.,
2000; Darby and Leverenz, 2004), organic wastewater contami-
nants (e.g., Hinkle et al., 2005; Swartz et al., 2006; Conn et al.,
2006), and pharmaceuticals (Seiler et al., 1999; Godfrey et al.,
2007; Carrara et al., 2008). Subsurface pathways of nutrients and
wastewater organic compounds (including pharmaceuticals and
personal care products) are increasingly recognized as a problem
for aquifers (e.g., Barnes et al., 2008; Focazio et al., 2008).
Reasonable estimates of the nitrogen (N) fl ux from septic tanks
to groundwater exist, but this fl ux is mitigated to an unknown
degree by physical, chemical, and microbiological processes
occurring in the shallow subsurface and in the unsaturated zone
above the water table. Apparent loss or attenuation of N beneath
septic tank drainfi elds has been observed in varied hydrogeo-
logic settings (e.g., Robertson et al., 1991; Ptacek, 1998; Seiler,
2005; Hinkle et al., 2008). Apparent N loss has been attributed
to several processes, including ammonia volatilization, sorption
of ammonium (NH4+) and organic N on soil particles, denitri-
cation, and dilution from infi ltrating rainfall and mixing with
ambient groundwater. Without knowing the amount of N, phos-
phorus, and organic wastewater compounds stored in the unsatu-
rated zone beneath on-site waste disposal sites, it is not possible to
Abbreviations: bls, below land surface; cfu, colony-forming units; DOC, dissolved
organic carbon; KN, Kjeldahl nitrogen; OWCs, organic wastewater compounds; RT-PCR,
reverse transcriptase–polymerase chain reaction; STE, septic tank e uent; UFA, Upper
Floridan aquifer; WKP, Woodville Karst Plain.
B.G. Katz and D.W. Gri n, U.S. Geological Survey, Florida Water Science Center, 2639
N. Monroe St., Tallahassee, FL 32303; P.B. McMahon, U.S. Geological Survey, Colorado
Science Center, Lakewood, CO 80225; H.S. Harden and J. P. Chanton, Florida State Univ.,
Dep. of Oceanography, Tallahassee, FL 32306; E. Wade and R.W. Hicks, Florida Dep.
of Environmental Protection, Tallahassee, FL 32399. Supplemental material available
online for this article. Assigned to Associate Editor Lakwhinder Hundal.
Copyright © 2010 by the American Society of Agronomy, Crop Science
Society of America, and Soil Science Society of America. All rights
reserved. No part of this periodical may be reproduced or transmitted
in any form or by any means, electronic or mechanical, including pho-
tocopying, recording, or any information storage and retrieval system,
without permission in writing from the publisher.
J. Environ. Qual. 39:1181–1195 (2010)
doi:10.2134/jeq2009.0244
Published online 18 Mar. 2010.
Supplemental  les available online for this article.
Received 30 June 2009.
*Corresponding author (bkatz@usgs.gov).
© ASA, CSSA, SSSA
5585 Guilford Rd., Madison, WI 53711 USA
SPECIAL SUBMISSIONS
1182 Journal of Environmental Quality • Volume 39 • July–August 2010
accurately model the movement of contaminants to groundwa-
ter and springs from these systems.
A steady increase in nitrate-nitrogen (NO3–N) concen-
trations (0.25–0.9 mg L1) during the past 30 yr in Wakulla
Springs, a large regional discharge point for water from the
Upper Floridan aquifer (UFA), has increased concerns about
groundwater contamination in the Woodville Karst Plain
(WKP) in northern Florida (Katz et al., 2004).  ere are an
estimated 20,000 septic tanks in the WKP (Fig. 1), an area
highly susceptible to groundwater contamination due to
numerous sinkholes and a relatively thin layer of highly perme-
able sands and clays overlying the karstic UFA. Although septic
tanks have been presumed to be a signifi cant source of N to the
unconfi ned groundwater system (Chelette et al., 2002), little
direct information exists on the impact of STE on groundwa-
ter quality. Studies in other mantled karst systems have shown
that septic tank systems contributed to elevated NO3 concen-
trations in groundwater (e.g., Panno et al., 2006; Harden et al.,
2003, 2008). Chemicals in many household products can enter
the subsurface from STE in more concentrated amounts, such
as optical brighteners (Murray et al., 2007). Pharmaceutical
compounds have been found in shallow zones of the UFA in
areas where septic tanks are presumably aff ecting groundwater
quality (Katz and Griffi n, 2008; Katz et al., 2009).
Our study had two main objectives: (i) to estimate N load-
ing to groundwater from septic tanks in the WKP and (ii) to
determine the attenuation of microorganisms, organic waste-
water compounds (OWCs), and pharmaceutical compounds
in the subsurface beneath septic tank drainfi eld sites in the
WKP. Multiple chemical indicators are used to elucidate pro-
cesses controlling the transport and fate of nutrients, OWCs,
and pharmaceutical compounds as effl uent moves downward
from septic tank drainfi elds to the groundwater system. In
addition, fecal indicators and enteroviruses are used to examine
the movement of microorganisms beneath the drainfi eld sites.
e multitracer approach used in this study can be applied to
a variety of locations in diff erent soil types to improve model
predictions of impacts from NO3 and other contaminants
from septic tanks on groundwater quality.
Description of Study Area and Sites
e study area (approximately 1100 km2) lies within the WKP
and is located in southern Leon and Wakulla Counties (Fig.
1).  e WKP is characterized by numerous wet and dry sink-
holes, conduit networks, and disappearing streams (Rupert and
Spencer, 1988).  e study area is mostly rural, with low-den-
sity residential areas (most lot sizes >0.4 ha). In the WKP, the
UFA is overlain by a thin veneer of highly porous quartz sand
with highly variable amounts of silt and clay (Fig. 1). Streams
owing into this area fl ow into sinkholes and disappear, and
other streams originate in the WKP from spring discharge.
Land-surface altitudes in the WKP are generally less than 15 m
above sea level.  e climate is humid subtropical with an aver-
age annual air temperature of 19.5°C and average annual pre-
cipitation of 161 cm (1971–2000), measured at the National
Weather Service station at the Tallahassee Regional Airport
(http://cdo.ncdc.noaa.gov/ancsum/ACS).  e largest amount
of monthly rainfall typically occurs in June, July, and August,
and the smallest monthly amount in occurs April, October,
and November. During August 2007 through July 2008, rain-
fall was about 28 cm below average. Recharge to the UFA in
the study area is about 46 cm yr1 (Davis, 1996), and ground-
water fl ow in the UFA generally is from north to south (Fig. 1)
based on previous potentiometric maps (Chelette et al., 2002).
Fig. 1. Location map showing septic tank sites and generalized lithology at each site.
Katz et al.: Contaminants beneath Septic Tank Drain elds 1183
Site Characteristics
e three sites selected for this study (Fig. 1) represent a range
of residences and water usage that likely are representative of
septic tank systems in the WKP. Water meters were installed
at each house, and the septic tank at each site was pumped
before the start of the study. Drainfi eld pipes are located at
approximately 0.6 m below land surface.  e northernmost
site (LT) had two to three adult residents who had lived in the
house since it was built in 1987; the household used the origi-
nal septic tank and drainfi eld. Average daily water use at the
LT site was 394 L d1. No fertilizers had been applied by the
current residents of the house. Depth to groundwater ranged
from 3.0 to 3.6 m during the study.
e HK site had four residents (two adults and two teenage
children), and the septic tank had been in use for approximately
30 yr. Average daily water use during the study period was
1630 L d1, although water discharge to the septic system was
reduced in early 2008 through the use of water conservation
measures.  e septic tank was installed in the mid 1970s, and
the drainfi eld was repaired or replaced on three occasions.  e
original drainfi eld, consisting of terra cotta pipe into a gravel
bed, was extended in 1984.  e extension was two gravel-fi lled
trenches with perforated black plastic pipe.  at drainfi eld
failed in 2005 and was replaced with infi ltrators in February
2005. Due to excessive loading of water, that drainfi eld failed
in December 2007 and was replaced in January 2008 by the
current one, which also has infi ltrators.  e turf over the new
septic tank drainfi eld (approximately 80 m2) was seeded with
rye and given a few handfuls of fertilizer in January 2008. In
May 2008, it was seeded with a summer grass mix, and about
2 kg of fertilizer (10–10–10) was added. In addition, <2 kg of
10–10–10 fertilizer was applied to a garden (about 70 m2) next
to the new drainfi eld. Depth to the water table ranged from 2.6
to 2.7 m during the study.
e YG site had two adult residents that have lived in the
household for 4 yr.  e house was built around 2003.  e
original septic tank system was in use at the time of the study.
Average daily water use was 610 L d1. No fertilizers had been
applied by the residents of the house. Depth to the water table
ranged from 4.1 to 4.4 m during the study.
Subsurface Lithology
All three sites have variable amounts of sands and clays that
overlie limestone of the Upper Floridan aquifer. Soils have low
slopes (0–5%), have low organic matter content (<2%), and
are classifi ed as Alpin sand and Blanton fi ne sands (thermic,
coated Typic Quartzipsamments), Alpin fi ne sands, and Ortega
sands (thermic, uncoated Typic Quartzipsamments) at the LT,
HK, and YG sites, respectively (Allen, 1991). Based on obser-
vations of core material collected during well installation, there
are important diff erences between the lithology of subsurface
material at the three sites (Fig. 1). Sand comprises the upper 2
to 3 m of the subsurface at each site and overlies clayey sand
at the LT and YG sites but overlies weathered limestone at the
HK site. Depth to limestone was variable at each site; the great-
est depth to limestone was at the YG site (Fig. 1). Particle size
analyses of samples from various depth intervals (from land
surface to 123 cm below land surface) at each site indicate vari-
ous mixtures of very fi ne to coarse sand and an increasing per-
centage of silt and clay with depth at each site.
Materials and Methods
e sampling design at each site (Supplemental Fig. S1) con-
sisted of two sets of lysimeters, a background well located
upgradient from the septic tank drainfi eld, a well beneath the
drainfi eld, and STE. A sampling port was installed by a septic
tank contractor in the pipe that connects the septic tank to the
drainfi eld to allow sampling of the STE.
Four suction-cup lysimeters were installed by hand auger
beneath the drainfi eld at each of the three sites to collect pore
water samples moving in the unsaturated zone. Details about
lysimeter construction are included in Supplemental Fig. S1.
Each of the shallow lysimeters had a total length of 110 cm and
collected soil water samples from a depth of 92 to 118 cm below
land surface. At this depth the shallow lysimeters were immedi-
ately below the bottom of the drainfi eld. Each of the deep lysim-
eters had a total length of 186 cm, sampling from a depth of 168
to 194 cm below land surface. To obtain samples, a suction of
60 kPa was applied to the lysimeters for a period of about 24 h
before sample collection. Nitrogen gas was used to apply a pres-
sure of 5 to 10 kPa to force the water into a collection vessel.
Drainfi eld and background wells were installed at each site
using a direct push rig.  ese wells consisted of a 1.9-cm-diameter
PVC casing with a 1.5-m-length prepacked screen that was placed
below the water table. A peristaltic pump was used to purge three
well volumes and to collect water samples for analyses.
Water Samples
Water samples were collected at each site on three occasions:
December 2007, March 2008, and July 2008. On each occa-
sion, samples were collected from the STE, the shallow and
deep lysimeters, the drainfi eld well, and the background well.
Two lysimeter samples from each depth were combined to
obtain the required volume of sample for laboratory analyses.
Sample collection and preservation methods are described in
the National Field Manual for the Collection of Water-Quality
Data (U.S. Geological Survey, 1999). Specifi c conductance,
pH, dissolved oxygen, and temperature were measured in
water samples in the fi eld. Major ions, nutrients, boron, and
dissolved organic carbon (DOC) were analyzed at the U.S.
