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
COG
Maumee River
      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 NO^-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	     ii

ACKNOWLEDGEMENTS	     1;L1

ABS TRACT	     iv

LIST OF TABLES	     vi

LIST OF FIGURES	     viii

1.  INTRODUCTION 	     1

    1.1  Study Approach	     3
    1.2  Study Methods 	     3

         1.21  Monitoring Sites in Defiance County 	     3
         1.22  Surface Runoff and Tile Drainage
               Measurement - Defiance County Sites 	    14
         1.23  Surface Runoff and Tile Drainage
               Measurement - Hoytville Plots 	    21
         1.24  Analysis of Watershed and Plot
               Water Samples	    2^

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 Trends of Precipitation,  Flow and  Soil Loss
         (1975-1980) y  .'..."	     45
    2.4  Crop Yields  on Hoytville Tillage  Plots (1975-1980)  	     50

3.  DISCUSSION	     52

4.  LITERATURE CITED	     56

-------
                                   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	38

     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  	     15

      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_  al, 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 characteristics 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 Pbts
•x— Continuous mass
     transport stations
           10
              IS   20
   j\ Bowling
6*}Green
       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 Slope
Material (%)
Lacustrine Clays 3-16
Lacustrine Clays 1
Lacustrine Clays
Clay Till < 1
Clay Till
Clay Loam Till 3-4
Clay Loam Till
Drainage
Area
(ha)
3.2
1.0
0.1
0.04
0.04
0.9
0.9
Drainage Systems
Monitored
Surface Runoff
Surface Runoff
Subsurface Tile
Surface Runoff
Subsurface Tile
Surface Runoff
Subsurface Tile
                                                                                                                             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-
            «•     -                    5



                                     . t>l

                                    . -^r..
                  •v."^.^-:v^
-------
                                  -9-
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, NWV4
               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-
Flgure 5.  Heisler Blount watershed looking downslope.

-------
                              -12-
                  Speiser Paulding (SOX)
                  Location: Delaware township,
                  T4N, R3E, Sec. 15. SW*/4
                   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-
                                                                 2*" __^*^"     *

                                                              ^•^•'t „•••-'.  >.' .  ••"
                                                              ' *     "  *~*jt3jttl~    :f^£.' jfei
                                                                  •-•' .f ''*' '"'    -I*.-.' -.
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 urn 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 urn filter was effective




in retaining fine clay.  The filtered sample was routinely analyzed for:




(NO., + NO^), NH3, 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 NH^ 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 arid 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).

Ill
401
402
501
502
641/661t
642/662
611/651
621/671
622/672
631/681
Flow
23.5
13.8
12.2
52.1
3.7
19.9
38.9
28.4
25.1
22.5
29.1
1978
Ppt
70.9
67.5
67.5
61.6
61.6
65.8
65.8
65.8
65.8
65.8
65.8
Flow
71.5
17.0
9.7
4.2
13.7
23.5
51.9'
26.0
30.2
32.4
25.5
1979
Ppt
87.6
83.1
83.1
89.8
89.8
98.7
98.7
98.7
98.7
98.7
98.7
Flow
27.7
9.7
6.4
0.0
8.5
6.3
16.8
6.8
3.5
16.5
7.8
1980*
Ppt
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.
                        1978
1979
                                                                                             1980
Concentration (ug/ml) Load Concentration (ug/ml) Load Concentration (ug/ml) Load
High Low FWM* (kg/ha) High
Sediment 3741 16 6f>7 1595 5100
Filtered
reactive-F 0.47 0.02 0.09 0.18 0.15
Total-P 2.15 0.00 1.37 2.86 2.05
(Nitrate
+ nitrite)-N 2-3 1-1 2.0 4.3 12.0
Ammonia-N °-6 °-° °-2 <0.1 1.0
Low FWM* (kg/ha) High Low FWM* (kg/ha)
0 755 4801 5133 868 2375 5848
0.00 0.02 0.09 0.95 0.04 0.35 0.86
0.16 0.41 2.59 — — — — i
o
i
0.0 1.2 7.2 13.3 3.3 5.4 13.3
0.0 0.1 0.1 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/ml)    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
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

0.00
0.26




2.3
0.0

0
0




5
0

.03
.31




.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
bJ
ho



*Flow weighted mean concentration  (FWM)

-------
 Table 6.,  Concentrations  and  pollutant loads from Heisler Blount (402) tile drainage.
Sediment
                       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)
                 66
                           0
34
37    222
                                                                    131
                                                                             112
                                                    197
                                                                                              0
                                                                                                      73
                                                                                                             41
Filtered
reactive-P
Total-P

(Nitrate
+ nitrite) -N
Ammonia-N
0.28
0.79 .

