EPA-660/2-74-005
February 1974
Environmental Protection Technology Series
Quantification of Pollutants
In Agricultural Runoff
Office of Research and Development
U.S. Environmental Protection Agency
Washington. D.C. 20460
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and
Monitoring, Environmental Protection Agency, have
been grouped into five series. These five bread
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 ares
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 ENVIRONMENTAL
PROTECTION TECHNOLOGY series. This series
describes research performed to develop and
demonstrate instrumentation, equipment and
methodology to repair or prevent environmental
degradation from point and non-point sources of
pollution* This work provides the new or improved
technology required for the control and treatment
of pollution sources to meet environmental quality
standards.
EPA REVIEW NOTICE
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.
-------
EPA- 660/2- 741-005
February 1974
QUANTIFICATION OF POLLUTANTS
IN
AGRICULTURAL RUNOFF
by
James N. Dornbush
John R. Andersen
Lei and L. Harms
Contract No. 68-01-0030
Project No. R-800400
Program Element 1BB039
Project Officer
Ronald R. Ritter
Region VII
U. S. Environmental Protection Agency
Kansas City, Missouri 64108
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D. C. 20460
For sale by the Superintendent of Documents, U.S. Government Printing Office
Washington, D.C. 20102 - Price $1.90
-------
ABSTRACT
Surface runoff from snowmelt and rainfall in eastern South Dakota vas
measured during a three year period. The size of the research sites
ranged from 7.18 to 18.69 acres, and all sites had crops of corn, oats,
pasture or hayland. Composite samples of the runoff vere used for
various chemical, physical and "biological determinations.
Runoff samples from 108 snowmelt events and 36 rainfall events vere
collected. Equipment fabrication and installation resulted in some
incomplete data for the initial year, but successful monitoring of
each runoff event vas accomplished thereafter.
Sediment losses vere considerably lover than anticipated. Pesticide
concentrations vere lov in both vater and sediment samples, and vere
usually less than the analytical test limits. Coliform and fecal levels
vere consistently greater than accepted surface vater quality criteria.
Most of the nutrients vere found to be soluble and/or associated with
snovmelt runoff.
This report vas submitted in fulfillment of Project Number R-800UOO,
Contract Kumber 68-01-0030, by the Civil Engineering Department,
South Dakota State University, Brookings, South Dakota under the
sponsorship of the Environmental Protection Agency. Work vas com-
pleted as of July, 1973.
ii
-------
CONTENTS
Page
Abstract ii
List of Figures iv
List of Tables vi
Acknowledgments viii
Sections
I Conclusions 1
II Recommendations 3
III Introduction U
IV Review of Literature 6
V Research Sites 22
VI Field Methods 39
VII Laboratory Methods ^7
VIII Data and Results 55
IX References 90
X Appendices 96
iii
-------
FIGURES
Ho« Page
1 General Site Locations 21*
2 Site 1 26
3 Site 2 28
^ . Site 3 30
5 Site U 32
6 Site 7 31*
7 Site 8 36
8 Site 9 38
9 Front Viev of Field Installation Uo
10 Side View of Field Installation HO
11 Interior View of Sampler W»
12 View of Solenoid Controlled Clamping System for the
Sampler fcU
13 Type F Water Level Recorder With Modifications to
Activate Automatic Sampler *»6
lU View of Level Recorder Showing Mounting of Plastic
Strip !»6
15 Flow Diagram For Treatment and Analysis of Samples
During Phase I W
16 Flow Diagram For Snowmelt Determinations During
Phase II 50
17 &Tinii»i Runoff Patterns for Phase II 58
iv
-------
FIGURES (continued)
No« Page
18 Distribution of Rainfall Events 60
19 Hydrograpn Comparison of a Low Intensity Storm and
a High Intensity Storm 62
20 Total Coliforms in Snowmelt Runoff 73
21 Fecal Coliforms in Snowmelt Runoff 75
22 Fecal Streptococcus in Snowmelt Runoff 76
23 Bacteriological Indicators in Rainfall Runoff
From Cultivated Fields 77
2U Breakdown of Total Runoff Contributions For
1971 and 1972 87
-------
TABLES
Ho. Page
1 General Quality of Surface Runoff from Nonpoint
Discharges 7
2 Nutrients in Rural Runoff 11
3 Summary of Analysis of Variance of Freezing Effects
on Analytical Determinations of Snowmelt Samples 51
U Autoanalyzer Methodology 53
5 Precipitation Summary at Research Sites From
Mid-March to Mid-November 56
6 Frequency of Runoff from Rainfall 56
7 Biochemical Oxygen Demand and Related Factors
for Agricultural Runoff 6U
8 Concentrations of Characteristics of Snowmelt
Runoff for Phase I • 66
9 Recommended Limits of Bacteriological Indicators
in Surface Waters 71
10 Bacteriological Data from Rainfall Runoff on
Uncultivated Fields 78
11 Number of Samples Grouped in Ranges of Pesticide
Concentrations for Filtered Runoff Samples 80
12 Number of Samples Grouped in Ranges of Pesticide
Concentrations for Sediment Samples 82
13 Mean Concentrations of Runoff Parameters by Landlse 83
lU Yearly Runoff Contributions of Runoff Parameters
by Land Use 86
vi
-------
TABLES (Continued)
Ho. Page
15 Soluble Fraction of COD, Phosphorus and TKN of
Snowmelt and Rainfall Runoff from Research Sites 89
vii
-------
ACKNOWLEDGMENTS
Many people added their contributions to this study and their help
vas sincerely appreciated. Mr. T. Alvin Biggar, Mr. Warner Mostad,
and Mr. Rick Benson assisted in the development of the automatic
sampler, among other services. Dr. Yvonne Greichus and her staff
from Station Biochemistry were responsible for the pesticide analyses
on the vater and sediment samples. Dr. Paul Middaugh and his lab-
oratory personnel from the Bacteriology Department cheerfully ran
the many bacteriological determinations required. Several part-
time workers conscientiously labored sometimes long and unusual
hours during sample collection and data acquisition, particularly
Mr. Roger Patocka, Mr, John Wellner, and Mrs. Sandy Kuchta.
Implementation of this project would not have been possible without
the cooperation of the individuals who owned and operated the land
where the research sites were located. Mr. Marvin Lamb was the land-
owner for Sites No. 1, 2, 3, and li; Mr. Dean Duff owned the land for
Site No. T» and Mr. M. G. Olson farmed the drainage areas for Sites
No. 8 and 9. Mrs. Kayt Daum owned the land which contains Sites
No. 8 and 9.
viii
-------
SECTION I
CONCLUSIONS
1. Considerable quantities of nutrients were present in the agri-
cultural runoff, and these have important implications regarding
lake eutrophication. Nutrient losses ranged from 0.03 to 3.0
Ib/acre/yr of nitrogen and from 0.01 to 0.72 l"b/acre/yr of phos-
phorus. The form of the nutrients is also of consequence. Most
of the annual nutrient load comes from snovmelt runoff, and a
large percentage is in a soluble form. In areas comparable to
those of this study, soil conservation practices aimed primarily
at reducing soil losses without substantially reducing surface
runoff, particularly from melting snow, will probably not "be
effective for retarding lake eutrophication from the nutrients
in agricultural runoff.
2. Annual soil losses were much lower than anticipated. Soil losses
ranged from less than 10 Ib/acre/yr to a maximum of less than
1000 Ib/acre/yr. While runoff due to rainfall accounted for
nearly all (93.750 of the sediment losses, it accounted for only
about one-half of the soluble nutrient species.
3. Most of the sediment in runoff waters was from cultivated fields
with relatively small amounts originating on areas in permanent
grass. The bulk of the soil losses, as well as losses of those
constituents associated with the soil, occurred during short
duration, high intensity rain storms. Almost all the soil lost
during the 2 year study can be attributed to approximately the
1/3 of the runoff which was caused by rainfall. The instantan-
eous concentration of soil in the rainfall was directly related
to the flow rate.
U. Total coliform and fecal coliform densities in runoff frequently
exceed those limits specified in some surface water quality
standards. However, runoff waters from agricultural lands are
probably not a potential health hazard and FC/FS ratios should
also be considered. Total coliform counts are usually greater
than fecal streptococcus levels. The highest total coliform and
fecal streptococcus counts for snowmelt runoff were attributed
to cultivated fields.
-------
5. Pesticide concentrations in runoff vaters from rainfall and snow-
melt vere very low with most samples having concentrations "below
the analytical test limits of 0.05 ppb for aldrin, DDE, dLeldrin,
lindane, heptachlor, heptachlor epoxide, and endrin and 0.10 ppb
for DDT, DDD, atrazine and methoxychlor. Most sediment samples
also had pesticide concentrations below the analytical test limits.
6. Each site consistently had more snowmelt runoff events than rain-
fall runoff events. Fields with permanent grass cover had runoff
quantities comparable to those of cultivated fields although al-
most all of the runoff occurs from snowmelt for fields in permanent
grass.
7. Fall plowing reduces the amount of snow retained on the fields
over the winter months. Although runoff was reduced, this
practice appeared to increase the erosion potential from wind.
8. Rainfall runoff was an infrequent and unpredictable event and
can be expected to occur only a few times each year, generally
during spring and early summer when crop cover is light and
rainstorms are more frequent.
9* When comparing individual parameter concentrations from rainfall
and snowmelt runoff, the runoff from cultivated areas showed
wider variations than runoff from uncultivated areas.
-------
SECTION II
RECOMMENDATIONS
This investigation has demonstrated that surface runoff from cultivated
fields, permanent grassland and pasture in the upper great plains area
is frequently of a quality that vould exceed standards of many receiving
streams and lakes. For some constituents such as phosphorus, concen-
trations exceeded recommended standards for treated vastevater effluents,
Recognizing that the quality of agricultural surface runoff is highly
variable and dependent on many factors indigenous to the watershed area,
the following recommendations are offered.
1. Additional monitoring of surface runoff from large agricultural
areas should be performed throughout the nation. Estimates of
some pollutant losses from small plot studies do not appear to
be representative of larger drainage areas.
2. Evaluation of routine stream monitoring data should emphasize
periods of predominately surface runoff as contrasted vith
periods of appreciable groundvater inflow to establish pollutant
contribution from non-point sources. Review of basic hydrologic
information for any region should reveal the periods when surface
runoff is most frequent.
3. In predominantly agricultural regions, consideration should be
given to establishing treated effluent quality limits at con-
centrations equal to that of surface runoff from the indigenous
area depending upon the quantity and frequency of runoff events.
This consideration would apply particularly to concentration
limits on plant nutrients for control of undesirable algal growths.
-------
SECTION III
INTRODUCTIOH
GENERAL
The problems of pollution from agricultural practices are indeed com-
plex but those of defining pollution from agriculture may be even more
involved. Agricultural runoff from large areas collects to create
the rivers or lakes, the pollution of which is the cause of concern.
Consequently, agricultural runoff serves as the baseline of available
vater quality whether it originates from forest, pastures, or crop-
lands. Yet, very little information is presently available concerning
the concentrations of pollutants in this baseline, agricultural runoff.
Information regarding the quality of agricultural runoff has been rel-
atively limited in comparison to that relating to most other industrial
sources of vater pollution. Much of the research with agricultural
runoff has been conducted with small test plots which simulated agri-
cultural lands and often related to only one particular water-carried
pollutant such as sediment. Information concerning runoff quality in
the upper great plains region was almost entirely lacking, and re-
search to more fully determine the characteristics of agricultural
runoff was needed. This research project was initiated to meet that
need.
OBJECTIVE
The general objective of this project was to quantify the pollutional
constituents of agricultural surface runoff. It was desired to deter-
mine contributions of constituents in terms of quantity per unit acre
per unit time for various land uses.
It was also considered important to know the reduction in waste dis-
charges that may be obtained by sedimentation. Determination of the
fraction of pollutional constituents associated with the suspended
solids or sediment load carried with the surface runoff would be an
objective aimed at relating the potential of soil conservation
practices to the improvement of the runoff quality.
In as much as possible, it was also an objective of this study to re-
late the quality of surface runoff to numerous factors peculiar to
-------
specific drainage "basins. This vould allow extrapolation of the
findings to drainage "basins in other areas having different climatic
conditions, other land uses, and various physical factors.
SCOPE OF THIS RESEARCH
This project pertained to surface runoff from cultivated fields,
permanent grass and alfalfa lands, and grazed pasture. The runoff
from confined feeding operations vas not evaluated. Ail field con-
ditions were natural and uncontrolled. Simulated rainfall was not
used, and an attempt was not made to influence the land operators
with respect to land management practices.
Sampling was executed any time that runoff occurred* Manual sampling
and flow measuring were employed for spring runoff, vhile automatic
sampling and flow measuring equipment were used during rainfall events.
Samples were brought into the laboratory and a composite sample was
made for each site. All the bacteriological, physical and chemical
determinations were then conducted on these composite samples.
-------
SECTION IV
REVIEW OF LITERATURE
GENERAL QUALITY
Estimates of amual losses of sediment and nutrients from rural areas
tend to stagger one's imagination. Approximately U billion tons of
sediment are washed into United States streams each year (l). ELiassen
and Tehobanoglous estimated nitrogen and phosphorus contributions from
rural agricultural land at 1,500 to 15,000 and 120 to 1,200 million
pounds per year, respectively. Annual contributions from rural non-
agricultural land vere estimated as HOO to 1900 million pounds of
nitrogen and 150 to 750 million pounds of phosphorus (2).
The majority of investigators vho have studied agricultural runoff
have conducted stream surveys and the data are generally limited to
nutrients contributions. However, a few researchers have compiled gen-
eral quality data from agricultural surface runoff (3) (b) (5), and
these data are presented in Table l along with data from an urban
area for comparison (6).
A 173 acre cultivated site in eastern South Dakota was sampled by
Benson (3) and McCarl (1») during 1969 and 1970. Samples were pre-
served by freezing without presenting data to verify the method used.
The quality of both snowmelt and rainfall runoff was recorded. A
summary of their data is shown in Table 1. Total annual contributions
were computed by McCarl for the 1970 runoff season (b). Annual loads
of 2,0^0 Ib/acre of suspended solids, 6.9 Ib/acre of BOD, 2U6 Ib/acre
of COD, 8.U Ib/acre total nitrogen and 0.07 Ib/acre of total phosphorus
were reported.
Weibel «£*!• (6) reported on the general quality of runoff water from
a 27 acre residential area in urban Cincinnati. About 730 Ib/acre/yr
of SS were measured of which 22% were volatile. The BOD and COD were
33 Ib/acre/yr and 2Uo Ib/acre/yr, respectively.
Weidner et, aJU (5) obtained runoff quality data from two 1.5 acre
watersheds near Coshocton, Ohio and also from a 5-acre apple orchard
at Ripley, Ohio. Most of the material losses occurred during the
summer months as a result of high intensity rainstorms. The organic
losses, as measured by BOD and COD, were small. BOD values ranged
-------
TABLE 1. - General Quality of Surface Runoff from Nonpoint Discharges
Item Benson (3)
Land use Farmland
BOD, mg/1 5-30
BOD, Ib/acre/yr -
COD, mg/1 50 - 360
COD, Ib/acre/yr
Solids, mg/1 90 - 5000 SS
Solids, Ib/acre/yr
Total phosphorus, 0.26 - 2.1*
mg/1 Soluble P
Total phosphorus,
Ib/acre/yr
Org. N + NH3,mg/l 1.3 - 20.3
Total nitrogen
mg/1
Total nitrogen, -
McCarl (1*)
Farmland
3-15
6.9
70 - 780
21*6
180 - 6000 ss
2,01*0 SS
o.oU - 0.60
0.07
2.8 - 17
12.9 - 33.2
8.U
Weibel, et, al . (6)
Residential
2-81*
33
20 - 610
21*0
5 - 1200 SS
730 SS
0 - 1.1*
Soluble P
0.8
Soluble P
0.1 - 7
0.2 - 9
8.9
Weidner, et^ al . (5)
Research plots and
apple orchard
3 -
3.7 -
1*0 -
27.8 -
500 -
185 -
0.1*2 -
0.36 -
-
U.9 -
0.8 -
8.U
120
68
1300
575 TS
13,200 TS
0.98
9.0
9.0
237
Ib/acre/yr
-------
from 3«7 to 120 Ib/acre/yr and COD losses ranged from 27.8 to 1300
Ib/acre/yr.
SOIL LOSSES
Soil erosion by vater is a common geological phenomenon. It accounts
for many of the variations in topography of the land about us, Kormal
processes of erosion work slowly and continuously. Accelerated
erosion, which is more rapid than normal erosion, has become a cause
for concern in recent decades.
Walker and Wadleigh discussed vater pollution as it relates to land
runoff* They estimated that the sediment yield of the Mississippi
River Basin averages 390 ton/sq. mi/yr. Soil losses are not the only
consideration. They also estimated that each 1000 tons of sediment
carries about 1000 pounds of phosphorus fixed to its surface (7).
Nutrients lost from the land are often greater than expected because
the finer and more fertile soil particles are sorted out and vashed
away (8).
Several investigators have obtained data regarding soil losses from
various sized fields and plots. McGuinness et^al. described the eight
small natural watersheds which the Agricultural Research Service had
been evaluating near Coshocton, Ohio from 19^5 to 1956 (9)* The
watersheds were matched into four pairs and subjected to a four year
rotation of corn, wheat, and two years of meadow.
The study compared the effects of improved farming practices with
prevailing farming methods. The improved practices decreased soil
and water losses except during extremely intense storms . Average
peak rate reductions of water losses for the improved methods were
0.82 in./hr for corn, 0,22 in./hr for wheat, and O.OU in./hr for
meadow (9) . Weidner et^ al. reported long term, average soil losses
as high as, 7.7 ton/acre/yr" for these same watersheds. (lU).
Data from small plots (13.3 ft by 72.6 ft) in western Minnesota were
reported by Holt (21) and Timmons et al. (22), Two-year average soil
losses were given by Holt at 0 to 7 ton/acre/yr (10) , and average
losses of 2.2 to 21. U ton/acre were reported for 1961 to 1967
(11).
Sediment from some larger watersheds was measured by McCarl (U) and
Dragoun and Miller (12). An «n™"«T load of about 1 ton/acre was
received from a 173 acre area under cultivation (U). A U8l acre
watershed being farmed by conventional methods was compared with a
Ull acre watershed on which Improved farming practices such as
terraces and grassed waterways were used. About 15% of the unimproved
acreage was cultivated, while 60% of the improved land was culti-
vated (12). Over a four year period, the sediment yield for the un-
improved area was 8.1* ton/acre and U.5 ton/acre for the improved
field. They concluded that sediment reductions of 50Jf or more could
8
-------
be realized by implementing soil conservation measures. The reduction
was primarily due to a decrease in water yield. Runoff on the unim-
proved land averaged 3.9 in./yr while the average runoff from the
improved acreage was 2.9 in./yr (12).
The Department of Agronomy at Cornell University conducted a three
part study to evaluate the factors which control soil and nutrient move-
ment from the land. An artificial rainfall simulator was used on small
test plots for one part of the study, and the largest erosion loss was
from continuous corn. Another part used larger test plots (0.8 acres
each) and natural rainfall conditions. The use of mineral fertilizers
resulted in physical deterioration of the soil and subsequently in-
creased surface runoff and soil losses. Incorporating crop residues
into the soil was recommended, as well as proper timing of fertilizer
applications. The third part of the project was an algal nutrient
study. Results from the algal nutrient study were inconclusive (13).
In an attempt to define soil erosion and allow prediction of soil
losses, researchers studied specific aspects of the erosion process.
The severity of raindrop erosion was investigated by Ellison (lU).
In particular, he studied the effects of raindrop impact and splash.
He speculated that raindrop splash caused most of the severe sheet
erosion near hilltops. He concluded that infiltration and runoff
were dependent upon raindrop erosion. The use of vegetal canopies
and mulches to prevent raindrop erosion was proposed.
Epstein and Grant evaluated some of the factors which define the
different credibility characteristics of soils by using a rainfall
simulator on six soil types (15). Soil losses reached a maximum
during the first 10 minutes of simulated rainfall for four of the
soils, but no peaking effect was noted on the other two soils. They
also cast doubts on obtaining realistic erodibility values by using
simulated rainfall.
In the early 196ofs, researchers were attempting to formulate a soil-
loss equation. Rogers et,^. used an artificial rainmaking device
to generate runoff on small plots, 35 ft to 75 ft long (l6). The
purpose was to further define coefficients in a soil-loss equation.
Most of the variations in soil-loss data were explained by rainfall
amounts multiplied by rainfall intensity.
Finally, after more than 20 years of development and based on the work
of many contributors, a generally accepted soil-loss equation came
into being. This equation is commonly known as the "universal" soil-
loss equation. The history of its development, as well as a detailed
explanation of the equation, can be found in a handbook compiled by the
Agricultural Research Service (17).
-------
The basic soil-loss equation is:
A = RKLSCP Eqn. (l)
Where: A = Soil-loss per unit area
R = Rainfall factor
K • Soil-erodibility factor
L « Slope-length factor
S = Slope-gradient factor
C « Cropping-management factor
P ~ Erosion-control practice factor
This equation is based on the data obtained from many small plots
scattered throughout the United States and Puerto Rico. It is gener-
ally accepted by agricultural people as representing acreage soil
losses for cropland east of the Rocky Mountains (17)*
NUTRIENTS
The nutrients carried to surface vaters by agricultural runoff are
an important facet of eutrophication. Information regarding the
nutrients in agricultural or rural runoff; and in particular, that
vhich provides quantitative data on nitrogen and phosphorus in agri-
cultural runoff vas reviewed.
A major problem of nutrients is the stimulation of grovth of algae and
other aquatic plants in streams and reservoirs. A report on the
various sources of these nutrients was completed by an American Water
Works Association Task Group (18). They reported on various sources
including domestic, industrial, rural, urban, and farm animal wastes.
It was concluded that the single largest contributor of nitrogen and
phosphorus to water supplies was agricultural runoff.
Bauman and Kelman warned about requiring cities and industries to in-
stall advanced waste treatment equipment without identifying agricul-
ture's contribution to the waste load of a stream (19) (20). They
made weekly stream measurements for flow, BOD, COD, SS, turbidity,
nitrogen, and phosphorus in the Des Mbines River between Boone and
Des Moines, Iowa. Exact quantities of nutrients from various sources
could not be determined, but they were able to conclude that nitrogen
and phosphorus removal from municipal and industrial discharges would
be beneficial for flowing streams. However, removal of nitrogen and
phosphorus may not benefit impounded waters because the major sources
of nutrients for these bodies of water are nonpoint discharges.
Various investigators have evaluated the nutrient contributions from
rural areas. These data are presented below and summarized in Table 2.
Three frequently quoted works were authored by Sawyer (21), Slyvester
(22), and Englebrecht and Morgan (23).
10
-------
TABLE 2. - Nutrients in Rural Runoff
Land Use
Stream Surface
Flov or Runoff
Nitrogen,
Ib/acre/yr
Phosphorus,
Ib/acre/yr
Remarks
Reference
Forest and X
Grazing Area
Farmland
Forested
Diversified
Farming
Farmland
Research
Plots
Farmland
Research
Plots and
Orchard
0.65 N03-N
0.90 Total N
7.0 Total H
1.3-3.0 Total N
X 2.5-2^.0 Total N
0.33 Total P
O.k Total N
0.3-0.8 Total P
0.9-3.9 Total P
0-15 Ortho plus
Hydrolyzable
P205, as P
X 3.1-6.H Total N 0.03-0.21 Total
Total N
0.07 Total P
X 0.8-237 Total N 0.36-9.0 Total P
Little cultivated 33
land, Data extrap-
olated from 2
months sampling
General farming 21
conditions
Some logging and 22
road construction
Irrigation return 22
flows
Heavily cultivated 23
land. Data extra-
polated from 6
summer months
Natural conditions, 10
small plots
173 acre cultivated k
site in corn and oats
Natural conditions, 5
1.5 acre plots, 5
acre orchard
-------
TABLE 2. - (continued)
StreamorSurface Nitrogen,
Farmland X
Farmland X
Rangeland X
Farmland and
Pasture
Farmland and X
Pasture
Irrigated
Field
Research
Plots
7.35 N03-N
•J
3.8 Total H
0.56 N03-N
X 3.6 Total N
U.O Total N
3.1 Total N
X 3.9 Total N
Phosphorus ,
0.05 Ortho
and Poly POi,
0.2U Total P
0.021 Soluble P
0.067 Total P
1.1 Total P
0.2 Total P
0.07 Total P
1.2 Total P
No significant point
discharges
20% forest, Q0% crop
land and pasture
Primarily grazing, no
chemicals added
Higher than normal
runoff, frozen sample
storage
Mostly pasture, 6 mo.
data, frozen sample
storage
Subsurface drainage, no
fertilizer applied
Small control plots
with no applied manure,
29
30
26
31
28
27
32
frozen sample storage
-------
Sawyer (21) estimated the nitrogen and phosphorus present in agricul-
tural runoff by sampling streams which did not receive municipal or
industrial discharges. The higher nutrient values came from areas
which contained marsh lands. Annual contributions of 7 Ib/acre/yr of
nitrogen and O.U Ib/acre/yr of total phosphorus were reported. Critical
nutrient levels in support of algal blooms were listed at 0.30 ppm for
inorganic nitrogen and 0.01 ppm for inorganic phosphorus.
