Content uploaded by Christopher W. Sinton
Author content
All content in this area was uploaded by Christopher W. Sinton on Oct 11, 2017
Content may be subject to copyright.
Volume 33, pages 1 – 11 www.northeasterngeoscience.org 2015
INTRODUCTION
During the late 18th and early 19th centuries, the northeastern United
States became industrialized due in large part to the harnessing of
hydropower along the many streams and rivers in the region (Hunter,
1979). For example, the Merrimack River of New England became a
hub of textile manufacturing in a matter of a few decades due to its
abundant hydropower (Steinberg, 2003). In addition to textiles, water-
power was used for other industries including lumber production, grain
milling, leather tanning, and metal forging. In order to take advantage
of hydropower, factories and mills were built adjacent to waterways.
The proximity of rivers and streams also provided an easy method of
getting rid of waste material. As a result, organic and inorganic waste
materials migrated into the waterways, either by direct dumping or by
surface runoff that entrained and transported material from the factory
site. After hydropower was largely displaced by coal-fueled steam
power in the latter part of the 19th century (Hunter, 1979), some
factories remained at their original sites and persisted as potential
sources of water contamination. While some material that was
transported off site, particularly organic wastes, can quickly biode-
grade, some persistent contaminants, such as heavy metals, can be of
concern long after the factory site has been abandoned.
The Ithaca Gun Company factory in Ithaca, New York is an
example of an old manufacturing operation that is adjacent to a water-
way. The site is located on a ledge overlooking the Ithaca Falls gorge of
Fall Creek, approximately 2 km upstream from where the creek
empties into southern Cayuga Lake (Figure 1). In 1813, the site was
first developed for industry and by the 1830s several mills were operat-
ing using hydropower from a tunnel and raceway constructed by Ezra
Cornell (Snyder, 1991). In 1888, the site was occupied by the Ithaca
Manufacturing Company, which made agricultural equipment, and the
W.H. Baker & Company gun factory. W.H. Baker soon expanded to the
entire site, was renamed the Ithaca Gun Company, and manufactured
firearms and munitions until 1986 (Snyder, 1991). On-site activities
included metal machining operations and testing of firearms in indoor
and outdoor shooting ranges (Prescott Assoc., 2002).
In 1995 the New York State Department of Environmental
Conservation (NYSDEC) first observed lead shot near the site and,
shortly after that, studies conducted by Cornell University and
NYSDEC found elevated levels (up to 40,000 ppm) of Pb (O’Brien
and Gere, 2013). Between 2001 and 2004, the US Environmental
Protection Agency (USEPA) removed 6,000 tons of soil containing
Pb and slag from the site. In 2009, the buildings from the site,
including old pits and foundations, were demolished and removed.
In 2014, a third round of site cleanup was initiated after detection of
elevated lead in some of the remaining unconsolidated surficial
material.
The company’s buildings were located on level surfaces,
but were surrounded by sloped terrain with the north side of the site
bounded by the sheer drop into Ithaca Falls gorge. It is conceivable
or perhaps likely that Pb migrated into Fall Creek during the 150+
years of operations at the site, particularly during flooding events.
Erosion from the site could have moved sediment downslope into
Fall Creek and into the delta where the creek meets southern
Cayuga Lake. The goal of this study was to determine if the heavy
metal concentrations, particularly Pb, in sediments downstream
from the Ithaca Gun site are elevated compared to background
(derived from natural sources such as bedrock weathering) and
ambient (background plus anthropogenic non-point sources)
concentrations.
METHODS
Study Area
The study area includes the ~2 km length of Fall Creek
between the Ithaca Falls gorge and the delta where the creek
ABSTRACT:
Industrial activity during the 18th and 19th centuries relied heavily on hydropower and, as a result,
mills and factories were constructed near rivers and streams. Unfortunately, the placement of industrial facilities
adjacent to waterways resulted in the contamination of some rivers and streams in the northeastern United States. The
Ithaca Gun Company of Ithaca, NY operated for almost a century adjacent to Fall Creek and after it closed in 1986, it
became an EPA-designated Superfund site due to excessive Pb concentrations in the soil. This study was designed to
determine if lead from the Ithaca Gun site has migrated downstream and is present in the sediments of Fall Creek and
its delta in Cayuga Lake. Our results show that recent sediments in the stream channel and delta in Cayuga Lake do
not have Pb concentrations above ambient levels. Some of the older lake sediments and some samples from the
floodplain have elevated Pb concentrations that are close to the ambient lake sediment compositions prior to the 1970s.
Positive correlations between the trace metals do not indicate is that the Pb is derived from a point sourcepreferentially
enriched in Pb, like the Ithaca Gun site. Based on New York State threshold levels, none of the Pb concentrations pose
a threat to human health.
Northeastern Geoscience
Volume 33
EVALUATION OF HEAVY METALS
IN SEDIMENTS DOWNSTREAM FROM
THE ITHACA GUN SUPERFUND SITE
Christopher W. Sinton1 , Lee Ann Hill1,2, and Camron Zerbian1
1 - Environmental Studies and Sciences, Ithaca College, Ithaca, New York
2 - now at University of California-Berkeley, Berkeley, California
© 2015 northeasterngeoscience.org
All rights and permissions beyond publication in this issue of Northeastern Geoscience
are held by the authors.
