United States
Environmental Protection
Agency
Great Lakes National
Program Office
536 South Clark Street
Chicago, Illinois 60605
EPA-905/9-79-005-C
May 1981
Volume III
coa
xvEPA
Maumee River
Pilot Watershed Study
Continued Watershed
Monitoring (1978-80)
-------
EPA-905/9-79-005-C
May 1981
THE MAUMEE RIVER BASIN PILOT WATERSHED STUDY
Volume III
Continued Watershed Monitoring (1978-80)
by
Terry J. Logan
Principal Investigator
(Grant R005353 01)
Ohio State University, Columbus, Ohio 43210
Ohio Agricultural Research and Development Center
Wooster, Ohio 44691
for
U.S. Environmental Protection Agency
Chicago, Illinois
Project Officer
Ralph Christensen
Great Lakes National Program Office
GREAT LAKES NATIONAL PROGRAM OFFICE
U.S. ENVIRONMENTAL PROTECTION AGENCY, REGION V
536 SOUTH CLARK STREET
CHICAGO, ILLINOIS 60605
-------
DISCLAIMER
This report has been reviewed by the Region V Office, U. S. Environ-
mental Protection Agency and approved for publication. Approval
does not signify that the contents necessarily reflect the views and
policies of the U. S. Environmental Protection Agency, nor does men-
tion of trade names or commercial products constitute endorsement or
recommendation for use.
-------
ACKNOWLEDGEMENTS
Work on this project was funded by a grant from the Great Lakes
National Program Office, U.S. Environmental Protection Agency, Region V,
Chicago, with Mr. Ralph Christensen, Project Officer.
Special thanks are due to Dr. Tom Oloya for his work on soluble
phosphorus transport, and to Mr. Bob Rettig for his technical support in
the field. The support of the staff at the Hoytville Branch of the Ohio
Agricultural Research and Development Center is also gratefully acknowledged.
iii
-------
ABSTRACT
Monitoring, which was started in 1975 as part of the PLUARG
Task C Pilot Watershed Study in the Maumee River Basin of Ohio, was
continued in 1978-1980 on three small watersheds in Defiance County and
eight plots in Wood County. Runoff and tile drainage were monitored for
flow, suspended solids, total P, filtered reactive P(FRP), NH3-N and N03-N.
Runoff and soil loss (4388 kg/ha) continued to be greatest on
the poorly drained Paulding soil compared to other sites with spring-
seeded crops, but winter wheat on the Paulding site greatly reduced
runoff in 1979 and there was no runoff with wheat in 1980. Wheat in
1980 on the other two Defiance watersheds did not affect runoff. No
till soybeans on the Blount soil in 1978 reduced erosion to near zero
(66 kg/ha) compared to previous years with fall-plowed soybeans where
sediment yields ranged from 900 to 2500 kg/ha. Runoff volume with no till,
however, was not measurably different than with fall plowing and filtered
reactive P loads with no till were no different than fall-plowed FRP loads.
Tile drainage had no effect on runoff volume from Hoytville soil, and
no till contineud to have no effect on soil loss compared to fall plowing
on this soil where soil losses have been low (< 750 kg/ha) throughout the
study. Phosphate fertilizer broadcast in the fall on the no till and fall-
plowed Hoytville plots every year from 1975-1979 steadily increased total and
FRP concentrations and loads in this period. FRP loads decreased rapidly in
1980 after fall fertilization was terminated.
iv
-------
TABLE OF CONTENTS
Page
DISCLAIMER .............................. **
ACKNOWLEDGEMENTS ........................... ll:L
ABS TRACT
LIST OF TABLES ............................ vi
LIST OF FIGURES ............................ viii
1. INTRODUCTION
1 . 1 Study Approach ........................ 3
1.2 Study Methods ........................ 3
1.21 Monitoring Sites in Defiance County
1.22 Surface Runoff and Tile Drainage
Measurement - Defiance County Sites
1.23 Surface Runoff and Tile Drainage
Measurement - Hoytville Plots
1.24 Analysis of Watershed and Plot
Water Samples
2. RESULTS .............................. 27
2.1 Precipitation and Flow (1978-1980) .............. 27
2.2 Soil and Nutrient. Losses (1978-1980) ............. 29
2.21 Hammersmith Roselms (111) ............... 29
2.22 Heisler Blount (401, 402) ............... 31
2.23 Speiser Paulding (501, 502) .............. 31
2.24 The Hoytville Plots (611-681, 612-682) ......... 36
2 . 3 Seasonal Tr^dj_gJ_Precip_jjatj.on, Flow and Soil Loss
(1975-1980)T~. ' ....................... 45
2.4 Crop Yields on Hoytville Tillage Plots (1975-1980) ..... 50
3. DISCUSSION .......................... 52
4. LITERATURE CITED ......................... 56
v
-------
LIST OF TABLES
Table Number Page
1. Characteristics of the Defiance County watersheds (111, 401, 501)
and Hoytville plots monitored in the period 1978-1980 5
2. Summary of crop management practices on Hoytville plots (1977-1980). 25
3. Precipitation and flow from Defiance watersheds and Hoytville plots
(1978-1980) 28
4. Concentrations and pollutant loads from Hammersmith Roselms (111)
surface runoff 30
5. Concentrations and pollutant loads from Heisler Blount (401) surface
runoff 32
6. Concentrations and pollutant loads from Heisler Blount (402) tile
drainage 33
7. Concentrations and pollutant loads from Speiser Paulding (501) sur-
face runoff 34
8. Concentrations and pollutant loads from Speiser Paulding (502) tile
drainage 35
9. Concentrations and pollutant loads in runoff from Hoytville plots
(621, 671). Plots were no tilled and tile drained. Mean of two
plots 37
10. Concentrations and pollutant loads in tile drainage from Hoytville
plots (622, 672). Plots were no tilled and tile drained. Mean of
two plots 3g
11. Concentrations and pollutant loads in runoff from Hoytville plots
(631, 681). Plots were no tilled with no tile drainage. Mean of
two plots 39
12. Concentrations and pollutant loads in runoff from Hoytville plots
(641, 661). Plots were fall plowed and tile drained. Mean of two
plots 40
13. Concentrations and pollutant loads in tile drainage from Hoytville
plots (642, 662). Plots were fall plowed and tile drained. Mean of
two plots 41
14. Concentrations and pollutant loads in runoff from Hoytville plots
(611, 651). Plots were fall plowed with no tile drainage. Mean of
two plots 42
15. Changes in concentration and unit area loads in runoff of total and
filtered reactive phosphate with fertilization of fall-plowed and no
till Hoytville soil (1975-1980) 44
vi
-------
16. Crop yields (bu/acre) on the Hoytville plots for the period 1975-
1980. Mean of two plots
51
vii
-------
LIST OF FIGURES
Page
Figure Number
1. Watershed and plot locations in the Maumer River Basin.
2. Hammersmith Roselms watershed. Heavy line denotes the monitored
area 7
3. Hammersmith Roselms watershed showing the sampling shelter 8
4. Heisler Blount watershed. Heavy line denotes the monitored area
and dotted lines are tile 10
5. Heisler Blount watershed looking downslope 11
6. Speiser Paulding watershed. Heavy line denotes the monitored area
and dotted lines are tile 12
7. Speiser Paulding watershed showing the sampling shelter 13
8. Sediment drop box used to collect runoff from Defiance County water-
sheds 2.5
9. System for monitoring and sampling surface runoff at Defiance County
watersheds 16
10. Sample containers for runoff and tile drainage at Defiance County
watersheds 17
11. System for monitoring and sampling tile flow 20
12. Runoff and tile drainage plots at OARDC research station, Hoytville,
Ohio 22
13. Analytical scheme for water samples 26
14. Monthly precipitation, flow and soil loss from Roselms watershed
(1975-1980) 46
15. Monthly precipitation, flow and soil loss from Blount watershed
(1975-1980) 47
16. Monthly precipitation, flow and soil loss from Paulding watershed
(1975-1980) 48
17. Monthly precipitation, flow and soil loss from Hoytville (621, 622)
plot (1975-1980). 49
viii
-------
-1-
1. INTRODUCTION
The Maumee River was chosen by PLUARG to be one of four pilot water-
sheds to be studied on the U. S. side of the Great Lakes drainage basin
as part of Task C - pilot watershed studies. Since there was already an
ongoing PL-92-500 Sec. 108 demonstration project in Black Creek basin, an
Indiana tributary to the Maumee, the Task C project was directed to the Ohio
portion of the Maumee to supplement the work being done in Black Creek.
