EPA-600/3-77-097
August 1977
Ecological Research Series
        IMPACT  OF  NEARSTREAM VEGETATION AND
                  STREAM MORPHOLOGY ON WATER
                        QUALITY AND  STREAM BIOTA
                                     Environmental Research Laboratory
                                    Office of Research and Development
                                    U.S. Environmental Protection Agency
                                           Athens, Georgia 30805

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology.  Elimination  of  traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.   Environmental  Health Effects Research
      2.   Environmental  Protection Technology
      3.   Ecological  Research
      4.   Environmental  Monitoring
      5.   Socioeconomic Environmental Studies
      6.   Scientific and Technical  Assessment Reports (STAR)
      7.   Interagency Energy-Environment Research and Development
      8.   "Special" Reports
      9.   Miscellaneous Reports

This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on  humans, plant and animal spe-
cies, and  materials. Problems are assessed for their long- and short-term influ-
ences. Investigations  include formation, transport, and pathway studies to deter-
mine the fate of pollutants and their effects. This work provides the technical basis
for setting standards to minimize undesirable changes in living organisms in the
aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                        EPA-600/3-77-097
                                        August   1977
 IMPACT OF NEARSTREAM VEGETATION AND STREAM
MORPHOLOGY ON WATER QUALITY AND STREAM BIOTA
                  by
    James R. Karr and Isaac J. Schlosser
Department of Ecology, Ethology and Evolution
           University of Illinois
         Champaign, Illinois 61820
          Contract No. 68-01-3584
               Project Officer
              George W. Bailey
Associate Director for Rural Lands Research
     Environmental Research Laboratory
           Athens, Georgia 30605
     ENVIRONMENTAL RESEARCH LABORATORY
    OFFICE OF RESEARCH AND DEVELOPMENT
  U. S. ENVIRONMENTAL PROTECTION AGENCY
           ATHENS, GEORGIA 30605

-------
                                 DISCLAIMER
     This report has been reviewed by the Environmental Research Laboratory,
EPA, Athens, Georgia, and approved for publication.  Approval does not
signify that the contents necessarily reflect the views and policies of the
Environmental Protection Agency, nor does mention of trade names or commer-
cial products constitute endorsement or recommendation for use.
                                     ii

-------
                                   FOREWORD
     As environmental controls become more costly to implement and the penal-
ties of judgment errors become more severe, environmental quality management
requires more efficient analytical tools based on greater knowledge of the
environmental phenomena to be managed.  The development of management or
engineering tools to help pollution control officials achieve water quality
goals through watershed management is an integral part of this Laboratory's
research on the occurrence, movement, transformation, impact, and control of
environmental contaminants.

     Water quality in streams receiving pollution from nonpoint sources is of
great concern to EPA.  Through the literature review presented in this report,
gaps in scientific knowledge of the ways in which nearstream vegetation and
stream morphology affect water quality and of the manner in which NPS pollu-
tion affects stream biota are identified.  Particular stress is given to the
beneficial use of greenbelts, but important questions are raised concerning
their effectiveness in controlling NPS pollution from agricultural lands.
In addition, the review suggests research to be accomplished prior to the
adoption of rational management strategies to effectively control water pollu-
tion in streams receiving NPS pollutants, which should be of particular inter-
est to persons in a wide variety of disciplines.
                                      David W. Duttweiler
                                      Director
                                      Environmental Research Laboratory
                                      Athens, Georgia
                                     111

-------
                                   PREFACE

     We have attempted a synthesis of research from several fields to evalu-
ate the effects of nearstream vegetation and channel morphology on water
quality and stream biota in agricultural watersheds.  An inevitable result of
such a synthesis is inadequate coverage of the literature from each field. We
acknowledge this deficiency and solicit comments on weaknesses in our cover-
age.  However, we are less concerned about further references supporting our
suggestions than about references which add new perspectives or disagree with
our conclusions.  Since our goals are wise use of biological, soil, and water:
resources, we hope readers will bring relevant studies to our attention.   (A
glossary is included (p. 85-90) for readers unfamiliar  with the terminology
of all fields discussed in this report.)

     From our perspective a major problem in the effort to control non-point
sediment pollution is the common philosophy that it is an agricultural pro-
blem.  This is an error on two counts.  First, not all sediment pollution
derives from agricultural, activities and second, fields other than agricul-
ture, including agricultural engineering, have expertise which should be
brought to bear on sediment pollution problems.  The two problems are only
indirectly related.  The long-term objective of improved water quality can
best be served with multi-disciplinary efforts which transcend the classical
disciplines, such as agriculture, engineering, and ecology.   We are particu-
larly concerned that more thought be given early in the planning phase of pro-
jects to the long-term ecological consequences of management alternatives.
Water quality degradation can be reversed if the past emphasis on control of
soil erosion is supplemented with a broader approach to water quality pro-
blems.

     An early draft of this paper was read by a number of scientists from
many backgrounds.  The most intriguing comments came from a biologist and
from an agriculturist.   One suggested that we attempted to generalize too
broadly; that is, since most of our examples come from midwestern situations
we should, perhaps, explicitly state that our conclusions might apply only to
that region.  Another individual commented that the general principles might
apply to some regions but they probably could not be used in the high inten-
sity agriculture of central Illinois.  These comments illustrate the problems
which result from differing perspectives.  In addition, they are related to
the problems involved in defining areas of applicability.  For example, do
our discussions pertain to highly modified agricultural drains, to natural
stream channels, to drains and channels of specific sizes, or to all of
these?  We feel they apply primarily to headwater streams in most watersheds,
but as we hope to demonstrate in this review, definitive answers to these
questions are not available.  That is an alarming statement after decades of
research effort.  Rigorous scientific investigations addressing the problems
posed in this report are long overdue.


                                     iv

-------
                                  ABSTRACT

     Like all functional parts of landscape units, streams have dynamic
equilibria in nutrient and sediment loads and biota.  As man modifies water-
sheds by removal of natural vegetation and stream channelization, disequili-
bria in both the terrestrial and aquatic environments result.  These dis-
equilibria are the major problem in controlling sediments and nutrients from
non-point sources and improving the quality of the stream biota.  Unfortu-
nately, most attempts to control non-point pollutants in agricultural water-
sheds emphasize reestablishing the terrestrial equilibrium via tillage prac-
tices such as minimum tillage, winter crop cover, and terraces.  These
efforts, which depend on erosion control as measured by the Universal Soil
Loss Equation, utilize technologies for preserving soil productivity.  This
approach must be replaced by one in which improvement in water quality and
quality of the stream biota is a primary objective.  This requires erosion
control on the general landscape plus an increased understanding of the link
between terrestrial and aquatic systems and the effect of stream morphology
on the dynamics of sediment transport and quality of the stream biota.

     In this report we review the literature dealing with (1) the possible
use of near stream vegetation to reduce the transport of sediment and nutri-
ents from the terrestrial to the aquatic environment and decrease stream
temperature fluctuations, (2) the effect of stream morphology on sediment
transport, and (3) how near stream vegetation and stream morphology affect
the biota of streams.  The results of this review suggest proper management
of near-stream vegetation and channel morphology can lead to significant
improvements in both the water and biological quality of many streams.  How-
ever, critical research outlined in this report is still necessary if we are
to properly use this management alternative to attain the objectives of the
Federal Water Pollution Control Act of 1972 (Public Law 92-500).

     This report was submitted in fulfillment of Contract No. 68-01-3584 by
the University of Illinois,  Champaign, IL, under the sponsorship of the U.S.
Environmental Protection Agency.   This report covers a period from January
1976 to July 1977, and work was completed as of July 1977.
                                      v

-------
                               CONTENTS

Foreword ........... .  .................  ±±±
Preface  .  .  ......  ,  ............ „ ..... „  .   iv
Abstract .......  .  ...........  „  .........    v
Figures  ........................  ..... viii
Tables   ......................... .   , .  .    x
Acknowledgments   ..........  . ...........   ...   xi
      1.  Introduction  ......................    1
      2.  Conclusions   ......................    3
      3.  Recommendations   ....................    7
               Specific research problems  ... .........    7
               Large scale  research   ...............   10
               What type of research  approach is best? ......   11
      4.  Vegetation As a Nutrient Filter  . .  ..........   12
      5.  Vegetation As a Sediment Filter  ....... .....   16
               Overland flow  .  . .................   17
               Channel  flow  ...................   20
      6.  Effect of Channel Morphology On Water Quality  .....   30
      7.  Determinants  of Sediment Loads in Flowing Waters ....   40
      8.  Effect of Streamside Vegetation on Water Temperature .  .   44
      9.  Impact of Nearstream Vegetation and Channel
          Morphology on Stream Biota  ...............   49
               Effects  of sediment ................   49
                  Fish  ......................   49
                  Invertebrates  .................   50
                  Stream Productivity ..............   52
                  Summary   ....................   53
               Effects  of temperature ..............   54
               Effects  on stream energetics  ...........   56
               Effects  of stream channelization  .........   57
               Conclusions  ....................   63
     10.  Recreational Benefits of Greenbelts   ..........   65
     11.  Other Advantages  of Greenbelts .......... ...   66
     12.  Discussion and Conclusions  ...............   67

References ............................   73
Glossary  .............................   85
                                  Vll

-------
                                   FIGURES
Number                                                                  Page

  1   Relationship between slope of a vegetative (sediment) filter
        and effective filter width	 .   18

  2   Effect of land use on discharge rates on two watersheds  ....   20

  3   Silt volume as percent of basin capacity for Lake McMillan
        near Carlsbad, New Mexico	   21

  4   Water velocity  through vegetation as a function of depth ....   22

  5   Relationship between depth of flow and amount of retardance
        with bermudagrass	   22

  6   Hypothesized relationship between filter width and percent
        of sediment remaining for several vegetation types 	   25

  7   Hypothesized relationship between distance solution is
        filtered and  percent of sediment remaining for several slopes .   26

  8   Relationship between water surface slope and total sediment
        discharge  	 .....   31

  9   Relationship between velocity and total.sediment discharge ...   32

 10   Relation between effective unit stream power and measured
        suspended sediment concentration 	   33

 11   Bed and water surface profiles for several flow conditions in
        the Middle Fork of the Vermillion River	   37

 12   Changing suspended solids loads in agricultural and forested
        sections of a headwater stream	   38

 13   Variation in suspended sediment load for a given unit stream
        power	   43

 14   Seasonal changes in stream temperatures in streams from
        forested and  farm watersheds	   45
                                    Vlll

-------
Number                                                                 Page

 15   Hypothetical relationship between angular canopy density
        and thermal input to a stream	   46

 16   Relationship between stream size and effectiveness of buffer
        strip in reducing thermal inputs to a stream	   46

 17   Effect of temperature on the release of phosphorus from
        sediments during a 12-week period 	   48

 18   Regression of fish species diversity on substrate diversity .  .   51

 19   Species diversity vs. substrate diversity for molluscs
        collected at 650 sampling stations in New York	   52

 20   Flow chart illustrating the basic pathways of energy flow
        in organisms	-	   54

 21   Theoretical effect of temperature change on food consumption
        and energy budget of two hypothetical fish species  	   55

 22   Percent of species lost versus temperature in warm and cold
        water streams	   57

 23   Major components of a headwater stream foodweb  	   58

 24   Factor train analysis of the effects of channelization on
        the physical environment and biota of streams 	   59

 25   Estimated total standing crop of fish in channelized and
        unchannelized sections of the Chariton River, Missouri  ...   61

 26   Estimated standing crop of catchable-size fish in channelized      61
        and unchannelized sections of the Chariton River,  Missouri.  .

 27   Regressions between sinuosity  index and several physical and
        biological properties of streams  	   62

 28   Eegression of fish species diversity on habitat diversity  ...   63

 29   Range on three environmental gradients under which
        hypothesized generalist (a)  and specialist (b) could exist.  .   64

 30   General model of the primary factors governing the quality and
        quantity of outflow water from an ecosystem	   68
                                      IX

-------
                                    TABLES
Number                                                                     Page

  1   Percent of nitrogen and phosphorus attached to sediment in
        surface runoff ............ 	 ...  15

  2   Effect of a bluegrass sod strip on sediment concentration
        in runoff	  19

  3   Percent removal of sediment after varying lengths of
        filtration through bermudagrass  ......... 	  24

  4   Percent of sand, silt, and clay in sediment deposited after
        filtration through coastal Bermudagrass  	  25

  5   Land uses immediately adjacent to modified streams and drainage
        ditches in the Black Creek watershed .... 	 .....  28

  6   Roughness coefficients (n) for various stream conditions • 	  35

  7   Unit stream power for three flow conditions  	  36

  8   Relationship between temperature and maximum dissolved oxygen
        concentration in water	47

  9   Effects of particle size of substrate on egg survival in
        sockeye salmon 	  51

 10   Energy budgets for Red Cedar River before and during periods
        of heavy siltation,  summer, 1961  	  53

 11   Effect of varying "management practices" on equilibria of
        equivalent watersheds  	  .  70

 12   Costs and benefits of more effective management of nearstream
        vegetation and channel morphology  	  72

-------
                               ACKNOWLEDGMENTS

     Any attempt to integrate information from a variety of fields depends on
cooperation and help from many individuals.  Many persons and agencies (too
many to list individually) have responded to our inquiries and provided us
with valuable reference material.

     The Institute for Environmental Studies and Water Resources Center at
the University of Illinois at Urbana-Champaign have provided support in a
number of ways.  The following persons have read and commented on earlier
drafts of this report:  W. D. Seitz, Institute for Environmental Studies;
G. C. Sanderson, R. W. Larimore, P. W. Smith, Illinois Natural History Survey;
R. W. Bachman, Iowa State University; R. R. Schneider, University of
Wisconsin; N. G. Benson, National Stream Alteration Team; D. R. Dudley,
Allen County, Indiana, Soil and Water Conservation District; H. Kuder, Soil
Conservation Service; D. Toetz, USEPA; and 0. Gorman, Purdue University.
Errors remaining in the present draft are our responsibility.  Stephanie
Heard, Barbara Jauhola, and Margaret LeGrande have provided valuable steno-
graphic services, and Marcia Clark assisted in finding many obscure refer-
ences.  Preparation of this report was initiated as a subproject under con-
tract No. 68-01-3584 to Dr. W. D. Seitz, Associate Director, Institute for
Environmental Studies at the University of Illinois.
                                    XI

-------
                                  SECTION  1

                                INTRODUCTION


     Sediments are largely non-point sources of pollution.  Because of this
diffuse nature, erosion and sediment control methods are inherently more com-
plex than those used to control point sources of pollution and will probably
involve several control strategies acting  synergistically.  The Environmental
Protection Agency and the United States Department of Agriculture have sug-
gested several feasible methods of control which should be investigated
(Agricultural Research Service 1975, 1976, Becker and Mills 1972).  These
methods can generally be divided into two broad classes: (1) those which
attempt to prevent erosion from occurring, i.e. minimum tillage, winter crop
cover, terraces and other soil conservation practices and (2) those which
attempt to prevent the sediment from entering streams once erosion has
occurred.

     Early conservation efforts emphasized soil erosion prevention (1 above)
with little emphasis on maintenance of water quality.  Incorporation of water
quality as another objective of conservation practices represents a major
shift in conservation policy.  To meet the water quality standards set for
1983 by the federal government an even broader perspective will have to be
established, including more detailed understanding of (1) the link between
terrestrial ecosystems and water quality and (2) the dynamics of sediment and
nutrient transport within streams.  This increasing breadth of perspective
requires a more demanding synthesis of several bodies of knowledge, including
hydraulic engineering, geology, the study of stream morphology and dynamics,
and terrestrial and aquatic ecology.

     We do not suggest that past management efforts have been in vain. Rather
our main point is that continued efforts to stabilize the terrestrial environ-
ment (minimum tillage, rotational practices, etc.) must be supplemented
with improved management of in and near channel areas before significant
improvements in water quality and quality of the stream biota will occur.
Furthermore, there is a serious deficiency in our knowledge of the dynamics
of in and near channel characteristics.  Hopefully, this report will stimu-
late research addressed to the solution of land-water resource problems.

     In this report we evaluate (1) existing data regarding the ability of
vegetation to act as a nutrient and sediment filter, (2) the effects of
stream morphology on water quality, (3) the effects of streamside vegetation
on water temperature, and (4) the potential impact of greenbelts  (a combina-
tion of nearstream vegetation and stream morphology) on stream biota and
their possible uses for other recreational activities.  We then use this
information to judge the feasibility of using greenbelts for  improving

-------
water quality and quality of the stream biota.  Finally, we propose several
major questions which must be addressed before the social and economic bene-
fits of greenbelts can be fully evaluated.

-------
                                  SECTION 2

                                 CONCLUSIONS
1.  Vegetation as a Nutrient Filter.

    a.  Past efforts to use vegetation to filter nutrients in solution from
        surface runoff have yielded conflicting results.  The reasons for
        this variability are largely unknown due to poorly controlled experi-
        ments.  Important variables include type of vegetation, detention
        time, volume of water treated, and soil type.

    b.  Conflicting data exists regarding the importance of subsurface in-
        puts of phosphorus into streams.  The magnitude of subsurface inputs
        appears to be primarily related to the quality of the tile line and
        the presence of surface intakes.

    c.  Nearly all of the phosphorus (greater than 85%) and most of the nitro-
        gen (greater than 70%) in surface runoff is attached to sediment.

    d.  Nutrients are usually absorbed to the smaller-sized particles,
        especially the clay fraction.  Hence, removal of nutrients from sur-
        face runoff by vegetation will be achieved primarily by removing the
        sediments to which the nutrients are attached.

2.  Vegetation as a.Sediment Filter.

    a.  Recent data from Black Creek watershed in Indiana indicates the major
        fraction of sediment loss in an agricultural watershed is via surface
        flow.  Therefore, sediment inputs into streams can be controlled by
        adequate treatment of surface runoff.

    b.  Data from field and laboratory studies in forestry and agriculture
        indicate vegetation can effectively filter sediment from both sheet
        and shallow channel flow.  Sediment reduction is less likely during
        channel flow exceeding the height of herbaceous vegetation.  Several
        variables appear to act in an interrelated manner to determine the
        effectiveness of vegetation filters.  These include filter length,
        slope of the filter, vegetation characteristics, size distribution
        of the incoming sediments, degree of submergence of the filter,
        application rate of the water to be filtered, and initial concentra-
        tion of the sediment.

    c.  The placement of a vegetative filter and the width or length of  the

-------
        filter required to remove a given fraction of the incoming sediment
        and the duration of its effectiveness is dependent on the inter-
        action of physical factors, biological factors, and the specifica-
        tions of water quality, all of which are yet to be thoroughly evalu-
        ated under normal agricultural conditions and quantitatively related
        to each other in any predictive manner.

3.  Effect of Channel Morphology on Water Quality.

    a.  The ability of a stream to transport sediment appears to be directly
        related to its Unit Stream Power (USP) —the rate of energy expendi-
        ture by a stream as it flows from a higher to a lower point.  Math-
        ematically USP is a function of water velocity and surface slope of
        the water.

    b.  Unit stream power is useful in predicting sediment loads only when
        available material equals or exceeds the amount that can be carried
        in a channel.

    c.  A stream naturally decreases its USP and sediment transporting capa-
        bility by making lateral (meandering) and vertical (pools and
        riffles) adjustments.

    d.  USP is reduced by pools and riffles only at low and intermediate
        flows.  The effect of meanders on USP in conjunction with pools and
        riffles at various flow rates does not appear to have been quantified.

    e.  Channelization destroys nearly all characteristics of "natural"
        stream morphology, i.e. pools and riffles and meanders.   As a result,
        the USP of these streams is increased and their ability to transport
        sediment and erode stream banks is enhanced.

    f.  Allowing streams to maintain their natural morphology or reinitiate
        meandering may be a feasible management  alternative for improving
        water quality, especially during periods of low and intermediate
        flows.

4.   Effect of Nearstream Vegetation on Water Temperature and Water Quality.

    a.  Removal of vegetation from along headwater streams in agricultural
        watersheds can result in water temperature increases of 6-9°C.

    b.  As water temperature increases, its capacity to hold oxygen decreases,
        which exaggerates the impact of each additional unit of organic waste
        added to the system.

    c.   Laboratory studies have demonstrated an  exponential increase in the
        rate  at which nutrients attached to sediment  are converted to readily
        available  forms  as temperature increases.   Slight increases in tem-
        perature above 15°C produced substantial increases in the amount of
        phosphorus released.

-------
    d.   Removal of nearstream vegetation along streams in agricultural water-
        sheds results in blooms of nuisance algae and periphyton due to ele-
        vated temperatures, rate of nutrient release from sediment, and light
        availability.

    e.   An examination of temperatures in various streams with different
        types of vegetation indicates that angular canopy density (a measure
        of the shading capability of the vegetation) is the only character-
        istic of the vegetation correlated with temperature control.  Width
        of the vegetation strip is not important.

    f.   Small streams in agricultural watersheds have the greatest temperature
        problem but they are also the easiest to control with nearstream
        vegetation, due to the inverse relationship between temperature
        change and stream discharge for a given input of thermal radiation.

5.  Effect of Nearstream Vegetation and Stream Morphology on Stream Biota.

    a.   Sediments reduce the structural complexity and productivity of
        aquatic plant, invertebrate and vertebrate communities.   Their detri-
        mental effect results when they settle out of suspension, cover the
        periphyton and essential spawning grounds of fish and decrease
        bottom (substrate) diversity.

    b.   When vegetation is removed along streambanks and water temperature
      '  increases from 6-9°C, it may become energetically impossible for
        species with lower temperature optimums to continue living in an area,
        regardless of changes in sediment load, habitat structure or other
        environmental conditions.  A shift in community structure may occur
        with resident species being replaced by less desirable species which
        are more tolerant of the increased temperatures.

    c.   Headwater streams (stream orders 1, 2, and 3) are dependent on near-
        stream vegetation as a major source of energy and are extremely
        important as spawning and nursery grounds for many commercial and
        sport species which spend their adult lives in rivers or large lakes.
        Removal of nearstream vegetation in these areas results in signifi-
        cant reductions in invertebrate and fish production because of the
        loss of allochthonous (terrestrial) energy inputs.

    d.   Channelization results in a significant reduction in invertebrate
        and vertebrate production due to the destruction of habitat complex-
        ity.   Recent research relating stream morphometry to habitat charac-
        teristics of aquatic organisms indicates that maintaining stream
        sinuosity and a diversity of depths, water velocities, and substrate
        types will lead to substantial improvements in the quality of stream
        biota in agricultural watersheds.

    e.   Because organisms are simultaneously adapted to a number of different
        environmental gradients, any attempt to improve the quality of the
        stream biota must utilize a broad-based, multi-purpose approach.
        Future stream management practices must optimize for a number of

-------
         environmental  variables,  including flow characteristics,  water
         temperature, oxygen concentration,  habitat  diversity,  food  avail-
         ability and water  quality,  all  of  which interact  to  determine  the
         quality of the stream biota.  Data in  the literature suggest a
         more realistic management of  nearstream vegetation and stream
         morphology can provide the  basis for such an  interactive  approach
         and lead to a  more productive,  diverse,  and stable biotic community.

6.   Feasibility of Greenbelts.

     a.   The conclusions listed above  suggest that proper  management of  near-
         stream vegetation  and stream  morphology may produce  substantial
         improvements in water quality and  the  stream  biota of  agricultural
         watersheds.  Vegetation may reduce sediment and attached  nutrient
         inputs to streams  and temperature  fluctuations.   A more natural
         stream morphology  can reduce  sediment  loads and provide suitable
         habitat for both fish and invertebrates.