Corresponding Author:
Christopher W. Sinton
Environmental Studies and Sciences
Ithaca College
953 Danby Rd, Ithaca, NY 14850
csinton@ithaca.edu
empties into Cayuga Lake (Figure 1). From the Ithaca Falls gorge
downstream to where Fall Creek flows under State Route 13, the
streambed consists of gravel and cobbles and was therefore not
suitable for sampling the finer sediments we needed for composi-
tional analysis. In addition, there is no undisturbed floodplain and
the banks of this section of the stream have been reinforced by stone
riprap. As a result, the floodplain was not sampled in this area.
After the creek passes underneath Route 13, the stream
channel begins to broaden and the streambed becomes siltier. At
this point, the floodplain to the north consists of the wooded Fuertes
Bird Sanctuary and the adjacent Stewart Park (Figure 1). On the
south side of Fall Creek the floodplain consists of the Newman golf
course (built in 1935) and to the west of that is a 20 acre parcel of
woods and marshland that housed the old Cornell University
biological field station that is often referred to as Jetty Woods. At
this point, Fall Creek meets Cayuga Lake to form a delta. The
mouth of the delta is shallow (1-2 m depth) and sandy for the first
~200 m into the lake where consistent waves keep the finer
sediments suspended. Further into the lake the depth to the lake
bottom increases and the sediments become finer.
Sample Collection and Processing
Surface sediment grabs and sediment cores were taken
during the summers of 2012 and 2013 from 18 locations in Fall
Creek, the adjacent floodplain, and Cayuga Lake (Figure 1; Table
1). Location coordinates were determined using an e-Trex handheld
GPS unit.
Six stream bed surface samples (76-81) were recovered
from the center of the Fall Creek channel and the mouth of the delta
using a stainless steel Wildco Ekman sediment dredge deployed
from a motorboat. Samples were placed in sealed plastic bags and
labeled. Six delta and lake sediment cores were taken from a motor-
boat using a gravity corer with plastic liners (Aquatic Research
Instruments) and deployed with an aluminum extension bar to drive
the core into the sediment. It was not possible to take cores from the
shallow, sandy portion of the delta because the sediments would not
remain in the core barrel. The cores were refrigerated until they
were extruded from the plastic sleeves, cut in half lengthwise with a
stainless steel blade, photographed, and described.
Floodplain core samples were recovered using a handheld
soil corer (ESP Plus, Clements Assoc.) with plastic core liners. The
Fuertes bird sanctuary was not sample in the wooded portion but a
shallow core was taken in a marsh adjacent to a small embayment of
Fall Creek (site 62). The core hit coarse cobbles at ~20 cm depth –
these cobbles appear to cover the shallow embayment. It is unclear
if these were deposited naturally or were dumped as fill as we
observed piles of old bricks and rock on the margin of the nearby
woods. Because the golf course is a disturbed part of the floodplain,
it was not sampled. Rather, a core was taken at the edge of the
channel (site 64) adjacent to the golf course. The Jetty Woods area
appears to be relatively undisturbed so three cores were taken in the
marshlands (sites 71, 73, and 74).
Two control samples were taken from areas that would not
have been influenced by the Ithaca Gun site. One grab sample (83)
was taken upstream from the site in the shallow western end of
Beebe Lake (Figure 1), which is a reservoir created by a dam in Fall
Creek. This sediment should represent weathering and erosion from
the Fall Creek watershed. The second control sample is a core
(CID2) taken to the west of the Fall Creek delta and close to where
the Cayuga Inlet flows into Cayuga Lake. The combined outflow
from Enfield Creek, Six Mile Creek, and Cascadilla Creek pass
through the Inlet and into Cayuga Lake so sediment at this location
should represent eroded material from these watersheds. In general,
the Inlet discharges a higher level of suspended sediment and Figure
1 shows the extent of this sediment plume.
After collection, the samples were described. Particle size
descriptions were based on feel and were categorized as silt/clay,
sandy silt, and sand. Composite samples for the cores were made by
homogenizing material from 3 to 10 cm thick layers (See appendix,
Table 2). Large lithic pieces and large pieces of organic material
such as twigs were removed prior to making the composite samples.
Total Organic Carbon
Total organic carbon (TOC) was measured using the
semi-quantitative method from Schumacher (2002). Dried, clean
crucibles were weighed, loaded with 50-60 g of moist sample, and
then placed in a 110°C drying oven for 12 hours. After drying, the
samples were again weighed to determine the dry mass of the
sample and then placed in a 400°C muffle furnace for 8 hours. At
this temperature, the organic materials will combust but the carbon-
ate minerals should not decompose. The samples were weighed
again and the reduction in mass was calculated by subtracting the
ignited mass from the dried mass. The estimated TOC was
calculated by dividing the mass loss by 1.724. The combusted
samples were placed in a dessicator in preparation of trace element
analysis.
Trace Element Analysis
The combusted material from the grab samples, the
floodplain cores, three of the lake cores, and the control sites were
powdered using a tungsten carbide ring and puck mill. The samples
were pressed into pellets using a wax binder and hydraulic press and
trace element compositions were determined by X-ray fluorescence
(XRF) using the Philips PW2404 system at Colgate University
using methods based on those described in Norrish and Chappell
(1977). Standard curves for each element are based on USGS
standard reference materials. Accuracy is determined by running the
calibration standards resulting in a deviation from standard values
ranging from 1% to 10%. Samples 79 and 79R are replicate samples
taken from the same composite sample. Both were processed with
the other samples, from combustion, powdering, and analysis. This
was done to determine how well each analyzed sample represented
the bulk composition of the composite. Two of the lake cores taken
close to site FCD4 were not analyzed because of budget limitations.
