United States
         Environmental Protection
         Agency
Great Lakes National    EPA-905/9-81-003
Program Office      May, 1981
536 South Clark Street, Room 932
Chicago, IL 60605
v>EPA
environmental impact
of land use
on water quality
                          Final Report on
                          the Black Creek
                          Project
                          Phase II

-------
                                      FOREWORD

The U.S. Environmnental  Protection Agency was created because of increasing
public and governmental  concern about the dangers of pollution to the
health and welfare of the American people.  Noxious air, foul water, and
spoiled land are tragic  testimony to the deterioration of our natural
environment.

The Great Lakes National Program Office(GLNPO) of the U.S. EPA was
established in Region V, Chicago, to provide specific focus on the water
quality concerns of the Great Lakes.  The Section 108(a) Demonstration
Grant Program of the Clean Water Act(PL 92-500) is specific to the Great
Lakes drainage basin and thus is administered by the Great Lakes National
Program Office.

Several sediment erosion-control projects within the Great Lakes drainage
basin have  been funded as a result of Section 108(a).  This report describes
one such project supported by this office to carry out our responsibility
to improve  water quality in the Great Lakes.

We hope the information and data contained herein will help planners and
managers of pollution control agencies to make better decisions in
carrying forward their pollution control responsibilities.


                                         Madonna F. McGrath
                                         Director
                                         Great Lakes National Program Office

-------
                                        EPA-905/9-81-0'
                                        May, 1981
 ENVIRONMENTAL IMPACT OF

          LAND USE ON

         WATER QUALITY

               Final Report
                On the
            Black Creek Project
                Phase II

               Prepared for

      U.S. ENVIRONMENTAL

      PROTECTION AGENCY

      Great Lakes National Program Office
           536 South Clark Street
           Chicago, Illinois 60605

RALPH CHRISTENSEN          CARL D. WILSON
Section  108a Program             Project Officer
     UNDER U.S. EPA GRANT NO. G 005335
                 To

ALLEN COUNTY SOIL & WATER

   CONSERVATION DISTRICT

     Purdue University, University of Illinois

-------
                              TABLE OF CONTENTS

                                                                         Page

FINAL REPORT - BLACK CREEK II
    J.B. Morrison	     1

WATER QUALITY: SEDIMENT AND NUTRIENT  LOADINGS  FROM CROPLAND
    D.W. Nelson, D.B. Beasley, E.J. Monke, R.A.  Dorich	    11

MAINTENANCE OF BMPs
    R.Z. Wheaton	    57

FIELD EXPERIENCES AND PROBLEMS
    R.E. Land	    59

EVALUATION OF SELECT BMPs
    E.J. Monke, L.F. Huggins, D.B. Beasley,  D.W.  Nelson,  T.A.  Dillaha,
    S. Amin, M.A. Purschwitz, R.E. Land	    63

THE ANSWERS MODEL
    D.B. Beasley, L.F. Huggins, E.J.  Monke,  T.A.  Dillaha,  III,
    S. Amin	    67

TILE DRAINAGE STUDIES
    E.J. Monke, A.B. Bottcher, E.R. Miller,  L.F.  Huggins,  D.B. Beasley
    D.W. Nelson, R.E. Land	    81

ALGAL AVAILABILITY OF PHOSPHORUS ASSOCIATED  WITH SUSPENDED STREAM
SEDIMENTS OF THE BLACK CREEK WATERSHED
    R.A. Dorich, D.W. Nelson	   115

ACCOUNTING FOR NITROGEN DISPOSITION WITHIN A WATERSHED
    R.F. Davila, L.F. Huggins, D.W. Nelson	   141

TEMPORAL INSTABILITY IN THE FISHES OF A  DISTURBED AGRICULTURAL
WATERSHED
    L.A. Toth, J.R. Karr, O.T. Gorman, D.R.  Dudley	   165

DECLINE OF A SILVERJAW MINNOW  (ERICYMBA  BUCCATA)  POPULATION IN
AN AGRICULTURAL WATERSHED
    L.A. Toth, D.R. Dudley, J.R. Karr, O.T.  Gorman...	,   231

THE SOCIOLOGICAL STUDY OF SOIL EROSION
    S.B. Lovejoy, F.D. Parent	   245

BLACK CREEK DATA MANAGEMENT SYSTEM
    P.K. Carter, D.B. Beasley, L.F. Huggins, S.J.  Mahler	

-------
                            LIST OF CONTRIBUTORS
S. Amin, Graduate Research Assistant, Department of Agricultural Engineering,
         Purdue University

D. B. Beasley, Assistant Professor, Department of Agricultural Engineering,
         Purdue University

A. B. Bottcher, Assistant Professor, Department of Agricultural Engineering,
         University of Florida (formerly, Graduate Instructor, Purdue
         University)

P. K. Carter, Programmer, Department of Agricultural Engineering, Purdue
         University

R. F. Davila, formerly Graduate Research Assistant, Department of Agricultural
         Engineering, Purdue University

T. A. Dillaha, Graduate Instructor, Department of Agricultural Engineering,
         Purdue University

R. A. Dorich, Graduate Research Assistant, Department of Agronomy, Purdue
         University

D. R. Dudley, Division of Surveillance, Ohio Environmental Protection Agency

O. T. Gorman, Museum of Natural History, University of Kansas

L. F. Muggins, Professor, Department of Agricultural Engineering, Purdue
         University

J. R. Karr, Department of Ecology,  Ethology, and Evolution, University of
         Illinois

R. E. Land, Field Coordinator, Department of Agricultural Engineering, Purdue
         University

S. B. Lovejoy, Assistant Professor,  Department of Agricultural  Economics,
         Purdue  University

S. J. Mahler,  Systems Analyst, Department of Agricultural Engineering, Purdue
         University

E. R. Miller,  Department of Agricultural Engineering, Purdue  University

E. J. Monke,  Professor,  Department of  Agricultural  Engineering, Purdue
         University

J. B. Morrison,  Department  of Agricultural  Information,  Purdue  University

D. W. Nelson,  Professor,  Department of Agronomy,  Purdue  University

-------
 F.  D. Parent, Graduate Research Assistant, Department of Agricultural Economics,
         Purdue  University

 M.  A. Purschwitz, Staff Engineer, American Society of Agricultural  Engineers
          (formerly Graduate Research Assistant., Purdue  University)

 L.  A. Toth,  Department of Ecology, Ethology, and  Evolution,  University  of
         Illinois

 R.  Z. Wheaton, Associate Professor, Department of Agricultural  Engineering,
         Purdue  University
                                ACKNOWLEDGEMENT
     The Board of  Supervisors of the Allen County  Soil and  Water Conservation
District wishes to  express its thanks to  the  investigators represented in this
report. In  addition,  special  thanks  is due  to Reg  Warner and James  Lake who
served as  administrators during  the  project,  and to Dan  McCain,  Soil Conseva-
tionist,  in Allen County for the Soil Consevation Service.

                                                           Mick Lomont, Chairman
                                                           Board of Supervisors
                                                           Allen County SWCD

-------
                                     - 1  -
                         FINAL REPORT - BLACK CREEK II
                                      by
                                J. B. Morrison

                                 INTRODUCTION

     The Black Creek Watershed, located in Allen County Indiana, is  the  site
of an intensive study of the impact of agricultural land use on water quality.

     Under the direction of the Allen County Soil and Water Conservation  Dis-
trict,  a  program of land treatment, complemented by water quality monitoring
and supporting scientific studies, has been carried out since 1972, when  par-
ticipants  in  a  conference  on  the Maunee River suggested that agricultural
practices in the Maumee River Basin were contributing to  the  degradation  of
the river and of Lake Erie.

     Work on the Black Creek watershed, during the first  five  years  of  the
project  was  reported  in  detail in the four volume "Environmental Impact of
Land Use on Water Quality," EPA 905/9-77-007 (A-D).  The current report, which
covers  a period of three years from 1977 through 1980, is intended to summar-
ize the findings of the initial effort, and  to  update  those  findings  with
results  of  water  quality monitoring and supporting studies conducted during
this period.

Purpose of the Black Creek Project

     Although the Black Creek Watershed is located in the  headwaters  of  the
Maumee  River,  the focus of the Black Creek Project, was, from its inception,
directed at Lake Erie.  In the early years of the last decade, it was fashion-
able  to  talk about the "death of Lake Erie." Lake Erie was easily identified
as the most polluted of the Great Lakes.  It's shallow  depth,  combined  with
the  existence  on  its shores of highly urbanized areas such as the Cleveland
and Detroit areas, and its role as the receiving body for drainage water  from
agricultural basins such as the Maumee, threatened the viability of the lake.

     Particularly troublesome in the Western Basin were algal blooms which had
as their eventual impact reduction of available oxygen for fish life.

     In 1972, prior to the adoption by Congress that year of the Water Quality
Act Amendments, much work to control pollution in the Lake Erie Basin had been
accomplished.  Municipal  treatment  plants  throughout  the  basin  had  been
updated and other improvements, including removal of phosphorus from effluent,
were under construction or being planned.  Regulations on industrial polluters
had been tightened and were due to be tightened even more with the Water Qual-
ity Act Amendments.

     Although municipalities and industries had from time to time  pointed  an
accusing finger at agriculture as a co-equal polluter of the lake, the concept
of nonpoint source pollution was not a generally understood term.  Few studies
had  been done which attempted to relate fertilizer, cropping practice, pesti-
cides and herbicides, and soil erosion to water quality.  Generally,  however,
a relationship was being hypothesized.  The elements of this relationship were
as follows:

-------
                                    - 2 -
 1.   In bodies of water like Lake Erie, a major water quality problem  is  the
     growth of algae.

 2.   In most fresh water lakes, the availability of phosphorus is limiting  to
     the growth of algae.

 3.   Phosphorus loading from point sources can be estimated and controlled.

 4.   It may be possible to eliminate all phosphorus from point sources without
     effecting  algal  growth,  if phosphorus from nonpoint sources,  primarily
     agriculture, is not also controlled.

     Since the 1930"s, agricultural conservation programs have been  aimed  at
controlling  soil  erosion.  Ihe primary purpose of these programs has been to
preserve the soil resource for the production of food for the current  popula-
tion  and  for  future  generations.   Ihe primary question posed by the Black
Creek study is as follows:

     Can traditional soil conservation practices be applied in such a  way  as
to improve water quality?

     Although the project concentrated on a relatively small watershed, a goal
has  been  to  apply  the knowledge gained in the Black Creek Watershed to the
larger area of the Maumee River Basin, so as to understand its impact on  Lake
Erie.

     Simultaneously, the watershed was studied in several ways.   Loadings  of
sediment  and  chemicals  were monitored throughout the watershed to determine
their impact on the chemical and physical properties  of  water  entering  the
Maumee  River.   tonitored water quality parameters were also used to evaluate
the success of individual Best l^anagement Practices on the reduction  of  non-
point  source pollution.  Additionally, the monitored data were used to verify
a generalized model by which the present capabilities of other  watersheds  to
control  agricultural nonpoint source pollution can be assessed and with which
best management practices can be planned to  reduce  further  pollution  in  a
cost-effective  manner.   At  the  same time, biological studies were directed
toward measuring the impact of the practices installed in the first  phase  of
the  project  on  the  biological  integrity of the streams.  Certain of these
practices, those which altered the streams or the riparian lands  adjacent  to
the streams, tended to disrupt the biological community in the aquatic system.

     In addition, efforts have been made to  assess  the  economic   impact  of
agricultural pollution control programs and their social impact on the commun-
ity of the Black Creek Watershed.


                     SELECTION  OF THE BLACK CREEK WA1ERSHED

     In selecting the  Black Creek Watershed as a study area, care was taken to
select  a  watershed   to reflect as closely as possible the characteristics of
the haumee Basin.  Comparisons of  Black  Creek and the Naumee are contained  in
detail   in  Volume   II   of   the  Black Creek Einal Report  (EPA  905/9-77-07-B).
Here,  it  is sufficient  to observe  that proportions of land  use,  soil   types,

-------
                                   - 3 -
land capability classes,  and population characteristics in the  watershed are
similar to those of the total basin.
                            PURPOSE OF THE REPORT

     The report is intended to consolidate and update materials collected dur-
ing the  eight year period, covered  by the Black Creek  Project.   However,  it
concentrates primarily on  the years  between 1977 and  1980,  and  represents  a
major interim report in the total project.  The organization of this report is
a collection of  research  papers presented by project  investigators.   A final
report, which synthesizes all of the work covered during these eight years and
an additional two-year period, will be published in 1983.

     The following sections are intended to  summarize the  research  findings
and implications as reported by the project investigators.  Details which sup-
port their conclusions are set forth in the individual papers.


                      FINDINGS OF THE BLACK CREEK PROJECT

     The question most often asked about the Black Creek  Project  is "How much
did  you  improve water  quality?"  The answer  to  that question  is   at  once
straight forward but extremely complex.  Clearly, based on water quality moni-
toring results extending from 1975 through 1980, there was a reduction in sed-
iment and sediment-related pollutants leaving the Black Creek Watershed, which
can  be attributed  in part  to the application of Best  l"ianagement Practices
throughout  the project  period.   (For  a  complete  report  of  the  practices
installed see EPA 905/9-77-007-B).

     Specifically, sediment losses from the watershed declined after 1975.  In
addition,  mean  concentrations of sediment,  adjusted  to  account for  flow
difference,  also declined after  1975.   At  the  same  time,  reductions  in
sediment-associated nutrients  —  sediment phosphorus and sediment nitrogen —
also declined.

     Losses of sediment-bound  phosphorus were  relatively high  in 1975  (4.9
kg/ha) but  averaged only about 1.1 kg/ha as the project continued.  Sediment-
bound nitrogen losses were also very high in 1975  (30  kg/ha)  but declined to
around 5 kg/ha thereafter.  Some of these reductions have been shown by simu-
lation to have been due to favorable weather patterns but an encouraging trend
toward better water quality nevertheless exists.

     In line with the discussion of the  critical  role phosphorus  plays  in a
receiving body of water  like Lake Erie, the reduction in sediment-bound phos-
phorus loss is significant, exceeding the reduction, if  applied  to the Niaumee
Basin, generally suggested  as necessary to achieve from  nonpoint  sources.
This suggests  that the land treatment applied  to  Black Creek, if  applied  to
the Basin, would achieve a worthwhile water quality purpose in addition to its
soil conservation benefits.

-------
     This conclusion is reinforced by studies of the availability of  sediment
bound  phosphorus  to  algae,  also conducted in association with the project.
Algae are unable to mine all of the phosphorus attached  to  sediment.   As  a
result  of  these  studies it was determined that most of the phosphorus which
will become available to algae becomes available  in  the  laboratory  in  two
days.   Ihis represents 25 percent of the total sediment bound phosphorus.   In
two weeks, an additional 5 percent of total phosphorus  becomes  available   to
algae.   Ihus  30  percent of the phosphorus carried from areas like the Black
Creek Watershed to Lake Erie are available to contribute to algal blooms,  and
the  reduction  of sediment bound phosphorus entering the lake.  Such a reduc-
tion, which can be achieved by the implementation  of  Best  Management  Prac-
tices,  is  significant because up to 90 percent of the total phosphorus yield
from agricultural lands is attached to sediment.

     Ihe amount of sediment lost is related to  rainfall,  while  the  average
concentration  is  related  to  land  conditions  at the time rainfall occurs.
Bebruary, March and June were the months of highest total  sediment  loss  and
were  also  the months of highest total loss of sediment bound nutrients.  May
and June were the months when sediment concentrations were  highest  in  water
leaving the watershed.

     Although definite reductions were shown in the loss of sediment and sedi-
ment  related  nutrients.  Comparable reductions were riot shown in the loss of
soluble forms of the nutrients.  Losses of soluble inorganic phosphorus  (SIP)
from  the  watershed have increased each year since 1976.  SIP losses were  not
correlated with runoff volumes.  Ihis finding suggests that  the  installation
of  BMPs had little effect on SIP.  Ihe increased yearly loss of SIP was prob-
ably due to increased septic tank flow in the watershed.

     Soluble forms of nitrogen lost from the watershed were  directly  related
to  runoff  volume.   Concentrations  of most forms of nitrogen did not change
during the project.  Ihe installation of Best Management Practices  apparently
did not reduce the loss of soluble forms of nitrogen from the watershed.

     In fact, the installation of Best Management Practices may have increased
the  loss of nitrogen as nitrate, largely as a result of the greater infiltra-
tion produced by the management practices, resulting in a higher proportion of
the  water  leaving  the watershed as a result of subsurface flow.  Subsurface
water provides a mechanism for greater leaching of nitrate.

     Greatest losses of most soluble forms of nitrogen and  soluble  inorganic
phosphorus  occurred   in  the  months  of Eeburary, March, April and December.
Snowmelt  runoff contributed significant amounts of soluble nutrients in Febru-
ary and March.

     Ihus,  it must be  concluded that as far as two of the more important  pol-
lutants   associated  with the Maumee River  from the standpoint of Lake Erie —
phosphorus  and sediment — that, as demonstrated by the  Black Creek project, a
reduction  can  be  achieved  through the installation of Best Management Prac-
tices and that this reduction is sufficient to have an  impact on the  lake  if
applied  in  a basin-wide program.

-------
                                    - 5 -
     These potential benefits have not been achieved without  costs,  however.
Ihese costs are of two fundamental kinds — economic and environmental.

     As detailed in Volume II of the Black Creek Final  Report,  the  economic
cost  of  installing  Best  Management  Practices was not trivial amounting to
$709,791.  An analysis, performed on a subjective basis on the costs of  prac-
tices  which it was believed would have produced a comparable result indicated
that the water quality benefit could have been achieved at a cost of $323,460.
Projected  to  the  Maumee Basin, this translates to a cost of $125,000,000 in
1979 dollars.  Simulation studies have indicated that at least  a  significant
portion of this cost could be avoided by greater reliance on conservation til-
lage as a management tool.  Although relatively small amounts of  funding  can
achieve  worthwhile  conservation  purposes, programs to improve water quality
most deal with a total system.  Cn the scale of the I»iaumee River Basin,  costs
will be significant, even with a well designed and tightly administered plan.

     Ihe environmental costs of the project were largely  experienced  by  the
aquatic  life  in  Black  Creek,  and derivitively, to some extent, the Maumee
River.  As a headwater stream, the natural environment for aquatic life in the
stream  system of the Black Creek Watershed is expected to be harsh.  However,
some of the construction activities such as  channel  grade  stabilization  or
clearing  of  tall vegetation on or along the channel banks are man caused and
clearly had detrimental effects on the aquatic community some of which may  be
long  lasting.   For example, the removal of near stream vegetation can result
in increased water temperature, increased sunlight falling on the stream,  and
a resulting increase in microscopic plant life such as algae in the stream.

     The fish population in the Black Creek Watershed was highly variable dur-
ing  the  eight years for which data was collected in the project.  During the
period, 44 species of fish were identified, although it was  unusual  to  find
more than 20 species represented during a single fish collection period.

     Very few fish species are resident in the Black Creek Watershed  continu-
ously throughout the year.  Most species leave the rather harsh environment of
the headwater stream during certain periods of the year and return when condi-
tions have become more favorable.

     Based on the sampling work conducted in the Black Creek, it is clear that
some  species  declined during the construction period, while others seemed to
thrive on the conditions made possible by the altered habitat.  At  any  rate,
the composition was altered and the stability of the population present before
construction activities begun has probably not yet been achieved.

     The conclusion is that practices which include significant  channel  work
will  have an impact on the aquatic life in the watershed and the duration and
length of this impact will vary with individual species.  Disturbances in  the
black Creek Watershed were probably largely masked because of the availability
of the Maumee River as a source of individuals to re-colonize the Black  Creek
Watershed.   If  all  of the tributary streams in a significant stretch of the
river had been simultaneously altered as was  the  Black  Creek,  the  overall
impact  could  have  been greater.  Moreover, these practices were shown to be
only marginally effective in reducing sediments and channel pollution  of  the
Black  Creek stream system.  In retrospect, these practices, desirable as they

-------
                                    - 6 -
might seem to landowners, should not have been classified  as  Best  Management
Practices in the first place.
              PERMANANCE OP TREATMENT PRACTICES IN THE WATERSHED

     A question equally important to the success of the Black Creek Project in
achieving  water quality goals has been a question about whether water quality
practices would be maintained after the cost-sharing  money  which  encouraged
their installation had been exhausted.

     In general, the answer to this question has been different for structural
practices than for management practices such as reduced tillage and crop resi-
due management.

     Structural practices — waterways,  drop  structures,  terraces  —  have
largely  been  maintained, while management-oriented practices are more likely
to be abandoned.  This is true despite the  fact  that  management  practices,
where  adapted  to the conditions of the watershed, have been considered to be
of less cost than structural practices.

     It has been suggested that the structural practices have been  maintained
more faithfully because they furnish a visual reminder of a commitment made to
the project.  Cultural and other management related practices are more  easily
ignored since no permanent visual reminder is present.

     Practices such as crop residue management were determined to be much more
likely to be maintained by person identified as "opinion leaders" in sociolog-
ical investigations than by others.  This leads to  two  conclusions.   First,
the  practices  adopted  by  opinion  leaders  did  not "filter down" to other
residents of the watershed as easily as might have been hoped.  Secondly,  the
tendency   of  the  opinion  leaders  to  continue  practices  may  well  have
represented the greater involvement of these individuals in  project  planning
and implementation, suggesting that in future projects, it should be a goal to
involve as many landowners as possible in the planning of the project.

     In general,   there are indications  that  the  awareness  of  individual
landowners  within  the  watershed  have increased throughout the project, and
there is a slightly greater tendency on the part of  watershed  landowners  to
consider that water quality may be a problem on which they are capable of con-
tribution to the solution.

     Willingness to participate in the project and maintain project  practices
has  also been shown  to be correlated with the perception of individuals about
whose responsibility  soil conservation ultimately is.   Landowners who  believe
conservation  to  be   the responsibility of the individual landowner were much
more likely to participate in the project, suggesting that education  programs
can  have  an impact  if successfully carried out in conjunction with watershed
projects.

-------
                                    — 7 —
               CONDUCTING A LAND RELA1ED WA1ER QUALITY PROGRAM

     Conducting the Black Creek Project was, to a large extent,  an exercise  in
gaining  new insights into the most efficient way to plan watershed management
projects and to evaluate their results.  Ihese  insights  can be  phrased   in
terms of how similar projects can be most efficiently planned and managed, and
in terms of relationships that can be used to project results to  larger and
more complex areas.

     A fundamental conclusion, based on work conducted during the  first  five
years  of  the  project,  is that a voluntary program, supplemented with cost-
sharing incentives, can be successful in encouraging the establishment of Best
fanagement  Practices  in a watershed area.  A corollary of this conclusion  is
that a locally based unit of government, such as a soil and water conservation
district, can effectively manage a project which has as its ultimate objective
results which extend beyond the local area.

Monitoring and Modeling

     A competent laboratory can accurately access the components  of  a  water
sample  delivered to it.  However, the interpretation of these results is less
straight forward.

     Losses of sediment and related nutrients in a watershed  are  very  event
oriented.  Thus, if there is no runoff, there are no runoff associated losses.
Since an ultimate goal of Black Creek Project research was focused on the Mau-
mee  River  and  Lake  Erie,  an  estimate of annual loadings to the river was
essential, thus it was critical that runoff events not be  missed.   Automated
samplers were thus considered a necessity.

     It can be observed that if the focus of the study had been   only  on the
biology  of  the  Black Creek itself unterminated samples would  have been less
important.  Aquatic organisms do not "see"  loadings;  they  "see"  concentra-
tions.   If  the materials entering the stream are not highly toxic to aquatic
life, wide temporary variations in concentrations can be tolerated  by  fishes
which  inhabit  headwater  streams.   for  determining  the suitability of the
aquatic environment to fishes, a few, well chosen grab samples may  very  well
be  sufficient, since peak concentrations which pass so rapidly  that they can-
not be found by grab sampling are probably well within the time   limits  which
can be tolerated by fish.

     Both a grab sampling and an automated sampling  program  were  simultane-
ously  conducted  in  the Black Creek Watershed.  Cost of the sampling program
was not trivial, and in fact exceeded the cost of land treatment.  While  this
can  be justified in terms of an experimental - demonstration project like the
Black Creek Project, it cannot be considered as a part of routine water  qual-
ity improvement programs.

     Black Creek investigators thus recommend that watershed projects be  car-
ried  out  with a combination of limited monitoring including appropriate fish
assessments and computer based simulation modeling.  A product  of  the  Black
Creek project is such a model — the ANSWERS model — which has  been described
in detail in previous reports.

-------
                                    - 8 -


     Briefly, the ANSWERS model is a spatially distributed model  which  simu-
lates  surface  and  subsurface  flow  from  a  variety  of  elements within a
watershed.  Ihe model  is  event  oriented.   It  is  written  in  a  computer
language,  fORIRAN,  which  is  available at most computing facilities.  Since
initial publication of the model, in 1977,  several  refinements  and  improve-
ments  have  been  made,  increasing  the scope of the model and adding to its
reliability.

     Ihe model is capable of aiding in the evaluation of a water quality  pro-
ject.   It  is also useful in planning water quality programs,  a utility which
has  been  demonstrated  in  the  planning   of  water  quality    measures   in
subwatersheds under a special conservation program in Allen County, Indiana.

     One effort, which continues within the Black Creek effort, is the  updat-
ing of the ANSWERS model by incorporation of more reliable data concerning the
impact of various management practices  on  sediment  loss  and  the  loss  of
related  nutrients, especially the individual BMP studies.  Initial results of
these studies indicate that all BMPs under  evaluation can result  in  signifi-
cant  reductions in sediment loss over those associated with conventional til-
lage,  towever, quantification of results awaits the collection of more data.

Other Studies

     Other studies have resulted in a deeper  understanding  of  some  of  the
dynamics of sediment and nutrient movement within a watershed.

     A study of tile drainage, began as a portion of the initial  Black  Creek
study  and discussed in the final Black Creek Report, indicates that appropri-
ately installed drainage in soils typical of the Maumee Basin can result in  a
reduction in sediment associated pollution, such as phorphorus.

     In conjunction with further development of the ANSWERS model,  investiga-
tions  were  begun into methods of accounting for the distribution of nitrogen
within a watershed.  Results which give good  agreement  with  monitored  data
have been obtained.

     Certain relationships have been established which  should   be  useful  in
future  studies,  for example, as a part of the study relating  to availability
of phosphorus to algae, it was determined that this amount, for a given situa-
tion,  could be readily estimated by two simple chemical tests.  Ihe amount of
phosphorus extractable by ammonium floride is roughly equivalent to the amount
extracted  by  algae in two-day incubation and the additional amount extracted
by sodium hydroxyde approximates the additional amount extracted  in  two-week
incubations.

-------
                                    - 9 -
                   IMPLICATIONS OF THE BLACK CREEK PROJECT


     For the agricultural nutrient of most  concern to water quality planners,
phosphorus,  Best Management Practices such as were applied in the Black Creek
Watershed can provide a  significant  reduction.   Best  Management  Practices
offer  less promise of controlling soluble forms of nutrients such as SIP,  and
may, in fact, increase the amounts of soluble nitrogen which  enter  receiving
bodies  of  water.   Sediment,  if  it is a problem, can also be significantly
reduced by the installation of Best Management Practices.

     Intensive treatment of areas like the Black Creek Watershed would have to
be  coordinated on a large scale basis to have much impact on ultimate receiv-
ing waters such as Lake Erie.  This implies careful planning both on a  basin-
wide  and  individual  watershed levels.  Computer models, such as the ANSWERS
model can provide assistance in both planning and evaluating  the  results   of
treatment  programs  on  small watersheds within the framework of a total land
treatment program.

     Educational efforts, and the involvement of as many landowners as  possi-
ble  in  planning and implementing treatment programs can help insure the suc-
cess of the initial efforts, and more importantly, help  assure  that  project
components  will be maintained after the first flush of activity has been com-
pleted.  Furthermore, proper location of bMPs can simultaneously obtain  water
quality  benefits as well as soil conservation benefits by maintaining produc-
tivity levels.

     Ihe economic cost of treating an area as large as the Maumee Basin is  not
trivial and can be measured in terms of hundreds of millions of dollars. How-
ever, these costs are not uncompetitive with the equally large costs  involved
in the removal of comparable amounts of phosphorus from point sources of water
pollution.

     It is the concensus of the investigators in the Black Creek Project that
the  ultimate result of land treatment programs will be an environmental bene-
fit.  This benefit is not achieved without the risk of  environmental  damage.
In  fact, an agricultural watershed is always a disturbed watershed. Practices
can be found which can minimize damage to the physical and chemical components
of  water quality.  But these practices may in themselves not be beneficial to
the aquatic life of the stream. If enough streams are disturbed along a water-
way  like  the  Maumee,  the  cumulative effect may be damage to fish life  and
other aquatic organisms.  Some thought should therefore be given,  when  water
quality  improvement  projects  are  planned, to designating certain areas  for
wildlife within the total project.

     Ihe most satisfactory areas for this purpose will probably be areas which
are currently relatively undisturbed.  Maintenance of small wooded areas, such
as the Wertz Wood, discussed in detail in previous Black Creek reports, should
be  given  a high priority.  Parklands and natural areas, where they currently
exist, should be maintained to help minimize the damage done  by  agricultural
activities  within  the total basin.  Popular support in the agricultural com-
munity for programs which  do  not  make  some  improvements  to  agricultural
drainage  is  unlikely,  but,  as  has  been shown in this project, subsurface

-------
                                    - 10 -
drainage at least can also be important to maintaining improved water quality.
Hovvever,  projects  should be planned with the intention of minimizing distur-
bances of stream channels and near  stream  vegetation.   These  modifications
should  only  be  undertaken when it is clear that the benefits will outweight
the environmental risks they entail.

-------
                                   - 11 -


         WA1ER QUALITY:  SEDIMENT AND NUTRIENT LOADINGS FRCfi CROPLAND


                                      by


            D.W. Nelson, D.B. Beasley, E.J. Monke, and K.A. Dorich

     Public Law 92-500 passed in 1972 mandated that each state prepare a water
quality management plan which encompasses nonpoint as well as point sources of
pollution.  In attempting to prepare strategies and/or plans  for  control  of
nonpoint  pollution,  most  state  and  federal  planning/regulatory officials
became aware that relative little is known about the amounts of  water  pollu-
tants originating from agricultural land or the effectiveness of techniques to
control or minimize pollutant deliveries.  Preliminary studies suggested  that
the  most significant water pollutants originating from cropland are sediment,
plant nutrients, and pesticides (1).

     Although a number of small watersheds (<30 ha) at  various  locations  in
the  eastern U.S. had been periodically monitored during the past 20 years, no
monitoring of a medium size ( 1000 ha)  agricultural watershed  had  been  con-
ducted,  furthermore, a long term study of the effects of agricultural activi-
ties on water quality was started in 1973 on a  5000  ha  watershed  in  Allf;n
County, Indiana.  The project was funded under the Great Lakes Program, Region
V  U.S. Environmental Protection Agency and involved  coordinated  efforts  of
the  Allen  County  Soil Conservation District, the Soil Conservation Service,
Purdue University, and the University of Illinois.  The objective of the  pro-
ject  was  to  determine  if  water quality in the watershed and in the Maumee
River could be improved by implementation of a wide range of soil conservation
practices  in  the drainage area.  For details of the project consult Lake and
Morrison (2).
                            MATERIALS AND METHODS

Study Area

     The 5000 ha black Creek watershed  (Figure  1)  was  selected  for  study
because  it  was  representative  of the soils and land uses prevailing in the
Maumee River drainage basin.  Table 1 provides information on  the  soils  and
land  use  in  the watershed.  About two-thirds of the area consists of nearly
level lake plain and beach ridge soils,  whereas  one-third  of  the  area  is
gently  sloping (3-6%) glacial till soils.   Land use in the watershed is about
60% row crops, 30%  small  grain  and  pasture,  and  10%  woods,  roads,  and
developed  areas.   The  drainage  pattern in the area consists of one natural
stream (Black Creek) running from west to  east and discharging into the  Kiu-
mee  River  (Figure 1).  A number of constructed drainage ditches intersecting
with Black Creek are used as outlets for surface and tile drains.  Most of the
lake plain soils in the watershed are tile drained.

-------
                                   - 12  -
           APPROXIMATE  SCALE  IN
                  KILOMETERS
Bigure 1.   Nap of the  Bleck Creek study area.

-------
                                    - 13 -
  Table 1.  Characteristics of the Black Creek Watershed and two intensively
            studied drainage areas within the watershed.

                               Black Creek     Smith-Fry         Driesback
       Characteristics	Watershed    Drain (Site 2)   Drain (Site 6)

  Drainage area, ha               4950            942              714

  Soils:
    Lake plain & beach ridge        64%            71%              26%
    Glacial till                    36%            29%              74%

  Land use:
    Row crops                       58%            63%              40%
    Small grain & pasture           31%            26%              44%
    Woods                            6%             8%               4%
    Urban, roads, etc.               5%             3%              12%

  Number of homes:                 	             28              143
Monitoring Systems

     Grab sampling stations were established at 14 sites within the  watershed
and  on  the  Maumee River to provide weekly data on the quality of water ori-
ginating from soils and land  uses  in  the  drainage  aree  above  the  site.
Automated  samplers (PS 69) and flow measuring devices were installed at three
locations (Sites 2, 6, and 12) in the watershed (tigure 1) to provide continu-
ous  flow data and to permit calculation of loadings on a storm or time period
basis.  Meteorological conditions in the  watershed  were  continuously  moni-
tored.  A complete hydrometeorological station with automatic data acquisition
and remote transmission capabilities was established at Site 6.  The amount of
rainfall  was measured at seven other locations in the watershed and rainwater
samples were collected for chemical analysis at two locations.

     Temperature and dissolved oxygen concentration of water were measured  _in
situ  and shortly after collection pH, turbidity, and alkalinity were measured"
in grab samples.  Water samples taken by grab or automated methods were frozen
soon  after  collection, transported to the Water Quality Laboratory at Purdue
University and analyzed  for  suspended  solids,  NH.-N,  soluble  organic  N,
sediment-bound  N,  soluble inorganic P (filtered reactive P), soluble organic
P, and sediment-bound P.  Pesticides, alkaline earth cations, and heavy metals
were  measured  in selected samples.  Methods used for analysis of al] samples
were those prescribed by the American Public Health  Association  (3)  or  the
U.S. Environmental Protection Agency (4).

Data Processing

     Loadings of sediment and nutrients were calculated by integration of flow
and  concentration data on a storm, monthly, quarterly, or yearly basis.  Flow
weighted mean concentrations (monthly basis) were calculated by  dividing  the
monthly  load  of  sediment  or nutrient by the monthly volume of runoff.  The
average total N and P concentrations in suspended sediment were calculated  by

-------
                                   - 14 -
dividing  the  monthly  sediment-bound  N -and P loads by the monthly sediment
load.  The enrichment ratios for total N and P were calculated by dividing the
total  N and P concentrations in sediment by the average total N and P concen-
trations in soils present in the drainage area.  Linear regression and  corre-
lation techniques (5) were used to determine the relationships between monthly
ot quarterly runoff volume, sediment losses,  nutrient  losses,  and  nutrient
concentrations in sediment.
                            RESULTS AND DISCUSSION

     Reconnaissance sampling within the watershed revealed that no significant
amounts  o£ hex?ne-soluble pesticides were present in water, sediment, or fish
tissue.  Specific pesticides evaluated  included  aldrin,  dieldrin,  DDT  and
metabolites,  atrazine, trifluralin, and 2,4,5-T (2).  Analysis of weekly grab
samples established that the dissolved oxygen,  temperature,  pH,  alkalinity,
and  alkaline  earth  cations  levels  exhibited trends which were typical for
medium size agricultural watersheds (6).  Heavy metals were  present  in  only
trace concentrations (6).

Sediment and Nutrient Loads

     Table 2 provides information on rainfall, runoff, and sediment lost  from
the  two  major  drainage areas in the Black Creek Watershed during the period
1S75 to 1978.  Precipitation was above normal in 1975, below normal  in  1976,
and  near normal in 1977 and 1978.  Runoff volumes tended to be highest during
years with greatest rainfall, however, the percentage of precipitation appear-
ing  as  runoff  varied over the years  (26% in 1975, 17% in 1976, 20% in 1977,
26% in 1978, 27% in 1979, 26% during the first half of 1980, and an average of
23%  over  5  years).   During  the  period  of  1975  through  1979, sediment
discharges averaged 844 and 1100 hg/ha for Sites 2 and 6, respectively.   How-
ever , sediment losses in 1975 were 4 to 8 times higher than the yearly average
of the other 4 years.  Sediment losses during 1977 and 1978 were low (range of
380  to  544  hg/ha  including  both  Sites 2 and 5 during 1977 and 1978) even
though rainfall during these years was normal.  This finding suggests that the
best  management  practices  implemented in the watershed during 1975 and 1976
resulted in reduced sediment losses in subsequent years.

     Data on amounts of  sediment-bound  N  and  P  discharged  from  the  two
drainage  areas  during 1975 to 1980 is also given in Table 2.  The quantities
of sediment-bound nutrients lost from the drainage  areas  decreased  markedly
after  1975, generally in proportion to reductions in sediment loss.  Applica-
tion of best management practices in the watershed  was,  at  least  in  part,
responsible  for  reductions  in  amounts of sediment-bound nutrients observed
during the course of the study.

-------
                                    - 15 -


Table 2.  Rainfall, runoff, and sediment and nutrient loss occurring in tvo
          drainage areas of the Black Creek watershed during the period 1975
          to 1980.
Parameter                  Site                       Year
                            no   	,—
                                  1975   1976  1977   1978  1979   198CT  Ave.
Rainfall, cm               2 & 6   108    66    96     77    79     39     85
Runoff, cm
Sediment loss, kg/ha
Sediment P loss, kg/ha
Sediment N loss, kg/ha
Sol. inorg. P loss, kg/ha
Sol. org. P loss, kg/ha
NH*-N loss, kg/ha
NO~-N loss, kg/ha
Sol. org. N loss, kg/ha
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
29.1
26.0
2126
3735
5.24
4.51
31
28
0.
0.
0.
0.
1.
1.
19
11
2.
2.
.25
.98
14
34
11
13
51
82
.01
.63
33
51
12.4
10.1
637
384
0.98
0.73
4.82
2.b6
0.06
0.18
0.04
0.04
0.60
0.85
5.55
2.39
0.93
0.74
18.5
19.4
435
452
1.67
1.78
4.55
4.71
0.14
0.47
0.06
0.10
0.58
1.30
15.42
12.73
1.10
1.78
18.5
21.3
380
544
0.65
0.79
6.
6.
0.
0.
0.
0.
0.
4
8.
5.
1.
2.
10
91
21
68
08
35
75
06
27
96
66
89
23
20
640
437
0.94
1.07
6.
5.
0.
0.
0.
0.
0.
1.
20
9.
1.
1.
08
00
22
59
07
08
69
32
.92
36
82
98
12
8
392
263
1.19
1.41
3.70
4.78
0.15
0.16
0.02
0.05
0.57
0.56
13.68
5.41
1.05
0.77
20
19
P44
1110
1.90
1.78
10.56
9.69
0.15
0.48
0.07
0.15
C.83
1.67
13.83
8.41
1.57
2.13
 a   1980 averages  include data only from first 6 months of 1980
 b   Average excludes 1980 data
     Table  2 also provides data on the amounts of soluble nutrients discharged
 from   the two drainage areas during a five and one half yeai period.  Although
 the amounts of soluble inorganic P annually discharged from the drainage areas
 were   low   (<0.7  kg/ha/year) ,  there  is  no indication that the amounts lost.
 decreased with the time during  the  study.   In  fact,  there  was  a  marked
 increase in soluble  inorganic P during 1978 and 1979.  One explanation for the
 increase in soluble  P loss at Site 6  during  1978  and  1979  is  that  large
 volumes  of untreated  household  wastewater was discharged into the drainage
 ditches near Harlan  during the  time  an  interceptor  sewer  was  being  con-
 structed.   Previous  studies have shown that septic tank effluents were a major
 source of soluble P  measured at Site 6  (7).   Ihe  amounts  of  NH.-N,  NC/I-N,
 soluble  organic  N,  and soluble organic P discharged from the drainage areas

-------
                                   - 16 -


each year were directly related to the volume of runoff.   Losses  of  soluble
organic F were very low (<0.13 kg/ha/year)  except at Site 6 in 1978 where sep-
tic tank effluent likely contributed to the load.  Soluble  organic  N  losses
were  significant  (0.74 to 2.78 kg/ha/year)  during all years at each site and
the higher losses measured at Site 6  probably  reflect  septic  tank  inputs.
Losses  of NH Ij-N    were relatively low (0.58-1.82 kg N/ha/yr) throughout the
period of study except for Site 6 during 1978.   Septic  tank  effluents  were
likely  responsible  for  the  higher NH.-N losses observed at Site 6 in 1978.
Ihe amounts of NO-^-N in drainage water appeared to be related  to  amounts  of
rainfall  in  the  watershed; i.e., loss of NG^ -N were highest in 1975, 1977,
and 1979, the three years with highest rainfall.  Losses of NCu-N  were  rela-
tively  large:-  (average  of  13  and 8 kg N/ha/year for Sites 2 and 6, respec-
tively) arid likely reflect the fact that the watershed  is  tile  drained  and
that  the soils are maintained in a high state of fertility by applications of
manure and inorganic N fertilizers.  Although the amounts of NO,-N  discharged
from  the watershed were substantial, the annual flow weighted mean NO--N con-
centration never exceeded the U.S. Environmental  Protection  Agency  drinking
water standard (10 nxj/liter) .

     Ihe data in Table 2 suggest that adoption of best management practices to
control soil erosion has not resulted in a reduction in the discharge of solu-
ble forms of N and P from the watershed.  In fact, there is an indication that
losses  of  soluble  N and P increased slightly as soil conservation practices
were implemented during the study.  In future projects some  attention  should
be  given  to  implementation  of best management practices which minimize the
transport in drainage water of soluble nutrients originating from cropland.

     Ihe annual discharges of sediment, sediment-bound nutrients, and  soluble
N from Site 2 (nearly all cropland) and Site 6  (affected by sewage) v*ere simi-
lar to those from large river basins and  some  small  watersheds  (Table  3).
However,  the  annual  sediment  losses measured at Sites 2 and 6 tended to be
lower than sediment losses reported for  several  small  (<33  ha)  watersheds
planted   to  row  crops.   Ihere  is  little  sediment  deposition  in  small
watersheds, but considerable deposition in the Black Creek area.  The  soluble
inorganic  P loadings measured at Sites 2 and 6 were similar to^those reported
from both river basins and for small watersheds.  Soluble N  (NH.-N plus N03~W)
loadings  at  Sites  2  and  6  were higher than_those reported for many small
watersheds.  Ihis finding likely results from NCu-N  in  tile  drainage  water
present   in  the  Black  Creek  Watershed,  but  absent  in  most of the small
watersheds previously studied.  Ihe higher soluble N levels also reflects  the
influence of septic inputs upon NH.-N levels  in samples from Site 6.

-------
Table 3. Annual sediment and nutrient loading from selected agricultural watersheds in the United States.
Watershed
Location Size
ha
Ohio (Maumee 1.639 x 106
River Basin)
Ohio (Portage 111 x 103
River Basin)
Ohio (Plot lll)d 3.2
Mich. (Ave. of 0.8
Plots) e
Georgia , 1.3
(Watershed P2r
Iowa 3. 3
(Watershed 2)9
Oklahoma 17.9
(Watershed C3)
(Watershed 2)1 1.5

Ohio (I»iaumee) -*
Michigan
(Mill Creek)3
Ag Watersheds-'
Land
Use
Mixed
Mixed
Soybeans
Row
Crops
Corn
Corn
Cotton
Corn
Wheat
Pasture
Cropland
Cropland
Cropland
Pollutants transported
Sediment Sed. P Sol. Pa Sed. N Sol. N.

— — — fcg/ha/year— — 	 	 __
950 1.53 0.29 — 13.4
658 0.84 0.30 — 13.1
1.09 0.13 — 12.3
12940 — 0.71 25.8 2.8
6022 — — 10.3 3.7
9980 — 0.09 14.8 1.4
3900 5.6 1.1 9.7 1.9
0.27 — 12.08

80-5100 0.7-4.3 0.05-0.3
20-70 0.1-0.3 0.1-0.3 4.3-10°C
400-800 0.6-0.9 0.3-0.4 16-31°c
References
9
9
9
10
11
12
13
14

1
1
1
 (a) Sol.  P  is  soluble  inorganic  P  (filtered  reactive  phosphate);  (b)  Sol.  N is (NH^-N + NOZ-N;
 (c) Sediment-bound  and soluble N combined;  (d) Average  of data  from 1975-1976;  (e) "Average of data
 from 1S74-1975;  (f) Average of data from 1973-1975;  (g) Average of  data  from 1969-1975;  (h)  Average
 of data from 1966-1976;  (i) Average of data  from 1968-1972;  (j) Average  of two years data  from
 1S75-1977.

-------
                                    - 18 -


Sediment and Nutrient Concentrations

     Kainfall, runoff, and flow-weighted mean concentrations of  sediment  and
nutrients measured at Sites 2 and 6 during the period of 1975 through 1980 are
presented in Table 4.  Suspended sediment concentrations measured at  Sites  2
and 6 over the 5 year period averaged 392 and 504 mg/1, respectively, and were
maximal in 1975 at sites 2 and 6 (732 and 1435 mg/1, respectively).   However,
suspended sediment concentrations measured at Site 2 and 6 remained relatively
constant throughout the rest of the study (206 to 515 and  216  to  380  mg/1,
respectively).   Ihe  average concentrations of sediment-bound nutrients meas-
ured at Sites 2 and 6 remained relatively constant over the period  from  1976
through  1980, but were very high during 1975  (Table 4).  The average sediment
P concentrations measured at Sites 2 and 6 over 5 years  were  0.85  and  0.65
mg/1,  respectively,  while  sediment  N concentrations averaged 4.6 and 4.42,
respectively.  Although peak concentrations of sediment and  sediment-bound  N
and  P were measured during 1975, soluble inorganic and organic forms of N and
P did not follow the same trend.  Measured soluble constituents (soluble inor-
ganic P, organic P, NH -N, and soluble organic N) remained relatively constant
for each constituent, measured at each site  (Table 4).  The exception to  this
is  the relatively high concentrations of soluble inorganic P, organic P, NHy-
N, and organic N occurring at Site 6, during  1978,  even  though  runoff  was
about  average  during  the year.  When concentrations of soluble inorganic P,
organic P, NH4~N, and organic N measured at Sites 2 and  6  are  compared  for
each,  of the six years (Table 4), the concentrations measured at Site 6 always
exceeded that measured at Site 2 except in one  case   (soluble  organic  P  in
1976).   Ihe trend is evidenced by each of the five and one-half year averages
(Table 4).  Just the opposite is true for NG~-N measurements at the two  sites
over  the  6 year period.  In very year the average NO.,-N concentrations meas-
ured at Site 2 exceeded that at Site 6.  As  indicated  by  this  data,  septic
effluents  which  enter  the  Site 6 subwatershed had  a largipr effect upon the
average concentrations of soluble inorganic P, organic P, NH^-N and organic  N
measured  in  the watershed streams than did agricultural input as measured in
Site 2.  On the other  hand,  the  system  (Site  2  subwatershed)  vfoich  was
comprised  primarily of agricultural drainage contained higher average concen-
trations of NOZ-N than was measured in a  subwatershed  influenced  by  septic
effluents.   Inis  higher NO~-N concentration measured in the largely agricul-
tural watershed  is possible evidence for the influence of tile drainage  water
on  stream water  quality.

-------
                                     - 19 -
    Table 4.  Rainfall,  runoff, and  flow weighted mean  concentrations, sediment
             and  nutrient  loss occurring  in  two drainage areas of  the Black
             Creek Watershed during the period 1975 to 1978.	
    '""""   ;'J'= =	' "site ' :    	  :L"""   ;  '         :ir::":-
         Parameter


Rainfall, cm
Runoff, cm
Sediment, mg/1

Sediment P,

Sediment N,

Sol. inorg.

Sol. org. P,

NH+-N, mg/1

NH~-N, mg/1

Sol. org. N,


mg/1

mg/1

P, mg/1

mg/1





mg/1

no.
2 & 6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
1975
108
29.1
26.0
732
1435
1.
1.
80
73
10.75
11.13
0.
0.
0.
0.
0.
0.
6.
4.
0.
0.
05
13
04
05
52
70
54
47
80
96
1976
66
12.4
10.1
515
380
0.79
0.72
3.89
2.82
0.05
0.18
0.03
0.04
0.48
0.84
4.49
2.36
0.75
0.73
1977
96
18.5
19.4
236
232
0.90
0.92
2.47
2.42
0.07
0.24
0.03
0.05
0.31
0.67
8.35
6.55
0.60
0.91
1978
77 '
18.5
21.3
206
256
0.35
0.37
3.30
3.24
0.12
0.32
0.04
0.17
0.40
1.44
4.48
2.80
0.90
1.35
1979
79
23
20
273
216
0.40
0.53
2.59
2.47
0.10
0.29
0.03
0.04
0.29
0.65
8.91
4.62
0.78
0.98
1980a
39
12
8
328
325
1.
1.
3.
5.
0.
0.
0.
0.
0.
0.
11,
6.
0.
0.
97
74
10
90
13
20
02
07
48
69
.45
69
88
95
Aveb
85
20
19
392
504
0.85
0.65
4.60
4.42
0.08
0.23
0.03
0.08
0.40
0.86
6.55
4.16
0.77
0.99
   a   lybO averages include data only from first 6 months of 1980
   b   Average excludes 1980 data

Average Monthly Loads

     Average monthly rainfall, runoff, sediment loss and nutrient loss  values
measured  at  Sites  2  and  6  are  given in Table 5 and plotted in figures 2
through 6.  Rainfall was reasonably  well  spread  throughout  the  year  with
April, June and August having the highest monthly averages.  However, as would
be expected due to soil and plant cover conditions, runoff volumes were  larg-
est during the winter and early spring months (December through April) (figure
2).  Sediment losses measured at Site 2 were maximal in Bebruary,  March,  May
and  June,  (Bigure 2), while at Site 6 greatest sediment loss was measured in
March, April, toy and June (Bigure 5).  Highest sediment N and P losses  meas-
ured   at  Sites  2 and 6 (Bigure 3) generally occurred during the same time of
year as high sediment losses.  Ihe highest monthly losses  of  NH*-N,  soluble
organic  N,  inorganic  P,  and  soluble organic P were observed in the period
Bebruary through April (Bigures 4 and 5).  This finding suggests that snowmelt
runoff  and  early  spring rains were responsible for significant transport of
soluble N and P.  Some of the soluble N and P transported likely  was  leached
from   residues  from the previous crop present on the soil surface at the time
of snownelt.  Nitrate-N losses closely paralleled runoff measured at  Sites  2
and 6  (Bigure 6) indicating the importance of surface runoff, subsurface flow,
and tile drainage in the export of NCL-N from subwatersheds within  the  Black
Creek Watershed.                     J

-------
                                  - 20 -
       Table 5.  Average monthly rainfall, runoff, and sediment and nutrient
                       losses from  the drainage area during 1975-1980.a
tenth !

Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
TOTAL
Bite

2
6
2
6
2
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
Rain-
fall

—en
3.2
4.0
7.6
8.9
7.3
10.6
5.6
11.5
6.0
5.0
6.4
5.8
8.19
Total !
runoff

\~- ™
1.01
1.00
2.21
2.79
5.67
5.41
3.15
1.32
1.41
1.32
1.61
1.24
0.16
0.12
0.17
0.29
0.34
0.31
0.13
0.01
0.59
0.52
3.30
2.40
19.8
18.2
Sediment
lost

23
46
121
101
140
139
93
181
166
181
121
312
7
7
2
8
10
10
1
1
11
13
93
65
788
985
NH+-N
lost

0.04
0.13
0.14
0.27
0.26
0.61
0.11
0.07
0.08
0.07
0.06
0.05
0.01
0.01
<0.01
0.01
0.01
0.01
<0.01
0.01
0.02
0.08
0.09
0.14
0.82
1.52
NO--N !
lost

0.83
0.54
1.27
1.19
3.52
2.02
2.55
0.58
0.91
0.58
1.68
0.67
0.12
0.07
0.02
0.05
0.13
0.09
0.02
0.02
0.34
0.17
3.04
1.51
14.43
8.25
Bed iment
N lost

- ——Kg/ na
0.25
0.47
0.86
0.68
1.63
1.48
0.94
1.11
2.38
1.11
2.01
2.61
0.06
0.07
0.01
0.08
0.08
0.12
0.01
0.01
0.14
0.15
0.99
0.82
9.36
9.09
Sol. inorg.
P lost

0.004
0.017
0.020
0.079
0.065
0.169
0.017
0.010
0.007
0.010
0.010
0.012
0.002
0.002
0.001
0.006
0.003
0.010
<0.001
0.002
0.003
0.011
0.032
0.057
0.164
0.419
Sediment
P lost

0.038
0.080
0.231
0.193
0.344
0.344
0.171
0.236
0.356
0.236
0.299
0.229
0.013
0.015
0.004
0.023
0.026
0.038
0.002
0.003
0.039
0.048
0.325
0.355
1.848
1.784
a  first 6 months averages include  6 years of data while last 6 months include 5 years.

-------
                                    - 21 -
   6.5
5.5
                    6      9
                    MONTH
                  69
                  MONTH
Figure 2.  Average monthly runoff and sediment  losses treasured  at Sites
           2 and 6 in the watershed during  the  j-eriod 1975-iySO.

-------
                                     - 22 -
     2. 5
                                            2.7
  cr
  I
  x
  x
  C9
  V

  CO
  CO
  o
 Z
 LoJ
 Q
 U)
 CO
 cr
 CO
 CO
 o
     1.5
69
MONTH
                                 12
                                                           6     9
                                                           MONTH
Bigure 3.  Average monthly sediment-bound N  and P losses measured  at  Sites 2
           and  6  in the Watershed during  the period 1975-1980.

-------
                                      _ 23 -
  oc


  to
  CO
  co
  o
      .3
.24
      18
  ^  . 12

  o

  r;
  cr
     .06
                                  SITE 2
                                    *
                     6      9

                     MONTH
                            12
                                              .7
cr

^  .56
CO
                                    to
                                    w
                                    9  .42
                                        .28

                                    o
                                    r

                                    (T

                                        . 14
                                                                      SITE 6
                                                 369

                                                       MONTH
                                12
  cc

  •s
  CD
  cn
  CO
  o
 
-------
                                    -24-
    .07
                                            .18
6     9
MONTH
                                                           6      9
                                                           MONTH
\
CD
CO
to
o
Q.

O
a:
o
    .02
    .016 .
   .012
   .008
   . 004
                                           .065
                                          .052
                                          .039
                                                                        SHE 6
                    6     9
                    MONTH
                                 369
                                       MONTH
tigure b.   Average monthly  soluble inorganic  P and  soluble organic  P losses
           measured  at Sites  2  and 6 in the K'atershed  during the period 1975-
           1980.

-------
                                    _ 25  _
                      2.1
                                      6      9
                                      MONTH
                                      6      9
                                      MONTH
Figure 6.  Average monthly nitrate N  losses measured et  Sites 2 and 6  in  the
           Watershed during the period 1975-1980.

-------
                                    -  26-


Average Monthly Flow Weighted Mean Concentrations

     Average monthly flow weighted mean suspended solids and nutrient  concen-
trations  measured in drainage water at Sites 2 and 6 are given in Table 6 and
figure 7-10.  Suspended solids  and  sediment-bound  nutrients  concentrations
measured at Sites 2 were highest in February, April and May (figures 7 and 8},
while losses at Site 6 were highest in l^ay, June and July (Figures 7  and  8).
Fortunately,  the poorest water quality from the standpoint of suspended sedi-
ment and insoluble nutrients occurred during the  annual  periods  of  highest
stream  flows.   Soluble organic N concentrations remained relatively constant
throughout the year with ranges of 0.58 to 0.99 mg/1 and 0.89 to 1.21 mg/1 for
Sites 2 and 6, respectively.  For Site 2 the highest soluble organic N concen-
trations occurred in January and September, Figure 8, while the  highest  con-
centrations measured at Site 6 occurred in Narch and June (figure 8).

     Ammonium N concentrations measured at Site 2 were highest in February and
l^ay  and lowest in August and October (figure 9).  At Site 6, NFK-N concentra-
tions tended to be relatively low between April and September, and higher from
October to Inarch (figure 9).  Low NB^-N concentrations in late summer reflects
assimilation by algae and aquatic weed growth in the ditches  throughout  this
period.  The monthly average NH.-N concentrations at Site 6 was usually higher
than that in Site 2 demonstrating the effect of the input of sewage (Table 6).
Nitrate N concentrations measured at Site 2 were highest (>8 mg/1) in January,
April, June and December.  Relatively low NO--N concentrations (<2 mg/1)  were
observed in Site 2 during August and October when tile drains were not running
and little water was present in the ditches (Figure 9).  While  NOZ-N  concen-
trations  measured  at Site 2 ranged widely between 1.18 and 10.44 mg/1, those
at Site 6 remained relatively constant ranging  between  1.63  and  6.29  mg/1
(figure 7).

     Despite the algae and weed growth in the  streams,  soluble  inorganic  P
concentrations  at Site 2 fell below 0.04 mg/1 only in October, and reached as
high as 0.125 mg/1 in July  (figure 10).  Soluble inorganic P concentrations at
Site 6 averaged below 0.163 mg/1 during only 2 months and never averaged below
0.070 mg/1  (figure 10).  Ihe ability of the stream at Site 2 to maintain  both
relatively  high concentrations of soluble inorganic P and growth of algae and
aquatic weeds may be due to supply of soluble inorganic P to the water through
equilibrium  processes by stream sediment enriched with P.  High soluble inor-
ganic P concentrations measured at Site 6 are probably due  to  septic  system
effluents.  Soluble organic P concentrations ranged from 0.001 to 0.0063 mg/1,
and 0.0021 to 0.117 mg/1 for Sites 2 and 6, respectively.  Ihe highest soluble
organic P concentrations occurred in July and Narch for Sites 2 and 6, respec-
tively .

-------
                                   - 27 -
 Table 6.  Average  flow weighted mean concentrations of sediment and
          nutrients measured  in two drainage areas of the Black
          Creek Watershed during the period from 1975 - 1980.a
Month

Jan.

Feb.

Mar.

Apr.

May

June

July

Aug.

Sept.

Oct.

Nov.

Dec.

Total

Site

2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
Suspended
Solids NH*-N


227
479
548
366
247
257
295
376
1177
1375
752
2518
438
550
118
282
294
333
77
124
186
242
282
271
399
551


2.40
1.29
0.63
0.98
0.46
1.13
0.35
0.49
0.57
0.51
0.37
0.41
0.63
0.48
0.12
0.44
0.29
0.30
0.08
0.67
0.34
1.45
0.27
0.60
0.42
0.84
Sol. org
NO..-W N


8.22
5.54
5.74
4.27
6.21
3.74
8.10
4.92
6.45
4.43
10.44
5.39
7.50
5.43
0.18
1.66
38.2
2.87
1.53
1.63
5.76
3.30
9.21
6.29
7.31
4.53


0.99
0.97
0.81
0.89
0.78
1.21
0.79
0.90
0.85
0.93
0.81
1.07
0.63
0.93
0.59
0.94
0.88
0.90
0.77
1.00
0.85
0.94
0.58
0.89
0.77
1.01
. Sed. Sol. inorg,
N P
-mg/1 	
2.48
4.81
3.89
2.44
2.88
2.73
2.98
5.49
16.88
8.41
12.48
21.05
3.75
5.73
0.59
2.86
2.35
3.72
0.77
0.51
2.37
2.94
3.00
3.44
4.74
4.99


0.040
0.17
0.090
0.283
0.5
0.313
0.054
0.163
0.050
0.072
0.062
0.098
0.125
0.200
0.059
0.190
0.088
0.315
0.001
0.164
0.050
0.219
0.097
0.240
0.083
0.230
. Sol. org Sed.
P P


0.030
0.043
0.041
0.043
0.034
0.117
0.035
0.077
0.028
0.054
0.043
0.051
0.063
0.050
0.059
0.054
0.029
0.058
0.001
0.021
0.017
0.038
0.027
0.050
0.030
0.070


0.376
0.829
1.045
0.692
0.607
0.636
0.543
0.918
2.525
1.791
1.8 57
1.847
0.813
1.2J7
0.235
0.776
0.765
1.212
0.154
0.306
0.661
0.919
O.SP5
1 . 397
O.W
0.980
only 5 years of data.

-------
                                      - 28  -
  X
  CD
  51

  co
  Q
  O
  co

  Q
  Ld
  a

  LJ
  CL
  CO

  CO
     1200
960
      720
460
      240
                                    SITE 2
                      6
                      MONTH
                             12
      2.6
     2.08  .
  o

  8
  5  1.04
  s:

  a
  LJ
  CO

      .52
                                    SJTE 2
                                             2600
                                       2080
                                             1560
                                           CD
                                           z:
                                     o

                                     o
                                     o
                                     a
                                     UJ
                                     co
                      6     9

                      MONTH
                             12
                                              520
                                                                            SJTE 6
                                                         6
                                                         MONTH
                             12
                                              1.6
1.2
                                         .8
                                                                            SJTE 6
                                                        6
                                                        MONTH
                             12
Bigure 7.   Average monthly flow weighted  mean suspended solids  and sediment P
            concentrations measured  in drainage  water at  Sites  2  and 6  during
            the period 1975-1980.

-------
                                    - 29 -
                     6     9
                     MONTH
figure 8.   Average monthly flow weighed mean sediment N and soluble organic  N
           concentrations  measured  in drainage water at Sites 2 and 6 during
           the period 1975-1980.

-------
                                      _ 30 _
  to
  z;
  u
  -z.
  o
  o
  •z.
  o
  z:

  a:
  
-------
                                      _ 31 -
  ta
  £1
  O
  o
  O
  0.
  o

  i
  ID
     .15
     ,12
    .09
     .06
     .03
                                  SITE Z
                                        O

                                        O
                                        O
                                        CE
                                        CD
                                        QC
                                        O
                                        O
                                        to
                                            .35
                                            .28
                                           .21
                                             .14
                                             .07
                                                                          SJTE 6
                     6     9
                     MONTH
                                  12
                                                           6      9
                                                           MONTH
                                                                         12
 x
 CO
 a
 o
 o
 o
I
co
tt
o
     .01
    .056
     .042
     .028 .
8  .014
                                             .11
                                            .088
                                            .066
                                            .044
                                            .022
                                                                          SJTE 6
                    6      9
                    MONTH
                                                             6     9
                                                             MONTH
                                                                          12
Figure 10.  Average monthly flow weighted mean soluble  inorganic P and soluble
            organic P concentrations measured in drainage  water at Sites 2 and
            6  during the period 1975-1980.

-------
                                    - 32 -
Nutrient.Concentrations in Sediment

     'ihe average monthly total M  and  total  P  concentrations  in  suspended
utrooii.  sediments  and  N  and  P  enrichment  ratios  for  the  Site  2 and 6
subwatersheds are presented in lable 7 and figures 11  and  12.   The  6  year
average  total N concentrations in suspended sediments from Sites 2 and 6 were
1.190 and 0.894% respectively,  fconthly averages for sediment total N  concen-
trations  at  Sites  2  and 6 ranged from 0.500 to 1.660% and 0.101 to 1.460%,
respectively, with the highest monthly  averages  occurring  June  and  April,
respectively  (figure  11).   Ihe  6  year  average  total P concentrations in
suspended sediments at Sites 2 and 6 were 0.235 and 0.175% respectively.   The
monthly average sediment P concentration remained relatively constant for Site
2  (ranging iron 0.166 to 0.355%), (figure 11), but varied more widely at  Site
6  (ranging from 0.073 to .516%)(tigure 11).  The two highest monthly average P
concentration for both sites occurred during November and December.

     The 6 year average total N enrichment  ratios  for  the  Sites  2  and  6
drainage  areas were 7.2 and 5.8, respectively.  The highest monthly average N
enrichments ratio for Sites 2 and 6 occurred in June and April,  respectively,
(figure  12)  and,  as  might be predicted, corresponds to the highest monthly
average N concentration.  The 6 year average total  P  enrichment  ratios  for
Sites 2 and 6 were 3.5 and 3.8, respectively (Figure 12).

Relationships Between Monthly Runoff, Sediment Loss and Nutrient Loss

     Table 8 and figures 13 through 19 provides data on the  degree  of  rela-
tionship  between  runoff, sediment loss, and nutrient losses from the Sites 2
and 6 areas over the period 1975 to 1980.  Statistical significance is  demon-
strated  for monthly and quarterly regressions when r  values are greater than
0.098 and 0.388, respectively.  Data from both monthly and  quarterly  periods
were used in the correlation studies.

     Losses of Nht-N, NO~-N, soluble organic N and soluble  inorganic  P  were
highly correlated  (r  > 0.60) with the total volume of runoff measured at both
Sites 2 and 6 when both quarterly and monthly data  was  examined   (Table  8).
Regression plots of the above monthly parameters measured at Sites 2 and 6 are
given in figures 13-15.  Average monthly soluble organic P losses measured  at
Sites   2 and 6 were highly correlated with monthly runoff  (r  = 0.89 and 0.65,
respectively) as presented  in  figure 15.  Average quarterly-soluble organic   P
loss  at  Site  2 were also highly correlated with runoff  (r  = 0.85).  Monthly
and quarterly sediment P losses measured at Site 2 were only weakly correlated
with runoff  (r  =  0.33 and 0.43, respectively), as was monthly sediment P loss
measured at  Site 6  (figure  15).

