EPA-905/9-81-004
                                                            June 1981
                   SUMMARY OF THE BLACK CREEK PROJECT
                           (Progress Report)

                    Report through 1980 Project Year
                            Based on Seminars

                                   in


                   'Washington, D. C., February 1980

                     Chicago, Illinois, March 1980



                                   by


           Allen County Soil and Water Conservation District

                           Purdue University

                          University of Illinois


                            Grant No. S005335
Ralph G. Christensen                                      Carl D. Wilson
Section 108a Program                                     Project Officer
                              Prepared for
                  Great Lakes National Program Office
                  U.S. Environmental Protection Agency
                    536 South Clark Street, Room 932
                        Chicago, Illinois 60605

-------
                             CONTENTS




OVERVIEW OF THE BLACK CREEK PROJECT                               1




MONITORING OF CHEMICAL ASPECTS OF WATER QUALITY IN BLACK CREEK    4




IMPACT OF CROP SEQUENCE AND TILLAGE ON SOIL LOSS                 13




ANSWERS                                                          20




PRACTICAL USES OF THE ANSWERS MODEL IN BMP PLANNING          '    25




BIOLOGICAL PERSPECTIVE ON WATER QUALITY GOALS                    30




BLACK CREEK IMPLICATIONS: PRESENT AND FUTURE                     57

-------
                                     -1 -


                      OVERVIEW OF THE BEACK CREEK PROJECT


                            Sy James. B. Morrison  1

     The Black Creek Watershed is located  in the  Maumee River Basin, a  highly
productive  agricultural  area,  geologically dominated by the former floor of
glacial laKa Maumee.  The Maumee basin drains into the  western  end  of  Lake
Erie through the gently flowing Maunee River.

     Black Creek, located in north central Allen  County, Indiana, is a  tribu-
tary  of  the  Maumee.  The  12,000-acre   Black   Creek watershed was cnosen to
represent the Maumee Basin in a water quality study, because its physical  and
economic character so closely mirror  that of the  basin.-

     Like the Maumee Basin, Black Creek is largely an agricultural  area.   If
you define the town of Harlan as "urban" then the watershed has about the same
proportion of rural and urban areas as does  the  basin.   It  probably  under
represents  the  lake  plain, and overrepresents  upland areas, but in general,
the watershed provides a satisfactory model for the basin which has been iden-
tified as the largest single contributor of silt  to Lake Erie.

     If you  fly over the Black Creek watershed, you will be surprised to hear
that  it has an erosion problem.  The basin appears to be flat.  There are not
obvious areas vvhere soil loss would be expected to be great, but at the begin-
ning  of  the  Black  Creek  project,  close inspection revealed many areas of
potential water quality problems.  Water had  damaged  roadsides,  cattle  had
damaged  ditch  banks, the erosion from many small rills and gullies in fields
produced tons of soil to be carried away toward the lake.

     In 1972, a conference on the Maunee River was  held  by  Rep.  J.  Edward
Roush'  in Fort Wayne.  The Maumee was at that time being considered for inclu-
sion in the Wild and Scenic Rivers system.   During  the  conference,  speaker
after  speaker spoke to the question of pollution of the Maunee.  Tney were in
agreement that although problans of industrial and municipal pollution of  the
river  remained,  these  were at least capable of solution by known and tested
methods.  The problem of agriculture, pollution from what would soon be  Known
as nonpoint sources, was another question.  Soil erosion, resulting from agri-
culture, may turn out to be the number one killer of rivers like  the  Maunee,
the speakers concluded.

     Those remarks, which kindled the interest of the Allen  County  Soil  and
Vfeter  Conservation  District, were the conception of the Black Creek project.
There seemed little doubt that soil erosion and the  related  fertilizers  and
pesticides  which might be carried away with soil particles, could represent a
water quality problem.  There was, at the same time,  a rather  extensive  body
of  experience and knowledge about controlling soil erosion.  There was little
understanding about how erosion control related to water quality.