The objectives of PLUARG were to determine the effects of prevailing
land use practices on pollution entering the Great Lakes. Specifically,
the PLUARG Task C objectives were to answer the following questions:
1. From what sources and from what causes (under what conditions,
management practices) are pollutants contributed to surface and
ground water?
2. What is the extent of pollutant contributions and what are the
unit area loadings by season from a given land use or practice
to surface or ground water?
3. To what degree are pollutants transmitted from sources to
boundary waters?
4. Are remedial measures required? What are they and how effective
might they be?
5. Were deficiencies in technology identified? If so, what is
recommended?
The Maumee River Basin is primarily agricultural in land use, and the
intensive crop production in the Basin accounts for most of the sediment
and a major part of the nitrogen and phosphorus delivered to Lake Erie
(Corps of Engineers, 1975; Sonzogni et^ a_l, 1978). Because of the importance
of agriculture as a source of pollutants in the Maumee Basin, it was decided
-------
-2-
to place emphasis in the Task C project on soil and nutrient loss from
small agricultural watersheds and on specialized studies on sediment
transport.
Specific objectives of this study were:
1. To determine the effects of land use practices on the loss of
sediment and associated chemicals from representative small
agricultural watersheds in the Basin and to compare these
data with downstream reference samples.
2. To study and determine the physical, chemical, and mineralogical
properties of major soils in the Basin and relate these data
to their susceptibility to erosion and fluvial transport.
3. To determine the physical, chemical, and mineralogical properties
of suspended sediments and bottom sediments in order to identify
fluvial transport mechanisms and to evaluate equilibrium
stabilities of minerals in suspended and bottom sediments.
4. To determine phosphate sorption-desorption and precipitation
interactions with sediment characteristics and concentration
levels.
5. To determine heavy metals leaving small agricultural watersheds
as contrasted to downstream reference sources.
The results of this study (1975-1977) have been published previously
(Logan and Stiefel, 1979; Logan, 1979) and the reader should consult them
for more complete details of the study results. This report presents the
results of the continued monitoring of three of the Defiance County water-
sheds and the Hoytville plots for the period 1978-1980.
-------
-3-
1.1 Study Approach
The basic approach of this study was to measure the generation of
sediment and nutrients from intensively cultivated cropland under prevailing
management practices. The study investigated the differences in pollutant
generation on several of the major soils of the Maumee Basin and determined
the effects of season and soil characterist'ics on sediment and nutrient
generation. Pollutant transport by tile drainage was also studied because
of the extensive use of underground tile for drainage in the Basin.
1.2 Study Methods
Five sites were chosen in Defiance County on four major soils of
the Basin (Figure 1 and Table 1) ranging from 0.6 to 3.2 hectares in the
area. Surface runoff was monitored at all.sites and tile drainage on the
Paulding and Blount sites. A continuous-flow monitoring system and
integrated sampler were used so that all events were monitored and sampled.
The sampling period was from January, 1978 - May 1980. Rainfall was
monitored at each site. At the OARDC branch research station in Wood County,
eight plots (0.04 ha) on Hoytville soil were subjected to a number of
different tillage treatments, and runoff and tile drainage were monitored.
1.21 Monitoring Sites in Defiance County
Five small agronomic sites were chosen in Defiance County to monitor
soil and nutrient loss under prevailing crop management practices. The
sites represent four of the more important series in the Basin: Paulding,
Blount, Roselms and Lenawee (similar to Latty). The sites were selected
with the following criteria:
1. Topography was typical for that series
2. The watershed was dominated by a single series
-------
The Maumee River Basin
Water samples
Watersheds
1 Hammersmith Roselms
2 Crites Roselms
3 Lenewee
4 Blount
5 Paulding
6 Hoytville Plots
fc Continuous mass
transport stations
10 15 20 25
MICHIGAN
OHIO
ir~i *
Figure 1. Watershed and plot locations in the Maumee River Basin.
-------
Table 1. Characteristics of the Defiance County watersheds (111, 401, 501) and Hoytville plots monitored in the period
1978-1980.
Site
Code
111
501
502
611-681
612-682
401
402
Dominant
Soil Series
Roselms
Paulding
Paulding
Hoytville
Hoytville
Blount
Blount
Soil Taxonomy
Aerie Ochraqualf
Typic Haplaquept
Typic Haplaquept
Mollic Ochraqualf
Mollic Ochraqualf
Aerie Ochraqualf
Aerie Ochraqualf
Physiographic
Region
Lake Plain
Lake Plain
Lake Plain
Lake Plain
Lake Plain
Till Plain
Till Plain
Parent
Material
Lacustrine Clays
Lacustrine Clays
Lacustrine Clays
Clay Till
Clay Till
Clay Loam Till
Clay Loam Till
Slope Drainage
(%) Area
(ha)
3-16 3.2
1 1.0
0.1
< 1 0.04
0.04
3-4 0.9
0.9
Drainage Systems
Monitored
Surface Runoff
Surface Runoff
Subsurface Tile
Surface Runoff
Subsurface Tile
Surface Runoff
Subsurface Tile
i
Y1
-------
-6-
3. The watershed _could be defined hydrologically
4. There were no septic tank or livestock waste discharges within
the watershed
5. Cooperation from the landowner was available
6. Site was accessible from the road, had adequate flow outlet,
and electrical service could be brought to the site.
Using these criteria, a large number of sites were examined and five
were selected. These were described in detail by Logan and Stiefel (1979)
in their report on the 1975-77 monitoring period. In the 1978-80 period
reported here, only the Hammersmith Roselms (111), Heisler Blount (401, 402)
and Speiser Paulding (501, 502) watersheds and the Hoytville plots (60X)
were monitored. A detailed description of the properties of the watershed
soils has been previously given by Logan (1979).
Table 1 summarizes the site characteristics and Figure 1 identifies
their location. A more detailed description of each site is given next.
A 3-digit code was used to identify the sites and for identification of
samples from each site:
First digit: 1-6 identifies the primary site
Second digit: 0-8 identifies the sub-site within the primary site
Third digit: 1 refers to surface runoff and 2 to tile drainage, which
were monitored separately.
Hammersmith Roselms (111); This site is located in the central area
of Defiance County and in the lake plain. The soil and plot map is given
in Figure 2, and the area is shown in Figure 3. The drainage area is 3.2 ha
(8.0 acres) and is composed of Roselms on most of the area with Broughton
on the steep slopes. The watershed has a well-defined drainageway (Figure 3),
-------
-7-
Hammersmith Roselms (10X)
Location: Noble township, T4N, R48, Sec. 6, NW*/4
Abandoned Road
BvB - Broughton scl
BvC2 - Broughton scl
BwDa - Broughton Clay
Pa - Paulding Clay
RsA - Roselms scl
RsB - Roselms scl
RsB2 - Roselms scl
Figure 2.
Layout of Hammersmith Roselms watershed.
the monitored area).
1 inch = 165 feet
(Heavy line denotes
-------
-8-
^^^^^&^^^^*^j^v^^
A: Sampling shelter. Front end is open to allow runoff to enter the
sediment drop box.
B, C: Component plots on different slope positions, showing the V-shaped
flumes.
Figure 3. Hammersmith Roselms watershed showing the sampling shelter.
-------
and the monitoring system is placed at the point where the drainageway exits
the watershed. Slopes vary from 1-3% on the more level part of the water-
shed to as high as 15 % where the landscape breaks into the drainageway.
Heisler Blount (401, 402): This site is located in the northwest
corner of Defiance County and is in the till plain region of the Maumee
River Basin. The soil and plot map is given in Figure 4 and the watershed
is illustrated in Figure 5. The area is bermed on the upslope perimeter
and on the lower side to channel the flow toward the flume. The upper
part of the site is Blount loam while the lower end is Mermill loam, which
represents the unconsolidated soil eroded from the top of the slope and
deposited downslope. The surface drainage area (401) is 0.8 ha (2.1 acres).
A previously installed tile system was also monitored (402), and the drainage
area has been estimated to be between 2 and 4 acres. The tile drainage
pattern shown in the plot diagram (dotted lines) (Figure 4) is only
speculative.
Speiser Paulding (501, 502): This site is located in the southcentral
area of Defiance County in the lake plain region. The soil and plot map is
given in Figure 6 and the area is illustrated in Figure 7. The major part
of the plot is occupied by Paulding-Roselms clay, a series which has all the
characteristics of a typical Paulding clay but whose clay content is minimal
for Paulding. About a third of the plot is Paulding clay itself. The surface-
drained area (501) is 0.9 ha (2.5 acres) and was defined by throwing up a berm.
