United States
Environmental Protection
Agency
Environmental Research
Laboratory
Duluth MN 55804
EPA 600 3 80 063
July 1980
Research and Development
Determination of
Atmospheric
Phosphorus
ion to
Michigan
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4, Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9 Miscellaneous Reports
This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal spe-
cies, and materials. Problems are assessed for their long- and short-term influ-
ences. Investigations include formation, transport, and pathway studies to deter-
mine the fate of pollutants and their effects. This work provides the technical basis
for setting standards to minimize undesirable changes in living organisms in the
aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfieid, Virginia 22161.
-------
EPA-600/3-80-063
July 1980
DETERMINATION OF ATMOSPHERIC PHOSPHORUS ADDITION
TO LAKE MICHIGAN
by
S. J. Eisenreich
P. J. Emmling
A. M. Beeton
Center for Great Lakes Studies
University of Wisconsin-Milwaukee
Milwaukee, Wisconsin 53201
Grant No. R 803238
Project Officer
Michael D. Mullin
Large Lakes Field Station
Environmental Research Laboratory-Duluth
Grosse He, Michigan 48138
ENVIRONMENTAL RESEARCH LABORATORY-DULUTH
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
DULUTH, MINNESOTA 55804
-------
DISCLAIMER
This report has been reviewed by the Environmental Research Laboratory-
Duluth, U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or recommendation
for use.
11
-------
FOREWORD
The accelerated rate of eutrophication, or aging, of the Great Lakes is
a topic that has been the object of extensive research and study. Increased
concentrations of phosphorus in the lakes have been identified as making a
major contribution to this undesirable process. It is necessary to have an
understanding of the modes of interaction between the lake ecosystems and
phosphorus to determine more intelligently the required steps to minimize
man's impact on this valuable resources.
This report presents the results of a detailed study to determine the
amount of phosphorus entering Lake Michigan from atmospheric sources. Sea-
sonal as well as variations in geographical deposition rates are determined.
Michael D. Mullin, Ph.D.
Project Officer
Large Lakes Research Station
Environmental Protection Agency-Duluth
Grosse He, Michigan
i* •
11
-------
ABSTRACT
Bulk precipitation was collected on a monthly basis in the Lake Michigan
basin for 18 months during 1975-1976 to determine atmospheric P loadings to
Lake Michigan. The sampling network consisted of bulk collectors at 23 land
stations and 2 in-lake buoys located off urban and rural areas. Annual TP
loading to Lake Michigan for 1976 based on loading rates of 0.184 and 0.303
vg/onVmo for the north and south basins, respectively was 1.69xl06 Kg/year,
representing ^ 16% of the total P budget. North and south basins showed
different TP loading properties with ~ 62% depositing in the south basin.
Seasonal variations in loading rates showed maxima in late spring and early
suomer, and minima in winter. In-lake buoy samplers yielded enchanced load-
ing of TP compared to land stations. Wet-only precipitation concentrations
for TP averaged 15% of bulk deposition values. Weighted-average TP con-
centrations in bulk precipitation were 0.050 and 0.064 mg/1 in the north and
south basins, respectively. Atmospheric TP loading was closely correlated
to Ca and Mg loading, independent of precipitation amount, and exhibited
a large dry fall component. Wind-blown soil and re-entrained dust are believed
to be the major sources of atmospheric phosphorus addition to Lake Michigan.
Annual loadings (1976) of various chemical components to Lake Michigan
in units of 106 Kg are:
TP, 1.69; TOG, 153.5; Si 7.99; S04, 135.4; Cl, 82.5; Ca, 103.4; Mg, 22.5;
Na, 16.2; K, 9.60; participates, 818.5.
IV
-------
CONTENTS
Foreword ill
Abstract iv
Figures vi
Tables vii
Acknowledgements vi i i
Section
1. Introduction 1
2. Conclusions '. 2
3. Recommendations * 3
4. Experimental Procedures 4
Description of study area 7
Precipitation sampling 9
Chemical analysis 11
5. Results and Discussion 12
Loading calculation 12
Atmospheric loading to Lake Michigan 13
Loading rate comparisons 17
Estimated present and future phosphorus trends.... 22
Spatial and temporal variations in loading rates.. 25
Precipitation composition relationships 30
References 38
Appendix 4l
-------
FIGURES
Number
1 Location of Lake Michigan Precipitation Collection Sites. ,_
2 Comparison of 1976 and Historical Precipitation Amounts at 15
Atmospheric Sampling Sites near Lake Michigan - 1976.
3 Atmospheric Loading Rates of Total and Total Dissolved 16
Phosphorus at Precipitation Sampling Sites near Lake
Michigan - 1976.
4 Atmospheric Loading Estimates of Phosphorus, Organic Carbon, 26
Silica, Sulfate, Chloride, Calcium, Magnesium, Sodium,
Potassium and Particulates to the Southern and Northern
Basins of Lake Michigan - 1976.
5 Seasonal Variations in Atmospheric Loading Rates Measured at 27
Beaver Island, Michigan, Milwaukee, Wisconsin and Chicago,
Illinois - 1976.
6 Seasonal Variations in Atmospheric Phosphorus Loading Rates 28
to the Southern and Northern Basins of Lake Michigan - 1976.
7 Seasonal Variations in Atmospheric Loading Rates of Calcium and 29
Silica to the Southern and Northern Basins of Lake Michigan -
1976.
8 Seasonal Variations in Atmospheric Loading Rates of Sulfate and 31
Organic Carbon to the Southern and Northern Basins of Lake
Michigan - 1976.
9 Relationship of Atmospheric Total Phosphorus and Calcium Loading 32
' Rates to Precipitation Amount in Lake Michigan - 1976.
10 Relationship of Atmospheric Silica, Sulfate and Organic Carbon 33
Loading Rates to Precipitation Amount in Lake Michigan - 1976.
11 Relationship of Atmospheric Phosphorus and Calcium, Magnesium 35
Loading Rates to Lake Michigan - 1976.
-------
TABLES
Number Page
1 Atmospheric Sampling Site Locations 6
f
2 Precision of Analytical Methodology 10
3 Comparison of Precipitation Collection 14
4 Atmospheric Loading Rates of Total Phosphorus IB
5 Bulk Precipitation Concentrations and Atmospheric 19
Loading Rates for Lake Michigan
20
6 Atmospheric Loading Estimates of Chemical Components
to Lake Michigan
7 Comparison of Atmospheric Loading 'Rates at Beaver 21
Island, Michigan
8 Comparison of Buoy and Adjacent Land Station Loading 23
Rates
9 Estimated Phosphorus Inputs to Lake Michigan 24
10 Comparison of Bulk and Wet-Only Precipitation 37
vn
-------
ACKNOWLEDGEMENTS
Special thanks is given to cooperating personnel who assisted in the
construction of precipitation samplers and collection of atmospheric pre-
cipitation samples. This research was funded under Grant No. R 803238 from
the Environmental Protection Agency. The support and cooperation of the
Great T.akps Research Facility and the Center for Great Lakes Studies of
the University of Wisconsin, Milwaukee, and the University of Minnesota,
Minneapolis, is acknowledged. (lontribution number 166, Center for Great
Studies, The University of Wisconsin, Milwaukee, Wisconsin.
vi 11
-------
SECTION 1
INTRODUCTION
Considerable effort has been expended In recent years investigating
the sources and cycling of potential limiting nutrients (P, N, C) in marine
and freshwater environments. However, only recently have atmospheric inputs
of nutrients, especially phosphorus (P) and nitrogen (N) been considered as
important sources to small and large lake systems. Most effort and funding
have concentrated on restricting nutrient inputs from point sources such as
industrial and municipal discharges. Lakes Michigan and Superior have large
surface areas, but small drainage basins suggesting the atmosphere may be a
significant source of natural and anthropogenic components representing a
sizeable portion of the total nutrient budget.
Recent studies by Canada Centre for Inland Waters (1975), Murphy (1974),
Murphy and Doskey (1976), Shiomi and Kuntz (1973), Casey and Salbach (197*0
and IJC (1976) indicate clearly that atmospheric P inputs to the Great Lakes
may represent a sizeable portion of the overall nutrient budget. Murphy and
Doskey (1976) using event, wet-only precipitation data at six sites estimated
the atmospheric loading of P to Lake Michigan as 1x106 Kg/year, which rep-
resented about 18$ of the 1974 P budget of the lake. Approximately 40$ of
the atmospheric P was in the "dissolved reactive" form and 0.6x106 Kg/year
was in a form which could be mobilized by aquatic organisms.'
Casey and Salbach (1974) calculated the atmospheric P input to Lake
Ontario to be 1.7xl06 Kg/year representing ^ 10$ of the total P budget.
Similar studies on the Upper Great Lakes (IJC, 1976) found that ^ 12.1$ and
19-3$ of the total P budget of Lakes Huron and Superior, respectively, were
derived from the atmosphere.
Enhanced loading of major chemical parameters and nutrients to Lake
Michigan on an historical basis has resulted in a steady increase in lake
water concentrations of P (Vaughn and Reed, 1974), SO,., Cl, Na, K and total
dissolved solids (Beeton, 1965, 1969). Schelske and Stoermer (1971) contend
that accelerated eutrophication of Lake Michigan stimulated by phosphorus
inputs has resulted in an increased rate of silica depletion by diatoms in
the surface waters during summer stagnation leading to undesirable changes
in biological species' diversity. As P from point sources decrease as a
result of treatment advances, other P sources become more important.
The objective of this investigation was to determine the atmospheric
contribution of P to the nutrient budget of Lake Michigan and to delineate
possible sources and atmospheric removal mechanisms.
-------
SECTION 2
CONCLUSIONS
Annual P (TP) loading to Lake Michigan from atmospheric sources was
1.69xl06 Kg based on loading rates of 0.184 and 0.303 yg/cm2/mo for the
north and south basins, respectively in 1976. This represents 15.7$ of
the total P budget based on 1975 data.
In 1976, atmospheric deposition of TP in Lake Michigan was dominated
by a dry fall component likely derived from wind-blown soil. Actual in-
lake loading of TP may exceed loading calculated from land-based
stations by up to 50%.
The weighted-average concentration of TP in bulk precipitation was
0.050 and 0.064 mg P/l in the north and south basins, respectively, as
compared to the average lake water concentration of 0.01 mg P/l. Only
5-6J6 of the P collected in urban areas and 18 to 38$ of the P collected
in rural areas on a monthly basis was "dissolved inorganic P".
Loadings of all chemical parameters measured were greater in the south-
ern compared to the northern basin. Atmospheric loading of P and other
chemicals was generally greatest in spring and summer and minimum in
winter.
Atmospheric P inputs were closely correlated with soil-derived com-
ponents. Atmospheric loading of TOC and SO^ exhibited evidence of regional
transport and deposition.