-------
                              -33-
Table 7.  Average  monthly enrichment ratios and totel N and P
          concentrations in suspended sediments in drainage water
          collected at Sites 2 and 6 in the Black Creek Watershed
          (all data calculated from average monthly loadings).
Month
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Overall
Site
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
Nutr ients
Total N
in sediment
Total P
It*-! /I
mg/ j.— — —
1.093 0.166
1.004 0.173
0.710
0.668
1.166
1.063
1.010
1.460
1.434
0.612
1.660
0.836
0.856
1.042
0.500
0.101
0.799
1.117
1.000
0.410
1.274
1.211
10.64
1.270
1.190
0.894
0.191
0.189
0.246
0.247
0.184
0.244
0.215
0.130
0.247
0.073
0.186
0.22
0.199
0.276
0.260
0.364
0.200
0.246
0.355
0.379
0.349
0.516
0.235
0.175
Nutrient .enrichment ratic
Total ND Total Pc
6.5
6.0
4.2
4.0
6.9
6.4
2.7
8.7
8.5
5.0
9.9
5.0
5.1
6.2
3.0
6.1
4.8
6.7
6.0
2.4
7.6
7.2
6.4
7.6
7.2
5.8
2.4
2.5
2.8
2.8
3.6
3.6
2.7
3.6
3.1
1.1
3.6
1.1
2.7
3.3
2.9
4.0
3.8
5.3
2.9
3.6
b.2
5.6
5.1
7.P
3.5
3.8
a) First 6 months averages include 6 years of data,
   while the last 6 months include 5 years of data
b) Average total N concentrations in Sites 2 and 6
   drainage area soils was 1670 and 1674 pg/g, respectively.
c) Average total P concentrations in sites 2 and 6 drainage
   area soils was 680

-------
                                      - 34 -
      1.8
 *   1.44
     1.08
 Q
 UJ

 CO
 
-------
                                      -  35 -
   a:
   Of.
   .


  O
  •—»

  ae.

  LU
     5.5
 a:
 a:
 LJ



 O
 uj

 a.
     3.3
     2.2
     1 . 1
                     6     9

                     MONTH
                                             1.8
                                   SITE 2
12
        
-------
                                    -36  _
     Quarterly and monthly sediment-bound nutrient (N and P)  losses were  well
correlated (r  > 0.65)  with sediment loss measured at Sites 2 and 6 (Table 8).
fcigure 6 presents average monthly sediment N and P losses measured at Sites  2
and  6  plotted  against  monthly sediment loss.  Monthly soluble NH,, soluble
organic N and soluble organic P losses measured at Site 2 were  weakly  corre-
lated  with  sediment loss (Table 8, Figure 17), as were quarterly averages of
the seme parameters measured at Site 2.  Monthly (Figure 18)  sediment  N  loss
and  sediment  P loss measured at Sites 2 and 6 were strongly correlated (r  >
0,67).  While soluble inorganic P and sediment P losses were not significantly
correlated,  soluble  organic  P  and NH.-N losses were highly correlated with
soluble organic N loss when quarterly and monthly averages measured at Sites 2
and 6, (tigure 19) were examined (r  > 0.74).

     f*onthly NH.-N losses were weakly correlated with sediment losses measured
at  Site  2   (figure 18) but not at Site 6, indicating the influence of septic
effluents upon Nh.-N levels at Site 6.  The NH^-N losses measured at  Site  2,
on  the  other hand, appeared to somewhat dependent on sediment N losses meas-
ured at Site 2.  The relationship between monthly soluble organic P and  Nh.-N
losses  measured at Sites 2 and 6 to soluble organic N losses are presented in
tigure IS.

     These findings suggest that monthly or annual loadings of  NH.-N,  NCu-N,
soluble  organic  N,  soluble  inorganic  P,  and  soluble organic P leaving a
watershed can be approximated by multiplying the  annual  flow  weighted  mean
concentrations  by  the  volume  of  runoff during the period.  This procedure
predicted the five-year average monthly  loadings  with  reasonable  accuracy.
however,  when  applied to individual year monthly data significant deviations
from observed values were obtained.  This approach may prove useful in  models
of soluble nutrient transport from watersheds.

     The above findings also indicate  that  monthly  or  annual  loadings  of
sediment-bound N and P leaving a watershed can be estimated by multiplying the
mean total N and P concentrations, respectively, in sediment by the amount  of
sediment  discharged.   This procedure predicted the average monthly sediment-
bound nutrient discharges with reasonable accuracy.   However,  when  used  to
calculate monthly losses of sediment-bound nutrients for individual years sig-
nificant differences from measured values were obtained.  It  is apparent  that
this  approach  will  be  valid  for   use in models which predict transport of
sediment-bound nutrients.

-------
      Table 8.
                                     - 37 -
Relationship between total runoff, sediment loss, and nutrient
 losses from the Sites 2 and 6 drainage area during 1975-1980

Variable 1

Total runoff
Total runoff
Total runoff
Total runoff
Total runoff
Total runoff
Total runoff
Total runoff
Sediment loss
Sediment loss
Sediment loss
Sediment loss
Sediment loss
Sediment loss
Sediment loss
Sediment loss
Sediment loss
Sediment loss
Sediment loss
Sediment P loss

Variable 2

Sediment loss
NH4-N loss
NOv-N loss
Sol. org. N loss
Sediment N loss
Sol. inorg. P loss
Sol. org. P loss
Sediment P loss
NH4-N loss
NO~-N loss
Sol. org. N loss
Sediment N loss
Sol. inorg. P loss
Sol. org. P loss
Sediment P loss
Total N in sediment
Total P in sediment
NHT-N loss
NH*-N loss
Sol. inorg. P loss
Sol. org. N loss Sol. org. P loss
Sol. org. N loss NH^-N loss
a) Statistical
is indicated


Site

Quarterly"

0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
c.
0.
0.

51
81
70
94
399
60
85
43
57
26
46
94
09
61
92
06
00
48
20
08
85
81
2



tonthly"

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
C
0
0
0
0
0

.39
.75
.75
.95
.31
.78
.89
.33
.45
.23
.33
.91
.16
.39
.90
.01
.00
.36
.16
.15
.98
.74


Site

Quarterly
2
I,
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
significance for monthly data at the 1% confidence

25
66
76
P8
29
72
49
39
02
19
17
9P
01
03
83
02
09
23
23
04
76
84
level
6



Konthly

0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.


21
72
75
90
27
83
65
43
03
14
14
95
03
CA
65
00
02
IB
18
14
87
P9

by r > 0.099 (66 replicates)
b) Statistical significance for2quarterly data at the 1% level of
   confidence is indicated by r  > 0.388
Quarterly Trends in Sediment and Nutrient Losses

     Quarterly runoff losses of sediment and nutrients measured at Sites ? and
6  are  presented  in  figures  20-24.   As seen in Figure 20, annual peaks in
runoff volumes measured at Site 2 seem to have been on a slight  decline  over
the  22  quarter  sampling  period, possibly as a result of runoff and erosion
control practices implemented in the Site 2 subv-etershed.  however, there   (Hcjurs; 20).  ^u
-------
                                       - 38 -
     11.5 _
          SITE 2
          K2=0.15
 u.
 u.
 o
 2
     9.2 .
     6.9 .
     4.6
             51
     !fl
     ffl
ffl     Q


 H  B
              Ijl
                   fij
         0      .1     .2     .3     .4
               fWIONIUM  N  LOSS.KG/HH
     11.5
      9.2
  o   6-9
  u.
  u.
      2.3
           SITE 2
           R2=0.75
                     °ffl O
                                  13
                                              10. 4  .
                                                    SITE 6
                                 c 5,     O
                                 3. t . o
                                               2.6 t
                                     0      .5     1      1.5    2
                                           flMNONIUM N LOSS,KG/Hfl
                                                13
                                              10. 4 .
                             g   7.8
                                               2.6
                                      SITE 6
                                      R2=0.75
                1.6    3.2    4.8    6.4
                NITRHTE N LOSS.KG/HH
                                            1.2   2.4    3.6   4.6
                                           NlTRflTE N LOSS.K6/Hfi
tigure  13.   Relationships between monthly  runoff volimes and monthly losses  of
             ammonium  N and  nitrate  N at Sites 2 and  6  during the  period 1975-
             1980.

-------
                                      _  39 _
  o
  u.
  o
      11.5
       9.2
       6.9
       4.6
       2.3
            SJTE 2
            R2=0.95
                &P
           0      .21    .42    .63    .84
               SOL.  ORGflNIC N LOSS.KG/HH
       11.5
        9.2  .
    o
    u.
    u.
    o
    i   4-6
        2.3
             SJTE 2
             R2=0.78
0      .04    .08    .12    .16
   SOL.  INORGflN]C P LOSS,KG/HH
                                                 13
                                                10.4
                                 O
                                 1   5.2
                                                2.6
                                          SITE 6
                                          R2=0.90
                                         0     .44    .88    1.32   1.76
                                            SOL. ORGflNJC N LOSS.KG/Hfl
                                                 13
                                                10. 4
                                 5   7.8
                                 u_
                                 u.
                                 o
                                 1   =.2
                                                2.6
                                          SJTE 6
                                          R2=0.83
                                                    0      .11    .22    .33    .44
                                                       SOL.  JNORGRNJC  P  LOSS.KG/HH
Figure  14.   Relationships between monthly runoff volumes and  monthly  losses of
             soluble  organic  N and soluble inorganic P rt  Sites 2 and 6  during
             the period  1975-1980.

-------
RUNOFF,CM.
                                         RUNOFF,CM.
c
CD
1— '
( n
\j <
9
S2?
to
f Oi
o' CT rr
O-' I—* *~"*
m o
(-• 3
V ^ V>
•~* 2 v
01 12 *""
£§^
00 o' 5"
o ° m
• rt
*o €
o
Qj **t
G, g


UI *"^
a M
§'•<
(5 »
3 S
O
*T3 ^i
f-ti
Q)
fT ^*
f-«
tX g
^" CD
fl)
w Q»

O 3
D O
*•** ^^
IT
a\ ^
Q*^
C i *
^"t D
3* W
i£) (D
tn
^0

i\> * o> to " r° * ? 5°
sLoO)(DWUlBu)0'(D^01
,^^ •• 	 1 	 1 • " 	 '[ •— 'l 	 , 	 	 	 1
S1
(n
m •
O •*•
i- r\>
m
•z
—{ ,

o
cn
(n ^
A po
^ 0)
I
3)
0)
00


^f'8,,8
8 B B
a B
.8 G

B a



B


30 (n
CS3 »-
II -1
O rn

to 10
U)
a
rf8 » (n
cn li o m
2 B s « M
.n • h • B a a o
S 1
o —
cn
3)
2 .
"8
^
0 •
«" cn ro oo -*» u)
uD



-------
                                     - 41 -
yaw


^ 760
to
% 570
o
_j

2
£ 380
O
Ul
to
190



D
SJTE 2
R2=0.91


s>
0





- *D • •
0 «
"IB ™

0 iBr iiii



cr 1440
CD
co 1080
o
_i
h-
Z
£ 720
a
Ul
to
360



0 i

SJTE 6
R2=0.95


0







ffl o
o o
• I,0
f • D
f ' 	 1 1 	 1 	
0 3 6 9 12 0 3 6 9 12
SEDIMENT N LOSS.KG/Hfl SEDIMENT N LOSS.KG/Hfl
otrr, 1 Rflfl
£?*3I0


<£ 760
~"
CD

*
%> 570
o
i—
•7
£ 380
3
LJ
CO
190



0
ID
SJTE 2
R2=0. 90





a

a
.



n „ °
D * _

% ° 0
fc* "



K 1440
\
to
^
to
lo 1080
o
t-
^ 720
a
Ul
CO
360



0

SJTE 6
R2=0. 65





O






.
o o
O (D
* =ta
DOCD % 0 CD 0
IjL BQ O , ,
               .42    .64    1.26  1.68
              SEDIMENT P LOSS.KG/Hfl
 .3    .6    .9
SEDIMENT P LOSS.KG/Hfl
figure 16.   Relationships between monthly sediment losses  and  monthly  losses
            of  sediment N and  sediment P  at  Sites 2 and  fi during the  period
            1975-1980.

-------
                                     -  42  _
     950
                                             950


H 760
\
CO
co 570
o
_i
>—
z
£ 380
Q
UJ
CO
190

0 ,
SJTE 2
R2=0. 45



D

a





' - " " B° ' " "
g ^ ^
JK°I ja.,0. m1"0 o , ,


^ 760
V.
CD
co 570
o
_i
t—
z
£ 380
Q
UJ
co
190

0 .
SJTE 2
R2=0.33



g>

n





-.:-••
0 ^P-B"°

0 .1 -2 .3 .4 0 .21 .42 .63 .84
flMMONIUM N LOSS.KG/Hfl SOL. ORGflNJC N LOSS.KG/Hfl
                         cc

                         N
                         CD
                            950
760
                         S  570
                         o
                         2
                         UJ


                         Q
                         UJ
                         CO
380
                            190
                                                         SJTE 2
                                                         R2=0.39
                                      .01    .02    .03    .04
                                   SOL.  ORGflNJC P LOSS.KG/Hfl
tigure 17.   Relationships between monthly sediment losses and  monthly losses
            of  ammonium N, soluble organic Mf and  soluble organic P at Site  2
            during  the  period 1975-1980.

-------
SEDIMENT N LOSS.KG/HR
fl>
,_,
00
•


VC Hi ?D
^•J 1— i
TjBft
^ 0- «'
ID M. O
00 g D
•°s s-
rt- »—
T
^D-

s> u) o> co w

Q^ '
Ipe » <2
j 8 it -i
,„ B a 01
en i o
m-
D u)
3
n
2
-o'°>

SEDIMENT N LOSS.KS/HR S
3 | - - ?«
O<^ s oo o> co i\) eji £J
m -
g 3 s>'

3 3
O O
2-R
J- C. m •
i M 2s
•< 3
2: m
•• W 2
JD -1.
S M ^°°
ft g' *
CO ^ 0
(-• 2. w
ri- rr cn^
fC 1. •
w ?- SK
M % I
O 3>
n, cn —
3 cn
K ID o>
a w °°
j>
-" cu
ag,
c a
-^
s§
rf
rr 3"
br t-1
(C •<
^6 Q1
i-r cn
M- cn
O CD
Oi tn

IB
SB . S£
Ba II -1
B B B o m
B a » w
8*


S
'




8





e
^
3)
»-•

l\)

BBI
8 e

e
s
B




B


e



SEDIMENT N LOSS.KS/HR

(s to o> ID ro

-------
                                     -  44
   1.05

^
o5
Q . 18
Q-
o
z -12
c
UJ
g

J .06
o
0)

0 <
SJTE 6
R2=0. 87

*










D


V
0
iffi^*' ff
              .21   .42    .63    .84
            SOL. ORGflNJC N LOSS.KG/Hfl
0     .5     1      1.5    2
    SOL. ORGflNIC N LOSS.KG/Hfl
figure 19.  Relationships  between monthly soluble organic  N losses  and  losses
            of  ammonium N and soluble organic  P at Sites 2 and 6  during the
            period  1975-1980.

-------
     14
   10.5
o

2   3.5
    1600
    1200
     800
z

£   400

3
Ul
co
                     8      -12

                       .QURRTER
                     8      12

                       QUflRTER
                                        SJTE 2
                                 16    20
                                         SJTE 2
                                  16     20
                                                      16
     12
  u_
  O



  \   4
                                                   3000
^  2250
W

CO
O
_J
                                                Ul
                                                s:

                                                Q
                                                Ul
                                                cn
                                                    1500
    750
                     8     12     16

                       QURRTER
                                                                                         SJTE 6
                                                                                        20
                                                                                         SJTE 6
               4      8     12     16    20

                       QUflRTER
                                                                                                                   UT


                                                                                                                    I
        Eigure kO.   Quartetlv  runoff  volumes  and  sediment  losses  at Sites  2  and

                     beginning with the first calender  quarter of 1S75.

-------
      25

-------
CO
CO
o
_J


z


LU
6.75
2-25
                    8      12

                      QUflRTER
                              16
                                    20
                                                 cr

                                                 x
                                                 CO
                                             to
                                             to
                                             o
                                             2


                                             LJ
                                             
-------
I
X
CO
co
CO
o
o


cc
o
CO
    1.1
    .825
    .55
   .275
                                         SJTE 2
                                             CC


                                             X
                                             CO
                                             co
                                             co
                                             o
                                                 z

                                                 o
                                                 cc
                                                 CO
                                             O
                                             CO
                                                     2.4
                                                     1.8
                                                     1.2
                     8      12

                       QtlflRTER
                                  16
                                        20
                                                                 8      12

                                                                   QUflRTER
20
 cc

 N
 CD
 to
 co
 a
 CL

 O


 Z

 CO
 a:
 o
 _
 o
 CO
.08
                     8      12

                       flUFIRTER
                                                                                                            oo

                                                                                                            I
                                                                 8      12

                                                                   QURRTER
      figure 23.   Quarterly losses of  soluble organic N and soluble inorganic P at

                   Sites 2 and 6 beginning with  the first calendar quarter  of 1975.

-------
I
to
to
CO
O
Q-

O


cc
co
a:
o
o
co
   .064
   .048
   .032
.016
                                              a:

                                              x
                                              (D
                                              CO
                                              CO
                                              a
a:
CD
a:
o
                                              o
                                              co
                                                     .28
                                                     .21
                                                    .14
                                                 .07
                    8      12
                      OUflRTER
                                                           10


                                                           I
                                                                 8      12
                                                                   QUflRTER
                                       20
      tigure 24.  Quarterly losses of  soluble organic P at  Sites 2  and 6  beginning
                  with  the  first calendar quarter of 1975.

-------
                                    -50 -


high (usually > 6 fold)  when compared to peak losses occurring in later years.
There  is  a  suggestion  that quarterly losses of sediment and sediment-bound
nutrients have declined during the course of the study in response  to  imple-
mentation of bhPs.  Even though runoff has declined slightly over the 22 quar-
ter sampling period, NO.--N losses measured, at Sites 2  and  6  appear  to  be
increasing  (figure 22).  The increasing NO~-N losses over the sampling period
may be an indication of increases  in  N  fertilization  in  the  subwatershed
and/or  increase  in  NO.^1  leaching from soils and loss via tile lines.  The
increased in NO~-N leaching may be a symptom of  runoff  and  erosion  control
measures  taken  within  the  watershed  which allow increased infiltration of
rainwater.

     Quarterly losses of other soluble constituents (NH.-N, soluble organic N,
soluble  inorganic  P,  and  soluble organic P) measurea at Sites 2 and 6 gen-
erally followed similar trends.  That is,  most  soluble  constituents  showed
peaks  during  the  2nd, 5th, 9th, 13th, 17th, and 21st quarters with maximums
measured during the 2nd, 13th and 17th quarters (figures 23-24).  The  maximum
losses  of  the  soluble  constituents  corresponded to the periods of maximum
runoff.

Trends _in Quarterly Sediment and Nutrient Concentrations jin Drainage Water

     Quarterly flow weighted  mean  concentrations  of  suspended  solids  and
sediment-bound and soluble nutrients measured in drainage water at Sites 2 and
6 over the  22 quarter sampling period are given in figures 25-28.   Concentra-
tions  of   solids, sediment N, and sediment P were normally highest during the
first and second quarters of each year; however,  in  some  cases  the  fourth
quarter  also had high concentrations of sediment and sediment-bound nutrients
(figures 25 and 26).  Highest concentrations of  sediment  and  sediment-bound
nutrients   in  drainage  water were measured during the second quarter of 1975
where the large rainfall event transported large  amounts  of  soil  material.
Interestingly, a large runoff event which occurred during the 22nd quarter did
not result  in high quarterly suspended solids and sediment-bound nutrient con-
centrations.   During the period 1978 to 1980 the quarterly sediment P concen-
trations tended to be considerably higher at Site 6  as  compared  to  Site   2
likely as a result of septic tank effluents.

     Ihe quarterly nitrate N concentrations tended to increase  with  time  at
both  monitoring sites  (figure 26).  This  increase in nitrate concentration in
drainage water may  reflect increasing rates of fertilizer or  manure  applica-
tions  in   the watershed or may result  from increased nitrate in tile drainage
water.  Ihe implementation of best management practices  in the  watershed  may
have increased the  proportion of rainwater which reaches streams by subsurface
flow at the expense of surface runoff.   Tile drainage water normally has a two
or  three-fold  higher  nitrate N concentration than does surface  runoff.

     Ihe  quarterly  ammonium N and  soluble  organic N concentrations  in drainage
water  remained  relatively constant  throughout the period of measurement  (Fig-
ure 27).   Highest concentrations were normally observed  during  the   first  and
fourth  quarters  of  each  calendar year.  Drainage water  flowing  past  Site  6
always had  higher ammonium N concentrations than did Site   2   drainage  water.
Ihe concentrations  of soluble  organic N were  similar at  Sites  2 and  6.

-------
N
co
u

o
o

to
Q
O
CO
Q
UJ
Ul
OL
   1200
    900
600
300
                                        SJTE 2
                                                CO
                                                z:
                                            o

                                            o
                                            u
                                            o
                                            to
                                            Q


                                            Ul
                                            Q.
                                                   2600
                                                   1950
                                                   1300
                                                650
                                                                                        SJTE 6
                           12
                      QUflRTER
                             16
                                                                8     12     16
                                                                  QUflRTER
20
  x
  CO
  z:
  o
  CJ
  Q
  Ul
  CO
                                             X
                                             CD
                                             O


                                             §
                                             O
                                             Ul
                                             s:

                                             3
                                             Ul
                                             CO
                           12

                      QUflRTER
                                   20
                                                                8     12     16     20
                                                                                                               1


                                                                                                               Ul
     tigure 25.   Quarterly flow  weighted  mean  concentrations of  sus[«nded  solids

                 and sediment N measured  in  drainage waters at Sites 2 and  6  begin-

                 ning with  the first calender  quarter of  1975.

-------
CD
z:
u
o
u
Q
UJ
co
   2.8
   2.1
1. 4
                                                  2. 4
                          12
                     QUflRTER
                                                               8      12
                                                                 QURRTER
                                                                                     20
 N
 IS
 O
 o
 u
 
-------
 O

 o
 u
    1.2
     .9
 .6
 =;   .3
 o
 z:
 (T
                                        SJTE 2
                    8     12
                      QUflRTER
                                   20
                                                  x
                                                  CO
   O

   O
   O
                                              o

                                              
-------
N
to
o
o

CL

O



-------
                                    - 55 -
     There was a tendency for quarterly flow weighted mean soluble inorganic P
concentrations  to  increase  with  time  at Site 2  (ligure 28).  However, the
soluble inorganic P concentrations measured at Site 6  exhibited  no  apparent
trend  except  that  quarterly  data for 1979 and 1978 appeared to be somewhat
lower than that for 1976 and 1977  (tigure  28).   The  new  interceptor  sewer
installed in Harlan may be somewhat responsible for lowering soluble inorganic
P concentrations measured at Site 6 during 1978 and 1979.   Quarterly  soluble
inorganic P concentrations at Site 6 were always at least two fold higher than
those at Site 2.  The concentrations of soluble organic P were remarkably con-
stant  at  both  sampling  sites  except for a shfrp increase in concentration
measured during the first three quarters of 1978.  Ihere is no apparent expla-
nation for the more than two fold increase in soluble organic P concentrations
measured during 1978.
                                  REFERENCES

 1.  International Reference Group on Great  Lakes  Pollution  from  Land  Use
     Activities.  1978.  Environmental management str^t^-cy for the Great Lakes
     system.  International Joint Commission, Windsor, Ontario.

 2.  Lake, J. and J. Morrison.  1977.  Environmental impact  of  land  use  on
     water  quality.  Final report on the Black Creek Project (summary).  U.S.
     Environmental Protection Agency, Chicago, IL.  KPA-905/9-78-001.  p. 3-9.

 3.  American Public Health Association.  1971.  Standard Methods for Examina-
     tion  of  Water and Wastewater.  13th ed. Am. Public Health Assoc., Wash-
     ington, DC.

 4.  U.S.  Environmental  Protection  Agency.   1971.   Methods  for  Chemical
     Analysis of Water and Wastes.  U.S. Environmental Protection Agency, Cin-
     cinnati, Olio.  16020—07/71.

 5.  Steel, R.G.D. and J.H.  Torrie.   1960.   Principles  and  Procedures  of
     Statistics.  McGraw-Hill Book Co., Inc., New York.  482 p.

 6.  Nelson, D.W. and D.B. Beasley.  1978.  Quality of  Black  Creek  drainage
     water:  Additional  parameters.   In  Environmental impact of lend use on
     water  quality-supplemental  comments.   U.S.  Environmental   Protection
     Agency, Chicago, IL.  EPA-905/9-77-007-D.  p. 36-83.

 7.  Nelson, D.W., E.J. Monke, A.D. Bottcher, and L.E. Sommers.  1979.   Sedi-
     ment  and nutrient contributions to the Maumee River from an agricultural
     watershed.  Ir\ R. C. Loehr (ed.).  Best Management Practices for Agricul-
     ture and Silviculture.  Ann Arbor Science, Ann Arbor,  M.  p. 491-505.

 8.  Logan, T.J. and R.C. Stiefel.  1979.   Maumee River pilot watershed study.
     Watershed  characteristics  and pollutant loadings, refinance Aret, Chio.
     U.S.  Environmental Protection Agency, Chicago,  IL.   EPA-905/9-79-005-A.
     135 p.

-------
                                    - 56  -
 9.   Ellis, B.C., A.E.  Erickson, and A.R.  Wolcott.  1978.   Nitrate  and  phos-
     phorus  runoff  losses  from small watersheds in Great lakes Basin.  U.S.
     Environmental Protection Agency, Athens, GA.   EPA-600/3-78-028. 84 p.

10.   Langaale, G.V».'., R.A. Leonard, W.G.  Eleming,   and  W.A.  Jackson.   1979.
     Nitrogen  and chloride movement in small upland Piedmont watersheds.  II.
     Nitrogen and chloride transport in runoff. J. Environ. Qual. 8:57-63.

11.   Alberts, E.L., G.E. Schuman, and R.E. Burnnell.  1978.   Seasonal  runoff
     losses of nitrogen and phosphorus from Missouri Valley losses watersheds.
     J. Environ. Qual.  7:203-208.

12.   Mtnzel, H.G., E.D. Rhoades, A.E. Olness, and  LJ.J. Smith.   1978.   Varia-
     bility of annual nutrient and .sediment discharges in runoff from Oklahoma
     cropland end rangeland.  J. Environ.  Qual. 7:401-406.

13.   Kilmer, V.J., J.Vv. Gillian, J.t. Lutz, R.T. Joyce and C.D. Eklund.  1974.
     Nutrient  losses from fertilized grassed waterways in North Carolina.  J.
     Environ. Qual. 3:214-219.

-------
                                   -  57 -



                             MAINTENANCE OF BMPs

                                 R.Z. Wheaton

     At  the start  of the  project  over  30   land  treatment  practices  were
included in the planning process.  The intent was to give each practice a fair
trial.  With experience it was found that only a limited number were suited to
the soil type,  topographic  conditions, and production operations of the Black
Creek Watershed and were also important in making water quality improvements.
Many of  the remaining practices were improvements to land mostly unrelated to
water quality.   In  general,  structure practices were preferred  over  cultural
ones even if the structure measure had a somewhat higher initial cost.

     Some of the most  acceptable measures were  field borders,  animal  waste
holding  tanks,  sediment  basins,  grass waterways, tile outlet terraces, criti-
cal  area planting,  livestock exclusion  (fencing),  and  pasture  renovation.
Also  important  were erosion  control  structures,   tile outlet pipes,  rock
chutes,  stream  bank sloping  and seedling, and  riprap for channel  stabiliza-
tion.  Conservation tillage (no-till or low-till)  and other cultural practices
were encouraged where applicable.

     Because drainage is an  essential production  practice  for  much of  the
watershed,  landowners were  amenable  to practices which reduced erosion in the
streams since they also hoped these same practices would  improve the  drainage
outlets.  On the other  hand it was  necessary  to  demonstrate that, when prop-
erly selected, many cultural practices could be used  without significant loss
in crop production.

     Even with  the  higher initial cost, structural  measures were  often pre-
ferred  because  of  their  low maintenance and low annual  costs.   A structural
BMP program  is  easier to administer than one containing  cultural BMPs.   When
properly designed  and installed,  structural  measures  provide  the protection
for which they  were planned.  Cultural measures are  more influenced  by soil
properties and varying climatic and economic conditions.  It is also more dif-
ficult  for  both the  landowner  and  planner  to assure  compliance  with  water
quality goals.

     The land  treatment measures  have performed well.   Their  acceptance  by
landowners, their maintenance and the continued acceptability after the end of
the project will be discussed later.   Stream channel  stabilization  measures
were well  accepted and  performed satisfactorily even  though there  was some
early reluctance by landowners to give up the land necessary for ditches with
2:1 or  flatter  sideslopes.   Planners and researchers were  especially pleased
with the rock drop structures and the channel containment accomplished by lin-
ing the  toe of  the channel banks  in critical  areas with rock.  Both of these
measures were somewhat experimental when initially used.

     The in-channel destining  basin was essentially  filled after the  first
few years.  It would now be very expensive to be cleaned out.  There is also a
problem of where the  soil (sediment material)  could be  spread.   However,  the
need for a basin  has largely disappeared now that  upstream construction has

    Acknowledgment:  Data and assistance were provided by the SCS District
    Conservationist, Mr. T.D. McCain.

-------
                                   - 58 -
been  completed.   On the  other  hand, the  sediment pond  is  still very  func-
tional.   No measurable  sediment accumulation  in  the sediment pond  occurred
between 1977 and 1979 because (1) land treatment measures above the  pond were
completed  and  (2)  major  runoff events such  as the  occurrence  of a  50 year
storm soon  after  the sediment pond was constructed  have not occurred.   This
pond  should  continue to  serve its function for many years.   The present rate
of sediment accumulation would suggest that its life would be at  least 50-100
years.

     Planned land  treatment  in  the Black Creek Watershed (1973-1977) can  be
evaluated simply by  observing what is "on the land" for the permanent (struc-
tural) practices.  These  "landmarks"  serve as  reminders  to  farmers  and have
been  generally  regarded  as necessary and  worth  keeping.   Unfortunately,
management-type (cultural) practices  such  as rotations,  conservation tillage
or waste disposal, lacking any "landmarks," have lost their "reminder" status.
Cultural  practices are easily overlooked and adherance to contractual agree-
ments as  during the  project years is now not necessary.  SCS and SWCD efforts
did make  inroads in changing the attitudes of the  farmers toward  making  water-
quality improvements, but cultural practices are difficult to implement year-
after-year when weather,  crop prices, machinery changes,  etc.  require ongoing
changes in management decisions.

      Failure to implement cultural practices to control erosion can also shor-
ten  the  life of  terraces, waterways,  sediment basins, and  other structural
practices.  Then when reconstruction is needed  it may be  more costly than for
the original construction and excessive soil erosion also may have resulted in
reduced productivity.

      Cultural practices are usually maintained by their continued utilization.
However,  structural measures such as pipe inlets,  dropped spillways and chutes
need  to be  observed  for  signs of erosion or  shifting.   Noted  problems should
be  corrected immediately.   Grassed waterways  and  field borders need  to  be
maintained  by fertilizing and clipping.  Small wet  spots, scouring  or  other
problems  in the waterways  or borders  should be  corrected and any  silt bars
removed.  Terraces will required periodic rebuilding of ridges and removal  of
sediment  from the  channels.   All practices should be maintained  as  nearly as
possible  to their original condition.  Recognizing the need for maintenance is
just  the  first step.  Accomplishment  may be more difficult and  some type  of
incentive program may be needed.

      Plans  for  land treatment  systems  should  include  maintenance schedules.
Plans could even hint that failure to maintain  structural practices may result
in  the  return  of  some of  the original  cost-share money.  Agreements between
landowners  and local  districts  should include  provisions  from maintenance
beyond the  life of a particular project.  Economic benefits need  to  be better
developed and described to the operator.

      As already mentioned,  incentives  for maintenance  are  probably needed.
They  can  range  from  financial aid  to just  recognition of jobs well  done.
Periodic  contact of  landowners by district personnel and yearly questionnaires
about their practices have  proven beneficial.  A strong educational program
including printed materials about maintenance should be carried out.   Finally,
there is  a need to  develop new "ideas" for encouraging landowners to maintain
their land  treatment practices.

-------
                             -  59 -
                 FIELD EXPERIENCES AND PROBLEMS

                           R. E. Land

     There has been an expression of interest in an essay type report out-
lining some of the field experiences and particularly the problems in-
volved in setting up, maintaining, and servicing monitoring sites and
equipment on the Black Creek Project.  In writing this report, it is assumed
that the reader is familiar with the Black Creek Project—its location and
purpose.

     When I became a part of this project as Field Coordinator  on April
1, 1973, most of the plans, organizational structure, area to be investi-
gated, etc. for the project had been established.  It became one of my
primary functions to take some leadership in locating and establishing moni-
toring sites, and then servicing and maintaining those sites.

     It wasn't planned or the project area was not selected for this reason,
but the watershed was of such shape and topography so as to have excellent
and convenient monitoring locations.  There are five main tributaries to
Black Creek.  Entrance of these drains into Black Creek lie along the same
county  road—mostly in a straight east-west line.  So it became no task
at all to select easily accessible and strategically located sites for moni-
toring these subwatersheds.  Easily accessible sites were also avaiable in
the upland areas for monitoring special BMP's, however, in spite of this
fact, and over my objections, some monitoring stations were selected in the
upland area that were remote, not easily accessible, thereby making servic-
ing and maintaining of those sites a most difficult task—especially during
wet periods when intense monitoring was important—needless to say, some
data was lost due to this very fact.  Other sites, more accessible, were
available for monitoring the same selected BMP's.  We were never equipped
to monitor remote stations.  Mention of this problem is made for the bene-
fit of those who may at some time work with a field monitoring program, and is
not intended as a reflection on the decisions already made at "Black  Creek."

     At the start of the project—using Agricultural Handbook No. 2241 as a
guide—six weighing type rain gauges were established to cover the 12,000
acre watershed.  Those stations were located so as to give, in our opinion,
the best rainfall readings for specific areas being monitored for water
quality.  The   rain gauges have worked well over the project period; main-
tenance was simple—consisting mainly of periodical recalibration, keeping
oil in the receiving bucket to prevent evaporation, changing charts every
six days, and other small maintenance chores.  Originally, some of the rain
gauges were equipped with box type recording pens which I found did not work
well where dust was a factor.  These pens were exchanged  for  trough type
recording pens which did the job with few problems.  We found that it was
necessary to mount these instruments on a solid base to prevent vibration by
the wind and etc.   We have collected good, reliable precipitation records
throughout the project.

-------
                              -  60 -

     Early in the program, flow measuring stations were set up at the exit
of each of the five tributaries flowing into Black Creek plus three others—
one on lower Black Creek, one on a reference drain outside the study area,
and one on a tile drain study.  It was requested that I consider methods
of constructing "control sections" across the streams at each of the sites.
Having had construction experience, I felt that sheet piling was the best
material for this particular job.  Bid documents were drawn up, and the con-
tract was let to a local contractor who did an excellent job in driving and
cutting the piling to form V-shaped weirs.  We found that  for this type
"control section," it was necessary to place rip-rap immediately down stream
from the weirs to dissipate flow energy to prevent deep "wash-outs" of the
stream bed.  Foxboro continuous air-type stage recorders were placed at
each of the flow measuring sections.  To prevent flooding of the instru-
ments, it was necessary to place the recorders farther from the stream than
recommended.  The length of air hose required to reach the stream caused
an excessive pressure drop in the line when an air bubble was emitted which
in turn caused the recording pen to deflect in a wide tracing—thereby caus-
ing some undesirable results, such as, difficulty in chart reading, ink runs
on the charts, etc.  A bubble emitter pipe was connected to the air hose and
placed in the middle of the streams above the weirs—this method did not
work since the pipe was continuously catching debris.  The bubble pipe was
eventually placed in a sump.  The recorders required considerable service
in order for us to obtain acceptable readings.

     With one of the biological investigator, we set up fourteen grab sample
stations which covered water sampling over the entire watershed plus the pre-
viously mentioned reference station outside the watershed.  Grab samples were
taken during each runoff event and an effort was made to sample the streams
on the rising side of the runoff hydrographs.  The grab samples (500 ml.)
were simply taken with a line attached to a bucket which was dropped into
the stream.  Two separate samples were taken at each stop.  In most cases,
the stream's flow at the sampling sites was in a rolling motion during events
and mixing appeared good—so it is believed that we have taken representaive
samples.  Weekly "low flow" samples were also taken at these sites.  Stage
was recorded at the time of sampling, as well as recording water temperature,
dissolved oxygen, turbidity, pH,  plus other pertinent  data.  In addition
to the grab sampling stations, twenty-one tile sampling stations were estab-
lished with samples taken on a weekly basis.  During the project, many "sets"
of special grab samples were taken at various stations.  Rain and snow
samples were also taken.

     A year or so after the start of the program, three automatic samplers
(PS-69 samplers purchased from the U.S. Interagency  Sedimentation Project)
were placed on the main streams of the watershed.  These samplers held
seventy-two 500 ml sample bottles, however, if serviced, they were capable
of continuous sampling and on occasion have operated eighty hours straight.
The machines were originally timed to take samples every 15 minutes.  A
float switch, placed in the stream, was set to activate the samplers at
approximately one foot of stage.  For us, these samplers were a high main-
tenance machine requiring careful attention to all mechanical and electroni-
cal details.  However, many of the problems were not directly attributed  to
the machines, such as frozen intake lines, debris over the  intake lines,
float switch failure, and on and on.  Twice the samplers were modified in an

-------
                              - 61 -

attempt to make the machine do extra work, but the net result was a com-
pounding of the problems.

     In addition to the three large samplers, eight smaller automatic sam-
plers were placed at select BMP sites, plus one sampler at a special drain
tile study.  These machines held forty-nine 500 ml. sample bottles  and
were also capable of continuous sampling if serviced during an event.   But
it should be pointed out that these samples were placed on small watersheds
for which the time of the event is usually short—so forty-nine samples,
in most cases, is more than adequate for the purpose.  These samplers were
also set originally to take samples every 15 minutes.  You are saying,
this is a lot of sampling, and you are right.  I would estimate that we
are approaching 100,000 water samples taken during the life of this project.

     It is most difficult to imagine the problems that occur when servicing
these monitoring stations during runoff events.  "Murphy's Law" definitely
takes over.  It seems as though most events occur at night and especially
on week-ends and holidays.  Long hours are involved and extra help is needed.
My wife has assisted me on many occasions, for which she is deserving of
much credit.  Also in thinking back on my experiences, I would have to say
that I could not recommend servicing field equipment at night alone—chances
for accidents are too great.

     The project proposal called for a sediment basin study for the improve-
ment of water quality—a site was selected.  With help from SCS personnel,
a survey was made (I wish to say that I always received the best cooperation
from SCS personnel throughout the project).  After the survey I designed,
drew up the plans, wrote the specifications, advertized for bids, selected
a contractor, and supervised  construction of the sediment basin.  Due to
the topography of the area, the basin had a very attactive, long, slender,
7 acre surface area.  Its watershed consisted of approximately 450 acres of
nearly level farm land.  For a period, I collected grab samples at the en-
trance and exit of the basin at each storm event and also on a weekly basis.
Flow volume was also recorded.  Two separate sediment deposit measurements
were made in the basin, the results of which were discussed in a Black Creek
report^.

     A desilting basin (an in stream basin) was also constructed.  The same
procedures were followed in setting  up the contract as with the sediment
basin.  The contractor did excellent work in constructing the basin—holding
the finished elevations to a very tight tolerance (we were fortunate in all
our construction projects to have had contractors who did the job in a pro-
fessional manner at very reasonalbe prices).  The basin was designed3,4,5
with two primary objectives:  (1) to trap particles of high specific gravity
and (2) to gain knowledge of bed load movement.  Some water samples were
taken during runoff events at the inlet and exit of the basin.  The basin
was periodically surveyed for sediment deposit.  A report on this investi-
gation has been published in a Black Creek report.6

     It also became my assignment to construct, evaluate, and report on two
mulch study areas located on the Black Creek drains.   Each of the areas was
constructed on ditch banks having 2:1, 3:1, and 4:1 slopes and each slope
contained five treated plots.  The mulch or treatment used was #4 crushed

-------
                              - 62 -

stone, straw, wood chips, Aquatain treatment, and a section with no mulch.
Previously, data regarding these studies was also published in another Black
Creek Report^.

     Each fall I recorded on an aerial photograph the existing ground
cover for the complete watershed.  Plus, an attempt was made to report
any events that might affect water quality.

     I have just skimmed the surface on reporting the experiences and prob-
lems involved with the Black Creek Project,  A more detailed report cover-
ing these past seven and one half years would be voluminous; however,
it is hoped some useful information may be gained from what is written.
                           REFERENCES

1.  Field Mannual for Research in Agricultural Hydrology Ag. Handbook No.
    224, Agricultural Research Services, U.S. Department of Agriculture.

2.  Sediment Reduction by Stream Bank Modification and Sediment Traps.
    Best Management Practices for Non-Point Source Pollution Control
    Seminar, R.Z. Wheaton and R.E. Land.  November 1976, pp. 155-163.

3.  Textbook of Water Supply,  Twort.

4.  Water Quality and Treatment, Am. Water Work Association.

5.  Settling Velocities of Gravel, Sand, and Silt Particles.  Paper by
    William W. Rubey, Published with the permission of the Director of  the
    U.S.  Geological Survey.

6.  Sediment Trap for Measuring Sediment Load, Non-Point Source Pollution
    Seminar.  Pp. 93-98, R.E. Land and R.Z. Wheaton.  November 20, 1975,
    Chicago, Illinois.

7.  Streambank Stabilization, Non-Point Source Pollution Seminar.  R.Z.
    Wheaton, pp. 86-92.  November 20, 1975, Chicago,  Illinois.

-------
                                  -  63 -
                          EVALUATION OE' SELECT BMPs

      E.J. Monke, L.F. Huggins, D.B. Beasley, D.W. Nelson, T.A. Dillaha,
                      S. Amin, M.A.  Purschwitz, R.E. Land
                                 INTRODUCTION

     Previous work in the Black Creek Watershed has  involved  continuous moni-
toring of  runoff,  sediment,  nutrient losses, and other  water quality parame-
ters at  the  outlet of  the  watershed and  several subwatershed  outlets.   The
primary  emphasis was  directed toward  evaluating  landowner  acceptance of  a
variety  of pollution control  practices and  their  overall  influence on  the
quality of water being discharged from Black Creek into the Maumee River.  The
influence of single practices  was  evaluated only on plot-sized  areas using  a
rainfall simulator.

     The  extensive collection of  environmental  data  on  the  Black  Creek
Watershed  was  useful  as  a  direct measure  of the  effectiveness of composite
management practices which were applied to the land  to  reduce nonpoint source
pollution.   In  so far  as the  Black Creek Watershed is  representative of  the
Maumee Basin, the  results are  useful for assessing  the  contribution of agri-
cultural  nonpoint  source pollutants   from  the  basin  into  Lake  Erie.   The
results were also  useful  for developing a simulation model called  ANSWERS by
which other  dissimilar  land  areas and storm events can be investigated.  This
model was  carefully designed to utilize  fundamental relationships  which  are
applicable to widely differing locations.

     The amount of usable field data on agricultural nonpoint source pollution
is quite limited.  Data concerning water quality particularly  at the outlets
to  areas  with  selected  best  management  practices (BMPs)  are  essential  to
assessing  the accuracy  with  which a watershed model can characterize the com-
plex interactions  of  various ground cover,  soils  and  farming  practices pre-
valent in  a  region.   Furthermore,  it is important that data be made available
concerning the response of small areas with  uniform  cover  and treatment prac-
tices  so  that  a  rational  selectivity  between  BMPs  is  possible.   Taken
together,  the monitored behavior  of small  sub-catchments  subject  to uniform
land treatments  will  provide a basis for building and refining models and, at
the same time, will provide  useful  benchmark data by which planners can rank
the relative effectiveness of different BMPs.
                                  BMP SITES

     Eight BMP sites ranging in size from 2 to 28 ha were established at vari-
ous  locations  in the  Black Creek Watershed  area as  shown in Figure  1.   In
addition, the  previously established tile drainage site  could also be con-
sidered  in a best management practice category.  The selected sites including
the drainage site were:

-------
                            - 64 -
                                                  61
                                                   t

                                              •AUTOMATED SAMPLER
                                                 & STAGE RECORDER

                                              • RAINGAGE
                                                               20
                                                MAUMEE RIVER
Figure  1.  Monitoring  locations on the  Black Creek Watershed.
          Stations 2,  6 and 12 are for the Smith-Fry Drain,
          the Dreisbach Drain and Black Creek, respectively.
          Seven BMP sites are located  on the watershed  as
          shown and two are located a  short distance away on
          similar soil types.

-------
                                  - 63 -
       Site No.    Farm Location                    BMP

          20      Becker          Subsurface drainage system
          51      Bennett         Conservation tillage using a no-till
                                  planter; grassed waterway outlet
          52      Schmucker       Pasture
          53      Gorrell         Parallel, tile-outlet (PTO) terraces
          54      Schaeffer       Conventional tillage; pipe
                                  spillway outlet
          55      Dean            Parallel, tile-outlet (PTO) terraces
          56      Armbruster      Conventional tillage with fall
                                  turn-plowing except for wheat;
                                  grassed waterway outlet
          57      Wolfe           Conventional tillage with mostly
                                  spring turn-plowing; grassed waterway
                                  outlet
          61      Hoeppner        Conservation tillage with special
                                  tillage tool; grassed waterway
                                  outlet
     An automatic sampler capable of collecting 48 discrete water  samples was
located at  the  outlet to each of  the  BMP sites.  This sampler  using  some of
the best features of the PS-69 sampler  was designed and built  in  the  Depart-
ment  of  Agricultural  Engineering  at  Purdue.    Except  for  the  subsurface
drainage site, all are powered using  a 12-volt heavy duty battery.  A simple
D.C. electrical  motor driven pump was used  to  both collect samples  and, by
reversing polarity, to purge the intake line.  The  water  stage over  control
sections was  also recorded  using  H-flumes, weirs  cut into sheet piling, or
existing overfall  structures except  for  the  PTO terrace  and   pipe  spillway
outlets (53,54,55).  At these latter sites, an elbow extension was attached to
the outlet pipe so that the pipe would always flow full.  The velocity of flow
was then to be  measured  using a magnetic flow meter.  However,  some reliabil-
ity problems are still being experienced with this particular instrument.
                                   RESULTS

     The  results  to date  are  still relatively  incomplete or  inconclusive.
Approximately 1200  samples  are still to be analyzed  from 1980.   However, the
primary difficulty is that  during  the spring  and early summer months of both
1979 and  1980 few runoff-producing rainfall events occurred.  Those which did
occur were also small events and somewhat erratic over  the experimental area.
Also since the events were small they were then subject to minor perturbations
which would become masked with larger events.   However, on  the basis of sedi-
ment  concentrations, one  PTO  terrace site  (No.  55)  produced  as  clean  or
cleaner runoff then  the  subsurface drainage system (No.  20).  The  field con-
taining the  PTO terrace system  was  also  chisel plowed and  at  the  particular
time samples were  being  collected  was planted  to a legume cover crop.   Both
systems delivered  an almost  inconsequential  sediment yield (on  the order  of
100 ppm).   The field containing  the  other  PTO terrace site  (No.  53)  was con-
ventionally tilled  and  planted  to corn.   As  a consequence of this and other

-------
                                   - 66 -
factors not readily apparent, sediment yields were considerably higher (on the
order of  5  to  10 times depending on  crop stage)  than the terrace system with
the  additional  BMPs.   One conservation  tillage system  (No.  51) appeared  to
produce sediment  concentrations about equal  to the  worst PTO  terrace  site.
However, this catchment was  rather  long  and, in addition  to  the conservation
tillage treatment, contained  a long, moderately sloping grassed waterway.  The
other conservation tillage system (No. 61), this one for a rather small,  slop-
ing catchment, gave around twice the sediment concentrations as for  the con-
servation tillage system on the more moderately sloping  and  longer  catchment.
The one  conventional  tillage  system  which was analyzed most  completely (No.
56) produced sediment concentrations much higher (on the order of three times)
than the  concentrations for  the worst-performing conservation tillage system.
This latter site  was  also located  on a   soil  combination of Hoytville  silty
clay and Nappannee silt loam  both of which contain slopes only between 0  and 1
percent.  The lower part  of  the catchment was planted  to soybeans  in  1980.
During the  spring months  in 1980,  this  particular catchment also  produced a
peak sediment concentration  of 10,000 ppm.   On the  other hand, the highest
sediment concentration measured  in  the principal outlet drains  (No. 2 and No.
6) was only around 1,500 ppm  for the same time period.
                                   SUMMARY

     The results from  the  BMP sites allow a  relative  ranking  of the  various
practices.   In  comparison  to conventional tillage all  BMPs  give considerable
benefits as  far as  sediment reduction is concerned.  From prior analyses,  we
can expect sediments in the Black Creek Watershed to account for approximately
90 percent of the total phosphorus loadings and 50 percent of the total nitro-
gen loadings.  The BMP sites are not replicate sites (replicate sites can only
be  approached  with  plots)  and  differences  in soils,  soil  configuration,
slopes, cropping sequence,  size, etc. are readily apparent.  Ultimately a nor-
malizing process through watershed simulation will have to be  accomplished  in
order  to separate  out the above variables from  what is  really  wanted—an
evaluation of BMPs themselves.

-------
                                  - 67 -
                              The ANSWERS Model
    D.B. Beasley, L.F. Huggins, E.J. Monke, T.A. Dillaha, III and S. Amin
     During the  second  year  of the project reported  herein,  several substan-
tial  changes  were made  to  the  ANSWERS  (Areal  Nonpoint  Source  Watershed
Environment Response Simulation) program, originally developed as  part  of the
Black Creek Project  (Lake  and Morrison, 1977).  Changes  included  the ability
to describe and  simulate  certain structural BMP's and  their  impact  on water
and sediment yield.

     It must  be noted  that  developmental work  on improving  the  utility  of
ANSWERS is still on-going.   In particular, developmental work is nearing com-
pletion on a  version  which  includes  direct  simulation  of  nutrient  losses
(phosphorus)   in  addition  to   the  sediment  yield discussed  herein.   Also,
improved sediment  detachment  and transport routines should give the user the
ability to model changes  in particle size distributions with changes in space
and time.
                               Acknowledgments

     The ANSWERS simulation model development was  financed  with  Federal funds
from the  U.S.   Environmental  Protection Agency under Sections•108a and 208 of
PL 92-500 and  by  the Purdue Agricultural Experiment Station.  The EPA grants
were administered by the Soil and Water Conservation District of Allen County,
Indiana and by the Indiana Heartland Coordinating  Commission in  Indianapolis.
Special recognition  is  due the U.S. Dept. of Agriculture—Science and Educa-
tion Administration, Agricultural  Research, for  technical  assistance  with
field  rainulator  experiments  conducted in cooperation with the  Department of
Agronomy, Purdue University.  These field tests, together with other plot data
and  professional  consultation made available  by  AR personnel,  provided  the
basic  information for development of the erosion  and  sediment  transport com-
ponents of the ANSWERS model.  Earlier research supported in part by the Dept.
of Interior, Office  of  Water  Resources Research provide a  foundation for  the
hydrologic portion of the model.

     Numerous graduate students of the Department  of Agricultural Engineering
and professional colleagues have contributed greatly to various components and
programming algorithms.   Individuals deserving  special mention  include:  J.R.
Burney, G.R. Foster, H.M. Galloway, J.V.  Mannering, T.D.  McCain,  and D.W. Nel-
son.
                             ANSWERS Improvements

     The  specific  component  relationships  used  in  the  ANSWERS  model  are
detailed  in Beasley, et  al.  (1980)  and Beasley  and  Huggins  (1980).  However,
since certain improvements have been made to the model during and as a part of
this project, they will be detailed below.

-------
                                   - 68 -
     Land  use  changes,  tillage techniques  and management  procedures  which
qualify as  Best  Management Practices (BMPs)  for controlling  non-point source
pollution are simulated with ANSWERS by using appropriate parameter values for
the component  relationships discussed above.  For example, conservation til-
lage generally results  in a rougher surface,  reduced C-factor and  increased
infiltration.  Gully stabilization  structures  such a drop spillways or chutes
may be  simulated  by reducing  the  slope steepness of the associated  channel
segments.   Certain structural  BMPs cannot  be  adequately accommodated  with
these component  relationships.   Currently, four specific BMPs which  require
special  computational   provision  have  been  included:  ponds, parallel  tile-
outlet terraces, grass waterways and field borders.

     Both ponds and PTOs are handled in a similar manner using a trap effi-
ciency concept.  Sediment trapped from the water flowing into  a pond or PTO is
diverted into a  special psuedo element which provides  a means of  tabulating
the combined effectiveness of  all  such BMPs.   Water  is  also assumed  to  be
diverted, in the  same  ratio as  sediment  is  trapped,  into the tile  drainage
system.  In this manner, effects of both reduced sediment loads and downstream
overland flow rates are simulated.

     Giass waterways and field border strips  are also treated  similarly to one
another.  It is assumed that the vegetated area within the affected element is
no longer subject to any sediment detachment.  Computationally, this is accom-
plished by adjusting the specified slope steepness  by an amount which produces
the desired  change  in  sediment detachment rate  for  the  element.   Deposition
within  the  vegetation of  a grass waterway  is  deliberately prohibited, since
any waterway that effectively traps sediment  would  soon  fill  and  become inef-
fective.  Specifying that an element has a grassed  waterway  forces the pres-
ence of a shadow channel element if none was  already present.


                            Using the ANSWERS Model

     The  following information  is  presented  in  an effort  to acquaint  the
potential  user with  input requirements and  output capabilities  of ANSWERS.
More detailed  information is in the ANSWERS User's  Manual (Beasley  and Hug-
gins, 1980).

     The data  file  used by the ANSWERS model  provides  a detailed description
of  the watershed  topography,  drainage networks, soils,  land uses,  and BMPs.
Most of the  information can be readily gleaned  from  USDA-SCS  Soil  Surveys and
land use  and cropping  surveys or summaries.   Also,  aerial photographs of the
area,  USGS  topographic  maps, and BMP construction or implementation data are
quite useful in developing descriptions of actual watershed areas.

     Input  information  for  the  ANSWERS model  contains six  general  types of
data:

     1)   Simulation requirements (measurement units and output control),

     2)   Rainfall  information  (times and  intensities),

-------
                                   - 69 -
     3)   Soils  information   (antecedent   moisture,   infiltration,  drainage
          response and potential erodibility),

     4)   Land use  and  surface information  (crop type,  surface  roughness and
          storage characteristics),

     5)   Channel descriptions  (width and roughness),

     6)   Individual  element   information   (location,  topography,  drainage,
          soils, land use and BMPs).

     The  individual  element information is the largest  body of data  and the
most time consuming to collect.  However, once the topography, soils, land use
and drainage patterns have been determined for all of the elements, changes in
watershed  management or BMPs  can  be  added  very easily  without  having  to
totally reconstruct the  input file.

     Figure 1 shows the configuration of a typical ANSWERS data file.  Each of
the six data  areas listed above are noted and will be covered individually in
succeeding sections.  The ANSWERS data  file was designed  to be self explana-
tory.  The  information  contained in the soils, land  use,  and individual ele-
ment  information sections  are  physically measurable  and  can be  checked for
validity without having  to go through a complicated process of differentiating
one or more lumped parameters.

     By using a very descriptive data file and  the  distributed parameter con-
cept,  the ANSWERS  model is capable of  producing  a detailed accounting of the
erosion and hydrologic  response of a watershed subjected  to  a precipitation
event.  The output listing consists of five basic sections:

     1)   An "echo" of  the  input data.(can be  suppressed  by removing "PRINT"
          parameter in line 2 of input data),

     2)   Watershed characteristics  (calculated from elemental data),

     3)   Blow and sediment information at the watershed outlet and effective-
          ness of structural BMPs,

     4)   Net transported sediment yield or deposition for each element,

     5)   Channel deposition.

     Several plotting programs  have  been constructed to use  the  input to and
the output  from the  ANSWERS model  to  provide visual  enhancement and better
understanding of the  information provided by  this  program.  The  QCKPLT pro-
gram, mentioned  in  a previous section,  uses the elemental data portion of the
input file to produce the map shown in Figure 2.  The arrows indicate the flow
direction  for each  element.   The  shaded areas  indicate  "dual"  elements  or
overland elements with  companion channel elements.  This map can  be produced
so that it will fit any scale.  This feature allows one to check input data on
maps with different scales by producing  overlays at the specific scele needed.
The grid  is also very useful in physically  locating  the predicted "hot spot"
areas.

-------
                -70 -
STANDARD PREDATA FILE FOR ALLEN CO., INDIANA— -800823
ENGLISH UNITS ARE USED ON INPUT/OUTPUT
PRINT
1

RAINFALL DATA FOR 2 GAUGECS) FOR EUENT OF: TEST
GAGE NUMBER Rl
0 0. 0.00
0 9. .52
0 15. 1.55
0 20. 2.40
0 30. 1.59
0 35. .85
0 45. .50
1 300. 0.00 9
GAGE NUMBER R2 «•
0 0. 0.00
0 7. .45
0 14. 1.25
0 18. 2.66
0 25. 1.G5
0 33. .GO
0 42. .35
1 300. 0.00
SOIL INFILTRATION, DRAINAGE AND GROUNDWATER CONSTANTS
NUMBER OF SOILS = 8
S 1, TP =.46, FP =.75, FC = .40, A = .80, P =.65, DF
S 2, TP =.46. FP =.65, FC = .40, A = .80, P =.65, DF
S 3, TP =.46, FP =.70. FC = .40, A = .80, P =.75, DF
5 4, TP =.42, FP =.70, FC = .60, A = 1.0, P =.65, DF
5 5, TP =.44, FP =.30, FC -- .60. A = 1.0, P =.65, DF
5 S, TP =.46, FP =.75, FC = .60, A = 1.0, P =.65, DF
S 7, TP =.46, FP =.75, FC = .40, A = .80, P =.65, DF
S 8, TP =.35, FP =.65, FC = .90, A = 1.6. P =.60, DF
DRAINAGE COEFFICIENT FOR TILE DRAINS =0.25 IN/24HR
GROUNDUATER RELEASE FRACTION = .005
SURFACE ROUGHNESS AND CROP CONSTANTS FOLLOW
NUMBER OF CROPS AND SURFACES = 6
C 1, CROP= SI CORN, PIT=.01, PER=0.0, RC=.47, HU= 2.0,
C 2, CROP= CORN-NT, PIT=.06, PER=.75, RC=.55, HU= 2.5.
C 3, CROP=BEANS TP, PIT=.01, PER=.45, RC=.47, HU= 1.5.
C 4, CROP=S. GRAINS, PIT=.04, PER=.90, RC=.55» HU= 2.0,
C 5, CROP= PASTURE, PIT=.03, PER=1.0, RC=.40, HU= 1.5,
C 6, CROP= WOODS , PIT=.10, PER=.90, RC=.5S, HU= 3.5,
CHANNEL SPECIFICATIONS FOLLOW
NUMBER OF TYPES OF CHANNELS = 4,
CHANNEL 1 WIDTH =15.0 FT, ROUGHNESS COEFF.(N) = .035
CHANNEL 2 WIDTH =10.0 FT, ROUGHNESS COEFF.(N) = .040
CHANNEL 3 WIDTH =7.0 FT, ROUGHNESS COEFF.(N) = .045
CHANNEL 4 WIDTH = 4.0 FT, ROUGHNESS COEFF.(N) = .050
ELEMENT SPECIFICATIONS FOR MIDDLETOWN WATERSHED
EACH ELEMENT IS 528. OFT SQUARE
OUTFLOW FROM ROW 21 COLUMN 3
1 15 4 270 1 1 Rl 0
1 16 5 259 3 1 Rl 0
• • *»•• • » •
7 20 9 186 2 1 Rl 6 2
7 22 2 135 1 2 Rl 0
7 23 3 180 3 5 Rl TILE 0
8 11 20 270 201 2 R2 TILE 5 4 32
8 12 17 180 302 2 R2 TILE 5 4 20
8 13 16 180 303 1 KB TILE 5 4 20
8 14 12 148 1 1 R2 TILE 0 1
• i •*•« • • - •
20 19 8 ISO 1 2 R2 TILE 6 3 20
20 20 8 380 1 3 R2 TILE 0
21 3 9 270 102 1 R2 TILE 3 4 40
21 4 28 180 108 1 R2 TILE 3 4 40
• * ••«• • •*
22 13 2 0 3 4 R2 "6
22 14 3 5 90 i 4 R2 0
FOLLOW
= 4.0,
= 3.0,
= 3.0,
= 4.0,
= 5.0,
= 5.0,
= 3.0,
= 6.0,
N=.075,
N=.120.
N=.OSO,
N=.120,
N=.150.
N=.200,
5

ASM =.70, K =.35
ASM =.70, K =.32
ASM =.70, K =.17
ASM =.70, K =.36
ASM =.70, K =.32
ASM =.70, K =.36
ASM =.70, K =.38
ASM =.70, K =.35
C=.SO
C=.30
C=.60
C=.15
C=.04
C=.15

827.8
828.8
•
822.8
835.3
835.3
799.7
799.3
802.2
809.2
•
802.6
803.0
766. 5
762.0
•
799.5
798.0
4
6
Figure 1.   Typical ANSWERS input  data  file

-------
                                     10  11  12  15  14  15  16  17  18  IS 20 21 22 25 24
          Figure 2.   Example elemental map produced by QCKPLT program
     Figure  3  shows the graphic  output from  the HYPLT program.   HYPLT uses
standard  CALCOMP-compatible  calls  and plots the  rainfall  hyetograph,  the
runoff hydrograph and the sediment concentration  curve.   The program directly
uses the hydrograph portion of the output listing.

     The information presented  in Figure 4 comes directly  from the net tran-
sported sediment yield  or  deposition section  of  the  ANSWERS output.  Several
programming steps are required to "reconstruct" an elemental  data  format (row
and column coordinates)  and input the individual element information to a pro-
gram called  CONTUR.  CONTUR  allows  the user  to set  the  levels of sediment
yield or deposition at which contours are desired.  The program produces a map
at any desired scale (up to the maximum size the plotter allows).  The shading
in Figure  4  is for additional  visual effect.  The portions  of the watershed
with closely spaced contours  show areas with  excessive  transported soil loss

-------
                                   - 72 -
   .000 -i
  2.000-
  1.000-
  6.000 J
         2. TOO
                                                BRUNSON DITCH
                                             8  YR. - 1.5 HR. STORM
                                                 Sediment Concentration
          .000
              .0     40.0
                           80.0
               ~i	r
120.0    160.0     200.0    2tO.O
      TIME - MINUTES
                                                                        -I- 18.00
                                                                         - 15.00
                                                                         - 12.00
                                                                              0-
                                                                              0_
                                                                         -  9.00
                                                                              CD
                                                                              CJ
                                                                         1-  6.00 ,
                                                                              UJ
                                                                              CO
                                                                         -  3.00
                                                                            .00
                                                                880.0
                                                                       320.0
              Figure  3.   Example output of HYPLT plotting program
or deposition.   In  general, high  transported soil  loss areas  will be  found
near high deposition  areas.  Ibis is due  to the fact  that  steep slopes blend
into flat slopes near the channels.
                     Planning and Evaluation Case Studies

     The following case studies are examples of  the application of ANSWERS  in
two different  areas  - planning and evaluation.  Continuing  development of the
model should allow even more descriptive and useful outputs  in the future.

     The relationship between monitored  information and simulation results  is
very important.   In  earlier sections, it has been  pointed out that monitoring
studies cannot provide the detailed  information necessary to  determine cause-
effect relationships.   Also,  the use of unvalidated models  or models with too
many simplifying assumptions or undescriptive  relationships  leads to  the same
result —  a lack of complete understanding of  the causes and  effects.   This
section will detail  the use of the  ANSWERS model  coupled with data  gathered
from an extensive monitoring system  for  use  in both the  planning and evalua-
tion roles.

     The planning example utilizes the model on  an  ungaged watershed  in Allen
County,  Indiana.  The  topography, soils,  land uses,  management systems and
precipitation  inputs are very similar to monitored  information  from  the Black

-------
                                  - 73 -
                                                                EROSION
      EROSION
                                                                DEPOSITION
                Figure 4.  Example output  from CONTUR program
       (Closely spaced  contours  indicate high deposition/erosion areas)
Creek Study area,  which is about 35 kilometers away.  The a priori ability of
ANSWERS to describe the responses of the various  hydrologic  and  erosion com-
ponents is used to produce water, sediment and nutrient yield  predictions for
a variety of hypothetical management strategies.

     The evaluation example gives some insight  into  using  the ANSWERS  model
for the purpose of interpreting monitored data.  Two different comparisons are
made concurrently: (1)  the improvement in water quality from 1975 to 1978 and
(2) the  improvement  in water  quality  from  the west side of the  watershed to
the east side.   All of  the simulations used monitored data to verify and  give
credence to the results.

