<>EPA
United States
Environmental Protection
Agency
Great Lakes National
Program Office
536 South Clark Street
Chicago. Illinois 60605
EPA-905/4-79-029-F
Volume 6
The IJC Menomonee
River Watershed Study
Dispersibility of Soils and
Elemental Composition of
Soils, Sediments, and
Dust and Dirt from the
Menomonee River Watershed
Menomonee River
-------
FOREWORD
The Environmental Protection Agency was established to coordinate adminis-
tration of the major Federal programs designed to protect the quality of our
environment.
An important part of the Agency's effort involves the search for information
about environmental problems, management techniques, and new technologies
through which optimum use of the nation's land and water resources can be
assured and the threat pollution poses to the welfare of the American people
can be minimized.
The Great Lakes National Program Office (GLNPO) of the U.S. EPA, was
established in Region V, Chicago to provide a specific focus on the water
quality concerns of the Great Lakes. GLNPO also provides funding and
personnel support to the International Joint Commission activities under
the U.S.- Canada Great Lakes Water Quality Agreement.
Several land use water quality studies have been funded to support the
pollution from Land Use Activities Reference Group (PLUARG) under the
Agreement to address specific objectives related to land use pollution to
the Great Lakes. This report describes some of the work supported by this
Office to carry out PLUARG study objectives.
We hope that the information and data contained herein will help planners
and managers of pollution control agencies make better decisions for
carrying forward their pollution control responsibilities.
Madonna F. McGrath
Director
Great Lakes National Program Office
-------
EPA-905/4-79-029F
December 1979
Dispersibility of Soils and Elemental
Composition of Soils, Sediments and Dust
and Dirt From The Menomonee River Watershed
VOLUME 6
by
A. Dong
G.Chesters
G.V. Simsiman
Wisconsin Water Resources Center
University of Wisconsin-Madison
Madison, Wisconsin
for
U.S. Environmental Protection Agency
Chicago, Illinois
Grant Number R005142
Grants Officer
Ralph G. Christensen
This study, funded by a Great Lakes Program grant from the U.S. EPA,
was conducted as part of the TASK C-Pilot Watershed Program for the International
Joint Commission's Reference Group on Pollution from Land Use Activities.
Great Lakes National Program Office
U.S. Environmental Protection Agency, Region V
536 South Clark Street, Room 932
Chicago, Illinois 60605
U.S. Environmental Protection
-------
DISCLAIMER
This report has been reviewed by the Great Lakes National Program Office
of the U.S. Environmental Protection Agency, Region V Chicago, and approved
for publication. Mention of trade names of commercial products does not
constitute endorsement or recommendation for use.
11
-------
PREFACE
Elemental composition (Al, Cd, Cr, Cu, Fe, Mn, Ni, P, Pb and Zn) in the
sand-, silt- and clay-sized fractions of major soil types, bottom sediments,
suspended sediments and dust and dirt samples from the Menomonee River Basin,
Wisconsin, were analyzed. Sediments and dust and dirt samples with elemental
compositions greater than levels found in the major soil types were suspected
of receiving additional inputs of pollutants from sources other than soils.
Locations of possible pollutant input into the Menomonee River were identified
by comparing the elemental composition of the clay-sized fraction of bottom
sediments. A method for estimating the dispersibility of soils was
developed. Soil samples were dispersed by shaking with water to simulate
natural water erosion conditions and by ultrasound to provide complete
dispersion. The ratio of the amount of clay-sized particles dislodged by
shaking to the amount obtained by ultrasound treatment measured the
dispersibility of soils, an indirect measurement of soil erosion potential.
iii
-------
CONTENTS
Title Page i
Disclaimer ii
Preface iii
Contents iv
Figures v
Tables vi
1. Introduction 1
2. Conclusions 3
3. Materials and Methods 5
Soil Sampling 5
Sediment Sampling 5
Bottom sediments 5
Suspended sediments 8
Urban Street Dust and Dirt Sampling 8
Ultrasound Dispersion and Fractionation of Soil and
Urban Street Dust and Dirt Samples 9
"Dispersion" and Fractionation of Soil by Shaking
in Water 9
Ultrasound Dispersion and Fractionation of Sediments 10
Bottom sediments 10
Suspended sediments 10
Digestion of Particle-Size Fractions 10
Elemental Analysis 11
4. Results and Discussion 12
Particle-Size Distribution of Samples 12
Phosphorus in Soils 15
Phosphorus in Bottom Sediments 15
Phosphorus in Suspended Sediments 18
Phosphorus in Urban Street Dust and Dirt 21
Metals in Soil 21
Metals in Bottom Sediments 27
Metals in Dousman Ditch Bottom Sediments 30
Metals in Suspended Sediments 31
Metals in Urban Street Dust and Dirt 35
Pollutional Classification of Sediments 37
Dispersibility of Soils 42
Extractability of Metals and Phosphorus 45
Correlation Analysis 49
References • • 53
iv
-------
FIGURES
Number
Sampling sites of soils, bottom sediment and suspended
sediment within the Menomonee River Watershed 6
Sampling sites of Dousman Ditch bottom sediments 7
Simple correlations between concentrations of lead and
cadmium, zinc and cadmium and zinc and lead for the
clay fractions of suspended and bottom sediments of
the Menomonee River 51
v
-------
TABLES
Number
1 Particle-size distribution and phosphorus concentrations
in each particle-size fraction of soils, sediments and
urban street dust and dirt in the Menomonee River
Watershed 13,14
2 Flow and time of sampling during collection of suspended
sediment samples on June 30 and July 18, 1977 storm
events 19
3 Antecedent rainfall in solid loading of urban street dust
and dirt samples in 1977 22
4 Lead, cadmium, zinc and copper concentrations in each
particle-size fraction of soils, sediments and urban
street dust and dirt in the Menomonee River Watershed 23,24
5 Aluminum, iron, manganese, chromium and nickel
concentrations in each particle-size fraction of soils,
sediments and urban street dust and dirt in the
Menomonee River Watershed 25,26
6 Particle-size distribution and element concentrations
in each size fraction of Dousman Ditch bottom sediment 32
7 Distribution of elements in each particle-size fraction
of urban street dust and dirt samples 38
8 Comparison of total elemental composition of soils,
sediments and urban street dust and dirt with U.S.-EPA
pollutional classification guideline for unfractionated
dredge spoils 39,40
9 "Dispersibility," by shaking, of soils in the Menomonee
River Watershed 43
10 Dispersion ration of the clay-sized fraction (shaking:
ultrasound) 44
11 Linear, inverse and log correlation coefficients
between soil organic carbon content and clay-sized
fraction dispersion ratio 44
vi
-------
Number Page
12 Elements extractable from soils by end-over-end shaking
and utlrasound treatment 47
13 Extractability (%) of elements contained in soils and
bottom sediments as affected by different dispersion
techniques • 48
14 Increasing order of extractability of various elements
using different dispersion techniques 50
VI1
-------
1. INTRODUCTION
Concern for the effects of various land use activities on Great Lakes
water quality has prompted the governments of the United States and Canada,
under the Great Lakes Water Quality Agreement of April 15, 1972, to direct the
International Joint Commission to conduct studies of the impact of land use
activities on the water quality of the Great Lakes Basin and to recommend
remedial measures for maintaining or improving Great Lakes water quality (1).
The International Reference Group on Great Lakes Pollution from Land Use
Activities has implicated sediments and several elements including phosphorus
(P), lead (Pb), cadmium (Cd), zinc (Zn) and copper (Cu) as contaminants of the
Great Lakes. Sediments pose a serious problem to water quality and
navigation. Sediments are potential carriers of pollutants—nutrients, toxic
organic compounds and toxic metals—e.g., 40 to 80% of the total P load to the
Great Lakes is associated with sediment (1). Accumulation of sediments in
harbors of the Great Lakes requires dredging and over $100 million are spent
annually to dredge Great Lakes harbors (1). The sediment loading to the Great
Lakes from tributaries is about 11 million Tonnes/yr. Such large amounts of
sediment reaching the Great Lakes annually indicates the need for some control
measures to minimize sediment transport.
The Menomonee River Watershed was one of the six study sites in the Great
Lakes basin used to evaluate the magnitude of non-point source pollution
reaching the Great Lakes. The 35,200 ha Menomonee River Watershed is located
in the southeastern corner of Wisconsin and discharges to Lake Michigan at the
Milwaukee Harbor and carries an average sediment loading of about 10,000
Tonnes/yr. The lower portion of the Watershed—including part of the city of
Milwaukee—is a commercial-industrial complex with some residential land
uses. The center half of the Watershed is primarily residential with a few
light industries in the City of Wauwatosa. The upper portion is mostly
agricultural with scattered areas that are rapidly changing from rural to
urban-residential. The Menomonee River Watershed therefore provides a
situation of dynamic change in land use and serves as a focus of
investigations on the effects of urban land use on water quality.
In the past, total unfractionated sediment samples have been analyzed for
pollutant content and insufficient consideration has been given to the
importance of particle-size distribution in evaluating pollutant carrying
capacity of sediments. Using ultrasound, it is possible to maintain dispersed
particles with their associated polluants unchanged. Thus, it is possible to
determine pollutant associations with particles as they appear in the field.
Soils and sediments were dispersed by ultrasound into sand-, silt-, and clay-
sized fractions and the elemental composition of each size fraction was
determined to characterize the material and indicate possible pollutant
sources. Of particular importance is the amount and elemental composition of
the clay-sized frctions because most of the surface reactive sites of the
-------
sediments are found in this fraction. A comparison of the amount and
elemental composition of the clay-sized fractions of bottom sediments from
different locations along the Menomonee River helps to identify those places
where pollutant input occurs. The effectiveness of these depositional areas,
e.g., siltation ponds, dams or harbors, can be measured in terms of the amount
of clay-sized particles contained in the bottom sediments.
The dispersibility of soils is one of the factors that contributes to
their erosion potential. Soils that disperse readily in water under natural
conditions are considered to be more hazardous than soils that remain in an
aggregated form, because dispersed fine particles stay in suspension longer
and are transported more readily over land surfcaes to streams, rivers and
lakes. Once reaching the river and the final receiving waterbody, they stay
in suspension longer. This longer time that the particles remain in
suspension longer. This longer time that the particles remainin suspension
contributes to an increased biological availability of associated
pollutants. Soil particles that remain intact as aggregates when subjected to
the erosive force of water are more difficult to dislodge or transport over
land surfaces. If they are transported into streams, the stable aggregates
and their associated pollutants tend to deposit along rivers more quickly than
dispersed soil particles. Dispersion of soils in the laboratory can be
accomplished by shaking the soil in water to simulate a natural erosion
condition or by ultrasound to provide complete dispersion of the soils. The
ratio of the amount of clay-sized particles dislodged by shaking in water to
the amount obtained after ultrasound treatment provides a means of evaluating
soil dispersibility.
As a portion of the Menomonee River Pilot Watershed study, this
investigation was designed to measure the water dispersibility of the major
soil types in the Watershed and the chemical and mechanical properties of the
soils (seven soil types), bottom sediments (24 sites), suspended sediments (12
sites) and urban street dust and dirt (two sites). Furthermore, attempts were
made to identify sources of pollutants based on composition of clay-sized
fractions and, where appropriate, to identify alternative measures for
pollutant control.
-------
2. CONCLUSIONS
Sediments are important carriers of pollutants. The Cd, Cr, Cu, Fe, Mn,
Ni, Zn and P concentrations in sand-, silt- and clay-sized fractions of soils,
bottom sediments, suspended sediments and urban street dust and dirt were
analyzed. Sediments and dust and dirt samples with elemental composition
greater than the levels found in the major soil types of the watershed were
suspected of receiving additional inputs of pollutants from sources other than
soils.
The Cd, Pb and Zn concentrations in some bottom- and suspended-sediment
samples were found to be higher than in soils. Concentrations of these
elements were correlated significantly with each other in the clay-sized
fraction of sediments but not in soils. This indicates that soils were not
the primary source of these metals and other sources such as vehicular
emission and atmospheric fallout were major inputs.
Locations of pollutant input to the Menomonee River can be identified by
comparing elemental composition of the clay-sized fractions of bottom
sediments collected at different locations. Total elemental composition of
unfractionated bottom sediment samples were less precise in identifying the
location of pollutant input.
In an agricultural land-use area, bottom sediment samples (Dretzka Creek)
with P levels greater than the soil level but without a corresponding increase
in metal composition was found. In the urban area, a sediment sampling site
located below the outfall of a sanitary treatment plant (STP) with secondary
treatment capability (Nor-X-way-B) showed an increase in P as well as Cd, Cr,
Cu, Ni, Pb and Zn levels. Clay fractions of bottom sediments from sites
located below the outfall of STPs with tertiary treatment capability (River
Lane and Lily Creek) showed lower levels of P as well as metals than those
found at the sampling site located below an STP with secondary treatment
capability. Apparently the waste water treatment for the removal of P also
removed metals for the effluent.
The average P, Pb and Cd concentrations in suspended sediment samples of
the Menomonee River collected during storm events were: 1840 pg/g P, 350 yg/g
Pb and 1.9 yg/g Cd in the clay-sized fractions; 780 yg/g P, 180 yg/g Pb and
0.48 yg/g Cd in the silt-sized fractions; and was calculated to contain 1620
yg/g P, 290 yg/g Pb and 1.4 yg/g Cd in the unfractionated sample. The average
annual storm event loadings from suspended sediments in the Menomonee River to
Lake Michigan was calculated to be 16,200 kg/yr P, 3000 kg/yr Pb and 15 kg/yr
Cd with about 90% of the P, Pb and Cd in the clay-sized fraction.
The Al, Fe and Mn concentrations in the clay-sized fraction of urban
street dust and dirt samples were found to be lower than in the major mineral
soil types of the watershed, while Cd, Cr, Cu, Ni, Pb and Zn levels were
-------
higher. Distribution of elements into the particle-size fractions were found
to divide into two main groups. One group had 78 to 87% of the metals in the
sand fraction (Cr, Cu, Fe, Mn and Ni); and the other with 41 to 58% in the
sand fraction (Al, Cd, P and Pb); while Zn was intermediate between these two
groups (70% in the sand fraction).
The Cr, Cu, Fe and Ni concentrations in the coarse particles of the dust
and dirt samples occasionally were found to be nearly equal to the
concentrations in fine particles (sand vs silt and silt vs clay-sized
particles). Similarly, Ni concentrations in the silt-sized fractions of
suspended sediments were occasionally found to be nearly equal to the
concentration in the clay-sized fraction. This may result from their presence
in large particles such as metal chips from abrasion of vehicular parts or
from disintegration of impervious surfaces.
Soil dispersibility—a contributing factor to soil erosion and sediment
loading to waterways—was evaluated for the Menomonee River Basin,
Wisconsin. Soil samples were dispersed by shaking with water to simulate
natural water erosion conditions and by ultrasound to provide complete
dispersion. The shaking treatment consisted of agitating a 1:10 w:v
soilrwater mixture for 0.5 to 128 hr. The ratio of the amount of clay-sized
particles dislodged by shaking to the amount obtained by ultrasound treatment
measured the dispersibility of soils. Organic carbon content (0.5 to 44%) was
best correlated with soil dispersion ratio in a negative inverse
relationship. If the 4 hr shaking treatment simulates the onset of soil
erosion conditions in the field, as much as 90% of the primary clay-sized
particles remain in silt-sized or larger aggregates during the overland
transport. Thus, the amount of clay reaching the waterways can be controlled
by retaining aggregates containing a high amount of clay-sized particles.
Resuspension of bottom sediments as simulated by end-over-end shaking (1
to 128 hr) was found to desorb about 0.06% Pb and 0.7% Cd of the total in the
solid phase. Under extreme agitation as simulated by 15 min of ultrasound
treatment, the desorption was 1.5% Pb and 2.0% Cd of the total in the solid
phase. Thus, resuspension of bottom sediments to the overlying water possibly
permits desorption of elements from the solid surfaces.
-------
3. MATERIALS AND METHODS
Soil Sampling
Of the total area of the Menomonee River Watershed, 85% was mapped as
soil; the remaining 15% was highly urbanized. Seven major soil types in the
Watershed—representing 74% of the mapped soils—were selected: Ozaukee sil
(represented 41%), Mequon sil (16%), Hochheim sil (4.0%), Pella sil (3.5%),
Theresa sil (3.5%), Ashkum sicl (2.6%) and Houghton muck (3.0%). The sampling
locations are shown in Fig. 1 and identified from a computerized soil survey
map of the Watershed prepared by the Southeastern Wisconsin Regional Planning
Commission (2). The mineral soils were sampled in the northern, rural part of
the Watershed, which is more exposed to wind and water erosion than the
highly-urbanized southern portion. Soil samples (top 20 to 40 cm) were taken
at least 50 m away from the nearest road to minimize automobile exhaust
contamination. The soil samples were dried in a forced-air dryer at 70 to
80°C, pulverized in a porcelain mortar, passed through a 10-mesh polypropylene
sieve and stored in glass jars for later use.
Sediment Sampling
Bottom sediments
Whenever possible, bottom sediment sampling sites (Figs. 1 and 2) were
selected near mainstem river monitoring stations and at sites subject to
significant inputs of pollutants of anthropogenic origin. No samples were
taken at Honey or Underwood Creeks because it was thought that the severely-
modified channels would prevent the collection of representative bottom
sediments. The River Lane, Nor-X-way B, Donges Bay Road, Capitol Drive and
70th Street sites were located on bedrock bottoms and samples were obtained by
searching for places where sediments had been deposited. The remaining sites
(County Q at the Milwaukee-Ozaukee County line, near intersection of Roads F
and B, Appleton Avenue, Friestad Road, Maple Road, Menomonee Falls, Menomonee
Falls Dam, Nor-X-way A, Lily Creek, Dretzka Creek, 124th Street and Dousman
Ditch site Nos. 1 to 6) were located in reaches where the river flow is low or
where the river widens so that flow decreases sufficiently to permit
deposition of fine-textured sediments. However, in spite of the selective
method of sampling for bottom sediments, meaningful data can be obtained if
the samples are fractionated and the elemental composition of each particle-
size fraction is determined. Clay-sized particles are the most important
because of their mobility and ease of transportation over considerable
distances. Except for the Nor-X-way A and Dousman Ditch site Nos. 1 to 6, all
samples were taken from the mainstem of the river. The Nor-X-way A sample was
taken before the confluence of an intermittent creek (Northern Crossway Creek)
5
-------
463001
673001
• Soils
A Bottom sediments
* Suspended sediments
0 1
I I
- Miles ,
024
I I I i I
Km
Fig. 1. Sampling sites of soils, bottom sediment and suspended sediment
within the Menomonee River Watershed.