The pH of wet-only precipitation averaged one pH unit lower than bulk
deposition.
-------
SECTION 3
That phosphorus be included as an air quality parameter and criteria
established.
That the relationship of land use to atmospheric phosphorus addition
to the Great Lakes be determined.
That phosphorus concentrations in rainfall, dry fallout and aerosol
particles over Lake Michigan be determined.
That sanpling for Total Organic Carbon (TOC)3 Calcium (Ca), Chloride (Cl),
Sulfate (SOj.) and pH be included in the air quality monitoring program for
the Great Lakes.
-------
SECTION 4
EXPERIMENTAL PROCEDURES
DESCRIPTION OF STUDY AREA
The precipitation collection network for determination of the atmospheric
P addition to Lake Michigan comprised the Lake Michigan basin and two sites
in the northern half of the lake which are land-locked and served as remote
stations. Precipitation stations were located in the states of Wisconsin,
Illinois, Indiana and Michigan. The largest fraction of the land component
in the basin is contributed by Wisconsin and Michigan which also provide
the greatest tributary inputs from the Grand and St. Joseph Rivers (Michigan)
and the Pox and Menominee-Milwaukee Rivers (Wisconsin) . Illinois and
Indiana in the Chicago-Gary area contribute negligible surface flow to the
lake, but represent major sources of atmospheric input as a result of intense
urban and industrial activity (Winchester and Nifong, 1971).
Lake Michigan has a surface area of S-SOxlO4 Kin2 as compared to a drain-
age basin of 1. 178x10 5 Km2 and lies wholly within the boundary of the con-
tiguous United States. Average depth and volume of Lake Michigan are 84m
and 4.87xl03 Kin3, respectively (Mortimer, 1976). Average annual precipita-
tion is 78.7 cm (Schelske and Roth, 1973).
Figure 1 depicts the 23 precipitation collection sites for the determina
tion of atmospheric P addition to Lake Michigan and Table I depicts their
location. The precipitation network was designed to collect bulk precipita-
tion (wet and dry fall) representative of rural, municipal and Industrial
regions of the basin adjacent to the lake so that atmospheric loading could
be derived. Urban and industrial contributions to atmospheric inputs were
investigated with a heavy density of collection sites in the Milwaukee,
Wisconsin and Chicago, Illinois-Gary, Indiana metropolitan centers. Loading
values from these areas represent averages of monthly-collected precipitation
samples for sites 7, 9 and 10 (Milwaukee) and 12 through 15 (Chicago). Av-
erage loading values were derived for intensely sampled areas so as not to
bias regional atmospheric inputs.
Sampling sites in the northern half of the lake represent rural and
regional inputs while collection sites in the southern half are more closely
related to urban and industrial sources. The northern and southern basins
were defined as one-half of the total lake surface area which parallels
closely geographical distributions. Sampling sites expected to be more
representative of actual in-lake precipitation were located at Beaver Island,
Michigan (#18) and two in-lake buoys located off the western shore opposite
the Milwaukee metropolitan area (#10B) and Kohler-Andre State Park (#6B).
The buoys were moored in ^ 23 m of water ^ 3-6 Kin offshore.
4
-------
14
CHICAGO
ILLINOIS
INDIANA
Scale in kilometers
0 25 50 75 100
Figure 1. Location of Lake Michigan Precipitation Stations
-------
TABLE 1.
ATMOSPHERIC SAMPLING SITE LOCATIONS
Site Number
Name
Location
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Watersmeet, Mich.
Stevens Point, Wise.
Green Bay, Wise.
Sister Bay, Wise.
Kewaunee, Wise.
Sheboygan, Wise.
Milwaukee, Wise.
Milwaukee, Wise.
Milwaukee, Wise.
Cudahy, Wise.
Zion, 111.
Wilmette, 111.
Chicago, 111.
Chicago, 111.
Calumet, 111.
Indiana Dunes
Ludington, Mich.
Beaver Island, Mich.
Douglas Lake, Mich.
Hammond Bay, Mich.
Sleeping Bear Sand Dunes
Grand Haven, Mich (Mlch)
St. Joseph, Mich.
45° 17' N, 89° 07' W
44° 31.6' N, 89° 34' W
44° 30'N, 88° 05' W
45° 09' N, 87° 04' W
44° 17' N, 87° 32.5' W
43° 39' N, 87° 44' W
43° 04.5' N, 87° 5V W
43° 02' N, 87° 50' W
42° 57' N, 87° 54' W
42° 57' N, 87° 48' W
42° 25.51 N, 87° 48' W
42° 04.6' N, 87° 4T W
41° 52' N, 87° 36.9' W
41° 55.3' N, 87' 39' W
41° 43' N, 87° 31' W
41° 49.8' N, 87° 04' W
43° 57' N, 86° 25.5' W
45° 44.8' N, 85° 30.5' W
45° 55' N, 84° 68' W
45° 30' N, 84° 02' W
44° 72.3' N, 86° 01.8' W
43° 03.8' N, 86° 14.8' W
42° 05.3' N, 86° 29.5' W
-------
Bulk-precipitation collectors were deployed in the field 1-4 July, 1975
and operated through 31 December, 1976. In-lake buoy collectors were oper-
ational for three months from August through October, 1976. Twenty-three
sites were equipped with bulk collectors of which 19 were monitored for the
entire study period. Two stations, Watersmeet, Michigan (#1) and Stevens
Point, Wisconsin (#2) were used to compare loading rates distant from Lake
Michigan. Two Atomic Energy Commission (AEC) wet-only samplers (Galloway,
1976) were deployed at sites #5 and #8 to obtain an estimate of the dry fall
component. Except for only a few samplers, all precipitation was collected
at sites within 0.5 Km of Lake Michigan in areas which would receive minimal
influence from local sources.
DESCRIPTION OP SAMPLING SITES
A brief description of each sampling site follows:
Site #1 consisted of a bulk collector located on the lawn of the
Ottawa National Forest headquarters in Watersmeet, Michigan. Entrance to
the building and collector was via a base gravel road.
A second bulk collector (#2) was located on the roof of the Science
Center of the University of Wisconsin-Stevens Point approximately 15 meters
above ground level. Stevens Point has a population of 23,471 and possesses
paper mill and light manufacturing capabilities.
A third bulk collector (#3) was located at the meteorological field of
the University of Wisconsin-Green Bay. Several weather monitoring devices
were enclosed within a fenced area of the university grounds about 1 Km east
of Green Bay and Lake Michigan. Green Bay (population 87,809) has light and
heavy manufacturing including several of the largest paper companies in the
United States.
A fourth bulk collector (#4) was located on a private lawn near Sister
Bay, Wisconsin adjacent to an agricultural area about 2 Km from Lake Michigan.
The fifth bulk collector (#5) and one AEC wet-only sampler was located
on the roof of the Point Beach Nuclear Power Plant visitor's center about
4 m above street level and 0.1 Km from Lake Michigan. Agriculture, commer-
cial fishing and tourism represent the major activities of nearby Kewaunee
(population 2.901) and Two Rivers, Wisconsin (population 13,732).
The sixth bulk collector (#6) was located on the beach of Kohler-Andre
State Park about 3-5 Km north of Sheboygan, Wisconsin (population 48,484).
The area is bordered by scotch pine (Pinus strobus) with agriculture predominant
inland. A buoy tower and bulk collector (#6B) was located 3.2 Km east of the
beach in about 23 m of water from 2 August, 1976 to 3 November, 1976.
Bulk collectors 7, 8, 9 and 10 were located in the Milwaukee metro-
politan area (population 717,372) on the south lawn of the Linwood Avenue
Water Filtration Plant on the roof of the University of Wisconsin Great Lakes
Research Facility. On the lawn of the U.S. Weather Service at Billy Mitchell
-------
Field-International Airport and on the roof of the Cudahy, Wisconsin
Senior High School. Most bulk collectors were within 0.5 Km of Lake
Michigan. Precipitation data at site #8 (GLRP) was influenced by a local
coke plant and foundry 0.5 Km to the west and was not used to calculate
loading values to Lake Michigan. An AEG wet-only sampler was deployed at
site #8 and operated from April-December, 1976 to investigate wet-only
deposition in an urban area. The second buoy tower and bulk collector was
located 6.4 Km east of Milwaukee opposite site #10 and operated from 2 August-
3 November, 1976.
Bulk collector #11 was located at Illinois Beach State Park, Zion, Illinois
approximately 4 m above ground level and 50 m from Lake Michigan. Light industry
and the nuclear power plant at Zion are in the area. The park is heavily wooded
with a deciduous flora.
Bulk collector #12 was located on the lawn of the U.S. Coast Guard
Station at Wilmette Harbor north of Chicago about 50 m from Lake Michigan.
Bulk collectors were located at sites 13, 1*J and 15 in the Chicago area
on the roof of the John G. Shedd Aquarium, on the roof of the science build-
ing at DePaul University and on the lawn of the U.S. Coast Guard Station at
Calumet Harbor near the Indiana-Illinois border, respectively. The collector
at DePaul University was located about 2 Km from Lake Michigan, but #13 and
#15 were within 0.5 Km of the lake. Site #15 was influenced by nearby steel
manufacturing while #14 reflected urban activity.
Bulk collector #16 was located on the beach pavillion roof at Indiana
Dune State Park. Large sane dunes and beach succession flora compose the
surrounding area.
Bulk collector #17 was located on the lawn of the Biological Station
of the U.S. Bureau of Sport Fisheries in Ludington, Michigan about 5 Kin from
the lake.
2
Bulk collector #18 was located on Beaver Island (area of 1^0 Km ) in
the northern reaches of Lake Michigan adjacent to a similar collector maintained
by the Michigan Department of Natural Resources. Local activity consists of
tourism, comnercial fishing and some agriculture.
Bulk collectors #19 and #20 were located at the Douglas Lake Biological
Station of the University of Michigan and Hammond Bay Biological Station of
the U.S. Bureau of Sport Fisheries, respectively, from July-November, 1975.
Sampling at these locations terminated in November 1975.
Bulk collectors 21, 22 and 23 were operated throughout 1976 at Sleeping
Bear Sand Dunes National Lakeshore near the mouth of the Grand River at
Grand Haven State Park and on the roof of St. Joseph High School, St. Joseph,
Michigan, respectively. Most of the bulk collectors were located within 0.5
Km of Lake Michigan. The combined population of the St. Joseph-Benton Harbor,
Michigan area is 27,523.