Slyvester reported on nitrogen and phosphorus loads from irrigation
surface return flows in the Washington Yakima Valley. Nitrogen values
ranged from 2.5 to 2U.O Ib/acre/yr and phosphorus varied from 0.9 to
3.9 Ib/acre/yr. Also listed were some nutrient values from forested
areas which were subject only to some logging and road construction.
Based on stream samples, total nitrogen varied from 1.3 to 3.0 lb/acre/
yr and total phosphorus ranged from 0.3 to 0.8 Ib/acre/yr (22).
Engelbrecht and Morgan estimated the amount of phosphorus related to
land drainage by deducting the phosphorus contributed from sewage
treatment plant effluents. Samples were collected from the Kaskaskia
River in Illinois and reflected phosphorus contributions from six
heavily cultivated drainage areas ranging in size from 12 to 5220
square miles. Determinations were made for orthophosphate and hydro-
lyzable phosphate as pp°5* Total phosphorus was not determined but
was estimated at 20% to 30/C more than the ortho plus hydrolyzable
P20c. Reported values of ortho plus hydrolyzable ?2®5 ranged from
0-15 Ib/acre/yr as P (23).
Nutrient losses from small plots in western Minnesota were reported
by Holt (10), Holt et, al. (2k), and Timmons et al. (ll). Based
on data collected for two years, Holt lists measured values of 3.1
and 6.k Ib/acre/yr of nitrogen and 0.03 to 0.21 Ib/acre/yr of phos-
phorus (10). Timmons e£ al. extrapolated these data based on soil-
loss information for seven years and estimated annual nutrient losses
at 31 to 183 Ib/acre of nitrogen and 0.85 to 1.1 lb/acre of phosphorus
(ll). Large quantities of soluble phosphorus were observed in snow-
melt runoff from alfalfa plots* This loss of soluble phosphorus was
verified by a laboratory investigation conducted by Timmons et al.
(25). Alfalfa and bluegrass crops contributed substantial amounts of
soluble nutrients.
Recent data can be found from several sources. McCarl measured 0.07
Ib/acre/yr of total phosphorus and 8.U Ib/acre/yr total nitrogen from
a 173 acre cultivated field in South Dakota (k). Weidner et, al. re-
ported annual losses of 0.8 to 237 lb N/acre/yr and 0.36 to 9.0 Ib
P/acre/yr from two 1.5 acre watersheds and a 5 acre apple orchard in
Ohio (5).
A 30,000 acre drainage basin in southwestern Ontario used primarily for
summer pasture was the subject of research by Campbell and Webber (26).
13
-------
Little or no chemicals or fertilizers were used on the land. As ex-
pected, low nutrient loads were measured. Annual values of 0.56
Ib/acre/yr of HOy-H, 0.021 Ib/acre/yr of soluble phosphorus, and
0.067 Ib/acre/yr of total phosphorus were reported.
An indication of the nutrients in subsurface irrigation drainage is
apparent from data presented by Johnson et^ aU (27). The drainage
beneath four irrigated fields in the San Joaquin Valley of California
was measured. One of the fields did not receive any fertilizer and
its drainage contained 3.1 lb N/acre/yr and 0.067 Ib P/acre/yr.
The Department of Biological and Agricultural Engineering at North
Carolina State University at Raleigh sampled the discharge from
Site F, an area draining 75 acres of pasture, corn, and orchard (28).
Samples were frozen prior to analysis. Data were not presented to
verify the sample preservation method used. Extrapolating six months
of data yields results of U.O Ib/acre/yr of nitrogen and 0.6 Ib/acre/yr
of phosphate.
Vang and Evans reported on the nutrient observations made in an exten-
sive study of Lake Bloomington in central Illinois (29). The average
precipitation and runoff are 36.5 and 9.05 in./yr, respectively. The
drainage area did not have any important point discharges. The total
annual runoff of nitrate-nitrogen was given as VfOO Ib/sq mile and the
annual contribution of phosphorus was 32 Ib/sq mile.
Waste discharges in the Potomac River Basin were evaluated by Jaworski
and Hetling (30). An estimated 5,8Uo sq miles were in cropland and
pasture. This area yielded 0.2U Ib/acre/yr of phosphorus and 3.8
Ib/acre/yr of nitrogen. The major source of all nutrients was from
wastewater. However, agricultural runoff contributed 65? of the
nutrients attributed to land runoff although agricultural land repre-
sented only 38>C of the total area. The remaining area was in forest
or urban land.
Personnel from the University of Wisconsin instigated surface runoff
studies from both natural watersheds and snail research plots (31)
(32). Samples were frozen and stored prior to analysis with a Tech-
nicon Autoanalyzer at a University of Wisconsin laboratory. Data
verifying the preservation method used were not presented.
Surface runoff from natural watersheds of 22.8, 52.5, and 171 acres
was sampled by Witzel et^&1 (31)* Runoff data for one year, including
some snowmelt runoff, were obtained. The smallest site was in pasture,
and the remaining sites were in pasture, hayland, and cultivated crops.
Commercial fertilizers and animal manures were used. The annual aver-
age surface runoff is about 1.75 in. of which 15% results from snowmelt
or rainfall on frozen soil. The particular year in question had about
twice the average annual runoff and gave nutrient loads of 3.6 Ib/acre
-------
of nitrogen, 1.1 Ib/acre of phosphorus, and 1.6 Ib/acre of potassium.
Average annual contributions were estimated at 2 Ib/acre/yr, 0.6
Ib/acre/yr, and U Ib/acre/yr of nitrogen, phosphorus and potassium,
respectively.
Minshall et al« used various manure applications on eight small plots
(10 ft by"TTo ft) and evaluated the effects of the applications on the
quantities of nutrients in the surface runoff (32). Data vere collected
for three years. Annual nitrogen losses were 11.3 Ib/acre for fresh
manure applications, 3.59 Ib/acre for fermented manure, and 3.20 Ib/acre
for liquid manure. This was compared to 3.89 Ib/acre/yr of nitrogen
vhich was obtained from a control plot which had not received any
manure. Annual phosphorus losses were 2.62 Ib/acre for fresh manure,
0.72 Ib/acre for fermented manure, and 0.86 Ib/acre for liquid manure.
Phosphorus lost from the control plot was 1.17 Ib/acre/yr.
Winter nutrient losses on unmanured plots were found to be less than
losses from manured summer plots. These unmanured plots lost more
nutrients than the manured plots during the summer, however. They con-
cluded that manure should be spread only on unfrozen ground, and the
manure should be incorporated into the soil as soon as possible (32).
Jennelle and Grizzard studied the eutrophication of a Virginia lake.
Stream samples were taken at various tributaries and nutrient sources
were identified as wastewater discharges, rural runoff, and urban run-
off. Rural runoff contributed 710 Ib POj^/day, U$6 Ib NO^-N/day, and
176 Ib total kjeldahl nitrogen/day. Sampling was for a two month
period, and the data shown in Table 2 are extrapolated to an annual
basis. They concluded that even if expensive nutrient removal from
municipal and industrial discharges was practiced that enough nutrients
would still be added to cause rapid eutrophication to continue (33).
BACTERIOLOGICAL INDICATORS
Indicator organisms are usually used to designate the bacteriological
quality of a water or wastewater. The bacteriological examinations
are important in that they imply the presence or absence of a potential
health hazard. Fecal contamination of a water resource is particularly
important, especially if the contamination can be attributed to human
sources.
In a recent symposium Middaugh et al« discussed the possibilities of
using some very selective indicators to distinquish between animal or
human fecal pollution* These were indicators such as Streptococcus
bovis or certain Salmonella species (3^)4 Some of the procedures
appear promising, but additional work remains to be done in unknown
areas. One such area is the determination of survival characteristics
of Streptococcus bovis after animal discharges reach a surface water.
15
-------
As early as the mid-1950 *s investigators recognized the problem of
distinguishing between animal and human fecal contamination of water.
Cooper and Ramadan studied the physiological and biochemical character-
istics of fecal streptococci taken from the feces of humans, bovine
animals, and sheep. The isolated organisms were split into six groups.
One group represented the typical Streptococcus^ faecalis group, while
the other groups vere something other than typical S. f aecallis .
Their results shoved that tvo of the groups of fecaT streptococci
were derived entirely from humans, two groups were traced directly to
animal sources (over 90£); but that two groups were mixed and were not
characteristic of a particular source
Geldrich discussed the use of bacteria as indicators of the sanitary
quality of water, and the techniques involved in testing for fecal
coli forms (36). He presented data (36) relating the average densities
of fecal coli forms and fecal streptococci present in the feces of
humans, ducks, sheep, chickens, cows, turkeys, and pigs. He advocated
a fecal coli form to fecal streptococcus ratio' (FC/FS) to indicate
whether contamination was from a human or animal source. A FC/FS ratio
less than 0«6 would indicate that the fecal source was non-human.
A FC/FS ratio for a water sample of 1.0 or more is now thought to be a
result of human sources, while ratios below 1*0 are thought to result
from animal sources. Evans et al«tend to verify this ratio by their
study of urban stormwater (37T. Hine rainfall runoff events were
sampled for total coli form, fecal coliform, and fecal streptococcus
organisms from a 27 acre residential district in Cincinnati. Fecal
contamination was evident from the biological indicators, although the
stormwater source did not have any domestic waste contamination. In
seven of the nine samples, the ratio of fecal coli forms to fecal strep-
tococcus was less than 0.8H.
Several investigators have attempted to determine the bacteriological
quality of rural areas by sampling streams which drained areas devoid
of municipal or industrial point discharges (38) (39) (bo) (Hi) . In-
vestigations of this nature are intended to define the pollutions!
load from non-point discharges. Ground water which would enter the
stream was not considered separately.
Walter and Bottman collected weekly samples from two watersheds in
a recreational area in Montana to determine concentrations of coliform
and enterococcus organisms (38). One watershed, with its reservoir,
was open to public recreational activities while the other watershed
had been closed to the public since 1920. Common recreational activi-
ties practiced on the open watershed were swimming, boating, camping,
and fishing. Coliform counts increased as the summer progressed,
and enterococci counts increased as the water flowed downstream from
the reservoirs. Coliform counts from the closed watershed were
greater than those from the open watershed in lk% of the tests. Simi-
lar results were obtained for enterococci organisms in 59% of the
16
-------
tests. Animals were thought to have been nearer the vater in the area
closed to the public, and this proximity was offered as a reason for
higher counts from the closed watershed. No actual animal counts were
made in either area.
Runoff from a O.T5 sq. km watershed in Vermont was sampled by Kunkle
for two years (39). The watershed had 31% of its area in hayfields,
3&% in pasture and 25% in forested land. About 150 head of cattle
were grazed at the headwaters of the watershed during the first year,
but they were not on the site after the spring of the second year.
Samples were obtained from a stream Just below the watershed during
normal runoff as well as when storm water runoff was occurring. Much
of the annual runoff resulted from spring snowmelt, but data were
reported only for rainstorms.
During periods of storm runoff, total coliform and fecal coliform den-
sities would rise. Maximum densities of 50 or more times the non-
storm runoff levels were obtained. Seven summer storms showed median
total coliform concentrations from 5,500 to 80,000/100 ml and median
fecal coliform levels of 1,100 to lU,000/100 ml. Over 9055 of the
storm runoff observations for both total and fecal coli forms were
greater than acceptable criteria for swimming waters. Kunkle con-
cluded that fecal coli forms were the better indicator of pollution
in his study (39).
Kunkle has also sampled some mountain streams for bacteria in con-
junction with Meiman (Uo) (Ul). One study obtained counts on fecal
coliform, total coliform, and fecal streptococcus bacteria from 6oU
samples taken at ten stations in the Colorado Rocky Mountains (bo).
The other investigation recorded bacterial densities for the same
indicators from two sites on a mountain stream located 1.5 miles
apart. A total of 3,102 observations were made in this later study
Both studies revealed fluctuations in numbers of bacterial indicators.
Total coliforms varied from 9 to 300 colonies/100 ml, fecal coliform
and fecal streptococcus values normally ranged from 0-75 colonies/
100 ml, and FC and PS densities would increase to several hundred per
100 ml during extreme flooding. The fecal coliform organisms were
found to be most sensitive for detecting animal fecal contamination.
Fecal coliform to fecal streptococcus ratios greater than one were
obtained from locations below cattle grazing areas. The fecal strepr
ococcus organisms were thought to be drastically affected by the cold
water temeratures involved in the studies (Uo)
A study to estimate the contribution of agriculture and/or urban runoff
to part of a northern lake was undertaken by Claudon et^ al . (1*2). The
work was done on Lake Mendota near Madison, Wisconsin. The major
sources of organisms were from a residential storm sewer and a wash-
water drain at the University of Wisconsin Experimental Farm. They
IT
-------
found 27 of 53 samples to be positive for Salmonella, They concluded
that runoff, even diluted runoff, can regularly add Salmonella to a
recreational lake.
Published results of only three investigations were found vhich
examined natural surface drainage from agricultural lands before the
runoff enter a watercourse. Two of these investigations were conducted
on a 173 acre watershed near Brookings, S. Dak. Benson (3) sampled
the watershed during one storm in 19&9, and McCarl (U) sampled the
same watershed during the 1970 season. Benson obtained samples for
total coliform, fecal coliform, and fecal streptococcus throughout a
rainfall runoff event. Densities of the organisms generally varied
with the flow for the first quarter of the event, but random densities
were obtained throughout the remainder of the event (3).
McCarl obtained microbial densities for total coliform, fecal coliform,
and fecal streptococcus organisms for seven runoff events in 1970 (U).
Six of the runoff events resulted from rainfall, but one was a snow-
melt runoff event which occurred between rainfall events. McCarl
concluded that the concentrations of the organisms generally varied
with the suspended solids and organic matter present in the runoff.
Analysis of the fecal coliform to fecal streptococcus ratio led him
to believe that animals rather than humans were the source of the
fecal pollution. Manure had previously been spread on the drainage
basin.
Results of the runoff from six watersheds near Coshocton, Ohio, were
reported by Weidner et^ al. (5)» One watershed contained 303 acres,
three had 1.5 acres each, and two were sized at 7.5 acres each. The
303 acre basin had mixed cover, and two of the 1.5 acre sites were in
a U-year rotation of corn, wheat, and meadow. One of the 7.5 acre
watersheds was strip-cropped, and the other was not. The remaining
1.5 acre watershed was used to study special pesticide applications
and mulch planting techniques.
The researchers determined the densities of total coliform, fecal
coliform, and fecal streptococcus organisms in the runoff from five
watersheds. For some of the watersheds, 90£ of the samples exceeded
1000 total coliforms per 100 ml; and this figure was exceeded in 50jf
of the samples from all of the watersheds. Densities of fecal strep-
tococci were greater than fecal coliform organism densities so they
concluded that ". . .pollution is from animals rather than humans, as
one would expect" (5).
PESTICIDES
The production and use of pesticides has grown throughout the United
States and the world. During the period from 1963 to 1967, for example,
the United States production of synthetic organic pesticides increased
by more than 37 percent. The synthetic organic pesticides include
almost all of the known pesticides (U3).
18
-------
Agriculture has continued a large demand for pesticides, particularly
for corn crops. In a recent year, 51% of the corn planted in the
United States vas treated vith herbicides, and 33% with insecticides.
Sometimes large areas are treated from the air, "but weather and other
practical considerations occasionally limit this method (^3).
The personnel at the Southeast Water Laboratory of the U. S. Environ-
mental Protection Agency in Athens, Georgia have intensively studied
insecticide runoff from agricultural land. Nicholson, Grzenda, and
others have reported this work in several Journals and at various
meetings (M») (U5) (U6) (1*7) (U8) (Ji9).
Nicholson lists three main problems that must be considered when re-
lating pesticides to water quality. The first problem stems from
very high pesticide concentrations which result in fish kills or other
aquatic damage. The second problem is the outcome of exposing aquatic
life to low-level, long-term dosages of pesticides; and the third
problem results when pesticides are processed through municipal water
treatment facilities. Land runoff is a prime source of the entry of
pesticides into surface water resources (
At a recent conference, Nicholson expressed the public's concern over
pesticides. The focal point of this concern has become DDT. Several
countries have banned the use of DDT, and Hungary has banned all
organochlorine insecticides (U5).
Nicholson and his co-workers showed that several insecticides can
enter a watercourse in conjunction with runoff water* He discussed
two fish kills which occurred and mentioned instances where fish-
eating birds have been poisoned. This poisoning vas thought to have
occurred from the biological magnification process (U6).
Greichus also traced the build-up of pesticides in the food chain by
examining the fat of different South Dakota fish, birds, and animals.
The various pathways of insecticide residues in an aquatic environment
were discussed, and data regarding the concentrations of insecticides
in the Lake Poinsett ecosystem were present (50).
Weibel e£ al. collected rainfall and runoff samples at some watersheds
near Coshocton, Ohio. Specific examples of rainfall transmitting
pesticides were documented. In one storm, chlordane, heptachlor
epoxide, DDE, DDT, dieldrin, and 2,U,5-T were all detected. Runoff
water from a field of winter wheat contained O.U3 yg/1 of organic
chlorine. The concentration of organic chlorine gives an estimate
of the pesticide concentration because a large number of pesticides
contain organic chlorine (51).
The fate of agricultural chemicals after they are applied to the land
is an important consideration in regard to the quality of agricultural
19
-------
surface runoff. If chemicals are loosely held by the soil, they could
be leached into the groundvater or dissolved into surface runoff. If
they are tightly held by the soil particles, soil losses become doubly
important.
Several researchers have investigated pesticides in the soil environ-
ment (52) (53) (5fc) (55) (56). Lichtenstein stated that some insecti-
cides such as aldrin and parathion are tightly attached to soil parti-
cles and that only small amounts may be removed by vater. He recognized
that surface vaters could become contaminated from insecticides being
adsorbed by the soil and subsequently being washed from the field by
runoff vater (52).
McCarty and King correlated the extent of pesticide adsorption with
the clay content of the soil used. They found that the rate of move-
ment of a pesticide in the soil was inversely related to the amount
of adsorption (53). The actual method of clay adsorption and the
factors affecting this adsorption became topics for research by several
investigators. White and Mortland discussed clay-organic interactions
and decided that the soil minerals most responsible for adsorption were
those found in clay, and that the attraction of the organic cations to
the clay was proportional to their molecular weight. They listed sev-
eral mechanisms for bonding, and said that the clay particles were
active because of their small particle size and relatively large specif-
ic surface (51*).
Huang and Idao (55) also studied clay mineral adsorption of pesticides,
and they present a wealth of data regarding individual pesticide ad-
sorption rates with various adsorption media. They disagree with White
and Mortland, however, as they state that the ". . .adsorptive capaci-
ties of the clay minerals are not correlated to their ion exchange
capacities or specific surface areas".
Bailey et al. (56) also studied the adsorption mechanisms of a clay,
montmorillonite. They stressed that the pH of the clay system plays
an important role in the adsorption process. Adsorption mechanisms
for both basic and acidic organic compounds were given.
Actual data concerning the level of pesticides in streams are impor-
tant to a runoff study particularly if the streams drain predominantly
agricultural areas. Several investigators have reported field in-
formation from some stream studies.
One of the problems previously mentioned by Nicholson dealt with
pesticides being passed through municipal water treatment facilities
(1»T). A four-year field study where this problem occurred was re-
ported by Thoman and Nicholson (U8). The Flint Creek watershed of
Alabama drains a bOO square mile cotton farming area. Agriculture is
the basic industry and cotton is the principal crop. Toxaphene, BHC,
20
-------
and DDT vere the most commonly used insecticides in the "basin. Insec-
ticides were not applied from the air to any extent.
Just belov the watershed, a town of 7,000 people drew its water for
the community. An extensive sampling program revealed that toxaphene
and BHC were present in both the raw and finished water throughout
each year of the study. DDT was not recovered. The lack of DDT in the
water was attributed to ". . .its strong affinity for organic matter
in the soil, and to its extreme insolubility in water" (U8).
Grzenda et^ al« reported on stream samples taken from a mountain stream
which carried the runoff from a U,000 acre hardwood forest in North
Carolina. DDT was sprayed by airplane in 1961 to halt an infestation
of a hardwood defoliating insect, Ennomos subsignarius. DDT residues
were detected in the stream and ranged from 0.3*i5 to 0.005 ppb.
Controlled spot-applications were practiced the following year on k9%
of the basin after which DDT residues were not detected in either
stream or sediment samples (U9).
Endrin is often used in sugar cane production. Numerous fish kills
in Louisiana were attributed to endrin without substantiating data.
Consequently, Lauer ej^ al. studied the surface waters to determine
if the charges were Justified. Endrin was recovered from all the
streams sampled and surface runoff was listed as the main source of
endrin. Heaviest recoveries were reported during the first runoff
event after an endrin application (57).
Data from a survey of 3 of the Great Lakes and 56 major drainage
basins of the United States were discussed by Nicholson (U5) and
Hill (WO. The original data for the discussion were obtained in
196U by Weaver e£ al. (58). The widespread distribution of dieldrin,
endrin, DDT and DDE was noted. However, concentrations of all pes-
ticides in water were less than 1 ppb. Additional stream surveys
since then have verified these results (U5).
The literature pertaining to pesticides seems to indicate that they
are widespread in the environment, being spread by water, soil, and
air. Most existing data were obtained from stream samples, although
the personnel at the Southeast Water Laboratory have insecticide
information for agricultural runoff. Weibel et/al. also related pes-
ticides to some organic chlorine measurements from some small water-
sheds near Coshocton, Ohio (51). Data regarding pesticide levels in
agricultural runoff waters for the upper mid-west were not found.
21
-------
SECTION V
RESEARCH SITES
INTRODUCTION
When this research to quantify the pollutants in agricultural runoff
was first proposed, the plan of operation was to determine concentra-
tion of pollutants in the runoff from drainage basins selected from
studies initiated by the South Dakota Department of Highways and the
U. S. Geological Survey. A total of 80 small drainage basins, with
areas from 0.5 to 10.5 square miles, were to be instrumented with dig-
ital stage and rainfall recorders for determining the rainfall-runoff
relationships. Consequently, measurements of flow and precipitation
collected with the digital recorders could hare been utilized. Several
objections to this plan were foreseen. Travel distances to maintain
automatic samplers and collect samples might have been excessive.
Compositing of samples without readily available flow data would be
difficult. The drainage areas would be large and probably include
numerous cropping practices and soil types. For these reasons, the
decision was made with the advice of FWPCA personnel to select smaller
drainage areas and to utilize flow measuring and sampling equipment
provided within the project.
The starting date for the project was proposed initially as July 1,
19^9. Delays to November 1, 19&9 were recognized as acceptable to
allow selection and instrumentation of the drainage basins for study
during the following spring and summer runoff season. The initial
notification of the approval of the project was received in December
1969 and the project was actually initiated in February 1970. The
later-than-planned starting date necessitated selection of drainage
areas and sampling sites during the winter when the area was covered
with snow.
Because of winter conditions, roadside sites with culverts were se-
lected as study sites to avoid the earth moving that might be nec-
essary for installing the flow measuring equipment. When selecting
the study sites for this research, the ideal drainage area was con-
sidered to possess the following characteristics:
1. Relatively uniform and well-defined soil type,
2. Farmed uniformly with one crop cover annually,
22
-------
3. Located near South Dakota State University to avoid
excessive travel,
U. Adjacent to an all-weather road,
5. Have a known history of previous land use including
applications of pesticides,
6. Have a drainage culvert that would lend itself to in-
stallation of flow measuring equipment.