RESULTS
The types of sediment recovered during this study range from
silt/clay to gravel. The sediment grabs from the Fall Creek channel
are a mix of clay, silt, and organic debris. Samples from delta sites
80 and 81 are sandy as this area is subjected consistently to waves
and as a result the delta floor is sandy as far out as site FCD1. The
cores from the delta and lake are all silt/clay with the exception of a
layer of sand and gravel about 16 cm from the sediment surface at
site FCD4 and the two other cores that were recovered but not
analyzed (Figure 2). This coarse layer may represent a flood deposit,
possibly from Hurricane Agnes in 1972. If this is accurate, then that
would represent an average sedimentation rate of 4 mm/year, which
is within the range of the 2.4-8.1 mm/year calculated by Yager
(2001) for southern Cayuga Lake.
The tops of the lake cores contain coal fragments which
could be from the years when coal was transferred from train to
barge when the Ithaca-Owego line transported coal to the Erie Canal
or, more likely, it is from a 1999 train derailment and coal spill into
2
www.northeasterngeoscience.org Volume 33 (2015)
Northeastern Geoscience
Sinton, Hill & Zerbian
3
www.northeasterngeoscience.org Volume 33 (2015)
Northeastern Geoscience
Evaluation of Heavy Metals from Ithaca
Figure 1. Map showing the sample location sites. The former Ithaca Gun site is noted in the red box southeast of Cayuga Lake. Fall Creek is shown with a white
dotted line. Surface grab samples are shown in yellow circles; green circles are control samples; and red circles are sediment cores. The easternmost
control sample (83) is a surface grab from Beebe Lake which is a reservoir of Fall Creek upstream from Ithaca Gun. The inset map is a close-up of the
delta where Fall Creek empties into Cayuga Lake. The Jetty Woods (blue outline) is adjacent to the Newman golf course and Stewart Park is the north.
The map was made using ArcGIS with data sourced from Esri, DigitalGlobe, i-cubed, USDA, USGS, AX, Getmapping Aerogrid, IGN, IGP, swisstopo,
and the GIS User Community.
Table 1: Fall Creek Sample Locations
Surface Grabs Latitude Longitude
62 42°27' 26.62'' N 76°30' 09.72'' W
76 42°27' 30.95'' N 76°30' 21.16'' W
77 42°27' 34.21'' N 76°30' 27.99'' W
78 42°27' 34.68'' N 76°30' 30.31'' W
79, 79R 42°27' 36.31'' N 76°30' 36.59'' W
80 42°27' 37.84'' N 76°30' 38.22'' W
81 42°27'38.78"N 76°30'41.10"W
83 42°27'08.66'' N 76°28'45.98'' W
Cores
64 42°27' 29.97'' N 76°30' 21.77'' W
71 42°27' 33.60'' N 76°30' 32.42'' W
73 42°27' 30.90'' N 76°30' 30.90'' W
74 42°27' 32.45'' N 76°30' 32.02'' W
FCD1 42°27'49.80"N 76°30'31.80"W
FCD4 42°28'4.44"N 76°30'40.80"W
FCD6 42°28'18.30"N 76°30'25.62"W
CID2 42°28'8.58"N 76°31'4.56"W
the Cayuga Inlet (Harper and Peckarsky, 2005). The three cores
from Jetty Woods are composed of silt/clay in the top 20 cm but
below that they consist of sandy silt and sand. This suggests that this
area at one point was a shallow section of the lake swept by waves.
Table 2 reports the TOC and the measured trace element composi-
tions of the sediment samples. Throughout the sample area, the Pb
concentrations range from 8 to 81 ppm; Cr 11 to 86 ppm; Cu 11 to
58 ppm; and Zn 26 to 193 ppm. TOC values range from <1% to 9%.
The samples with the highest concentrations of Pb (81 and 69 ppm)
are the top two composite samples of core 74 in the Jetty Woods
(Figure 1). After that, the next highest Pb concentrations (<44 pm) are
found in the lake cores FCD6 and FCD4. The samples from the Fall
Creek streambed have Pb concentrations ranging from 21 to 37 ppm.
DISCUSSION
In this section, we examine the trace element analyses of the
sediment with a particular focus on lead (Pb) because this was the
main contaminant at the Ithaca Gun site. Chromium (Cr) is also
highlighted as this was reported as slightly elevated in one sample
from the industrial site (O’Brien and Gere, 2013). Copper (Cu) and
zinc (Zn) concentrations are compared to the lead as we did not
expect these metals to be elevated from ambient concentrations. The
results are then compared to the control samples that were analyzed
in this study as well as lake sediment analyses from previous work.
These concentrations are then evaluated for potential ecosystem and
human health hazards using established New York State threshold
values.
Trace Element Patterns and Relationships
Relationships between the different trace metals were
examined by plotting the metal concentrations in two-element
diagrams (Figure 3). In these, Pb is plotted against Cr, Zn and Cu
and linear regressions were calculated using all of the data except
the control sites. All three plots show positive linear trends with
r-squared values of 0.77, 0.89, and 0.71 for the plots of Pb vs. Cr,
Zn, and Cu, respectively. That is, the concentrations of these trace
metals increase and decrease together.
There are several possible explanations for the positive
correlations. One is the relationship between particle size and trace
metals adsorbed to particle surfaces. Studies have shown that finer
particle sizes, and therefore higher surface area, in stream sediments
have higher trace metal concentrations (e.g., Horowitz and Elrick,
1987). Higher cation exchange capacity from clay minerals has
been shown to concentrate Pb in sediments (Hem, 1976). As noted
in the previous section, the highest Pb levels were found in the top
layers of core 74 but the deeper sections of the core have much
lower Pb (<13 ppm), coincident with the change from clay to sand.