         Therefore, before  major improvements in water quality  and stream
         biota can be realized,  the  present  emphasis on erosion control  on
         the land surface must be  combined with the  management  of  nearstream
         vegetation and stream morphology.   This  approach  will  require a sub-
         stantial new research effort  on a  number of questions  outlined  in
         the recommendations section of  this report.   These problems must  be
         dealt with at  individual  drainage  and  watershed levels before the
         environmental  and  economic  feasibility of this management strategy
         can be fully evaluated.

-------
                                  SECTION 3

                               RECOMMENDATIONS
     We have divided our recommendations for future studies into two broad
categories:  (1) research aimed at providing answers to specific questions
dealing with the possible use of nearstream vegetation and/or stream morphol-
ogy for improving water quality and the stream biota and (2) large-scale
research projects to be initiated at the single drainage or watershed level
to clarify the economic costs and benefits of the greenbelt management alter-
native.  We also briefly comment on the benefits and costs of these two
research approaches with respect to solving water quality problems in the
non-point source area.

SPECIFIC RESEARCH PROBLEMS

1.  Vegetation as a Nutrient Filter

    a.  Substantially more research needs to be done elucidating variables
        important in determining the levels of nutrients in subsurface flows.
        Emphasis should be placed on the effects of soil type, slope, drain-
        age characteristics, and type and quality of tile line.

    b.  Four factors should be more rigidly controlled in attempts to deter-
        mine the efficiency of vegetation for filtering nutrients from
        surface runoff:
        (1)  Detention time:  should be accurately quantified so that
             filtering success can be more precisely related to this
             variable.

        (2)  Vegetation effects:  controls should be maintained to clarify
             the relationship between type of vegetation and nutrient removal
             with specific emphasis on the effects of (a) structure and phys-
             iological capabilities of the plants and (b) single plant
             species versus more complex plant communities.

        (3)  Volume of water treated:  the relationship of volume of water
             treated to efficiency of nutrient removal should be studied in
             more detail.

        (4)  Soil type:  soil type affects filter efficiency, with sandy loam
             soils being more efficient than clay loam soils.  More attention
             should be directed to this variable, especially with respect to
             its effect on shallow percolation into the soil versus rapid
             overland flow of the solution.

-------
2.  Vegetation as a Sediment Filter

    Before nearstream vegetation can be  efficiently  utilized  as  a land manage-
ment alternative for improving water quality,  the  following problems  must be
addressed:

    a.   The importance of runoff over streambanks  as a  sediment  source ver-
        sus runoff entering streams from intermittent tributaries and water-
        ways needs to be established.  How does  the  contribution from the two
        sources change with stream  size  (stream  order)?

    b.   The relationship between efficiency of sediment  filtration and volume
        of runoff,  initial sediment concentration, size  distribution  of the
        incoming sediment,  filtration length, slope  of  the filter,  character-
        istics of the vegetation, and degree of  filter  submergence should be
        established for runoff  conditions encountered in typical agricultural
        watersheds.

    c.   The quantitative relationship  between the  retardance  of  flow,  vegeta-
        tion type,  and filtering  efficiency should be established and its
        implications  for an  economic   cost and benefit model  be  evaluated.

    d.   For different  vegetation  types the shape of  the  curve relating dis-
        tance  the water has  passed  through the filter to percent of sediment
        remaining  should  be  established  and an evaluation of  why vegetation
        types  differ  should  be  performed.

    e.   The  duration  of  effectiveness  of vegetative  filters should be
        established along with  how  arrangement of  the strip(s) of vegetation
        affect  its efficiency and/or  duration of effectiveness.

    f.  Harvestable crops should be evaluated to determine their value as a
       vegetative filter and their potential as an  economic  incentive for
       filter use.

   g.  The effect of  surface litter  on  filter efficiency and the longevity
       of effectiveness of  the filter should be evaluated.

   h.  The effect tillage practices have on filter  efficiency by reducing
       peak rates of runoff should be established.

   i.  The effect of filters along streambanks in reducing erosion from
       flood waters should be examined along with their  effect on sediment
       deposition by flood waters.

   j.   The effect of nearstream vegetation on drainage rates   in agricul-
       tural watersheds should be systematically established  and evaluated
       w.-th respect to the economic costs and benefits  of this  decreased
       drainage rate.

   k.   The amount of land to be taken out of production to minimize nutrient
       and sediment loads must be determined.

-------
3.   Effect of Channel Morphology on Water Quality

    a.  The relationship between stream morphology, Unit Stream Power  (USP),
        and suspended sediment concentration should be rigorously evaluated
        in agricultural watersheds to establish the effect of present stream
        management practices on USP, suspended sediment load and total sedi-
        ment load leaving a watershed.

    b.  A systematic evaluation of the effect of stream morphology on
        drainage rates in headwater streams should be performed to establish
        what options are available, depending on the balance desired between
        the goals of agriculture (rapid drainage) and those of water quality
        (slower drainage, lower USP's, reduced suspended sediment loads,  and
        reduced downstream flood and silt damage).

4.   Predicting Sediment Loads in Flowing Waters

    a.  A realistic model for predicting sediment losses from agricultural
        watersheds is critically needed.  Among other things, the model
        must consider the magnitude of erosion from the watershed, how and
        where eroded material is removed before the runoff reaches waterways,
        the addition of sediment from underground (drain) tile systems and
        the energy the stream has available to transport the sediment
        entering it.  Future research should emphasize combining the Univer-
        sal Soil Loss Equation, information concerning the nature of the
        erosion-deposition equilibrium between the terrestrial and aquatic
        environment, and the Unit Stream Power model.

5.  Stream Biota

    a.  Systematic experiments should be performed on headwater streams,
        evaluating the effect of removal of nearstream vegetation on the
        distribution and abundance of organisms in these areas.

    b.  A multivariate analysis should be performed relating invertebrate and
        vertebrate abundance and diversity to several relevant environmental
        variables.  This would provide significant insights into what com-
        bination of water quality and stream morphometric variables are of
        prime importance in determining the quality and quantity of stream
        biota.  This data could then be utilized to make significant improve-
        ments in future engineering designs for alteration of the nearstream
        vegetation and channel morphology.

    c.  A rigorous economic re-evaluation of all aspects of channelization
        should be performed.

6.  Floodplain Ecology

        Ecosystems which have evolved in the regular flooding regimes of
        small stream or river floodplains require periodic flooding for
        normal growth and reproduction.  Modifications of stream  channels
        by channelization and by construction of dams modifies  the

-------
        periodicities of flood and result in shifts in species composition of
        floodplain floras.   Such problems may preclude management for natural
        vegetation along streams and the benefits deriving from their pre-
        sence.   Any efforts to assess the value of greenbelts should consider
        both "artificial" and "natural" assemblages of species.  Intuitively,
        it seems likely that more natural assemblages will optimize a wider
        range of societal objectives.

LARGE SCALE RESEARCH

    Before the use of nearstream vegetation and channel morphology can be
fully evaluated as a management alternative, its costs and benefits must be
assessed at the level of a single drainage and at the watershed level.  The
design of this research must explicitly recognize the functional interrela-
tionships of the terrestrial and aquatic components of watersheds. For
example, research examining the effects of tillage practices alone compared
with other research on the effects of greenbelts and/or more natural channel
morphology is inadequate.  Significant benefits may derive from combinations
of several practices that do not accrue from either practice used alone.

    1.  Changes in nearstream vegetation could be instituted in a watershed
        by preventing plowing along one or several of the drainages and
        letting vegetation recolonize the area or by planting some selected
        species.  The effects of nearstream vegetation on water quality, rate
        of stream discharge, and total sediment discharge could be evaluated.
        Unless this type of land treatment could be performed on a number of
        different drainages with different stream morphologies, substantial
        insights' into the interaction between nearstream vegetation and
        s.tream morphology will not be forthcoming.   However, it could pro-
        vide a sound data base for the economic evaluation of the costs and
        water quality benefits from alternate management strategies of the
        nearstream vegetation.

    2.  To evaluate the combined effect of nearstream vegetation and channel
        morphology on water quality and quality of the stream biota, a
        number of streams differing in distance to nearest cropland and in
        the length of time since removal of nearstream vegetation and chan-
        nelization should be intensively studied.  Depending on the length of
        time since alteration,  the streams would differ in type of vegeta-
        tion along their streambanks and the degree to which new meanders
        and pool-riffle sequences have developed.  An intensive examination
        of water quality, sediment transport, and the structure of biotic
        communities in these streams would provide a substantial data base
        and significant insights into how the terrestrial-aquatic interface
        and stream morphology can be more efficiently managed to maximize the
        quality of the water and biota in agricultural watersheds and mini-
        mize the costs.
                                     10

-------
WHAT TYPE OF RESEARCH IS BEST?

    In this section we have outlined a number of both specific and broad
research problems which could be undertaken to more fully evaluate the feasi-
bility of the greenbelt management alternative.  Although it would be best
to extensively investigate each of these problems,  time and resource limita-
tions demand that those problems with the highest benefit-cost ratio be
investigated first.  As we have documented in this report, the quality of
water leaving a watershed and the quality of the stream biota are based upon
complex interactions on the land surface, in the stream, and between the
land and stream.  The classical reductionist approach of controlling as many
variables as possible and examining the effect of one or a series of treat-
ments will be difficult (if not impossible) and expensive to do at the water-
shed level and will probably have limited success in identifying feasible,
comprehensive management strategies for controlling non-point pollution; the
interactions between the variables are too important and the variables are
too numerous.  Rather, we emphasize the need for a more holistic approach
towards the solution of non-point pollution problems in which emphasis is
placed on identifying groups of interacting variables which, if properly
managed, can lead to significant improvements in a broad range of water
quality problems.  We believe the nearstream vegetation and stream morphology
are two such variables.  Future research should document the magnitude of the
benefits and costs produced through the interaction of these variables along
with their feasibility of implementation.
                                      11

-------
                                  SECTION 4

                      VEGETATION AS A NUTRIENT FILTER


      Previous  studies investigating the  use of vegetation to remove nutrients
 from solution  are  largely  associated with  the renovation of secondary  efflu-
 ent  by spray irrigation  techniques.  These methods rely on rapid  infiltration
 of the solution with both  the vegetation and soil acting as a filter medium.
 Nutrients  are  removed through uptake by  plant roots and microorganisms and
 by chemical adsorption of  the ions onto  the surface of soil particles.
 Studies of the use of combined soil and  vegetative filters suggests consider-
 able effectiveness at phosphorus removal (98-99%), with lower efficiencies
 for  removal of nitrogen  (10-60%) and other nutrients (Lehman and  Wilson 1971,
 Kardos et  al.  1974, Sopper and Kardos 1973, Younge et al. 1976, Inst, Water
 Research,  Michigan State University 1976).  The effectiveness of  vegetative-
 soil filters appears to be related to soil and vegetation type and quantity
 of effluent applied to the system.  Unfortunately, long term effects on soil
 nutrient loads are still not clear.

      We will not consider  "infiltration  filters" in further detail in  this
 report,  since  we are primarily interested  in the feasability of using vege-
 tation to  remove nutrients from shallow  overland flow and not percolation of
 solution into  the  soil.  (We were not able to find experimental studies on
 the  use of vegetation to remove nutrients  from channel flow.)  An excellent
 example of the use of grass filters to remove nutrients from overland flow
 was  conducted  by the Campbell Soup Company in Paris, Texas (Mather 1969,
 Law  et  al. 1970).  Wastewater was sprayed  on the top of grass plots 60 to
 90 m long with a slope of  one to 12 percent.  Because of the impermeability
 of the  soil, an average of 61% of the water applied left the area as surface
 runoff.  Chemical  analyses of the runoff showed that BOD and nitrogen removal
 varied  from 85-99% and 76-90%, respectively.  However, the efficiency of
 phosphorus removal was dependent on application frequency, with the percent
 removed increasing from 50 to 80% by a change in the spray schedule from
 once  a  day to  three times  a week.  Detention time was a key factor control-
 ling  both nutrient and BOD removal but quantitative information relating
 detention  time to  percent  reduction was  not given.

      Other studies using grass filters to remove nutrients from municipal
 effluents were not as successful.  In one study, effluent from an oxidation
 pond was applied to grass plots 300 m long (Wilson and Lehman 1966 in Butler
 et al.  1974).   Apparently only one trial was performed and nitrogen and
 phosphorus concentrations were reduced 4 and 6%, respectively.   Information
 regarding application rate and slope were not available and detention times
were not given.


                                      12

-------
     A similar study (Butler et al. 1974) applied secondary effluent to reed
canarygrass plots 150 feet long with a 6% slope.  Reductions in phosphate
and nitrate concentrations were only 0-20%.  Insufficient contact time, due
to high flow rates and a short flow distance, was suggested as the reason
for the inefficiency of the system.

     Popkin (1973) performed a series of tests using a grass covered soil
filter for renovating urban runoff in Tucson.  The filter was 60 m long,
1.2 m wide and 1.5 m deep.  The volume of water treated per acre of grass
per day varied from 3.8 to 25.1 acre feet.  For the grass filter, the fol-
lowing percent reductions compared to untreated runoff occurred:  COD-62%,
suspended solids-35%, turbidity-97%, total coliforms-84%, and fecal
coliforms-87%.  The efficiency of the filter for removing nutrients was
not reported.

     The treatment of feedlot wastes by grass filters is a rapidly expanding
area.  The systems used usually consist of a lagoon for settling of solid
wastes followed by a grass waterway through which the lagoon effluent is
allowed to filter.  Much of the research in this area has just been initi-
ated and extensive sets of data are not yet available (D. H. Vanderholm,
Associate Professor of Agricultural Engineering, University of Illinois,
personal communication, 1976).  The grass waterways used are usually long,
i.e. 260-800 meters (Sievers et al. 1975, Swanson et al. 1975) but even
extremely long waterways, up to 3600 meters, have not been successful at
reducing nutrient concentrations (Kreis et al. 1972).  Furthermore, when
systems are successful (Edwards et al. 1971, Swanson et al.  1975) it is
difficult to determine if the reduction in nutrients is due to some action
of the vegetation or is merely a result of dilution by water entering the
waterway from the cropland area.

     From this brief literature review it is clear that the use of vegetation
for removing nutrients in solution from surface runoff is not always success-
ful.  The reasons for success, or lack of success are not always clear.  We
suggest four factors should be more rigidly controlled and evaluated in
future research:

    1.  Detention time:  it should be accurately quantified so that filtering
        success can be related to detention time and slope of the filter.

    2.  Vegetation effects:  when possible, controls should be maintained to
        more fully elucidate the relationship between type of vegetation
        and nutrient removal with special emphasis on the effects of (a)
        structure and physiological capabilities of the plants and (b)
        single plant species versus more complex plant communities.

    3.  Volume of water treated:  the relationship of volume of water
        treated to efficiency of nutrient removal should be studied in more
        detail.
                                     13

-------
     4.   Soil type:   soil  type  affects  filter  efficiency,  with sandy loam
         soils being more  efficient  than  clay  loam soils  (Mather,  1969).   More
         attention should  be  directed to  this  variable, especially with res-
         pect to  its effect on  shallow  percolation into the  soil versus rapid
         overland flow  of  the solution.

      Even if these  variables are  properly  evaluated,  certain problems must be
 clarified before final recommendations for use  of vegetation to control
 nutrient inputs  from agricultural runoff can  be made.  First,  a common
 assumption is that  nutrients like phosphorus  are  entering streams via sur-
 face runoff rather  than subsurface  or  groundwater runoff.   This is probably
 true under most  circumstances.  Some studies  (Jones et al.  1975,  Hanway and
 Laflen  1974)  have found very small  quantities of  phosphorus in tile flows
 and they have been  much less than concentrations  in surface flows.   However,
 when tile lines  have surface intakes,  including along road  ditches,  they may
 carry significant amounts of nutrients and sediments  (R.  Bachman,  Iowa State
 University,  pers.  comm.).

      Other recent reports (Ryden  et al.  1973, Sommers et al.  1975a,b)  also
 suggest substantial inputs of  phosphorus from subsurface  flows.   In some
 cases concentrations of N and  P were comparable to or greater  than those in
 surface runoff in the  Black  Creek watershed (Sommers  et al.  1975a).   Based
 on  average concentrations, tile effluents  and surface runoff were equally
 important sources of nutrients entering  surface waters.   Variability in the
 importance of subsurface  inputs among  watersheds  is due to  differences in
 soil, slope,  drainage  characteristics, and quality of tile  lines.   Observa-
 tions at  Black Creek suggest that very efficient  surface  filters,  if acting
 alone,  may have  limited potential for  controlling eutrophication  when
 inefficient  tile systems  are present.  This is  especially true in streams
 where small  concentrations of nutrients  from  subsurface flows  could  have
 substantial  effects  when  moderate flow velocities are involved (Phaup and
 Gannon  1961).  The  conflicting results outlined above suggest  the  need for
 more  research on  the variables important in determining the levels  of nutri-
 ents  in subsurface  flows.

      Second,  the distribution of  nutrients between the solid and  aqueous
 phases of  runoff must be  examined.  Again  data  from Black Creek are  rela-
 vant  to this  question.   Sommers et  al.  (1975a)  and Nelson et al.  (1976)
 found that a  large proportion of  nitrogen  and phosphorus  in surface  runoff
 is attached to sediment (Table 1).  With fertilizer additions, a  slightly
 larger proportion of the nutrients  are in  the solution phase of the  runoff
but even then nearly all of  the phosphorus  (greater than  85%)  and most of
 the nitrogen  (greater than 70%) is  attached to  sediment.  (Although  nutrients
adsorbed to sediment particles are  not directly available to plants,  removal
of nutrients  in solution by  plants will  result  in movement  of  some of the
adsorbed nutrients from sediment  particles  to solution.   In  solution  they
are directly available  to biological systems.)

     Soil  types differed only slightly,  with  clay loams typically having  ,
a smaller  fraction of the nutrients in solution.   Nutrients are usually
adsorbed to the smaller sized particles, especially clay  (Sommers  et al.
                                     14

-------
1975a,b, Scarseth and Chandler, 1938, Monke et al. 1975b) and this is also
the soil fraction that is selectively lost during the erosion process
(Stotlenberg and White 1935, J. V. Mannering, personal communication,
Professor of Agronomy, Purdue University, 1976).


TABLE I. PERCENT OF NITROGEN (N) AND PHOSPHORUS  (P) ATTACHED TO SEDIMENT IN
         SURFACE RUNOFF FROM FOUR SOIL TYPES UNDER FERTILIZED AND UNFERTIL-
	IZED CONDITIONS  (Adapted from Sommers et al. 1975a)	

Soil Treatment              % N as Sediment N           % P as Sediment P

Fertilized
Unfertilized
Average
75
88
Range
49-91
69-98
Average
92
96
Range
85-98
91-100
     These observations  suggest  four main conclusions regarding the use of
vegetation as a nutrient filter:

    1.  There has been variable  success  in the use of vegetation to filter
        nutrients in  solution  from  surface runoff.  The reasons for this
        variability are  largely  unknown  due  to poorly controlled experiments.

    2.  Even at high  efficiencies of nutrient removal, in some agricultural
        areas surface filters  acting alone may not control eutrophication
        because of the inputs  of nutrients into streams via subsurface path-
        ways.  This does not imply  that  surface filters should not be used.
        Rather, we emphasize that no single  factor is likely  to solve the
        problem.
                  *
    3.  Conflicting data exists  regarding the importance of subsurface inputs
        of phosphorus into  streams.  More research is needed  to clarify the
        effects of soil  type,  slope, drainage characteristics and type and
        quality of tile  lines  on this variable.

    4.  Substantial amounts of nutrients can be removed from  surface runoff
        by removing sediments, especially the clay fraction, to which the
        nutrients are attached.
                                     15

-------
                                  SECTION 5

                       VEGETATION AS A SEDIMENT FILTER
     Three potential sources of sediments exist:   (1) streambank erosion,
 (2) subsurface inputs  (i.e. tile), and  (3) surface inputs.  The value of
vegetation as a sediment filter depends on the relative amounts of sediment
from these three sources.  Data from Black Creek suggest that streambank
erosion is a minor source of sediment inputs  (Wheaton 1975, Mildner, 1976)
especially if treatment is applied to the few sites where most bank erosion
occurs.  Studies of surface versus subsurface inputs at Black Creek yielded
conflicting results.   Some tile effluents contained low sediment concentra-
tions while concentrations in others were large enough to suggest tiles may
contribute a significant amount to the  sediment yield from watersheds
(Monke et al. 1975a).  There are probably two reasons for the variability of
the data obtained.  The samples were random "grab samples."  As a result
they varied depending  on when they were taken relative to the last storm
event.  Furthermore,, the concentration  of sediment in tile outflows is signi-
ficantly influenced by the quality of the tile line (Karr, personal observ.).
This latter point, along with more recent data from Black Creek indicating
that surface runoff accounts for 50% of the water and 99% of the suspended
solids loss in the watershed  (Nelson et al. 1976), suggests maintenance of
tile lines will insure that surface runoff will be the main source of sedi-
ment inputs into agricultural drainages.  Sediment inputs into streams can
therefore be controlled by adequate treatment of surface runoff.

     Wischmeier (1962) examined the effect of storm intensity on erosion and
demonstrated that about three-fourths of the total soil loss from agricul-
tural land occurs during an average of  four storms each year.  A more recent
study suggests caution in such conclusions (Piest 1963).  He found that the
sediment contributions of large storms  (with a return period greater than
2 yrs.) varied from 3-45% of total suspended-sediment yield; the yield from
moderate storms (1-2 yr. return period) ranged from 3-22% of total yields
and storms with a return period of less than 1 yr. carried 34-92% of total
suspended sediments.   Since small storms carried more than one-half of the
soil losses in most watersheds, small scale conservation practices with low
cost can result in significant reductions in downstream sedimentation
(Piest 1963).  Piest also concluded that the soil losses resulting from
large storms were higher in the semi-arid Great Plains than in the humid
Southeast.   Furthermore, relative sediment contributions for a given large
storm decrease with increasing watershed size.

     Therefore,  in most watersheds, filters can contribute significantly to
the reduction of sediments if they are efficient during low storm intensi-
                                     16

-------
ties.  If they can also be demonstrated  to be efficient at high storm
intensities an even greater contribution  to  the reduction of sediments will
re -;ult.

      This section of the report examines  (1) the ability of vegetation to
filter sediment from water which spreads  as  a shallow layer more or less uni-
formly over the surface of the land  (overland or sheet flow) and (2) the
ability of vegetation  to filter sediment  from shallow channel flow entering
streams as intermittently flowing  tributaries or waterways (channelized flow).

OVERLAND FLOW

     Unfortunately only a minimal  amount  of  data has been found regarding
this  topic.  Much of the work in this area is descriptive, relating to the
effectiveness of filter strips between logging roads and mountain streams.
In Idaho, Haupt and Kidd (1965) observed  that sediment flows off logging
roads during major storm events reached  streams if the undisturbed filter
strip was only 2.5m wide but did not if  they were 9 m wide.  In Montana,  silt
flows off skid and logging roads fanned  out  and deposited sediment soon
(6-9 m) after reaching an undisturbed mat of pine needles and other forest
litter (Johnson 1953).