-------
A Bottom sediments
• Brookfield station 683089
•*• Direction of drainage
0 200 400 600
i i i i
Feet
0 100 200
i i i
Meters
Bluemound Road
6
•A
T3
n)
Fig. 2. Sampling sites of Dousman Ditch bottom sediments.
-------
with the Menomonee River which drains part of Highway U.S. 41/45 and an area
which constitutes about 3.9% of the area of the Menomonee River Watershed
(3). Dousman Ditch (sites 1 to 6) is a natural drainage ditch which connects
the storm sewer outfall of Brookfield Shopping Center to Underwood Creek (Fig.
2).
Most of the river sites were either shallow enough to approach the
midstream with rubber waders or had sufficient ice to permit walking to
midstream. The top 1 to 15 cm of sediments were sampled using an acrylic tube
(4.5 cm I.D. x 120 cm long) to avoid metal contamination. A No. 11 rubber
stopper was used to seal one end after inserting the tube into the river
bottom sediments. A ring clamp was attached to the upper portion of the
acrylic tube to facilitate gripping the tube. However, at two sites—124th
Street and the Harbor—where the water was too deep to use the acrylic tube, a
brass Eckman dredge was used and the samples were taken from the center
portion of the dredge thereby avoiding the metallic edge. The samples were
stored in a frozen condition in polypropylene bottles (1 L) until used.
Suspended sediments
Runoff water samples containing suspended sediments from two storm events
(June 30 and July 18, 1977) were collected by the U.S. Geological Survey
(USGS) using depth-integrated samples, and the samples were stored in plastic
containers. The June 30 samples included only 3 sites (Falk Corp., Honey
Creek and 70th Street), while the July 18 sampling included all mainstem
monitoring stations (Fig. 1).
Suspended sediments were separated from river water by centrifugation in
polypropylene bottles (250 ml) for 30 min at 750.0 rpm using a Sorval RF-2
refrigerated centrifuge fitted with a GSA head. Temperature was maintained at
20° C during centrifugation, and those samples which could not be centrifuged
immediately were frozen. To further concentrate the sediment samples, they
were subjected to centrifugation at 10,000 rpm for 10 min using 50 ml Oakridge
polypropylene tubes; the frozen samples were stored in these tubes.
Urban Street Dust and Dirt Sampling
Urban street dust and dirt samples were collected by the Wisconsin
Department of Natural Resources (WDNR) Southeastern District from two
locations—13th Street Bridge and 91st Street—during September and October
1977. The sampling sites—located in front of fire hydrants to avoid possible
interference from parked cars during sample collection—were in an area not
sewpt by municipl street sweepers. Sampling sites—0.91 m wide and 4.57 m
along the street curb—were carefully swept three times with a portable
electric vacuum cleaner and the collected samples were weighed. Whenever
possible, samples were collected immediately after a rainstorm and 2 to 4 days
thereafter. It should be noted that rainfall recorded at Mitchell Field
Airport, Milwaukee, Wisconsin, was not necessarily the same as that occurring
at the sampling sites.
8
-------
Ultrasound Dispersion and Fractionation of Soil
and Urban Street Dust and Dirt Samples
Samples (3 g) were placed in pre-weighed glass beakers (30 ml) and soaked
overnight with enough water to submerge the sample. Although Genrich and
Bremner (4) indicated that overnight soaking was unnecessary, this step was
included as a precautionary measure against incomplete dispersion. After
soaking the samples overnight, more water was added to approximately 30 ml and
the beakers were placed in an ice bath for ultrasound treatment. The
ultrasound treatment continued for 15 min at 100 to 125 watts using a Braun
model 1510 ultrasonic generator with a standard probe. Preliminarly
dispersion tests indicate that 15 min is sufficient to disperse the soil
particles. After ultrasound treatment, the silt- and clay-sized fractions
were decanted along with some sand into polycarbonate tubes (110 x 15 mm) for
centrifugation. The sand fraction remaining in the beaker was rinsed several
times to ensure complete removal of silt- and clay-sized particles. The silt-
plus clay-sized fractions in the centrifuge tubes were resuspended and
centrifuged at appropriate speed to separate the silt and sand from the
clay. The silt and sand fractions were rinsed several times with distilled
water, resuspended and centrifuged until the water became clear. The clay
suspension was collected in 50-ml Oakridge centrifuge tubes. The particle-
size fractions chosen were: Sand—2000 to 62 pm, silt—62 to 4 um and clay—
<4 pm as used by the USGS. Centrifugation time was calculated using Stoke's
Law assuming a particle specific gravity of p = 2.65 for soil and p = 1 for
water. For clay-sized particles «4 um), centrifugation time was 99 sec at
600 rpm. Usually the first two centrifugations were conducted for longer time
periods and/or at higher speeds than the computed values to compensate for the
increase in viscosity due to the larger amounts of clay in suspension.
Additional centrifugations (1 to 3 times) were performed at the calculated
time and speed. Clay suspensions were concentrated further by centrifugation
at 2.10,000 rpm for _XLO min in 50 ml Oakridge centrifuge tubes and transferred
to pre-weighed beakers (30 ml) prior to drying and weighing. Sand was
separated from silt by resuspending the sample and allowing for gravity
settling of the sand at 26 sec for a 10 cm fall. Usually the first two
partitionings of silt from sand were conducted with <10 cm fall and longer
time periods. Silt particles were transferred to pre—weighed beakers (30 ml)
to be dried and weighed. The remaining sand was transferred to the original
pre-weighed beakers used in the ultrasound treatment which contained sand not
initially decanted. Each fraction was oven-dried for 2.24 hr at 110° C, cooled
in a dessicator and weighed. Supernates arising from fractionation of soils
were collected for analysis of elements in solution. In some samples, the
supernates were filtered through a 0.4 um polycarbonate membrane filter using
an all-plastic Nucleopore filter holder and were designated as filtered
supernates. The second set was designated as unfiltered supernates.
"Dispersion" and Fractionation of Soil by Shaking in Water
Soil samples (1.5 g) in 15 ml deionized, distilled water were placed in
glass centrifuge tubes (150 x 15 mm) with screw caps and shaken on an end-
over-end shaker for 1, 4, 16, 32, 64 and 128 hr at 20 to 24 cycles/min using a
Scientific Industries Inc. Multipurpose rotator. After "dispersion," the
9
-------
samples were fractionated by centrifugation and gravity settling as described
for ultrasound-dispersed soil samples. The supernates obtained after
centrifugation were collected and stored. The samples—shaken for 1 to 16
hr—were left unfiltered, while those shaken for 32 to 128 hr were filtered
through 0.4 ym membrane filters.
Ultrasound Dispersion and Fractionation of Sediments
Bottom sediments
Bottom sediments were thawed, mixed and a subsample (2 to 12 g) placed in
a pre-weighed glass beaker (30 ml); a larger sample size was used if the clay
content of the sediment was known to be low. The ultrasound treatment and
fractionation procedure were the same as those for soil samples. Fractionated
samples were oven-dried and weighed. All supernates arising from
fractionation were collected and stored for analysis.
Suspended sediments
Frozen suspended sediment samples were thawed and subjected to ultrasound
treatment at 80 to 100 watts for 3 min in a test tube to disperse the
sediments aggregated due to high ionic strength resulting from eutectic
freezing. A shorter ultrasound treatment time with lower energy was used
since the suspended sediments were not as tightly-bound as the soil samples.
Suspended sediment samples were transferred to polycarbonate centrifuge tubes
(110 x 15 mm) and fractionated by centrifugation and gravity settling as
described for soil samples. Fractionated samples were oven-dried and weighed.
Digestion of Particle-Size Fractions
After recording the weights of the dried particle-size fractions 6 to 8
ml of concentrated redistilled HNO-j (to remove metals) were added slowly to
the beakers to allow time for C02 evolution. The beakers were covered and
heated on a hot plate (5). This procedure was adequate for soil and suspended
sediment samples but not for silt- and clay-sized fractions of bottom
sediments because the organic matter was more resistant to HNCU oxidation. In
these cases, samples were treated alternately with 2 ml of redistilled 1:1 HC1
and concentrated redistilled HNOo and were heated to near dryness prior to
each new addition of acid. The digests were transferred to 30 ml plastic
bottles after gravity settling of the solids. The solids were rinsed twice
with deionized distilled water and the rinse water was added to the digest to
bring the volume to 20 to 25 ml.
The supernates collected during the fractionation procedures were placed
in 250 ml erlenmeyer flasks and digested with 3 ml concentrated redistilled
HNOg. Further digestion was continued after addition of 3 ml each of
concentrated redistilled HN03 and redistilled 1:1 HC1.
10
-------
Elemental Analysis
The metals—Cd, Cr, Cu, Ni and Pb—were analyzed by flame atomic
absorption spectroscopy (AAS) using a Perkin Elmer model 603 AA
Spectrophotometer. Sodium interference was sufficiently high—especially for
bottom sediment samples—to require the use of the deuterium arc background
corrector in a single beam mode (the Perkin Elmer model 603 does not have a
double beam option when using the background corrector). Determination of Al,
Fe, Mn and Zn concentrations in the digests required dilution which served to
diminish salt interference. These elements were analyzed by flame AAS using a
double beam mode without deuterium arc background correction on the Perkin
Elmer model 306 AA Spectrophotometer. An aliquot of the HNOo-HCl digest was
used for "total" phosphorus analysis by the molybdenum blue method (5). The
concentrations of Cd, Cr, Ni and Pb in the supernate obtained from dispersion
of soils by shaking in water were below flame AAS detection. These elements
were determined by the Perkin Elmer model 603 AA Spectrophotometer equipped
with a Perkin Elmer graphite furnace model HGA 2100A. For Cd and Cr
determination, 0.25 mg (NH^^HPO^) were added to 20 yl samples injected into
the furnace to stabilize the volatilization temperature of Cd and to reduce
the salt interference associated with Cd and Cr determination (6).
Organic carbon in soils was determined using a Leco Induction Furnace and
Carbon Analyzer after removal of inorganic carbon with H-SOo or HC1 (7).
Elemental compositions and particle-size distributions of soils,
suspended seidments and urban street dust and dirt samples were analyzed in
duplicate. Bottom sediment samples were analyzed in triplicate.
11
-------
4. RESULTS AND DISCUSSION
Particle-Size Distribution of Samples
The concentration of elements sorbed onto surfaces of soils, sediments
and urban street dust and dirt particles tend to increase with increasing
surface area. The clay-sized particles have the highest concentration of
sorbed elements due to their high surface area per unit mass. Removal of the
effect of different clay contents in soil, sediment and dust and dirt samples
on the elemental composition of the samples can only be achieved by
fractionation into sand-, silt- and clay-sized separates.
Dispersion of samples was accomplished by ultrasound treatment which
allows the organo-mineral complex and their associated elements to remain
intact. The commonly-used hydrogen perioxide-"Calgon" (Na-hexa metaphosphate)
dispersion method was avoided because I^C^ destroys the organic component and
alters the chemical composition of the dispersed particles while "Calgon"
interferes with phosphorus determinations. Fractionation into different
particle sizes was performed using the gravity settling and centrifugation
techniques. This method was preferred because it approximates the conditions
of particle transport and deposition better than sieving and filtration
techniques. The particle-size distribution of soils, sediments and urban
street dust and dirt are given in Table 1.
Using the average particle-size distribution of the six mineral soils as
a reference, it can be seen that the average particle-size distribution of
suspended sediment was skewed toward the clay-sized fraction as a result of
the deposition of coarser materials; on average, suspended sediments contained
77% clay-sized particles compared to 27% for soils. The average particle-size
distribution of the urban street dust and dirt samples was skewed toward the
coarser size particles; only 5% of the particles were in the clay-sized
range. The low amounts of fine particles on urban streets may result from
wind and runoff transport of fine particles, leaving coarser materials
behind. The average particle-size distribution of bottom sediments is
difficult to determine from the data obtained because sampling sites were
selected for high clay content. Some sampling sites were located on bedrock
bottoms and samples were collected by searching for places where sediments had
been deposited. Other sampling sites were chosen at segments of the river
where it widens and flow rate decreases sufficiently to allow deposition of
fine sediments.
Low amounts of clay-sized particles (2 to 14%) were found in the
Milwaukee Harbor bottom sediments (8), even though river flow decreases
sufficiently at the river mouth to permit sediment deposition; occasional
dredging of the Harbor has been necessary because of sediment deposition. In
contrast, bottom sediments from several upstream sites—in "pools" and behind
12
-------
Table 1. Particle-size distribution and phosphorus concentrations in each particle-size
fraction of soils, sediments and urban street dust and dirt in the Menomonee River
Watershed
Sample/ sampling
location*
Ozaukee sil (Ou)
Mequon sil (Mt)
Hochheim sil (Hm)
Ashkum sicl (As)
Pella sil (Ph)
Theresa sil (Th)
Houghton muck (Ht)
Pooled standard deviation
Little Menomonee River
Donges Bay Road (1)
County Q Road (2)
Road F near Road B (3)
Appleton Avenue (4)
Upper Menomonee River
Friestad (5)
River Lane (6)
Maple Road (7)
Menomonee Falls (8)
Menomonee Falls Dam (9)
Northern Crossway A*** (10)
Northern Crossway B*** (11)
Lily Creek (12)
Dretzka Creek (13)
124th Street (14)
Lower Menomonee River
Capitol Drive (15)
70th Street (16)
Falk Corporation (17)
Harbor (18)
Pooled standard deviation
Particle-size distribution, %
Sand
24
35
29
21
14
22
1
2.4
59
25
64
17
36
65
58
27
16
46
67
21
30
18
46
80
46
84
5.1
Silt
Soils
57
36
44
44
49
62
38
2.1
Bottom Sediment
25
25
19
32
37
19
23
30
26
19
19
46
23
45
28
8
33
10
6.8
Clay
19
29
27
35
37
16
61
1.5
16
50
17
51
27
16
19
43
58
35
14
33
47
37
26
5
21
6
6.3
Total P, ug/g
Sand
119
186
82
491
426
79
302
19
149
404
81
154
258
108
69
365
385
216
180
872
529
641
128
141
289
42
112
Silt
241
443
154
397
290
128
327
61
394
524
290
200
561
490
350
570
406
288
894
1,949
700
1,839
210
418
1,257
1,930
157
Clay
2,757
2,668
1,821
2,775
1,517
2,336
503
206
2,092
1,557
1,782
426
1,434
1,564
1,410
1,688
**
888
5,529
4,214
5,115
2,943
2,203
1,584
2,683
2,360
603
13
-------
Table 1. Continued
Sample/ sampling
location*
Little Menomonee River
Donges Bay Road (463001)
Noyes Creek (413011)
Appleton Avenue (413008)
Upper Menomonee River
River Lane (673001)
Pilgrim Road (683002)
124th Street (683001)
Lower Menomonee River .
Underwood Creek (413007)
Honey Creek (413006)"""
Honey Creek (413006)
70th Street (413005)++
70th Street (413005)
Schoonmaker Creek (413010)
Falk Corporation (413004)
Falk Corporation (413004)
Harbor
Pooled standard deviation
Particle-size distribution, % Total P, yg/g
Sand
0
0
0
0
0
0
0
1
0
0
0
0
6
0
0
2.0
Silt
Suspended Sediment
9
6
17
18
19
28
18
35
20
29
18
27
54
17
28
1.4
Urban Street Dust and
13th Street Bridge, 9/22/77
13th Street Bridge, 9/26/77
91st Street, 9/22/77
91st Street, 9/26/77
Pooled standard deviation
86
90
86
82
1.6
9
6
8
12
1.9
Clay Sand
+
91
94
83
82
81
72
82
64 26
80
71
82
73
40 550
83
72
3.3 12
Dirt
5 79
4 100
6 120
5 39
0.6 12
Silt
760
286
402
1,030
1,366
903
724
300
864
499
640
770
980
873
1,260
108
317
317
421
471
45
Clay
1,795
1,109
1,061
4,016
2,023
2,142
1,376
900
1,705
1,570
1,414
2,179
2,190
1,148
3,010
258
713
730
883
934
110
*Numbers or letters in parentheses correspond to sampling sites shown in Fig. 1.
**No sample.
***A-before confluence; B-after confluence.
+Samples were collected on July 18, 1977 except where noted.
-H-Samples were collected on June 30, 1977.
Blanks indicate no data due to the absence of sand fraction.
14
-------
dams—showed high contents of clay-sized particles (33 to 58%). Apparently,
fine particles were trapped in the upper reaches of the river. It was
believed that clay-sized particles occur in aggregates which settle in the
upstream portion of the river. By the time aggregates of clay-sized particles
reach the harbor, they were dispersed and remained in suspension for a longer
time before depositing on the lake floor. Lake seiche and turbulence
generated by ship passage and wind can facilitate scouring and resuspension of
clay-sized particles in the harbor and retard their accumulation. Thus, if
the Milwaukee Harbor were to function as a large flow-through lagoon to settle
out fine sediments from the Watershed or from the Jones Island STP, it would
not be very efficient. The ability to trap and retain clay-sized aggregates
decreases with increasing distance from the source due to aggregate
dispersion.