While we did not conduct any quantitative analysis of particle size
distribution or particle surface area, we can observe that all samples
described as sandy have lower trace metal concentrations. How-
ever, the samples characterized as silt/clay show a range of Pb
concentrations with some <25 ppm. Thus particle size alone cannot
account for the range of metal concentrations in the finer sediments.
Another factor that can affect trace metal concentration is TOC as
ionic metals and ionic metallic compounds can adsorb preferentially
to organic matter (Lollar, 2005; Salim, 1983). In fact, the two
composite samples from the top of core 74 have both the highest Pb
and the highest TOC concentrations of the analyzed samples.
However, a plot of TOC and trace metal concentration (Figure 4)
does not show a discernible positive correlation between the two
and a linear regression through the data has an r-squared value of
0.04.
The biogenic deposition of Mn and Fe oxide and hydrox-
ide coatings on particle surfaces can increase the adsorption of trace
metals to sediments. Studies in Cayuga Lake (Wilson et al., 2001;
4
www.northeasterngeoscience.org Volume 33 (2015)
Northeastern Geoscience
Figure 2. A photograph of FCD-2 (located next FCD-1 on Figure 1) showing a 2 cm-thick layer of coarse sand that we speculate as being a flood deposit from
Hurricane Agnes in 1972.
Sinton, Hill & Zerbian
Evaluation of Heavy Metals from Ithaca
Dong et al., 2003) demonstrated that Fe oxide and biogenic Mn
oxides from biofilms are important adsorption sites for Pb and likely
other trace metals in lake sediments. While the presence of these
oxides was not part of this study, we can at least look for a correla-
tion between the trace metal concentration and Mn or Fe. Because
the XRF method produced low weight percent Mn concentrations
with poor resolution (0.03-0.06 wt. %), we plotted the trace metals
against the more abundant Fe concentrations (0.5-4.5 wt. %) (Figure
4). Here we see a distinct positive correlation with an r-squared
value of 0.85 for the linear regression. While this does not uniquely
show that Fe and Mn oxides play a role in the sequestration of trace
metals, our data show that this is at least a possible factor.
The patterns displayed in Figure 3 and 4 all show positive
correlations between Pb and other metals. This could be due to a
combination of the effects of particle surface area, mineral composi-
tion, organic material content, and adsorption of metals during
deposition or post-depositional processes such as the growth of Mn
derived from a point source, like Ithaca Gun, that is preferentially
enriched in Pb. If this were the case, then Pb would not show
positive correlations with other metals and any samples preferen-
tially enriched in Pb would plot to the right of the graphs in Figure
3.
Comparison to Background and Ambient Pb Concentrations
We wanted to determine if the Pb concentrations of the
collected samples were higher than background and ambient levels.
According to Rice (1999), background “concentration of a trace
element in [streambed] sediment is defined as the concentration that
is the result of natural processes, including weathering and subse-
quent erosion of local soil and bedrock, and atmospheric deposition
unaffected by anthropogenic activity”. We consider ambient levels
to be concentrations of background elements in the sediments with
the addition of non-point source anthropogenic additions. Ambient
anthropogenic Pb inputs can include deposition from the burning of
gasoline containing tetraethyl lead (phased out in the 1970s) and
combustion of coal from regional powerplants. There is one coal
plant currently operating 20 km north of the study site on Cayuga
Lake and the Dresden/Greenidge plant on Seneca Lake 45 km to the
northwest of the study site was shut down in 2011.
In order to assess background levels, we considered the
composition of the bedrock in the watershed and analyses of lake
sediment cores that are of pre-industrial age. The geology of the Fall
Creek watershed is composed of bedrock that is Devonian mudstone
(shale and siltstone) of the Genesee, Sonyea, and West Falls groups
that, in places, is covered in glacial till or post-glacial lacustrine
sediments. There is no published trace metal composition that
5
www.northeasterngeoscience.org Volume 33 (2015)
Northeastern Geoscience
Figure 3. Variation diagrams of Pb plotted against Cr, Cu, and Zn. The blue diamonds
represent the Fall Creek and Cayuga Lake delta sediment analyses reported
in Table 2. Regression lines for these data have r2 values of 0.77, 0.71, and
0.89, respectively. The yellow circles are the control samples from upstream
(83) and west of the delta (CID1). The red squares are sediment sampled
north of the Fall Creek delta as part of the evaluation for the Cornell Lake
Source Cooling (LSC) system (Stearns & Wheeler, 1997). The average New
York shale composition is taken from 68 analyses from 61 to 140 m of the
Akzo core reported in Ver Straeten et al. (2011). The median US stream
sediment values are from Rice (1999).
Figure 4. Plot showing summed trace metal concentrations of the samples with
total organic carbon (TOC) and Fe. A linear regression through the
TOC data has a r2 value of 0.04 and a value of 0.85 for the Fe data.
and Fe oxide coatings. What the data do not indicate is that the Pb is
represents the bulk composition of the mudstones in the Fall Creek
watershed so we used a composite average of mudstone analyses
from a drill core into central New York Devonian bedrock (Ver
Straeten et al., 2011) that are of the age of the Genesee and Sonyea
groups. The Pb concentration average is 7 ppm but ranges from 0 to
72 with the highest concentrations in black shale (high organic)
layers. In Figure 3, the composite sample falls to the left of the trend
observed for the samples of the study area, indicating that Pb is
lower for a given concentration of Cu, Cr, and Zn. Background Pb
concentrations can also be estimated from a study by White (2001)
that reported Pb concentrations in a core of Cayuga Lake sediment
taken about 4 km north of the Fall Creek delta (near the Yacht Club).