Geological Survey National Water Quality Laboratory (USGS
NWQL) in Denver, Colorado (Fishman and Friedman, 1989;
Patton and Truitt, 1992; Brenton and Arnett, 1993; Fishman,
1993; Patton and Truitt, 2000).
Stable isotopes of N, oxygen, and hydrogen were used to
help diff erentiate between sources of N in water samples and to
evaluate N transformation processes. Nitrogen isotope analyses
of dissolved NH4+ (δ15N-NH4) were performed at the USGS
Stable Isotope Laboratory in Reston, Virginia, using continu-
ous-fl ow isotope ratio mass spectrometry (Brenna et al., 1997;
Hannon and Böhlke, 2008) and are reported in parts per thou-
sand (‰) relative to N2 in air.  e 2-sigma (σ) uncertainty of
δ15N-NH4 analyses is 0.8‰.
e stable isotope composition of NO3 (δ15N-NO3 and
δ18O-NO3) in water samples was analyzed at the USGS Stable
Isotope Laboratory in Reston, Virginia, by bacterial conversion
1184 Journal of Environmental Quality • Volume 39 • July–August 2010
of NO3 to nitrous oxide and subsequent measurement on a
continuous fl ow isotope ratio mass spectrometer (Sigman et
al., 2001; Casciotti et al., 2002; Coplen et al., 2004). Values
of δ15N-NO3 are reported in ‰ relative to N2 in air (Mariotti,
1983). For samples with NO3–N concentrations ≥0.06 mg L1,
the 2σ uncertainty of N isotopic results is 0.5‰; for samples
with NO3 concentrations <0.06 mg kg1 as N, it is 1‰. Values
of δ18O-NO3 are reported in ‰ relative to Vienna Standard
Mean Ocean Water (reference water) and normalized on a
scale such that Standard Light Antarctic Precipitation reference
water is 55.5‰ (Coplen, 1988; Coplen, 1994). For water
samples with NO3–N concentrations ≥0.06 mg L1, the 2σ
uncertainty of oxygen isotopic data is 1.0‰, and for samples
with NO3–N concentrations <0.06 mg L1, it is 2‰.
Water samples were analyzed for various groups of OWCs
(Supplemental Table S1) to assess the movement of these com-
pounds into groundwater beneath septic tank drainfi elds.  e
OWCs included fragrances and fl avorants, fl ame retardants,
antioxidants, fuel-related compounds, detergent metabolites,
plasticizers, disinfectants, solvents and preservatives, poly-
nuclear aromatic hydrocarbons, pesticides, plant and animal
steroids, and caff eine. Samples were collected using protocols
to avoid fi eld contamination of the sample (U.S. Geological
Survey, 1999) because several compounds are present in com-
monly used products, including soaps, fragrances, insect repel-
lants, and beverages. Water samples were fi ltered in the fi eld
through 0.7-μm nominal pore size glass fi ber lters, chilled
and maintained at 4°C, and shipped overnight to the labo-
ratory. Samples were analyzed at the USGS NWQL for 63
compounds using capillary-column gas chromatography/mass
spectrometry methods (Zaugg et al., 2002; Burkhardt et al.,
2006; Zaugg and Leiker, 2006). Quality assurance information
for OWCs is included in Supplemental Table S1.
Water samples were collected and analyzed for 16 pharma-
ceutical compounds at the USGS NWQL (Cahill et al., 2004).
Supplemental Table S2 includes a list of these compounds
(selected prescription and over-the-counter drugs and certain
degradates/metabolites) along with laboratory quality assur-
ance information for these compounds.
Subsurface Material Samples
Core samples of unsaturated zone material were collected
beneath the septic tank drainfi elds (in between drain lines)
using direct push techniques.  is technique resulted in
minimal disturbance to subsurface material due to the small-
diameter holes created by the rig because no cuttings are
produced, and greater sample recovery can be achieved com-
pared with other methods. Subsamples of oven-dried (105°C)
core material from discrete depth intervals were analyzed for
exchangeable NH4+ in a solution extracted with 2 mol L1 KCl.
Ammonium was analyzed colorimetrically using the salicylate
method (Nelson, 1983; McMahon et al., 2006). Nitrate and
other anions (nitrite, chloride, bromide, and sulfate) were ana-
lyzed in a solution extracted using deionized water (McMahon
et al., 2006). Replicate samples were run after every fi fth envi-
ronmental sample to assess the eff ects of fi eld and laboratory
procedures on measurement variability.
Microbiology Analyses
Microbiological samples were collected in sterile 1.0-L bottles
and stored on ice until processed. Samples were analyzed for
the presence of fecal indicator bacteria in duplicate in 5.0 μL
(septic effl uent) to 200-mL volumes. Samples were processed
using membrane fi ltration per Standard Method (SM) 9222D
(American Public Health Association, 1998). Indicator sample
results are expressed in the text and tables as the number of
colony-forming units (cfu) per 100.0 mL of water (average of
the duplicates).
Samples also were screened for enteroviruses (polio, cox-
sackie A, coxsackie B, echoviruses, etc.) using 50.0- to 100.0-
mL volumes (volume dependent on particle load) following
a modifi ed reverse transcriptase–polymerase chain reaction
(RT-PCR) and dot blot protocol (Lipp and Griffi n, 2004).  e
modifi cation was direct concentration of water samples with
Centriprep YM-50 centrifugal fi lter devices (i.e., no acidifi ca-
tion or membrane fi ltration–based preconcentration) follow-
ing the Centriprep protocol in 15-mL increments to a fi nal
volume of ~500.0 µL. RNA was purifi ed from one half of the
Centriprep concentrate using Qiagens RNeasy Kit to a fi nal
eluent volume of 30 μL (7 μL was used for RT-PCR).  e
combined enterovirus RT-PCR–dot blot results are expressed
as presence (+) or absence (). Negative and positive controls
were used with all prokaryote and virus assays.
For screening of soil samples for microbial indicators and
viruses, 10 mL of sterile water was added to ~5.0 g of soil and
shaken overnight on a table top rotator at room temperature.
On the following day, samples were centrifuged at 1500 rpm
for 2 min, and 4.5-mL aliquots (×2) were used for membrane
ltration following SM 9222D. Indicator sample results are
expressed as the number of cfu per gram of soil (average of
the duplicates). RNA was extracted from ~2.0 g of soil using
the RNA PowerSopil Total RNA Isolation Kit (MO BIO,
Carlsbad, CA). Kit protocol was followed, and 7 μL of eluent
was used for RT-PCR.
Results and Discussion
Distribution of Nitrogen Species in Septic Tank E uent,
Soil Pore Water, and Groundwater
Concentrations of N species varied considerably among
and within sites during the three sampling periods (Fig. 2).
Kjeldahl nitrogen (KN) comprised essentially 100% of the
total N in the STE at each site. About 80 to 94% of the total
N consists of NH4+, which is slightly higher than the range of
75 to 85% in previous studies (McCray et al., 2005). Kjeldahl
nitrogen concentrations in the STE were lowest at the HK site
(17–39 mg L1) and ranged from 42 to 65 mg L1 and from
53 to 58 mg L1 at the YG and LT sites, respectively. High and
low KN concentrations in STE were not consistent temporally
among the three sites (Fig. 2). Nitrate was the dominant form
of N in water samples from the shallow lysimeters, deep lysim-
eters, and the drainfi eld well at each site, with median NO3–N
concentrations comprising 94, 95, and 99% of the total N,
respectively.  is is consistent with other studies that show
nearly complete and rapid nitrifi cation in septic system effl u-
ent within short distances from the distribution lines (Wilhelm
Katz et al.: Contaminants beneath Septic Tank Drain elds 1185
et al., 1994; Ptacek, 1998; Barringer et al., 2006). Ammonium
and organic N comprised almost 44% of the total N in water
from the HK drainfi eld well in December 2007, indicating
that the drainfi eld was not functioning properly.  e hom-
eowner replaced the clogged drainfi eld with a new drainfi eld
in January 2008, and a new set of lysimeters and a drainfi eld
well were installed.
Median NO3–N concentrations were higher in the July
2008 water samples from the shallow and deep lysimeters and
from drainfi eld wells at the three sites than in samples from
December 2007 and March 2008. Median chloride concentra-
tions also increased in the July 2008 samples from the deep
lysimeters and drainfi eld wells from the three sites but decreased
in the shallow lysimeters. Because the median chloride and KN
concentrations in STE were lower in the July 2008 samples
than in the December 2007 and March 2008 samples, it is
likely that the chloride and NO3–N that were stored beneath
the drainfi eld lines moved deeper due to increased rainfall
and recharge during late spring and early summer.  e higher
median NO3–N concentration in groundwater samples from
July 2008 also could be biased high due to fertilizer application
at the HK site (see below).
Isotope Ratios of Ammonium, Nitrate, and Water
Values of δ15N-NH4+ in STE samples showed little varia-
tion among sites and temporally, ranging from 4.4 to 4.9‰,
from 4.5 to 5.3‰, and from 5.0 to 5.5‰ at the HK, YG,
and LT sites, respectively.  ese values are similar to those
(4.2–5.1‰) for STE from a study of septic tank systems
and packed bed sand fi lters (Hinkle et al., 2008). Values of
δ15N-NO3
in water samples were more variable and ranged
from <2‰ (YG background well) to 13.4‰ (old drainfi eld
well at HK site) (Supplemental Table S3).  e low δ15N-
NO3 values (<6.0‰) correspond to values observed else-
where, where NO3 in groundwater is recharged beneath land
receiving inorganic fertilizers (Kendall and Aravena, 2000).
Previous studies have found that NO3–N concentrations and
δ15N-NO3 values were highly variable in groundwater from
sites throughout the WKP and could be attributed to sev-
eral sources, including inorganic fertilizer applied to lawns,
gardens, cropland, and turfgrass, and organic wastes from
domestic wastewater and animal wastes (Katz et al., 2004).
e HK homeowner indicated that fertilizer was applied to
new grass planted on top of the new drainfi eld before the col-
lection of the July 2008 samples. Fertilizer applied upgradient
from the YG site may have resulted in the low δ15N-NO3 sig-
nature in the background well, as the homeowner indicated
that no fertilizer had been applied recently.
Additional information about processes aff ecting the trans-
formation of N species can be gleaned from N and oxygen
isotopes of NO3. Values of δ15N-NO3 and δ18O-NO3 increase
systematically in water samples from shallow and deep lysim-
eter samples at the LT and YG sites and plot along trend lines
with a slope of 0.5, which is consistent with denitrifi cation
(Fig. 3).  is process is related to the ratio of enrichment of
oxygen to N in NO3, which has been shown to be close to 1:2
(Aravena and Robertson, 1998; Kendall and Aravena, 2000).
Additional evidence for this ratio of enrichment of oxygen to
N in NO3 is based on a comparison of plotted values of δ18O-
H2O and δ18O-NO3 (Supplemental Fig. S2 and Table S3).  e
higher δ15N-NO3 and δ18O-NO3 values in the LT deep lysim-
eter site could be related to higher clay content in the sub-
surface, which could create conditions (perched-water; slower
water movement, low-dissolved-oxygen microsites) more
favorable to denitrifi cation. Problems with ponding and poor
drainage over time at the old HK drainfi eld likely resulted in an
enriched δ15N-NO3 and δ18O-NO3 signature in the December
2007 sample from the drainfi eld well; however, similarly high
isotopic values were not seen in the shallow lysimeter at this
time. Mixed redox conditions at the water table at the old HK
drainfi eld site likely were present in December 2007, as indi-
cated by elevated NH4+, NO3, and dissolved oxygen concentra-
tions (Table 1).  e July samples at the new HK drainfi eld site
clearly show the infl uence from the application of synthetic
fertilizer, with low δ15N-NO3 values <6‰. Elevated δ15N-NO3
(8–10‰) and δ18O-NO3 (5–6.5‰) values for water from the
HK background well may indicate the infl uence from upgradi-
ent septic tanks in the area.