      The relationship  between the  slope  of a vegetative filter and minimum
effective greenbelt width for treatment  of surface runoff is direct (Fig.l;
Trimble and Sartz 1957).  Generally minimum width of filter strips increases
with  slope of land and varies with water  quality objectives.  Where mainte-
nance of the highest possible water  quality  is important, such as in munici-
pal watersheds, recommended filter widths are higher (Trimble and Sartz
1957).  However, the scatter of their original data suggests that other
variables are also determining the efficiency of the filter.  Apparently,
increased flow velocity results from increased slope and enhances the ability
of the surface flow to transport sediment.  Although some recommendations
for greenbelt width are in the literature, little or no data are given.
Recommendations, then, must be viewed as  tentative.  Significantly, in fil-
ters  observed for long periods of  time (i.e., 15 months), sediments did not
clog  the loose surface litter nor  decrease the efficiency of the filter.
Trimble and Sartz suggested that leaves  fall each year and form a fresh sur-
face  for deposition of new sediment.  This indicates that sediment filters
may be effective for long periods  of time without sediment buildup along
streambanks due to natural variability in the plant community from year to
year  (Gosz et al. 1972).  However, the length of the effective period must
be examined by more detailed, long-term  research.

     More elaborate equations were later  developed for predicting the sedi-
ment  flow distances using multiple regression techniques (Haupt 1959, cited
in Broderson 1973).  Sediment flow distance  is primarily a function of (1)
slope of the filter and the density  of obstructions within the filter, i.e.
grass, brush, tree stumps, depressions,  etc. and (2) the slope and length of
the embankment which the runoff passes down  before it enters the filter.
More recently, Ohlander (1976) developed  an  equation to predict the size of
                                      17

-------
 buffer required to trap sediment  exitting a logging road drainage.   His  pre-
 diction was based on soil erodibility  (K—see page 40), slope of the filter,
 and the infiltration rate of water into the filter.
        _C
        -*->
        T3
             75
            60
            45
            30
         £_
         (D
        Li.   15
                       Recommended
                        widths for:
                   ^Weighted  means
                          for  observed
                          data
                           1
             I
I
1
I
10    20   30   40   50
          Filter  Slope
                                                   60
 Figure  1.  Relationship between slope of a vegetative  (sediment) filter and
           effective filter width.   Note that the  recommended widths are
           "evaluated guesses" rather than empirically defined widths.  As
           noted in the text the scatter of data is broad suggesting that a
           number of factors may be important in determining optimal filter
           width.  (Modified from Trimble and Sartz 1957).

     Although much work evaluating the use of natural  vegetation for filter-
 ing  sediments from overland flow has been conducted by foresters, relatively
 little  has been done in agricultural watersheds.   One  study relevant to this
 problem (Mannering and Johnson 1974) passed sediment-laden water through a
'15 m strip of bluegrass sod (Table 2) .  They found  in a single trial that 54%
 of the  sediment was removed from the water.

     Studies of surface runoff through heavy cornstalk residue on the lower
 3 m  of  a 11 m erosion plot carried only 3-5% of the sediment expected from a
 bare surface (G. R. Foster, pers.  comm.).   The study was done with simulated
 rainfall (6.2 cm per hour).  Foster, an hydraulic  engineer with USDA-ARS in
 West Lafayette, Indiana, felt that  well-designed and maintained grass strips
                                     18

-------
would be more effective at  sediment  removal  than  is  reflected  in  the P values
of the Universal  Soil Loss  Equation  (USLE)  (Wischmeier  and  Smith  1965).


TABLE 2.  EFFECT  OF A BLUEGRASS  SOD  STRIP ON SEDIMENT CONCENTRATION IN RUNOFF
	(Modified  from  Manner ing and Johnson 1974)	

  Distance  into                  Slope between       Sediment concentration
  sod strip                      sample  points       expressed  as  percent of
      (M)	(%)	original
0
3
8
15
4
20
4

100
78
66
46

     The  effect  of  grass  and  presumably  other vegetation  types on sediment
load is really a two-fold phenomenon.  The  studies  already mentioned docu-
ment the  value of vegetation  or  surface  litter  for  reducing  the  amount of
sediment  in  a given volume of water.   A  second  benefit  is the reduction  in
volume of runoff from vegetated  areas.*   USDA studies on  demonstration plots
in Wisconsin (0.04  hectares-400  sq. meters)  have  shown  that  runoff  from  grass
is one-fifth of  that from continuous  row crop plots (Glymph  and  Holtan 1969).
Rotations with grain or hay had  intermediate delivery rates.  They  concluded
that runoff  is inversely  related to the  density of  vegetation and frequency
of cultivation.   When forested watersheds are compared  to areas  under intense
cultivation,differences in runoff may be as great as 40X  (Hornbeck  et al.
1970).  Hewlett  and Nutter (1970)  have shown that in natural watersheds  over-
land flow is a rare phenomenon.   The  retardance of  the  flow  by the  vegetation
allows the soil  to  absorb much of the water, which  then goes on  yielding it
to the stream for extended periods of time  (Fig.  2).  The results of this are
(1) the water is passed through  a soil filter before entering the stream and
(2) there is a more controlled release of water from the  watershed, resulting
in a more stable water supply and aquatic environment  (Curry 1975).  On  land
with reduced vegetation cover, the probability  of flooding increases and
seasonal  shortages  of water are  more  common. This  suggests  that incorpora-
tion of greenbelts  along  streams might not  only reduce  sediment  loads but
also produce small  reductions in runoff.  Vegetation, with associated litter
and soil  characteristics, might  serve as a  sponge to hold water  for a slow
release amd  more stable aquatic  environment. Incidentally,  the  same object-
ives (slow release  and reduced sediment  loads)  may  be accomplished  with
some construction practices such as parallel tile outlet  terraces.
*Actually a  third  factor  is  also important.   Rainfall impact on soil struc-
ture  is reduced when it strikes vegetation before reaching the soil surface.
These factors  are  dealt with implicitly in the C and P terms of the USLE.   A
clearer definition of their  relative importance is required.


                                      19

-------
                  £  60
                  L-
                  (D
                  Q.
                  if)
                  0
                  Q)
                  £_
                  CO


                 I
                 Q
                     40
                                       -Farmed
                        /''jX^Forested
Figure 2.
                     40     80      120
                   Time  (min.)

Effect of land use  on discharge  rates  on two watersheds (Modified
from Borman et al.  1969).
     In summary,  limited  data indicates that vegetation and surface litter
can effectively remove  sediment, and perhaps reduce runoff, from surface flow
before water reaches  channels.  However, the size and type of vegetative strip
needed to remove a  given  fraction of the overland sediment load or its
effect on rate of water release cannot be established at this time.
CHANNEL FLOW

     Early observations documenting the ability of vegetation to act as a sed-
iment filter in channels  are primarily descriptive.  When tamarisk invaded an
area along a stream above a reservoir in Carlsbad, New Mexico, the stream
spread across a broader flood plain, decreased its velocity, and deposited
sediment around the tamarisk  (Tamarix gallica) rather than in the reservoir
(Brown 1943).  This resulted  in  a decrease in the rate of silt deposition in
the reservoir (Fig.  3).   Similar observations were made by others (Taylor
1930) but quantitative  measurements of the effect of vegetation on flow
velocity and efficiency of sediment removal were not performed.

     Other more quantitative  field experiments examined the effect of vegeta-
tion on the flow of water in  open channels (Cook and Campbell 1939,  Palmer
                                    20

-------
1946, Ree and Palmer 1949, Ree 1949).   Although designed  to evaluate scouring
problems, they provide quantitative documentation  of how  vegetation affects
the velocity of water.
'u 80
05
§"60
          O
   40

   20
          or
                                        i  Tamarisk
                                        '_/  colonizes
                       Water
                                                   ediment
         1895    1905
                                     1915
                                    Year
1925     1935
Figure 3.  Silt volume as percent of basin capacity for  Lake McMillan near
           Carlsbad, New Mexico.  Flattening of the curve after 1915 is
           attributed to the effect of tamarisk in the valley above the
           head of the reservoir (Modified from Brown 1943)

     When water depth is less than the grass height, velocities tend to be
low  (Fig. 4).  From the channel bottom to the top of the grass the veloci-
ties do not  exceed two feet per second.  At the upper surface of the vegeta-
tion the velocity increases rapidly.  These results suggest  (1) a substan-
tial reduction in the flow velocity due to the retarding action of the
vegetation,  (2) deposition of a fraction of the sediment the runoff is
carrying, (3) the effectiveness of the vegetation at retarding the runoff
and allowing deposition of sediments depends on the relationship between
the depth of runoff and the height of the vegetation and (4) reduced water
velocities retard drainage.  When water velocities are  slowed enough to
retard drainage the potential for crop damage exists.  Vegetative filters
should be designed so that water is not detained long enough to damage corn
or soybeans, i.e. 4-6 days (Hamilton 1969), but sufficiently long to provide
a more regulated release of water from the land surface  and  allow sediment
deposition to occur.  This may also be significant at reducing downstream
flooding and erosion due to flooding outside the streambank.  This latter
point has recently been documented below the washed-out  Teton Dam where iso-
lated tree stands prevented excess erosion (N.G. Benson, National Stream
Alteration Team, pers. comm., 1976).

    The importance of depth of runoff is further illustrated in Figure 5.
Maximum flow retardance is obtained during low flows since vegetation  is
still upright (neither bent nor submerged).  The gradual increase in retarda-
tion in this range is probably due to the greater surface  area of vegetation
                                     21

-------
Figure 4.
                        Water  surface^
                         60       120      180

                           Velocity (cm/sec)

          Water velocity as a function of depth.  Note  that velocity
          increases rapidly near  the top of  the grass and even faster for
          channel flows exceeding grass height (Modifled from Ree 1949).
       5 1.0

       (J
       CD
       O
      u

       (D
       u
       c
       03  0.1
      -Q
       L.
       03
            .01
                     Low Flows
Intermediate
                                           Submer'g
                                             begins
                                Depth (ft)
Figure 5.
          Relationship between depth of flow and amount  of retardance with
          bermudagrass (Modified from Ree 1949) .
                                  22

-------
encountered with increasing depth.  As the depth is further increased, the
plants start to bend and retardance of the flow is decreased.  Finally the
vegetation is shingled or flattened and offers little or no resistance to
the flow of water.  This indicates that flexible vegetation may be ineffi-
cient at filtering sediment from channel runoff of large depths, i.e., 0.1 m
or larger.  The exact depth will of course depend on both the height and
flexibility of the vegetation.  This effect of depth on filtering efficiency
has been partially substantiated by Wilson (1967) and the importance of it
will be discussed later in this report.  (Note that when water depths are
low relative to grass height the channel characteristics are similar to those
of overland flow filters discussed above.)

     More recently, a mathematical model of the sediment filtering capability
of simulated vegetation in shallow channel flows suggested that the impor-
tance of depth of flow and the effect of vegetation on flow velocity are only
two of several factors determining the efficiency of vegetative filters
(Trollner et al. 1973, 1975, 1976; Barfield et al. 1975).  Using nails to
simulate vegetation and glass beads of varying sizes to simulate sediment,
the percent of "sediment" trapped in the channel was correlated with the  inde-
pendent variables:  (1) slope of the nails, (2) spacing of the nails, (3)
particle size, (4) velocity of flow, (5) filter length, and (6) input sedi*-
ment concentration.  Evaluation of the effect of the independent variables
was achieved by "optimizing" a linear regression so that the standard error
for predicting the percent of sediment trapped from the independent variables
was minimized and the correlation coefficient was maximized.  The final model
revealed that (1) a large percent of the input sediment in a shallow channel
is filtered by the "vegetation" and (2) the most important independent vari-
ables determining the  fraction of sediment trapped are, in order of importance,
flow velocity, depth of flow, spacing of "grass" blades, sediment size, and
filter length.

     In a similar experiment stiff, medium, and high flexibility strips of
plastic proved to be efficient sediment traps (Kayo et al. 1975).  At high
velocities filter efficiency decreased due to bending of the flexible strips.
This suggests that the relationship between depth of flow and retardance
illustrated in Figure 5 is also a function of flow velocity.

     These simulations and models cannot be used to predict the efficiency of
real vegetation as a sediment filter.  They are restricted to an artificial
set of independent variables thought to be important and are evaluated in a
non-random and artificial sequence.  As a result they ignore other variables
which may be more important in the real world, i.e. the effect of duration
of input into the filter, and they fail to evaluate the frequently random
and synergistic nature of these independent variables, i.e. a large storm
event while the vegetation is still dormant.  Therefore their usefulness is
limited.  Their importance is their clarification of relationships which re-
quire more research under natural conditions.

     Unfortunately only a minimal amount of data exists to evaluate the
efficiency of real vegetative filters in open channels.  However, the data
are encouraging.  Wilson (1967) attempted to develop economical methods  for
                                     23

-------
 sediment removal from flood waters to allow use of the water for artificial
 recharge.  He allowed turbid flood water to filter through various types of
 vegetation and analyzed their efficiency at sediment removal.  The major
 findings of his study are discussed and illustrated in the next several pages,.

    1.  Filter efficiency varies with vegetation type.  The most efficient
        species studied to date, coastal and common bermuda grass (Cynodon
        dactylon), usually removed 99% of the initial sediment concentration
        (5000 ppm) in 300 m of filtering (Table 3).  The percent removed
        increased with distance.  Sediment removal seems to be a decreasing
        exponential function, differing with vegetation type.  This is
        schematically illustrated by the set of curves in Figure 6.  Varia-
        tion in the efficiency of sediment removal among grasses is corre-
        lated with retardance of flow by the vegetation.  This finding has
        two important implications:

        a.  Intensive studies are needed to document variation among plants.
            Limited data already available (Ree 1949, Ree and Palmer 1949)
            may provide first approximations to suggest which vegetation
            types should be examined for use as sediment filters.

        b.  There will be a trade-off between rapid drainage of the land and
            amount of sediment removed from the runoff.  Clarification of
            patterns of variation within and between vegetation types must
            be a major objective of research in this area.
TABLE 3.  PERCENT REMOVAL OF SEDIMENT AFTER VARYING LENGTHS OF FILTRATION
	THROUGH BERMUDA GRASS*  (Modified from Wilson 1967)
                 „     c   .              	Distance (M)
                 Grass Species           —	*-^-
                                              90     215       300
Common Bermuda
Coastal Bermuda
0
0
50.0
55.5
90.4
97.5
97.0
98.5

*Initial concentration - 5000 ppm.

    2.  Filter efficiency varies not only with vegetation type but also with
        distribution of particle sizes in the suspended material (Table 4).
        There is an inverse relationship between the filtration length re-
        quired to remove a given percent of a particle size and that particle
        size.  This inverse relationship will have two important implications
        for sizing vegetative filters:

        a.   The size distribution of particles in the runoff will have a
            definite effect on how wide the strips of vegetation must be to
            filter a given percent of the total load and

                                     24

-------
        b.  Emphasis of water quality regulations will have a definite
            effect on their required width; that is, it may be desirable to
            remove various volumes of sediment or a given particle size.
            The preferred alternative will depend on water quality objectives
            and the local situation.
oilOO
c
c
'ni
E
(L)
            80
(D
E
T5
(D
O
-*->
C
u
(L)
Q_
            60
            40
            20
                                Distanced  filtered

Figure 6.  Hypothesized relationship between filter width and  percent  of
           sediment remaining for several vegetation types.  The  four  curves
           are hypothetical.  We expect differences among vegetation types
           but data are not available to show how sediment loads  and filter
           distance will vary among vegetation types.
TABLE 4.  PERCENT OF SAND, SILT, AND CLAY IN SEDIMENT DEPOSITED AFTER FIL-
	TRATION THROUGH COASTAL BERMUDA GRASS (From Wilson  1967)	
Distance from head
    of filter (M)
                            Sand
Silt
Clay
3
8
15
30
60
90
120
51.6
43.2
3.6
3.4
20.2
31.2
21.6
40.0
46.6
77.0
69.4
47.0
37.6
39.6
8.4
10.2
19.4
27.2
32.8
31.2
38.8
                                     25

-------
        The rate of sediment deposition is constant over a range of slopes.
        After a critical (threshold) slope is reached,  the efficiency of
        the vegetative filter declines (Table 2).   These data indicate the
        section of the filter with the largest slope (jsection 2) had the
        smallest proportion of sediment deposited  within it,  even though its
        length was longer than that of section 1  and only slightly shorter
        than that of section 3.   This suggests that not only  will vegetation
        types differ in their relationship between distance runoff is fil-
        tered and percent of sediment remaining (Figure 6) but also, for a
        given type of vegetation, the shape of the curve will differ depend-
        ing on the slope of the filter.  This hypothesized relationship is
        illustrated by the theoretical curves of Figure 7. More extensive
        data are needed to substantiate this relationship. The critical
        slope depends on several factors, including application rate of the
        water to be filtered, grass characteristics, (surface area, height,
        etc.) and the particle size distribution of the incoming sediments
        (Wilson 1967).   If the relationship in Figure 7 is supported by field
        data, the width of the filter required will depend on the slope of
        the land.  Like most environmental technologies this  will produce
        varying effects among farmers depending on the  slopes of their land.
        Under some topographic,  soil, or other conditions, filter width may
        be impossibly large for effective implementation.
       en  100
            80
       C
       
       E
      o
      c
      (b
      u
      Q_
Figure 7.
 60
           40
            20
            below
             critical
              slope Sc
                     Distance   filtered
Hypothesized relationship between distance solution is filtered
and percent of sediment remaining for several slopes.   These ar
hypothisized curves and assume vegetation type is constant.
                                                                        are
                                     26

-------
    4.  When grasses are clipped and flow rates are high enough to submerge
        the grass, filtering efficiency declines to zero.  This correlates
        well with retardance coefficients being lower for shorter grasses
        than for taller ones and retardance decreasing with increased sub-
        mergence as shown in Figure 5.  It further emphasizes the importance
        of examining the filter efficiency of vegetation under normal storm
        runoff conditions.  It also suggests that mowing of grass waterways
        and streamside vegetation should be discouraged, especially during
        periods when storms are likely.

    5.  Grasses used for filters should have the following characteristics:

        a.  deep root systems to resist scouring in swift currents;

        b.  dense, well-branched top growth;

        c.  resistance to flooding;

        d.  ability to recover growth subsequent to inundation by sediment;

        e.  if possible, it should yield an economic return.

     It is apparent that several variables are important in determining how
effective real vegetation will be at filtering sediments from shallow channel
flow.  These variables appear to act in an interrelated manner and include
filtration length, slope of the filter, grass characteristics, size distribu-
tion of the incoming sediments, degree of submergence of the filter,  applica-
tion rate of the water to be filtered and initial concentration of the sedi-
ments.  Fortunately, many of these same variables were judged to be important
by the laboratory studies previously discussed (Trollner et al.  1973, 1975,
1976) and the minimal amount of data regarding filtering from overland flows.
Unfortunately there is no quantitative information available on how these
variables are interrelated when real vegetation is used as a filter under
runoff conditions normally encountered in agricultural watersheds.

     An important problem yet to be discussed concerns the placement of
these vegetative filters.  It is unlikely that a significant amount of sur-
face runoff (and sediment) enters directly into the main channel of a river.
Rather it most likely comes in through smaller, often intermittent tribu-
taries, gullies, and drainage ditches which dissect the landscape.  There-
fore, any use of vegetative filters must concentrate on these areas,
    especially areas where tributaries enter larger channels.  More precise
statements on the placement of these filters will require more detailed
understanding and quantification of the dynamics of sediment movement at the
land-water interface.  Only then can a full evaluation of the economic costs
and benefits of filters be completed.

     The data obtained by this literature review suggests that vegetation
can serve as an effective sediment filter.  However, the placement of the
filter and the width or length of the filter required to remove a given
fraction of the incoming sediment, and the duration of its effectiveness is
dependent on the interaction of physical factors, biological factors, and


                                     27

-------
the specifications of water quality standards,  all of which are yet to be
thoroughly evaluated under normal agricultural  conditions and quantitatively
related to each other in any predictive manner.   Present land use practices
immediately adjacent to streams (Table 5)  indicates these relationships
should be investigated.  The following questions must be answered before
informed nearstream vegetation management  programs can be implemented.
TABLE 5  LAND USES IMMEDIATELY ADJACENT TO MODIFIED STREAMS AND DRAINAGE
         DITCHES IN THE BLACK CREEK WATERSHED  (Data from Mildner 1976)	


 Land Use Type                Modified Stream                 Drainage Ditch
Crops
Pasture
Forest
Urban
69.9%
4.7%
17.7%
3.6%
74.2%
0.0%
0.0%
0.0%
  Other (Farmsteads,  roads,
         etc.)                       4=1%                           25.8%
    1.   How important  is  runoff  over  streambanks  as  a  sediment  source  com-
        pared  with runoff entering  from intermittent tributaries?   How do
        the contributions from these  two sources  change  among land  uses and
        with stream volume?

    2.   What general relationship exists under  normal  runoff conditions
        between efficiency of  sediment  filtration and  volume of runoff,
        initial concentration  of sediment,  size distribution of incoming
        sediment,  filtration length,  slope  of filter,  characteristics  of
        the vegetation, and  degree  of filter submergence?

    3.   Is  there a quantitative  relationship between the retardance of flow,
        vegetation,  and filtering efficiency?   If so,  what  are  the  implica-
        tions  for an economic  cost  and  benefit  model?

    4.   For different  vegetation types  what is  the shape of the curve
        relating distance the  water has passed  through the  filter  to percent
        of  sediment  remaining?  Why do  vegetation types  differ  in  the  shape
        of  this curve?

    5.   What is the  duration of  effectiveness of  the vegetative filter?
        Does arrangement  of  the  strip(s)  of vegetation effect their effi-
        ciency and/or  duration of effectiveness?

    6.   Could  a sediment  filter  be  harvested and  still be effective?   If so,
        this will  provide another economic  incentive for filters.

    7.   How much phosphorus  can  be  removed  by filtering  the sediments?
                                     28

-------
    8.   What is the effect of surface litter on filter efficiency and the
        longevity of effectiveness of the filter?

    9.   What effect will tillage practices (i.e. contouring vs.  straight row)
        have on filter efficiency by reducing the peak rates of runoff
        (Glymph and Holtan 1969)?

   10.   How much will filters along streambanks reduce erosion from flood-
        waters which overflow stream banks?  Will they result in sediment
        deposition by the floodwaters  (Parsons 1963)?

   11.   How much will greenbelts reduce drainage rates of agricultural land?
        Will it be economically detrimental?

   12.   How much land will be taken out of production to minimize sediment
        and nutrient loads?

We believe the last two questions regarding the economic impact of managing
for improved water quality by the use of vegetative filters can only be
addressed when more comprehensive information regarding the first ten ques-
tions is available.
                                      29

-------
                                  SECTION 6

                EFFECT OF CHANNEL MORPHOLOGY ON WATER QUALITY
     Studies of sediment movement and deposition in aquatic environments have
largely been from either a geological or hydraulic engineering perspective.
Geologists have been primarily interested in sediment deposition (Leopold
et al. 1964,,Friedkin 1945, Bridge 1975, Jackson 1975) and have directed
little attention towards the dynamics of suspended solids.  Hydraulic engi-
neers have emphasized the governing processes in the sediment environment,
with most emphasis on the shearing flow of fluids, turbulences, and vortices
(Chow 1959, Graf 1971).

     The relationship between environmental factors like stream morphology
and sediment discharge appears to have only recently been examined in any
detail in the natural environment (Stall and Yang 1972, Yang 1972).  The
data found regarding this, relationship will be presented here, along with
the hydraulic theory relevant to the feasibility of maintaining a more
"natural" stream morphology to improve water quality.