Seven sites were selected. Four of the sites were located during
January of 1970 and the remaining sites were selected later in the
summer. Figure 1 shows the general location of the sites which are
all located in Brookings County, and are within 20 miles of Brookings,
S. Dak.
The chosen sites generally follow the criteria of the ideal site. They
all have one crop cover and drain to a single point. The approach roads
were not ideal, but did allow sampling as necessary. Fertilizer infor-
mation was obtained from the land operators.
23
-------
o
o
NORTH DAKOTA
SOUTH DAKOTA
NEBRASKA
SITE NOS. 8 AND 9
VOLGA
WHITE
HWY
BROOK INGS
SITE NOS.
1,2,3,4
AND 7
_•
BUSH NELL
SCALE: I" = 4 MILES
FIGURE I. - GENERAL SITE LOCATIONS
-------
SITE NO. 1 (See Figure 2)
Site No. 1 was cultivated and seeded in an oats and corn rotation.
The landowner preferred oats because of the additional soil pro-
tection it offered. The land vas severely eroded in spots*
Legal Description:
Area:
Average Slope:
Soil Texture:
Soil Type:
Crop Cover:
Fertilizer:
NW 1/U, Sec. 23, RU8W, T111N,
5th PM
7.18 acres
Approximately b,l%
Sandy clay loam
H8.8JJ sand, 2k.1% silt, 26.5* clay
Vienna Loam
1970 - Corn stubble in spring,
oats in summer
1971 - Oat stubble in spring,
oats in summer
1972 - Plowed previous fall,
corn in summer
1970 - Uo Ib N/acre and 26.h Ib P/acre
1971 - 52.2 Ib N/acre and 11.5 Ib P/acre
1972 - 60 Ib N/acre and 26.U Ib P/acre
25
-------
18' DIA. CULVKRT
'|00« KI.KV. OK H-FLUME
SIM)
j
FIGURK 2.- SITK
-------
SITE HO. 2 (See Figure 3)
Site No. 2 is adjacent to Site No. 1, and both sites are contained
within the same larger area* Consequently, the crop rotation and
the fertilizer application vas always the same for each site.
Legal Description:
Area:
Average Slope:
Soil Texture:
Soil Type:
Crop Cover:
Fertilizer:
NW1/U, Sec. 23, R**8W, THIN, 5th PM
8.77 acres
Approximately U.I?
Sandy clay loam
U7.8* sand, 2U.7* silt, 27.5* clay
Vienna Loam
1970 - Corn stubble in spring,
oats in summer
1971 - Oat stubble in spring,
oats in summer
1972 - Plowed previous fall,
corn in summer
1970 - Uo Ib N/acre and 26. U Ib P/acre
1971 - 52.2 Ib N/acre and 11.5 Ib P/acre
1972 - 60 Ib N/acre and 26.U Ib P/acre
27
-------
'.-15* DIA. CULVERT -
MOO* LIP OK H-H.UMK
<>
I
500'
I
F1GURK 3.- SITK 2
-------
SITE NO. 3 (See Figure U)
Site No. 3 vas a hayfield with a permanent grass cover of mixed
brome grass and alfalfa* The cover vas cut and put up for hay
two or three times each growing season. Cattle were sometimes
pastured in the fall.
Legal Description:
Area:
Average Slope:
Soil Texture:
Soil Type:
Crop cover:
Fertilizer:
SW1/U, Sec. 23, RU8W, T111N, 5th PM
10.12 acres
Approximately H.0$
Sandy clay loam
U6.8J5 sand, 26.7? siat, 26.5* clay
Vienna Loam
1970 - Brome grass and alfalfa
1971 - Brome grass and alfalfa
1972 - Brome grass and alfalfa
1970 - 100 Ib If/acre and 13.2 Ib P/acre
1971 - 91.5 Ib N/acre
1972 - 250 Ib N/acre
29
-------
1
- -i—18'> CULVERT-/-^ r "-'
iobo
LIP OF H-FLUME
ROAO
• 00*
FIGURE 4.-SITE 3
-------
SITE HO. fc (See Figure 5)
Site Ho. U vas adjacent to Site No. 3 and a single hay field contained
both sites. The fertilizer applied and land management practices vere
the same as those for Site Ho. 3.
Legal Description:
Area:
Average Slope:
Soil Texture:
Soil lype:
Crop Cover:
Fertilizer:
SWlA, Sec. 23, RUoV, T111H, 5th PM
8.77 acres
Approximately U.l£
Loam
1*8.8* sand, 36.1% silt, lU.5* clay
Vienna Loam
1970 - Brome grass and alfalfa
1971 - Brome grass and alfalfa
1972 - Brome grass and alfalfa
1970 - 100 lt> H/acre and 13.2 ID P/acre
1971 - 91.5 lb H/acre
1972 - 250 lb H/acre
31
-------
J
I '
^ ( — t- HOADT7 — LlS- D1A. COVERT
-100* LIP OF H-KLUME
500
FIGURE 5. - SITE 4
-------
SITE HO. 7 (See Figure 6)
Site No. 7 served as summer pasture for a livestock feeder. The
pasture vas left vacant until mid-summer vhen it vould be heavily
pastured for 3 to k weeks. The cattle vere then removed to
another location.
Legal Description:
Area:
Average Slope:
Soil Texture:
Soil Type:
Crop Cover:
Fertilizer:
NWlA, Sec. 26, RU8W, THIN, 5th PM
15.51 acres
Approximately 5.^#
Sandy clay loam
1*9.8% sand, 26.1% silt, 23.5# clay
Vienna Loam
1970
1971
1972
None
Grazed pasture
Grazed pasture
Grazed pasture
33
-------
-I U
500
_)
FIGURE fc. - SITE 7
-------
SITE NO. 8 (See Figure 7)
Site No* 8 vas cultivated and seeded in an oats and corn rotation,
A fairly impervious layer lay "beneath the soil surface. Erosion
vas most noticeable at the upper elevations of the basin.
Legal Description:
Area:
Average Slope:
Soil Texture:
Soil Type:
Crop Cover:
Fertilizer:
SEl/U, Sec. 29, R51W, T111N, 5th PM
18.68 acres
Approximately 3.0/J
Sandy clay loam
U6.8* sand, 26.755 silt, 26.55? clay
Poinsett-Buse-Pierce
1970 - Corn in summer
1971 - Corn stubble in spring
oats in summer
1972 - Ploved previous fall,
idle acres in summer
1971 - 29 Ib N/acre and 6.2 Ib P/acre
1972 - None
35
-------
<••
0
ino' KIKV Oh H-KLI'MK
FK;URK 7.- SITK «
-------
SITE NO. 9 (See Figure 8)
Site No. 9 is adjacent to Site No. 8, and both are contained within
the same larger area. Consequently, the crop rotation and the
fertilizer application were always the same for each site.
Legal Description: NE1/U, Sec. 29, B51W, T111N, 5th PM
Area:
Average Slope:
Soil Texture:
Soil Type:
Crop Cover:
Fertilizer:
9.79 acres
Approximately 2.1%
Sandy clay loam
U9.8# sand, 25.7% silt, 2k.5% clay
Poinsett-Buse-Pierce
1970 - Corn in summer
1971 - Corn stubble in spring,
oats in summer
1972 - Plowed previous fall,
idle acres in summer
1971 - 29 lb N/acre and 6.2 Ib P/acre
1972 - None
37
-------
-...
00
\00«
00« KLEV. OK H KLl'ME
a
FIGURE 8.- SITE
-------
SECTION VI
FIELD METHODS
SITE INSTRUMENTATION
Flow measuring equipment, vater level recorders, and automatic
samplers were placed at each site. Figures 9 and 10 show front and
side views of the field installation. The drainage from each site
passed through a culvert and the installation of the field equipment
was on the downstream side of the culvert. All equipment shown was
in the road ditch within the right-of-way.
The flow measuring device was an H flume constructed of I1* gauge
galvanized steel with a "black iron supporting frame. The flume was
mounted on a plywood headboard which extended approximately 20 in.
underground to eliminate the possibility of water flowing beneath the
measuring flume. The flume and headboard were bolted to the side-
boards, and the joints were caulked to prevent leakage.
Sideboards were extended beyond the culvert exit for a distance of at
least five times the height of the measuring flume. This was in
accordance with installation instructions given in Field Manual for
Research in Agricultural Hydrology (59) to avoid velocity disturbances
upstream from the flow measuring device. The distance between the
sideboards was equal to the flume width. Sideboards were U ft by 8 ft
sheets of exterior plywood and extended a minimum of 15 in. below the
ground surface to eliminate leakage during runoff events.
Sites No. 1, 2, 3, and U were equipped with a 1.5 H flume; while
Sites No, 7* 8, and 9 were provided with a 2.5 H flume. Dimensions
and rating tables for the flumes are listed in "Agricultural Handbook
No. 22V (59).
A Leupold Stevens Type F water level recorder was used to measure the
depth of flow in the flume during rainfall runoff events. A contin-
uous record of the depth was traced on a graph by the recorder. The
depths were later converted to flow measurements using the rating
table previously mentioned. The recorder was equipped with a self-
starting device which actuated when the depth of flow through the
flume reached 0.05 ft.
39
-------
: -.
FIGURE 9. - Front view of field installation.
FIGUPE 10. - Side viev of field installation.
-------
During spring snovmelt, the water in the recorder's stilling veil
vas prone to freeze even though the vater vas still flowing freely
in the culvert and flume* Consequently, spring snowmelt runoff
measurements were obtained manually using a staff gauge. The
recorder vas used for rainfall events only.
As flow would commence from the culvert and proceed through the flume,
the float vould start to rise in the stilling veil. The movement
of the float actuated the self-starting mechanism which started the
recorder. The recorder vas modified to send a signal to the auto-
matic sampler at predetermined time intervals. The sampler utilized
a vacuum principle to obtain a runoff sample each time a signal vas
received.
A vooden catch-basin vas designed and fabricated to assist in obtain-
ing representative samples when using the automatic sampler. The
catch-basin vould intercept about one-half of the flow, and allow mix-
ing to occur as the sample vas being taken. The automatic sampler head
was located 1 1/U in. above the floor of the catch-basin. A drainage
notch in the front of the catch-basin assisted in sample collection by
retarding the runoff vater at low flows. The notch and catch-basin
were fashioned to maintain a self-cleansing unit, however. Figure 10
shows the positioning of the sampler head and the catch-basin.
Verification of the sampling equipment was accomplished by obtaining
samples at both the catch—basin and the culvert outlet. Solids deter-
minations conducted on the analogous samples were used for comparison.
Plastic rain gages which could measure rainfall of up to 6 in. were
placed at Sites Ho. 1, kt 7, and 9. These plastic rain gages vere
manufactured by Edwards Manufacturing Co. of Albert Lea, Minnesota.
Sites No. 2 and 8 vere equipped with a recording rain gage which
recorded precipitation with respect to time* Rainfall intensity
could then be calculated. The recording rain gages were manufactured
by the Bedford Instrument Co. of Baltimore, Maryland. These recording
or weighing rain gages are of U. S. Weather Bureau approved design
with an 8 in. diameter opening. They also had the capability of
measuring rainfall of 6 in. or less.
SAMPLING SNOTOffiLT RUNOFF
Samples of snowmelt runoff vere taken manually from each site at
periodic intervals throughout each day. On a normal day, about five
or six samples were taken from each site. More frequent sampling was
deemed unnecessary because of the relatively uniform qualities of the
runoff.
Each time that sampling occurred, two individual samples were collected.
One sample vas obtained in a sterile plastic bag for bacteriological
testing. Another sample vas obtained in a clean glass container.
Ul
-------
All samples vere marked for identification. Two composite samples
vere made after the day's sampling vas completed; one from the plastic
hags and one from the glass containers.
Depth measurements vere taken from the H flumes throughout each runoff
day. Usually depths vere measured and recorded at one-half hour inter-
vals from about 8 AM to 6 PM. Readings vere taken at less frequent
intervals thereafter, since the flov would normally have diminished by
this time. Readings vere taken vith a staff gage and recorded to the
nearest one-hundredth of a foot.
SAMPLING RAINFALL RUNOFF
Because of the short duration and unpredictable occurrence of a summer
rainstorm, it vas not possible to collect these runoff samples man-
ually. An automatic sampler vas therefore used to obtain rainfall
runoff samples. Manual sampling vas sometimes employed to supplement
those samples collected by the automatic sampler. Manual sampling vas
especially necessary for those rainfall events vhich had a duration of
more than four hours*
Desirable criteria developed for the automatic sampler for the project
included that it be self-starting, obtain a large volume sample, and
collect representative samples of the total flov. The self-starting
feature vas necessary because the research sites vere remotely located
and personnel vere unavailable at short notice to collect samples. A
sample volume in excess of 1,500 ml vas needed for the laboratory deter-
minations. Because of the anticipated quality changes that would occur
throughout a runoff event, it vas necessary that the final composited
sample depict the entire runoff period. Also since electrical power
vas not available at the remote locations, any sampler vhich needed an
electrical hook-up vas unsuitable.
A satisfactory sampler to meet these criteria could not be purchased.
Many of the commercial samplers required 110 volt electrical energy
to drive a sampling pump. Other samplers obtained a composite sample
comprised of individual aliquots collected at fixed time intervals and
because the volume of each individual aliquot vas the same, the com-
posite sample vould not be representative of the runoff event. Still
other samplers could not be adapted to become self-starting when flov
began. Because of these shortcomings, an automatic, self-starting
sampler vas designed to satisfy the requirements of the project.
The sampling unit designed for this project incorporated the flov
measuring device, an H flume equipped vith a vater level recorder,
as an integral part of the sampling unit. Flov through the flume
started the sampling sequence.
1*2
-------
The sampler contained 12 sample bottles (about 2 liters each) vhich
had been evacuated. As the vacuum was released on one of the bottles,
a water sample vas drawn into the bottle. The sampler vas connected
electrically to attachments on the Leupold & Stevens Type F water
level recorder equipped with a self-starting clock. Power was
supplied by two 6 volt dry cell batteries wired in series to make a
12 volt system. The operational sequence of the sampler was as
follows:
1. The float on the water level recorder would rise when
water started to flow and an automatic clock starter
actuated the recorder clock.
2. As the clock ran, it caused the recording pen to move
across the sheet. An attachment to the recording pen
periodically completed an electrical circuit as the
pen traveled across the recorder.
3. Upon completion of the electrical circuit, a solenoid
was actuated which tripped a mechanism which released
the vacuum held in one of the bottles.
U. As the vacuum decreased, the water sample was collected
through individual hoses connecting each bottle to the
sampler head.
Figure 11 shows an overall view of the interior of the sampler. Each
bottle had an intake hose with a solenoid operated clamp which pinched
off a rubber hose to contain the vacuum in the bottle until released.
A vacuum of about 2U in. Eg was applied to the 12 individual bottles,
contained within the sampler. The water sample entered the bottle
which was under negative pressure. The sampler head consisted of glass
tubing connected to Tygon tubing* The glass tubing was protected by
an aluminum tubular shield.
Figure 12 shows some of the details of the hose clamping system. The
hose clamps were manually latched and an electrical solenoid released
the clamp as the solenoid received a signal from the recorder. With
all of the solenoid-operated clamps engaged vacuum was applied to the
bottles using a vacuum pump and a portable generator, through a common
manifold. The manifold was located on the backside of the crosspiece
at the top of the sampler through which the 12 hoses pass. After the
proper amount of vacuum had been attained, a Hoffman "H" clamp was
tightened on each hose to seal each bottle individually until the
vacuum was released by the solenoid clamp. For the top row of the
solenoid-operated clamps shown in Figure 5, the two hoses on the left
are pictured in the clamped position whereas the four clamps on the
right have been opened by the solenoid.
U3
-------
FIGURE 11. - Interior view of sampler.
' 9*11101111
- i i- ...- k
FIGURE 12. - Viev of solenoid controlled clamping system for
the sant>ler.
-------
Figures 13 and lk shows the modifications made to the Leupold and
Stevens Type F water level recorder. A multi-conductor cable connected
the sampler, where the battery and solenoids were contained, to the re-
corder by the multi-connector plug shown in Figure 13. A plastic
strip with two rows of screws was mounted directly above the pen path
on the recorder. These screws served as electrical connections which
were wired to the battery and solenoid clamps in the sampler through
the multiconductor dable. An attachment fastened on the pen carriage
served as a switch to make contact with both screws as the pen traveled
the length of the chart drum.
The spacing of the contacts along the plastic strip determined the
tine intervals at which samples were collected. The particular level
recorder shown in Figure lU was equipped for a four hour runoff event,
and after starting would collect 12 individual samples in succession.
Time intervals for collection were two samples each at 5 minute inter-
vals, 2 at 10 minute intervals, U at 20 minute intervals, and k at 30
minute intervals.
After each runoff event the samples and the stage level chart were re-
moved from the sampler and the sampler was reset. A single liquid com-
posite sample was made using individual aliquots representing the flow
volume at the time of collection of the individual samples. This
single composite was used for bacteriological, liquid pesticide, chemi-
cal, and physical tests. A separate mud sample for pesticide determina-
tions was usually collected. This mud sample was obtained after the
runoff was finished by scraping mud deposits from various points where
fresh sediment was apparent. The mud sample was used as an indication
of the pesticide associated with sediment. If fresh sediment accumu-
lations did not seem obvious, mud samples were not taken.
1*5
-------
FIGURE 13. - Type F water level recorder with modifications
to activate automatic sampler.
FIGUBE I1*. - Viev of level recorder showing mounting of
plastic strip.
k6
-------
SECTION VII
LABORATORY METHODS
PHASE I
Phase I is the designation given to those activities which were carried
out during the initial year of the project. These activities included
searching for the selection of the field research sites, purchasing and
installing field and laboratory equipment, and collecting and storing
runoff samples until the purchased laboratory equipment was available
for some of the analytical determinations.
Background
Funding for the project was authorized as of February 1, 1970 and a
search for acceptable research sites was immediately initiated, al-
though the search was hampered by the snow cover present at this time.
Four suitable sites were soon located and the field measuring equip-
ment was ordered in an attempt to secure acceptable flow data for the
spring snowmelt events. Unfortunately, a preseason rainfall on frozen
ground preceded the installation of the field equipment. Consequently,
total quantities of runoff pollutants could not be computed for the
four research sites which were instrumented during the initial year.
Procedures
The flow diagram which was followed for the treatment and analysis of
all the runoff samples obtained during 1970 is shown in Figure 15.
Sterile plastic bags were used to collect the snowmelt runoff samples
for bacteriological tests and total coliform (TC), fecal coliform (FC)
and fecal streptococcus (FS) analyses were performed on each discrete
sample. For purposes of comparison, composite MPN values for TC, FC
and FS were calculated from the individual snowmelt samples and the
flow information. During rainfall events, the bacteriological deter-
minations were conducted on a composite sample which was obtained by
proportioning the individual samples collected by the automatic sampler.
Determinations for total coliform, fecal coliform, fecal streptococcus
suspended solids, pH, and specific conductance were made on fresh
samples* All remaining determinations were conducted on frozen samples
U7
-------
Discrete
Sample
Discrete */
Bacteriological
Sample
I '
»«l /
I -'
Total Coliform
Fecal Coliform
Fecal Streptococcus
Discrete
Sample
COMPOSITE SAMPLE
Composite
Pesticide
Frozen
Storage
F
Suspended
Solids
I
pH
I
Specific
Conductance
Frozen Storage
-Chemical Oxygen Demand
-Nitrate H
— Ammonia H
_Total KJeldahl
Hitrogen
— Total Phosphorus
* Rainfall runoff only
** Snovmelt runoff only
Discrete
Sample
Composite Sample For
Chemical & Physical
Tests
Centrifuge
I
O.U5 u
Filter
Frozen Storage—'
. Biochemical
Oxygen Demand
_ Chemical Oxygen
Demand
— Nitrate N
— Ammonia N
_ Total KJeldahl
Nitrogen
— Total Phosphorus
*— Total Solids
Figure 15. - Flow diagram for treatment and analysis of samples
during Phase I»
1*8
-------
which had been stored for about six months. It was subsequently shown
that several of these determinations are affected by frozen storage and
therefore some of the results from Phase I are not directly comparable
to those obtained during Phase II. The primary reason for using the
frozen storage technique vas the unavailability of the necessary anal-
ytical equipment until the fall of 1970.
Pesticide values reported during Phase I are not valid because of the
storage method used. Composite pesticide samples were frozen in heavy
duty plastic bags prior to analysis. Known standard samples were
stored in the same manner and revealed that a majority of the pesticide
in the sample could not be recovered after storage.
PHASE II
Phase II is the designation given to those activities which were carried
out during the final two years of the project. Phase II activities
were basically the routine field maintenance required, collection of
samples, and data acquisition and interpretation. Field and laboratory
equipment had been purchased, installed and tested for reliability.
Prolonged storage of the samples vas not a problem during this phase.
Sample Storage and Handling
The methods of sampling runoff were the manual collection of individual
samples during snowmelt runoff, and the automatic and manual collection
of samples during rainfall runoff. A single composite sample for each
event was then made from the individual samples. Each individual
sample represented a certain percentage of the total flow, therefore
the volume taken from the individual sample was this percentage multi-
plied by the required volume of the composite sample.
Figure 16 depicts the processes followed during snowmelt runoff. Dis-
crete samples were held at U°C prior to compositing. Compositing was
completed within 12 to 28 hours after sample collection. Part of the
composite sample was then frozen for later analysis. Laboratory deter-
minations on the remaining sample portion were finished within one week
of initial collection. Samples were stored at U°C during this period.
Passing the sample through the 0.1*5 micron filter allowed the deter-
mination of the soluble fraction of certain constituents; namely,
chemical oxygen demand, total kjeldahl nitrogen, and total phosphorus.
Certain determinations were thought to be affected by freezing and
these were conducted on fresh, unfrozen aliquots. All determinations,
which were carried out on a sample which was preserved by freezing,
were verified by utilizing a test set of samples to determine the
concentration of the particular parameter before and after frozen
storage. Because of time limitations during snowmelt runoff, the
ammonia determination was not performed as this constituent would be
included in the total kjeldahl nitrogen results.
-------
BACTL
(SAMPLES
DISCRETE
SAMPLE
,
DISCRETE
SAMPLE
DISCRETE
SAMPLE
1 1
TO
BACT1.
LAB
CENTRIFUGE
If
FILTER THRU
GLASS FIBER
FILTER
POSSIBLE
ADDITIONAL
COMPOSITING
TO PESTICIDE
LAB
TOTAL
P*
*I97I ONLY
** 1972 ONLY
COMPOSITE SAMPLE
SPEC.
COND.
TOTAL
P*
FREEZE
[NO* |
TUIAL
P**
1
TOTAL
RESIDUE
CENTRIFUGE
0.45 u
FILTER
FIGURE 16.
- Flow Diagram for snowmelt determinations
during Phase II.
50
-------
Nitrate-nitrogen, soluble chemical oxygen demand, and total residue
vere determined on snowmelt samples which had been preserved by
freezing. In addition to these three determinations, the effects of
freezing on total phosphorus and total kjeldahl nitrogen vere also
investigated. A t-test based on a procedure presented by Steel and
Torrie (60) was used to determine the significance of the freezing
effects, and the results are shown in Table 3. The results are based
on the average of duplicate determinations for 10 samples. It can be
seen from the table that only total kjeldahl nitrogen was significantly
affected by frozen storage of snowmelt samples. All values for total
kjeldahl nitrogen reported herein for Phase II were obtained by deter-
minations performed on fresh, unfrozen samples.
TABLE 3. - Summary of Analysis of Variance of Freezing
Effects on Analytical Determinations of Snowmelt Samples
Determination
t g 0.05 Level
Remarks
Total
phosphorus
Total Kjeldahl
0.083
2.262
Not significant
nitrogen
Nitrate
Soluble COD
Total residue
2.U1
0.316
1.92
0.91
2.306
2.262
2.262
2.365
Significant
Hot significant
Not significant
Not significant
Initially during 1971 the pesticide determinations were made on each
snowmelt composite sample. Because the concentrations were below the
analytical limits of the tests, later pesticides analyses vere made
on a sample representing a composite of two or three days flow.
The analytical determinations of the rainfall runoff samples were made
on fresh, unfrozen composites. All determinations made of the snow-
melt samples vere also made of the rain runoff samples. In addition,
ammonia vas determined, and pesticide analyses vere conducted on a sed-
iment sample as veil as the vater portion of the runoff. Rainfall run-
off samples vere stored at U°C until tested. All determinations on
rainfall runoff samples vere completed vithin one veek of collection.