Phosphorus in Soils
Phosphorus concentration in each particle-size fraction of the seven
major soil types from the Menomonee River Watershed (Table 1) were analyzed to
provide an estimate of the range of P concentration. These soils comprise the
majority of the surface area material that potentially can erode and
contribute sediment and associated pollutants to the waterways. The soil
types were scattered randomly through the Watershed so that the source of
sediment could not be related to particular soil types. The six major mineral
soil types showed similar average P contents in their clay-sized fractions and
covered a small range of values, with an average of 2310 pg/g. The organic
soil—containing 44% of organic C—was developed in a bog area where water was
impounded and represents a sink for sediment and pollutants rather than a
source. This soil was included for completeness sake but was not included in
the data interpretation because of its low erosion potential. The P
concentration in the organic soil (503 yg/g) was found to be nearly constant
for all particle sizes. It was believed that the organic constituents in each
particle-size fraction were largely plant material at various stages of
decomposition. All seven soil sampling sites were located in cultivated
fields or pastures, and their P contents may have been affected by fertilizer
additions or animals.
Phosphorus concentrations in the six mineral soils were used as a
reference for comparison with bottom and suspended sediments and urban street
dust and dirt samples. Those samples with P levels greater than the average
soil level (X = 2310 yg/g) were suspected of receiving P input from sources
other than soil. Normally, P concentration in bottom and suspended sediments
would be lower than the P level in soils because of leaching of P.
Phosphorus in Bottom Sediments
Phosphorus concentrations in each particle-size fraction of the bottom
sediments (sampled November 1976, Table 1) were considered to be the result of
long-term sediment deposition—residence time of the bottom sediments in the
Menomonee River is probably several years and a "memory effect of pollutant
additions" might be recorded in the sediments.
15
-------
The 18 bottom sediment sampling sites were divided into three groups:
namely, the Little Menomonee River (Donges Bay Road, County Q Road, Road F
near B and Appleton Avenue); the Upper Menomonee River (Freistadt Road, River
Lane, Menomonee Falls, Nor-X-way A and B, Lily Creek, Dretzka Creek and 124th
Street); and the Lower Menomonee River (Capitol Drive, 70th Street, Falk
Corporation and the Harbor—the latter two were actually part of the
estuary). Bottom sediments were sampled selectively at Donges Bay, River
Lane, Nor-X-way B, Capitol Drive and 70th Street sites to obtain sufficient
amount of clay for P analysis because the river bottom rests on bedrock at
these sites. Since clay-sized particles stay in suspension longer and are
dispersed over a greater area than coarse particles, the clay-sized fraction
was considered to be more uniformly distributed throughout the river system.
Phosphorus concentrations in the clay-sized fraction of the bottom
sediments varied from 246 to 5529 yg/g (Table 1) due to the diverse sources of
P input to the Menomonee River at different locations. Therefore, each bottom
sediment sample and site must be considered separately.
The Little Menomonee River bottom sediment sampling sites were located in
a predominately agricultural area (58%, ref. 9). Phosphorus concentration.in
the clay-sized fraction of the bottom sediment from the Donges Bay site (2092
Mg/g, Table 1) was nearly equal to the average P concentration in the clay-
sized fraction of the six mineral soils (2310 yg/g). This was attributed to
the relative ease with which surface soils erode to the river at this site.
Some sections of the river bank near the sampling site were separated from the
cultivated area by only a thin 2 m grass strip; whereas, other parts were
separated by a large woodlot. There was a decrease in P concentration in the
clay-sized fraction of bottom sediment taken from the three sites downstream
from the Donges Bay sampling site. Each of the sites was surrounded by a
wooded area, with the Appleton Avenue site being the most protected, likely
accounting for it having the lowest P concentration of all bottom sediments in
the river system (426 yg/g) since forests and woodlots provide very little
pollutant loadings.
The Upper Menomonee River drainage area had about 55% of the area in
agricultural and related land use (9). Two villages (Germantown, 5500
population, and Menomonee Falls, 17,000 population) discharge ther STP
effluents to the Upper Menomonee River. The Germantown STP No. 1 outfall was
located between the Freistadt Rd. and River Lane sampling sites (Germantown
STP No. 2 was removed from service in 1973, ref. 10). The industrial waste
water received by the Germantown STP was supplied primarily from a milk
processing and a metal plating plant. This STP treats approximately 380
m /day of water and is classified as providing advanced level treatment for P
removal from pickling liquor. The final effluent was held in a 3780 m
sedimentation pond before being discharged into the Upper Menomonee River
(10). Phosphorus concentration in the clay-sized fraction of the bottom
seidments downstream from the STP outfall (River Lane) was 1560 yg/g. This
value was similar to the P concentration in the clay-sized fraction of the
bottom sediments located upstream, e.g., Freistad Rd.—1434 yg/g, and in the
clay-sized fractions of three upstream sites of the Little Menomonee River,
namely Donges Bay Road, County Q Road and Road F near B. Thus, the amount of
P contributed by the Germantown STP to the bottom sediments was similar in
amount to P contributed by soil eroding from agricultural land use areas in
the Watershed. Average P concentrations in the clay-sized fractions of the
16
-------
sediments from the upper reaches of the Menomonee River Basin (Donges Bay
Road, County Q Road, Road F near B, Appleton Avewnue, Freistadt Road, River
Lane, Maple Road and Menomonee Falls) was 1490 yg/g and somewhat lower than
the average P concentration in the clay-sized fraction of the six major
mineral soil types (2310 yg/g). These values indicate that some leaching or
desorption of P occurs from the soil during its transport to and in the river
system.
The Nor-X-way A bottom sediment sampling site was located in Northern
Crossway, an intermittent creek before its confluence with the Menomonee
River. The creek drains about 1340 ha or 3.9% of the Watershed area, and
about 60% of the areas in agricultural and related land use (3,9). The
drainage area includes part of the village of Germantown, the city of Mequon,
and the village of Menomonee Falls as well as portions of Highway U.S.
41/45. The immediate surrounding area of this sampling site includes Highway
U.S. 41/45, frontage road property and a woodlot, none of which were likely to
have received P fertilizer. As with the Appleton Avenue sampling site, the P
concentration in the clay-sized fraction of the Nor-X-way A bottom sediment
(888 yg/g) was lower than the levels found at the upstream site and was second
lowest of all sites sampled in the river system.
The Nor-X-way B sampling site was located after the confluence of
Northern Crossway Creek with the Upper Menomonee River. It was located
downstream from the Menomonee Falls STP No. 1 outfall and upstream from the
Menomonee Falls STP No. 2 outfall (Fig. 1). A sharp increase occurred in the
P concentration of the clay-sized fraction of the Nor-X-way B bottom sediment
sample (5530 yg/g) which is about twice as high as the average P concentration
in the clay-sized fraction of the six major soil types. It was the highest P
concentration of all 18 bottom sediment sampling sites. This sharp increase
in P concentration was attributed to its location downstream from the
Menomonee Falls STP No. 1 outfall. The STP was old (last major modification
in 1962, ref. 10) and provides a secondary level of wastewater treatment—it
does not include P removal from pickling liquor. The average hydraulic
loading from the STP was 640 nr/day.
The next downstream bottom sediment sampling site (Lily Creek) was
located 3 km from Nor-X-way B and receives effluent input from Menomonee Falls
STP Nos. 1 and 2. However, the P concentration in the clay-sized fraction was
4210 yg/g, which was lower than that found at Nor-X-way B. This was contrary
to the expected trend in P concentration since the Lily Creek site was only 3
km downstream from Nor-X-way B and located below two STP outfalls, while Nor-
X-way B was below only one outfall. It was expected that Menomonee Falls STP
No. 2 would add more P to the already polluted water from upstream, thereby
increasing the P concentration in the sediment sample from the Lily Creek
site. Both plants receive the same type of wastewater, since a common valve
regulates their flow rates. The hydraulic loading at Menomonee Falls STP No.
2 was 265 m /day. The No. 2 plant was newer (constructed in 1969, ref. 10)
and provides tertiary treatment including a settling pond for phsophate
removal. Apparently STP No. 2 made little or no additional P contribution to
the Lily Creek bottom sediment. The decrease in P concentrations in the clay-
sized fraction of the Lily Creek bottom sediment could be explained on the
basis that the effect of the No. 1 plant diminished with distance from the
outfall, and this effect was not compensated for by P loading from the No. 2
plant. Although the Menomonee River resuspends, transports and disperses the
17
-------
sediments from Nor-X-way B to its Lily Creek site, a difference in P
concentration in the two sediment samples could be distinguished. It was
concluded that lagoons and settling ponds located near the source of P input
might remove P from solution by sorption and coprecipitation mechanisms and
improve the water quality of the river emptying into Lake Michigan.
The next sampling site downstream from Lily Creek was Dretzka Creek
located in an agricultural area. It showed an increase in P concentration in
the clay-sized fraction (5120 yg/g)—although slightly lower than the level
found at the Nor-X-way B site. Some P was added at this site to the already
polluted sediments from upstream. After Dretzka Creek, the P concentration in
the clay-sized fraction of the bottom sediments from 124th Street to the
Harbor averaged_2350 yg/g i.e., levels similar to those found in the six
mineral soils (X - 2310 yg/g). It was concluded that P sorption by sediment
in this portion of the Menomonee River was balanced by leaching or desorption
of P from particle surfaces.
The highest P concentration in the clay-sized fraction of the 18 bottom
sediment samples occurred at the Nor-X-way B sampling site, located below a
STP outfall which had secondary treatment capability. Phosphorus
concentration in the clay-sized fractions of sediment samples from two other
sites located below STP outfalls (River Lane and Lily Creek) were lower than
values found at Nor-X-way B due to the tertiary treatment capability of ;these
plants with large settling ponds for coprecipitation of P. Sediment samples
with low P concentration in the clay-sized fraction were found in the upper
part of the Watershed (Appleton Avenue and Nor-X-way A) which was surrounded
by woodlots and in areas where little P fertilizer is used. Apparently, the
land use close to the river (woodlots) had an impact on lowering the P
concentration in sediments when compared with these agricultural areas in
which the woodlots were absent. This reemphasizes the importance of
controlling agricultural P input into rivers through proper management of land
immediately adjacent to the waterways.
Phosphorus in Suspended Sediments
Suspended sediment samples were collected and analyzed to provide a
measure of the immediate and transient P input into the Menoonee River.
Depth-integrated water samples were taken by USGS (Waukesha Branch) on June
30, 1977—2 days after a rainstorm—and on July 18, 1977—11 days after a
rainstorm. In general, P concentrations in the clay-sized fraction of the
suspended sediments were dependent on site location, time of sampling as it
related to position on the hydrograph, and the nature of storm event. The
time of sampling varied from 1 hr before peak flow to as much as 13 hr after
peak flow (Table 2). The lack of uniformity in rain intensity and duration of
storms in the Watershed made it difficult to compare the composition of
suspended sediment samples with each other or with bottom sediments.
Suspended sediments from the Donges Bay Road site was sampled 1.5 hr
after peak flow. Phosphorus concentration in the clay-sized fraction of
suspended sediment from this site (1800 yg/g, Table 1). However, P
concentrations in the clay-sized fraction of the bottom sediments were
generally lower than in suspended sediments at the same location possibly due
18
-------
Table 2.
Flow and time of sampling during collection of suspended sediment samples on June 30 and July 18, 1977 storm events*
Major hydrograph peak***
During sampling
Stations**
Lower Menomonee River
Honey Creek (413006)
70th Street (413005)
Falk (413004)
Upper Menomonee River
River Lane (673001)
Pilgrim Road (683002)
124th Street (683001)
Little Menomonee River
Donges Bay (463001)
Noyes Creek (413011)
Appleton Avenue (413008)
Lower Menomonee River
Underwood Creek (413007)
Honey Creek (413006)
70th Street (413005)
Schoonmaker (413010)
Falk (413004)
Harbor
Flow, cms
15.9
24.8
26.9
0.51
6.23
5.38
0.22
3.54
3.65
7.33
32.3
66.0
1.98
67.7
n. a.
Time
1015
1020-1025
1020-1025
1320-1410
0315
0845-0955
0900-1040
0410-0415
0800-0900
0410-0425
0445
0455-0505
0130
0455-0505
n.a.
Flow, cms
June 30, 1977+
8.98
16.2
16.8
July 18, 1978++
0.51
0.79
3.40
0.20
1.13
0.85
2.55
2.83
7.79
0.28
8.50
n.a.
Time
1100
1200
0930
1400
1530
1520
1130
1030
1500
0930
1100
1730
0800
1730
1400
fraction or
peak flow sampled
0.57
0.65
0.63
1.00
0.13
0.63
0.90
0.32
0.23
0.35
0.09
0.12
0.14
0.13
lime ui seunpiiug aiuei.
peak flow, hr
1
2_
-I4-**
0
12
7
1.5
6
5.5
5
6
12.5
6.5
12.5
*Data made available by USGS.
**STORET numbers correspond to sampling sites shown in Fig. 1.
***Data given corresponds to the major hydrograph peak; which may actually be a plateau as indicated by the time ranges.
+Storm event on June 30, 1978 occurred 2 days after last measurable precipitation.
-H-Storm event on July 18, 1978 occurred 11 days after last measurable precipitation.
-H-+Sampled before peak.
n.a. Not available.
-------
to more extensive leaching of bottom sediments. This effect may not be
obvious if the sediment composition is controlled by STP inputs. This
indicates that the long term rate of P input into this portion of the river
was greater than the P concentration found in the July 18 suspended sediment
sample.
The Noyes Creek suspended sediment sampling site was the next site
downstream from the Donges Bay Road site. This site was located in the Noyes
Creek subwatershed and receives storm water runoff from a medium density
residential area. Phosphorus concentrations in the suspended sediments from
the residential area vary with season—because of leaf litter and grass
clippings deposited on street surfaces and from fertilizer application.
Phosphorus concentration in the clay-sized fraction of the Noyes Creek
suspended sediment sample was 1110 pg/g, which was the second lowest
concentration in the July 18, 1977, suspended sediment samples—lower than
Donges Bay Road site. The suspended sediment sample with the lowest P
concentration in the clay-sized fraction (1060 pg/g) occurred in the July 18
event at the Appleton Avenue site, the next site downstream from Noyes
Creek. Low P contents of suspended and bottom sediments occurred at the same
locations.
Suspended sediment from the River Lane site—located in the Upper
Menomonee River—was sampled during peak flow (Table 2) and had the highest P
concentration (4020 pg/g) in the clay-sized fraction of all suspended
sediments sampled. This could be attributed to the Germanstown STP and/or an
adjacent golf course. The settling pond at the Germantown STP selectively
removes sediments capable of depositing in a short period of time. Sediments
that remain in suspension are transported over a greater distance without
significant contribution to the bottom sediment at the River Lane site.
Hence, a corresponding high P concentration in the bottom sediment did not
occur even though P in the suspended sediment was high. Another possibility
was that the event was of sufficient intensity that sediments of high P
content were removed from the settling pond of the STP and transported to the
river. Downstream from the River Lane sampling site is the Pilgrim Road site,
located in the village of Menomonee Falls and located above both Menomonee
Falls treatment plants. Phosphorus concentration in the clay-sized fraction
of the suspended sediment at the Pilgrim Road site was about half of that
found at River Lane site. This difference may be due to a lower P input at
the Pilgrim Road site or to a difference in sampling time (12 hr after peak
flow vs. during peak flow, Table 2). Downstream from the Pilgrim Road
sampling site was the 124th Street sampling site. Phosphorus concentration in
the clay-sized fraction of suspended sediment at 124th Street was lower than
that of the bottom sediment but slighly higher than the suspended sediment
from the upstream site—Pilgrim Road. Difference in time of sampling relative
to peak flow may acount for differences in P concentration of the two
suspended sediments or a possible P input along the intervening waterway.
The Lower Menomonee River was in the industrial, commercial and
residential area of the Watershed characterized by a higher degree of
impervious area. No municipal STP outfalls are located in this portion of the
river. Except for Harbor sampling site, P concentrations in the clay-sized
fraction of the suspended sediments generally were lower than in soils. With
only one event sample from the Harbor site, it is difficult to identify the
locations of P input. Suspended sediment from the Falk site was found to be
20
-------
less polluted with P than the Harbor site while the bottom sediments from
these sites had similar P composition.
Suspended sediment from Honey Creek, 70th Street and Falk sites were
collected on June 30 and July 18, 1977. Although the July 18 storm was more
intense than the June 30 storm, the flow rate at the time of samping was
greater for the June 30 sample (Table 2) because it was sampled closer to peak
flow. Consequently a higher percent of coarse particles was found in the June
30 samples than in the July 18 samples (Table 1). Phosphorus concentration in
the suspended sediments from the June 30 samples showed no anamolous deviation
from the July 18 samples.
Average P concentrations of suspended sediment samples (June 30 and July
18) was: 1840 ug/g in clay-sized fraction; 780 yg/g in silt-sized fraction;
and 1620 yg/g in unfractionated sample. Average annual event P loading from
suspended sediments in the Menomonee River to Lake Michigan was calculated to
be 16,200 kg/yr with 90% of P associated with the clay-sized fraction. The
calculation was based on a 3 yr average of 10,455,000 kg/yr suspended solid
loading from the Menomonee River to Lake Michigan (11) and assumed that the
June 30 and July 18 suspended sediments were representative samples.
Phosphorus in Urban Street Dust and Dirt
Urban street dust and dirt samples were analyzed for P in each particle-
size fraction to provide an estimate of P contributed by urban storm water
runoff. Total solids and P concentration in the urban street dust and dirt
samples were dependent on the source of particulate matter, time of sampling
and location of sampling site. The 91st Street sampling site was located in a
predominantly commercial-residential area in the central portion of the
Watershed and is located closer to agricultural areas that is the 13th Street
site. The 13th Street sampling site was located in the lower part of the
Watershed in the Industrial Valley of Milwaukee and receives atmospheric
fallout from the industries. The difference in surrounding land use between
the two sampling sites was reflected in the amount of solids removed per curb
meter and in the geometric mean suspended particulate matter concentration at
each site. The 13th Street site consistently received more total solid
loading than the 91st Street site (Table 3).