Fig. 2. Plots showing concentrations of nitrogen species in water
samples from the background well, septic tank, lysimeters, and
drain eld well at each site. Nitrogen species include nitrate (NO3–N),
organic nitrogen (ORG-N), and ammonium (NH4–N).
1186 Journal of Environmental Quality • Volume 39 • July–August 2010
Depth Pro les of Nitrogen Species and Chloride in
Unsaturated Zone Material
Analyses of N species and chloride in soil core material from
various depth intervals provide snapshots of N and Cl fl uxes
to the water table before and after the water samples were col-
lected. Ammonium N concentrations (expressed as mg L1 by
accounting for moisture content of the oven-dried samples)
were substantially higher in core samples from depths <5 cm
at the YG site compared with the two other sites (Fig. 4). At
depths >50 cm, NO3–N generally was the dominant N spe-
cies at the three sites, which corresponds to the oxic conditions
measured in soil pore water from lysimeters and in groundwa-
ter (top of UFA). Nitrogen concentrations were substantially
higher in the July 2008 samples compared with November
2007 samples at all three sites. At the LT site, the highest N
(NO3) concentration (about 110 mg L1) at the 180-cm depth
corresponded to the highest chloride concentration.  is depth
interval also contains a higher amount of clay (~5%) compared
with other depth intervals analyzed for particle size distribu-
tion. Depth profi les of chloride concentrations were similar at
the YG site between November 2007 and July 2008. In con-
trast, higher chloride concentrations were measured in core
samples beneath the old drainfi eld compared with those col-
lected beneath the new drainfi eld at the HK site (Fig. 4). At all
three sites, N concentrations in the unsaturated zone generally
were higher than concentrations measured in STE but were
similar to N concentrations measured in water samples from
shallow lysimeters at the YG and HK sites. Concentrations at
the LT site were considerably higher in extracts of unsaturated
zone material than in water from shallow lysimeters. High vari-
ability of N concentrations in soil and unsaturated zone pore
water among sites has been reported in other studies (Rosen et
al., 2006; Hinkle et al., 2008).  e high
variability of N concentrations between
the three sites likely is due to a combi-
nation of factors, such as N loading, soil
characteristics and lithology, and dilu-
tion from household activities such as
washing clothes and bathing.
Large diff erences in concentrations
of N species and chloride with depth
between the November 2007 and the
July 2008 samples appear to be related to
diff erences in rainfall and recharge con-
ditions before sampling. In December
2007, the previous 30-, 60-, and 90-d
rainfall amounts (7, 13, and 21 cm)
were substantially lower than before the
March (22, 29, and 36 cm) and July
(15, 25, and 27 cm) sampling events.
Increased recharge in the spring and
early summer of 2008 likely resulted in
downward movement of N (NH4+ and
NO3). With increased rainfall/recharge,
one would expect that concentrations
in the unsaturated zone would be lower
(i.e., more dilution) than during periods
of lower rainfall. However, it is likely that
NH4+ and organic N in water retained beneath the drainfi elds
may have been converted to NO3 as it moved toward the water
table. At the LT site, the perched-water conditions created by
the presence of higher amounts of clay at lower depths may
account for the increase in chloride and N at the 170 cm depth
and possibly allowing more water–sediment contact time for
NH4+ desorption reactions.
Attenuation of Nitrogen beneath Septic Tank Drain elds
Median and mean N (dissolved) concentrations in STE
decrease with depth as the STE percolates downward from
the drainfi eld to the water table, as indicated by progressively
lower concentrations in shallow lysimeters, deep lysimeters,
and drainfi eld wells. Mean N concentrations (± 1σ) for all
samples of STE, shallow lysimeters, deep lysimeters, and drain-
eld wells were 45 ± 14, 30 ± 21, 22.6 ± 16.5, and 21 ± 6.8
mg L1, respectively (Fig. 5). Also, N concentrations decrease
with depth to about 25 to 40 mg L1 in pore water samples
from the unsaturated zone. Dilution accounts for some of
the decrease in N concentrations (15–25%) based on median
and mean chloride concentrations, respectively, in STE, water
from shallow and deep lysimeters, and drainfi eld wells (Fig.
5) (assuming that chloride is conservative and that there are
no naturally occurring sources of chloride in the unsaturated
zone). Nitrogen loss is related to denitrifi cation, sorption of
NH4+ (includes adsorption and ion exchange) on soil particles,
and ammonia volatilization.
Denitri cation and Ammonium Volatilization
Decreased N concentrations in soil water and groundwater may
result from denitrifi cation or ammonia volatilization. At the
YG and LT sites, there were average increases of 3.3 and 3.6‰
between δ15N-NH4 in water samples from STE and δ15N-NO3
Fig. 3. Plot of nitrate isotope data for water samples from the background well, septic tank, lysim-
eters, and drain eld well at each site. Values of δ15N and δ18O are given with respect to air N2 and
Vienna standard mean ocean water, respectively, in parts per thousand.
Katz et al.: Contaminants beneath Septic Tank Drain elds 1187
Table 1. Physical properties and concentrations† of major dissolved chemical constituents in water samples from septic tanks, background (back.) wells, shallow (shal.) and deep lysimeters (lys.), and drain-
eld (drain.) wells at the three studied sites.
Site Date Smpl.
t‡ Temp. Sp.
Cond.,
at 25°C‡
Diss.
O2pH Amm.,
as
N‡
Amm.
plus ON,
as N‡
Nitrate
plus
nitrite, N
Ortho-
phosphate,
as P DOC‡ Ca Mg Na K ClSulfate FSiO2BAlk., as
CaCO3Bromide Cl/Br
mass
ratio
°C μS cm1μg L1
YG
Septic
tank 18 Dec
2007 1430 19.0 725 2.2 6.5 39‡ 47 <0.04 5.8 33 50 1.8 27 9.6 27 E0.12§ 0.14 14 22 293 0.04 680
Septic
tank 13 Mar.
2008 1200 23.6 887 1.0 6.6 56 65 <0.04 7.8 46 46 2.2 42 15 41 <0.18 0.12 9.6 253 346 E0.07 591
Septic
tank 16 July
2008 1140 20.2 738 6.6 38 42 <0.04 6.2 26 48 1.7 34 9.9 35 4.5 E.09 9.5 191 285 0.05 694
Septic
tank
(DUP¶)
16 July
2008 1145 20.2 738 6.6 38 43 <0.04 6.2 26 50 1.7 35 10 35 4.4 0.1 11 207 287 0.05 692
Back.
well 18 Dec.
2007 1200 18.4 20 6.2 8.3 <0.10 1.2 <0.04 0.13 7.1 0.15 0.42 1.8 0.17 2.6 1.1 <0.12 8.5 7.9 <5 <0.02 ND
Back.
well 13 Mar.
2008 940 17.8 182 5.3 5.3 <0.02 E0.08 0.2 0.12 1.1 33 0.96 1.8 0.40 2.6 1.3 <0.12 5.8 8.7 84 E0.01 263
Back.
well 16 July
2008 930 21.7 225 6.5 <0.02 E0.14 0.39 0.13 0.8 39 1.1 2.8 0.37 5.6 2.5 <0.12 5.3 7.9 103 0.03 187
Drain.
well 18 Dec.
2007 1350 21.0 288 5.6 5.7 0.08 0.29 18.7 0.02 2.2 3.7 7.9 29 11 27 7.6 0.14 6.3 200 <5 0.04 665
Drain.
well 13 Mar.
2008 1030 20.4 324 4.8 5.0 0.07 0.14 20.5 0.02 1.3 5.4 8.3 29 11 32 7.7 0.17 5.7 164 <5 0.04 805
Drain.
well 16 July
2008 1000 24.0 294 5.5 0.13 0.21 16.2 0.02 0.7 4.3 6.8 27 8.9 35 7.1 0.14 5.2 142 <5 0.04 865
Shal.
lys. 18 Dec.
2007 1340 17.0 438 5.9 5.7 0.02 0.49 26.2 <0.006 15 42 3.4 26 4.8 31 17 NS 13 31 27 0.03 1043
Shal.
lys. 13 Mar.
2008 1130 18.6 439 3.3 5.7 0.11 0.77 18.6 0.02 14 35 1.7 45 5.0 39 22 <0.12 11 186 44 0.04 970
Shal.
lys. 17 July
2008 1110 29.6 1450 6.2 0.13 4.9 51.9 0.04 24 105 1.6 195 0.96 196 58 <0.12 14 470 153 0.21 933
Deep
lys. 18 Dec.
2007 1400 17.2 421 5.3 5.7 0.04 0.58 22.7 <0.006 NS NS NS NS NS 31 20 NS NS NS 34 E0.02 1560
Deep
lys. 13 Mar.
2008 1100 17.2 332 4.0 5.5 E0.02 0.45 4.97 E0.003 14 29 1.4 35 1.1 27 19 <0.12 8.6 100 79 E0.02 1370
Deep
lys. 16 July
2008 1050 28.4 467 6.0 0.04 0.61 14.4 0.01 16 35 1.1 53 2.3 40 28 <0.12 12 234 78 0.07 571
1188 Journal of Environmental Quality • Volume 39 • July–August 2010
Site Date Smpl.
t‡ Temp. Sp.
Cond.,
at 25°C‡
Diss.
O2pH Amm.,
as
N‡
Amm.
plus ON,
as N‡
Nitrate
plus
nitrite, N
Ortho-
phosphate,
as P DOC‡ Ca Mg Na K ClSulfate FSiO2BAlk., as
CaCO3Bromide Cl/Br
mass
ratio
HK
Septic
tank 19 Dec.
2007 1130 19.0 1050 1.5 6.8 26 30 <0.04 5.4 43 63 15 86 11 136 24 0.21 15 61 257 E0.25 544
Septic
tank 12 Mar.
2008 1300 22.2 566 1.7 7.0 14 17 0.2 4.0 22 69 4.6 17 4.3 20 7.1 E0.09 7.6 99 242 0.02 1010
Septic
tank 15 July
2008 1200 29.8 725 6.6 35 39 <0.04 7.2 56 56 6.3 26 10.8 30 4.2 0.15 10 115 290 0.05 598
Back.
well 19 Dec.
2007 900 21.0 435 1.6 6.2 0.06 0.15 0.37 E0.006 2.1 91 2.3 2.1 0.20 3.1 4.4 0.13 7.6 11 222 <0.04 77
Back.
well 12 Mar.
2008 1000 18.9 397 3.5 6.3 <0.02 E0.08 0.53 E0.004 0.9 81 1.9 2.1 0.12 4.3 4.2 <0.12 6.0 8.6 198 E0.02 213
Back.
well 15 July
2008 920 25.0 400 6.7 <0.02 <0.14 0.39 0.01 1.2 77 2.1 2.1 0.23 3.4 4.5 E0.07 7.2 12 201 E0.01 339
Drain.
well,
old
19 Dec.
2007 945 21.0 894 7.3 6.3 6.9 7.6 9.9 0.02 3.1 139 5.7 28 5.1 43 18 E0.09 11 154 338 0.1 427
Drain.
well,
new
12 Mar.
2008 1110 19.4 560 2.1 6.6 1.00 1.5 8.2 0.11 2.7 93 4.2 11 2.6 18 12 <0.12 6.6 53 211 0.04 453
Drain.
well,
new§
12 Mar.
2008 1115 19.4 560 2.1 6.6 0.81 1.2 8.0 0.06 2.6 93 4.1 11 2.5 18 12 <0.12 6.7 54 211 0.03 603
Drain.
well,
new
15 July
2008 1020 25.4 763 6.5 <0.02 0.26 30 1.0 2.1 105 6.3 29 5.7 46 19 E0.06 11 51 180 0.11 420
Shal.
lys. 19 Dec.