51
-------
Bacteriological Determinations
Total coliform, fecal coliform, and fecal streptococcus determinations
for all composite samples vere made tinder the direction of Dr. Paul
Middaugh. The density (MPN/100 ml) vas found using the multiple-tube
fermentation technique described in the 12th and 13th Editions of
"Standard Methods" (6l) (62). Both the total coliform and fecal
streptococcus technics utilized the confirmed test; the total coliforms
vere confirmed by using "brilliant green lactose bile broth, and the
fecal streptococci vere confirmed "by using ethyl violet azide broth.
Fecal coliforms vere determined using technics requiring ED medium and
a temperature of U5°C.
During periods of snovmelt runoff, bacteriological samples vere
collected in sterile plastic bags, stored at U°C, and then composited
according to flov. Tests vere started on the composite samples
approximately l6-2l» hours after initial collection.
Individual bacteriological samples vere not collected for rainfall run-
off events* A single composite sample vas made from the discrete sam-
ples vhich the automatic sampler had collected, and the bacteriological
sample vas a portion of this composite sample. Tests vere initiated
on the bacteriological samples vithin approximately 12 hours after ini-
tial collection by the automatic sampler.
Pesticide Determinations
Pesticide analyses vere performed under the direction of Dr. Y. A.
Greichus. All vater samples vere analyzed oy the procedures described
in The^ Identification and Measurement of Chlorinated Hydrocarbon Pesti-
cides in Surface Waters (63). The follovine insecticides vere in-
cluded: lindane, heptachlor, aldrin, heptachlor epoxide, DDE, DDD,
DDT, and dieldrin; vhile herbicides included the triazines, simazine
and atrazine.
On all snovmelt samples, pesticide data vere obtained on only the run-
off vaters; sediment vas excluded by filtration according to "Standard
Methods" (62). On rainfall runoff samples, data vere obtained on both
the vater and sediment portions of runoff.
Pesticide composites for rainfall runoff represented a single event.
Hovever, because of the lov concentrations of pesticides in the snov-
melt runoff; the snovmelt samples vere composited over a longer time
period vhich vas normally of tvo to three days duration.
Determinations Using the Auto Analyzer
Several of the analyses vere made by using a basic autoanalyzer system
as manufactured by the Technicon Corporation. Components used included
proportioning pumps and manifolds, heating baths, a continuous digester,
52
-------
colorimeter, and recorder. Tests conducted using the auto-analyzer
were soluble chemical oxygen demand, raw and soluble total phosphorus.
raw and soluble total kjeldahl nitrogen, ammonia, and nitrate.
Methods used on the autoanalyzer are based on "Standard Methods" (61).
They are basically the same procedures as used in the laboratories
of the Environmental Protection Agency and listed in Methods for Chemi-
cal Analysis of Water and Wastes 1971 (6U). Specific laborat^y" Methods
used were obtained from the Technicon Corporation and are shovn in
Table k.
Industrial Method 1-68W vas used for total phosphorus although the
method is specified for orthophosphate. Total phosphorus samples
were prepared by sample digestion on a hot plate as described in
Methods for Chemical Analysis of Water and Wastes 19T1 (6U).
Glassware was cleaned with 1:1 HC1. —
TABLE k. - Autoanalyzer Methodology
Test
Technicon Methodology
Soluble chemical
oxygen demand
Raw and soluble
total
phosphorus
Raw and soluble
total kjeldahl
nitrogen
Aanonia
Nitrate
Industrial Method
Industrial Method
Industrial Method
Industrial Method
Industrial Method
26-69W
1-68W
30-69A
19-69W
32-69W
Other Physical and Chemical Determinations
All remaining analytical determinations were conducted in accordance
with "Standard Methods" (6l) (62). Parameters so measured were total
suspended matter, total residue, chemical oxygen demand, and specific
conductance.
53
-------
Total suspended natter vas determined by passing the sample through
glass fiber filter disks positioned on a membrane filtering apparatus.
The filters vere dried at 103°C and allowed to cool in a desiccator
before weighing. Blanks vere handled in the same vay as the filters
to account for any weight loss of the filters upon drying. Suspended
matter was computed from the weight gain obtained after drying.
The test for total residue on evaporation specifies a drying temper-
ature of 103°C to minimize losses of volatile materials which may be
present. Coors porcelain evaporating dishes were used, and were
predried to a constant weight* Sample volumes of 100 ml were evapor-
ated on a water bath and the dishes were dried to a constant weight in
an oven, cooled and weighed* The weight gain represented the total
residue of the sample.
Chemical oxygen demand determinations were conducted using the standard
dichromate reflux method with sample \yolumes of 20 ml. Specific con-
ductance values were obtained using a Type PC conductivity bridge as
manufactured by Industrial Instruments', Inc. of Cedar Grove, N. J. The
cell constant was checked periodically and the samples vere allowed to
warm up to room temperature before the conductivity vas measured. Con-
ductivity valves vere adjusted to a value at 25°C by using a graph
based on a 0.01 M KC1 solution.
-------
SECTION VIII
DATA AND RESULTS
CLIMATOLOGICAL SUMMARY
All the research sites are located within Brookings County and within
20 miles of Brookings, South Dakota; an area that annually receives
about 20.k inches of precipitation. Over 80/5 of the precipitation
occurs as rainfall, and the annual snowfall averages 23 inches.
Brookings is located in east central South Dakota and enjoys a con-
tinental climate. The only nearby water is the Big Sioux River about
six miles west of Brookings. Some small lakes are 15 to 30 miles
west and northwest (65).
Temperatures annually rise above 100 degrees in summer and drop to
20 degrees below zero or lower in the winter. The last frost will
usually occur before May 18, and the average date for the first frost
in the fall is September 22 (65).
Prevailing winds average about 10 mph from the south during the
summer months and usually average about the same speed from the north-
west during the winter months. Wind velocities of over 50 mph are
common and may occur at anytime, "but are most common during a summer
thunderstorm (65).
Evaporation normally exceeds the precipitation with an annual average
of Uj inches from a Weather Bureau Class A evaporation pan. The
nearby shallow lakes average about 33 to 31* inches of evaporation per
year (65).
Table 5» a precipitation summary, indicated that 1971 vas probably
about an average year with regard to precipitation, while 1972 was
considerably above average. The rainfall data collected at Sites No.
2 and 8 are considered to be more accurate because recording rain gages
approved by the United States Weather Bureau were used at these sites.
The other sites were equipped with plastic Tru-Chek gages as manu-
factured by the Edwards Manufacturing Co. of Albert Lea, Minnesota.
Because Sites No, 1 and 2 are adjacent, and Sites No. 8 and 9 are
adjacent; it appears that the recorded rainfall from the plastic rain
gages averaged about 3% to 5% high.
55
-------
TABLE 5. - Precipitation Summary at Research
Sites Prom Mid-March to Mid-November*
Site No.
1
2
3 & U
7
8
9
Type of Gage
Plastic
Recording
Plastic
Plastic
Recording
Plastic
1971 Rainfall (in.)
19.96
19.05
19.70
19. 6U
18. 11
18.58
1972 Rainfall (in.)
26.56
21U9
25.88
26.35
27.93
28.93
* Average annual precipitation for Brookings, S. D. from mid-March to
mid-November is about 18.U".
Table 6 shows the increase in the number of runoff events for the 1972
season. Comparing the total runoff events for the three years causes
speculation regarding the amount of runoff vhich results from rainfall
during a normal year*
TABLE 6. - Frequency of Runoff from Rainfall
Year
1970
1971
1972
Average No. of
Days of Rainfall
1*6
1*9
6k
Actual Days of
Rainfall Runoff
2
2
1*
Number of Events
of Rainfall Runoff
U »
2
30
* Four of seven sites vere operable during 1970
56
-------
Actually 1971 can not be considered as a normal year with respect to
rainfall. While the total amount of rainfall vas normal, the seasonal
distribution of the rainfall was quite abnormal. Almost one-third of
the total rainfall came after August vhich resulted in limited runoff
because of the established crop cover.
Considerably more rainfall than usual was recorded during 1972. A
substantial amount of this rain was received in late spring and early
summer before cover became established on the cultivated fields which
was reflected in the large increase of rainfall runoff events for
1972. In May of 1972, a total of 9.35 in. of rain was recorded at
Site No. 8 and 7.97 in. was recorded at Site No. 2. This May of 1972
was the wettest May on record and ranks as the second or third wettest
month since record keeping began in 1893 (65).
Therefore, as far as rainfall runoff is concerned; two of the three
years of study can be considered as approaching the maximum and mini-
mum conditions. The second year, 1971, was a minimal rainfall runoff
year because the precipitation occurred when the ground cover was well
established. The third year, 1972, approached a maximum rainfall run-
off condition because much of the rainfall occurred when the ground
was unprotected. Both of these years occurred during Phase II and
complete runoff data were obtained.
Several factors affect both the quantity and quality of surface runoff.
Some of these factors are surface and subsoil type and formation,
ground cover, intensity and frequency of precipitation, the total amount
and duration of rainfall, the topography of the area, land management
practices, and the time of the year.
To fully evaluate the effect of each factor was beyond the scope of
this project. However, some of the above factors are quite pertinent
to the runoff patterns obtained for a particular site; and they will
be referred to as the runoff results are presented.
RUNOFF ANALOGIES
Figure 17 illustrates the runoff patterns for each of the two project
years in Phase II. A comparison between the two years regarding the
relative proportion of snowmelt and rainfall runoff is quite interest-
ing.
The first year had an almost negligible amount of rainfall runoff.
Figure 17 shows that all of the sites, with the exception of Site No. 8,
had snowmelt as their only surface runoff. Rainfall runoff accounted
for only about 0.2% of the year's runoff volume on this single site.
Figure 17 also indicates that the rainfall runoff for all the sites
during the second year was substantially greater than for the first
57
-------
72 71 "72 "71
12347
SITE AND YEAR
FIGURE 17.- ANNUAL RUNOFF PATTERNS FOR PHASE II.
-------
y!ear% ?? Primary reason for this difference vas the change in season
tlr±S distrlbuf ' aese figures also point out variations T
the amount of snovmelt runoff. Only Sites No. 3 and 7 experienced the
6
not
previous year site N°- 9 had
Obviously the conditions affecting snovmelt runoff vill not be the
same from year to year. Snovfall vill vary, evaporation vill chance
the rate of thaving for the soil and the snov viL differed £* '
ESS* • + dlff"^ to correlate conditions from year to year.
However at vas very apparent to even the most casual observe/that
1 2 S L^ V ^elt rUn°ff WOUld be ****™* on Sites Ho.
1, 2. 8. and 9 during 1972. All four of these sites had been in the
ete e ° C PreVi°US year "* the ^d had kept
the sites almost clear of snov throughout the vinter. The effects of
vind erosion vere evident. «iects or
A small amount of snov vas retained on Sites 1, 2, and 9. The snov
°nS the -asur^ a«f became
« present. Some snov vas retained near
the summit of the drainage basin on Site No. 8 as this particular
segment of the area vas not ploved. particular
SthM!?'-3 Tl7 ShW the effects of a Permanent grass cover as they
Doth retained the snov and had a volume of snovmelt runoff vhich vas
very similar. Although Site No. 1> has snovmelt runoff conslderab^
in excess of the cultivated fields in its vicinity, and because it too
has a permanent grass cover; it vould be expected to have a snovmelt
volume resembling that of Sites No. 3 and 7. This vas true during
1971 but not during 1972. A reason for the lesser volume of snovLlt
runoff for Site No. k for 1972 vas not apparent.
Depending upon veather conditions, snovmelt runoff may occur in one or
tvo day spurts or for a sustained period of several days. During the
second year three separate snovmelt runoff periods of tvo, three, and
four day s duration occurred. Snovmelt runoff came in one period during
ten Consecutive days for the third year. The snov vas usually gone by
the middle of March and may have begun to thav in late January or Feb-
ruary. After the ground thaved, additional snovfall did not produce
surface runoff, but vas absorbed into the soil.
Figure 18 shovs the distribution of rainstorm events based upon the
magnitude of the event. The minimum amount of rain vhich caused run-
off to occur vas 0.1+0 in. The average amount of rain causing runoff
vas 1.31 in. From the figure, this average of 1.31 in. implies that
surface runoff from rainfall can only be expected about five times a
year. The total amount of rainfall in a rainfall event is only one
factor, but certainly rainfall runoff from agricultural lands is not a
59
-------
U4
csx
O UJ
tJL,
Of
C/3 O
UJ U]
>-l
UJ -3
<
UU O
O O
LEGEND' • SITE 8
X SITE 2
RAINFALL PER EVENT, In.
FIGURE 18.-Attribution of rainfall •v«ntt.
60
-------
frequent occurrence for the geographic area studied. It is important
to remember how infrequently runoff occurs if possible control measures
are being considered.
Figure 19 illustrates the different hydrographs obtained -when the storm
intensity vas the main variable. Both runoff events took place on the
same site in the same year so the topography is identical and the crop
cover vas similar. Both storms applied about the same amount of rain
to the drainage area, about 1 in. of rainfall. The event of May 29,
1972 had 0.8? in. of rain, and the event of July26, 1972 had 1.20 in.
of rain. The runoff resulting from both events vas nearly identical,
being vithin about k% of each other.
The hydrographs reflect a change in the ground condition and a differ-
ence in rainfall intensity. The rainfall of May 29 vas a low intensity
storm on a relatively saturated soil which resulted in a long duration
runoff event. The maximum intensity for a 10 minute period vas 0.6
in.hr. This same intensity and rainfall may not have caused runoff if
the ground had been drier. The rainfall event of July 26 had a much
higher intensity and the ground vas considerably drier. The maximum
10 minute intensity for this storm vas 2.7 in./hr. The average rain-
fall intensity for the entire rainfall period vas 0.16 in./hr for the
May 29 event and 1.1 in./hr for the July 26 event.
Differences betveen the tvo events vith respect to the amount of
material which vas vashed off the field is presented. The high inten-
sity storm of July 26 contributed almost 25 times the suspended solids
load of the May 29 event. The impact of the raindrop as it strikes an
unprotected soil is undoubtedly one of the important contributing fac-
tors to soil erosion (8). Pollution may result vhen the dislodged
soil and other materials are vashed off the field in runoff.
The amount of runoff, the total rainfall, and the periods of maximum
rainfall intensity were extremely variable for those storms which
caused runoff, as well as for some similar storms which did not cause
runoff. It is virtually impossible to predict if a rainfall event will
cause runoff when considering only amount and intensity of rainfall.
Other factors such as ground cover, antecedent rain, and condition of
the soil must also be considered.
PHASE I
A complete summary of all the data collected during Phase I can be
found in the appendices* The primary Tcnowledge obtained during this
phase centered around biochemical oxygen demand (BOD), pH, and infor-
mation regarding pollutional constituents in snowmelt runoff.
61
-------
SITE 8, 7/2«/72
Q
HIGH INTENSITY *TORM-2.7 in./hr
TOTAL VOL.* 139,000 GALLONS
TOW, SUSPENDED^ SOLIDS« 6320 LB8,
SUSPENDED SOLIDS* 5450 mg/littr
i i
7 S »
TIME, Hr.
10 II 12 13 14 15
FIGURE
i9.""Hydrogroph comparison of o low intensity ttorm and a high kitsusity
-------
Biochemical Oxygen Demand
Both 5-day and ultimate BOD values were obtained for the runoff sam-
ples collected during Phase I. BOD determinations vere carried out
on frozen samples which had been stored for several months. All sam-
ples were essentially run in duplicate using 100$ and 50$ sample
portions and a Weston - Stack DO probe. The DO was monitored in a
300 ml BOD bottle with the samples being reaerated in the bottle as
required. Settled primary effluent was used to seed the dilution
water. Therefore, the two BOD values obtained for each runoff sample,
one value from a 100J& sample bottle and one value from a 50$ sample
dilution, should give an indication of the effects of seeding on a
previously frozen runoff sample.
Fogarty and Reeder (66) evaluated the BOD results from frozen and fresh
samples, as well as seeded and unseeded samples. Storage periods for
various periods up to six months were utilized. They concluded that
there wasn't any difference between BOD values obtained by using fresh
and frozen samples, as long as the frozen samples were seeded prior to
analysis. A difference in the BOD values for the seeded and unseeded
samples could not be detected for the runoff study being reported here-
in, and the tvo values were simply averaged before data interpretation.
Table 7 shows the important BOD relationships which were obtained from
some selected samples. Complete data for all the runoff samples from
Phase I can be found in Appendix A.
A normal domestic waste is usually considered to have satisfied its
carbonaceous oxygen demand during the first 20 days. The oxygen de-
mand which is exerted during these first 20 days is called the first-
stage or ultimate demand. A normal domestic waste will frequently
have a deoxygenation factor of from 0.1 to 0.2 Referring to Table 7,
it can be seen that the deoxygenation factor was much lower than this.
Therefore, the ultimate BOD was calculated from the oxygen consumed
by the organisms until the oxygen exertion rate became very small.
This time period was usually about 50 days and ranged from U2 to 62
days. The ultimate first-stage BOD values were corrected slightly for
nitrification by using blanks for the same period of time. This
correction for nitrogenous demand was small because of the low ammonia
concentrations present in the runoff samples.
The 5-day BOD values were always quite low with most values at 15 mg/1
or less. The maximum 5-day value obtained was 21 mg/1 from Site No. U
on March 23. The low BOD to COD ratios indicate that much of the or-
ganic matter is not readily available for biological oxidation. This
is probably representative of the cellulose and hemi-celluloses which
would degrade very slowly. The slow degradation is also substantiated
by the low deoxygenation factor.
63
-------
TABLE 7 - Biochemical Oxygen Demand and
Related Factors for Agricultural Runoff
Sample,
Site & Date
1970
U, 3/23
2, U/l
1, V5
3, 3/23
2, 3/23
1, 3/23
1, 5/31
2, 5/31
5-day BOD
mg/1
21
11
17
18
16
13
10
9
Ultimate BOD,
mg/1
75
32
U8
70
M
53
U7
1*2
5-day BOD as a
Percentage of
the Ultimate
BOD. %
28.0
3U.U
35, U
25.7
36.U
2U.5
21.3
21. U
Deoxygenation
Factor
0.03
O.OU
O.OU
0.03
O.OU
0.03
0.02
0.02
COD,
mg/1
131
62
91
105
82
105
1780
980
BOD/COD
0.16
0.18
0.19
0.17
0.19
0.12
0.006
0.01
-------
£H
A laboratory pH meter was used to determine pH on the runoff samples
•which were collected during Phase I. Only minor variations in pH
were detected, all the values ranging from 6,8 to 7.8, and this test
was discontinued after the initial project year.
One interesting aspect of the pH measurements which were taken was the
two narrow ranges of pH which were established for snowmelt and rain-
fall runoff samples. All snowmelt runoff samples had a pH measurement
of between T.1* and 7.8. All rainfall runoff samples had a pH which
measured either 6.8 or 6.9.
Because both of the above pH ranges are near neutral and are accept-
able to most water quality standards, additional research to determine
the exact cause of the pH drop was not conducted. Very few solids
were present in the snowmelt runoff and the slight alkalinity present
is thought to be a result of bicarbonates dissolved in the runoff.
Precipitation which occurred after the ground was thawed promoted
percolation which would allow leaching of the cations. Accumulated
organic matter would gradually decompose and generate organic and
other acids. Hydrogen ions, supplied by these acids, would replace the
leached cations on the topsoil's cation exchange complex. The pH of
the rainfall runoff sample is probably lower than its snowmelt counter-
part because more soil is lost from a rainfall runoff event than from
a snowmelt runoff event.
Snowmelt Runoff
The major portion of the overland runoff from drainage basins in the
geographic area studied originated with the melting of accumulated
snow. The quality of the samples collected on the individual days
was usually quite comparable from day to day even though the quantity
of runoff on a particular day may "be several times that of the pre-
vious day. Consequently, a tentative method of describing the rela-
tive quality of the snowmelt runoff was desired.
Because the snowmelt runoff occurs in a cyclic manner with flow in-
creasing during the warm daylight period and diminishing frequently to
zero flow during the night when below freezing temperatures occur, the
individual samples are representative of discrete runoff events. Con-
sequently, a series of samples collected from a drainage basin can be
considered as a statistical distribution of samples and the various
characteristics of the samples can be similarly grouped. The median
values of the distribution of the various characteristics such as sus-
pended solids, phosphorus, nitrates etc. were selected as being most
representative of the overall quality of the snowmelt.
Table 8 contains the median and range of concentration for the chemical
characteristics of the snowmelt runoff during Phase I. The number of
samples collected and the flow range are also included.
65
-------
TABLE 8 - Concentrations of Characteristics
of Snowmelt Runoff for Phase I
Site No.
Characteristic
No. of Samples
Flow, Hundreds
of Gallons
Total Residue,
ng/1
Suspended Solids
ng/1
Specific Con-
ductance ,
UMHOS/cm 625C
Rav Total Phos-
phorus, ng/1
Soluble Total
Phosphorus (
ng/1
Rav Total KJel-
dahl Nitrogen,
Kg/1
Soluble Total
KJeldahl Nit-
rogen, fflg/1
Nitrate,
ng/1 H
pH
5-day BOD,
fflg/1
Ultimate BOD,
Bg/1
Median
Range
Median
Range
Median
Range
Median
Range
Median
Range
Median
Range
Median
Range
Median
Range
Median
Range
Median
Range
Median
Range
Median
Range
i
6
71
5-136
302
179-890
lUo
38-58U
125
96-172
O.Ul
.27-1,13
0.2U
.12-. 68
2.7
2.5-»». 1
2.1
1.9-3*0
1.1*
0.8-1.9
7.1*
7.W.5
11
9-17
32
27-53
£
5
6U
1-72
187
128-175
38
6-71
130
110-160
0.1*7
.36-.66
0.3l*
.17-39
3.3
2.1*-3.7
2.k
0.7-2.7
0.8
0.3-1.1*
7.1*
7.U-7.5
15
11-16
1»2
32-1*1*
.3
3
56
11-228
225
222-227
31*
33-75
202
153-208
0.60
.l*8-.72
0.17
.08-. 30
5.1
1*. 0-6.1
3.1
3.1-3.5
0.2
0.2-1.2
7.6
7.6-7.7
16
1U-18
59
52-70
1*_
3
2
1-9
230
223-258
21
19-21*
202
178-225
0.91*
.68-1.35
0.62
.37-.T7
5.6
U.5-5.6
2.1
1.2-2.U
0.3
0.1-0.9
7.7
7.6-7.8
18
18-21
69
69-75
66
-------
TABLE 8 - (continued)
Site No.
Ch aracter i st ic
Rav COD,
mg/1
Soluble COD,
mg/1
Median
Range
Median
Range
1
82
56-125
70
U8-93
2
79
62-82
83
62-86
3
103
97-129
121
93-127
U
131
105-131
130
87-1^0
67
-------
From Table 8, it can "be seen that the total residue of the runoff from
the various sites has most generally been between 100 and 300 mg/1.
The suspended solids concentrations vere found to be 75 mg/1 or less in
the snovmelt runoff from all of the drainage basins regardless of crop
cover with the exception of several samples from Site 1. The corn crop
cover on Site 1 prior to the spring of 1970 was probably a contributing
factor to the higher suspended solids concentrations of the snowmelt
runoff from that site for that period.
Vhen the suspended solids concentrations of less than 75 mg/1 for snow-
melt runoff are compared to the suspended solids concentrations found
for rainfall runoff, the solids carrying capacity of the snowmelt
appears somewhat inconsequential. Rainfall runoff events from Site 1
in 1970 ranged up to 15,200 mg/1 of suspended solids. The cyclic nature
of the snowmelt process probably prevents the development of substantial
flows with sufficient velocity to transport large quantities of sedi-
ment. Also the snowmelt runoff is not subjected to an energy input to
dislodge the soil particles similar to that which occurs when raindrops
create a splashing effect during a thunderstorm causing rainfall runoff.
The nutrient properties of the snowmelt runoff were examined by deter-
mining the concentrations of phosphorus and nitrogen of the runoff
samples* The phosphorus concentrations have been determined by two
methods in an attempt to evaluate the relationship of the suspended
solids to the available phosphorus in the runoff. The total phosphorus
values were obtained after digestion with the silt present and the
soluble phosphorus values were determined after the samples were fil-
tered to remove the silt.