-------
                                  -  74 -
Planning Example

     As an example of using ANSWERS as a tool for planning BMP systems, let us
consider  an  actual   situation,   the  Marie  Delarme  watershed  located  in
northeastern Indiana.  This watershed is composed of almost  500  ha of predom-
inately  (60  percent)  poorly  drained Blount, Crosby and  Hoytville silty clay
loams, with  the  remainder being moderately permeable Haskins  and Rensselaer
silt loams.  Element  slopes range from 1  to 6  percent and have an average of
1.9 percent.  Because  of the moderate relief,  an element size of  2.6  ha was
chosen as adequate for modeling purposes.  The resulting watershed representa-
tion is shown in Figure 5.
                                    MARIE DELARME WATERSHED
                                            0 1/4 Mil. 1/2 Mil.
                             PTO Terrace £rea

                             Chisel Plow Area
                 Figure 5.   Elemental watershed representation
     In  order to  rank  the effectiveness  of alternative  control  strategies,
some  frame  of  reference  or "baseline  condition"  was required.   To  remove
effects of particular land use and management practices from the  baseline con-
dition,  all  tillable land (in this case,  the entire watershed) was assumed  to
be planted to conventionally tilled corn.

     In  addition  to choosing a land  use pattern,  a time frame  must also  be
selected.   Average annual conditions are  usually used.  Since ANSWERS  is  an
event-based  model,  simulations are performed on a storm-by-storm  basis.   While
it  is  certainly possible  to  simulate all  the  storms  of a  "typical"  year and
then sum the results, this is not necessary in  most situations.   Many research
results have shown  that most of the sediment and associated  chemicals are pro-
duced by the largest one or two storms for the  year.  Monitored data  from the
Black  Creek watershed  in northeastern  Indiana  also  indicate  that  average
annual yields can be approximated, for that region, by simulating a single 1.5
hour  duration, B-distribution storm which has a  recurrence interval  of  8
years.   This hypothetical storm was  assumed to occur  approximately one  month
after  planting, with antecedent soil moisture  at  field capacity.   Simulating
these  conditions  will  approximate  average annual yields  for  sediment and

-------
                                   -  75 -
sediment-related  nutrients.  However,  soluble chemical  transport is underes-
timated.   Because this single  storm yields  only  10  percent of  the expected
annual  water  yield,   annual   soluble  nitrogen  yields  were not  accurately
predicted  by the  single storm simplification.

     Having decided upon a set of baseline conditions,  the next step  was  to
simulate the response  of  the Marie Delarme watershed to those conditions.  The
result of  that effort  is  shown  in Figure 6.  It shows  the distribution of net
sediment transported from each  element for the baseline condition.  Note that
the values, ranging  from  a  loss in excess of 5000 kg/ha  to deposition of more
than 1000  kg/ha,  represent  net  transport from each 2.6 ha element.  Local ero-
sion rates would  generally  be much higher that the  rates of transported sedi-
ment.  These net  transport  rates are predicted by ANSWERS without resorting  to
a delivery ratio concept which is very difficult to  quantify.   Instead, the
specified  watershed topography,  land use and surface  runoff rates determined
transport  capacity within the watershed.
                                        MARIE DELARME WATERSHED
                                          ALLEN COUNTY, INDIANA
                              LFOEND

                            YIELD IN EXCESS OF , TON ACHfc

                            YIELD IN fcXCESS OF 1 TON/ACHt

                            AREAS INSIDE OASHE.D LINES SHOW

                            DEPOSITION OF SEDIMENT IN
                            EXCESS OF '/, TON/ACHE
                         figure 6.  Net sediment yield
     Figure  6 gives  information important  for  devising  alternative control
strategies.   It shows that  the  highest sediment and associated nutrient yields
occur  in  the upper third of  the watershed and  gradually decrease  toward  the
outlet.  While anyone knowledgeable  about  soil erosion  and  familiar with  the
watershed  could  have  predicted  this  general  trend  without  a  simulation
analysis,  a  distributed  model   is  required to  quantify actual  yields in  the
manner shown.  More importantly, as the following  results  demonstrate, such a
model can predict relative  impacts  of alternative control strategies.

     Figures 7a through 7d  depict four alternative control strategies.  While
many other strategies,  possibly even more  effective  than  those chosen, could
have been  selected,  they illustrate  the scope  of  information  made available
and the manner in which simulation  can be used as an effective planning tool.

-------
                                            - 76  -
STRATEGY #2
                MARIE DELA3ME WATERSHED
                     Al I,EM COUNTY, INDIANA
                           O  1X4 Mite tJZ Mil*
                                                             STRATEGY
      PTO Terrace Area





      Ght'.ol Plow /" i on
                                                                             MARIE DELARME WATERSHED
                                                                                  ALLEN COUNTY, INDIANA
PTO Terrace Area





Chisel Plow Aroa
                   (A)
                                                                                (B)
 STRATEGY *«
                 MARIE BELARME WATERSHED
                      ALLEN COtlNTY, INDIANA
                            0  1/4 Mlla 1/2 Mil*
                                                            STRATEGIES *5S6
                                                                              MARIE DELARME WATERSHED
   •  PTO Terrace Area




   Q  Chisel Plow Area
 PTO Terraco Area




 Chisol Plow Aroa
                    (C)
                                                                                  (D)
           Figure   7.    Locations  of  BMPs  for Alternative Strategies.

-------
                                  -  77-
     Table 1 summarizes simulation results from all strategies considered.   It
illustrates  the complicated  nature  of ranking  alternative programs  for  NFS
pollution control.  The  position of  a particular  strategy is very  dependent
upon the  ranking  criteria used.   B'or example, Strategies 2-5 have been listed
in terms of decreasing effectiveness for reducing sediment yield at the outlet
of the  watershed.  However,  the ranking would  be quite  different  if annual
unit cost of achieving a  sediment  yield reduction  is employed.  Still  dif-
ferent results would be obtained if nutrient yields or concentration levels in
the stream  are  chosen.  All of  these water  quality improvement  criteria  and
others are valid for developing a control program.  Generally, several of them
would be given some consideration.  It is the ability of a comprehensive simu-
lation model to provide information on such a wide range of factors that makes
it such an attractive and even essential tool for planning  NFS  pollution con-
trol.

     The ranking  of strategies is also influenced by  the choice  of  baseline
conditions,  as  illustrated  in  Table  1  by Strategy  6.   The only difference
between results for Strategies 5 and 6  is the  severity of  the  hypothetical
storm used  to simulate the baseline condition.  For Strategy 6,  a storm with
25 percent lower intensities and total volume was used.  This gave a  sediment
yield of  640 kg/ha for the same land  use as in Strategy  1.  When simulation
results from Strategy  5 are compared  to  that baseline instead of Strategy 1,
they  show  lower   absolute reductions  (110  kg/ha  vs.   170  kg/ha) ,  but  an
increased percentage reduction.  The unit cost of reducing  sediment yield  was
also higher  when  the  less intense baseline storm was used.  This again illus-
trates the complexity of analyzing NFS pollution and its control.

Evaluation Example

     As an example of the use of ANSWERS for  interpreting  monitored data,  let
us consider  its use  in the Black Creek Project.  The 4860 ha study watershed,
located in northeastern Indiana, is composed of relatively heavy soils associ-
ated  with glacial  till  and  an  old  glacial  lake.   The  land  use is almost
entirely agricultural except for a small community of about 500 persons.   Its
soils and  land  use  distribution are  representative of  those  in the Maumee
Basin.

     The Allen County Soil and Water Conservation District, with the  technical
assistance of the  Soil Conservation  Service, developed a cost-sharing program
to encourage installation of  appropriate BMPs.   Purdue  University   and  the
University  of  Illinois were responsible for  monitoring  the physical/chemical
and biological  water  quality  impacts  of installed  BMPs.  Figure 8  depicts
locations at which physical/chemical  monitoring has been conducted for periods
ranging from 3 to  6 years.  An even more extensive network of  biological moni-
toring locations was  established.

     The Black Creek  project has utilized automated, continuous monitoring of
stream water  conditions near  the  outlet and at selected  subwatershed points
for about 5  years.  This comprehensive data base  has  established a  reliable
indication of water quality conditions  within the watershed.  However,  it is
impossible from  such data to answer  the question:  "What have been  the  benefits
from  individual  classes  of  BMPs  installed   during  the  project?" This is  a
result of  the  many  uncontrolled  factors   which  influence   levels   of  NFS

-------
                         Table  1.  Simulation Results for Alternative Strategies
Area Affected
by BMPs
Strategy* PTO Chisel


1
2
3
4
5
6

(ha)
0
285
85
91
0
0

(ha)
0
104
213
272
321
321
Sediment

(kg/ha)
1530
650
1080
1220
1340
520
Total Yield at Watershed Outlet
Total Avail. Sed. Sol.
P P N N

(kg/ha)
2.1
.8
1.5
1.6
1.8
.6

(kg/ha)
.6
.2
.3
.4
.4
.1

(kg/ha)
13
6
10
11
12
4

(kg/ha)
1.0
.6
.8
.8
.9
.4
Sed iment
Reduction Cost**

(%)

57
29
20
13
19
($/tonne
reduced)

34.70
22.00
35.30
7.50
11.60
 *1.  Baseline condition: Fall moldboard plowed,  no BMPs.
  2.  PTO terraces installed where most of the sediment yield was in excess of 2.25  tonne/hectare.   In
     addition, those areas  with  sediment yield  in excess of  1.12  tonne/hectare not benefited by ter-
     races were chisel plowed.
  3.  PTO terraces installed in the upper 1/3 of  the watershed only.  All areas with sediment  yield in
     excess of 1.12 tonne/hectare not benefited  by terraces  were chisel plowed.
  4.  PTO terraces installed in the lower 1/3 of  the watershed only.  All areas with sediment  yield in
     excess of 1.12 tonne/hectare not benefited  by terraces  were chisel plowed.
  5.  All areas with sediment yield greater than  1.12 tonne/hectare  chisel plowed.
  6.  Same as Strategy 5 except that a storm with 25%  lower  intensity and total  volume was  used.   The
     "baseline condition" for this storm gave a  total sediment yield of 640  kg/ha.

* * Cost information was  based  on 1979  construction costs for  PTO terrace systems in Allen County,
  Indiana.  The  cost  is based on total area benefited  (both above and below terraces).   The figure
  used in these  calculations  was $510.80 per hectare benefited.   A 10-year life  was assumed, which
  yielded an annual cost of $51.08 per hectare benefited. The  chisel plow  was  also assumed to have a
  10-year life.  The average  annual  cost per hectare, based  on the cost of  a  new plow, was  $2.17.
  Since the "design storm"  used  in this example produced approximately the  annual  sediment  yield,  the
  cost per tonne of reduced yield at the watershed outlet is,  essentially, the annual cost.   However,
  due to  simplifying  assumptions and unique local conditions,  these  cost figures  should not be con-
  sidered to be generally applicable to other planning situations.   They were  included in  an effort to
  give the reader a feeling for the type of analysis which can be performed  by ANSWERS.
(X>
I

-------
                                   - 79 -
     Figure  8.   Black Creek monitoring locations and primary subwatersheds
pollution and the diversity of BMPs, crops and annual management changes which
occur on a  watershed  scale.  Such a question can be answered using simulation
analysis because it is possible to hypothetically hold all factors, especially
hydrologic conditions, constant and change only the applied BMPs.

     Figure 8  shows the  Black  Creek watershed  subdivided  into three  major
subwatersheds of  approximately equal size.   A deliberate effort was made  to
encourage the  installation of  BMPs within  the  western subwatershed.   This
decision was made to more clearly differentiate BMP impacts and to  demonstrate
the magnitude of water quality change which was feasible.  As a result,  a pat-
tern of decreasing  practice density occurs as one goes eastward.  The  initial
development of the ANSWERS model was undertaken  as a part of  the  Black Creek
Project as  it  became apparent that  monitored data  alone  was  inadequate  to
quantify impacts of the combinations of  installed BMPs.

-------
                                  - 80 -
     The  BMPs  installed  within  Black  Creek were  primarily  structural  in
nature: parallel  tile-outlet terraces  (PTO),  field borders, grass waterways
and livestock exclusion.  Despite  the demonstrated water quality benefits  of
reduced tillage systems, only limited utilization was achieved.  This was the
result of farmer concern about wetness during the spring on the heavy soils of
the area.

     The area of land directly  affected  by installed BMPs ranged from 6 per-
cent  in  the western  subwatershed  to  less  than  2  percent  in the  eastern
subwatershed.  ANSWERS simulations, using patterns of land use change and BMPs
installed between  1975  and  1978 with  assumed  constant  hydrologic conditions,
indicated that  EMPs installed  within the western  subwatershed would reduce
annual sediment yield about  30 percent.  The reduction for the medium density
middle subwatershed was  about 20 percent, dropping to  only about  10  percent
for the minimum  treatment  level on the eastern subwatershed.  Watershed scale
impacts from a  single  installation are also available,  but such results are
extremely location dependent.

     One  additional result  of  general  interest  was  determined.   When  the
installed  structural  BMPs  for  the  western subwatershed were  hypothetically
augmented with chisel plowing in selected critical areas, the projected reduc-
tion of annual sediment yield was increased to 50 percent.
                                  REFERENCES

 1.  Beasley, D.B., L.P. Huggins and E.J. Monke.  1980.   ANSWERS:  A model for
     watershed planning.  Transactions of the ASAE  (23)4:938-944.

 2.  Beasley, D.B.  and L.F.  Huggins.   1980.   ANSWERS user's  manual.  Agric.
     Eng. Dept.  Purdue Univ., 55p.

 3.  Lake, J. and  J.  Morrison.   1977.   Environmental  impact  of  land  use on
     water quality: final report of the Black Creek project—technical report.
     U.S.  Envir.  Prot. Agency,  Region V,  Chicago,  IL.   EPA-905/9-77-007-B.
     274 p.

-------
                                  - 81 -

                            TILE  DRAINAGE  STUDIES

             E.J. Monke,  A.B.  Bottcher,  E.R.  Miller,  L.F.  Huggins,
                     D.B. Beasley, D.W. Nelson, R.E. Land

                                   ABSTRACT

     In the  first  study, sediment pesticide  (1977 only)   and  nutrient losses
were measured  from a  17 ha subsurface drainage system for the years 1976-79
using  automatic  sampling  equipment.   The  monitored  drainage  system  was
installed  in the  early 1950's on a nearly flat Hoytville  silty clay with lim-
ited surface runoff due  to  raised field  borders.  Dynamic responses  of the
drainage  system are graphically  presented and discussed as  they  related to
field management practices and climatic variations.  Sediment yields were gen-
erally low averaging  81 Kg/ha for the four years  of record.  A comparison was
also made  between  this system with its  low  surface runoff and a  more normal
situation  with  much greater surface runoff.  For  the 17 ha system, runoff per
unit area was substantially lower resulting also in less sediment and nutrient
losses.

     In the  second  study, the  sediment movement  in the  backfill  profile of
Hoytville  silty clay was compared to  that  for two other soils  found in the
Maunee Basin.   One  soil was Latty clay  which is  a true  lacustrine  soil  from
the  center  portion of  the  basin.   It is  marginally suited  for subsurface
drainage.  The  other  soil was Blount  silt loam,  a  glacial till  soil,  found
around the periphery  of the basin.  It  is mostly drained by random tile sys-
tems since the topography is gently rolling.  The  study was conducted using a
laboratory set-up to measure the sediment yield from vertically downward move-
ment of water  through the backfill profiles.   The average sediment  loss for
Latty clay was about  20 percent  greater  and  that  for the Blount  silt loam
about 75  percent  less than for Hoytville  silty clay.   If  the  sediment yield
for  Hoytville  silty clay in  the  first study is  representative  of the Maumee
Basin, the yields  and associated  chemical transport for  the other  two  soils
are also  likely to  be low.  As another part of the second study, none of five
envelope materials  had  any effect on  sediment  movement  from  Hoytville silty
clay backfill profiles.


                                 INTRODUCTION

     The major  cause  of accelerated eutrophication  in many of our  lakes and
streams has  been identified as nutrient enrichment, particularly phosphorus.
Nutrient enrichment in many areas has been attributed  to  runoff from agricul-
tural lands which is related to fertilizer usage,  cropping practices and water
management.  A  large  portion  of the nutrients  are associated  with  sediments
being transported  from fields.   Therefore, an  obvious  abatement  procedure is
the use of practices to reduce soil erosion which  generally has  the  effect of
increasing subsurface  drainage.   However, before any such  practice can be
identified for  nonpoint source pollution control,  we need to  know how a  par-
ticular practice  affects the type,  form  or  amount of nutrients  being tran-
sported and  how water  yields  and  concentration differences  impact  on loading
rates.

-------
                                  -  82  -


     With respect  to  subsurface drainage,  Baiter  and Johnson  (1976)  indicate
that  nitrate  concentrations  are higher  in subsurface  waters  than surface
waters.  Also,  Schwab,  Nolte,  and  Brehm  (1977)  found  significant  sediment
losses from  tile  drains.  Nutrient data collected during the Black Creek pro-
ject  (Lake,  1977)  showed  that in  all  cases  the average  soluble  forms  of
nitrogen and phosphorus were higher in subsurface drainage waters than in sur-
face  waters.   The high  nutrient concentrations  likely to be  found  in tile
drainage waters and the extensive and rapidly increasing acreage of subsurface
drainage reinforces the need to better understand  the  transport and transfor-
mation  processes  involved  with  this  practice.   In  addition,  subsurface
drainage may have to be associated with water quality or erosion control prac-
tices in order to maintain productivity.

     In the  past,  most methods of reducing  sediment  movement have  been con-
cerned with  sand and  coarse silt  in  preventing clogging  of drainage lines.
Installation practices were developed to lengthen and  improve the operational
life of subsurface drainage systems.  In recent  years, however,  the emphasis
on reducing surface sediment movement has been extended to include the overall
subsurface contribution  of sediments  and nutrients to our  rivers and lakes.
This additional emphasis on improving water quality has  shown the  need for
better practices  of  stabilizing  silts and  clays near drain  lines since the
small soil particles have large surface areas which can  adsorb and transport
soil nutrients  into  the drains.  The stabilization of heavy soils surrounding
drainage lines  involves  recognizing some of  the possible  causes for  their
movement.

     Schwab  (1975),  working with heavy  soils  in the Maumee  Basin,  proposed
that the main mechanism of sediment movement is due to suspended particles in
the soil water  that  move through the  soil  profile  or the  backfill  material
into subsurface drains.   His results also showed that the sediment concentra-
tion increases significantly with antecedent moisture content of the soil pro-
file.  Monke  and Beasley (1975)  also noted turbid discharge  from some tile
outlets during  spring  thaws when the seepage water  in the  largely saturated
profile was quickly released to the tile drains.

     A mechanical analysis of the sediment losses during a field evaluation of
drain  envelopes  by  Taylor  and  Coins  (1976)   indicated  that  the  sediment
reassembled the A horizon more than any other part of the profile.  Its compo-
sition was  probably  developed by  sorting  during  transport in  the  drainage
channels of the horizons above the drain.  They also noted  extensive channel-
ing  (possibly  caused  by incomplete  settling of  the  backfill   rather  than
shrinkage cracks)  in the  soil  adjacent to the tile lines  installed in Crosby
silt loam.

     In both cohesive and non-cohesive soils, the mechanism of piping has also
been a  significant contributor  to  sediment movement  into  subsurface drains.
Zaslovsky and  Kassiff  (1965)  defined  piping as  the condition  in which  the
forces of drag  and gravity on soil  particles overcame their maximum resisting
force of cohesion.   Such an unstable hydraulic condition occurs  at locations
of excessive hydraulic  gradients  resulting from the effects of convergence at
drain pipe perforations.   The limiting  form of  soil  bridging from  edge and
convergence effects tends toward  a  hemisphere which allows the gradients over
the surface of the hemisphere to become lower and more uniform.

-------
                                  - 83 -


     Walker  (1978)  noted that the  hemisphere formed  in  well-graded cohesive
soil by bridging collapsed when the  exit  gradient for the flow rate exceeded
the characteristic critical failure gradient of the soil.  WalKer demonstrated
experimentally  that  the critical failure gradient was a  function of soil pro-
perties and  not that of the  restraining envelope  material.   Thus, for filter
and  envelope materials, criteria  which reduce  excessive exit gradients from
convergence  or  the hydraulic  gradients at  the soil-envelope  interface to lev-
els  that prevent  initial  soil  movement  is very appropriate in determining
their application.
                 I.   MOVEMENT OF NUTRIENTS AND SEDIMENT FROM A
                          SUBSURFACE DRAINAGE SYSTEM
                                   OBJECTIVE

     The objective  of this study was  to  determine the  overall  water quality
impact  associated with the waters discharged during the period 1976-79 from a
subsurface drainage system on a field with minimum surface runoff.
                               SITE DESCRIPTION

     The monitored tile system drained a 17.4 ha field located two miles south
of Woodburn,  Indiana.  The  field was very flat  (<  1%  slope)  and had raised
field borders.  The field borders were the  result of ditching around the field
and effectively prevent most surface runoff.  Ninety-five percent of the field
consisted of a Hoytville silty clay  (fine,  illitic,  mesic Mollic Ochraqualfs)
with the remaining area being a Nappanee silt loam (fine, illitic, mesic Alric
Ochraqualfs).

     The drainage system was installed in  the early 1950's at  a depth of one
to two meters.   The  drain lines were clay drains with topsoil blinding except
under observed wet spots where stone envelopes were placed.  The main line was
thirty-centimeters in  diameter.   A layout of the  drainage  system, as obtained
from construction records Kept by  the  owner is shown in  Figure  1.   The crop-
ping and fertilizer history for the field is given in Table 1.
                               DATA COLLECTION

     The subsurface drainage  system  outlet was monitored  for flow  and water
quality parameters using  automatic  sampling and recording equipment  (see Fig-
ure 2).  The flow was determined by recording the depth of water behind a weir
crest with a  bubble-tube  stage recorder.  Free fall was assured over the weir
by pumping the discharge into the outlet ditch.

-------
o
a>
•^

O
3
O
fi>
a
           Tile Monitoring
                                          Completely Tiled

                                            20m Spacing
                                          Completely
                                            Tiled
Farm

Build-

 ings
                                                                       20m Spacing
            Completely Tiled

              20m Spacing
 If
                                                                    i

                                                                    00
Miller Ditch
                                                              State Route 101
Figure 1.   Subsurface Drainage System Layout  (Hatched Area Rspresents Portion of Field Which Drains

          to the JVbnitored Outlet)

-------
                              -85-
Table 1.  Field Cropping and Fertilization for Subsurface Drainage Field
Year
1971
1972

1973
1974



1975



1976


1977


1978
1979
1980
Crop
Oats
Corn

Soybeans
Corn

Wheat

Beans
Corn


Wheat

Soybeans
Soybeans
Corn


Soybeans
Wneat
Wheat
(Clover)
Corn
Area
(hectare)
17.4
17.4

17.4
8.7

8.7

8.7
8.7


8.7

8.7
17.4
17.4


17.4
17.4
17.4
17.4
Fertilization
(kg/ha) (Type)
150
50
300
60
100
100
300
100
300
100
100
100
300
300
50
100
400
180
170
10
200
120
300
175
400
250
240
16-16-16
N-45 urea
0-26-26 fall plow-down
10-34-0 starter
10-34-0 starter
N-28 solution
0-26-26 fall plow-down
N-28 solution
5-24-24 fall plow-down
10-34-0 starter
10-34-0 starter
N-45 urea
0-26-26 fall plow-down
16-16-16
N-45 urea
10-34-0 starter
0-25-26 fall plow-down
10-34-0 starter
N-45 Urea
Manganese
0-0-60 fall plow-down
10-34-0 starter
0-26-26 fall plow-down
N-56 urea (56%)
0-9-48 fall plow-down
82-0-0 (11/14 with N-Serve)
10-34-0 starter

-------
                  - 86  -
                                             PUMP SAMPLER





                                             FLOW INTEGRATOR
v?/:-3H	     ^*»» e>    r Ul
    Figure  2.   Tile  Flow Sampling Station

-------
                                  - 87 -

     Discrete  (500 ml) water samples were collected  at  a rate proportional to
the tile outflow.  A flow integrater was constructed to vary the sampling  rate
with flow.   The sampler (designed by  authors)  had  a  capacity of  72 samples
before  servicing  was  required.  The sampling rate  varied from one sample for
each forty minutes at maximum  flow to one sample for each twelve  hours at low
flow  rates.   Samples  were  frozen within  24 hours of  collection  for later
laboratory  analysis.   Standard  laboratory  procedures  were  used  and  are
described in the  final report  of the Black CreeK Project  (Lake, 1977).


                            RESULTS AND DISCUSSION

     Selected  flow  periods  during 1977 were  chosen for  detailed  analyses of
the flow  rate, rainfall and various nutrient and  sediment concentrations and
yields.  However, data for the remaining years of record  are also available in
the  same  detail.  Summaries  of  all  the  data  from  1976  through  1979  are
presented.

Water Yield

     The  flow  response  of the subsurface drainage  system for the  months of
February, March,  and April 1977 is shown in Figure 3.  This is the period  when
the soil  profile  is normally  the wettest resulting in the highest  tile  out-
flow.   Later  in the spring and on into the summer and early fall, little  tile
flow occurs as  the result of high evapotranspiration rates which increases the
available water storage in the profile and lowers the water table.  In several
cases, two or  four  centimeter  rainfall events during  this period resulted in
no tile flow at all.   In general, tile flow in the Midwest can be considered
mostly a winter and early spring phenomenon.

     The freezing and thawing  conditions during this high flow period may  com-
plicate the mechanics of water flow.  Flows can go from zero to near full  pipe
flow within a few hours during an initial flush in  early spring.  The magni-
tude of the initial  flush  depends on  the depth  of the  freeze  line and the
amount of water stored  in and above the  soil profile.  Then when  the freeze
line  "breaks",  which it  does uniformly,  large  amounts of  free  water  are
released to  the tile  drains.  Furthermore,  temporary  freezing of  the ground
surface may reduce  the tile  flow rate  by  interfering  with  the groundwater
pressures.   Since temporary freezing  likely occurs during  the night,   flow
fluctuations show up  as a diurnal cycle.  An example of  this phenomena can be
seen in Figure 3 following the initial flush event.

     Another characteristic of the tile flow was the rapid hydraulic response
of the tile system which can be noted by the sharpness of the leading edges of
the hydrographs.  To simulate  these responses, an average soil profile conduc-
tivity of 3.0  cm/hr was used  (Bottcher, Monke and Huggins, 1978).  This indi-
cates that  the reported hydraulic  conductivity 2-6 cm/hr above and  0.1-2.0
on/hr below the plow layer)  of Hoytville soil in the county soil survey report
may be either too low or that  the  years of subsurface  drainage and  good  soil
management have changed the hydraulic characteristics of this soil profile.

-------
                     SUSPENDED SOLIDS  Cmg/1)
u>
       ro
       o'
m

<—>
Q.


CO
FT.
sr
       05
              -j.    ro
              o    o

             J2	2.
                     CO
                     o
                     o
 Oi    O)
 o    o
 o    o
_J	1—
-J
o
                           FLOW Cm3/hr)
             ro
             en  en
                               CO

                               9
           en

           9
                                              m
                                              CD
 00
 o
 o
_JL_
                                               O)
(O
o
o
I
o
o
o
                                                      (O
                                                        •Mr
                                                RAIN  Com/12 hrs)
                             -nco

                             Oco

                             ^m
                               z
                               o
                               m
                               a

                               c/>
                               O
                               r;
                               D
                               CO
                                     m
                                     CO
                                -88  -

-------
                                  - 89 -

Sediment Data

     The sediment concentration  (total suspended solids) data showed two major
characteristics.   The first  was the  relatively high  sediment concentration
(all concentrations were  rather low)  at  the start  of  most tile  flow events
which  began from zero  flow,  and secondly,  the relatively  constant sediment
concentration  after  a rapid  decay  of the  initial  concentrations.   However,
some exceptions  occurred  during the early spring or late winter period possi-
bly due to freezing and thawing action.  The  high  sediment concentration dur-
ing  the  initial flow  may have resulted  from some  water flowing directly
through the unsaturated zone  to the tile drains as  the result of channeliza-
tion.  This could  easily occur especially for that region of the soil profile
directly above the  tile  line.  This rapid  flow of  water  directly  to  a tile
line is substantiated by the detection of both a herbicide and an insecticide
in the tile water shortly after surface application  of  these chemicals.  Deep
cracking of the soil was also observed  during the  drier  summer  months which
could account  in part for direct channelization of  flow.   The rapid movement
at times  of nutrients through the  soil  profile also  is consistent  with this
observation.

     The initial high sediment concentrations may be  caused partially because
the cohesive strength of the soil is a  function  of the water content.  A dry
soil profile would exhibit lower cohesive bonding and therefore might be more
erosive until the soil matrix became wetter.

     The rather  consistent  sediment concentration which occurs after  an ini-
tial event  (see Figure 4) is in line with surface erosion characteristics, but
exceptions to this observation make it very difficult to explain  actual tran-
sport  mechanisms.   However,  the  particle  detachment theory  presented  by
Bottcher, Monke, and Huggins (1978)  does address this problem.

     The actual  loadings and  concentrations of  sediment  in  the  subsurface
drainage waters were  very small as seen in  Tables  2 and 3.  The average loss
per year was less than 100 kg/ha.  The sediment concentration in the tile out-
flow was not well  correlated to flow rate  for  the four years of data.  Sedi-
ment concentration was more of a function of the antecedent  soil  conditions
than flow.

-------
                         TILE FLOW  (m3/hr)
a.
I?

5

H-
CO
to
         SEDIMENT CONC. (mg/liter)
RAINFALL (cm/12 hrs)
                                      - 06 -

-------
                                  -  91 -

       Table 2.  Water,  Sediment and  Nutrient Yields from the Subsurface
        	Drainage system	

        Component	1976	1977	1978	1979     Average
                      	cm	

       Rainfall         66         98          75        82         78
       Runoff           1.2       12           9.9       5.2        7.1

                                                kg/ha
Sediment
Sol. Inorg. P
Sol. Org. P
Sediment P
Ammonium N
Nitrate N
Sol. Org. N
Sediment N
21
.002
.005
.02
.01
.68
.05
.11
120
.073
.040
.32
.32
14.
3.6
1.1
140
.029
.034
.15
.11
4.8
.32
.91
43
.013
.006
.05
.07
5.0
.27
.21
81
.029
.021
.14
.13
6.1
1.1
.58
       Table 3.   Sediment and  Nutrient Concentrations from the  Subsurface
                 Drainage System
Component

Sediment
Sol. Inorg. P
Sol. Orgn. P
Sediment P
Ammonium N
Nitrate N
Sol. Org. N
Sediment N
1976

170
.02
.04
.22
.09
5.6
.44
.92
1977

98
.06
.03
.26
.27
12
2.9
.94
1978
mg/1
140
.03
.03
.15
.11
4.9
.33
.93
1979

83
.03
.01
.10
.13
9.6
.52
.40
Average

123
.035
.03
.18
.15
8.0
1.0
.80
Nitrogen Data

     Only seven percent  of the  total nitrogen  loss  was associated  with the
sediment.  This would be expected because of the relatively low total sediment
yield  from  the tile  drainage system.   However,  the sediment  bound nitrogen
data does show that the source of the particles being transported varies dur-
ing a storm event.  Figure 5 shows sediment N to be initially  very high which
indicates nitrogen  rich surface  particles are  reaching  the  tile line.  How-
ever, later in the event when the soil profile was closed off  to  direct chan-
nelization paths, the sediment nitrogen concentration went to near zero.  This
indicates the particles in the drainage water after direct  channelization has
stopped are originating from the nitrogen starved lower profile.

     As seen  in Table  2,  the majority  of the  soluble  nitrogen  being  tran-
sported was  in the  form of nitrate.  Nitrates accounted  for  seventy-ninety
percent of the total nitrogen  loss for  each  of the  four  years  even  though

-------
                                                                                                          0,0   _
 j=
 V.
IO
 E
 O
 Ul
                                 	AMMONIUM  N

                                 	—  SEDI'MENT  N
                     I*  • /»x*\
                     '   '
                     *l  IA * "   •

                     ll .ty
                                   8          10          12          14

                                       TIME AFTER FERTILIZATION (DAYS)

               Figure 5.   Ammonium N and Sediment N losses vs Tims (Day Zero is Jtoril 19, 1977)

-------
                                   - 93 -

total  nitrogen varied  significantly between  years.   The high  nitrogen loss
during  1977 was mostly  the result  of heavy  rainfall  occurring immediately
after fertilizer application.

     The difference between total  and nitrate nitrogen  in  1977  was mostly in
the  form of  soluble  organic  nitrogen.   The higher  percentages  of organic
nitrogen loss in 1977  was caused by very  heavy  rains  occurring  shortly after
surface  application of  urea  on April  19.  Approximately two percent of the
applied urea was lost  as urea  during this one storm.   The  urea   (organic form
of  nitrogen)  obviously  moved  directly through  the soil profile during this
high flow period.  This  is  shown in  Figures  6 and 7 by the  rapid response to
flow  of the  soluble organic  N concentrations, while  nitrate concentrations
were  only  moderately  affected.   However,  about  fifteen days later  in early
May, a  large  event occurred again and this time the majority of nitrogen lost
was in a nitrate form  (Figure 6).

     The  high concentration  level  of nitrate  during  the  May  event  was  a
delayed  reaction  resulting  from  nitrification  of the April  19 fertilizer
application.  The fifteen day period between April  19th when urea was applied
and  the May  4th  storm event provided  ample  time  for  absorption, ammonifica-
tion, mineralization,  and  nitrification  to  tie  up or  transform the organic
forms of N  to inorganic forms, primarily  nitrate.   As also seen  in Figure 6,
the organic nitrogen  was almost  totally  bound  or  transformed within 6 days
following  the  urea application,  allowing  the  later  high  nitrate losses.
Therefore soluble organic nitrogen will likely be very low unless  a rain event
occurs very close to the application date.

     The low  yields of  ammonia  resulted  from  the  ammonium ion  being easily
nitrified or  attached to  the cation  exchange complex.  The  rise in ammonia
concentrations during the late April event (see Figure  7) was  probably due to
direct  passage  of  ammonia  to  the  tile drains from  the ammonification of the
applied urea fertilizer.  Also, since ammonia  is not highly absorbed, equili-
brium of the ammonium ion within the lower soil profile would not  have time to
occur at the higher flow rates.

     The higher concentration  of all nutrients  (except  for  soluble inorganic
phosphorus)  during  1977 was  primarily the result  of  higher total fertilizer
application for that  year.  Corn  was grown during  1977 which  requires very
high  fertilization  rates.  Soybeans  during  1976 and  1978  on the other hand
required almost no nitrogen and reduced amounts of phosphorus  (see Table 1).

Phosphorus Data

     The  phosphorus concentration  data   is  quite  different  from  nitrogen
because approximately  seventy percent  of  the phosphorus  lost was associated
with sediments compared to 10 percent or less for nitrogen (see Table 2).  The
same percentage  occurred every  year.   As seen  in Figure 8,  sediment bonded
phosphorus varied similarily  to the sediment  curve.   However,  soluble inor-
ganic phosphorus concentration correlate well with tile outflow rate.

     The soluble organic phosphorus varied very little both during and between
different storm events.  Both the soluble organic and inorganic forms of phos-
phorus were very low in concentration and yielded insignificant loadings.  The

-------
IO
 u.
 lit
                                         TILE FLOW

                                         SOLUBLE  ORGANIC N
                                         NITRATE N
                                    8          10           12           14
                                       TIME  AFTER  FERTILIZATION  (DAYS)
 I
VD
          Figure 6.  Soluble Organic and Nitrate N Concentrations vs Time (Day Zero is April 19,  1977)

-------
 O)
Q

<

UU
O
O
DC

Z
LU
CD

O
CO
80


72-\


64


56-


48-


40


32-


24-


16-


 8-


 0-
                0
                 0
30    40   50

  TIME (days)
                                                                         CO
                                                                         cc
                                                                         CVJ
                                                                         •^

                                                                         o
                                                        80   90
                                        SYMBOLS:
                                         o SOLUBLE NITROGEN
                                         + NITRATE
                                        — FLOW
Ul
I
      Figure 7.  Soluble and Nitrate N vs Tine (Soluble and Nitrate N are Equal Before Late

-------
sediment bound  phosphorus was also  quite  low and would not  be considered an
environmental concern.

     During certain storms,  such  as one occurring  in  the  latter part  of May
1977, the  sediment-bound phosphorus  to sediment  ratio  was much  higher than
normal even though  the majority  of the phosphorus  lost was  still sediment-
associated.  The sediment-bound phosphorus to sediment ratio for the two storm
events  (late  April  and  early May)  shown  in  Figure  8  were  .005  and  .007,
respectively, whereas the  late  May event  (not shown)  had a  ratio  of.016.
Since the  soil  in the upper part  of the profile is more  phosphorus enriched
than the soil in the lower part  of  the profile, it seems likely that surface
particles are reaching the drain lines.

     Part of the high phosphorus concentrations in the late May event may have
been caused by  a delayed migration of the heavy applications of fertilizer in
April.  The first major  rainfall  event  (late April)  following the  fertilizer
application moved a  substantial   amount of phosphorus  into the  tile drains
because the sediment rate was then also high.  On the  next major event  (early
May) very  little sediment and consequently very little phosphorus attached to
the sediment occurred over that previously attached.  This was probably caused
by a response lag due to the slow movement of soil fines in the profile.

Pesticide Data

     A herbicide  (Lasso II) and an insecticide  (Furadan) were also applied on
April 19,  1977.  As  seen  in Figure 9, these chemicals  were  also detected in
the  tile drainage waters.  Direct passage  of some surface water  to the tile
drains must have occurred since these chemicals are normally absorbed quicKly
by the soil.  The total loss of the pesticides  was on the order of one  tenth
of one percent of the applied, 8 and 13 kg/ha of Lasso II and Furadan,  respec-
tively.  Concentrations  of  the herbicide were  not detected after  the  April
storm event;  however, concentrations of the insecticide, which were initially
much higher than  for  the herbicide, were still  detectable  during an early May
storm.
                             WATER QUALITY IMPACT

     The  water quality  impact  of  the  well-managed  17  ha  field  with  tile
drainage  can  be evaluated by comparing water, sediment and nutrient yields on
a unit area basis with  a more typical drainage  saturation  in the area.  Sur-
face  runoff  from  a 942 ha watershed  (Smith-Fry  Drain) just 12 km  to the north
of  the  tile  drained field had been  measured and  sampled over  the  same time
period  as part of the Black  Creek  Project (Lake, 1977).  However,  difference
between  these drainage  areas should be  noted.   First,  there  is  an obvious
difference in size.   Also the larger area has a more  diversified  land use and
is  in general  not as  well managed.   Approximately 70  percent of the soils in
the  larger drainage  area are similar  to those  in  the  17  ha field, but the
remainder are  mostly gently rolling glacial till soils.   Also the larger area
is  only 50 percent tile  drained while complete  subsurface drainage  exists for
the studied field.  Still such a comparison may  be beneficial because it will
reveal  what  ultimately  might be attained  from subsurface drainage or total

-------
    50
   40
   30
10

 E
 Q
 UJ
   20
    10
   TILE FLOW


   SEDIMENT P


•- SOLUBLE  INORGANIC P


   SOLUBLE  ORGANIC  P
                                                                                                   O.O   -»
                                                                                                        «

                                                                                                        JB

                                                                                                   1.0   £!



                                                                                                   2.0   -S.
                                                                                                        <
                                                                                                        o:
                                                                                                   .30
                                                                                                   .15
                                                                                                        u

                                                                                                        o
                                                                                                        u
           4          6          8         10          12           14         16


                                    TIME AFTER  FERTILIZATION  (DAYS)


    Figure  8.  Sedirrent, Soluble and Inorganic  Soluble P vs Tine (Day Zero is April 19,  1977)
                                              18

-------
                        TILE  FLOW   tm3/hr)
                                 ro
                                 o
                                                                    Ul
I
P-

8-
en
rt

9
 H-
 cn
 to
                     Ol

                     O
o
o
01
o
          PESTICIDE CONCENTRATION
        liter)



     -86-
         ?*   X    °
         O   O    b


RAINFALL (cm/12 hrs)

-------
                                   - 99 -

elimination of  surface soil erosion to  abate nonpoint source  pollution from
cropped fields.

Water Yield

     The water yield was much lower  (57 percent for  1976-1978)  from the 17 ha
field  with  its  extensive subsurface drainage  system  than  for  the  larger
drainage area although total annual  rainfall  for  the two  locations were simi-
lar  (see  Table 4).   The  reduced  water yield from  the  17 ha field  indicates
that more water was being  stored in  the soil  profile for  potential later eva-
potranspiration.   Some of  the  moisture difference  may also be accounted for
through deep seepage,

Table 4.  Percent Difference in Unit Loadings and Concentrations of Rainfall, Water
          Loss, Sediment and Nutrients Between Discharge  from the Well-Managed 17
	ha Field and a More Normal Drained  Area (Smith-Fry Drain) .	

  Component                             Percent Difference
1976


Rainfall
Water Loss
Sediment
Sol. Inorg. P
Sol. Org. P
Sediment P
Ammonium N
Nitrate
Sol. Org. N
Sediment N
Load-
ing
0
-89
-97
-97
-83
-98
-98
-88
-84
-97
Cone.

_
-
-67
-60
+33
-72
-81
+24
+76
-71
1977
Load-
ing
+2
-35
-72
-47
-27
-81
-45
-9
+230
-76
Cone.

—
-
-58
-14
0
-71
-13
+42
+380
-62
1978
Load-
ing
N/A
-46
-83
-86
-58
-77
-85
-42
-81
-85
Cone.

_
-
-32
-74
-31
-57
-73
+9
-63
-72
Average
Load-
ing
+1
-57
-83
-79
-63
-87
-80
-37
-10
-89
Cone.

—
-
-62
-56
-13
-74
-62
+38
+33
-75
i.e.,  increased  recharge.  However,  this would  be small  since  most  of the
runoff difference occurs during the summer months when et is high and there is
little tile outflow.   Two  to  four centimeter  rainfall  events  during July and
August  resulted  in sufficient  surface runoff  in the  Smith-Fry  Drain but no
subsurface drainage from the studied field.  The reduced water yield will also
have  a direct  impact on  sediment and nutrient  loadings  since  loadings are
equal  to  concentration times  flow.  Therefore, even  if the concentrations of
soluble nutrients  are  increased,  as well they might with subsurface drainage,
loadings may still be  reduced.

Sediment and Nutrient Yields

     As shown in  Table 4,  sediment and  nutrient  loadings were  usually much
lower from the 17 ha field than from the larger drainage area even though some
of the soluble nutrients  had  higher concentrations.  Although  there was evi-
dence that some soil  fines had migrated  through  the  soil profile to the tile

-------
                                  - 100 -
drains, the amount  was small and  in  no way  approximated the  sediment loads
found in the surface runoff.  Loadings of nitrate and soluble organic nitrogen
were not significantly  reduced  during 1977 mostly because of  the heavy rain-
fall event which  occurred shortly after the 170 kg/ha urea application.  How-
ever, the average fertilization rates were twice as high for  the tiled field
than for the surface drained area.  In general, the studied field had a signi-
ficant reduction of sediment and nutrient loadings and moderate reductions for
concentration of  nutrients,  except for nitrate  and  soluble  organic nitrogen.
Phosphorus loading reductions were more pronounced than for total nitrogen and
especially  for  nitrates.  These  findings point out that good fertilization
management  including  the use of  less susceptible  nitrogen  forms  to runoff,
better  placement,  and  timely  application  are  also  needed  to enhance  the
already good water quality characteristics of  the subsurface drainage system.
A note of caution should be given here about looking only at loadings in judg-
ing  the water  quality benefits  of  a particular  practice.   For  instream
effects, concentration  levels  may  be more  important than  loadings, whereas
large water bodies generally are more impacted by total loadings.
                           SUMMARY AND CONCLUSIONS

     Flow, sediment and nutrient data were collected for four years  from a 17
ha  field with  complete  subsurface  drainage.   These  data  were subsequently
analyzed.  The field had limited surface runoff which made the drainage system
a very effective water quality management practice.  The data from the subsur-
face drainage system were also compared to data collected  from  a more typical
drainage  area with  more surface  runoff in  order  to  evaluate  the  potential
impact of drainage practices for nonpoint source pollution abatement.

     The following conclusions are drawn from the analysis of the outflow from
the subsurface drainage system:

 1.  The majority of the subsurface drainage from the 17 ha  field occurred as
     the result of winter moisture accumulation.

 2.  Hydraulic conductivity of  a soil profile may be  increased by  long term
     good soil management which includes subsurface drainage.

 3.  Over seventy percent of  the nitrogen lost  in  the  drainage water  was in
     nitrate form.

 4.  Approximately seventy percent of the phosphorus losses were  in the form
     of sediment-bound phosphorus.

 5.  Sediment, sediment-bound phosphorus and pesticides were all able to move
     through the soil profile, indicating the presence of direct flow channels
     during initial periods of a storm.

 6.  Heavy rainfall occurring shortly after fertilizer  is  applied can greatly
     increase losses of  nitrogen and to a lesser  degree  phosphorus through a
     subsurface drainage system.

-------
                                  -  101 -

 7.  Good  fertilizer  management can reduce the  amount of  soluble nutrients
     reaching tile drains.

     The following conclusions  were  drawn from a comparison of the data from
the  17  ha  subsurface drainage system with  data from a nearby watershed where
surface runoff was also an important factor:

 1.  The 17 ha field with a complete subsurface drainage system and restricted
     surface  runoff  significantly reduced  water and  sediment losses as com-
     pared to the more  normal drainage situation for  this  area of the Maumee
     Basin.

 2.  The  better  drained  area  also  provided  a  significant  reduction  in
     sediment-bound nutrient  loadings particularly as affecting phosphorus.

 3.  Concentrations  of nitrate-nigrogen  and  soluble  organic nitrogen  were
     higher  in  the runoff water from the well-drained 17  ha field than from
     the more normal  drained area.  However,  these  higher  concentrations may
     not lead to increased loadings.

 4.  A complete subsurface drainage system on  recommended soil types may well
     be  thought  of in  terms of a  water quality management practice, except
     when instream nitrogen concentrations are of a concern.
                 II. SEDIMENT MOVEMENT INTO SUBSURFACE DRAINS
                            FROM BACKFILL PROFILES
                                   OBJECTIVE

     The  objective  of this  study was  to  investigate the  effect  of various
envelope  materials  and soil  conditioners  on  sediment movement from backfill
profiles  of  Hoytville silty  clay.  The soil  was contained  in  a laboratory
apparatus  which closely  duplicated  the  opening  between two  adjoining tile
drains and the  constructed  trench above the drains  which would be backfilled
with the excavated soil material.  In addition, sediment losses from the simu-
lated  trenches  backfilled with  Hoytville  silty  clay, Latty clay and Blount
silt loam were also evaluated and compared.


                                     SOILS

     Hoytville, Blount, and  Latty soils were  selected for the  experiment on
the basis  of  their  clay content and  location  within the Maumee Basin (Figure
10).  These soils provided an opportunity to observe  the  contrast in sediment
losses from soils with a relatively wide  range  in clay contents.  A quantity
of each soil was collected  in the fall of 1977  to a depth of 90 cm and then
mixed  similarly to  a field  trenching operation.    Later,  the soils  were
screened to ensure an aggregate size of 5 cm or less.

-------
                                   - 102 -
                                         SOIL TYPES
                                          A- Blounf
                                          B-Hoytville
                                          C-Latty
                          MAUMEE RIVER BASIN
                          Figure 10.  Soil  Locations
     Hoytville silty  clay soils are depressional  and nearly  level with  very
poor drainage.   They  are classified as Mollic Ochraqualf  and  have a clay con-
tent of  33  to 48 percent  with a moisture  content  of  19 to  26 percent  (dry
basis)  over  the  90 cm depth.   Initial  flows from a  monitored drainage system
in this soil after a storm event sometimes  have  a milky appearance indicating
fine sediment in the effluent.

     Blount silt loam  soils  are somewhat poorly  drained with nearly  level  to
gently undulating  relief in upland positions.  Blount  is classified  as Aerie
Orhraqualf with a clay content of 13 to 36  percent and moisture range of 13  to
24 percent (dry basis) over the 90 cm profile.

     Latty clay is classified as a Typic Halplaquept with  a  clay content of  40
to 53 percent and moisture range of 24 to 34 percent (dry  basis)  over the pro-
file.  These  soils were developed in  a heavy  lacustrine clay  layer and are
very poorly drained.

-------
                                  -  103 -
                              ENVELOPE MATERIALS

      In  this study,  five different envelope  materials were  tested with  the
Hoytville  soil to evaluate their effectiveness for reducing sediment  losses.
Topsoil and  its combination with two different soil conditioners, gravel  and  a
synthetic  fabric, were selected as the envelope and filter envelope  materials.

      Topsoil  is  probably the most common  envelope material  used  around  tile
and  plastic  subsurface drain lines  in the Midwest to  prevent misalignment and
damage during backfilling operations.  Because the organic content  of  topsoil
stabilizes the aggregate structure, Hoytville topsoil with 6 percent  organic
matter has the potential  of providing and maintaining  a porous envelope around
a  drain  tube.   Topsoil from  the upper 18 cm of the Hoytville profile was col-
lected, air  dried for two days,  and then crushed  so  that all the soil passed
through a  3  cm screen.  An appropriate quantity was then  placed by hand at the
bottom of  selected soil bins  and packed so that the 8  cm  enveloped at the bot-
tom  of  the  bin  was under  an equivalent  overburden  pressure of   100 gm/cm
 (equal to  a  67 on depth of Hoytville backfill).

      Two  soil  conditioners,  Petroset  and  Portland  cement,  were  applied to
screened,  air-dried  Hoytville topsoil.  Soil  conditioners are additives  which
artificially stabilize  soil  aggregates  externally  rather  than  internally.
Although  these  conditioners  have  been usually  applied on  soil  surfaces to
prevent wind and water erosion from freshly  earth cuts  during construction,
similar materials have been demonstrated by Diericfcx et al.   (1976)  and Bishay
et al.   (1975) to improve the permeability and aggregate  stability around tile
drains under saturated conditions in clay loams.

      Petroset  is a  commercially  available   rubber  emulsion that  possesses
hydrophobic properties.   An emulsion was sprayed on the topsoil at a 5  percent
•ratio of emulsion/soil by weight for saturated conditions and according to the
aggregate  size  range.   The conditioned topsoil was  then placed at  the bottom
of a  soil  bin as previously described.

      Portland cement has  also been employed successfully  in the past to modify
and   stabilize  soils  for construction  purposes.   Ahuja  and Swartzendruber
 (1972) observed  significant  structural  aggregate  stability  of Russell   silt
loam  under saturated conditions when the cement  was  applied at low rates and
allowed to cure properly.  In this experiment, Portland cement was  applied to
the screened air-dried topsoil at the rate of  1 percent by weight with  15 per-
cent  by weight of water  sprayed on  the coating to start  the  curing process.
The conditioned soil was  then placed in two soil bins,  packed to uniform  depth
of 8  cm,  and  allowed to cure  for  fourteen  days before Hoytville soil  was
placed on  top of the envelope.

      Of the  many envelope materials available, sand  and gravel  are probably
used  more  extensively than other materials to improve hydraulic entry condi-
tions, bedding conditions, and/or filtration.   They are usually pit-run rather
than  specifically designed  according to some  standard criteria.   Both Winger
and  Ryan   (1971)  and  Luthin   et al.   (1967)  proposed  a  design criteria  for
gravel envelopes that is  based on gradients being  less  than one.  According to
them, the graduation of the envelope did not play  as significant a role as the
thickness and permeability of the envelope in  reducing  the convergence  effects

-------
                                  -1 04  -

of ground water on drainage lines.  Walker  (1978)  reinforced their conclusion
by observing  that the  critical  failure gradient  was a  function  of the soil
properties rather than the retaining envelope once  minimum mechanical support
for soil bridging  over  the envelope voids was provided.  In the Maumee Basin,
#11 pea gravel  has been frequently used whenever  a gravel  envelope  has been
recommended.  Although  it  is  not generally used with soils having a high clay
content, an 8  cm thickness of #11 pea  gravel  was nevertheless placed  in two
soil bins to evaluate its effect on sediment losses.

     In the past 20 years, synthetic fabrics have been  developed  and utilized
as  filter   envelopes for  subsurface  drains.   Where plastic  drain pipe  is
installed and soil conditions permit, fabric filters  have  emerged  as the com-
mon alternative to conventional aggregate envelopes of sand and gravel because
of their reasonable cost and handling convenience.

     Although  many nylon  socks  and  other  commercially available  synthetic
fabric sheets  have been used to protect drains  in the sand soil areas of the
Maumee Basin, Mirafi 140, a product of  the  Celanese Corporation,  was selected
to evaluate its potential  for reducing sediment  losses with Hoytville silty
clay.  Mirafi  140  is composed of nylon covered  polypropylene fibers randomly
fused together  into  a thin sheet with an average thickness of 0.75 cm, effec-
tive pore size of  0.085  mm,  density of 140 gm/nr,  and  permeability of 5x10
cm/sec.  A  15  cm by 30 cm rectangle was cut and glued around the outside edge
to the soil bin bottom to prevent sediment and water  from by-passing the sheet
at  the tile  crack.  Marks  (1975),  testing Mirafi 140  against  woven fabric
sheets of polyester and aggregate envelopes of sand  and  gravel  in soils rang-
ing from fine sands to silty clays, indicated the development of a filter cake
layer at the boundary of the fabric.  The over-all permeability of both fabric
and filter cake was comparable to the tested aggregate envelope.
                           APPARATUS AND PROCEDURE

     Sixteen backfill profile soil bins, 90 cm deep with base dimensions of 15
cm by  30  cm, were constructed side by side to represent portions of the back-
fill trench  in the  field.   The base dimensions  resulted  from the symmetry of
the  streamlines  longitudinally between  butted 15  cm diameter by  30 cm long
drain tile and the approximately parallel streamlines above the tile drain.  A
1.6  mm (one-sixteenth  inch)  crack width  across the bottom  of  each soil bin
represented  the nominal opening between  tile sections.  A schematic drawing of
a single backfill profile model is shown in Figure 11.

-------
                                  -105  -
           BACKFILL
           TENSIOMETER-
                                                  PONDED WATER

                                                      SOIL BIN
                                               \   MOISTURE
                                               _^	EXTRACTORS
                        „  ^=.^_J.,I             / (POROUS CUPS)
            MANIFOLDS   ^ ^^3^^            /

                                                     ENVELOPE
                                                      MOISTURE
                                                     XTRACTORS
                                                   CPOROUS PLATES)
                                               SEDIMENT COLLECTOR
                              VACUUM PUMP |  "TV-OVERFLOW SEDIMENT
                                                  RESERVOIR
             VACUUM RESERVOIR

                      Figure 11.  Backfill  Profile  Model

     Each of  the sixteen soil bins  was filled in  a  random order with  either
Latty; Blount; Hoytville; one of the five envelope  materials under a Hoytville
backfill; or  one of the replications  for  each treatment.   Each  backfill  was
then  compacted  slightly to  reduce the  initial  settlement  after  water  was
applied.

     Initially,  3200 ml  (equal  to  a 3-inch ponding depth)  of snow  melt water
was  applied twice  to  the  top  of  each  profile over  a two  week conditioning
period to bring  all backfill replications to approximately the same antecedent
moisture condition  and  to  induce settlement.  At the end the two week period,
ceramic porous cup moisture extractors were inserted  into the  profile along
with  tensiometers  to  monitor  moisture changes  (see  Figure 11).  Four  days
after their installation,  suction was  applied to  the  porous cups  and  previ-
ously installed porous  plates  at  the bottom of  each soil bin  in preparation
for further experimentation.

     The study consisted of five cycles with a four-day wetting  phase followed
by a  ten-day  drying phase  to  simulate the wetting and drying  process  in the
field.  Such a wetting and  drying process could cause sediment movement in and
through the profile by possible slaking, piping, and turbulence.  The quanti-
ties of gravitational water and sediment loss from  the bottom of each profile
were collected, measured, and recorded  after each application of water.

-------
                                 -106  -
                            RESULTS AND DISCUSSION

     The sediment yield, water, and water transient times through  the profile
after each application  of  water for the  initial  conditioning  period and five
wetting and drying  cycles  are shown in Tables  5  and  6.  A water  balance  for
each cycle  was also maintained on the assumption that all the  applied water
was removed.  After the first complete cycle of wetting and drying, approxi-
mately one-half  of the  water added during  each  of the remaining cycles  was
recovered as gravitational water.   The rest of the water was either removed by
extraction or  evaporation.  The  evaporation from the  soil surface and water
vapor loss through  the  vacuum pump tended to increase  with  successive cycles
because of an average ambient temperature rise of 13 C during the test period.
Since transient times after the second cycle had  become fairly constant, dis-
tinct channels through the profile apparently were established.
Table 5.  Sediment Loss (gm), Water Loss (ml) and Percolation Time (sec or min)
          for Wetting and Drying Cycles on Hoytville Silty Clay, Blount Silt
          Loam, and Latty Clay
Bin and
soil*
2H


10H


IB


13B


6L


16L


Unit
gm
ml
sec
gm
ml
sec
gm
ml
min
gm
ml
min
gm
ml
sec
gm
ml
sec
1
0.75
2130
32
0.93
2250
29
0.12
2170
280
0.29
2160
235
0.92
2370
32
1.27
2550
24
Wetting
2
0.41
2090
46
0.41
1740
37
0.072
610
40
0.082
860
30
0.46
1950
24
0.49
1950
25
and Drying
3
0.32
1670
55
0.33
1690
44
0.080
780
65
0.026
990
38
0.39
1850
35
0.43
1850
23
Cycle
4
0.27
1650
65
0.30
1670
44
0.051
760
75
0.045
960
41
0.39
1840
40
0.40
1840
23
5
0.29
1670
73
0.33
1640
44
0.072
870
72
0.026
1020
40
0.42
1910
36
0.43
1910
23
 *H=Hoytville silty clay,  B=Blount silt loam,  L=Latty clay

-------
                                -107  -
Table 6.  Sediment Loss  (gin) , Water  Loss  (ml)  and  Percolation Time (sec)  for
          Wetting and Drying Cycles  on  Hoytville Silty Clay with Different
          Envelope Materials
Bin and
material*
2H


10H


3T


7T


4P


9P


5C


11C


8G


15G


12F


14F


*H=control ,
Unit
gm
ml
sec
gm
ml
sec
gm
ml
sec
gm
ml
sec
gm
ml
sec
gm
ml
sec
gm
ml
sec
gm
ml
sec
gm
ml
sec
gm
ml
sec
gm
ml
sec
gm
ml
sec
T=topsi
1
0.75
2130
32
0.94
2250
29
0.65
2230
35
0.50
2220
41
0.61
2290
36
0.62
2220
37
0.94
2400
40
0.61
2370
46
0.92
2440
37
0.82
2332
33
0.93
2410
50
0.53
2370
37
Wetting
2
0.41
2090
46
0.41
1740
37
0.35
1770
37
0.27
1590
45
0.38
1690
40
0.34
1760
60
0.51
1730
49
0.31
1670
48
0.46
1960
56
0.43
1840
36
0.22
1660
70
0.32
1710
51
and Drying
3
0.32
1670
55
0.33
1690
44
0.31
1570
42
0.28
1560
40
0.25
1680
55
0.32
1700
53
0.34
1720
53
0.27
1680
65
0.42
1860
55
0.30
1820
41
0.25
1520
65
0.31
1750
53
Cycle
4
0.27
1650
65
0.30
1670
44
0.31
1560
44
0.25
1540
43
0.20
1630
49
0.2S
1700
49
0.36
1720
47
0.28
1750
47
0.38
1810
55
0.43
1800
36
0.29
1540
52
0.28
1680
55
5
0.29
1670
73
0.33
1640
44
0.32
1620
53
0.27
1600
58
0.17
1660
58
0.38
1690
48
0.32
1680
68
0.28
1690
72
0.43
1830
65
0.47
1700
40
0.32
1640
83
0.26
1700
68
oil, P=Petroset, C=cement, G=qravel, F=fabric

-------
                                  _ 108 _

                              ENVELOPE ANALYSIS

     The average sediment  losses  and concentrations for the  experiments with
envelope materials are plotted in Figures 12 and 13.  Statistical tests at the
5 percent level  showed no  significant  effects on  the reduction  of sediment
losses by the  envelopes,  especially once sediment  losses  approached a steady
state condition.  Moreover, gravel might have even  shown a detrimental effect
on the stability of  the backfill and envelope interface if the experiment had
been continued.

     For near steady state conditions, the sediment losses and concentrations
averaged about 0.3 gm and 175 mg/1, respectively.   The amount of sediment loss
seemed to increase with the  amount of gravitational  water passing  from the
backfill profile.

     Although sediment losses through the fabric filter were about the same as
for  the control  of  Hoytville  without  an  envelope,  the  fabric  pore  size
apparently limited  the sediment  size to  less than 0.085  mm.   However,  no
noticeable reduction  in permeability over time occurred from  blockage of the
fabric pores.

     The effect  of successive wetting  and  drying  cycles  had a  significant
influence on sediment losses initially but diminished as the losses approached
a steady state condition.   As the soil profiles settled with successive cycles
of operation,  the  transient time of water flow through the profile increased
indicating a decrease in permeability.


                              SOIL TYPE ANALYSIS

     The average sediment losses and concentration  for  Hoytville,  Blount, and
Latty soils  are plotted in Figure 14.   Sediment losses from  the  three soils
showed a similar trend with time.  A statistical analysis of the data showed a
significant difference in sediment losses between Hoytville and the other soil
types, especially once the sediment losses approached a near steady state con-
dition.

     The average sediment loss for Blount was about 75  percent less than that
for Hoytville  with an average sediment loss and concentration over the steady
state conditions of  0.05  gm and  60 mg/1, respectively.   The low  losses for
Blount may  be partially attributed both to the greater consolidation of the
profile and to the smaller quantity of effluent during the wetting phase.  The
finer pores of Blount reduced the erosive potential of the gravitational water
by causing the water to percolate through the profile slowly.

     Both Hoytville and Blount  had extensive surface cracking at the  end  of
each drying  phase  in the  laboratory.  These cracks extended  into the profile
from a few millimeters to over 15 cm.  During the collection  of the Hoytville
soil, drying cracks  were  noted  from the  soil surface  to the bottom of the
trench.  The most recent cracks had an  average width of 6 mm.  Older cracks,
which  were   sometimes  filled with  the  dark topsoil,  had  similar  widths and
depths.  Such  cracks  intercepting the drainage line could affect  the overall
sediment and nutrient loadings in the drain effluent.

-------
                         SEDIMENT LOSS  (gms)
  to
t
O
  Q
  (C
  H-

  8
c
O
•<
3 "-^ ro co '4^ 01 b> ^ CD co o
1 1 1 1 1 1 I 1 T — T"
^ Ka
1 	 " T
             •n
       m
       o
          IV)
          Ol
  rt
  rt
o
M

cn


31
  o
  6
  0)
  M
  cn
                 SEDIMENT CONCENTRATION (mg/1)
               01
               o
                o
                o
Ol
o
N>   rv>
O   Ol
o   o
CO   CO  -U
O   Ol  O
o   o  o
Ol
o
Ol
o
o
       o
       m
     O  rv>
                             V
Ol   O)
Ol   O
o   o
                                                    I
                                                   I  I  I
                                                   D • >

                                                   H2O
                                                   ° o r;
                                                   X)   »
                                                   w m<
                                                   03®.
                                                   =: <
                                                     9.
                                                     o
                                                     •a
                                                     to
                               - 60 L -

-------
                           SEDIMENT LOSS  (gnt«)
  OJ
             O   -*
            ro   co
        en   o>
        00   »
                                                        II
                                   5"
                                   T3
                                   (D
                                  Oil

-------
                         SEDIMENT LOSS (gms)
            o  '_t
                      GO
                      6i  o>  '-g   bo  co  b  :_i
en en
H- (D
  01
  01
MO,

us
en
  §
  01
O


?  °
m
CO
       01
       o  -t*
  rt

  ft-
                SEDIMENT CONCENTRATION  (mg/1)
  te1
  01
 I

 O
 I
o

o

m
c*
o
   r>o
          CO
          Ol
                  o
                  o
              -*  ro  ro
              01  o  ui
              o  o  o
CO  CO
o  en
o  o
Ol

O
Ol
O
o
Ol   O)
Ol   O
o   o
                             I


                                               I
                                            n • >

                                            »IS
                                            §55
                             - ILL -

-------
                                  - 112 -
     For Latty, the average sediment loss and concentration was  about 20 per-
cent more  for each  cycle of wetting  and drying  than for Hoytville with an
average sediment loss of  0.4  gm and concentration of  225  mg/1.   These larger
losses were probably due  in part to the highly porous condition of the back-
fill caused by the heavy clay aggregates maintaining  their  blocky structure.
This condition allowed  water  to move through the  profile  rather rapidly pro-
viding opportunity for  less absorption of the water and greater soil detach-
ment than with the other soils.
                                   SUMMARY

     The average sediment loss from Blount silt loam was about 75 percent less
than  from  Hoytville.  This difference  may  be attributed in part  to the com-
plete settlement of  the  Blount profile and the subsequent  low hydraulic con-
ductivity of  the  column.  However, Latty averaged about 20 percent more sedi-
ment than Hoytville.  This difference  was probably due in  part  to the incom-
plete  breakdown of  the  large clay  aggregates in  the  Latty profile  and the
associated  large  channels which  then  occurred  around  the aggregates.   The
effluent from the Hoytville  and  Latty was always  cloudy  in appearance while
that from the Blount was clear.