8
-------
Precipitation Sampling
Precipitation was collected primarily with bulk collectors (wet and
dry fall) based at either 23 land-based stations or two in-lake buoys. Bulk
precipitation data was augmented with limited amounts of wet-only collected
samples. The bulk collector was similar to that used by Kramer (CCIW, 1975)
and consisted of a central collection standpipe (25 cm diameter, 506.45 cm2
surface area) surrounded by an Alter windshield and a bird-off to limit
sample contamination. The standpipe collector was designed originally as a
snow sampler which also operates effectively as a rain sampler with some
modification. The sampler is similar to that used previously in determining
atmospheric inputs of chemical components to Lakes Superior and Huron (IJC,
1976). The bulk collector has been evaluated by Berry (1975) and Galloway
(1976) for use in collecting rain and dry fallout for chemical analysis.
Rawls et_al. , (1975) compared the rainfall collection efficiency of stand--
pipe-type samplers with a modified Alter windshield and a rigid Alter wind-
shield. The windshields did not produce variable collection efficiency on
the average, and either model can be used to compute actual precipitation.
Bulk precipitation was collected in an inner liner made of Jj-mil thick-
ness polyethylene. In the summer of 1975, a 2 mm, Nytex Nylon screen was
placed in the liner to limit introduction of coarse debris and large insects.
In 1976, a 0.5 mm screen replaced the 2 nrn screen. In addition, a poly-
ethylene funnel was placed at 0.3 m above the bottom of the liner and the
liner restricted by tying at the funnel outlet to limit evaporation. Summer
evaporative losses with the modified system were limited to > \% under even
extreme temperature conditions.
The buoy samplers were modified bulk collectors deployed on oceano-
graphic buoys as indicated above except three, polyethylene funnels drain-
ing into three-10 I polyethylene containers were used instead of the stand-
pipe and liner. Rainfall entered the collection containers located in wooden,
waterproof containers through FVC tubing. Two funnels were equipped with
bug/debris screens (0.5 mm Nytex Nylon) and one was left open for comparison.
The wet-only precipitation samplers were of the AEG design and constructed
from blueprints. The polyethylene containers were 29.5 cm in diameter and
684.9 cm2 in surface area. Galloway (1976), in evaluating collectors concluded
that the AEG sampler was the most reliable of seven automatic collectors tested.
Gear problems limited applicability in this project.
Bulk precipitation was collected on a monthly-basis by cooperative
personnel at most land stations. The data resulting from such a sampling
system represents total loadings over a one month period. Although event,
weekly or bi-weekly samples are preferred to increase time-resolution of
loading, monthly samples were selected to 1) permit comparison of our data
with recent atmospheric loading studies conducted with similar samplers and
frequency on the Great Lakes and 2) to require as little attention as possible
by volunteer personnel maintaining the collection systems.
-------
Table II
PRECISION OF ANALYTICAL METHODOLOGY
COMPONENT
TP
TOP
DRP
DRSi
so4
~
Cl
TOC
TIC
Ca
Mg
Na
K ,
COND.
Particulates
CONCENTRATION
mg/1
0.119
0.050
0.019
0.032
0.011
0.003
0.029
0.014
0.004
1.40
0.68
0.17
10.0
4.3
7.3
3.0
24.5
16.1
6.1
15.2
5.9
2.6
4.32
0.88
0.69
0.41
102.2
66.9
31.9
151.2
59.4
22.9
STD. DEV.
0.007
0.001
0.006
<001
<001
<001
<001
<001
<001
0.01
<001
0.006
0.2
0.1
1.0
0.7
0.4
2.6
0.5
1.2
0.4
0.3
0.08
0.01
0.01
0.01
0.3
1.1
0.6
25.9
2.8
1.5
SIZE
10
10
10
10
10
10
10
10
10
5
5
5
10
10
10
10
5
5
5
5
5
5
10
10
10
10
5
5
5
5
5
5
S/crn
10
-------
Precipitation samples were collected at the end of each month and mailed
to the Great Lakes Research Facility (GLRF) in Milwaukee or retrieved by
project personnel. All sanples were collected according to the same procedure
with particular emphasis on sarrple homogeneity and contamination problems.
Visual observations of sample condition were included in sample mailing and
evaluated upon receipt. Occasionally, monthly sanples were not received
due to mailing difficulties which sometimes limited the proper statistical
evaluation of loading data. However, the large number of sampling sites in
operation ininimized the inpact of sanple loss from individual sites.
Chemical Analysis
Bulk and wet-only precipitation samples were analyzed for total (TP),
total dissolved (TOP) and dissolved reactive P (DRP) using the heteropoly
blue method as modified by Eisenreich ejb aL., (1975). Samples were filtered
the day of receipt through washed 0.4 ym cellulose nitrate membranes (Sartorius
113 06) which have low P blanks and analyzed within 96 hours. Dissolved
reactive Si (DRSi) was determined on membrane filtered sanples within 96
hours by the method described in Standard Methods (1971). Sulfate (SOL)
was determined by membrane-filtered sanples by the turbimetric method and
chloride (Cl) by the mercuric nitrate procedure (Standard Methods, 1971).
Ca, Mg, Na and K were determined using a Perkin Elmer 503 Atomic Absorption
Spectrophotometer with (Ca, Mg) or without (Na, K) La addition. Total
organic and inorganic carbon were determined on unfiltered sanples pre-
served by freezing using a modified Beckman Infrared Carbonaceous Analyzer
(Maier e_t al., 197*0. Dissolved organic and inorganic carbon were deter-
mined in a similar manner on samples filtered through pre-combusted Whatman
GF/C filters. Particulates were determined by a gravimetric procedure and
specific conductance measurements were made with a Leeds and Northrup
electrical conductivity bridge at 25° C. A combination glass electrode
was used to determine pH.
No chemical preservatives were added to collected samples in the field
or laboratory to minimize biological transformations so as to limit con-
tamination. Contamination due to polyethylene liners was negligible as
determined by leaching experiments. Analytical precision typically ranged
from 1 to 10% depending on parameter and concentration.
11
-------
SECTION 5
RESULTS AND DISCUSSION
LOADING CALCULATION
The principal objective of this investigation was to determine the
atmospheric loading of P and other chemical couponents to Lake Michigan.
To this end, 23 land-based sites were operated for 18 months in 1975-1976
in which monthly bulk precipitation (wet and dry) was collected and analyzed.
Chemical loading rates were calculated by multiplying the analytical con-
centration of a monthly-collected sample by the volume of precipitation and
dividing by the collector surface area.
LOADING RATE = Cone. (gg/O x Volume (A)
505.45 cirr
Average monthly loading rates for each site were obtained by summing individual
monthly loading rates over all months and dividing by the number of months
sampled. To convert to loading rates expressed as g/m^/yr, multiply the
loauing rate expressed as ug/cnvmo by 0.12.
Annual lake-wide loading is calculated by multiplying the mean monthly
loading rate (all sites) by the number of months per year, the surface area
of Lake Michigan (58016 Km2) and a unit conversion factor (10~9pg/Kg):
ANNUAL LOADING = LOADING RATE x 12 x 5-80 x 10 cm x 10" Kg/yg
(Kg/year) (yg/cm2/mo)
Calculation of annual loadings based on mean loading rates determined from
all sites heavily biases the loading value to the southern half of the lake
where municipal and industrial activity is greatest. Two modifications of
the above loading approach were implemented: 1) a mean loading rate was
determined for each chemical parameter expressed in yg/cm2/mo for northern
snd southern Lake Michigan separately; and 2) the mean monthly loading rates
for the Milwaukee and Chicago areas were averaged to obtain a value typical
of the two metropolitan areas. Collection sites used in computation of the
northern Lake Michigan loading were 1, 2, 3, 4, 5, 6, 17, 18 and 21.
Likewise, southern sites were (7, 9, 10), 11, (12-15), 16, 22 and 23. Site
#8 was not used in the Milwaukee averages due to local source contamination.
These modifications permit a better estimate of lake loading as well as an
improved evaluation of atmospheric source impact due to the generally rural
and forested northern half of the basin versus the urban-industrial south.
12
-------
Atmospheric Loading to Lake Michigan
The efficiency of precipitation collection by the bulk collectors was
evaluated by placing a bulk sampler within 10 m of a weather service collector
at Milwaukee's Billy Mitchell Field. Table III lists the precipitation col-
lected in the two collectors for 1976 on a month by month basis. The bulk
sampler collected 86 ± 5% of the rain gage precipitation from January through
December. The constant percentage throughout 1976 encompassing rain and
snow months suggests differences noted were due to differences in samplers
rather than sample loss, and that the bulk collector estimated adequately
precipitation falling in either form.
Figure II shows the amount of precipitation collected in 1976 and the
historical mean annual precipitation amounts for nearby meteorological sta-
tions at all sampling sites. The data clearly indicate that 1976 was a dry
year compared to historical precipitation records in which only 59% and 70%
of the mean annual rainfall was observed in the north and south basins of
Lake Michigan, respectively. Local variability in rainfall amounts did not
permit comparison of weather service and bulk collector data for individual
sites in 1976. The near-drought conditions experienced in the upper Lake
Michigan basin suggest that dustfall inputs from the atmosphere may be sig-
nificantly greater than rainfall inputs.
Bulk precipitation was collected from July, 1975 through December, 1976
but only 1976 data were used in calculating atmospheric loading rates.
JBlgure III shows the 1976 mean monthly loading rates given in (yg/cm2/mo)
for 1976 at each land-based sampling station for TP and TDP. The observed
loading rates for TP indicate a definite north-south gradation as the mean
TP loading rates were 0.184 and 0.303 yg/cm2/mo for north and south basins
of the lake, respectively. For comparison, whole-lake loading rate for TP
with no differentiation of north-south inputs was 0.231 yg/cm2/mo. Increased
loading rates were also observed for the eastern shore of Lake Michigan versus
the western side. This behavior likely signifies differences in soil types
whereas north-south variations were due to urban-industrial activity.
A compilation of mean loading rates for TP derived from the literature
converted to units of yg/cm2/mo is given in Table IV. Values range from
0.074 to 0.850 yg TP/cm2/mo. The TP loading rate determined for the northern
Lake Michigan basin in this study (0.184 yg/cm2/mo) compares favorably with
the value reported for the Upper Great Lakes (0.140 yg/cmz/mo) and Lake
Michigan (0.144 yg/cm2/mo) determined from wet deposition only. In compar-
ison, the less populated and developed Lake Superior basin exhibited a load-
ing rate of only 0.081 yg/cm2/mo reflecting the relative lack of urban-
industrial activity. The large TP loading rates reported by Allen et_ al., .
(1968), Gore (1968), Pearson and Fisher (1971) and Kluesner (1972) reflect
urban and agricultural activities. Brezonik and Hendry (1977) have reported
recently a TP loading rate of 4.50 yg/em2/mo for north central Florida cor-
responding to intense atmospheric loading from phosphate mining and an annual
rianfall of 145 cm.