Phosphorus concentration in the clay-sized fraction of the dust and dirt
sample from 91st Street was slightly higher than the 13th Street sample. This
may result from a higher vegetation density in the 91st Street area as well as
being located closer to an agricultural area. Phosphorus concentration in the
clay-sized fraction of the dust and dirt samples were found to be appreciably
lower than in the mineral soils and the bottom and suspended sediments.
Metals in Soils
Metal composition of each particle size fraction—sand, silt and clay—of
the seven major soils of the Menomonee River Watershed were analyzed (Tables 4
and 5). They provide an estimate of the concentration of metals in the soils
21
-------
Table 3. Antecedent rainfall and solid loading of urban street dust and dirt
samples in 1977
Date of sample
collection or
rainfall
Sept. 12
15
16
17
18
19
22
23
24
26
28
30
Oct. 1
3
5
Rainfall
intensity,
cm/yr
1.
0.
1.
0.
0.
0.
1.
2.
0.
0.
30
76
02
84
08
38
54
64
66
18
Solid
13th Street
8/"f
169
34
30
22
18
37
35
kg/curb km
154
31
27
20
17
34
32
loading*
91st Street
g/ma kg/curb km
82
8.6
19
18
4.5
18
6.1
75
7.8
17
17
4.1
16
5.6
o
*Geometric mean of atmospheric suspended particulate matter is 104.4 yg/m
for 13th Street and 56.6 yg/m for 91st Street.
22
-------
Table 4.
Lead, cadmium, zinc and copper concentrations in each particle-size fraction of soils, sediments and urban street dust
and dirt in the Menomonee River Watershed
Metal, pg/g
Sample/ sampling
location*
Ozaukee sil (Ou)
Mequon sil (Mt)
Hochheim sil (Hm)
Ashkum sicl (As)
Fella sil (Ph)
Theresa sil (Th)
Houghton muck (Ht)
Pooled standard deviation
Sand
N.D.
4.7
5.5
9.0
9.8
N.D.
N.D.
0.66
Pb
Silt
9.5
11
9.8
14
10
6.0
12.3
1.1
Clay
58
39
56
36
39
55
13
1.9
Sand
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
Cd
Silt
Soils
0,16
0.25
0.11
0.53
0.23
0.12
N.D.
0.10
Clay
0.78
0.73
0.35
1.24
0.81
0.44
N.D.
0.08
Sand
12
12
6.9
9.8
10
10
N.D.
1.2
Zn
Silt
32
29
41
71
40
20
34
23
Clay
110
64
192
154
204
345
31
88
Sand
2.3
4.0
1.8
2.0
1.9
2.1
N.D.
0.37
Cu
Silt
17
15
7.3
27
8.2
4.8
17
2.7
Clay
90
82
41
106
44
36
15
2.2
Bottom Sediment
Little Menomonee River
Donges Bay Road (1)
County Q Road (2)
Road F near Road B (3)
Appleton Avenue (4)
Upper Menomonee River
Friestad (5)
River Lane (6)
Maple Road (7)
Menomonee Falls (8)
Menomonee Falls Dam (9)
Northern Crossway A*** (10)
Northern Crossway B*** (11)
Lily Creek (12)
Dretzka Creek (13)
124th Street (14)
Lower Menomonee River
Capitol Drive (15)
70th Street (16)
Falk Corporation (17)
Harbor (18)
Pooled standard deviation
2.5
9.6
4.1
20
4.1
7.4
4.8
12
21
17
32
36
17
14
32
16
170
277
10
7.3
17
16
21
7.8
16
17
18
42
65
101
64
55
33
35
92
412
771
16
36
25
65
41
25
41
44
55
**
176
512
438
334
208
115
487
1,439
2,210
53
0.06
N.D.
0.06
N.D.
N.D.
N.D.
0.02
N.D.
0.07
N.D.
0.08
N.D.
0.11
N.D.
0.19
0.07
1.88
0.35
0.11
0.21
N.D.
0.45
0.16
0.20
N.D.
0.13
N.D.
0.33
0.36
0.72
0.31
0.59
0.26
0.44
0.52
4.98
3.8
0.14
1.1
0.86
1.6
0.58
1.3
0.98
1.1
0.54
**
1.3
3.2
2.9
2.5
1.7
1.8
3.8
33
12.2
0.49
8.8
16
12
25
11
14
9.7
15
26
32
29
61
23
26
23
37
127
75
11
42
43
80
48
34
32
28
34
50
150
153
96
98
69
50
188
408
469
11
236
213
484
171
168
174
181
213
**
695
1,390
1,240
949
637
283
849
2,160
1,400
101
2.4
2.4
1.7
3.0
2.1
1.9
1.5
3.6
3.5
3.1
4.1
9.4
4.1
6.7
6.6
5.7
102
37
3.2
8.5
11
8.1
6.6
9.8
8.2
12
7.1
12
17
27
11.8
20
17
13.8
42
219
198
5.6
48
36
48
29
52
44
44
38
**
64
149
145
122
85
108
110
475
304
11
-------
Table "f. Continued
to
-p-
Sample/ sample
location*
Little Menomonee River
Donges Bay Road (463001)
Noyes Creek (413011)
Appleton Avenue (413008)
Upper Menomonee River
River Lane (673001)
Pilgrim Road (683002)
124th Street (683001)
Lower Menomonee River
Underwood Creek (413007)
Honey Creek (413006)++
Honey Creek (413006)
70th Street (413005)"H"
70th Street (413005)
Schoonmaker Creek (413010)
Falk Corporation (413004 V1"
Falk Corporation (413004)
Ha rbo r
Pooled standard deviation
13th Street Bridge, 9/22/77
13th Street Bridge, 9/26/77
91st Street, 9/22/77
91st Street, 9/26/77
Pooled standard deviation
Pb
Sand Silt
N.D.
139
31
N.D.
50
60
348
314
158
339
125
967
301 441
104
363
9 26
142 370
245 813
540 1,655
777 2,570
62 47
Clay Sand
Suspended
43
166
63
83
244
204
515
720
333
477
165
1,513
1,030 1.2
118
715
26 0.5
Urban Street
1,120 0.95
2,319 2.0
2,891 1.3
5,081 2.7
166 0.38
Cd
Silt
Metal, ug/g
clay Sand
Zn
Silt
Clay
Cu
Sand Silt
Clay
Sediment"1"
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
0.9
0.81
1.3
N.D.
N.D.
2.3
0.77
1.1
0.83
Dust
6.2
11
12
17
0.61
0. 34
0.58
0.37
2.4
N.D.
0.90
1.7
1.7 132
1.4
2.3
0.88
4.4
6.4 193
0.75
4.3
0.21 46
and Dirt
19 396
29 495
15 637
22 907
1.3 130
70
218
117
692
235
221
330
180
246
214
138
2,910
327
124
2,390
74
811
1,070
1,070
1,320
106
306
495
328
392
380
409
623
556
619
618
297
3,280
1,670
312
1,239
101
2,590
2,890
1,720
1,800
248
20
41
8.3
29
22
19
40
24
39
42
50
50
59 85
37
531
4 22
164 166
275 268
270 357
241 478
128 24
47
41
38
37
51
71
78
84
76
98
70
104
188
69
346
9
319
433
356
376
34
*Numbers or letters in parentheses correspond to sampling sites shown in Fig. 1.
**No sample.
***A-before confluence; B-after confluence.
+Samples were collected on July 18, 1977 except where noted.
-H-Samples were collected on June 30, 1977.
N.D. Not detected.
Blanks indicate no data due to the absence of sand fraction.
-------
Table 5.
Aluminum, iron, manganese, chromium and nickel concentrations in each particle-size fraction of soils, sediments and urban street dust and dirt in
the Menomonee River Watershed
Sample/ sampling
location*
Metal, ug/g
Sand
Al
Silt
Clay
Sand
Fe
Silt
Clay
Sand
Mn
Silt
Clay
Sand
Cr
Silt
Metal
Clay
. Pg/8
Sand
Ml
Silt
Clay
Soils
Ozaukee sil (Ou)
Mequon sil (Mt)
Hochheim sil (Hm)
Ashkum sicl (As)
Pella sil (Ph)
Theresa sil (Th)
Houghton muck (Ht)
Pooled standard deviation
Little Menomonee River
Donges Bay Road (1)
County Q Road (2)
Road F near Road B (3)
Appleton Avenue (4)
Upper Menomonee River
Friestad (5)
River Lane (6)
Maple Road (7)
Menomonee Falls (8)
Menomonee Falls Dam (9)
Northern Crossway A*** (10)
Northern Crossway B*** (11)
Lily Creek (12)
Dretzka Creek (13)
124th Street (14)
Lower Menomonee River
Capitol Drive (15)
70th Street (16)
Falk Corporation (17)
Harbor (18)
Pooled standard deviation
2,460
866
2,450
5,500
6,580
2,200
3,590
761
1,350
4,640
1,160
2,370
2,000
792
1,490
3,130
2,640
3,500
1,330
2,970
1,840
2,890
1,300
1,470
4,500
**
492
5,970
7,860
6,160
6,870
5,680
6,380
4,390
783
5,920
7,500
3,660
4,430
5,610
4,690
4,900
7,120
5,780
7,140
4,000
5,230
3,250
5,880
3,600
8,890
8,210
**
594
62,400
50,700
85,400
61,200
65,500
79,200
4,500
4,498
43,200
35,100
34,200
16,500
35,860
26,200
23,500
41,600
**
24,900
29,900
38,300
33,800
41,300
34,600
4-2,900
34,600
**
5,590
4,280
10,100
7,520
2,770
5,210
10,300
10,370
711
2,840
6,320
2,320
5,110
12,200
2,700
4,750
4,920
4,810
4,490
3,890
7,930
4,410
8,850
3,080
3,310
11,800
6,550
1,510
10,000
18,900
9,550
6,180
6,270
8,750
8,180
1,240
Bottom
8,550
12,300
7,760
7,720
12,600
8,480
11,900
8,890
7,620
8,280
13,000
11,100
7,280
12,800
6,650
15,800
23,100
25,100
4,470
53,100
56,600
61,000
40,600
57,100
60,600
8,960
4,660
Sediment
55,200
33,000
44,300
16,600
46,700
37,000
41,900
33,400
A*
20,300
52,100
47,800
49,100
44,700
40,000
50,840
46,900
48,100
7,390
215
191
111
47
41
100
104
7.7
137
230
227
449
186
256
212
497
315
203
189
377
334
457
343
185
325
194
64
647
984
575
149
323
477
77
77
239
317
939
651
206
313
250
316
382
348
789
469
741
599
942
560
704
706
31
2,610
1,160
2,270
238
1,300
1,640
130
104
499
665
1,170
772
363
479
397
657
**
587
1,200
321
1,080
361
1,540
722
903
894
111
11
9.4
7.0
3.7
4.2
14
N.D.
1.4
1.9
6.1
1.3
3.7
3.5
1.6
1.9
5.2
5.0
4.0
3.6
14
4.0
7.2
3.4
2.9
40
12
6.0
17
14
15
15
13
13
8.6
2.6
11
13
7.1
6.9
10
11
11
11
14
13
26
25
18
17
16
18
303
140
18
75
63
75
71
96
69
11
15
43
48
51
21
40
43
45
52
**
35
347
260
198
115
151
139
329
556
84
7.2
8.9
4.2
2.3
3.2
2.1
N.D.
0.64
2.4
4.9
**
4.2
2.7
2.3
2.9
6.7
4.5
3.8
**
15
**
11
**
3.7
21
6.2
2.7
9.1
15
7.7
7.8
5.0
6.0
12
2.0
7.0
13
**
9.0
9.0
9.3
8.8
9.1
10
10
**
35
**
44
**
16
62
33
2.7
49
48
50
35
30
45
12
1.0
36
32
**
27
35
33
35
34
**
26
**
235
**
204
**
63
117
83
6.9
-------
Table 5. Continued
NJ
Metal, [ig/g
Sample /sampling
location*
Al
Sand Silt
Fe
Clay
Sand Silt
Clay Sand
Mn
Silt
Clay
Cr
Sand Silt
Metal,
Clay
Mg/g
Ni
Sand Silt
Clay
Suspended Sediment
Little Menomonee River
Donges Bay Road (463001)
Noyes Creek (413011)
Appleton Avenue (413008)
Upper Menomonee River
River Lane (673001)
Pilgrim Road (683002)
124th Street (683001)
Lower Menomonee River
Underwood Creek (413007)
Honey Creek ( 413006 )++
Honey Creek (413006)
70th Street (413005)++
70th Street (413005)
Schoonmaker Creek (413010)
Falk Corporation (413004)
Falk Corporation (413004)
Harbor
Pooled standard deviation
12,260
13,960
8,610
12,00
11,000
7,700
12,800
3,180 10,610
11,450
13,800
9,210
15,000
3,700 6,890
9,470
20,700
890 2,405
52,500
40,400
42,900
37,800
43,700
48,500
51,100
47,600
58,800
52,400
40,700
60,900
93,000
33,400
53,600
4,481
13
14
11
138
11
11
19
7,600 15
14
16
13
161
11,100 19
12
145
1,493 5
,200
,500
,300
,000
,000
,800
,000
,000
,900
,600
,700
,000
,900
,000
,000
,574
Urban Street
13th Street Bridge, 9/22/77
13th Street Bridge, 9/26/77
91st Street, 9/22/77
91st Street, 9/26/77
Pooled standard deviation
3,186 5,560
2,900 6,170
1,520 8,600
1,640 8,550
809 2,405
15,410
16,790
32,160
28,000
4,480
45,980 34
63,340 47
69,730 49
57,070 67
1,490 5
,310
,360
,880
,750
,570
49J700
39,400
35,400
50,300
44,500
46,400
50,500
46,600 330
46,100
44,600
39,100
163,000
98,900 441
25,500
68,300
5,395 46
Dust and Dirt
43,940 283
55,490 348
65,540 439
77,860 554
5,390 46
323
655
503
528
846
480
447
379
438
805
438
903
596
551
817
68
332
437
770
790
68
768
578
613
1,620
1,020
783
767
683
558
581
604
804
1,830
526
1,640
284
614
764
1,090
1,310
284
27
51
16
**
21
23
39
N.D. 25
34
N.D.
28
28
33 59
28
96
36 9.4
33 43
41 53
53 43
27 44
6.8 5.4
59
49
47
45
63
154
77
56
105
51
104
98
181
57
239
13
87
108
72
80
11
48
167
33
**
97
36
78
54 12
40
41
32
33
12 25
26
148
5.2 6.7
37 57
66 64
49 69
67 80
13 7
45
54
53
34
59
79
54
52
53
70
52
65
66
51
140
9.3
75
73
76
82
5
*Numbers or letters in parentheses correspond to sampling sites shown in Fig. 1.
**No sample.
***A-before confluence; B-after confluence.
+Samples were collected on July 18, 1977 except where noted.
-H-Samples were collected on June 30, 1977.
N.D, Not detected.
Blanks indicate no data due to absence of sand fraction.
-------
subject to erosion in the Menomonee River Watershed. These values were used
for comparison with bottom and suspended sediments and urban street dust and
dirt. The clay-sized fraction was the most important fraction for this
comparison because of its large surface area per unit mass on which surface
sorption and desorption of metals might occur. It was expected that metal
concentrations in sediments would be lower than the levels found in soils
because of leaching during resuspension, transport and deposition by storm
water runoff and in waterways. Furthermore, Fe and Mn may undergo chemical
reduction and dissolution in an aqueous environment, thereby lowering their
concentrations in sediments as compared to surface soils. Sediments with
higher metal concentrations in the clay-sized fraction than found in the clay-
sized fraction of soils were suspected of receiving pollutant input from
external sources. However, the high concentrations of Al, Fe and Mn in the
six mineral soils tend to mask other sources of these metals in sediments.
Metals in Bottom Sediments
Metal compositions for each particle-size fraction of bottom sediments
are given in Tables 4 and 5. Cadmium, Cr, Cu, Ni, Pb and Zn concentrations in
the clay-sized fractions of bottom sediments from the Little Menomonee River
were low compared to concentrations in bottom sediments from the Upper and
Lower Menomonee River and the six mineral soils (Tables 4 and 5). A similar
trend also was indicated in the silt-sized fractions for some of the metals.
Low metal concentrations in the clay-sized fraction of bottom sediment samples
from the Little Menomonee were probably due to the predominantly agricultural
and related land use in the drainage area at these sites (58%, ref. 9); Little
or no industrial surface water discharge occurs into the Little Menomonee
River. Except for bottom sediments sampled at Road F near B the other samples
collected in the Little Menomonee River were considered to be unpolluted with
respect to Cd, Cr, Cu, Ni, Pb and Zn. The metal contents of these samples
were used as a basis for comparison with other bottom sediment samples.
Bottom sediment samples from the Road F near B site had higher Cd, Mn, Pb
and Zn concentrations than bottom sediments from other Little Menomonee
sites. However, no measurable increase in Cr, Cu and Ni concentrations was
found at the Road F near B samping site. Increases in Cd, Pb and Zn
concentrations at this site could be attributed to its close proximity to two
well-traveled county roads. The Pb presumably was from use of leaded
gasoline, the Zn from vulcanized rubber tires and oil lubricant and the Cd as
an impurity of Zn (12,13).