     The five envelope materials did not  significantly  reduce  sediment losses
from  the Hoytville  soil.  The gravel envelope even  showed a  potential for
increasing sediment losses.  The gravel may have reduced soil  bridging at the
envelope  and backfill  interface   allowing  soil particles to  move  into and
through the envelope more rapidly.


                                  REFERENCES

 1.  Ahuja,  L.R.  and D.  Swartzendruber.  1972.  Effect of  Portland cement on
     clay aggregation and hydraulic properties.  Soil Sci.  114(5):359-366.

 2.  Baker,  J.L.  and H.P. Johnson.  1976.   Impact of subsurface  drainage on
     water  quality.   3rd National  Drainage  Symposium.   Amer.  Soc.  of Agr.
     Engrs.,  St. Joseph, MI.  pp. 91-98.

 3.  Bishay,  E.G. and W. DiericKx.  1975.  Drainage efficiency in  a  low perme-
     able clay-loam soil  through physical modification of the trench bacKfill.
     Pedologie  25(3): 179-189.

 4.  Bottcher,  A.B., E.J. Monke and  L.F.  Huggins.   1980.   Subsurface drainage
     model with associated sediment transport.  Trans. ASAE 23(4): 870-876.

 5.  Dierickx,  W. and D. Gabriel.  1976.  Stabilizing  backfill of drain  pipes
     and  drainage efficiency.  Third  International Symposium  of  Soil Condi-
     tioning, Mededelingen Fakultett  Landbouw-Wetenschappen Rijksuniversitet
     Gent.   41(1):293-301.

 6.  Lake,  J.   1977.    Environmental  impact of  land  use  on  water quality.
     EPA-905/9-77-007-B,  Final Report on the Black Creek Project  (Technical).
     Region  V,  USEPA, Chicago.  280 p.

-------
                                  - 113 -

 7.   Luthin, J.N., G.S. Taylor and C.  Prieto.  1968.   Exit gradients into sub-
     surface drains.   Hilgardia 39(15):418-428.

 8.   MarKs, B.D.   1975.  The behavior  of aggregate and fabric  filters  in sub-
     surface applications.   Report,  Department of Civil  Engineering,  Univer-
     sity of Tennessee.  149p.

 9.   Monke, E.J., D.B.  Beasley and A.B. Bottcher.  1975.   Sediment contribu-
     tions to the Maumee  River.  EPA-905/975-007, Proc.  Non-Point Source Pol-
     lution Seminar.   Region V, USEPA, Chicago.  71p.

10.   Schwab, G.O., B.H. Nolte and R.D.  Brehm.  1977.  Sediment  from drainage
     systems for  clay soil.  Trans. ASAE 20(5):866-868.

11.   Taylor, G.S.  and T. Coins.  1967.  Field evaluations of tile drain filter
     in a  humid  region soil.   Res.  Ctr.  154,  Ohio Agr. Res. and  Dev.  Ctr.,
     Vfooster, Ohio.

12.   WalKer, R.E.   1978.  The interaction of synthetic envelope materials with
     soil.  M.S.  Thesis.  Utah State University.

13.   Winger, R.J.  and  W.F.  Ryan.   1971.  Gravel envelopes  for pipe drains —
     Design.  Trans.  ASAE 14(3):471-479.

14.   Zaslavsky, D. and G.  Kassiff.  1965.  Theoretical formulation of piping
     mechanism in cohesive soils.   Geotechnique 15(3):305-314.

-------
                                - 114 -


           ALGAL AVAILABILITY OF PHOSPHORUS ASSOCIATED WITH

       SUSPENDED STREAM SEDIMENTS OF THE BLACK CREEK WATERSHED

                                  by

                     R. A. Dorich and D.W. Nelson
     During the past 2 decades the quality of water in Lake Erie has come
under severe scrutiny, and review of pertinent data has shown a steady de-
cline.  Although other data have demonstrated the steady decline in water
quality, algal cell numbers, P loading history and hypolimnial oxygen levels
data seem to exemplify one of the major problems.  Davis (1964) presented
data that showed that between 1919 and 1963 there has been a consistent
increase in the average number of phytoplankton in Lake Erie.  The average
number of cells increased from 81/ml in 1929 to 2423/ml in 1962.  The
intensity and length of the periods of maximum cell counts have also in-
creased, while the minimums have become shorter, less pronounced,  and in
some cases, failed to appear.  Williams et al. (1976) analyzed P in sedi-
ment cores from six sites in Lake Erie and correlated depth below the
sediment:water intefface   with time of deposition.  Results show a gradual
increase in the concentration of sediment P since 1948.  Some samples
showed a doubling in non-apatite P in the last 10 years.  However, there
were indications of vertical migration of P upward.  Dobson and Gilbertson
(1972) presented hypolimnial oxygen level data for the years 1929, 1949,
and 1969.  When 1929 data and 1969 data are compared, the results show
that from the beginning of the stratified period the rate of oxygen deple-
tion has significantly increased, and the onset of deoxygenated conditions
in the bottom water has occurred earlier.

     The dangers of anoxic conditions include the death of fish and other
aerobic organisms, the production of odorous and unpalatable water, the
fouling of water treatment facilities, and because nutrients are released
from sediments during anoxic or reduced conditions, the possibility exists
of cyclic self-fertilization process being initiated (Burns and Ross,
1972a).

     The indicators of serious eutrophy discussed above did, in fact,
point directly to approaching problems in Lake Erie.  The Federal Water
Pollution Control Administration  (FWPCA) estimated in 1968 that approx-
imately 2600 square miles of the 5650 square miles of hypolimnion of the
central basin of Lake Erie was oxygen deficient  (< 2 mg/1).  A  1970 study
was initiated to determine the causes and effects of oxygen depletion in
Lake Erie, and the results were presented in a  1972 report (Burns and
Ross, 1972b).

      In July of 1970 a massive algal bloom occurred in  the Central basin
of Lake Erie which depleted  the phosphorus concentration to near
undetectable levels in 80% of the surface water  of the  basin, and sub-
sequent  sedimentation and  death caused a layer of algae about 2  cm thick
to blanket approximately  70% of the basin floor.  Aerobic decomposition
of the  July bloom combined with additional blooms accounted  for  88% of
observed oxygen depletion  during  the month of August,  1970.  The onset of

-------
                                - 115 -

anoxic conditions in mid-August brought an eleven-fold increase in P
regeneration rates  (from 22 ymolesP m ^ day"-'- to 245 ymoles P nf^ day~^).
The indirect cause of the extensive oxygen depletion was massive algal
blooms.  Furthermore, since P is often the limiting nutrient for algae in
Lake Erie, it was projected that if P inputs were decreased so that algal
glooms were limited, oxygenated conditions would be maintained for a
longer period and the Lake would return to an "acceptable state" (Burns
and Ross,  1972c).

     The sources of P input into Lake Erie were also investigated
(Gilbertson, £t al., 1972).  Although agricultural runoff inputs quite
probably did contribute a portion of the total P, this study based its
judgements only on municipal and residential inputs.  However, a more
recent report by the International Joint Commission (IJC) (1980) pre-
sented data which showed "land use" activities contributed nearly half
of the total P load to Lake Erie in 1976, 20% of which was unrelated to
agricultural activities.

     Inputs of P up to the present have largely been based upon total
P transported, but as indicated above, the important portion of the total
sediment P to the Lake Erie system is that portion which becomes available
for algal growth.  Therefore, in order to properly assess the impact of
P in agricultural drainage or runoff upon an aquatic ecosystem the algal
availability of the total P transported must be determined.

     In regard to the P availability problem Ryden et al. (1973) states:

     "At present it is difficult to estimate the impact of runoff-
     and stream-derived P on standing waters, and such consider-
     ations can only be made if the forms of P relevant to bio-
     logical productivity are measured".

The IJC (1980) concurred when it stated:

     "...the Commission recommends a reassessment of surveillence
     and research activities to ensure the development of a data
     base adequate to address the question of relative biological
     availability  of phosphorus in the Great Lakes from the vari-
     ous direct and tributary point and nonpoint sources, so that
     the efficacy of point versus nonpoint source control measures
     can be more precisely determined".

     In order to properly address a major objective of the Black Creek
Watershed Project; that is,to assess the role of agricultural activities
along the Maumee River in the pollution of Lake Erie,  the availability
to algae of P derived from eroded soils within the watershed should be
determined.  The general purpose of this study, therefore, is the
determination of the quantities of algal available P in drainage water
of the small agricultural watershed in northeastern Indiana, the Black
Creek Watershed (Fig.  1) which is typical of subwatersheds within the
Maumee River basin.   This data will indicate the water quality of the
effluent of the watershed in regard to one parameter of water quality.
The effect of the addition of this water to the receiving body,  in this
case,  the western basin of Lake Erie can be,  therefore, more easily

-------
                          - 116 -
    APPROXIMATE SCALE
          KILOMETERS
  1/2     0      1/2
IN
Figure 1.  The Black Creek study area,  Allen County,  Indiana.

-------
                                _ 117  -


evaluated.   Other aspects of this study will ensure that the data collected
can be more easily evaluated with respect to data presented by investi-
gators associated with other similar projects.

     The specific objectives include  (1) the determination of the avail-
ability of sediment-bound P to algae  (2) the determination of the pro-
portions of sediment Pi and total sediment P  (TP) which are available for
algal assimilation and (3) the comparison of various methods of assessing
algal available P in sediments.
                           LITERATURE REVIEW
     The most available form of P to algae in lakes and streams is soluble
inorganic P (Sol Pi) (Vollenweider, 1968; Bartsch, 1972), although not all
Sol Pi as determined by chemical analysis of drainage water of the Black
Creek watershed is available to algae (Dorich1).  Sediment P most likely
behaves as a buffer or "pool" for replenishment of Sol Pi when it is
removed from solution (Porcella et^ ad,, 1970; Li et_ al., 1974; Fillos and
Swanson, 1975; Sagher j2t a_l., 1975; Cowen and Lee, 1976a; Golterman, 1976;
Moshiri and Crumpton, 1978; McCallister and Logan, 1978; Oloya and Logan,
1980; Dorich ejt al_., 1980).  The major obstacle to overcome in studying
the availability of sediment P to algae directly is the separation of P
associated with algae and that associated with sediment when sediment is
acting as the only source of P during incubation with algae.  Although
methods have been proposed earlier, the majority of studies which mark
advances in the determination of sediment P availability to algae have
occurred during the past decade, and their method of overcoming the ob-
stacle mentioned .above have varied from ignoring it completely to separa-
tion of algae and sediment by semipermiable membranes to correction techni-
ques.

     Gerloff and Skoog  (1954) proposed a method for addressing the problem
of nutrient availability to algae which circumvented the necessity for
incubating algae and sediment together.  The method consisted of  direct
analysis of algae removed  from  their native environment and relating the
cell content of a nutrient necessary for maximum  growth to  that found in
the organism.  In other words,  the  cell P  content was  found to increase
with external supply over  a wide range, but over  a  significant portion
of this range, growth remained  constant.   Therefore, levels of P  inside
the  cell in excess of the  critical  level necessary  for maximum growth re-
flected the abundance of the external  supply.   However,  the method has a
disadvantage.  The  cell  content reflects only  the conditions  under which
the  cell developed,  and  yields  no  information  concerning maximum protentially
available nutrient  levels, which is a  primary  concern.
 ^.A.  Dorich.   1978.   Algal availability of soluble and sediment phosphorus
  in drainage water of the Black Creek watershed.   M.S.  Thesis.   Purdue
  University,  West Lafayette, IN.   76 p.

-------
                                - 118  -

     Fitzgerald and Nelson (1966) evaluated a different procedure which
enabled conclusions to be drawn about nutrient status in an aquatic system
by analysis of the algae growing in it.  Fitzgerald (1966) found that the
amount of Pi removed from algae by a 60 minute boiling water extraction
was proportional to the level of Pi of the growth medium when Pi levels
limited algal growth.  Data indicated that when < 0.008 mg P/100 mg algae
was extracted by the 60 minute boiling water procedure, growth of algae
had been likely limited by P availability.  In the same study, alkaline
phosphatase activity was measured as a function of Pi level in the growth
medium.  Alkaline phosphatase is an extracellular enzyme which serves to
cleave Pi from molecules or substrates which do not yield Pi otherwise
(i.e., organic forms of P).  Its production by the cell is induced by low
levels of Sol Pi.  The activity of alkaline phosphatase was found to be
5 to 25 fold higher in algae whose growth was limited by Pi.

     Porcella et al.  (1970) conducted a long-term (164-209 days) microcosm
study in which sediments were placed in the bottom of plexiglass cylinders
and incubated with algae.  The sediments were analyzed prior to and follow-
ing incubation for P according to Jackson's (1958) extraction scheme, in-
cluding dilute acid-fluoride soluble phosphate.  Porcella et_ al. (1970) found
that between 60 and 80% of the dilute acid-fluoride extractable Pi in sediments
was lost from sediments during incubation indicating uptake by algae.  How-
ever, reagents used to extract Pi from sediments following incubation may
have removed Pi associated with biomass produced in the sediments during
incubation.  Therefore, sediment P at the end of the incubation period could
have been overestimated and the available fraction underestimated.

     Fitzgerald  (1970) attempted to estimate available P  (AP) by utilizing
a dialysis  tubing to contain the sediment during incubation with algae.
No algal response was  evident even when 2000 yg of sediment P was present.
Similar results were reported later by Golterman  (1976).  Conversely,
Wildung and  Schmidt  (1973) used a similar system in which algae and  sedi-
ment were incubated  separately in two glass half-cells with a membrane
filter between.  Algae assimilated 11  to  25% of the sediment Pi present.

      Sagher  et al. (1975) assessed the availability of sediment P in
Wisconsin lake sediments by growing P-deficient algae  (Selanastrum capri-
cornutum) in contact with  sediment over a 4 week period, and determining
the P  incorporated into  algal biomass  through  fractionation of  sediment Pi.
Correction  of Pi levels  in various fractions for Pi removed from algal cells
by extraction of the sediment:algal mixture was made by measuring Pi
extracted from a sediment-free cell culture.   Sagher e^t al.  (1975) con-
cluded that NaOH-extractable Pi  (i.e., Al- and Fe-bound Pi) was the  most
available for algal  growth over  a 28 day  incubation period, and 53 to  83%
of the Pi was AP.  Sagher  used  a similar procedure to determine AP  in
 2
 A.  Sagher.   1976.  Availability  of  soil  runoff  phosphorus  to  algae.
 Ph.D.  Thesis.   University  of Wisconsin,  Madison, WI.   162  p.

-------
                                -  119 -

the clay-size fraction of soil to S.  capricornutum over a 2 day incuba-
tion period.   The clay fraction and short incubation time were used be-
cause these particles are most subject to erosive processes, typically
contain a large proportion of the P transported by erosion, and are in
the euphotic zone of a lake for a sufficient period of time (24-48 hrs.)
to provide P to plankton.  The amount of AP in clay fractions in short-
term incubation closely agreed with levels of NaOH-extractable Pi.

     Huettl et_ al_. (1979) using Sagher's^ sediment samples showed that
an hydroxy-Al resin removed amounts of P similar to that shown to be
assimilated by algae in Sagher's^ sediment:algal incubation.  Specifically,
Huettl et al. (1979) demonstrated that resin removed an amount of P that
was, on the average, 98% of that assimilated by algae in water systems.

     Cowen and Lee (1976a) used a strict bioassay technique which resembled
the basic approach of the algal assay bottle test (Miller, 1978) to evalu-
ate urban runoff particulate P  (total P-total soluble P) availability to
algae.  The method involved the incubation of S. capricornutum with particu-
late material and direct counting of cells after 19-22 days.  Comparison
of cell numbers resulting from the incubation of algae with particulate P to
cell numbers resulting from incubation of algae grown in medium containing
known amounts of Pi resulted in their AP estimate.  The method makes the
assumption that algal growth in a strictly solution culture is similar to
that in sediment:media culture.  Cowen and Lee  (1976a) then compared
their AP estimate to levels of NaOH-, resin-, and HCl-extractable P found
in sediments prior to incubation, although analysis of incubated sediments
was not conducted.  Cowen and Lee (1976a) found that levels of NaOH- and
resin-extraetable  Pi agreed most closely with the quantity of AP (30%
of the particulate P).  Cowen and Lee (1976b) used a similar procedure to
estimate the algal availability of particulate P in the Genessee River
and several Lake Ontario tributaries.  The amounts of particulate P found
to be available to algae in the Genessee River samples by bioassay were
very similar to the amounts of resin-  and NaOH-extractable P.  The range
of particulate P which was available to algae was 1 to 24%  (ave. = 9%).
However, Lake Ontario tributary samples showed availabilities of < 6%.
Bioassay analyses of the tributary sediments were comfounded with inter-
ferences from the native microbial populations.  Autoclaved samples showed
bioassay-determined AP levels of 36  to 41% of TP.  The resin extraction
which was similar to the algal  available fraction in other  samples, re-
moved 6 to 31% of the particulate P.

     Fekete j£t a_l. (1976) utilized a direct bioassay technique with the
aquatic plant, Lemna minor, or common duckweed, to measure the available P
in sediments.  Lemna minor is a free floating, vascular plant typically
found in high numbers in shallow, protected areas of lakes high in nutrients.
The total number of fronds/plant, frond diameter, and dry weight consist-
ently reflected P concentration in solution.  Sediments were analyzed for
Bray Pi and TP, and bioassays were conducted on solutions of medium which
had been incubated over sediment for a one week period under both aerobic
and anaerobic conditions.  The incubation of solution over sediment occurred
three times and Lemna grown on each of the three solutions to evaluate the
available P level.  As would be expected, more Pi was released to the medium
from the sediment under anaerobic conditions than was released under aerobic
conditions.

-------
                                - 120 -


     Golterman (1976) measured available P in shallow lake sediments of the
Netherlands using the green alga, Scenedesmus  quadricauda.  Golterman's
(1976) method for separating algae and sediment consisted of dispersing
sediment in agar which was sliced into blocks and incubated with algae.
Algal growth rates were less in some cases with this method than with algae
in direct contact with sediment.  The amounts of P taken up by algae was
estimated by cell counts (1 mg phosphate-P= 10^ cells), and compared to
several extraction schemes.  In most cases, the use of strongly alkaline
or acid extractants as in that of Jackson (1958) proved unsatisfactory in
both replication and correlation with algal availability.  Golterman (1976)
suspected that milder extractants which would chelate the cations Fe+3 and
Ca+2 would be more appropriate.  After trials with NTA (nitrilotriacetic
acid), EDTA (ethylene diaminetetraacetic acid), and DTPA (diethylene-
triaminepentaacitic acid), NTA was found to satisfactorally separate out
the Fe- and Ca-bound Pi Golterman was striving for and best simulate the
quantities of Pi removed by algae.  Algae removed an amount of P which was
ca.99% of the amount of P removed with 0.01M NTA in 3 sequential extractions
Furthermore, Golterman (1976) found that an additional, nearly equivalent
amount of sediment P could be assimilated by algae when the agar:sediment
blocks which had been incubated with algae were removed and reinoculated
with algae a second time.

     Verhoff et al. (1978) studied the rate of P availability from suspended
river sediments by allowing the growth and natural succession of the in-
digenous microbial population in large capacity test vessels (12-14 A),over
a long period (9 months), and attempting a mass balance for P in the system.
Samples (1-1.5 £) were removed initially and periodically thereafter for
analysis of P and volatile solids.  It was assumed that all volatile solids
were algal and that the sediment associated P originally in the sample was
evenly distributed among the suspended solids.  Ve-rhoff et al. (1978) found
that the indigenous population was able to removed between 0.092 and 0.191
mg P/gm solids, and between .087 and .268% of TP/day.  This data is also
reported by Logan ^t_ jil. (1979).  However, Logan et^ al^.  (1979) went on to re-
port data on the chemical fractionation of the sediments prior to and
following incubation.  Logan et al. (1979) observed a decrease in the NaOH-
extractable fraction.

     Williams &t al_.  (1980) studied the availability of  Lake Ontario and Lake
Erie sediment P to the green alga, in  12 day incubations.  Available P in
sediment samples incubated with algae was estimated both by cell numbers,
as well as by decreases in a single HCl-extractable fraction, considered a
measure of  the sediment Pi.  Corrections for Pi extractable from algae by
the HC1 extraction of the sediment:algal mixture was accomplished by the
method of Sagher et_ al.  (1975) discussed previously.  Williams et al.  (1980)
went on to  correlate  the maximum  cell numbers  achieved  in  sediment:algal
incubations with the  quantities of TP, NTA-extractable  P, non-apatite P
(CDB + NaOH-extractable Pi), apatite P  (HCl-extractable),  and organic P
added.  Williams et. al.  (1980) reported that the relationship of cell numbers
and TP was  linear at  levels greater than 90 yg added TP  little algal growth
was observed in TP levels  less than 90 yg added P.  Pronounced linearity
was shown in the relationship between  cell counts  and nonapatite P with
concentrations greater  than  25 yg added P, and little growth shown below
25 yg added P.   Similar  results were shown for plots of  cell numbers vs.
NaOH-, resin-, and NTA-extractable P.  Williams ejt al.  (1980) concluded from

-------
                                 -  121  -


 this  data  that  apatite P  was  not utilized by algae,  while only a portion
 of  the non-apatite P  was  assimilated by  algae.   Available P data arrived
 at  by following the decrease  in the single HCl  extractable P level and
 comparison to initial levels  of Pi,  non-apatite Pi,  and NaOH-extractable
 Pi  indicated an average uptake of  ca.  75% (38 to 83%)  of the amount of
 non-apatite Pi,  nearly all  of the  amount of NaOH-extractable Pi (which was
 ca. 69% of the  non-apatite  Pi) and 8 to  50% of  the TP.

      Dorich et:  al.  (1980) studied  the  AP levels in suspended stream sedi-
 ments in runoff water of  an agricultural watershed (the Black Creek water-
 shed)  in the Maumee River basin in much  the same manner as Sagher et al.
 (1975).  The study of Dorich  ^t a^.  (1980) differed  from that of Sagher"
 et  al.  (1975) in the  length of incubation (2 rather  than 4 weeks) and in
 the extraction  procedure  used to fractionate P  in sediment:algal incubations
 (sequential NityF,  NaOH, and HCl rather than sequential  NaOH and HCl).
 Dorich et  al. (1980)  found  that the two  week incubation resulted in the
 assimilation of all the AP  and an  additional 2  weeks of incubation resulted
 in  increases -in Sol P and sediment P.  Dorich et ad.  (1980) found that the
 majority of the AP (30% of  Pi and  21%  of TP) originated in the NH4F-
 extractable fraction  (43%)  while NaOH- and HCl-extractable fractions
 accounted  for less (37 and  20%,  respectively).   Ammonium fluoride-,  NaOH-,
 and HCl-extractable P fractions contributed 60,  27,  and 13%,  respectively,
 of  their amounts present  initially to  the pool  of available P.   Dorich's
 ejt  al.  (1980) finding that  HCl-extractable P contributed significantly to
 the AP pool was  in conflict with most  results of other  studies (Sagher eit
 al.,  1975;  Sagher1; Logan et  al.,  1979;  Williams et_  al. , 1980),  but  other"
 results  reported show HCl-extractable  Pi availability  (Sagher1;  Logan et al.,
 1979).

      Studies have  been reported in the literature in which an "available"
 fraction is evaluated based upon a simple chemical extraction.   However,
 caution  must be  exercised in  the interpretation  of such studies.   For
 instance,  Wentz  and Lee (1969a)  presented a procedure in which a dilute
 HC1-H2S04  extractant  was used to estimate "available P".   The dilute
 HC1-H2S04  extraction  was recommended due to the  fact that P sorption is
 maximized  at neutral  to slightly acidic  conditions,  and minimized at low
 and high pH's.   Therefore,  the acid  extractant would supposedly release
 sorbed Pi,  which is available to algae.   As shown above in later studies,
 neutral  NlfyF- and/or  NaOH-extractable  Pi provides the largest portion of the
 AP  (Sagher £t al.,  1975; Sagher1;  Logan  iet  al. ,  1979; Williams  et_ al.,  1980;
 Dorich ^t.  a_l. ,  1980).   On the other  hand,  the HC1-H2S04  extractant would
 remove most of the  Pi extractable  with the  NH4F  and  NaOH,  as  well as most
 of  the Pi  associated  with Ca"1"2, which is considered to be  largely  unavailable.
 Therefore,   the "available"  P  as  outlined by Wentz and Lee (1969a)  would
 overestimate available P,  and,  drastically  so, in calcareous  sediments.   In
 short, unless a  fraction of Pi  (i.e.,  NaOH-extractable)  has been shown  to
 provide Pi  to algae in a bioassay  study  such as  Sagher _et  al.  (1975),
 Logan et al. (1979) or Dorich  et al. (1980)  it is  questionable  to  use  the
method as a measure of AP.  Wentz  and  Lee  (1969b)  evaluated the  depositional
history of "available P" (estimated by their  dilute HC1-H2S04 extraction)  in
 '.ake Mendota,  a eutrophlc, calcareous lake in Madison, Wisconsin.   Wentz and
Lee (1969b) found 50%  of the TP to be  extractable with HC1-H2S04.

     Since a number of studies  (Sagher  et al., 1975; Sagher2; Golterman, 1976;

-------
                                - 122 -


Williams ej; al., 1980; Dorich et_ ad.,  1980) have made direct measurements
of decreases in specific fractions of sediment P as a result of algal up-
take, the chemically extractable fractions found to supply P to the AP
pool have been taken to be an estimate of the AP.  Allan and Williams (1978)
studied the historical levels of various forms of sediment P present. Allan
and Williams (1978) determined CDB-extractable P as the bioavailable
fraction and cite that the concentration would be slightly greater than
that determined by direct bioassay as determined by Golterman (1976).  Sur-
prisingly enough, the levels of bioavailable or CDB-extractable P in the
presettlement era of some lakes was higher than that occurring in present-
day, culturally eutrophied sediments in Lake Erie.

     Logan et al. (1979) in his P extraction studies with various suspended
stream sediments evaluated sequential NaOH-, CDB-, and HCl-extractable P.
Logan et al. (1979) cited the NaOH-extractable fraction as an estimate of
short-term available fraction and the NaOH + CDB-extractable fraction as
the long-term or total, potentially available fraction.  Logan et^ al_. (1979)
found NaOH-extractable P to range from 14 to 42% and NaOH + CDB-extractable
from 42 to 89% of the TP.

     Armstrong et al. (1979) evaluated AP in several rivers with access to
the Great Lakes.  Two chemical extraction methods, 0.1N NaOH and anion
exchange resin desorption, were used to estimate maximum algal AP and readily
available P, respectively.  Sodium hydroxide extractable P ranged from 14
to 37% of TP and resin extractable from 7 to 17% of TP.  Maximum AP in the
clay fraction, which may remain in suspension indefinitely, ranged from 16 to
53% of the TP.  He continues by stating that about 50% of the U.S. tributary
loadings of P to the Great Lakes is in the available form (50% of which is
particulate and 50% dissolved).

     The effects of agricultural activities  along  the  Maumee  River upon  the
pollution of Lake Erie relative to  P availability  has  been  a  point of conten-
tion.  However,  the effect of  agriculture must  be  studied in  a watershed
which  is unique  in its domination by such  activities.   The  opportunity to
study  availability of P  in a  strictly agricultural watershed  presents itself
in the Black  Creek watershed.  The  amount  of work  related to  the  direct
measurements  of  AP in suspended river sediments is minimal  (Logan et al.,
1979;  Armstrong jet ail. ,  1979;  Dorich e*  al., 1980),  and those making direct
estimates of  the availability  of  sediment  P  in  runoff  from  strictly  agricul-
tural  watersheds is  even less  (Dorich e£ al.,  1980).   Furthermore, compari-
sons of various  methods  of AP  determinations have  been made in lake  sediments
 (Williams £t  al^.,  1980),  but  not  in suspended  stream sediments.   In  view of
these  facts,  the specific objectives of  this study will be  concerned with
determining in samples  from several sites  within the Black  Creek Watershed:
 (1)  The algal  availability  of  P associated with suspended  sediments  as
determined  in both long-term (2 weeks)  and short-term (2 days) sediment:algal
incubations in studies  similar to  Sagher et al. (1975), and Dorich et al.,
 (1980).   (2)  The proportions  of Pi  and  TP  which are  available for algal  up-
 take as determined in long- and  short-term sediment:algal  incubations  and
 (3)  The relationships between the quantity of  AP and resin-,  NTA-, and
sequential  NH^F-,  NaOH-,  and HCl-extractable P  levels.

-------
                                - 123 -


                         MATERIALS AND METHODS
Basic Experimental Design

     The objectives of this study were (1) to determine algal availability
of suspended stream sediment P in drainage water of the Black Creek water-
shed in both 2 week and 2 day bioassays (2) to determine the proportions
of Pi and TP in suspended stream sediments which are available to algae
in 2 week and 2 day bioassays and (3) to determine the relationships be-
tween AP and resin-, NTA-, and sequential NH^F-, NaOH, and HCl-extractable
Pi in suspended sediments in drainage water of the Black Creek watershed.

     Availability of sediment P has been determined in all samples by
direct measurement of P fractions in sterile sediments prior to and follow-
ing incubation with S. capricornuturn.  The difference in the quantity of
P in various Pi fractions initially and at the end of the incubation period
is assumed to have been assimilated by S. capricornutum.  Availability of
sediment P to algae has been determined in both 2 week and 2 day incubation
periods similar to those systems described first by Sagher et al. (1975).
The value of AP determined in incubation of sterile sediment with algae
has been compared to levels of resin-, NTA-, and sequential NlfyF-, NaOH-,
and HCl-extractable P in sterile sediments which have not undergone incuba-
tion with algae.

Sediment Collection and Treatment

     Suspended sediments were collected as water grab samples in 2.5 liter
sterile glass containers at the peak of the hydrograph immediately following
rainfall events on 4/14/80, 6/2/80, 7/22/80, and 8/20/80.  Sampling included
7 sites within the Black Creek watershed (Figure 1).  Sites 2, 3, and 4
are primarily drainage from cropland while sites 5 and 6 are affected by
sewage from the town of Harlan, IN.  Sites 12 and 14 represent the Black
Creek and Maumee River, respectively.  Samples were returned to Purdue
University and stored at 4°C until processed.

     Because the concentration of suspended sediments is not normally high
enough to conduct bioassays, it was necessary to concentrate the sediments.
Suspended sediments were concentrated by slow rate continuous flow centri-
fugation (9,000 x g_), and diluted to between 100 and 500 ml depending upon
the relative concentration of sediment in the water sample.  The concen-
trated sediment samples were then sterilized by 3 megarads of y radiation
(6°Co source, ca. 7200 rads/minute, and an exposure time of ca. 8 hrs.).
Preliminary studies found this exposure was adequate to ensure sterilization.
Following sterilization, concentrated sediment samples were stored at 4°C
until used in bioassay measurements.

Stock Culture

     A stock culture of Selanastrum capricornutum (a single-celled member
of the Chlorophyceae  family) was acquired from the U.S. Environmental
Protection Agency, Pacific Northwest Water Laboratory, Corvallis, Oregon.
Algal cells were cultured in 200 ml of synthetic nutrient medium (PAAP)

-------
                                - 12 4 -


(Miller, 1978) in 1000 ml Erlenmeyer Flasks at 26 ± 1°C with fluorescent
light intensity of ca. 5500 lux for 2-3 weeks.  The pH of the culture was
adjusted periodically with HC1 to pH 6.8.

P-deficient Inoculum for 2 Week Sediment; Algal Bioassays

     When cultures of S. capricornutum achieved maximum cell densities, the
cells were havested by centrifugation, rinsed in P-free PAAP and resuspended
in P-free PAAP.  The cells were then incubated for 3-5 weeks (Sagher, 1975).
Before used as an inoculum for the sediment:algal (SA) incubation, the cells
were again rinsed in P-free PAAP medium to remove Pi which may have been
released from senesced cells.  Cells were then counted.

P-deficient Inoculum fror 2 Day Incubations

     When cultures of S. capricornutum reached maximum cell densities, cells
were harvested by centrifugation, rinsed in PAAP, and resuspended in 600 ml
of PAAP in a 2 liter Erlenmeyer flask and incubated until cell densities were
sufficient for use as an inoculum.  This procedure achieved the high cell
densities that were necessary for the massive inoculum required for the 2
day incubation experiments.  Cells were again havested by centrifugation,
rinsed in P-free PAAP, and resuspended in 600 ml of P-free PAAP in a 2000 ml
Erlenmeyer flask.  The cells were incubated for 3-5 weeks (Sagher, 1975).
Before used as an inoculum for SA incubations, cells were agin rinsed in
P-free PAAP, and counted.

Bioassay Conditions

     Sediment:algal   incubations were conducted to evaluate the availability
of sediment P  to algae.  All incubations were conducted  in 50 ml of P-free
PAAP to provide all essential nutrients but P, an aliquot of sterile suspended
sediments containing  35-45 ]Jg total sediment P, diluted  to 60 ml with de-
ionized water  in a 250 ml Erlenmeyer Flask, and stoppered with a cotton plug.
Each flask was inoculated with P-deficient  S. capricprnutum to arrive at an
initial cell  density  in  the bioassay flask  of 5 x  104  cells/ml for the 2
week SA bioassays and 2  x 10^ cells/ml for  the  2 day  SA  bioassays.

General Information Concerning Extraction and Analyses

      The  following  information  pertains  to  the  extractions  and  analyses  per-
 formed  on all inoculated and uninoculated  sediment samples.   The  analysis
 of  inoculated sediment  or  SA  cultures  initially  and after the incubation
 period  served as  a  direct measurement of the decrease in sediment P  as a
 result  of uptake by algae (Sagher jet  al.,  1975;  Sagher2; Dorich jet al.,  1980).
 Sediment-free algal incubation flask  contents were extracted  in order to
 determine extractability of P in algal cells.   This data was  used to correct
 the amount of P extracted from sediment:algal mixtures to arrive at  the
 actual  amount of P  extracted from sediment.  Uninoculated sediments  were
 extracted with various reagents reputed to remove amounts of  P from sedi-
 ments similar to that used by algae.   These values were compared to  avail-
 able P  measured directly in inoculated sediments.

      All  sediment P extractions  were  carried out  in tared 50  ml polypropylene
 centrifuge tubes.   Following extraction  (shaking  on reciprocating shaker),
 centrifugation (12000 rpm at 9,000 x  £ for  ca.  20  minutes), and decanting

-------
                                - 125 -
by suction, the amounts of liquid carryover to the next extraction was
determined gravimetrically.  The quantity of P determined in the sub-
sequent extraction was corrected accordingly.

     Solution:sediment ratio (v/w) were maintained as near to 500 as
possible.  Therefore, the volume of extractant varied among samples, and
depended on the weight of sediment in the tube.  The sediment pellet in
the tube and extractant was shaken by hand prior to being placed on the
shaker to ensure complete dispersion of the sample in the extractant.

     All colormetric Pi determinations were conducted in 25 ml volumetric
flasks.  The phosphomolybdate color was developed according to Murphy and
Riley  (1962) for all extracts and digests following neutralization with
HC1 or NaOH with p-nitrophenol as the indicator with the exception of the
NH^F extract.  Aliquots (10 ml) of the NH^F extracts were treated with 7.5
ml of H3B03(50 g/&) and the pH adjusted with HCl using 2,4-dinitrophenol
as the indicator.  Phosphomolybdate color was developed with SnCl2 as the
reductant.  The determination of Pi in NIfyF extracts was as recommended
by Jackson (1958).

     When the decision has to be made whether to use one large single
aliquot or two smaller replicate aliquots for Pi determination in the avail-
able extract of a replicate flask contents, the decision was made in favor
of the larger aliquot for a more precise single measurement.  Therefore,
the Pi levels reported were a result of the averaging of the single measure-
ments made in each of 3 replicate flasks.  Color intensity was measured as
absorbance with a Beckman model 24 visible spectrophotometer equipped with
a automatic filling 1 cm cell at 850 nm.

     Residual P following Pi fractionation was determined following sequen-
tial HN03 and HC104 digestion (Sommers and Nelson, 1972) in 50 ml calibrated
Folin-Wu digestion tube.

Extraction and Analysis of Sediment:Algal Bioassay Samples

     Sediment:algal bioassay solutions and sediment-free algal solutions
were sequentially analyzed for soluble, NaOH-, and HCl-extractable, and
residual P initially and at the end of the incubation period (i.e., 2 weeks
and 2 days).

     Fractions of sediment Pi were determined initially and at the end of
the incubation period colormetrically following sequential extraction with
NaOH and HCl, and digestion of the residue.  In detail, the contents of the
SA incubation flasks was transferred into a 50 ml polypropylene centrifuge
tube and centrifuged.  The solution phase was decanted and analyzed for
Sol Pi.  To the sediment in the centrifuge tube the appropriate volume (to
give an extraction ratio of 500:1) of O.lN^ NaOH was added and the solution
shaken for 17 hrs.  Following extraction with NaOH, the sample was centri-
fuged, the extract decanted, and Pi determined in an aliquot.  To the
sediment remaining in the centrifuge tube, the appropriate volume of 1 N_
HCl was added and shaken for 1 hr.  Following extraction the solution was
centrifuged, the extract decanted and Pi determined in an aliquot.  The
residue in the centrifuge tube was transferred to a Folin-Wu digestion tube
and the Pi determined colormetrically following digestion.

-------
                                - 126 -


Extraction and Analyses of Uninoculated Sediment Samples

     Sediment samples which had not been incubated with algae, but treated
identically otherwise, were subjected to various Pi extractants and the
quantity of Pi in the extracts determined for comparison to quantities of
AP.  Sequential NltyF-, NaOH-, and HCl-extractable Pi was determined as
outlined by Dorich £t_ al_. (1980).  An amount of the sterilized, concentrated
sediment containing between 35 and 45 ug TP was added to the centrifuge tube,
the solution centrifuged, the liquid decanted and Sol Pi determined.  To
the sediment remaining in the centrifuge tube, an appropriate volume of
neutral 0.5K[ NItyF was added and the contents shaken for 1 hr.  Following
extraction, the tube was centrifuged, the extract decanted, and Pi determined
in an aliquot of the extract as indicated earlier, according to Jackson
(1958).  To sediment in the centrifuge tube, the proper volume of O.lN NaOH was
added and the tube shaken for 17 hrs.  Following centrifugation and decanting
of extract, Pi was determined in the extract.  The appropriate volume of
1 _N HC1 was added to the sediment pellet in the centrifuge tube and shaken
for 1 hr.  Following centrifugation and decanting of extract, Pi was deter-
mined in the extract.  Residual P was determined following HN03 and HC104
digestion as discussed previously.

     Sediments were also subjected to Goltermanls  (1976) sequential 0.01 M
NTA extraction.  Golterman  (1976) found that 3 successive 20 ml extractions
with 0.01 M NTA (pH 7) removed amounts of sediment Pi similar to that re-
moved by algae.  An aliquot of concentrated sediment solution containing
between 35 and 45 yg TP was added to a centrifuge  tube.  The  tube was centri-
fuged, the supernatant decanted, and an aliquot of the supernatant analyzed
for Sol Pi.  Twenty ml of 0.01 M NTA was added to  the centrifuged sediment
and the tube shaken for 2 hrs.  Following centrifugation supernatant was
decanted.  Two additional such extractions were performed on  the same sample
and the 3 supernatants combined.  The combined extracts were  titrated to
pH 1.5 with HC1 to precipitate organic matter and  NTA and  the volume of
titrant recorded  (Nnadi  and Tabatabai, 1975).  The  titrated  extract was
allowed to stand overnight  and then vacuum filtered  through a 0.45 u
Nucleopore membrane.   Inorganic P was determined  in  an aliquot colormetrically
according to the method of  Murphy and Riley  (1962)   with a 2.5 hr. color
development rather than 30  min., although others  have indicated NTA inter-
feres with Murphy and  Riley (1962) color development (Nnadi   and Tabatabai,
1962; Golterman,  1976; Williams,  1980).  Unpublished studies  conducted  in
our laboratory show  that although color development  according to Murphy
and Riley  (1962)  is not  complete after the recommended 30 minutes, full and
linear color development is attained after 2.5 hrs.  with up  to 10 ml of
0.01 M NTA present.   After  the supernatant from the  third  extraction has
been decanted, the sediment was  transferred  to a  digestion tube and
digested with HN03 and HC104.  Phosphorus was determined in  the digest  as
discussed  previously.

     Non-incubated sediments were also subjected  to  the  resin extraction
procedure  of Huettl  et^ al^.  (1979).   Resin preparation was  according  to  that
specified  by  Corey  (Professor  of Agronomy, University of Wisconsin,  personal
communication,  1979),  and will be discussed  below.   The  resin used was
Dowex  50W-X2,20-50 mesh  (wet)  and wet  sieved to  remove particles  smaller  than
40 mesh  prior  to  beginning  chemical  preparation.   Following sieving,  %
moisture was  determined  on  a sample  of  resin (ca. 5  g)  to  obtain  an accurate

-------
                                - 127 -
dry weight.  Cation exchange capacity of the resin was determined by satur-
ation of a 10 g (wet) sample with acid (stirring for 15 minutes with 1 N^
HC1).  The resin placed in a vacuum filter holder and washed several times
with deionized water.  The resin was then rinsed several times with separate
portions of a 25 ml aliquot of 0.5 IJ KC1.  The acidity of two 10 ml aliquots
of the KC1 eluent was titrated with standardized NaOH (ca. 0.5 lp with
phenophthalein as the indicator.  Once the CEC of the resin had been calcu-
lated based upon the N_ and volume of the standardized NaOH titrant, pre-
paration of the bulk of the resin was initiated.

     In a column the resin was leached with dilute AlCl3*5H20 (1 N_ Al or
more dilute).  Since the purpose of the AlCl-j'SH^O leaching was to ensure
saturation of the resin with Al, excess- was applied.  The excess Al was
removed by copious leaching with several bed volumes of deionized water.
     The Al-saturated resin was transferred to a 4 £ beaker where a
solution was slowly added (over several hours) with stirring to allow for
diffusion into the beads.  The total NaHC03 added was sufficient to
neutraluze 2/3 of the original sites, and stirring was not ceased until C02
evolution had stopped.

     The resin was transferred back to the column, where it was leached with
enough 0.5 ^ A1C13-5H20 to saturate the entire CEC with Al.  The resin was
then rinsed with several bed volumes of deionized water.  This concluded
chemical preparation of the resin.

     Concentrated sediment samples containing between 35 and 45 yg TP were
placed in centrifuge tubes, and the suspension centrifuged.  Following de-
cantation of the supernatant Sol Pi was determined.  A quantity of resin con-
taining a total CEC of 2.5 meq and 20 ml of 0.001 M CaCl2 was added to the
sediment in the tube and the mixture shaken for 24 hrs.  Following the 24-
hour extraction period, the contents of the centrifuge tube were wet sieved
through a 60 mesh sieve to separate the resin and sediment.  Sediment passing
through the sieve was trapped in a 50 ml digestion tube.  Followed by evapor-
ation and HN03~HC104 digestion, the amount of P remaining in the sediment
was determined.  The resin trapped on the sieve was transferred to a
vacuum filter apparatus where the extraction of Pi from the resin took place.
The resin in the filter apparatus was rinsed several times with separate
portions of a 50 ml aliquot of 0.2 N^ HC1 and entire rinse volume recovered.
Inorganic P was determined in an aliquot of the rinse following neutraliza-
tion.

Calculations

     Corrections for P removed from algae were made upon the amount of P
extracted from incubated sediment: algal mixtures in the following manner.
The proportion of Pi extracted by NaOH and HC1 from sediment-free algal
cultures was determined based upon known levels of P in cells.  The amount
of P in cells in sediment: algal mixtures or available sediment P was calcu-
lated based upon the difference in fractions initially and after incubation
period (2 days or 2 weeks) and proportions removed from cells in the sediment-
free algal cultures.

-------
                                - 128 -


                        RESULTS AND DISCUSSION
     Sediments used in incubations with algae were sequentially extracted
with NaOH and HC1 initially and after the incubation period (2 weeks and
2 days). Decreases in inorganic P fractions over the incubation period
served as the basis for the determination of the quantities of sediment
P removed by algae during incubation.  Initial levels of P in various
fractions of stream sediment P are presented in Table 1.  As indicated in
Table 1 significant quantities of inorganic P (Pi) are desorbed when the
sediment is added to the P-free PAAP medium.  As a % of total P (TP) in
the system, averages of 9 to 14% were desorbed.   The average concentration
of NaOH-extractable P for the four sampling times ranged between 295.5
and 390.5 yg/g (avg. = 326.1 yg/g), or as a % of TP between 26 and 34%
(avg. = 30%).  Average HCl-extractable P levels ranged between 154.9 and
284.7 yg/g (avg.  = 206.2 yg/g) or as a % of total P between 17 and 22%
(avg. = 18%).  Average Pi levels (the sum of soluble Pi, NaOH-, and HCl-
extractable P) ranged between 554.7 and 827.1 yg/g (avg. = 671.3 yg/g),
or as a % of total P, between 58 and 65% (avg. = 61%).  The significance
of the various Pi fractions and total Pi in sediment relative to algal
available Pi has been demonstrated in the past (Sagher, 1975; Safher3;
Dorich £t al., 1980).  Both Sagher (1975) working with lake sediment and
Dorich et_ al_. (1980) working with Black Creek watershed stream sediments
have found that only Pi is available to algae, and that the majority of
the available Pi originates in the NaOH-extractable fraction, which is
defined as P sorbed on the surfaces of hydrous Fe and Al oxides (Syers
et^ al. , 1973).  Dorich e_t al. (1980) also demonstrated contributions of the
HCl-extractable fractions to the pool of available P.

     Following correction for P extracted by NaOH and HC1 from algal cells
which populated the sediment:algal mixtures after the incubation period,
available P was calculated based upon differences in levels of NaOH- and HCl-
extractable P initially and after 2 days or 2 weeks of incubation.  Available P
in stream sediments is presented in Table 2.  The average concentration
of available P in stream sediments for the 4 sampling dates ranged between
190.3 and 349.2 yg/g (avg. = 289.8 yg/g) and between 278.8 and 430.2 yg/g
(avg. = 333.4 yg/g) as measured in 2 day and 2 week incubation periods,
respectively.  As a % of Pi, these values represent a range of 33.9 to
45.7% (avg. = 40.8%) and 48.3 to 51.5% (avg. = 50.4%) as measured in 2 day
and 2 week incubation periods, respectively.  As a % of TP, average avail-
able P concentration, in stream sediments ranged between 19.7 and 27.5%
(avg. = 24.6%) and between 27 and 34.4%  (avg. = 30.3%) as measured in 2
day and 2 week incubation periods, respectively.  As indicated in Table 2,
algae consistently assimilated more P from sediment when the incubation
period was increased from 2 days to 2 weeks.  However, in most cases, as
the averages bear out, the vast majority of available sediment P was
assimilated within the first  2 days of incubation.  The average additional
sediment P which was removed by algae in the period between 2 days and 2
weeks ranged  from 5.1 to  16.8%  (avg. = 9.6%) and  1.0 to 11.6%  (avg. = 5.7)
of the  Pi  and TP, respectively.  If  the  assumption is made that most of
the P-bearing particles  (small diameter) remain in the photic  zone of a
lake  for a period of 2 days where algal  assimilation is maximum  (Sagher,
1976),  then  it appears that  sediments transported from  the Black  Creek water-
shed  and deposited  in Lake Erie are  capable of releasing the majority of

-------
                                -  129 -

 Table  1.  Quantities of  P  present  in various  fractions  in  stream sediment
          used  in  2 day  and  2 week incubations.

Date


4/14







6/2







7/22







8/20








Site No.


2
3
4
5
6
12
14
Avg.
2
3
4
5
6
12
14
Avg.
2
3
4
5
6
12
14
Avg.
2
3
4
5
6
12
14
Avg.
I
SIP


109.7
150.4
101.1
106.3
108.3
150.9
141.9
124.1
150.7
60.1
153.2
189.0
222.5
186.1
131.1
156.1
69.8
85.3
77.6
82.2
101.7
68.4
113.0
85.4
138.7
137.8
120.0
185.2
176.3
160.5
144.4
151.8
!nitial se
NaOH


396.7
371.9
277.9
363.3
353.2
258.6
306.3
332.6
265.0
131.2
336.6
326.4
476.5
288.3
244.6
295.5
285.0
317.9
349.7
273.2
392.7
284.8
297.4
314.4
386.5
389.7
245.7
377.0
559.3
346.5
428.6
390.5
idiment
HC1


271.0
175.4
113.6
159.9
137.9
241.0
195.4
184.9
191.1
84.7
127.0
309.0
191.5
275.9
222.1
200.2
145.2
154. ,6
110.4
121.2
149.4
161.3
241.9
154.9
236.5
256.5
314.6
311.3
265.5
286.0
322.6
284.7
P found in
Inorg
1 1 cr I fr _ _
Mg/g —
777.4
697.7
492.6
629.5
599.4
650.5
643.6
646.6
606.8
276.0
616.8
824.4
890.5
750.3
597.8
651.8
500.0
557.8
537.7
476.6
643.8
514.5
652. .3
554.7
761.7
784.0
680.8
873.5
1001.1
793.0
895.6
827.1
fractions
Org


586.1
476.8
308.6
475.7
418.7
521.9
423.8
458.8
535.5
236.4
476.2
521.1
589.8
515.8
414.7
469.9
393.7
365.5
366.0
357.6
374.5
393.7
323.7
367.8
462.6
514.9
404.5
449.4
419.3
432.0
458.6
448.8
.* .
Total


1363.5
1174.5
801.2
1105.2
1018.1
1172.4
1067.4
1100.3
1142.3
512.4
1093.0
1345.5
1480.3
1266.1
1012.5
1121.8
893.7
923.3
903.7
834.2
1018.3
908.2
976.0
922.5
1224.3
1298.9
1085.3
1322.9
1420.4
1225.0
1354.2
1275.9
«
 Values shown here are overall averages found initially in 2 day and 2
 week incubations.

-------
                                - 130 -
Table 2.  Levels of available P in stream sediments as measured by bioassay
          (2 day and 2 week incubations).
Bioassay incubation period
Date

4/14







6/2







7/22







8/20







Site No.

2
3
4
5
6
12
14
Avg.
2
3
4
5
6
12
14
Avg.
2
3
4
5
6
12
14
Avg.
2
3
4
5
6
12
14
Avs.
2
Mg/g
740.3
345.1
282.7
250.3
263.0
212.4
265.8
337.1
222.7
129.7
279.0
367.9
496.5
296.9
184.4
282.4
111.7
215.9
134.0
200.0
216.4
175.5
278.3
190.3
305.6
373.9
266.4
307.5
471.1
345.3
374.9
349.2
day available P
% of Pi
70.6
50.5
43.5
38.6
39.8
37.3
39.5
45.7
35.7
46.6
44.1
43.6
51.9
37.3
31.5
41.5
22.2
38.3
25.0
42.3
33.2
34.7
41.3
33.9
38.9
44.9
38.9
39.0
46.6
42.8
41.6
41.9
% of TP
43.1
30.0
27.3
23.
25.0
21.1
23.1
27.5
19.8
25.7
25.4
28.1
31.8
23.5
18.7
24.7
12.0
21.8
14.2
23.1
20.6
19.6
27.2
19.7
23.8
27.7
24.0
23.8
31.6
27.3
26.9
26.4
2 week available P
yg/g
284.3
411.9
162.7
362.2
373.0
337.6
255.0
312.4
263.6
112.2
308.2
407.6
495.5
318.4
278.6
312.0
253.5
284.0
286.9
252.7
342.1
257.5
275.0
278.8
365.8
356.1
304.9
551.3
578.5
398.4
456.3
430.2
% of Pi
51.1
60.3
48.4
59.4
50.7
46.1
41.5
51.1
44.8
41.0
51.2
50.0
60.1
45.2
45.6
48.3
51.0
51.5
53.2
52.6
53.8
49.3
43.6
50.7
49.6
48.4
45.0
57.6
57.9
51.1
51.2
51.5
% of TP
26.0
34.4
28.8
32.3
27.8
25.2
25.2
28.5
22.7
21.5
28.3
29.0
35.4
25.1
26.8
27.0
29.6
32.6
32.9
31.4
34.7
28.4
29.6
31.3
31.5
28.6
28.8
40.8
42.5
33.6
34.7
34.4

-------
                                - 131 -


their available P within that period of time if sufficient algal popula-
tions are present.  The maximum differences in available P between sampling
dates as a % of Pi and TP, are 11.8 and 3.2%, and 7.8 and 7.4% as measured
in 2 day and 2 week incubations, respectively.

     Several chemical extraction methods have been proposed to remove
quantities of P from sediments similar to that assimilated by algae.  Of
these suggested extractants which have been correlated with available P,
Sagher's (1975, 1976) NaOH-extraction, Dorich's £t ail. (1980) sequential
OTtyF, NaOH, and HC1 extraction, Huettl's jst_ al. (1979) resin extraction,
and Golterman's (1976) NTA extraction appear to be most promising.  The
quantities of P removed by NaOH from stream sediments for the four sampling
dates are presented in Table 1 while that extracted by the sequential
NH4F, NaOH, and HC1 extractions, resin, and NTA are presented in Table 3.
As a fair basis for comparison in this description the sum of soluble Pi
+ NaOH-extractable P concentrations will be used because (1) essentially all
the Sol P in this experimental system is assimilated by algae in addition
to the majority of NaOH-extractable P .(Sagher, 1975; Sagher, 1976; Dorich
£t al_., 1980), and (2) the Pi found as soluble Pi initially in this experi-
mental system would be detected as NaOH-extractable P in routine extraction
procedures.  The average NaOH-P + soluble Pi (heretofore referred to as
NaOH-extractable P) ranged between 399.8 and 542.3 yg/g (avg. = 455.5 yg/g).
These concentrations represent a range between 61 and 72% (avg. = 67%) and
between 39 and 43% (avg. = 41%) of the sediment Pi and TP, respectively.
Average Pi levels for the 4 sampling periods determined as the sum of solu-
able Pi, NaOH- and HCl-extractable P ranged between 58 and 65% (avg. = 61%) of
sediment TP.  Sodium hydroxide-extractable P appears to slightly over-
estimate the actual quantity of available P as determined in both 2 day and
2 week incubations of the sediments sampled.  The overestimation of avail-
able P by the NaOH-extractable fraction makes sense, because not all of this
P should be readily available to algae.

     Presented in Table 3 are the concentrations of Pi extractable with
sequential OTfyF, NaOH, and HC1, resin, and NTA.  For the same reasons as
indicated for comparisons of NaOH-extractable to available P, levels of
NH4F-extractable P discussed will include the amounts of soluble Pi detected.
Average NH4F-extractable P for the 4 sampling dates ranged between 205.2
and 409.0 yg/g Cavg. = 281.1 yg/g).  As a % of inorganic P and TP, NfyF-
extractable Pconstituted between 33.8 and 42.0% (avg. 37%), and 18.8 and 27.6%
(avg. = 22.9%), respectively.  On a quantitative basis NH4F appears to
simulate the 2 day available P level in stream sediments which averaged 40%
of Pi and 24.6% of TP, and underestimate the 2 week available P which
averaged 50.4% of Pi and 30.3% of TP.  The average sums of the Pi extract-
able with NltyF- and the subsequent NaOH reagent ranged between 309.7 and
498.2 yg/g ( avg. = 386.5 yg/g) which constituted between 48.3 and 54.7%
(avg. =52.1%) and between 27.9 and 36.6% (avg. = 31.8%) of the Pi and TP,
respectively.  The sum of NIfyF- and NaOH-extractable P may be considered as
similar to available P levels (50.4% and 30.3% of Pi and TP, respectively)
measured in 2 week incubations of the stream sediments.

     Resin extractable P (the sum of soluble Pi + resin-extractable P)
averages for the 4 sampling days ranged between 145.7 to 316.8 yg/g (avg. =
228.3 yg/g).   As an average this underestimates both 2 day and 2 week avail-
able P.   However, it should be noted that there does not appear to be a

-------
                                - 132 -

Table 3.  Quantities of P in stream sediments extracted with sequential
          NH,,  NaOH and HC1; resin, and NTA.


Date


4/14







6/2







7/22







8/20









Site No.


2
3
4
5
6
12
14
Avg.
2
3
4
5
6
12
14
Avg.
2
3
4
5
6
12
14
Avg.
2
3
4
5
6
12
14
Avg.

i
NH.F+
4


170.8
206.8
200.9
200.5
196.8
216.2
244.3
205.2
258.8
105.0
260.8
305.2
398.6
253.3
167.9
249.9
217.2
238.9
220.0
236.4
351.9
225.7
331.3
260.2
381.6
378.6
261.1
430.5
566.7
388.0
456.2
409.0

Sequential
NaOH


89.6
97.2
223.8
171.3
111.3
44.8
46.0
112.0
53.8
26.9
92.5
62.9
123.7
32.0
27.0
59.8
143.1
160.1
236.3
167.5
216.1
143.2
57.7
160.6
178.6
93.7
80.5
65.0
62.4
84.2
46.1
87.2
Extractai
1
HC1
.
yg/g -
280.7
275.6
233.2
306.3
184.0
353.4
316.0
278.5
298.3
131.4
216.0
411.8
284.9
296.2
359.2
285.4
264.1
354.8
277.1
262.5
324.6
362.1
490.8
333.7
571.7
426.8
683.9
456.7
440.6
662.8
513.5
536.6
it

Resin+


101.2
152.7
113.5
188.0
319.1
197.9
322.2
199.2
547.5
78.8
278.8
195.8
360.3
135.5
164.8
251.6
151.0
145.0
110.5
139.3
194.7
184.3
95.0
145.7
219.2
250.5
190.3
415.6
505.1
271.8
365.0
316.8


NTA+


489.2
628.5
320.7
482.1
344.8
840.1
721.6
546.7
652.8
290.8
657.9
894.1
827.4
747.2
490.0
651.5
355.0
481.4
359.1
341.0
546.1
389.0
524.6
428.0
653.3
673.5
535.6
790.7
915.0
695.2
886.2
735.6
+Includes  SIP

-------
                               - 133 -


consistant relationship between the quantity of available P and the
quantity of resin-extractable P in these sediments.  The reason for what
appears to be erratic extraction which removes widely varying proportions
of sediment P, might actually be more a result of poor recovery of P
sorbed on resin after extraction for several reasons:  (1) Poor recovery
of P sorbed on resin by the HC1 rinse, and/or (2) Poor recovery of the
resin itself after extraction, since some resin did appear to pass through
the 60 mesh sieve or become trapped in the sieve during separation of
sediment and resin, and/or (3) High concentrations of precipitate are
formed in the aliquot of HC1 rinse taken for analysis when the pH is
adjusted to 7 prior to addition of colorimetric reagent.  The source of
the precipitates are probably compounds or ions (i.e., Al^"*") removed
from the resin by the acid.  Although these precipitates redissolve when
colorimetric reagents are added (pH dropped), an ionic chemical inter-
ference may be occurring.  Although Huettl et al.  (1979) reported a very
good relationship between available P and resin-extractable P, the erratic
results we obtained and potential for problems in  the handling of the
resin make the resin-extraction procedure (Huettl  et al., 1979) a question-
able choice for a routine estimator of available P.

     The range in the average concentration of P extractable with NTA
(+ soluble Pi) for the 4 sampling periods was between 428.0 and 735.6 yg/g
(avg.  = 540.5 yg/g).   These concentrations represent 38 to 47% (avg.  =
42.5%) of the sediment TP,  which is a sizeable overestimatlon of measured
available P (avg. = 30.3% of TP).

     Linear regression was used to define statistical relationships be-
tween sediment available P measured in 2 day and 2 week incubations and
chemically extractable P fractions.  Linear regression correlation co-
efficients from such comparisons are presented in Table 4.  A reasonably
significant (r = 0.75 and 0.77, respectively) relationship existed between
NH^F-extractable P (+ soluble Pi)  and both 2 day and 2 week available P
levels in stream sediments.  Therefore, it appears that NH^F removes amounts
of P similar to that removed by algae in 2 day incubation periods over the
range of concentrations and sediments tested in this study, and even though
NH^F-extractable P underestimated 2 week available P, it appears that the
quantity extracted by IttfyF was linearly related to 2 week available P.
Furthermore, even though averages presented earlier indicate a substantial
relationship between quantities of Pi extracted by algae in 2 weeks and
sequential IttfyF and NaOH linear regression of values obtained for each
site in 4 sampling periods showed only a mildly significant relationship.
The relationship between Pi resulting from sequential NH4F, NaOH, and HC1
extraction was only a weakly significant one.

     The sequential NaOH and HC1 extraction performed prior to the 2 day
incubation resulted in two high correlation coefficients in relation to 2
day available P.   Both NaOH-extractable P and total sediment Pi resulted
in correlation coefficients of 0.90 or greater,  indicating that although
both NaOH-extractable Pi and Pi overestimate available P, they are both
linearly related to 2 day available P.  Linear regression analysis of the
same variables obtained initially in the 2 week incubations versus 2 week
available P produced correlation coefficients greater than 0.9.  The same
conclusions may be drawn.  Linear  regression comparisons of NTA-extractable

-------
                                - 134 -
Table 4.  Relationships between available P measured by bioassay (2 day
          and 2 week incubation periods) and chemically extractable P
          initially present in stream sediments.

                                       Incubation period
                                           *
Parameter                             2 day         2 week
Sequential NH.F:
  NH4F-P + SIP+                        0.75          0.77
  NH4F-P + NaOH-P + SIP                0.57          0.59
  Inorganic P                          0.52          0.61

Sequential NaOH  (2 day):
  NaOH-P + SIP                         0.90
  Inorganic P                          0.92

Sequential NaOH  (2 week):
  NaOH-P + SIP                          —           0.91
  Inorganic P                           —           0.94

NTA:
  NTA-P + SIP                          0.75          0.78

Resin :
  Resin-P + SIP                        0.52          0.59

Total  Sediment P                      0.83          0.80
  All correlations  involving the  available P levels  obtained  from 2  day
  incubations were  calculated with the omission of  the site 2 sample taken
  on 4/14.
  SIP = soluble inorganic P
^At the 0.1 confidence level a r > 0.463 indicates  statistical signifi-
  cance.

-------
                                - 135 -


P and 2 day and 2 week available P resulted correlation coefficients of
0.75 and 0.78 indicating that although NTA-extractable P overestimates
available P, there appears to be a relatively consistant relationship.  The
low r values obtained in regressing resin-extractable P against 2 day and
2 week available P  (r = 0.52 and 0.59, respectively) bear out conclusions
drawn earlier concerning the use of the resin extraction in estimating
available P.  There appears to be direct relationship between the concentra-
tion of TP and available P as indicated by r values of 0.83 and 0.80 for
2 day and 2 week incubations, respectively.
                              CONCLUSIONS
     The following conclusions may be cited from the data collected during
the course of this study:

1) Available sediment P as measured by incubation for 2 days with algae
   was ca. 41% of Pi and 25% of TP.  The 2 day available P represents the
   amount of P which might become immediately available upon entering the
   receiving body of water if conditions are optimum for the growth of
   algae.

2) Most (87%) of maximum, potentially available sediment P became available
   within the first 2 days of incubation.

3) Maximum available sediment P was measured in sediments incubated for
   2 weeks with algae and amounted to 50% of Pi and 30% of TP.

4) Of the extractants tested, NffyF extraction of sediment best  simulated
   the quantity of Pi removed by algae during the 2 day incubations.
   Therefore, it appears that a single NIfyF extraction of sediment could
   be used to estimate immediately available sediment P.

5) Sequential NtfyF and NaOH was,  in turn, a good estimator of the total,
   potentially available sediment P.

6) The single NaOH-extractable fraction, a triple NTA-extractable fraction,
   inorganic P, and TP all contain levels of P which are linearly related
   to 2 day and 2 week available P.

-------
                               - 136 -


                          LITERATURE CITED
1.   Allan, R. J. and J. D. H. Williams.  1978.  Trophic status related
    to sediment chemistry of Canadian prairie lakes.  J. Environ. Qual.
    7:99-106.

2.   Armstrong, D. E., J. J. Perry, and D. E. Flatness.  1979.  Avail-
    ability of pollutants associated with suspended or settled river
    sediments which gain access to the Great Lakes.  Great Lake National
    Program Office, U.S. Environmental Protection Agency.  Chicago,
    Illinois.  EPA-905/4-79-028-.  106 p.

3.   Bartsch, A. F. 1972.  Role of phosphorus in eutrophication.  U.S.
    Environmental Protection Agency, Office of Research and Monitoring.
    Washington, B.C.  EPA-R3-72-001.  45 p.

4.   Burns, N. M. and C. -Ross.  1-972a.  Project Hypo- - an introduction
    p. 1 In. N. M. Burns, C. Ross (Eds.).  Project Hypo...An  intensive
    study of the Lake Erie Central Basin hypolimnion and related surface
    water phenomena.  Canada Centre for Inland Waters  (Paper  No. 6).
    Burlington, Ontario and U.S. Environmental Protection Agency, Region
    V.   (Technical Report TS-OS-71-208-24).  182 p.