The loading rate for TDP averaged 0.041 ± .014 yg/cm2/mo in the northern
13
-------
TABLE III
COMPARISON OF PRECIPITATION COLLECTION
Month
1976
Jan
Feb
Mar
Apr
May
June
Julyb
Augb
Sept
Oct
Nov
Dec
Amount
Bulk Collector cm
2.35
6.32
15.64
11.25
8.08
4.96
9,08
3.71
5.92
1.30
0.69
U.S.W.S.3
2.95
6.73
17.60
12.73
9.58
5.77
10.60
4.32
7.16
1.65
0.74
Aug:
%
80
94
89
88
84
86
86
86
83
79
93
86 + 5%
a U.S. Weather Service gage was located 10 m from bulk collector at
. Milwaukee's Billy Mitchel Field.
Rainfall volumes are for July and August.
14
-------
PRECIPITATION
(cm)
MILWAUKEE
ANNUAL AVERAGE
73.8
CHICAGO
ANNUAL AVERAGE
85.3
ILLINOIS
Scale in kilometers
0 25 50 75 100
Figure 2. Comparison of 1976 and historical precipitation amounts
at atmospheric sampling sites near Lake Michigan.
-------
LOADING RATE
(M9/cm2/mo)
-N-
MILWAUKEE
0.310
CHICAGO
12-15
ILLINOIS
INDIANA
Scale in kilometers
0 25 50 75 100
Figure 3. Atmospheric loading rates of total and total dissolved
phosphorus at precipitation sampling sites near Lake
Michigan - 1976.
16
-------
lake basin and 0.061 + .046 yg/cm2/mo in the south, corresponding to 22 and
20% of the TP values in the north and south, respectively. The low percent-
age of dissolved compared to total P may be due to conversion of dissolved
to particulate forms in biotic or abiotic reactions upon standing in the
field. DRP represented less than 5% of TP in nearly all cases. As a result
of sampling on a monthly bas.is and collecting wet and dry deposition, P
speciation was not feasible. Collection of wet-only precipitation on an
event basis combined with rapid analysis should give best estimates of P
forms in rainfall.
The annual loading of TP to Lake Michigan by the atmospheric route has
been estimated by Murphy and Doskey (1976) to be l.OxlO6 Kg P/year based
on 6 collection sites and analyzing event, wet-only precipitation. The
approach taken was to determine a weighted-average TP concentration in rain-
fall for the entire lake and multiply by the mean annual precipitation in
the Lake Michigan basin (7*1 cm) and the surface area of the lake. The
approach taken in this study was to employ a larger number of samplings
sites, determine mean monthly loading rates for the north and south basins
and calculate annual loadings based on lake surface area. This strategy
minimizes errors due to variations in annual precipitation, deposition type
or geographical location. The annual loading of TP to Lake Michigan was
estimated as 1.69xl06 Kg/year, of which 0.64xl06 and 1.05xl06 Kg/year depos-
ited in the north and south basins, respectively. Table V shows the loading
rates and volume-weighted average concentrations of TP, TOG, Si02, SO^, Cl,
Ca, Mg, Na, K and particulates for 1976 from which annual loading values
were calculated as reported in Table VI..The calculated range of southern to
northern inputs is 1.64 for TP and 1.39 for Mg to 3.04 for SiO^. The en-
hanced loading of all chemical parameters measured in the south verifies the
often-made estimate that the urban-industrial complex of southern Lake
Michigan contributes significant quantities of major and trace elements to
the lake via the atmospheric route (Winchester and Nifong, 1971). Deposi-
tion of chemical elements other than P will be limited in this paper to pro-
viding source and removal mechanisms for P.
The difference between Murphy and Dosky's (1976) estimate of TP load-
ing from the atmosphere and this study is due to differences in sampling
strategy (bulk vs wet), number of sites, calculation method and year. Pre-
cipitation data verifies that 1976 was significantly more dry than the
1973-1974 period. Phosphorus loading has a significant dry-fall component
(Pearson and Fisher, 1971) which would enhance loadings in dry years.
Loading Rate Comparisons
Atmospheric loading rates for TP observed in this study can be compared
to loading data generated in the Upper Great Lakes Reference Study (CCIW,_
1975) under the auspices of the International Joint Commission. Table VII
lists the atmospheric loading rates for TP and other chemical components
determined at Beaver Island, Michigan (Site #18) in the northern Lake
Michigan basin for the period 1973-1975 and 1976. Bulk collectors using the
central standpipe design were employed in both studies. The loading rates
for all chemical parameters except sulfate for the period 1973-1975 were 21
to 60$ of the 1976 values. Sulfate exhibited 157% greater loading in the
earlier period. The near drought conditions in the area during 1976 may have
17
-------
Table IV
ATMOSPHERIC LOADING RATES OF TOTAL PHOSPHORUS
LOCATION
England
Northern England
(1959-1965)
Northeastern U.S.3
Northwest Ontario
Central Wisconsin3
Central Finger Lakes
Region
Upper Great Lakes3
Lake Erie3
Lake Ontario3
Lake Huron3
(1973-1975)
Georgian Bay
(1973-1975)
North Channel3
Lake Superior3
(1973-1975)
Lake Michigan0
Lake Michigan3
Northern
Southern
(1975-1976)
LOADING RATE
2
(ug/cm /mo)
.167-667
.667
.394
.333
.850
.240
.140
^.40
.298
.219
.074
.076
.081
.144
.184
.303
REFERENCE
Allen et al (1968)
Gore (1968)
Pearson and Fisher (1971)
Armstrong and Schindler
(1971)
Kluesner (1972)
Likens (1972)
CCIW (1975)
Matheson (1974)
Shiomi and Kuntz (1973)
IJC (1976)
IJC (1976)
IJC (1976)
IJC (1976)
Murphy (1976)
This Study (1977)
Bulk Deposition
Snow-Only Deposition
18
:Wet-Only Deposition
-------
TABLE V
BULK PRECIPITATION CONCENTRATIONS AND ATMOSPHERIC LOADING
RATES FOR LAKE MICHIGAN
Component
Phosphorus
(Total as P)
Organic Carbon
(Total as C)
Si 1 i ca
(Reactive as Si02)
Sulfate
Chloride
Calcium
Magnesium
Sodium
Potassium
Parti culates
PH
North
0.050
3.8
0.146
3.2
2.5
2.38
0.67
0.59
0.23
19.9
5.28
Concentration3 .
South Lake Michigan
mg/1
0.064
6.2
0.356
5.6
2.9
4.32
0.78
0.64
0.41
34.0
5.56
0.01
4.9
0.51
16
6
32
10
3
1
-
8.0
Loading Rate0
North 2 South
yg/cm /mo
0.184
(.038)
13.9
(2.2)
0.56
(.47)
12.2
(4.7)
9.7
(3.8)
9.0
(3.5)
2.78
(2.3)
1.5
(1.1)
0.88
(.34)
70.7
(33)
—
0.303
(.096)
30.2
(2.3)
1.70
(.95)
26.7
(5.5)
14.0
(4.8)
20.7
(5.5)
3.8
(0.6)
3.1
(1.5)
1.88
(.95)
164.5
(89)
—
Weighted-average concentration in bulk precipitation for north and south basin.
'Component mean concentrations for open Lake Michgain water; from Upchurch (1975)
and Schelske and Roth (1973).
•^
'Numbers in parenthesis represent one standard deviation from the mean.
19
-------
TABLE VI
ATMOSPHERIC LOADING ESTIMATES OF CHEMICAL COMPONENTS TO LAKE MICHIGAN
COMPONENT
NORTHERN1
SOUTHERN11 TOTAL
(10° Kg/year)
RATIO1
Phosphorus, Total as P
Organic Carbon (Total as C)
Silica (Reactive as Si02)
Sulfate (as S04)
Chloride (as Cl)
Cal ci urn
Magnesi urn
Sodi urn
Potassium
Parti culates
0.64
48.4
1.98
42.5
33.7
31.4
9.40
5.29
3.06
246.
1.05
105.
6.01
92.9
48.8
72.0
13.1
10.9
6.54
573.
1.69
154.
7.99
135.
82.5
103.
22.5
16.2
9.60
819.
1.6
2.2
3.0
2.2
1.5
2.3
1.4
2.1
2.1
2.3
aNorthern and southern basin areas were assumed equal (29008 Km each)
Ratio: Southern Inputs/Northern Inputs
20
-------
TABLE VII
COMPARISON OF ATMOSPHERIC LOADING
RATES AT BEAVER ISLAND, MICHIGAN
Parameter
TP
Tnr
1 UL
Si02
so4
Cl
Ca
Mg
Na
K
Part.
PHC
Loading Rates
1 973-1 975a 2
pg/cm /mo
0.100
0.335
16.6
4.26
7.03
1.22
0.304
0.304
18.3
5.95
1976b
0.180
IK T
1.63
10.6
10.9
11.8
2.19
0.88
0.85
43.0
5.14
Ratio
0.56
0.21
1.57
0.39
0.60
0.56
0.35
0.36
0.43
— — — —
SCCIW (1975); IJC (1976)
bThis Study
cWeighted-Average pH
21
-------
resulted in increased short and long range transport of soil particles which
were ultimately removed by sedimentation or washout. The similar loading
behavior shown for TP, Ca and Mg in the two time periods indicate that
atmospheric P may be associated with soil particles. However, Na and K
exhibit a lower, but constant percentage loading in the two time periods
suggesting a different source or atmospheric removal mechanism. Sulfate
in remote areas is generally associated with biotic processes such as
sulfide to sulfate conversion in wetland areas, long distance transport from
urban-industrial centers or soil particles. Gas-to-particle conversion
results in sub-micron sulfate aerosols while production of soil-derived
sulfate is associated with large particles greater than 5 ym. The greater
sulfate loading rate in 1973-75 compared to 1976 may be caused by efficient
washout by rainfall in the former wet period.
The chemical loading rate data derived from bulk precipitation data
shown in Table V were developed at land-based stations on the perimeter
of Lake Michigan. Coastguard buoys outfitted with three sampling units
were deployed in August, 1976 about 3-6 Km offshore to ascertain whether
land-based loading rates were representative of in-lake loading. Table "VIII
lists the mean monthly loading rates for August-October, 1976 for land
station and buoys in rural and urban areas. Buoy collectors outfitted with
bug screens of a type different from land collectors contaminated rainfall
and could not be used.