This soil is normally surface-drained by using shallow field ditches, and
in this instance, the ditches were used to carry surface runoff to the
sampler. Three tile drains were installed 12.7 m apart and 1 m deep.
-------
-10-
Heisler Blount (40X)
Location: Farmer township, T5N. R2E, Sec. 19, NW*/4
BnA - Blount loam
BnB - Blount loam
GIB - Glynwood loam
Md - Mermill loam
Pm - Pewamo silty clay loam
1 inch = 165 feet
Figure 4. Layout of Heisler Blount watershed (Heavy line denotes the
monitored area and dotted lines are tile.)
-------
-11-
Figure 5. Heisler Blount watershed looking downslope.
-------
-12-
Speiser Paulding (SOX)
Location: Delaware township,
T4N, R3E, Sec. 15. SWV4
Pa - Paulding clay
RsA - Paulding - Roselms clay
1 inch = 165 feet
Figure 6. Layout of Speiser Paulding watershed (heavy line denotes
monitored area and dotted lines are tile.)
-------
-13-
Figure 7. Speiser Paulding watershed showing the sampling shelter.
-------
-14-
The central tile, 55.7 m (220 feet), was monitored with a drainage area of
0.09 ha (0.23 acres).
1-22 Surface Runoff and Tile Drainage Measurement - Defiance County Sites
Surface Runoff: It was decided early in the development of this
research that sophisticated instrumentation of the sites in Defiance
County was not feasible or warranted. A number of physical restraints
guided the selection of monitoring devices: both small and large events
must be monitored; equipment would have to be automatic because events
on small areas are very rapid and the sites had to be serviced by a single
technician; it was important to be able to operate in the winter because
much of the runoff occurs in the initial storms after thawing in the early
spring; there was a general lack of hydraulic head at all sites. The
system that was developed had the following basic principle: the runoff was
channeled over a drop structure and a known fraction of the flow was inter-
cepted. The intercepted flow was then passed over a Coshocton wheel, which
intercepted another fraction. This water then discharged into a sump. A
sump pump of known discharge rate (gallons per minute) was activated when
water in the sump reached a given level. The pump was connected to a timer,
which recorded time of pumping. The water was pumped up into a container
from which a sample could be taken. By knowing the fraction of total
runoff intercepted and the pump rate and time of pumping, total runoff in
a given interval was calculated. The sample taken from the pump discharge
represented runoff for that interval. Samples were taken after each event.
A diagram of the equipment used is given in Figures 8, 9 and 10.
Figure 8 shows a standard SCS concrete drop-box, which is used to carry
runoff from surface drains to the stream or drainage ditch without causing
-------
-15-
fry^f ^ * t^sm .
""'"*'/^kjl
V
Figure 8. Sediment drop box used to collect runoff from Defiance County
watersheds.
-------
-16-
A: Variable slit flume which diverts fraction of runoff into Goshocton
wheel (B).
C: Sump collects discharge from Coshocton wheel; discharge is then
pumped into sample container (Figure 13).
Figure 9. System for monitoring and sampling surface runoff at Defiance
County watersheds.
-------
-17-
A: Sump for collecting runoff. Contains sump pump which discharges into
sample container (B).
C: Sample container for tile drainage.
Figure 10. Sample containers for runoff and tile drainage at Defiance
County watersheds.
-------
-18-
undue erosion of the bank. A similar structure was used at all five
sites in Defiance County. The perimeter of the box was levelled so that
flow would be uniform around it. A flume with adjustable vertical slit
(Figure 9) was bolted to the front rim of the drop-box. The runoff from
the slit fell over a Coshocton wheel (Figure 9) and then from the Coshocton
wheel into a sump which was bolted to the floor of the drop-box (Figure 9).
The runoff was pumped by a "Haynes Demon Drainer" submersible pump. This
particular model was used because it would pump to near dryness and this
prevented an accumulation of sediment in the sump. Recovery of sediment
was tested in the laboratory during the development and calibration of this
equipment and was found to be acceptable. The pump was activated by
electrodes set to turn on when approximately 0.1 inch of runoff was recieved,
The pump could also be activated manually. The pump was connected to a
timer, which could either accumulate pumping time or be reset between events,
The runoff was pumped into a 20-gallon plastic garbage can with a fitted
lid (Figure 10). After each event, a subsample (usually 1 gallon) was
taken from the container by a faucet at the bottom after thorough mixing.
The remaining sample was discarded. The entire system was housed in a
shed open only at the front, where the drop-box faced the field. The
equipment was winterized by the use of heat lamps directed onto the
Coshocton wheel and mounted in the sump and garbage lids. Heating tape
was used for all pipes. Even during the extremely low temperatures of
1977, the system never failed to operate during winter events.
Tile drainage. In all cases, a single tile line was monitored,
except for the Blount site (402), where a small tile system was monitored
by intercepting the main at the point where it discharged into the drainage
-------
-19-
ditch. The tile was usually at a depth of 1 meter, and a specially
constructed fiberglas sump was set into the ground in the same sampling
shelter used for surface runoff. The sump (Figure 11) intercepted the
tile and collected all discharge. As in the case of surface runoff, a
calibrated sump pump was used to pump the water out of the sump. A
timer was used as before to measure pumping time, and the pump was activated
at a given water level by electrode; it could also be activated manually.
An orifice inserted into the discharge pipe from the pump delivered a sample
of the water to a 20-gallon plastic garbage can, where it was subsampled
as described previously. This sample was considered to be representative
of the tile flow for a given time interval, since all of the flow was
sampled. The amount of sample taken by the orifice was adjusted by a valve.
Sampling Handling and Processing - All sites in Defiance County were
serviced by a technician every 48 hours or sooner if significant precipitation
occurred. A 1-gallon subsample of the sample in the garbage can was taken
after thorough mixing and the remainder discarded. Sumps were pumped dry
manually after subsampling, time of pumping was recorded and rainfall at the
site was measured from a manual rain gauge. Samples were stored in a
refrigerator at 4° C at field headquarters until they could be transported
to the laboratory at Columbus. Samples usually reached the laboratory within
7 days or less. Additional measurements taken in the field included depth
of snow cover, depth to frozen soil and all pertinent details on
field operations (times of planting, plowing, harvesting, rates of fertili-
zation, etc.) .
Cropping Practices - The following cropping practices were employed by
the cooperating farmers on the Defiance County watersheds in 1978-1980:
-------
-20-
B:
Fiberglas sump which intersects field tile. Contains submersible
sump pump with flow-activating electrodes.
Sampling valve which diverts portion of sump pump discharge into
sample container.
Figure 11. System for monitoring and sampling tile flow.
-------
-21-
Hammersmith Roselms (111) - Fall chisel every year. In 1978 and 1979
soybeans was the crop. In fall of 1979 wheat was planted. Grassed water-
way was established in fall of 1979.
Heisler Blount (401, 402) - In 1978, no till soybeans were grown.
Residue cover was good because of a heavy weed infestation in the previous
year which was killed with paraquat prior to no till soybeans. In fall of
1978, the watershed was chisel plowed and soybeans grown in 1979. In fall
of 1979, the watersehd was planted to wheat.
Speiser Paulding (501, 502) - Fall molboard plowed every year. Oats
were grown in 1978, followed by wheat in 1979 and 1980. The oat crop was
fertilized with 33 kgN/ha and 15 kgP/ha just after seeding.
1.23 Surface Runoff and Tile Drainage Measurement Hoytville Plots
In 1974, a research facility was constructed at the NW Branch, Ohio
Agricultural Research and Development Center (OARDC), located at Hoytville
in Wood County (Figure 1), to study the loss of soil and nutrients by
runoff and tile drainage. Eight plots, each 30.5 m (100 ft) x 12.1 m (40 ft),
were laid out, four in a block, with a sampling house in the center (Figure
12). Each plot was trenched to a depth of four feet and heavy plastic
sheeting was placed against the plot wall; the soil was then backfilled
to hold the plastic in place. Earth berms (15-30 cm high) were raised on
the sides of the plots and seeded with fescue. The backs of the plots were
left open to allow passage of equipment; a berm was then formed after each
operation to enclose the plot. A concrete gutter was built on the other
end of the plots with a 10 cm (4 inch) diameter drain to collect runoff.
The drain was connected by 10 cm (4 inch) plastic pipe (placed at 90 cm depth)
to the sampling house. A 10 cm (4 inch) perforated corrugated plastic tile
-------
-22-
Figure 12. Runoff and tile drainage plots at OARDC Research Station,
Hoytville, Ohio.