5.   Burns, N. M. and C. Ross.  1972b.  Project Hypo - discussion of  findings
    p. 124,  In. N. M. Burns, C. Ross  (Eds.)  Project Hypo	 An intensive
    study of the Lake Erie Central Basin hypolimnion and related surface
    water phenomena.  Canada Centre for Inland Waters  (Paper  No. 6).
    Burlington, Ontario and U.S. Environmental Protection Agency, Region
    V.   (Technical Report TS-OS-71-208-24).  182 p.

6.   Burns, N. M. and C. Ross.  1972c.  Oxygen-Nutrient relationships with-
    in the central basin of Lake Erie  p. 118.  In.  N. M. Burns,  C.  Ross
    (Eds.)  Project Hypo...An intensive study of the Lake Erie Central
    Basin hypolimnion and related surface water phenomena.   Canada  Centre
    for  Inland Waters  (Paper No. 6).   Burlington,  Ontario and U.S.  Environ-
    mental Protection Agency, Region V.  (Technical Report TS-OS-71-208-
    24).  182 p.

7.  Cowen, W. F. and G. F. Lee.   1976.  Phosphorus availability  in  particu-
    late materials  transported by urban runoff.  J. Water Pollut. Control
    Fed. 48_: 580-591.

8.  Cowen, W. F. and G. F. Lee.   1976b.  Algal nutrient  availability and
    limitation  in  Lake  Ontario during  IFGYL.  Part 1.   Office of  Research
    and  Development, U.S.  Environmental Protection Agency.   Duluth,
    Minnesota.   EPA-600/3-76-094a.   218 p.

9.  Davis,  C.  C.   1964.   Evidence for  the  eutrophication of  Lake Erie
    from phytoplankton  records.   Limnol.   Oceandg.  9:275-283.

-------
                                - 137 -


 10.   Dobson,  H.  H.  and M.  Gilbertson.   1972.   Oxygen depletion in the
      hypolimnion of the central basin of Lake Erie, 1929 to 1970.  p. 6.
      In N.  M.  Burns and C.  Ross (Eds.)   Project Hypo...An intensive study
      of the Lake Erie Central Basin hypolimnion and related surface
      water  phenomena.   Canada Centre for Inland Waters (Paper No. 6),
      Burlington,  Ontario and U.S.  Environmental Protection Agency Region
      V.   (Technical Report,  TS-05-71-208-24).   182 p.

 11.   Dorich,  R.  A.,  D.  W.  Nelson  and L.  E.  Sommers.  1980.  Algal avail-
      ability of  sediment phosphorus in drainage water  of the Black Creek
      watershed.   J.  Environ.  Qual.  9:557-563.

 12.   Fekete,  A.,  D.  N.  Riemer,  and  H.  L. Motto.  1976.  A bioassay to
      evaluate the release of available phosphorus from pond sediments.
      Journal of  Aquatic Plant Management.  14:19-25.

 13.   Fellos,  J.  and W.  R.  Swanson.   1975.  The release rate of nutrients
      from river  and lake sediments.   J.  Water Pollut.  Control. Fed.   47:
      1032-1042.

 14.   Fitzgerald,  G.  P.  and T.  C.  Nelson.  1966.  Extractive and enzymatic
      analyses  for limiting or surplus  phosphorus in algae.  J. Phycol.
      2:32-37.

 15.   Fitzgerald,  G.  P.   1970.   Aerobic  lake muds for the removal of  phos-
      phorus from lake waters.   Limnol.  Oceanog. 15:551-555.

 16.   Gerloff,  G.  C.  and F.  Skoog.   1954.  Cell contents  of nitrogen  and
      phosphorus  as  a measure of their availability for growth of
      Microcystis  aerugenosa.   Ecology" 35:348-353.

 17.   Gilbertson,  M., H.  H. Dobson and T. R. Lee.   1972.   Phosphorus  and
      hypolimnial  dissolved oxygen in Lake Erie.   In. N.  M. Burns, C.  Ross -
      (Eds.).   Project Hypo...An intensive study of the Lake Erie Central
      Basin  hypolimnion  and related  surface  water phenomena.   Canada  Centre
      for  Inland Waters  (Paper No. 6).   Burlington, Ontario and U.S.  Environ-
      mental Protection  Agency,  Region V.  (Technical Report TS-05-71-208-
      24).   182 p.

 18.   Golterman, H.  L.   1976.   Sediments  as  a  source of phosphorus for algal
      growth.   In. H. L.  Golterman  (Ed.).   Interactions between sediments
      and  freshwater.  Proc.  51L-UNESCO  Conf.  Junk and  Pudoc,  The Hague,
      Netherlands.

 19.   Huettl, P. J.,  R.  C. Wendt, and R.  B.  Corey.   1979.   Prediction  of
      algal-available phosphorus in runoff suspensions.   J.  Environ. Qual.
      8:130-132.

20.   International Joint Commission.  1980.  Pollution in the  Great Lakes
      Basin  from land use activities.  International Joint Commission,
     Great  Lakes Regional Office.   Windsor, Ontario.   141 p.

21.  Jackson, M.  L.  1958.  Soil Chemical Analysis.  First course.  Prentice-
     Hall,  Inc.  Sixth printing.  1970,  by  the  author, Dept. of  Soil  Sci.,
     Univ. of Wisconsin.  Madison, WI  498 p.

-------
                                - 138 -


22.   Li, W. C., D. E. Armstrong, J. D. H. Williams, R. F. Harris, and J.
     K. Syers.  1972.  Rate and extent of inorganic phosphate exchange in
     lake sediments.  Soil Sci. Soc. Amer. Proc. 36:279-285.

23.   Logan, T. J., F. H. Verhoff, and J. V. Depinto.  1979.  Biological
     availability of total phosphorus.  Lake Erie Wastewater Management
     Study.  U.S.  Army Engineer District, Buffalo, NY.  62 p.

24.   McCallister,  D. L. and T. J. Logan.  1979.  Phosphate adsorption-
     desorption characteristics of soils and bottom sediments in  the
     Maumee River basin of Ohio'.  J. Environ. Qual. 7:87-92.

25.   Miller, W. E., J. C. Greene, and T. Shiroyama.   1978.  The Selanastrum
     capricornutum Printz algal assay bottle test.  Office of Research and
     Development,  U.S. Environmental Protection Agency, Corvallis,  Oregon.
     EPA-600/9-78-018.  126 p.

26.   Moshiri, G. A. and W. C. Crumpton.  1978.  Certain mechanisms  affecting
     water column-to-sediment phosphate exchange in a bayou estuary.  J.
     Water Pollut. Control Fed. 51:392-394.

27.   Murphy, J. and J. R. Riley.  1962.  A modified single solution pro-
     cedure for determination of phosphate in natural waters.  Anal.  Chem.
     Acta. 27:31-36.

28.   Nnadi, L. A., M. A. Tabatabai, and J. J. Hanway. 1975.  Determination
     of phosphate extracted from soils by EDTA and NTA.   Soil Sci.  119:
     203-209.

29.   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.

30.   Porcella, D. B., S. K. Kumazar, and E.  J. Mlddlebrooks.  1970.  Bio-
     logical  effects on  sediment-water nutrient interchange.  J.  Sanitary
     Eng.  Div., Am.  Soc. Chem.  Eng. 96:911-926.

31.  Ryden, J. C.,  J. K. Syers,  and R. F. Harris.   1973.   Phosphorus  in
     runoff and stream.  Adv.  Agron.  25:1-45.

32.  Sagher,  A.,  R.  Harris, and D. A. Armstrong.   1975.   Availability of
     sediment phosphorus to microorganisms.  Water  Resource  Center,
     University of  Wisconsin,  Madison, WI.   Wis WRC 75-01.   56  p.

33.  Sommers,  L.  E., and D. W.  Nelson.   1972.   Determination of  total
     phosphorus in  soils:  A  rapid perchloric  acid  digestion procedure
     Soil Sci. Amer. Proc. 36:902-904.

34.  Syers,  J. K.,  R.  F. Harris,  and  D.  E.  Armstrong.  1973.   Phosphate
     chemistry in lake  sediments.   J.  Environ.  Qual.  2:1-13.

-------
                                - 139 -
35.   Verhoff,  F.  H.,  M. Eeffner, and W. A. Sack.  1978.  Measurement of
     availability rate for total phosphorus from river waters.  LEWMS.
     U.S.  Army Corps of Engineers, Buffalo District.  Buffalo, NY.  32 p.

36.   Vollenweider, R. A.  1968.  Scientific fundamentals of eutrophica-
     tion of lakes and flowing waters with particular reference to nit-
     rogen and phosphorus as factors in eutrophication.  Organization for
     Economic Cooperation and Development, Paris.  Report DAS/CSI/68.27.
     159 p.

37.   Wentz, D. A. and G. F. Lee.  1969a.  Sedimentary phosphorus  in lake
     cores-analytical procedure.  Environ. Sci. Tech. 3:750-754.

38.   Wentz, D. A. and G. F. Lee.  1969b.  Sedimentary phosphorus  in lake
     cores-observations on depositional pattern in Lake Mendota.  Environ.
     Sci.  Tech. 8:754-759.

39.   Wildung,  R.  E. and R. L. Schmidt.  1973.  Phosphorus release from lake
     sediments.  Office of Research and Monitoring, U.S. EPA.  Washington,
     B.C.   EPA-R3-73-04.  185 p.

40.   Williams, J. D.  H., T. P. Murphy, and T. Mayer.  1976.  Rates of
     accumulation of phosphorus forms in Lake Erie sediments.  J. Fish.
     Res.  Board Can.  33:430-439.

41.   Williams, J. D.  H., H. Shear, and R. L. Thomas.  1980.  Availability
     to Scenedesmus quadricauda of different forms of phosphorus in sedi-
     mentary materials from the Great Lakes.  Limnol. Oceanog. 25:1-11.

-------
                                   - 140 -
            ACCOUNTING FOR NITROGEN DISPOSITION WITHIN A WATERSHED

                                 R.F. Davila
                                 L.F. Huggins
                                 D.W. Nelson
     Tne adaptability of existing nitrogen cycle models to be used in conjunc-
tion with the ANSWERS watershed simulation model to directly evaluate the fate
of this agricultural nutrient was investigated.  Field data for several of the
processes in the nitrogen cycle were collected from three of the Black Creek
subwatersheds and used to evaluate the accuracy of the selected nitrogen
model.
1.  Introduction

     Tne availability of nitrogen is of prime importance to growing plants,
since they are dependent on an adequate supply of nitrate and ammonium for
synthesis of their nitrogenous constituents.  When plant and animal residues
are added to the soil, the nitrogen containing compounds in these residues
undergo numerous transformations.  Some of these work in opposite direction;
the net result being that not all of the total nitrogen is in an available
form at any one time.

     Nitrogen in the soil exists in organic and inorganic forms.  The inor-
ganic fraction is the one used by plants.  Inorganic nitrogen rarely exceeds 2
to 3 percent of the total soil nitrogen (Bear, 1964).  Transformations from
one form to the other  occur continuously as a result of biochemical reac-
tions.  Fertilizers are added to supplement the nitrogen in the soil, espe-
cially when the rate of conversion of organic to inorganic nitrogen is not
great enough to satisfy plant needs.  Figure 1 is a diagram of the inputs,
outputs and losses of nitrogen in the soil-plant system.

     As seen in the figure, sources of nitrogen are: fertilizers, fixation of
atmospheric nitrogen, precipitation and residues. The organic nitrogen is
mainly the result of manure and plant residues.  Nitrogen not used by plants
is lost; the principal losses being denitrification, ammonia volatilization
and leaching of nitrate.  Additionally, considerable losses of nitrogen may
result from runoff.

     Probably the most undesirable losses of nitrogen result from runoff and
leaching.  These not only represent an economic loss, since the nitrogen is
never used by plants, but also create a pollution problem in the receiving
waters.

     A lot is known about nitrogen in the soil, water and atmosphere (Porter,
1975; Bartholomew et al.,1965; Nielsen et al., 1978).  However, this body of
knowledge has been concentrated on individual processes. During the last few
years these processes have been studied together and the system has been
looked at as a whole rather than by parts (Endelman et al., 1972).

-------
  CROP RESIDUES
   ORGANIC
    NITROGEN
        V
     RUNOFF
     N2- FIXATION

   RAINFALL
                                     FERTILIZER

                                           4,
                      MINERALIZATION
                     IMMOBILIZATION
INORGANIC
      NITROGEN
        v
     LEACHING
•PLANT UPTAKE

 RUNOFF

 VOLATILIZATION

 DENITRIFICATION
Figure 1.  Inputs, Outputs and losses of Nitrogen in the Soil-Plant System

-------
                                  - 142 -

1.1  Fertilizer

     Fertilizer is by far the largest of the nitrogen inputs to the soil-plant
system.  It may be added in the ammonia, ammonium, nitrate or urea form.
Ammonia and urea react with water to form ammonium which, under aerobic condi-
tions, is oxidized to nitrate which is one of the forms of nitrogen used by
plants.

     Due to the inconvenience of applying fertilizer on a grown crop, it has
traditionally been applied all at once in an early stage of growth.  Of the
fertilizer applied this way, less than 50% is assimilated by the crop.  The
rest is lost through various transport mechanisms (Harmsen et al., 1965).
This problem has been addressed by researchers in trying to develop slow
release fertilizers (Gould et al., 1978).  Although some products are commer-
cially available, the technology has not been completely developed yet.
Essentially, these products prevent nitrification from occurring all at once.

1.2  Precipitation

     The quantity of nitrogen added to the soil in rain and snow depends on
location.  Extreme values have been reported between 1.8 and 22.3 Kg of inor-
ganic nitrogen per hectare per year (Allison, 1965).  In rural areas the
values will be closer to the lower end.  Most of the nitrogen contributed by
rain is in the inorganic form.

     Acid rainfall refers to rainfall produced by absorption of oxidized sul-
fur and nitrogen compounds by moisture in the air.  The resulting rainfall is
a weak acid.  In a study conducted over the Great Lakes by the PLUARG group it
was reported that acid precipitation had no measurable effect at the time of
the study except in two isolated embayments over the area covered by the study
(PLUARG Study, 1978).

     Measurements in the Black Creek Watershed show the amount of ammonium and
nitrate in rainwater to be 0.46 to 0.59 mg/1 (3.72 to 4.77 Kg/ha/yr) respec-
tively (EPA-905/9-77-007-B).

1.3  Nitrogen Fixation

     Fixation is the biological transformation of nitrogen gas from the atmos-
phere to organic forms.  Historically, leguminous crops such as alfalfa are
known for their ability to synthesize their own nitrogen fertilizers.  Purdue
researchers estimate tha 70,000 kg of nitrogen were fixed annually within the
Black Creek watershed during 1976 and 1977 by soybeans and forage legumes
(EPA-905/9-7-007-B).  A complete discussion of bacterial nitrogen fixation is
available from Nutman (1965).

1.4  Crop Uptake

     Crop uptake is the principal mechanism by which nitrogen is removed from
the soil.  Each kind of crop and soil situation results  in a unique removal
pattern.  Estimates indicate that, on the average, 50% of the applied nitrogen
fertilizer is used by plants  (Harmsen et al., 1965).

-------
                                   - 143 -


     Most of  the nitrogen absorbed by plants  is  in the nitrate form.  Applied
 nitrogen, if  not in the nitrate form, is  transformed to  it and is made avail-
 able for uptake by plants.  Generally,  fertilizers are applied at one time
 early  in the  cropping period and are made available to plants within the next
 few days after application; at this time  the  mineralization rate is the
 fastest.  Depending on the crop's demand  at this particular time, part of the
 nitrate  will be absorbed; the rest will  either  be stored as nitrate or in
 some other  form or lost.

     Besides  being dependent on the particular crop, the amount of nitrogen
 uptake will also depend on environmental  factors such as temperature, pH,
 moisture and  aeration.  Much work have  been done in this area (Hargrove et
 al., 1979;  Terman and Allen, 1978; Shumway and Atkinson, 1978) in order to
 optimize fertilizer applications, but crop uptake is hard to predict and most
 recommendations are based upon practical  knowledge of the area rather than on
 theoretical reasoning.

     In Indiana, a corn crop is expected  to utilize from 136 to 269 kgs. of
 nitrogen per  hectare, depending on the  yield.  Wheat is expected to uptake
 from 79 to  209 kgs. per hectare (Extension Service, Purdue University #PIH-
 25).

 1.5 Runoff

     One of the most undesirable losses of nitrogen is in runoff.  Highly
 soluble nitrate as well as ammonium and organic  forms are all removed by water
 flowing over  the soil.  Studies have shown that  reduction in rainfall amount
 reduces runoff which in turn reduces nutrient losses (Black Creek Study,
 1977).  The amount of the loss is dependent on land use, conservation prac-
 tices  and season of the year.

     Allison  (1965) cites Lipman and Conybeare with an average of 27.2 kgs.
 per hectare per year of nitrogen lost by erosion from cropland in the United
 States.  An additional problem with nitrogen, besides the high solubility of
 nitrate, is that rain may float organic matter present in the soil and the
 nitrogen content of this organic matter lost may be higher than what is left
 behind.

 1.6  Gaseous Losses

     Gaseous nitrogen losses from the soil may occur as ammonia,  elemental
 nitrogen, oxides of nitrogen and by plants as organic compounds.   Of these
 forms, ammonia is by far the most severe.  Under natural conditions, ammonia
 does not occur in great quantities in the soil;  however, under certain condi-
 tions ammonia which escapes to the atmosphere can be produced in  significant
 amounts.  Ammonia production is favored by high temperature and high pH
values.  Two conditions under which ammonia could be produced in  considerable
 quantities are application of urea and of manure to a field.

     Denitrification is the process by which nitrate is microbially reduced to
elemental  nitrogen and to nitrous oxide.  From the standpoint of  the microbial
population,  denitrifing microorganisms compete for nitrate with plants (Smith,
1979).   The  gaseous products of denitrification  are volatilized,  hence lost

-------
                                   - 144 ~

from the soil.  Denitrification is affected by environmental factors.  It has
been reported that denitrification is affected by excessive moisture; a criti-
cal level being 60% of the water holding capacity of the soil (Allison, 1965).
Under moisture conditions which favor denitrification, the limiting factor is
the amount of organic material percent.  Denitrification losses are favored in
an anaerobic soil environment, i.e. waterlogged soils.

1.7  Mineralization and Immobilization of Nitrogen

     A large portion of the soil nitrogenous material is in organic form.  In
order for this material to be used by plants it must be mineralized to ammonia
or nitrate (inorganic forms).  These inorganic forms of nitrogen, besides
being used by plants, are subject to leaching and volatilization.  In order to
Keep the soil reserves of nitrogen from being depleted, at the same time that
the organic fraction is being mineralized, the inorganic fraction is immobil-
ized; the inorganic nitrogen not used or lost is returned to the organic form.

     The reate of these two opposing processes, besides being controlled by
the total nitrogen status of the soil and the inorganic nutrient supply, is
also dependent on a number of environmental factors such as moisture, pH,
aeration and temperature.  The carbonrnitrogen ratio of added organic residues
also affects mineralization and immobilization.  Usually the term mineraliza-
tion implies net mineralization.  Optimum moisture for mineralization is
between 50 and 75% of saturation (Beas, 1964).  In waterlogged soils, aeration
is reduced and soils tend to go anaerobic; under anaerobic conditions immobil-
ization does not occur and net mineralization is highly positive.

     Temperatures most suitable for mineralization have been reported at about
35 degrees C (Nielsen and McDonald, 1978).  When temperatures drop to freez-
ing, mineralization stops (Endelman et al., 1972).  Mineralization is also
favored by pH near neutral.  The carbon:nitrogen ratio of organic matter in
the soil has also been used as a criteria for mineralization.  This criteria
is usef\ul in cases when crop residues or manure are added to the soil.  The
C:N ratio of stable soil organic matter is about 10:1.  It has been reported
that, as a general rule, when residues yielding a C:N ratio greater than 30 is
added to the soil, immobilization is favored in the initial stage of the
decomposition process (Gilliam et al., 1978).  If the C:N ratio is less than
20, mineralization is favored.  For values between 20 and 30 there may be
immobilization or mineralization.
2.  Model Description

     The soil-plant-nitrogen system is a complex one.  As previously men-
tioned, knowledge about individual processes maKing up the system were not
integrated until recently.  It wasn't until the last few years that the system
has been simulated as a whole.

     ANSWERS is an event-oriented simulation model.  That is, it simulates
dynamic watershed processes occurring during and immediately following a storm
event.  In order to incorporate nitrogen transport into its computations, the
antecedent levels of the various nitrogen forms must be specified.  Thus, the
search for a suitable existing nitrogen model was concerned with finding one

-------
                                    - 145 -


which would  simulate  long-term seasonal  trends of each nitrogen form
throughout a watershed.  The output of such a model could then serve to define
antecedent values  for subsequent ANSWERS storm simulations.

     After reviewing  several models (Davila, 1980), the model developed by
Mehran and Tanji  (1974) was selected.  This model ties a long-term hydrologic
submodel  to  a nitrogen transport submodel to provide the moisture gradient for
nitrogen  transformations.  The model is  based on mass balance and steady state
conditions.  First order kinetics are assumed for the nitrogen transforma-
tions.  In its most recent version,  the  model includes a storage term which
makes it  suitable  to  simulate sites in which permeability is low (Tanji et al,
1979).

     The  latest published version of the Mehran-Tanji model assumes that
effective precipitation (actual precipitation - runoff losses) not lost
through evapotranspiration is lost  through leaching; hence, no moisture accu-
mulation  terms are included.  For this study, the model was subsequently modi-
fied to include a  water accumulation term.  The other major modification made
to the published version of the model was to reduce its time scale from an
annual to a  monthly basis.

     The  nitrogen  transport submodel is  also based on the principle of mass
balance.  This submodel uses flows predicted by the water submodel as a gra-
dient for nitrogen movement.  Nitrogen transformations are assumed to follow
first order  kinetics.  In reducing  the time scale from yearly to monthly,
nitrogen  transformations were neglected  in the months of November, December,
January and  February.  The nitrogen  submodel includes a nitrogen storage term
to account for organic nitrogen not mineralized and inorganic nitrogen not
taken up  by  the crop or demineralized.

     The  plant uptake component has been  rearranged so that instead of supply-
ing an annual value, plant uptake is calculated from the percentage of total
nitrogen  uptake and the available nitrogen in the soil on a monthly basis.
Neither nitrogen gaseous losses of fertilizer nor losses due to nitrogen car-
ried in runoff were independently included in the model.  Some of these losses
are implicit in the "denitrification and other losses coefficient".

     The depth of  soil used through this study has been one meter.  The model
considers this one meter depth as a unit volume on a per hectare basis.  The
mass balance takes place in this unit volume. Nitrogen processes simulated are
assumed to take place uniformly within this unit volume.  Leaching concentra-
tions predicted represent concentrations just below the one meter depth.
Remaining nitrogen concentrations represent the average nitrogen concentration
in the first meter of soil (surface to depth one meter)  at the end of the
month.   After the leaching component is calculated,  the flows are divided into
interflow and deep percolation.   In a field with a subsurface drainage system,
interflow represents expected effluent from the tile system.  Deep percolation
represents flows to the water table.  Schematic descriptions of the water and
nitrogen submodels are given in  Figures 2 and 3.

-------
                         - 146 _
SOIL SURFACE
 BOTTOM OF ROOT ZONE
 Figure 2.   Schematic Description of Water Submodel

-------
                         -  147-
                             NAIF         NAOF
SOIL SURFACE NPRO
-------
                                   - 148 -
3.  Field Investigation
     Soil samples, approximately 500 gins, wet basis, were collected using a
hand sampler from the surface at 33 centimeters intervals to 1 meter of depth
from each of the three subwatersheds studies.  Where row crops were present,
sites 51 and 55, three cores were taken at each sample location: one in the
row and the other two to each side of the row. The side cores were sampled at
the surface, 33 cm. and 66 cm., but not at 1 meter.  In places where row crops
were not present three cores were also taken, but the side cores were taken
about 66 centimeters from the center core, at depths up to 66 centimeters.

     Instead of designing a grid system for sample locations, which would have
made the number of samples extremely large, an S pattern was followed on each
site and sample locations were chosen arbitrarily within the S pattern. This
system kept the number of samples reasonable and at the same time gave an
acceptable representation of the site. The number of locations within a site
varied from 9 (site 20) to 11 (site 51).  Figure 4 shows this methodology.

     The samples were collected in paper bags in the field.  In the laboratory
each sample was divided into two parts, one part was frozen and the other was
air dried.  The samples to be frozen were put in plastic bags.  After being
air dried, the second part was also kept in plastic bags until ready to be
analyzed.  Sampling was done in June, August and October of 1979.  Sample
locations were measured so that sampling could be done in the same place at a
later date.

     Analyses performed on the samples were: net mineralization rate, ammonium
content and nitrate content.  The first was done on the frozen samples and the
last two on the air dried samples.  Mineralization rate tests were done using
the aerobic method proposed by Bremner (Methods of Soil Analysis, Part II,
1965). The freezing of the samples was used to stop mineralization until the
samples were ready to be analysed as discussed by Gasser (1958).  Ammonium and
nitrate were determined using the steam distillation method (Methods of Soil
Analysis, Part II, 1965).

     Figures 5 to 7 show the inorganic nitrogen content of the soil vs time
for the three sites studied at the four levels measured.  The farming opera-
tion is also given in the top portion of each figure.  The time scale for the
farming operation is the same as for the inorganic nitrogen content.  Inor-
ganic nitrogen is the sum of ammonium and nitrate measured.  The values
presented are average values.  The samples at each location within each site
were composited before analysis.

     From these three figures the following general observations are obtained:
inorganic nitrogen content decreases with depth, inorganic nitrogen depletion
is slow when the crop is just planted and is faster when the crop gets closer
to maturity.

     The data for site 20, depicted in Figure 5 shows the inorganic nitrogen
content decreasing with depth for samples taken in June and November.  Samples
taken in August show the amount of inorganic nitrogen higher in the second
level than in the first.  A reason for this might be the fact that samples
were taken from a field which had been harvested and the sampling was done

-------
                      _ 149 _
LAYOUT OF SAMPLING POINTS WITHIN A WATERSHED
      A'A
              33 CM
              33 CM
               33 CM
SAMPLES TAKEN AT  EACH SAMPLING  POINT
           Figure  4. Sampling Methodology

-------
  OPERATION
  PERFORMED
                  */
   45

   40

   35

21 30
o
g 25
z
 •  20
Q.
a.
15


10

5
          SURFACE
      •   33 CM.
      X   66 CM.
      A   IOOCM.
                                                                                      H
                                                                                      Ul
                                                                                      o
         MAMJJASO

                          CALENDAR  MONTH (1979)

      Figure 5.  Soil Inorganic Nitrogen Content vs. Calendar Month for Site 20
                                                                     N

-------
  OPERATION
  PERFORMED
a:
Q.
CL-
35

30

25

20

15

10
           •  SURFACE
           •  33  CM.
           X  66  CM.
           A  ICO CM.
         M
                                                                                        i

                                                                                        Ln
                                                                                        t-»
                                                                                        I
                      M
                           J       J       A       S      0

                       CALENDAR  MONTH (1979)

Figure 6.  Soil Inorganic Nitrogen Content vs. Calendar Month for Site 51
N

-------
OPERATION V
         i/
PERFORMED,
*/
v>
       40


       35


       30
     CO
     or
     020
     z

     •  15
     2

     fc 10
        5


        0
               •  SURFACE

               •   33 CM.
               X   66 CM.
               A   100 CM.
             M      A       M      J       J       A       S

                                CALENDAR  MONTH  (1979)


          Figure 7.  Soil Inorganic Nitrogen vs. Calendar MDnth for Site 55
                                                                                         i

                                                                                         M

                                                                                         to

                                                                                         I
                                                                           N

-------
                                   _ 153 _


during the period of highest precipitation in the year.  During this period
surface loses of nutrients are expected to be the higher.

     Samples taken at site 20 in November also show a considerably higher
inorganic nitrogen content than in the previous sampling.  This is understand-
able since the field had been fertilized in early September and no crops; were
planted; therefore, depletion was minimal during the period since the previous
sampling.

     At site 51, data shown in Figure 6, the inorganic nitrogen content
decreases with depth during the first and second samplings and in the last
sampling the results for the first and second level are the same.  At this
site fertilization was done in the first part of May.  At this time the crop
was in its early stage and the depletion during the subsequent period until
the next sampling was not as fast as the inputs to the inorganic nitrogen
pool, thus leading to higher inorganic nitrogen values in the second sampling.
As the crop matured the nitrogen requirements increased and the inorganic
nitrogen pool was depleted, as seen in the results for November.

     At site 55, Figure 7, the same general trend is followed in the first and
second samplings, but in November inorganic nitrogen in the fourth level is
higher than expected although very close to the values for the second and
third levels.

     Figure 8 is a plot of net mineralization rate vs depth for the three
sites studied.  The results shown are averages for the field and the three
measurements done during the growing season.  Normally mineralization will be
expected to decrease with depth in a field of uniform permeability charac-
teristics.

     The soil at site 20 is a deep soil of low permeability decreasing with
depth.  The top layer, the layer of cultivation, is more permeable.  As a
result of this permeability, mineralization in the second level is higher than
in the first, although subsequent levels follow the expected decrease with
depth.

     Mineralization rate at site 51 decreases with depth as expected. At site
55 the pattern is as expected until the third level where it is slightly
higher.  This variation in the third level amounts to 0.050 ppm/day more than
in the second level.

     Table 1 shows organic-N and mineralization rate data for three sites.
Data used to generate the optimum N mineralized is included in appendix B.
The soil bulK density used was 1.5 gms/cm .   The organic nitrogen concentra-
tions were measured in the top 33 cm.  Optimum mineralization rate constants
(K) were calculated assuming a first order decay rate.

-------
            SITE  4*20
         SITE #51
         SITE *55
     Sr
o
Q.
UJ
O
   66
   100
                              33
 66
100
                          33
 66
100
      0     Q50    LOO         0     0.50     LOO         0     0.50    LOO

                          MINERALIZATION  RATE,   PPM / DAY

      Figure 8.  Net Mineralization Rate vs. Depth.

-------
                                   - 155 -
                      Table 1.  Mineralization Rate Data
                                       Site 20    Site  51       Site 55
     Organic-N  cone.  (0-33  on),  ug/g    2095       1033         1677
     Organic-N  in  profile,  Kg/ha       10470       5165         8385
     Optimum N  mineralized, ug/g/day    0.29       0.32         0.36
     Optimum K,  weeks l                 9.7x10  4    2.17x10 J    1.48x10   J

     In theory, a first order decay rate is expected for most nitrogen
transformations; the decay constant can be expected to change through the
year.   In order to fully characterize this first order decay with a sampling
procedure like the one used in this study, the time between samplings needs to
be greatly  reduced.  In a laboratory type situation in which the nitrogen
inputs could be controlled in experimental plots and the number of samplings
during the growing season increased, this decay could be observed.


4.  Model Testing

     A discussion of all model evaluations completed is available elsewhere
  (Davila, 1980).  Results from final tests are given in Tables 2 to 4.  The
measured inorganic nitrogen content in the soil is compared to simulation
results.

     At site 20, predicted and observed inorganic nitrogen agreed within 20%
in July and September.  In May there is considerable difference between
observed and predicted data.  Leaching losses took place at site 20 in March
and April.  In March and April the soil water capacity was exceeded causing
leaching.  The amount of leaching depends on the amount of nitrogen available
to be leached and on the amount of water that exceeds the soil moisture capa-
city.

     Results for site 51 show agreement between predicted and measured inor-
ganic nitrogen data within 20% in July and September.  May values agree within
30%.  Leaching took place in March and April.  The peak concentration of
nitrogen leached was in March which coincides with results from site 20.

     At site 55, predicted and observed results of inorganic nitrogen agreed
within 20% in July and September, the agreement in May is within 40%.   Leach-
ing at this site started in February and continued until April.

     The amount of nitrogen leached at site 20 for the year was less than 3%
of the total fertilizer applied.  At site 51, 13% of the applied fertilizer
was leached.  At site 55, more than 50% of the fertilizer was leached.

     Nitrogen percolated for all sites was calculated as 5% of the amount
leached.  The amount of nitrogen in interflow is the difference between nitro-
gen percolated and leached.  Variations of nitrogen in interflow and nitrogen
percolated follow the same pattern as variations in leaching.

-------
Table 2.  Final Simulation - Site 20
Month Fertili- N-Leached Dentrifi- Crop Soil Inorg-N
zation cation and uptake Organic N
other losses mineralized Predicted measured
(kg/ha) (kg/ha) (kg/ha) (kg/ha) (kg/ha) (kg/ha)
Jan
Feb
Mar
Apr
May
June
July
Aug
Sept
Oct
Nov
Dec
0
0
44.3
28.9
0
0
0
57.8
0
0
234.6
0
0
0
9.4
4.1
0
0
0
0
0
0
0
0
0
0
9.3
10.2
8.3
6.2
6.0
7.5
9.3
10.0
0
0
0
0
14.4
42.6
62.0
39.3
14.0
0
0
0
0
0
0
0
14.1
17.1
28.5
39.8
42.4
39.6
33.0
19.9
0
0
75.4
75.5
99.6
89.2
48.0
43.0
66.1
99.1
122.9
133.2
368.5
-
-
-
-
-
85.0
-
55.0
-
135.0
-
-
-

-------
Table 3.  Final Simulation - Site 51
Month Fertili- N-Leached Dentrifi- Crop Soil Inorg-N
zation cation and uptake organic N
other losses mineralized Predicted measured
(Kg/ha) (Kg/ha) (kg/ha) (Kg/ha) (Kg/ha) (kg/ha)
Jan
Feb
Mar

Apr
May
June
July
Aug
Sept
Oct
Nov
Dec
0
0
0

0
51.6
0
0
0
0
0
0
0
0
6.0
0.8

0
0
0
0
0
0
0
0
0
0
0
6.9

7.4
13.5
15.6
16.5
11.6
9.2
8.3
0
0
0
0
0

0
3.9
19.6
88.4
53.0
23.6
7.9
0
0
0
0
15.4

18.9
31.4
43.3
45.9
42.6
35.2
21.3
0
0
70.3
70.5
73.5

84.7
150.8
159.5
101.1
79.8
82.3
87.9
88.4
-
-
_ i
Ul
I
-
110.0
-
120.0
-
65.0
-
-
-

-------
Table 4.  Final Simulation - Site 55
Month Fertili- N-Leached Dentrifi Crop Soil Inorg-N
zation cation and update organic N
other losses mineralized Predicted measured
(Kg/ha) (Kg/ha) (Kg/ha) (Kg/ha) (Kg/ha) (Kg/ha)
Jan
Feb
Mar
Apr
May
June
July
Aug
Sept
Oct
Nov
Dec
0
40.4
0
0
9.2
0
0
0
0
3.0
0
0
0
3.9
19.8
6.9
0
0
0
0
0
0
0
0
0
0
10.3
9.2
9.0
9.8
11.7
10.6
9.1
9.4
0
0
0
0.8
5.5
30.6
31.0
18.1
50.6
47.9
14.8
3.7
0
0
0
0
17.1
21.0
34.7
48.1
51.3
47.6
39.6
24.1
0
0
75.5
111.4
93.5
68.4
72.9
94.0
83.9
74.4
90.2
104.6
105.4
-
-
-
-
-
125.0
-
70.0
-
80.0
-
-
-
                                                                               00

-------
                                     159
4.1  Parameter Sensitivity
     Site 51 was chosen as a case study to illustrate how model predictions
change as a result of incremental changes for individual parameter values.
The parameters changed were: mineralization rate constant (K),  denitrification
and other losses coefficient (C)  and initial soil water content (STOO).

     The response of nitrogen leached to variations in the mineralization rate
constant is shown in Table 5.  Results of these simulations show that the
amount of nitrogen  leached is fairly sensitive to order of magnitude changes
in K.  Increasing K causes an increase of inorganic nitrogen production,
therefore an increase in nitrogen available to be leached.  This situation  is
favored by high temperature in the summer months.  A decrease in K works  the
opposite way, decreasing the inorganic nitrogen produced and the nitrogen that
could be subject to leaching.  This situation is favored by low temperatures.
          Table 5.   Leaching Response to Variations in K for Site 51.
K (Week !)
0.000217 0.00217 0.0217
Month N - Leached, (kg/ ha)
Jan
Feb
Mar
Apr
May
Dec
0
0
5.0
0.5
0
•
•
0
0
0
6.0
0.8
0
•
0
0
0
15.4
3.2
0
•
0
     The response of nitrogen leached to variations in C is shown in Table 6.
As expected, an increase in C decreases the nitrogen amount in leaching,  but
the magnitude of change is relatively small.  An increase in C increases
nitrogen losses and as a result a decrease in the nitrogen pool in the soil.
A decrease in the nitrogen pool in the soil reduces the nitrogen that could be
leached.  A decrease in C decreases nitrogen losses and, as a result, nitrogen
that could be leached.
          Table 6.   Leaching Response to Variations in C for Site 51.
                       0.02
 C
0.08
0.20
Month N - Leached, (kg/ha)
Jan
Feb
Mar
Apr
May

Dec
0
0
6.4
0.9
0
I
»
0
0
0
6.0
0.8
0
•
»
0
0
0
5.2
0.6
0
i
0

-------
                                   - 160  -

     Variations in the initial soil water parameter (STOO) on nitrogen leached
are shown in Table 7.  These results demonstrate the very substantial effect
of the initial soil moisture condition on the amount of nitrogen leaching that
occurs.  Unfortunately, this quantity was not measured for the field test
subwatersheds.  Thus, the uncertainty associated with this parameter could
account for some of the discrepancy between simulated and measured conditions.
          Table 7.  Leaching Response to Variations in Initial Soil
                          Water Content  for Site 51.
                           Initial Soil Water Content (on)
                        8.0             14.0              20.0
Month N - Leached, (kg/ha)
Jan
Feb
Mar
Apr
May
Dec
0
0
0
0
0
0
0
0
6.8
0.8
0
0
13.2
4.5
6.6
0.6
0
0
5.  Summary and Conclusions

     A data base was developed which provided inputs and outputs of nitrogen
in the soil-plant system of three subwatersheds in the Black Creek area.  This
data base was analysed, supplemented with data from the literature and used
with a slightly modified form of an existing nitrogen fate model (Tanji,
et.al., 1979).  The most significant modifications made to the model were the
addition of a water storage term in the water submodel and changing from a
yearly to a monthly time scale.

     Some of the model parameters were varied and the effect of these varia-
tions on the predicted amounts of nitrogen leached observed.  Parameters
varied were: the mineralization rate constant (K), the denitrification and
other losses coefficient (C)  and the initial soil water content (STOO).  Vari-
ations in K and STOO produced the most significant changes in the amount of
nitrogen leached.

     K proved to be a fairly sensitive parameter in determining the amount of
nitrogen leached.  A positive variation in K increased the inorganic nitrogen
pool and, as a consequence, the amount of nitrogen leached.  The importance of
STOO is related to the timing with which nitrogen is available for leaching.
A high value of STOO will cause leaching earlier in the year.  Early in the
year the crop's nitrogen requirements are low.  Most leaching occurs at this
time and nitrogen required to meet the crop's need at a later time is lost.

     The modified version of the Tanji model gave good agreement between
predicted and observed data.  However, uncertainties in some of the model
parameters and the neglect of others precludes conclusive evidence on the
validity of the model.  The effect of having only estimated the initial soil
water suggests deficiencies in the data base.  The initial soil water proved
to be a key parameter in predicting nitrogen leached.

-------
                                   -161  -

     Approaching the problem of simulating the soil-plant nitrogen cycle with
the principle of mass transfer was a very effective approach.  Acceptable
results were obtained at a computer cost of $0.25 per run.  From an economic
standpoint this tool is good.

     Other deficiencies encountered in the model toward which future worK
should be directed were:

     The model does not take into account farming practices.  The effect of
farming practices on the amount of nitrogen in the field should be considered.

     The effect of precipitation is treated as a single unit in the model.
Modifications should be made so that rainfall and snowfall are treated
separately.  In reality, higher values of leaching are expected in the period
of thawing than in the rest of the year.  Interflow does not occur when the
soil is frozen.  Precipitation as snow in December, January and February usu-
ally does not become soil water until late March in this part of the country.

     In most cases, fertilizer doesn't become nitrate immediately after being
applied.  The model does not account for this effect.  This deficiency causes
overprediction of nitrogen concentration in leaching at the end of the month
in which fertilization takes place.

     Future work should also be directed toward strengthening the data base by
making more frequent field measurements.  As done for this study, it is recom-
mended that the data base be obtained from watersheds in which more than one
crop is cultivated and different farming practices are used.

-------
                                  -162  -
                              LIST OF REFERENCES

Allison, F.E.  1965.  Evaluation of Incoming and Outgoing Processes that
Affect Soil Nitrogen.  Soil Nitrogen, Monograph No. 10.  American Society of
Agronomy, Inc., Madison, Wisconsin.

Bartholomew, W.V.  1965.  Mineralization and Immobilization of Nitrogen in the
Decomposition of Plant and Animal Residues.  Soil Nitrogen, Monograph No. 10.
American Society of Agronomy, Inc., Madison, Wisconsin.

Bear, F. E.  1964.  Chemistry of the Soil.  American Chemical Society, Mono-
graph Series No. 160.

Bremmer, J.M.  1965.  Inorganic Forms of Nitrogen.  Methods of Soil Analysis
Part II: Chemical and Microbiological Properties, C.A. Black (ed.), American
Society of Agronomy, Madison, Wisconsin.

Davila, R.F.  1980.  Predicting the Nitrogen Leached in the Black Creek
Watershed.  MS Thesis.  Purdue Univ.  W. Lafayette, IN.

En dleman, F.J., M.C. Northup, D.R. Kenney, J.R. Boyle and R.R. Hughes.  1972.
A Systems Approach to an Analysis of the Terrestial Nitrogen Cycle.  Journal
of Environmental Systems, Vol. 2,(1) pp. 3-19.

Gasser, J.K.R.  1958.  Use of Deep-Freezing in the Preservation and Prepara-
tion of Fresh Soil Samples.  Nature Vol. 181.

Gilliam, J.W., S. Dasherg, L.J. Lund and D.D. Focht.  1978.  Denitrification
in Four California Soils: Effect of Soil Profile Characteristics.  Journal of
Soil Science Society of America, Vol. 42: 61-66.

Gould, W.D., F.D. Cook and J.A. Bulat.  1978.  Inhibition of Urease Activity
by Heterocyclic Sulfur Compounds.  Journal of Soil Science of America, Vol.
42.

Hargrove, W.L. and D.E. Kissel.  1979.  Ammonia Volatilization from Surface
Applications of Urea in the Field and Laboratory.  Journal of Soil Science,
43:359-363.

Harmsen, G.W., G.J. Kolenbrander.  1965.  Soil Inorganic Nitrogen.  Soil Nito-
gen, Monograph No. 10.  American Society of Agronomy, Madison, Wisconsin.

International Joint Commission, PLUARG.  1978.  Nitrogen Transformation
Processes in Agricultural Watershed Soils, Windsor, Canada.

International Joint Commision, PLUARG.  1978.  Environmental Management Stra-
tegy for the Great Lakes Systems, Final Report, Windsor, Canada.

-------
                                   _ 163 _

Mehran, M. and K.K. Tanji.  1974.  Computer Modeling of Nitrogen Transforma-
tions in Soils.  Journal of Environmental Quality, Vol. 3, No. 4.

Nielsen, D.R. and J.G. MacDonald (eds.).  1978.  Nitrogen in the Environment,
Vol. 1 and 2.  Academic Press, New YorK.

Nutman, P.S.  1965.  Symbiotic Nitrogen Fixation.  Soil Nitrogen.  Monograph
No. 10.  American Society of Agronomy, Madison, Wisconsin.

Porter, K.S.  1975.  Nitrogen and Phosphorus Food Production, Waste, The
Environment.  Ann Arbor Science Publishers, Inc.

Purdue University, Cooperative Extension Service, Fertilizer Value of Swine
Manure.  Publication # PIH-25.

Shumway, J. and W.A. Atkinson.  1978.  predicting Nitrogen Fertilizer Response
in Unthinned Stands of Douglas-Fir.  Communications in Soil Science and Plant
Analysis, 9  (6) 529-539.

Smith, O.L.  1979.  Application of Soil Organic Matter Decomposition Model.
Soil Biology and Biochemistry, Vol. 2. pp. 607-618.

Tanji, K.K., F.E. Broadbent, M. Mehran and M. Fried.  1979.  An Extended Ver-
sion of a Conceptual Model for Evaluating Annual Nitrogen Leaching Losses from
Croplands.  Journal of Environmental Quality, Vol. 8, No. 1.

Terman, G.L. and S.E. Allen.  1978.  Crop Yield - Nitrate-N and Total N and K
Concentration Relationship: Corn and Fescuegrass.  Communications in Soil Sci-
ence and Plant Analysis, 9 (9): 827-841.

U.S. Environmental Protection Agency.  1977.  Environmental Impact of Land Use
on Water Quality.  Black Creek Project Report, EPA-905/9-77-007-B.

U.S. Environmental Protection Agency.  1978.  Simulation of Nitrogen Movement,
Transportation and Uptake in Plant Root Zone.  EPA-600/3-78-029.

-------
                              -  164 -



                  TEMPORAL INSTABILITY IN THE PISHES OF A

                    DISTURBED AGRICULTURAL WATERSHED

                                  by
                             11                2
                Louis A. Toth /  James R. Karr , Owen T. Gorman

                         and Daniel R.  Dudley


                               ABSTRACT

     Eight years of data on the fishes cf Black Creek are analyzed to
evaluate potential effects of stream alterations and other watershed
modifications.  During this period 44 species of fish were captured
but most samples contained less than 20.  Benthic (bottom feeding)
species numerically dominated the community.  Species composition was
very unstable, particularly during summer months.  Only three species,
Semotilus atromaculatus, Pimephales promelas, and P_. notatus
consistently maintained populations in the watershed during the entire
year.  Others migrated to and from the Maumee River.  The most
important migrants were Notropis spilopterus, Dorosoma cepedianum,
Catastomus commersoni, Cyprinus carpio, and Carpiodes cyprinus.  Adults
of the latter three species generally remained in the watershed only
long enough to spawn, but their young dominated the fauna along with
young D_. cepedianum during summer months.  Recent trends suggest
increases in abundances of Notropis stramineus, N_. cornutus- t£.
chrysocephalus, and N_. umbratilis.  These increases are matched by
earlier declines in populations of Etheostoma spectabile, Campostoma
anomalum, and especially Ericymba buccata.  Factors responsible for
these instabilities include short-and long-term effects of stream
alterations and other watershed modifications resulting in extreme
variation in flow regimes and choking algal blooms.  In addition,
fish kills, fish migration and within-stream movements, and temporal
and spatial variation in recruitment and mortality result in highly
variable fish communities.  The relative importance, duration and time
of effect of these factors varies among fish species.  Knowledge of
these patterns is essential for more informed programs of watershed
management, and, thus, better water resource systems.

Key words:  Agricultural impacts, Black Creek, channelization, fish
            communities, perturbation,  stability, streams, stream
            alteration, water resources*
1.  Department of Ecology, Ethology, and Evolution,  606 E. Healey
    Vivarium Bldg., University of  Illinois, Champaign, IL 61820

2.  Museum of Natural History, University of Kansas,  Lawrence, KA
    66045
3.  Division of  Survellance, Ohio  Environmental  Protection Agency,
    361  E. Broad, Columbus, Ohio 43215

-------
                              - 165 -
                             INTRODUCTION
     The primary goals of the Black Creek project were to develop and
implement plans for controlling soil erosion and to evaluate the
effectiveness of traditional conservation programs in improving the
quality of a water resource.  Recent clean water legislation defines
the water quality goal as physical, chemical, and biological integrity.
Hence, the success of the Black Creek project must be viewed not only
in light of sediment and nutrient loads but also with respect to effects
on the biota of the water resource system.

     Fish were selected as the primary group for evaluation of project
impacts on the aquatic biota.  In this report we present a summary of
fish population data  from the Black Creek watershed.  We summarize
data on community structure beginning with preliminary sampling conducted
in 1973 and continuing to the collection of the most recent samples
(June 1980).  In addition, we discuss the spatial and temporal patterns of
variation in species richness and species diversity in the fishes of the
watershed.

     One of the problems of this type of analysis is the complex array
of variables, both man-altered and natural, that govern structure and
function in the aquatic community.  Thus, we are often not able to
definitively demonstrate causal links between community change and
specific human activities.  However,      strong inferences can often be
made.  In general, we       explore those links in this paper where
possible but we leave many of the details of such links to an analysis
in an integrative Black Creek Report to be completed in 1982.

     Although it is not the intent of this report to evaluate effects
of project activities on the physical integrity of Black Creek and its
tributaries, it is cJear that the intensive application of structural
conservation practices modified stream channels throughout the watershed.
A number of studies (see Karr and Gorman 1975 for a review) have shown
that similar perturbations, particularly ditching and channel straightening,
severely alter  the structure and stability of existing fish communities,
at least over the short term.  Advocates of these practices commonly
assume that these effects are  temporary; the biota will recover and
perhaps even be enhanced as water-quality benefits accrue.

     Few studies have evaluated the long-term effects of stream modification.
Gorman and Karr (1978) suggest that community complexity  (diversity) may
recover relatively quickly while stability either requires a longer
period or may never be attained.  Hence, we attempt to summarize eight
years of sampling  in  the  Black  Creek
watershed and    evaluate the origins of     changes that have occurred
in the fish community during this period.

-------
                                166 ~

                                METHODS
 Fish Sampling.

     Twenty-five fish sampling stations were established in the Black
 Creek watershed (Fig.  1).   However,  since most of the sampling effort
 focused upon the main channel, only data from stations 6, 18, 17, 28,
 29, 15, and 12  were intensively analyzed.  Each station was 100 m in
 length.  Extensive algal blooms and extremely low flows (Table 1)
 sometimes prevented the entire station from being sampled.  Samples
 covering less than 100 m were rather infrequent, however, and did not
 distort the data significantly enough to warrant special treatment.

     Sampling frequency was not uniform among stations but generally
included at least four main channel sites on each sampling date.
Sampling frequency also varied among years.  Initial samples were taken
at sites 6 and 12 during July 1973.  From 1974 to 1978 samples were
taken at monthly intervals although some bi-weekly samples were taken
during 1975 and 1976.  Sampling was less frequent in 1979 but included
collections from spring, summer, and fall.  The last sample was taken
in June 1980.

     Sampling was conducted using 3,1 or 6.2 mm mesh minnow seines
with block nets at the upper and lower ends of the station,  However,
in April 1979 fish were sampled with an electric seine powered by a
gasoline geierator.  After capture, fish were either identified,
counted, and released or preserved in 10% formalin for laboratory
analysis.  In addition, fish were often measured to the nearest 1 mm
(total length) to monitor the age structure of populations.

Data Analysis.

     Fish species diversity patterns were evaluated during the course
of the study using the Shannon-Weiner index  (H).
                          N
                     H = -£     P± loge P.
                          i=l

where P^ is the proportion of species i in a sample of N species.  This
index is sensitive to shifts in the number of species and distribution of
individuals among the species.  H increases as the number of species
or their equitability increases.

     To evaluate short-term (i.e., month to month) changes in community
structure, a sample similarity index (PS) was calculated whenever
samples were taken during consecutive months from any given station on
the main channel of Black Creek.  The index expresses the degree  (%)
to which two samples are alike in quantitative representation of  species
(Whittaker 1975) and was calculated as follows:

-------
                                                                                                    I

                                                                                                    H-1
                                                                                                    ON


                                                                                                    I
                                                                    0
                                                                       SCALE  (km)
Figure 1.   Map of  the Black Creek watershed showing  the  locations of streams and sampling


           sites discussed in text.

-------
                               - 168 -

Table 1.  Environmental conditions (based upon temperature and precipitation
          records)  and perturbations in the Black Creek watershed.
                                        Season
Year
1973
Winter
Mild
Spring
Normal
Summer
Normal
Fall
Normal
1974
1975
1976
1977
1978
1979
Mild
Mild
Mild
Severe
Severe
Severe
                   Some channel
                   alterations upstream

 Wet               Dry             Dry
 - Extensive channel alterations	
                   Algal blooms upstream
 Normal            Normal
Bridge construe-   Algal blooms
tion downstream    upstream
                                  Normal
                                   Dry
Dry               Dry
                  Some channel
                  alterations
                  upstream
   - - - Extensive algal blooms - - --
 Normal
 Algal blooms

 Wet
 Dry
                                  Wet
                                                                  Normal
                  Dry             Dry
                     	 Algal blooms 	

                  Normal          Dry
                     Algal blooms upstream

-------
                              _ 169 _
                      PS = 1 - 0.5Z P  - P,  Iwhere
                                   la    hi
                                     a

P  = the number of individuals of a given species expressed as its
     proportional importance in the community during month  (a)

P  = the number of individuals of the same species expressed as its
     proportional importance in the community during the following
     month (b)

When more than one sample was taken during a given month, similarity
values were calculated for each combination of samples from consecutive
months.  Sample similarity was then given as the mean of all
similarity values for that period.

     Temporal variability and diversity patterns were dissected by
evaluating these parameters among benthic and pelagic guilds, delimited
primarily according to feeding locales  (Table 2).  Independent diversity
and sample similarity measures were calculated based upon the
proportional representation of a given species within its respective
guild.

     Habitat conditions during the study period are summarized in Table 1.

                         COMMUNITY STRUCTURE

     In this section we outline the status of fish populations at each
of the main channel stations.  We begin at upstream areas and proceed
downstream discussing the data year-by-year with a brief summary for
each station.  A list of the species known from the watershed
and their guild assignments is provided in Table 2.
Figs 2-10 and Table 3 show the changing populations for several more
abundant  species.  A more detailed analysis of the Ericymba buccata
population  is provided in another paper   (Toth et al. 1981).

Station 6-26.

1973 - Pimephales promelas was the most abundant species and E_. buccata,
P_. notatus and C_. carpio  (recruits) were common in a preliminary
sample taken in July 1973.  The presence of numerous  young C_. carpio
is signficant since it indicates that this species spawned  in the
watershed prior to the project-related habitat alterations.

1974 - Pimephales promelas was even more abundant and again dominant
in March 1974.  Large adults of N^. cornutus-N. chrysocephalus,
C_. commersoni, and especially S_. atromaculatus were also numerous.
Although there was a marked decline in fish density in April, P_. promelas
was still fairly abundant.  Only ten fish were caught here  in May—
soon after this stretch of stream was channelized and the fauna
remained rather depauperate in July.  However, a large number of E_.
buccata yearlings were caught in algae-choked water during this month.
Although no quantitative samples were taken during the fall, dip-net
samples indicated that KL spilopterus was common in September and October.

-------
                               - 170 -
Table 2.  Guild assignments (based upon feeding habitat)  of all
          fish species that were caught in the Black Creek watershed
          from 1973 - 1980.  Benthic species feed on or near the
          substrate while pelagic species forage higher in the water
          column or at the surface.
                                  Benthic Guild
         Dorosoma cepedianum
         Umbra limi
         Campostoma anomalum
         Carassius auratus
         Cyprinus carpio
         Ericymba buccata
         Notropis stramineus
         Phenacobius mirabilis
         Pimephales notatus
         Pimephales promelas
         Carpiodes cyprinus
         Catostomus commersoni
         Erimyzon obloncjiis
         Ictiobus cyprinellus
         Minytrema melanops
         Moxostoma spp.
         Ictalurus melas
         Ictalurus nebulosus
         Ictalurus natalis^
         Etheostoma blenniodes
         Etheostoma nigrum
         Etheostoma spectabile
         Percina maculata
         Aplodinotus grunniens
Gizzard Shad
Central Mudminnow
Common Stoneroller
Goldfish
Carp
Silverjaw Minnow
Sand Shiner
Suckermouth Minnow
Bluntnose minnow
Fathead Minnow
Quillback
White Sucker
Creek Chubsucker
Bigmouth Buffalo
Spotted Sucker
Redhorse
Black Bullhead
Brown Bullhead
Yellow Bullhead
Greenside Darter
Johnny Darter
Orangethroat Darter
Blackside Darter
Freshwater Drum
                                 Pelagic  Guild
         Esox  lucius                    Northern  Pike
         Notemigonus  crysoleucas        Golden Shiner
         Notropis atherinoides          Emerald Shiner
         Notropis chrysocephalus        Striped Shiner

-------
                                _ 171 _
Table 2. (continued)
                                   Pelagic Guild Cont.
          Notropis cornutus
          Notropis spilopterus
          Notropis umbratilis
          Semotilus atromaculatus
          Fundulus notatus
          Labidesthes sicculus
          Ambloplites rupestris
          Lepomis cyanellus
          Lepomis gibbosus
          Lepomis humilis
          Lepomis macrochirus
          Lepomis microlophus
          Micropterus salmoides
          Pomoxis annularis
          Pomoxis nigromaculatis
          Perca flavescens
Common Shiner
Spotfin Shiner
Redfin Shiner
Creek Chub
Blackstripe Topminnow
Brook Silverside
Rock Bass
Green Sunfish
Pumpkinseed
Orangespotted Sunfish
Bluegill
Redear Sunfish
Largemouth Bass
White Crappie
Black Crappie
Yellow Perch

-------
                             - 172 -
             3-
         to
        iZ
         O)
        jQ
         £
         13
         O>
         O
                STATION  12
                                   I1  ' '  'I
STATION 15
                 STATION  29
               I TTI  I I Til I  T II    I I  I I   I I I I I  I I I I  I I I  I
               AN'FMAN'FMAN'FMAN'FMAN'FMANFMAN'FM
              1973 1974  1975  1976  1977  1978  1979 I960

                             Sample  Date
Figure 2a,,  IjC"3-,n number of fish at Station 12,  15,  and 29 in the Black

           Creek Watershed,  1973-1980.

-------
       4r
x:


il   I

**—

 0   3
 w_
 CD
X>

 E
 13
                I TTI I  I  TM  i I  Tjl I  I Tjl  I I  ITT i  I I


        h STATION 18
   cn
   O
                      - 173



         STATION  17-28
               STATION
                 nji-. ri|  i



         h STATION 6-26
         I  III I  T FIT  I I  I I I  I  I III I  I  III  I I  ITI F I  Ml  I
         AN'FMAN'FMAN'FMAN'FMAN'FMAN'FMAN'FM

        1973  1974   1975   1976  1977   1978   1979   1980


                         Sample  Date
Figure 2b.  L°9ln number of fish at Stations 17-28, 18, and 6-26 in the


          Black Creek Watershed, 1973-1980.  Samples were taken at


          Station 26 from February - May 1976 only.

-------
               I
                O>
               JCl
                e
                O>
                O
                    2         - 174

                       STATION 12
                                       , jyjl.
                                       ill  i i 11i  i i i 11
                       STATION 15
                        IT 11 TTT I  I MlI I Mlin

                       STATION  29
                        M1 '  ' 'I1 '  ' 'I1 ' ' M1
                       STATION 17-28
                                           1 i
                          n«c n  «	I
                       i 111  i r T 11  i i ri i  i i i ri i i ri i

                       STATION  18
                                          JLri
                       STATION 6-26
                         TT i I TTT I I Tp I I  I I I I I  III I I  If' I I  '}< I
                       A N F M A N F M A N F M A N F M A N FM AN FM A  N T M
                      1973 1974    1975    1976    1977  1978   1979  I960
                                   Sample  Date
Figure 3.   Lo<310 number of  Pimephales promelas at main channel stations

           in the Black Creek Watershed, 1973-1980.  Station 26

           sampled from February - May 1976 only.

-------
                            - 175 ~
        I

        H—
         o


         CD


        "1
         13
         CD

         O
                 STATION  12
                 STATION  15
         1
|l
      1 'I1 '  '  'I1  ' '  '


      STATION  29
                                            n_
                 iii  i i  i 11  i  i i 11 i  i  i 11  i i  i 11  i  i i  i i  i 11 \  i

                A N F M A  N F M A N F M A  N F M A N F M A N F M A N F M

                1973  1974   !975    1976   1977    1978    1979   1980
                                Sample   Date
Figure 4a.
Log1Q nvunber of Pimephales notatus at Stations 12, 15,  and 29



in the Black Creek Watershed,  1973-80.

-------
to    2
•
 o


 CD
JQ

 £
 13
 cn
 o
_J
                           - 176 -
          STATION  17-28
                 1  '1  '  '  '
          STATION   18
                                            0
                    I I I  I  I  11 I  I  I  I I I
                  R
J
          1  'I1 '  '  T

          STATION  6-26
  IXXX   1111X   XX
I  II11 I I 11II
              XXX X
            X  J  X
                                             K XX
          I  III  I I  TTT   I  III  I I  I I I  I  I  I I I  I I  I I I  I  I  I I I  I
          A  N  F M A NT M A  N F M A N F M A NT M A N F M A  NT M

         1973   1974     1975    1976     1977    1978     1979    1980
                           Sample   Date
   Figure 4b.  IJ°^-,Q number of Pimephales notatus at Stations 17-28, 18, and

             6-26 in the Black Creek Watershed, 1973-80.  Station 26 sampled

             February - May 1976 only.

-------

-------

  o


  0>
  CJ>
  o
         4              _178 _


            STATION  17-28
i  111 *iffli xi nx i i
                                   JUJ
            STATION   18
            1 ill  1 1  111 1  1  III


            STATION  6-26
        :J|
                                          xx
           L
           •WVrW
                                                  i  1  1 11
                                      n
               i  i i  in i  i  in i  r IIT ri rrr i i  111 i  i  m n
           AN'FMANFMANFMAN'FMAN'FMANFMAN'FM
           1973  1974
           1975
1976
1977    1978   1979  1980
                            Sample   Date
Figure 5b. Log   number of Notropis spilopterus at Stattions 17-28, 18,  and

         6-26 in the Black Creek Watershed, 1973-1980.  Station 26 sampled


         February - May 1976 only.

-------
                           -  179 -


                   STATION  17-28
          I    2
yyy y [1 WXW * , ,X f »
1 . .*. < ..."


I mWni"l I ' | i"T"' ' | l ^ ' 1 ' ' ' ' 1 '
           CD
           _Q
           E
            o
           O
STATION  18
                                                 10!
STATION  6-26
                   1973  1974    1975   1976   1977    1978   1979   I960
                                   Sample  Date
Figure  6.  Log,n number of Notropis  cornutus/Notropis chrysocephalus  at main
          channel stations in the Black Creek Watershed, 1973-1980.
          Station 26  sampled February - May 1976 only.

-------
                               - 180 -
                 s

                 I
                 I
                 Z   2
                 -Q

                 E
                  o

                 CT
                 o
                         STATION 12
                        i  iii r i  MI  r r



                        STATION 15
                                           J
                                                           i rn\
                                 m
                                 i  i
                                   floc
1  'I' '  ' T


STATION 29
               xxyrf
               1 T
i  in  I i rp I  I TTI I i rri



STATION  17-28
                                          ^ _ nfirT
                        y  II  y
                        r 1 1 i  i
                         STATION 18
                         i 1)1  r i rji ^r rji T i T|,^ i ^( i



                         STATION 6-26
                          M iii iii  ii i|i r i TJI ri iii

                        ANFMANFMANFMANFMANFMANFMANFM

                        1973 1974   1975   1976  1977   1978   1979  1980
                                      Sample   Date
Figure 7.   Log   number of  Notropis stramineus at main  channel dtations


            in the Black Creek Watershed,  1973-1980.  Station 26 sampled



            February - May 1976 only.

-------
                    - 181 -
J
2
^5 i
.5 i
t
•s
i_
JD
1 i
~Z-
0
CP
.3 i
i
STATION 12


, .,. , , ,,,"T
STATION \5



«J\ 1
j- ,
T''1 	 '"





n
PI


xx xx xxxxxxxx xxxxrfl x x xxx 'I
1 .
1 1 I 1 I 1 II 1 1 1 1 1 I ~\ 1 I 1 1 1 1 1 1 1 1 III 1 1 1 1 1
STATION 29 r


1 l|l 1 1 l|l 1 ! J 1
STATION 17-28
XXX X X
,,,,,, ,r ,





xxxxx x x xdllll x
1 Mill 1 1 1 1 III


''•
X
1 1 1 1
n
X
i i i i i i 1 1 i
STATION 18

X X XXX)

KXXXX XXXXXXXX IKXX XX
I Ijl 1 1 ipl <  1 III . < III 1
STATION 6-26
LT-- XXX X. 	 _..Jl ,

XX II XX




X X
i i I i i i i i
xx NX
1 1 1^ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
ANFMANFMANFMAN FMANFMA^
973 1974 1975 1976 1977 1978
III 1
1 F M A N F M
1979 I98C
Sample Date
rure 8. Log10 number of Notropis umbratilis at main channel stations i,
Black Creek Watershed, 1973-1980.  Station 26 sampled February -



May 1976 only.