The loading rates calculated from buoy data exceeded corresponding land
stations for TP only by a factor of 1.2. In general, other chemical load-
ing rates from buoy data were approximately equal to or slightly less than
land stations except for two parameters. In-lake particulate loading ex-
ceeded land loading off the urban area while in-lake TOG loading at the
rural site was about 37$ of the land loading. As expected, the buoy load-
ing rates obtained off the urban area were greater than the rural site
demonstrating the effect of urban-industrial centers on atmospheric loading
of adjacent water bodies. The increased deposition of TP at buoy sites may
be due to greater scavenging efficiency over water. Enhanced chemical load-
ings for in-lake buoys compared to land stations has been reported pre-
viously for Lake Huron (CCIW, 1975). This behavior suggests that land-
derived atmospheric loading rates may underestimate TP inputs to Lake
Michigan although caution should be observed in extrapolating only three
months of in-lake precipitation data.
Estimated Present and Future Phosphorus Inputs
/
The phosphorus inputs from point and non-point sources have been de-
termined for Lake Michigan and are compared to atmospheric inputs in
Table IX for 1974 and 1976, and forecasted for the year 2000. The input
of P in precipitation to Lake Michigan in 1974 was 1.0 x 106 Kg/year based on
Murphy and Doskey's estimates (1976) and represented 11.8% of the total in-
put. The percentage contribution of atmospheric P differed from the quoted
value of 18% because source loading data prior to 1974 was used in the
earlier calculation.
22
-------
TABLE VIII
COMPARISON OF BUOY AND ADJACENT LAND STATION LOADING RATES3
COMPONENT
TP
TOC
Si02
so4
Cl
Ca
Mg
Na
K
Part
RURAL
Land
.12
{.07-. 16)
29.
(21. -38.)
.30
(.18-. 42)
16.
(7.7-25.)
12.
(7.8-17.)
9.4
(6.1-13.)
1.7
(.80-2.5)
.96
(.73-1.2)
.54
(.51-. 56)
58.
(37. -79.)
Buoy
.15
(.12-. 21)
11.
(6.8-15.)
.34
(.18-. 54)
12.
(8.3-16)
3.
(1.7-4.3)
9.5
(5.9-14.)
1.0
(.54-1.9)
.77
(.37-1.1)
.81
(.64-. 91)
48.
(40. -55.)
LOADING RATES b
URBAN
2 Land
yg/cm /mo
.29
(.22-. 33)
30.
(26-38)
.60
(.36-. 78)
23.
(14. -29.)
17.
(10. -24.)
19.
(12. -26.)
3.0
(2.0-3.9)
1.5
(.28-2.8)
.63
(.28-. 89)
319.
(71. -785.)
Buoy
.34
(.28-. 44)
30.
(20. -39)
.56
(.30-. 90)
20.
(17. -25.)
14.
(4.8-23.)
15.
(11. -18.)
1.8
(1.4-2.2)
.88
(.67-1.1)
.67
(.50-. 92)
146.
(41. -244)
aValues represent mean monthly loading rates for August-October, 1976.
bNumbers in parenthesis represent range of monthly loading rates measured.
Assuming an annual P loading to Lake Michigan of 1.69 x 106 Kg/year, the
percentage atmospheric contribution to total loading was 15.7$. With a 10$
net decrease in industrial, municipal and tributary loadings, and a 13-1$
decrease in erosional inputs by the year 2000 (PLUARG, 197*0, the contribution
23
-------
TABLE IX
ESTIMATED PHOSPHORUS INPUTS TO LAKE MICHIGAN3
Source
Direct Industrial
Discharge
Direct Municipal
Discharge
Tributary
Erosion
Atmospheric
TOTAL
1974
0.045
(.5)
1.09
(13)
4.97
(58.7)
1.35°
(16.0)
1.00f
(11-8)
8.46
1976
-.in J'n/umi-
i u Kg/year
0.061
(-6)
1.07
(10)
4.23
(39.4)
3.7d
(34.3)
1.699
(15.7)
10.75
2000
0.055
(-6)
0.96
(9.7)
4.00
(40.3)
3.2e
(32.3)
1.69
(17.1)
9.91
aNumbers in parenthesis represent percent of total.
b4th Annual Great Lakes Water Quality Report (1976).
CU.S. EPA (1974)
dSonzogni and Monteith (1977)
International Reference Group on Great Lakes Pollution from Land Use
Activities (1974); 13.1% decrease in erosional inputs by year 2000.
fMurphy (1976)
9This study (1977) 24
-------
of atmospheric P to the budget of Lake Michigan will be * 15.756 if atmo-
spheric Inputs remain constant. Similar estimates have been arrived at for
the Great Lakes by Chapra (1977). The accuracy of the predicted Inputs will
depend heavily on the 80-90$ removal of P from wastewater, reduction or
elimination of P in detergents and implementation of minimal air quality
standards for P. The latter action may be difficult or impossible to leg-
islate .
Spatial and Temporal Variations in Loading Rates
The atmospheric loading of TP and other chemical components to Lake
Michigan is heavily influenced by the north-south orientation of the lake
and the large industrial complex in the Chicago-Gary area. Thus, 62% of
the TP loading to Lake Michigan is deposited in the southern basin. The
relative contribution of each chemical component to the total loading in
the northern and southern basin is given in Figure IV. The percentage of
southern loading of the total ranges from 58$ for Mg to 75% for Si02- The
increased loading in the southern basin may be attributed to urban-industrial
activity, agricultural practices in Iowa, Illinois and Indiana, and regional
soil transport from the south and southwest (Murphy, 1974; Gatz, 1975;
Nifong and Winchester, 1971).
Seasonal variations in TP loading rates from the atmosphere may be used
to suggest sources and importance of rain versus snow deposition. Figure V
shows the mean monthly loading rates in 1976 for atmospheric TP for the
Chicago and Milwaukee metropolitan areas and Beaver Island, Michigan. The
latter site was used to indicate relative background loading in the upper
basin. Loading rates for Chicago exhibit erratic behavior, but reach maxima
in March, July and early winter while Milwaukee shows a biomodal distribution
in loading rates with maxima in May and October. The lag in spring loading
peaks between Milwaukee and Chicago may be due to colder spring temperatures
in the former. Beaver Island shows a small peak in March with major loading
noted in July, matching Chicago. In general, loading decreases in the order
Chicago > Milwaukee > Beaver Island. There is no apparent relationship
between loading rate maxima and monthly precipitation volume indicating the
importance of dry deposition. Observed spring-ear.ly summer maxima may be
caused by seeds, pollen, insects (Kluesner, 1972) and agricultural fertil-
izing and tilling while winter peaks may result from urban activities,
judging by the absence of a winter peak at Beaver Island. Snow cover in-
hibits atmospheric loading by reducing wind-blown dust levels. Seasonal
variations in TP loading have been summarized previous ly by Brezonik (1976),
but appeared too random for interpretation. Kramer (1976) has also observed
that TP loading shows a seasonal variation increasing in spring and summer.
Figure VI summarizes the seasonal variation in TP loading rates for the
north and south basins of Lake Michigan. In general, loadings were dominated
by spring-summer maxima with greater fluctuations in the south. Winter load-
ings are decreased by ^ 50$ in the south as compared to summer loading rates
and were minimal in the northern basin. In contrast, Figure VII shows the
temporal variations in north-south loading rates for SiOp and Ca. Ca and
SiOp are derived primarily from soil in remote areas or reentrained dust in
urban areas, but both may also receive contributions from industrial activity
in the south. Ca and SiOp atmospheric loading rates are relatively constant
25
-------
ro
ATMOSPHERIC LOADING ESTIMATES OF CHEMICAL
COMPONENTS TO LAKE MICHIGAN
1.0
>
O)
(O
O
_- 0.5
O
0
Figure 4
1976
SOUTHERN BASIN
NORTHERN BASIN
1
n
1.69 153.5 7.99 135.4 82.5 103.4 22.5 16.2 9.60 818.5
I
I
m
n
TP TOG SiOo SO,
Cl
Ca
Mg
Atmospheric loading estimates of P, TOC, Si02, S04, Cl, Ca,
particulates to the southern and northern basins of Lake Mi
Na
Mg, Na
chigan
K
, K and
- 1976.
Partic-
ulates
-------
8
LLJ
<
DC
9 T
< o
o *-
_i x
D ^
c E
i?
35!
O TO
O
0
1976
CHICAGO
./
BEAVER ISLAND
0
468
MONTH
10
12
Figure 5. Seasonal variations in atmospheric loading rates measured
at Beaver Island, Michigan, Milwaukee, Wisconsin and
Chicago, Illinois - 1976.
27
-------
6.4
ul 5.6
H
1976
QC
4.8
o 4.0
—5 O Q O
5: £ 3>2
CO
DC
O rT
Q- O O /I
CO ^ ^-^
O 5s
O
1.6
0.8
SOUTHERN
BASIN
NORTHERN
BASIN
0
1
1
1
0
4 6 8
MONTH
10
12
Figure 6. Seasonal variations in atmospheric loading rates
to the southern and northern basins of Lake
Michigan - 1976.
28
-------
SOUTHERN
BASIN
2 4 6 8 10 12
MONTH
Figure 7. Seasonal variations in atmospheric loading rates
of calcium and silica to the southern and
northern basins of Lake Michigan - 1976.
29
-------
throughout the year except for a spring maximum In the north basin while
greater fluctuations occur for both in the south. Two chemical parameters
more closely associated with anthropogenic sources are TOG and SCL. The
observation is made from Figure VIII that TOG and SO^, exhibit parallel load-
ing behavior in the north and south basins of Lake Michigan. One explanation
for the observed variations is that TOG and SO^ deposition result from
regional transport processes and have similar sources or removal mechanisms.
Both TOG and SQ^, if derived from anthropogenic sources, are associated with
aerosol-size particles less than 2 pm diameter and may have long residence
times in the atmosphere.
If the assumption is made that the difference in chemical loading rates
for TP, Ca and Si02 between the north and south basins on the average is due
to anthropogenic sources, then the contribution of man's activities to their
loading rates may be estimated as: TP > 40$; Ca > 5Q%-3 SiCU > 10%; SCL > 40$;
TOG > 60$. These values represent gross estimates based on summer loading
rates and should be interpreted with caution.
Precipitation Composition Relationships
The chemical substances comprising bulk precipitation originate from
two atmospheric sources: dust particles removed by sedimentation or impac-
tion, or soluble gases or salts which are scavenged by rainfall. Chemical
components originating from the two atmospheric sources would behave differently
with respect to loading characteristics as a function of precipitation volume.
Thus, atmospheric components which deposit on water primarily by dry fallout
will be independent of precipitation amount while components removed by wash-
out or rainout will vary with precipitation amount and number of events.
Figure IX depicts the relationship between TP and Ca loading rates and annual
precipitation amount. Each data point represents the mean monthly loading
rate for one sampling station considered for 1976. The drawn lines represent
the linear regression fit for northern sampling sites only. The southern
basin data points were not considered in the calculation, but increased load-
ings were noted in urban-industrial areas for large precipitation volumes.