     There are a number of hydraulic parameters determining the amount of
sediment transported within a stream, including intensity of turbulences,
velocity of water flow,  slope of the water surface, bed roughness, shear
stress and many others (Morisawa 1968, Blench 1972).  Because these variables
are not independent, the precise relationship between them as a group and
the sediment load of a stream is difficult if not impossible to establish.
As a result, hydraulic engineers have determined the relationship between
single variables and sediment transport. Interrelationships between total
sediment load and velocity, slope, shear stress or water discharge have been
examined (Yang 1972).   In many cases predictions of sediment transported by
a given flow are possible but conflicting results are also common.  For
example, for a given water surface slope two different discharges can be
observed (Fig. 8) and for nearly identical flow velocities large differences
in sediment discharges can occur (Fig. 9).  It is apparent that more indepen-
dent variables and a more process oriented approach needs to be utilized to
provide a better predictive capability.

     Because of these inconsistencies the concept of unit stream power was
proposed to predict total suspended sediment concentrations (Yang 1972,
Stall and Yang 1972, Yang and Stall 1974).  The concept is an expansion of
previous equations and employs two variables, velocity and slope, to predict
sediment loads.   Basically, they suggest that the rate of sediment transfer
is related to unit stream power (USP)—the rate of energy expenditure by a
stream as it flows from a higher to a lower point.  It is defined as the time


                                     30

-------
rate of potential energy expenditure per unit weight of water in an alluvial
channel and can be expressed mathematically in  terms of average water velo-
city (V) and, under steady uniform flow conditions, the surface slope of the
water  (S) (Yang 1972, Stall and Yang 1972).


                    UNIT STREAM POWER = — = — — = VS
                                        dt  dt dX

    where t = time

          Y = elevation above a given datum and is equivalent
              to the potential energy per unit  weight of water
          X = longitudinal distance

More complex considerations of lift force,  critical velocity, drag force,  and
particle diameter were added to this basic concept to more precisely estab-
         u
             1  0
         1/1
         -Q

         (L)
         cn

         -^  irf 1
         u  1 U
         1/1
         ~°  m
         CD  IU

         "m
         o
              '2
                             10
                               -3
  -2
10
  -3
10
                        Water  surface  slope(feet/foot)
Figure. 8.   Relationship  between water surface slope and total sediment dis-
           charge for  one  set  of  experimental conditions (Adapted from Yang
           1972).
                                     31

-------
             1.0
        u
        
        in
        CO
       -Q
CD
CD

03
_C
U

TJ
     10
        -1
       52   10
              r2
       0)
       E
       TJ
       03
       -f->
       O
       h-
Figure 9.
            1<53
            id4
                1
                      2            46
                        Velocity  (ft/sec)
.   ,  I
8   10
    Relationship between velocity and  total  sediment discharge for
    one set of experimental conditions (Adapted  from Yang 1972).
blish the relationship between USP and sediment  transport  (Yang 1972, Yang
and Stall 1974).   For a detailed discussion  of these  considerations those
publications should be examined.   The  suspended  sediment concentration in a
stream or agricultural drainage is to  a large degree  determined by the
stream's USP (Fig.  10)-

     Before we discuss the concept of  Unit Stream Power in more detail, it
is important to  discuss a very important point.  The  Unit Stream Power con-
cept was developed to predict  the amount of  energy available to transport
sediment.   A stream will only  carry that amount  of sediment which can be
transported by the available energy.   However, the stream may carry less than
                                     32

-------
that amount if no source of sediment is available.   In watersheds with little
or no surface erosion and stable banks, sediment loads may be well below the
level predicted from consideration of USP-  The energy available to transport
sediment is only useful in predicting actual sediment loads if other things
(sediment availability and other factors discussed  below)  are held constant.
                     104F
E
Q.
Q.

00
C


I
TJ

CO

"O
(D

C
(U
Q.
LO
ZJ
CO
                     10
                     10
                                                    1 n
                                      0.01          0.1
                          Effective  unit  stream
                            power (ft-lbs/1 b/sec)

Figure 10. Relationship between effective unit stream power and measured sus-
            pended sediment concentration for four streams.  (Modified from
            Yang and Stall 1974).

     Although  these cautions should always be kept in mind, the concept of
unit stream power provides a tremendously useful tool as a first step to
understanding  a variety of characteristics of stream channels.   These in-
clude bed roughness, bedform, width adjustments, pool-riffle frequency,
meandering characteristics, stream bed profile, and sediment load.  (See
Blench 1972 for a good, brief introduction into this field.)  Basically, a
flowing stream has  an energy potential which must be dissipated.  Man must
learn to recognize the extent to which he can minimize the effects of that
energy dissipation on water quality and, therefore, on human society.
                                      33

-------
     Two questions will be examined in this section of the report:

    1.  What characteristics of natural streams or channels control
        flow velocity, slope, unit stream power, and the potential of the
        stream to transport sediment?

    2.  What are the effects of present stream management practices on sedi-
        ment dynamics in watersheds?

     Only the most relevant and "manageable" variables will be discussed hera
A more detailed account of other variables which control velocity and slope
can be found in Graf (1971) or Chow (1959) .

    1.  Area of Flow:  This parameter is critical in determining the velocity
of flow for a stream at a given discharge and slope.  Its effect is illus-
trated by the continuity equation from hydraulic engineering;

                                   Q = VA

                             where Q = discharge Ccfs)
                                   V = velocity (jfps)
                                                2
                                   A = area (ft  )

               or from our perspective a more meaningful form
Under constant discharge conditions (Qi = Q2 = Qn) differences in velocities
(V^, V2, Vn) are a function of differences in area of flow.  Therefore, for
a given discharge under steady and uniform flow conditions, a drainage with
a restricted area of flow, such as by confinement in a ditched channel, will
have a higher flow velocity than a drainage which spreads its discharge over
a larger area.  This implies that by channelizing small drainages and con-
centrating the flow in a minimal area with a maximal velocity, one increases
the USP and enhances the capability of the water to erode the stream banks
and transport sediment.

    2.  Roughness Factor (n) :  The n value for a stream or channel is depen-
dent on factors which affect the roughness of the drainage area, including
size and shape of grains on the bed surface, sinuosity of the channel, and
obstructions in the channel such as vegetation, logs and sandbars.  As rough-
ness increases, n increases and the flow velocity and USP decrease.  Values
for n have been empirically determined for a number of channel conditions
(Chow 1959).  Some effects of soil type, channel morphology and vegetation
on the magnitude of n are illustrated in Table 6.  These data have two irapor^
tant implications :

    a.  Clean straight channels normally found in agricultural drainages have
        very low n values.  They will have high flow velocities, slopes, and
        a high capability of carrying large loads of suspended sediment (high
        USP) .

                                      34

-------
        Options in stream management practices are extensive in that a
        broad range of n values, flow velocities, and USP magnitudes can be
        obtained.  For example, preservation of  natural stream morphology
        (meanders, pool and riffle topography) can increase n and decrease
        flow velocities, slopes, and USP.  Therefore numerous options are-
        available depending on the balance desired between the goals of
        agriculture (rapid drainage) and those of water quality (lower USP's
        and reduced suspended sediment concentrations).
    TABLE 6.  ROUGHNESS COEFFICIENTS  (n) FOR VARIOUS STREAM CONDITIONS
	(From Chow 1959)	


Description of Stream                                   Average n
Clean straight channel, full
  stage, no riffles or pools                              0.030

Same as above, but with more
  stones and weeds                                        0.035

Clean winding channel, some
  pools and shoals                                        0.040

Sluggish reaches, weedy,
  deep pools                                              0.070

Natural channel, variation in
  depth, some logs and dead
  fallen trees                                            0.125

Natural meandering river,
  many roots, trees, brush
  and other drift on bottom                               0.150
It should be emphasized  that no data has been  found directly relating the
effect of altering  the roughness  of a natural  stream on its USP and sus-
pended sediment concentration.  From a  stream  management perspective, further
research should be  done  to more precisely establish the relationship between
the two variables.

    3.  slope or Gradient of Stream Channel: The main way in which a stream
can adjust its~~slope is  via meandering.  By doing so, a channel decreases not
only its slope but  also  its USP and potential  for transporting sediments
(Wertz 1963).  Unfortunately, present stream management practices emphasize
                                      35

-------
maintenance of straight channels which will have larger flow velocities,
slopes, USP's and increased potential for transport of sediment and erosion
of stream banks (Emerson 1971, Hansen 1971).

     The magnitude of the effect of several of these variables on USP was
demonstrated by a study of the Middle Fork of the Vermillion River  (Stall and
Yang 1972).  Measurements of area of flow, roughness coefficients, width,
depth, velocity, and slope were made on a 3360 feet test reach containing
three riffles and two pools.  Measurements were made at low, medium, and high
discharges and USP was calculated for the natural stream and an "equivalent"
channel without pools and riffles (Table 7).  USP was reduced by 23-26% in
a pool and riffle stream during medium and low flow conditions when compared
to an equivalent uniform channel of the type normally formed by present chan-
nelization practices.  Pools and riffles served as an effective means for the
channel to reduce USP and its erosive energy and sediment transporting capa-
bility during low and medium flows.   At high flows the pools and riffles were
obscured and had no effect on stream gradient or USP (Figure 11).  Unfortu-
nately, suspended sediment concentrations were not measured and similar types
of data were not collected for an area of pools and riffles in conjunction
with meandering.
TABLE 7.  UNIT STREAM POWER FOR THREE FLOW CONDITIONS, MIDDLE FORK VERMILLION
          RIVER NEAR OAKWOOD (Modified from Stall and Yang 1972)



Amount
Q
(cfs)

18.9
43.6
681

Flow
Frequency
F
(% of days)

0.80
0.60
0.15
Unit
(foot-pounds

Pools
and
Riffles
0.000282
0.000452
0.002526
Stream Power
per pound

Uniform
equivalent
channel
0.000382
0.000589
0.002522
per second)


reduction
%
26.2
23.3
0

     However, data collected at Black Creek (Karr and Gorman 1975) relating
suspended sediment concentrations to stream morphology correlates well with
the results of Stall and Yang (1972)-  Karr and Gorman measured suspended
solid concentrations in a channelized section above a forest, a meandering
pool-riffle section within forest and a channelized section below forest.
The meandering pool-riffle section acted as a sediment trap during low and
medium flow conditions, resulting in a decrease of 28% in suspended solids
by the time the flow reached the lower end of the forest (Fig.  12).   This
reduction is similar to the reduction in USP caused by pools and riffles
shown in Table 7.   As the flow left the forest, suspended solid concentra-
tions attained the same level as the channel above the forest.   The roughness
                                     36

-------
(n) is the likely factor responsible for the decreased sediment loads -since
the slopes (S)  are lower (.25) above and below the woodlot than in the
woodlot (.40).

     We conclude that  sediment reduction in the Karr and Gorman study
results from the effect of channel morphology instead of the impact of the
near stream biota, at  least under normal flow regimes.  However, during per-
iods of high runoff, the forest or other vegetation along the stream might
act in conjunction with the stream morphology to improve water quality.   To
separate the effect of these two variables, one would have to compare streams
with and without vegetative filters along a major portion of their drainage
length, rather than a  mere patch of natural vegetation as observed by Karr
and Gorman.  Such a comparison should be  made for a wide variety of flow
regimes.

     At very high flow rates  (100 yr. storm) no reduction in suspended sedi-
ment concentrations occurred at Black Creek, probably because pools and rif-
fles were obscured and no reduction in USP occurred.  This result, which is
similar to the result  of Stall and Yang (1972), suggests that knowing the
flow (Q) at which no reduction in USP occurs and its frequency (F), one
could predict what fraction of the year pools would act as sediment traps.
    110
                      = Riffle
                   P=Pool
    105
 05
UJ
     100
                  R
                  P
R
                                                  P
R
             Water  surface profiles
                   High  flow
Medium f I ow
            _L
   J_
                                Streambed
                                                  _L
               800
                              2400
                          stream
              3000
                                                        (ft)
Figure 11.
                   1600
         Distance  along
Bed and water  surface profiles for several  flow conditions in
the Vermillion River.  Note that the pool and riffle character of
the stream is  obscured as flow volume increases.  (Modified from
Stall and  Yang 1972)
                                    37

-------
—
g^oo
(f)

."5 90
O
(f)

-o 80
(D
c
g. 70
(/)
D
i/> <*

-
• i

-

'• — .<








-

—


—

7 i
Fo


i^^
*^<



^e
ct

1
i
i
i
i
i
i
\
\
t






\
i



t

i
\
\
*


1


I
>\)

1
0—0
/
A

1
^

I
/
/
/



Direction of
flo\
A/ tl
/V F
1 1
1
                 12  11   10 9    87    6543
                          Station   number
                                                          2 1
Figure 12.
            Changing suspended  solids  loads  in agricultural and forested
            sections of  a headwater  stream.  Dashed lines indicate margin
            of forest.   High  sediment  loads  at station 5 are due to an ero-
            sion problem at the outlet of a  field tile.  Data from July 1974
            to October 1975.  Sample size varies from 12 to 16 at each sta-
            tion.   (Modified  from Karr and Gorman 1975)

     From this brief review of  the literature and data relevant to the rela-
tionship between the "manageable" parameters of stream morphology and their
effects on water quality,  four  critical points have been established:

    1.   A stream naturally decreases its USP and sediment transporting capa-
        bility by increasing  its bed roughness and by making lateral (mean-
        dering)  (Hussey  and Zimmerman  1953) and vertical (pools and riffles)
        adjustments  (Yang 1971a,b, Stall and Yang 1972).

    2.   USP is reduced by pools and  riffles only at intermediate and low
        flows.

    3.   The effect of meanders  on USP  in conjunction with pools and riffles
        at various flow  rates does not appear to have been quantified.

    4.   Channelization in agricultural areas destroys nearly all characteris-
        tics of  "natural"  stream morphology.  As a result USP is increased
                                     38

-------
        and the ability of the stream to transport sediment and erode its
        banks is enhanced.

These four points, along with the data of Karr and Gorman (1975), indicate
that allowing streams to maintain their natural morphology or allowing them
to reinitiate meandering to reduce USP and suspended sediment concentrations
is a feasible management alternative for improving water quality, especially
during periods of low and intermediate flows.  During these periods pools
act as sediment traps to reduce suspended sediment concentrations and in-
crease the suitability of the water for human uses.  The overall effect on
aquatic organisms needs more detailed investigation.

     This raises the question:  Are materials deposited in low flow periods
flushed downstream during later storm events?  The quantity of material
reaching the final sediment trap such as a lake or  reservoir would not be
affected (unless of course stream bank erosion is a major sediment input)
but the sequence of input would be altered.  Instead of a continual input
into the lake or reservoir, trapped sediment inputs would be restricted to
high flow periods when pools are flushed (Lane and Borland 1954, Straub
1942).  The effect of flushing versus a continual input of attached nutrients
may have significant implications for nutrient dynamics in the receiving
body of water, especially during low flow periods in mid and late summer when
nutrient concentrations are frequently reaching limiting levels for algal
uptake.  Furthermore, the nutrients attached to the sediments trapped during
these periods may have their eventual avilability altered by aquatic inverte-
brates (Davis 1974) before being flushed into the lake or reservoir.  Also,
flushing of the sediment versus a continual input may have a significant
effect on the fraction of sediment trapped by the lake or reservoir via
decreasing its detention time (Rausch and Heinemann 1975) .  All of these
possibilities should be examined by more detailed research on (1) the quanti-
ty of sediment trapped by the stream, (2) the size fraction of sediment
trapped by the stream, (3) timing of sediment transport, and (4) the biologi-
cal and physical interactions involved both on the stream bottom and in the
lake.

     However we feel the main benefit, with respect to water quality, from
a more natural stream morphology would be a substantial reduction in the sus-
pended sediment concentration in a stream during low and intermediate flow
periods, resulting in monetarily important improvements in water quality.
The costs of such a management practice would include (1) taking some land
out of production to maintain a natural meandering channel and (2) the eco-
nomic effect of decreasing the rate of land drainage.  Rough estimates for
(1) could be obtained by using one of the empirically determined relation-
ships between meander amplitude and channel width (Leopold and Wolman 1960,
Leopold et al. 1964).  One should expect a large error in this type of esti-
mate (see Leopold et al.  1964).  Determining the possible effect of decreas-
ing 'the rate of land drainage (#2 above) may also be difficult.  However, if
natural stream morphology was used in conjunction with a vegetative filter
along the streambank  and high quality tile lines were maintained.any detri-
mental effects would probably be minimal, especially in light of the length
of time (4-6 days)  corn and soybeans can withstand flooding (Hamilton 1969).
                                      39

-------
                                  SECTION 7

              DETERMINANTS OF SEDIMENT LOADS IN FLOWING WATERS
     As noted in the introduction to this review early efforts in the soil
conservation field emphasized the control of erosion to insure the mainte-
nance of the productive capacities of soils.  Recently, conservationists have
recognized that meeting this objective does not necessarily result in mainte-
nance of water quality.  For example, in many situations soil losses of 5
tons/acre are tolerable as the soil replaces itself at that rate.  Although
that loss from the soil may be tolerable, the addition of 5 tons/acre into
flowing waters has serious consequences for water resources.  As a result of
this problem, conservationists are attempting to develop a set of water
quality criteria.

     The theoretical foundation for calculation of soil loss is the Universal
Soil Loss Equation (USLE).  The present form of the equation is a result of
more than 20 years of research by many scientists (Wischmeier and Smith 1965).
It has remained relatively unchanged for the past decade.

     The equation is used by agriculturists to predict the average soil loss
in tons per acre per year.  This is compared to the "soil-loss tolerance",
the maximum rate of erosion that will result in sustained production indefi-
nitely.  This model is presented in the form of an equation:

                              A=RKLSCP-

     Each of the terms of the equation are defined and discussed briefly
below.   For more detailed discussion of the model and its use, the reader
is referred to Agriculture Handbook No. 282 (Wischmeier and Smith 1965).

    Computed Soil Loss (A).   The average soil loss in tons per acre per year
    as computed from the six factors of the Universal Soil Loss equation.

    The Rainfall Factor (R).  This is the number of erosion-index units in
    the average year of rain.  An erosion-index unit is a measure of the
    erosive force of specific rainfall reflecting the combined effect of
    rainfall impact to dislodge soil particles and runoff to transport
    dislodged particles.

    The Soil-Erodibility Factor (K).  This component of the equation measures
    the susceptibility of the soil to erosion.  Since erodibility varies with
    s.1 ope, cover, management, and other factors it is essential that erodi-
    bility be measured under controlled conditions.  Generally, it is
    expressed as a relative value for a specific soil in cultivated continu-
                                     40

-------
    ous  fallow on a 9% slope 72.6  feet long.   Soil characteristics  which
    affect  the K-factor include,  among others,  infiltration rate, permeabil-
    ity  and total water-holding capacity.

    The  Slope-Length Factor (L).   This is  the value obtained by computing the
    ratio   of soil loss from a field of any length to that  of a standard field
    with a  length of 72.6 feet.  Soil type and gradient are assumed to be
    constant.

    The  Slope-Gradient Factor (S).  This is the ratio of soil loss  from the
    field to that from a 9% slope, other factors held constant.

    The  Cropping Management Factor (C).  The factor C in the soil-loss equa-
    tion is the ratio of soil loss from land cropped under  specified condi-
    tions to the corresponding loss from tilled, continuous fallow.  This
    factor  attempts to evaluate the combined effect of cover, crop  sequence,
    and  management practices.  Crop residues may be left on the surface,
    removed, chopped, or plowed under.  Crops may be grown  continuously or
    rotated in various combinations.  These and other variations in land
    management which change erosion rates are incorporated  into factor C.

    The  Erosion-Control Practice Factor (P).  The factor P  is the soil loss
    with various practices relative to soil loss with straight-row  farming,
    up-and-down slope, contour  tillage, stripcropping on the contour, ter-
    race systems, and stabilized waterways are the most important practices
    involved in the P factor of  the USLE.  A number of practices such as
    improved tillage regimes, sod-based rotations, and fertility treatments
    contribute to erosion control, but these are considered conservation
    cropping and management practices.  They are, therefore, incorporated in
    the  factor C discussed above.

     Recently, a number of studies  (Stall 1964, Karr and Gorman 1975, Great
Lakes Basin Commission 1976) have shown that the USLE may not be useful in
predicting the sediment transported by a stream or river.  A regional analy-
sis of lakes and watersheds in central Illinois showed a predicted  soil loss
of 3.2 tons/acre, while deposition  in lakes averaged only about 1 ton per
acre per year  (Stall 1964).  The proportion of the eroded soil that reached
the lake bed varied among lakes from 10 to 50%.  In the Maumee River Basin
only about 11% of total erosion is yielded annually as suspended sediments
to Maumee Bay  (Great Lakes Basin Commission 1976).  Clearly, much of the
material eroded from the land  surface, using the assumptions of the USLE,
does not reach lakes or reservoirs.

     This raises the question  of where the sediments are deposited.  Perhaps
they are deposited in  the channels  of  the  streams and rivers that  flow  into
the lakes.   Perhaps the soil is deposited  on the land surface  before  it
reaches the  stream or  on the banks  of  the  stream.  Or if it  is in  fact
reaching the lake, it may be carried  out by the discharge water.

     Alternatively, the techniques  for measuring and calculating amounts  of
sediment deposited in  the lake or reservoir may not be accurate.   In  the
following discussion we shall  comment  on the problems associated with each of
these possibilities.

                                      41

-------
     1.   Errors in calculating volumes of lake sediment.   There certainly is
 some error associated with this factor but it is probably minor when compared
 to the  range of variation observed among lakes.

     2.   Sediment removed from lakes by discharge water.   The potential for
 error here is larger.   No doubt some small particles move downstream by this
 route.   Variables that might be considered here  include  the size, turnover
 time, and cross-sectional area of the lake (Rausch and Heinemann 1975).  Long,
 narrow  lakes or small lakes might have rapid flowthrough rates which result in
 considerable fine particle material being carried out of the  lake before it
 is deposited.   However, as Stall (1964)  reported, a large share (39%)  of the
 sediment in lakes is clay, suggesting that much  of the fine material is de-
 posited in the lake bed.   In summary, there is reason to believe that  this
 factor  may be of some significance in the variation among lakes but more
 careful study is needed.

     3.   Soil deposited on land surface at varying distances from the channel.
 This problem is certainly dealt with in the USLE in the  form of erosion con-
 trol practices (P).   However,  a larger problem is the difficulty involved in
 scaling differences between models and the real  world.   For example, erosion
 determined from study of a small plot can only approximately be expanded to
 very large areas.   The problem is even more complex when the larger area
 has a diversity of  slopes, for example,  and includes some pockets from which
 there is no outlet  except percolation down through the soil.   Furthermore,
 even small discontinuities such as fence rows, field borders,  concave  vs.
 convex  slopes as the floodplain is approached (Young and Mutchier 1969) are
 commonly not incorporated into USLE determinations.   This general factor—
 sediment deposition before runoff reaches the stream—needs more investiga-
 tion and is the larger class in which greenbelts should  be included.