The Upper Menomonee River subwatershed had about 55% of the land in
agricultural and related land use (10). The bottom sediment at the uppermost
sampling site in this subwatershed—Freistadt Road—was surrounded by
agricultural land. Cadmium, Cr, Cu, Pb, Ni and Zn concentrations in the clay-
sized fractions were similar to the levels found in the "unpolluted" bottom
sediment samples from the Little Menomonee River. Downstream from the
Freistadt Road sampling site was the River Lane site, located below the
Germantown STP No. 1 outfall and adjacent to a golf course. No measurable
increase in Cd, Cr, Cu, Pb, Ni and Zn concentrations occurred in the bottom
sediments even though the Germantown STP No. 1 receives wastewater from a
metal plating company. Low metal concentrations also were found in the clay-
27
-------
sized fractions of bottom sediments from the Maple Road and Menomonee Falls
sites—both located downstream from the River Lane site. Thus, bottom
sediments from the four uppermost sampling sites (Freistadt Road, River Lane,
Maple Road and Menomonee Falls) also were considered to be "unpolluted" with
respect to Ca, Cr, Cu, Ni, Pb and Zn. Metal concentrations in the clay-sized
fraction of the "unpolluted" bottom sediments showed similar ranges even
though their clay-sized fraction contents varied from 14 to 51% (Table 1) and
the sampling sites extended a distance of 24 km along the Upper Menoonee River
and the Little Menomonee River. Metal concentrations in the clay-sized
fractions of the bottom sediment samples were unaffected by clay content even
though some samples were collected selectively to obtain high clay content.
The Nor-X-way A bottom sediment sampling site was located in the Northern
Crossway intermittent creek, just before its confluence with the Menomonee
River. It receives storm water runoff from a frontage road and U.S. Highway
41/45 which is about 100 m upstream. Except for Pb and Zn, the metal
concentrations in the clay-sized fraction of the bottom sediments were similar
to the "unpolluted" bottom sediment samples from the four Upper Menomonee
River sites and three Little Menomonee River sites. This was also the case
for the Road F near B site, which was located in close proximity to well-
traveled roads and showed an increase in Pb and Zn concentrations in the clay-
sized fractions of the bottom sediments as compared to the background bottom
sediment samples. A slight increase in Cu concentration may have occurred
above the background level at Nor-X-way A. A possible source of Cu to this
site is direct industrial surface water discharge to the creek—an outfall is
located upstream from the sampling site. Further analyses would be necessary
to establish positively the possible increases in Cu levels and to reveal the
source. The sediment sample from the Nor-X-way A site was considered to be
polluted with Zn and Pb.
The village of Menomonee Falls had two STPs that discharged effluent into
the Menomonee River. The first STP was located upstream from the Nor-X-way B
site which was located immediately after the confluence of the intermittent
Northern Crossway Creek and Menomonee River (Fig. 1). The second STP outfall
was located about 3 km downstream from STP No. 1, below the Nor-X-way B and
above the Lily Creek sites. In addition to domestic sewage, these plants
treat wastewater from metal foundries and electroplating plants. Both plants
received the same type of wastes because they have a common valve that
regulated flow rates. Menomonee Falls STP No. 1 had secondary treatment
capability while STP No. 2 had tertiary treatment capability. Cadmium, Cr,
Cu, Pb and Zn concentrations in the clay-sized fraction of the bottom
sediments from the Nor-X-way B site were significantly higher (2 to 10 times)
than the "unpolluted" sites. Furthermore, they were higher than the Lily
Creek sample which was downstream from Nor-X-way B and received discharges
from both treatment plants. As with P concentration, the Nor-X-way B site had
the highest concentrations of Cr, Cu, Ni, Pb and Zn of any site in the uppr
reaches of the Menomonee River down to 70th Street. It also had the second
highest Cd concentration in the clay-sized fraction of this portion of the
Menomonee River. It was concluded that the level of sanitary treatment was
responsible for the observed differences in P and metal concentrations in the
bottom sediments. The Menomonee Falls STP No. 2 with teritary treatment for
the removal of P by co-precipitation also removed metals from the effluent.
The clay-sized fraction of the bottom sediments could be used to distinguish
the extent of pollutant loading at these two sites which were 3 km apart, even
28
-------
though the river transports and diffuses the pollutants from the Nor-X-way B
site downstream to the Lily Creek site. The portion of the Menomonee River
from Dretzka Creek to 70th Street did not receive continuous sewage discharge
even though an overflow bypass was located at the village of Butler just
upstream from the 124th Street site which occasionally discharged raw sewage
into the river (10). Since this segment of the river did not receive
industrial or municipal wastewater discharge, a general decline in Cd, Cr, Cu,
Ni, Pb and Zn concentrations in the clay-sized fraction of the sediments was
observed with increasing distance from the Nor-X-way B site to the Capitol
Drive site. The slight increase in P concentration in the clay-sized fraction
of Dretzka Creek bottom sediment without a corresponding increase in metal
concentrations indicated an agricultural input of P rather than industrial or
municipal sewage discharge. The occasional discharge of raw sewage at the
village of Butler did not measurably affect metal concentrations in the clay-
sized fraction of the bottom sediments from 124th Street. The level of
pollutant input may have been too low or the duration of the discharge was
insufficient to change the elemental composition of the bottom sediments.
However, the general decline in metal contents downstream from this point
never decreased levels to those found in the upper reaches of the Menomonee
River.
Normally a decrease in metal concentration was expected in the bottom
sediment at the Capitol Drive sampling site because is was located after the
confluence of the Upper Menomonee River and the Little Menomonee River (Fig.
1). Thus, there should be evidence of a dilution effect on the metal
composition of the bottom sediments particularly since bottom sediments from
the Little Menomonee River were found to be "unpolluted." Furthermore, the
site was located at a greater distance from the Menomonee Falls STP. However,
the clay-sized fraction of the bottom sediment sampled at Capitol Drive showed
an increase in Cd, Cr, Cu and Mn over the levels found at the 124th Street
site. Therefore, it was suspected that another source for these metals
existed. It is generally difficult to interpret slight changes in Al, Fe and
Mn concentrations in bottom sediment samples because Eh and/or pH changes in
the aquatic environment might influence the concentrations of these
elements. However, the increase in Mn concentration in the Capitol Drive
sample was quite large. It had been reported (10) that an electroplating
company had one of its wastewater discharge outfalls leading directly to the
Menomonee River about 1.6 km upstream from the Capitol Drive site and about
1 km downstream from the 124th Street site. In each of the three sampling
surveys taken by the WDNR and Southeastern Wisconsin Regional Planning
Commission (SEWRPC) from April 1973 to August 1974 (10), the concentration of
Cd, Cr and Pb in the plant's wastewater discharge were found to be in excess
of the maximum concentrations recommended for surface water supporting fish
and other aquatic life. Other metals used by the electroplating company
included Cu and Ni. Thus, three of the five metals (Cd, Cr and Cu) that were
discharged by this company to the river showed increased concentrations in the
clay-sized fraction of the bottom sediment sampled 1.6 km downstream from the
outfall—Capitol Drive—as compared to an upstream site—124th Street.
However, as of June 1975, the plating company claimed to have stopped
discharging wastewater directly into the Menomonee River (14). This indicates
either that a "memory effect" for metals is present in the sediments and/or
other sources of metals are present. Several other industrial surface
discharge sites leading to the Menomonee River in this area have been reported
(10).
29
-------
The next site downstream from Capitol Drive is 70th Street, which showed
an increase in Cd, Cr, Pb and Zn concentrations in the clay-sized fractions of
the bottom sediments. This section of the river traverses an industrial park
in the city of Wauwatosa and receives water from Underwood and Honey Creeks.
The Underwood and Honey Creek subwatersheds include some heavy industries
which may contribute to pollutant loading at the 70th Street site. The
traffic density and surface imperviousness of these subwatersheds increased as
the river approached the inner city of Milwaukee. Sources of metal pollutants
are diverse, although automobiles are generally important contributors of Pb
from leaded gasoline; Zn from oil lubricants and vulcanized rubber; Cd as an
impurity in Zn; and Ni from corrosion of Ni-containing parts (11,12,15). A
sharp increase in metal concentrations in the clay-sized fraction of bottom
sediments was observed at the Falk and Harbor sites, located downstream from
70th Street. Both of these sites were in the Industrial Valley of the city of
Milwaukee and receive storm water runoff from a railroad yard, chemical
plants, foundries and interstate highway 94. The Harbor site showed an
increase in Cr and Pb and a decrease in Cd, Cu, Ni and Zn concentrations in
the clay-sized fraction of the bottom sediment as compared to the Falk site.
However, these differences in the bottom sediments may not represent actual
differences in pollutant input from surface discharges due to disturbance of
the sediments. The Harbor site was located in that part of the river used by
ships and was dredged by the Army Corps of Engineers. The most recent
dredging operation—performed in 1975—may have affected the metal
concentrations in the Harbor sediment sample. Both sites were located In the
estuary part of the river and were affected by lake seiche.
Elemental composition of clay-sized fractions of bottom sediments were
found to be useful in identifying possible sites of pollutant input. These
values were compared with values found in the major mineral soil types of the
Watershed. The elemental composition of soils was expected to be higher than
"unpolluted" bottom sediments due to leaching and/or desorption of elements
from particle surfaces. Metal concentrations in the clay-sized fraction of
bottom sediments that were greater than levels in "unpolluted samples" were
considered to have received metal inputs from sources other than soils.
,In agricultural land use areas, bottom sediments with P levels greater
than the soil levels but without a corresponding increase in metal levels were
found. In urban areas, sediments which received effluent from a STP with
secondary treatment capability showed an increase in Cd, Cr, Cu, Ni, P, Pb and
Zn concentrations. Sediment samples from the inner city and Industrial Valley
of Milwaukee have high metal concentrations presumably from atmospheric
fallout and storm water runoff, since there are no STP outfalls in this
segment of the river. The Al, Fe and Mn concentrations in "unpolluted" bottom
sediment samples were slightly lower than those found in soils.
Metals in Dousman Ditch Bottom Sediments
Oftentimes, sediments that are transported from a subwatershed to a
receiving body of water traverse through natural drainage ditches or marshes
where sediments are trapped. The Brookfield Shopping Center specific study
site was chosen to evaluate such an effect. It had a storm sewer outfall
which was connected to Underwood Creek by a natural drainage ditch
30
-------
(Dousman Ditch) and traverses through a bog. Location of bottom sediment
sampling sites are shown in Fig. 2 and the particle-size distribution and
elemental composition of each particle-size fraaction are given in Table 6.
The Mn concentration was found to be lower than the soil level for bottom
sediments. The high organic matter content of the bottom sediment and the
surrounding muck soil likely contributed to the chemical reduction of Fe and
Mn and their subsequent depletion from sediments. Manganese concentration in
the clay-sized fraction was correlated significantly with Fe concentration
(r = 0.95) in the six Dousman Ditch bottom sediment samples, indicating that
the two elements were depleted simultaneously by chemical reduction.
Disproportionately high levels of Cd, Cr, Fe, Ni, Pb and Zn in the clay-
sized fraction at the Dousman No. 1 site than at the other Dousman sites
indicate that considerable portions of these metals were deposited in the
neighborhood of the sewer outfall and showed little tendency to migrate.
Metals that remain in suspension were transported through and exited the
drainage ditch and did not accumulate in the intervening channel from Dousman
site No. 2 to Dousman site No. 6. The high concentration of Cd, Pb, and Zn at
Dousman No. 1 (3, 18 and 15 times greater than in "unpolluted" bottom
sediments, respectively) were presumed to arise from storm water runoff of
traffic-related pollutants in the Brookfield Shopping Center parking lot.
Thus, a natural drainage ditch traversing through a bog was found to retain
metal pollutants in the neighborhood of the outfall under low flow rates.
Metals in Suspended Sediments
Metal concentrations in suspended sediments were measured to determine
the short-term sediment and associated pollutant input into a receiving body
of water. Concentration of pollutants associated with the sediments are
affected by many variables so that long-term trends are difficult to
extrapolate from a small set of suspended sediment data. Metal concentrations
in urban suspended sediment samples are affected by the length of the dry
period preceding the storm event, the amount of dust and dirt accumulation on
streets, street sweeping practices, the intensity of the storm event which
influences the amount and particle size of dust and dirt removed from the
surfaces, and the stage of the hydrograph when the samples are taken.
Generally, the concentration of pollutants are highest during the first flush
stage of the rising part of the hydrograph. Therefore, the concentrations of
elements in each particle-size fraction of the suspended sediment samples
(Tables 4 and 5) represented discrete samples for a specific event and for a
particular portion of the hydrograph.
Two sets of suspended sediment samples (June 30 and July 18, 1977) were
analyzed for concentration of elements in each particle-size fraction for a
general comparison with the bottom sediments and soils. Unlike the bottom
sediments, no general pattern or association of metal concentration with their
sampling sites was observed for the suspended sediment samples. Differences
in time of sampling relative to the hydrograph peak of a storm event was a
source of wide variation in elemental composition of suspended sediments.
Time of sample collection varied from 1 hr before peak flow to as much as
13 hr thereafter (Table 2). With the exception of samples affected by STPs
and industrial discharges, the Al, Fe and Mn concentrations in the clay-sized
31
-------
Table 6.
Particle size distribution and element concentrations in each size fractions of Dousman Ditch bottom sediment
Sample*
Dousman 1
Dousman 2
Dousman 3
Dousman 4
Dousman 5
Dousman 6
Pooled
standard
deviation
Particle size
Fraction distribution,** ;
Sand
Silt
Clay
Sand
Silt
Clay
Sand
Silt
Clay
Sand
Silt
Clay
Sand
Silt
Clay
Sand
Silt
Clay
Sand
Silt
Clay
20
55
25
3
42
55
7
35
58
27
32
41
22
45
33
23
40
37
Element, yg/g
I Total P
314
573
2,220
938
1,170
2,460
652
430
1,190
328
971
1,940
294
1,620
2,090
284
501
1,570
152
84
153
Pb
141
208
932
***
36
60
55
62
163
11
46
134
10
53
124
16
38
167
6.1
4.7
25
Cd
0.68
1.0
5.0
3.9
2.0
2.6
0.91
0.72
1.5
0.48
0.37
1.6
0.14
0.71
1.7
0.09
0.34
1.3
0.20
0.16
0.27
Cu
7.9
24
31
24
14
54
5.8
12
44
4.8
17
46
4.6
21
38
6.4
12
54
0.80
1.3
3.1
Cr
7.4
21
102
17
15
21
9.8
9.1
59
4.6
11
53
4.7
12
41
4.5
9.5
69
1.5
1.5
5.6
Ni
4.3
12
62
***
23
28
13
13
36
4
13
31
3.4
12
27
4.2
9.0
36
2.3
1.4
5.9
Zn
39
104
671
34
45
113
42
56
200
23
54
209
37
61
184
21
37
234
7.2
7.1
22
Al
2,890
8,840
41,800
5,020
10,300
103,000
6,200
8,170
40,600
3,830
5,900
28,800
3,070
5,340
23,400
2,880
4,890
39,600
1,050
1,570
11,470
Fe
7,190
12,300
54,800
679
7,960
10,200
6,650
9,180
29,500
6,090
14,700
25,700
6,520
13,200
23,800
6,420
10,900
34,400
1,370
890
1,025
Mn
261
392
574
92
138
126
378
442
361
170
220
245
202
325
344
392
289
413
27
25
10
*See Fig. 2 for location of sampling points along Dousman Ditch.
**Samples dispersed by ultrasonic treatment without prior removal of organic matter; clay size fraction is < 4 ym.
***No sample.
+Sp = / Mean square error.
-------
fractions of the suspended sediments were found to be higher than in the clay-
sized fractions of bottom sediments and lower than in the clay-sized fractions
of the surface soils. Apparently, suspended sediments were less subject to
leaching during aqueous transport than bottom sediments. High concentrations
of Al, Fe and Mn in soils generally mask other possible sources of input for
these metals in the suspended sediments. The few sediment samples that
deviate from this trend and show high Al, Fe or Mn concentrations in their
clay-sized fractions as compared to other sediment samples were suspected of
receiving significant inputs of these metals from external sources. The
concentrations of Cd, Cr, Cu, Ni, Pb and Zn in the major soil types from the
watershed were sufficiently low to permit the detection of additional input of
these metals in the suspended sediment derived from surface soils.
The elemental composition of the clay-sized fraction of suspended
sediment from the Donges Bay Road sampling site (Tables 4 and 5)—located in
the upper reaches of the Little Menomonee River—was similar to the levels
found in soils, and was considered to be "unpolluted" with these metals. The
River Lane sampling site—located in the upper reaches of the Upper Menomonee
River—was found to be polluted with Cd and Mn. This might be caused by
inputs from the Germantown STP and/or an adjacent golf course. The settling
pond at the Germantown STP selectively removes sediments that were able to
deposit quickly and in so doing removed those sediments more likely to
contribute to bottom sediment pollution; suspended sediments and associated
pollutants from the STP were not removed at the settling pond. Another
possibility was that the storm was of sufficient intensity to cause overflow
and/or a more rapid flow of effluent which subsequently lowered the
depositional capability of the STP settling pond. Pollutants generated in
this fashion are more likely to be detected in suspended than in bottom
sediments.
The Pilgrim Road suspended sediment sampling site was in the village of
Menomonee Falls above the two Menomonee Falls treatment plants and drains 8739
ha of land, 78% of which is in agricultural and related land uses (9). It
also receives storm water from the village of Menomonee Falls, which could
explain the sharp increase in Pb arising from automobile traffic inputs. Lead
concentration in the clay-sized fraction was significantly higher than the
level in soil even though the sample was taken 12 hr after the peak of the
hydrograph had been reached (Table 2). Although Zn is associated commonly
with vehicular emission, no measurable increase over the soil level was
observed. Possible explanations are 1. Zn is more soluble or more readily
removed from streets during the initial phase of the storm than is P (12),
2. Pb in the sample arose from atmospheric fallout of automobile exhaust
materials and contributes to pollutional loading for a longer time than Zn and
3. another unknown source of Pb exists which was not associated with Zn.