-------
-23-
was installed in the center of each plot at a depth of 90 cm. The tiles
were also connected by 10 cm (4 inch) solid pipe to the sampling house.
Additional field tile was placed outside the plot area to keep water other
than that intercepted by the plots from entering the area. The hydraulic
conductivity of the soil (Hoytville clay) was low enough to prevent any
significant water movement between plots. The area between the plots and
sampling house was seeded with fescue to prevent erosion.
The sampling procedure used was similar to that used to measure tile
drainage on the Defiance County sites. Fiberglas sumps intercepted the
flow from the surface runoff and tile drain lines. Sump pumps (Hydromatic
submersible pump) and timers were used to measure flow as described pre-
viously, and water was sampled as before by placing an orifice in the dis-
charge line from the sump pump. The sampled water was collected in 1-gallon
or 5-gallon plastic bottles housed in a refrigerated (4° C) compartment
so that the samples were refrigerated immediately. Samples were returned
to the laboratory at Columbus within 1 week or less. Samplers were serviced
daily and sumps were pumped dry between events. Precipitation records were
kept by the personnel at the research station which has a 20-year weather
record.
The facility was completed early in 1975, and some flow and sediment
monitoring was initiated in April 1975; water quality sampling was begun in
May 1975. The previous fall, the plots were fall plowed and left bare until
planting in May 1975. The area had been in sod for at least 10 years prior
to construction of the plots and had received no fertilizer during that period.
In May 1975, soybeans were planted and all plots were treated the same
through November 1975 to measure variability among the plots. From 1976-1977,
-------
-24-
soybeans were grown with varying tillage treatments ranging from fall
moldboard plow to no till (Logan and Stiefel, 1979). In November 1977,
the tillage treatments were consolidated into fall moldboard plowing and
no till, with the previous no till and fall plowing plots continued, and the
intermediate tillage plots no tilled or fall plowed. One half of the plots
had their tile drainage pump systems disconnected to give four tile drained
and four non-tile drained plots. The plots and their treatments were:
Plots Tillage Drainage
641, 661 Fall plow Tile drained
642, 662
651, 611 " Not tile drained
621, 671 No till Tile drained
622, 672
631, 681 " Not tile drained
Crop management practices in 1977-1980 are given in Table 2. In 1978
and 1979 corn was grown and then soybeans again in 1980.
1.24 Analysis of Watershed and Plot Water Samples
As soon as samples were received in the laboratory, the 1-gallon poly-
ethylene bottles were shaken thoroughly and a 250 ml sample was placed in
another bottle and refrigerated (Figure 13). A 100-ml aliquot of the
unfiltered sample was filtered through a preweighed 1.0 um Nucleopore
membrane filter. The sediment and filter were oven-dried, reweighed, and
sediment concentration calculated. The filtered solution was refrigerated
until further analysis. Tests showed that a 1.0 um filter was effective
in retaining fine clay. The filtered sample was routinely analyzed for:
(NO,, + N02), NH-j, and filtered reactive-P. The unfiltered sample was
analyzed for total P. Methods of analysis were discussed in detail by
Logan and Steifel (1979).
-------
-25-
Table 2. Summary of crop management practices on Hoytville plots (1977-1980)
1977
1. Tillage -
2. Fertilization -
1. Fertilization -
2. Tillage -
3. Planting -
4. Pesticides -
5. Harvest -
6. Fertilization -
7. Tillage -
1. Tillage -
2. Fertilization -
3. Pesticides -
4. Planting -
5. Harvest -
6. Fertilization -
7. Tillage -
Half of the plots (1, 4, 5 and 6) were moldboard
plowed November 7. The other plots (2, 3, 7 and 8)
were in no till.
19 kgP/ha was broadcast October 5.
1978
180 kgN/ha as urea broadcast April 14.
Plowed plots were field cultivated twice April 28.
DeKalb XL 64 corn planted in 30 inch rows April 28.
1 kg/ha Furadan (AI) with planter; 2 kg/ha Atrazine,
2 kg/ha Lasso and 2 kg/ha Roundup (no till only) April
29.
October 5.
86 kgP/ha broadcast October 30.
Fall moldboard plowed (plots 1, 4, 5 and 6) October 31.
1979
Plowed plots were field cultivated twice April 23.
179 kgN/ha as anhydrous NH3 injected April 27. Also
112 kg/ha of 6-24-24 was applied through the planter
April 23.
17 kg/ha Counter applied at planting; 2 kg/ha
Atrazine, 2 kg/ha Lasso and 1.5 kg/ha Roundup (no till
only) applied May 1.
Landmark C747X corn planted in 30 inch rows April 23.
October 24
30 kgP/ha broadcast December 13.
Fall moldboard plowed (plots 1, 4, 5 and 6) December 14,
1. Tillage -
2. Planting -
3. Fertilization
4. Pesticides -
Plowed plots were field cultivated twice May 2.
Williams soybeans planted in 30 inch rows May 2.
112 kg/ha 6-24-24 was applied through the planter May 2,
2 kg/ha Dual 6E, 3 kg/ha Amiben and 1.5 kg/ha Roundup
(no till only) applied May 3.
-------
SAMPLER
1 gallon
unfiltered
water
100 ml
unfiltered
sample
Sediment
concentration
1.0 ym
Nucleopore
filter
Total P
100 ml
filtered water
Filtered
reactive P
NH/.-N
N03-N
Figure 13. Analytical scheme for water samples.
-------
-27-
2. RESULTS
2.1 Precipitation and Flow (1978-1980)
Table 3 gives annual precipitation and flow (surface runoff and tile
drainage) for the Defiance watersheds and Hoytville plots. The 1980 data
is through May only. Precipitation was lower in 1978 than in 1979, and this
is reflected in the flows for these years. Surface runoff was very high on
the Roselms site (111) in 1979, but there is no explanation for this increase.
However, this is the most precipitation received on this site since
monitoring began in 1975.
Runoff and tile flows on the Blount site (401, 402) in 1978-1980 were
similar to these monitored in 1975-77. In 1978, no till soybeans were
grown on this site as compared to fall-plowed soybeans in previous years
and again in 1979. No till appeared to have no effect on runoff and tile
drainage flows. The wheat crop in 1980 also appeared to have little effect
on drainage flows.
Surface runoff was very high on Paulding soil (501, 502) in 1978 and
tile flow very low. This was similar to the results of the 1975-77
monitoring, although the 1978 surface runoff was higher than in previous
years. The crop in 1978 was spring seeded oats, and since much of the
runoff occurs in late winter and early spring (Logan and Stiefel, 1979),
the oats crop would have had about the same effect on surface soil condi-
tions as the soybean crops grown in previous years. In 1979 and 1980,
winter wheat was grown on the Paudling watershed and this dramatically
decreased surface runoff to 4.2 cm in 1979. In 1980 there was no runoff
through May when monitoring stopped. There was also a slight increase in
tile flow in 1979-1980 compared to previous years. It would appear that
-------
-28-
Table 3. Precipitation and flow from Defiance watersheds and
Hoytville plots (1978-1980).
1978
Flow Ppt
111
401
402
501
502
641/661t
642/662
611/651
621/671
622/672
631/681
23.5
13.8
12.2
52.1
3.7
19.9
38.9
28.4
25.1
22.5
29.1
70.9
67.5
67.5
61.6
61.6
65.8
65.8
65.8
65.8
65.8
65.8
1979
Flow Ppt
71.5 87.6
17.0
9.7
4.2
13.7
23.5
51.9'
26.0
30.2
32.4
25.5
83.1
83.1
89.8
89.8
98.7
98.7
98.7
98.7
98.7
98.7
1980*
Flow Ppt
27.7
9.7
6.4
0.0
8.5
6.3
16.8
6.8
3.5
16.5
7.8
31.4
32.4
32.4
28.2
28.2
28.6
28.6
28.6
28.6
28.6
28.6
* Through May 31.
t Mean of duplicate plots.
-------
-29-
this large of a decrease in surface runoff must be due to increased moisture
removal with the fall-seeded wheat crop in addition to any increased
infiltration capacity that the increased vegetative cover might have
provided in the winter-spring runoff period.
In 1978-1980, half of the Hoytville plots were tile drained, while
the other half had only surface drainage. There were no significant
differences in surface runoff as a result of tile drainage, with either no
till or fall plowing. There were also no significant differences in either
surface runoff or tile flow between no till and fall plowing.