     4.   Deposition  in the stream channel.   Recognition of this possibility
 stimulated the excellent  research of  Stall and Yang that resulted in the idea
 of Unit Stream Power to predict suspended sediment concentrations.   As dis-
 cussed  above (Section 6),  this theory suggests that sediment  concentration
 is an increasing function of unit stream power (Figs.  10 and  13),  and  the
 variation in suspended sediment loads for a given stream power should  be low
 (Fig. 13,  stippled  area).   As  is often the case,  the real world is only a
 rough approximation to the theory.   For  any USP  there is commonly consider-
 able variation in sediment loads (Fig.  13,  cross-hatched area).   That  is,
 sediment concentration is an increasing  function of energy in the stream,
%which is an incresing  function of USP.   This is  true as  long  as a myriad of
 other factors  are held constant:   stability of bank,  erodibility of soils,
 bedload availability,  variation in land  use,  and many others.   A different
 line could be  necessary in Fig.  13 for each such factor,  therefore accounting
 for the variation in sediment  load with  USP which is frequently observed.

      Clearly,  the unit stream  power concept is useful but some variation in
 sediment loads is unaccounted  for when this model is used.  An example will
 illustrate the weaknesses of the single  factor approach  of USP to account for
 variation in sediment  loads.   Two identical forested watersheds  are con-

                                      42

-------
trasted  in  the  following way:  one  is  cleared of forest and planted with a
row crop such as  corn, while the  second  is left as a forested control.  The
streams  in  these  two watersheds would  carry markedly different sediment loads
despite  identical unit stream powers.  Clearly, velocity will vary due to
differing runoff  rates but  the general point here is that with the same
slopes and  at similar flow  rates  sediment loads will not be the same in the
two streams.

     What is needed then is a model, analogous to the universal soil loss
equation, which will predict sediment  loads for a flowing stream.  This
model, dare we  call it the  Universal Sediment Load Model (Equation), must
consider a  complex of factors.  It  is  too early to specify all the parameters
of this model.  However, the previous  discussion and comments in following
sections show that the situation  is more complex than the Universal Soil
Loss Equation might lead us to believe.  Among other things a realistic sedi-
ment model  must consider the magnitude of erosion from a watershed, how and
where eroded material is removed  before runoff reaches waterways, and the
addition of sediment from underground  (drain) tile systems.   Finally, as
discussed in a  later section, the nature of the erosion-deposition equili-
brium between the terrestrial and aquatic environments must be clearly
understood.
            ID
            03
            O
            c
            
-------
                                  SECTION  8

             EFFECT OF  STREAMSIDE VEGETATION ON WATER  TEMPERATURE
      This  report has emphasized  the feasibility  of using vegetation  along
 streams  to reduce sediment and nutrient inputs into agricultural  drainages.
 Another  potential use of  this vegetation is  to control water  temperature.
 Past  evaluations of the effect of streamside vegetation on water  temperature
 will  be  reviewed here, along with an examination of the significant  effects
 of  temperature on dissolved oxygen levels and nutrient dynamics,  especially
 the release of nutrients  from sediments and  their uptake by nuisance algae
 and periphyton.  The importance  of water temperature in maintaining  a  "fish-
 able" water quality will  be discussed in a later section of this  report.

      Because of the importance of temperature in regulating the physical and
 biotic characteristics of streams, a number of studies have documented  the
 effect of  streamside vegetation  on water temperature.  Green  (1950), in a
 study at the Cowetta Hydrologic  Laboratory, documented the effect of land use
 on water temperature.  For a period of one year  he compared the temperature
 of a  stream on a farm which had  originally supported a hardwood forest  but
 for eight  years had been  under cultivation and pasturage and  that of a
 stream in  mature hardwood forest  (Fig. 14).  The  weekly maximum temperatures
 of the farm stream ranged from 5.0 to 12.8°C above the forest stream (average
 6.4°C).  Both streams were coldest during the month of February but  the
 temperature of the forest stream frequently ranged as high as 3.9°C  above
 the farm stream.  Similar effects of vegetation  on temperature extremes were
 found in studies of a stream before and after vegetation was removed from the
 streambank (Gray and Eddington 1969).  In another study, stream temperatures
 inside a small woodlot (19°C) were much lower than in unshaded areas (28°C)
 nearby (Karr and Gorman 1975).   These data indicate vegetation serves as an
 effective  buffer against  temperature extremes; shaded streams are cooler in
 the summer and warmer during winter.

     A more detailed analysis of the result of several types of land treat-
ment on stream temperature was performed in the  Appalachian Mountains
 (Swift and Messer 1971).   Clear-cut streams averaged (5.5-6.5°C) warmer than
 forested streams.   During periods of highest temperatures, temperature mini-
mums were  also increased.   This  indicates that the streams did not have a
 "normal" cooling-off period during the night and that periods of elevated
 temperatures may be prolonged.  Long-term elevations in temperatures can
have significant effects on the  energetic "balance sheet" of aquatic eco-
 systems,  resulting in major alterations in their biotic components.   From a
management perspective,  one other important observation was made by  Swift and
Messer.   If streamside vegetation was left to shade the channel, while  the
                                      44

-------
    u_

    GL
    £
80

60

40
                                                  i     i     i     i
             JFMAMJJASON    D

                                  Month
Figure 14.  Seasonal changes in water temperatures of streams from forested
            and farm watersheds (From Greene 1950).

rest of the woods was clearcut, only minor changes in stream temperature
were observed.  This latter point suggests that if a minimal vegetative
"buffer strip" is left along agricultural drainages, significant decreases
in stream temperature and improvements in water quality could be expected.

     More recently, a detailed analysis of the use of "buffer strips" to
control temperature has been made in the field of forestry (Brown and Brazier
1972).  Previous mathematical analyses by Brown (1969) and Brown and Krygier
(1970) demonstrated that net thermal radiation in relation to stream dis-
charge was the primary determinant of stream temperature.  When strips of
brush or  trees were left along the stream, no increase in temperature
occurred.  From an economic and management perspective it is therefore desir-
able to know whether controlling the temperature of the stream is a function
of the volume of vegetation in the buffer strip, the strip width, or the
density of the canopy of the strip which is perpendicular to the suns rays.
An extensive examination of temperature in various streams with different
types of  buffers indicated that angular canopy density (ACD - a measure of
the shading  ability of the vegetation) is  the only buffer strip parameter
correlated with temperature.  The generalized relationship between ACD and
heat input into a stream is illustrated in Figure 15.  Buffer strip width
was not important.  A buffer strip 5-10 feet wide was as effective at temper-
ature control as a strip 50 feet wide, as  long as its ACD was high.  Brown
and Brazier  (1972) also observed that buffer effectiveness decreased with
increasing stream size  (Figure 16).  Small streams have  the  greatest tempera-
ture problem but they are  also the easiest to control, due to the  inverse
relationship between  temperature change and  stream discharge for a given
input of  thermal radiation.  These results indicate  that little  or no land
would need to be taken out of production  for  temperature control on  small
streams.  Further, natural processes will  provide sufficient ACD levels on
small  streams  through succession as  cottonwoods,  willows,  or other types  of
vegetation invade  the stream banks.  Finally,  if  temperature control is
accomplished in the  upper  reaches  of drainages,  a reduction in temperature
associated problems  will  result  in upstream  areas as well  as in downstream
areas,  including  small  lakes and  reservoirs.
                                      45

-------
           Q.
           C
           03
           E
Figure 15.
          Angular canopy  density

Hypothetical  relationship between angular canopy density and
thermal input to streams (Modified from Brown  and Brazier 1972)
           10

           to
           C
           u
           0)
          M—
          H—
           
-------
     The importance of temperature in determining various water quality para-
meters and regulating biotic communities cannot be overemphasized.  As tem-
perature increases, the capacity of the water to hold oxygen decreases
(Table 8).  Since oxygen is utilized during the decomposition of organic mat-
ter, at elevated temperatures the ability of the stream to assimilate organic
wastes without oxygen depletion is reduced.  This exaggerates the impact of
each additional unit of waste added to the system.


TABLE 8.  RELATIONSHIP BETWEEN TEMPERATURE AND MAXIMUM DISSOLVED OXYGEN
	CONCENTRATION IN WATER	


           Temperature  C             Maximum oxygen concentration (jpm)

                  0                                  14.6
                 10                                  11.3
                 20                                  10.7
                 30                                   7.6
     Even more  important with respect  to water quality and eutrophication is
the effect of temperature on the release of .nutrients from sediments.
Sommers et al.  (1975b) studied  the effect of incubation temperature on solu-
ble nutrient  concentrations.   They found significant effects of temperature
on the rate at  which  insoluble  (attached nutrients) were converted to soluble
and readily available forms.  Figure 17 illustrates the relationship between
temperature and amount of phosphorus released from the sediments in a 12-week
period.  This plot  indicates an exponential increase in phosphorus released
with an increase  in temperature.  Slight increases in temperature above 15°C
produce substantial increases in the amount of phosphorus released.

     These data,  along with those previously discussed,  indicate  that by  re-
moving vegetation which  shades  agricultural drainages,  several  detrimental
patterns will  develop:   (1) Increases  in  temperature will  occur during  sum-
mer periods  (5.5-9.0°C),  resulting  in  increasing rates  of  phosphorus dis-
association  from  sediments.  (2)  Increases  in  phosphorus concentrations  in
the drainages  result  in  higher  nutrient concentrations  in  receiving  bodies
such as lakes  and reservoirs.   (3)  Increasingly large blooms of nuisance
algae  and periphyton  will appear because  of  elevated nutrient concentrations,
temperatures,  and light  availability.   The  effect of all these  will  be  to
decrease water  quality and the  quality of biotic communities.

     The  importance of streamside vegetation  clearly goes  beyond  its use  in
filtering sediments and  nutrients  from surface runoff.   Its  potential  for
temperature  control,  enhancement  of  the oxygen carrying capacity  of  the
stream, and  reducing  nutrient  availability  and utilization is evident.   Its
significant  economic  impact on  fishery resources will be discussed in  a
later  section  of  this report.   These benefits  appear  to outweigh  the negli-
gible  cost  of  this management  practice.  Little.or no  land would  have  to be
taken  out  of production  and natural processes  would provide the vegetation


                                      47

-------
for shading the  streams by processes of natural succession.  To fully evalu-
ate this management alternative, more detailed analyses of its benefits to
water quality and  the stream biota in relation to the cost of decreased rate
of drainage must be performed.
               to
               D
               O 4
               Q.
               CO
              Q_
Figure 17-
                10      20       30
               Temperature  (C)

Effect of temperature on the release of  phosphorus from sedi-
ments over a 12-week period (Data from Sommers et al. 1975b).
                                   48

-------
                                  SECTION 9

   IMPACT OF NEAR STREAM VEGETATION AND CHANNEL MORPHOLOGY ON STREAM BIOTA


     Streams may be narrowly viewed as a means for transporting water axvay
from the land, as collecting points for a water supply, or as outlets to
carry away the refuse of modern society.  This combination of uses relates to
the needs for water quality and is therefore relevant to the subject of
greenbelts—especially when increasing demands are being placed on resource
managers to improve the biological aspects of environments and to provide
areas suitable for use by an increasingly recreation oriented public.  This
trend is demonstrated by the federal government's call for "fishable and
swimmable" water by 1983.  Therefore, alternatives proposed for controlling
non-point pollution from agricultural areas must also be evaluated in light
of their effects on the biota of streams and on recreational opportunities.

     A surprising fact to most non-biologists is that small headwater drain-
ages in agricultural watersheds are extremely important in maintaining fish-
able populations (sport and commercial) in larger streams, rivers, or lakes.
They are important as breeding grounds for many valuable species which mi-
grate into these areas from lakes or large rivers to spawn (Hall 1972, Smith
1972, Karr and Gorman 1975).  Also, they serve as breeding grounds and/or
permanent habitats for the smaller fish used for food by these species.  If
these environments are disrupted or destroyed, a major source of food for
the commercially valued species will be destroyed and many of the valued
species will be prevented from reproducing.  This happened in the Lake Erie
basin and was one of the major causes of the destruction of a multi-million
dollar fishery in that region (Regier and Hartman 1973, Smith 1972, Ryder
and Johnson 1972).

     In this section we briefly examine the effects on the biota in headwater
streams of (1) elevated sediment loads, (2) increased water temperature due
to removal of vegetation along stream banks,  (3) disruption of the aquatic
food chain by removal of allochthonous (terrestrial) inputs of energy, and
(4) decreased habitat diversity due to stream channelization.  With this
information we then evaluate the potential of greenbelts for improving the
biological quality of streams.  In Section 10 we briefly outline some of the
possible recreational benefits of greenbelts.

EFFECTS OF SEDIMENT

Fish

     Most research on the effects of sediment on fish emphasizes commercially


                                     49

-------
important species such as the salmonids.  For adult fish only very high con-
centrations of sediment (>20,000 ppm) cause mortality, primarily by clog-
ging the opercular cavity and the gill filaments (Wallen 1951).  This pre-
vents normal water circulation and aeration of the blood.  Such high sedi-
ment levels are rarely encountered in streams.

     However, indirect effects on adults may occur at much lower sediment
concentrations.  In areas with slightly increased suspended solids loads
(35 ppm) adult cutthroat trout showed no increase in distress or mortality,
but the fish did cease feeding and move to cover (Bachman 1958 cited in
Cordone and Kelly 1961).   Similar subtle changes in behavior were noted in
smallmouth bass and green sunfish when turbidities were elevated (14-16 JTU)
for 30 days (Heimstra et al. 1969).  Furthermore, the normal social hier-
archy was disturbed with elevated turbidities.  When in search of spawning
areas, adult salmon swim through areas of high turbidity in search of clear
areas for spawning (Cooper 1956) .  These results suggest that adult fish
tend to be physiologically resistant to the direct effects of elevated
sediment levels but that their behavior can readily be altered by only slight
increases in turbidity levels (Swenson et al. 1976).

     The major effect of sediment on fish populations is the disruption of
normal reproduction (Cordone and Kelly 1961).  When sediments settle out of
suspension they frequently cover essential spawning grounds of fish, cover
eggs, or prevent emergence of recently hatched fry.  One of the earliest
experiments to document the effects of sediment on egg survival was that of
Harrison (1923, Table 9).   As the particle sizes in the bottom decrease, the
survival rate of eggs declines.   A number of other studies have documented
similar effects of sediment on egg survival and fry emergence (Shapavalov
1937, Shapavalov and Berrian 1940, Shapavalov and Taft 1954).  The cause of
the reduced survival is a decrease in circulation of water around the devel-
oping eggs, resulting in the inhibition of oxygen uptake and carbon dioxide
release.  This results in reduced metabolic rates which are lethal if inhi-
bition is prolonged (Snyder 1959).  Thus, the effect of sediment on spawning
success has been one of the major causes for the decrease in the quality of
fisheries throughout the United States and has prompted some (Langlois 1941)
to suggest that changes in bottom type are of utmost importance in modifying
the breeding grounds and determining fish diversity.  As bottom type is sim-
plified by the deposition of sediment, species diversity decreases.  Recent
research (Gorman and Karr 1977)  suggests a direct relationship between
species diversity and bottom diversity (Fig. 18).  However, the low correla-
tion coefficient suggests that other factors are also important in determining
species distributions.  As will be pointed out below, the effect of sediment
on bottom type is only one of several physical factors which interact to
determine the structure of fish communities in streams.  Thus, the single
problem-single solution approach frequently used in the  past (.e.g. reduction
in sediment inputs to improve fisheries) will have limited usefulness for
improving the quality of the stream biota.

Invertebrates

    Many aquatic invertebrates spend a large proportion of their life on
or in the bottom substrate.   Sediment deposition, resulting in decreased

                                     50

-------
TABLE 9.  THE EFFECTS OF PARTICLE SIZE OF  SUBSTRATE ON EGG SURVIVAL IN SOCK-
          EYE SALMON (Oncorhynchus nerka)  (From Harrison 1923).
Number of
eggs planted
      Description of Nest
Number of
eggs hatched
    500


    500

    500


    500


    500
       Large  gravel, very little sand, no clay
         or top  covering of silt

       Small  gravel, some clean sand

       Small  gravel, top coating of silt
         1/2  inch  deep

       Very fine gravel, sand, and small amount
         of clay in  sand

       Very fine gravel and much clay or mud
         in sand
    420


    350

    325


    200


    170
>>
£2*5
CD
"^2.0

-------
substrate types, will inevitably cause changes in the species diversity and
numerical  abundances of benthic organisms (Ellis 1931,  Tebo 1955,  Wilson
1957).  Commonly these changes can be related to changes in substrate diver-
sity  (Smith and Moyle 1944, Barman 1972,  Fig. 19).   There is a general rela-
tionship between substrate diversity and  species diversity of molluscs but
the scatter, as in the fish data discussed above, suggests other environ-
mental variables are also important.   Other studies  have shown the  importance
of depth diversity and water velocity (Zimmer and Bachman 1976),  temperature
(Sprules 1947)  and organic inputs from terrestrial environments  (Ross 1963,
Woodall and Wallace 1972, Cummins 1975) in determining the distribution and
abundance of aquatic invertebrates.   Clearly, reduction  in sediment inputs
must be accompanied by improvements in other water quality characteristics
for maximal benefits to invertebrate populations.
                      15
                    (S)
                    Qj
                    U
                    
                    Q.
                      10
                    —
                    O
                    t_
                    
-------
ecosystems will often decline.  A comprehensive study of the energetics of
Red Cedar River in Michigan showed that stream productivity declined signifi-
cantly following siltation (King and Ball 1967).  When turbidity levels
shifted from 20-30 JTU to 380 JTU, Aufwuchs production declined by 68% and
heterotrophic energy consumption declined by 58% (Table 10) .  Sediment increases
resulted in reduced ecosystem productivities across all trophic levels.


TABLE 10.  ENERGY BUDGETS FOR THE RED CEDAR RIVER BEFORE AND DURING PERIODS
           OF HEAVY SILTATION, SUMMER, 1961. Units are cal uT2 day'1 (King
	and Ball 1967)	


Trophic level                                       Energy fixed

                                      Before siltation       During siltation
Auto trophic Aufwuchs
Macrophytes
Total primary producers
1,140
127
1,267
368
127
495
Heterotrophic Aufwuchs                        360                    170

Insects:
     Herbivores  (net yield)                   43                     -9

     Carnivores  (net yield)                   13                      6

Tubificid worms                               65                    -39

                                Energy Required

Heterotrophic Aufwuchs                      4,000                  1,889

Total  insects                                 283                      4

Tubificid worms                               200                   -115*
Total energy  required                       4,483                  1,893


*Negative values  are not  included  in  the  total  energy  required.

Summary

     Clearly,  the  impacts  of  sediments on  the  stream biota  include alteration
of  the structure  and productivities of plant, invertebrate,  and vertebrate
communities.   Reductions  in  sediment  loadings will improve the biota  of
streams  but,  as we  will demonstrate below,  more comprehensive management

                                      53

-------
programs will be necessary to improve stream biotas;  that is, efforts to re-
duce sediment loads must be accompanied by more  informed management of other
stream characteristics.

EFFECTS OF TEMPERATURE

     Temperature has a significant effect on oxygen concentrations and nu-
trient dynamics within streams (.see pp.  44).   Elevated temperature, due to
the removal of vegetation along stream banks,  increases the susceptibility
of biological systems to inputs of nutrients and other pollutants.  Tempera-
ture changes, regardless of other physico-chemical changes, may also cause
shifts in the entire structure of aquatic communities.

     A clear understanding of this problem depends on knowledge of the ener-
getic processes of organisms.  The use of energy by an organism can be repre-
sented by a flow chart (Figure 20).  Energy is ingested but only a fraction
of what is ingested is actually available to the organism as net energy for
utilization. This net energy must then be divided among standard metabolism
(maintenance of body functions),  active metabolism (providing energy for
swimming, feeding, etc.), growth, and reproduction.  The ability of an organ-
ism to survive and reproduce is dependent on the organism being in an envi-
ronment where some fraction of energy input is available for growth and
reproduction.
Food Energy
                              J-
                     Feces
                        [Assimilated   Energy!
         I
                                                    NH
                        [Net  Energy
     Standard
     Metabolism
Active
Metabolism
[Growth I    [Reproduction!
Figure 20.   Flow  chart illustrating the basic  pathways of energy flow in
            organisms.

     Because fish are cold-blooded organisms  (poikilotherms), temperature
is extremely important in determining their standard metabolic rate (Brett
1972).  As  temperature increases, standard metabolism increases but other
factors,  such as  feeding rates (energy intake),  also increase.  As a result,
each species has  an optimal temperature for feeding, general activity,
                                    54

-------
growth, and reproduction.   This  can be  illustrated by comparing two hypothe-
tical fish species (Figure 21).    For both species, activity, food consump-
tion, and in turn growth,  decrease at temperatures below the optimum.   At
temperatures above the optimum,  food consumption starts to decrease but more
importantly, increasing standard metabolism demands more and more of the
energy value of the food with less being available for growth and reproduc-
tion.  This is of course a very  simplified explanation of the effect of
temperature on the growth  of fish.  Numerous other factors such as food
availability, age of the fish, and oxygen concentration will affect the
exact form of the relationship.   A more detailed discussion of these vari-
ables can be found in Warren (1971) or  Brett  (1964, 1970).  Hypothetical
species 1 (Fig. 21) has a  temperature optimum of 20°C.  At this temperature
it has the maximum amount  of energy available for growth and reproduction,
while species 2 has only a minimal amount of energy available for these acti-
vities because of decreased food consumption.  If the temperature of the
water is increased to 25°C, species 2 would then have maximum energy avail-
able for growth and reproduction while  species 1 would have little energy
available because of its increasing standard metabolic requirements.
                      03
                     "O
                     To
                      u
                     _*:
                      en
                      t_
                      
                      C
                     UJ
      Food  consumption
              \
       Active  metabol.
                                       Standard  metabol.;
                                       ,/^C. jsy TL /^^ >^y~v 7^/^A. /^/ >- /%/ f^y -ic**
          spec ies 1
          species 2
15              20             25
20              25             30
      Temperature  (C)
Figure 21.  Theoretical effect of temperature change  on  food consumption and
            energy budget of two hypothetical fish species  (Adapted from
            Warren 1971).
                                     55

-------
     This illustrates an important point.  When vegetation is removed from
along streambanks and water temperatures increase from 6-9°C (Gray and
Eddington 1969, Karr and Gorman 1975), it may become energetically impossible
for species with lower temperature optimums to continue living in the area,
regardless of changes in sediment loads, habitat structure or other environ-
mental conditions.  A shift in community structure may occur with resident
species being replaced by the frequently less desirable species which are
more tolerant of increased temperatures.

     A common misconception is that shifts in species composition will not
occur unless temperature increases are substantial, especially in warmwater
streams.  It is believed that since warmwater species have higher tempera-
ture optimums than species living in coldwater streams, they have a greater
"assimilative capacity" for temperature increases.  To test if this was true,
predictions of changes in fish communities of a coldwater (Columbia) and
warmwater (Tennessee) river were made based on the preferred (optimal) and
lethal temperatures of their respective species CBush et al. 1974).  The per-
cent of species lost versus temperature is shown in Figure 22.  In both
streams, significant individual losses start to occur after temperature
changes of 5-6°C.  However, as a community, the preferred temperature is
closer to the lethal limit as one moves from cold to warmwater streams.
Warmwater species do not have a greater assimilative capacity for tempera-
ture increases.   These observations,  along with a
recent extensive field and laboratory study demonstrating that temperature is
an extremely important factor determining the distribution of fish species
(Stauffer et al. 1976), indicate that removal of vegetation along small
streams will result in substantial changes in fish community composition by
causing temperature changes which exceed the preferred and lethal limits for
species with lower temperature optimums and favor those with higher optima
and lethal limits.  Similar arguments could be made for the effects of tem-
perature on invertebrate distributions.  If a major goal of present research
on non-point pollution in agricultural watersheds is an improvement in the
biota of streams and their "fishability," substantially more effort must be
directed towards the effect of near stream vegetation on water temperature
and the distribution of organisms.