The data show that the Pb concentration of sediments prior to ~1850
was about 20 ppm. The concentrations increased progressively to
40-50 ppm by 1970, reflecting increased emissions from anthropo-
genic activity. The top of the core is missing so the estimated young-
est age of this core is 1972. Based on these two lines of evidence,
we can conclude that the background Pb concentrations of Cayuga
Lake sediments is no greater than 20 ppm.
To determine the ambient Pb concentrations in the
Cayuga Lake sediments, we can use the controls from this study and
from previously published sources. Analyses from the two control
sites show trace element compositions that fall within the range of
those in the study area (Figure 3). The upstream Beebe Pond sample
(83) has a Pb concentration of 24 ppm while the two composite
samples from the top 6 cm of the CID2 core near the outflow of the
Inlet have Pb levels of 33-36. Published results of other Cayuga
Lake sediment analyses can also be used as controls. Sediments
were collected north of the Fall Creek delta and analyzed in 1994
and 1996 as part of the environmental impact assessment for the
Cornell lake source cooling (LSC) system (Stearns & Wheeler,
1997). The sampling sites were along the intake pipeline route that
begins just east of site FCD6. The samples included composites of
20 cm depth box cores and composites of cores 1 to 3 m long. The
values for Pb ranged from 10 to 123 ppm (Figure 4) with the highest
value from one of the 1 m core composites. With the exception of
this one sample, the LSC samples fall below 36 ppm, which is
comparable with the control sites for this study. Taken together, the
control data indicate that the maximum ambient Pb concentration is
36 ppm and concentrations above that could be considered elevated.
Samples in the study area with elevated Pb are found in
the tops of the cores from the Jetty Woods, the 6-14 cm section of
delta core FCD4 and the 19-52 cm sections of delta core FCD6
(Figure 5). The most recent sediments represented by the tops of the
lake cores and the sediment grab samples from the Fall Creek
streambed have Pb concentrations between the background and
ambient levels and are therefore not considered elevated. The
sections of the lake cores that are elevated have Pb concentrations
range of 40-50 ppm, which is the same as those in the pre-1972
sediments of White (2001). This could indicate that, at least for
FCD6, the higher lead Pb below 20 cm depth (or older than 1972
using a sedimentation rate of 5 mm/year) represents ambient Pb
levels at that time. This idea is supported by studies of Lake Erie
sediments (Graney et al., 1995) that show a decrease in Pb content
starting in the early 1970s.
Hazard Assessment
In order to assess if a hazard is posed by the metal concen-
trations of the stream and lake sediments, they can be compared to
established threshold levels. In New York State, the Screening
and Assessment of Contaminated Sediment (NYSDEC, 2014) is
used to determine whether or not sediment “is toxic and poses a
risk to aquatic life.” This document sets maximum contaminant levels
of selected chemicals within three classes: Class A sediments
pose low risk to aquatic life; Class B are “slightly to moderately
contaminated and additional testing is required to evaluate the poten-
tial risks to aquatic life”; and Class C sediment should be considered
highly contaminated (Table 3). None of the samples
fall within
Class C. The majority of the samples, including the
background
mudstone composite, are classified as Class B based onthe >43 ppm
threshold for Cr. The threshold for Pb between Class A and Class B
is 36 ppm, exceeded by the lower layers of the three lake delta cores.
The recent stream and lake delta sediments are either at or below
this threshold.
The hazard assessment of the floodplain samples should
be done using established threshold levels for soil. In this case, the
400 ppm Pb concentration stated in the soil cleanup objectives for
residential use (Chapter IV subpart 375-6) can be applied. Based on
this, the floodplain samples in the Jetty Woods do not pose a hazard
to human health. However, when compared to the published thresh-
old for ecological health, the Pb threshold of 63 ppm is slightly
exceeded by the top layer of core 74.
Fate of Lead from Ithaca Gun
Our data indicate that the recent sediments in the areas we
studied downstream from Ithaca Gun do not have Pb concentrations
above ambient levels and older sediments can have slightly higher
Pb levels but not to the level that pose an immediate hazard to
human health. Nevertheless, it still seems reasonable to consider
that the Ithaca Gun site was a source of Pb into Fall Creek through
erosion. Because the main source of lead contamination on the
Ithaca Gun site was elemental Pb in the form of bullets and shot
pellets, we expect that some of the potential transport of Pb into Fall
Creek would be in this form. If shot entered directly into Fall Creek,
we would expect that its high specific density would cause it to drop
out of suspension relatively quickly along with coarse clastic
sediments. If this is the case, then our sampling methods were not
adequate to find lead particles as we did not take samples of the
cobble-gravel streambed directly downstream from the site.
Another possibility is that shot or bullet fragments collected in the
sandy areas at the mouth of the Fall Creek delta. While we did not
see any evidence of shot or bullet fragments in the surface grabs of
this area, we were unable to sample below the sediment surface with
the gravity corer. Therefore it is possible that shot or bullet
fragments have collected in the delta but, if so, they are likely to be
well dispersed.
An alternative transport mechanism for Pb from the Ithaca
Gun site into the creek is from weathering reactions onsite that
6
www.northeasterngeoscience.org Volume 33 (2015)
Northeastern Geoscience
Figure 5. Diagram showing Pb with depth for the three Fall Creek delta cores.
The circles show actual data points.