35.8
0.9
0.01
0.10

0.0
0.0
0
0

8
0
.09
.30

.3
.2
0.
0.

8.
0.
10
32

9
2
0.20
0.56

27.0
0.6
0.00
0.00

0.9
0.0
0.08
0.27

13.9
0.6
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


i
i

*Flow weighted mean concentration  (FWM)

-------
 Ta"ble 7. Concentrations and pollutant loads from Speiser Paulding (501)  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)
2519
0
946
4388  1733
381
                                                                    1204
453
Filtered
reactive-P
Total-P

(Nitrate
+ nitrite) -N
Ammonia— N

0.
6.

2.
1.

97
35.

3
5

0.02
0.31

0.0
0.0

0.57
3.35

0.6
0.1

2.65
15.50

2.6
0.7

0.
1.

0.
0.

10
89

0
0

0.00
0.92

0.0
0.0

0.
1.

0.
0.

03
49

0
0

0.01
0.56
i
0.0 — — -- -- *,
0.0
*Flow weighted mean concentration  (FWM)
tThere was no flow in 1980.

-------
Table 8.  Concentrations and pollutant  loads  from Speiser  Paulding  (502)  tile  drainage.
 Sediment
                        1978                              1979                                1980



                 Concentration   (ug/ral)   Load    Concentration  (ug/ml)    Load     Concentration  (ug/ml)   Load



                 High     Low    FWM*     (kg/ha)  High     Low      FWM*   (kg/ha)    High     Low     FWM*   (kg/ha)
847
                            0
289
94
510
                                                                     133
                                                            163      247
                                                                                                0
                                                                     138     104
Filtered
reactive-P
Tbtal-P

(Nitrate
+ nitrite) -N
Ammonia-N
1.31
1.63

27.8
1.5
0.03
0.11

1.3
0.0
0.18
0.47

5.6
0.6
0.06 0.08
0.15 1.33

1.8 23.8
<0.1 0.0
0.00
0.00

0.6
0.0
0.03
0.24

13.0
0.0
0.03 0.15 0.01 0.06 0.04
0.29

15.9 14.0 3.0 8.6 6.5
0.0 0.6 0.0 0.0 0.0


i
u
i

 *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
1980
Concentration

Sediment
Filtered
reactive-P
Total-P
(Nitrate
+ nitrite) -N
Ammonia-N
High
2927
1.87
5'45.
31.9
6.5
Low
0
0.00
0.00
0.0
0.0
(ug/ml)
FWM*
57
0.48
0.78
2.9
0.1
Load
(kg/ha)
94
1.37
1.87
8.1
0.1
Concentration
High
1647
5.02
7.09
12.7
2.0
Low
0
0.01
0.11
0.0
0.0
(ug/ml)
FWM*
45
1.79
2.28
1.9
0.5
Load
(kg/ha)
101
4.76
5.92
4.6
0.9
Concentration (ug/ml) Load
High Low FWM* (kg/ha)
111 0 48 15
2.24 0.07 1.00 0.30
—
i
4.8 0.0 1.8 0.5 V
0.6 0.0 <0.1 <0.1
*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                               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
126
0
                                           18     377
                                            0
                                          12
                                                                            35
                                                                    45
                                                                   0
Filtered
reactive-P
Tdtal-P

(Nitrate
+ nitrite) -H
Airanonia-N
1.04
1.68 ..

37.1
2.0
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 0.42 0.00 0.05 0.07
0.72
I
13.4 5.3 0.0 2.5 3.7 -J
0.1 1.9 0.0 0.2 0.1
*Flow weighted mean concentration  (FWM)

-------
Table 11..Concentrations and pollutant loads  In runoff  from Hoytvllle  plots  (631,  681).  Plots were no tilled with
          no tile drainage.   Mean of two  plots-.
  Sediment
                         1978                              1979                                1980

                  Concentration  (ug/ml)   Load    Concentration   (ug/ml)    Load     Concentration  (ug/tnl)   Load

                  High     Low    FWM*    (kg/ha)  High     Low     FWM*    (kg/ha)    High     Low     FWM*   (kg/ha)
1373
0
                                   29
55
1877
0
                                                                     118
                                                            164
                                                          6467
                                                           267
  *Flow weighted mean concentration (FWM)
                                                                                                                72
Filtered
reactive-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
0.0 0.1 <0.1


t
i


-------
Table 12.  Concentrations and pollutant loads in runoff from Hoytville plots (641, 661).  Plots were
           fall plowed and tile drainedr* Mean of two plots.