EFFECTS ON STREAM ENERGETICS

     Up to now, this report has emphasized the indirect impact of near stream
vegetation on stream biota via its effect on sediment and nutrient inputs and
temperature fluctuations.   An important direct effect of the removal of this
vegetation is disruption of the aquatic food web, especially in those areas
where terrestrial inputs are a major source of energy for the stream.

     Headwater streams (stream orders 1, 2, and 3) make up about 85% of the
3.25 million miles of running waters in the continental United States
(Cummins 1975).  These areas represent the maximum interface between the
terrestrial and aquatic environments.   It is here that most sediment enters
streams and here that extensive channelization and removal of near stream
vegetation is occurring.   These are the areas which are most dependent on
near stream vegetation for an energy source and are also extremely important
as spawning and nursery grounds for many commercial and sport species which
spend their adult life in lakes or large rivers (Smith 1972, Karr and Gorman
                                     56

-------
1975, Hall 1972, Hynes 1970).   In these  areas most of the energy utilized by
the aquatic invertebrates and  fish is  terrestrial in origin (Cummins 1975,
Lotrich 1973, Chapman  and Demory 1963,  Hynes 1963, Minshall 1967 - Figure
23).  Once the coarse particulate organic matter (CPOM) like leaves, twigs,
etc. is in the stream, either  the dissolved  organics leached from it are
utilized or the CPOM is directly ingested by shredders which have powerful
jaws to shred the organic matter and then digest the microflora growing on
it.  The shredders then egest  a fine particulate organic matter (FPOM)  which
is utilized by a group of invertebrates  called  collectors.  At the top  of
this food web are the fish predators.
              GO
              O
              £100

              s.80
              00
  o
  -t-<
  c
  
  u
  t_
  CD
  Q_
                      Cold water
                         stream
                                              Warmwater
                                                  stream
                                                    (30°)
Figure 22.
                18     24     30    36    42
                   Temperature (C)
Percent of fish species lost versus  temperature in warm and cold
water streams (Data from Bush et  al.  1974).
     The implication of this web is that in these upstream areas removal of
nearstream vegetation will result in significant reductions in invertebrate
production and, in turn, fish production because of  the loss of allocthonous
energy inputs.  Studies have indicated low diversity and numbers of inverte-
brates in streams flowing through areas lacking deciduous vegetation
(Minshall 1968).  Therefore, if we are to maintain productive fisheries in
streams, we must not only maintain suitable instream habitat for the fish
and their invertebrate food source but we must also  maintain a major source
of energy on which both invertebrates and vertebrates depend:  the nearstream
vegetation.

EFFECTS OF STREAM CHANNELIZATION

     Channelization is one of the most detrimental factors affecting the use
of streams for recreation (including fishing).  The  primary goals of channel-
ization are to prevent flooding of crops and increase the amount of tillable
agricultural land by allowing rapid drainage.  These are desirable goals but
the substantial number of detrimental effects  on water quality, fisheries
                                     57

-------
            Inputs to  Headwater   Stream
Coarse  participate   Dissolved organic
  organic  matter            matter
      (CPOM)                 (DOM)
                                                        .  . ,  ,
                                                        Ll9ht
 I
                                            Producers
                                        Macro       Micro
  Microbes
                           /
                    Flocculation
                                                       Scrapers
       Shredders
                 -*Fme participate
                    organic  matter
                       (FPOM)
                          Collectors

                               \
                              Predators
                                                                  i

  Functional   Groups   of   Headwater  Stream

Figure 23.  Major components of a headwater  stream foodweb.  Note the
          systems  dependence on inputs of  organic matter from the
          terrestrial environment and processing of  that organic matter
          by stream animals (Modified from Cummins 1975).

resources,  and the recreational potential of streams suggest caution in  the
application of technology involving channel  modifications.   From a number of
esthetic, biological, and water quality perspectives the costs and benefits
of channelization should receive a rigorous  economic reevaluation.  Included
among its effects are (Fig. 24):

   1.  Increased stream temperatures and their associated problems due  to
       removal of vegetation along stream banks (Hansen 1971);

   2.  Increased bank erosion (Emerson 1971) and turbidities (Hansen 1971)
       due to higher slopes and flow velocities, resulting in higher unit
       stream-power;
                                 58

-------
U3





STREAM
CHANNELIZATION





Destruction of
pools £ riffles
Cutting off of
meanders &
shortening of
stream length


Deepening of
channel

Removal of
near stream
vegetation

//
Increase in Increase
	 > stream 	 > flow
gradient velocity



Lowering of
	 » floodplain
water table

Increased
temperature

\
in Increased Increased Increased
	 f unit stream 	 * channel s 	 * sediment
power bank erosion loads \
\ N \
\iMore rapid Widening of \
drainage of the channel \
the land \
^il Increased downstream
flood hazard


Loss of habitat
* diversity
V
\
\
\v
^ Loss of potential
aquatic habitat

\
\
V
Decrease in
water quality





BIOLOGICAL
EFFECTS





               Figure 24.  Factor train analysis of the effects of channelization of the physical
                           environment and biota of streams  (Modified from Darnell, 1976).

-------
    3.  Increased flooding downstream (Campbell et al. 1972);

    4.  A reduction in habitat complexity, i.e. riffles and pools are des-
        troyed, meanders are removed, the substrate is made uniform, and
        cover in the stream channel is eliminated.

     One result of the synergistic action of all these effects on the physi-
cal environment is a drastic reduction in both fish and invertebrate popula-
tions.  A number of studies have documented this.  Congdon (1971) examined
the effect of channelization on fish populations in the Chariton River in
Missouri.  He observed an 83% reduction in total standing crop per acre
(Figure 25) and an 86% reduction in standing crop of catchable size fish per
acre (Figure 26) after channelization.  Bayless and Smith (1967) reported a
90% reduction 'in catchable size fish (over 6 inches) per acre in 23 channel-
ized streams compared to 36 more natural streams.  A 98% reduction in fish
standing crop in the Tippah River in Mississippi was observed immediately
after channelization (Wharton 1970 cited in Congdon 1971).  Similar observa-
tions of reductions in fish populations in smaller agricultural drainages
after channelization have also been documented (Karr and Gorman 1975).  Other
studies have demonstrated reductions in invertebrate populations due to
channelization (Morris et al. 1968, Moyle 1976).   For more references con-
cerning the effects of channelization on the physical environment and fishery
resources, the review by Henegar and Harmon (1971) should be consulted.

     Although these studies have provided us with substantial information
concerning the effects of channelization on the biota, there have been very
few attempts to identify and quantify those environmental variables altered
by channelization which affect the distribution and abundance of organisms.
Because the main effect of channelization is the alteration of stream mor-
phology, any relationship which can be established between morphometric
variables and the habitat characteristics of aquatic organisms or morpho-
metric variables and biological parameters should be useful in predicting the
impact of channelization upon aquatic communites and should provide valuable
information for improving engineering designs to minimize the biological
impact of channelization (Zimmer and Bachman 1976, Menzel and Fierstine 1976,
Gorman and Karr 1977).

     Zimmer and Bachman (1976) examined the relationship between channel
morphology, habitat diversity, and drift density of invertebrates in 11
natural and channelized streams in Iowa.  They found a significant positive
correlation between channel sinuosity and variability of stream depth
(Figure 27a) and stream velocity (Figure 27b).   The more the stream meandered,
the more diverse the available habitats were with respect to water depth and
velocity.  In conjunction with this, they observed that as sinuosity
increased (i.e. habitat diversity increased)  the biomass (Figure 27c) and
number of organisms (Figure 27d) in the invertebrate drift increased.  These
data indicate that by designing channels with more sinuosity or leaving
streams with natural meanders, significant improvements in the quality of the
invertebrate biota ar\d availability of food for fish would result.
                                     60

-------
Figure 25.
Figure 26.
              CD
              L_
              U
              0)150
              .0
   §"100
   U
   O)

   150
   ro
                              Unchanneled
                     Channeled
CA  CS
                    CC  B    D
                     Species
                                            RH  FC  Oth.
Estimated  standing crop  for fish* of unchannelized (total 304
Ibs./acre) and channelized (total 53 Ibs. /acre) sections of the
Chariton River, Missouri (Adapted from Congdon 1971).
              _Q
                 150
                 100
   QL
   O
   L.
   U

   cn   50
   c
  ID
   C
   CO
                                Unchanneled
                                  Channeled
             CA  CC    D   CR   FC  LMB
  v /
                    Species

Estimated  standing crop of catchable size fish* in unchannelized
(187 Ibs./acre) and channelized (total  27 Ibs./acre)  sections of
the Chariton River, Missouri  (Adapted from Congdon 1971).
*Codes for  fish abbreviations:
CA - Carp  (Cyprinus carpio)               RH - Redhorse (Moxostoma)
CS - Carpsucker (Carpiodes)               FC - Flathead Catfish (Pylodictis
CC - Channel Catfish (Ictalurus punctatus)       olivaris)
B  - Buffalo (Ictiobus)                   CR - Crappie (Pomoxis)
D  - Drum  (Aplodinotus grunniens)         LMB - Large-mouth  Bass (Micropterus
                                              salmoides)
                                   61

-------
     A similar type of study (Gorman and Karr 1977)  established  a  direct
relationship between habitat diversity and the quality  of  fish communities.
Streams were examined in three different areas and  included  both recently
channelized and natural streams.   For each sample station, fish  species
diversity and diversity of bottom type,  depth of water,  and  velocity of flow
were determined.   The relationship between habitat  diversity and fish  species
diversity (Figure 28) indicates fish are habitat specialists; as the diver-
sity of bottom types, water depths, and  velocity increases  (or any combina-
tion of them), fish diversity increases.
           500
         d
         0)
         TJ
         O
         U
         C  100
         05
         >   50
                     1.1
                              = .62
                  1.5
                            u
                            O
                            o
                            CD
                            (J
                            C
                            03
                                             '(b)
                                               r= 78
                1.1
    1.5
         CO
         £
         D)
         O)
         CD
Figure 27.
            0.7
0.5
            0.3
                tc)
                0.9
             1.3
1.7
                            LO
                            £
                                         120
      C
      03
      O)
      s_
      O 80

      O

      (L>
      -Q 40
      £
      D
                                                           r^.72
0.9
1.3
1.7
                            Sinuosity   index
Regressions between sinuosity index and  (A) variance  of  depth
between cross sections (expressed as a percentage  of  mean  depth),
(B) variance of current velocity (expressed as  percentage  of the
mean), (C) mean drift sample weight, and (D) mean   number  of
organisms per drift sample.   (From Zimmer and Bachman 1976).
                                     62

-------
    In the broader perspective of the control of non-point pollution,  the
studies by.Ziflmer and Bachman (1976) and Gorman and Karr (1977)  have extremely
important implications.  Millions of dollars spent on preventing sediments
from entering streams will have minimal return value in improving the quality
of biota if present channelization practices continue to destroy  the  habitat
of stream organisms.  High water quality is necessary for fishable streams
but insufficient in itself if suitable habitat is not maintained.
Figure 28.
            1.0     2.0      3.0     4.0
               Habitat   diversity
Regression of fish species diversity on  habitat  diversity.
Habitat diversity is calculated from a combination  of  substrate,
depth, and current characteristics (From Gorman  and Karr 1977).
CONCLUSIONS

     It should be evident from this brief review of selected aspects of the
effects of land use on stream biota, that any attempt to improve the quality
of those resources must involve a broad based, multi-purpose program.   The
reasons for this can be best understood by examining a graphical representa-
tion of how species-are adapted to the physical environment (Figure 29a,b).
An organism's physical environment can be represented by a number of differ-
ent environmental gradients.  These gradients might represent factors such
as water temperature, water Velocity, oxygen concentration, size of substrate,
food type, depth of flow and others.  For ease of representation only three
environmental gradients are plotted.  Each organism can survive under a
limited range on each gradient, e.g. from 15°C to 25°C on the temperature
gradient.  With all of these gradients defined for a species, a hypervolume
of conditions under which the organism can survive will be defined.  For an
organism that is a generalist, the range of conditions will be large and in
turn the niche volume  is larger    (Figure 29b).  For a specialist, the
range of conditions is reduced and hence the volume is smaller.  In either
case a number of environmental conditions must be met before the organism
will be able to survive.  If only one of the environmental variables is suf-
ficiently altered, the organism will not be able to tolerate the change and
                                     63

-------
will be displaced, regardless of other environmental conditions.  On an
average, the magnitude of the change must be larger to displace a generalist
than a specialist.

     Therefore, eliminating only one of the factors detrimental to fish popu-
lations or other stream biota will result in very little improvement in these
resources.  For example, if sediments are prevented from entering streams in
agricultural watersheds but vegetation is removed from streambanks, elevated
temperatures will probably prevent any substantial increase in the quality of
the fish community.  Or if sediments and nutrients are prevented from enter-
ing streams but channelization is still performed, the lack of suitable habi-
tat will prevent any improvement in fisheries resources.  Before substantial
improvements in the biota of streams can be obtained, realistic management
of nearstream vegetation and stream morphology must be used in conjunction
with improved tillage practices.  Data presented here suggest that nearstream
vegetation may reduce sediment and attached nutrient inputs and temperature
fluctuations.  Further, a more natural stream morphology may reduce sediment
loads and provide suitable habitat for both fish and invertebrates.  Data
collected in forested watersheds (Hall and Lantz 1969) and in agricultural
areas (Karr and Gorman 1975), indicates that these factors acting in concert
result in a more productive, diverse, and stable stream biota.
        (a)
(b)
                  x
      X
                                     Y|
                        Y
Figure 29.  Range on three environmental gradients (X, Y, and Z) under which
            hypothetical generalist (a) and specialist (b) could exist.  Note
            that the range of conditions under which a generalist can survive
            is broader than that of a specialist.
                                     64

-------
                                 SECTION 10

                    RECREATIONAL BENEFITS OF GREENBELTS
     Although space does not permit an extensive evaluation of all possible
benefits to recreation from greenbelts, a few of these should at least be
mentioned.  Maintaining a strip of "natural" vegetation along the upper
reaches of streams in agricultural watersheds will provide a substantial
amount of habitat suitable for a number of terrestrial game species.  The
improved cover and forage would result in significant increases in pheasants,
dove, quail, rabbits, and many non-game species in these areas (Ferguson
et al. 1975).  The magnitude of increase could probably be estimated from
existing data or a minimal field survey.  On larger streams, greenbelts
could have a number of significant uses if their development was planned
properly, including fishing, hunting, sightseeing, picnicking, camping,
nature study and canoeing.  A study done by Fleener (1971) found that 96,500
trips and over 384,000 hours were spent in these activities in one year on
a greenbelt 30 m wide and 57 miles long on each side of the Platte River in
Missouri.  Furthermore he found 67% of the people travelled 25 miles or less
and 31% travelled only 26-50 miles to get there.  These data indicate the
potential of greenbelts for not only improving water quality and the stream
biota, but also for providing multiple-use recreational areas on larger
streams within close proximity to the majority of the people.  The latter
point is particularly relevant in the present time of rapidly increasing
energy costs.
                                     65

-------
                                 SECTION 11

                       OTHER ADVANTAGES OF GREENBELTS
     This report has addressed several major advantages which might accrue
from the use of greenbelts along streams.   A number of other potential advan-
tages have not been specifically addressed.   Some are mentioned briefly in
the following comments.

    1.   Suspended sediments can cause considerable damage including wear
        and tear on metal parts wherever machinery contacts flowing water.
        Reduction in sediment loads with greenbelts can reduce the magni-
        tude of this problem.

    2.   Changes in water temperature and sediment and nutrient loads can
        precipitate major shifts in algal communities.   These shifts can
        affect the taste and appearance of water.

    3.   High water quality and the associated rich biotic communities can
        reduce the problem of pathogens surviving in water supplies.

    4.   Frequency and cost of removing sediment from drainage ditches or
        streams could be decreased by filtering sediments from surface runoff
        before it reaches the stream bed.

    5.   Cost for removing the excess turbidity from water used for human
        consumption is reduced.

    6.   Flood damage, i.e., the cost of cleaning up the sediment deposited,
        is reduced.

    7.   Probability of flooding is reduced due to less clogging of the
        channel by sediments and because of the more controlled release of
        runoff.

    8.   Costs for storage space destroyed by silting in of reservoirs is
        diminished.
                                     66

-------
                                 SECTION 12

                         DISCUSSION AND CONCLUSIONS
     Any attempts to understand the dynamics of sediment transport and the
biological communities of streams must recognize the interrelationships
between water bodies and the land and atmosphere which surrounds them.
Rivers are functional parts of much larger landscape units, and they receive
most of their individual characteristics from the landscapes of their drain-
age basin (Fig. 30; Sioli 1975).  The channel bottom and suspended solids
loads are derived from the landscape and the form of the bed is at least
partially determined by the geological history of the surrounding area.  The
biota of a watershed also determines many of the characteristics of a stream,,
including the amount of organic matter and the molecular characteristics of
that organic matter (Janzen 1974).  The complex interplay of biological,
geological, chemical, and physical phenomena in both the terrestrial and
aquatic environments are of major importance in determining stream character-
istics (Hynes 1975, Likens and Bormann 1974).

     In an undisturbed watershed both the terrestrial and aquatic environments
are in an equilibrium, albeit a dynamic equilibrium.  In natural watersheds,
drastic fluctuations in water levels are uncommon (unless extremes in rainfall
occur).  Under most circumstances rainfall is absorbed by the land surface
and subsequently released from the soil to the stream over a long period
(Hewlett and Nutter 1970).  Furthermore, there is little surface runoff from
natural watersheds during periods of normal rainfall.  Nutrient cycles are
"tight" in natural watersheds with very few nutrients being lost to the
drainage waters (Likens and Bormann 1974).  Under most circumstances the
small amounts of nutrients lost from the terrestrial environment are readily
assimilated by the biotic communities of the stream.  Erosion in this equili-
brium state is minimal (Hobble and Likens 1973).

     When man arrives in the area the natural vegetation is removed and
instabilities in the terrestrial environment are an inevitable result,
especially if conservation practices are not employed.  These instabilities
have repercussions which affect the aquatic environment and disturb the equil-
ibrium in that section of the biogeocoenoses, to use the European terminology
(Sukachev  and Dylis 1964) or ecosystem, in the American literature (Tansley
1935).  Often, the response is to modify the stream channel to (1) improve
drainage of the land surface, and (2) reduce natural bank erosion and  other
bank instabilities stimulated by the modification of the land surface with
the advent of agriculture and urban development.  These channelization activ-
ities create more instabilities  in the aquatic environment.  The combined
effects of modifications on the land and restructuring of channels result in
                                      67

-------
   Climate
   Atmosphere
     Sun
                               TERRESTRIAL   ENVIRONMENT
                 /Parent
                  Material

                 Topography
Natural
Biological *-"
rCommunities.
                                                          Human
                                                           Influences
                                         .Nutrients,

                               Sedimentf___   \
                                              Energy
                                               Sources
                           RUNOFF _WAJE_R_
                                Quality
                                Quantity
                                Tinning
                      AQUATIC

                  WATER
                   ^Quantity*	
                   [ Physico -chemical*
                   I  Conditions

                  BOTTOM TYPE

                  'GRADIENT
                                                ENVIRONMENT
                                                         BIOTA
                                                         •Producers
                                                         /  Herbivores
                                                         I  Carnivores
                                                         Decomposers
Figure 30.
                                 QUALITY
                                   and
                                 QUANTITY
                                   of
                                 OUTFLOW
                                  WATER

General model of the primary factors governing the quality and
quantity of outflow water from -an  ecosystem (From Karr and
Gorman 1975).
                                    68

-------
disequilibria  in both  the  aquatic  and  terrestrial  areas.  Readily observed
signs of this  disequilibrium  include:
    1.  Rapid  runoff resulting in  drastic fluctuations in the water levels of
        streams.  These include floods during heavy rains and nearly stagnant
        conditions during  dry periods.

    2.  Large  volumes  of nutrients and sediments are lost from terrestrial
        ecosystems to  aquatic ecosystems, often over short time periods
        (Hobble and Likens 1973, Likens and Bormann 1974).

    3.  Increased fluctuations in  stream temperature (Likens 1970).

    4.  Increased streambank erosion as the stream attempts to re-establish
        its equilibrium by forming pools and riffles and meanders (Yang
        1971a,b).

    5.  Decreased diversity and stability in the biotic component of the
        aquatic ecoeystem  due to the less stable environment produced by a
        complex of sediment, nutrient, temperature and stream morphology
        effects (Margalef  1968, Odum 1969, Karr and Gorman 1975, Gorman and
        Karr 1977).

     If we compare two hypothetical, identical forested watersheds,  these
patterns can be illustrated in a study of disequilibrium in sediment loads
and biotic communities (Table 8).  Clearing of the land without conservation
activities and channel modifications produces considerable instabilities.  If
equilibrium is attained in the terrestrial component of the watershed but the
streams are channelized, increased sediment loads and major disequilibria in
the biotic and abiotic components  of the aquatic ecosystem will result.

     The natural tendency  is for the ecosystem (terrestrial and aquatic) to
return to equilibrium  by natural successional processes (Margalef 1968, Odum
1969).  However, the activities of man—agriculture and construction in the
terrestrial area, and  channel "maintenance" in the stream—tend to maintain
a disequilibrium.  Regrettably, this is the situation throughout much of the
U. S., especially in the heavy agricultural areas of the Midwest.

     Stabilization of  the  terrestrial environment within the requirements of
maintaining an intensive agricultural system is an essential component of
achieving water quality.  With careful management in fields (i.e., minimum
tillage, rotational practices, etc.) and along streams (i.e., greenbelts)
effects of the terrestrial disequilibrium can be minimized.

     However,  major efforts in this area will yield only partial success in
achieving water quality goals as long as the present approach to stream
management prevails.   That is, before major improvements in water quality and
stream biota can be realized, the  present philosophy of erosion control on
the land surface must  be combined  with a more reasoned approach on the
management of  stream channels (Table 11)-  A coordinated management program
should, first, reduce  sediment input and, second, manage the stream channel
to reduce USP.  The stream channel will be filled with sediment transported
from the land  surface  if USP control is emphasized first.


                                     69

-------
TABLE 11.  EFFECT OF VARYING "MANAGEMENT PRACTICES" ON EQUILIBRIA OF EQUIVALENT WATERSHEDS.  These
           are best estimates of relative effects for a variety of watershed conditions, including
	sources and amounts of sediment.	

                                EFFECT ON EQUILIBRIUM
                      TERRESTRIAL
                   AQUATIC
                                       ABIOTIC
                                    TRANSPORT  EROSION
                             BIOTIC
                                SUSPENDED
                                  SOLIDS

                                   LOAD
                                         SOURCE
                                           OF

                                        SEDIMENT
Natural Watershed
Clear Land for Row
None
None
None
None
                            Moderate-
Low
Crop Agriculture
Channelize Stream
Clear Land and
Channelize Stream
Best Land Surface
Management with
Channelization
Best Land Surface
and "Natural" Channel
Major
None
Major
Low
Low
Major
Major
Major
Major
Low
Minor*
Major
Major
Major
Low
Major
Major
Major
Major
Low
Medium
High
Very
High
Medium
High
Low-
Medium
Land Surface
Channel Banks
Equilibrium Between
Land Surface and
Channel Banks
Channel Banks
Equilibrium Between
Land and Channel
*Will increase if hydrograph peaks (floods) more severe.

