EPA-600/3-75-005
December 1975
Ecological Research Series
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ECOLOGICAL RESEARCH series This series
describes research on the effects of pollution on humans, plant and animal
species, and materials. Problems are assessed for their long- and short-term
influences. Investigations include formation, transport, and pathway studies to
determine 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, Springfield, Virginia 22161.
-------
EPA-600/3-75-005
December 1975
INPUTS OF PHOSPHORUS FROM PRECIPITATION
TO LAKE MICHIGAN
by
T. J. Murphy
P. V. Doskey
DePaul University
Chicago, Illinois 60614
Grant No. 80261*7
Project Officer
Michael D. Mull in
Large Lakes Field Station
Environmental Research Laboratory Duluth
Grosse Me, Michigan ^8138
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
ENVIRONMENTAL RESEARCH LABORATORY —-DULUTH
DULUTH, MINNESOTA
-------
DISCLAIMER
This report has been reviewed by the Office of Research and Development,
EPA, and approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
ii
-------
ABSTRACT
Precipitation samples were collected at six locations around Lake
Michigan and analyzed for the different forms of phosphorus present.
It was found that the atmosphere is presently contributing one million
kilograms per year of phosphorus or about 18 percent of the phosphorus
budget of the Lake. As the phosphorus removal program on sewage
effluents becomes fully implemented in the Lake Michigan basin, the
contribution to the Lake of phosphorus from particulate matter
scavenged by precipitation could increase to about 30 percent of the
total.
The phosphorus concentration of precipitation was found to be higher
at the southern end of the Lake. More than 40 percent of the phos-
phorus in precipitation is dissolved reactive phosphates and the amount
of the dissolved reactive phosphates in precipitation was found to be
somewhat dependent on the pH of the sample.
The washout ratio for phosphorus by precipitation was determined. It
was also found from the analysis of some glacial samples that phosphorus
has long been a component of precipitation.
This report was submitted in fulfillment of Grant Number 802647 by
DePaul University under the sponsorship of the Environmental Protection
Agency. Work was completed as of December 31, 1974.
iii
-------
CONTENTS
SECTIONS Page
I. Conclusions 1
II. Recommendations 2
III. Introduction 3
IV. Experimental Design 5
V. Results 10
VI. Discussion 21
VII. References 25
-------
FIGURES
Number
1 Precipitation weighted concentrations (mg/1.) of
the forms of phosphorus in rainfall in different
locations around Lake Michigan. 13
2 Distribution of pH in precipitation samples in Chicago. 16
3 Inputs of Phosphorus per Precipitation Event. 23
VI
-------
TABLES
Number Page
1 Total Phosphorus in Precipitation 10
2 Dissolved Reactive Phosphates in Precipitation 11
3 Total Reactive Phosphates in Precipitation 12
4 Species of Phosphorus Present in Precipitation 13
5 Estimated Inputs of Phosphorus to Lake Michigan 14
6 Phosphorus Concentrations in Snow Samples 15
7 Rainfall on April 4, 1974 16
8 Percent Total Phosphorus in Particulate Matter from
"Red Rains" 18
9 Washout Ratio 19
10 Phosphorus in Glacial Samples 19
vii
-------
ACKNOWLEDGMENT S
The Authors would like to express their thanks and appreciation to the
following co-operating personnel who did the remote sample collecting.
Mr. Bernard Wagner and Mr. Roy Viterna of the Michigan Department of
Natural Resources, Mr. Darrell Johnson of the Illinois Department of
Conservation, Mr. Phillip Peterson of the Wisconsin Department of
Natural Resources and the eight men on the Harrison crib and the
Chicago Department of Waters and Sewers. Without their unselfish
efforts, this project would not have been possible.
We wish to thank Dr. Chester C. Langway and the U.S. Army Cold Regions
Research Laboratory for generously supplying the glacial samples for
this project.
We would like to acknowledge the help of Mr. Charles P. Rzeszutko and
Mr. Lawrence M. O'Conner with some of the experimental aspects of the
project and the support of DePaul University during the early phases
of this project.
viii
-------
SECTION I
CONCLUSIONS
One million kilograms per year of phosphorus is being brought into
Lake Michigan each year by precipitation scavenging. This is a signi-
ficant factor in the phosphorus budget of the Lake.
The average concentration of phosphorus in precipitation is three times
the 0.008 mg/1. concentration of phosphorus in the Lake. About half of
the phosphorus in precipitation is in the form of dissolved reactive
phosphates and thus should be immediately available to the organisms
in the Lake.
That the concentration of phosphorus in precipitation is higher at the
southern end of the Lake than at the northern end.
That the pH of the precipitation can affect the amount of dissolved
reactive phosphates in precipitation.
That there has been phosphorus in precipitation in the northern
hemisphere for at least the last three centuries.
-------
SECTION II
RECOMMENDATIONS
That measurements of the phosphorus concentration in precipitation over
the Lake be made.
That phosphorus inputs to Lake Michigan from dry fallout be determined.
That a pH be stipulated for the equilibrium of precipitation samples
before filtration for the dissolved reactive phosphates determination.
-------
SECTION III
INTRODUCTION
This project had as its primary objective the determination of the
magnitude of phosphorus inputs to Lake Michigan in precipitation and to
determine the form in which this phosphorus is present in precipitation.
Other objectives were to determine the distribution of phosphorus in
precipitation around the Lake, to develop indirect methods of estimating
the precipitation inputs and to see if it could be determined whether
the inputs via this route had changed through time.
Lake Michigan is the second largest by volume of the Great Lakes and the
only one of the Great Lakes entirely within the United States. It has
enormous value for sport and commercial fishing, for shipping, as a
recreational area and as a source of fresh water supplies for industrial
and municipal uses. It is still considered to be in an oligotrophic
state although some areas show serious signs of degradation. In recent
years a number of studies have shown that phosphorus is the principal
limiting nutrient in the Lake (Miller, 1974; Schelske et al., 1974).
There are indications that the phosphorus concentrations in the Lake
are increasing (Vaughn, 1974; Lee, 1974). This presumably is the cause
for the observed increases in the number of algal cells and changes
in the species of algas in the Lake in recent years (Schelske et al.,
1972; Stoermer, 1974). Because of the changing conditions in the
Lake and the indications that phosphorus inputs are the cause,
controls on the discharge of phosphorus in sewage effluents to the
Lake or its tributaries were put into effect starting in 1973. In
order to be able to evaluate the effect of this program and to formulate
subsequent programs, the phosphorus budget of the Lake must be known. A
determination of the phosphorus inputs to the Lake from tributaries
and direct discharges in 1971 was made by the U. S. Environmental Pro-
tection Agency (1974). To date, no determination of inputs of phosphorus
from precipitation have been made.