The next suspended sediment sampling site downstream from Pilgrim Road
was 124th Street. It receives effluent from both treatment plants in the
village of Menomonee Falls. Chromium, Cu, Ni, Pb and Zn in the clay-sized
fraction of the suspended sediment were higher than the soil levels even when
the smaple was taken 7 hr after the peak of the hydrograph (Table 2). This
was in agreement with bottom sediment analysis of this portion of the
Menomonee River—a large contribution of metals from the Menomonee Falls STP
No. 1 outfall which was located after Pilgrim Road and before 124th Street.
However, the Pb level at the 124th Street site was slightly lower than in the
33
-------
Pilgrim Road site, even though the former was sampled closer to the time of
peak flow (7 hr compared to 12 hr). The reduction of Pb concentration may be
due to a lower density of streets near the 124th Street site than near the
Pilgrim Road site. An effect due to dilution and deposition of Pb may occur
as the river flows through the 4 kmn of agricultural land separating the two
sites.
Although bottom sediments from the upper reaches of the watershed
generally have lower metal concentrations in the clay-sized fractions than
samples from the lower part of the Watershed, this was not always the case
with suspended sediments. Suspended sediments from the next two sites below
124th Street—70th Street and Falk Corporation—had low Zn concentrations with
values similar to those found at the Donges Bay Road and River Lane sites
located in the upper part of the Watershed. These two samples were collected
12.5 hr following peak flow and dilution effects may have been significant
during this part of the storm water runoff hydrograph.
Elemental composition of suspended sediments collected from the same
sites for different storm events may show different trends resulting from
differences in antecedent rainfall conditions and the stages of the hydrograph
when the samples were collected. The suspended sediment sample collected at
the Falk Corporation site on July 18, 1977 had lower Cd, Cr, and Pb
concentrations in the clay-sized fraction than those found in the 70th ^Street
sample; the opposite relationship was true for the June 30, 1977 sample. Flow
rates during collection of the June 30 suspended sediment samples were higher
than during collection of the July 18 suspended sediment samples (Table 2).
The higher flow rate during the collection of the June 30 sample showed a
correspondingly higher sand-sized fraction in the suspended sediments.
A suspended sediment sample from the Schoonmaker site—a highly urbanized
medium density residential area—showed unusually high Al, Cd, Cu, Fe, Pb and
Zn concentrations. One municipal sewage flow relief valve was located in this
subwatershed but it was rarely in operation, and no industrial surface
discharges existed (10). A combination of factors may explain these unusually
high levels of pollutants 1. a low flow rate (1.98 cms peak) at Schoonmaker
Creek and long contact time with pollutants, 2. a high degree of connected
impervious areas (54%, ref. 16) and 3. a close enough proximity to the
Industrial Valley of Milwaukee to receive atmospheric fallout from this
source. Another medium density residential area—the Noyes Creek
subwatershed—also showed an increase in Pb and Zn levels over the soil level
but not nearly as high as the suspended sediment sample from Schoonmaker
Creek. The Noyes Creek sampling site was located 8 km north of the
Schoonmaker Creek site and was thus further away from the Industrial Valley of
Milwaukee and also had a lower traffic volume.
Generally, the concentration of an element sorbed on the surfaces of
clay-sized particles is higher than for silt-sized particles because the
amount of element sorbed is proportional to the surface area per unit mass.
In this investigation, this hypothesis was nearly always true with bottom
sediments and soil samples (Tables 1, 4 and 5). However, pollutant sorption
on silt-sized particles relative to clay-sized particles appeared higher for
suspended sediments than for soils and bottom sediments. This may be due, in
part, to the non-uniform distribution of silt-sized particles in the suspended
sediment over the size range of 62 to 4 pm. The particle sizes were probably
34
-------
skewed toward the 4 ym end of this range due to gravity settling of the
coarser silt particles during transport (17). However, the skewed
distribution of silt-sized particles cannot explain cases where the ratio is
>1—i.e., where the element concentration in the silt-sized fraction is higher
than in the clay-sized fraction. This was found for Cu and Zn in the Harbor
samples and Ni in the Noyes Creek, Underwood Creek and Pilgrim Road samples.
The high Ni concentrations in the silt-sized fractions may be due to the way
in which metal abrades from metallic surfaces, resulting in metal chips that
are not as finely divided as those which have passed through a vapor phase
prior to sorption on solid surfaces. Dust and dirt remaining on impervious
surfaces also may have been depleted selectively of its clay-sized fraction by
wind and water erosion. During a storm, dust and dirt on impervious surfaces
are washed into the river and become mixed with the suspended material from
upstream. Thus, if the selective depletion of clay in urban street dust and
dirt is an important factor for the enrichment of Ni in suspended sediments,
then other elements should also show a concentration enrichment in the silt-
sized particles. Aluminum, Pb and Zn concentrations in the silt-sized
fraction of the suspended sediments were found to be higher than in the silt-
sized fraction of bottom sediment or soil, but this enrichment was not as
great as for Ni. The higher concentrations of certain elements in silt-sized
fractions than in clay-sized fractions (without prior chemical or ultrasonic
dispersion treatment in the fractionation procedure) also were observed for
some suspended sediments (18) and on urban street dust and dirt samples (19).
Comparisons of elemental composition of suspended sediments from
different locations and storm events were more difficult to make than for
bottom sediments. Additional factors that must be included for a comparison
of suspended sediments are the time of sampling relative to the peak of the
hydrograph, antecedent rainfall conditions and street sweeping practices. The
ratios of element concentrations in the silt-sized fractions to element
concentrations in the clay-sized fractions were found to be generally higher
in suspended sediment than in bottom sediment and soil. In some cases, metal
concentrations in the silt-sized fractions were found to be greater than in
the clay-sized fractions.
Excluding the anomalously high Pb content of the sample from Schoonmaker
Creek, the average Pb and Cd concentrations for the June 30 and July 18, 1977
suspended sediment samples were 350 yg/g Pb and 1.9 yg/g Cd in clay-sized
particles; 180 ug/g Pb and 0.48 yg/g Cd in silt-sized particles and 290 yg/g
Pb and 1.4 yg/g Cd in the unfractionated sample. The average annual event
loading of Pb and Cd from suspended sediment in the Menomonee River basin to
Lake Michigan was calculated to be about 3000 kg/yr Pb and 15 kg/yr Cd with
~90% of the Pb in the clay-sized fraction. The calculation was based on a
3-yr average suspended solids loading of 10,455,000 kg/yr (11) and assuming
the June 30 and July 18 suspended sediments were representative samples.
Metals in Urban Street Dust and Dirt
Urban street dust and dirt samples were dispersed and analyzed for their
chemical composition in sand-, silt- and clay-sized fractions to provide an
estimate of pollutant they contribute to storm water runoff. Concentrations
of Al, Fe and Mn in the 91st Street and 13th Street dust and dirt samples were
35
-------
lower than in the six mineral soils. However, Al, Fe and Mn concentrations in
sand, silt and clay fractions of the 13th Street samples were lower than in
the 91st Street samples for September 22 and 26, 1977. This relationship was
consistent with surrounding land use. The 91st Street sampling site was
located in the central portion of the Watershed, in a commercial-residential
area nearer to agricultural land use areas. The Al, Fe and Mn concentrations
in dust and dirt at 91st Street were closer to the values found for soils than
was the case at 13th Street. The 13th Street sampling site was located near
the Industrial Valley of Milwaukee and receives atmospheric fallout from it.
An apparent higher Zn and lower Pb loading rate existed at 13th Street than at
the 91st Street sampling site. Other metal concentrations in dust and dirt
samples showed similar ranges at both sites (Tables 4 and 5).
Cadmium, Cr, Cu, Ni, Pb and Zn concentrations in sand, silt and clay
fractions of urban street dust and dirt samples were found to be higher than
the levels found in the respective fractions of the six mineral soils.
Therefore, Cd, Cr, Cu, Ni, Pb and Zn input into the river is expected to
increase as the source of sediment input shifts from agricultural soil to
urban street dust and dirt. When comparing metal concentrations of urban
street dust and dirt samples with bottom sediments, Cr and Ni concentrations
in the clay-sized fraction of the bottom sediments sampled at the Nor-X-way B
site and further downstream were as much as five times greater than the levels
found in the dust and dirt samples. Thus, another other source of Cr and Ni
was present, probably industrial sewage from the Menomonee Falls STP No. 1.
It was noted that Cr, Cu, and Ni concentrations in the coarser fractions
of urban street dust and dirt samples were occasionally greater than the
concentration in the finer-textured material—sand vs. silt and silt vs.
clay. A similar situation was noted in the suspended sediment samples. Also
the pooled standard deviations for Cu and Ni in the sand-sized fraction were
greater than in the silt-sized or clay-sized fractions. These results
indicate that a simple surface sorption model showing a higher concentration
of sorbed element in the fine particle size fraction is insufficient to
account for the concentrations of Cr, Cu and Ni in urban street dust and dirt
samples. However, Cr, Cu and Ni originating as metallic chips or particles
without passing through a vapor phase before being sorbed on to larger
particles could explain the large pooled standard deviations of metal
concentrations in the sand-sized fractions; and the high Cr, Cu and Ni
concentrations in the coarser particles of urban street dust and dirt.
Other elements (Cd, P, Pb and Zn) were found to be less concentrated in
the coarser fractions of urban street dust and dirt samples than in the fine
fraction. The coefficients of variation for Cd, P, Pb and Zn in sand also
were lower than for Cr, Cu and Ni in the sand fraction of the street dust and
dirt samples. Much of the Cd, P, Pb and Zn from combustion of leaded gasoline
and oil lubricants probably passes through a vapor phase prior to sorption on
particle surfaces and tends to be concentrated on particles with largest
surface area per unit mass. Thus, the surface sorption model was adequate to
explain the distribution of Cd, P, Pb and Zn in the sand, silt and clay
fractions of the urban street dust and dirt samples.
The efficiency of street sweeping for the removal of metals and P could
be estimated by comparing total elemental composition of the urban street dust
and dirt with the elemental composition of the fraction removed by street
36
-------
sweepers. Table 7 lists the percentage of elements distributed in each
particle size range. Thus, the elements can be divided into two main groups:
those elements with a high distribution in the sand fraction (Cr, Cu, Fe, Mn,
Ni and Zn), varying from 70 to 87%, and those elements with a lower
distribution in the sand fraction (Al, Cd, P and Pb), varying from 41 to
58%. If street sweeping is efficient in removing all particles of the sand-
sized fraction (>62 ym) but leaving behind silt- and clay-sized particles,
then the removal of Cr, Cu, Fe, Mn and Ni would be more efficient than the
removal of Al, Cd, P, and Pb—of which Cd, P and Pb were considered to be more
hazardous to the environment (1). However, the distribution and removal
efficiencies of elements should be viewed with caution. Two factors which
work counter to each other were used in obtaining the distribution of elements
in the particle sizes. The dust and dirt samples were collected in areas
which were not swept. Therefore, the particle-size distribution of the
samples is biased toward coarser-sized particles because coarse particles are
less readily transported by wind and traffic movement. Antecedent rainfall
conditions also affect particle-size distribution. This bias of particle-size
distribution toward the coarser sizes was probably slight because of the
frequency of rainstorms (Table 3). However, samples also were dispersed by
ultrasound to ensure aggregate breakdown. Complete dispersion would lead to a
greater proportion of the pollutants residing on fine particles. Ideally the
elemental composition of the dust and dirt samples should be analyzed in two
sets, i.e., with and without ultrasound dispersion prior to fractionation.
The Al, Fe and Mn concentrations in urban street dust and dirt samples
were found to be lower than in the six mineral soils, while Cd, Cr, Cu, Ni, Pb
and Zn were higher. Occasionally Cr, Cu and Ni concentrations in the coarse
particle-size fractions of the dust and dirt samples were found to be greater
than the concentration in the fine fraction. This was presumed to result from
their presence in metal chips. Distribution of Cr, Cu, Ni, Fe and Mn was
found to be associated to a greater extent with the sand fraction while Al,
Cd, P, and Pb were associated with finer particles. A similar distribution
pattern of elements in urban street dust and dirt particle size fractions with
Cd, N, P, Pb and Zn in one group and Cr and Cu in another group was reported
by Pitt (19).
Pollutional Classification of Sediments
Presently (January 1980) no national guidelines exist for identifying
polluted bottom sediments or dredge spoils. However, Region V of the U.S.
Environmental Protection Agency (U.S.-EPA) has developed interim guidelines
for classifying Great Lakes harbor dredge spoils into three levels of
contamination: non-polluted, moderately polluted and heavily polluted (Table
8, ref. 20). This classification scheme was based on statistical analysis of
260 dredge samples from 34 Great Lakes harbors which were analyzed for
chemical composition of intact, unfractionated samples. Application of the
guidelines was intended for dredge samples from harbors in this area and not
for samples collected away from the harbor where particle-size distribution
differs from that of harbor samples. However, a guideline for unfractionated
samples cannot isolate the effect of particle-size distribution on elemental
composition of the sediments. For example, suspended sediments are enriched
37
-------
Table 7. Distribution of elements in each particle-size
fraction of urban street dust and dirt samples
Parameter
Distribution, %
Sand
Silt
Clay
p
Pb
Cd
Zn
Cu
Al
Fe
Mn
Cr
Ni
48(17.9)*
58(4.0)
41(6.6)
70(6.0)
81(6.8)
52(20.7)
87(5.4)
78(5.1)
79(7.1)
82(5.2)
23(12.1)
18(5.2)
27(8.4)
13(2.5)
11(6.6)
16(7.2)
8(4.2)
11(3.4)
10(4.8)
11(3.5)
28(6.2)
25(2.3)
31(8.3)
17(6.4)
8(1.5)
32(15.3)
5(1.5)
11(2.1)
11(2.4)
7(2.4)
*Standard deviation in parentheses.
38
-------
Comparison of total elemental composit ion of soils, sediments and urban street dust and dirt with U.S. — EPA
pollutional classification guideline for unfractionated dredge spoils
Sample /samp ling
locat ion*
Moderately polluted
Heavily polluted
Ozaukee sil (Ou)
Mequon sil (Mt)
Hochheim sil (Hm)
Ashkum sicl (As)
Pella sil (Ph)
Theresa sil (Th)
Houghton muck (Ht)
Element, ug/g
Total P
U.S.
420
650
690
998
583
1,250
760
470
1,130
Pb
EPA
40
60
16
17
21
20
21
12
7
Cd
Zn
Dredge Guidelines for
6.0
0.24
0.30
0.14
0.66
0.41
0.14
N.D.
90
200
42
34
72
81
96
70
65
Cu
Pollution
25
50
Soils
27
31
15
49
20
9
34
Cr
Levels
25
75
26
27
29
32
42
22
19
Ni
Fe
of Unf ractionated
20
50
16
22
18
16
14
11
25
17,000
25,000
16,800
27,000
22,900
17,500
24,900
17,400
27,500
Mn
Sample
300
500
916
757
900
159
645
580
311
Bottom Sediment
Little Menomonee River
Donges Bay Road (1)
County Q Road (2)
Road F near Road B (3)
Appleton Avenue (4)
Upper Menomonee River
Friestad (5)
River Lane (6)
Maple Road (7)
Menomonee Falls (8)
Nor-X-way A*** (10)
Nor-X-way B*** (11)
Lily Creek (12)
Dretzka Creek (13)
124th Street (14)
Lower Menomonee River
Capitol Drive (15)
70th Street (16)
Falk Corporation (17)
Harbor (18)
Dousman Ditch
Dousman 1
Dousman 2
Dousman 3
Dousman 4
Dousman 5
Dousman 6
521
1,010
410
307
687
414
333
995
465
1,060
2,470
2,720
2,030
609
235
1,100
370
933
1,870
855
1,190
1,480
847
9
19
17
31
11
14
15
32
82
112
181
174
94
54
45
487
440
376
48
117
73
67
81
0.26
0.43
0.39
0.35
0.42
0.16
0.25
0.32
0.50
0.64
1.0
1.3
0.74
0.68
0.29
9.0
1.4
1.9
2.4
1.2
0.9
0.9
0.6
53
121
105
107
62
43
46
106
286
243
466
475
271
98
90
647
193
233
82
135
109
96
106
11
21
11
17
18
9.8
12
19
27
29
55
63
40
35
14
219
69
23
36
29
26
23
26
11
29
11
14
16
10
12
27
17
56
100
98
51
45
11
187
58
39
18
37
26
20
30
9
20
17
14
8
10
19
13
**
97
**
97
**
8
55
13
23
25
26
18
15
18
12,600
21,000
11,000
11,800
21,600
9,300
13,400
18,400
10,700
12,400
22,500
26,000
23,900
13,700
6,700
22,900
10,900
21,900
8,970
20,300
16,900
15,200
18,600
220
469
522
678
241
303
256
515
365
444
401
776
485
820
240
570
287
411
130
396
217
304
359
-------
Table 8. Continued
Sample/ sampling
location*
Total P
Pb
Cd
Zn
Element ,
Cu
Pg/g
Cr
Mi
Fe
Mn
Suspended Sediment
Upper tlenomonee River
River Lane (673001)
Pilgrim Road (683002)
124th Street (683001)
Little Henomonee River
Donges Bay Road (463001)
Noyes Creek (413011)
Appleton Avenue (413008)
Lower Menomonee River
Underwood Creek (413007)
Honey Creek (413006)"*^"
Honey Creek (413006)
70th Street (413005)"1"1"
70th Street (413005)
Schoonmaker Creek (413010)
Falk Corporation (413004)^
Falk Corporation (413004)
Harbor
13th Street Bridge, 9/22/77
13th Street Bridge, 9/26/77
91st Street, 9/22/77
91st Street, 9/26/77
3,500
1,900
1,790
1,700
1,060
950
1,260
1,180
1,540
1,260
1,270
1,800
1,440
1,100
2,520
134
138
190
140
68
207
163
39
164
57
485
570
298
437
158
1,360
668
115
616
214
362
769
1,227
1.9
N.D.