     Primarily, the stated purpose of the project  was  to   determine  if  the
traditional   and  well  known  techniques of soil erosion control could have a
 1. Information Specialist, Purdue University

-------
                                    - 2 -   .


significant impact on water quality.  Recklessly,  we said we would apply those
practices,  find  out  how water quality improved, find out how much money was
spent in getting the improvement, and  finally  predict how  mucn  loadings  to
Lake Erie could be reduced for certain expenditures in the Maumee Basin.

     Much of the history of the Black  Creek project has involved more realist-
ically defining the goals.

     Other authors in this report will speak in detail  about  the  monitoring
and  modeling efforts, biological studies,  application of technology, and til-
lage trials.  This overview will concentrate on some of the changes in  think-
ing that have accompanied nearly 10 years of work  on Black Creek.

     LAND TREATMENT PLANNING — Initial  plans for   conservation  treatment  of
land  in  the  Black  Creek project were detailed. They involved many alterna-
tives, many practices, and considerable  effort. Unfortunately, often  project
administrators,  and  even  new  planners  could not understand what was being
planned and what kind of commitments had been made by the project and  by  the
landowner.  They were difficult to use in a voluntary program,  they would have
been impossible as a basis for determining  compliance in a mandatory  program.
As the project continued the planning  process was  simplified, and an agreement
form was developed that made it readily  apparent what kind of work was  to  be
accomplished  each year, what the responsibility of the landowner was and what
the responsibility of the project was.

     LAND TREATMENT GOALS — The initial concept of the  Black   Creek  project
was   "wall to wall" conservation.  Apply as many  different kinds of treatment
as possible to every acre of land.  Although this  was  not  accomplished,  the
project  and  the  landowners  of  the  watershed   managed  to  spend more than
$750,000 on land treatment.  At the end  of  the treatment phase, it was  deter-
mined  that the same water quality benefits could  have been achieved by spend-
ing less than half of that amount —$325,000,  concentrating the expenditure on
critical  areas, and limiting those practices involved to those wnich, in this
particular area, had an obvious water  quality impact — field borders, holding
tanks,  sediment  basins, critical area  planting,  grassed waterways, livestocx
exclusion, pasture renovation, and terraces.

     IMPORTANCE OF DITCH BANKS —At the  beginning  of the project, no  one  was
certain  how  much  of the erosion problem  was represented by tne condition of
the ditch banks.  However, everyone was  certain that work nad to be done there
because  the  erosion  problem was most  visible there. Much money was spent on
channel reconstruction, seeding, shaping, etc.  This despite objections  from
the  biological  group  that  some  of the  work was doing more  harm than good.
Finally, it was done in the face of findings that  less than 7 percent  of  the
erosion  could  be  attributed  to  ditch banks.  Project personnel now say we
would have spent less on the ditches if  the work were to be done  over.   How-
ever,  teams  starting  new projects continue to want to put initial effort on
ditches for the same reasons outlined  in Black Creek.  Ditch bank  erosion  is
visible, even if it is not too important.

     COLLECTING AND ANALYZING DATA —  As investigators sought to  analyze  the
results of the Black Creek land treatment efforts, some significant changes in
the thinking of most of them about data  analysis  occurred.  Initially,  most

-------
                                     - 3  -
were  convinced  that  an  adequate picture of the impact of land treatment on
water quality could be obtained by periodically collecting grab  samples  from
the  stream  and  its  major  tributaries.   As the nature of the storm runoff
events became more clear, the need for the collection of samples on a continu-
ous  basis  during a storm and the value of automated samples became apparent.
At the same time, the need for detailed analysis and predicting, such as could
be  provided  only  by  a computerized model, became apparent.  Other speakers
will elaborate on these points.