From Table 8 it can "be seen that the median values for total phosphorus
ranged from O.Ul to 0,9k mg/1 and soluble phosphorus from 0.17 to 0.62
mg/1 considering all sites. It appears that snowmelt runoff is con-
sistently low in phosphorus concentration, usually below 1.0 mg/1 total
phosphorus and 0.5 mg/1 soluable phosphorus; however, even these con-
centrations are substantially in excess of the 0.1 and 0.01 mg/1 of
organic and inorganic phosphorus levels which have been quoted as
critical values to support nuisance algae growths in lakes. The land
cover or cropping practice on the individual basins did not seem to
have an important influence on the overall phosphorus concentration
of the saowmelt runoff*
The Median KJeldahl nitrogen concentrations of the snowmelt runoff from
the various sites ranged from 2.7 to 5»6 mg/1 for those samples con-
taining suspended solids (raw total KJeldahl nitrogen in Table 8) and
from 2.1 to 3.1 mg/1 after the solids were removed by filtration .
Median nitrate nitrogen concentrations of 0.2 to l.U mg/1 were found.
These concentrations would not appear to be limiting to algal growths
if the runoff reached lakes and other conditions became optimal.
68
-------
Concentrations of other constituents such as COD and BOD are similarly
lov substantiating a general observation that runoff from snowmelt is
of comparatively high chemical quality compared to runoff resulting
from rainfall. Nevertheless, in this area, it vould appear that even
though sediment transport does not appear to be a major problem, the
nutrient concentrations of the snowmelt runoff are sufficient to sup-
port unwanted aquatic growths if the vater is impounded.
PHASE II
A complete listing of the data collected during Phase II can be found
in the appendices* Sampling and laboratory procedures were sufficient-
ly established to allow interpretation and comparison from year to
year and season to season. Investigations during this phase yielded
information regarding the bacteriological quality of runoff water,
the pesticides carried with runoff, and the chemical and physical
characteristics of land surface drainage*
Indicator Organisms in Runoff
Indicator organisms have long been used to designate the bacteriologi-
cal quality of a water or wastewater. Pathogenic bacteria are rarely
isolated routinely, and the sanitary quality of the vater or waste-
water sample is usually based on the test results for certain indicator
organisms.
Routine bacteriological examinations are important mainly in vhat they
imply and not in what they actually determine. The general health im-
plication or possibility of disease transmission is very important.
The degree of health hazard is implied from the routine bacteriological
determinations.
The most common bacteriological test for water and wastewater samples
is the total coliform test* Coliforms are generally the preferred
indicator of fecal contamination in water. The majority of the organ-
isms vhich give a positive coliform test can be grouped into three
species: Esherichia coli, Aerobacter aerogenes« and Aerobacter
cloacae•
Esherichia coli are normally found in the intestinal tract of man and
animals, and they represent about 90% of the coliforms discharged in
fecal matter. Aerobacter aerogenes occur naturally on plants, grain,
and soil; but they may also be found in the feces of man and animals.
Aerobacter cloacae are also found in the feces of man and animals as
veil as "In the soil. The primary disadvantage of the coliform group
as an indicator of fecal contamination is that their presence does not
always indicate fecal contamination but may be the result of other
foreign matter such as grain, plants or soil. This disadvantage is
69
-------
particularly important vhen evaluating bacteriological determinations
from agricultural runoff sample.s.
Because of the aforementioned disadvantage, other bacteriological
indicators of fecal contamination have been proposed. Two of these
indicators are those bacteria belonging to the fecal coliform (FC)
and fecal streptococcus (FS) groups. Fecal coliforms are the members
of the coliform group associated vith the feces of man and animals.
The predominant member of the fecal coliform group is £« eoli. Those
streptococci vhich belong to the fecal streptococcus group are,
according to the 13th edition of "Standard Methods" (62), as follows?
(l) Streptococcus faeealis
(2) Streptococcus faeealis var. liquefaciens
(3) Streptococcus faeealis var. zymogenes
(U) Streptococcus durans
(5) Streptococcus faecium
(6) Streptococcus bovis, and
(7) Streptococcus equinus
A comparison of fecal coliform and fecal streptococcus densities from
the feces of warm blooded animals was made by Geldreich (36). He
found that the fecal coliform to fecal streptococcus ratio (FC/FS) for
all sources other than man vas less than one. It has become apparent
that the use of both fecal coliform and fecal streptococcus indicators
will probably provide more reliable information about the sanitary
quality of a water than information based on total coliform data alone.
Water or wastewater samples vith FC/FS ratios of more than one can be
said to be contaminated by human feces and should be regarded as con-
taining possible pathogenic organisms. Samples with FC/FS ratios of
less than one are said to be contaminated from a nonhuman source, and
the resulting health hazard is less than the hazard resulting from
human sources.
All of the research sites are located in fairly remote areas and human
fecal contamination was not expected. About 10% of the 123 runoff
samples had fecal coliform counts higher than the fecal streptoccus
enumerations. The previous statement regarding the FC/FS ratio being
less than one for a sample from a nonhuman source would appear to have
about a 90/f confidence level. Because of uneven bacterial distribution
in a sample, the precision of the multiple tube fermentation test is
considered to be rather low. A 90/f confidence level is actually quite
high, and should be considered as satisfactory. Even though the
indicator organism counts became quite high at times, it would appear
that the actual health hazard was low.
Most water quality standards specify total coliform and/or fecal col-
iform as parameters to be considered when evaluating surface waters.
Some recommended limits of these indicators are shown in Table 9.
70
-------
TABLE 9 - Recommended Limits of Bacteriological
Indicators in Surface Waters
Beneficial Use
Public Water Supply
(Minimal treatment)
Public Water Supply
(Conventional treatment)
-j Recreation
(Limited Contact)
Recreation
(Primary Contact)
Total Coliform Fecal Coliform
50/100 ml
10,000/100 ml 2,000/100 ml
2,000/100 ml
1,000/100 ml
1,000/100 ml
200/100 ml
2UO/100 ml
Reference
McKee and Wolf (6?)
FWPCA (68)
(67)
(68)
(67)
(68)
(67)
Irrigation
5000/100 ml
1000/100 ml
(67)
-------
A method described in Steel and Torrie (60) vas used to assess the
bacteriological data* The logarithm of the density of the indicator
organism vas regressed on the percent of time that the density vas
equalled or exceeded. The regression lines vere computed for the
various crop covers representing all the samples obtained from runoff
from a particular ground cover* In other vords, data for the fall-
ploved regression lines include results from Site No. 1, 2, 8, and 9
for 1972; vhile data for the pasture regression lines are from samples
from only Site No. 7, but the samples vere collected over a tvo year
period.
Figure 20 displays the effect of different crop covers on the fre-
quency of total coliform counts present in snovmelt runoff. In gen-
eral, the runoff from fields which had minimum cover shoved higher
total coliform densities. The plots of the coliform counts from
those fields vith heavier cover are similar. Fields vith heavier
cover are those vith oats stubble, permanent brome grass and alfalfa,
and permanent pasture; and the regression lines from these fields
exhibit lov coliform densities and have similar slopes. One regression
line has a disparate slope, and the plot for this line vas based upon
data taken from fields vhich vere fall ploved and remained barren
through the vinter.
When comparing the criteria of Table 9 vith the actual data for
Figure 20,it can be seen that the total coliform limits for the
different beneficial uses are often exceeded. Criteria for only the
beneficial use category of domestic vater supply needs to be considered
because the criteria for the beneficial uses of recreation and irri-
gation vould apply only during their respective seasons. It is doubt-
ful that criteria for recreation and irrigation vould apply vhen snov-
melt runoff vas occurring.
The average total coliform density to be generally considered is about
5,000/100 ml, from Table 9. The criteria listed in Table 9 apply only
to stream vater quality. Hovever, it is important to compare these
criteria to the quality of agricultural runoff vater because the runoff
may contribute vater to the stream and the runoff vill affect the
stream vater quality. Referring to Figure 20, this value vas exceeded
about 50JC of the time for those fields vith heavy ground cover and more
than 50JC of the time for those fields vith minimum cover. For the
fields in fallow, the 5,000/100 ml count vas surpassed about 90% of
the time; and for those fields vith corn stubble, this value vas ex-
ceeded 100J6 of the time. Coliform counts vere greater than 20,000/100
ml only about 10JK of the time for the fields vith permanent cover, but
exceeded this value about 75/C to 100% of the time vhen only minimum
ground cover vas maintained,
Snovmelt runoff should be some of the better surface vater vhich vould
be added to a vater course. The fact that snovmelt runoff has total
72
-------
i—i i i i i i—i—i—r
LEGEND:
• OATS STUBBLE
x PASTURE
o CORN STUBBLE
• BROME GRASS
A ALFALFA
•FALL PLOWED
10
I I
I i i i I i I i i i
B 10 20 30 40 SO GO 70 BO 9099tB
PERCENT OF TIME MPN COUNT IS EQUALLED OR EXCEEDED
FIGURE 20. -Total coUfoffBt in snowntGlt runoff.
73
-------
coliform counts which are normally higher than the total coliform
vater quality criteria may not be too important, because the ben-
eficial uses vary for the individual surface vaters. It may mean,
however, that the limits for total coliform organisms should be
reexamined* Because some of the total coliform organisms are commonly
found on plants and the soil, the potential health hazard does not
appear to be adequately reflected in these criteria.
Figure 21 shovs regression lines for fecal coliform counts from snov-
melt runoff for the different crop covers. Some variations betveen
crop covers exist, but no distinct separation is apparent. One factor
which tends to remove any distinction because of crop cover would be
the common practice of pasturing cattle on most of the fields in the
fall* Certainly this practice would be expected to affect the level
of fecal coliforms present in the runoff. Referring to Table 9, it
would appear that the critical level for fecal coliform organisms is
2000/100 ml during the time that snowmelt runoff would occur. The
limits established for recreation and irrigation would not normally
be enforced when snowmelt runoff was occurring. This level was ex-
ceeded less than 10JC of the time for snowmelt runoff from fields which
had oats stubble, corn stubble, or were fall plowed; and it was ex-
ceeded 20% to kQ% of the time for pasture and hay land. The greater
densities from the pasture and hayfields were attributed to the addi-
tional time that cattle were pastured on these sites.
Figure 22 exhibits the regression lines for the frequency of fecal
streptococcus occurrence for the various crop covers for snowmelt
runoff* All the lines have similar slopes and the runoff from the
sites with minimum ground cover show higher densities. The relative
positions of the regression lines with respect to each other are
definite, but reasons for the placements are not apparent.
The densities of total coliform, fecal coliform, and fecal streptococcus
organisms in rainfall runoff from cultivated fields are shown in
Figure 23. The data for the total coliform and fecal streptococcus
plots provide nearly parallel regression lines and some of the data
points are identical* The data for the fecal coliform organisms
result in a plot somewhat lower. This is as expected because the
fecal coliforms represent only a portion of the total coliforms.
Discharges to a watercourse which have total coliform counts of
1,000 to 5,000 per 100 ml and/or fecal coliform levels of 200 to 2,000
per 100 ml can be considered as sources of pollution (See Table 9)*
Figure 23 indicates that these levels are exceeded about 90Jf of the
time, or more, for total coliforms and 65% to 90% of the time when
fecal coliforms are considered. These data suggest that runoff from
agricultural lands may at times be a significant source of pollution
for the surface waters of the state*
-------
i
X
0.
£
2
I
-I
U
III
n.
LEGEND< • OATS STUBBLE
PASTURE
o CORN STUBBLE
• BROKE GRASS
• FALL PLOWED
ft ALFALFA
10
S10 20 30
PERCENT OF TIME MPN COUNT IS EQUALLED Oil
FIGURE 21,-Fscd cofiforms in tnow»«K runoff.
75
-------
i—n—i—i—i
LEGEND* • OATS STUBBLE
* PASTURE
o CORN STUBBLE
• BROME GRASS 8 ALFALf*
FALL PLOWED
PERCENT OF TIME MPN COUNT IS EQUALLED OR EXCEEDED
FIGURE 2 2.-Fecal streptococcus in snow melt runoff.
76
-------
I ' I — I I I I I — I - 1 — I
LEGEND: •
o
•
TOTAL COLIFORMS
FECAL COL I FORMS
FECAL STREPTOCOC-
CUS
10 20 30 4050 6O 70 80 90
PERCENT OF TIME MPN COUNT IS EQUALLED OR EXCEEDED
FIGURE 23.-Bocterioiogjcal indicators in rainfall runoff from cul-
tivated fields.
77
-------
Lack of rainfall runoff from uncultivated fields resulted in an in-
sufficient amount of data on which to base any conclusions. These
data are shown in Table 10 and generally seem to indicate levels of
organisms vhich could cause the vater quality criteria to be exceeded,
TABLE 10. - Bacteriological Data from Rainfall
Runoff on Uncultivated Fields
Site
3
H
7
7
Date
7/28/72
7/28/72
5/29/72
7/28/72
Total Coliform
(MFK/100 ml)
161,000
161,000
5*» ,000
1,610,000
Fecal Coliform
(MPB/100 ml)
790
330
l*,6oo
161,000
Fecal Streptococcus
(MPK/100 ml)
161,000
161,000
17,200
91,800
Summarizing the bacteriological data, it vould appear that runoff
waters have indicator organism counts which frequently exceed the
bacteriological limits established by water quality standards.
Addition of these runoff waters to a watercourse can cause degradation
and be construed as pollution. The actual health hazard has not been
determined, and should be considered in the light of the FC/FS ratios
as well as Just total and fecal coliform densities.
Total coliform densities are usually somewhat greater than the fecal
coliform or fecal streptococcus levels. This relationship would prob-
ably be influenced by the coliform organisms which are found in the
soil and it does not necessarily indicate fecal contamination or the
presence of pathogenic organisms. In the past, it has frequently been
found that polluted water resulting from farm animal discharges give
total coliform and fecal streptococcus densities of the same order of
magnitude (69).
It would appear that the current water quality standards should recog-
nize the quality of agricultural runoff as providing a basis for
water quality improvement. Agricultural runoff is a nonpoint source of
vater pollution; and consequently, it is virtually uncontrollable.
Criteria which establish water quality parameters at levels less than
the levels present in agricultural runoff or other non point sources
may be unrealistic. Perhaps a better approach vould be to establish
the water quality criteria at levels which reflect the quality of
78
-------
streams from agricultural areas and then attempt to improve the quality
of the runoff. Subsequent sampling vould establish any improvements
in agricultural runoff quality vhich could be obtained.
Pesticides
Concern over degradation of the environment by pesticides has developed
in recent years. Organochlorine insecticides have "been outlawed in
Hungary. Other countries, including the United States, have expressed
their concern by restricting or banning certain pesticides.
The vide acceptance and usage of some insecticides and herbicides
has resulted in their becoming virtually ubiquitous. Determinations for
these pesticides are often positive even vhen there has been no knovn
usage of the chemicals in the area. Some investigations report the
travel of pesticides in the atmosphere and also the finding of these
compounds in rainfall (51).
The United States Public Health Service conducted a study to determine
the distribution of chlorinated hydrocarbon insecticides in the major
drainage basins of the United States and collected vater samples for
analysis from kl states. They found widespread distribution of
dieldrin, endrin, DDT, and DDE (58). Samples in the Brookings County,
S. Dak. study vere also investigated for these four insecticides
which the USPHS has found to be widespread. In addition, analyses vere
made for aldrin, lindane, heptachlor, heptachlor epoxide, DDD, atrazine,
and methoxychlor.
Because most of these pesticides are relatively insoluble in vater, it
vas quite likely that they might be associated vith the soil particles
in runoff. Consequently, sediment or mud samples vere investigated
vhen samples could be obtained. Only small amounts of soil vere pre-
sent in snovmelt runoff. Therefore, only filtered runoff samples vere
used for pesticide determinations during snovmelt. If enough soil vas
vashed from a site vhen rainfall runoff occurred, mud samples as veil
as filtered vater samples vere examined.
Data for all the pesticide samples vere evaluated together vithout
any designation regarding the site from vhich the sample vas obtained.
The results of the pesticide analyses are shovn in Tables 11 and 12.
In general, the level of all the pesticides present-in the runoff
seems to be quite lev, and the majority of the concentrations vere
below the analytical test limits.
Table 11 presents the findings for 75 filtered samples. These 75
samples include the filtered snovmelt samples and the filtered rain-
fall runoff samples. The vast majority of the data are below the
analytical test limits and all values are less than one part per billion
(ppb). These results agree vith the USPHS study vhich reported that
79
-------
TABLE 11. - Number of Samples Grouped in
Ranges of Pesticide Concentrations for Filtered Bunoff Samples
Concentration
(ppb)
Aldrin
DDT
DDE
ODD
Atrazine, Diedrin,
Lindane, Heptachlor,
Heptachlor Epoxide
Endrin, Methoxychlor
Below limits11
Snovmelt
RainfaU
.05 - .09
Snovmelt
Rainfan
.10 - .30
Snovmelt
Rainfan
.31 - .50
Snovmelt
Rainfan
.51 - .75
Snovmelt
Rainfan
2U
0
1
0
0
0
0
0
1
uu
2U
Does
not
apply*
k
2
1
0
0
0
15
0
3
1
8
0
0
0
0
UU
25
Does
not ^
apply
5
1
0
0
0
0
U9
26
0
0
0
0
0
0
0
0
Total
75
75
75
75
75
Analytical test limits:
0.05 ppb for Aldrin, DDE, Dieldrin, Lindane,
Heptochlor, Heptachlor Epoxide, and Endrin.
0.10 ppb for DDT, DDD, Atrazine and
Methoxychlor.
80
-------
all the pesticide concentrations were less than one ppb (58). A dis-
tribution of the pesticide concentrations for the sediment samples is
shovn in Table 12. A total of 2k samples vere collected, and the ma-
jority of the samples had concentrations of pesticides which were less
than the analytical test limits*
Chemical and Physical Characteristics
The basic project objectives were to find the concentrations of the
runoff's constituents, and also to determine the total contribution
of these constituents on an annual basis. The chemical and physical
characteristics of runoff are related to the drainage area, the nature
of the runoff and the land management practices.
Two broad categories of runoff were observed. They resulted from snow-
melt runoff and rainfall runoff. Runoff from snowmelt generally had a
low suspended solids content, often less than 20 mg/1, and exhibited a
characteristic yellow-tan color. Snowmelt runoff typically started
slowly and reached a peak during the early afternoon hours when the
effect of the sun's rays was most pronounced.
Rainfall runoff generally had a much higher suspended solids content
than snowmelt runoff, sometimes reaching the several thousand mg/1
range. It too showed color; but usually not of the same intensity
as snowmelt runoff, particularly if allowed to settle. Punoff result-
ing from rainfall tended to be a more violent event of much shorter
duration. The runoff hydrograph may have reached its peak only a few
minutes after runoff began. Runoff then continued to taper off until
the event was finished.
The dissolved material present was fairly uniform as measured by the
specific conductance determination. Values ranged from 69 to 352
umhos/cm for the two year study. Median specific conductance for
snowmelt runoff was 12U ymhos/cm. The range of specific conductance
for rainfall runoff was wider being ^5 to 538 umhos/cm. The median
specific conductance value was 210 pmhos/cm. A general observation
would be that the quality of rainfall runoff as measured by specific
conductance, is more variable than snowmelt runoff.
The mean concentrations of some runoff parameters are presented in
Table 13. These are averages of concentrations and are not weighed
with respect to now. Sites No. 1, 2, 8 and 9 were combined to give
an indication of the runoff characteristics from cultivated land, Site
Ho. 7 represents the pasture, and Sites No. 3 and U yielded data for
the land use category of alfalfa and brome grass. Mean concentrations
of all the runoff parameters are broken down by site in the Appendix.
Note the prominent differences between rainfall and snowmelt runoff
data. In all cases, the number of snowmelt events exceeds the number
81
-------
TABLE 12. - Number of Samples Grouped in
Ranges of Pesticide Concentrations for Sediment Samples
Concentration
(ppb Dry Wt.)
Belov
«
limits"
5.0 -
10 -
16 -
26 -
51 -
76-
100 -
150 -
200 -
250 -
•7
Total
9.9
15
25
50
75
100
150
200
250
300
300
Aldrin
11
2
3
0
1
1
3
0
1
0
2
0
2U
DDT
20
*
1
2
1
0
0
0
0
0
0
0
7
Dieldrin, DDD, Lin-
dane, Heptachlor,
Heptachlor Epoxide
DDE Atrazine Endrin, Methoxychlor
15
2
3
3
0
0
0
0
1
0
0
0
2U
23
0
0
0
0
0
0
0
0
0
0
1
7
2h
0
0
0
0
0
0
0
0
0
0
0
21*
Analytical test limits:
5.0 ppb for Aldrin, DDE, Dieldrin, Lindane,
Heptachlor, Heptachlor Epoxide, and Endrin.
10.0 ppb for DDT, DDD, and Methoxychlor
100 ppb for Atrazine.
82
-------
TABLE 13. - Mean Concentrations of Runoff
Parameters by Land Use
Parameter
Cultivated Land
Snov Rain
Total Residue, 187
mg/1
Suspended
Solids, mg/1
Total
Phosphorus,
mg/1
Nitrate,
mg/lU
COD, mg/1
No. Samples
12Ul
51 1021
O.U1* 1.05
1.0
Total KJeldahl 2.1
Nitrogen mg/lK
UU
Crop Cover
Pasture
Alfalfa & Brome Grass
Snov Rain Snov
150
18
222
38
131*
1*2
0.67 0.1*9 0.1*3
Rain
108
1*0
0.35
1.5
2.6
1U8
28
0.9
3.3
69
16
0.1*
1.7
-9
2
0.8
2.8
62
31
0.3
0.8
22
2
83
-------
of rainfall events* Fev rainfall runoff events occurred on unculti-
vated lands. For these sites, the uncultivated fields averaged less
than one rainfall runoff event per site per year.
Another notable comparison.between rainfall and snovmelt runoff is
the difference in the concentration level which occurred for some of
the parameters. This difference is particularly true vith respect to
solids data. A substantial change in magnitude is apparent when the
mean suspended solids for cultivated land during rainfall runoff is
compared to its counterpart for snovmelt runoff, for example. Com-
paring the average mean values of 51 mg/1 of suspended solids with
1,021 mg/1 reveals about a 20 fold increase for rainfall runoff.
The magnitude of the suspended solids variation varied from use to
use, but this increase was especially conspicuous for the cultivated
fields. Apparently, runoff from rainfall has better quality if it
results from permanent grassland.
The uncultivated sites (Sites No. 3, fc, and 7) show relatively little
change in parameter concentrations when comparing snovmelt to rainfall
runoff. In fact, some of the mean concentrations are much less for
snovmelt. The nitrogen and phosphorus levels for uncultivated sites
tended to be higher for snowmelt events than for rainfall runoff. Just
the opposite was true for the cultivated sites* Freezing and thawing
help rupture organic molecules vhich produces material more easily
dissolved or carried in the runoff. Because more organic material was
available on the noncultivated sites, they often had snovmelt nutrient
levels higher than the cultivated fields. Rainwater does not remain
on the fields as long as water resulting from melted snow. Any in-
crease in nutrient levels for rainfall runoff was probably associated
with suspended matter washed from the fields. This would account for
the increase in nutrients from rainfall runoff for the cultivated
lands.
Nutrient levels are of interest because of the continuing emphasis on
eutrophication problems. Levels of nitrogen and phosphorus which have
often been quoted as causing nuisance algal growths in lakes are 0.01
mg/1 of inorganic phosphorus and 0.3 mg/1 of inorganic nitrogen. It
is believed that these particular levels were first quoted by Sawyer
(21). Total phosphorus was measured during this study instead of in-
organic phosphorus because in a lake's ecosystem there is a continuing
conversion from one phosphorus form to another. Even organic phosphorus
which is bound up in the bottom sediments of a lake may later be used
to increase biological activity (70). In a recent article, Loehr also
agrees with this concept as he states that: "The phosphorus of concern
to environmental quality is associated with that adsorbed on sediment
and with the interchange of phosphorus from bottom deposits in bodies
of waters with the upper waters11 (71). This same reasoning applies to
the nitrogen values although some differentiation in nitrogen forms
was made.
-------
All mean concentrations of nitrate alone equal or exceed the 0.3 mg/1
level, sometimes reaching 1.5 mg/1. Adding ammonia to the nitrate
amplifies the inorganic nitrogen contribution of runoff vaters.