Inputs of nutrients in precipitation are a component of the nutrient
budget of bodies of water. These inputs should be most significant
in those cases where the lake is oligotrophic and occupies a significant
portion of its drainage basin. This is true of Lake Michigan which
occupies one third of its drainage basin, receives half of its water
budget directly from precipitation, is relatively deep and has a
hydraulic residence .time greater than 100 years.
The chemical form of the phosphorus entering the Lake in precipitation
is important since organisms utilize the phosphorus primarily in the
form of ortho phosphate. Not all of the phosphorus entering the Lake
can become available as ortho phosphate. Thus ortho phosphate and phos-
phates which can be hydrolyzed to ortho phosphate in the Lake are
important while phosphorus which is strongly bound in organic or
-------
inorganic compounds may never become available as ortho phosphate and
therefore has no effect on the biota of the Lake. Most of the phos-
phorus present in eroded materials and much of the phosphorus dis-
charged to tributaries far from the Lake is thought not to become
available as ortho phosphate in the Lake.
To carry out this project, the following plan was developed. Precipi-
tation samples would be collected in several locations around the Lake.
These samples would be analyzed for the different forms of phosphorus
present and the inputs of the different forms to the Lake would be
calculated from the concentrations and the amount of precipitation.
These inputs would be compared to inputs of phosphorus to the Lake from
other sources.
The washout ratio for phosphorus containing particulate matter would be
determined. The washout ratio is the ratio of the phosphorus in
precipitation to the amount of phosphorus in the atmosphere. Knowing
this ratio would permit the estimation of the phosphorus concentration
in precipitation by the determination of the particulate phosphorus
concentration in the atmosphere. Since high volume air particulate
samples are routinely collected in many areas, this indirect method of
determining phosphorus inputs could have wide applicability to present
and historical samples. Finally, a series of glacial samples collected
in Greenland was obtained. It was hoped that the analysis of these
samples would indicate whether the concentration of phosphorus in the
atmosphere had changed over time and, if so, by what f actoil.
-------
SECTION IV
EXPERIMENTAL DESIGN
COLLECTION
Most atmospheric sampling has been of bulk precipitation. That is, both
wet and dry fallout were collected. The problem with this is that it
does not differentiate between the wet and dry inputs. This can be
important because the particulate matter deposited by these two
different routes can be very different in origin and composition. It
was decided that the project would concentrate on the determination of
the precipitation input component of the atmospheric loading. This was
for several reasons. The first is that the dry fallout component is
chiefly of particles larger than 20 micrometers, giant size aerosols.
These have short atmospheric residence times and impact on the surface
close to their source and therefore would be of only local significance.
They still, however, could be of importance to the phosphorus budget
and may contribute to the increased concentration of nutrients observed
in the inshore waters of the Lake. The size distribution studies of
particulate matter (Lee and Patterson, 1969; Cunningham et al., 1974)
show that the phosphorus is present chiefly on aerosols with a mean
diameter of a.,few micrometers. Thus, these particules will have a long
atmospheric residence time and will be removed from the atmosphere
chiefly by precipitation scavenging. To the extent that precipitation
is uniform on the surface of the Lake, the inputs determined close to
the shore should be a good approximation of the input on the surface
of the Lake.
Trying to collect just wet fallout samples presented a number of
problems. Ideally, the collector would be exposed only during the
precipitation event to minimize the collection of dry fallout. Because
of the activity of microorganisms in the samples, the samples would
have to be collected and then immediately analysed or preserved for
later analysis if the different forms of phosphorus present were to be
accurately determined.
In order to collect the samples, people who lived or worked in the areas
where sampling was desired and who were willing to help with the project
were located. These people were asked to put out a clean collector each
day, or on days when rain was anticipated, and to take in the sample
after the precipitation event or, if no precipitation had fallen, to
change collectors within 24 hours. It was realized that these people
could not be expected to go very far out of their way on a regular
basis to change the collectors and handle the samples for this project
and so a suitable sampling location was found where it would be con-
venient for the person doing the sampling.
The washout ratio was determined by sampling the air before and/or
during precipitation events to determine the phosphorus concentration
-------
in the air and then dividing the phosphorus concentration found in the
precipitation by this number.
Finally, glacial samples from Greenland were obtained from the U. S.
Army's Cold Region Research and Engineering Laboratory (CRREL). The
samples had been collected in an isolated area where snow continually
deposits but where little or no melting occurs. Hence, non-volatile
material deposited with the precipitation accumulates and leaves a
permanent record with each years accumulation of snow. Samples of snow
deposited approximately in the years 1270, 1650 and. 1966 were obtained.
These were analysed for the different forms of phosphates to see if it
could be determined if the phosphorus concentration in the atmosphere
had changed through time.
LOCATION
Six sampling locations were used for the project. The first of these
was located on the roof of the science building at DePaul University
in Chicago. The building (41° 55.4'N; 87° 39.3'W) was in a densely
populated urban area about five km north of the central business area of
Chicago and about two km west of the shore of the Lake. There was light
manufacturing in the immediate vicinity. The roof was about fifteen
meters above street level. The location of the collector on the roof
undoubtedly decreased the number of traffic related aerosol particles
incorporated in the samples. Also, it was above the normal habitat of
insects and birds in the area and no problems with contamination from
these sources were encountered during the project.
A second sampler in the Chicago area was located at the Harrison water
intake crib of the City of Chicago (41° 55'N; 87° 34'W). This crib is
located about four km from the west shore of Lake Michigan, just east
of the City of Chicago. It is manned 24 hours a day by a caretaker and
maintenance crew of 3-4 men. This station was chosen in order to obtain
samples in the Chicago area at a location where it was anticipated that
local inputs would be at a minimum. It was hoped that bias in the
DePaul samples due to very local inputs would be discernable when
compared to the samples collected at the crib.
The third collection site was located at Illinois Beach State Park, about
75 km north of Chicago along the Lake. The sampler was placed in an open
area about 100 m from the shore (42° 26'N; 88° 48'W). The Park extends
along the shore for about six km and is about 1.5 km deep. Beyond the
Park on the west is a suburban area with light manufacturing. Five km to
the south of the collection site is a large coal fired power plant and
a nuclear power plant is located two km to the north. There is a resort
motel and the maintenance buildings for the park in the vicinity of the
sampling site.