     Although this document concentrates on biological and physical aspects of
the  Black  Creek project, it is important to recognize that there were colla-
teral studies which provided useful information.  Rather detailed  studies  of
the  sociology  and  economics of the Black Creek area led to some interesting
conclusions.  An interesting economic point was that  it  costs  fanners  much
more  to  apply  conservation practices when farm prices are good than it does
when farm prices are less satisfactory.

     As a part of the project, a computer network was established, controlling
the  collection  of  water samples and recording weather data on a 24-hour per
day basis throughout the year.

     Studies of the Black Creek ecology evolved from  simple  studies  of  the
kinds  and number of fish present to detailed investigations of the interrela-
tionships within the stream-land-river system.

     Details of these and other supporting studies are available in  a  series
of reports on the project published by USEPA Region V.

-------
                                   - 4  -
        MONITORING OF CHEMICAL ASPECTS OF WATER QUALITY IN BLACK CREEK
                             By  Darrell Nelson
     Figure 1.  is  a graphical representation of the 5000  hectare   BlacK  CreeK.
study area.
                                                     •AUTOMATED SAMPLER
                                                        & STAGE RECORDER

                                                     • RAINGAGE
                                                       MAUMEE RIVER
                       Figure 1.  BlacK Creeic Watershed
Table 1. provides information on soils and land use  in  the  watershed.    The
drainage  pattern  in  the   area consists of one natural stream (Blac:< Creex) ,
running from east to west and discharging into the Maumee River, and a number
 1. 1. Agronomy Department,  Purdue University

-------
                                    - 3 -


of drainage ditches used as outlets for surface and tile  drains.   Grab  sam-
pling stations were established at numerous sites within the watershed to pro-
vide water quality data from the various soils and land uses on  drains  above
the  station.   Automated samplers were installed at three locations (Stations
2,6, and 12) in the watershed (Figure 1) to provide  data  on  loadings  being
discharged  during  storm  events.   Snail  subwatersheds were established and
equipped with automated samplers to allow careful monitoring of the  influence
of  selected  best management practices on sediments and nutrients in drainage
water.  Samples were collected from each of the tile outlets in the  watershed
to  provide information on the quality of subsurface drainage water and a tile
system draining a 23 hectare field was continuously monitored  by  use  of  an
automatic sampling system.

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

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

  Drainage area, ha               4950            942        -     714

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

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

  Number of homes;			28	143	

     Flow was continuously monitored at all stations and water  samples  taken
by  grab  or automatic methods,  were analyzed for suspended solids, nutrients
and other water quality parameters.  Pesticides alkaline  earth  cations,  and
selected  heavy  metals were occasionally monitored in water samples to deter-
mine if unusual conditions existed in the watershed.  All tile  drainage  sam-
ples  were  analyzed for suspended solids and nutrients.  Whereas samples col-
lected by the automatic sampler from the defined tile system were analyzed for
pesticides and flow was continously monitored.

     Meteorological conditions in the watershed were continously monitored.  A
complete hydrometerological station with automatic data acquisition and remote
transmission capability was established at site 6 (Figure 1).  The  amount  of
rainfall  was measured at seven other locations in the watershed and rainwater
samples were collected for chemical analysis at two locations.

     After collection, samples were transported  to  Purdue  University  where
analyses  were  conducted.  Data processing was also accomplished at Purdue to
obtain loadings through multiplying measured flows by concentrations of solids
or  nutrients.   Remotely collected meteorological data was stored in computer
format until it was needed for correlation with measured flow data.

-------
                                     - 6 -
                       Results of Water Quality Analyses

     Reconnaissance sampling  early in  the project revealed that no significant
 amounts  of  hexane — soluble pesticides where present in water, sediment, or
 fish tissue collected  from   the  watershed.   Specific  pesticides  evaluated
 included aldrin, dieldrin, DDT atrazine, trifluralin, and 2, 4, 5-T.

     Table 2 provides information on the rainfall, runoff, and  sediment  lost
 from the two major subwatersheds in the Black CreeK area form 1975 to 1978.