Phosphorus loading in the northern Lake Michigan basin appears indepen-
dent of precipitation in agreement with the limited results of Pearson and
Fisher (1971). The TP loading rate averaged 0.184 ± .04 yg/cmVmo. Southern
basin precipitation sites showed significant data scatter and could not be
interpreted readily. TP loading thus may have a large dry fallout component.
Ca loadings are dependent on precipitation volume and range from ^
6 ug/cm2/mo at 30 cm to ^ 12 yg/cm2/mo at 60 cm. SunrL et_ al., (1959)
identified gypsum (CaSOj,.2HpO) in urban precipitation and attributed con-
sistently high Ca levels to its dissolution. The precipitation dependency
of Ca is inconsistent with the hypothesis that soil or dust is its major
source to the atmosphere. Pearson and Fisher (1971) reported that Ca load-
ing in bulk precipitation in the northeastern U.S. was independent of pre-
cipitation amount.
Figure X portrays similar loading-precipitation relationships for
Si02, SO^ and TOC. Sulfate in precipitation originates primarily from
30
-------
56
£ 48
oc
o 40
z: o
9 ^ 32
<< CN
g |
LU 05
H- =*-
< 16
_J
3 8
0
1 56
° 48
<
CJ
o
CD
§
o
I-
NORTHERN
b 5/4S//V
1
1
SOUTHERN BASIN
Figure 8.
4 6 8
MONTH
Seasonal variations in atmospheric loading rates
of sulfate and organic carbon to the southern
and northern basins of Lake Michigan - 1976.
31
-------
30
20
CM
10
LLJ
h-
o
o
o
0.610
0.4
0.2
A. Calcium
O NORTHERN SITES
• SOUTHERN SITES
t
1
B. Total phosphorus
I
1
20
30 40 50
PRECIPITATION, cm
60
70
Figure 9. Relationship of atmospheric total phosphorus and
calcium loading rates to precipitation amount in
Lake Michigan - 1976.
32
-------
DU
40
20
o
^60|0
CJ
O5
^ 40
LU
1-
cc
5
g eoio
40
20
n
A. Silica
_ 0 NORTHERN S/TES
• SOUTHERN S/TES
o •
P— o-do7T-T°TT «>*
B. Sulfate
—
• ,
* *o /*
0^2^o-o^0-Tr^
"" ^ I ° I 1
C. Total Organic Carbon
—
. * • ...
0 oCL-Ao-0-0- *
— — - "°o
1 1 1 1
20
30 40 50 60
PRECIPITATION, cm
70
Figure 10. Relationship of atmospheric silica, sulfate and
organic carbon loading rates to precipitation
amount in Lake Michigan - 1976.
33
-------
solution of aerosol sulfate (<2ym) and gaseous H?S (wetlands) and S0p (fossil
fuel combustion). The physical state of both forms suggests rainout as a
major removal mechanism which is supported by its dependency on precipitation.
Andren (personal communication, University of Wisconsin, 1977) has found
that a large fraction of the atmospheric organic carbon collected over Lake
Michigan occurred in aerosol-size particles. The small particle size suggests
that atmospheric residence times may be long. The dependency of TOG loading
on precipitation amount shown in Figure X supports the hypothesis that
aerosol particles are removed by washout. SiCL loading appears to be in-
dependent of precipitation amount, but some scatter exists limiting inter-
pretation. SiCL may be derived from soil particles or power plant facilities
in the form of fly ash.
Further information on sources of atmospheric P may be derived by com-
paring paired-element loading rates. If the principal source of Ca and Mg
to the atmosphere is soil or re-entrained dust, then elements also derived
from soil ought to be closely correlated. Figure XI shows the relationship
of TP loading rate to that of Ca and Mg in units of yg/cm2/mo. Each data
point is the mean monthly loading rate average over all months of 1976.
The strong relationship between TP and Ca, Mg loading as judged by the high
correlation coefficient of the linear regression plots Implies that all
have a coranon source and/or removal mechanism. In addition, the near-zero,
y-intercept of the TP-Mg plot suggests that other sources are quantitatively
unimportant. The non-zero, y-intercept of the TP-Ca plot can be explained
by a depletion of Ca relative to P in precipitation. Possible explanations
include formation of participate Ca (i.e., CaCO-0 on standing which would
be removed by filtration prior to measurement, or to scatter in data result-
ing from comparison of urban and rural areas. However, soil and re-en trained
dust are significant sources of atmospheric P in the Lake Michigan Basin.
Alternate techniques of source determination include comparison of
elemental ratios and source enrichment factors, and determination of elemental
mass balances. Calculation of P/Ca ratios in bulk precipitation yielded
values ranging from0.011 to0.035 compared to the crustal value of 0.032
(Poldervaart, 1955). Soil would be depleted in Ca due to natural weathering
increasing the P/Ca ratio. However, the average P/Ca ratio for all sites of
0018 was similar to the ratio determined for bulk precipitation in the Upper
Great Lakes (CCIW, 1976). Further evaluation of these techniques was aban-
doned until better soil tracers such as Al are determined. Data thus far
presented suggests that wind-blown soil or re-entrained dust in the case of
urban areas may be the principal source of P in bulk precipitation. If
true, a comparison of monthly bulk and wet-only precipitation should dem-
onstrate the importance of dry deposition. Table X lists the weighted-
average concentrations of bulk and wet-only precipitation obtained with.AEC
samplers at an urban (Milwaukee) and rural (Kewaunee) site. Loading rates
are not reported for wet-only samplers because collection efficiency de-
creased to ^ 25% of the bulk collector in winter months.
At both the rural and urban sites, TP in bulk precipitation exceeded
wet-only concentrations by 600 to 700/S, with higher TP concentrations
observed at the Milwaukee site which is heavily influenced by local industry.
34
-------
o
JE
CM
E
o
O5
V
LLJ
I —
r^
DC
•z.
Q
O
_i
CO
DC
0
X
Q_
CO
O
X
-!
1-
o
0.5
0.4
0.3
0.2
0.1
0
(
0.5
0.4
0.3
0.2
0.1
n
•
A. Magnesium
_ •
•
>^ •
^k ^r
^^
•
* ^
s m
'•
m. • S •*
• /+• y = O.OSx + 0.02
_ / n = 19
.S r = 0.826
^ 1 1 1 1 1
) 1 234 56
•
B. Calcium *
/ •
/
• /
s
~ / •
• /
• /
- /*:
^^•* K = 0.01x + 0.110
/ • n = 20
r = 0.875
1 I 1 I 1
0 10 20 30 40 50 60
CATION LOADING RATE, jug/cm2/mo
Figure 11. Relationship of atmospheric phosphorus and calcium,
magnesium loading rates to Lake Michigan - 1976.
35
-------
TOP and DRP concentrations at the urban site were not significantly
different demonstrating the refractory nature of the particulate P de-
rived from industrial sources (i.e., fly ash), Small, but measurable
differences were observed between wet-only and bulk precipitation con-
centrations of dissolved P forms at the rural site suggesting that partic-
ulate P may be partially solubilized by water.
Lower pH and particulate concentrations were observed for wet-only
precipitation. The .free acidity noted in the wet-only samples is partially
neutralized by particulate matter.
In summary, atmospheric P loading to Lake Michigan is correlated closely
with Ca and Mg loading, is independent of precipitation amount in the north-
ern basin, exhibited a large dry fallout component and is much greater in
the southern than the northern basin. Wind-blown soil and re-entrained
dust represent a significant source of atmospheric P addition to Lake
Michigan.
36
-------
Table X
COMPARISON OF BULK AND WET-ONLY PRECIPITATION*1
MILWAUKEE (Site #8) KEWAUNEE (Site #5)
Wet-Only Bulk Wet-Only Bulk
TP
TOP
DRP
PART
PH
0.031
0.007
0.002
14.2
4.36
0.190
0.009
0.003
225.5
5.34
mg/1
0.008
0.003
< 0.001
6.5
4.48
0.057
0.010
0.004
25.6
5.96
aWeighted-average concentrations for April-November, 1976 (#8) and
May-December, 1976 (#5)
37
-------
REFERENCES
1. Allen, S. E., Carlisle, A., White, E. J., and Evans, C. C., 1968.
The plant nutrient content of rainwater. J. Ecol., 56: 497-504.
2. Armstrong, P.A.J. and Schindler, D. W., 1971. Preliminary chemical
characterization of waters In the Experimental Lakes Area, North-
western Ontario, J. Fish. Res. Bd. Can., 28: 171-187.
3. Beeton, A. M., 1965. Eutrophication of the St. Lawrence Great Lakes.
Limnol. Oceanogr., 10: 240-254.
4. Beeton, A. M., 1969. Changes in the environment and biota of the
Great Lakes. In Eutrophication: Causes, Consequences, Correctives,
pp. 150-187, National Academy of Sciences, Washington, B.C.
5. Berry, R. L., 1975. Report on the preliminary phase of the TOPS
project. Atmos. Environ. Service, Internal Report, ARQS-1-75,
Downsview, Ontario, Canada.
6. Brezonik, P. L., 1975. Nutrients and other biologically-active
substances in atmospheric precipitation. Proc. First Spec. Symp.
on Atmospheric Contribution to the Chemistry of Lake Waters. Intern.
Assoc. Great Lakes Res., p. 166.
7. Brezonik, P. L. and Hendry, C. D., 1977. Chemical investigations of
precipitation in north central Florida. Presented 40th. Ann. Mtd.
Amer. Soc. Limnol. Oceanogr., June 20-23, East Lansing, Michigan.
8. Canada Centre for Inland Waters, 1975- Atmospheric Loading of the
Upper Great Lakes, Vols. 2 and 3, Acres Consulting Services, Ltd.
9. Casey, D. J. and Salbach, S. E., 1974. IFYGL stream materials balance
study. Proc. 17th. Conf. on Great Lakes Res., 668-681.
10. Chapra, S. C., 1977. Total phosphorus model for the Great Lakes.
J. Env. Eng. Div., ASCE, 103, l47-l6l.
11. Eisenreich, S. J., Barmerman, R. T., and Armstrong, D. E., 1975-
A simplified phosphorus analysis technique. Environ. Lett., £ (1):
43-53.
12. Galloway, J. N., 1976. Critical factors in the collection of precipi-
tation for chemical analysis. Proc. First Spec. Symp. on Atmospheric
Contribution to the Chemistry of Lake Waters. Internat. Assoc. Great
Lakes Res. 1^ Supplement 1:65-81.
38
-------
13. Gatz, D. P., 1975- Relative contribution of different sources of
urban aerosols: application of a new estimation method to multiple
sites in Chicago, Atmos. Env., £: 1-18.