2007 1100 15.0 1270 3.0 6.7 0.08 0.81 57 0.87 13 171 20 68 11 103 104 E0.11 13 102 122 0.21 490
Shal.
lys. 12 Mar.
2008 1230 18.0 673 5.0 6.9 0.62 1.7 23 0.22 11 98 8.4 22 11 27 21 <0.12 24 130 204 0.05 534
Shal.
lys. 15 July
2008 1120 27.0 1080 6.4 0.06 2 50 0.64 30 156 17 9.7 20 9.1 34 E0.07 13 65 288 <0.02 ND
Deep
lys. 12 Mar.
2008 1140 18.5 680 4.2 6.9 0.32 1.2 21 0.16 16 93 8.5 19 7.6 27 23 <0.12 17 105 181 0.06 447
Deep
lys. 15 July
2008 1050 26.6 783 6.4 <0.02 1.7 42 0.24 40 118 13 10 14 14 27 <0.12 8.7 52 202 <0.02 ND
LT
Septic
tank 19 Dec.
2007 1730 16.0 1030 1.6 7.1 52 58 <0.04 13 50 46 12 57 21 41 2.4 0.24 25 71 391 E0.05 812
Septic
tank 11 Mar.
2008 1300 18.4 923 2.8 7.1 48 53 <0.04 11 49 50 9.7 41 20 35 2.0 0.20 23 57 373 E0.03 1150
Septic
tank 17 July
2008 1200 25.7 1100 6.9 51 54 E0.02 6.3 31 49 11 85 18 31 9.2 0.18 14 83 491 E0.04 773
Back.
well 11 Mar.
2008 1030 18.8 195 4.9 6.8 <0.02 E0.07 0.11 0.01 1.1 34 2.4 1.5 0.22 2.7 3.5 <0.12 5.6 9.0 92 E0.01 270
Back.
well 17 July
2008 910 21.3 193 7.0 <0.02 E0.13 0.08 0.01 1.4 34 2.6 2.0 0.23 2.7 3.4 E0.06 7.3 9.1 93 E0.01 271
Table 1. Continued.
Katz et al.: Contaminants beneath Septic Tank Drain elds 1189
in lysimeters and the drainfi eld well samples.  ese
increases in N-isotopic composition, along with the
decrease in N at these sites, are consistent with iso-
topic fractionation resulting from denitrifi cation or
volatilization of ammonia. At the HK site, the lower
average increases of 1.7‰ (0.4‰ median) between
δ15N-NH4 in STE samples and δ15N-NO3 in lysim-
eters and the drainfi eld samples likely were aff ected
by fertilizer applications in April and May to turf-
grass above the drainfi eld. However, samples showed
an increase of 8.5‰ between δ15N-NH4 and δ15N-
NO3 in water samples collected from the old drain-
eld well in December 2007 and an increase of 3.7
to 4.0‰ in the new drainfi eld well in March 2008
before fertilizer applications.
Denitrifi cation likely occurs directly beneath the
drainfi eld lines in all three systems. Other studies
report about 10 to 25% N loss due to denitrifi cation
and ammonia volatilization before the leachate leaves
the bottom of the drainfi eld, depending on the initial
N concentration in STE (Bauman and Schafer, 1985;
Hantzsche and Finnemore, 1992). Septic tank effl u-
ent from the three sites has elevated concentrations
of DOC, with a median concentration of about 42
mg L1 (Fig. 5; Table 1). Median concentrations in
soil water from lysimeters decrease to about 15 mg
L1, indicating that DOC was consumed (or sorbed)
during infi ltration. However, water samples generally
are oxic, but denitrifi cation can occur in anaerobic
microsites beneath septic tank systems in otherwise
oxic systems (Koba et al., 1997). Concentrations
of DOC in the drainfi eld well decrease to slightly
greater than those in the background wells. Evidence
for some denitrifi cation can be seen from enriched
values of δ15N and δ18O of NO3 (Fig. 3).  ere is a
shift toward more enriched values for water samples
from shallow and deep lysimeters at the LT and YG
sites and for water from the old drainfi eld well at the
HK site.  ese slightly enriched values of δ15N and
δ18O of NO3 plot along denitrifi cation trend lines
with a slope of 0.5 (Kendall and Aravena, 2000).
Sorption of Ammonium
Ammonium strongly sorbs to mineral surfaces, partic-
ularly clay minerals (Drever, 1988); however, Hinkle
et al. (2008) found essentially complete sorption of
NH4+ on coarse sand with only 0.4% silt and clay in
column experiments. Almost 100% of sorbed NH4+
was recovered using 2 mol L1 KCl in column desorp-
tion experiments (Hinkle et al., 2008). Even though
soil cores were not collected directly beneath drainfi eld
lines (to prevent damage to the existing system), the
cores collected and analyzed for N species between the
drainfi eld lines provide information about NH4+ sorp-
tion by soils. Based on 2 mol L1 KCl extracts of core
samples, NH4+ concentrations ranged from 0.2 to 8.2
μg gsoil
1 (~5–190 μeq NH4+ gsoil
1) in the November
2007 samples (Fig. 6) below the bottom of the drain-
eld (depth of about 80–100 cm). Higher amounts of
Site Date Smpl.
t‡ Temp. Sp.
Cond.,
at 25°C‡
Diss.
O2pH Amm.,
as
N‡
Amm.
plus ON,
as N‡
Nitrate
plus
nitrite, N
Ortho-
phosphate,
as P DOC‡ Ca Mg Na K ClSulfate FSiO2BAlk., as
CaCO3Bromide Cl/Br
mass
ratio
Drain.
well 19 Dec.
2007 1430 20.0 733 6.5 6.9 0.14 0.29 23 0.18 3.1 78 19 55 4.2 27 21 0.22 42 304 252 0.08 335
Drain.
well 11 Mar.
2008 1115 19.3 794 5.8 6.9 0.04 0.19 24 0.17 2.4 85 19 46 4.3 29 21 0.13 30 285 254 0.05 584
Drain.
well 17 July
2008 1010 21.3 762 6.8 0.03 0.19 26 0.31 1.9 80 20 47 4.4 28 20 0.14 31 210 237 0.05 550
Shal. lys.19 Dec.
2007 1600 12.0 716 5.5 7.3 0.04 1.5 24 3.3 NS NS NS NS NS 30 40 0.20 NS NS 177 <0.02 ND
Shal. lys.11 Mar.
2008 1200 16.9 312 5.6 7.2 E0.02 0.5 0.47 1.1 13 41 12 3.7 1.8 1.6 10 0.14 12 92 146 <0.02 ND
Shal. lys.17 July
2008 1130 27.1 942 6.9 <0.02 1.6 6.5 3.1 40 91 25 60 1.6 18 43 0.30 33 182 359 E0.01 1830
Deep
lys. 19 Dec.
2007 1620 16.0 724 4.5 7.2 E0.01 1.6 22 3.2 13 69 20 48 11 27 33 0.24 21 59 204 E0.02 1350
Deep
lys. 11 Mar.
2008 1230 16.7 365 5.5 7.3 0.02 0.83 1.9 3.1 9.8 45 13 12 3.4 2.6 8.9 0.19 18 39 172 <0.02 ND
Deep
lys. 17 July
2008 1100 27.4 1120 6.6 0.03 1.4 33 5.6 20 90 25 79 17 54 53 0.21 25 70 267 0.05 1086
† Concentrations are in mg L1 unless otherwise noted.
‡ Smpl. t., sample time; Sp. cond., speci c conductance; Diss., dissolved; Amm., ammonia; ON, organic nitrogen; DOC, dissolved organic carbon; Alk., alkalinity.
§ E, estimated value between method reporting level and detection limit; NS, no sample collected due to insu cient volume of water; ND, not determined as bromide concentration is below method reporting level.
¶ Duplicate sample.
Table 1. Continued.
1190 Journal of Environmental Quality • Volume 39 • July–August 2010
NH4+ were found in KCl extracts of soil material in the July 2008
samples (2–10 µg gsoil
1).  ese values fall within the range of
total cation exchange capacities for glacial outwash sands (5–20
μeq g1; Bohlke et al., 2006) to various types of clays (28–8800
μeq g1) (Busenberg and Clemency, 1973).  us, there would be
suffi cient opportunity for additional NH4+ sorption, particularly
with the increasing relative amounts of clay minerals with depth.
Also, NH4+ sorption by soils was likely greater beneath the new
HK drainfi eld because this material was not at steady state with
respect to NH4+ sorption.  e source of the higher amounts of
NH4+ in samples from the shallow subsurface (<100 cm) at the
YG site is unknown but may result from ammonia volatilization,
nitrifi cation of organic N, or fertilizer applications.
Nitrate-Nitrogen Loading to Groundwater from Septic Tanks
Based on average daily water use at each household (metered)
and NO3–N concentrations at the water table (adjusted for
dilution), N loading (kg N yr1) to groundwater ranged from
4.3 to 5.4 at the YG site, 3.9 to 4.3 at the LT site, and 6.6 to
12 at the HK site.  e higher loading at the HK site compared
with the two other sites results from a higher average daily
water use (~1,630 L d1) and the larger range in NO3–N con-
centrations over the three sampling events.  e July 2008 NO3
data from the HK drainfi eld well were not used to calculate
septic tank loads because fertilizer applications in May resulted
in a large increase in NO3 in water
from the drainfi eld well (39 mg
L1) (and the lysimeter samples).
Annual per capita N loading rates
to groundwater at the LT and YG
sites are lower than other estimates
of 4.5 kg N yr1 (Cox and Kahle,
1999), which would yield 9 and
13.5 kg N yr1 for households
with two and three people, respec-
tively. Based on septic tank stud-
ies in central Florida (Sherman
and Anderson, 1991) and Oregon
(Morgan et al., 2007), about 9.5
to 9.7 kg N yr1 per home are
released to the drainfi eld, and it is
estimated that an additional 25%
reduction in N occurs beneath the
drainfi eld (Anderson et al., 1994;
Morgan et al., 2007).  e annual
N loading for the HK site (6.6–12
kg N yr1) also is lower than the
estimated loading to groundwater
of 18 kg N yr1 based on per capita
N loading (Anderson et al., 1994;
Cox and Kahle, 1999).
Although only three septic tank
systems were studied, measured N
loading rates to groundwater are
consistent with results from other
studies. Furthermore, the number
of people living in each house-
hold, per capita water use, loca-
tion, and age of the studied septic
tank systems are representative of the area. Given that there are
many uncertainties in extrapolating from the studied sites to
the approximately 20,000 septic tanks in the WKP, it is esti-
mated that annual NO3–N loading to groundwater from septic
tanks in the WKP would range from 78,000 to 240,000 kg.
is estimated range of N loading to groundwater can be con-
sidered conservative because NO3–N loads from septic tanks in
this study are lower than those from other studies (due to lower
per capita water use). Even with the conservative estimate of N
loading from septic tanks in the WKP, this source represents a
substantial contribution compared with the estimated N load-
ing to the land surface (about 2 million kg N yr1) from other
sources (atmospheric deposition, the land application of treated
wastewater effl uent, commercial fertilizers, livestock, and sink-
ing streams) (Chelette et al., 2002). After N losses are considered
for these other sources (e.g., uptake by vegetation, soils, denitri-
cation, and volatilization), the relative contribution from septic
tanks would be considerably higher and would represent a sig-
nifi cant source of NO3 to the groundwater system.