2.2 Soil and Nutrient Losses (1978-1980
2.21 Hammersmith Roselms (111)
Table 4 gives the mean annual soil and nutrient losses for this site
for 1978-80. This soil is on moderate to steep slopes (2-15%) and is high
in clay. In 1975-77, soil loss varied from 1284 to 3714 kg/ha. Loss in
1978 was similar to this, but 1979 and 1980 losses were higher, especially
the 1980 soil loss which was only for the period January-May. In fall
1978, an attempt was made to establish a grassed waterway in the natural
draw which drains this watershed (Figure 3). The fescue stand was only
partly established by spring of 1979, and was reseeded in fall 1979 and
again in spring 1980. In addition, wheat was seeded on this watershed in
fall 1979 and the wheat was seeded across the waterway to increase
vegetative cover. By winter 1979, there was an adequate stand of
wheat and fescue in the waterway. However, neither the grassed waterway
nor the wheat crop had any effect on soil loss; in fact soil loss increased
in 1979-80.
-------
Table 4. Concentrations and pollutant loads from Hammersmith Roselms (111) surface runoff.
Sediment
1978
1979
1980
Concentration (ug/ml) Load Concentration (ug/ml) Load Concentration (ug/ml) Load
High Low FWM* (kg/ha) High Low FWM* (kg/ha) High Low FWM* (kg/ha)
'3741
16 6t,7
1595 5100
0
75-5
4801 5133
868
2375 5848
Filtered
reactive-F
Total-P
(Nitrate
+ nitrite) -N
Ammonia-N
0.
2.
2.
0.
47
15
3
6
0.
0.
1.
0.
02
00
1
0
0.09
1.37
2.0
0.2
0.
2.
4.
<0.
18
86
3
1
0.15
2.05
12.0
1.0
0.
0.
0.
0.
00
16
0
0
0.
0.
1.
0.
02
41
2
1
0.
2.
7.
0.
09 0.95 0.04 0.35 0.86
59 u,
o
2 13.3 3.3 5.4 13.3
1 0.6 0.0 0.3 0.2
*Flow weighted mean concentration (FWM)
-------
-31-
Nitrogen and phosphorus losses were similar to those in 1975-77 and
were quite low. This watershed has been in soybeans and wheat for the six
years of the study and has received no nitrogen fertilizer in this period
and very little P and K.
2.22 Heisler Blount (401, 402)
Mean annual soil and nutrient losses in surface runoff (401) and tile
drainage (402) are given in Tables 5 and 6, respectively. Soil losses in
1975-1977 varied from 890-3400 kg/ha. In 1978, no till soybeans were grown
in a mixture of soybean and killed quackgrass sod (surface residue was
> 50%). No till reduced soil loss to 66 kg/ha, essentially zero at the
level of detection of this study. As discussed previously, this reduction
was not a result of runoff volume, which was not greatly different than
previous or subsequent years, but was due to the greater protection of the
soil surface by the residue cover. In 1975-77, total P loads varied from
1.14 to 2.33 kg/ha, and in 1978 this was reduced to 0.37 kg/ha as sediment
load was reduced. However, filtered reactive P (FRP) loads were unchanged
with no till. Loads ranged'from 0.02-0.08 kg/ha in 1975-1977 when the water-
shed was plowed, and was 0.08 kg/ha with no till in 1978.
2.23 Speiser Paulding (501, 502)
Nutrient and sediment losses for 1978-1980 are given in Tables 7 and 8.
In the period 1975-1977 this watershed had the highest soil loss of all
sites studied and ranged from 3849 to 4576 kg/ha/yr. Soil was fall plowed
and soybeans were grown in these three years, and the high soil loss was
attributed to the high clay content and poor structure of this soil together
with the lack of subsurface drainage (Logan and Stiefel, 1979). As previously
discussed, oats were grown in 1978 followed by wheat in 1979 and 1980.
Table 3 showed that runoff volume in 1978 was similar to previous years, but
-------
Table 5. Concentrations and pollutant loads from Heisler Blount (401) surface runoff.
1978 1979 1980
Concentration (ug/ml) Load Concentration (ug/ml) Load Concentration (ug/ral) Load
High Low FWM* (kg/ha) High Low FWM* (kg/ha) High Low FWM* (kg/ha)
Sediment
95 0 53 66 1184 48 390 515 1365 511 1265 1092
Filtered
reactive-P
Total-P
(Nitrate
+ nitrite) -N
Ammonia-N
0.13
0.37
4.8
0.9
0.04
0.16
0.9
0.0
0.07
0.30
2.2
0.5
0.08
0.37
2,7
0.1
0.07
0.51
8.0
0.0
0.
0.
2,
0.
00
26
3
0
0.03
0.31
5.0
0.0
0
0
7
0
.04 0.72 0.04 0.49 0.43
.47
.5 0.0 0.0 0.0 0.0
.0 0.0 0.0 0.0 0.0
i
UJ
M
*Flow weighted mean concentration (FWM)
-------
Table 6.. Concentrations and pollutant loads from Heisler Blount (402) tile drainage.
1978
1979
1980
Concentration (ug/ml)
Sediment
Filtered
react ive-P
Total-P
(Nitrate
+ nitrite) -N
Ammonia-N
High
66
0.28
0.79 .
35.8
0.9
Low FWM*
0 34
0.01 0.09
0.10 0.30
0.0 8.3
0.0 0.2
Load
(kg/ha)
37
0.10
0.32
8.9
0.2
Concentration (ug/ml)
High
222
0.20
0.56
27.0
0.6
Low FWM*
0 131
0.00 0.08
0.00 0.27
0.9 13.9
0.0 0.6
Load Concentration (ug/ml) Load
(kg/ha) High Low FWM* (kg/ha)
112 197 0 73 41
0.07 0.14 0.02 0.07 0.04
0.23
11.9 18.0 1.7 13.7 7.7
<0.1 0.8 0.0 0.4 0.1
u>
w
*Flow weighted mean concentration (FWM)
-------
Table 7. Concentrations and pollutant loads from Speiser Pauldlng (501) surface runoff.
1978
1979
1980
Sediment
Filtered
reactlve-P
Total-P
(Nitrate
+ nitrite)-N
AmmoniaN
Concentration (ug/ml) Load Concentration (ug/ml) Load Concentration (ug/ml) Load
High Low FWM* (kg/ha) High Low FWM* (kg/ha) High Low FWM* (kg/ha)
2519 0 946 4388 1733 381 1204
0.97 0.02 0.57 2.65 0.10
6.35.. 0.31 3.35 15.50 1.89
2.3 0.0 0.6 2.6 0.0 0.0 0.0
1.5 0.0 0.1 0.7 0.0 0.0 0.0
0.00 0.03
0.92 1.49
453
0.01
0.56
0.0
0.0
i
o
£
*Flow weighted mean concentration (FWM)
tThere was no flow in 1980.
-------
Table 8. Concentrations and pollutant loads from Speiser Paulding (502) tile drainage.
1978
1979
1980
Concentration (ug/ml)
Sediment
Filtered
reactive-P
Tbtal-P
(Nitrate
+ nitrite) -N
Ammonia-N
High
847
1.31
1.63
27.8
1.5
Low FWM*
0 289
0.03 0.18
0.11 0.47
1.3 5.6
0.0 0.6
Load Concentration (ug/ml)
(kg/ha) High
94 510
0.06 0.08
0.15 1.33
1.8 23.8
<0.1 0.0
Low
0
0.00
0.00
0.6
0.0
FWM*
133
0.03
0.24
13.0
0.0
Load Concentration (ug/ml) Load
(kg/ha) High Low FWM* (kg/ha)
163 247 0 138 104
0.03 0.15 0.01 0.06 0.04
0.29
15.9 14.0 3.0 8.6 6.5
0.0 0.0 0.0 0.0 0.0
i
u
v
*Flow weighted mean concentration (FWM)
-------
-36-
was drastically reduced in 1979 and stopped completely in 1980 (through
May when sampling was terminated). These reductions were in spite of near
normal precipitation. The reduced runoff greatly reduced soil and nutrient
loss in 1979. The very high total P load in 1978 was associated with events
in March and April and were, therefore, not affected by the fertilizer
application in May after the oats were planted. Total P loads were also
high in 1976 (A.02 kg/ha) and 1977 (6.89 kg/ha) and these may reflect the
higher P content of clay-sized soil particles compared to coarser particles.
2.24 The Hoytville Plots (611-681. 612-682)
In the first three years of this study, surface runoff and tile
drainage were monitored on the eight plots with varying degrees of tillage
ranging from fall moldboard plowing to no till. The 1975-1977 data showed
that surface runoff volume was much lower than tile flows on this Hoytville
soil, and as a consequence, soil loss never exceeded 750 kg/ha/yr. Tillage
had no effect on soil loss at these low levels. In 1978, the treatments
on the plots were changed. The tillage treatments were reduced to two: fall
moldboard plow and no till, and on half of the plots the tile drainage was
stopped and surface runoff was the only means of drainage. The results are
summarized in Tables 9-14.