0.6
0.3
0.5
0.3
1.4
1.4
1.3
2.0
0.7
3.2
3.9
0.7
3.4
2.4
3.6
3.0
5.5
440
352
356
285
478
292
570
420
544
500
268
3,180
856
280
1,560
Urban
548
626
737
1,000
35
45
56
44
41
33
71
62
69
82
66
89
125
63
398
Street Dust
172
281
282
277
37
55
117
56
49
42
70
45
99
103
90
79
106
52
199
and Dirt
37
44
53
32
28
66
67
45
61
50
58
38
50
62
48
56
41
47
142
41
66
53
70
66,000
38,000
36,800
46,400
37,900
31,300
44,800
35,100
36,900
36,480
34,500
162,000
51,000
23,200
89,700
44,800
62,000
68,000
59,500
1,400
987
700
730
582
594
710
573
534
646
574
831
1,080
530
1,410
305
370
504
624
*Numbers or letters in parentheses correspond to sampling sites shown in Fig. 1; for Dousnian Ditch sample
locations see Fig. 2.
**No sample.
***A-before confluence; B-after confluence.
+Samples were collected on July 18, 1977 except where noted.
-H-Samples were collected on June 30, 1977.
N.D. Not detected.
-------
with clay as compared to bottom sediment samples due to selective deposition
of coarse particles. Bottom sediments collected from segments of the river
where water flow rate is lower tend to have higher amounts of clay-sized
particles than at sites where flow rate is higher. An enrichment in the
content of clay-sized particles increases the concentration of surface sorbed
elements in unfractionated samples. This apparent increase in element
concentration of unfractionated samples can be isolated by fractionation of
samples prior to determination of their composition.
The Region V of U.S.-EPA dredge spoil guidelines showed greater tolerance
for polluted samples when the guidelines were applied to the harbor bottom
sediment samples collected in this study. Chromium and Zn in the harbor
bottom sediments were classified as moderately polluted according to the
Region V of EPA scheme but had the highest and second highest concentration of
these metals in the clay-sized fraction.
In general, unfractionated suspended sediment samples had higher total
concentrations of surface-sorbed elements than unfractionated bottom sediments
or surface soils (Table 8). However, many suspended sediment samples if they
were classified on the basis of elemental composition in the clay-sized
fraction.
A more important advantage in analyzing fractionated rather than intact
samples can be seen in the greater precision of locating areas that received
high pollutant inputs, even though the river tends to disperse pollutants from
upstream and carry them downstream. For example, the gradual decline in
elemental composition in the clay-sized fraction of bottom sediments from
Nor-X-way B to Dretzka Creek and the slight increase at Capitol Drive (Tables
4 and 5) would not be detected using unfractionated samples (Table 8). The
relationship of elemental composition in Lily Creek compared to Nor-X-way B
unfractionated bottom sediment samples was the reverse of the relationship in
the clay-sized fraction.
It is recommended that the pollutional guidelines for bottom sediment be
based on elemental composition of the clay-sized fraction of unpolluted bottom
sediments or major soils in the watershed which potentially erode and
contribute to the sediment loading. Comparisons of soil composition with
sediment composition should take into account the greater leaching that
sediments are subject to and the differences in pH and Eh between soils and
sediments.
It was noted that the elemental composition of the clay-sized fractions
of "unpolluted" bottom sediments were lower than soils except for Cd, Pb, and
Zn. It was suspected that all bottom sediment samples received C, Pb and Zn
inputs from atmospheric fallout in sufficient amounts to balance off losses
due to leaching, Eh and/or pH changes. Each of these metals is associated
with vehicular emission.
Dispersibility of Soils
Sediment is the major water pollutant. Soils with the greatest potential
to be dislodged by rain impaction tend to contribute more sediments to the
41
-------
storm water runoff. The finer-textured materials that are dispersed stay in
suspension longer than coarse-textured materials and are transported a greater
distance overland. Once transported into the waterways the finer-textured
materials tend to be the carriers of pollutants. However, clay particles form
aggregates and the aggregates deposit quicker than the dispersed particles of
which they are comprised. Consequently, nutrients and pollutants sorbed by
aggregates are less available to the biota. The dispersibility of soils may
be an indirect method of evaluating the availability of nutrients and
polluants sorbed on the surface as well as the potential of the soil to erode.
Six surface mineral soils, one organic soil, and two subsurface soils
were shaken with water to simulate their dispersion and transport in an
aquatic system. Their particle-size distribution was determined (Table 9).
The ratio of the particle size distribution by shaking and by ultrasonic
treatment provided a number which measures the ease of dispersibility of a
particular soil (Table 10). This number could serve as a ranking factor for
nutrient or pollutant availability to the lake biota and for the potential of
the soil to erode.
A significant negative linear correlation between the dispersion ratio of
the clay-sized fraction and the organic C content of six surface mineral soils
existed for short periods of shaking: P - 5% for 16 hr and P = 1% for 4 hr
shaking (Table 11). Organic matter facilitates aggregation of soil particles
(21). As the length of the agitation period increased (32 to 128 hr), the
binding properties of organic matter were reduced and the linear correlation
was less significant. It was thought that the six surface mineral soils were
too homogeneous and represented a narrow range of soil organic carbon (1.2 to
5.7% C, Table 11). Therefore, two subsurface soils with lower organic carbon
content (^0.5%) and one organic soil (44% organic C) were included in the
dispersibility study. The significance levels of the linear correlations
remained unchanged when the two subsurface soils were included in the analysis
for the 4- and 16-hr dispersion treatments. Linear correlation between
organic C and dispersion ratio became insignificant when the organic soil also
was included in the analysis.
The correlation between dispersibility and organic carbon content also
was analyzed for "log x" and "1/x" relationships. In both cases the level of
significance was high for the 1-, 4- and 16-hr shaking treatments. When two
subsurface soils were included in the analysis between organic C and clay-
sized particle dispersion ratio, it was noted that the "1/x" relationship was
consistently better correlated than "log x;" which was consistently better
correlated than "x" (linear) for dispersion treatments of 0.5 to 64 hr.
Similarly, when the organic soils as well as the two subsurface soils were
included in the correlation analysis, the "1/x" relationship was consistently
better correlated than "log x", which was better than "x" for 4 to 64 hr
dispersion treatments. Thus "1/x" and "log x" relationships between clay-
sized particle dispersion ratio and organic C content of soil were more
applicable to a wider diversity of soils than the linear relationship. The
"1/x" relationship predicts that increasing organic C content of soil—e.g.,
with the addition of sewage sludge—would benefit the aggregate stability of
soils with low organic C more than soils with high organic C.
A low correlation between soil erodibility and organic C content of soils
with >4% organic C was noted by Wischmeier et al. (22). The preceding results
42
-------
Table 9. Dispersability, by shaking, of soils in the Menomonee River Watershed
Amount of fraction dispersed, %
Time of
shaking, hr
0.5
1
4
16
32
64
128
Particle-size
fraction
Sand
Silt
Clay
Sand
Silt
Clay
Sand
Silt
Clay
Sand
Silt
Clay
Sand
Silt
Clay
Sand
Silt
Clay
Sand
Silt
Clay
Ozaukee
sil
61.7
37.0
1.2
51.8
46.3
1.9
48.4
48.0
3.6
37.6
57.4
5.0
41.7
52.2
6.1
39.8
52.5
7.7
38.8
54.1
7.1
Ozaukee
subsoil
65.7
29.3
4.9
60.4
32.1
7.4
57.0
31.6
11.5
50.5
30.2
19.3
49.2
29.5
21.3
45.4
27.6
27.0
43.7
29.7
26.6
Mequon
sil
77.0
22.2
0.8
63.3
35.3
1.4
64.6
32.3
3.0
58.5
37.0
4.4
49.8
42.3
7.9
54.1
34.5
11.4
49.4
38.5
12.0
Meq uon
subsoil
73.4
24.1
2.5
69.0
27.3
3.6
68.0
26.5
5.6
64.8
25.4
9.7
66.5
23.0
10.5
67.6
19.4
13.0
66.1
21.5
12.4
Hochheim
sil
70. ,6
27.6
1.8
56.8
41.2
2.0
55.5
40.3
4.2
46.6
46.8
6.6
46.0
45.8
8.2
44.8
44.4
10.8
36.5
50.5
13.0
Ashkum
sicl
64.8
33.5
1.7
45.2
52.4
2.4
52.7
43.3
4.0
45.6
47.9
6.4
45.2
46.5
8.3
39.5
50.2
10.2
36.1
51.5
12.4
Pella
sil
61.9
35.7
2.3
51.4
45.6
3.0
50.6
44.6
4.8
42.3
50.7
7.0
32.7
58.6
8.7
32.9
56.7
10.4
32.7
57.7
9.6
Theresa
sil
53.3
45.3
1.4
37.7
60.2
2.1
43.0
54.0
3.0
34.3
61.6
4.1
35.3
60.2
4.6
35.6
58.2
6.1
30.4
62.7
6.9
Houghton
muck
86.1
11.9
2.0
81.8
15.0
3.2
64.6
29.1
6.3
66.8
24.4
8.8
58.0
28.3
13.6
Blanks indicate no data.
-------
Table 10. Dispersion ratio of the clay-sized fraction (shaking/ultrasonic)
Soils*
Time of
shaking, hr
0.5
1
4
16
32
64
128
Ozaukee
sil
0.06
0.10
0.19
0.26
0.32
0.40
0.37
Ozaukee
subsoil
0.15
0.22
0.35
0.59
0.65
0.82
0.81
Mequon
sil
0.03
0.05
0.10
0.15
0.27
0.39
0.41
Mequon
subsoil
0.15
0.21
0.33
0.57
0.62
0.76
0.73
Hochheim
sil
0.07
0.07
0.16
0.24
0.30
0.40
0.48
Ashkum
sicl
0.05
0.07
0.11
0.18
0.24
0.29
0.35
Pella
sil
0.06
0.08
0.13
0.19
0.24
0.28
0.33
Theresa
sil
0.08
0.13
0.19
0.26
0.29
0.38
0.43
Houghton
muck
0.03
0.05
0.10
0.15
0.22
*0rganic carbon contents of soils are: Ozaukee sil - 1.8%, Ozaukee subsoil - 0.50%, Mequon sil - 4.5%,
Mequon subsoil - 0.49%, Hochheim - 2.5%, Ashkum sicl - 5.7%, Pella sil - 3.4%, Theresa sil - 1.2% and
Houghton muck - 44.2%.
Blanks indicate no data.
Table 11. Linear, inverse and log correlation coefficient between soil organic carbon content and clay-sized fraction
dispersion ratio
6 Surface soils +
Time of
shaking, hr
0.5
1
4
16
32
64
6
X
-0.762*
-0.775*
-0.935**
-0.883**
-0.763*
-0.580
Suface soils
1/x
0.790*
0.928**
0.904**
0.853*
0.642
0.492
log1Q x
-0.797*
-0.874**
-0.951**
-0.898**
-0.730*
-0.553
6 Surface
X
-0.804**
-0.822**
-0.859**
-0.789**
-0.740*
-0.736*
soils + 2
1/x
0.981**
0.987**
0.985**
0.989**
0.976**
0.963**
subsoils
log1Q x
-0.940**
-0.959**
-0.970**
-0.932**
-0.902**
-0.892**
x
-0.612
-0.547
-0.569
-0.555
+ 1 organic
1/x
0.968**
0.967**
0.967**
0.953**
2 subsoils
soil
1°810 x
-0.930**
-0.879**
-0.879**
-0.867**
*Significant at P = 0.05.
**Significant at P = 0.01.
Blanks indicate no data.
-------
may provide an improvement to the Wischmeier et al. equation (23 and 24) for
predicting soil credibility. The equation assumed a negative linear
relationship between soil erodibility and organic C content in the range of
0.9 to 4% organic C. The present findings also showed a significant linear
correlation in this range for 4 to 16 hr shaking treatments. However, when
soils with a wider range of organic C (0.5 to 44% C) were included, a "1/x"
relationship provided better correlation than the linear relationship.
From the above results, it is clear that soils eroding to waterways do
not disperse completely during the early stage of transport. For example,
after 4 hr of shaking 19% of the clay in the Ozaukee soil was dispersed into
clay-sized particles while the remaining 81% of the clay remained intact as
silt-sized or larger aggregates (calculated from Table 10). Settling ponds
that capture silt-sized or larger particles and aggregates while allowing
clay-sized particles to pass through are able to capture 80 to 90% of the
clays for the six mineral surface soils for conditions simulated by a 4 hr
shaking period. In certain parts of the upper reaches of the Menomonee River,
e.g., behind dams (Menomonee Falls Dam and Lily Creek) and widening areas in
the river (Dretzka Creek and 124th Street), high contents of clay-sized
particles (33 to 58%) were found in bottom sediments. For sediments deposited
in the Milwaukee Harbor there is a longer transport distance for eroded
materials from the upper part of the watershed and a correspondingly greater
dispersion of aggregates during transport. Dispersed clay-sized particles
will remain in suspension and exit the harbor area; hence, the clay-sized
fraction of the Milwaukee Harbor bottom sediments are quite low (2 to 14%,
ref. 8).
One of the methodological problems with the procedure for estimating soil
erodibility is soil preparation. Soil samples were forced-air dried and
ground to pass a 10-mesh sieve prior to analysis. Such treatment undoubtedly
affects the soil aggregates. However, without some drying and sieving,
samples would be less uniform, and the reproducibility of the results would be
diminished. The organic soil and subsurface soils were not forced-air dried
to avoid conversion of the materials into hydrophobic aggregates, although
they were sieved prior to the dispersion treatment. The end-over-end shaking
treatments to simulate transport of soil in an aqueous medium was not fully
investigated, i.e., the optimum soil:water ratio and rotation speed were not
evaluated. A 1:10 w:v soil:water ratio as well as the shorter time period (1
to 16 hr) may approximate the onset of erosion conditions but not for long
term transport to deposition and resuspension of bottom sediments in the
river.
Extractability of Metals and Phosphorus
During aqueous transport of soils and sediments, metals and P may be
desorbed from surfaces. The amount of element that could be desorbed rapidly
(in minutes to a few hours) into the aqueous phase was estimated by analyzing
the supernate obtained from end-over-end and ultrasound dispersion
procedures. The amount of metals and P exracted from the solid surface during
the laboratory dispersion and fractionation procedures was also a measurement
of the error in determining element concentrations in solids caused by
desorption. Those elements with the greatest tendency to be desorbed from
surfaces would show the greatest error. This may have an effect on the
biological availability of specific elements. Thus, the supernates from the
45
-------
dispersion treatments (end-over-end shaking and ultrasound) were collected and
analyzed (Table 12).
Supernate samples from the shaking treatment of the six mineral soils for
1 to 16 hr were digested without prior filtration; while supernate samples
from the 32 to 128 hr shaking treatment were passed through a 0.4 pm
polycarbonate membrane filter using an all plastic (polycarbonate) filter
appratus. Element concentrations (normalized to 1 g of solids) in the
unfiltered supernate following 1 to 16 hr shaking treatments were found not to
be significantly different from each other for the different shaking periods
as shown by the pooled standard deviations of the individual means (Table 12).
Thus, the mean amounts of elements removed from the solids by 1 to 16 hr
shaking treatments for each of the six soils were computed and designated as
the "unfiltered shaking treatment group" (Table 12). Similarly, the mean
value of elements removed from the solids by 32 to 128 hr shaking treatments
for each of the six soils were computed and designated as the "filtered
shaking treatment group" (Table 12). The amount of Fe was found to be higher
in the supernates of the "unfiltered group" than in the "filtered group."
This was attributed to the presence of microcrystalline and amorphous Fe
oxides of low specific gravity that were not removed from the supernate by
centrifugation but were removed by filtration. The removal of
microcrystalline and amorphous Fe oxide by filtration did not significantly
reduce the amount of Al, Cu, Mn and P found in the supernate even though some
elements may have been sorbed by the Fe oxides. The amount of Zn extracted
apparently increased with dispersion time, i.e., a greater dissolution of Zn
in the "filtered" than "unfiltered" samples.
The amount of element extracted per gram of sample for "filtered" and
"unfiltered" supernate samples of ultrasound treated soils were determined
(Table 12). Ultrasound treatment of soils represents an extreme erosive
condition as compared to end-over-end shaking. The amount of Fe extracted by
the supernate per gram of sample was greater for ultrasound-treated than for
shaken soils; and greater for "unfiltered" than for "filtered" samples.
The relative amount of element extracted from the solids during
dispersion and fractionation was calculated by taking the ratio of the total
element in the supernate to the total element in the solids (sand-, silt- and
clay-sized particles). This ratio was computed from values in Tables 1, 8 and
12. The ratio of element loss from each of the six mineral soils were pooled
together to obtain a mean value and standard deviation for each element (Table
13). Ratios of element loss from bottom sediments were determined similarly
(Table 13). This was done on the assumption that the amount of element
released into the supernate was proportional to the amount present at the
solid surfaces. Thus, the ratio would remain relatively constant even though
the total amount of a given element extracted and the amount present in the
solid phase varied for different samples. The amount of element removed from
the solid phase into the supernate during ultrasound fractionation of soils
varied from 1 to 8%; and from 0.3 to 2% from bottom sediments (Table 13).
Lower ratios of element loss from bottom sediments were probably caused by
prior release of the more soluble components during their transport in the
river.
If a sequence of increasing extractability were constructed from the
"filtered supernates" of soils receiving a shaking treatment (Table 13) the
46
-------
Table 12.
Elements extractable from soils by end over end shaking and ultrasonic treatment
Element, yg/g
Treatment
Shaking, unfiltered
Shaking, filtered
Ultrasonic, unfiltered**
Ultrasonic, filtered**
Sample
Ozaukee silt loam
Mequon silt loam
Hochheim silt loam
Ashkum silty clay loam
Pella silt loam
Theresa silt loam
Pooled std. dev.