The fourth collection site was located at Rock Island State Park (45°
25'N; 86° 49'W), an island maintained as a wilderness to the north of
the Door Peninsula of Wisconsin. This site was about one km from the
-------
nearest inhabited area, a large island with a farming and fishing
community, and more than 30 km from the nearest cities and industrial
centers. The park superintendant did the collecting. Since the pre-
vailing winds are from the west, it was hoped that this site would give
information on the phosphorus in precipitation coming over the Lake in
the northern part of the lake away from the population and industrial
centers.
The fifth collection site was located on Beaver Island, the largest
island in Lake Michigan. The island is located in the northern part of
the Lake about 20 km from the nearest shore. It has an area of about
140 km^ and a permanent population of about 200 people. Its industries
are hunting, tourism and commercial fishing. The island is in a natural
state with much of it covered with forests. Only one farm is being
worked on the island. The sampler was located in an open area about 0.5
km from the center of town in the northeast corner of the island (45°
45'N; 85° 31'W). The forest conservation officer who did the sampling
for the project has been a cooperative weather observer for the National
Weather Service for a number of years. This was an extremely useful
sampling site for the project as it was the closest approximation
arrived at for a sampling site in mid-Lake. With prevailing westerly
winds, the air masses pass over more than 60 km of open water before
getting to Beaver Island. The Beaver Island samples should be lower
in concentration than any samples collected around the shores for two
reasons. First, it is far from the inputs from the industrialized
regions of the Lake and, secondly, being far from any shore some of the
particulate matter has been scavenged by the time those air masses
reach Beaver Island.
The last sampling site was at Silver Lake State Park along the east side
of the Lake at a place where the land juts out into the Lake (43° 37'N;
86° 36'W). The area is rural with fruit the main crop. This site was
expected to give information on the phosphorus concentration of air
masses after they had passed over the Lake. The park superintendant did
the sampling for the project.
The glacial samples were from ice cores taken at Camp Century in
Greenland (77° 10'N; 61° 08?W).
The samplers were, located in open areas and as far from roads as
possible. But since these collectors had to be located near the home
or office of the volunteers, there could be a problem with dust from
roads and other local sources. This problem is minimized by the fact
that the collectors are exposed only during precipitation events when
the raising of dust from roads should be at a minimum.
Polyethylene buckets 25 cm high and with an opening 20 cm in diameter
were used for the collectors. These were free of leachable materials
and were easily cleaned and shipped. A holder was constructed which
raised the bucket about a meter to prevent splash from the ground from
getting into the collector. An adjacent rain gage was used to measure
-------
precipitation. No attempt was made to collect all precipitation events
although as many as possible were collected.
Precipitation samples collected in Chicago were brought in as soon as
practical after the event, the pH of the sample was determined, the
sample was placed in a polyethylene bottle and frozen.
Samples collected at the other locations were brought in by the parti-
cipants as soon as practical, preserved with mercuric chloride (40 mg/1)
and sodium chloride (50 mg/1), placed in a polyethylene bottle and
mailed to Chicago. All necessary containers, boxes, labels, funnels,
holders, etc. were supplied to the participants. Buckets were to be
used only once and then returned to Chicago to be cleaned. This was
to keep the participants from having to clean the buckets, but also
to avoid problems with contamination with detergents and from buckets
being used multiple times between cleanings, etc. A quantity of buckets
was supplied such that an accumulation of 25 or so could be made before
it was necessary that they be returned.
Evaporation of the sample after the precipitation event but before the
sample is collected would concentrate the sample and lead to high
estimates of the inputs. To correct for this, the participants were
requested to return all of sample that was in the collector. From the
amount of rain which fell and knowing the area of the collector, 300 cm2,
the amount of sample which should have been in the collector could be
calculated. Any difference between this amount and the amount actually
returned was assumed to have evaporated and was corrected for.
ANALYSIS
The basic analytical procedure used for phosphorus determinations was
the single reagent method of the EPA (1971). A Bausch and Lomb
spectrophotometer with a 2.5 cm path length sample tube was used for
most analyses. The glacial samples and some other samples of low
concentration, were analysed using a Beckman DK-2 recording spectro-
photometer with a ten cm pathlength cell. Total reactive phosphates
(TRP) analyses were done on unfiltered 12.5 ml samples. Dissolved
reactive phosphates (DRP) determinations were done on 12.5 ml samples
which had been filtered through a 0.45 nm cellulose nitrate (Sartorius
113 06) membrane filter. These membrane filters have been reported to
be free of extractable phosphorus. This was checked on several
occasions and no extractable phosphorus was found. Hydrolyzable phos-
phorus was determined on 25 ml samples digested with 0.5 ml of 11 N
sulfuric acid. Total phosphorus was determined on 25 or 50 ml samples
digested with ammonium persulfate (0.4 gm/50 ml) and sulfuric acid
(11 N, 1.0 ml/50 ml). When possible, total phosphorus and total and
dissolved reactive phosphates analyses were run in duplicate on all
samples. When insufficient sample was available for all of the analyses,
only total and/or dissolved reactive phosphates analyses were run.
A set of standards using ortho phosphate was run each day. The results
from these analyses were used to determine the constants for a linear
equation which related the absorbance of the samples to their phosphorus
8
-------
concentration. When total phosphorus was determined,two or more phos-
phorus standards were run in addition to the ortho standards. To deter-
mine the relationship between the absorbance and the concentrations for
the total phosphorus samples, it was assumed that the slope of the line
was the same for both the ortho and total samples but that the intercept
could be different. The total phosphorus standards were used to deter-
mine a new intercept for the correlation line.
The pH of the samples collected in Chicago was measured with a pH meter
immediately after the sample was collected. The meter was standardized
before use with a 0.05 molar solution of potassium acid phthalate. This
is an NBS certified buffer, is readily prepared, is at a convenient pH
(4.CO at 20°) and contains no phosphorus.
The concentration of the forms of phosphorus found in each sample and
the amount of precipitation that the sample represented were used to
calculate an input in kg/km2 of the different forms of phosphorus in
that sample. The summary results from a group of samples include some
of the amounts of precipitation and inputs in kg/km2, and a precipitation
weighted average concentration is calculated by multiplying the con-
centration of phosphorus in mg/1. by the amount (cm) of precipitation
and converting these units (mg/1000 cm2) to kg/km2 by multiplying by 10.