Sinton, Hill & Zerbian
Evaluation of Heavy Metals from Ithaca
formed other Pb-containing compounds. There have been studies on
the fate of Pb bullets from shooting ranges (Cao et al., 2003;
Jørgensen and Willems, 1987; Lin et al., 1995; Vantelon et al., 2005;
Darling and Thomas, 2003) and these can be applicable to our study.
Metallic Pb in soils is subject to oxidation, carbonation, and hydra-
tion reactions to form secondary minerals such as hydrocerussite
(Pb
3
(CO
3
)
2
(OH)
2
) and cerussite (PbCO
3
). The reaction rate
depends on variables such as climate and soil pH. As an example,
elemental Pb shot will decompose on the order of 100 to 300 years
in a Danish climate (Jørgensen and Willems, 1987). These second-
ary minerals form as crusts on the exterior of the shot and are
relatively soluble. When they dissolve, the resulting Pb2+ ions can
enter into solution or react to form soluble organometallic
compounds (e.g., Cao et al., 2003). As a result, Pb could have
entered Fall Creek as either suspended particles of secondary Pb
minerals, as dissolved Pb2+ or Pb-organometallic compounds, or as
Pb2+ adsorbed to clay or organic particles.
The lack of a clearly elevated lead signal in the down-
stream
sediments can also be attributed to dilution of sediments from
the upstream watershed. The Fall Creek watershed is approximately
340 km2 so the contaminated site is only a fraction of the potentially
erodible surface. However, there is a dam that creates Beebe Pond
(see site 83 in Figure 1) and an impoundment upstream in Forest
Home used by the Cornell water treatment plant. Both of these
structures retain sediment and, therefore, the suspended load at
Ithaca Falls is reduced compared to what it would be if the stream
flowed without impediment. In order to clearly calculate the effect
of dilution, the suspended load at Ithaca Falls would need to be
measured, particularly during high stream discharge.
CONCLUSIONS
In this study, sediment samples taken downstream from the
Pb-contaminated Ithaca Gun site were analyzed for heavy metals.
Our results show that recent sediments in the stream channel and
delta in Cayuga Lake do not have Pb concentrations above ambient
levels. Some of the older delta sediments and some samples from
the floodplain in Jetty Woods have elevated Pb which are close to
the ambient lake sediment composition prior to the 1970s. Positive
correlations between the trace metals do not indicate is that the Pb in
the sediments is derived from a preferentially enriched in Pb point
source, like the Ithaca Gun site. Based on New York State threshold
levels, none of the Pb concentrations pose a threat to human health.
ACKNOWLEDGEMENTS
The authors would like to thank Di Keller and Dave Linsley for their
assistance with the trace element analysis. Diana Sinton helped
create our maps and two anonymous reviewers gave helpful
comments. This work was supported by the Ithaca College Ford
Science Fund.
REFERENCES CITED
Cao, X., Ma, L.Q., Chen, M., Hardison Jr., D.W., and Harris, W.G.,
2003, Lead transformation and distribution in the soils of
shooting ranges in Florida, USA: Science of The Total
Environment, v. 307, p. 179–189, doi: 10.1016/S0048-
9697(02)00543-0.
Darling, C.T.R., and Thomas, V.G., 2003, The distribution of
outdoor shooting ranges in Ontario and the potential for
lead pollution of soil and water: Science of The Total
Environment, v. 313, p. 235–243, doi: 10.1016/S0048-
9697(03)00328-0.
Dong, D., Derry, L.A., and Lion, L.W., 2003, Pb scavenging from a
freshwater lake by Mn oxides in heterogeneous surface
coating materials: Water Research, v. 37, p. 1662–1666,
doi: 10.1016/S0043-1354(02)00556-0.
Graney, J.R., Halliday, A.N., Keeler, G.J., Nriagu, J.O., Robbins,
J.A., and Norton, S.A., 1995, Isotopic record of lead
pollution in lake sediments from the northeastern United
States: Geochimica et Cosmochimica Acta, v. 59, p.
1715–1728, doi: 10.1016/0016-7037(95)00077-D.
Harper, M.P., and Peckarsky, B.L., 2005, Effects of pulsed and
pressed disturbances on the benthic invertebrate commu
nity following a coal spill in a small stream in northeast
ern USA: Hydrobiologia, v. 544, p. 241–247, doi:
10.1007/s10750-005-0862-5.
Hem, J.D., 1976, Geochemical controls on lead concentrations in
stream water and sediments: Geochimica et Cosmochi
mica Acta, v. 40, p. 599–609, doi: 10.1016/0016-
7037(76)90106-X.
Horowitz, A.J., and Elrick, K.A., 1987, The relation of stream
sediment surface area, grain size and composition to trace
element chemistry: Applied Geochemistry, v. 2, p.
437–451, doi: 10.1016/0883-2927(87)90027-8.
7
www.northeasterngeoscience.org Volume 33 (2015)
Northeastern Geoscience
Table 3. NYSDEC Sediment Guidance Values for Freshwater Sediment and Soil
Metal (ppm) Class A Class B Class C Soil -
Residential
i
Soil –
Ecological
ii
Chromium < 43 43 – 110 > 110 180
iii
41
Copper < 32 32 – 150 > 150 270 50
Lead < 36 36 – 130 > 130 400 63
Nickel < 23 23 – 49 > 49 310 30
Zinc < 120 120 – 460 > 460 10,000 109
i - These values are from NYS environmental law chapter IV subpart 375-6 for restricted-residential levels for
public protection in residential areas.
ii - These values are from NYS environmental law chapter IV subpart 375-6 for protection of ecological resources.