In 1978-1980, tile drainage had no effect on surface runoff volume,
as has already been shown (Table 3), and Tables 9, 11, 12 and 14 show that
soil loss remained very low (< 300 kg/ha/yr), regardless of drainage or
tillage.
In 1975-1977, soybeans were grown and no nitrogen fertilizer was added.
In this period, N03~N in runoff was < 5 kg/ha and < 21 kg/ha in tile
drainage. NH^-N loads were generally < 1 kg/ha. In 1978 and 1979, corn
-------
Table 9. Concentrations and pollutant loads in runoff from Hoytville plots (621, 671). Plots were
no tilled and tile drained. Mean of two plots.
1978 1979 198Q
Concentration (ug/ml) Load Concentration (ug/ml) Load Concentration (ug/ml) Load
High Low FWM* (kg/ha) High Low FWM* (kg/ha) High Low FWM* (kg/ha)
Sediment
2927
0
57
94 1647
0
45
101
111
48
15
Filtered
reactive-P
Total-P
(Nitrate
+ nitrite) -N
Anmonia-N
1.87
5.45_
31.9
6.5
0.00
0.00
0.0
0.0
0.48
0.78
2.9
0.1
1.37
1.87
8.1
0.1
5
7
12
2
.02
.09
.7
.0
0.01
0.11
0.0
0.0
1.79
2.28
1.9
0.5
4
5
4
0
.76 2.24 0.07 1.00 0.30
e 92
.6 4.8 0.0 1.8 0.5
.9 0.6 0.0 <0.1 <0.1
i
u>
i
*Flow weighted mean concentration (FWM)
-------
Table 10. Concentrations and pollutant loads in tile drainage from Hoytville plots (622, 672). Plots were no
tilled and tile drained. Mean of- two plots.
1978 1979
Concentration (ug/ml) Load Concentration (ug/ml) Load
High Low FWM* (kg/ha) High Low FWM* (kg/ha)
1980
Concentration (ug/ml) Load
High Low FWM* (kg/ha)
Sediment
126
18 377
0
12
35
45
0
Filtered
reactive-P
T6tal-P
(Nitrate
-1- nitrite) -N
Amraonia-N
1
1
37
2
.04
.68 ..
.1
.0
0.00
0.00
0.0
0.0
0.10
0.22
10.5
0.1
0.21
0.47
21.9
0.1
1.23
1.63
21.2
1.4
0.00
0.00
0.0
0.0
0.19
0.25
4.7
0.4
0.52
0.72
-
13.4
0.1
0.42 0.00 0.05 0.07
__ ~
5.3 0.0 2.5 3.7
1.9 0.0 0.2 0.1
i
u>
. oo
I
*Flow weighted mean concentration (FWM)
-------
Table 11. Concentrations and pollutant loads in runoff from Hoytville plots (631, 681). Plots were no tilled with
no tile drainage. Mean of two plots-*
1978 1979 1980
Concentration (ug/ml) Load Concentration (ug/ml) Load Concentration (ug/ml) Load
High Low FWM* (kg/ha) High Low FWM* (kg/ha) High Low FWM* (kg/ha)
Sediment 1373 0 29 55 1877 0 118 164 6467 0 267 72
Filtered
react ive-P
Total-P
(Nitrate
+ nitrite)-N
Ammonia N
1.74
4.46
21.9
6.5
0.04
0.05
0.0
0.0
0.57
0.83
2.1
0.2
1.79
2.38
5.9
0.4
5.38
7.34
15.1
4.1
0.01
0.00
0.0
0.0
0.95
1.48
2.3
0.6
3.
4.
4.
0.
01 2.34
30
3 6.2
-
7 1.6
0.01 0.35 0.29
0.0 1.3 0.8
"X
0.0 0.1 <0.1
1
1
*Flow weighted mean concentration (FWM)
-------
Table 12. Concentrations and pollutant loads In runoff from Hoytvllle plots (641, 661). Plots were
fall plowed and tile drainedf" Mean of two plots.
1978 1979 1980
Concentration (ug/ml) Load Concentration (ug/ml) Load Concentration (ug/ml) Load
High Low FWM* (kg/ha) High Low FWtf* (kg/ha) High Low FWM* (kg/ha)
Sediment 1303
78 135 1246 0 54 111 838 0 91 37
Filtered
reactlve-P
Total-P
(Nitrate
+ nltrlte)-N
AnnnonlaN
0.98
4.6J
19.0
3.6
0.02
0.00
0.0
0.0
0.25
0.62
3.2
0.1
0.44
1.11
5.8
0.2
1.84
3.20
13.6
1.4
0.00
0.00
0.0
0.0
0.28
0.67
3.4
0.4
0.
1.
7.
0.
59 0.52
39
2 7.2
4 1.4
0.00 0.15
__ .
0.0 2.1
0.0 <0.1
0.07
1.2
<0.1
*Flow weighted mean concentration (FWM)
-------
Table 13. Concentrations and pollutant loads In tile drainage from Hoytville plots (642, 662). Plots were fall
plowed and tile drained. Mean of two plots.
1978 1979 198Q
Concentration (ug/tnl) Load Concentration (ug/ml) Load Concentration (ug/ml) Load
High Low FWM* (kg/ha) High Low FWM* (kg/ha) High Low FWM* (kg/ha)
Sediment
100
16
77
15
72
123
0
11
17
Filtered
reactive-P
Tbtal-P
(Nitrate
+ nitrite)-N
Ammonia-N
1.12
3.30.
20.9
7.9
0.00
0.00
0.0
0.0
0.16
0.28
8.2
0.2
0.57
1.39
27.1
0.6
0.46
0.42
20.0
1.0
0.00 0.12
0.00 0.15
0.0 3.9
0.0 <0.1
0.49 0.34 0.00 0.07 0.10
0.62
i
17.1 5,9 0.0 2.5 3.9 7
<0.1 1.9 0.0 0.2 0.1
*Flow weighted mean concentration (FWM)
-------
Sediment
1979
Hlgh
1443
63
153 2274
0
122
307
997
0
37
Filtered
reactive-P
Total-P
(Nitrate
+ nitrite) -N
Ammonia-N
1,
3
26
5
.87
15 ..
.4
.6
0.00
0.02
0.0
0.0
0.20
0.46
4.6
0.1
0.50
1.09
10.6
0.2
2.31
4.25
13.3
0.9
0.00
0.00
0.0
0.0
0
0
3
<0
.25
.61
.6
'l
0.55
1.40
8.6
1.05 u.uu v.j.v
8.7 0.0 1.7
1.6 0.0 <0.1
.L \S ~
*Flow weighted mean concentration (FWM)
23
0.06
LI
fO
I
-------
-43-
was grown with nitrogen fertilizer (anhydrous NH-) applications of 180 kgN/ha.
In these two years NO-j-N loads in runoff were < 11 kg/ha and ranged from
13-27 kgN/ha in tile drainage. This indicates small but not significant
increases in nitrate losses with fertilized corn as compared to nonfertilized
soybeans. In 1980, soybeans were again grown and N03~N loads decreased
slightly.
\
Total phosphorus loads have increased steadily since 1975 with increased
fertilization of these plots. Table 15 gives concentrations and losses of
total and filtered reactive P in runoff from fall-plowed and no till plots
since 1975. Also given are the P fertilizer applications during that period,
and the increase in Bray PI extractable P which is a measure of crop-
available phosphate. Flow weighted mean (FWM) concentrations of filtered
reactive P (FRP) remained steady from 1975-1979 and decreased in 1980 after
fall application of P fertilizer was stopped on the fall-plowed plots, while
unit area loads increased through 1979 and then decreased in 1980. Total
P showed the same trend. Bray extractable P increased from 18 yg/g in
1975 prior to P fertilizer application to 47 yg/g in 1979.
On the no till plots, concentrations and loads of total P and FRP
increased dramatically from 1975 to 1979. FRP then decreased rapidly in
1980 after fall P fertilizer applications were terminated. The higher losses
of soluble and total P are due to the accumulation of fall broadcast fertilizer
at the surface. This is indicated by the increase in Bray PI extractable
P in the surface 5 cm of soil frbm 18 yg/g in 1975 prior to P fertlization
to 168 yg/g in 1979. Oloya and Logan (1980) have shown a very high corre-
lation (r2 > 0.98), on this Hoytville soil, between Bray PI extractable P and
P that can desorb into water. They also showed that a large fraction of the
desorbable P was desorbed instantaneously, and this fraction may represent
-------
Table 15. Changes In concentration and unit area loads in runoff of total and filtered reactive phosphate with fertilization of fall-plowed
and no till Hoytvllle soil (1975-1980).