This gives the input of phosphorus for each sample. The average
concentration then was determined by summing the inputs for each sample,
dividing this total by the total amount of precipitation for those
samples and multiplying by 0.1 to give the concentration in mg/1. This
should be equivalent to the concentration that one would find if a
sampler were placed out for the entire period, no evaporation occurred
and the sample was collected at the end of the period, mixed well and
analyzed.
The glacial samples were analyzed with the same reagents and under the
same, conditions as were used for the other samples. A 10 cm pathlength
absorption cell was used for the measurements and an absorption spectrum
of each sample, blank and standard was run from 600 nm to 950 nm. The
absorption peak at 880 nm was used to calculate the phosphorus
concentration.
Two of the glacial samples had a large absorption at 750 nm which
obscurred the phosphorus absorption. Attempts were made to extract
the phosphorus-molybdate complex from these samples but all the methods
used (Isaeva, 1969; Goings and Eisenreich, 1974; Pakalns, 1972) gave
high and variable blanks with the phosphorus concentrations of about
one ppb.
Arse.nate forms a similar complex to the phosphate under the conditions
used for analysis and its absorption maximum is at 865 nm. This then is
a possible interference in the phosphorus analysis. To check the
possible magnitude of this interference, a number of precipitation
samples were analyzed for arsenate by the method of Johnson (1971).
-------
SECTION V
RESULTS
The weighted average concentration of phosphorus in the forms of total
phosphorus and total and dissolved reactive phosphates for the preci-
pitation samples collected in different regions of Lake Michigan are
shown on Tables 1, 2 and 3. Included is information on the average
amount of precipitation per sample, the average time the collector was
exposed and statistical information. In addition the geometric mean
concentration of the total phosphorus and dissolved reactive phosphate
samples collected at DePaul was .0344 and .015. The distribution around
the Lake of the phosphorus concentrations shown on Tables 1, 2 and 3 is
shown in Figure 1.
Table 1. TOTAL PHOSPHORUS IN PRECIPITATION
DePaul Beaver Silver Illinois Rock Harrison
Univ. Island Lake Beach Island Crib
Total Precipitation (cm)
Phosphorus in Preci-
pitation (kg/km2)
Weighted concentration
of Phosphorus (mg/1.)
Number of Samples
Average Precipitation
per Sample (cm)
Average Sample Exposure
(hr)
Arithmetic Average
Concentration (mg/1.)
Standard Dev. of the
Arithmetic Mean
101.7
32.77
0.032
98
1.04
13.1
0.043
0.034
19.5
3.16
.016
17
1.14
17.6
.032
.038
33.9
8.56
.025
34
1.00
13.3
.034
.024
18.6 11.0
3.71 3.97
.020 .036
14 4
1.33 2.75
18.3 20
.028 .034
.024
29.0
9.15
.032
21
1.38
24
.037
.024
10
-------
Table 2. DISSOLVED REACTIVE PHOSPHATES IN PRECIPITATION
DePaul Beaver Silver Illinois Rock Harrison
Univ. Island Lake Beach Island Crib
Total Precipitation (cm) 55.0 25.7 27.8 15.2 11.0 16.9
Phosphorus in Preci-
pitation (kg/km2) 7.57 1.53 3.88 1.50 1.23 2.30
Weighted concentration
of Phosphorus (mg/1.) .014 .006 .014 .010 .011 .014
Number of Samples 53 16 29 13 4 16
Average Precipitation
per Sample (cm) 1.04 1.6 0.96 1.17 2.75 1.06
Average Sample Exposure
(hr) 12.3 18.4 12.9 17.9 20.0 24.0
Arithmetic Average
Concentration (mg/1.) .018 .007 .020 .014 .012 .016
Standard Dev. of the
Arithmetic Mean .011 .005 .015 .009 - .014
A series of rainfall samples was analyzed in order to be able to cal-
culate how much of the phosphorus was present in the form of hydrolyzable
and organic phosphates. These results are shown in Table 4. Since the
analysis for hydrolyzable phosphates yields the sum of hydrolyzable plus
dissolved reactive phosphates, the values shown have had the dissolved
reactive phosphates subtracted out. Likewise, the organic phosphates
value shown is total phosphorus minus the hydrolyzable and dissolved
reactive phosphates. Values of total reactive phosphates, which are the
dissolved reactive phosphates plus a portion of the hydrolyzable
phosphates, are also shown.
A loading for the Lake in kilograms of phosphorus per year may be
obtained by multiplying the appropriate weighted average concentration
by the area of the Lake, 58,040 km2, by the average precipitation on
the Lake, 74 cm/yr (Jones and Meredith 1973) and by 10 to convert the
units. This factor to convert a concentration in precipitation in mg/1.
to an input in kg/yr to Lake Michigan is then 4.3 x 10'.
A reasonable calculation from the data in Figure 1 would be to use
the average of the DePaul and the crib data for the southern quar-
ter of the Lake, the average of the Illinois Beach and the Silver
11
-------
Table 3. TOTAL DISSOLVED PHOSPHATES IN PRECIPITATION
DePaul Beaver Silver Illinois Rock Harrison
Univ. Island Lake Beach Island Crib
Total Precipitation (cm) 89.6 10.5 21.5 8.1 7.0 30.3
18.4 0.97 2.54 0.99 1.05 8.37
Phosphorus in Preci-
pitation (kg/km2)
Weighted concentration
of Phosphorus (mg/1.) .020 .009 .012 .012 .015 .027
Number of Samples
Average Precipitation
per Sample (cm)
Average Sample Exposure
(hr)
Arithmetic Average
Concentration (mg/1.)
Standard Dev. of the
Arithmetic Mean
92
10
20
21
0.97 1.05 1.08 1.35 2.3 1.44
12.8 17.1 15.5 16.0 21.3 24.0
.025 .018 .017 .023 .016 .034
.017 .012 .014
.012
Lake data for the middle half of the Lake, and the Beaver Island data for
the northern quarter of the Lake. For total phosphorus this would give an
an average concentration of 0.023 mg/1. and an input of 1.0 x 10 kg/yr.
For dissolved reactive phosphates, the average concentration would be
0.011 mg/1. for an input of 0.5 x 106 kg/yr.