Values are for trivalent Cr .
+3
Hunter, L.C., 1979, A History of Industrial Power in the United
States 1780-1930: Volume One: Waterpower in the
Century of the Steam Engine: University Press of
Virginia, 606 p.
Jørgensen, S.S., and Willems, M., 1987, The Fate of Lead in Soils:
The Transformation of Lead Pellets in Shooting-Range
Soils: Ambio, v. 16, p. 11–15.
Lin, Z., Comet, B., Qvarfort, U., and Herbert, R., 1995, The chemi
cal and mineralogical behaviour of Pb in shooting range
soils from central Sweden: Environmental Pollution, v.
89, p. 303–309, doi: 10.1016/0269-7491(94)00068-O.
Lollar, B.S., 2005, Environmental Geochemistry: Elsevier, 652 p.
Norrish, K., and Chappell, B.W., 1977, X-ray fluorescence
spectrometry, in Physical methods in determinative
mineralogy, Academic Press London, p. 201–272.
NYSDEC, 2014, Screening and Assessment of Contaminated
Sediment:.
O’Brien and Gere, 2013, Remedial Investigation Work Plan Former
Ithaca Gun Factory Site:.
Prescott Assoc., 2002, Voluntary Site Clean-up Program Investiga
tion Report, Frmer Ithaca Gun Factory Property:.
Rice, K.C., 1999, Trace-element concentrations in streambed
sediment across the conterminous United States: Environ
mental Science & Technology, v. 33, p. 2499–2504.
Salim, R., 1983, Adsorption of lead on the suspended particles of
river water: Water Research, v. 17, p. 423–429.
Schumacher, B.A., 2002, Methods for the determination of total
organic carbon (TOC) in soils and sediments: US EPA,
1–23 p.
Snyder, W.C., 1991, The Ithaca Gun Company: From the Begin
ning: Cook and Uline Pub., 246 p.
Stearns & Wheeler, 1997, Draft evironmental impact statement :
lake source cooling; Cornell University:.
Steinberg, T., 2003, Nature Incorporated: Industrialization and the
Waters of New England: Cambridge University Press,
310 p.
Ver Straeten, C.A., Brett, C.E., and Sageman, B.B., 2011, Mudrock
sequence stratigraphy: A multi-proxy (sedimentological,
paleobiological and geochemical) approach, Devonian
Appalachian Basin: Sea-level cyclicity, climate change,
and bioevents in Middle Devonian marine and terrestrial
environments, v. 304, p. 54–73, doi:
10.1016/j.palaeo.2010.10.010.
Vantelon, D., Lanzirotti, A., Scheinost, A.C., and Kretzschmar, R.,
2005, Spatial Distribution and Speciation of Lead around
Corroding Bullets in a Shooting Range Soil Studied by
Micro-X-ray Fluorescence and Absorption Spectroscopy:
Environmental Science & Technology, v. 39, p. 4808–
4815, doi: 10.1021/es0482740.
White, W.M., 2001, Lead isotope ratios and concentrations in
Cayuga Lake sediment as chronostratographic tools, in A
symposium on environmental research in the Cayuga
Lake Watershed, Cornell University.
Wilson, A.R., Lion, L.W., Nelson, Y.M., Shuler, M.L., and Ghiorse,
W.C., 2001, The Effects of pH and Surface Composition
on Pb Adsorption to Natural Freshwater Biofilms:
Environmental Science & Technology, v. 35, p. 3182–
3189, doi: 10.1021/es001701z.
Yager, R., 2001, Estimating sedimentation rates in Cayuga Lake,
New York from sediment profiles of 137Cs and 210Pb
activity, in A symposium on environmental research in the
Cayuga Lake Watershed, Cornell University, p. 78– 102.
8
www.northeasterngeoscience.org Volume 33 (2015)
Northeastern Geoscience
Sinton, Hill & Zerbian
Evaluation of Heavy Metals from Ithaca Northeastern Geoscience
9
www.northeasterngeoscience.org Volume 33 (2015)
Table 2: Calculated Total Organic Carbon and Analyzed Trace Elements from Fall Creek Delta Sediments
Sample ID texture
Fe
(wt. %)
TOC
(%) Ba Ce Cr Cu Ni Rb Sr Y Zn Zr Pb
Surface Samples
62 silt/clay 2.74 0.16 347 60 55 36 40 87 114 31 96 397 31
76 silt/clay 3.17 1.58 386 61 60 31 41 91 123 31 92 337 32
77 silt/clay 2.86 2.61 344 59 51 37 40 86 117 31 96 370 29
78 silt/clay 2.91 2.52 362 55 57 33 40 83 121 32 96 392 28
79 silt/clay 2.94 3.91 354 61 53 30 40 86 119 31 96 433 37
79R silt/clay 2.90 3.55 355 55 56 33 39 81 123 32 96 447 29