-------
                               - 182 -
                 ^


                 
-------
                        - 183 -

-------
                                - 184 -

Table 3.  Seasonal abundance of migrant species in Black Creek.   The
          data given are the mean number of individuals (±  SE)  per
          sample, based upon the average seasonal abundance of each
          species at the six main channel sampling stations.
1974
Mar - May
Jun - Aug
Sept - Nov
1975
Mar - May
Jun - Aug
Sep - Nov
1976
Mar - May
Jun - Aug
Sept- Nov
1977
Mar - May
Jun - Aug
Sept- Nov
1978
Mar - May
Jun - Aug
Sept- Nov
1979
Mar - May
Jun - Aug
Sept- Nov
Cyprinus
carpio
0
0
0

0
0.2 + 0.1
0.3± 0.3

0.7± 0.4
6.9± 4.2
0

1.2± 1.1
35. 7± 11.0
7.0± 2.4

0.6± 0.3
44.0+ 22.0
50.7+ 36.6

8.6 ± 3.3
0.5 ± 0.5
0.2 ± 0.2
Catostomus
commersoni

7.3+ 1.9
3.4± 2.9
0.8± 0.8

0.6+ 0.5
0.1± 0.1
0

0
13.5 ± 6.9
14.8 ± 7.5

4.7 ± 1.7
8.0 ± 2.7
12.8 ± 3.1

4.5 ± 1.6
33.6 ±12.0
37.7 ±27.3

34.4 ±11.4
27.0 ±20.0
2.0 ± 1.1
Carpiodes
cyprinus
0
3.0± 2.4
0.7± 0.7

0.1± 0.1
0.4± 0.3
0

0
14.7+ 14.6
4.1 + 2.0 .

0.4 ± 0.4
58.9 ± 17.9
16.1 ± 5.7

0
11.9 + 4.8
6.7 ± 3.8

0
0
0
Dorosoma
cepedianum

0
0
4.3+ 1.2

0
5.5± 3.1
25. 6± 25.5

0
21.8+ 19.1
1.3± 1.0

0
67.8+ 19.3
50. 1± 13.3

0
4.1 ± 2.5
0

0
0
1.0 ± 1.0

-------
                              -  185 -
1975 - Very few fish were caught during the early spring of  1975,
apparently because a newly installed weir below the  station  prevented
fish from moving into this region during the prevailing low  flow
conditions.  As water levels rose in May, this area  was repopulated
by P_. promelas, £. notatus, and N. umbratilis.  The  density  of  each
of these species remained about the same  (i.e.,<50)  through  July but
all other- species were rare.  A substantial decrease in the  abundance
of P_. notatus in August preceded a similar decline by P. promelas  and
N^. umbratilis in September.  During this month, 12. buccata and  _L.
cyanellus were the most common species.  Only 20 fish, 11 of which were
L. cyanellus, were caught in October.

1976 - The density of fish, particularly that of P_.  promelas, was
Hghfrom February through April 1976.  Two other species, P_.  notatus
and E_. buccata, were also abundant during this period; however, the
density of ,1E. buccata declined considerably in late  March.   A fair
number of large adult S_. atromaculatus were caught during February
and March.  The density of fish was much lower in May when this entire
section of stream was algae-choked.  During this time, P_. promelas
and S. atromaculatus were the dominant species and P_.  notatus and
1C. buccata were common.  In contrast  to the early spring samples, most
of the S_. atromaculatus caught in May were yearlings.  Algae was
flushed out by July when P_. promelas was again dominant while S_.
atromaculatus, P_. notatus, and E^. buccata remained common.   This
region nearly dried up completely during August and  September and  only
ten fish were caught in October when algae again choked the  stream.

1977 - Fish densities were not particularly high but more species  were
caught in the late spring of 1977 than in any previous sample at this
station.  From March through June, P_. promelas was again the most
abundant species but P_. notatus, S_. atromaculatus, NL  spilopterus, and
N. stramineus were also common.  In addition, adult  C_. commersoni,
(2. carpio, and C^. cyrpinus were also caught during this period.  In
July, the community was dominated by a very large number of  C^.  carpio
recruits and the only other common species were P_. promelas  and
S_. stromaculatus.  Another major shift in community  structure occurred
in August when there was an influx of D_. cepedianum  migrants.   Notropis
spilopterus also began to invade this area during August and P_. promelas
remained common.  During November, D_. cepedianum was still abundant but
was replaced by N. spilopterus as the dominant species.  Although only
a few P_. promelas were caught, S_. atromaculatus, N_.  cornutus-N.
chrysocephalus, N. umbratilis, P_. notatus, and young of C_. carpio  and
C. commersoni were common.

1978 - The fish community was again rather depauperate in 1978.
Pimephales promelas, I?, notatus  (March only) , and S_. atromaculatus were
the only abundant species in two spring samples and  C_. carpio recruits
represented 84% of the fish collected in July and 73% of those  found in
September.  Pimephales promelas and S_. atromaculatus were also  common  in
September.

-------
                              - 186 -

1979 - The only sample at this station in 1979 was taken in October
and consisted of a small number of N^. spilopterus, L^. cyanellus,  P_.
notatus, P. promelas, 14. umbratilis, and N_. cornutus-N. chrysocephalus.
The last sample was taken during June 1980 when the community was
overwhelmingly dominated by P_. promelas.

Summary.

     With the exception of 1977, the fish community at this station
was very depauperate.  Pimephales promelas was clearly the most
abundant species, particularly during the spring months.  Pimephales
notatus and both large adult and yearling S_. atromaculatus were  also
consistently common during the spring.

     During the summer months, P_. promelas remained relatively  abundant;
however, the fauna was dominated by recruits of C. carpio and p_.
cepedianum migrants in 1977, and by young £. carpio again in 1978.
Cyprinus carpio recruits were also present, though not as abundant,  in
a July 1973 sample.  Other common species found during the summer
months included E_. buccata, P_. notatus, S_. atromaculatus, and  less
frequently, N_. umbratilis and N. spilopterus.

     No samples were taken during the fall in  1973 and 1974 and fish
densities were extremely low during this period  in 1975  and 1976.  Fish
were more abundant during the fall of 1977-1979 but only one  sample
was taken during each of these years.  Notropis  spilopterus was
dominant in November 1977 but I), cepedianum was  also  fairly abundant.
Young_C. carpio were dominant in September 1978.  No  species was
particularly abundant in October 1979 but_N. spilopterus was  common.
Other common species that were frequently  collected during the fall
months  included I1. cyanellus, S_. atromaculatus,  P. notatus, P_.  promelas,
H. umbratilis, and :N. cornutus-N. chrysocephalus.

Station 18.

1974 -  A large percentage of  young E_. buccata  were  found in  a somewhat
unreliable sample taken in algae-choked waters in July  1974.   Although
E. buccata remained  common in September, N_.  spilopterus  migrants
dominated  a rather depauperate community.

1975 -  The fauna was still sparse during the  early spring of  1975 when
P_. promelas and P_. notatus were the  most  common species.   Pimephales
notatus emerged as the  most  abundant species in May but E_.  buccata,
S_. atromaculatus  and N_.  stramineus  also increased in density.   The
influx  of  N.  stramineus continued  through June and  it temporarily
became  the dominant  species.  However,  a  large number of primarily
one-year-old  E^. buccata dominated  the samples  from July through  October.
Notropis  stramineus  and especially P_.  notatus  were  also fairly common
during  this period.

-------
                               - 187 -


 1976  -  Although more than twice as many fish were caught here in
 March 1976  than during the early spring of 1975, two species, P.
 notatus and E_.  buccata,  comprised about 85% of the sample.  Ericymba
 buccata continued to dominate the community through July while the
 density of  P_.  notatus declined.  Adult I?,  promelas became temporarily
 abundant during May but most of these fish were gone by July.  However,
 a  large number of small P_.  promelas and young-of-the-year S_.  atromaculatus
 were  found  in  August, making these species co-dominants with E_. buccata.
 A  dramatic  change in community structure occurred in September when
 N.-  spilopterus migrated into this region and IP. promelas, S_.  atromaculatus,
 and IE.  buccata all apparently moved out.  In fact N. spilopterus
 consistently made up 98% of the monthly fish samples from September
 through November 1976.   Much of the impetus for this shift was undoubtedly
 tied  to the dense algal  blooms and low water level conditions that began
 in August and prevailed through this period.

 1977  -  Like the previous spring, a large number of fish were again caught
 here  during April 1977.   Pimephales promelas was the most abundant
 species but P_.  notatus an 13.  buccata, the dominant species during the
 spring  of 1976,  were also very abundant.  In addition,  N. stramineus,
 N.-  cornutus-N.  chrysocephalus, and especially S_. atromaculatus were
 fairly  common.   A major  shift in community structure occurred in July
 when  the density of P_. promelas and P_.  notatus declined significantly
 and the fauna  was dominated by £.  cyprinus recruits.  Although E. buccata
 remained abundant in July,  it exhibited a marked decrease in density
 along with  £.  cyprinus in August.   No species was particularly abundant
 during  this sampling period;  however, S_. atromaculatus, N. spilopterus,
 and I),  cepedianum were common.

 1978- Very  few fish were caught here during March and April 1978, but
 many  adult  £.  commersoni, £.  carpio, and £. cyprinus were sighted in this
 region  in late April and May.

 1979  -  Fish densities,  particularly those of J?. notatus, P_. promelas, s_.
 atromaculatus,  and N_. stramineus were high again during the spring of
 1979.   In addition,  for  the first time during the course of sampling
 at this station, NL  umbratilis was also abundant.  Other common species
 collected in April included t>J. cornutus-N. chrysocephalus, F. notatus,
 E_.  spectabile,  adult £.  commersoni, and young £. carpio.  In October ,N_.
 spilopterus, P_.  notatus, and P. promelas were the most abundant species,
 while N_.  stramineus was  also common.

 1980  -  A sample taken in June 1980 revealed a somewhat depauperate,
 though  seasonally typical,  fish community dominated by P_. promelas
 and a fair  number of P_.  notatus.                       ~

 Summary.

      Although  densities  of individual species fluctuated somewhat, the
 spring  fauna showed very little structural variation between  years.
 Pimephales  notatus, J>. pronelas,and, prior to 1978,  E.  buccata were
 consistently the most abundant species.   However, during two  recent
 years (1977  and 1979) !3.  .atromaculatus  has become much  more abundant
 in  spring samples.   Furthermore,  a similar increase  in  the density of
JJ.  stramineus  and N.  umbratilis occurred during 1979.   The schools of
 adult (2.  commersoni, £.  carpio,  and  C.  cyprinus observed during the

-------
                               - 188 -


spring of 1978 is also noteworthy,  since it indicates that these  species
probably spawn in this region.

     During the summer months fish  community structure was much more
variable both within and between years.  Furthermore, there were  no
general patterns to this variability  except that one of  the dominant
species from 1974 through 1976, E.  buccata, declined in  importance
thereafter.  Species that were dominant for short periods of  time
included N. stramineus (June 1975), P_, promelas and S. atromaculatus
(August 1976), and C_. cyprinus recruits  (July 1977).  Pimephales
promelas was also the dominant species in the sample taken in June
1980 and _P. notatus was abundant on limited occasions.   In contrast to
the lower stations, IX cepedianum migrants and recruits  of £.  commersoni,
C_. carpio, and C_. cyprinus did not  form a significant component of the
summer fauna even though the latter three species appear to spawn
nearby.

     With the exception of 1975 when  E_. buccata was dominant, 1£.
spilopterus was consistently the most abundant species during the
fall months; however, only one fall sample was taken here after
1976.  In this sample  (October 1979) , P_. promelas, N_. stramineus, and
P_. notatus were also fairly common.   Pimephales notatus  was also
abundant in the fall of 1975.

Station 17.

1974 - Fish densities were low and community composition was  poor
during each of the four sampling dates in  1974.   There was  also
considerable temporal  instability,  with  S_. atromaculatus dominating
the  fauna  in April, P_. notatus  in  June,  and E_. buccata in July
and  October.  It  is likely that extensive habitat perturbations
throughout this year and an algal   bloom in the  fall were largely
responsible  for the depauperate fish community.

1975 - Considerably more fish  were found here  in  March  1975 and,
although E_. buccata remained the domariant species, S_. atromaculatus
and  P_. notatus were also abundant.   The  presence  of numerous  one-year-
old  S_. atromaculatus during the spring   of 1974  and  1975 suggests that
these young  fish  may be overwintering in this  region.  The  final
sample taken  at this station was in May and yielded  about the  same
number of  fish as were caught  in March;  however,  the  species
composition was not recorded.

Station  28.

1976 - In  contrast to  the  temporal instability typically exhibited by
the  Black  Creek  fish  fauna, community structure  at  this  station
underwent  only minor  changes  in monthly  samples  from March through
July 1976.   During this period,  E_. buccata was the  dominant species
and  P_. notatus  and P_.  promelas were consistently abundant;  however,  the
density  of P_. promelas did decline considerably  in  July.  In
addition,  owing  to the recruitment of young,  the number of S_. atromaculatus

-------
                              -  189  -
increased steadily during June and July when it emerged as the  second
most abundant species.                      .   Other  fairly  common
species, whose relative abundance also  fluctuated significantly during
this period, included N. stramineus  in  April,  F_.  notatus  in  April
and June, and C_. anomalum and E_. spectabile  in June.   The presence of
about 80 mostly young E_. spectabile  in  June  is particularly noteworthy
since it represents the greatest number of darters ever captured in
a sample during the course of the study.  In October,  this  area appears
to have supported the highest density  f>3000)  of  fish  in  the watershed.
Although this included 14 species of fish, eight  were  represented by
less than 20 individuals.  Notropis  spilopterus was  the dominant species
but over 700 P_. notatus, 500 E_. buccata,  and about 180 individuals
each of _P. promelas, C_. anomalum and S_. atromaculatus  were  also
captured.

1977 - Relative to the preceding fall,  the fauna  was rather  sparse
during the spiring of 1977.  Pimephales  promelas,  P_.  notatus, and
S. atromaculatus were the most abundant species with the  latter two
species exhibiting substantial increases  in  density  in May.   The number
of fish continued to increase, and the  community  underwent  a
significant change in July.  Although S_.  atromaculatus maintained the
same density as in May, equivalent numbers of  E_.  buccata, ID. cepedianum,
and C_. cyrpinus (young) were also found,  while P_.  promelas,  P_.  notatus,
and young of C_. carpio were common.  In August the community was
dominated by IX cepedianum and N_. spilopterus  migrants as well  as
recruits of C_. cyprinus and C. carpio.  The  densities  of  S_.  atromaculatus
and IE. buccata declined during this  month but  both of  these  species
remained fairly common along with P_. promelas,  .P.  notatus,  N_.  cornutus-
N. chrysocephalus, and young of C^. commersoni.  Migrants  from the
Maumee River continued to invade this area in  September and included
large numbers of N. spilopterus, N_.  stramineus, and £.  cepedianum.
Pimephales notatus and N_. cornutus-N. chrysocephalus were also  among
the most abundant species captured.  In fact,  the density of
N^. cornutus-N. chrysocephalus was the highest  recorded for  this species
to date.  Other common species included £L atromaculatus, P_. promelas,
E_. buccata, and young of £. cyprinus.   Although the  fauna was probably
decimated by a fish kill in late September,  N.  spilopterus  and
E>. cepedianum were still abundant in November.  Semotilus atromaculatus,
N. cornutus-N. chrysocephalus, P_. notatus, and N.  umbratilis were
also common but N. stramineus showed a  marked  decline  in  density.

1978 - The density of fish was lower in April  1978 but N_. stramineus
re-invaded   and was the dominant species along with P. notatus
and P_. promelas.  Fifteen large, adult  C^. commersoni were also  caught
during this month.  Semotilus atromaculatus  was the  dominant species
and P_. promelas, E_. buccata and young of  C.  commersoni and C. carpio
were common in August, the only other sampling date  in 19787

-------
                             - 190-

1979 - In contrast to the previous springs, a large number of fish,
about half of which were p. notatus, were found here during April
1979.  Pimephales promelas, N_. stramineus, and S_. atromaculatus were
also very abundant while N_. cornutus-N. chrysocephalus, N_. umbratilis,
N_. spilopterus, and E_. buccata were relatively common.  A small number
of young C_. commersoni were also captured.  With the exception of  a
significant decline in the density of N_. stramineus, fish community
structure was very similar in June when P_. notatus, P_. promelas, and
S_. atromaculatus were again the most abundant species.  As was the
case during October 1976, 14 species were again found here in October
1979, but only three species, N_. spilopterus, N_. cornutus-N. chrysocephalus,
and N_. stramineus, were abundant and only two others, P_. notatus and N_.
umbratilis, could be considered common.  The rest were represented by
fewerthan 20 individuals each.

1980 - Community composition during June 1980 was similar to that  of
June 1979 except N_. stramineus and, to a lesser extent, KL cornutus-N.
chrysocephalus were more abundant.  Notropis stramineus was a
co-dominant with P_. notatus and P_. promelas while S_. atromaculatus
and N_. cornutus-N. chrysocephalus were the next most abundant species.
In addition, about 20 large adult C_. commersoni were still present.

Summary  (17 - 28).

   Fish densities were generally lower at Station 17 than at Station
28, particularly during 1974 when this area was subjected to extensive
habitat perturbations.  Although sampling was conducted somewhat
infrequently in recent years, densities appear to have been consistently
higher during 1979-80 than during previous years.   In  addition, more
frequent samples in 1976 and 1977 reveal a seasonal pattern, characterized
by an increase in density from spring to summer.

    Species compositions and relative abundances were  remarkably
similar during the spring months both within and between years.   The major
long-term change that has occurred  involves the population decline by
E. buccata and a recent increase in the relative importance of
N. stramineus, N. cornutus-N. chrysocephalus, and N_. umbratilis.   Ericymba
buccata was particularly abundant here prior to  its decline, and  was
the dominant species during the spring of 1975 and  1976.  Pimephales
notatus, P_. promelas, and S_. atromaculatus were  also abundant during
the  first three years of sampling and remained so through the last
spring  sample in 1979.  The density of P_. notatus was  particularly high
during April, 1979.  Although N_. stramineus was  fairly common in  a
spring  sample in 1976, it was much  more abundant in samples taken during
April of 1978 and 1979.  Notropis umbratilis and N. cornutus-N.
chrysocephalus were also significantly more  abundant  in April,  1979 than
during  any previous spring.

     To a large extent  community  structure during  the summer months of
1974  and 1976 mirrored that of  the  spring of these  years.  Ericymba
buccata, for example, was  the dominant  species  and  P_.  notatus  and S_.

-------
                                - 191 -
 atromaculatus were  also abundant.   From 1977 through 1980 the latter
 two species  played  an ever more prominent role in the community although
 the density  of P_. notatus generally declined as the summer progressed.
 In  contrast,  12.  buccata,  though it was still a dominant species in
 July 1977, was never very abundant thereafter.  The increase in fish
 density that occurred during the summer of 1977 and 1978 was largely
 due to  the recruitment of C_.  cyprinus, C_.  carpio, and £, commersoni,
 and immigration of  IN.  spilopterus  and 13.  cepedianum, all of which were
 dominant species in samples during these years.  Pimephales promelas,
 another common species during the  summer months of 1976 through 1978,
 showed  a signficant increase in density in June samples during 1979
 and 1980.  Notropis cornutus-N.  chrysocephalus and especially N.
 stramineus were also more abundant in June 1980 than in any previous
 summer  sample.

      After October  1974,  when E_. buccata was the dominant species, N.
 spilopterus  was consistently the most abundant species in the fall
 samples.  However,  an unusually large concentration of fish were found
 here during  the fall of 1976 and also included high densities of E_.
 buccata,  P.  notatus,  p. promelas,  C_.  anomalum,and S_. atromaculatus.
 All of  these  species except C.  anomalum were also common during the
 fall of 1977  and/or 1979.   Co-dominant species with N_.  spilopterus
 in  recent years included  £.  cepedianum during 1977 and N. stramineus
 and N.  cornutus-N.  chrysocephalus  during both 1977 and T979~]

 Station 29.
1976 - During March  - April  1976, more  than  eight times as many fish
were caught here than at nearby  Station 15.   This large concentration
of fish consisted primarily  of two-year-old  E_.  buccata and marked the
height of this species' population  explosion in the watershed.   A large
number (>200) of P_.  notatus, N_.  spilopterus,  ttf.  stramineus, and P.
promelas were also caught during this period.   The density of P."promelas,
although not very high in March, increased substantially in April when
N_. stramineus apparently moved downstream.   The density of fish declined
significantly during May - June  but E.  buccata  remained the dominant
species and P_. notatus, p_. promelas,and N. stramineus  were still common.
The number of E_. buccata rose again during July with the recruitment
of the 1975 year class and influx of adults  from upstream.   Pimephales
notatus and N. stramineus also exhibited slight increases in abundance
and recruits of C_. carpio and C.commersoni became common.   A significant
change in community  structure occurred  in August when,  in addition  to
E_. buccata, four other species reached  dominant status.   These  included
a school of young ID. cepedianum migrants from the Maumee River,
recruits of s_. atromaculatus and C_. cyprinus, and an influx of  P_. notatus
adults from upstream.  Evidence of  downstream movement  was also"
exhibited by p_.  promelas and may have been a  response to deteriorating
habitat conditions upstream brought about by  the severe  drought.
Notropis stramineus  and young C_. anomalum, C. commersoni,  and F.  notatus
were also common.  A marked decline in  the E_. buccata population  was

-------
                                -  192-


evident by September but it remained a co-dominant species with
S_. atromaculatus.  Although about 350 £. cepedianum were captured
here in August, none were found in the September sample, illustrating
the transient nature of this species in Black Creek.  Pimephales
notatus and £. cyprinus also exhibited substantial declines in density
but I?, notatus and P_. promelas were still fairly abundant.  Due to
sampling inadequacies, a November sample give an unreliable picture of
community structure, but documents the immigration of N_. spilopterus
and suggests that it was probably the dominant species at that time.

1977 - Relative to March - April 1976, the density of fish was much lower
during the early spring of 1977.  By May, community structure was more
comparable but notable  contrasts with the previous year were still
evident.  In 1977, for example, the abundance of IJ. stramineus increased
from March through May when it became the dominant species, but
declined through the same period in 1976.  The low numbers of E. buccata
and N._ spilopterus found during the entire spring of 1977 is even a more
striking contrast.  In addition to N_. stramineus, P_. promelas and
P_. notatus were also particularly abundant during May 1977.  Furthermore,
like N_. stramineus, these species along with S_. atromaculatus and IN.
umbratilis all increased in density from March through May,  The
community changed drastically in July when, coincident with an influx
of p_. cepedianum and recruitment of young C_. carpio and C_. cyprinus,
all cyprinids became scarce.  A fairly large number of N. spilopterus
immigrated in August but p_. cepedianum remained dominant and C_. carpio
and C_. cyprinus young were still common.  Many D_. cepedianum appear to
have left the watershed by the end of September but were replaced by
an immigrating school of N. stramineus which, along with N. spilopterus,
became the dominant species.  Although a devastating fish kill affected
this area a day after the  September sample, representatives of most
of the same species were found in November.  However, their densities
(particularly that of N. stramineus) were much lower.
1978 - Fish density remained rather low during  April  - May  1978  and,
as was the case during the spring of 1977, P_. notatus, N.  stramineus,
and P. promelas were the most abundant species  (though not nearly as
abundant as during the previous spring).  About 30  large  adult
C. commersoni were also caught during late April, apparently as they
were moving upstream to spawn.  The density of  P_.  notatus,  P_. promelas,
and N. stramineus remained the same during July and August when a similar
number of s_. atromaculatus adults and C_. commersoni and C_. carpio
recruits were also caught.  Consequently, during this  period the  eveness
component  of species diversity was high.  This changed  somewhat  during
November, however, when a much larger number  of fish were caught  than
during any of the prior sampling dates in 1978.  At that  time, S_.
atromaculatus, P_. notatus, and P_. promelas emerged  as  dominant species,
but N. spilopterus, E_. buccata, C_. anomalum,  N. cornutus-N.  chrysocephalus,
and C. commersoni  (recruits) were also abundant.

-------
                               - 193 -
1979 - Although the density of fish declined  in April  1979,  it  was still
considerably higher than during the previous  two  springs.  Pimephales
notatus and S_. atromaculatus remained the dominant  species but  P_.
promelas and N_. stramineus were also fairly abundant.   As was also the
case at stations 15 and 12, both young and adult  C_.  commersoni  were
also captured, providing evidence that some individuals of this species
overwintered in Black Creek.

Summary.

     Fish densities in the spring of 1976 approached the highest levels
ever recorded in Black Creek, but were not maintained  or duplicated on
subsequent sampling dates.  Furthermore, this unusually high density
of fish was largely due to a population explosion by E_.  buccata, which
was particularly abundant here prior to its decline in the watershed.
With the exception of E_. buccata' s numerical  dominance and  the presence
of a large number of IN. spilopterus in 1976,  the  structure of the fish
community was remarkably similar during the spring  of  each year,
Pimephales notatus, P_. promelas, and IN. stramineus  were the  most
abundant species but S. atromaculatus was a co-dominant in April 1979.
However, monthly variation in the density of  these  species was  somewhat
inconsistent between years and is likely attributable  to different flow
regimes.

     Relative to the next two years, the density  of fish was also
abnormally high during the summer of 1976.  However, in contrast to
the spring of that year, four other species in addition to E_. buccata
formed significant  components of this density.  These  included  P_.
notatus, I), cepedianum, and recruits of 
-------
                               -  194  -


Adult S_. atromaculatus and IS. connutus-  N.  chrysocephalus were also
common, though not abundant, in April and May samples.   The density of
fish doubled in September relative to May,  mainly due to the immigration
of N_. spilopterus.  Young L^. cyanellus,  S_.  atromaculatus, and adult
N_. connutus - N. chrysoc ephalus and E_. buccata were also common.

1975 - In contrast to the previous spring,  the most aburidant species
during March and April 1975 were  P_. notatus and E_.  buccata.  Notropis
spilopterus, NL cornutus- N_. chrysocephalus,  and one-year-old S_.
atromaculatus were also common.   The  density of I£.  spilopterus was
similar to that found during the  spring  of  1974 (ca.  25 - 35) .  All
species that were common in the spring declined in abundance during
May - June and were replaced by immigrating IN.  stramineus.  During
July, N_. stramineus declined in abundance and a large number (>100) of
E_. buccata fry were caught.  Pimephales  notatus and S_.  atromaculatus
were also common.  The fauna was  depleted by a fish kill in early
August and did not begin to recover until October when there was an
influx of P_. notatus, E_. buccata,  and  especially N_,  spilopterus.

1976 - Although E_. buccata was very abundant, the fish community remained
rather poor through the spring of 1976.   The only other common species
were N. spilopterus  (March) and P_. notatus  (March and May) .  Ericymba
buccata remained dominant through July when P_. notatus, P_. promelas,
N. stramineus, F_. notatus, and recruits  of  C^. carpio and C_. commersoni
also became abundant.  During August  the density of E_.  buccata declined
sharply and the community was dominated  by  a large number of S^.
atromaculatus recruits.  Pimephales notatusf F. notatus, and D_. cepedianum
were also fairly abundant during  this month, but the density of N_.
stramineus, C. commersoni and C^.  carpio  fell dramtically.  Ericymba
buccata, along with N_. spilopterus, became  dominant again  in October
as the density of S_. atromaculatus declined.  Caropostoma anomalum  and
F_. notatus were also fairly common during this month.

1977 - The only sample during 1977 was taken in September when N.
spilopterus and N_. stramineus were the dominant species.  Pimephales
notatus, P_. promelas, N_. cornutus- N. chrysocephalus, D_. cepedianum, and
C_. cyprinus were also fairly abundant.

1978 -  The  community was again  somewhat sparse during the  spring of
1978.   The most common  species  were  S_. atromaculatus, N_. cornutus  - N_.
chrysocephalus, P_. notatus,  P_.  promelas, and N. stramineus.  During
July the fauna was dominated by a mixture of both young  and adult
S. atromaculatus.  Pimephales promelas,  P_.  notatus, C_.  anomalum, N_.
cornutus -  N.  chrysocephalus  and C^.  commersoni recruits  were also  common.

1979 -  In contrast to all  previous years, a large concentration of fish
was  found here  in April  of  1979 but was primarily composed of  P_.
notatus.  Other very abundant  species included P_. promelas, N_. stramineus,
and  S_.  atromaculatus.   Catostomus commersoni was also abundant and
included both young  and large  adults.  Notropis spilopterus, N_. cornutus  -
N. chrysocephalus, E_.  buccata,  P_. maculata, E_. spectabile, C_.  anomalum,

-------
                               -  195 -
the end of April indicating  that  their emigration back into the Maumee
River was nearly complete.   During  March and early April all captured
S_. atromaculatus were one-year-olds;  however, older individuals were
caught later in April.   It is  therefore likely that many of the young
were leaving Black Creek as  the adults enter to spawn.  Pimephales
promelas, E_. buccata, IJ.  stramineus,  and N.  cornutus-N.  chrysocephalus
were also common during  the  spring  months.   A fish kill decimated the
fauna at the end of May  and  recolonization  proved to be very slow
during June.  A fairly large number of E_. buccata recruits (ca. 100)
were caught during July  but  no other  species was particularly
abundant.  The region was depleted  by another fish kill in early
August and recolonized by N_. spilopterus, N_. stramineus, P_. notatus, and
13. cepedianum.  The community was  still depauperate in September and
dominated by E). cepedianum.  In October only P_.  notatus and Ifl.  spilopterus
were common.

1976 - As in the spring  of 1975,  a  large number of fish were caught in
March-April 1976 and N;.  spilopterus was again a dominant species.
Pimephales notatus was also  abundant;  however, unlike the previous year
S_. atromaculatus was not abundant and N_. stramineus was a co-dominant
species during the early spring in  1976. Notropis umbratilis,  E_.
buccata, P_. promelasfand I,,  cyanellus were  also common during March-
April.  Notropis stramineus  remained  veryabundant during May while the
density of IS. spilopterus decreased significantly.  High densities of
13. buccata and S_. atromaculatus were  found  in late May and N. umbratilis,
P_. promelasfand L^. cyanellus remained common.  With the apparent
emigration of N_, stramineus  and adult S_. atromaculatus during June, E_.
buccata   emerged as the dominant species.   The density of N_. umbratilis
also decreased during this month  but  eight  other species, including
recruits of (?. commersoni were common.   The major changes in July were
a marked decrease in density of P_.  notatus,  P_. promelas, E_. buccata,
N. stramineus, N. spilopterus,  and  L_.  cyanellus, and the recruitment of
young C_. carpio, I_. natalis, and  C_. commersoni.   Young S_. atromaculatus
were also common.  Recruits  of £. commersoni and !_. natalis increased
in abundance in September.   This  sample also consisted of large numbers
of £. anomalum and both  young  and adult S_.  atromaculatus and a fair
number of P_. promelas and P_. notatus.   October was marked by the
immigration of large numbers of t£.  stramineus and N. spilopterus.  Over
100 individuals of S_. atromaculatus,  E_.  buccata, and P.  notatus were
also caught.  P_. promelas and  C_.  anomalum remained common but most
I_. natalis and C_. commersoni appeared to have emigrated into the Maumee.
It is likely that many of the  S_.  atromaculatus,  P_. promelas,  P_. notatus,
and E_. buccata found here during  September  - October were seeking
refuge from severe drought conditions in upstream tributaries.

1977 - Possibly as a result  of a  severe winter,  only about half as
many fish were found here during  the  spring of 1977 as were caught at
this time during the two previous years.  Large  numbers of N. spilopterus,
in  particular, were notably absent.   The dominant species during April
and May were P_. promelas  and P_. notatus,  respectively.   Also common
during these months were  S_.  atromaculatus,  N_.  stramineus, N.  umbratilis,

-------
                               - 196 -
N_. umbratilis, and L. cyanellus were common.  The dominant  species  during
October were N. stramineus and P_, notatus.  Also abundant were  S_.
atromaculatus, N^. cornutus - K[. chrysocephalus, N_.  spilopterus,  T3.
umbratilis, and especially P. promelas.

1980 - During June 1980 Ifl. stamineus was again dominant but N_.  cornutus  -
N. chrysocephalus, P. promelas, and P. notatus were also abundant.
Other common species included S_. atromaculatus, F\  notatus  and  E_.
buccata.

Summary.

     Up until recently, the fish fauna was  generally rather sparse  during
the spring months and rarely consisted of large numbers of  any  species.
Pimephales notatus and IS. buccata  (prior to its decline in  the  watershed)
were the most abundant species during this  period and S_. atromaculatus,
N. cornutus - N^. chrysocephalus, N_. spilopterus, and N_. stramineus  were
common.  Fish were much more abundant in the  spring of 1979 when P_.
notatus was the dominant  species.  The density of P_.  promelas,  S_,
atromaculatus, and N. stramineus was also significantly higher  in April
1979 than during any of the previous springs.  In addition, the presence
of a large number of both young and adult £.  commersoni suggests that
perhaps for the first time during  the course  of sampling,  a small
population of this species may have overwintered in Black  Creek.

     Analysis of long-term trends  in community structure during the
summer months is hindered by infrequent  sampling from 1977  to 1980
and the fish kill that affected this area in  1975.   The most  abundant
species during this period were N_. stramineus, both young  and adult
S_. atromaculatus, P_. notatus, P_. promelas,  and on limited  occasions
D_. cepedianum, F_. notatus, and  recruits  of  C. commersoni  and  C_. carpio.

     During the  fall months, N_.  spilopterus was consistently  one of the
most abundant  species but never  reached  particularly high  densities
 (e.g.  >200).  However, the fall density  of  IN.  stramineus  increased
steadily from  1977 to 1979 when over  500 individuals were  captured.
This trend  coincides with the population decline of E^. buccata - the
dominant species during  the  fall of  1976.   Pimephales notatus was also
relatively  abundant  during most  years  and emerged  as a co-dominant
species with N_.  stramineus in  October  1979.  Other  common  fall species
included S_. atromaculatus, P_.  promelas and  N_.  cornutus -  N_. chrysocephalus.

Station 12.

1973  - 1974 -  Initial  samples  in July 1973  and April and July 1974 did
not yield  a very large number  of fish (90-130)  and  no species was
particularly  dominant  on any of the  three dates.

1975  - During March  -  April  1975 there was  a large concentration of fish
 (500  - 1000)  consisting  primarily  of 13.  spilopterus, P_.  notatus, and  S_.
atromaculatus.   The  number  of  N.  spilopterus decreased significantly  by

-------
                               - 197 -

I,, cyanellus, JL. macrochirus, and ~L. microlophus.   The  density of fish
declined considerably during June, probably in response to  a dense algal
bloom affecting this region.  Both Pimephales species,  N_. stramineus,
and N_. umbratilis showed significant decreases in  abundance and no
species was dominant.  July was marked by the recruitment of young
(2. carpio and £. cyrpinus and the immigration of D_.  cepedianum.
Notropis stramineus and N_. spilopterus invaded this region  along with
more D_. cepedianum during August.  By September, N.  stramineus was the
dominant species but P_. promelas, P_. notatus, IX cepedianum, N_. spilopterus,
and L. cyanellus, and recruits of C_. commersoni  and C_.  cyprinus were all
common.  The species composition was similar in  November but N. stramineus,
P. promelas, P. notatus, and C. cyprinus were considerably less abundant.

1978 - As in 1977, large number of II. spilopterus  and fish  in general
were again not found during the spring of 1978.  Community  structure was
also very similar to the previous spring with P_. promelas,  P_. notatus,
N_. stramineus, N_. umbratilis, and N_. cornutus-N. chrysocephalus the most
common species.  However, the sunfish  (Lepomis spp)  were considerably
less abundant than they were in 1977.  Two samples in July  clearly
illustrate the vicissitude of the Black Creek fish fauna.   During the
first week in July a large number  (>400) of N. stramineus were caught
along with a fair number of E_. buccata  (47) , P_.  notatus (64) , IT. notatus
and recruits of C_. carpio, C. commersoni, and C_.  cyprinus.   At the end of
July the abundance of the latter four species remained about the same;
however, only 12 P. notatus, 1  E_. buccata,and no  N.  stramineus were
caught.  In addition, a small number of 1^. natalis and I_. me las recruits
were found and P_. promelas and S_. atromaculatus  showed significant
increases in density.  In contrast, with the exception of an influx of
a few EK cepedianum and a decrease in the abundance of C_. commersoni
and C_. cyprinus young, the fish community showed little change through
September.  However, a tremendous number of fish,  especially P_.
notatus recruits, were found in November.  Pimephales promelas, S_.
atromaculatus  (recruits), N. spilopterus,and N.  umbratilis  were abundant
and N_. stramineus and N_. cornutus-N. chrysocephalus were common.  This
was the first year that such a large number of young (particularly P_.
notatus, P_. promelas, and N. umbratilis) were found here during the fall.

1979 - The species composition and density was similar in April 1979,
except S_. atromaculatus, N. umbratilis, and N_. cornutus-N.  chrysocephalus
were not as abundant while the number of F. notatus and ]L.  cyanellus
increased slightly.  The presence of young !_. melas and C_.  commersoni
is also noteworthy for this time of the year.  It  appears that for the
first time young C_. commersoni overwintered in Black Creek.  The density
of fish, particularly P_. notatus, P_. promelas, and  N.  stramineus dropped
precipitously in June.  However, relative to this  date during previous
years, there was an unusually large number of N_. spilopterus, N. cornutus-
N_. chrysocephalus, and adult C_. commersoni present.   In addition to
these species, S_. atromaculatus, F_. notatus, and N.  umbratilis were
fairly common.  During October, ttf. spilopterus was the dominant species
but largenumbers (>100) of !N. stramineus and young of P_. notatus and
N. umbratilis were also caught.  Other common species included S_. atromaculatus,
N. cornutus-N. chrysocephalus, F_. notatus, and L.  macrochirus (young) .

-------
                               - 198 -
1980  ~ During June 1980 the dominant  fish were N. umbratilis  and N.
stramineus.  Pimphales notatus, IJ.  spilopterus,  F.  notatus, and cT
commersoni were common.

Summary.

     Limited sampling during 1973 and 1974 yielded  considerably less  fish
than were usually caught at this station from 1975-1980.   During these
latter years fish density was generally high in  the spring, lower in  the
summer, and high again in the fall.   Lower densities than  usual were
encountered during the spring of 1977 and 1978,  possibly due  to heavy
mortality during the severe winters that preceded these samples.   Large
numbers of N. spilopterus, a dominant species during other springs, were
notably absent in 1977 and 1978.  Other abundant species found  during the
spring months (March - May) include P_. notatus,  P.  promelas,  S_.  atromaculatus,
N. stramineus, and E_. buccata.  The degree of dominance exhibited by  these
species and NL spilopterus varied from early to  late spring as  well as
between years.  Other fairly common species found during the  spring months
include N. umbratilis, N[. cornutus-N.  chrysocephalus,  L_. cyanellus, and
less frequently, L_. macrochirus, !L. microlophus, and F_. notatus.   The summer
months (June - August) were characterized by the presence  of  numerous
recruits and a decrease in abundance  of most species that  were  common during
the spring.  The decline in density of N_. spilopterus  and  N.  stramineus
is particularly striking as these species migrate out  of Black  Creek  and
back into the Maumee River.  The dominant species during the  summer include
recruits of C_. carpio, C_. commersoni,  C_. cyprinus,  I_.  natalis,   I. melas,
and S_. atromaculatus.  In addition, large numbers of D_. cepedianum young
commonly migrate into Black Creek during July or August.   Another
significant change in community structure occurs  in  the late summer or fall
and is marked by the immigration ofN.  spilopterus and  N. stramineus.
Other abundant species found during the fall include P_. notatus,  P_.
promelas, D_. cepedianum, S_. atromaculatus, and less  freqeuently,  EL buccata,
C. anomalum, C_.  commersoni (young) , !_. natalis  (young) , C_.  cyrpinus
(young) , N. cornutus- N. chrysocephalus, NL, umbratilis (young) ,  L_.
cyanellus, and L_. macrochirus  (young) .  The presence of numerous young
—• notatus, P_. pomelas, L_. macrochirus, and especially N^.  umbratilis  is
a recent  (1978-1979) occurrence.  An  increase in the abundance  of
F_. notatus  (especially during the summer months) and the presence of
young C_. commersoni during the spring are also noteworthy  recent
developments.

                   SPECIES RICHNESS AND DIVERSITY

     As in the previous section we describe the  fish fauna from the
headwaters to downstream areas.  Species richness is the number of
species in a sample while diversity is indexed by the  Shannon-Weiner
function  (see Methods).

Station 6 - 26 - During most years this station  supported  an  average  of
7-8 species; however, species richness was much  lower  in 1974 after
channelization  (Fig. 11).  It was also above normal in 1977 when 14 - 19

-------
    20
i_ 4)
     10
     0
                            JBl
         II I I I III I I I I I
         A   N I  F   M   A   Nip   M'A   N I  F  M  A   NlpM'A'"Nlp   MA  Nip  M  A
   'i i i i i|i i rriT'rrn i IP i fnVri i i i i|i i i i i i i i i i i ip i i i 11 i i i i i mil i i i i i i t \ m i i i r
                                                                   N I  r   M
                    I I I I I I III I I I M I I I I I III I I I I I M I T I III I I I I I I I I I I III I I M I I I I I I III I I I I I I I I I I III I I I I
         A  N!F  MA   N'F  MA   Nip   MA   N'F   MA   NIFMAN'F   MA   N'FM
         1973
1974
1975        1976        1977

           Sample Date
1978
1979
1980
                                                                       Pelagic Species

                                                                    t^j] Benthic Species





                                                                    STATION 6-26
                                                                                                  —A All Species
                                                                                                  	• Benthic Species
                                                                                                  	O Pelagic Species
    Figure 11.  Number of species  and species diversity for fish  at Station 6-26 in the Black  Creek

                Watershed, 1973-1980.   Station 26  sampled February - May 1976  only.

-------
                               - 200-

species were caught on all six sampling dates from May through November.
No seasonal trends were discernable and the species composition was
generally evenly divided between benthic and pelagic forms.  The most
common benthic species were P. promelas, P_. notatus, E_. buccata, C_.
carpio, and C_. commersoni.  The most common pelagic species were S_.
atromaculatus, N. cornutus - N. chrysocephalus, N_. umbratilis, N_.
spilopterus, and .L. cyanellus.

     Species diversity was maintained between 1.2 and 1.9 during about
70% of the  sampling dates but fell to fairly low levels  (<1.0) during
May and July 1974, April 1975, April and October 1976, July and
September 1978 and June 1980.  The highest levels of diversity (2.0 -
2.2) occurred during May and June 1977, owing primarily to the relatively
large number of species present.  There were clearly no definitive
seasonal patterns.  The diversity of benthic and pelagic species was
remarkably similar during most sampling  dates and henae mirrored the
total species diversity curve.  Extremely low levels of species diversity
were either due to low numbers of fish present  (April and July 1974,
April 1975, and October 1976) or dominance by a benthic species  (P_.
promelas - 22 April 1976 and 26 June 1980, £. carpio - July and September
1978).

Station 18.

     Infrequent sampling after 1977 hinders long-term evaluations of
species richness patterns at this .station, but  it appears that the number
of  species present has increased in recent years  (Fig.  12).  Although
species richness was very low during the spring of 1978, 11-15 species
were caught here during every other sampling date from  1977  - 1980.
During extensive sampling in 1975 and  1976 an average of only about 8
species/sample were caught.  During most sampling dates there were
slightly more benthic species than pelagic species.  The most common
benthic species were P_. notatus, P_. promelas, E_. buccata,  and N^.
stramineus.  The most common pelagic  species were S_. atromaculatus,
N.  cornutus - N. chrysocephalus, N. spilopterus, N. umbratilis,  and
F.  notatus.

     Species diversity fell below 1.0  during July 1974, September  -
October 1975  and September - November 1976 and  reached  its highest  levels
 (1.8 - 2.1) during 1977 and  1979.  From  1974 -  1976 benthic diversity  and
pelagic diversity  differed markedly,  but were  fairly  similar from  1977-
1980.  Relative  to the downstream stations benthic diversity was more
consistent and lacked distict  seasonal peaks.   Low levels  of benthic diversity
were  caused by the dominance  of E_. buccata  (July  1974  and  September  -
October 1975) and  P_. promelas  (June 1980)  or low  densities of benthic
fish  (September  1976 and  March  1978).  Extremely  low  levels of pelagic
diversity  recorded during 1974  - 1976 were due  to low pelagic species
richness and dominance by N.  spilopterus during  the fall months.   Given
the low numbers  of individuals  and  species found  here,  it  does  not appear
that this  station  provided  very good  habitat for  pelagic  species (at
 least  prior to 1977)  and accounts for  the dramatic fluctuations  in pelagic
diversity.   The  greater stability exhibited  by pelagic diversity measures

-------
o „
      20
Is.   I0
n 
-------
                               - .202 -


in recent years appears to be due to improved conditions  for pelagic
species and has contributed to higher levels of total  species diversity.

Station 17 - 28.

     Species richness at Station 28  proved to be considerably higher  than
in Station 17  (Fig. 13) but could have been due to the habitat  alterations
that occurred during the first two years of sampling.   Species richness  at
Station 28 ranged from 9-17 species with 17 species  being captured  on
three separate occasions (8 August 1977, 19 April 1979, and 28  June 1979).
Average species richness during 1976 appeared to be lower than  during
each of the following years and there was a slight increase in  the number
of species found during the summer months of 1976-1978.   During most
sampling periods benthic species were more numerous than  pelagic  species.
The most common benthic species were P_. promelas, P_. notatus, £.  buccata,
£. anomalum, N. stramineus, C_. carpio, £. commersoni, £.  cyprinus,
—• "igrum, E_. spectabile, and £. cepedianum.  The most  common pelagic
species were S_. atromaculatus, N^. cornutus - N. chrysocephalus, N.
spilopterus, N. umbratilis, F. notatus, and L. cyanellus.

     Species diversity at Station 28 was consistently  above 1.4;  however,
much lower levels were recorded at Station 17 during 1974,  The highest
level of diversity  (2.41) occurred on 24 August 1977.  Benthic  diversity
basically mirrors the total species  diversity curve, but  exhibits more
distinct peaks during the summer months of 1977 and 1978.  These  high
levels of benthic diversity were brought about by the  influx of D_. cepedianum
and recruitment of (2. commersoni, (X carpio, and £. cyprinus.   In contrast to
benthic diversity, pelagic diversity showed rather sharp  and erratic  fluctuations.
This can be attributed to the paucity of the pelagic fauna  (at  least
through 1976) and shifts in dominance by S_. atromaculatus and N_.  spilopterus.
Although limited sampling in 1978 would indicate otherwise, the diversity
of the pelagic fauna appears to have improved in recent years.

Station 29.

     During each of the three years  of intensive sampling (1976 - 1978)
at this station species richness patterns were remarkably consistent,
ranging from 10 - 18 species/sample  with an average of about 13 species
(Fig. 14).  There was a clear seasonal trend with the number of  species
present, particularly benthic species, increasing from spring to  summer
and decreasing again during the fall.  This was due to the recruitment
of young C_. commersoni, £. carpio, and £. cyprinus, and the influx of
D_. cepedianum.  The most common benthic species found  here were P_.
promelas, P_. notatus, E_. buccata, £. anomalum, N^. stramineus, C_.  carpio,
—• commersoni, £. cyprinus, and I), cepedianum while the most common
pelagic species included S_. atromaculatus, N_. cornutus -  N_. chrysocephalus,
NL spilopterus, N^. umbratilis, F_. notatus, and L. cyanellus.

     As was the case at nearby station 15, species diversity was  relatively
low during the first half of 1976, but rose sharply in August 1976 and
was maintained above 1.7 thereafter.  The highest level  (2.37)  was
recorded on 16 August 1978.  Total species diversity measures also show a

-------
      20-
 0) "o

|«   10

 3(/>
   2.4


^ 2.0


    1.6


    1.2
-



M
L
rn

i i i i in i i i i M in TT
A N ' F M A N


T'
-

m

7

1 1 1 1 1 1 1 1 1 1 1 1
F M A N



-

7
-
TT

_
[



r.

-.

pi

1 1 1 1 1 1 1 1 1 1 1 III 1 1 1 1 1 M 1 1 Ml 1 1 1 1 M T 1 1 IT
FMAN'FMAN'FMAN
_.
i m TTT
F M
  5
   o>
   o 0.8
   o>

  co 0.4-
^8
           I I I I  I I MM I I I I I I I I I I III I I I I I I I I I I II I I I I I I I I I I I I II I I I I I I I I I I  III

           MA  N'F   M   A   N'F   MA   N'F   M   A   N'F   M  A  N1
                                                             i i i i l MI l i i i i i i i i  i ill i i i i i
                                                              A   N'F   M   A  N'F   M

         1974        1975        1976        1977        1978         1979      1980
[	I Pelagic Species



|H Benthic Species







STATION 17-28
                 i


                o


	A AM Species    '

—•• Benthic Species

--O Pelagic  Species
                                          Sample  Date
Figure 13.  Number of species and  species diversity for fish at Station 17-28 in the Black  Creek


            Watershed, 1973 1980.    Station  17  sampled in 1974 - 1975 and Station 28 sampled


            from 1976 - 1980.

-------
        20
  i_ 0)
"fS.10
 3CO

        0
             • i 11 111 i  i ill i 11  11 11  11 11111 i
               M   A  N'F  M   A   NlF   M   A   N'F
                                                 lllMIIII
                                                                     Pelagic Species


                                                                     Benthic Species
      2.4


  £2.0


   i   1.6


  S   1.2

   (D
   o  0.8

   O.
  CO  0.4
         OL
               M  A   N'F   M   A  N'F   M  A   N'F
                                      Ii i  i i i i i  i i i i 111 i ir
                                       P   M  A   M I  P
               1976
                            1977          1978
                              Sample  Date
1979
                                                                 STATION  29
               All Species
               Benthic Species
             -O Pelagic Species
                                                                                     NJ
                                                                                     O
Figure 14.  Number of species and species diversity for fish at Station 29 in the Black Creek Watershed,

         1973-1980.

-------
                               - 205 -

clear seasonal pattern but benthic diversity and pelagic diversity curves
are more revealing.  For example, owing primarily to an increase  in
species richness, benthic diversity exhibited a distinct peak in  August
of each year from 1976 - 1978.  The low levels recorded from March
through July 1976 were due to extreme dominance by E^. buccata.  In
contrast, pelagic diversity reached its highest seasonal levels a month
or two earlier than benthic diversity during a "turnover" period  when
there was a change in the dominant species within the pelagic fauna.  For
example, high levels of pelagic diversity were recorded when there was
a shift in dominance from N_. spilopterus in April 1976 to S_. atromaculatus
in August 1976, and from  S_. atromaculatus and JN. umbratilis in May 1977
to N^. spilopterus in August 1977.  Although no pelagic species was
dominant in early 1978, pelagic diversity rose with an increase in species
richness in late April but fell again when S_. at r omacu 1 at us became
dominant in July.

Station 15.

     Twenty-two species were found at this station when it was sampled
with an electric seine on 20 April 1979  (Fig. 15); however, during years
when more than two samples were taken, an average of only about 9-11
species was caught.  In fact, species richness was consistently low from
October 1974 through the spring of 1976, and during each of the three
sampling dates in 1978.  A seasonal trend, marked by an increasing number
of benthic species occurring during the summer months, was evident during
1976 and 1978.  This was due to the immigration of £. cepedianum, and
recruitment of young £. carpio , C_. commersoni , C_. cyprinus and during
1976, E. nigrum and E_. spectabile.  In addition, the number of benthic
species usually exceeded the number of pelagic species.  The most common
benthic species were P_. promelas , P_. notatus, IS. buccata , C_. anomalum ,
—' stramineus, and C. commersoni and the most common pelagic species were
S_. atromaculatus, N_. cornutus - 14. chrysocephalus , N. spilopterus ,
N. umbratilis, L. cyanellus and F_. notatus .

     Species diversity was fairly high during the initial  sampling in
1974 but declined precipitously from June 1975 until 8 April 1976 when
only three species were caught and diversity measured Q.39.  Thereafter
species diversity returned to previous levels and a high of 2.20 was
recorded on 7 October 1976.  Both benthic and pelagic diversity reflected
the total species diversity curve through its decline in early 1976;
however, during the recovery period benthic diversity generally exceeded
pelagic diversity and was largely responsible for the shape of the total
species diversity curve.  Low levels of pelagic diversity  from 1976 -
1978 were caused by the dominance of S. atromaculatus and  N_. spilopterus
over a rather sparse pelagic fauna.  Note that the large number of
species (22)  captured on 20 April 1979 did not produce a particularly
high level of species diversity, reflecting extreme numerical dominance
of P_. notatus .

Station 12.

     Relative to 1973 - 1975 there was a definite increase in species
richness at this station from 1976 - 1980 (Fig. 16) .  The  fewest

-------
^ o> 20
.a °
E Q. 10
z
0
-
~
— •



-



1
1






] ri
U^



'1



/Ml EUji

r




^



-







-











r



J



]R
^












«
-

-t












          M  A  NlF  M   A   NTp  M1 A " N T F " M " A " ^'^F " ^' ' A '  '^ T ^ ' ^ ' ' A ' '^ T ^
                                                                                     M
1975
                                  1976         1977        1978
                                          Sample Date
       N'''F ' 'M1
1979      1980
                                                                                             I	I Pelagic Species

                                                                                             h]J Benthic Species
                                                                                             STATION 15
                                                                                                               O
                                                                                                               •O1
                                                                                            —A All Species      (
                                                                                            	• Benthic Species
                                                                                            —O Pelagic Species
Figure 15.  Number of  species  and species  diversity for fish at Station 15  in  the  Black Creek Watershed,
            1973 - 1980.

-------
o w 20
0) 'o
£ Q. 10
z o
-
-
-
-,
T

-

-
a
' Tl M 1 III 1 I'l'l I'l'l 1 1 1 III I'VI
A N 1 F M A N 1 F
-.
"-
-
M
ri
r
-r

J-
-. \




l-i
-


-
.-
_

-
'
-1



-
^


-
-


till III 1 1 1 1 1 1 1 1 1 1 III 1 1 1 1 1 IT TIM
A N'F M A N 1 F M A N

-n L


III
F M

Fh



-

A N


-









~
-


F M A N
(-,

7
F M
>»
35
k.
>
15
V)
F
 A  N
 1973
                  M  A
                  1974
                             1975
1976        1977
   Sample  Date
I I M M M III I
 M   A  N'F
 1978
                                                                               MAN
                                                                               1979
F  M
 1980
                                                                                           Fj Pelagic Species

                                                                                           f^j Benthic Species



                                                                                           STATION 12



                                                                                           —A All Species
                                                                                           	•Benthic Species
                                                                                           —O Pelagic Species
                                                                          I
                                                                          10
Figure  16.  Number of species  and species  diversity  for fish at  Station 12  in the Black  Creek
             Watershed,  1973-1980.

-------
                               - 208 -

species were caught during August 1975 - October 1975, largely due to
the fish kill that affected this area in early August 1975.  However,
during 1976 and each of the following years an average of about 15
species was caught and on three occasions (23 August 1977, 20 April
1979, and 26 June 1980) 20 species comprised the samples.  In addition,
during 1976-1978 when intensive sampling was terminated, a clear
seasonal trend emerged with more benthic species caught during the
summer months.  However, in general, pelagic species slightly exceeded
the number of benthic species at this station.  The most common benthic
species captured were P_. promelas, P_. notatus, E_. buccata, N^ stamineus,
C. commersoni, £. cyprinus, C. carpio, and IX cepedianum while the most
common pelagic species included S_. atromaculatus, N^. cornutus - .N.
chrysocephalus, N_. spilopterus, N. umbratilis, F_. notatus, L. cyanellus,
and I,, macrochirus.

     As with species richness, species diversity reached its lowest
levels at this station during 1975 and the effect of the August fish kill
during that year is clear. Thereafter, about 78% of the sampling dates
had diversity measures above 1.7.  Species diversity also showed well
defined seasonal patterns with high levels occurring during the summer
months contrasting with the low levels found during the spring and fall.
Benthic diversity closely paralleled the total species diversity curve
but shows an even more pronounced seasonal pattern.  This trend was
primarily due to the recruitment of young C_. commersoni, £. carpio,
C. cyprinus, I_. natalis, and I_. melas, and the immigration of IX
cepedianum during the  summer months, and the numerical dominance of
either E_. buccata, N.  stramineus, or, most commonly, IP. notatus during
the spring and fall.   Pelagic diversity also exhibited a seasonal pattern
but generally had highs a month or two in advance of the benthic diversity
curve.  During these pelagic diversity peaks the pelagic fish fauna had
a more equitable species composition and could be considered more
stable than before and after these periods when the pelagic fauna was
usually dominated by migrant N. spilopterus or S_. atromaculatus.

Summary.

      Average species richness was highest at Station 12 and decreased
steadily upstream.  The maximum number of species caught on a single
sampling date was 22 at Station  15 on 20 April 1979.  Relative to the
first few years of sampling, species richness appears to have improved
during recent years, particularly at Station 12  and 28.  Long-term
trends of this nature  at the upper regions of the watershed  (e.g. ,
Stations 6 and 18) are obscured by the overriding influence of annual
variation in flow regimes.  Seasonal changes in  species richness
occurred at Stations 12, 15, 29, and 28 and were largely attributable
to the influx of migrant individuals of D. cepedianum from the Maumee
River and recruitment  of young C. carpio, C_. commersoni ,and C_. cyprinus
during the summer months.

      In general,  species diversity at Stations 12,  15,  29, and 28 was
similar, an<^  consistently  higher than at  Stations 18  and  6, where diversity

-------
                               - 209 -

frequently fell below 1.0.  The highest diversity level at each station
was recorded during or after 1976 and ranged from 2.08 at Station 18 to
2.41 at Station 28.  The benthic guild generally exerted major control
over patterns of species diversity, particularly seasonal trends.  Summer
peaks in species diversity observed at Station 12, 29, and 28, for
example, reflected recruitment and immigration of benthic species.
Furthermore, low levels of diversity were most often due to dominance by
a single benthic species.  Seasonal trends in pelagic diversity also
occurred at Stations 12 and 29 but had peaks a month or two in advance
of benthic diversity during "turnover" periods when dominance relationshps
were shifting within the pelagic fauna.  However, up until very recently
when pelagic diversity has shown signs  of marked improvement, the
pelagic guild, with the exception of S_. atromaculatus and 14. spilopterus
has formed a weak component within the Black Creek fish community.

                    SHORT-TERM TEMPORAL VARIABILITY

     Mean values of sample similarity for all main channel stations from
1974 through 1978 were considerably lower from June - July through
August - September (Figure 17).  In addition, similarity values for
samples taken from June through October appear to be highly variable.
Since these measures include samples taken at seven stations over a
five-year period, a good deal of variability should be expected, but
it does not explain the apparent seasonal trends.

     In view of the asynchronous changes in diversity displayed by
benthic and pelagic species and detailed in a previous section, further
analysis of short-term variability in community structure was augmented
by calculating independent sample similarity values for these guilds.
The separate indices then, were based upon the proportional
representation  of a given species within its respective guild and as
such, could behave somewhat differently both from one another as well
as from the index that is based upon the entire community.  This not
only facilitated more accurate determinations of the nature and
underlying causes of temporal variability patterns but also permitted
detection of more subtle changes in community composition.

     Mean values of sample similarity for both the benthic and pelagic
guilds proved to be very similar to measures based on the entire
community and were clearly within one standard deviation of that
index (Fig. 17).   However, the mean similarity value for pelagic
species during August-September did not follow the general seasonal
pattern displayed by the benthic guild as well as the overall community.
Furthermore, from March through October temporal differences in the
mean similarity index for the pelagic guild were relatively small,
ranging from .528 to .634 while those of benthic species ranged from
.509 to .742.   In addition, during the spring months mean similarity
values for benthic species were consistently higher than those of
pelagic species,  indicating that the benthic guild was more stable during
this period.   There were also differences between benthic and pelagic
guilds in the variability of their respective similarity indices (Fig.
18).   Similarity values for pelagic species, for example, were more

-------
         i.o-
        0.9-
    ^  0.8H
        0.7
    ^  0.6-
    CO
0.4-
    <  0.3-
    GO
    _  0.2-1
    UJ
         0.1-
          0
                                                   All species
                                              •—•  Benthic species
                                              o~-o  Pelagic species
       —,	,	1	1	1	1	1	1     I     I    I     r-
        JAN FEB  MAR APR  MAY  JUN  JUL  AUG SEP OCT  NOV DEC
                                                                                     i
                                                                                     S3
                                                                                     H
                                                                                     O
                                                                                     I
                              SAMPLE  PERIOD
Figure 17.  Mean sample similarity values for all species  combined (+ SD) between months
         for main channel stations in the Black Creek Watershed, 1973-1980.

-------
            0.40-1
            0.35-
     O ^  0.30
     5
0.25-
     UJ h-  0.20
     Q ££
        <  0-15
     or =!
            O.I OH
            0.05-
    Q 7^
    CO
                                         •—• Benthic species
                                         o—o Pelagic species
                     JAN FEB  MAR  APR MAY  JUN  JUL  AU6  SEP OCT  NOV DEC

                                   SAMPLE    PERIOD
                                                                       i
                                                                      H
                                                                      M
                                                                       I
Figure 18.  Standard deviations of similarity values for benthic and pelagic guilds at main channel

         stations in the Black Creek Watershed, 1973-1980.

-------
                               - 212 -
variable than those of benthic species during March-April, August-September,
and September-October.  However, during June-July, sample similarity
values among benthic species were more inconsistent than those of pelagic
species.

     In order to further dissect these patterns, significant changes
in community structure, arbitrarily  defined as changes resulting in
sample similarity values less than or equal to 0.50, were examined in
detail  (Tables 4 and 5).  Temporal variability within the benthic guild,
at least according to the categorization criteria used here (see methods)
was clearly of greater significance to overall community structure than
were changes among pelagic species.  Of the 25 significant changes in
community structure that were detected from 1974 through 1978, 13 were
due to short-term changes within both the benthic and pelagic guilds,
10 involved primarily benthic species, and only 2 were solely attributable
to pelagic species.  During 18 other periods, significant changes were
measured within the pelagic guild while sample similarity values for the
community exceeded 0.50.  The frequency of these changes as well as their
apparent insignficance was due to the low densities of pelagic species
that were often encountered.  The similarity index is particularly
sensitive to changes in the relative abundance of species when densities
are low.

     The relative frequency of "significant" changes in the composition of
benthic species (Table 6) largely reflected the temporal patterns
displayed by the mean similarity values of this guild.  For example, the
frequency of these major changes ranged from 50% to 57% for sampling
periods from June through October but did not exceed 21% during the
other months.  However, note that the frequency of these major changes
during  September - October  (57%) was higher than what might be expected
based upon the mean similarity index of the benthic guild for this
period.   There was less congruence between the mean similarity index
and the frequency of major changes within the pelagic guild.  Among these
species, similarity values less than or equal to 0.50 occurred most often
from June through August and during March-April and September-October
(Table  6).

     There were also considerable differences  in the frequency of major
changes within the benthic guild between years  (Table 7).  Temporal
stability among these species appeared to be relatively low in 1974, 1975,
and 1977 and high  in 1976 and 1978.  In addition, during 1974 and 1975
major changes in benthic guild structure occurred during the spring as
well as the summer and  fall months, but were limited to June through
October from 1976  through 1978.  Although the relative frequency of major
short-term changes within the pelagic guild was similar between years,
low sample similarity measures were more numerous during 1976  (Table 7).

     Factors contributing to these changes  in community structure include:
fish kills, algal  blooms, flow regimes, channelization and other habitat
perturbations, fish migrations and within-stream movements, recruitment,
and natural mortality  (Table 4 and 5).  Note that each of these factors

-------
        Table 4,
Changes in benthic guile1 resulting in similarity values (PS) less than or equal to 0.50.  Code
numbers show the relationship between changes within this guild and overall community structure.
1)  Indicates that short-term shifts in overall community structure were primarily due to changes
in species composition within this guild.   2)   Indicates that short-term shifts in overall
community structure were due to changes in species composition within both the benthic and pelagic
guilds.  3)  Indicates that changes within  this guild had a relatively small impact on overall
community structure.
Sampling Period

February-March
 Year
Station
  PS
Code
Nature of Change
March-April
April-May
 1974
                   1974
  15
.478
                  .115
             Low fish densities during both sample periods and some
             sampling problems due to high water in May.  The major
             change involved the presence of large, adult C_, commersoni
             during their spawning run in April and their absence in way.
             Due to the scarcity of fish following channelization in early
             May.

May-June
1975
1975
1975
12
6
15
.258
.444
.366
1
2
1
Due to a fish kill that decimated the fauna in May.
Absence of fish in April due to a weir below the station which
prevented fish from colonizing this area during prevailing low
flow conditions.
Immigration of a fairly large number of N, stramineus during
                                                                                              I
                                                                                              ro
                                                                                              (->
                                                                                              CO
                                                                                              I
                   1978
           18     .496
                              June.

                              Movement of P,  notatus and E.  buccata out of this area in June,
                              possibly in response to the immigration of a large number of
                              N.  stramineus.

June-July


1974

1975
17

12
.219

.237
2

2
Recruitment
in July.
Recruitment
of

of
one-year-old

one-year-old
E.

E.
buccata

buccata
into the

into the
population

population in
                                                         July.   This area was also still recovering from the May fish
                                                         kill.

-------
          Table 4  (continued)
Sampling Period    Year    Station    PS

June-July          1975      15      .249
                         Code
                   1975
                   1977
                   1977
          18     .416
          12     .133
                 .305
                                                    Nature of Change

                            Emigration of N_.  stramineus and recruitment of one-year-
                            old E_.  buccata in July.  Algae may have also contributed
                            to low densities  of some species.

                            Recruitment  of one-year-old E_. buccata and emigration of
                            IN. stramineus in  July.

                            Recruitment of young C_, carpio and C_. cyprinus—and
                            immigration of D_. cepedianum in July.