These inputs, inputs from tributaries and direct discharges to the Lake
(U. S. EPA 1974), and estimates of changes in these inputs in the future
are shown in Table 5. These data indicate that about 18 percent of
the phosphorus presently going into Lake Michigan is being brought in
with precipitation. The program to remove 80 percent of the phosphorus
from waste water discharges is now 63 percent implemented (Adamkus 1975).
When it is completed, and combined sewer overflows are eliminated, inputs
from precipitation could constitute about 30 percent of the phosphorus
going into Lake Michigan.
The concentration of the different forms of phosphorus were also deter-
mined in the snow samples which were collected. However, principally
because of the problem of accurately determining the amount of snowfall
at the sampling site in Chicago due to uneven snow deposition caused by
structures on the roof of the building and by the presence of adjacent
12
-------
Table 4. SPECIES OF PHOSPHORUS PRESENT IN PRECIPITATION
Total Precipitation (cm)
Dissolved Total
Reactive Hydrolyzable Organic Reactive
37.89
Phosphorus in Precipitation
(kg/km2) 3.79
Weighted Concentration of
Phosphorus (mg/1.) 0.010
Percent of Total Phosphorus 44
Number of Samples 23
Average Precipitation per
Sample (cm) 1.65
Arithmetic Average
Concentration (mg/1.) 0.011
37.89
1.65
.0044
19
23
1.65
0.006
37.89 37.89
3.29 4.68
0.009 0.012
38 53
23 23
1.65
1.65
0.008 0.014
.010
Dissolved Reactive
Phosphates
Total Reactive
Phosphates
.020
.032
O 5O 1OO
km
Total
Phosphorus
Figure 1. Precipitation weighted concentrations (mg/1.) of the forms
of phosphorus in rainfall in different locations around
Lake Michigan.
13
-------
Table 5. ESTIMATED INPUTS OF PHOSPHORUS TO LAKE MICHIGAN
1971°
December 1974'
Future
Waste Waters
Discharges to Lake
Discharges to Tributaries
Combined Sewer Overflows
Erosion
Precipitation0
Total
Million Percent
kg/yr
1.7 20
4.2 48
0.35 4
1.35 16
1.0 12
8.6
Million
kg/yr
0.95
2.1
.15
1.35
1.0
5.5
Percent
17
38
3
25
18
Million
kg/yr
0.35
0.8
1.35
1.0
3.5
Percent
10
23
39
29
aU. S. EPA (1974)
bAdamkus (1975)
GThis work
-------
buildings of similar height, the amount of snow which fell was not
measured. Thus inputs in kg/km^ and weighted concentrations could not
be determined. The concentrations found in a number of snow samples are
shown in Table 6. They seem to be similar (average concentrations for
DePaul samples: total phosphorus = 0.038; total reactive phosphates =
0.023) to those found in rainfall and should not significantly affect
the conclusions based on the results from the rainfall samples.
Table 6. PHOSPHORUS CONCENTRATIONS IN SNOW SAMPLES
Sample Phosphorus (mg/1.) Collection
Number Total DRP TRP Site
001 .041 DePaul
004 .051 .012
012 .036 .020
103 .026
110 .054
113 .034 .019
125 .030 .017
127 .021 .006
128 .061 .027
216 .016 .002 .016
234 .054 .004 .016
2012 .011 .004 .007 Beaver Island
2020 .007 .003
2021 .010 .010
2022 .012 .002 .003
2041 .010 .010
3001 .011 .003 .004 Silver Lake
3002 .007
3006 .012 .007
3007 .021 .019
5010 .058 .024 .035 Rock Island
5011 .021 .014 .016
The pH of 60 rainfall samples collected in Chicago over a two year
period was determined. The precipitation weighted average hydrogen
ion concentration corresponded to a pH of 3.85. The pH distribution
of these samples is shown in Figure 2.
An unusual sample was collected on April 4, 1974. On that day a number
of tornados did severe damage in Kentucky, Southern Illinois, Indiana
and Ohio. At DePaul that day, 2.62 cm of rain fell. The data for this
rain are shown in Table 7. Sample number 244 collected the entire rain
15
-------
20-
NUMBER of SAMPLES
) 0
n_
u
•-
3
—
i
a
^
n n rn
b I 1
4567
pH
pH corresponding to the precipitation weighted
average hydrogen ion concentration, 3.85.
bpH of pure water in equilibrium with atmospheric
carbon dioxide, 5.70.
Figure 2. Distribution of pH in precipitation samples in
Chicago.
Sample
Number
244
244Aa
247
244-247b
Table 7.
Hours
Exposed
4
3
(1)
RAINFALL ON APRIL 4, 1974
DRP TRP Total
0.023 0.090 0.117
0.046
0.014 0.024 0.044
(0.032) (0.156) (0.190)
£H
6.0
3.85
4.5
aSample 244 adjusted to pH 3.85. See Text.
"Values in parenthesis are calculated.
16
-------
event while sample number 247 collected the last 1.31 cm. That sample
247 collected exactly one-half of the rain was fortuitous.
That this was not a typical rain event was evident when sample number
244 was brought in. The bottom of the white polyethylene container
was covered with a very noticable layer of reddish-brown particulate
matter. Evidently the high winds associated with the storms that day
were able to transport a large amount of particulate matter. Such
"red rains" are infrequent but not unusual occurrences (Winchell and
Miller, 1924; 1922; Robinson, 1936; Weibel et al., 1966) and cover
hundreds of thousands of square kilometers. The concentration of
materials in the first half of the storm, 244-247, was calculated by
substracting the amounts of materials in sample 247 from the amount in
sample 244 and determining the concentration for the remainder of the
material. Filterable particulate matter in sample 244 was 75 mg/1. The
concentrations of and the inputs from this sample are included in the
data in Tables 1 to 3.
The substantial effect that each of these "red rain" events can have is
illustrated by this particular storm. Of the annual precipitation
inputs in the Chicago area calculated from the data in Tables 1 to 3,
this one precipitation event contributed 13 percent of the total
phosphorus, six percent of the dissolved reactive phosphates and 16
percent of the total reactive phosphates. In addition, 2 x 1(P kg/km^
of particulate matter was deposited.
Of more interest though are the pH and the types of phosphates in these
samples. The pH of sample 244 was 6.0, quite high for precipitation
samples in this area, while sample 247 had a more normal pH of 4.5.
