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
------- |