-------
     In summary, before the admirable objectives of the Federal Water Pollu-
tion Control Act of 1972  (Public Law 92-500)—fishable and swimmable water
by 1983—can be attained, it is essential that an equilibrium concept be
implemented in the control of non-point source pollution.  Furthermore, this
philosophy must involve equilibria in both terrestrial and aquatic ecosystems.

     We believe this report has demonstrated the value of a perspective
which emphasizes the link between terrestrial and aquatic environments and
the dynamics of stream behavior.  The suggestion that near-stream vegetation
along with soil conservation practices and "natural" stream morphology can
produce substantial improvements in water quality is supported for areas of
intensive agriculture.  Furthermore, this management alternative should
enhance the quality of fishery resources and provide a variety of recreation-
al benefits.  The magnitude of these benefits must, of course, be weighed
against the costs  (Table  12).  Perhaps the techniques of decision theory
recently used in studying buffer strips along streams in the field of for-
estry (Sadler 1970, Gillick and Scott 1975) should be applied.  This approach
will require substantial  new research efforts on a number of questions
raised in this paper.  These problems must be dealt with at the individual
drainage and watershed levels before the environmental and economic value
of these management strategies can be fully evaluated.
                                      71

-------
TABLE 12.  COSTS AND BENEFITS OF MORE EFFECTIVE MANAGEMENT OF NEARSTREAM VEGETATION AND CHANNEL MORPHOLOGY.
           This is not meant to be a comprehensive list.  Furthermore, the magnitude of the suggested costs
	and benefits have not been adequately evaluted.	
                        Costs
                     Benefits
     1.  Land taken out of production to maintain
         a vegetative filter.

     2.  Land taken out of production to allow
         meandering.

     3.  Reduced drainage rates.

     4.  Maintenance of greenbelts.

     5.  Reservoir areas for various pests.

     6.  Need for management of recreational
         areas.
1.  Reduced sediment, nutrient, and pesticide inputs
    into streams.

2.  Increased shading resulting in decreased water
    temperature; this will alleviate problems
    associated with release of nutrients from sedi-
    ments, oxygen carrying capacity, temperature and
    nutrients in downstream reservoirs, and fewer
    algal blooms.

3.  Improved habitat for fisheries and terrestrial
    wildlife.

4.  Increased recreational opportunities.

5.  Decreased cost of channel construction and
    maintenance activities since nature will pro-
    vide the vegetation along streambanks via natu-
    ral succession and the stream itself will
    initiate meandering and pool-riffle formation.

6.  Reduced downstream flooding.

7.  May allow more intensive agriculture with
    reduced effects on the aquatic ecosystem when
    best land management practices (i.e. minimum
    tillage) are not feasible.

-------
                                  REFERENCES

Agricultural Research  Service.   1975,  1976.  Control of water pollution from
     cropland. Volume ,1. A manual for  guideline development. Volume II.  An
     overview. ARS-H-5-1, ARS-H-5-2. Agricultural Research Service and U.S.
     Environmental Protection Agency,  Washington, B.C. Ill pp., 187 pp.

Bachman, R. W.   1958.   The ecology  of  four north Idaho trout streams with
     reference to  the  influence  of  forest road construction.  M. S. thesis,
     University  of Idaho, Moscow, Idaho.  97 pp.

Barfield, B. J., D. T.  Y. Kayo,  and E. W. Tollner.  1975.  Analysis of sedi-
     ment filtering action of grassed  media.  Res. Rep. No. 90, University of
     Kentucky Water Resources Research Institute, Lexington, Kentucky,  50 pp.

Bayless, J. and  N. B.  Smith.  1967.  The effects of channelization upon the
     fish populations  of lotic waters  in eastern North Carolina.  Proc. Ann.
     Conf.  S. E. Assoc.  Game and  Fish Comm. 18:230-238.

Becker, B. C. and T. R. Mills.   1972.  Guidelines for erosion and sediment
     control planning  and implementation.  EPA-R2-72-0105. U.S. Environmental
     Protection  Agency, Washington, D. C.  228 pp.

Blench, T.  1972.  Morphometric  changes.  In;  River Ecology and Man, R. T.
     Oglesby, C. A. Carlson, and J. A. McCann, eds. Academic Press.  N. Y.
     pp. 287-308.

Bormann, F. H.,  G. E.  Likens and J. S. Eaton.  1969.  Biotic regulation of
     particulate and solution losses from a forest ecosystem.  Bioscience
     19:600-610.

Brett, J. R.  1970.  Fish - the  energy cost of living.  Conference on Marine
     Aquaculture.       William J. McNeill, ed.  Oregon State University
     Marine Science Center, Corvallis, Oregon, pp. 37-52.

Brett, J. R.  1964.  The respiratory metabolism and swimming performance of
     young sockeye salmon.  J. Fish Res. Bd. Canada.  21:1183-1226.

Bridge, J.  S.  1975.   Computer simulation of sedimentation in meandering
     streams.  Sedimentology.  22:3-43.

Broderson,  J. M.   1973.  Sizing  buffer strips to maintain water quality.
     M. S.  Thesis, University of Washington, Seattle, Washington.  84 pp.
                                      73

-------
Brown, C.  B.   1943.   The control of reservoir silting.  Miscellaneous Publi-
     cation No.  521, U. S.  Department of Agriculture, Washington D. C.
     166 pp.

Brown, G.  W.   1969.   Predicting temperatures of small streams.  Water
     Resources Res.   5:68-75.

Brown, G.  W.  and J.  T.  Krygier.  1970.  Effects of clear-cutting on stream
     temperature.   Water Resour. Res. 6:1131-1139.

Brown, G.  W.  and J.  R.  Brazier.  1972.  Controlling thermal pollution in
     small streams.   EPA-R2-72-083, U. S. Environmental Protection Agency,
     Washington, D.  C., 64 pp.

Bush, R. M.,  E.  B.  Welch, and B. W. Mar.  1974.  Potential effects of thermal
     discharges on aquatic systems.  Env. Science and Tech.  8:561-568.

Butler, R. M., E.  A. Meyers, M. H.  Walter, and J. V.  Husted.  1974.  Nutrient
     reduction in wastewater by grass filtration.  Paper No. 74-4024.  Pre-
     sented at the 1974 winter meeting,  ASAE, Stillwater, Oklahoma.
     12 pp. mimeo.

Campbell,  K.  L. , S.  Kumar,  and H.  P. Johnson.  1972.   Stream straightening
     effects on flood runoff characteristics.  Trans. Amer. Soc. Agr. Eng.
     15:94-98.

Chapman, D. W.,  and R.  L. Demory.   1963.  Seasonal changes in the food
     ingested by aquatic insect larvae and nymphs in two Oregon streams.
     Ecology 44:140-146.

Chow, V. T.  1959.   Open Channel Hydraulics.  McGraw-Hill Book Co. New York.
     680 pp.

Congdon, J. C.  1971.  Fish populations  of channelized and unchannelized
     sections of the Chariton River, Missouri.  In:  Stream Channelization:
     A Symposium.   E. Schneberger and J. L. Funk, eds. Spec. Publ. 2,
     North Central Division, American Fisheries Society, Omaha, Nebraska.
     pp. 52-62.

Cook, H. L. and F.  B. Campbell.  1939.  Characteristics of some meadow strip
     vegetation.  Agr.  Eng. 20:345-348.

Cooper, A. C.   1956.  A study of the Horsefly River and the effects of placer
     mining operations on sockeye spawning grounds.  Publ. No. 3, Inter.
     Pac.  Salmon Fisheries Comm, New Westminster, British Columbia.  58pp.

Cordone, A. J. and D. W. Kelley.  1961.   The influences of inorganic sediment
     on the aquatic life of streams.  Calif. Fish and Game.  47:189-228.
                                      74

-------
Cummins, K. W.  1975.  The use of macroinvertebrate benthos in evaluating
     environmental damage.  In:  Proceedings of Nuclear Regulatory Commission
     Workshop on the Biological Significance of Environmental Impacts.  R. K.
     Sharma, J. D. Buffington, and J. T. McFadden, eds., NR-CONF-002.  U. S.
     Nuclear Regulatory Commission, Washington, D. C.  pp. 139-149.

Curry, R. R.  1975.  Limiting downstream effects of watershed manipulation:
     A proposed NEPA guideline.  In:  Proceedings of Nuclear Regulatory Com-
     mission Workshop on  the Biological Significance of Environmental
     Impacts, R. K. Sharma, J. D. Buffington, and J. T. McFadden, eds.,
     NR-CONF-002, U. S. Nuclear Regulatory Commission, Washington, D. C.
     pp. 251-262.

Darnell, R. M.  1976.  Impacts of construction activities in wetlands of
     the United States.   EPA-600/3-76-045, U. S. Environmental Protection
     Agency, Corvallis, Oregon.  392 pp.

Davis, R. B.  1974.  Effects of burrowing tubificid worms on the exchange of
     phosphorus between lake sediments and overlying water.  Progr. Rep.
     U. S. Office Water Resour. Res., Land Water Res. Center, University of
     Maine, Orono. Proj.  A-022-ME.  7 pp.

Edwards, W. M., F. W. Chichester, and L. L. Harrold.  1971.  Management of
     barnlot runoff to improve downstream water quality.  In:  Livestock
     water management and pollution abatement.  Publ. PROC-271, Amer. Soc.
     Agr. Eng., St. Joseph, Michigan,  pp. 48-50.

Ellis, M. M.  1931.  Some factors affecting the replacement of the commercial
     fresh-water mussles.  Fish. Circ. 5, Department of Commerce, Bureau of
     Commercial Fisheries, Washington, D. C. , 18 pp.

Emerson, J. W.  1971.  Channelization; a case study.  Science 173:325-326.

Ferguson, H. L., R. W. Ellis, and J. B. Whelan.  1975.  Effects of stream
     channelization on avian diversity and density in Piedmont Virginia.
     Paper presented at 1975 annual meeting of the S. E. Assoc. Game Fish
     Commissioners, St. Louis, MO.  15 pp. mimeo.

Fleener, G. G.  1971.  Recreational use of the Platte River Missouri.  In:
     Stream Channelization:  A Symposium.  E. Schneberger and J. L. Funk,
     eds. Spec. Publ. 2,  North Central Division, American Fisheries Society,
     Omaha, Nebraska,  pp. 63-78.

Friedkin, J. F.  1945.  Meandering of alluvial rivers.  U. S. Waterways Exp.
     Sta., Vicksburg, Mississippi.  40 pp.

Gillick, T. and B. D. Scott.  1975.  Buffer strips and the protection of
     fishery resources:   an economic analysis.  Rep. No. 32, Department of
     Natural Resources, State of Washington, Olympia, Washington.  30 pp.
                                      75

-------
Glymph, L. M. and N. H. Holtan.  1969.  Land treatment in agricultural water-
     shed hydrology research.  In:  Effect of Watershed Changes on Streamflow,
     W. L. Moore and E. W. Morgan, eds. University of Texas Press, Austin,
     Texas pp. 44-68.

Gorman, 0. T. and J. R. Karr.  1977-  Diversity and stability in the fish
     communities of some Indiana and Panama streams.  Submitted to Ecology.

Gosz, J. R., G. E. Likens, and F. H. Bormann.  1972.  Nutrient content of
     litter fall on the Hubbard Brook Experimental Forest, New Hampshire.
     Ecology 53:769-784.

Graf, W. H.  1971.  Hydraulics of sediment transport.  McGraw-Hill.  New
     York.  513 pp.

Gray, J. R. and J. M. Eddington.  1969.  Effect of woodland clearance on
     stream temperature.  Jour. Fish. Res. Bd. Can.  26:399-403.

Great Lakes Basin Commission.  1975.  Preliminary Report.  Erosion and Sedi-
     mentation Technical Paper.  MRB Series No. 11, Maumee River Basin Level
     B Study, Great Lakes Basin Commission.  Ill pp.

Greene, G. F.  1950.  Land use and trout streams. J.Soil Water Conserv.
     5:125-126.

Hall, C. A.  1972.  Migration and metabolism in a temperate stream ecosystem.
     Ecology 53:584-604.

Hall, J. D. and R. L. Lantz.  1969.  Effects of logging on the habitat of
     Coho salmon and cutthroat trout in coastal streams.  In:  Symp. on
     Salmon and Trout in Streams, T. G. Northcote,  ed., University of British
     Columbia, Vancouver, British Columbia.  pp. 355-375.

Hamilton, R. E.  1969.  Tolerance of grasses and legumes to flooding.
     Agronomy No. 7-  Technical Note.  USDA Soil Conservation Service.
     Indianapolis, Indiana.

Hansen, D. R.  1971.  Stream channelization effects of fishes and bottom
     fauna in the Little Sioux River, Iowa.  In:  Stream Channelization:  A
     Symposium, E. Schneberger and J. L. Funk, eds., Spec. Publ. 2, North
     Central Division, American Fisheries Society,  Omaha, Nebraska.
     pp. 29-51.

Hanway, J. J. and J. M. Laflen.  1974.  Plant nutrient losses from tile-
     outlet terraces.  J. Env. Qual. 3:351-356.

Harman, W. N.  1972.  Benthic substrates:  their effect on fresh water
     mollusca.  Ecology 53:271-277.

Harrison, C.  W.  1923.  Planting eyed salmon and trout eggs.  Trans. Amer.
     Fish Soc. 53:191-200.
                                     76

-------
Haupt, H. F.  1959.  Road and slope characteristics affecting sediment move-
     ment from  logging roads. J. Forest. 57:329-332.

Haupt, H. F. and W. J. Kidd, Jr.  1965.  Good logging practices reduce sedi-
     mentation  in central Idaho.  Forest. 63:664-670.

Heimstra, N. W., D. K. Damkot, and N. G. Benson.  1969.  Some effects of
     silt turbidity on behavior of juvenile largemouth bass and green sun-
     fish.  Tech. Pap. 20, U. S. Fish and Wildlife Service, Washington, D. C.
     9 pp.

Henegar, D. L.  and K. W. Harmon.  1971.  A review of references to channeli-
     zation and its environmental impact.  In:  Stream Channelization:  A
     Symposium, E. Schneberger and J. L. Funk, eds., Spec. Pub. 2, North
     Central Division, American Fisheries Society, Omaha, Nebraska.
     pp. 79-83.

Hewlett, J. D.  and W. L. Nutter, 1970.  The varying source area for stream-
     flow from  upland basins.  Proc. Symp.    Interdisciplinary Aspects of
     Watershed  Management.  Amer. Soc. Civ. Eng.  New York.  pp. 65-83.

Hobble, J. E. and G. E. Likens.  1973.  The output of phosphorus, dissolved
     organic carbon, and fine particulate carbon from Hubbard Brook water-
     sheds.  Limnol.  Oceanogr. 18:734-742.

Hornbeck, J. W., R. S. Pierce, and C. A. Federer.  1970.  Streamflow changes
     after forest clearing in New England.  Water Resources Research
     6:1124-1132.

Hussey, K. M. and H. L. Zimmerman.  1953.  Rate of meander development as
     exhibited  by two streams in Story County, Iowa.  Proc. Iowa Acad. Sci.
     60:390-392.

Hynes, H. B. N.  1963.  Imported organic matter and secondary productivity in
     streams.   Proc. 16th Internat. Cong. Zool. 4:324-329.

Hynes, H. B. N.  1970.  The ecology of running waters.  University of
     Toronto Press, Toronto, Ontario. 555 pp.

Hynes, H. B. N.  1975.  The stream and its valley.  Verh. Internat. Verein
     Limnol. 19:1-15.

Institute of Water Research, Michigan State University.  1976.  Utilization
     of Natural ecosystems for wastewater renovation.  Comprehensive Project
     No. Y005065, East Lansing, Michigan.  195pp.

Jackson, G.  R.  1975.  A depositional model of point bars in the lower
     Wabash River.  Ph. D. Thesis, University of Illinois, Urbana, Illinois.
     262 pp.

Janzen, D. H.   1974.  Tropical blackwater rivers, animals, and mass fruiting
     by the Dipterocarpaceae.  Biotropica 6:69-103.


                                     77

-------
Johnson, F. W.  1953.  Forests and trout.  J. Forest. 51:551-554.

Jones, J. R.,  B. P. Borofka, and R. W. Bachman.  1975.  Factors affecting
     nutrient loads in some Iowa streams.  Water Research 10:117-122.

Kardos, L. T., W. E. Sopper, E. A. Myers, R. R. Parizek, and J. B. Nesbitt.
     1974.  Renovation of secondary effluent for reuse as a water resource.
     EPA 660/2-74-016, U. S. Environmental Protection Agency, 495 pp.

Karr, J. R. and 0. T. Gorman.  1975.  Effects of land treatment on the
     aquatic environment.  In:  Non-Point Source Pollution Seminar, EPA
     905/9-75-007, U. S. Environmental Protection Agency, Chicago, Illinois.
     pp. 120-150.

Kayo, D. T. Y., B. J. Barfield, and A. E. Lyons.  1975.  On site sediment
     filtration using grass strips.  In:  Nat. Symp. on Urban Hydrology and
     Sediment Control.  Kayo, D. T. Y.   and R. W. DeVore, eds., University
     of Kentucky, Lexington, Kentucky,  pp. 73-82.

King, D. L. and R. C. Ball.  1967.  Comparative energetics of a polluted
     stream.  Limnol. Oceanogr. 12:27-33.

Kreis, R. D. ,  M. R. Scalf, and J. F. McNabb.  1972.  Characteristics of rain-
     fall runoff from a beef cattle feedlot.  EPA-R2-72-061, U. S. Environ-
     mental Protection Agency, Corvallis, Oregon.  43 pp.

Lane, E. W. and W. M. Borland.  1954.  River bed scour during floods.  Amer.
     Soc. Civil Eng. Trans. 119:1069-1079.

Langlois, T. H.  1941.  Two processes operating for the reduction in abundance
     or  elimination of fish species from certain types of water areas.
     Trans. N. Amer. Wildl. Conf. 6:189-201.

Law, J. P., R. E. Thomas and L. H. Myers.  1970.  Cannery wastewater treat-
     ment by high-rate spray on grassland.  J. Water Poll. Cont. Fed.
     42:1621-1631.

Lehman, G. S.  and L. G. Wilson.  1971.  Trace element removal from sewage
     effluent  soil filtration.  Water Resources Research.  7:90-99.

Leopold, L. B. and M. G. Wolman.  1960.  River meanders.  Bull. Geol. Soc.
     Amer. 71:769-796.
«
Leopold, L. B., M. G. Wolman and J. P- Miller.  1964.  Fluvial processes in
     geomorphology.  W. H. Freeman  and Co.  San Francisco.  522 pp.

Likens, G. E.   1970.  Effects of deforestation on water quality.  Proc.
     Sympos.  Interdisciplinary Aspects of Watershed Management.  Amer. Soc.
     Civ. Eng. New York.  pp. 133-140

Likens, G. E.  and F.^H. Bormann.  1974.  Linkages between terrestrial and
     aquatic systems.  Bioscience 24:447-456.

                                     78

-------
Lotrich, V. A.  1973.  Growth, production and community composition of
     fishes inhabiting a first, second and third order stream of eastern
     Kentucky.  Ecol. Monogr. 43:377-397.

Mannering, J. V. and C. B. Johnson.  1974.  Report on simulated rainfall
     phase.  Appendix No. 9.  First Annual Report, Black Creek Study Project,
     Allen County, Indiana, Soil and Water Conservation District, Fort Wayne,
     Indiana.

Margalef, R.  1968.  Perspectives  in ecological theory.  Univ. of Chicago
     Press.  Chicago.  Ill pp.

Mather, J. R.  (ed.).  1969.  An evaluation of cannery waste disposal by over-
     land flow  spray irrigation.   Publications in Climatology.  22(2):l-76.

Menzel, B. W. and H. L. Fierstine.  1976.  A Study of the Effects of Stream
     Channelization and Bank Stabilization on Warmwater Sport Fish in Iowa:
     Subproject No. 5.  Effects of Long-Reach Stream Channelization on Dis-
     tribution  and Abundance of Fishes.  FWS/OBS-76-15, U. S. Fish and Wild-
     life Service, Washington, D.  C. 75 pp.

Mildner, W.  1976.  Streambank erosion in Black Creek Watershed, Indiana.
     Prep, by the USDA Soil Conserv. Service.  An assignment of the U. S.
     Task C work group of the International Reference Group on Great Lakes
     Pollution  from Land Use Activities.  5 pp. +2 appendices, Mimeo.

Minshall, G. W.  1967-  Role of allochthonous detritus in the trophic struc-
     ture of a  woodland spring brook community.  Ecology 48:139-149.

Minshall, G. W.  1968.  Community  dynamics in a woodland spring brook.
     Hydrobiologia 32:305-339.

Monke, E. J., D. B. Beasley, and A. B. Bottcher.  1975a.  Sediment contribu-
     tions to the Maumee River.  In:  Non-point pollution seminar EPA-905/
     9-75-007,  U. S. Environmental Protection Agency, Chicago, Illinois.
     pp. 71-85.

Monke, E. J., H. J. Marelli, L. D. Meyer, and J. F. DeLong.  1975b.  Physical-
     chemical composition of eroded soil.  Paper 75-2584, presented at the
     1975 winter meeting Amer. Soc. Agr. Eng., Chicago, Illinois.

Morisawa, M.  1968.  Streams:  their dynamics and morphology.  McGraw-Hill
     Book Co. New York.  173 pp.

Morris, L. A.,  R. N. Langmeier, T. R. Russell, and A. Witt.  1968.  Effects
     of main stem impoundments and channelization upon the limnology of the
     Missouri River, Nebraska.  Trans. Amer. Fish. Soc. 97:380-388.

Moyle, P- B.  1976.  Some effects  of channelization on the fishes and
     invertebrates of Rush Creek,  Modoc County, California.  Calif. Fish  and
     Game 62:119-186.
                                      79

-------
Nelson, D. W., L. E. Sommers and A. D. Bottcher.  1976.  Nutrient contribu-
     tions to the Maumee River.  In:  Best Management Practices for Non Point
     Source Pollution Control Seminar, EPA 905/9-76-005, U. S. Environmental
     Protection Agency, Chicago, Illinois,  pp. 141-154.

Odum, E. P.  1969.  The strategy of ecosystem development.  Science
     164:262-270.

Ohlander, C. A.  1976.  Defining the sediment trapping characteristics of a
     vegetative buffer.  Special Case:  Road erosion.  NTIS PB-245100 Proc.
     Fed. Interagency Sedimentation Conf.j  Denver, Colorado,  pp. 77-82.

Palmer, V. J.  1946.  Retardance coefficients for low flow in channels lined
     with vegetation.  Trans. Amer. Geophysical Union 27:187-197-

Parsons, 0. A.  1963.  Vegetative control of streambank erosion.  In:  Proc.
     Fed. Interagency Sedimentation Conf. Misc. Publ. 970, U. S. Department
     of Agriculture, Washington, D. C.  pp. 130-136.

Phaup, J. D. and J. Gannon.  1967.  Ecology of Sphaerotilus in an experi-
     mental outdoor channel.  Water Research.  1:523-541.

Piest, R. F.  1963.  The role of the large storm as a sediment contributor.
     Proc. Fed. Interagency Sedimentation Conf., Misc. Publ. 970, U. S.
     Department of Agriculture, Washington, D. C.  pp. 98-108.

Popkin, B. P.  1973.  Evaluation of a mixed grass cover and native soil
     filter for treatment of Tucson urban runoff.  Completion Report Water
     Resources Center.  University of Arizona.  Grant B-OZ3-ARIZ.