Attenuation of Organic Wastewater
and Pharmaceutical Compounds
Given the relatively low analytical spike recoveries for several
OWCs (<60%) and for surrogate compounds (Supplemental
Fig. 4. Depth pro les of ammonium-N, nitrogen (sum of nitrate-N, nitrite -N, and ammonium-N), and chloride in
subsurface samples from various depth intervals collected from each site in November 2007 and July 2008.
Katz et al.: Contaminants beneath Septic Tank Drain elds 1191
Table S4), detections for individual compounds only were
considered when the measured concentrations were at least
three times the method reporting level. No OWCs were
detected in blank water samples submitted to the NWQL
during the study or in samples from background wells. In STE
samples, 25 OWCs were detected out of the 63 compounds
analyzed (Supplemental Table S4).  e concentrations for
most of these compounds were below 10 μg L1; however,
50% or more STE samples had fi ve compounds with concen-
trations >10 μg L1: caff eine, 1.7-dimethylxanthine (caff eine
degradate), menthol, p-cresol, and phenol. Concentrations of
detergent metabolites and disinfectants (e.g., triclosan) were
considerably lower than those found in other studies (e.g.,
Hinkle et al., 2005; Conn et al., 2006) and may result from
uptake by solids and removal by sedimentation in the septic
tank (Conn et al., 2006).
Substantial attenuation or removal of OWCs was observed
in the subsurface beneath the drainfi elds. Phenol and cotinine
were the only compounds detected in samples from lysimeters
(Supplemental Table S4). Five OWCs were detected in the
drainfi eld well at the HK site (caff eine, 1.7-dimethylxanthine,
phenol, galaxolide, and tris(dichloroisopropyl)phosphate).
Tris(dichloroisopropyl)phosphate was detected twice in sam-
ples from the YG site drainfi eld well. Concentrations of these
OWCs in groundwater samples were <1.0 µg L1. Multiple
detections of tris(dichloroisopropyl)phosphate in groundwa-
ter samples may indicate that this compound would be useful
as an indicator of STE in shallow groundwater.  e higher
number of OWCs detected in drainfi eld wells compared with
those for shallow and deep lysimeters may be indicative of the
ineffi ciency of lysimeters as collectors of these compounds.
Leachate could be moving along preferential pathways that
Fig. 5. Boxplots showing concentration distribution of nitrogen, chloride, and
dissolved organic carbon in water samples from the background well, septic
tank, lysimeters, and drain eld well at each site.
Fig. 6. Depth pro le of ammonium in soil samples based on extracts
of 2 mol L1 KCl solution.
1192 Journal of Environmental Quality • Volume 39 • July–August 2010
would lower the eff ectiveness of using lysimeters to collect
OWCs in the unsaturated zone. Another possibility is that
some compounds may have been excluded (i.e., removed)
during passage through the ceramic cup.
Two pharmaceutical compounds (acetaminophen and sulfa-
methoxazole) were detected in water samples from septic tanks,
lysimeters, and drainfi eld wells. Median recoveries for these
two compounds in spike samples were 78 and 63%, respec-
tively (Supplemental Table S2). Acetaminophen (an over-the-
counter analgesic) was detected in eight of nine STE samples in
concentrations ranging from 0.18 to 78.2 μg L1 at the HK site
(Supplemental Table S5). Sulfamethoxazole (an antibiotic) was
detected in only one STE sample (HK site; 0.04 μg L1).  is
compound was detected in all three water samples from the LT
site drainfi eld well but not in any of the STE samples from this
site. Sulfamethoxazole also was detected once in concentra-
tions below the method reporting level in drainfi eld well water
samples from the YG site (0.02 μg L1) and the HK site (0.04
μg L1). Low-level concentrations of these compounds in water
samples from lysimeters and shallow groundwater (drainfi eld
wells) are consistent with the movement of sulfamethoxazole
from septic systems and its persistence in shallow groundwater
in other studies (Godfrey et al., 2007; Carrara et al., 2008).
Sulfamethoxazole, which has a low sorption affi nity in soils
(Cordy et al., 2004; Drillia et al., 2005), has been found in
shallow zones of the Upper Floridan aquifer in areas where
septic tanks likely have aff ected groundwater quality (Katz and
Griffi n, 2008; Katz et al., 2009).
Microbial Indicators and Enteric Viruses
Numbers of fecal coliforms were highly variable temporally in
STE samples from each site and among sites (Table 2). For
example, fecal coliforms in STE samples from the HK site
ranged from 120,000 cfu per 100 mL in July 2008 to 380,000
cfu per 100 mL in March 2008. Fecal coliforms in STE
from the LT site ranged from 1.4 million cfu per 100 mL in
December 2007 to 7.8 million cfu per 100 mL in July 2008
and from 150,000 in July 2008 to 2.7 million cfu 100 per mL
in March 2008 at the YG site. Rapid attenuation of fecal coli-
forms was observed at all sites; however, occasional detections
of fecal coliforms (e.g., high concentrations in the HK drain-
eld well in March 2008) appear to be related to high rainfall
amounts during the previous 30 to 60 d.
Enteroviruses were detected in STE samples from all three
sites in December 2007 and in March 2008 but not in any of
the July 2008 samples (Table 2). Enteroviruses were detected in
an unused domestic well (i.e., an irrigation well) at the HK site,
in the LT deep lysimeter in December 2007, in the LT back-
ground well and in both YG lysimeters in March 2008, and
in the LT shallow lysimeter in July 2008. Sporadic detections
and transport of enteroviruses in soil water and groundwater
samples beneath the three septic tank drainfi elds are consistent
with other studies (Darby and Leverenz, 2004; DeBorde et al.,
1998; Abbaszadegan et al., 1999).  e absence of enteroviruses
in the July 2008 STE samples may indicate seasonality for this
region because previous studies have shown that the entero-
virus season runs from June through October in temperate
Table 2. Microbiological data for samples of septic tank e uent, soil water from lysimeters, and water from drain eld and background wells at each
site.†
Site† Fecal coliforms Enteroviruses
Dec. 2007 Mar. 2008 July 2008 Dec. 2007 Mar. 2008 July 2008
——————— c f u 100 mL−1‡ ———————
HK site
Background well 0.0 16.8 0.5 –§
Septic tank e uent 220,000 380,000 120,000 + +
Drain eld well 0.0 tntc¶ 6.0
Lysimeter, shallow 0.0 0.0 0.0
Lysimeter, deep nl 0.0 0.0 nl
Domestic well used for irrigation 0.5 0.0 3.0 +
LT site
Background well 0.0 0.0 0.0 +
Septic tank e uent 1,450,000 4,660,000 7,800,000 + +
Drain eld well 0.0 0.0 0.0
Lysimeter, shallow 0.0 0.0 0.0 +
Lysimeter, deep 0.0 0.0 0.0 +
YG site
Background well 0.0 0.3 0.3
Septic tank e uent 210,000 2,720,000 150,000 + +
Drain eld well 5.7 0.0 0.0
Lysimeter, shallow 0.0 0.0 0.0 +
Lysimeter, deep 2.0 0.0 0.0 +
† Data are for culturable fecal coliform and reverse transcriptase–polymerase chain reaction dot blot detection of human enteroviruses.
‡ cfu, colony-forming units.
§ +, presence; , absence.
¶ nl, no lysimeter installed on this date; tntc, too numerous to count.
Katz et al.: Contaminants beneath Septic Tank Drain elds 1193
regions and may be continuous in tropical and semitropical
environments (Moore, 1982; Melnick, 1989).
Generally there was little relation between detections of
fecal coliforms and OWCs and pharmaceutical compounds. In
samples where enteroviruses were detected, there were occa-
sional detections of OWCs and pharmaceuticals; for example,
in soil water from shallow lysimeters, caff eine was detected at
the YG site in March 2008, and phenol was detected at the LT
site in July 2008. In soil water from deep lysimeters, cotinine
was detected at the YG site in March 2008, and phenol and
camphor were detected at the LT site in December 2007.  e
slightly higher number of detected compounds in the March
and July samples may in part be related to increased rainfall
and recharge that may enhance the transport of these com-
pounds in the subsurface.
Movement of Chloride and Bromide
Chloride/bromide (Cl/Br) ratios have been used to discriminate
between wastewater and other sources of chloride from anthro-
pogenic and naturally occurring contaminants in groundwater
(Vengosh and Pankratov, 1998; Panno et al., 2006; Brown et
al., 2009).  e Cl/Br ratios in samples of STE ranged from
540 to 1150, with a median Cl/Br ratio of 694 (n = 9).  is
median value is similar to the median Cl/Br ratio (769; n =
29) reported by Panno et al. (2006) for STE. Groundwater
samples from wells beneath the three drainfi elds had a slightly
smaller range of Cl/Br ratios (335–805) and a median of 518
(Supplemental Fig. S3). Based on the small number of samples
from this study, it appears that groundwater samples with a Cl/
Br range of 400 to 1100 may be useful as a screening tool for
evaluating septic tank infl uence on groundwater quality.
Conclusions
Septic tank effl uent represents a substantial source of N load-
ing to groundwater in the WKP, an area in northern Florida
that is highly vulnerable to contamination due a thin veneer
of sands and clays overlying a karstic aquifer. Given that mea-
sured N loading to groundwater from three studied septic tank
systems is consistent with results from other studies and that
the studied septic tank systems are representative of other sys-
tems in the area, it is estimated that N loading to groundwater
from the approximately 20,000 septic tanks in the WKP would
range from 78,000 to 240,000 kg N yr1.
e use of multiple chemical and isotopic indicators pro-
vided important information about N transport to ground-
water, which was aff ected by three main processes.  e 50%
apparent loss of N as drainfi eld STE moves to the water table
is related to (i) dilution, which accounted for only about 10 to
25% of the decrease in N concentrations from STE to ground-
water; (ii) denitrifi cation, indicated by a systematic increase in
δ15N and δ18O of NO3 in soil water from lysimeters; and (iii)
sorption of ammonia and retention on soil particles, indicated
by the release of NH4+ in 2 mol L1 KCl extracts of soil core
samples. Fertilizers applied to lawns and gardens also may rep-
resent an important (but unknown) source of N contamina-
tion in this area.
For the most part, microbial indicators, viruses, organic
wastewater compounds, and pharmaceutical compounds are
highly attenuated or removed as the STE percolates through
about 5 to 7 m of unsaturated zone material. Only sporadic
detections of fecal coliforms, enteroviruses, organic wastewater
compounds, and pharmaceutical compounds were found in
soil water from shallow and deep lysimeters and in ground-
water (drainfi eld wells). Higher detections of fecal indica-
tors, enteric viruses, OWCs, and pharmaceuticals were found
in groundwater at the site where the depth to the limestone
was the shallowest and average daily water use was highest.
Enterovirus data from this study are consistent with the move-
ment of viruses from septic tanks found in other studies, indi-
cating that there is a potential for contamination of drinking
water supplies from septic tanks and that there is a human
health risk in karstic aquifer systems.
Further work is needed to assess the impacts of septic sys-
tems on groundwater quality in higher density areas in the
karst plain, particularly where limestone is near the land sur-
face, to protect drinking water supplies from on-site waste dis-
posal practices. Current health department regulations require
a 23-m separation between a septic tank drainfi eld and a pota-
ble well, which may not be suffi cient in areas of higher septic
tank density or where the top of the limestone aquifer is near
the surface.
Acknowledgments
is study was funded by the Florida Department of Environmental
Protection and the USGS.  e authors thank D. Blum and H. Davis
for assistance with sampling; T. Coplen for stable isotope analyses; the
USGS National Water Quality Laboratory for analyses of wastewater
and pharmaceutical compounds; and S. Hinkle, S. Panno, L. Hundal,
and several anonymous reviewers, whose comments were very helpful
in revising this manuscript.