The high pH of sample 244 is attributed to the buffering by and
ion-exchange with the particulate matter in the sample. While only 19.8
percent of the total phosphorus in sample 244 is dissolved reactive
phosphates, 77 percent of it is total reactive phosphates. Evidently,
the contact with 0.4 N sulfuric acid during the analysis, about 15
minutes, increases the amount of dissolved reactive phosphates by a
factor of four! To check this effect of acid further, the following
experiment was done. An aliquot of sample 244 was taken in duplicate,
acidified with sulfuric acid to pH 3.8 and let stand for 30 minutes.
When this treated sample, number 244A, was filtered and analyzed, the
dissolved reactive phosphates concentration had doubled. Thus contact
with a pH typical of rainfall for a time period typical of the lifetime
of a rain droplet greatly affected the dissolved phosphates concentration
in this sample.
If it is assumed that no appreciable amount of the particulate matter
in sample 244 dissolved, then knowing the amount of particulate matter
and of tKe total phosphorus present in the sample, the percent phosphorus
in the participate matter can be calculated. The results of this cal-
culation and the results of determinations performed many years ago
are shown on Table 8. Whether the present number is different from the
early ones is unclear but it may bo Higher due to the heavier use of
phosphorus containing fertilizers in recent yoar-s.
17
-------
Table 8. PERCENT TOTAL PHOSPHORUS IN PARTICULATE MATTER FROM "RED RAINS"
Total
Date Phosphate
4 April 1974 0.15%
13 February 1923a 0.07%
19 March 1920b 0.06%
aWinchell and Miller, 1924.
bWinchell and Miller, 1922.
To check on the possible interference of arsenate, Aso|~, in the phos-
phorus analysis, the concentration of arsenate was determined in 24
samples. Seventeen of these were from DePaul and seven were from other
locations around the Lake. The average concentration found (3.75 ±3.6 ppb
As)corresponds to a phosphorus concentration of 1.55 ppb with a standard
deviation of 1.49 ppb. The intensity of the color of the complex with
the arsenate developes more slowly than that from the phosphate, there-
fore, only about two-thirds of the arsenate present would appear as
phosphate. Since the arsenate correction was low, it was less than
other variables in these determinations, and because other reported
values for phosphorus are not corrected, the phosphorus values reported
here are also not corrected.
Five precipitation samples were collected in conjunction with high
volume air particulate samples. Each was analyzed for total phosphorus
and dissolved reactive phosphates and the concentrations of these
determined in the rain and in the air. These concentrations and the
washout ratio calculated from them are shown in Table 9. The ratio
for the total phosphorus and the dissolved reactive phosphates may
be different, indicating that, at least partially, these have different
sources. The washout ratio indicates an average particle diameter of
about three microns (Gatz, 1973).
18
-------
Table 9. WASHOUT RATIO
Phosphorus Concentration
Sample
Number
142
220
297
298
299
average
Precipitation
(ug/1.)
DRP
7.5
15
11
12
19
Total
12
25
29
56
43
Atmosphere
(ug/m3)
DRP
0.037b
0.022
0.030
0.050
0.062
Total
0.066b
0.067
0.055
0.110
0.102
Washout
Ratioa
DRP
261
879
473
310
395
464
Total
234
1040
680
656
543
631
a= (ug/kg in precipitation)(ug/kg in atmosphere)
= 1.29 x (ug/l.)/(ug/m3)
bGeometric mean of 25 samples from City of Chicago network.
Four glacial samples were available for analysis. Three were of sections
of cores melted at the time of collection and stored as water in poly-
ethylene containers. The last sample, number 81, was stored as a frozen
core until a portion was taken and thawed into a clean glass container
for this project. Upon reaction with the combined reagent for the
phosphorus analysis, two of the samples had an absorption at 750 nm
which interfered with the analytical absorption peak at 880 nm. The
results obtained from the other two glacial samples are shown in
Table 10.
Table 10. PHOSPHORUS IN GLACIAL SAMPLES
Year of
Deposition
1650
1966-7
DRP
(mg/1.)
0.0015
0.0017
Total
Phosphorus
(ntt/1.)
0.0024
To check on the precision and accuracy of the analyses, two things were
done. First, the results from the analysis of the 0.010 mg/1. phosphorus
standard on 100 different occasions showed an average concentration of
0.00976 mg/1. phosphorus with a standard deviation of 0.00096 mg/1.
Secondly, a group of samples containing three different levels of phos-
phorus were analyzed as part of a laboratory intercomparison study by
19
-------
the Upper Lakes Reference Group. All of our results were well within one
standard deviation of the group averages.
On several occasions, multiple samplers were exposed to check the
variability in samples collected at one site from the same rain event.
For six different events and a total of 17 different collectors, the
average variation from the mean was 16% for total phosphorus, 11% for
dissolved reactive phosphates and 6.5% for total reactive phosphates.
For the dissolved reacti/e and total reactive phosphates samples, much
of the variability noted is probably due to variability in sub sampling
and in the analyses. All of the percentages are well below that found
for the variability between individual precipitation events.
Since it was not expected that all precipitation events would be sampled,
the samples came in with some degree of irregularity from all of the sites.
The total number of samples obtained was: 191 from DePaul; 41 from Beaver
Island; 52 from Silver Lake State Park; 21 from Illinois Beach State Park;
11 from Rock Island State Park and 120 from the Harrison Crib. For the
reasons to be discussed, not all of these samples were able to be used for
the final calculations.
20
-------
SECTION VI
DISCUSSION
The results shown in Tables 1, 2 and 3 and in Figure 1 are based on the
samples collected in this project. However, not all of the samples
collected were analyzed nor were the results of all of the samples
analyzed used in the calculations. Samples, or the results from samples,
were rejected for the following reasons:
1) The collection container was exposed for more than 24 hours
2) The sample contained insects, leaves or a large amount of
particulate matter.
3) The total phosphorus concentration of a sample from an event
of more than one centimeter of precipitation was higher than
0.100 mg/1. and there was no reason to suspect that the sample
was not contaminated in some way.
The largest number of samples and the most conplete set of data were
obtained in Chicago at the DePaul site. It was anticipated that the
greater concentration of particulate matter in the atmosphere in the
urban area would lead to an increase in the phosphorus concentration of
the samples collected in Chicago (Murphy 1974). Such an effect was
found, but it is not as large as expected. One reason for the effect
being small may be that the collecter was more than fifteen meters above
the ground and thus above much of the traffic related particulate matter.
The site at the Harrison crib in Chicago was chosen to try and correct
for some of the effects of the urban area. The results obtained were
very different from those expected and can only be partially explained.