Rausch, 0. L. and H. G. Heinemann.  1975.  Controlling reservoir trap effi-
     ciency.  Trans. Amer. Soc. Agr. Eng.  18:1105-1113.

Ree, W. 0.  1949.  Hydraulic characteristics of vegetation for vegetated
     waterways.  Agr. Eng. 30:184-189.

Ree, W. 0. and V. J. Palmer.  1949.  Flow of water in channels protected by
     vegetative linings.  Tech. Bull. 967, U. S. Department of Agriculture.
     115 pp.

Regier, H. A. and W. L. Hartman.  1973.  Lake Erie's Fish Community:  150
     years of cultural stresses.  Science 180:1248-1255.

Ross, H. H.  1963.  Stream communities and terrestrial biomes.  Arch. Hydro-
     biol.  59:235-242.

Ryden, J. C., H. K. Syers, and R. F. Harris.  1973.  Phosphorus in runoff
     and streams.  Adv. in Agron. 25:1-41.

Ryder, R. A. and L. Johnson.  1972.  The future of salmonid communities  in
     North American Oligotrophic Lakes. J. Fish. Res. Bd. Canada. 29:941-949.
                                     80

-------
Sadler, R. R.  1970.  Buffer strips:  a possible application of decision
     theory.  BLM Technical Filing code 5000-6512, U. S. Department of
     Interior, Portland, Oregon.  11 pp. mimeo.

Scarseth,  G. D. and W. V. Chandler.  1938.  Losses of phosphate from a light
     textured soil in Alabama and its relation to some aspects of soil con-
     servation.  Amer. Soc. Agron. J.  30:361-374.

Shapavalov, L.  1937.  Experiments in hatching steelhead eggs in gravel.
     Calif. Fish and Game 23:208-214.

Shapavalov, L. and W. Berrian.  1940.  An experiment in hatching silver
     salmon (Onchorynchus kisutch) eggs in gravel.  Trans. Amer. Fish Soc.
     69:135-140.

Shapavalov, L. and A. Taft.  1954.  The life histories of  the steelhead rain-
     bow trout (Salmo gairdnerii gairdnerii) and silver salmon (Onchorynchus
     kisutch).  Calif. Dept. Fish and Game, Fish Bull. 98.  375 pp.

Sievers, D. M., G. B. Garner and E. E. Pickett.  1975.  A lagoon-grass ter-
     race system to treat swine waste.  In:  International Symp. of Livestock
     Waste Manag.  Pub. PROC-275, Am. Soc. Agr. Eng., Urbana, Illinois.
     pp. 541-548.

Sioli, H.  1975.  Tropical rivers as expressions of their terrestrial
     environments.  In:  Tropical Ecological Systems:  Trends in Terrestrial
     and Aquatic Research, F. B. Golley and E. Medina, eds.,  Springer-Verlag.
     N. Y., pp. 275-288.

Smith, L. L.  and J. B. Moyle. 1944.  A biological survey and fishery manage-
     ment plan for the streams of the Lake Superior north shore watershed.
     Tech. Bull. 1.  Minn. Dept. of Cons., Div. of Game and Fish, St. Paul,
     Minnesota.  228 pp.
                   *
Smith, S. H.  1972.  The future of salmonid communities in the Laurentian
     Great Lakes.  J. Fish. Res. Bd. Canada 29:951-957.

Snyder, G. R.  1959.  Evaluation of cutthroat trout reproduction in Trappers
     Lake Inlet. Colo. Coop. Fish. Res. Unit, Quart. Rep. 5:12-52.

Sommers, L. E., D. W. Nelson and D. B. Kaminsky.  1975a.  Nutrient contribu-
     tions to the Maumee River.  In:  Non-Point Pollution Seminar, EPA-905/
     9-75-007, U. S. Environmental Protection Agency, Chicago, Illinois.
     pp. 105-119.

Sommers, L. E., D. W. Nelson, E. J. Monke, D. Beasley, A. D. Bottcher, and
     D. Kaminsky.  1975b.  Nutrients are more concentrated in subsurface
     flow.  EPA-905/9-75-006.  U. S. Environmental Protection Agency,
     Chicago, Illinois,  pp. 63-154.
                                     81

-------
Sopper, W. E.  and L.  T. Kardos Ceds.)  1973.  Recycling treated municipal
     wastewater and sludge through forest and cropland.  The Pennsylvania
     State University Press.  University Park, Pennsylvania.  480 pp.

Sprules, W. M.  1947.  An ecological investigation of stream insects in
     Algonquin Park,  Ontario.  Univ. Toronto Studies, Biol. Series, No. 56.,
     Publ. Ontario Fish. Res. Lab., No. 69.  Toronto, Ontario.  81 pp.

Stall, J. B.  1964.  Sediment movement and deposition patterns in Illinois
     impounding reservoirs.   J. Amer, Water Works Assn. 50:755-766.

Stall, J. B. and T. C. Yang.  1972.  Hydraulic geometry and low stream flow
     regimen.   Res. Rep. 54, Water Resources Center, University of Illinois,
     Urbana, Illinois.  31 pp.

Stauffer, T. R., K. L. Dickson, J. Cairns and D. S. Cherry.  1976.  The
     potential and realized influences of temperature on the distribution of
     fishes in the New River, Glen Lyn, Virginia.  Wildlife Monographs No.
     50.  40 pp.

Stotlenberg, N. L. and J. L. White.  1953.  Selective loss of plant nutrients
     by erosion.  Soil Sci.  Soc.  Amer. Proc. 17:406-410.

Straub, L. G.   1942.   Mechanics of rivers.  In:  Meinzer,0. E. ed. Hydrology.
     Daber Public. Inc. N. Y.  pp. 614-636.

Sukachev, V. and N. Dylis.  1964.  Fundamentals of Forest Biogeocoenology.
     Oliver and Boyd.  London.  672 pp.  (Transl. from Russian.)

Swanson, N. P., C. L. Linderman,  and L. N. Mielke.  1975.   Direct land dis-
     posal of feedlot runoff.  In:  International Symp. on Livestock Wastes.
     Manag. Publ. Proc-275,  Am. Soc. Agr. Eng., Urbana, Illinois.
     pp. 255-257.

Swenson, W. A., L. T. Brooke, and P. W. Devore.  1976.   Effects of red clay
     turbidity on the aquatic environment.  In:  Best Management Practices
     for Non-Point Source Pollution Control Seminar, EPA 905/9-76-005, U. S.
     Environmental Protection Agency, Chicago, Illinois.  pp. 207-230.

Swift, L. W. and J. E. Messer.  1971.  Forest cuttings raise temperatures of
     small streams in the Appalachians.J.Soil Water Conserv. 26:111-116.

Tansley, A. G.  1935.  The use and abuse of vegetational concepts and terms.
     Ecology 16:284-307.

Taylor, T. U.   1930.   Silting of Reservoirs.  Bull. 3025,  University of
     Texas, Austin, Texas.  170 pp.

Tebo, L. B.  1955.  Effects of siltation, resulting from improper logging,
     on the bottom fauna of a small trout stream in the southern Appalachians.
     Prog. Fish. Cult. 17:64-70.
                                     82

-------
Trollner, E. W., B. J. Barfield and T. Y. Kayo.  1973.  Modeling the sus-
     pended sediment filtration capacity of simulated vegetation.  Paper
     No. 73-2553, presented at the 1973 Winter meeting of the Am. Soc. Agr.
     Eng., Chicago, Illinois.  9 pp. mimeo.

Trollner, E. W., B. J. Barfield, and C. T. Hoan.  1975.  Vegetation as a
     sediment filter.  National Symposium on Urban Hydrology and Sediment
     Control.  University of Kentucky, Lexington, Kentucky,  pp. 61-64.

Trollner, E. W., B. J. Barfield, C. T. Hoan, and T. Y. Kayo. 1975.  Sus-
     pended sediment filtration capacity of simulated vegetation.  Trans.
     Am. Soc. Agr. Eng. 19:678-682.

Trimble, G. R. and R. S. Sartz.  1957.  How far from a stream should a
     logging road be located?J. Forest. 55:339-341.

Wallen, E. I.  1951.  The direct effect of turbidity on fishes.  Biol.
     Series No. 2, Okla Agric. and Mech. Col. Arts and Sciences Studies.
     Stillwater, Oklahoma. 48:1-27.

Warren, C. E.  1971.  Biology and water pollution control.  W. B. Saunders
     Co. Philadelphia.  434 pp.

Wertz, J. B.  1963.  Mechanisms of erosion and deposition along channelways.
     J. Ariz. Acad. Sci. 2:146-163.

Wharton, C. H.  1970.  The southern river swamp:  a multiple use environment.
     Bureau of Business and Economic Research, School of Business Administra-
     tion.  Georgia State University, Atlanta, Georgia.  48 pp.

Wheaton, R. Z.  1975.  Streambank stabilization.  In:  Non-point Source
     Pollution Seminar, EPA-905/9-75-007.U. S. Environmental Protection
     Agency, Chicago, Illinois,  pp. 86-98.

Wilson, J. N.  1957.  Effects of turbidity and silt on aquatic life.  In:
     Biological Problems in Water Pollution, C. M. Tarzwell, ed., U. S.
     Department of Health, Education, and Welfare, Cincinnati, Ohio.
     pp. 235-239.

Wilson, L. G.  1967.  Sediment removal from flood water by grass filtration.
     Trans.  Amer. Soc. Agr. Eng. 10:35-37-

Wilson, L. G. and G. S. Lehman.  1966.  Grass filtration of sewage  effluent
     for quality improvement prior to artificial recharge.  Presented at  the
     1966 Winter meeting, Amer. Soc. Agr. Eng., Chicago, Illinois.  Cited
     in Butler et al. 1974.

Wischmeier, W. H.  1962.  Storms and soil conservation.  J. Soil Water
     Conserv. 17:55-59.
                                      83

-------
Wischmeier, W. H. and D. D. Smith.  1965.  Predicting rainfall-erosion
     losses from cropland east of the Rocky Mountains.  Agric.  Handbook
     No. 282, U. S. Department of Agriculture, Washington, D. C.  47 pp.

Woodall, W. R. and J. B. Wallace.  1972.  The benthic fauna in four small
     southern Appalachian streams.  Amer. Midi. Nat. 88:393-407.

Yang, C. T.  1971a.  On river meanders.  J. Hydrology 13:231-253.

Yang, C. T.  1971b.  Formation of riffles and pools.  Water Resources
     Research 7:1567-1574.

Yang, C. T.  1972.  "Unit stream power and sediment transport.  Am. Soc. Civ.
     Eng., Jour, of Hydraulics Division 98 (HY 10):1805-1826.

Yang, C. T. and J. B. Stall.  1974.   Unit stream power for sediment transport
     in alluvial rivers.  Research Rep. 88, Water Resources Center, Univer-
     sity of Illinois, Urbana,  Illinois.  38 pp.

Young, R. A. and C. K. Muchler.   1969.  Effect of slope shape on erosion and
     runoff.  Trans. Amer. Soc.  Agr.  Eng. 12:231-239.

Younger, V. B., T. E. Williams and L.  R. Green.  1976.   Ecological and physio-
     logical implications of greenbelt irrigation.  Contrib. No. 157,  Water
     Resources Center, University of  California, Davis, California.  104 pp.

Zimmer, D. W. and R. W. Bachman.   1976.  A Study of the Effects of Stream
     Channelization and Bank Stabilization on Warm Water Sport Fish in Iowa:
     Subproject No. 4.  The effects of Long Reach Channelization on Habitat
     and Invertebrate Drift in Some Iowa Streams.  FWS/OBS-76-14, U.  S. Fish
     and Wildlife Service, Washington, D. C.   87 pp.
                                     84

-------
                                  GLOSSARY

Active metabolism—metabolic requirements of an organism  undergoing normal
   activity, such as swimming, feeding, etc.

Allochthonous—Material, generally an energy source such as leaves or insects,
   which falls into the water from the nearby terrestrial environment.

Angular canopy density—a measure of the ability of vegetation along a chan-
   nel to shade the water in that stream.

Attached nutrients—elemental materials required by organisms (N, P, K, etc.)
   that are adsorbed to the surface of sediments.

Autochthonous-material.,  generally an energy source such as algae or insects,
   which are produced in the stream.

Autotroph—organisms which obtain their nourishment by oxidizing simple
   chemical elements such as iron, sulfur, or via photosynthesis.

Benthos—bottom-dwelling organisms in stream, lake, and marine environments.

Biological oxygen demand—the oxygen required to reduce the organic material
   (e.g. in a water sample).  Measure of amount of organic pollution.

Biota—the plant and animal life of a region or period of time.

Buffer strip—a strip of vegetation along a streambank which serves to iso-
   late the channel itself from the primary land use in the region.  May be
   a field border of grass or a strip of forest along the stream channel fol-
   lowing clear cutting of forest.

Channel—an open conduit (natural or man-made) in which water flows from a
   higher to lower point.

Channel gradient (slope)—the incline of a channel which may be expressed as
   the number of feet of fall per 100 feet of horizontal distance.

Channelization—the process of straightening and deepening a stream along
   with removal of nearstream vegetation in order to increase the rate of
   drainage from the land.

Channel morphology—the structural characteristics of a channel including
   such things as bank slopes, depth, sinuosity, etc.  Important in determin-
   ing runoff rates, channel capacity, sediment carrying ability, biotic
   diversity, etc.

                                     85

-------
Chemical oxygen demand—similar to biological oxygen demand defined above
   but referring to the volume of oxygen-demanding chemical wastes in water.

Coarse particulate organic matter—organic matter in a stream, such as fallen
   leaves, which is present in large size units; that is, before decomposi-
   tion or invertebrates have fragmented the material.

Community—An association of interacting populations, commonly defined by
   their co-occurrence in space.

Community structure—a complex of parameters (species diversity, trophic
   structure, etc.) which can be used to identify the organizational charac-
   teristics of a community.

Cold blooded—an organism whose temperature conforms with the temperature
   of its environment - "poikilothermic".

Continuity equation—Hydraulic equation resulting from the principle of con-
   servation of mass.  For steady flow, the mass of fluid passing all sec-
   tions in a stream of fluid per unit of time is the same, so that as area
   of flow decreases, velocity of flow increases.

Critical slope—slope threshold above or below which efficiency of vegetation
   to remove sediment from flowing water declines rapidly.

Delivery ratio—ratio used to describe the proportion of material eroded from
   the land surface which actually reaches stream channels, lake beds, or
   other locations.

Detention time—the time during which water is stored or held by, for example,
   a PTO terrace before it is released slowly to a nearby channel.

Ecosystem—the totality involved in the plants and animals (biota) plus the
   physical environment of an area and their interactions.

Environmental gradient—A physical or biological factor varies in its effect
   over space; e.g., temperature changes as one climbs up a mountain.

Erosion—wearing away of land surface by detachment and transport of soil and
   rock materials.   May occur through action of wind, water, or other agents.

Eutrophication—the accumulation of nutrients in a body of water more
   rapidly than natural biological processes can accommodate them.

Filter (vegetative)—the use of living vegetation or surface litter to reduce
   sediment content of flowing water.  Results from changing velocity, turbu-
   lence, and other characteristics of flowing water.

Flow rate—The volume of water moving past a location per unit time.
                                     86

-------
Flow retardance—situation when debris or other material in a channel
   or other area of moving water inhibits the flow of that water.

Food chain—sequence of organisms which depend on each other as sources of
   food; more or less linear sequence, (see food web)

Food web—complex of organisms with many pathways (food chains) interdigi-
   tating from plants to top carnivore.

Fry—recently hatched young of fish.

Greenbelt—Narrow band of vegetation along the channel of a stream.  May be
   either natural assemblage of plant species or a selected species or group
   of species.

Habitat—the natural environment of an organism.

Habitat diversity—as the number of habitats (pools, riffles; bottom types,
   etc.) in a reach of stream increases, the habitat diversity of the stream
   increases.  It is important in determining the species diversity of the
   resident biotic community.

Headwater—the upstream end of a drainage complex; usually considered to be
   stream orders 1-3.

Heterotroph—organisms which obtain their nourishment from organic matter.

Holistic—the philosophy that for complete description of a system the
   behavior of the individual components must be combined with knowledge of
   the components joined as units; very simply, the whole is equal to more
   than the sum of its parts.

Infiltration filter—area of vegetation where nutrients and/or sediment are
   removed as water infiltrates down through the soil profile; there is no
   surface runoff out of such an area.

Invertebrate drift—aquatic invertebrates which release their hold or are
   torn loose from a substrate and move downstream with the moving water.

Jackson Turbidity Unit (JTU)—see turbidity.

Meander—the winding of a stream channel

Metabolism—the totality of chemical activity in a plant or animal which
   results in the provision of energy and nutrients to the organism.

Microflora—microscopic organisms such as bacteria, fungi, and other groups
   in a biotic community.  Frequently used to refer to those organisms
   colonizing decomposing organic matter.

Nursery area—the region where  immature fish initiate growth and development.
                                     87

-------
Nutrient—a substance required by an organism for normal growth and mainte-
   nance.

Organic matter—materials which are produced by living organisms which con-
   tain carbon and are not in a completely oxidized condition.

Opercular cavity—the region under the gill cover (or opercle) of a fish
   where the gills and gill arches are located.

Overland flow—water flowing on the surface of the land; in contrast with
   channel flow or underground movement of water.

Parallel tile outlet terrace (PTO terrace)—a system of low terraces con-
   structed to impede the runoff of water from the land surface while pro-
   viding for a slow release of water through a tile system; results in
   improved water quality by reducing sediments carried to stream channels.

Periphyton—The plants and animals attached or clinging to stems and leaves
   of rooted plants or other surfaces projecting above the bottom of a body
   of water.

Productivity—rate at which energy or nutrients are assimilated by an indi-
   vidual, population or community (amount per unit time).  Important to
   distinguish between net production which is the rate of accumulation in
   a system and gross production which involves both accumulation and the
   amount metabolized in the maintenance of the organisms.  Primary produc-
   tivity is productivity of plants, while secondary productivity is pro-
   ductivity of animals.

Reductionist—the philosophy which suggests that a complex system can be
   fully described by decomposing it into successively smaller components
   and describing their behavior.

Roughness factor (n)—Measure of the irregularity in a drainage channel which
   will reduce velocity of flow of water in the channel.

Salmonids—fishes in the family Salmonidae, including trout and salmon.

Sediment—fragmentary material that is transported by water or air or is
   accumulated in beds by natural processes.

Sediment deposition—the placement of sediment in a location after its move-
   ment by wind, water or other geological agents.

Sediment load—all particulate matter carried or transported by water.

Sediment trap—a specifically constructed area to reduce flow velocity of
   water thereby reducing its sediment carrying capacity.

Shallow channel flow—water flowing in a channel when the depth of the water
   does not exceed the herbaceous vegetation in the channel.

-------
Sinuosity—the ratio of channel length to down-valley distance.  That is,
   an index to the amount of meandering by a stream.

Social hierarchy—a social organization among animals in which dominance
   position within the group determines use of available resources.

Spawning—the act of laying and fertilization of eggs in fishes and shell-
   fishes, especially.

Spawning grounds—area or location where fish reproduce (lay eggs).  Typi-
   cally characterized by specific bottom type, current, etc. for each
   species.

Species diversity—a measure of the complexity of a community, which may be
   simply the number of species (species richness) or a more complex index
   utilizing information on number of species and their relative abundances.

Standard metabolism—metabolic requirements of an organism for maintaining
   normal physiological functions (e.g. heartbeat, muscle tone, respiration,
   etc.) .

Streambank—the sides of a channel which are exposed during most normal flow
   periods.

Stream order—A method for numbering streams as part of a drainage network.
   The smallest stream channel is first order.  A second order stream is
   reached when two first order channels join, etc.

Subsurface runoff—underground movement of water either through natural
   drainage pathways or through tile systems.  Subsurface runoff may or may
   not reach an open channel.

Succession—the process of replacement of one biotic community by another.
   Each successive community changes the physical and biotic environment by
   its presence.

Surface litter—plant debris deposited on the soil surface which helps to
   protect that surface from disturbance by raindrop impact or overland flow.

Surface runoff—movement of water overland until it reaches a channel.

Suspended sediment—fine particulate material which remains in suspension and
   without contact with the channel bottom.

Tile line—an underground conduit installed by man to facilitate drainage of
   water from an area and thereby lower the water table.

Transfer coefficient—amount of material moved from a system component to
   another component relative to the amount in the donor compartment usually
   expressed as a ratio.
                                     89

-------
Turbidity—condition of water resulting from the presence of suspended
   material.   Typically expressed in Jackson Turbidity Units, a measure of
   interference with light transmission.

Unit stream power (USP)—the energy available in flowing water; a function
   of the slope and velocity of the channel in which the water is flowing.

Universal soil loss equation (USLE)—equation developed by agricultural
   scientists to predict the amount of soil eroded from a land area.
                                     90

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
1. REPORT NO.
  EPA-600/3-77-Q97
                                                           3. RECIPIENT'S ACCESSI ON-NO.
4. TITLE AMD SUBTITLE
 Impact  of Nearstream Vegetation and Stream Morphology
 on Water  Quality and Stream  Biota
             5. REPORT DATE
               August 1977 issuing date
             6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)

 James R.  Karr and Isaac J.  Schlosser
             8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 Department of Ecology, Ethology  and Evolution
 University of Illinois
 Champaign,  IL 61820
             10. PROGRAM ELEMENT NO.

                1HB617
             11. CONTRACT/GRANT NO.

               68-01-3584
 12. SPONSORING AGENCY NAME AND ADDRESS
 U. S. Environmental Protection Agency - Athens, GA,
 Environmental Research Laboratory
 College  Station Road
 Athens.  GA.   30605	
             13. TYPE OF RE PORT AND PERIOD COVERED
               Final         	
             14. SPONSORING AGENCY CODE

               EPA/600/01
 15. SUPPLEMENTARY NOTES
 16. ABSTRACT
      As man modifies watersheds by removal of  natural vegetation and stream channeliza-
 tion, disequilibria  in both the terrestrial  and aquatic environments result.   These
 disequilibria are  the major problem in controlling sediments and nutrients  from non-
 point sources and  improving the quality of the stream biota.

      In this report  we review the literature dealing with (1) the possible  use of
 near stream vegetation to reduce the transport of sediment and nutrients  from the
 terrestrial to  the aquatic environment and decrease stream temperature  fluctuations,
 (2)  the effect  of  stream morphology on sediment transport, and  (3) how  near stream
 vegetation and  stream morphology affect the  biota of streams.  The results  of this
 review suggest  proper management of near-stream vegetation and channel  morphology
 can lead to significant improvements in both the water and biological quality of
 many streams.   However, critical research outlined in this report is still  necessary
 if we are to properly use this management alternative to attain the objectives of the
 Federal Water Pollution Control Act of 1972  (Public Law 92-500).
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.IDENTIFIERS/OPEN ENDED TERMS
                           c.  cos AT I Field/Group
Watersheds
Pollution
Water pollution
Agriculture
  Nonpoint sources
   06F
   13B
   08H
   08M
18. DISTRIBUTION STATEMENT

RELEASE  TO PUBLIC
19. SECURITY CLASS (ThisReport!
   UNCLASSIFIED
21. NO. OF PAGES

	103
                                              20. SECURITY CLASS (This page)

                                                 UNCLASSIFIED	
                                                                         22. PRICE
EPA Form 2220-1 (9-73)
                                           91
                                                           * U.S. GOVERNMENT PRINTING OFFICE  1977 0-241-037/76

-------