Ammonia had a nearly constant level, varying only from 0.1 to 0,^
mg/1 H for all rainfall events. Ammonia vas not measured for snovmelt
runoff, but vas included with organic nitrogen in the total kjeldahl
nitrogen determination.
Phosphorus data show concentrations many times in excess of 0,01 mg/1,
Even soluble phosphorus levels in snowmelt runoff were in the 0.2 to
0.3 mg/1 P range and consequently, practices aimed at retaining the
soil, such as sediment traps, would not greatly decrease the phosphorus
load discharged from a field.
The above nitrogen and phosphorus levels suggest that agricultural run-
off may be an important contributing factor to lake eutrophication.
Further expenditures for advanced treatment for municipal and industrial
wastewater plants may be superfluous if a significant reduction in lake
eutrophication is the intention of such expenditures and agricultural
runoff is not controlled.
Table 1^ gives the total yearly contributions of the runoff constituents
in lb/acre/yr» The annual soil loss for each site did not exceed 1/2
ton/acre/yr, (See Appendix). The average annual loss for cultivated
land was less than 300 Ib/acre/yr, This is in sharp contrast to some
long-term reported values of Missouri and Ohio studies. Reported cul-
tivated field soil losses from a small plot study in Missouri were
from 2.78 to 19.72 ton/acre/yr (8). Soil losses from the 1.5 acre
fields in Ohio averaged 2.16 ton/acre/yr. The maximum annual loss
occurred on corn fields at a rate of 7.70 ton/acre/yr (5).
Agricultural experts seek to establish soil conservation practices to
allow a tolerable erosion of less than 3 to k ton/acre/yr (72). This
would seem to indicate that the erosion rate is satisfactory for the
sites under consideration. Yet most of the cultivated sites exhibit
some of the common signs associated with soil erosion such as sub-soil
protrusion on the ridges, silting in road ditches, some gullying, and
rock outcrops. The land owner of two of the cultivated sites is
presently considering a permanent grass cover to decrease his soil
losses.
Climatic factors undoubtedly provide at least a partial explanation for
the differences in soil losses between this study and the two studies
mentioned above. Both Ohio and Missouri annually average about twice
the precipitation and about ten times the runoff as the research sites
evaluated in this study (73). Differences in rainfall intensity,
amount of snowmelt runoff, soil type, slope, and fanning practices
would be considerations. Also, the studies may not be comparable be-
cause of the size of the research areas being investigated. Small 90 ft
85
-------
TABLE lU, - Yearly Runoff Contributions
of Runoff Parameters by Land Use
Parameter Crop Cover
(Ib/aere/yr) Cultivated Land Pasture Alfalfa & Brome Grass
Total Residue
Suspended Solids
Total Phosphorus
Nitrate - N
Total KJeldahl
298
255
0.27
0.33
0.81
51.9
10.5
0.22
0.36
1.00
28.9
3.6
0.09
0.21
0.65
Nitrogen
COD 1»3 25 12
long plots and 1.5 acre areas may tend to measure soil movement in-
stead of actual soil losses.
A pictorial summary of all the runoff for both years of Phase II is
expressed in Figure 2k. The number beneath each pie chart, vith the
exception of the area chart, represents the total quantity contributed
during the tvo year study. For example, 23, 687 lb of suspended solids
vas vashed off the seven sites in both years.
The total runoff area vas divided into cultivated land, pasture, and
alfalfa and brome grass. Sites No. 1, 2, 8, and 9 were under culti-
vation and the summation of their areas vas 56.3% of the total re-
search area. In addition, Site No. 7 vas the pastured area, and Sites
No. 3 and No. U comprised the alfalfa and brome grass segment.
Some interesting observations vere made. One of the most striking vas
the relationship betveen rainfall runoff and its total contribution.
While rainfall runoff comprised only 32.2J& of the total runoff, it vas
responsible for 93.7# of the suspended matter and 6l.8% of the COD lost
in the runoff. Thus, almost all of the soil loss vas caused by only
about one-third of the runoff.
86
-------
56.3% / 19.7%
CULTIVATED/ PASTURE
38.2%T •*
SNOW X,
SOLUBLE
24.0%
ALFALFA
ft BftOME
67.8%
SNOWMELT
78.8 ACRES
4945 Ibt. COO
/ Sl.7% RAIN \
5.88 MG RUNOFF
v /
29,213 Ibi. TOTAL
RESIDUE
44.9% \ 'SOLUBLE
SNOWMELAl
47.9 IbS. NITRATE
1276 Ibs. TOTAL KJELDAHL
NITROGEN
33.93 Ibt. PHOSPHORUS
FIGURE 24.-
Breokdown of total runoff contributions
for 1971 and 1972.
87
-------
It vould appear that pollution from agricultural runoff would be
effectively reduced if rainfall runoff was eliminated, or if complete
sedimentation occurred. Yet, the nutrient contributions indicate Just
the opposite. Snovnelt runoff accounted for 65.8£ of the total kjeldahl
nitrogen, 62.2% of the nitrate, and UU.9J& of the phosphorus lost during
the two years. Important quantities of nutrients would still be lost
annually even if all rainfall runoff was eliminated.
A large proportion of the nutrients was found to be soluble. All of
the nitrate, 69% of the total kjeldahl nitrogen, and 27.5# of the phos-
phorus were independent of any sediment. Soil conservation practices
could be implemented to hold soil losses to some acceptable minimum,
but such practices could not be construed as limiting nutrients as
well.
Soluble Fraction
The data in Table 15 represent the mean percentages of chemical oxygen
demand, phosphorus, or total kjeldahl nitrogen which were soluble.
For example, 75.1? of the COD contributed from Site No. 1 during the
snowmelt runoff events of 1971 vas soluble. During 1971 rainfall run-
off was recorded only on Site No. 8, and snowmelt runoff did not occur
on Site Ho. 9 during 1972.
As a general observation, most of the three constituents are soluble
for snowmelt runoff, and a substantial percentage of rainfall runoff
was also soluble. However, the amount of soluble components diminishes
for rainfall runoff. In fact, most of the data for the cultivated
lands (Site No. 1, 2, 8 and 9) indicate that major portions of the
constituents were associated with the sediment. The sites which are
permanently covered with grass continued to have a high percentage
of the three parameters remaining soluble.
Soil - Losses
This study is unique in that published data from other studies which
evaluated soil-losses in surface runoff from melting snow could not
be found. The total soil-losses were much less than expected, espec-
ially when considering the universal soil-loss equation. The present
"universal" soil loss equation did not appear to be suitable for esti-
mating the potential of water pollution from agricultural lands for
the geographic area studied. Results from future studies in other
areas with dissimilar climatic conditions will be required before the
adequacy of the universal soil-loss equation for larger areas can be
evaluated.
-------
TABLE 15. - Soluble Fraction of COD, Phosphorus and TKN of
Snovmelt and Rainfall Runoff from Research Sites
Mean Percentage
Snovmelt - 1971
Site
1
2
3
It
7
8
9
COD*
75.1
75.1
67.3
72.0
67.8
81.8
82.1
P*
72.8
69.9
1*3,1
U8«6
50.0
62.2
66.»»
TKN*
75.9
82.8
72.2
79.5
77.6
89.1*
98.2
Mean Percentage Mean Percentage
Snowmelt - 1972 Rainfall - 1971
COD
90.9
1*8.1
5M
52.6
61.3
55.5
mm
P
97.
72.
31*.
27.
52.
U7.
™*
7
1*
2
2
0
5
TKN COD P TKN
100.0 ...
82.1 .
67.1 ...
67.0 ...
70.2 ...
72.8 37.9 21.1 31.1
. . . / .
Mean Percentage
Rainfall - 1972
COD
U8.8
7.1»
72.7
90.9
63.9
26.7
1*6.2
p
56
12
53
67
61*
21*
1*3
.1
.6
.3
.5
.1
.9
.3
TKN
69.5
16.3
87.5
87.5
73.3
1*1.9
57.9
* COD - Chemical oxygen demand
P « Phosphorus
TKN - Total kjeldahl nitrogen
-------
SECTION IX
REFERENCES
1. Wadleigh, C. H. , "Wastes in Relation to Agriculture and Forestry."
U. S. Department of Agriculture Pub. No. 1065, Washington, D. C.
(1968).
2. Eliassen, R., and Tchobanoglous, G«, "Removal of Nitrogen and
Phosphorus*" Proc. 23rd Ind. Waste Conf . « Purdue Univ. Ext.
Ser. 132. 35 (ipSff).
3. Benson, R. D.t "The Quality of Surface Runoff from a Farmland
Area in South Dakota During 1969." M. S. Thesis, S. D. State
Univ., Brook ings, S. D. (1970).
U. McCarl, T. A., "Quality and Quantity of Surface Runoff From A
Cropland Area in South Dakota During 19TO." M. S. Thesis, S. D.
State Univ., Brookings, S. D. (1971).
5. Weidner, R. B. , e£ al. , "Rural Runoff as a Factor in Stream
Pollution," Jour. Water Poll. Control Fed.. Hi, 377 (1969).
6. Weibel, J. R. , e£ al,, "Urban Land Runoff as a Factor in Stream
Pollution. Jour. Water Poll. Control Fed.. 36. 915* (196U).
7. Walker, K. C. , and Wadleigh, C. H. , "Water Pollution from Land
Runoff." Plant Food Rev. . jU, 2 (1968).
8. Buckman, H. 0., and Brady, N. C., "The Nature and Properties of
Soils." Macmillan Co., Nev York, N. Y., (1969).
9. McGuinness, J. L. e£ al., "Some Effects of Land Use and Treatment
on Small Single Crop Watersheds." Jour. Soil Water Conserv..
, 65 (I960).
10. Holt, R. P., "Runoff and Sediment as Nutrient Sources," Presented
at 1969 Annual Meeting of Minnesota Chapter Soil Conservation
Society of America, Bui. No. 13, Univ. of Minn., Minneapolis,
Minn. (1969).
90
-------
11. Timmons, D. R,, e£ al,t "Loss of Crop Nutrients Through
Runoff." Minn. Sci.» 2^, 16 (1968).
12. Dragoun, F. J.t and Miller, C. R., "Sediment Characteristics of
Tvo Small Agricultural Watersheds." Trans. Amer. Soe. Agric.
Engr. £, 66 (1966).
13. "Management of Nutrients on Agricultural Land for Improved Water
Quality." Dept. of Agronomy, Cornell Uhiv., Ithaca, N. Y.t
EPA Rept. No. 13020 DPB 08/71, Washington, D. C. (1971).
lU. Ellison. W. D,, "Studies of Raindrop Erosion," Agrie. Eng.« 25,
131 and 181 (19M).
15. Epstein, E., and Grant, W. J., "Soil Losses and Crust Formation as
Related to Some Soil Physical Properties." Soil Science Soe.
Amer. Proc.. 31, 5^7 (1967).
16. Rogers, J. S., et al., "An Evaluation of Factors Affecting Runoff
and Soil Loss From Simulated Rainfall." Trans. Amer. Soe. Agric.
Engr.. £, 1*57 (196U).
17. Wischmeier, W. H., and Smith, C. C., "Predicting Rainfall-Erosion
Losses from Cropland East of the Rocky Mountains." Soil and
Water Cons. Res. Div., Agric. Res. Serv., U. S. Dept. of
Agric., Washington, D. C. (1965).
18. "Sources of Nitrogen and Phosphorus in Water Supplies" Task
Group Report, Jour. Amer. Water Works Assn.. g?. 3kb (1967).
19. Baumann, E. R.t and Kelman, S., "Sources and Contributions of Nit-
rogen and Phosphorus to an lova Stream." Presented at 26th
Annual Purdue Ind. Waste Conf., Purdue Univ. (1971).
20. Baumann, E. R., and Kelman, S., "Effects of Surface Runoff on the
Feasibility of Municipal Advanced Waste Treatment" In "Agricul-
tural Practices and Water Quality." T. L. Willrich and G. E.
Smith, (Eds.), lova State Univ. Press, Ames, 3kk (1970).
21, Savyer, C. N., "Fertilization of Lakes by Agricultural and Urban
Drainage." Jour. Nev. En£. Water Works Assn.. 61, 109 (19U7).
22. Slyvester, R. 0., Nutrient Content of Drainage Water From Forested,
Urban, and Agricultural Areas." In "Algae and Metropolitan Wastes,"
U. S. Dept. of Health, Educ. and Welfare, Pub. No. SEC-TR-W61-3.
Cincinnati, Ohio, (1961).
91
-------
23. Engelbrecht , R, C. , and Morgan, J. J., "Land Drainage as a Source
of Phosphorus in Illinois Surface Waters." In "Algae and
Metropolitan Wastes," U. S. Dept. of Health, Educ. and
Welfare, Pub. No. SEC-TR-W61-3, Cincinnati, Ohio, (I96l).
2 U. Holt, R« £., e£ al . "Accumulation of Phosphates in Water."
Agric. and Food Chem.. 18, 78l (1970).
25. Timmons, D. R., et^al., "Leaching of Crop Residues as a Source of
Nutrients in Surface Runoff Water." Water Resources Res., 6A
1367 (1970).
26. Campbell, F. R., and Webber, L. R., "Contribution of Range Land
Runoff to Lake Eutrophication." Presented at 5th Intern. Water
Poll. Res. Conf., San Francisco, Calif. (1970).
27* Johnston, W. R, , et^ al, "Nitrogen and Phosphorus in Tile Drainage
Effluent." Soil Sci. Soc. Amer., Proc., 29., 287 (1965).
28. "Role of Animal Wastes in Agricultural Land Runoff." Dept. of Biol.
and Agric. Eng. , N. C. State Univ. at Raleigh, Raleigh, N. C. ,
EPA Rept. No. 13020 DGX 08/71, Washington, D. C. (1971).
29. Wang, W. L. and Evans, R. L. , "Nutrients and Quality in Impounded
Water." Jour. Amer. Water Works Assn. . 62, 510 (1970).
30. Javorski, N. and Hetling, L. J., "Relative Contributions of
Nutrients to the Potomac River Basin from Various Sources."
In "Relationship of Agriculture to Soil and Water Pollution."
Conf. of Agric* Waste Management, Cornell Univ., Ithaca, N. Y.,
(1970).
31. Witzel, S. A., e£al., "Surface Runoff and Nutrient Losses of
Fennimore Watersheds." Trans. Amer. Soc. Agric. Engr., 12,
338 (1969).
32. Minshall, N. E., e£ al. "Stream Enrichment From Farm Operations.
Jour. San. Eng. Div.« Proc. Amer. Soc. Civil Engr. . 96. 513
T1970).
33. Grizxard, T. J., and Jennelle, E. M., "Will Wastevater Treatment
Stop Eutrophication of Impoundments." Presented at 27th Annual
Purdue Ind. Waste Conf., Purdue Univ., (1972).
3U. Middaugh, P. R. , e£ al . "Differentiation of Ruminant From Non-
Ruminant Fecal Sources of Water Pollution by Use of Enteric
Bacteria.11 Proc. Intern. Symp. Livestock Wastes, Amer. Soc. Agri.
Eng., St. Joseph, Mich., 126 (1971).
92
-------
35. Cooper, K. E.f and Ramadan, F. M., "Studies in the Diffentiation
Between Human and Animal Pollution," Jour, Gen. Microbiol..
12, 180 (1955).
36. Geldreich, E. E., "Sanitary Significance of Fecal Coliforms in the
Environment." Pub. Ho. WP-20-3, Water Poll. Control Res. Ser.
U. S. Dept. of the Int., Washington, D. C. (1966).
37. Evans, F. L., e£ al., "Treatment of Urban Stormwater Runoff,"
Jour. Water Poll. Control Fed.. U0_, R162 (1968).
38. Walter, W. F. and Bottman, R. P., "Microbiological and Chemical
Studies of an Open and Closed Watershed." Jour. Environ»
Health. 30, 157 (1967).
39. Kunkle, S. H., "Concentrations and Cycles of Bacterial Indicators
in Farm Surface Runoff." In "Relationship of Agriculture to Soil
and Water Pollution," Conf. on Agric. Waste Management,
Cornell, Univ., Ithaca, N. Y., U9 (1970).
UO. Kunkle, S. H., and Meiman, J. R., "Water Quality of Mountain Water-
sheds." Hydrology Paper No. 21, Colorado State Univ., Ft. Collins,
Colo. (1967).
4l. Kunkle, S. H., and Meiman, J. R., "Sampling Bacteria in a Mountain
Stream." Hydrology Paper No. 28, Colorado State Univ., Ft.
Collins, Colo. (1968).
J»2. Claudon, D. G«, jet^ aL., "Prolonged Salmonella Contamination of a
Recreational Lake by Runoff Waters." Appl. Microbiol.. 21,
875 (1971).
1*3. Mahan, J. N., e_t al., "The Pesticide Review - 1968." Agric. Stab.
and Conserv. Serv., USDA, Washington, D. C. (1968).
M. .Nicholson, H. P., and Hill, D. W., "Pesticide Contaminants in Water
and Mud and Their Environmental Impact." In Relationship of
Agriculture to Soil and Water Pollution." Conf. on Agric. Waste
Management, Cornell Univ., Ithaca, N. Y., 171 (1970).
1»5. Nicholson, H. P., "The Pesticide Burden in Water and Its Signifi-
cance." In "Agric. Practices and Water Quality." T. L. Willrich
and G. E. Smith (Eds.) Iowa State Univ. Press, Ames, 183 (1970).
U6. Nicholson, H. P. "Pesticide Pollution Control." Science. 158. 871
(1967).
1»7. Nicholson. H. P., "Pesticides: A Current Water Quality Problem."
Trans. Kansas Acad. Sci.t 70, 39 (1968).
93
-------
1(8. Thoman, J. R., and Nicholson, H. P., Pesticides and Water Quality."
Presented 2nd San. Eng. Conf., Vanderbilt Univ., Nashville,
Tenn. (1963).
1*9. Grzenda, A. R., et al., "DDT Residues in Mountain Stream Water as
Influenced by Treatment Practices." Jour. Econ. Ent., 57,
615 (1961*).
50. Greichus, Y. A., "Importance of Agricultural Biocides in Water
Pollution." In "South Dakota Agriculture and Water Quality."
Symp. on Water Poll., S. D. State Univ., Brookings, S. D., 26
(1970).
51. Weibel, S. R., et al., "Pesticides and Other Contaminants in Rain-
fall and RunoffT11^ Jour. Amer. Water Works Assn., 58, 1075 (1966).
52. Lichtenstein, E. P., "Fate and Movement of Insecticides in and from
Soils." In "Pesticides in the soil: Ecology, Degradation &
Movement." Intern. Symp. on Pesticides in the Soil, Michigan
State Univ., 101 (1970).
53. McCarty, P. L., and King, P. H., "The Movement of Pesticides in
Soils." Proc. 21st Ind. Waste Conf.. Purdue Univ., Ext. Ser.
121. 156 1356*6).
5U. White, J. L., and Mortland, M. M., "Pesticides by Soil Minerals."
In "Pesticides in the Soil: Ecology, Degradation & Movement,"
Intern. Symp. on Pesticides in the Soil, Michigan State Univ.,
95 (1970).
55* Huang, J., and Liao, C., "Adsorption of Pesticides by Clay Minerals."
Jour. San. Eng. Div., Proc. Amer. Soc. Civil Engr., 96, 1057
(1970).
56. Bailey, G. W., et al., "Adsorption of Organic Herbicides by
Montmorillonite: Role of pH and Chemical Character of Adsorbate."
Soil Sci. Soc. Amer. Proc., 32, 222 (1968)
57. Lauer, G. J., et al., "Pesticide Contamination of Surface Waters
by Sugar Cane Farming in Louisiana." Trans. Amer. Fisheries Soc.,
25_, 310 (1966).
58. Weaver, L., et al., "Chlorinated Hydrocarbon Pesticides in Major
U.S. River Basins." Pub. Health Repts.. 80, U8l (1965).
59. "Field Manual for Research in Agricultural Hydrology," Agric. Hdbk.
No. 22It, Agric. Res. Ser., USDA, Washington, D. C., (1962).
60. Steel, R. G. D., and Torrie, J. H., "Principles and Procedures of
Statistics." McGrav-Hill Cook Co., Inc., New York, N. Y. (I960).
-------
6l. "Standard Methods for the Examination of Water and Wastewater."
12th Ed., Amer. Pub. Health Assn., New York, N. Y., (1965).
62. "Standard Methods for the Examination of Water and Wastevater."
13th Ed. , Amer. Pub. Health Assn., Washington, D. C. (1971) •
63. Breidenbach, A. W. , e_t al. , "The Identification and Measurement of
Chlorinated Hydrocarbon Pesticides in Surface Waters." Fed.
Water Poll. Control Adm. , Washington, D. C. , (1966).
6U. "Methods for Chemical Analysis of Water and Wastes 1971."
Envirn. Prot. Agency, Cinn. , Ohio (1971).
65. "Climatological Summary No. 3 for Brookings, South Dakota."
State Climatologist for South Dakota, South Dakota State Univ. ,
Brookings, S. D. , (1969).
66. Fogarty, W. J., and Reeder, M. E. , "BOD Data Retrieval Through
Frozen Storage." Public Works, ££, No- 3» 88 (196H).
67. McKee, J. E. , and Wolf, H. W. , "Water Quality Criteria." 2nd Ed.,
Pub. No. 3-A, Calif. State Water Quality Control Board, Sacramento,
Calif. (1963).
68. "A Report of the Committee on Water Quality Criteria." Fed. Water
Poll. Control Adm., U. S. Dept. of Interior, Washington, D. C.
(1968).
69. Geldrich, E. E. , and Kenner, B. A., "Concepts of Fecal Streptococci
in Stream Pollution." Jour. Water Poll. Control Fed ., Hi, R336
(1969).
70. Clark, J. W. , ejt al . , "Water Supply and Pollution Control."
International Textbook Co., Scranton, Pa. (1971).
71. Loehr, R. C. , "Agricultural Runoff -Characteristics and Control."
Jour. San. Eng. Div. , Proc. Amer. Soc. Civil Engr. , 98, 909
71972).
72. Martin, W. P., et al. , "Fertilizer Management for Pollution Control."
In "Agricultural Practices and Water Quality." T. L. Willrich and
G. E. Smith (Eds.). Iowa State Univ. Press, Ames, 1^2 (1970).
73. Miller, D. W., et al . , "Water Atlas of the United States." Water
Info. Center, Port Washington, N. Y. , (1963).
95
-------
SECTION X
APPENDICES
Page
A. Chemical and Physical Quality Data 97
B. Bacteriological Quality Data 123
C. Pesticide Data 137
D. Mean Concentrations of Parameters lU3
£. Annual Contributions
-------
APPENDIX A
Chemical and Physical Quality Data
97
-------
Site No. 1
Area » 7.18 acres
Cover: Corn stubble and oats
Table Al. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings County, South Dakota - 1970
Characteristic*
Volume (gal)
pH- units
Specific Conductance
(y rohos/cw)
oa
Suspended Solids
Total Residue
5-day BOD
COD Rav
Soluble
Total KJeldahl Rav
Nitrogen Soluble
Total Phosphorus Rav
as P Soluble
Nitrate Rav
as N Soluble
Mar 23
1,000
7.U
125
ll+O
302
12
105
70
3.8
2.5
0.68
0.36
l.U
1.6
Mar 31
550
7.1*
150
38
179
9
56
1*8
2.5
2.0
0.27
0.18
1.8
2.1
Apr 1
7,100
7.5
125
21*2
293
10
73
60
2.5
2.2
o.i+o
0.21+
1.3
1.6
Apr 2
9.700
7.1*
100
281*
1+18
10
82
53
3.0
1-9
0.1+1
0.12
1.9
2.0
Apr 1*
13,600
T.U
96
58U
890
11
125
71*
2.7
2.1
0.95
0.20
0.8
0.8
Apr 5
600
7-5
170
92
315
17
91
93
l+.l
3.0
1.13
0.68
1.5
1.5
May 31
29,300
6.9
1*6
15,200
22,100
9
1,780
25
U-5
1.1
1+.30
0.35
2.1
2.1
June 15B
-
-
-
3,330
U.810
-
610
—
2.6
0.6
2.20
0.08
0.6
0.6
-------
Table Al (continued)
Characteristic*
Ammonia
Raw
Soluble
Total Nitrogen
Mar 23
0.75
1.17
5.2
Mar
0.
0.
*
31
96
96
•3
Apr 1
0.75
1.00
3.8
Apr
0.