80 sand 1.12 336 58 52 22 38 81 122 25 70 318 23
81 sand 2.44 0.75 347 52 45 21 34 73 129 26 67 397 21
Core Samples
64
0-10 cm silt/clay 2.55 1.56 336 52 46 28 40 81 112 29 81 396 36
10-20 cm silt/clay 2.60 0.15 336 60 53 29 38 83 103 33 82 453 37
20-30 cm silt/clay 2.54 1.93 308 50 46 27 39 76 117 28 80 430 26
71
0-7 cm silt/clay 2.90 2.86 369 73 60 42 74 102 95 35 107 418 43
7-14 cm silt/clay 3.47 0.16 447 76 77 43 62 113 97 38 127 381 46
14-21 cm silt/clay 3.39 3.15 443 79 85 45 67 118 99 37 128 379 49
73
10-20 cm silt/clay 3.74 5.88 496 92 83 37 54 121 100 45 112 328 40
20-30 cm silt/clay 2.58 4.68 403 74 57 24 39 83 95 35 73 343 21
30-40 cm
sandy
silt 1.20 3.65 321 44 31 18 28 53 99 23 43 328 12
40-50 cm sand 1.02 1.83 364 32 20 11 27 48 130 16 35 223 8
50-60 cm sand 0.71 1.53 295 37 13 16 21 40 114 15 29 199 10
60-70 cm sand 1.04 0.04 347 38 19 13 29 51 117 16 33 179 12
70-80 cm sand 1.03 1.70 358 37 20 15 28 50 126 16 32 197 11
80-90 cm sand 1.00 0.46 338 36 17 14 27 50 125 15 33 293 10
90-94 cm sand 0.52 7.77 296 16 11 13 18 35 110 13 26 192 8
Northeastern Geoscience
10
www.northeasterngeoscience.org Volume 33 (2015)
Sample ID texture
Fe
(wt. %)
TOC
(%) Ba Ce Cr Cu Ni Rb Sr Y Zn Zr Pb
74
0-10 cm silt/clay 3.94 9.12 508 92 86 58 58 151 109 45 193 346 81
10-20 cm silt/clay 4.48 6.49 519 93 86 50 58 137 100 43 144 315 69
30-40 cm sandy silt 1.56 3.82 328 56 30 20 34 62 106 25 50 350 13
40-50 cm sandy silt 1.40 2.73 351 34 35 13 29 56 118 22 45 365 10
50-60 cm sand 1.64 0.85 282 43 33 12 34 58 79 21 46 293 10
60-70 cm sand 1.66 1.02 352 44 49 16 36 62 105 20 48 249 13
FCD1
0-3 cm silt/clay 1.95 1.10 284 44 49 15 22 58 142 31 59 415 22
3-6 cm silt/clay 1.92 0.64 276 38 47 17 21 59 121 31 61 418 20
6-9 cm silt/clay 1.76 0.29 275 47 42 13 20 57 128 30 57 488 19
9-14 cm silt/clay 2.18 0.58 327 41 48 21 25 69 117 31 73 394 26
14-19 cm silt/clay 2.45 1.15 337 57 59 23 27 74 118 36 86 563 38
19-24 cm silt/clay 2.22 0.29 324 53 57 21 26 73 118 34 89 530 37
24-28 cm silt/clay 2.15 0.74 300 48 55 16 25 70 115 36 76 686 33
FCD4
0-3 cm silt/clay 2.45 1.14 NA NA NA NA NA NA NA NA NA NA NA
3-6 cm silt/clay 2.67 1.93 313 63 60 25 29 72 136 35 84 449 32
6-9 cm silt/clay 2.77 1.05 352 52 71 27 31 84 120 35 104 370 44
9-14 cm silt/clay 2.13 1.02 347 64 67 30 32 87 124 36 100 353 49
14-19 cm sandy silt 2.38 1.79 305 57 53 17 26 72 151 35 73 404 31
19-24 cm sandy silt 2.25 2.70 380 52 48 21 27 80 131 33 75 407 33
24-29 cm silt/clay 1.92 1.44 347 64 58 20 27 78 125 35 67 450 30
29-34 cm silt/clay 1.87 1.79 313 50 43 14 25 70 126 32 59 418 23
34-38 cm silt/clay 2.31 1.91 322 36 38 13 22 66 108 31 54 442 20
Table 2 (cont.)
Sinton, Hill & Zerbian
Evaluation of Heavy Metals from Ithaca Northeastern Geoscience
11
www.northeasterngeoscience.org Volume 33 (2015)
Sample ID texture
Fe
(wt. %)
TOC
(%) Ba Ce Cr Cu Ni Rb Sr Y Zn Zr Pb
FCD6
0-3 cm silt/clay 2.99 2.61 326 56 53 17 27 77 103 36 64 413 22
3-6 cm silt/clay 2.96 2.76 365 58 63 31 33 85 144 35 105 295 36
6-9 cm silt/clay 2.81 2.47 362 59 64 31 33 84 143 34 105 287 36
9-14 cm silt/clay 2.72 1.57 351 56 62 28 31 81 163 34 101 312 35
14-19 cm silt/clay 2.12 332 59 59 29 30 78 137 35 96 317 38
19-24 cm silt/clay 3.07 2.10 379 71 64 37 37 93 122 37 122 335 52
24-29 cm silt/clay 2.93 1.95 366 69 65 28 35 93 141 36 107 309 49
29-34 cm silt/clay 2.91 2.45 372 66 63 31 35 93 153 36 108 325 50
34-39 cm silt/clay 2.96 2.10 383 64 62 30 34 95 150 37 96 300 48
39-44 cm silt/clay 2.21 1.51 307 49 57 19 26 73 169 32 70 328 37
44-49 cm silt/clay 3.25 3.12 406 69 71 31 39 102 122 41 97 305 48
49-52 cm silt/clay 3.47 2.99 411 77 77 33 41 108 119 41 103 288 47
Control Sites
83 silt/clay 2.56 364 63 56 30 40 81 116 33 89 398 24
CID2
0-3 cm silt/clay 2.72 2.08 403 61 58 27 30 84 145 33 92 395 36
3-6 cm silt/clay 2.26 2.16 346 54 52 20 23 73 155 31 68 337 33
N
A = not analyzed