On the crib, the sample containers were almost without exception, changed
at 8 a.m. Thus these are all 24 hour samples. It was surprising how
much particulate matter was in many of the samples. It seemed to be much
more than that obtained in 24 hour samples from any of the other
locations. In fact, more than 80 percent of the crib samples or the
results from these samples, were rejected because of the large amount of
particulate matter or insects in the samples, or because the analytical
results were very high. For example, 14 of 40 samples (35%) from the
crib analyzed for total phosphorus had concentrations higher than 0.100
mg/1. while only four of more than 200 of the samples collected at DePaul
(2%) had such high concentrations. About half of the high crib samples,
and only one of those collected at DePaul, occurred with precipitation
events of more than one centimeter of rain. It is not clear why the
crib samples contained so much particulate matter.
A problem of unforeseen magnitude on our part was that of insects. It
turns out that the crib, about 4 kilometers from shore is a mecca for
insects, many of which end up in the collecters. The men on the crib
confirmed that insects were a great nuisance during warm weather.
21
-------
The variability of the concentration of phosphorus between different
rainfall events is high. Thus one must obtain a number of samples in
order to get a good estimate of what the average concentration might be.
This is probably the reason the average value of the total phosphorus
concentration for the Rock Island samples (0.036 mg/1.) is higher than
expected. It is based on only four samples. It can be seen in Tables
1, 2 and 3 that the standard deviation of the arithmetic average
concentration is almost as high as the average itself in most cases.
Another variable which must be kept in mind is the fact that the concen-
tration of phosphorus generally decreases as the amount of precipitation
in an event increases. An illustration of this variability in concen-
tration and thus in phosphorus input, is shown in Figure 3. This figure
is a plot of the inputs of phosphorus from each of the samples of
total phosphorus and DRP used in the calculations shown in Tables 1 and
2, versus the amount of precipitation each sample represented. Also
shown in Figure 3 are iso-concentration lines which show, for any amount
of precipitation, the concentration of phosphorus which would give the
indicated input. It can be seen clearly that smaller amounts of
precipitation have higher concentrations of phosphorus. This means
that the small precipitation events contribute a higher proportion of
the phosphorus than their amount of precipitation would indicate. This
is why a precipitation weighted average concentration is lower than an
arithmetic average concentration and why one must try to obtain a
representative sample of rainfall events of different amounts of
precipitation and from different types of storms.
It is also generally true that the early precipitation in an event will
be more concentrated than the later precipitation. The reasons for this
are varied and have been discussed by Gatz and Dingle (1971). Thus it
is important to have the collector out and open at the beginning of each
event.
One of the important goals of this project was to contribute information
on the species of the phosphorus present in the precipitation and there-
fore on how significant these inputs might be to the Lake. The results
in Table 4 show that 43% of the phosphorus in precipitation is dissolved
reactive phosphates. This along with the information in Table 5 that
18% of the phosphorus presently going into Lake Michigan is coming from
the precipitation in the atmosphere, indicates that these inputs are an
important part of the phosphorus budget of the Lake. The magnitude of
these inputs and the fact that they are an uncontrollable source will
have to be taken into account when the present phosphorus control
programs are evaluated and when new ones are considered. It should also
be remembered that dry deposition is also contributing phosphorus to the
Lake and that the phosphorus inputs from tributary streams contain a
lower percentage of DRP.
The observation of the difference between the dissolved reactive phosphates
and the total reactive phosphates determination, and of the results of
the pH changes on the concentration of DRP in sample number 244 indicate
22
-------
x= Total P
o = DRP
°
cm of Precipitation
Figure 3. Input of phosphorus per precipitation event.
-------
that the amount of DRP determined is a function of the pH of the sample
at the time of filtration. Therefore, in order to obtain the most
meaningful results, the sample should be filtered at a standard pH value.
Whether this should be a pH of 8.0, the approximate acidity of the Lake,
a pH of 4.0, the approximate acidity of rainfall or some other value is
not clear. It is not known what happens to the DRP in pH 4.0 rainfall
when it enters the Lake. That is, is it taken up by participate matter
or by organisms.
The ready release of some additional DRP by the acidic rainfall indi-
cated that some of the adsorbed phosphorus is readily exchangeable. How
much of this ultimately becomes available in the Lake? Ideally, a
bioassay on precipitation samples would give this information. The
difficulty with doing this, however, is the variability in the results
when bioassays are performed on samples with low phosphorus concentrations,
Attempts were made with a number of samples to use the anion exchange
method of Cowen (1974), but we were unable to obtain consistent, repro-
ducable results. It was also not obvious to us that the ion-exchange
resin would successfully compete with the organisms present for all of
the ortho phosphate.
The next best method is to try to find a chemical determination which
would give a measure of the amount of phosphorus in the sample that
ultimately becomes available. Walton and Lee (1972), report that the
total reactive phosphates determination gives a good measure of the
amount of phosphate which ultimately becomes available in Lakes Mendota
and Wingra. A more conservative estimate would be the DRP at rainfall
acidity. The most conservative estimate would be the DRP determined
on a sample filtered at Lake water acidity while the maximum would be
the DRP plus the hydrolyzable and organic phosphorus.
The data in Table 4 permit estimates using the different assumptions,
with the exception of the dissolved reactive phosphates at Lake pH, to
be compared. It is felt that the dissolved reactive phosphates plus
some additional phosphorus becomes available to organisms in the Lake.
Therefore, with all of the variables in precipitation sampling, it is
felt that an estimate of 50% of the total phosphorus may be the best
estimate of the available phosphorus in precipitation.
24
-------
SECTION VII
REFERENCES
Adamkus, V. 1975. Private Communication.
Cowen, W. F. 1974. Available Phosphorus in Urban Runoff and Lake
Ontario Tributary Waters. Ph. D. Thesis. Univ. of Wisconsin, Madison.
Cunningham, P. T., S. A. Johnson and R. T. Yang 1974. Variations in
Chemistry of Airborne Particulate Matter with Particulate Size and Time.
Envir. Sci. and Tech. 8.: 131, (1974).
Gatz, D. F. 1973. Scavenging Ratio Measurements. In: Summary Report
of Metromex Studies, 1971-72, F. A. Huff, Ed. Report of Investigations
74, Illinois Water Survey, Urbana, 111. p. 133, (1973).
Gatz, D. F. and A. N. Dingle 1971. Trace Substances in Rain Water:
Concentrations Variations During Convective Rains, and their
Interpretation. Tellus 23:14, (1971).