Ozaukee silt loam
Mequon silt loam
Hochheim silt loam
Ashkum silty clay loam
Pella silt loam
Theresa silt loam
Pooled std. dev.
Ozaukee silt loam
Mequon silt loam
Hochheim silt loam
Ashkum silty _clay loam
Pella silt loam
Theresa silt loam
Ozaukee silt loam
Mequon silt loam
Hochheim silt loam
Askhum silty clay loam
Pella silt loam
Theresa silt loam
Total P
21.6
31
3.2
37
4.5
10.7
2.97
19
26
3.0
40
2.8
8.1
6.7
53
42
15
75
16
52
57
63
19
42
31
26
Pb*
0.079
0.078
0.022
0.010
0.035
0.027
0.044
0.67
0.85
0.097
0.017
0.040
0.005
0.077
0.047
Cd*
0.005
0.013
0.007
0.006
0.009
0.002
0-003
0.02
0.03
0.02
0.02
0.020
0.015
0.016
0.011
0.020
0.017
Cu
0.15
0.22
0.27
0.12
0.36
0.07
0.042
0.22
0.33
0.13
0.35
0.15
0.11
0.062
1.4
0.67
0.25
0.79
0.48
0.66
1.07
1.00
0-33
0.67
0.33
0.27
Cr*
0.022
0.031
0.017
0.032
0.026
0.028
0.017
0.67
0.24
0.24
0.32
0.44
1.34
0.57
0.33
0.45
0.12
0.23
0.40
Ni*
0.011
0.040
0.022
0.034
0.023
0.037
0.017
0.64
0.60
0.12
0.07
0.17
0.36
0.22
0.14
0.15
0.07
0.06
0.17
Zn
0.22
0.25
0.24
0.24
0.15
0.19
0.09
4.19
1.25
0.49
0.64
3. 31
1.94
0.86
2.05
0.62
0.59
0.57
0.58
2.35
2.2
1.1
1.0
0.67
0.47
0.93
Al
29
24
37
27
10
29
11.8
36
69
57
53
89
33
12
604
119
228
178
156
1,270
Fe
19
17
27
13
12
18
5.5
3.3
9.1
6.3
2.2
4.0
2.7
5.0
487
127
164
152
161
931
71
35
31
32
24
37
Mn
0.44
0.41
0.67
0.20
0.26
0.47
0.10
0.42
4.06
1.04
0.31
0.20
0.27
1.6
29
16
12
1.6
2.9
33
12
3.1
4.7
0.25
0.40
5.3
*Analyzed by graphite furnace AAS.
**Samples not replicated hence no pooled standard deviation for individual means.
Blanks indicate that concentrations are below detection limit.
-------
Table 13. Extractability (%) of elements contained in soils and bottom
sediments as affected by different dispersion techniques
Dispersion technique
Shaking
Element Unfiltered
P 2.0 ± 1.2(6)*
Pb
Cd
Cu 0.9 ± 0.6(6)
Cr
Ni
Zn 0.4 ± 0.2(6)
Al 0.1 ± 0.06(6)
Fe 0.08 ± 0.03(6)
Mn 0.04 ± 0.05(6)
Pb
Cd
Cu
Cr
Ni
Zn
Al
Fe
Mn
Filtered
Soils
1.9 ± 1.2(6)
0.2 ± 0.18(6)
2.6 ± 1.5(6)
0.9 ± 0.2(6)
0.1 ± 0.03(6)
0.2 ± 0.1(6)
1.8 ± 1.4(6)
0.3 ± 0.07(6)
0.02 ± 0.01(6)
0.2 ± 0.2(6)
Bottom sediment
0.06 ± 0.07(4)
0.7 ± 0.1(4)
0.04 ± 0.02(4)
0.3 ± 0.01(4)
0.03 ± 0.03(4)
0.3 ± 0.2(4)
Ultrasonic
Unfiltered
5.5 ± 3.2(6)
4.9 ± 2.4(2)
7.8 ± 5.2(4)
3.3 ± 2.2(6)
2.0 ± 2.0(6)
1.9 ± 1.4(6)
2.0 ± 1.7(6)
1.3 ± 1.2(6)
1.8 ± 1.9(6)
2.3 ± 2.0(6)
1.5 ± 1.8(13)
2.0 ± 1.0(10)
1.0 ± 5.0(14)
0.5 ± 0.4(13)
1.6 ± 1.0(10)
1.0 ± 1.0(15)
0.4 ± 0.3(14)
0.3 ± 0.3(15)
0.5 ± 0.3(15)
Filtered
5.0 ± 1.9(6)
0.3 ± 0.2(6)
7.1 ± 4.0(6)
2.4 ± 1.0(6)
1.3 ± 0.7(6)
0.8 ± 0.4(6)
2.0 ± 1.8(6)
0.2 ± 0.01(6)
0.6 ± 0.5(6)
*Figure in parenthesis corresponds to number of samples.
48
-------
order would be Fe < Cr - Mn = Nl = Pb _< Al _£ Cu < Zn _< P _< Cd (Table 14).
With the exception of Nl, this sequence resembles the decreasing order of
sorptlon of divalent metal Ions on to Fe gel which Is Pb > Cu > Zn > Ni > Cd
(25). The sequence of increasing order of extractability also corresponds
closely with the increasing order of first hydrolysis constants of metals with
the exception of Ni; Fe(III) < Cr(III) < Al < Cu(Il) < Ni(II) < Zn < Cd
(26). Extractability ratios of "unfiltered supernates" from soils receiving a
shaking treatment show a similar trend to the "filtered supernates"—Mn, Fe
and Al in the less extractable group and Zn, Cu and Pb in the more extractable
group. An increasing order of extractibility also was constructed for
ultrasound treated soils and bottom sediments (Table 14). Although the mean
values of the extractability ratios were lower for bottom sediment supernates
than for soil supernates after ultrasound treatment, only Fe and Mn ratios of
bottom sediments were significantly different from soils (90% confidence level
using Student's T test). Reducing conditions in the bottom sediments augment
the dissolution of Mn and Fe. It is important to note however, that the
dissolution and desorption of elements from bottom sediments is not
complete. If bottom sediments were resuspended, further release of elements
to the overlying water would occur with Cd and P being the most extractable.
This is of particular interest because Cd and P are considered to be very
important pollutants by the U.S.-Canada IJC (1). Although the concentration
of Cd was low in soil and sediments, it consistently had the highest
extractability ratio for soil and bottom sediments with either ultrasound or
end-over-end shaking dispersion. Thus, a higher proportion of total Cd and P
in the solids can be extracted by water in a relatively short period of time
(minutes to a few hours) as compared to other elements.
Correlation Analysis
Concentrations of elements in soil, bottom and suspended sediments and
their supernatant liquids may be correlated together due to a common origin or
association with a common source. To determine this, correlation coefficients
and plots were obtained for pairs of different elements in each particle-size
fraction of soils, bottom and suspended sediments and supernates; and the
amounts of elements recovered in the supernate were compared with their
concentrations in the solid phase of the soils and bottom and suspended
sediments. High correlation coefficients were considered significant if the
plots of data points were not controlled by a few points.
Six soils were considered to be too small a sample for making
generalizations about soil properties, however, they were useful as references
and for comparisons with bottom and suspended sediments. No significant
correlations existed between different elements in the sand-, and silt-sized
fractions of bottom and suspended sediments. Cadmium, Pb and Zn (Fig. 3) and
Cr, Cu and Ni were correlated significantly with each other in the clay-sized
fractions of the bottom sediments. Cadmium, Pb and Zn in the clay-sized
fractions of suspended sediments also were correlated with each other (Fig.
3). These correlations were not found in soils which suggests that soils were
not the primary sources of these metals, and vehicular emission and/or
industrial effluent contributed to their presence in the clay-sized fractions
of the sediments. The area surrounding several sediment sampling sites had
fairly high density automobile traffic: near well-traveled roads (Road F near
B and Nor-X-way A); parking lots (Dousman Ditch); urban areas with a higher
degree of impervious areas (Noyes Creek and Schoonmaker Creek); an area with
49
-------
Table 14. Increasing order of extractability of various elements using
different dispersion techniques
Dispersion technique
Order of extractability
Shaking, unfiltered
filtered
Ultrasonic, unfiltered
filtered
Shaking, filtered
Ultrasonic, unfiltered
Soils
Mn < Fe < Al < Zn < Cu < P
Fe < Cr - Mn - Ni _< Pb _< Al _< Cu < Zn < P < Cd
Al _< Fe - Ni - Cr - Zn - Mn < Cu < Pb < P < Cd
Fe - Pb _< Mn _< Ni < Cr < Zn _< Cu < P < Cd
Bottom sediment
Fe _< Cr £ Pb < Mn = Zn < Cd
Fe < Al < Mn - Cr < Cu - Zn < Pb < Ni < Cd
50
-------
SUSPENDED SEDIMENTS
1.0*
t - 0.819
~ y = 0.152 + 0 0038X
3.0+ R - 32
BOTTOM SEDIWNTS
36 0+
r - 0.727
y . 0.444 + 0.0096X
"3 18.0*
:a
s :
9 0+
- 5.85*32
.0. 26
1200. 2400.
600. 1800 3000.
4.0+
I r - 0.906
y - -0.293 + 0.0036X
3.0+ n " 32
r • 0.784
y - - 2 23 + 0 0101X
400. 100
200. 600. 1000.
Zn, ug/g
£ 300
r • 0.788
y - -30.2 + 0 675X
n - 32
400. 300.
200. 600. 1000.
•343
0 + 5+5
Fig 3. Simple correlations between concentrations of lead and cadmium, zinc and cadmium and zinc and lead for the clay fractions of suspended and
bottom sediments of the Menomonee River.
51
-------
large amounts of connected impervious surfaces in its immediate surrounding
(Pilgrim Road), indicating that vehicular emission was probably the major
source of these metals in the sediment samples.
All element concentrations from ultrasound "unfiltered" supernates of
bottom sediments and soils also were pooled together as a single group of 37
data points for correlation analysis. Only Al and Fe in the supernates were
correlated significantly with each other (r = 0.895). It was concluded that
soils and sediments derived from soil were the major contributors of Al and Fe
to the supernatant solutions which accounts for their degree of correlation.
No significant correlations were found between elements extracted in the
supernates and elemental concentration in the clay-sized fraction. The lack
of correlation is probably due to the diverse sources of pollutants in an
urban setting and the diverse solubilities of the elements.
52
-------
REFERENCES
1. Pollution from Land Use Activities Reference Group. Environmental
Management Strategy for the Great Lakes Systems. International Joint
Commission, Windsor, Ontario, 1978.
2. Southeastern Wisconsin Regional Planning Commission. A Comprehensive
Plan for the Menomonee River Watershed. Vol. I: Inventory Findings and
Forecasts. Waukesha, Wisconsin, October 1976. p. 71.
3. Southeastern Wisconsin Regional Planning Commission. A Comprehensive
Plan for the Menomonee River Watershed. Vol. 1: Inventory Findings and
Forecasts. Waukesha, Wisconsin, October 1976. p. 157.
4. Genrich, D. A. and J. M. Bremner. A Re-evaluation of the Ultrasonic
Vibration Method of Dispersing Soils. Soil Sci. Soc. Amer. Proc. 36:944-
947, 1972.
5. U.S. Environmental Protection Agency. Manual of Methods for Chemical
Analysis of Water and Wastes. U.S. Environmental Protection Agency
Report No. EPA-625/6-74-003. Environmental Research Center, Cincinnati,
Ohio, 1976.
6. Ediger, R. D. Atomic Absorption Analysis with the Graphic Furnace Using
Matrix Modifications. Atomic Absorption Newsletter 14:127-130, 1975.
7. Tabatabai, M. A. and J. M. Bremner. Use of the Leco Automatic 70 Second
Carbon Analyzer for Total Carbon Analysis in Soils. Soil Sci. Soc. Amer.
Proc. 34:608-610, 1970.
8. Bannerman, R., J. G. Konrad and D. Becker. Effects of Tributary Inputs
on Lake Michigan during High Flows. Final Report of the Menomonee River
Pilot Watershed Study, Vol. 10, U.S. Environmental Protection Agency,
1979.
9. Simsiman, G. V., J. Goodrich-Mahoney, G. Chesters and R. Bannerman. Land
Use, Population and Physical Characteristics of the Menomonee River
Watershed. Part III: Description of the Watershed. Final Report of the
Menomonee River Pilot Watershed Study, Vol. 2, U.S. Environmental
Protection Agency, 1979.
10. Southeastern Wisconsin Regional Planning Commission. A Comprehensive
Plan for the Menomonee River Watershed. Vol. I: Inventory Findings and
Forecasts. Waukesha, Wisconsin, October 1976. pp. 236-247.
53
-------
11. Bannerman, R. B., J. G. Konrad, D. Becker and G. V. Simsiman. Surface
Water Monitoring Data. Part I: Quality of Runoff from Mixed Land
Uses. Final Report of the Menomonee River Pilot Watershed Study, Vol. 3,
U.S. Environmental Protection Agency, 1979.
12. Shaheen, D. G. Contribution of Urban Roadway Usage to Water Pollution.
U.S. Environmental Protection Agency EPA-600/75-004, 1975.
13. Sylvester, R. 0. and F. B. Dewalle. Character and Significance of
Highway Runoff Waters: A Preliminary Appraisal. Dept. of Civil
Engineering, Univ. of Washington, Seattle, Washington, 1972.
14. Wisconsin Department of Natural Resources. NR 101 File Reports, 1975.
15. Lagerwerff, J. V. and A. W. Specht. Contamination of Roadside Soil and
Vegetation with Cadmium, Nickel, Lead and Zinc. Environ. Sci. Technol.
4:583-586, 1970.
16. Konrad, J. G. and G. Chesters. Menomonee River Pilot Watershed Study;
Summary Pilot Watershed Report. Submitted to PLUARG Task Group C (U.S.),
Activity 2. Windsor, Ontario, May 1978.
17. U.S. Geological Survey. Water Resources Data for Wisconsin, WateriYear
1976. U.S. Geological Survey, Madison, Wisconsin, 1977. pp. 144-177.
18. Armstrong, D. E., J. R. Perry and D. E. Flatness. Availability of
Pollutants Associated with Suspended or Settled River Sediments which
Gain Access to the Great Lakes. Final Report of the Menomonee River
Pilot Watershed Study, Vol. 11, U.S. Environmental Protection Agency,
1979.
19. Pitt, R. Demonstration of Nonpoint Pollution Abatement through Improved
Street Cleaning Practices. U.S. Environmental Protection Agency,
Municipal Environmental Research Laboratory, Cincinnati, Ohio, 1979.
20. Bowden, R. J. A Study to Develop Guidelines for the Evaluation of Harbor
Sediments. Great Lakes Surveillance Branch, U.S. Environmental
Protection Agency, Region V. Chicago, Illinois, 1976. Unpublished
report.
21. Harris, R. F., G. Chesters and 0. N. Allen. Dynamics of Soil
Aggregation. Adv. in Agron. 18:107-169, 1966.
22. Wischmeier, W. H., C. B. Johnson and B. V. Cross. A Soil Erodibility
Nomograph for Farmland and Construction Sites. J. Soil Water
Conservation 26:189-193, 1971.
23. Wischmeier, W. H. and D. D. Smith. Predicting Rainfall Erosion Losses—A
Guide to Conservation Planning. U.S. Department of Agriculture Handbook
No. 537, 1978.
24. Wischmeier, W. H. Soil Erodibility by Rainfall and Runoff. In:
Erosion: Research Techniques, Erodibility and Sediment Delivery. T. J.
Toy (ed.). Geo Abstract Ltd., Norwich, England, 1977.
54
-------
25. Kinniburgh, D. G., M. L. Jackson and J. K. Syers. Adsorption of Alkaline
Earth, Transition and Heavy Metal Cations by Hydrous Oxide Gels of Iron
and Aluminum. Soil Sci. Soc. Amer. Proc. 40:769-779, 1976.
26. Stumm, W. and J. J. Morgan. Aquatic Chemistry: An Introduction
Emphasizing Chemical Equilibria in Natural Waters. Wiley-Interscience,
New York, N.Y., 1970. pp. 168, 484.
55
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-905/4-79-029F
3. RECIPIENT'S ACCESSION-NO.
Dispersibility of Soils and Elemental Composition of
Soils, Sediments and Dust and Dirt From The Menomonee
River Watershed
5. REPORT DATE
December 1979
6. PERFORMING ORGANIZATION CODE
AUTHOR(S)
A. Dong, G. Chesters and G.V. Simsiman
8. PERFORMING ORGANIZATION REPORT NO.
PERFORMING ORGANIZATION NAME AND ADDRESS
Wisconsin Water Resources Center
University of Wisconsin-Madison
Madison, Wisconsin 53701
10. PROGRAM ELEMENT NO.
A42B2A
11. CONTRACT/GRANT NO.
R005142
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Great Lakes National Program Office
536 South Clark Street, Room 932
Chicago, Illinois 60605
13. TYPE OF REPORT AND PERIOD COVERED
Final -Report 1974-1978
14. SPONSORING AGENCY CODE
U.S. EPA
15. SUPPLEMENTARY NOTES
University of Wisconsin-Water Resources Center and Southeastern Wisconsin Regional
Planning Commission
16. ABSTRACT
This project was in support of the U.S./Canada Great Lakes Water Quality Agreement to
direct the International Joint Commission to conduct studies of the impact of
land use activities on the water quality of the Great Lakes Basin and to recommend
remedial measures for maintaining or improving Great Lakes water quality.
17.
KFY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Bottom sediments
Ultrasound dispersion
Pollutants
Total solids
Watershed
Urban
b.IDENTIFIERS/OPEN ENDED TERMS
COS AT I Field/Group
13. DISTRIBUTION STATEMENT
Document is available to the public through
the National Technical Information Service,
------- |