                       1978                               1979                                1980
Concentration (ug/ml)

Sediment
Filtered
reactive-P
Total-P
(Nitrate
+ nitrite)-N
Aimoonia-N
High
1303
0.98
4.61
19.0
3.6
Low FWM*
0 78
0.02 0.25
0.00 0.62
0.0 3.2
0.0 0.1
Load
(kg/ha)
135
0.44
1.11
5.8
0.2
Concentration (ug/ml)
High
1246
1.84
3.20
13.6
1.4
Low FWM*
0 54
0.00 0.28
0.00 0.67
0.0 3.4
0.0 0.4
Load Concentration (ug/ml) Load
(kg/ha) High
111 838
0.59 0.52
1.39
7.2 7.2
0.4 1.4
Low FWM* (kg/ha)
0 91 37
0.00 0.15 0.07
— — —
0.0 2.1 1.2
0.0 <0.1 <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.
Sediment

Filtered
reactive-P

Tbtal-P

(Nitrate
+ nitrite)-N

Ammonia-N
                       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)
  100


 1.12

 3.30,


20.9

 7.9
 0      4


0.00   0.16

0.00   0.28


0.0    8.2

0.0    0.2
  16
77
 0.57   0.46

 1.39   0.42


27.1   20.0

 0.6    1.0
         15
72
123
0
0.0     3.9     17.1     5,9

0.0    <0.1     <0.1     1.9
11
17
         0.00    0.12     0.49    0.34     0.00    0.07    0.10

         0.00    0.15     0.62
                                           0.0     2.5    J.9

                                           0.0     0.2     0.1
*Flow weighted mean concentration (FWM)

-------
Table „   concentration, an, poUutant
Table 14.               o tlje dralnage.  Mean of two plots.
                                                                                '  «>
 Sediment
   ..  MT.


Concentration  (ugM»   Load    Concentration



Hlgh     Lov
                                                                                   Concentration  (ug/*l)   L-d



                                                                                   High     Lev
1443
                                 63
                                           153   2274
                                           0
                                                                   122
                                                           307
                                                                                    997
Filtered
reactive-P
Total-P
(Nitrate
-1- nitrite) -N

Ammonia-N
1.87
3.15 ..
26.4

5.6
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
0.25
0.61
3.6
1
<0 . ]_
0.55 1.05 0.00
1.40
8.6 8.7 0.0
<0.1 1.6 0.0

                                                                                                     37
                                                                                                     1.7
                                                                                              23
                                                                                             1.1
NJ

I
 *Flow weighted mean concentration (FWM)

-------
                                  -43-
was grown with nitrogen fertilizer (anhydrous NH,) applications of 180 kgN/ha.



In these two years NO^-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 noiyfertilized



soybeans.  In 1980, soybeans were again grown and NC^-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 ££>.  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 (1975-1960).
Year
1975*
1976
1977
1978
1979
1980

1975
1976
1977
1978
1979
1980
P Fertilizer Applied (kg/ha)
Spring Fall
—
12
12
-
12
12

-
12
12
-
12
12
34
34
19
86
30
-

34
34
19
86
30
-
Filtered Reactive P
Concentration (ug/ml) Runoff Load
High Lou FWM* (kg/ha/yr)
3.92
0.78
2.18
1.87
2.31
1.05

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
0.00
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
No Till
0.01
0.22
0.94
1.58
3.89
0.29
Total P
Concentration dig/ml) Runoff Load
High Low FWM* (kg/h«/yr)
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
0.22
0.84
1.09
1.40
—

0.26
0.34
1.56
2.13
5.11
..
Bray Pit
extractable P
(vg/g)
18.1



46.6


18.1



168.0

* Flow weighted mean concentration (FWM).
t Sampled from the 0-5 cm depth.
$ Plot area was in aod 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
.2 e
o
'£• 10
u
u
^ 12
14
16
20"

16

16
14
12
o
- 10
3
i- 8
6


4
2
0
-
-
-

-

.