14. Gore, A.J.P., 1968. The supply of six elements by rain to an upland
peat area. J. Ecol., 56: 483-495.
15. International Joint Commission, 1976. The waters of Lake Huron and
Lake Superior. Vol. I. Summary and recommendations. 236 pp.
16. Kluesner, J. W., 1972. Nutrient transport and transformation in
Lake Wingra, Wisconsin. Ph.D. thesis, University of Wisconsin,
Madison.
17. Kramer, J. R., 1976. Assessment of the ecological effects of long-
term atmospheric material deposition. McMaster University, Hamilton,
Ontario, 83 p.
18. Likens, G. E., 1972. The chemistry of precipitation in the Central
Finger Lakes Region. Water Res. Marine Center, Tech. Rept. 50,
Cornell University, Ithica, New York, 47 p.
19. Maier, W. J., and McConnell, H. L., 1974. Carbon measurements in
water quality monitoring. J. Wat. Poll. Control Fed., 46: 623.
20. Matheson, D. H., 197^. Measurement of atmospheric inputs to the
Great Lakes. Unpublished project report, Canada Centre for Inland
Waters, Dept. of the Environment, Canada.
21. Mortimer, C. H., 1976. Physical characteristics of Lake Michigan
and its responses to applied forces. In: Physical Limnology
of Lake Michigan, Vol. 2, Environmental Status, Great Lakes Region,
Argonne National Laboratory.
22. Murphy, T. J., 1974. Sources of phosphorus inputs from the atmosphere
and their significance to oligotrophic lakes. Water Res. Center,
Univ. of Illinois, Urbana, 111., Res. Rept. No. 92.
23- Murphy, T. J., and Doskey, P. V., 1976. Inputs of phosphorus from
precipitation to Lake Michigan. J. Great Lakes Res., 2^, (1); 60-70.
24. Pearson, F. J. Jr., and Fisher, D. W., 1971. Chemical composition
of atmospheric precipitation in the northeastern United States. Geol.
Survey Water-Supply Paper 1535-P: 1-23.
25. Pollution from Land Use Activities Reference Group (PLUARG), Inter-
national Joint Commission, Volume I, November, 1974.
26. Poldervaart, A., 1955. The chemistry of the earth's crust. GSA Spec.
Paper No. 62: 119-144.
39
-------
27. Rawls, W. J., Robertson, D. C., and Zuzel, J. P., 1975. Comparison
of precipitation gage catches with a modified alter and rigid alter
type windshield. Water Res. Res., 11 (3): 415-417.
28. Schelske, C. L., and Roth, J. C., 1973- Limnological survey of Lakes
Michigan, Superior, Huron and Erie. Publ. No. 17, Great Lakes Res.
Div., University of Michigan, Ann Arbor, Mich.
29. Schelske, C. L. and Stoermer, E. F., 1971. Euthrophication, silica
depletion, and predicted changes in algal quality in Lake Michigan.
Science, 173: 423-424.
30. Shiomi, M. T., and Kuntz, K. W., 1973- Great Lakes precipitation
chemistry: Part 1. Lake Ontario basin. Proc. 16th. Conf. on
Great Lakes Res., IAGLR, 581-602.
31. Sonzogni, W. C., and Monteith, T. J., 1977- Great Lakes shoreline
erosion: chemical loading. Presented 20th. Conf. on Great Lakes
Res., IAGLR. University of Michigan, Ann Arbor.
32. Standard Methods for the Examination of Water and Wastewater, APHA,
13th. Ed., 1971.
33. Sumi, L., Corkery, A., and Monkman, J. L., 1959. Calcium sulfate
content of urban air: Baltimore, Md., Am. Geophys. Union Geophys.
Mon., 3: 69-80.
34. Upchurch, S. B., Chemical characteristics of the Great Lakes, 1976.
In Great Lakes Basin Framework Study, Appendix 4, Limnology of Lakes
and Bribayments, Great Lakes Basin Commission.
35- U.S. Environmental Protection Agency, 1974. Report of the Phosphorus
Technical Committee. In Proc. 4th. Lake Michigan Enforcement Conf.,
Sept., 1972. Chicago, 111., p. 209.
36. Vaughn, J. C., and Reed, P., 1974. Progress report on water quality
of Lake Michigan near Chicago. In Proc. 4th. Lake Michigan En-
forcement Conf., Sept., 1972, Chicago, 111., p. 88.
40
-------
APPENDIX A
SAMPLING STATION LOADING VALUES-1976a
Site Number 1: Watersmeet, Michigan
Number
Mln
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
TP
12
.039
.493
.161
.140
K
11
.073
2.73
.829
.825
TOP
12
.008
.210
.048
.054
Part.
12
5.13
156
48.5
43.4
Si02
11
.060
.661
.273
.185
Cl
11
ND
11.9
7.54
2.60
TOC
9
2.55
35.9
14.4
12.6
so4
n
• ND
15.6
6.2
4.28
Ca
11
.769
11.3
4.72
3.66
PH
n
4.30
5.80
4.69
4.86
Mg
11
.119
1.53
.653
.444
Vol.
43.7
Na
12
.282
2.50
.874
.591
a Loading values are given in units of ug/cm /month.
Volume represents total bulk precipitation in cm.
41
-------
APPENDIX A, Contd.
Site Number 2: Stevens Point. Wisconsin
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
TP
12
.038
.485
.176
.143
K
11
.242
1.50
.734
.439
TP
10
.055
.331
.141
.091
K
10
.170
1.70
.739
.538
TOP
12
.009
.287
.041
.078
Part.
11
4.53
144
65.1
45.3
TOP
10
.012
.124
.042
.035
Part.
7
5.69
65.8
40.1
22.8
Si02
11
ND
.541
.337
.161
Cl
10
ND
14.2
5.28
4.31
Site Number
sio2
10
.060
.841
.276
.227
Cl
7
ND
10.7
8.11
1.81
TOC
11
1.03
24.6
9.80
8.40
so4
9
ND
22.2
9.40
6.46
3: Green
TOC
9
4.80
17.0
12.9
11.5
so4
6
ND
20.9
12.7
7.11
Ca
11
2.36
14.8
6.62
3.77
PH
12
4.60
7.50
5.23
5.12
Mg
11
.332
3.12
1.42
.977
Vol.
41.9
Na
11
.498
69.6
7.44
20.6
Bay, Wisconsin
Ca
10
1.34
13.9
6.64
4.16
pH
10
5.0
6.9
5.62
5.45
Mg
10
.228
4.33
1.83
1.39
Vol.
37.6
Na
9
.474
1.70
.888
.454
-------
APPENDIX A, Contd.
Site Number 4: Sister Bay, Wisconsin
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
TP
9
.041
.901
.191
.268
K
9
.152
1.23
.587
.349
TP
10
.049
.227
.155
.051
K
8
.160
1.31
.523
.405
TOP
9
.006
.062
.028
.017
Part.
8
1.52
699
126
236
TOP
10
.010
.056
.025
.014
Part.
10
10.5
155
76.3
53.1
Si02
8
.060
1.26
.353
.406
Cl
8
ND
11.7
8.25
3.68
Site Number
Si02
10
.060
.541
.252
.141
Cl
9
ND
15.7
7.74
5.66
TOC
10
ND
42.9
14.5
16.0
so4
7
1.68
11.4
6.92
3.79
Ca
9
1.40
12.6
7.09
4.39
PH
11
5.0
6.9
5.62
5.45
Mg
9
.334
5.27
2.07
1.75
Vol.
30.4
Na
9
.369
1.05
.677
.244
5: Kewaunee, Wisconsin
TOC
10
.268
21.3
12.2
6.8
S04
9
3.61
49.8
16.2
14.5
Ca
9
1.96
16.3
8.22
5.29
PH
10
5.5
7.3
6.00
5.91
Mg
9
.480
4.31
2.05
1.23
Vol.
40.0
Na
8
.253
21.8
3.58
7.37
43
-------
APPENDIX A, Contd.
Site Number 6: Sheboygan, Wisconsin
Number
Kin
Max
Mean
Std.Dev.
Number
Win
Max
Mean
Std.Dev.
Number
Hin
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
TP
11
.052
.704
.189
.183
K
10
.101
5.64
1.62
1.75
Site
TP
12
.035
.527
.266
.148
K
12
.105
5.18
1.68
1.67
TDP
11
.006
.444
.066
.127
Part.
10
23.8
104
62.1
27.9
Number
TDP
11
.010
.171
.042
.044
Part.
12
20.5
398
141
113
Si02
11
.060
.661
.268
.198
Cl
10
ND
20.4
10.6
6.51
TOC
11
3.09
37.5
14.7
10.6
so4
11
ND
36.3
13.8
10.9
Ca
11
1.87
15.5
8.15
4.83
PH
11
4.6
6.8
5.23
5.04
Mg
11
.345
5.37
2.39
1.53
Vol.
55.9
Na
10
.379
3.36
1.49
1.04
7: Milwaukee, Wisconsin (Linwood Ave.)
Si02
11
.060
1.26
.612
.376
Cl
11
ND
19.2
13.6
3.37
TOC
12
3.59
35.5
20.5
9.7
so4
8
5.23
34.7
22.6
8.99
Ca
12
1.68
40.9
16.7
11.7
PH
11
5.0
7.2
5.49
5.39
Mg
12
.259
9.21
4.21
3,23
Vol.
65.4
Na
12
.569
11.8
3.61
3-72
44
-------
APPENDIX A, Contd.
Site Number 8: Milwaukee. Wisconsin (GLRF)
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
TP
12
.176
1.67
.760
.420
K
12
.362
2.62
1.26
.651
Site
TP
12
.136
.856
.388
.210
K
12
.517
4.32
1.83
1.04
TDP
12
.012
.105
.038
.025
Part.
12
279
4276
1252
1278
Number
TDP
12
.017
.255
.060
.067
Part.
12
5.87
938
340
307
Si02
12
.180
2.71
1.21
.710
Cl
11
14.7
31.9
22.3
6.43
TOC
11
77.0
862
327
249
S04
9
11.2
69.5
34.2
16.6
Ca Mg
12 12
6.40 .418
53.4 17.3
22.4 4.74
13.4 4.62
pH Vol .
11
3.7
6.8
4.73 64.1
4.22
Na
12
2.51
20.4
9.49
5.86
9: Milwaukee, Wisconsin (Mitchell Field)
Si02
12
.421
19.5
4.64
5.18
Cl
11
ND
40.0
21.2
11.1
TOC
12
10.1
62.6
29.1
16.0
so4
10
12.9
54.7
27.2
12.0
Ca Mg
12 12
11.0 .352
88.2 17.5
37.6 5.15
22.6 4.96
pH Vol .