References
Abbaszadegan, M., P. Stewart, and M. LeChevallier. 1999. A strategy for de-
tection of viruses in groundwater by PCR. Appl. Environ. Microbiol.
65:444–449.
Allen, W.J. 1991. Soil survey of Wakulla County, Florida. USDA Soil Conser-
vation Service, Washington, DC.
American Public Health Association. 1988. Standard methods for the exami-
nation of water and wastewater, 20th ed. American Public Health As-
sociation, Washington, DC.
Anderson, D.L., R.L. Otis, J.I. McNeillie, and R.A. Apfel. 1994. In-situ lysim-
eter investigation of pollutant attenuation in the vadose zone of a fi ne
sand. p. 209–218. In E. Collins (ed.) On-site wastewater treatment. v. 7.
ASAE Pub. 18-94. ASAE, St. Joseph, MI.
Aravena, R., and W.D. Robertson. 1998. Use of multiple isotope tracers to
evaluate denitrifi cation in ground water: Study of nitrate from a large-
ux septic system plume. Ground Water 36:975–982.
Barnes, K.K., D.W. Kolpin, E.T. Furlong, S.D. Zaugg, M.T. Meyer, and L.B.
Barber. 2008. A national reconnaissance of pharmaceuticals and other
organic wastewater contaminants in the United States: I. Groundwater.
Sci. Total Environ. 402:192–200.
Barringer, J.L., Z. Szabo, D. Schneider, W.D. Atkinson, and R.A. Gallagh-
er. 2006. Mercury in ground water, septage, leach-fi eld effl uent, and
soils in residential areas, New Jersey coastal Plain. Sci. Total Environ.
361:144–162.
Bauman, B.J., and W.M. Schafer. 1985. Estimating ground water quality
impacts from on-site sewage treatment systems: On-site treatment. p.
285–294. In Proceedings of the Fourth National Symposium on Indi-
vidual and Small Community Sewage Systems, December 1984, New
Orleans, LA. Pub.07-85. American Society of Agricultural Engineers, St.
Joseph, MI.
Böhlke, J.K., R.L. Smith, D.N. Miller. 2006. Ammonium transport
and reaction in contaminated groundwater: Application of iso-
tope tracers and isotope fractionation studies. Water Resources Res.
doi:10.1029/2005WR004349.
1194 Journal of Environmental Quality • Volume 39 • July–August 2010
Brenna, J.T., T.N. Corso, H.J. Tobias, and R.J. Caimi. 1997. High-precision
continuous-fl ow isotope ratio mass spectrometry. Mass Spectrom. Rev.
16:227–258.
Brenton, R.W., and T.L. Arnett. 1993. Methods of analysis by the U.S. Geo-
logical Survey National Water Quality Laboratory: Determination of
dissolved organic carbon by UV-promoted persulfate oxidation and in-
frared spectrometry. U.S. Geological Survey Open-File Report 92-480,
USGS, Denver, CO.
Brown, C.L., J.J. Starn, K. Stollenwerk, R.A. Mondazzi, and T.J. Trombley.
2009. Aquifer chemistry and transport processes in the zone of contri-
bution to a public-supply well in Woodbury, Connecticut, 2002–06.
USGS Scientifi c Investigations Rep. 2009-5051. USGS, Reston, VA.
Burkhardt, M.R., S.D. Zaugg, S.G. Smith, and R.C. ReVello. 2006. Deter-
mination of wastewater compounds in sediment and soil by pressur-
ized solvent extraction, solid-phase extraction, and capillary-column gas
chromatography/mass spectrometry: U.S. Geological Survey techniques
and methods. USGS, Reston, VA.
Busenberg, E., and C.V. Clemency. 1973. Determination of the cation ex-
change capacity of clays and soils using an ammonia electrode. Clays
Clay Miner. 21:213–217.
Cahill, J.D., E.T. Furlong, M.R. Burkhardt, D. Kolpin, and L.G. Anderson.
2004. Determination of pharmaceutical compounds in surface- and
ground-water samples by solid-phase extraction and high-performance
liquid chromatography-electrospray ionization mass spectrometry. J.
Chromatogr. A 1041:171–180.
Carrara, C., C.J. Ptacek, W.D. Robertson, D.W. Blowes, M.C. Moncur, E.
Sverko, and S. Backus. 2008. Fate of pharmaceutical and trace organ-
ic compounds in three septic system plumes. Environ. Sci. Technol.
42:2805–2811.
Casciotti, K.L., D.M. Sigman, M. Hastings, J.K. Böhlke, and A. Hilkert.
2002. Measurement of the oxygen isotopic composition of nitrate in
seawater and freshwater using the denitrifi er method. Anal. Chem.
74:4905–4912.
Chelette, A.R., T.R. Pratt, and B.G. Katz. 2002. Nitrate loading as an indica-
tor of nonpoint source pollution in the lower St. Marks-Wakulla Rivers
watershed. Water Resources special report 02-1. Northwest Florida Wa-
ter Management District, Havana, FL.
Conn, K.E., L.B. Barber, G.K. Brown, and R.L. Siegrist. 2006. Occurrence
and fate of organic contaminants during onsite wastewater treatment.
Environ. Sci. Technol. 40:7358–7366.
Coplen, T.B. 1988. Normalization of oxygen and hydrogen isotope data.
Chem. Geol. 72:293–297.
Coplen, T.B. 1994. Reporting of stable hydrogen, carbon, and oxygen isotopic
abundances. Pure Appl. Chem. 66:273–276.
Coplen, T.B., J.K. Böhlke, and K. Casciotti. 2004. Using dual-bacterial deni-
trifi cation to improve delta-15N determinations of nitrates containing
mass-independent 17-O. Rapid Commun. Mass Spectrom. 18:245–250.
Cordy, G.E., N.L. Duran, H. Bouwer, R.C. Rice, E.T. Furlong, S.D. Zaugg,
M.T. Meyer, L.B. Barber, and D.W. Kolpin. 2004. Do pharmaceuti-
cals, pathogens, and other organic wastewater compounds persist when
wastewater is used for recharge? Ground Water Monit. Rev. 24:58–69.
Cox, S.E., and S.C. Kahle. 1999. Hydrogeology, ground-water quality, and
sources of nitrate in lowland glacial aquifer of Whatcom County, Wash-
ington, and British Columbia, Canada. USGS Water Resources Investi-
gation Report 98-4195, USGS, Tacoma, WA.
Darby, J.L., and H. Leverenz. 2004. Virus, phosphorus, and nitrogen removal
in onsite wastewater treatment processes. Technical completion report
project no. W-953. University of California Water Resources Center,
Riverside, CA.
Davis, J.H. 1996. Hydrogeologic investigation and simulation of ground-
water fl ow in the Upper Floridan aquifer of north-central Florida and
southwestern Georgia and delineation of contributing areas for selected
City of Tallahassee, Florida, water-supply wells. USGSWater-Resources
Investigations Rep. 95-4296. USGS, Reston, VA.
DeBorde, D.C., W.W. Woessner, B. Lauerman, and P.N. Ball. 1998. Virus oc-
currence and transport in a school septic system and unconfi ned aquifer.
Ground Water 36:825–834.
Drever, J.I. 1988.  e geochemistry of natural waters. 2nd ed. Prentice Hall,
Englewood Cliff s, NJ.
Drillia, P., K. Stamatelatou, and G. Lyberatos. 2005. Fate and mobility of
pharmaceuticals in solid matrices. Chemosphere 60:1034–1044.
Fishman, M.J. (ed.). 1993. Methods of analysis by the U.S. Geological Survey
National Water Quality Laboratory: Determination of inorganic and or-
ganic constituents in water and fl uvial sediments. USGS Open-File Rep.
93-125. USGS, Denver, CO.
Fishman, M.J., and L.C. Friedman. 1989. Methods for determination of in-
organic substances in water and fl uvial sediments: U.S. Geological Sur-
vey Techniques of Water-Resources Investigations. Book 5. Chapter A1.
USGS, Denver, CO.
Focazio, M.J., D.W. Kolpin, K.K. Barnes, E.T. Furlong, M.T. Meyer, S.D. Za-
ugg, L.B. Barber, and M.E.  urman. 2008. A national reconnaissance
for pharmaceuticals and other organic wastewater contaminants in the
United States: II. Untreated drinking water sources. Sci. Total Environ.
402:201–216.
Godfrey, E., W.W. Woessner, and M.J. Benotti. 2007. Pharmaceuticals in on-
site sewage effl uent and ground water, western Montana. Ground Water
45:263–271.
Hannon, J.E., and J.K. Böhlke. 2008. Determination of the δ (15N/14N) of
ammonium (NH4+) in water: RSIL lab code 2898. Chapter C15. In
K. Révész and T.B. Coplen (ed.) Methods of the Reston Stable Isotope
Laboratory. U.S. Geological Survey, Techniques and Methods 10-C15.
USGS, Reston, VA.
Hantzsche, N.N., and E.J. Finnemore. 1992. Predicting ground-water nitrate-
nitrogen impacts. Ground Water 30:490–499.
Harden, H.S., J.P. Chanton, J.B. Rose, D.E. John, and M.E. Hooks. 2003.
Comparison of sulfur hexafl uoride, fl uorescein and rhodamine dyes
and the bacteriophage PRD-1 in tracing subsurface fl ow. J. Hydrol.
277:100–115.
Harden, H.S., E. Roeder, M. Hooks, and J.P. Chanton. 2008. Evaluation of
onsite sewage treatment and disposal systems in shallow karst terrain.
Water Res. 42:2585–2597.
Hinkle, S.R., R.J. Weick, J.M. Hohnson, J.D. Cahill, S.G. Smith, and B.J.
Rich. 2005. Organic wastewater compounds, pharmaceuticals, and
coliphage in ground water receiving discharge from onsite wastewater
treatment systems near La Pine, Oregon: Occurrence and implications
for transport. USGS Scientifi c Investigations Rep. 2005-5055. USGS,
Reston, VA.
Hinkle, S.R., J.K. Bohlke, and L.H. Fisher. 2008. Mass balance and isotope ef-
fects during nitrogen transport through septic tank systems with packed-
bed (sand) fi lters. Sci. Total Environ. 407:324–332.
Katz, B.G., A.R. Chelette, and T.R. Pratt. 2004. Use of chemical and isotopic
tracers to assess nitrate contamination and ground-water age, Woodville
Karst Plain, USA. J. Hydrol. 289:36–61.
Katz, B.G., and D.W. Griffi n. 2008. Using chemical and microbiological in-
dicators to track the possible movement of contaminants from the land
application of reclaimed municipal wastewater in a karstic springs basin.
Environ. Geol. 55:801–821.
Katz, B.G., D.W. Griffi n, and H.D. Davis. 2009. Groundwater quality im-
pacts from the land application of treated municipal wastewater in a
large karstic spring basin: Chemical and microbiological indicators. Sci.
Total Environ. 407:2872–2886.
Kendall, C., and R. Aravena. 2000. Nitrogen isotopes in groundwater systems.
p. 261–298. In P. Cook, and A.L. Herczeg (ed.) Environmental tracers in
subsurface hydrology. Kluwer Academic, Boston, MA.
Koba, K., N. Tokuchi, E. Wada, T. Nakajima, and G. Iwatsubo. 1997. Intermit-
tent denitrifi cation:  e application of a 15N natural abundance method
to a forested ecosystem. Geochim. Cosmochim. Acta 61:5043–5050.
Lipp, E.K., and D.W. Griffi n. 2004. Analysis of coral mucus as an improved
media for detection of enteric microbes and patterns of sewage contami-
nation in reef environments. EcoHealth 1:284–295.