Goings, J. E. and S. J. Eisenreich 1974. Extraction of Reduced Moly-
bdophosphoric and Molybdoantimonyphosphoric Acids with Oxygenated
Solvents. Anal. Chim. Acta, 71:393, (1974).
Isaeva, A. B. 1969. Determination of Small Amounts of Phosphates in Sea
Water after their Preliminary Extraction as Phosphomolybdic acid, and
Determination of Phosphates in Turbid Waters. Zhurnal Analiticheskoi
Khim. 2£:1854, (1969).
Johnson, D. L. 1972. Simultaneous Determination of Arsenate and Phos-
phate in Natural Waters. Envir. Sci. and Tech., .5:411, (1972).
Jones, D. M. A. and D. D. Meredith 1972. Great Lakes Hydrology by
Months. 1946-65. Proc. 15th Conf. of Great Lakes Res. p. 477, (1972).
Lee, G. F. 1974. Phosphorus, Water Quality and Lake Michigan. In:
Proceedings of the 4th Lake Michigan Enforcement Conference. Sept. 1972,
Chicago, 111. p.257, (1974).
Lee, R. E. and R. K. Patterson 1969. Size Determination of Atmospheric
Phosphate, Nitrate, Chloride and Ammonium Particulate in Several Areas.
Atmos. Envir., 3_:249, (1969).
Miller, W. E., T. E. Maloney, and J. C. Greene 1974. Algal Productivity
in 49 lake waters as determined by Algal Assays. Water Res. 8^:667, (1974)
Murphy, T. J. 1974. Sources of Phosphorus Inputs from the Atmosphere
and their Significance to Oligotrophic Lakes. Water Resouces Center,
Urbana, 111. Research Report No. 92 (1974).
25
-------
Pakalns, P. and B. R. McAllister 1972. Determination of Phosphate in
Seawater by an Isobutyl-Acetate-Extraction Procedure. J. Marine Res
30_:305, (1972).
(^36)'. ^ ^ Sn°Wfa11 ** ^ Hampshire and Vermont.
Schelske, C. L. and E. F. Stoermer 1972. Phosphorus, Silica and
Eutrophication of Lake Michigan. In: The Limiting Nutrient Controversy,
i». E. Likens, Ed. Allen Press, Lawrence, Kansas 66044, p. 157, (1972).
Schelske, C. L. , E. D. Rothman, E. F. Stoermer and M. A. Santiago 1974
Responses of Phosphorus Limited Lake Michigan Phytoplankton to Factorial
Enrichments with Nitrogen and Phosphorus. Limn, and Ocean., 19:409,
(.1974) . —
Stoermer, E. F. 1974. In: Proceedings of the 4th Lake Michigan
Enforcement Conference, Sept. 1972, Chicago, 111. p. 217, (1974).
U. S. Environmental Protection Agency. Report of the Phosphorus Techni-
cal Committee. In: Proceedings of the 4th Lake Michigan Enforcement
Conference, Sept. 1972, Chicago, 111. p. 209 (1974).
U. S. Environmental Protection Agency. Methods for Chemical Analysis
of Water and Wastes. National Environmental Research Center, Cincinnati
Ohio, EPA-16020-07/71. 298 p. (1971).
Vaughn, J. C. 1974. Progress Report on Water Quality of Lake Michigan
near Chicago. In: Proceedings of the 4th Lake Michigan Enforcement
Conference, Sept. 1972, Chicago, 111. p. 88, (1974).
Walton, C. P. and G. F. Lee 1972. A Biological Evaluation of the
Molybdenum Blue Method for Orthophosphate Analysis. Verh. Inter. Verein
Limn. 18^:676, (1972) .
Weibel, C. P. and R. B. Weidres, J. M. Cohen and A. G. Christianson
1966. Pesticides and Other Contaminants in Rainfall and Runoff.
J. A. W. W. A., 1075, (1966).
Winchell, A. N. and E. R. Miller 1924. The Dustfall of February 13,
1923. J. of Agricultural Research, 29.:443, (1924).
Winchell, A. N. and E. R. Miller 1922. The Dustfall of March 9, 1918.
Amer. J. Science (Ser. 4, Vol. 46) 196:599, (1922).
26
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
ni POT i NO.
_J2PA-600/3-75-Q05__
"lITLt AND SUBTITLE
INPUTS OF PHOSPHORUS FROM PRECIPITATION
TO LAKE MICHIGAN
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
December 1975 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
AUTHOR(S)
T. J. Murphy and P. V. Doskey
8. PERFORMING ORG>
"PERFORMING ORG ^NIZATION NAME AND ADDRESS
Chemistry Department
DePaul University
1036 West Belden Avenue
-111;
10. PROGRAM ELEMENT NO.
1BA026: RQAP 25ADR: Task 10!
11. CONTRACT/GRANT NO.
Grant R-8026^7
GENCY NAME AND ADDRESS
2. SPONSORING AC
Environmental Research Laboratory Duluth
Office of Research and Development
U.S. Environmental Protection Agency
Duluth, Minnesota 5580^
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
EPA-ORD
5. SUPPLEMENTARY NOTES
6 ABSTRACTPrecipitation samples were collected at six locations around Lake Michigan
and analyzed for the different forms of phosphorus present. It was found that the
atmosphere is presently contributing one million kilograms per year of phosphorus or
about 18 percent of the phosphorus budget of the Lake. As the phosphorus removal
program on sewage effluents becomes fully implemented in the Lake Michigan basin, the
contribution to the Lake of phosphorus from particulate matter scavenged by precipi-
tation could increase to about 30 percent of the total.
The phosphorus concentration of precipitation was found to be higher at the southern
end of the Lake. More than 40 percent of the phosphorus in precipitation is dissolve
reactive phosphates and the amount of the dissolved phosphates in precipitation was
found to be somewhat dependent on the pH of the sample.
The washout ratio for phosphorus by precipitation was determined. It was also found
from the analysis of some glacial samples that phosphorus has long been a conponent
of precipitation.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Phosphorus
Precipitation (chemistry)
b.lDENTIFIERS/OPEN ENDED TERMS
Eutrophication
Acid rainfall
Lake Michigan
COS AT I Field/Group
08/H
13. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
21. NO. OF PAGES
35
20. SECURITY CLASS (Thispage/
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
27
7:? U.S. GOVERNMENT PRINTING OFFICE: 1976- 657-695/5369
------- |