11
6.1
6.9
6.45 69.5
6.71
Na
12
.817
23.4
7.26
7.26
45
-------
APPENDIX A, Contd.
Site Number 10: Milwaukee. Wisconsin (CHS)
Number
Win
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
TP
12
.058
.743
.276
.173
K
12
.201
2.69
1.10
.916
TP
11
.066
.797
.292
.202
K
11
.107
1.93
.817
.475
TOP
12
.008
.061
.032
.018
Part.
12
28.5
785
187
213
TOP
12
.005
.068
.031
.021
Part.
11
33.9
390
136
123
Si02
11
.120
1.26
.623
.394
Cl
11
ND
23.8
16.4
4.32
Site Number
Si02
11
.120
2.34
1.19
.751
Cl
11
ND
12.1
9.07
3.00
TOC
12
8.29
57.3
26.1
12.7
S04
9
3.20
41.3
20.2
10.5
11: Zion
TOC
9
3.73
144
33.7
42.5
S04
9
2.37
88.1
28.6
24.8
Ca Mg
12 12
3.53 .368
27.6 14.9
14.8 4.04
8.43 3.82
pH Vol .
10
5.1
6.9
5.50 64.6
5.55
, Illinois
Ca Mg
11 11
2.19 .169
41.6 11.4
19.5 4.18
11.2 3.46
pH Vol .
10
5.2
6.9
5.83 59.1
5.73
Na
12
.210
11.0
3.54
3.19
Na
11
.906
13.5
3.03
3.78
46
-------
APPENDIX A, Contd.
Site Number 12: Wllmette, Illinois
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
TP
11
.099
.597
.298
.157
K
10
.308
4.47
1.84
1.32
TP
12
.272
.997
.591
.281
K
11
.874
5.01
2.19
1.48
TOP
10
.006
.469
.082
.138
Part.
10
15.4
215
102
55.9
Site Number
TOP
12
.011
.436
.110
.155
Part.
11
81.5
1009
287
261
sio2
10
ND
2.22
1.10
.695
Cl
8
ND
14.9
12.3
1.94
TOC
10
6.03
50.3
22.8
12.9
so4
9
6.18
31.8
21.2
8.72
Ca
10
5.24
32.5
15.4
8.46
pH •
11
4.8
6.8
5.51
5.36
13: Chicago, Illinois (Shedd
Si02
11
.240
5.35
2.43
1.53
Cl
10
ND
68.8
33.2
20.0
TOC
11
24.4
109
57.1
25.7
so4
10
18.4
52.5
31.1
11.3
Ca
11
14.4
150
39.3
38.7
PH
11
5.6
7.6
6.19
6.15
Mg
10
.744
6.88
3.32
2.25
Vol.
65.7
Aquarium)
Mg
11
.781
16.2
4.95
4.77
Vol.
65.3
Na
10
.483
7.41
2.74
2.45
Na
11
2.38
24.2
7.62
6.73
47
-------
APPENDIX A, Contd.
Site Number 14: Chicago. Illinois (DePaul Univ.)
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
TP
12
.184
.738
.452
.175
K
12
.685
2.51
1.35
.595
TP
12
.208
1.00
.518
.245
K
12
.582
6.92
2.76
1.79
TOP
12
.009
.066
.036
.015
Part.
12
26.2
3345
501
920
TOP
12
.024
.083
.051
.022
Part.
12
64.6
953
368
306
Si02
12
.240
3.25
1.78
1.05
Cl
11
ND
36.2
20.4
8.67
Site Number
S102
12
1.32
13.5
6.55
4.12
Cl
11
ND
38.0
20.2
9.63
TOC
11
22.5
61.3
36.4
13.2
so4
11
10.2
42.7
27.0
9.90
Ca
12
9.98
39.9
22.1
8.20
PH
12
5.8
7.5
6.21
6.36
Mg
12
1.10
7.47
3.92
2.48
Vol.
51.0
Na
12
1.94
11.8
5.08
3.84
15: Calumet, Illinois
TOC
12
27.2
241
93.9
70.2
so4
11
20.3
105
56.3
25.1
Ca
12
24.8
69.3
49.3
16.9
i
pH
12
6.1
7.8
6.54
6.64
Mg
12
1.11
11.4
5.12
3.50
Vol.
65.7
Na
12
1.38
12.1
4.58
3.38
48
-------
APPENDIX A, Contd.
Site Number 16: Indiana Dunes State Park
Number
Win
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
TP
9
.019
.566
.199
.176
K
9
.302
22.1
3.26
7.11
TP
11
.092
.317
.190
.077
K
11
.136
2.87
1.18
, .740
TOP
9
ND
.051
.029
.018
Part.
7
35.6
200
137
64.9
Site
TOP
11
.009
.067
.033
.020
Part.
11
3.00
551
123
153
Si02
8
1.14
5.35
2.56
1.44
Cl
8
5.69
29.1
14.9
9.39
Number
Si02
11
.060
1.92
1.01
.622
Cl
10
ND
39.7
17.7
10.2
TOC
9
11.4
55.8
24.9
14.5
so4
7
10.6
56.6
31.2
16.6
Ca Mg
9 9
6.68 .436
34.7 6.91
17.9 2.84
10.3 2.32
pH Vol .
9
5.5
7.4
6.22 46.0
6.00
Na
9
.354
6.06
2.42
2.12
17: Ludington, Michigan
TOC
11
4.32
36.5
17.5
9.30
SO
4
9
ND
35.6
21.2
9.58
Ca Mg
11 11
5.71 .635
34.2 23.8
2.54 8.69
1.77 8.65
pH Vol .
11
5.9
8.3
6.46 60.2
6.35
Na
11
.825
7.25
2.74
2.09
49
-------
APPENDIX A, Contd.
Site Number 18: Beaver Island. Michigan
Number
Min
Max
Mean
Std.Oev.
Number
Min
Max
Mean
Std.Dev.
IP
10
.032
.466
.180
.129
K
11
.217
2.73
.851
.788
TOP
11
.012
.056
.030
.017
Part.
10
4.54
141
43.0
42.0
Si02
11
.027
14.3
1.63
4.21
C1
10
ND
25.9
10.9
9.06
Site Number 21:
Number
Min
Max
Mean
Std.Oev.
Number
Min
Max.
Mean
Std.Oev.
TP
10
.055
1.15
.274
.359
K
10
.050
2.15
.863
.646
TOP
10
ND
.187
.056
.058
Part.
10
16.3
157
52.5
41.9
S102
10
.117
4.17
.671
1.24
Cl
9
ND
16.5
12.2
4.19
TOC
11
ND
59.6
15.3
16.5
so4
10
ND
23.1
10.6
7.15
Ca
11
3.68
44.1
11.8
11.9
PH
10
4.7
6.8
5.14
5.08
Sleeping Bear Dunes,
TOC
10
ND
59.0
14.2
17.0
so4
10
ND
30.9
13.1
7.45
Ca
10
6.47
25.7
12.9
5.44
PH
10
5.2
7.6
5.90
5.71
Mg
11
.526
9.13
2.19
2.48
Vol.
45.4
Michigan
Mg
10
.691
6.23
3.01
1.54
Vol.
52.7
Na
11
.064
2.52
.875
.662
Na
10
.351
1.98
1.12
.521
50
-------
APPENDIX A, Contd.
Number
Hin
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
Number
Min
Max
Mean
Std.Dev.
TP
n
.055
1.51
.346
.433
K
12
.355
18.1
2.60
4.97
TP
11
.023
.781
.228
.212
K
11
.178
2.05
1.00
.762
Site
TOP
11
.009
1.24
.149
.362
Part.
12
2.29
357
83.9
97.9
Site
TOP
11
ND
.092
.039
.028
Part.
11
7.48
248
92.4
74.6
Number 22: Grand Haven, Michigan
Si02
11
.110
1.23
.733
.369
Cl
11
ND
18.8
10.3
5.50
TOC
12
ND
36.2
16.7
11.3
so4
10
5.04
42.3
19.3
13.0
Ca
12
5.48 1
31.4 9
18.0 3
8.86 2
PH
12
4.7
7.1
5.50
5.16
Mg
12
.17
.96
.53
.83
Vol.
50.2
Na
12
.355
3.63
1.70
.987
Number 23: St. Joseph, Michigan
Si02
11
.247
1.73
.762
.472
Cl
10
ND
21.6
10.9
5.89
TOC
11
ND
69.1
28.2
23.7
so4
9
5.35
49.7
24.0
14.4
Ca
11
3.74
45.9 8.
15.1 3.
11.9 2.
PH
10
4.5
7.0
5.10
4.99
Mg
11
647
48
29
,56
Vol.
67.4
Na
11
.174
4.22
1.83
1.33
51
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/3-80-063
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Determination of Atmospheric Phosphorus Addition
to Lake Michigan
5. REPORT DATE
JULY 1980 ISSUING DATE.
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
S.J. Eisenreich, P.J. Enroling, A.M. Beeton
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Center for Great Lakes Studies
University of Wisconsin-Milwaukee
Milwaukee, WI 53201
10. PROGRAM ELEMENT NO.
1BA608
11. CONTRACT/GRANT NO.
R 803238
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Research Laboratory-Duluth
Office of Research and Development
U.S. Environmental Protection Agency
Duluth, MN 55804
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
EPA/600/03
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Bulk precipitation was collected on a monthly basis in the Lake Michigan basin
for 18 months during 1975-1976 to determine atmospheric P loadings to Lake
Michigan. The sampling network consisted of bulk collectors at 23 land stations
and 2 in-lake buoys located off urban and rural areas. Annual TP loading to
Lake Michigan for 1976 based on loading rates of 0.184 and 0.303 yg/cm2/mo for
the north and south basins, respectively was 1.69x10^ kg/year, representing
•^16% of the total P budget. North and south basins showed different TP loading
properties with *^62% depositing in the south basin. Seasonal variations in
loading rates showed maxima in late spring and early summer, and minima in
winter. In-lake buoy samplers yielded enhanced loading of TP compared to land
stations. Wet-only precipitation concentrations for TP averaged 15% of bulk
deposition values. Weighted-average TP concentrations in bulk precipitation
were 0.050 and 0.064 mg/1 in the north and south basins, respectively.
Atmospheric TP loading was closely correlated to Ca and Mg loading, independent
of precipitation amount, and exhibited a large dry fall component. Wind-
blown soil and re-entrained dust are believed to be the major sources of
atmospheric phosphorus addition to Lake Michigan.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Precipitation
Atmosphere
Dryfall
Phosphorus
Rain
Loading
Lake Michigan
08/H
18. DISTRIBUTION STATEMENT
Release to the Public
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
60
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
U.S. GOVERNHENT PRINTIN6 OFFICE: 1980—657-165/0039
52
------- |