Filtered Reactive P
Tear
1975*
1976
1977
1978
1979
1980
1975
1976
1977
1978
1979
1980
P Fertilizer Applied (kg/ha)
Spring F«U
-
12
12
-
12
12
-
12
12
-
12
12
34
34
19
86
30
-
34
34
19
86
30
-
Concentration (ug/ml) Runoff Load
High Low FWM* (kg/ha/yr)
3.92
0.78
2.18
1.87
2,31
1.0$
3.92
1.79
4.33
1.87
5.38
2.34
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Q.OO
0.04
0.00
0.01
0.01
0.31
0.25
0.29
0.23
0.27
0.12
0.31
0.37
0.82
0.53
1.37
0.67
Fall Plow
0.01
0.13
0.30
0.47
0.57
0.07
Ho Till
0.01
0.22
0.94
1.58
3.89
0.29
Total P
Bray Pit
Concentration (ug/ml) Runoff Load extractable P
High Low FWM* (kg/h«/yr) (jig/g)
5.85
2.95
9.18
4.61
4.25
__
3.96
1.81
10.70
5.45
7.34
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.10
0.00
0.00
0.57
0.43
0.82
0.54
0.64
1.03
0.54
1.46
0.81
1.88
__
0.10 18.1
0.22
0.84
1.09
1.40 46.6
0.26 18. 1
0.34
1.56
2.13
5.11 168.0
-_
* Flow weighted mean concentration (FWM).
t Sampled from the 0-5 cm depth.
$ Plot area was In sod prior to 1975.
-------
-45-
unreacted fertilizer P, soluble reaction products of soil and fertilizer,
and P in solution in soil pores. The 1980 FRP data show that this effect
of fertilization on dissolved P losses may diminish rapidly after P
fertilizer is no longer broadcast.
2.3 Seasonal Trends of Precipitation, Flow and Soil Loss (1975-1980)
Figures 14-17 give precipitation, flow and soil loss by month for the
period 1975-1980 for the Roselms (111), Blount (401, 402) and Paulding
(501, 502) watersheds and for one of the Hoytville plots (621, 622). On
the Roselms watershed, greatest runoff and soil loss were in the period
February-April. Runoff was minimal in the summer months and soil loss was
low even when monthly precipitation was high.
On the Blount watershed (401, 402), total flow (runoff and tile drainage)
was highest in February-April also, and soil loss generally corresponded to
runoff volume. In 1978 with no till, soil loss was greatly reduced although
runoff and tile flow volumes were similar to other years. There was very
little runoff or soil loss in the summer or early fall months.
Runoff continued to be highest in the early spring on the Paulding
watershed (501) as it was on the other watersheds. With the exception of
September 1975, summer runoff was low as was soil loss even though summer
rainfall was often as high as in spring months. In 1979 and 1980 with wheat
crops, runoff was low or absent while tile flow was somewhat higher than
in other years.
Runoff and tile drainage (621, 622) were also higher in the spring on
the Hoytville plot. On this soil, tile flow was often higher than surface
runoff and soil losses were quite low. Unlike the Defiance watersheds, tile
flow was common, although low, in the summer and fall months.
-------
u
2
4
6
c
° 8
p
.2 10
o
w
^ 12
14
16
-
-
-
-
.
-
-
-
-«**
20
ie
16
14
12
"E
o
10
?
o
^ 8
6
4
2
0
*-
^
s
IS,
s ^
1
1
-
K
S 5? M
mfl
(\MJJAS
0
N
P"^^
5;
c
5
K,
1
1
|
1
1
f\.
"1
SsnalT-'
D J F M A M
J J A S
0
1
N D J
rr
1
<0
t\l
5 «vl
IT
S
J\J.
F M A M
It
IL
J"J~T
I
M
$
1
A M
<0
8
F
n§
//
M A "~
Figure 14. Monthly precipitation, flow and soil loss from Roselms watershed (1975-1980),
-------
o
o
'§.
o
v
a
0
Z
4
6
8
10
12
14
16
14
12
10
8
6
4
2
0
O
o
Illlllllllimil
r
>O
V>
r
4h
BLOUNT SOIL
[ Surface runoff
Tile flow
Precipitation
A M J J A SONDJFMAMJJASON 0 J F M A M
1975 1976 1977
J F M A M J JA50NDJFMAMJJ A S 0 N- D
1978
1979
1980
Figure 15. Monthly precipitation, flow and soil loss from Blount watershed (1975-1980).
-------
c
o
o
I
E
o
0
2
4
8
10
12
14
16
22
20
IB
16
14
12
10
e
6
4
2
_
-
-
-
-
I
'
^
_
-
-
TJ-
_
-
<
\
2"
<*
\O
^ 0
A M J J A
1975
s
<0
Q «
*-*
"""I'D
_
S 0 M D
1
1
1
1
Ou 5
K) t\l
^
<§
J F
1
TT
1976
1
^ 5
ft 5:
i
if!
53
S
>c
1977
1
^ _
g
c
IT
mm
!
1
i
1978
1
1
1
'
1
PAUL
Ii i'h
1
.DING SOIL
H Surface runoff
Tile flow
Precipitation
Ol
O
ir>
1 1
N.
i **
II mn
[JIJIII HI
J^IH
lllj|l| ^^^Jll
"» illllllL ill
fc *
* »
imnHHH
oil HI 1111
1 ^ III Illlllll
bill III M
1979 1980
oo
Figure 16. Monthly precipitation, flow and soil loss from Paulding watershed (1975-1980),
-------
0
2
4
6
8
10
12
14
16
18
20
22
14
12
10
8
6
4
2
0
HOYTVILLE SOIL
Surface runoff
Tile flow
Precipitation
1975
NDJFMAMJJASOND
1976
JASON
1977
M J J A SONDJFMAMJJASON 0 J F M A M
1978 '979 I98°
VO
I
Figure 17. Monthly precipitation, flow and soil loss from Hoytville (621, 622) plot (1975-1980).
-------
-50-
2.4 Crop Yields on Hoytville Tillage Plots (1975-1980)
Table 16 presents crop yields for the six years of the study. In the
first three years, soybeans were grown on the plots with different tillage
practices. There were no yield differences between any of the tillage
treatments in any of the three years. The yields obtained each year
were representative of yields for that year in that area. The large
variation in yield between 1975 and the other two years is due to the more
favorable rainfall distribution of that year. Bone et_ al^ (1977) reported
lower soybean yields on Hoytville soil with no till compared to minimum
tillage (plow plant) or conventional tillage (fall moldboard plow). How-
ever, soybeans always followed corn in their study and the large amount of
crop residue provided by the previous corn crop may have contributed to the
yield reduction with no till by keeping the soil wetter and cooler in the
spring. In the study reported here, soybeans followed soybeans, and there
is very little residue from a soybean crop. This may explain why no till
yields were as good as other tillage treatments.
In 1978 and 1979, corn was grown with fall moldboard plow and no till
tillage treatments, and half of the plots had no tile drainage. In 1978,
there were no significant differences in corn yields due to differences
in tillage or drainage. These data confirm the findings of Bone et al
(1977) that no till corn yields are not significantly different than
yields with fall plowing when corn follows soybeans. They also showed
that tile drainage did not affect yields of corn following soybeans.
In 1979, corn yields were generally high because of a favorable growing
season. No till yields were generally lower than fall plowing. These
differences were not statistically significant because there were only two
-------
Table 16 .
-51-
Crop yields (bu/acre) on the Hoytville plots for the period
1975-1980. Mean of two plots.
1975
1976
1977
Mean
Fall Plow
59.7
38.1
41. 4
46. A
1975-1977
Fall Chisel
58.8
35.7
42.1
45.5
Soybeans
Fall Disk* No Till Mean
59.0 60.2
36.1 36.1
46.2 43.0
47.1 46.4
59.4
36.5
43.2
46.4
1978-1979 Corn
1978
1979
Fall
Tile drained
133.5
183.9
Plow
No tile
131.9
174.3
No Till
Tile drained
131.9
119.0
No tile
134.5
152.2
1980 Soybeans
Fall Plow
Tile drained
51.0
No tile
48.4
No Till
Tile drained
37.8
No tile
35.3
*In 1975 and 1976, a 15 cm strip was rototilled for the seedbed. In 1977,
these plots were disked once.