-














r*









„
<
O
§ 1

• m
i til i i A r*
1
1























&
(-* hi












•












r\
1
1




1











^













O
«










f-






^
—










Ot
<*>

























3










I
'























§


























"1

























1 1
1



































8
<0


^
ff






«8

— 11
























-r/f




























































?;




1




























^


















































1
•
























1
1























P
1















1
I









1
1




















I
1





























1










0 ROSELMS
'1










1



























j










SOIL
A ( I Surfoce












\
>


1
U Tile
runoff

flow
IMH Precipitation





IS
'O






t^ fti
^



s;






















n










?*








h
^




S!
p—










&
^




»>


















L.













1
•£
O
1










                                                      MAMJJ  ASONDJ FMAMJ  JASOND  J F
                                                            1978                   1979            1980
Figure 14.  Monthly precipitation,  flow and  soil  loss from Roselms watershed (1975-1980).

-------
-6
u
u
O)
0.  12 -
 o
                                                                                                      BLOUNT  SOIL
                                                                                                          Surface runoff

                                                                                                          Tile flow

                                                                                                          Precipitation
                5 0 N D J F  M AMJJASONDJFM  AM  "  J  F
   JASONOJFVAMJJA  SON- D  J


1978                     '979              1980
                                                            I
                                                            x>
    Figure 15.   Monthly  precipitation, flow  and soil  loss from Blount watershed  (1975-1980)

-------
0
2
4
.2 8
o
a 10
'o
i 12
14
16


-
-
~_

-

-








-



















\










-*"
22
20
18
16

14
— 12
e
u
~ 10
o
"- 8
6
4
2
0
*•
-
-
.

-
o
_

-
«;
\
O.
r^
r







§
0)
k








* 1
3_^-


2L

AMJJASONDJ F
1975




MAM























i























1.
1976

1

•



















i

























1 '























































r 5







>



§


-











8

1977




























r~j

























S j
tl









nnrim

!
i









,

































































































1

1







1

























1







Ot
IT)


^ 5




1

1




























1








































!• • • • mih

1








PflULDING SOIL
Hsurfoco runoff
Tile flow
••• Precipitation








N.
<0
1








..
•Lfl!
^ ' JFMAMJJA SOND JFMAMJJ ASUNUJ KVUIVI
1978 1979 1980
                                                                                                                 00
Figure 16.  Monthly precipitation, flow and soil loss from Paulding watershed  (1975-1980)

-------
                                                                                                                    SurfocB runoff
                                                                                                                    Tile flow
                                                                                                                    Precipitation
                                                                                        6 N D  J  F M   M  J  J  A  S 0 N 0 J
M A M J  J O  b
      1977
                       j T MAMJJASONl)
                                                                                                                                   vO
         1975
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 aJL  (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 aJL




(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.
1975-1977 Soybeans
1975
1976
1977
Mean
1978
1979
Fall Plow
59.7
38.1
41.4
46.4
Fall
Tile drained
133.5
183.9
Fall Chisel
Fall Disk* No Till Mean
58.8 59.0 60.2
35.7 36.1 36.1
42.1 46.2 43.0
45.5 47.1 46.4
1978-1979 Corn
Plow
No tile
131.9
174.3
1980
Fall Plow

Tile drained
51.0
No tile
48.4
No Till
Tile drained
131.9
119.0
Soybeans
No Till
Tile drained
37.8
59.4
36.5
43.2
46.4
No tile
134.5
152.2
No tile
35.3
*ln 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 et_ 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).
                                 •
     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 ug/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
                                 ffoJ IHUTUC lions on the rorm hrfon
  REPORT NO
   EPA-905/9-79-005-C
              3 RECIPIENT'S ACCESSION-NO.
  TITLE AND SUBT ITLE
   The Maumee River Basin Pilot  Watershed Study
                                                             5 REPORT DATE

                                                                 May 1981
                                                             6 PERFORMING ORGANIZATION CODE
7 AUTHORiS)
   Terry  J.  Logan
              B 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 Prograir
                  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.
17
                                 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
   Document is available through the National
   Technical Information Service, Springfield
   VA  22161
                                               19 SECURITY CLASS /This Reporij
20 SECURITY CLASS (Till! pafc/

  Unclassified
                            21. NO. OF PAGES
                               56
22. PRICE
EPA Form 2220-1 (9-73)
  US GOVERNMENT PRINTING OFFICE 1984—756-957/442

-------

-------