1.
*
2
77
20
•9
Apr b
0.58
1.35
3.5
Apr 5
0.98
1.75
5.6
May 31
0.8U
1.26
6.6
June 15B
0.29
3.2
* All concentrations in milligrams per liter except as noted
vo
-------
Site No. 2
Area « 8.77 acres
Cover: corn stxib"ble and oats
Table A2. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brooklngs, South Dakota - 1970
Characteristic*
Volume (gal)
pH-units
Specific Conductance
H *\i mhos/cm)
8
Suspended Solids
Total Residue
5-day BOD
COD
Total KJeldahl
Nitrogen
Rav
Soluble
Rav
Soluble
Total Phosphorus Rav
as P Soluble
Nitrate
as N
Rav
Soluble
Mar 23
700
7.U
11*6
38
187
16
82
3.7
2.7
0.66
0.3l»
1.1*
1.2
Mar 31
100
7.1*
160
6
168
15
79
81*
3.3
0.7
0.1*2
0.39
1.0
1.1
Apr 1
6,1*00
7.1*
130
17
128
11
62
86
2.8
2.6
0.36
0.27
0.8
1.2
Apr 2
7,200
7.1*
110
38
275
-
80
62
2.1*
2.2
0.1*7
0.17
0.3
1.7
Apr i*
6,900
7.5
111*
71
195
16
75
83
3.1*
2.1*
0.60
0.36
0.6
0.6
May 31
57,500
6.9
1*8
7,500
8,610
8
980
38
U.8
1.0
0.5!*
0.07
1.8
1.7
June 15A
725
6.8
27
1,200
1,360
7
202
23
1.5
0.9
1.17
0.18
0.7
0.6
-------
Tatle A2 (continued)
Characteristic*
Ammonia Raw
Soluble
Total Nitrogen
Mar 23
0.87
1.25
5.1
Mar 31 Apr 1
1.02 0.72
1.11 1.27
1*.3 3.6
Apr 2
0.51*
1.35
2.7
Apr 1*
0.69
1.67
i*.o
May 31
0.91
0.92
6.5
June 15A
0.92
O.U7
2.1
*A11 concentrations in milligrams per liter except as noted.
-------
Site No. 3
Area - 10.12 acres
Cover: Brome grass & alfalfa
Table A3. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings, South Dakota - 1970
Characteristic*
Volume (gal)
pH-units
Specific Conductance
(vi mhos /cm)
Suspended Solids
Total Residue
5-day BOD
COD Haw
Soluble
Total KJeldahl Rav
Nitrogen Soluble
Total Phosphorus Rav
as P Soluble
Nitrate Rav
as N Soluble
Mar 23
5,600
7.6
208
31
227
18
129
121
6.1
3.5
0.72
0.08
1.2
1.2
Apr 1
22,800
7.7
202
3l4
225
16
97
127
5.1
3.1
0.60
0.30
0.2
0.3
Apr 2
1,100
7.6
153
75
222
1U
103
93
lt.0
3.1
0.1*8
0.17
0.2
0.3
-------
M
O
Table A3 (continued)
Characteristic*
Ammonia Rav
Soluble
Total Nitrogen
Mar 23
1.81*
2.57
7.3
Apr 1
1.53
1.10
5.3
Apr 2
1.09
1.87
U.2
*A11 concentrations in milligrams per liter except as noted.
-------
M
O
Site No. U
Area * 8.77 acres
Cover: Brome grass & alfalfa
Table
Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings, South Dakota - 1970
Characteristic*
Volume (gal)
pH-units
Specific Conductance
(u mhos /cm)
Suspended Solids
Total Residue
5-day BOD
COD
Total KJeldahl
Nitrogen
Total Phosphorus
as P
Nitrate
as N
Mar 23
120
7.6
202.
21
223
21
Raw 131
Soluble 130
Raw 5.6
Soluble 2.U
Raw 0.9U
Soluble 0.62
Raw 0.9
Soluble 1.1
Aw 1
180
7.8
225
21*
258
18
131
5.6
1.2
0.68
0.37
0.3
0.1*
Apr 1*
900
7-7
178
19
230
-
105
87
2.1
1.35
0.77
0.1
0.2
-------
Table AU (continued)
Characteristic*
Ammonia
Total Nitrogen
Mar 23
Rav 1.37
Soluble 1.16
-
Apr 1 Apr k
1.52 1.16
1.12 1.26
- -
*A11 concentrations in milligrams per liter except as noted.
-------
o
o\
Site No. 1
Area * 7«l8 acres
Cover: Oat stubble and oats
Table A5. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings County, South Dakota - 1971
Characteristic* Feb. 16
Volume (gal) 1
Specific
Conductance
(y mhos/cm @ 25°C)
Suspended Solids
Total residue
COD Raw
Soluble
Total kjeldahl
nitrogen Raw
Soluble
Total phosphorus
as P Rav
Soluble
Nitrate as N
Total nitrogen
,900
190
77
220
51
25
l.U
1.1
0.98
0.82
1.0
2.U
Feb 17
9,100
170
25
. 186
3H
26
2.0
1.1
0.55
O.U8
0.6
2.6
Feb 18
2,UOO
179
11
171
32
32
1.0
1.0
0.39
0.38
0.5
1.5
Feb 25
7,200
106
18
132
28
28
0.9
0.8
0.28
0.21
0.5
1.3
Feb 26
6,800
107
5
102
26
13
0.9
0.5
0.25
0.18
0.3
1.2
Mar 11
6,200
97
27
93
23
32
1.3
1.2
0.28
0.17
0.5
1.8
Mar 12
18,200
107
90
202
Ul*
2l»
1.7
1.0
O.Ul
0.26
0.3
2.0
Mar 13
6,300
103
157
260
38
12
0.9
0.7
o.6l
0.26
0.2
1.1
*A11 concentrations in milligrams per liter except as noted
-------
H»
O
Site Ho. 2
Area = 8.77 acres
Cover: Oat stubble and oats
Table A6. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings County, South Dakota - 1971
Character! st ic*
Volume (gal)
Specific conductance
(y mhos/cm @ 25°C)
Suspended solids
Total residue
COD Raw
Soluble
Total kjeldahl
nitrogen Raw
Soluble
Total phosphorus
as P Raw
Soluble
Nitrate as N
Total nitrogen
Feb 16
2,000
192
63
232
51
27
1.6
1.6
0.93
0.70
1.0
2.6
Feb 17
7,900
11*8
1*5
188
36
27
1.2
1.1*
0.1*0
0.30
0.1*
1.6
Feb 18
2,100
151*
17
ll*9
36
30
1.1
1.0
0.30
0.27
0.3
1.1*
Feb 25
6,100
87
11*
136
28
25
1.3
0.6
0.22
0.13
0.3
1.6
Feb 26
5,000
90
5
87
20
15
1.0
0.6
0.18
0.09
o.J*
1.1*
*A11 concentrations in milligrams per liter except as noted.
-------
Site No. 3
Area • 10.12 acres
Cover: Brome grass & alfalfa
Table A7. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings County, South Dakota - 1971
Characteristic*
Volume (gal)
Specific conductance
(Umhos/cm g 25°C)
H Suspended solids
o
oo
Total residue
COD Rav
Soluble
Total kjeldahl
nitrogen Rav
Soluble
Total phosphorus
as P Rav
Soluble
Nitrate as N
Total nitrogen
Peb 16
11*1*, 000
99
17
129
31*
23
1.5
1.5
0.20
0.13
1.0
2.5
Peb 17
80,300
129
17
130
1*3
36
2.1*
1.6
0.37
0.20
0.8
3.2
Feb 18
2,300
177
17
155
63
1*0
2.6
1.9
0.55
0.31
1.2
3.8
Peb 25
37,800
82
6
92
52
Ul
1.9
1.5
0.27
0.12
o.i*
2.3
Peb 26
3,100
122
5
131
1*1*
1*1
1.7
1.3
0.28
0.10
o.i*
2.1
Mar 10
30,500
107
19
110
78
1*8
3.3
2.3
0.63
0.22
0.3
3.6
Mar 11
33,100
117
25
11*5
85
53
i*.o
2.5
0.77
0.18
o.i*
l*.l*
Mar 12
25,000
117
23
130
70
35
3.6
1.1*
0.66
0.21
0.5
l*.l
Mar 13
8,000
136
21
102
38
17
1.2
1.0
0.52
0.22
0.2
1.1*
*A11 concentrations in milligrams per liter except as noted.
-------
H
O
Site No. U
Area « 8.77 acres
Cover: Brome grass & alfalfa
Table A8. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings County, South Dakota - 1971
Characteristic* Feb
Volume (gal) 51,
Specific Conductance
(vimhos/cm @ 25°C)
Suspended solids
Total residue
COD Raw
Soluble
Total kjeldahl
nitrogen Raw
Soluble
Total phosphorus
as P Raw 0
Soluble 0
Nitrate as H
Total nitrogen
Ib
000
98
18
105
1+3
28
1.6
1.6
.21
.13
1.1
2.7
Feb 17
Ul,000
125
19
126
1+3
3k
1.9
1.8
0.36
0.17
1,0
2.9
Feb 18
2,500
178
16
161
63
38
2.5
1.6
0.1+2
0.23
1.1
3.6
Feb 25
19,200
69
7
87
36
33
1.3
1.0
0.18
0.07
0.6
1.9
Feb 26
19,900
9l»
3
156
3lt
30
1.3
1.2
0.16
0.09
0.6
1.9
Mar 10
W.MO
107
11
110
72
1+9
2.8
2.1+
0.52
0.26
O.U
3.2
Mar 11
1+7,600
101
17
109
68
53
3.5
2.5
0.72
0.23
0.1+
3.9
Mar 12
1+5,100
112
18
116
66
33
3.5
1.6
0.62
0.22
0.1+
3.9
Mar 13
11,800
115
11
91*
28
19
1.3
1.1
0.1+6
0.28
0.1
' 1.1+
*A11 concentrations in milligrams per liter except as noted
-------
Site No. 7
Area * 15.51 acres
Cover: Grassland - pastures
Table A9. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings County, South Dakota - 1971
Char acteri s t i cw
Volume (gal)
Feb 16
118,000
Feb 17
117,000
Feb 18
6,200
Feb 25
1*6,600
Feb 26
15,600
Mar 10
19,800
Mar 11
127,000
Mar 12
90 ,000
Mar 13
9,800
Specific conductance
(pmhOB/cm % 25°C)
Suspended solids
P
° Total residue
COD Raw
Soluble
Total kjeldahl
nitrogen Raw
Soluble
Total phosphorus
as P Rav
Soluble
Nitrate as N
Total nitrogen
121
37
157
55
39
2.3
2.3
0.36
0.27
1.2
3.5
111*
17
113
53
36
2.7
2.2
0.1*0
0.20
1.0
3.7
252
23
2U8
121
76
5.6
5.6
0.97
0.1*1
2.1*
8.0
115
18
119
62
»»5
2.5
1.6
0.53
0.26
0.6
3.1
116
1*
120
U8
1*1*
1.9
1.8
0.31
0.21
0.6
2.5
151
17
1*3
7*
50
3.1
2.1*
0.72
0.3U
0.1*
3.5
137
23
127
76
55
3.6
2.5
0.86
0.33
0.1*
i*.o
1U8
28
ll*8
86
1*1
J».9
2.0
l.OU
0.38
0.1*
5.3
229
23
208
85
1*8
l».l
2.9
1.1*3
0.63
0.3
l*.l*
•All concentrations in milligrams per liter except as noted.
-------
Site No. 8
Area = 18.68 acres
Cover: Corn stubble and oats
Table MO. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings County, South Dakota - 1971
Character! s t i c*
Volume (gal)
Feb 16
131,000
Feb 17
137,000
Feb 18
3,200
Feb 25
50,800
Feb 26
33,300
Mar 10
13,000
Mar 11
232 ,000
Mar 12
218,000
Mar 1,3
52,600
Specific Conductance
(vmhos/cm g 25°C)
H* Suspended solids
M
Total residue
COD Rav
Soluble
Total kjeldahl
nitrogen Rav
Soluble
Total phosphorus
as P Rav
Soluble
Nitrate as N
Total nitrogen
117
73
185
53
35
2.0
2.0
0.51
0.35
0.9
2.9
127
32
lUo
Uo
31
2.2
2.0
0.39
0.32
0.9
3.1
209
75
203
61
38
2.1*
2.0
0.56
0.50
1.6
i*.o
120
13
11*3
56
U5
1.7
1.3
0.30
0.18
0.8
2.5
111
2
82
l*l*
1*2
1.7
1.8
0.21
0.19
0.7
2.H
121
7
137
83
72
1*.6
3.6
O.U6
0.27
0.6
5.2
113
1*1
153
6U
55
3.0
2.1*
O.Ul
0.16
0.5
3.5
121
58
150
76
70
1.7
2.9
0.1*0
0.05
0.5
2.2
176
ll*
153
1*0
36
1.7
1.7
0.3»*
0.20
0.6
2.3
-------
ro
Characteristic*
Volume (gal)
Specific conductance
iynhos/cm g 25°C)
Suspended solids
Total residue
COD Rav
Soluble
Total kjeldahl
nitrogen Rav
Soluble
Total phosphorus
as P Rav
Soluble
Nitrate as N
Ammonia as N
Total nitrogen
Table A10. - Continued
1,500
127
675
1,110
9*
27
3.*
0.9
1.32
0.23
2.6
0.2
6.0
Tune 29
3UO
138
1*30
770
70
33
2.8
1.0
1.21
0.30
2.6
0.3
5.U
•All concentrations in milligrams per liter except as noted.
-------
U)
Site No. 9
Area « 9.79 acres
Cover: Corn stubble and oats
Table All. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings County, South Dakota - 1971
Characteristic*
Volume (gal)
Specific conductance
(Vimhos/cm % 25°C)
Suspended solids
Total residue
COD Raw
Soluble
Total kjeldahl
nitrogen Raw
Soluble
Total phosphorus
as P Raw
Soluble
Nitrate as N
Total nitrogen
Feb 16
50 ,000
157
27
205
67
55
2.0
2.1*
0.58
0.1*8
0.6
2.6
Feb 17
177,000
128
21*
125
57
37
2.5
2.5
0.1*1*
0.1*3
0.9
3.1*
Feb 18
1,1,00
189
16
155
1*5
35
2.2
2.0
0*.33
1.1
3.3
Feb 25
12>0
130
16
122
66
52
2.5
2.1
0.1*3
0.31
0.7
3.2
Feb 26
3,900
120
3
ll*0
1*2
1*1*
2.0
2.0
0.27
0.21
0.7
2.7
Mar 10
6,900
HO
11
135
93
71
|l C
|i C
0.56
0.39
0.5
5.0
Mar 11
71* ,800
152
30
155
86
69
i*.o
3.8
0.60
0.33
0.7
M
Mar 12
93,800
120
1*0
127
1*6
39
1.7
1.6
0.1*5
0.08
0.1*
2.1
Mar 13
19,700
161*
12
127
38
31*
1.7
1.7
0.32
0.17
0.2
1.9
*A11 concentrations in milligrams per liter except as noted.
-------
Site Ho. 1
Area * 7.18 acres
Cover: Plowed and oats
Table A12. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings County, South Dakota - 1972
Characteristic* Mar 11
Volume (gal)
Specific conductance
(umhos/cm % 25°C)
Suspended solids
*" Total residue
COD Raw
Soluble
Total kjeldahl
nitrogen Raw
Soluble
Total phosphorus
as P Raw
Soluble
Nitrate as N
Ammonia as N
Total nitrogen
170
115
12
107
22
20
1.3
1.3
O.M»
0.1*3
0.9
2.2
May 22
1,1*00
358
208
1*60
75
H5
2.1
1.5
0.78
0.58
0.2
0.2
2.3
May 28 AM
2,600
502
3»*
378
55
30
1.1»
1.3
0.1*9
0.37
0.2
0.2
1.6
May 28 PM
1,600
538
51
1*16
1*5
21*
1.3
1.1
0.38
0.35
0.1*
0.2
1.7
May 29
2,200
1*1*1*
28
31*6
31*
21*
1.2
0.9
0.26
0.09
1.5
0.1
2.7
July 28
51,800
136
6,700
6,730
660
30
3.8
- 0.9
3.50
O.ll*
0.8
0.2
1*.6
*A11 concentrations in milligrams per liter except as noted.
-------
Site No. 2
Area * 8.77 acres
Cover: Ploved and oats
Table A13. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings County, South Dakota - 1972
Characteristics*
Volume (gal)
Specific conductance
(ymhos/cm g 25°C)
Suspended solids
Total residue
COD Rav
Soluble
Total kjeldahl
nitrogen Rav
Soluble
Total phosphorus
as P Rav
Soluble
Nitrate as N
Ammonia as N
Total nitrogen
Mar 11
170
107
19
103
6
3
1.1
0.9
0.56
0.55
O.U
1.5
Mar 13
590
12%
28
162
31*
25
1.9
1.7
0.62
0.59
1.6
3.5
Mar 15
600
120
59
1U5
2U
5
1.2
0.9
0.1+2
0.10
1.*
2.6
May 29
5,000
97
1,360
1,560
221
22
7.0
1.3
2.11
0.1*2
1.8
0.1.
8.8
July 28
103,000
U5
2,91*0
2,950
328
16
It. 3
0.6
2,1*9
0.13
O.U
0.1
U.7
All concentrations in milligrams per liter except as noted.
-------
Site No. 3
Area « 10.12 acres
Cover: Brome grass & alfalfa
Table All*. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings County, South Dakota - 1972
Characteristic* Mar 7
Volume (gal) 257,000
Specific conductance
(pmhos/cm % 25°C) 108
Suspended solids
Total residue
COD Rav
Soluble
Total kjeldahl
nitrogen Rav
Soluble
Total phosphorus
as P Rav
Soluble
Nitrate as N
Ammonia as N
Total nitrogen
6
119
51
50
3.3
2.6
0.28
O.ll*
1.1*
M
Mar 10
16,1*00
116
13
129
62
25
3.6
2.6
0.33
0.19
1.3
1*.9
Mar 11
1*8,300
86
16
96
37
28
2.3
1.9
0.32
0.05
0.7
3.0
Mar 12
6,600
ITU
22
2l*9
150
76
5.6
3.9
0.86
0.06
1.8
T.k
Mar 13
23,600
159
28
212
123
65
l*.l*
3.2
0.91
O.lU
1.3
5.7
Mar 11*
1,700
137
22
11*8
63
27
3.5
1.7
0.83
0.27
0.5
U.O
Mar 15
1,1*00
93
1*1*
111*
21*
5
1.1
0.5
0.36
0.22
0.2
1.3
July 28
27,800
73
61*
lUl
22
16
0.8
0.7
0.30
0.16
0.3
0.2
1.1
•All concentrations in milligrams per liter except as noted.
-------
Site No. U
Area * 8.77 acres
Cover: Brome grass & alfalfa
Table A15. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings County, South Dakota - 1972
Characteristic*
Volume (gal)
Mar 7
1*0,300
Specific conductance
(umhos/cm g 25°C) 100
Suspended solids
Total Residue
COD Rav
Soluble
Total kjeldahl
nitrogen Rav
Soluble
Total phosphorus
as P Rav
Soluble
Nitrate as N
Ammonia as N
Total nitrogen
12
121
U9
UU
3.0
2.2
0.3U
0.08
1.1
U.I
Mar 10
10,700
109
13
130
56
21
2.9
2.2
0.31
0.22
1.0
3.9
Mar 11
21,UOO
92
16
108
U5
30
2.U
1.9
0.37
0.07
0.7
3.1
Mar 12
530
181
18
190
127
72
U.6
3.1
0.71*
o.lU
1.3
5.9
Mar 13
U.300
165
33
229
139
67
5.2
3.1
1.08
O.ll*
l.U
6.6
Mar 15
950
81
5U
116
30
5
1.5
0.7
0.56
0.10
0.2
1.7
July 28
1,900
60
17
75
22
20
0.8
0.7
o.Uo
0.27
0.3
0.2
1.1
*A11 concentrations in milligrams per liter except as noted.
-------
00
Site No. 7
Area » 15.51 acres
Cover: Grassland - Pasture
Table Al6. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings County, South Dakota - 1972
Characteristic* Mar 7
Volume (gal) 200
Specific conductance
(unhoB/cm % 25°C)
Suspended solids
Total Residue
COD Raw
Soluble
Total kjeldahl
nitrogen Raw
Soluble
Total phosphorus
as P Raw
Soluble
Nitrate as N
Ammonia as N
Total nitrogen
,000
130
3
126
53
3.2
2.5
0.30
0.31
2.0
5.2
Mar 10
38,500
131
5
139
63
33
3.3
2.6
0.36
0.18
1.1*
»».7
Mar 11
228,000
97
18
101
1*5
29
2.5
1.9
0.1*5
0.17
0.8
3.3
Mar 12
39,200
157
10
161
67
1*6
3.3
2.2
0.69
0.25
0.7
i*.o
Mar 13
91, **00
163
2U
198
no
63
4.7
3.6
0.90
0.44
1.0
5.7
Mar 14
18, too
117
18
11*1*
73
31
3.9
2.1
0.85
0.31
0.1*
4.3
Mar 15
2,600
IT7
15
147
39
23
2.6
1.6
0.62
0.32
0.6
3.2
May 29
1*60
1*23
15
319
38
27
1.6
1.1
0.1*8
0.37
0.3
0.2
1.9
July 28
295,000
79
60
125
60
31*
1.8
1.1*
0.1*9
0.25
0.4
o.i*
2.2
*A11 concentrations in milligrams per liter except as noted.
-------
Site Ho. 8
Area • 18.68 acres
Cover: Plowed and veeds
Table AIT. - Chemical and Physical Quality Characteristics of Agricultural Bunoff
Brookings County, South Dakota - 1972
Characteristic*
Volume (gal)
Mar 6
1*,1*00
Specific conductance
(ymhos/cm 6 25°C) 167
Suspended solids
Total residue
COD Raw
Soluble
Total kjeldahl
nitrogen Raw
Soluble
Total phosphorus
as P Raw
Soluble
Nitrate as N
Total nitrogen
152
332
59
37
3.2
2.1
0.58
0.25
2.1*
5.6
Mar 8
650
225
10
190
37
22
2.8
1.6
0.32
0.23
1*.3
7.1
Mar 9
2,300
190
21
183
33
11
U.3
1.6
o.3l*
0.20
2.9
7.2
Mar 10
36 ,800
152
278
711
129
35
3.8
2.3
0.83
0.25
1.6
5.U
Mar 11
11*7,000
136
1*63
560
105
30
2.5
2.0
o!l5
1.0
3.5
Mar 12
37,000
169
36
190
50
35
2.0
1.8
0.35
O.ll*
1.6
3.6
Mar 13
81,000
166
106
280
65
35
2.1*
2.2
0.31*
0.17
1.9
1*.3
Mar 11*
29,700
221*
18
211
1*1
33
2.1*
2.1
0.39
0.18
2.6
5.0
Mar 15
1*,000
352
7
258
31
26
2.1
1.8
0.28
0.18
l*.l*
6.5
-------
Table AIT. - Continued
Characteristic* May 1
Volume (gal) 7
Specific conductance
(umhos/cm § 25°C)
Suspended solids
Total residue
COD Rav
Soluble
Total kjeldahl
nitrogen Rav
Soluble
Total phosphorus
as P Rav
Soluble
Nitrate as N
Ammonia as N
Total nitrogen
,200
170
270
51*0
1*1*
1U
1.6
0.6
0.7l*
0.22
1.5
0.1
3.1
May 12
180
201*
220
578
55
5
1.3
0.9
1.07
0.25
1.2
0.1
2.5
May 22
1,800
206
192
1(22
1*1*
25
2.0
1.1
0.52
0.28
3.0
0.2
5.0
May 23
5,100
1*60
51*
1*01
53
2.0
1.1*
0.35
0.27
1*.3
0.2
6.3
May 21*
22,300
511*
32
1*32
1*6
31*
1.6
1.1*
0.2U
0.16
2.7
0.1
U.3
May 28
1*3,1*00
1*29
609
911*
115
20
3.1
1.0
0.76
0.13
1.2
0.2
U.3
May 29
131* ,000
376
232
550
67
1,8
0.8
0.1*2
0.12
1.1
0.1
2.9
-------
Table AIT. - Continued
ro
Characteristic*
Volume (gal)
Specific conductance
(umhos/cm @25°C)
Suspended solids
Total residue
COD Rav
Soluble
Total kjeldahl
nitrogen Paw
Soluble
Total phosphorus
as P Rav
Soluble
Nitrate as N
Ammonia as N
Total nitrogen
June 13
1*50,000
219
1,066
1,290
152
19
2.8
0.9
1.17
0.08
1.8
0.1
U.6
June 19
12U.OOO
280
1,1+30
1,610
182
13
3.1+
0.9
1.58
O.ll*
1.2
0.2
U.6
July 8
55,300
181
1,360
1,1*60
176
1+
2. It
0.9
0.93
O.Ql*
2.0
0.2
i*.i+
July 11+
1,900
213
1,803
1,605
258
27
5.8
1.1
1.60
0.06
3.3
0.1
9.1
July 21
99,100
281*
2,1+10
2,530
362
18
5.6
0.6
2.06
0.03
2.0
0.1
7.6
July 26
139,000
165
5,1*50
5,330
77U
18
1+.9
1.1
2.55
0.05
1.8
0.1
6.7
*A11 concentrations in milligrams per liter except as noted.