-------
-52-
plots in each treatment. Also, since the plots were established in 1975,
there has been some subsidence on several of the no till plots which
reduced surface runoff and caused some temporary ponding. This probably
overshadowed any effect of tile drainage on corn yields. Bone et al (1977)
found that no till corn following corn gave lower yields than fall plowing,
and although the data reported here for 1979 show no significant differences
due to tillage, the trend is to lower yields with no till.
In 1980, soybeans were grown and the results show that no till yields
were significantly lower than with fall plowing, a finding also reported
by Bone e_t al^ (1977) for soybeans following corn. Tile drainage had no
significant effect on yields.
The yield results from the Hoytville plots and the study of Bone et al
(1977) indicate that no till corn and soybean yields were the same as those
with fall plowing when soybeans was the previous crop, but no till corn and
soybean yields are reduced when corn is the previous crop. This yield
reduction has been attributed to the colder and wetter soil conditions
provided in the spring with the large amount of residue remaining after
harvest of a corn crop.
3. DISCUSSION
Logan and Stiefel (1979) reported on the first three years of this
study (1975-1977). They found that soil losses form Maumee River Basin
soils, although not excessive in terms of soil productivity effects, were
among the highest in the Great Lakes Basin. They showed that the internal
drainage and structural stability of these soils were important in deter-
mining their susceptibility to .erosion and sediment transport. The soils
of the Maumee Basin are all poorly-drained and fine-textured, and Logan and
-------
-53-
Stiefel (1979) found that those soils which could be effectively tile
drained (Hoytville and Lenawee) had little runoff and low soil loss. A
soil of similar slope (< 2%) and texture, Paulding clay, had poor internal
drainage and structure and produced low tile drainage volumes and had the
highest soil loss of the soils that were monitored in the study. Conser-
vation tillage and no till were compared to fall moldboard plowing on the
Hoytville soil but soil loss was so low on this soil that reduced tillage
had no effect on erosion.
Logan and Stiefel (1979) also showed that, in the 1975-1977 period, most
runoff occurred in the period February-May during spring thaw and spring
rains. Occasional summer storms did not produce as much runoff or soil loss
as precipitation in the early spring. This was attributed to the greater
water saturation of the soils in the spring months. A conclusion of this
finding was that soil protection was most needed in the period after crop
harvest in the fall and crop canopy development the following spring and
summer.
Watershed and plot monitoring (Logan and Stiefel, 1979) also showed
that unit area loads of total and filtered reactive phosphate were high compared
to other watersheds in the Great Lakes Basin, and this was attributed to the
high clay content of the soils in the Maumee Basin, the youthful nature
of the soils (^ 8,000 years), and the intensive cultivation and fertilization
of the area (Logan, 1979).
i
The results of the present study confirm some of the previous findings
and also provide new information about soil and nutrient losses from Maumee
Basin agricultural soils.
-------
-54-
Runoff continued to be highest in the early spring months. Soil loss
continued to be highest on the Paulding soil with spring-seeded crops, but
fall-seeded wheat greatly reduced runoff in 1979 and there was no runoff with
wheat in 1980. The effect of the wheat crop appeared to be related to drying
of the soil and increased water storage and infiltration capacity rather than
increased soil protection by the winter cover since runoff volume and not
just soil loss were drastically reduced.
No till soybeans in 1978 dramatically reduced soil loss on the Blount
soil without measurably changing runoff volume. Residue cover of the
previous soybean crop was enhanced by a heavy infestation of quackgrass which
was killed with herbicide prior to no till planting in 1978. Wheat in 1980
had no effect on runoff on Blount or Roselms soils, in contrast to the
runoff reduction on Paulding soil. It should be noted, however, that the
greatest reduction on Paulding occurred in the second year of wheat,
indicating that the effect might be cumulative.
Results from the Hoytville plots showed that tile drainage had no
measurable effect on surface runoff volume. This finding is not clear at
this time. Perhaps tile drainage capacity is not adequate in the early
spring months when most runoff occurs, or perhaps the effect of tile
drainage on surface runoff does not develop for several years.
Phosphate fertilizer broadcast on the Hoytville plots in the fall in
1975-1979 steadily increased flow weighted mean concentrations and annual
loads of total P and filtered reactive P in that period. Concentrations
and loads of FRP decreased rapidly in 1980 after fall fertilization was
terminated. Concentration and load increases were greatest on the no till
plots because broadcast fertilizer remained at the surface where it was
-------
-55-
most susceptible to washoff in dissolved form or attached to soil particles.
Oloya and Logan (1980) found a high correlation between Bray PI extractable P
of this soil and P that could be desorbed into water. In the period 1975-
1979, Bray PI of the 0-5 cm depth was increased from 18 to 168 yg/g soil,
while the same amount of fertilizer applied to the plowed soil prior to
plowing in the fall only increased Bray extractable P to 47 yg/g. This
suggests that, if no till is to be used to control phosphorus losses from
agricultural land (especially dissolved), then fertilizer management is
also required. This would entail keeping available P levels in the soil
no higher than needed for optimum crop production and also would involve
methods to place the fertilizer into the soil rather than on the surface.
-------
-56-
4. LITERATURE CITED
1. Bone, S. W., D. M. Van Doren and G. B. Triplett. 1977. Tillage research
in Ohio. A guide to the selection of profitable tillage systems. Coopera-
tive Extension Service. The Ohio State University. Bull. 620. 12 pp.
2. Corps of Engineers. 1975. Lake Erie Wastewater Management Study.
Preliminary Feasibility Report. Volume 1. Buffalo District, Buffalo,
N.Y.
3. Logan, T. J. 1979. The Maumee River Basin Pilot Watershed Study. Vol.
2. Sediment, phosphates and heavy metal transport. USEPA Region V.
Great Lakes National Program Office. EPA-905/9-79-005-B. 132 pp.
A. Logan, T. J. and R. C. Stiefel. 1979. The Maumee River Basin Pilot
Watershed Study. Vol. 1. Watershed characteristics and pollutant
loadings. USEPA Region V. Great Lakes National Program Office.
EPA-905/9-78-005-A. 135 pp.
5. Oloya, T. 0. and T. J. Logan. 1980. Phosphate desorption from soils and
sediments with varying levels of extractable phosphate. J. Environ.
Qual. 9:526-531.
6. Sonzogni, W. C., T. J. Monteith, W. N. Bach and V. G. Hughes. 1978.
U.S. Great Lakes Tributary Loadings. PLUARG Task D. Great Lakes
Basin Commission, Ann Arbor, Michigan.
-------
TECHNICAL REPORT DATA
(Plcasr reaJ lmuructn>ns or the rr\rn< bc/ort
1 REPORT NO
EPA-905/9-79-005-C
3 RECIPIENT'S ACCESSION'NO.
ITLE AND SUBTITLE
The Maumee River Basin Pilot Watershed Study
6 REPORT DATE
May 1981
6 PERFORMING ORGANIZATION CODE
7 AUTHORiS)
Terry J. Logan
e PERFORMING ORGANIZATION REPORT NO
Volume III
9 PERFORMING ORGANIZATION NAME AND ADDRESS
Ohio State University, Columbus, Ohio 43210
Ohio Agricultural Research and Development Center
Wooster, Ohio 44691
10 PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
Grant R00535301
12 SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agency
Great Lakes National Program Office
536 South Clark Street, Room 958
Chicago, Illinois 60605
13. TYPE OF REPORT AND PERIOD COVERED
Monitoring 1978-1980
14. SPONSORING AGENCY CODE
Great Lakes National Program
Office, U.S. EPA, Region V
15 SUPPLEMENTARY NOTES
Ralph G. Christensen
Project Officer
16. ABSTRACT
The Maumee River was chosen by PLUARG to be one of four pilot watersheds to be
studied on the U.S. side of the Great Lakes drainage basin as part of Task C-
pilot watershed studies.
This report represents the results of the continued monitoring of three of the
Defiance County watersheds and the Hoytville plots for the period 1978-1980.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS-OPEN ENDED TERMS
c. COSATI Field Group
Watershed
Monitored
Soil erosion
Tile drainage
Surface run off
Cropping
FEME
Document is available through the National
Technical Information Service, Springfield
VA 22161
19 SECURITY CLASS (THii Reporij
20 SECURITY CLASS (Ttii: page/
Unclassified
21 NC. OF PAGES
56
22 PRICE
EPA Form 2220-1 (9-73)
US GOVERNMENT PRINTING OFFICE: 1984756-957/442
------- |