Mariotti, A. 1983. Atmospheric nitrogen is a reliable standard for natural 15N
abundance measurements. Nature 303:685–687.
McCray, J.E., S.L. Kirkland, R.L. Siegrist, and G.D.  yne. 2005. Model pa-
rameters for simulating fate and transport of on-site wastewater nutri-
ents. Ground Water 43:628–639.
McMahon, P.B., K.F. Dennehy, B.W. Bruce, J.K. Bohlke, R.L. Michel, J.J.
Gurdak, and D.B. Hurlbut. 2006. Storage and transit time of chemicals
in thick unsaturated zones under rangeland and irrigated cropland, High
Plains, United States. Water Resour. Res. 42:1–18.
Melnick, J.L. 1989. Enteroviruses. p. 191–263. In A.S. Evans (ed.) Viral infec-
tions of humans, epidemiology and control. Plenum, New York.
Moore, M. 1982. Enteroviral disease in the United States, 1970–1979. J. In-
fect. Dis. 146:103–108.
Morgan, D.S., S.R. Hinkle, and R.J. Weick. 2007. Evaluation of approaches
for managing nitrate loading from on-site wastewater systems near La
Pine, Oregon. USGS Scientifi c Investigations Rep. 2007-5237. USGS,
Reston, VA.
Murray, K.E., D.R. Straud, and W.W. Hammond. 2007. Characterizing
groundwater fl ow in a faulted karst system using optical brighteners from
septic systems as tracers. Environ. Geol. 53:769–776.
Nelson, D.W. 1983. Determination of ammonium in KCl extracts of soils by
the salicylate method. Commun. Soil Sci. Plant Anal. 14:1051–1062.
Katz et al.: Contaminants beneath Septic Tank Drain elds 1195
Panno, S.V., K.C. Hackley, H.H. Hwang, S.E. Greenberg, I.G. Krapac, S.
Landsberger, and D.J. O’Kelly. 2006. Characterization and identifi ca-
tion of Na-Cl sources in ground water. Ground Water 44:176–187.
Patton, C.J., and E.P. Truitt. 1992. Methods of analysis by the U.S. Geologi-
cal Survey National Water Quality Laboratory: Determination of total
phosphorus by a Kjeldahl digestion method and an automated colormet-
ric fi nish that includes dialysis. USGS Open-File Rep. 92-146. USGS,
Denver, CO.
Patton, C.J., and E.P. Truitt. 2000. Methods of analysis by the U.S. Geological
Survey National Water Quality Laboratory: Determination of ammo-
nium plus organic nitrogen by a Kjeldahl digestion method and an auto-
mated photometric fi nish that includes digest cleanup by gas diff usion.
USGS Open-File Rep. 00-170. USGS, Denver, CO.
Ptacek, C.J. 1998. Geochemistry of a septic-system plume in a coastal barier
bar, Point Pelee, Ontario, Canada. J. Contam. Hydrol. 33:293–312.
Robertson, W.D., J.A. Cherry, and E.A. Sudicky. 1991. Ground-water con-
tamination from two small septic systems on sand aquifers. Ground Wa-
ter 29:82–92.
Rose, J.B., D.W. Griffi n, and L.W. Nicosia. 2000. Virus transport: From septic
tanks to coastal waters. Small Flows Quarterly 1:20–23.
Rosen, M.R., C. Kropf, and K.A.  omas. 2006. Quantifi cation of the contri-
bution of nitrogen from septic tanks to ground water in Spanish Springs
Valley, Nevada. USGS Scientifi c Investigations Rep. 2006-5206. USGS,
Reston, VA.
Rupert, F.R., and S. Spencer. 1988. Geology of Wakulla County, Florida. Bull.
60. Florida Geological Survey, Tallahassee, FL.
Seiler, R. 2005. Combined use of 15N and 18O of nitrate and 11B to evaluate
nitrate contamination in groundwater. Appl. Geochem. 20:1626–1636.
Seiler, R.L., S.D. Zaugg, J.M.  omas, and D.L. Howcroft. 1999. Caff eine
and pharmaceuticals as indicators of waste water contamination in wells.
Ground Water 37:405–410.
Sherman, K.M., and D.L. Anderson. 1991. An evaluation of volatile organic
compounds and conventional parameters from on-site sewage disposal
systems in Florida. p. 62–75. In J. Converse (ed.) On-site wastewater
treatment. v. 6. ASAE Pub. 1091. ASAE, St. Joseph, MI.
Sigman, D.M., K. Casciotti, M. Andreani, C. Barford, M. Galanter, and J.K.
Böhlke. 2001. A bacterial method for the nitrogen isotopic analysis of
nitrate in seawater and freshwater. Anal. Chem. 73:4145–4153.
Swartz, C.H., S. Reddy, M.J. Benotti, H. Yin, L.B. Barber, B.J. Brownawell,
and R.A. Rudel. 2006. Steroid estrogens, nonylphenol ethoxylate me-
tabolistes, and other wastewater contaminants in groundwater aff ected
by a residential septic system on Cape Cod, MA. Environ. Sci. Technol.
40:4894–4902.
U.S. Geological Survey. 1999. National fi eld manual for the collection of wa-
ter-quality data: U.S. Geological Survey techniques of water-resources
investigations, book 9, Chapters A1–A9, 2 Vol. Chapters originally were
published from 1997–1999; updates and revisions are available at http://
water.usgs.gov/owq/FieldManual/ (verifi ed 10 Feb. 2010).
U.S. Environmental Protection Agency. 2003. Funding decentralized waste-
water systems using the clean water state revolving fund. Publication
832-F-03-003. Available at http://www.scdhec.gov/environment/ocrm/
plan_tech/docs/srf_funding.pdf (verifi ed 10 Feb. 2010). USEPA, Wash-
ington, DC.
Vengosh, A., and I. Pankratov. 1998. Chloride/bromide and chloride/fl uoride
ratios of domestic sewage effl uents and associated contaminated ground
water. Ground Water 36:815–824.
Wilhelm, S.L., Sr., S.L. Schiff , and W.D. Robertson. 1994. Chemical fate and
transport in a domestic septic system: Unsaturated and saturated zone
geochemistry. Environ. Toxicol. Chem. 13:193–203.
Zaugg, S.D., and T.J. Leiker. 2006. Review of method performance and im-
provement for determining wastewater compounds (schedule 1433).
U.S. Geological Survey National Water Quality Laboratory Technical
Memorandum 2006.01. USGS, Reston, VA.
Zaugg, S.D., S.G. Smith, M.P. Schroeder, and M.R. Burkhardt. 2002. Meth-
ods of analysis by the U.S. Geological Survey National Water Quality
Laboratory: Determination of wastewater compounds by polystyrene-
divinylbenzene solid-phase extraction and capillary-column gas chro-
matography/mass spectrometry. Water-Resources Investigations Rep.
01-4186. USGS, Reston, VA.
... However, due to the anaerobic condition within the tank and microbial activities, organic N is converted to inorganic N (largely ammonium-N) with no reduction in total nitrogen (TN). A large range in total nitrogen concentrations 10 -210 mg l -1 have been recorded in septic tank discharges (Edwards and Withers, 2008;Humphrey et al., 2013;Katz et al., 2010;Lusk et al., 2017;Valiela et al., 1997). This has been attributed to the variability in daily water usage which directly influences dilution capacity of the tanks as well as dietary differences in population utilising the system. ...
... Ammonium (NH 4 ) is normally the dominant nitrogen species released from STS (>70%) (Gill et al., 2009;Katz et al., 2010;Lusk et al., 2017). However, the importance of organic-N has also been highlighted, normally accounting for between 10-30% of TN, although some studies have found it to be the dominant N species (O'Driscoll et al., 2014). ...
... N speciation in the soakaway is heavily dependent on the site conditions. When STS operate efficiently, almost all the TN discharged from the tank is converted to NO 3 by nitrification in the soakaway (94-99%) Gill et al., 2009;Katz et al., 2010;O'Driscoll et al., 2014). However, where the soakaway is not operating effectively, NH 4 and organic-N can account for a significant proportion of the drainfield TN (> 40%). ...
Technical Report
Full-text available
Septic tank systems (STS) are private sewage treatment facilities which typically serve the population not connected to main sewer networks. There is substantial uncertainty about the impact of septic tanks on water quality, primarily because of a lack of information about the location, number and condition, and inadequate monitoring of the effects of septic tank discharges to surface water and groundwater. Under the requirements of the Water Framework Directive, there is a need for SEPA to identify pressures contributing to water quality downgrades and to put in place appropriate and feasible measures to return waters to good status. SEPA currently use the SAGIS model to identify major pollutants including phosphorus (P) loads and concentrations at the waterbody scale. However, assumptions regarding parameterisation and uncertainty of SAGIS model outputs, particularly in rural areas, make it difficult to quantify the sources of P pollution from septic tanks. A new approach to modelling the contribution of STS P loading is therefore required to inform mitigation strategies, while accounting for uncertainties. The aim of this project was to review the literature and develop a new probabilistic modelling approach, based on Bayesian Belief Networks (BBN), to improve understanding of soluble reactive phosphorus (SRP) losses to surface water from STS. This was initially undertaken for seven pilot catchments and then at a national scale. The model included a range of risk factors related to SRP losses from STS, resulting in simulated SRP losses that were comparable with previously available estimates. The project also considered what factors would need to be applied for modelling nitrogen (N) and microbial (FIOs) pollution risk from STS. Key findings indicate that: (1) Poor performance of a large proportion of STS and hence the risk of SRP leaching to water bodies are related to the lack of proper soakaway system, poor soil quality, undersized STS, and lack of maintenance. Hence the literature review suggested multiple risk factors contributing to STS pollution risk to watercourses need to be considered when simulating SRP losses; (2) Model sensitivity analysis identified STS effluent concentration, STS density, and connectedness as the most important risk factors; (3) Upgrading of STS treatment to secondary or tertiary level, improving STS maintenance and/or disconnecting STS direct discharges to watercourses would be most effective in mitigating SRP losses at catchment scale; (4) The model could be adapted relatively easily to simulate microbial (faecal indicator organisms) and nitrogen losses from STS to surface water. A detailed catchment-based survey of STS condition would be beneficial to validate and improve BBN model parameterisation.
... conduits, fissures and fractures) (Dvory et al., 2018a(Dvory et al., , 2018bEinsiedl et al., 2010). Other studies have indirectly provided information on attenuation processes (Bekele et al., 2009;Doummar and Aoun, 2018b;Heinz et al., 2009;Hillebrand et al., 2012bHillebrand et al., , 2015Katz et al., 2010;Reh et al., 2013;Zemann et al., 2015). Einsiedl et al. (2010) used two pharmaceuticals, diclofenac and ibuprofen, from wastewater effluents in springs, to compliment (Dodgen et al., 2017;Huang et al., 2019;Kiefer et al., 2021 modelling and tracer testing studies of attenuation. ...
Article
Full-text available
A quarter of the world's population uses groundwater from karst aquifers. A range of emerging organic contaminants (EOCs) are considered a potential threat to water resources and dependant ecosystems, and karst aquifers are the most vulnerable groundwater systems to anthropogenic pollution. This paper provides the first global compilation (based on 50 studies) of EOCs in karst aquifers and explores EOC occurrence and the use of EOCs to understand karst systems. Of the 144 compounds detected in the reviewed studies, the vast majority in karst groundwater are pharmaceuticals and pesticides. Maximum concentrations of compounds varied over five orders of magnitude, and nearly half of the detected compounds exceed 100 ng/L. Karst groundwater is shown to have lower frequency of detection and lower concentrations compared to surface waters and local shallow intergranular aquifers, but overall higher concentrations compared to other major aquifer types. A growing number of studies have demonstrated the utility of EOCs and some legacy compounds for groundwater quality assessment and as tracers for characterising karst systems. They can improve understanding of vulnerability, storage, attenuation mechanisms, and in some cases have been used to assist with catchment delineation. This is a growing research area for karst hydrogeology, and more research is needed to understand EOC contamination of karst aquifers, and to develop EOCs as tracers within karst to improve our understanding of this critical water resource.