-------
Site No. 9
Area * 18.68 acres
Cover: Plowed and veeds
Table Al8. - Chemical and Physical Quality Characteristics of Agricultural Runoff
Brookings County, South Dakota - 1972
Characteristic*
Volume (gal)
May 1
2,200
Specific conductance
(umhos/cm 6 25°C) 139
Suspended solids
fo
w Total residue
COD Rav
Soluble
Total kjeldahl
nitrogen Rav
Soluble
Total phosphorus
as P Rav
Soluble
Nitrate as N
Ammonia as N
Total nitrogen
1*0
290
21
1*
0.9
0.3
0.36
0.19
o.i*
0.0
1.3
May 29
50,600
319
»»5
290
31*
16
1.2
0.9
0.21*
0.16
1.7
0.1
2.9
June 16
183,000
110
577
690
79
13
1.7
0.6
0.82
0.07
0.1
0.1
1.8
June 19
1*3,800
272
90
3M
23
17
l.U
0.9
0.37
0.11
0.3
0.1
1.7
June 21
170
253
105
385
3U
21
1.1*
0.9
0.69
0.1*2
2.5
0.3
3.9
July 26
35,700
199
182
372
1*1*
26
1.2
0.9
0.51
0.21
0.1*
0.2
1.6
*A11 concentrations in milligrams per liter except as noted.
-------
APPENDIX B
Bacteriological Quality Data
123
-------
Site No. 1
Area: 7.18 acres
Cover: Corn stubble and oats
Table Bl. - Bacteriological Quality Characteristics of Agricultural
Runoff, Brookings County, South Dakota
Date
Snowmelt
19TO
March 23
March 31
April 1
April 2
April 1*
April 5
Rainfall
1970
May 31
June 15B
Total Coliform
(MPN/100 ml)
2U7.000
32,1*00
22,000
1,910,000
12,500
13,600
79,000
3,300
Fecal Coliform
(MFff/100 ml)
7,370
1,680
1,800
7,600
36
15
7,900
790
Fecal Streptococcus
(MPN/100 ml)
35,800
230,000
18,200
16U.OOO
8,U80
57,500
1*90,000
13,000
121*
-------
Site Ho. 2
Area =8.77 acres
Cover Corn Stubble and oats
Table B2. - Bacteriological Quality Characteristics of Agricultural
Runoff, Brookings County, South Dakota
Date
Snowmelt
1970
March 23
March 31
April 1
April 2
April k
Rainfall
1970
May 31
June 15A
Total Coliform
(MPlf/100 ml)
71,100
28,000
1,517,000
208,000
1»,680
79,000
1*90,000
Fecal Coliform
(MPN/100 ml)
^,990
2,300
12,900
11,700
32
13,000
23,000
Fecal Streptococcus
(MPU/100 ml)
25,000
26,000
19,600
57,800
2,1*90
33,000
k9 ,000
125
-------
Site No. 3
Area = 10,12 acres
Cover: Brome Grass and alfalfa
Table B3. - Bacteriological Quality Characteristics of Agricultural
Runoff* Brookings County, South Dakota
Date
Snovmelt
1212.
March 23
April 1
April 2
Total Coliform
(MPH/100 ml)
23,000
6,530
27,600
Fecal Coliform
(MFW/100 ml)
3,100
2,ll»0
6,390
Fecal Streptococcus
(MPN/100 ml)
27,600
6,390
10,700
Site Ho. U
Area =8.77 acres
Brome Grass and alfalfa
Table Bfc. - Bacteriological Quality Characteristics of Agricultural
Runoff, Brookings County, South Dakota
Date
Snovmelt
1970
March 23
April 1
April k
Total Coliform
(MFH/100 ml)
5,600
U.550
2,300
Fecal Coliform
(MPN/100 ml)
2,050
3,082
15
Fecal Streptococcus
(MPH/100 ml)
51,000
l,8Uo
1,920
126
-------
Site No. 1
Area - 7.18 acres
Cover: Oat stubble and oats - '71
Plowed and corn - *72
Table B5. - Bacteriological Quality Characteristics of Agricultural
Runoff, Brookings County, South Dakota
Date
Snovmelt
19J1
February 16
February 17
February 18
February 25
February 26
Mar 11
March 12
March 13
Snovmelt
-| f\+Tf\
1972
March 11
Rainfall
1972
May 22
May 28 AM
May 28 PM
May 29
July 28
Total Coliform
(MPN/100 ml)
16,100
1,200
1,300
1,300
585
16,100
> 2*1,000
> 2^,000
5,^20
13,000
5*»,200
3,300
3,100
10,900
Fecal Coliform
(MPN/100 ml)
80
230
170
< 100
< 100
5
33
23
630
1*90
5^,200
330
1,090
33,000
Fecal Streptococcus
(MPN/100 ml)
220
1,720
330
1,300
230
2,UOO
790
1*90
H.900
5^,200
10,900
U.900
1,700
172,000
127
-------
Site No. 2
Area =8.77 acres
Cover: Oat Stubble and oats - '71
Plowed and corn - *72
Table B6. - Bacteriological Quality Characteristics of Agricultural
Runoff, Brookings County, South Dakota
Date
Snowmelt
1QJ1
February 16
February 17
February 18
February 25
February 26
Snowmelt
1972
March 11
March 13
March 15
Rainfall
1972
May 29
July 28
Total Coliform
(MEW/100 ml)
2,700
> Uoo
1,720
2,U90
715
330
3,300
2U,000
13,000
3^8,000
Fecal Coliform
(MPH/100 ml)
20
50
130
50
50
l»0
1,300
1,720
J|,900
8,000
Fecal Streptococcus
(MPN/100 ml)
790
330
130
870
50
10,900
2,300
5^,200
33,000
1*9,000
128
-------
Site No. 3
Area = 10.12 acres
Cover: Brome grass and alfalfa
Table B7. - Bacteriological Quality Characteristics of Agricultural
Runoff, Brookings County, South Dakota
Date Total Coliform
(MFN/100 ml)
Snowmelt
19J1
February 16
February 17
February 18
February 25
February 26
March 10
March 11
March 12
March 13
Snowmelt
1972
March 7
March 10
March 11
March 12
March 13
March I1*
March 15
2,210
3.U80
16,100
700
2,300
1,700
3,100
1,1*00
7,900
9,200
1,1*00
92,000
92,000
22,100
790
U90
Fecal Coliform
(MPN/100 ml)
1,720
3.H80
16,100
600
650
330
— -
1,1*10
330
21*0
500
330
u,6oo
700
no
0
Fecal Streptococcui
(MPN/100 ml)
2,UOO
2,1*00
2,210
2,700
2,300
2,200
2^,000
1,700
i.Uoo
1,600
3U,800
10 ,900
17,200
22,100
1,720
1,300
129
-------
Table B7. - Continued
Date Total Coliform Fecal Coliform Fecal Streptococcus
(MPN/100 ml) (MFH/100 ml) (MFH/100 ml)
Rainfall
1972
July~28 161,000 790 161,000
130
-------
Site No. 1*
Area = 8.77 acres
Cover: Brome grass and alfalfa
Table B8. - Bacteriological Quality Characteristics of Agricultural
Runoff, Brookings County, South Dakota
Date Total Coliform
(MPN/100 ml)
Snowmelt
19J1
February 16
February 17
February 18
February 25
February 26
March 10
March 11
March 12
March 13
Snovmelt
1972
March 7
March 10
March 11
March 12
March 13
March 15
Rainfall
1972
July 28
2,1*00
1,090
1,300
2,390
I,1* 50
3,1*80
2, 1*00
1,090
1,720
2,780
800
13,000
ll»,100
5U, 200
700
161,000
Fecal Coliform
(MPN/100 ml)
790
1*90
•1.300
595
330
1*90
51*2
1*90
3U8
2UO
200
330
10 ,900
700
0
300
Fecal Streptococcus
(MPN/100 ml)
3,300
i*,6oo
3,1*80
1,700
1,700
1*,900
3,300
1*,900
790
> 1,600
10 ,900
3,300
1*,900
It, 900
1*,900
161,000
131
-------
Site No. 7
Area = 15.51 acres
Cover: Grassland - pasture
Table B9. - Bacteriological Quality Characteristics of Agricultural
Runoff Brookings County, South Dakota
Date Total Coliform
(MPH/100 ml)
Snowmelt
1971
February 16
February 17
February 18
February 25
February 26
March 10
March 11
March 12
March 13
Snovmelt
1972
March 7
March 10
March 11
March 12
March 13
March lU
March 15
1,720
1,300
3^,800
U65
2,300
2,300
7,000
2,200
U,6oo
> 16,000
fc,900
27,800
10,900
7,000
7,000
1,720
Fecal Coliform
(MRf/100 ml)
1,090
1,300
9,180
280
395
1*90
k6
330
5, if 20
1,300
1,720
3,300
I* ,900
330
Uo
Fecal Streptococcus
(MPR/100 ml)
220
2,300
3.U80
1»50
1,800
700
2,300
2.UOO
900
920
2,700
1,750
U.900
7,000
2,300
^90
132
-------
Table B9. - Continued
Date Total Colifora Fecal Coliform Fecal Streptococcus
(MFK/100 ml) (MPK/100 ml) (MPH/100 ml)
Ralnfan
1972
May 29 5^,200 U,600 17,200
July 28 161,000 161,000 91,800
133
-------
Site No. 8
Area = 18.68 acres
Cover: Corn stubble and oats - '71
Plowed and idle acres - '72
Table BIO. - Bacteriological Quality Characteristics of Agricultural
Runoff, Brookings County, South Dakota
Date
Snovmelt
T f\*n
1971
February 16
February 17
February 18
February 25
February 26
March 10
March 11
March 12
March 13
Rainfall
1971
June 8
June 29
Snovmelt
March 6
March 8
March 9
March 10
Total Coliform
(MPN/100 ml)
> 16,090
> 16,090
>160,900
765,000
730,000
1*00,000
1,300,000
> 21*0,000
2,100,000
130,000
3U.800
> 16,000
160,000
> 160,000
> 160,000
Fecal Coliform
(MPN/100 ml)
130
50
1,1*10
< 100,000
< 100,000
20
31
33
79
17,200
U.900
U9
80
20
0
Fecal Streptococcus
(MPN/100 ml)
3,300
3.U80
9,180
U.100
33,300
U,900
11,000
i»,6oo
1*, 100
22,100
92,000
> 1,600
10,900
5^,200
6,300
-------
Table BIO. - Continued
Date
March 11
March 12
March 13
March lU
March 15
Rainfall
i r»TO
1972
May 1
May 12
May 22
May 23
May 2b
May 28
May 29
June 18
June 19
July 8
July lU
July 21
July 26
Total Co li form
(MPN/100 ml)
> 160,000
>i,6oo,ooo
>1, 600 ,000
1,600,000
3U8.000
3,300
U.900
3U.800
lU,100*
17,200*
3^,800
3U.800
160 ,900
2H.OOO
U9.000
161,000
3U8.000
3^8,000
Fecal Coliform
(MPN/100 ml)
110
110
80
80
0
200
170
700
700*
790*
10,900
1,720
3U.800
790
3^,800
91,800
8,000
33,000
Fecal Streptococcus
(MPN/100 ml)
17,200
3U,800
17,200
U,900
10 ,900
1,300
13,000
160,900
2,300*
7,000*
5^,200
91,800
27,800
91,800
91,800
3U.800
1*9,000
23,000
•Delayed coliform test.
135
-------
Site No. 9
Area =9.79 acres
Cover: Corn stubble and oats - *71
Plowed and idle acres - f72
Table Bll. - Bacteriological Quality Characteristics of Agricultural
Runoff, Brookings County, South Dakota
Date
Snovmelt
February 16
Febuary 17
February 18
February 25
February 26
March 10
March 11
March 12
March 13
Rainfall
1972
May 1
May 29
June 18
June 19
July 21
July 26
Total Coliform
(MPB/100 ml)
> 16,090
16,100
>160,900
979,000
730,000
1,090,000
31*0,000
1,300,000
1*60,000
110
31*, 800
91,800
13,000
91,800
161,000
Fecal Coliform
(MFH/100 ml)
230
330
220
< 10,000
< 10,000
70
33
2,1*00
k9
20
3,300
3U.800
1,090
7,900
7,900
Fecal Streptococcus
(MPN/100 ml)
1,300
1,300
3,1*80
2,9»»0
5,1*20
13,000
22,100
7,900
3,100
790
3,300
160,900
17,500
161,000
91,800
136
-------
APPENDIX C
Pesticide Data
137
-------
Table Cl. - Results of Pesticides Analyses For Phase I
(All values in ppb, unless noted.)
Date
Site 1
Site 2
Site 3
Site U
3/23/70
U/l/70
U/2/70
5/31/70
6/15/70
Aldrin 0.17
DDE 0.07
Heptachlor 0.25
Aldrin 0.29
DDT 0.13
(Lost)
(Duplicate)
Aldrin 0.17
Aldrin 0.30
Aldrin 0.07
Dieldrin 0.08
DDE 0.21
DDD 0.17
•Below analytical test limits.
138
-------
Table C2. - Results of Pesticides Analyses on Filtered Samples
Phase II
(All values in ppb, unless noted.)
Date Site 1 Site 2 Site 3 Site 4 Site 7 Site 8 Site 9
2/16/71 «»«»«**
2/17/71 »«»*«»*
2/18/71 «»•••»»
2/25/71 .15DDD
& .2 DDT * .12 DDD * .27 DDD .18 DDD .18 DDD
2/26/71 .I1* DDT .10 DDT .17 DDT
3/10/71 •!** DDE
to
3/13/71
3/11/71 .33 DDT
to
3/13/71
6/8/71 -2 DDE
.1 DDT
3/6/72 *
3/7/72 * * *
3/8/72
&
3/9/72
3/11/72 *
3AO/72 9
3/13/72
3/1U/72
4
3/15/72 * " *
3/15/72 *
139
-------
Table C2. - Continued
(All values in ppb, unless noted.)
Date Site 1 Site 2
5/1/72
5/12/72
5/22/72 »
5/23/72
5/2U/72
5/28/72 ,2U DDE
AM .13 DDD
.10 DDT
5/28/72 .22 DDE
PM
5/29/72 • *
7/8/72
7/21/72
7/26/72
7/28/72 » «
Site 3 Site U Site 7 Site 8 Site 9
• *
•
.06 DDE
*
»
.06
Aldrin
.17 DDE
.08 DDE
• * »
.61
Aldrin
.12 DDE
* «
.10 DDE
.15 DDE .19 DDE .06 DDE
•Below analytical test limits.
1*0
-------
Table C3. - Results of Pesticides Analyses on Filtered Samples
Phase II
(All values in ppb, unless noted.)
Date Site 1
6/8/T1
Site 2 Site 7 Site 8
10 Aldrin
Site 9
6/29/71
5/1/72
5/12/72
5/22/72
5/2U/72
5/28/72
AM
5/28/72
PM
5/29/72
6/18/72
&
6/19/72
7/8/72
100 Aldrin
23 DDE
U2 DDT
262 Aldrin
10 Aldrin
13 DDE
19 DDT
5 DDE
1 Dieldrin
10 DDT
3.U Lindane
5.8 Aldrin
22.9 DDE
5.3 Dieldrin
5 Aldrin
16 DDE
1*0 Aldrin 59 Aldrin 191* Aldrin
10 Aldrin
12 DDE
lU.17 ppm
Atrazine
290 Adrin
170 DDE
7/1V72
7/21/72
-------
Table C3. -Continued
Date Site 1 Site 2 Site 7 Site 8 Site 9
7/26/72 89 Aldrin
7/28/72 89 Aldrin
12 DDE »
17 DDT
•Below analytical test limits
-------
APPENDIX D
Mean Concentrations of Parameters
3*3
-------
Table Dl. - Mean Concentrations of Agricultural Runoff Parameters
Brookings County, South Dakota
Total
Site & Time Residue (out/l)
Site 1
Snowmelt '71
Snovmelt '72
Rainfall '72
Site 2
Snovmelt '71
Snovmelt '72
Rainfall '72
Site 3
Snovmelt '71
Snowmelt '72
Rainfall '72
Site 1*
Snovmelt '71
Snovmelt '72
Rainfall '72
171
107
1670
158
137
2260
125
152
11*1
118
ll*9
75
Parameter
Specific Total
Suspended Conductance Phosphorus (mg/l)
Solids (aw/l) (umhOB/cm % 25°C) Raw Soluble
51
12
11*00
29
35
2150
17
22
61*
13
21*
17
132
115
396
13U
117
71
121
125
73
111
121
60
0.1»7
O.UU
1.08
O.Ul
0.53
2.30
0.1*7
0.56
0.30
O.Ul
0.57
0.1*0
0.35
0.1»3
0.31
0.30
0.1*1
0.27
0.19
0.15
0.16
0.19
0.13
0.27
Nitrate
(lMC/1 N)
0.5
0.9
0.6
0.5
1.1
1.1
0.6
1.0
0.3
0.6
1.0
0.3
-------
Table Dl. - Continued
Parameter
Total Suspended
Site & Time Residue (mg/l) Solids (ms/l)
H
ff
VJl
Site 7
Snowmelt
Snowmelt
Rainfall
Site 8
Snowmelt
Snowmelt
Rainfall
Rainfall
Site 9
Snowmelt
Rainfall
'71
'72
'72
'71
'72
'71
'72
'71
•72
15U
1*5
222
150
32U
9UO
1360
ll»3
395
21
13
38
35
121
552
1160
20
173
Specific Total
Conductance Phosphorus (mg/l)
(umhos/cm g 25°C) Raw Soluble
151*
139
251
135
198
133
285
lU
215
0.
0.
0.
0.
0.
1.
1.
0.
0.
71*
60
1*9
1*0
U5
27
08
U6
50
0
0
0
0
0
0
0
0
0
.3*
.28
.31
.25
.19
.27
,1U
.30
.19
Nitrate
(mK/1 N)
0
1
0
0
2
2
2
0
0
.8
.0
.1*
.8
.5
.6
.1
.6
.9
-------
Table Dl. - Continued
ON
Parameter
Site & Time
Site 1
Snovmelt
Snovmelt
Rainfall
Site 2
Snovmelt
Snovmelt
Rainfall
Site 3
Snovmelt
Snovmelt
Rainfall
Site It
Snovmelt
Snovmelt
Rainfall
Site 7
Snovmelt
Snovmelt
Rainfall
'71
'72
'72
'71
'72
'72
'71
'72
'72
'71
•72
•72
'71
'72
•72
Total KJeldahl
Nitrogen (ng/1 N)
Rav Soluble
1.3
1.3
2.0
1.2
1.1*
5.7
2.5
3.U
0.8
2.2
3.3
0.8
3.1*
3.1*
1.7
0.9
1.3
1.1
1.0
1.2
2.0
1.7
2.3
0.7
1.6
2.2
0.7
2.6
2.2
1.2
COD (nw/1)
Pav
35
22
171*
31*
21
275
56
73
22
50
7!*
22
73
61*
1*9
Soluble
21*
20
31
25
11
19
37
39
16
35
1*0
20
1*8
39
30
Ammonia Number
(mg/1 N) Of Samples
8
1
0.2 5
5
3
0.2 2
9
6
0.2 1
9
6
0.2 1
9
7
0.3 2
-------
Table Dl* - Continued
H
ff
Parameter
Total KJeldahl
Nitrogen (mg/1 N)
Site t Time
Site 8
Snovmelt '71
Snovmelt '72
Rainfall '71
Rainfall '72
Site 9
Snovmelt '71
Rainfall '72
Raw
2.3
2.8
3.1
2.9
2.6
1.3
Soluble
2.2
1.9
1.0
1.0
2.5
0.7
COD
Raw
57
61
82
179
60
39
(mg/1)
Soluble
1*7
29
30
20
k&
16
Ammonia Number
(mg/1 N) Of Samples
9
9
0.2 2
0.1 13
9
0.1 6
-------
APPENDIX E
Annual Contributions
-------
Table El. - Yearly Runoff Contributions of Agricultural Runoff Constituents
Brookings County, South Dakota
Constituent (Ib/acre/yr)
H
VO
£
1
m
i
IS
Tl
3D
1
o
Site Year
1 1971
1972
2 1971
1972
3 1971
1972
I* 1971
1972
7 1971
1972
8 1971
1972
9 1971
1972
Total
Residue
11.7
1*08
3.3
29»*
37.1*
39.8
25.9
9.5
1*0.1*
63.1*
59.8
953
52.5
11*5
Chemical
Suspended Oxygen
Solids
3.9
1*01*
0.6
296
5.1
U.3
3.5
1.1
7.21
13.7
17.5
81*0
10.3
101
Demand
2.1*
1*0.1
0.7
33.2
ll*.7
16.9
11.1*
l*.l
19.8
29.7
23.1
129
22.7
16.0
Total
Total
Phosphorus KJeldahl
as P
0.03
0.21
0.01
0.25
0.11
0.11
0.10
0.03
0.19
0.2U
0.16
0.72
0.18
0.17
Nitrogen
0.10
0.2l*
0.02
0.1*5
0.68
0.98
0.65
0.22
0.95
1.0U
0.86
1.95
0.9l*
0.1*1
Nitrate-N
0.03
0.06
0.01
0.05
0.22
0.39
0.15
0.07
0.23
0.1*8
0.26
1.05
0.26
0.11
Total
0.13
0.30
0.03
0.50
0.90
1.37
0.80
0.29
1.17
1.52
1.12
3.00
1.20
0.52
-------
SELECTED WATER
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
w
4. Title
Quantification of Pollutants in Agricultural Runoff
James H. Dornbush, John R. Andersen, Leland L. Harms
Civil Engineering Department
South Dakota State University
Brookings, South Dakota 57006
5. R
6.
S. PfrfoTtnisr OTgax''iation
Rtjort No.
R800UOO
68-01-0030
13. Type < ' Repor: and
Period Coveted
_ . _ ^ *. , «_ *. _.... .
U. S. Environmental Protection Agency
r ^ . -
Environmental Protection Agency report number,
EPA-660/2-74-005, February 1974.
15. Ah^rnct
Surface runoff from snovmelt and rainfall in eastern South Dakota was measured during
a three year period. The size of the research sites ranged from 7.18 to 18.69 acres,
and all sites had crops of corn, oats, pasture or hayland. Composite samples of the
runoff were used for various chemical, physical and biological determinations.
Runoff samples from 108 snovmelt events and 36 rainfall events vere collected. Equipmeir
fabrication and installation resulted in some incomplete data for the initial year, but
successful monitoring of each runoff event was accomplished thereafter.
Sediment losses vere considerably lower than anticipated. Pesticide concentrations vere
low in both vater and sediment samples, and vere usually less than the analytical test
limits. Coliform and fecal levels vere consistently greater than accepted surface vater
quality criteria. Most of the nutrients vere found to be soluble and/or associated with
snowmelt runoff.
17a. Descriptors
Surface Runoff, *Agricultural Runoff, Erosion,'Nutrients, Phosphorus, *Nonpoint
pollution source, Runoff Chemistry, Sediments, "Water Pollution Sources,
Pesticides, Bacteriological Indicators
l"b. Identifiers
Eastern So. Dak., Runoff Pollutants
KRFn'UA- Group
-2G. Security Class.
21, tio.of
.Pages
22. Price
SeodTo:
WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON. D. C. 2O24O
Leland L. Harms
South Dakota State University
------- |