                            Recruitment of young C. carpio and, to a lesser extent, the
                            emigration of NL  stramineus in July.
July-August
1975
                   1975
                   1975
                   1976
12
.436
          15     .264
                 .445
          15
       .320
                   1976
          29
       .400
A fish kill decimated the fauna  in  early  August  but  there  was
an influx of 13. cepedianum later this month.
A fish kill decimated the fauna  in  early  August  but  there  was
an influx of D. cepedianum later this month.
Fish densities were low during the  period apparently due to
algae problems.  A slight increase  in the abundance  of  P.
notatus accompanied a decline in the number of P_. promelas
during August.
Sharp decrease in the density of E_. buccata in August,
possibly reflecting heavy mortality sustained by this
species during the prevailing drought.  In addition,  a  fairly
large number of C_. carpio and £. commersoni recruits  that
were found in July apparently emigrated out of the watershed
by August.  There was also a sizable influx of D_. cepedianum
during this month.
Tremendous influx of ID. cepedianum and recruitment of  £. cyprinus^
young in August.  Also evidence  of  downstream movement  of  fish,
particularly P_. notatus, into this  area in response to
deteriorating habitat conditions upstream brought about by the
drought.  Ericymba buccata exhibited a significant decline in
density in August, again likely  due to heavy mortality.
                                                                                             I

                                                                                             N3
                                                                                             H
                                                                                             -ps
                                                                                             I

-------
Table  4  (continued)

Sampling period    Year    Station    PS    Code

July-August        1977      18       .287    1
                   1977
                   1978
 12
         .192
.407
                      Nature of change

Downstream movement  (out of this region) of C. cyprinus
recruits and probably E_. buccata, accompanied by the
immigration of IX cepedianum in August.
Immigration of IX cepedianum and emigration of C_. carpio
recruits.
Emigration of N_. stramineus.
August-September   1975
                   1975
                   1977
                   1977
                   1977
 15
 6

12
29
28
.179
.469

.491
.394
.491
Very low densities of fish apparently due to low flow
conditions and algae problems in September.  This area was
also recovering from a fish kill that decimated the fauna
in early August.

Few fish present, apparently due to algae problems.
Immigration of N. stramineus and to a lesser extent, an
influx of p_. notatus (probably from upstream) , accompanied
by the emigration of IX cepedianum and C. cyprinus recruits
in September.

Immigration of _N. stramineus accompanied by the emigration of
D_. cepedianum and recruits of £. cyprinus and C_. carpio in
September.

Immigration of N. stramineus and to a lesser extent an
influx of P_. notatus accompanied by the emigration of
IX cepedianum and recruits of C_. carpio, C. commersoni, and
C. cyprinus in September.
                                                                                                                        I

                                                                                                                        N3
                                                                                                                        M
                                                                                                                        Ul
                                                                                                                        I
September-October  1974     15
                   1975     12
                   1976     12
         .493     2
         .274     1
         .480    2
                    Fish densities were low apparently due to frequent habitat
                    perturbations (channelization)  throughout this year.
                    Pimephales notatus declined in abundance during October.

                    Emigration of D.  cepedianum in October.  This region was
                    also still recovering from an August fish kill.

                    Emigration of C.  commersoni and I. natalis recruits in October
                    accompanied by the immigration of N. stramineus and influx of
                    P.  promelas, P.  notatus, and E. buccata from upstream.  The
                    latter three species were probably seekingrefuge from severe
                    drought conditions upstream.

-------
Table  4  (continued)
Sampling period    Year    Station    PS    Code

September-October  1976       18     .403     3
                         Nature of Change

Very few benthic fish present apparently due to drought
conditions and algae problems.
Oc tober-November

-------
        Table  5.   Changes in pelagic  guild resulting in  similarity values  (PS)  less than or equal to 0.50.
                  Code  numbers  show  the  relationship between  changes  within  this  guild and overall community
                  structure.  1)   Indicates that short-term shifts in overall  community structure were
                  primarily due to changes in species composition within  this  guild.   2)   Indicates that
                  short-term shifts  in overall community  structure were due  to changes in  species composition
                  within  both the  benthic and pelagic guilds.   3)   Indicates that changes  within this guild had
                  a  relatively  small impact on overall  community  structure.
Sampling Period    Year     Station    PS_    Code

February - March    —        —       —     —
                                                             Nature of  Change
March-April
1976
15
.000     3
                   1976
          28
        .241     3
                   1976

                   1977


                   1977

                   1978
          18

          29


           6

          18
        .398

        .485


        .479

        .500
         3

         3


         1

         3
Only one pelagic fish was found here  in April.  With  the
exception of N_. spilopterus, which was common in March,
no other pelagic species recolonized  this region since
the August 1975 fish kill.  Algae problems probably contri-
buted to this in April.

The density of pelagic fish was fairly low probably due to
a possible minor fish kill in March and algae problems in
April.  The major changes included the emigration of  N_.
spilopterus and influx of F. notatus  and S_. atromaculatus
in April.

Low density of pelagic species .

Immigration of N. umbratilis and movement of S_. atromaculatus
out of this area during its spawning  run in April.

Immigration of N_. umbratilis during April.

Low densities of pelagic species possibly due to high water
levels.
                                                                                                                        to
                                                                                                                        h->
                                                                                                                        -J
                                                                                                                         I
April-May
1975
                   1976
          12
        .000
        .292
                    Absence of fish in April due to a weir below the station
                    which prevented fish from colonizing this area during pre-
                    vailing low flow conditions.

                    Emigration of N.  spilopterus and apparent immigration of S_.
                    atromaculatus adults in May.

-------
 Table  5  (continued)

 Sampling  Period    Year

 April-May         1976

                   1976


                   1978



                   1978
         Station    PS     Code

           15       .192      3

           29       .165      3
           12
          29
          .444
          .416
                                            Nature of Change

                    Very low densities of pelagic fish probably due to algae

                    Emigration of N_. spilopterus and movement of S_. atromaculatus
                    out of this area during their spawning run in May.

                    Low densities of pelagic fish in April possibly due to heavy
                    siltation during the spring flooding.  Immigration of N.
                    cornutus-N. chrysocephalus and ttf. umbratilis in May.  ~

                    Low densities of pelagic fish during both months.  Minor
                    decrease in the number of S_. atromaculatus and influx of N.
                    cornutus-N. chrysocephalus in May.                       ~
May-June
1976
15
.450
 Low densities of pelagic fish probably due to algae.   Increase

June-July
1976
1974
1975
29
17
12
.442
.333
.195
3
2
2
in the number of F. notatus in June.
Very low densities of pelagic fish, probably due to algae..
Emigration of N. umbratilis in June.

Very low densities of pelagic fish possibly due to effects of
channelization .
Low densities of pelagic fish due to fish kill in late May.
i
S3
t->
00
1

                   1975


                   1976


                   1977
         18


         29


          6
         .424


         .482


         ,261
         3


         2
Apparent  influx  of  adult  S_.  atromaculatus from upstream in
July.

Low densities of pelagic  fish.   Emigration of  N.  cornutus-
N. chrysocephalus.
Very low  densities  of pelagic  fish probably due to  algae
and/or low  flow  conditions .
Influx of adult  S_.  atromaculatus and  downstream movement of
N^. umbratilis in July.
July - August
1975
12
.323
Low densities of pelagic fish due  to  a  fish kill  in early
August and slow recolonization  from the fish kill  in late
May.  Immigration of N. spilopterus in  late August.

-------
Table  5  (continued)

Sampling  Period     Year     Station    PS    Code

July - August       1975       18      .478      3



                    1976       15      .367      2


                    1976       29      .469      2


                    1977       12      .398      3

                    1977       29      .271      3
                                                  Nature of Change

                            Very low densities of pelagic fish probably due to  algae
                            and/or low flow conditions.  Slight increase in the
                            abundance of S_. atromaculatus in late August.

                            Recruitment of S_. atromaculatus and to a lesser extent F_.
                            notatus in August.  Low densities of pelagic fish in July.

                            Recruitment of S_. atromaculatus and to a much lesser extent
                            F_. notatus in August.  Low densities of pelagic fish in July.

                            Immigration of N^. spilopterus in August.

                            Immigration of N. spilopterus accompanied by upstream movement or
1977 28 .346 3
1977 6 .431 2
emigration of N. umbratilis in August.
Influx (from downstream) of N. cornutus-N. chrysocephalus and
immigration of N. spilopterus accompanied by the movement of
some S. atromaculatus young out of this region in August.
Immigration of N. spilopterus and downstream movement of S.
atromaculatus in August.

I
N3
IT-*
VO
1
August-Septebmer   1975
                   1975
                   1976
15     .372
       .229
18     .000
Very low densities of pelagic fish in August due  to  the  fish
kill.  Immigration of N. spilopterus in September.

Low densities of pelagic fish probably due to  algae.   Increase
in abundance of L_. cyanellus  (probably recruits)  and decrease
in abundance of N_. umbratilis.
This region was heavily choked with algae in August  and  a
series of isolated pools in September.  Only two  species were
captured during each sampling period.  During  August S_.
atromaculatus young and E\ notatus were abundant; however,
both of these species were absent in September when  N.
spilopterus immigrants were dominant.

-------
     Table 5  (continued)

Sampling Period    Year    Station    PS    Code
September-October  1974       15     .274
                   1975       18     .051
                   1976       12     .468
                       Nature of  Change

Movement of N_. spilopterus, L_. cyanellus, and  N_.  cornutus-N.
chrysocephalus out of this region  in  October.

Very low densities of pelagic fish.   Immigration of  N.
spilopterus in October.

Immigration of N_. spilopterus in October.
                                                                                                                          N3

                                                                                                                          O

                                                                                                                           I

-------
  Table 6.  Relative frequency of significant changes (PS<.50) in the structure of the benthic and pelagic

            guilds throughout the year.  These proportions are based upon the total number of sample

            similarity measures(n) during that period from 1974 - 1978.

Benthic
Guild
Pelagic
Guild
February
March
0
0
March
April
0
.43
April
May
.21
.32
May
June
.20
.20
June
July
.55
.46
July
August
.57
.57
August
September
.50
.30
September
October
.57
.43
October
November
0
0
n
                        14
19
10
11
14
                                              10
                                                                                                                         to
                                                                                                                         to

-------
Table 7.
Frequency and relative frequency (based upon n sample similarity measuresduring that year) of
significant changes (PS < .50) in the structure of the benthic and pelagic guilds during each of
the five years of intensive sampling.
                        1974
                               1975
1976
1977
1978
Benthic
Guild
Pelagic
Guild

4 (.80)
2(.40)

13 (.48)
8(.30)

4(.13)
13(.43)

7 (.39)
7 (.39)

1
3

(.13)
(.38)
                                         27
                                               30
                    18
                                                                                                                        NS
                                                                                                                        Ni

-------
                               - 223 -
did not always act alone or independently to produce the observed
instability.  In addition, their relative importance and timing often
varied between guilds.

     Three major fish kills are known to have occurred during the course
of the study and since they severely reduced the density of fish in the
affected region, only minor changes in species composition during the
recovery phase commonly resulted in low sample similarity values.  The
first two kills occurred in May and August of 1975 and were fairly
localized, directly affecting only stations 15 and 12.  The benthic guild
appeared to recover fairly rapidly (within a month) from the May kill but
somewhat slower during the fall.  In contrast, the structure of the
pelagic guild following the May kill remained rather poor through October.
The most devastating and widespread (affecting all stations downstream of
station 6) kill occurred in late September 1977.  Although no samples
were taken in October, fish densities and species composition were fairly
similar to pre-kill conditions by November at stations 12, 29, and 28.

     The rate of recovery from these fish kills clearly varied among
species and according to the time of the year during which they occurred.
That is, not only are some species more vagile than others and hence more
likely to recolonize a devastated area, but different species are also
more mobile during certain seasons.  In addition, it is likely that other
factors such as habitat conditions also influenced recoveryrates.  For
example, the 1975 fish kills occurred relatively soon after the watershed
was subject to massive habitat modifications, and flow regimes were
somewhat different during the recovery periods in the fall of 1975 and 1977.

     Algal blooms are common in disturbed agricultural  watersheds which
are enriched by nutrients in runoff and where solar input is high as a
result of the clearing of riparian vegetation.  Their occurrence and
persistence in Black Creek appears to be determined by temporal variation
in the amount of rainfall (Table 1).   During years with substantial
precipitation in the spring and early summer  (e.g., 1974 and 1975), algal
blooms are curbed by the flushing action of channel flow and are generally
limited in occurrence to the summer months.  However, during years with
little rainfall (e.g., 1976)  algal blooms develop as early as April and
persist through the summer and early fall.  Although the effects of algal
blooms could not be entirely distinguished from those of low water levels,
their occurrence generally resulted in a significant reduction in fish
densities.  In addition, on a few occasions community structure was
directly altered as species like sunfish (Lepomis)  attempted to avoid
algae-choked areas whereas the black-striped  top minnow, F_. notatus,
appeared to thrive in these conditions.  In general, pelagic species seemed
to be more adversely affected than benthic species.  Low sample similarity
values caused by reduced fish densities within the pelagic guild were at
least partially attributable to algal blooms from July through September
of 1975 and from April through August of 1976.  Major changes of this
nature within the benthic guild were less frequent and only occurred during
the summer and early fall of 1975.   It is likely, however, that more subtle

-------
                               - 224 -

effects of algal blooms on both guilds either went undetected due to less
frequent sampling after 1976 or were confounded by the effects of other
factors.

     While major short-term changes in community structure could not
always be unequivocally tied to flow conditions, direct and subtle
impacts appeared to be assoicated with both extremes.   High water levels
in Black Creek were most common during the spring months, but are a
somewhat unpredictable, yet integral, feature of the stream environment.
For example, some of the distributional shifts and especially migrations
to and from the Maumee River occurred during high water level conditions.
The timing of the immigration of D. cepedianum, in particular, seemed to
be linked to the occurrence of minor spates during the late summer or
early fall.  In addition, the success of spawning runs of many species
may be dependent on the timing and extent of high flow periods.  In this
regard, it should be noted that flooding, especially during the spring,
is generally more extreme in modified watersheds like Black Creek due to
runoff from the adjacent land surface.  The deposition of silt following
such runoff events also alters habitat structure and appears to have
contributed, in at least one instance, to a major shift in species
composition associated with low densities of benthic fish.  In view of all
of this evidence, it is perhaps no coincidence that the frequency of major
changes in the structure of the benthic guild was greatest during years in
which there were numerous high flow periods  (i.e., 1975 and 1977).

     Low flow conditions had a more obvious effect on the fish community
since they usually were accompanied by reduced fish densities, particularly
among pelagic species.  They also appear to have caused fish movements
within the watershed but, as mentioned previously, both of these effects
were often confounded with those of algal blooms.  Nevertheless, many of
the major changes in community structure resulting from downstream
movement of fish were likely associated with low water levels upstream.
Such conditions were most common during the summer months but varied in
timing and duration among years.  During the driest years  (1976 and 1978)
stability appeared to be lowest within the pelagic guild  (Table 7).  The
low densities of pelagic fish that were frequently found throughout the
year at station 18 were probably due to the low depths that are character-
istic of this region.  Hence, changes in the structure of the pelagic guild
associated with these low densities cannot be considered dynamic and were,
in fact, of minor import to the overall community.  Like high flows, low
flow conditions are generally more severe in modified watersheds due to the
absence of riparian vegetation and the straight, uniform channels.

     Short-term effects of the major channel modifications that were
completed during 1974 went largely undetected due to infrequent sampling
during  that year.  However, limited  sampling at stations 15,  17, and 6,
indicates that fish densities were very low  and species  richness was poor
immediately following channelization.  In addition, although  the recovery
period  was less than adequately monitored during this year, repopulation
by benthic  species appeared to be particularly  slow.  This should be
expected, perhaps, since guilds were delimited  according to feeding locales,
and substrates were highly modified,  if not  destroyed, by dredging operations.

-------
                               - 225 -


The effect of these channel modifications on flow regimes should be
reiterated and hence recognized as a contributing factor in temporal
changes in community structure associated with high or low water level
conditions.  The interaction between flow regimes and habitat
modifications was also evident when a weir below station 6 blocked fish
movements during low flow conditions.

     While movements of fishes within the watershed often directly resulted
in major shifts in community structure, they were usually caused by
environmental conditions.  Within-stream movements are distinguished here
from fish migrations in that the former did not appear to constitute
movement into or from the Maumee River and hence involves only those
species which maintained resident populations throughout the year.  Within
the benthic guild, major changes in species composition of this nature
most often involved the downstream movement of P. notatus during the summer
or fall months  (Fig. 4).   This was probably a response to unsuitable
habitat conditions  (e.g., low flows or algal blooms) upstream, but in a
few cases may have been caused by an interaction with immigrating species.
Ericymba buccata and I?, promelas (Fig. 3) also showed signs of downstream
movements for the same reasons.

     Relative to the benthic guild, changes in the structure of the pelagic
guild were more often due to movements by resident species.  Movements by
S. atromaculatus were particularly significant and included a spawning run
upstream during the spring followed by the redispersal of adults throughout
the watershed  (Fig. 9).  Additional distributional shifts by this species
also occurred during the summer months in response to deteriorating habitat
conditions (i.e., low flows) upstream.  Similar shifts were exhibited by
H- umkratilis  (Fig. 8) and N. cornutus-N. chrysocephalus (Fig. 6) during
low flow periods, but these species only appeared to maintain a resident
population when water levels did not get excessively low during the
summer months.  Algal blooms probably also contributed to these shifts in
pelagic community structure during low flow periods; however, whereas
S_. atromaculatus, N. umbratilis, and ttf. cornutus-N. chrysocephalus moved
out of algae-choked regions, F_. notatus appeared to invade such areas.

     The major cause of short-term variation in the community structure of
Black Creek fishes was migrations of both benthic and pelagic species into
and from the Maumee River.  Among the benthic species, spring immigrants
included 13. stramineus and large adults of C_. commersoni, C_. carpio, and
£. cyprinus.   The latter three species made a brief spawning run into the
the Black Creek watershed but quickly returned to the river.  Hence,
although large schools of these species were observed, they proved to be
difficult to capture and rarely contributed to significant changes in
community structure during the spring months.  However, when these spawning
runs were successful, their young generally dominated the benthic fauna
during the summer months.  Major shifts in community structure then
occurred when these recruits emigrated into the Maumee River during the
late summer and fall.  The immigration of IJ. stramineus appeared to occur
during both the fall and spring (Fig. 7), but the fall influx probably
had a greater impact on community structure.  Although it is not clear
whether these fall immigrants overwinter in Black Creek or perhaps leave

-------
                               - 226 -

during the winter and return again in the spring, individuals found in
Black Creek during the spring definitely emigrate back into the Maumee
during July or August.

     The most profound migrations among the benthic species were carried out
by E). cepedianum during the summer or early fall  (Table 3).  Large numbers
of young D^ cepedianum invaded Black Creek during minor spates as early as
July (e.g., in 1977) and depending on flow conditions, remained in the
watershed as late as November.  Since these individuals appeared to
travel in large schools they contributed to numerous significant changes
in community structure during this period.

     The emigrations of young of two other species, I_. natalis and C_.
anqmalum, were partially responsible for a few other shifts in species
composition within the benthic guild  (e.g., during the fall of 1976).

     Among the pelagic species the immigration of N. spilopterus had the
greatest impact on community structure.  This species invaded the watershed
during the late summer or fall, overwintered, and then emigrated back into
the Maumee during the following spring  (April-June)  (Pig. 5).  In addition,
there was some evidence that N_. umbratilis and N. cornutus-N. chrysocephalus
immigrate into Black Creek as N_. spilopterus leave.  During dry years
these species appeared to emigrate back into the Maumee soon after they
attempted to spawn in early summer.  However, when flow conditions were
favorable they probably remained in Black Creek.  It is also likely that
the  spring spawning run of S_. atromaculatus included immigrants from the
Maumee as well as resident adults.  Although these migrations by !N.
umbratilis, N. cornutus-N. chrysocephalus, and S_. atromaculatus were of
relatively little consequence to the overall community, they accounted
for much of the temporal variation within the pelagic guild during the
spring and summer.

     As mentioned earlier, the recruitment of young C_. commersoni, C. carpio,
and  C_. cyprinus  (Table 3) had a significant impact on the structure of the
benthic guild as well as on the overall community during  the summer months.
Prior to 1976, the recruitment of yearling E. buccata had a similar effect
in July, and although it was not detected in this analysis, the recruitment
of young P. notatus and P_. promelas may have potentially  altered community
structure radically during the  fall months.  Among-.the pelagic species,
the  recruitment of young S_. atromaculatus and, to a  lesser extent,  F_.
notatus, was important during the summer months  while recruits of L_. cyanellus
contributed to a significant change on  one occasion at an  upper station
during the  fall.

     Although non-catastrophic mortality  (i.e.,  mortality not related  to
obvious  fish kills) is difficult to  detect,  it  is another potential cause
of short-term changes in community structure.   For  example,  a detailed
analysis of the  decline of the  12. buccata population in Black Creek
strongly suggests that mortality of  this  species was heavy during the
drought  in  1976.  Hence,  it  is  also  likely that some of the  other  shifts
in community  structure that were attributed  to  movements  and/or migrations,

-------
                               _ 227 -

particularly those that appeared to be in response to severe habitat
conditions, were at least partially due to mortality.

     In summary, it is clear that very few species maintain resident
populations throughout the year in Black Creek.  Those that do must
shift their distributions in response to harsh environmental conditions
whereas other species leave the watershed as habitats begin to
deteriorate.  Although the headwater stream environment is typically
rigorous and a number of its habitat parameters undergo extreme temporal
fluctuations even in the natural state, the severity of living conditions
is magnified by habitat modifications like those in Black Creek.  However,
while these perturbations have contributed to unfavorable conditions for
some species, others have apparently been able to exploit the altered
state.  The influx of I), cepedianum, for example, appears to be linked to
this species' capacity to utilize the heavy growths of algae that
typically occur during the summer months and are a direct result of the
clearing of riparian vegetation in the watershed.  Similarly, the
occurrence and success of the spawning runs by C_. commersoni, C_. carpio,
and (:. cyprinus probably reflects the ability of their young to also
exploit this primary production.  While the invasion of Black Creek by
these species appears to be directly linked to the altered habitat
conditions in the watershed, the immigration of these and other species
is probably only possible because of the absence of a stable and integrated
resident fish fauna.  However, this loss of biological integrity is also
a consequence of the habitat modifications that therefore appear to have
bred much of the observed temporal instability in the fish community.

                                DISCUSSION

      Any attempt to summarize many fish samples  collected from several
 stations over a seven-year period must involve oversimplications.   We
 are certainly guilty of that here.   Thus, we add a few notes of explanation
 to outline some general conclusions about individual species and to briefly
 explore methods of simplifying the data on fish  communities of the Black
 Creek watershed.

 Status of Individual Species.

      The foregoing discussion generally focussed on species at each site0
 At this point we view each of the major fishes within the watershed and
 evaluate their population trends at the watershed level.

      Several species seem to be especially successful at maintaining
 populations in the watershed.  These include S_. atromaculatus, P_. promelas ,
 and P_. notatus.  The two Pimephales species are generally viewed as
 opportunistic omnivores that are regularly successful in headwater streams
 even after they are highly modified.  Semotilus generally feeds at higher
 trophic levels and is especially successful at feeding on insects that
 fall into streams from terrestrial areas.  This species seems to be present
 in the watershed both as a migrant and as a resident.  As a resident,
 individuals seem to persist year around in areas of high quality habitat
 (Gorman and Karr 1978, Karr and Dudley 1980) such as in the Wertz; Woods.
 In most other areas in the watershed, individual Semotilus are generally
 smaller  (except in spawning periods when migrants enter from the Maumee
 River) and their presence is more transitory.  Finally, they are often

-------
                               - 228 -

slower growing, and less vigorous looking.  Their color, for example,
is paler reflecting the less than optimal stream environment they
occupy.

     Three species seem to be relatively successful in using the Black
Creek watershed as a nursery area.  These are the white sucker (Catostomus
commersoni)ithe introduced carp  (Cyprinus carpio), and the carpsucker
(Carpiodes cyprinus).   During the spring, all three species migrate into
the watershed from the Maumee River and large schools of adult C_.
commersoni and C_. cyprinus, in particular, have been observed as they
search for spawning sites.  Spawning success is highly variable between
years and individuals experience considerable fin damage as they try to
clean spawning sites in gravel substrates of fine sediment and benthic
algal  growth.  Successful recruitment in the Black Creek watershed seems
to be linked to the ability of their young to exploit the rich algal
growth that commonly occur during the summer months.  The immigration of
young IX cepedianum appears to occur for the same reason.  Since most
young and adults of all of these species migrate into the Maumee River
during the fall, we suspect that small watersheds like Black Creek may
be of considerable importance to the fishes of many of our major rivers.

     Notropis umbratilis, N. cornutus- 14. chrysocephalus,and another
migrant species, N_. stramineus,      showed some signs of increasing
abundance in the watershed in recent years but have yet to establish a
strong resident population.  Ericymba buccata, once an established
resident, experienced a population increase early in the study but is now
in danger of extirpation  (see other paper on E_. buccata in this report) .
Other species that seem to have declined beyond the point of recovery
include the darter, E_. spectabile, and the stoneroller, £. anomalum.
Population declines by all of these species, but especially, E_. spectabile,
were at least partially due to increased siltation of substrates.

The Watershed Environment.

     Our experience in a number of midwestern states indicates that the
pattern of migration of fish into small streams is rather common.  However,
the  specific circumstances in Black Creek should be reiterated.  This
water shed is composed of a stream that reaches, at most, third order
before it enters a major river - the Maumee.  Thus, the main river .serves
as a reliable source of colonists to replace local mortality and we
expect that local extinctions due to natural or man-induced events may
be easily countered by dispersers moving up from the river.  Indeed, the
regularity with which C_. carpio, I), cepedianum,C. cyprinus, and N^.
spilopterus move well into the Black Creek watershed may be due to its
proximity to the Maumee River.  Unfortunately, sampling problems and
financial resources have prevented us from sampling that major river
component of the fishes which regularly utilize the Black Creek watershed.

Guild  Structure.

     For simplicity, we have chosen to describe only two guilds  in this
paper.  This oversimplification is necessitated by a variety of
circumstances.   First, we want  to try to minimize the detailed analysis

-------
                               - 229 -

that could be undertaken at this time.  It is our plan to dissect
selected patterns in more detail in the next year as the final
integrative project report is formulated.  Second, the extreme
perturbations imposed by the man-induced alterations in Black Creek
tend to reduce the differences in natural history among Black Creek
fishes.

     For example, it could be argued that our benthic guild is very
heterogeneous.  Indeed, species in the benthic guild feed on a variety
of resources  (herbivores, bottom filterers, carnivores) while
virtually all of the pelagic species are carnivores.  In addition, some
of the benthic species, particularly P_. promelas, may feed to a
significant extent not directly in association with the bottom.  Furthermore,
species such as C. cyprinus , C_. carpio, and C_. commersoni filter the
bottom ooze whereas other species such as Pimepnales, E_. buccata, and
tJ. stramineus pick food items off the bottom substrate.  For convenience
in this presentation, all were grouped into the benthic guild.

     Similarly, one might argue that the inclusion of Fundulus in the
pelagic guild is an oversimplification.  This is really a surface-feeding
species found in shallow edge environments.  It is also the major species
associated with the algae-choked areas during low flow periods of summer
and early fall.

     Despite these weaknesses, we feel that general patterns in the
watershed have emerged and plan to continue dissecting those patterns
in the months ahead.

                            LITERATURE CITED

Gorman, O. T. and J. R. Karr.  1978.  Habitat structure and stream
     fish communities.  Ecology. 59: 507-515.

Karr, J. R. and O. T. Gorman.  1975.  Effects of land treatment on the
     aquatic environment.  Non-point Source Pollution:  Pollution
     Control in Great Lakes.  USEPA-905/9-75-007.  pp. 120-150.

Karr, J. R. and D. R. Dudley.  1980.  Ecological perspective on water
     quality goals.  Environmental Management.  In press.

Toth, L. A., D. R. Dudley, J. R. Karr, and O. T. Gorman.  1981.  Decline
     of a silverjaw minnow (Ericymba buccata) population in an
     agricultural watershed.  This volume.

Whittaker, R. H.  1975.  Communities and ecosystems.  (2nd Edit.)
     Macmillan Publ. Co., N.Y.  385 pp.

-------
                                _  230  _
              DECLINE OF A SILVERJAW MINNOW (BRICYMBA

           BUCCATA) POPULATION IN AN AGRICULTURAL WATERSHED

                                  by

        Louis A. Toth , Daniel R. Dudley , James R. Karr , and

                          Owen T. Gorman
                              ABSTRACT

     During eight years of fish sampling in Black Creek, the silverjaw
minnow, Ericymba buccata, exhibited wide fluctuations in density, including
a population outbreak in 1976 followed by a rapid decline in abundance.
Accompanying these fluctuations were changes in population structure that
were linked to differential mortality and variations in recruitment.
Factors responsible for this species demise in the watershed included:
severe droughts during 1976 and 1978; consecutive harsh winters in
1976-77, 1977-78, and 1978-79; fish kills;and altered habitat conditions
brought about by channelization and the removal of riparian vegetation.

Key words:  Agricultural watershed, channelization, habitat, nonpoint
            pollution, population recuitment, silverjaw minnow
1.  Department of Ecology, Ethology, and Evolution, 606 E. Healey,
    University of Illinois, Champaign, IL  61820

2.  Division of Surveillance, Ohio Environmental Protection Agency,
    361 E. Broad, Columbus, Ohio 43215

3.  Museum of Natural History, University of Kansas, Lawrence, KA 66045

-------
                               - 231 -

                            INTRODUCTION
     Fish populations in freshwater streams vary in space and time due
to a variety of factors in both "natural" and man-altered environments.
Because of the complexity of factors responsible for this variation, it
is difficult to interpret the results of programs designed to minimize
negative effects of man's activities on the biological integrity of water
resources, particularly when only one or two years of data are collected.
This report examines the decline of a silverjaw minnow (Ericymba buccata)
population over an eight-year period in an agricultural watershed in
northeastern Indiana.  The study area was the target of an intensive
application of soil conservation practices designed to reduce soil
erosion and thereby improve water quality (Morrison 1977).    Thus it was
possible to evalute the population dynamics of Ericymba in light of
natural environmental variability and human induced perturbations in the
watershed.

     Ericymba buccata is a common inhabitant of small, headwater streams
in the midwest, where it lives and feeds in schools on or near the bottom
(Trautman 1957, Hoyt 1970, Pflieger 1975, Wallace 1976, Smith 1978).  It
is most abundant in areas with sandy substrates and occurs in low densities
over silt (Wallace  1972).  In Indiana, spawning takes place from late
April through July (Wallace 1973a).

                               METHODS
     The study was conducted in conjunction with an interdisciplinary
demonstration project (Black Creek Project) carried out on a 48.5 km
watershed in Allen County, Indiana.  The primary objective of the project
was to develop and implement plans  for controlling soil erosion and to
evaluate the effectiveness of traditional conservation methods in
improving water resources.  This included analysis of the short- and
long-term effects of project activities on the fish fauna of the watershed.

     Twenty-five fish sampling stations were established in the Black
Creek watershed (Fig. 1), but most of the sampling effort was focused upon
the main channel (particularly stations 6, 18, 17, 28, 29, 15, and 12).
Samples generally covered a distance of 100 m at each station and were
taken using 3.1 or 6.3 mm mesh minnow seines with block nets at the upper
and lower ends of the station.

     Sampling frequency was not uniform throughout the study period but is
believed to accurately reflect general population trends among adult
(>1 year) Ericymba.  Initial samples were taken at sites 6 and 12 during
July 1973.  From 1974  to 1978 samples were taken at monthly intervals,
although not all stations were sampled during each period.  Bi-weekly
samples were taken at some stations during 1975 and 1976.  Sampling was
less frequent in 1979 but included collections from spring, summer, and
fall.  The last sample was taken in June 1980.

     Captured Ericymba were either counted and released, measured to the
nearest 1 mm (total length) and released, or preserved in 10% formalin for

-------
                           - 232 -
                                                       -4-
                                                  SCALE (km)
Figure 1.  Map of the Black Creek Watershed showing the location of
           sampling sites.

-------
                               - 233 -
laboratory analysis.  Field processing of large samples in 1976 was expedited
by assigning groups of Ericymba to 3-5 mm size classes.  These fish were
later distributed equally among 1 mm increments within their respective
size classes.  Length-frequency distributions and scale readings were used
to establish the age structure of the population during the course of the
study.  Scales for age determination were taken from the first or second
row of scales above the lateral line and just anterior to the dorsal fin of
preserved fish, mounted in water between two microscope slides, and read
using a microprojector.

     Some environmental variables that may affect fish populations were
also monitored during the course of the study.  Stream discharge was
measured with a stage recording device at station 6 from 1975 through 1978
and temperature and rainfall data were taken from monthly records of the
Fort Wayne, Indiana weather station (approximately 20 km from the watershed)
Quantitative  substrate measurements were taken at fish sampling stations
on the main channel of Black Creek during 1975-76 and 1978-79 according
to methods outlined by Gorman and Karr (1978).  Point samples of bottom
types within these stations were grouped into either physical  (e.g., silt,
sand, clay) or biotic  (e.g., vegetation, litter) categories.

                                RESULTS

Environmental Perturbations.

     From 1973 to 1977 the stream environment of Black Creek underwent
major alterations as a result of the concentrated application of structural
conservation practices such as streambank protection and the establishment
of grassed waterways  throughout the watershed  (Table 1).  In the Black
Creek project streambank protection consisted of grading and stabilizing
streambanks on one or both sides of the channel, plus dredging and
straightening the channel in selected locations.  The stream had been
previously channelized in 1941 so these project activities amounted to
"re-channelization" that eliminated pool-riffle complexes, substrate
sorting, and riparian vegetation that had recovered since 1941.  Other
commonly observed effects of channelization included rapid fluctuations
in discharge, decreased water depths,  greater daily and seasonal changes in
water temperature, and increased turbidity (Karr and Gorman 1975).

Table 1.  The amount of grassed waterway constructed and streambank
          protection work conducted in the Black Creek watershed,
          1973 through 1976.
              Upstream	       Main Channel	
                        Waterway    Streambank                   Streambank
Year	Month (s)	(acres)  Protection  (ft.) Month(s)    Protection  (ft.)
1973
1974

1975

1976

September
April thru
November
August thru
September
June thru
August
9.6

2

11

10
5,300

7,175

0

9_,400
0
June thru
September 37,022

a

0
°Dridge construction work near stations 12 and 15.

-------
                               - 234 -

     The majority of the structural conservation practices installed
during the Black Creek project involved.disturbance of the land surface
near the main channel or its tributaries.  Sediment delivery from these
sites to the stream network was sometimes very high until vegetative
cover was re-established-.  In one instance, sheet and rill erosion was
so severe during the year following the construction of a grassed
waterway and streambank protection work, that sedimentdeposition reduced
the depths of pools one kilometer downstream in an unchannelized segment
(Wertz Woods, Fig. 1) by as much as one meter.  Substrate measurements
also indicated that the deposition of sediment     persistently altered
the structure of habitats in Black Creek (Fig. 2).  Gravel-sized
particles that were exposed in 1975 were covered by sand and silt by
1978; this trend toward finer substrates continued through 1979.  These
changes appear to be due to the combined effects of erosion from the land
surface and stream channel modifications that favor the extensive
accumulation of sediment.

     Substrate measurements also revealed that the distribution of silt
in the watershed was highly dependent upon discharge rates.  For example,
during a high discharge period in the early spring of 1976 silt was
apparently transported from upstream locations  (e.g., station 6) and
deposited  at station 12, whereas the mid- and upstream stations experienced
siltation during extended low flow periods (e.g., summer of 1978).

Water Quality and Fish Kills.

     Water quality conditions in Black  Creek have been reported elsewhere
(Karr and Dudley 1976, Morrison 1977, Nelson and Beasley 1978, Dudley
and Karr 1979, 1980).  The stream receives organic pollutants from septic
tank effluent and barnlot runoff.  Overall nutrient enrichment is
substantial and algal blooms occur where solar radiation is high as a
result of the clearing of riparian vegetation.  During prolonged low flow
periods algal blooms alter substrate characteristics  (Fig. 2) and reduce
available habitat space.  Solar radiation also raises water temperatures
in shallow stream segments as high as 34° C.  Diurnal dissolved oxygen
patterns during the  summer months suggested periods of high productivity
and  subsequent decay of  algal biomass  (unpublished data, DRD).  Daily
minimum dissolved oxygen concentrations were  frequently below 5 mg/1 in
July and August and  were depressed even further  as a result of  leaf litter
input when low flow  conditions persisted into the  fall months.

     Three major fish kills are known to have occurred during the course
of the study.  During the first incident  (28 May  1975), probably caused
by the application of herbicides, dead  fish were  found on the Smith-Fry
drain downstream  from station  19 and on the main  channel  from station  22
to the Maumee River  (Fig. 1).  A smaller kill of  undetermined cause
occurred  in August 1975  and affected all stations  downstream of the
entrance  of  the Smith-Fry drain  (i.e.,  stations  15 and 12).  The most
devastating  kill  occurred on 29 September  1977 when  several thousand gallons
of manure  slurry  were accidentally discharged into the main channel from
an animal waste holding  lagoon near  station 6.   Mortality was near  100%
in 9 kilometers of stream below the  spill  (Dudley and  Karr  1979).

-------
                            -  235 -
         dSILT SSAND  0GRAVEL  OBVEGETATION  DOTHER
IUU
LJ 80
PERCENTAG
ro -& 
D 0 O 0

_
-

*






i




::
i


m
I




$
\





^
i







1
^s
fy.
ft


1


j


i




in
\
1
fe


1
1

I
^
i




I
Hi
1


\
\
\
\

HI
1





\
?
1
f

1
%




^
1
1
I
1





m


i



                6
           JUNE
           1975
12 15 18 6
  SEPT
  1975
12 15 18 6
 MARCH
  1976
12 15  6    12 15 6
 MAY    JULY
 1978     1978
12 6    12 15 18
SEPT    MAY
1978    1979
Figure 2.   Proportional  representation of substrate categories at
           sampling sites on the main channel of Black Creek.

-------
                               -  236 -
Climatic Conditions and Stream Discharge.

     Severe climatic conditions occurred during the study period (Fig. 3).
Rainfall was below normal during each year except 1975 and 1977, and
resulted in extended periods of low discharge during the summers of 1976
and 1978.  A minimum base flow of approximately 0.1 to 0.5 cfs. was
generally maintained in the lower 7 km of Black Creek by groundwater
exfiltration.  Beginning with the winter of 1976-1977 this region also
experienced three consecutive harsh winters that caused many sections of
stream to freeze solid.  Snow and ice melt during the following springs
resulted in temporary periods of high discharge.

Population dynamics and distribution of Ericymba

     During the course of the study, Ericymba exhibited wide fluctuations
in density, highlighted by a population outbreak in 1976 followed by a
rapid decline in abundance in recent years (Fig. 3).  Prior to this decline
Ericymba was a dominant member of the Black Creek fish community, representing
an average of 22% to 32% of the total number of individuals in each sample
taken during 1973 through 1976  (Fig. 4).  Its numerical importance fell
below 5% in 1977 and measured only about 1% in 1979 and 1980 samples.

     Seasonal and year to year variations in the density of Ericymba
were linked to changes in population structure  (Fig 5).  Recruitment of
one-year-old fish, for example, accounted for the increase in abundance
during the summer and fall of 1974 and 1975 (Fig. 3).  One-year-olds were
first caught in July samples and are an indication of reproductive success
during the previous year.  Hence, it appears that the 1973 year class was
fairly successful and responsible for the increase in population density
during 1974.  However, overwinter mortality primarily within this year class
brought about a decline in density during the spring of 1975.  In contrast,
the highly successful 1974 year class survived well through the winter of
1975-76 and was solely responsible for the population explosion during 1976.
The population began to decline during the late summer and fall of 1976
despite the high recruitment of the 1975 year class.  Mortality during
the severe winter of 1976-77 was heavy among all age groups and brought the
population density to a low level from which it continued to decline.

     Early in the study Ericymba exhibited complex seasonal distribution
patterns with extensive use of upstream reaches.  Although two of the
five headwater streams  (Fig. 1) offered poor quality fish habitat due to
domestic pollution  (Richelderfer) or heavy siltation  (Dreisbach), and
another  (Smith-Fry) was subject to frequent fish kills  (see above),
numerous Ericymba, particularly yearlings, were found at upstream sites
 (e.g., stations 20, 5, 4, 16, 3 and WertzWoods) during the summer and fall
of 1974 and  1975.  Ericymba was also widely distributed throughout the
watershed during the spring of 1976 but, in contrast to the previous two
years, decreased in abundance upstream during the summer and fall.  Although
a fairly large number were caught at station 18  in April 1977, Ericymba
was rarely found above station 28 after July 1977.

-------
M-
3
H-
ft
O fD
O hj
   fD

   01
O  P

^ 'O
01  H1
rt fD
Hj  p,
ro
P  P

.   a

   P

y ni
fD
   fD
   X
o
o
tr fD
o  01
H-1 01
   fD
 = ft
     w
   fD

   0 fD
   rt
   fD h-
   01 O

   rtM
   ECO
   ^  P
      H-
      3
3

£ a.

H 01
^  O
   EC
Su  &*
<  Hi
fD  iQ
H  ro
p
:Q  p.
fD  01
01

0  H-
Mi <
   fD
M 3

H. fij
O  01

   tg
   o
P  3
   rt
O  3*
             rt 'O
             0 hj
                fD
             rt O
             EC H-
             fD ^
             H- H-
             i™i rt"
                p
             ft rt
             fD H-
             >t3 O
             p 3

             rt ft

             % %•
             fD P

             Mi P

             3 %
                           HI
                           fO
           O3O
           3  O  H

           rt. 5.  g'
        3    o
        P  3
        H-  C n

        3  IB
        O  fD ft
        EC  H; H-
        p
        3  O
        3  Mi 0
        fD    3
           M 01
                           rt
                           H-
fD
ft
P  CC
3  (D
   H!
P
<  13
fD  fD
H!  H!
P
C  P
H!  <
fD  fD
01  H!
   P
£  ^0
fD  fD
H!  01
fD
   Mi
n  H!
P  o
3  fD
        o
        C cn
        H rt
        P P
        rt ft
3
O
Hi


P
M

^- fD

P  3
01  01
fD


* fi
£  T3
>0  Hi
O  fD
3  01
   01
^  fD
P  ft
01
rt p
   O
H!  o
   0
n  H;
o  ft
Hi  H-
ft 3
01   "
                      P
                         o 01
                         P
                           fD
                   o
                   w  n
                      P  ft
                   n v  H-
                   HS  rt  01
                   fD  c  n
                   fD  hj  EC
                   !V  fD  p
                   •   ft  Hi

                      p  fD
                   H  rt  -

                   §
                         a
   01

rb
Hi
P
rt  H-  EC
C  3  fD

fD     3
   01  fD
p  H'  p
3  rt  3
ft  fD
   01
           fD  H- —'
           ft O •
                                                 LOG|0 NUMBER

                                                  OF Ericymba
                                        CD
                                        ->l<
                                            -0
                                        01
                       (/>
                       >

                       -o
                       r~
                       n
                                     m
                                        3
                             CD


                             00
                                        CD
                                        CD
                                        03>

                                        O
                              -F="
                                                                        MEAN DISCHARGE

                                                                               (CFS)
                                                                   OJ
                                                                     — >
                                                                     CO
                                                                     01
                                                D    — >
                                                     CD
                                                     -J
                                                                     CD
                                               — >
                                               CD
                                               ->lc_
                                               00
                                                                  — >
                                                                  CD

                                                                  CD"
                                                                  CD
                                                                  03 >
                                                                                Ui
                                                                                              Oi
                                                                                          DEPARTURE
                                                                                         FROM NORMAL
                                                                                             (cm)
                                                                                          i         	
                                                                                          o   o   o
                                                                                                                     DEPARTURE
                                                                                                                    FROM NORMAL
                                                                                                CD
                                                                          	>


                                                                          CD
                                                                                                CD

                                                                                                -si
                                                                                             CD
                                                                                             -N|=
                                                                                             CD
                                                                                      CD
                                                                                      C0>
                                                                                      O
                                                                                                                  CO
                                                                                                           ~U
                                                                                                                  -
I


N5

-------
                           - 238 -
   LJ
   O
   cr
   o
   Q.
   LJ
   >
   LJ
   or
      40
      30
      20
       10 -
       0
           1973  1974  1975  1976  1977  1978  1979 1980

                               YEAR
Figure  4.  Mean proportional representation (- 1  standard error) of

          Ericymba in fish samples taken at stations on the main

          channel of Black Creek.

-------
                          - 239 -
                                            AGE GROUP  I
                                            SUM WTR
SUM WTR
    SUM WTR  SUM WTR SUM WTR  SUM WTR
SPR FALL SPR  FALL SPR FALL SPR  FALL SPR FALL  SPR FALL SPR
  1974    1975    1976    1977    1978     1979   1980
                 SAMPLE  PERIOD
Figure 5.  Average number (log  )  of individuals within various age
          groups based upon themean number of Ericymba/ 100 m  (i.e.,
          data given in Fig.  3)  and the proportional representation
          of each age group in samples taken during that period.
          The symbol "x" denotes  that an age group consisted of less
          than 10 individuals/100 m during that sampling period.
          Data on the age structure of the population were not taken
          during the winter and spring of 1978.

-------
                               _ 240 _

                               DISCUSSION

     Conditions in the Black Creek watershed appeared to be favorable for
Ericymba during most of the first four years of sampling despite the
extensive habitat perturbations that occurred during 1974.  Except for
some overwinter mortality during 1974-75, healthy population densities were
maintained through the summer of 1976 and recruitment of one-year-olds
indicated that reproductive success was fairly good during 1973 and 1975
and excellent in 1974.  Little is known about the reproductive ecology of
Ericymba but it presumably requires gravel or sand substrates that are
free of silt (Pflieger 1975, Smith 1978).  Quantitative habitat measurements
in June 1975 indicated that these substrate types predominated in Black
Creek through the spring of that year.  The tremendous success of the .
1974 year class was likely due to relaxed competition from other species.
Project activities resulted in reduced population densities among other
Black Creek fishes during 1974 and 1975  (Karr and Gorman 1975) and Smith
(1978) has indicated that Ericymba is a pioneering species that quickly
invades newly dredged streams.

     A number of factors likely contributed to the rapid decline in the
Ericymba population.  Wallace (1972) and Smith (1978) indicated that
Ericymba is intolerant of siltation which became an increasing problem in
Black Creek as a result of the newly channelized stream morphology,
increased sediment loads from near-stream perturbations, and low stream
discharge rates.  The altered substrates may have both limited the
reproductive success of Ericymba and also affected its food supply
(Muncy et. al.  1979).

     Extreme environmental conditions caused mortality among adult
Ericymba and may have also limited recruitment.  For example, the heavy
mortality that occurred during the late summer and fall of 1976 likely
affected young of the 1976 year class as well as adults.  During this
period, population density was very high and habitat space was
deteriorating at a rapid rate due to the prolonged drought and accompanying
algal blooms.  Hoyt (1970) noted that an E_. buccata population experienced
a high degree of stress in similar conditions in Kentucky.  High water
temperatures (32  - 34 C), accentuated by the lack of near-stream
vegetation were an additional stress factor during the summer months.  In
a study on thermal discharges to the White Riverin Indiana, Proffitt
and Benda  (1971) found that Ericymba did not occur in areas where the
water temperature exceeded 31.1 C.  Thus shallow water and open canopies
probably limited survivorship of Ericymba in many regions in the Black
Creek watershed during the extended drought periods of 1976 and 1978.

     Mortality among adult Ericymba was also high during the winter of
1976-77.  This winter was hard on all resident fish populations due to the
extreme temperatures and ice cover coupled with the buildup of decaying
algae; however, Ericymba may have been particularly affected by the cold
temperatures, since the population is near the northern edge of its range
(Wallace 1973a).  Furthermore, temperature-related winter mortality may
have been even more severe among new recruits.  There is good evidence

-------
                               -  241 -


(Christie and Regier 1973) indicating that recruitment of smallmouth bass
(Micropterus dolomieui)  is limited by cold temperatures at the northern
boundary of its range.  Hence, during its population decline recruitment
of E_. buccata may have also been curbed by the three consecutive harsh
winters that occurred in this region.

     The large fish kill that occurred in September 1977 was another
contributing element in the decline of the Ericymba population.  Wallace
(1972) reported that young-of-the-year and yearlings tend to concentrate
in upstream areas in the summer months, but unfavorable flow conditions
and siltation restricted Ericymba's distribution in Black Creek to
mid-stream reaches after July 1977.  Thus, because of its constricted
distribution,  the Ericymba population was devastated by the 1977 fish
kill.

     It is clear that the frequency of extirpations in aquatic environments,
as well as the time-span required to produce them, has been influenced by
man  (Larimore and Smith 1963).  The rapid decline of the Ericymba
population in Black Creek illustrates the interplay between "natural"
limiting factors and man-induced perturbations.  Although it is difficult
to separate the effects of these factors in this particular case, the
deterioration of habitat brought about by man's activities in the Black
Creek watershed not only added other stress factors, but perhaps more
importantly, eliminated or at least severely taxed natural environmental
buffers.  The perturbations then can be thought of as a precondition for
the collapse of the Ericymba population, reducing its resiliency
(Rolling 1973) to extreme manifestations of natural limiting factors and
sources of mortality.

-------
                               - 242 -

                             REFERENCES

Christie, W. J. and H. A. Regier.  1973.  Temperature as a major factor
     influencing reproductive success of fish—two examples.  Rapports et
     Proces-Verbaux des Reunion, Conseil Int. pour 1'Exploration de la
     Mer 164:  208-218.

Dudley, D. R. and J. R. Karr.  1979.  Concentration and sources of fecal
     and organic pollution in an agriculture watershed.  Water Res. Bull.
     15:  911-923.

Dudley, D. R. & J. R. Karr.  1980.  Pesticides and PCB residues in the
     Black Creek watershed, Allen County, Indiana - 1977-78.  Pestic.
     Monit. J. 13: 155-157.

Gorman, 0. T. and J. R. Karr.  1978.  Habitat structure and stream fish
     communities.  Ecology 59:  507-515.

Holling, C. S.  1973.  Resilience and stability in ecological systems.
     Ann. Rev. Ecol. Syst.  4:  1-23.

Hoyt, R. D.  1970.  Food habits of the silverjaw minnow, Ericymba buccata
     Cope, in an intermittent stream in Kentucky.  Am. Midi. Nat. 84:
     225-36.

Karr, J. R. and D. R. Dudley. 1976.  Determinants of water quality in the
     Black Creek watershed.  In Best Management Practices for Non-Point
     Source Pollution Control Seminar.  USEPA - 905/9-76-005.  pp. 171-184.

Karr, J. R. and D. R. Dudley.  In press. Biological perspective on water
     quality goals.  Env. Management

Karr, J. R. and O. T. Gorman.  1975.  Effects of land  treatment on the
     aquatic environment.  EPA-905/9-75-007.  pp. 120-150.

Larimore, R. W. and P. W. Smith.  1963.  The fishes of Champaign County,
     Illinois, as affected by 60 years of stream changes.   111. Nat. Hist.
     Surv. Bull.  28:   299-382.

Morrison, J. B.   1977.   Environmental impact of land use on water quality  -
     final report on  the Black Creek project.  USEPA.  EPA-905/9-77-07.

Muncy,  R. J.,  G.  J. Atchison, R. V. Bulkley, B. W. Menzel,  L. G. Perry,
     &  R. C. Summerfelt.   1979.  Effects of  suspended  solids and sediments
     on reproduction  and early life of warmwater fish:  A review.  USEPA
     EPA-600/3-79-042.

Nelson,  D. W.  and D.  Beasley.  1978.  Quality  of Black Creek drainage water:
     additional parameters.   In  J.  Lake  and  J. Morrison  (eds.) Environmental
     Impact  of Land Use  on Water Quality:  Supplemental Comments.
     EPA-905/9-77-007-D.  pp.  36-78.

-------
                                -  243 -

Pflieger, W. L.  1975.  The fishes of Missouri.  Missouri Department of
     Conservation.  343 pp.

Proffitt, M. A. and R. S. Benda.  1971.  Growth and movement of fishes,
      and distribution of invertebrates, related to a heated discharge
      into the White River at Petersburg, Indiana.  Indiana Univ. Water
      Res, Ctr.  Rept. Invest. No. 5, 94 pp.

Smith, P. W.  1978.  The fishes of Illinois.  U. of Illinois Press,
      Urbana, IL. 314 pp.

Trautman, M. B.  1957.  The fishes of Ohio.  Ohio State Univ. Press,
      Columbus, Ohio.  683 pp.

U. S. Environmental Protection Agency.   1976.  Quality criteria for water.
      Office of Water Planning and Standards, USEPA.  Washington, D. C.
      256 pp.

Wallace, D. C. 1972.  The ecology of the silverjaw minnow, Ericymba
      buccata Cope.  Am. Midi. Nat. 87:  172-190.

Wallace, D. C.  1973a,_  Reproduction of the silverjaw minnow, Ericymba
      buccata Cope.  Tran. Am. Fish. Soc. 102:  786-793.

Wallace, D. C. 1973b.  The distribution and dispersal of the silverjaw
      minnow, Ericymba buccata Cope.  Am. Midi. Nat. 89:  145-55.

Wallace, D. C.  1976.  Feeding behavior and development, seasonal and diet
      changes in the food of the silverjaw minnow, Ericymba buccata Cope.
      Am. Midi. Nat. 95:  361-376.

-------
                            - 244 -



                 The Sociological  Study  of  Soil Erosion


                  Stephen B. Lovejoy  and F.  Dale  Parent


     Soil erosion is a national  problem  which  affects  all  citizens,  agri-
culturalists and non-agriculturalists.   The  erosion of soil  affects
everyone by reducing the fertility  and productivity of our agricultural
lands as well as by contributing  to serious  water quality  problems.

     The loss of valuable topsoil  from cropland may lead  to  long-term
productivity losses and possibly irrepairable  damage to more sensitive
land.  Erosion of topsoil is becoming more  serious  as  lower  quality  land
is brought into production  as well  as  the more intensive  and erosive
practices being utilized on  existing  cropland.  While  increased ferti-
lizer usage has ameliorated  the  impacts  of  erosion upon yields and pro-
ductivity in the short run,  the  long  term consequences of  significant
erosion cannot be escaped.

     In addition, eroded soil is transported into our  water  bodies
(rivers, lakes, streams) along with a variety  of  chemicals attached  to
the  soil particles.  Erosion of  agricultural land has  been pinpointed as
a primary contributor to the decline  in  water  quality (GAO,  1977).  The
transported soil creates sedimentation problems  in water  bodies as well
as promoting eutrophication  in lakes  and reservoirs as a  result of
increased levels of nutrients.   Added  to this  is  the problem of a variety
of harmful chemicals which may be  attached  to  the soil particles, thus
degrading the natural environment  as  well as posing potential hazards to
fish, wildlife and possibly  mankind.

     Many policies and programs  have  been enacted to deal  with the prob-
lems associated with soil erosion,  beginning with the  conservation pro-
grams in the 1930's.  The present  myraid of programs to reduce soil  ero-
sion are administered by a variety of  agencies including  Soil Conservation
Service, Agricultural Stabilization and  Conservation Service, Environmen-
tal  Protection Agency, Rural Clean Water Program, etc.  One  common attri-
bute of all these programs  is that the agencies  have little, if any,
coercive power to force participation in the program or compliance with
recommended practices.

     Programs designed to abate  soil  erosion have been primarily volun-
tary. Landowners who desired to  participate were  encouraged  to do so by a
variety of educational programs  and financial  incentives.   The educational
programs have been orientated toward  providing information on the causes
and  consequences of erosion and  thus  the need for soil conservation.  In
addition, a great deal of information has been provided on the implemen-
tation of specific practices designed to abate soil erosion.  The finan-
cial incentives have, historically, been offerred in the  form of cost-
shares, whereby  the government  agency shares with the  farmer the costs of
implementing the abatement  practice.   Both  education and  financial incen-
tives reinforce  the voluntary nature  of  these programs and strengthen the
antipathy toward mandatory  controls,  regulations  and coercive powers in
general.  However,  little  effort has  been  expended  in  determining the

-------
                            - 245 -
success of these programs  or  in  determining the "best" mix of educational
programs, technical  assistance and  financial  incentives.   Further, the
scant literature available  indicates  that  the effects of these programs
are not equally distributed among  agriculturalists and therefore the pro-
grams cannot be structured  identically for all farmers regardless of geo-
graphic location,  soil  type,  type  of  farm  firm, etc.

     Current programs have  not been spectacularly successful for a vari-
ety of reasons including  inadequate attention to the  incentive structure
established.  If landowners are  to  retain  their rights to voluntarily
participate or not participate and  we are  to  have successful abatement of
soil erosion, we must very  carefully construct incentive structures which
will induce them to  participate.   Questions have been raised concerning
the optimal rate of  cost-sharing (e.g. Bouwes & Lovejoy,  1980; Lovejoy,
et al., 1980) as well as  the  impact of the educational programs and tech-
nical assistance efforts  (e.g. Klessig,  and Lovejoy,  1980).  In general,
we have very little  information  concerning the effects of various program
structures upon individuals and  farm firms with differing predilections,
preferences, values, attributes, etc.  This type of information would be
essential to construct  policies  and programs  which would induce landown-
ers to voluntarily participate in  these  soil  erosion  abatement programs.

     While significant  research  has been done in the  adoption of commercial
practices in agriculture,  the practices  associated with erosion abatement
seem quite different (see Taylor and  Miller,  1979).  Essentially, most do
not have short-run financial  benefits for  the farm firm and therefore are
not as attractive  as other  practices.  This factor necessitates construc-
tion of a program which makes adoption an  attractive  option.

     The problems  associated  with  soil erosion will not diminish in the
near future and government  policies to control those  problems will not
dry up and blow away.   If  these  programs are  to be successful on a volun-
tary participation basis  they need  to be constructed  with more attention
to the desires and needs  of the  potential  adopters, the ability of the
individuals and farm firms  to incorporate  the practices into their farm-
ing operations as well  as  the consequences for the farmer and the farm
firm of utilizing  these practices.

     Excessive soil  erosion in the  U.S.  is the product of a social insti-
tution.  This social institution is composed  of the behavioral patterns
of American farmers.  The overall  purpose  of  the Black Creek project was
to reduce soil erosion  by effecting changes in this social institution,
or, in other words,  by  altering  the practices (behavior)  of farmers.

     However, changes in  social  institutions  are not  straightforward.
Alterations may be more difficult  to  enact than expected  and once enacted
may have secondary effects  that  were  not anticipated.  Such anticipated
effects may have implications (positive  or negative)  for  the soil erosion
program, local residents  and  the local community as well  as future pro-
grams by the agencies involved.

     The Black Creek Project has been rather  unique in that sociologists
have been involved since the early  stages  of  the project.   Not only has
this involvement provided a source  of data for evaluation of the project

-------
                            - 246 -
but also assisted in dissemination  of  information and establishment of
channels of communication in the  early  stages.   Sociologists have aided
in decisions concerning how to  involve  the  local residents and the proce-
dures to use in attempting to insure widespread  acceptance of new agri-
cultural practices.

     In 1974,  interviews were conducted with  eight-nine (89) landowners
in the Black Creek watershed.   These landowners  were questioned about
their land use practices as well  as numerous  attitudinal measures.  In
1976, a sample of those same landowners were  again contacted and inter-
viewed in an attempt to investigate the impacts  of the project and changes
in land use.   Results of both earlier  investigation are available in prior
reports from the Black Creek Project.

     The present study utilizes information obtained in interviews con-
ducted during  the summer of 1980.   Attempts were made to contact all 89
respondents  included in the 1974  study, but due  to outmigration (15),
mortality (5), illness (2), etc., only 54 were  interviewed in 1980.  The
present report will detail the  responses of those 54 respondents.  For
some measures, the 1980 responses will  be contrasted with the 1974
responses in order to assess shifts  in land use  patterns, attitudes, etc.
In other sections, the 1980 data  will  be utilized to assess differences
between respondent groupings.   Overall, the objectives of this report  are
as follows:  1) assess the consequences of  an erosion abatement program
for the local  community and its residents,  2) to provide for a better
understanding  of the adoption process,  especially in regard to environ-
mental innovations and 3) to provide  information necessary to a better
structuring  of similar projects.

     This project has been a post-program evaluation of the success or
failure of the program.  We need  to know the  success or failure of the
program as well as the reasons  for  the success or failure if the  lessons
learned from this demonstration project are to be useful.  The other sci-
entists on  this project have  indicated the  changes or stability  in a num-
ber of water quality parameters such  as dissolved nitrogen as well as
changes in  the eco-system of  the water bodies involved.

     However,  evaluation of the success or  failure cannot stop there.
Part of the  evaluation of a project,  especially a demonstration  project,
must deal with the acceptance  of the  project  by the designated population,
their  cooperation  and  their use of  recommended behavioral changes.  This
is essential  from  two  perspectives.   First, the costs and benefits of
such projects  cannot be  limited to  technological costs  and environmental
benefits but must  indicate  the  personal costs paid by the affected firms
and  residents.  Secondly,  the  knowledge gained  from  the social and eco-
nomic  evaluation will  illustrate the  aspects of the program which  facili-
tate  the  primary  goals  and  those that   do not.

Conservation Attitudes

     Although  changes  in  land  use paractices are  the  best  indicators of
the  success  of a  water  quality program, they are not  singular  indicators
of success.   Water quality  programs may effect  the  community  and  resi-
dents  in  a  myraid  of ways.   Residents  may  react negatively  to  a  program

-------
                            _ 247 _
pushed too hard  or  in  an  incorrect  manner,  thus reducing their propensity
to participate in any  environmental program.   The method of financial and
technical assistance may  not  meet with residents desires and expectations.
On the other hand,  the program may  predispose participants to enroll in
other programs.  Participation may  improve  the communication lines between
residents and agency personnel.  The environmental education may increase
awareness of environmental  problems.  These and other attitudinal effects
will affect how  residents will view future  programs as well as how they
will behave in relation to  adopted  practices.

     Let us begin looking at  some of those  attitudes (see Table 1).  In
1974, 46% of respondents  thought that conservation of soil was not a
problem.  By 1980,  that percentage  had dropped to 44%, indicating
increasing awareness of the problems of soil  conservation.  This effect
is larger when we consider  just  those respondents for whom we have data
from both surveys (50% down to 44.4%).  This  suggests that these has been
a slight change  in  views  concerning soil conservation as a problem
although fewer respondents  thought  stream pollution was a major problem,
possibly indicating a  view  that  the problem has been solved.

     Fewer respondents in 1980 thought that landowners would lose from
the soil and water  development programs and a alisghtly greater number
felt that landowners should pay for conservation practices adopted.  In
line with the above, fewer  respondents felt that the federal government
should play an important  role in local soil conservation programs.  The
trend seems to be toward  less federal involvement and greater monetary
burdens on landowners,  although this remains  a minority viewpoint.  Over
all, the respondents increasingly suggested that the cost of water qual-
ity projects should be borne  by state and local units of government as
well as landowners  (see Table 2).   A possible explanation of this trend
is the increased dissatisfaction with the opportunities for landowners to
express their opinion  in planning watershed projects.

     Another aspect of any  project  in a community is it's potential capa-
city building effect.   Some communities will  become more effective in
initiating and implementing local projects  because of new leadership,
improved organizational skills, etc. (see Klessig and Lovejoy, 1980).
The local community in the  Black Creek watershed seems to have increased
its capabilities.   Table 3  indicates a substantial change in attitudes
toward the willingness  of residents to get  involved and in the degree of
organization in  the community.  This suggests that the residents now have
a better organized  community  and more responsive neighbors than they had
prior to the Black  Creek Project.   This may prove to be a definite bene-
fit in coping with  future community problems  and projects.  The respon-
dents who thought that  soil and water development is a good investment
increased,  with  all but 2 respondents agreeing.   Contrary to the results
in Table 2,  fewer respondents  agreed that the watershed program was being
pushed too hard.   This  suggests that local  residents do not feel pres-
sured to enroll  in  this project but  that  control should remain at the
local level.

-------
                            - 248 -
               Table 1:  Attitudes Toward Erosion  Control
                                                       Percent Agreeing
                                                      1974   r974~*1930
Conservation of soil  is not  a  real  problem  in  this
area.                                                 46.1%   50.0%  44,4%

The average landowner  in this  county  stands  to lose
more than he will gain by soil  and  water  development
programs.                                              9.0    9.3    7.4

The cost of soil erosion reducing programs  (e.g.,
field borders, grassed waterways) should  be  borne
entirely by those who  adopt  them.                     28,1   22,2   29.6

The  federal government should  play  an important  role
in soil conservation  programs  in this county.         62.9   64.8   59.3

Pollution of the streams is  a  major problem in this
county.                                               40.4   42.6   25.9

Landowners have little opportunity  to express  their
opinions in planning  watershed projects.              25.8   22.2   33.3

                                        N  =            89     54     54
Table  2:  Who  Should  Pay  For  Efforts  To Clean Up Water
                                          1974*        1980
X percent  federal                  .       38.8%        34.9%
X percent  state                           26.21%       28.6%
X percent  local                           32.57%       36.7%
*0nly  includes  those 54 respondents also interviewed in.1980.