... Septic system effluent contamination of groundwater with fecal indicator bacteria and pathogenic viruses and bacteria is well documented in the literature (Hagedorn et al. 1981;Yates 1985: Nicosia et al. 2001Katz et al. 2010;Hynds et al. 2012;Lusk et al. 2017). In one study, vaccine poliovirus was introduced into the tank of a new conventional septic system, and the virus was cultured in multiple samples over time in a monitoring well 6 m down-gradient from the edge of the drainfield (Alhajjar et al. 1988). ...
Article
Full-text available
Background: Groundwater quality in the Silurian dolomite aquifer in northeastern Wisconsin, USA, has become contentious as dairy farms and exurban development expand. Objectives: We investigated private household wells in the region, determining the extent, sources, and risk factors of nitrate and microbial contamination. Methods: Total coliforms, Escherichia coli, and nitrate were evaluated by synoptic sampling during groundwater recharge and no-recharge periods. Additional seasonal sampling measured genetic markers of human and bovine fecal-associated microbes and enteric zoonotic pathogens. We constructed multivariable regression models of detection probability (log-binomial) and concentration (gamma) for each contaminant to identify risk factors related to land use, precipitation, hydrogeology, and well construction. Results: Total coliforms and nitrate were strongly associated with depth-to-bedrock at well sites and nearby agricultural land use, but not septic systems. Both human wastewater and cattle manure contributed to well contamination. Rotavirus group A, Cryptosporidium, and Salmonella were the most frequently detected pathogens. Wells positive for human fecal markers were associated with depth-to-groundwater and number of septic system drainfield within 229m. Manure-contaminated wells were associated with groundwater recharge and the area size of nearby agricultural land. Wells positive for any fecal-associated microbe, regardless of source, were associated with septic system density and manure storage proximity modified by bedrock depth. Well construction was generally not related to contamination, indicating land use, groundwater recharge, and bedrock depth were the most important risk factors. Discussion: These findings may inform policies to minimize contamination of the Silurian dolomite aquifer, a major water supply for the U.S. and Canadian Great Lakes region. https://doi.org/10.1289/EHP7813.
... According to Petitjean et al. (2016), about 20% of the population use onsite wastewater treatment in France. In the United States Katz et al. (2010) reported that about one in four households in the United States uses a septic tank for wastewater disposal. Approximately, 500,000 rural households depend on onsite wastewater treatment systems for their effluent disposal in Ireland Keegan et al. (2014). ...
Chapter
Full-text available
A residential housing is normally designed to serve the immediate need for human activities. Usually, provision for collection and disposal of sewage is a priority to enhance a healthy living environment. The millennium development goal on human health remains unrealistic in many developing countries due to poor disposal of sewage. Using a combination of primary and secondary data this study presents an overview of onsite sewage disposal and its impact on underground water and health issues in Nigeria. Findings show that onsite septic tank is the main method of sewage disposal. An average of 5 m was observed between the shallow dug well and septic tank. Hence, the underground water is highly susceptible to contamination but remains the primary source of domestic water supply. Reported cases of cholera traceable to fecal contamination of underground water remain very high. This paper argues for the involvement of professional in the building industry to present a proposal or a central sewage system particularly where the risk of infiltration of sewage into the water body is high.
... A wide range of compounds may be found in these waters, the most common of which may be nitrate (NO 3 -) (Pye et al. 1983). The sources of nitrogen (N) may include, but are not limited to, atmospheric deposition (National Atmospheric Deposition Program 2015), septic tanks (Katz et al. 2010), effluent water disposal (Warneke et al. 2011), agricultural fertilization (Schmidt and Clark 2012), or landscape fertilization (Erickson et al. 2008). Despite evidence to the contrary, some Floridians assume that N applied to turfgrass is a major contributor to water pollution (Shaddox et al. 2016a;Shaddox et al. 2016b;Telenko et al. 2015;Trenholm and Unruh 2005). ...
Article
The quality of Florida’s surface and ground waters is of utmost importance to the flora and fauna living in them, as their growth is directly related to the amount of nutrients in these waters. In order to make informed decisions regarding nitrogen (N) applications to turfgrass, it is important to understand the N cycle in the soil/turfgrass system. The objective of this 8-page publication is to identify and describe the sources and potential fates of N applied to Florida turfgrass. Written by T.W. Shaddox and J.B. Unruh and published by the UF/IFAS Department of Environmental Horticulture, January 2018. http://edis.ifas.ufl.edu/ep546
Article
As analytical capabilities in the early 2000s began to enable the detection of chemicals in environmental media at increasingly small concentrations, chemicals with the potential to cause adverse human and ecosystem health effects began to be found nearly ubiquitously worldwide. The types of chemicals that were targeted for analysis included natural and synthetic hormones, human and veterinary pharmaceuticals, chemicals in personal care products, novel pesticides, nanoparticles, microplastics, and other chemicals of natural and synthetic origin. The impacts of these chemicals on environmental and human health in many cases remain unknown. Collectively, these chemicals became known as “emerging contaminants” or “contaminants of emerging concern”. Much progress has been made toward understanding the sources of these contaminants in the environment, the processes that control their fate and transport once they are released into the environment, and the ability of technology and/or best management practices to mitigate their occurrence. As the Journal of Environmental Quality (JEQ) celebrates its 50th anniversary, we sought to understand how publications in the journal have made impactful contributions in the research area of emerging contaminants. Here, we present the trajectory of publications in JEQ that have shaped knowledge in this field, highlight the importance of these contributions, and conclude with opportunities for JEQ to continue attracting high‐quality emerging contaminants research. This article is protected by copyright. All rights reserved
Article
The presence of pharmaceuticals and personal care products (PPCPs) in the environment is primarily the result of discharge of waste, including from onsite wastewater treatment systems (OWTSs) which are employed by 25% of homes in the United States. However, the occurrence and removal of PPCPs in OWTSs is not well understood, particularly given the large diversity in PPCP compounds as well as in OWTS designs. In this study, we monitored 26 different PPCPs in 13 full-scale nitrogen removing biofilters (NRBs), an innovative/alternative type of OWTS that utilizes an overlying sand layer and an underlying woodchip/sand layer to simultaneously remove nitrogen and other wastewater-derived contaminants. The specific objectives of this study were (i) to measure the occurrence of PPCPs in septic tank effluent (STE) that served as an influent to NRBs, (ii) to quantify PPCP removal in three types of NRB configurations (n = 13), and (iii) to evaluate PPCP removal with depth and environmental conditions in NRBs. Aqueous samples were taken during 42 separate sampling events during 2016 – 2019 and analyzed by liquid chromatography tandem mass spectrometry. Analysis of the STE samples yielded detection of 23 of the 26 PPCPs, with caffeine being the most abundant and frequently detected compound at 52,000 ng/L (range: 190 – 181,000 ng/L), followed by acetaminophen and paraxanthine at 47,500 ng/L (190 – 160,000 ng/L), and 34,300 ng/L (430 – 210,000 ng/L) respectively. Cimetidine, fenofibrate, and warfarin were the only compounds not detected. The average removal of PPCPs by NRBs ranged from 58% to >99% for the various compounds. PPCP removal as a function of depth in the systems showed that 50 to >99% of the observed removal was achieved within the top oxic layer (0 - 46 centimeters) of the NRBs for 19 analytes. Seven of the compounds had >85% removal by the same depth. These results indicate that NRBs are effective at removing PPCPs and that a large portion of the removal is achieved within the oxic nitrifying layer of the NRBs. Overall, the removal of PPCPs in NRBs were comparable (n = 8) or better (n = 15) than that observed for conventional wastewater treatment plants.
Article
Full-text available
Rudimentary onsite sanitation systems (OSSs) are extensively used in low‐income countries despite the risk of groundwater contamination. The present study investigated the potential impact of community soak pits on sandy shallow aquifers in a rural area in Sri Lanka. Thirty‐two groundwater samples were collected and measured for traditional indicators, i.e., fecal bacteria Escherichia coli (E.coli), total coliforms (TCs), nitrate and chloride; seven pharmaceuticals were added as wastewater indicators. Analysis showed that the local groundwater is infected by fecal bacteria, but it was unable to determine whether the source of fecal bacteria is from OSSs. Similarly, nitrate and chloride provided unclear evidence. Conversely, detection of three pharmaceuticals at trace levels in groundwater evidenced wastewater impact from OSSs. Caffeine (CAF) was detected in 89% of the sample at maximum concentration 7.9 ng/L, indicating fecal pollution. Carbamazepine (CBZ) was detected in 42% of the samples with a concentration of up to 6.9 ng/L, while sulfamethoxazole (SFM) was detected in only two samples. The presence of CBZ and SFM were also consistent with recorded drug use of the residents. E. coli showed a moderate positive correlation with CAF concentration (Kendall's tau τ = 0.38, p = 0.017), indicating concurring short‐lifetime fecal bacteria and labile wastewater organic compounds. Nitrate showed a significant correlation with CBZ concentration (τ = 0.39, p = 0.016). Fecal bacteria and nitrate can be used in screening for micro‐pollutants in domestic wells impacted by OSSs. This case study emphasized the potential impact of poor sanitation on groundwater resource and the necessity of sanitation improvement in achieving Sustainable Development Goals in developing countries. This article is protected by copyright. All rights reserved.
Article
Full-text available
Eutrophication is one of the causes of the degradation of reefs worldwide. The aim of this research is to determine if sewage discharge reaches the fore reefs at northwest of Cuba using δ15N in tissues of the octocorals Eunicea flexuosa and Plexaura kuekenthali and the concentration of microbiological and physical-chemical variables. Thirteen reefs at 10-m depth were selected near river basins and far from the urban and industrial development of Havana City. Branch tips of both species were collected, the concentrations of nutrient and microorganisms in water samples were quantified, and horizontal visibility in the water (Vis) was determined. Overall, δ15N of E. flexuosa ranged from 1.5 to 6.3‰ and P. kuekenthali from 1.7 to 6.7‰. The tissue of both species was significantly enriched in 15N in reefs near polluted watersheds compared with reefs far from pollution by anthropogenic activities. The δ15N of both species showed a positive and significant correlation with the concentration of fecal and total coliform bacteria, heterotrophic bacteria, and NH4+ and a negative and significant correlation with the Vis. The δ15N of the two species and microbiological and physical-chemical variables evidenced water quality decline by sewage discharge that reached reefs near polluted watersheds.
Chapter
Full-text available
Nitrate contamination, often associated with agricultural activities, is a major problem in some shallow aquifers and is increasingly becoming a threat to groundwater supplies (Gillham and Cherry, 1978; Ronen et al., 1983; Spalding and Exner, 1991). The intake of high levels of nitrate can cause methemoglobinemia in infants, and there is substantial evidence collected from animal experiments that N-nitroso compounds are carcinogens. Similar conclusive evidence is not yet available for humans but many observations suggest that these compounds can function as initiators of human carcinogenesis. These findings are the basis for the maximum permissible limit of 10 ppm nitrate-N (50 ppm as NO3) in drinking water set by the World Health Organization and the U.S. Environmental Protection Agency. The impact of high loading of nutrients such as nitrate and phosphorous from agricultural practices via groundwater into surface water is also a major environmental concern, causing eutrophication of streams, rivers and lakes (Hill, 1978; Böhlke and Denver, 1995).