-------
                            -  249 -
               Table 3:  Attitudes  Toward  Local  Community
                                                        Percent Agreeing
                                                       19741974*T980
The people of this  community  are  usually quick to
respond when problems  arise requiring  action.          55.1   46.3   85.2

This community  is well  organized.                      32.6   29.6   79.6

Spending money  for  soil  and water  development  is a
good investment.                                       87.6   83.3   96.3

The watershed program  is  being  pushed  too hard in
this county.                                  .          9.0    9.3    7.4

	N     =	89    ^54     54 _

*0nly  includes  those interviewed  in 1974 and 1980.


Attitudes Toward Community and  Agencies

     Another important  aspect of  a project in  a community is the effects
upon the flow of information.   Many stress the self-reliance and indepen-
dence  of the American  farmer.   We  surveyed the farmers to determine how
they handle problems they encounter or may encounter in their farming
operations (see Table  4).  The  1974 survey indicated that for most prob-
lems,  the responsdents  relied upon themselves  or their neighbors.  In the
1980 survey, respondents  relied upon themselves or neighbors for only
three  (3) problem areas:  crop  rotation,  farm  management, and non-farm
land uses.  The role of  small businesses/professionals and local govern-
mental agencies has expanded  in all problem areas except non-farm land
uses.  These data suggest that  the farmers in  the Black Creek are
increasingly relying upon advice  from  professionals (public & private) in
the operation of their  farms.   This trend may  imply a movement away from
traditional farming practices,  a  trend essential to adoption of new tech-
nologies including  erosion control practices.   However, the adoption of
new practices does  not  necessarily imply that  erosion control measures
will be adopted.  The  new practices may  be more erosive or polluting than
traditional practices  (e.g.,  continuous  corn,  greater reliance on
insecticides and herbicides,  etc.).

     Another implication  of the above  trend is the increasing contact of
these  farmers with  government agency personnel.   Respondents were asked
how many times  in the  past year they had contact with personnel from sev-
eral governmental agencies.  Table 5 indicates that respondents are
increasingly having contact with  agency  personnel.  The only agency to
show decreases  in contact is Purdue University,  likely the result of
fewer  researchers in the  area as  the project has wound-down.  The most
dramatic increase in contact was with  the Cooperative Extension Service
(CES), whether  this resulted  from  a new  CES program or whether the
respondents were seeking  out CES  agents  is unknown.

-------
               Table  4:   Who  Do  Respondents  Contact for Assistance With Problems
Problem
Crop Decrease
Insect Control
Machinery
Livestock
Crop Rotation
Farm Management
Soil Management
Fertilizer Usage
New Crop Varieties
Non-Farm Land Uses
Potential Pollution
Handle
Myself
1974 1980
29%
28
64
37
83
80
51
53
46
50
41
7%
13
37
13
70
74
26
30
38
71
33
Friend/
Neighbor
1974 1980
14%
6
6
5
4
9
4
6
21
7
8
4%
6
6
10
7
9
6
8
9
4
2
Who Is
Contacted
Business/
Professional
1974 1980
30%
42
30
58
4
4
24
39
29
5
5
41%
44
57
77
4
2
9
53
51
—
6
Government
1974 1980
27%
24
—
—
10
7
21
2
4
38
47
48%
37
—
—
19
15
58
9
2
25
59
N*
1974 1980
77%
79
81
65
82
82
80
83
80
60
64
54%
54
54
31
54
54
53
52
53
52
51
                                                                                                              Ln
                                                                                                              O
*N size varies due to non-responses as well as lack  of  applicability.

-------
                              251  -
           Table 5:  Number of Contacts With Agency Personnel
Cooperative
Extension Service

Agricultural
Stabilization and
Conservation Service

Soil and Water
Conservation

Soil and Water Con-
servation Districts

Purdue University
                                                            1980
74.7% 14.0% 11.5% 100.0%  37.7% 15.1% 47.2% 100.0%



51.7  26.4  21.8  100.0   43.4  15.1  41.5  100.0


61.6  23.3  15.1  100.0   54.9  15.7  29.4  100.0


52.3  32.6  15.1  100.0   56.9   9.8  33.3  100.0

52.9  36.8  10.3  100.0   73.1  11.5  15.4  100.0
*0nly indicates those  interviewed  in  both  1974  and  1980.
     A change which may be a more subtle effect  of  the  project  is  the
shifting preferences for methods to get people  to cooperate  (see Table
6).  In 1974, 60 percent of the respondents  thought  education was  the
best mechanism, while  in 1980, only 44 percent  felt  education was  the
best method.  This suggests that some of those  for  whom education  was  a
preferred method of assuring cooperation no  longer  prefer  education.
Financial incentives rose in popularity, possibly a  result of the  per-
ceived success of the  financial incentives offered  by the  Black Creek
project.
Table 6:  Best Method For Getting People  to  Cooperate  in Soil  and Water
          Conservation Programs
Method
     1974
Number  Percent
    1974
Number  Percent
     1980
Number  Percent
Education
Financial Incentives
Laws and Controls
Combination of Above
Other
No Response
N =
35
8
8
—
2
4
57
61.4%
14.0
14.0
	
3.5
7.0
100.0
32
7
4
—
2
8
54
59.3%
13.0
7.4
	
3.7
14.9
100.0
24
11
4
4
5
6
54
44.4%
20.4
7.4
7.4
9.3
11.1
100.0

-------
                            _ 252 _
Land Use

     This section presents the  farmers'  responses  to  a variety of ques-
tions concerning their use of selected  land  use  practices.

     Table 7 provides a comparison  of  the  percentage  of farmers reporting
their use of selected structural  and management  land  use  practices in
both 1974 and 1980.  The first  column  of Table 7 reports  the responses of
all 89 respondents interviewed  in 1974,  while the  second  column includes
only those 54 individuals contacted  in  1974  who  were  again  contacted in
1980.  An overall look at the table  reveals  that  from 1974  to 1980,  land-
owners increased the use of  five  of  the  practices  and decreased use  of
four.  Apparently, there has not  been  a  steady increase in  the adoption
of all the management practices.  Three  of the four  practices which
declined in use, do not involve the  installation of  permanent structures.
Practices such  as conservation  cropping, crop residue management, and
livestock exclusion may be more conducive  to discontinuation than prac-
tices which require structural  modifications.
       Table 7:  Respondent's Use  of  Several  Management Practices,
        1974 and 1980 Percent of Respondent's Using The Practice
Management Practice       	       1974            1974*	1980
Conservation Cropping
Contour Farming
Crop Residue Management
Field Borders
Grade Stablization Structures
Grassed Waterway or Outlet
Holding Pond or Tank
Livestock Exclusion
Farm Pond
Strip-cropping
83.3%
2.2
41.6
36.0
18.0
33.7
13.5
18.0
18.0
—
88.9%
1.9
46.3
38.9
14.8
27.8
14.8
16.7
16.7
—
62.9%
5.6
35.2
48.1
18.5
44.4
9.2
9.2
18.5
—
*0nly  includes  those  landowners  who  also  responded in 1980.
     The  reduction  in  the  use  of  conservation cropping provides one of
the most  interesting findings  in  Table 7.   In 1974 over 80% of the land-
owners were using conservation cropping,  while in 1980, only 62.9% of the
landowners were  using  conservation cropping.   This may suggest a trend
towards mono-agriculture  in  the Black Creek area, where there is greater
reliance  on continuous  cropping.   Crop residue management, another prac-
tice that  experienced  a decline in use between 1974 and 1980, will be
discussed  in  detail  later  because of its  importance as a management tool
in  reducing soil erosion.

     The  use  of  grassed waterways or outlets illustrates the opposite
trend exhibited  by  conservation cropping  and crop residue management.
There has  been  a rather significant increase in the implementation of

-------
                            - 253 -
grassed waterways  since 1974.   This may,  in part, be the result of low
capital costs  and  80% cost-sharing by the Allen Co. SWCD.  When focusing
specifically on  those 54 respondents who  participated in both interviews
(1974  and  1980)  the  increase in the use of grassed waterways rose from
17.8%  in 1974  to 44.4% in 1980.

     We now turn our attention to Table 8 which focuses directly on
information gathered from the  farmers in  the 1980 survey.  The first col-
umn  of this table  indicates the number and percentage of farmers report-
ing  either  the current use of  or the past use of each land use practice.
The  practice most  widely adopted (100%) by the landowners is that of tile
drains.  While,  tile drains are an essential part of many of the farming
operations  in  the  Black Creek  area, they  have little value for water
quality.   Their  main value is  in terms of crop production.  Therefore, it
is not surprising  to have such a high adoption rate among Black Creek
area farmers especially when the district was cost-sharing at 70%.  The
The  use of  practices and structures depends, in part, upon local condi-
tions  and  inducements offered.

     The use of  tile drains can be contrasted with stripcropping.  Only
one  farmer  reported  using stripcropping on his farm.  Further, only six
individuals indicated that they had received any information about this
practices.  Such a low rate of use may be attributable to a lack of
applicability  of stripcropping to most of the area farms or lack of
information regarding procedures and consequences.

     Over  sixty  percent of the respondents indicated that they use a con-
servation  cropping system,  which makes it the second most utilized best
management  practice  (BMP) among the Black Creek farmers.  No other land
management  practice  was reported being used by more than 50% of the farm-
ers.   However, field borders (48%) and crop residue management (37%)
showed substantial rates of adoption.  The low rate of adoption for some
practices may  result from some practices  not being appropriate for all
the  farming operations.  Two examples of  this include, contour farming
and  livestock  exclusion.  Much of the land in the Black Creek area is
relatively  flat, thus not requiring contour farming.  In addition, many
of the hills are in  permanent  vegetation.  In a similiar manner, live-
stock  exclusion  has  no relevance to those farmers with a strictly cash-
grain  operation.   Another consideration in examining adoption rates for
various practices  is the amount  of assistance the farmer has received.

     Each  respondent was asked if he/she  had received any technical or
financial assistance in instituting each  of the land management practices
(see Table  8).   The  only practices in which less than 50% of the farmers,
using  the specified  practice,  had received technical assistance were:
conservation cropping,  contour  farming, holding ponds or tanks and tile
drains.  However,  conservation  cropping and tile drains were established
by many farmers, prior to the  Black Creek Demonstration Project.  Many of
the  landowners using conservation cropping indicated that this was the
way  they had learned to farm.   Therefore, few reported receiving techni-
cal  assistance in  starting  their conservation cropping system.   Similarly,
many respondents stated that tile drains  were already on their land when
they began  farming.   The percentage of those receiving technical assist-
ance for contour farming and holding ponds or tanks was low.   However,
there  were  very  few  farmers  utilizing these practices.

-------
Table 8:  Specific Information  on  Selected  Best  Management  Practices
Using
or
Used
Management
Practice
Conservation
Cropping
Contour Farming
Crop Residue
Management
Field Borders
Grade
Stabilization
Structures
Grassed Waterway
or Outlet
Holding Pond
or Tank
Livestock
Exclusion
Farm Pond
Strip Cropping
Surface Drains
Tile D.rains
N

35
3

20
26


10

2&

5

6
11
1
14
54
%

64.5
5.5

37.0
48.1


18.5

48.1

9.3

11.1
20.4
1.8
26.0
100.0
Received
Technical
Assistance
N

10
0

10
22


10

15

2

4,
8
-
10
26
%

28.6
0.0

50.0
84.6


100.0

57.7

40.0

66.7
7-2.7
—
71.4
48.1
Received
Financial
Assistance
N

7
0

7
21


9

14

2

2
8
-
10
23
%

20,0.
0.0

35.0
80.8


90.0

53.8

40,0

33.3
72.7
—
71.4
42,6
Increase
In Profit
N

25
2

11
5


1

14

4

1
1
-
10
45
%

71.4
66.6

55.0
19.2


10,0

53.8

80.0

16.6
9,1
—
71.4
83.3
Use Has
Effect
On Water
Quality
N

32
3

19
26


10

26

4

5
10
-
14
49
%

91.4
100.0

95.0
100.0


100.0

100.0

80.0

83.0
90.9
—
100.0
90,7
Have
Never
Used
N

19
51

34
28


44

28

49

48
43
53
40
T
%

35.5
94.5

63.0
51.9


81,5

51.9

90.7

88.9
79.6
98,2
74,0
0.0
Never
Used But
Received
Information
N

7
16

8
5


4

9

9

-
I?
6
5
-
%

36.8
31.3

23.5
17.8


9,0

32.4

20.4

—
28,0
U.3
12,. 5
— r
Never Used ,
But Thought
it Had Effect
on Water
Quality
N

4
16

6
4


4

9

9

-
7
6
5
-
%

57.1
100.0

75.0
80.0


100.0

100.0

100.0

—
58.3
100.0
100.0
—
                                                                                                                            I
                                                                                                                            <-n
                                                                                                                            I

-------
                            - 255 -
     For most  of  the  practices,  approximately the same percentage of
farmers who reported  receiving  technical assistance also said they had
received financial  aid  to  start  the practices.  The major discrepancies
from this pattern were  found  among the adopters of livestock exclusion
and crop residue  management.   This suggests that in the Black Creek area,
financial assistance  and technical assistance seem to have gone hand in
hand.  The need  for these  types  of assistance is often based upon ideas
of lack of knowledge  concerning  the practice and effects of adoption on
farm profits.

     Some land use  practices,  designed to control agricultural run-off,
may reduce farm  firm  profits.   This is obviously an effect which must be
considered in  persuading farmers to adopt such innovations.  In addition
to an objective  appraisal,  the  farmers'  own perceptions of how each prac-
tice has affected their income,  can be very important with regard to
their continued  use and in  the  information transmitted to non-adopters.
In order to determine adopting  farmers perceptions, the following ques-
tion was asked about  each  agricultural practice that had been adopted by
a particular farmer:   "How does  or did it affect income or profit?"

     The fourth  column  of  Table  8 indicates the number of respondents
reporting an increase in income  or profits due to adoption of each land
use practice.  Conservation cropping and tile drains were both cited by a
large percentage  of farmers as  increasing profits.  Slightly more than
half of those  farmers utilizing  crop residue management and grassed
waterways also reported an  increase in profit from the use of these prac-
tices.  However,  Table  7 indicated a decline in the use of conservation
cropping and crop residue management.

     The case  of  crop residue management is a particularly interesting
and important  one.   The environmental nature of this practice might sug-
gest that it negatively influences income.  However, 55% of those using
crop residue management attributed an increase in profit to its use.
This point could  be emphasized  to potential adopters of this practice.

     Very few  respondents who were using grade stabilization structures
or ponds said  they  had  experienced monetary gains from their adoption.
However, it should  not  be  interpreted that these practices negatively
affect income. Most users  of  both grade  stablization structures and ponds
said these practices  simply had  no influence on their income or profit.

     Each adopter of  a  particular land use practice was asked if he felt
the use of the practice had any  effect on the quality of the water in our
rivers, lakes, and  streams.  This was done in order to assess the farm-
ers' awareness of the environmental implications of each of the land use
practices.  Table 8,  indicates  that over 80% of the adopting farmers, for
any specific practice,  believed  it had affected water quality.   The last
two columns of the  table reveal  a similiar situation for nonadopters.
Most of the respondents who had  not adopted a specific management prac-
tice, but had  received  information about it, also felt the practice would
affect the quality  of our water  resources.   This demonstrates the success-
ful communication of  the impact  of these land use practices upon the
quality of our water  resources.  For most of those farmers  contacted about
a certain management  practice, whether they had adopted it or not, they
apparently understood its value  in maintaining water quality.

-------
                            _ 256 _
     Table 9 indicates  the  specific  reasons  given for initiating particu-
lar land use practices.  An  important  point  illustrated by the table is
the significant number  of persons  citing  erosion control as the reason
for starting many of the practices.  This  suggests  farmers who adopt
these practices have an understanding  of  the need to control erosion and
that it enters into this decision-making  process.

     We now turn our attention  specifically  to crop residue management,
one of the more important methods  for  controlling soil erosion in the
Black Creek watershed.  Of  the  54  respondents questioned in 1980, 19 said
they were currently using crop  residue management,  one person indicated
he had used this practice in the past, and the remaining 34 farmers waid
they had never used crop residue management.  Those persons who indicated
that they were using crop residue  management were asked if they were
using chisel plowing, minimum till or  no  till systems.  Fourteen (14)
reported using chisel plowing,  three were  using a minimum till system,
and one was using no till.   Most of  the farmers (84.2%) indicated they
had adopted the practice of  using  crop residue management during the
Black Creek Demonstration Project.

     The reason mentioned most  often by these farmers for adopting crop
residue management was  to reduce soil  erosion.  Two of the individuals
cited business reasons  for  initiating  the  practice, while four said they
started because it was  recommended by  others or because of financial aid
received.  When these individuals  were asked how crop residue management
specifically affected their  income or  profit the following responses
resulted:  four said it had  no  effect, three said it produced higher
yield/ better crops, four indicated  that  it  had generally reduced operat-
ing cost, one suggested it  saved labor, anmd one landowner indicated that
crop residue management had  improved drainage.  One respondent said crop
residue management had  contributed to  a loss in profit as a result of
later planting.  Nineteen of the twenty farmers who had adopted crop
residue management felt it  has  an  effect  on  the quality of our water
resources.

     Eight of the 34 farmers who reported  never using crop residue man-
agement, indicated that they had received  some information about it.
These eight farmers were asked  why they had  never used crop residue man-
agement.  They responded as  follows:  four said there was no need for it
on their land, one person thought  fall plowing was better, another indi-
vidual said it didn't work  for  a friend,  and the remaining two indicated
that they had no reason for  not using  the practice.

     The conservation practices covered by the Black Creek project have a
great deal of variance  in terms of conservation of soil and in terms of
wide applicability.  However, one  set  of  practices which seem to have
wide utility is that of crop residue management.  Crop residue management
refers to the use of residue on the  surface  to discourage the runoff of
water (which of course  carries  soil  with it).  The most common methods of
crop residue management include chisel plowing, minimum tillage cropping
and no-till cropping.   These practices were  promoted  in the Black Creek
project as BMP's which  the  farmer  should  adopt.  Apparently, there was
limited success at encouraging  the adoption of these  practices.  Of the
54  farmers contacted  in 1980, only 19  were using any  of these BMP's, 14

-------
 Table  9:   Respondent's  Reason for Starting  Several  Management Practices
Recommended
Keep or
Improve Reduce Land Business Financial Tradition/
Drainage Erosion Fertile Reason Aid Habit Misc.* Total
Management Practice N%N% N % N%N % N % N%N%
Conservation Cropping 1 2.9 2 5.6 18 51.4 3 8.6 3
Contour Farming - — 1 33.3
Crop Residue
Management - — 8 40.0 - — 5 25.0 5
Field Border - — 18 69.2 - — 1 3.8 5
Grad. Stab. Structure - — 7 70.0 - — - — 3
Grassed Waterway
or Outlet 8 30.8 13 50.0 - — 2 7.7 2
Holding Pond or Tank ' - — - — - — 4 80.0
Livestock Exclusion - — 2 33.3 - — 1 16.7 1
Farm Pond 1 11.1 1 11.1
Strip Cropping - — - — - — - — -
Surface Drains 8 57.2 4 28.6 - — - — 1
Tile Drains 45 85.0 1 1.9 1 1.9
8.6 7 20.0 1 2.9 35
----- 2 66.6 5

25.0 - 2 10.0 20
19.3 - — 2 7.7 26
30.0 - — _ — 10

7.7 - — 1 3.8 26
1 20.0 - — 5
16.7 - — 2 33.3 6
9* 81.8 11
—
7.1 - — 1 7.1 14
5 9.3 1 1.9 53
100
100

100
100
100

100
100
100
100
100
100
100
                                                                                                                   to
                                                                                                                   Ul
*Miscellaneous category  for  farm pond  includes:  Recreation  4,  drinking for livestock 1, sediment basin 2,
and needed the dirt, 2.

-------
                            - 258 -
were using chisel plowing, 3 minimum tillage  cropping and 2 no-till crop-
ping.  One other  farmer  indicated  that  he  was using crop residue manage-
ment, but the exact practice was not ascertained by the interviewer.
Sixty-five percent (65%) of the  farmers contacted had not adopted any of
the crop residue management practices.   While this has implications for
the water quality and soil conservation goals of the project, it also has
implications  for  the  socio-economic  consequences of participation in the
Black Creek project.  With that  in mind,  the  adopters of crop residue
management practices  will be contrasted with  non-adopters.

Adopters of Crop Residue Management  and Non-Adopters

     Respondents were, as illustrated previously, questioned about their
use of several  land use  practices.   For brevity, we decided to choose 1
practice for  intensive investigation of the consequences of adopting a
BMP.  Since crop residue management  practices are extensively involved
with water quality protection  and  it is often purported to have conse-
quences more  significant than  other  practices, it was selected for inten-
sive investigation.   We  begin  by contrasting  changing farm firm charac-
teristics for adopters and non-adopters.

     As indicated in  Table 10,  the average number of acres owned by our
respondents increased from 1974  to 1980.   Non-adopters, on the average,
increased their  land  holdings  by 24%, but  adopters increased their hold-
ings by 155%.  The average adopting  respondent farms over 4 times as many
acres as the  average  non-adopt ing  farmer.   The adopters tend to be larger
farmers and have grown more rapidly  from 1974 to 1980 than non-adopting
farmers.  While  non-adopters,  on the average, own increased acreage in
1980, they have  fewer acres in  crops, 3% less acreage in corn and 11%
less acreage  in  soybeans.  Adopters, on the other hand, have increased
their corn acreage by 55% and  their  soybean acreage by 86%.

     While yields seem to have  increased for  both groups of respondents,
the non-adopters have experienced  greater  percentage increases in yield
although their  corn yield  is still below the  adopters.  Whether this dif-
ference in percentage yield increases in due  to the use of crop residue
management or is  a spurious finding  is unknown.  Another anticipated bene-
fit of inducing  a farmer to adopt  a  conservation practice is the carry-
over to other practices.   In other words,  if he adopts one practice he
will, presumably, be  more  likely to  adopt  other conservation practices.
This anticipation seems  to hold for  adopters  of crop residue management
in  Black Creek.   For  each of the other selected practices, farmers using
crop residue  management  are more likely to be adopters of the other con-
servation practices.  These results  indicate  that there is a correlation
between use of  crop residue management and other conservation practices,
and  suggests  that getting a farmer to adopt one practice increases the
probability of  his adopting other  practices.

     The next question  to be addressed is the differences in environmen-
tal  and conservation  attitudes between adopters and non-adopters.  While
we  have indicated that  adopters of one practice seem to be more  likely  to
adopt other practices,  we  have not suggested  any causal  factor.  One such
causal  factor could be  that the adopters become more conscious of envi-
ronmental  and conservation  problems and therefore  change their  attitudes
about these problems  and their solutions.

-------
                            -  259  -
   Table 10:  Farm Firm Characteristics  for Adopters  and Non-Adopters
                                                       Respondents  not
                          Respondents using            using  crop
                          crop residue                 residue management
Characteristic	management in 1980	in  1980	

Acres Owned
   1974 (X)                    103.5                         85.8
   1980 (X)                    263.5                       106.5
Acres Farmed 1980 (X)          523.4                       116.7
Acres in Crops
   1974 (X)                    105.2                         97.7
   1980 (X)                    473.6                         91.9

Corn for Grain - Acres
   1974 (X)                     87.2                         37.0
   1980 (X)                    136.0                         35.9

Corn for Grain - Yield in Bushels/Acre
   1974 (X)                     85.9                         66.0
   1980 (X)                    125.3                       110.0

Soybeans - Acres
   1974 (X)                    103.7                         31.0
   1980 (X)                    193.2                         27.7

Soybeans - Yield in Bushels/Acre
   1974 (X)                     32.1                         29.8
   1980 (X)                     39.9                         42.4

-------
                            - 260 -
     Table 11:  Use of Other Conservation  Practices  by Adopters  and
                 Non-Adopters of Crop  Residue  Management
                                          Crop  Residue  Management
Other Conservation Practices
Conservation Cropping
Contour Farming
Crop Residue Management
Field Borders
Grade Stabilization Structures
Grassed Waterway or Outlet
Holding Pond or Tank
Livestock Exclusion
Farm Pond
Strip-cropping
Surface Drains
Tile Drains
Adopter
65%
15%
100%
60%
35%
60%
10%
20%
15%
30%
100%
Non-Adopter
62%
41%
9%
35%
9%
15%
21%
24%
97%
     Table 12 indicates that the adopters were more  pro-environment  in
terms of utilizing available technology  and Federal  taxation.   They  were
also less likely to agree that  farmers must,  primarily,  be  concerned with
profits.  However, a greater percentage  of non-adopters  agreed  that  it is
very important to clean up the  environment.
     Table 12:  Environmental Attitudes  of Adopters  and  Non-Adopters
Statement
Percent Agreeing with Statement
   Adopter	Non-Adopter
Even considering the cost, all available
pollution control techniques should be
used                                          40%

Federal taxation to clean up our water
completely wouldn't be too expensive  to
consider                                      90%
It is very important to clean up  the
environment
Farmers are businessmen and therefore must
be primarily concerned with profits            70%

Farmers have a responsibility to preserve
the land for future generations                95%
                     35%
                     60%
                     91%
                     85%
                    100%

-------
                            -  261 -
     Comparing the attitudes of  adopters  in  1974  and  1980,  a greater per-
centage indicated that conservation of  soil  is  a  problem (see Table 13).
This suggests that the adoption  of crop residue management  and the accom-
panying education have changed the attitudes  of these farmers.  However,
for non-adopters, fewer thought  that  soil  conservation was  a problem in
1980.  This suggests the possibility  that  these farmers think the project
has controlled or eliminated the  problem.  Adopters  are also more likely
to feel that pollution of streams is  a  problem  in the county and to indi-
cate that soil erosion contributes to water  quality.
     Table 13:  Conservation Attitudes  of  Adopters  and Non-Adopters
                                                     Percent  Agreeing
                                                	  with Statement
Statement		 	Adopter      Non-Adopter

Conservation of soil  is not  a  real problem
in this area
     1974                                          55%           47%
     1980                                          30%           53%

The average landowner  in  this  county  stands to
lose more than he will gain  by soil and  water
development programs                                5%            9%

The cost of soil erosion  reducing practices
(e.g., field borders,  grassed  waterways) should
be borne entirely by  those who adopt  them         30%           29%

The federal government should  play an important
role in soil conservation programs in this
county                                             70%           53%

Pollution of the streams  is  a  major problem in
this county                                        40%           18%

Landowners have little opportunity to express
their opinions in planning watershed  problems     20%           41%

Soil erosion contributes  to  water pollution
problems                                           90%           85%
     Table 13 also  indicates  that  adopters  are more likely to think that
the federal government should play a  role  in  soil  conservation programs
in the county.  Even  though adopters  thought  that  the federal government
should be involved, they were still less  likely to feel  that landowners
cannot express their  opinions in watershed  projects.   However, 20% of
adopters and 41% of non-adopters  felt  that  there was  little opportunity
for landowner input into projects  in  the  watershed.

-------
                           -  262 -
     The attitude toward soil erosion  and  local watershed  projects  may
also have effects upon the respondent's views  concerning responsibility
for the protection of water quality.   Two  important  questions  arise here,
who is responsible for the protection  of water quality  in  our  lakes,
rivers and streams and who should be?  Tables  14  and 15 indicate  the
respondent's answers to the questions  posed  above.   In  terms  of who is
responsible, adopters did not overwhelmingly choose  one answer, but T5%
selected the landowners.  Among non-adopters,  the most  conspicuous  answer
was "don't know", suggesting that they feel  uninformed. Another  inter-
esting response was that the various levels  of government  and  government
agencies were not selected by numerous respondents.


      Table 14:  Who is Responsible  for Protection  of Water Quality
Respo_nsible Party   	          Adopters	        Non-Adopters

Landowner                           35%                   18%
Local Unit of Government             5                     6
State Government
Federal Government                   5                     3
SCS                                 20                     3
ASCS                                --                     6
SWCD                                10                    12
Don't Know                          25                    53
     N =                            20                    34
  Table  15:  Who  Should  be  Responsible  for Protection of Water Quality
Responsible Party	Adopters	Norn-Adopters

Landowner
Local Unit of Government
State Government
Federal  Government
SCS
ASCS
SWCD
Don't Know
     N -
      The  more  interesting results are the differences expressed between
 who  is  and  should be responsible.  Table 15 indicates that 65% of adopters
 thin~that  the landowners should be responsible and 32% of non-adopters
 also feel that way.   This suggests that adopters have accepted more of
 the  responsibility for water quality protection than non-adopters.  This
 is  an especially important aspect for future environmental behavior,  since
 taking  personal responsibility is the first step in going from environ-
 mentally sound attitudes and values to environmentally sound behavior.
65%
at 5
5
20
—
—
5
20
32%
21
3
6
6
18
15
34

-------
                            - 263
     While  adopters  tend  to  indicate a greater sensitivity to environmen-
tal problems,  they also  tend  to  feel that  pollution is under control in
the Black Creek  watershed.   Table 16 indicates that 100% of adopters
thought pollution was  under  control, while only 76% of non-adopters
thought it  was controlled.   As would be expected, adopters were more
likely to have personally benefitted from  the Black Creek project and they
were slightly more likely to  feel that money spent for soil and water
development was  a good investment.   Another interesting aspect is that
adopters were more likely to  feel that major decisions should be made by
professional/technical staff.  Possibly the non-adopters were reacting to
their, previously discussed,  feeling that  landowners have little oppor-
tunity for  input into  the decision-making.  The next section will outline
the involvement  of the respondents  in the  Black Creek Project.
           Table  16:  Attitudes  Toward  the Black Creek Project
Question	^	   _ 	Adopters	Non-Adopters

Percent who  think  that  pollution control for
Black Creek  is now excellent  or  good             100%           76%

Percent who  personally  benefitted from Black
Creek Project                                      90%           56%

Percent who  thought  that  major decisions in
the demonstration  project  should be  made by
professional/technical  staff                       75%           62%

Percent who  felt that spending money for soil
and water development is  a good  investment        100%           94%
     The adopters  seem  to have  had  a great  deal of contact with project
personnel and attended  a number of  public meetings concerning the project,
Eighty percent  indicated substantial contact  and they attended, on the
average, 8 public meetings.  Non-adopters,  on the  other hand, attended
fewer than 3 public meetings and nearly 60% had little or no contact with
project personnel.  In  addition,  this  difference in involvement is indi-
cated by differences  in rates of discussions  with  neighbors.  85% of
adopters had discussed  the project  some or  a  great deal with neighbors,
while 44% of non-adopters had not discussed the project with neighbors or
had had very little such discussion.

     Holding public meetings and other educational activities are often
aimed at giving farmers enough  information  to enable them to make wise
decisions.  Of  course,  project  personnel hope that those decisions will
be to adopt conservation practices.  This process  is a result of our dom-
inant ideology which  stresses that  participation should be voluntary.
Government can  induce participation  through use of education, technical
assistance, and financial incentives.   In earlier  surveys among Black

-------
                            _ 264 _
      Table 17:  Involvement  and Discussion of Black Creek -Project
Question	        	   Adopters     Non-adopters

"How familiar are you with the Black  Creek
Demonstration Project in  this county?"
   Never heard of it
   Heard, but no cohtact                           10%            29%
   Little contact                                  10             29
   Contact w/various  project representatives      30             26
   Much contact and participation                  50             15

Have you discussed  the Black Creek  Project  with
your neighbors?
   A great deal                                    35%            18%
   Some                                            50             38
   Very little                                     10             38
   Not at all                                       5              6

How many public meetings  concerning the project
have you attended in  the  past 7  or  '8  years? (X)    8.0            2.7
Creek respondents,  they  indicated  that  education was the best way 'to get
people to cooperate in helping  to  protect water "quality.  However,, in
1980, we saw some  shifts  in  this attitude.   Table 18 indicates that more
respondents  felt  that  financial incentives were a better way.  Among
adopters, support  for education as a mechanism declined by 30% while sup-
port for financial  incentives doubled.   Among non-adopters, the major
shift was away  from a combined  approach to the use of financial incen-
tives.  These results  suggest some disillusionment with education as a
motivator and increasing  support for financial incentives as motivators.
                 Table  18:   How to Get People to Cooperate
"What do you think is the better way to
get people to cooperate in helping to
protect water quality in the Black Creek?"
Education
Finaricial
Laws and Controls
Combination of above
N =





Adopters
1974 1980
65%
15
10
10
20
45%
30
10
15
20
Non-adopters
1974 1980
56%
1.2
6
26
34
56%
26
3
1.5
34

-------
                            - 265 -
     We have  outlined  numerous differences between adopters and non-
adopters  including  farm firm characteristics, attitudes and other  land
use practices.   We  have seen that adopters seem to prefer different typos
of inducements  for  getting  cooperation.   Part of these differences may be
the result  of differences  in objectives.  In other words, adopters may
have different  goals  from non-adopters.   In order to assess this, we
asked  the respondents  to rank-order seven possible objectives which are
common for  farmers.   Table  19 reports on the mean (average) ranking given
each goal or  objective by both adopters  and non-adopters.  Adopters are
more likely to  have  stability and certainty of income as an objective
when compared to non-adopters and to be  slightly more concerned with
yields and  increases  in the value of the farm.  Non-adopters are more
likely to be  pursuing  the goals of condition of the farm, time for family
and non-work  and high  consumption.  The  differences in goals and objec-
tives  are vital for  project planning and management.  Programs designed
to help the farmer move toward certain goals will not be successful with
all farmers.  Some  will be  concerned with consumption and increased farm
values, while others  will be concerned with stability or certainty of
income, time  for family or  general condition or appearance of their farm.
Tailoring of  the product (e.g. the soil  and water project) to help the
client's  (farmers)  move toward his goals and objectives is a time honored
principle of  successful salesmanship and good business.
            Table  19:   Mean Ranks of Goals and Objectives of
                         Adopters  and Non-Adopters
Goal or Objective  for  Farm Firm
    Mean Ranking*
Adopters    Non—Adopters
Stability of Income
High Level of Consumption
Fast Increase in Value of Farm
Time for Family and Non-work Activity
Certainty of Income
Condition of Farm (e.g. equipment,
buildings, land)
Greatest Yields
2.6
5.6
4.6
4.9
2.5

4.0
3.8
5.4
5.4
5.0
4.6
3.2

3.0
4.0
*Respondents were  asked  to  rank these objectives from 1 to 7, with 1
being most important  and 7  being least important.
Social, Economic and Demographic  Characteristics.  Adopters are younger,
better educated and have higher  incomes"!   They are less likely to have
off-farm employment.   They  receive a greater proportion of their income
from farming and are more likely  to be  cash-grain farmers.  They also
expend more labor on their  farms,  partially a function of size.  They are,
more likely to have a  conservation plan with SCS although fewer indicated
they have them now than  in  1974.   Adopters are also using more fertilizer
herbicide and insecticide,  a  result expected from the technical studies
of crop residue management.

-------
                            - 266 -
     In general, adopters are 'bigger,  full  time farmers with larger
incomes and with less income from  off-farm  sources.

Leader-Nonleader Comparison

     Opinion leaders  informally  influence the attitudes and behavior of
other individuals in  a community.   Their  leadership  is  not  a result of
any official position held in the  community,  but rather it  is based on
respect from their followers.  Most  research  on adoption and diffusion of
agricultural practices stresses  the  importance of opinion leaders in the
dissemination of information.  Therefore, any effort directed towards
inducing change among a group of individuals  should  attempt to identify
the informal leadership structure  that  exists within the group.   Fortun-
ately, we were able to determine the opinion  leaders in the Black Creek
area by asking the following question:   "Who  do you  think is well
respected in this area for his general  agricultural  practices and abili-
ties?"  Through this  procedure 13  farmers were identified as opinion
leaders.

     Since the Black  Creek Demonstration  Project has officially ended,
the role of the opinion leaders  becomes increasingly important.   It is
through them that much of the continued success of the  project's goals
will depend.  Therefore, is  is imperative that we examine the current
attitudes and behaviors of these individuals  towards pollution control.
Also, in order to assess their influence  on the other farmers in the
area, a comparison of the two group's  attitudes and  behavior is essen-
tial.  However, before doing this  it is necessary to compare certain farm
related characteristics between  the leaders and non-leaders.  This is
done, in part, to enhance our understanding of differences  in attitudes
and behavior that may exist  between leaders and non-leaders.  In other
words, there may be certain  characteristics that leaders possess that
will tend to influence them  toward forming  different opinions about ero-
sion control than non-leaders.   One of  these  possible factors is the
amount of acreage farmed.

     There is a large difference in the average farm size of leaders and
non-leaders in the Black Creek area.  For  leaders and non-leaders the
average number of acres farmed  is  573  and  170 respectively.  Also, 10
percent more of the leaders  than non-leaders  reported farming more land
in  1980 than they had in 1974  (Table 21).   The percentage of time devoted
to  one's farming operation also  seems  to  contribute  to  the  leaders status
in  the agricultural community.   This was  affirmed by the farmers in the
Black Creek area.  Seventy percent of  the  non-leaders indicated having
off-farm employment,  while only  53.8%  of the  leaders reported off-farm
employment.

     In summary, these farmers perceived  as leaders  differ from the
remaining farmers.  Leaders  tend to farm more land,  on  the  average, than
non-leaders.  They have also shown a greater increase in acreage farmed
since 1974.  Opinion  leaders  tend  to be more  fully involved in farming
than non-leaders, as  revealed  by the percentage of off-farm employment.
These differences should be  kept in mind  as we compare  these two groups
in  terms of environmental  attitudes.  A logical place to begin this com-
parison is with the  farmers' perceptions of pollution as a problem.

-------
                       - 267  -
Table 20:  Social, Economic and Demographic Characteristics  of
     Adopters and Non-Adopters of Crop Residue Management
Characteristic
Age - 1980 (X)
Years of Formal Education (X)
# in Household (X)
# children age > 18 (X)
# children age < 18 (X)
Income, 1979, Gross, Median
% with off-farm employment
% of income from farming (X)
% of income from spouse (X)
% of Farming Income (X) From:
Crops
Livestock
Reported Market Value of Land
(X) $/acre
Hours of Operator Labor/week (X)
April to October
November to March
% with spouse providing farm labor
% using more in 1980 than in
fertilizer
herbicide
insecticide
% indicating that they had a
conservation plan with SCS
1974
1980
Adopters
49.4
12.1
6.4
1.1
.2
51749.65
55%
73.6%
3.0%
88.2
11.8
2287.5
53
34
50%
1974
35%
70%
40%
35%
30%
Non— Adopters
53.9
9.4
4.7
1.7
.7
27499.50
74%
54.8%
5.2%
61.0
38.6
2354.3
39
22
56%
29%
35%
12%
18%
15%

-------
                            - 268 _
       Table 21:  Are You Farming More  Land,  Less  Land or the Same
                               (Since  1974?)
Leader
Response
More
Same
Less

N
4
9
0
13
%
30.8
69.2
O.'O
100.0
Non-Le ader
N
8
31
2
41
%
19.5
75.6
4.9
100.0
     Awareness of the erosion  and  water pollution problem is an essential
first step in persuading  farmers to  adopt  and maintain best management
practices.  With this in  mind  there  was a major effort by .the Black Creek
Project to inform the landowners of  the existing pollution problem.  To
determine the current level  of awareness among the Black Creek farmers
they were asked to agree  or  disagree with  several statements -pertaining
to the pollution problems.   The following  statement, "Soil erosion con-
tributes to water pollution,"  was  presented to the ^respondents to  find
out  their 'knowledge of  the  lirtk between soil erosion and water pollution.
As shown in Table 22,, a large  percentage of both leaders and non-leaders
agreed with this Statement.
    Table 22.  Soil Erosion  Contributes to Water Pollution Problems.
Leader
•Response
Agree
Disagree
Don't know

N
11
1
1
T3"
%
84.6
7.7
7.7
100.0
Non-Leader
N
36
3
. 2
4T
%
87 .-8
7.3
4.9
100.0
     However, -knowledge  of  this  link does not necessarily imply  that 'the
farmers  are  aware .of  any local soil conservation or water .pollution  prob-
lems.  The  farmers' perceptions  of pollution as a problem in the  Black
Creek  area  are  presented in Tables 23 and 24.

-------
                            - 269 -
Table  23.   Conservation of Soil  is Not a Real Problem in This Area.
Leader
Response
Agree
Disagree
Don't know

N
4
9
0
13
%
30.8
69.2
0.0
100.0
Non-Leader
N
20
19
2
41
%
48.8
46.3
3.7
100.0
  Table  24.  Pollution of the  Streams is a Major Problem in the County
Leader
Response
Agree
Disagree
Don ' t know

N
6
6
1
T3
%
46.2
46.2
7.7
100.0
Non-Leader
N
8
30
3
4T
%
19.5
73.2
7.3
100.0
     As might be  expected,  a  larger  percentage of leaders than non-leaders
perceive  soil conservation  as  a real problem in the area (Table 23).
Unfortunately, knowledge  of the soil conservation problem has not com-
pletely filtered  down  from  the  leaders  to the non-leaders.   The two groups
differed  even greater  in  their  opinions  concerning pollution of the
streams in  their  county (Table  24).   Forty-six percent of the leaders
felt water  pollution was  a  major  problem in the area,  while only 19.5% of
the non—leaders agreed with this  conclusion.  Although both groups see
conservation of the soil  as more  of  a local problem than pollution of the
streams,  this difference  is much  greater among non-leaders  than leaders.
This again  shows  that  the communication  channels and levels of informa-
tion between the  two groups is  not  identical.

     In general,  those individuals chosen as leaders exhibit more aware-
ness of the local pollution problems.  As suggested above,  the informa-
tion has  not filtered down  to the non-leaders.   Therefore,  future efforts
designed  to educate the public  about pollution problem may  want to consi-
der more  direct communication with as many individuals as possible.

     Although awareness is  essential for controlling the pollution prob-
lem, it does not guarantee  the  farmers'  cooperation.   One way of encour-
aging the use of  pollution  control techniques by farmers is through gov-
ernment involvement.  However,  the proper role  of the  government is  not
easily determined.  The following section attempts to  assess the leaders'
and non-leaders' attitudes  towards government involvement in pollution
control.

-------
                            _ 270 _
     In order to understand  the  willingness  of farmers to accept outside
intervention into this problem the  farmers were  asked to agree or disagree
with the following  statement:  "The cost  of  soil erosion reducing prac-
tices should be borne entirely by  those who  adopt them."  Although a
majority of both groups  disagreed  with  this  statement, substantially more
leaders than non-leaders  thought that efforts  to control soil erosion
should be paid  for  by individual landowners  (Table 25).  However, this
may simply represent the  economic  differences  between leaders and non-
leaders .
    Table 25.  The Cost  of  Soil  Erosion Reducing Practices Should be
                 Borne Entirely  by  Those Who Adopt Them.
Leader
Response
Agree
Disagree
Don't know

N
6
7
0
13
%
46.2
58.8
0.0
100.0
Non-Leader
N
10
28
3
41
%
24.4
68.3
7.3
100.0
     As  seen  earlier,  opinion leaders generally have larger farms than
non-leaders.  Therefore,  they may be better able to afford the cost of
erosion  control  practices.   It is interesting to note however that while
a  larger percentage  of leaders felt  that soil erosion practices should be
borne  primarily  by  those  who adopt them, over three-fourths said the  fed-
eral government  should play  an important role in local soil conservation
programs (Table  26).   Among  non-leaders, 53.7% agreed with this statement.


   Table 26.  The Federal Government Should Play an Important Role in
                 Soil Conservation Programs in This County
Leader
Response
Agree
Disagree
Don ' t know

N
10
3
0
13
%
76.9
23.1
0.0
100.0
Non-Leader
N
22
16
3
41
%
53.7
39.0
7.3
100.0
      Government  involvement in pollution control programs can encompass  a
wide  range of activities including financial and technical  assistance.
In  order  to assess the farmers attitudes towrds the extent  to which  fed-
eral  taxation should be imposed to correct this problem  the  following

-------
                            -  271 -
statement was presented to the  farmers:   "Federal  taxation to clean up
our water completely wouldn't be  too  expensive to  consider."  The respon-
dents were asked if they strongly  agreed,  agreed,  disagreed or strongly
disagreed with this statement.  As seen  in Table 27 no one strongly agreed
with unlimited federal taxation to clean  up our  water resources.   However,
a large percentage (46.2%) of the  opinion leaders  expressed a positive
response to this idea.  This is contrasted with  non-leaders in which only
29.3% agreed with this statement.
      Table 27.  Federal Taxation  to  Clean Up Our Water Completely
                  Wouldn't be  Too  Expensive to Consider
Response
Strongly Agree
Agree
Disagree
Strongly Disagree
Don ' t Know

Leader
N %
0
6
5
2
0
T3
0.0
46.2
18.5
15.4
0.0
100.0
Non-Leader
N %
0
12
22
2
5
41
0.0
29.3
53.7
4.9
12.2
100.0
     While many  opinion  leaders  think that  pollution control practices
should be born primarily by  those  who adopt  them,  they are also more
inclined to favor  federal  involvement in soil and  water conservation pro-
grams.  This would  seem  to indicate  more of  an overall willingness by
leaders to actively support  soil  and water  development programs.  However,
the above findings  do  seem to  suggest some  question as to who should
actually be responsible  for  controlling  soil erosion and water pollution.
It is, therefore,  important  that  we  examine  more directly the farmers'
opinions as to who  should be responsible for controlling soil erosion.
Landowners in the  survey were  asked  to identify who they thought should
be responsible for  controlling the soil  erosion problem.  As illustrated
in Table 28, nearly 70%  of both  the  leaders  and non- leaders indicated
that either the  individual landowner, local  government or the local soil
and water conservation district  should be responsible for controlling the
soil erosion problem.  The leaders specifically mentioned the individual
landowners more  often  than non-leaders.   This finding suggests a clear
bias towards local  responsibility  for the soil erosion problem.  In sum-
mary, the notion of federal  involvement  in  local pollution control, espe-
cially among opinion leaders,  seems  to be an acceptable idea, but both
groups are more  favorable towards  local  control of the situation.  This
section has dealt  with government  involvement in general, while the fol-
lowing section will  focus specifically on attitudes towards the Black
Creek Project.

-------
                           -  272 -
        Table 28.  Who Should be Responsible  for Controlling  the
                         Soil Erosion Problems?
Response
Individual Landowner
Local Government
Local Soil and Water
Conservation Districts
State Government
Federal Government
Other
Don't Know

Leader
N %
7
0

2
1
0
3
0
TT
53.8
0.0

15.4
7,7
0.0
23.1
0.0
100.0
Non-Leader
N %
16
3

10
2
1
6
3
4T
39.0
7.3

24.4
4.9
2.4
14.6
7.3
100.0
     Since the Black Creek Demonstration Project began,  the  opinion lead-
ers have attended more public meetings  concerning  the  project  than non-
leaders.  Leaders have, on the average, attended over  three  times  as  many
public meetings.  As shown in Table 29, the  leaders  were also  much more
familiar with the project than their counterparts.   Sixty-one  percent of
the leaders indicated much contact and  participation in  the  Project.   The
percentage of non-leaders indicating similar  involvement was considerably
less (17.1%).  Another very distinct contrast  between  the two  groups
appeared when they were asked how much  they  had participated in the plan-
ning and implementing of the Project (Table  30).   Forty-six  percent of
the leaders indicated a great deal of participation  while only 7.3% of
the non-leaders  indicated the same.  Equally important is that 24  (58.2%)
of the non-leaders said they had not participated  at all in  planning  or
implement at ing the Project and only one (7.7%)  leader  indicated a  lack of
involvement. In  summary, it is clear that  the  informal opinion leaders in
                Table 29.  How Familiar Are You With  the
                   Black Creek Demonstration  Project?
                                    Leader             Non-Leader
     Response	N	%	N	%

  Never heard of  it
  Heard, but no contact          1         7.7        11       25.8
  Little contact                 1         7.7        11       25.8
  Contact w/ various
   Project representatives       3        23.1        12       29,3
  Much contact and
   participation                 8        61.5         7       17.1
                                TT     TWHT       ?r      100.0

-------
                            _ 273 _
the Black Creek area had an  active  role  in  the Project.   It is also
important to see how this  greater  involvement  of the leaders in the
Project may have affected  their  attitudes  towards  the Project.
    Table 30.  Participated  in  Planning  or  Implementing this Project
Response
Not at all
Very little
Some
A great deal

Leader
N %
1
2
4
6
13
7.7
15.4
30.8
46.2
100.0
Non-Leader
N %
24
3
11
3
41
58.5
7.3
26.8
7.3
100.0
     A larger percentage of both  leaders  and  non-leaders agreed that
almost everyone in the area would benefit  from the  Black Creek
Demonstration Project (Table 31).  However, when asked their overall
reaction to the program, while  almost  everyone indicated approval,
opinion leaders were more  likely  to  say  it was an excellent  program
(Table 32).  Deaaling with a more substantive  question,  a majority  of
both groups felt that pollution control  for Black Creek  is currently in
good shape (Table 33).  Therefore, both  leaders  and  non-leaders seem to
have an overall positive reaction to  the  project.
       Table 31.  Almost Everyone  in This Area Will  be  Benefitted
                     From This Demonstration  Project
Leader
Response
Agree
Disagree
Don ' t know

N
12
0
1
13"
%
92.3
0.0
7.7
100.0
Non-Leader
N
35
2
4
41
%
85.4
4.9
9.8
100.0
               Table 32.  Overall Reaction Towards Program
                                   Leader            Non-Leader
     Response                    N        %          N         %
  Excellent Program              5       38.5        4
  Good  Program                   7       53.8       33
  Not a Very Good Program
  Not a Good Program at All     —       —          l
  Undecided                      1        7.7        3
                                T3      TooTo       4T

-------
                            - 274 ~
     To gain insight into their  individual  perception of the project the
respondents were asked:  "Do you  feel  that  you  have  personally benefit ted
from the Black Creek demonstration  project?"  Responses  to this question
reveal a much greater variation between  the  leaders  and  non-leaders
(Table 34).  One hundred percent  of  the  leaders felt they had personally
       Table 33.  How Effective Do You Think  Pollution  Control  for
                           Black Creek is Now?
Leader
Response N %
Excellent -—• — -
Good 13 100.0
Fair
Poor — —
Don ' t Know — —
13 TOOVO
Won- Leader
N %
3
30
4
—
4
41
7.3
73.2
9.8
—
9.8
100.0
       Table 34.  Do You Feel That You  Personally  Have  Benefitted
               from the Black Creek  Demonstration  Project?
Response
Yes
No

Leader
N %
13 100.0
,,_ —
T3~ TooYo"
Non-Leader
N %
24
17
41
58.5
41.5
100.0
benefitted from the project  and  only  58.5%  of othe non-leaders felt they
had benefitted.  This large  discrepancy  between leaders and non-leaders
in perceived personal benefits may  be the  result of the patterns of
involvement with the project, especially the  adoption of various best
management practices.

     In summary, leaders were much  more  involved and familiar with the
Black Creek Project than non-leaders.  Nevertheless, both groups perceive
the project as generally successful.   However, there were many non-leader
who felt they had not personally benefitted from the project.

     While the attitudes and opinions of the  farmers toward pollution is
important, their actions towards actually  solving this problem are even
more crucial.  Thererfore, we compared the  difference in adoption
patterns between the leaders and non-leaders.  Table 35 indicates the
average number of land  use practices  adopted  by each group.  This average
is based on ten selectecd practices.   As can  be seen leaders have a
slightly higher rate of adoption than non-leaders.  This may be the
consequence of two  factors.  First,  the  opinion leaders tend to come from
larger  farms,  and their land may simply  be  in need of more structures and

-------
                            - 275 -
practices to adate  erosion.   However,  it  could also be a result of the
leaders' greate  awareness  of  the  pollution problems and methods to solve
them.  To help determine this we  now turn to Table 36 which illustrates
the  specific practices  adopted by each group of farmers.
     Table 35.  The Average  Number  of Land Use Practices Adopted by
                        Leaders  and Non-Leaders.*
     Response                       Leader            Non-Leader
  Mean                              3.538                2.195
*Based on ten selected  practices.
     As shown by Table 36,  a  significantly larger percentage of leaders
are currently using crop residue management.   This phenomenon is also
true for field borders and  grassed  waterways  or outlets.  These are three
practices that were heavily emphasized  in the Black Creek Project.   There
fore, the difference between  leaders  and  non-leaders is rather disappoint
ing.  The anticipated "filtering down"  effect from leaders to non-leaders
doesn't seem to have worked exceptionally well with regard to these prac-
tices.   This is not to imply  that the  leaders had no effect on the  non-
leaders' decision to adopt  these practices.   However, their influence was
not as  strong as might be hoped.

     Another important finding  revealed in Table 36 is the number of non-
leaders discontinuing to use  of certain land  management practices.   Among
the opinion leaders, this did not occur at all.  The closer involvement
of the  leaders in the project may have  influenced their decision to con-
tinue using the practices they had  adopted.   If this is the case, then
there is probably some value  in getting as many people as possible  dir-
ectly involved in the project.

-------
Table 36.  Selected Land Use Practices by Leaders  and Non-Leaders
Conservation Cropping
Contour Farming
Crop Res. Management
Field Borders
Grade Stab. Structures
Grassed Waterways
Holding Ponds or Tanks
Livestock Exclusion
Pond
Strip Cropping
LEADERS
U
N
8
1
8
9
4
8
3
2
3
-
sing
%
61.5
7.7
61.5
69.2
30.8
61.5
23.1
15.4
23.1
—
Hav
N
—
—
—
—
—
—
—
—
—
—
e Used
%
—
—

—
—
—
—
—
—
—
Neve
N
5
11
5
4
9
5
10
6
10
13
r Used
%
38.5
84.6
38.5
30.8
69.2
38.5
76.9
46.2
76.9
100.0
1.
N
-
1
-
-
-
-
-
5
-
-
I. A.
%
—
7.7
—
—
—
—
—
38.5
—
—
NON-LEADERS
E
N
26
2
11
17
6
16
2
3
7
40
sing
%
63.4
4.9
26.8
41.5
14.6
39.0
4.9
7.3
17.1
97.6
Hav
N
1
0
1
-
-
2
-
1
1
1
e Used
%
2.4
0.0
1.9
—
—
4.9
—
2.4
2.4
2.9
Neve
N
14
39
28
24
35
23
39
17
33
—
r Used
%
34.1
95.1
68.3
58.5
85.4
56.1
95.1
41.5
80.5
—
h
N
—
—
1
—
—
—
—
20
—
—
I. A.
%
—
—
2.4
—
—
—
—
48.8
—
—
                                                                                                I
                                                                                                NJ

-------
                            - 277 -
Summary

     Previous work  in  the Black  Creek  area  has  demonstrated the utility
of sociological work in the early  stages  of a water quality project
(Brooks and Taylor, 1974).  The  present  study has investigated several
dimensions of potential effects  of  such  a project on the local community,
and its residents as well as  future water quality projects.  The research
suggests that such  projects have substantial effects upon the attitudes
and behavior of local  residents  in  addition to  any direct effects upon
water quality.

     From an evaluation point  of reference, the Black Creek Project seems
to have been rather successful.  However, several areas  were pinpointed
which suggest that  with proper planning,  dissemenation of information and
implementation, the project could have been more successful.  In
designing and implementing future  projects, managers should careful1 note
the impacts the Black  Creek Project  had  on  the  local community and
residents.

-------
                            - 278 -
                             References Cited

Bouwes, Nicolaas,  Stephen Love joy,  Lowell L. Klessig, and Douglas Yangeen
   1980 "Benefits  of Lake Protection to a Small Urban Community," in pro-
        ceedings of International Lake Management Conference, P.orfLand,
        Maine,  October.

Bouwes, Nicholaas  and Stephen  8.  Lovejoy
   1980 "'Optimal' Cost  Sharing and Nonpoint Source Pollution Control"
        .Economic Is sue s,  May,  1980.

Brooks,, Ralph M. and David L,.  Taylor
   .1975 Impact  of  social  attitudes  on managing the environment.  Proceed-
        ings of the point source  pollution seminar..  Chicago, Illinois,

•Genera,! .Accounting Office
   1977 Report  to  the Congress  by the Comptroller General of the United
        States, Washington,  D.C.

Kl.essig, Lowell and Stephen Lovejoy
   19$Q "Necesgary Conditions  for Resource Allocation and Management,,'"  in
        proceedings of 45th North American .Wildlife and 'Natural Resources
        Conference, March 22-26,  1980, Miami Beach, Florida.

Lovefoy,, Stephen B., Lowell Klessig, and Nichoi-aas Bojjw.es
   1<980 "A  Forkful Each:  ,CO(St-Sharing for Manure Handling" Journal of
        Soil ;and Water Conservation,, January-February, 19$Q,

Lo-vejoy, Stephen B.., and  Nicolaas Bouwes
   197.9 "Subsidies and Agricultural Bo Hut ion 
-------
                                    _ 279 _



                      Black Creek Data Management System


            P.K.  Carter,  D.B.  Beasley,  L.B.  Huggins and  S.J. Mahler
     The  Black  Creek Data Management System  (BCDMS)  was  developed to provide
convenient, efficient,  flexible  retrieval of the data that has been gathered.
BCDMS  is  a  hierarchy  of  computer  programs,  macros,  and  data  handling
subroutines that  can manipulate  and retrieve information  that  has been col-
lected from the Black Creek Project since  late  in 1973.   Ihe BCDMS  is  user-
oriented, taking  into account the various levels of personnel that it will be
serving.  It  provides useful tools for all —  from  those  minimally familiar
with  terminal usage  to  the experienced  programmer.   BCDMS consists  of data
files, EZBASE,  and  "packaged" application programs.  The  subsequent sections
will discuss  each of these three components, individually.


                                  Data Files

     With large amounts of data being collected, it became necessary to devise
a  file structure  that would make application development  faster,  easier, and
more flexible.  To  accomplish this end,  the old  master files  were converted
into a new, event-oriented format.  Ihis conversion led to simpler, more effi-
cient access.

     Ihe data files consist of physical/chemical water  quality  (both grab and
pump samples),  stage, rainfall,  and in site quality  samples.   Each file con-
tains all the data  for one  site,  with  the  records stored  in  chronological
order.   Ihis  feature  (combining  all  the  years into  one  file)  permits one to
analyze any period of time without being constrained by the previously imposed
beginning and end of year boundaries.

     Previously written programs need not be abandoned with the development of
this new system.   Recognizing the need to provide an interface with the past,
a program was developed to convert the new files into STORET compatible files.
The availability  of  this  program is  intended  to ease transition, and allow
continued use of  existing programs whose performance would  not justify revi-
sion.  However, this  capability should  not  stifle the  system's growth  by
overemphasizing the "patching" of existing programs rather than development of
new and improved methods for processing the data.


                                    EZBASE

     EZBASE is  a  library of FORTRAN callable subroutines  that  interfaces the
application programmer with the physical data files.  It provides the program-
mer with powerful tools to retrieve and manipulate the data.

     EZBASE's use is  easy  and understandable.   It employs a  number of common
blocks that have been broken into logical  divisions.  These allow the program-
mer to use  only  those variables  essential to  his  application.   He  simply

-------
                                    - 280 -
declares the necessary  common  blocks in each of his  modules that uses EZBASE
routines.

     The user  is provided  the capability of  requesting that only  specified
data  types  (grab, pump,  etc.) be considered  in his  program.   This  is  done
merely by calling a routine with  the desired type numbers as parameters.   He
can easily find  the most recently obtained values for any data  element(s), at
any point in  time.  He  can also  find  interpolated values  for  any point  in
time.

     The usr always has access to the last  set  of  data values he  requested.
Thus, he  is able  to make comparisons, and compute differences without being
burdened with setting  up temporary storage facilities.   Another feature built
into  EZbASE permits the programmer  to  request  only  that  data which falls
within a designated time period.  By lengthening this  time period,  one  can
request all  the data.

     The features of EZBASE can be employed in a number  of different combina-
tions.  Clever use of  this library allows the application programmer  a great
deal of flexibility in data retrieval, serving a wide range of applications.

     For further  discussion and  detailed instructions  on using  EZBASE,  the
EZBASE user Document should be consulted.  This document explains how each of
the commands is used,  and the results it  produces.   Illustrative  examples are
included.
                              "Packaged" Programs

     To  provide  a facility  for  generating standard  reports, the  "packaged"
programs were developed.   Ihey can be used with minimal effort by anyone from
novice to expert.

     Some analyses  are done  repeatedly with  data  from  many different  time
periods  and/or  sites.   Bor  these  analyses,  the  "packaged" programs  prove
invaluable.  Their use is outlined below:

     1)   The user types  the name of  the macro designed  to run  the desired
          program.

     2)   he  responds to  the prompts displayed on  the  screen  (e.g.,  SITE
          NUMBER  (2/6/12/20)?, LINEAR  INTERPOLATION DESIRED  (Y/N)?, BEGINNING
          DATA — YEAR (2 Digit Integer)?, etc.).   These  inquiries permit the
          user to analyze the data in an infinite number of subsets.

The standard report created using macro BLKCR1 contains the following informa<-
tion:

     1)   Total  transport,  in  kilograms  and  kilograms  per   hectare  (of
          suspended  solids, ammonia,  nitrate, soluble organic  and sediment
          bound  nitrogen,  and inorganic,  soluble organic, and  sediment bound
          phosphorus)  passing the site during the specified time period.

-------
                                   - 281 -
     2)   Flow  weighted  concentrations,  in  milligrams per  liter  (for  the
          parameters listed above).

     3)   Blow characteristics, including: peak and mean flow rates, volume of
          runoff, total runoff, and total rainfall for the period.

     4)   Statistical  analysis of the  concentrations,  including:  maximum,
          minimum, mean, median, and standard deviation.

     This report can be varied by the user.  It can be generated for any site,
and for any  specified  period of time (from a span of days to years) .  Another
features that allows variance is  the ability for  the  user to  request  linear
interpolation throughout the  analysis.   When dealing with a short  time span,
interpolation can  give a  clearer  view of  the situation  at that  particular
time.

     Presently, analysis of the data is only available  with  tabular represen-
tations.  Packages  that graphically depict the data and  analyses are  in the
development and testing stages and will soon be available.


                                   Summary

     Ihe Black  Creek Data  Management System was organized to  support  diverse
applications with varied data requirements.  Designed with the increasing cost
of programming and decreasing  cost of  storage in mind,  BCDMS  promotes  simple
application  programming  by  hiding complexities.   It  was planned  such that
changes to it would not require revision of application programs (which can be
extremely costly).  This was accomplished by severing the application program-
mers'  view of the data  from  it's  physical representation.  All of  these fac-
tors considered, BCDMS  is  proposed to  be a solid foundation  for future appli-
cations to build on.

-------
                                  TECHNICAL REPORT DATA
                           (Please read Instructions on the reverse before completing}
 REPORT NO.
 EPA-905/9-81-003
                            2.
                                                          3. RECIPIENT'S ACCESSION-NO.
 TITLE AND SUBTITLE
 Environmental Impact of Land Use on Water Quality
 Final  Report on the Black  Creek Project  Phase II
                                                         |6. REPORT DATE
                                                         	May 1981
                                                         6. PERFORMING ORGANIZATION CODE
. AUTHORIS)
Jim B.  Morrison and James E. Lake
                                                          B. PERFORMING ORGANIZATION REPORT NO.
                                                          10. PROGRAM ELEMENT NO.
                                                              A42B2A    	
                                                          11. CONtRACt/GRANT NO!""
PERFORMING ORGANIZATION NAME AND ADDRESS
Allen County  Soil & Water
Conservation  District
Purdue University
University of Illinois
                                                              G005335
2. SPONSORING AGENCY NAME AND ADDRESS
Great Lakes National Program Office
U.S.  Environmental Protection Agency
536 South Clark Street, Room 932
Chicago, Illinois 60605
                                                          13. TYPE OF REPORT AND PERIOD COVERED
                                                           Final Report.    1977-1980
                                                          14. SPONSORING AGENCY CODE
                                                          US EPA-GLNPO
5. SUPPLEMENTARY NOTES
 ** Each individual report  has its own authors.
 This report is to provide  an update of all activites  on the Black Creek Project.
6. ABSTRACT

The report is intended  to  consolidate and update materials collected during
the eight year period,  covered by the Black Creek Project.  It concentrates
primarily on the years  between 1977 and 1980, and represents a major interim report
in the total project.   The organization of this report is a collection of  research
papers presented by project investigatdrs,   A final  report, which synthesizes all
of the work covered during these eight years and an additional two-year period,
will be published in 1983.

The following sections  are intended to summarize the  research findings and
implications as reported by the project investigators.  Details which support their
their conclusions are set  forth in the individual papers.
                               KEY WORDS AND DOCUMENT ANALYSIS
                 DESCRIPTORS
                                             b.IDENTIFIERS/OPEN ENDED TERMS
                                                                          COS AT I Field/Group
 Permeability
 Denitrification
 Percolated
 Peak concentration
 Silty clay soils
 Latty clay
 Blount silt loam soils
                      Rural runoff
                      Erosion
                      Agricultural watershec
                      Water quality
                      Groundwater pressure
18. DISTRIBUTION STATEMENT      ————
 Document ±s available to  the  public through
 the National Technical Information Service,
         eld. VA 22161
                                              19. SECURITY CLASS (ThisReport)
                                                 None
                                                                        21. NO. OF PAGES
                                                                          286
                                              20. SECURITY CLASS (This page)
                                                                         22. PRICE
EPA Form 2220-1 (9-73)
                                       282

-------