  Table 2.   Rainfall,  runoff, and sediment and nutrient loss occurring  in two
            drainage areas of the BlacK CreeK watershed during  the period 1975
            to  1978.
Parameter
Rainfall, cm
Runoff, cm
Sediment loss, Kg/ha
Sediment P loss, Kg/ha
Sediment N loss, Kg/ha
Sol. inorg. P loss, Kg/ha
Sol. org. P loss, Kg/ha
NH J-N loss, kg/ha
NO I-N loss, kg/ha
Site
no
2 & 6
2
6
2
6
2
6
2
6
2
6
2
6
2
6
2
6

1975
108
29.1
26.0
2126
3735
5.24
4.51
31.25
28.98
0.14
0.34
0.11
0.13
1.51
1..82
19.01
11.63

, 1976
66
12.4
10.1
637
384
0.98
0.73
4.82
2.86
0.06
0.18
0.04
0.04
0.60
0.85
5.55
2.39
Year
1977
96
18.5
19.4
435
452 .
1.67
1.78
4.55
4.71
0.14
0.47
0.06
0.10
0.58
1.30
15.42
12.73

1978
77
18.5
21.3
380
544
0.65
0.79
6.10
6.91
0.21
0.68
0.08
0.35
0.75
3.06
8.27
5.96

Ave.
86.8
19.6
19.2
895
1279
2.14
1.95
11.68
10.87
0.14
0.42
0.07
0.16
0.86
1.76
12.06
8.18
Precipitation was above normal during 1975, below normal  in  1976,  and  near
normal  in  1977 and 1978.  Runoff volumes increased with increasing rainfall,
however, the percentage of precipitation appearing as runoff varied  over  the
years  (26%  in  1975,  17%  in  1976, and an average of 22% for the four year
period).  Sediment discharges from the watershed averaged 895 and  1279  Kg/ha
for  station 2 and 6, respectively.  However sediment losses in 1975 were from
4 to 8 times higher than average of the other three years.  Adoption  of  best
management practices in the watershed during 1975 and 1976 apparently resulted
in decreased sediment losses during 1977 and 1978 as  shown  by  low  sediment

-------
                                     - 7 -
discharges  during  these  years  even though rainfall occurred in near normal
amounts.

     Table 2 provides data on amounts of sediment - bound N and  P  discharged
from  subwatersheds  in  the  Black  Creek area during 1975 through 1978.  The
quantities of sediment - bound nutrients lost  from  the  watershed  decreased
markedly  after 1975 generally in proportion to reductions in sediment losses.
Application of best management practices in the watershed  was,  at  least  in
part/  responsible for the reductions in amounts of sediment - bound nutrients
observed during the course of the study.

     Table 3 provides data on the amounts of soluble inorganic P  and  soluble
organic P discharge from the two subwatersheds during a four-year period.  The
amounts of soluble organic P discharged  were  generally  low.   however,  the
amounts lost from the watershed did not decrease during the

    Table  3.   Proportions of  total P and N leaving the Black Creek watershed
    	transported as  various nutrient forms.	

    Form of nutrient             Site no.                    Site no.
    transported       	:	,	
                           % of total  P lost*          % of total N lost*

    Sol.  inorg.  P          6.0           16.6            —          —

    Sol.  org.  P             3.0            6.3            —          —

    Sediment P           91.1           77.1            —          —

    NHj-N                  —             —             3.3         7.7

    NO~-N                  —             —            46.2        35.9

    Sol.  org.  N             —             —             5.8         8.7

    Sediment N             —             —            44.7        47.7


    *Average for  four years (1975-1978).

period of study.  In fact, it appeared  that  soluble  inorganic  P  discharge
increased during 1978.   One explanation for this finding may be that untreated
household wastewater was discharged into the ditches near  Harlan  during  the
time  an  interceptor sewer was being constructed.  Study had previously shown
the septic tank effluents were a major source of the  soluble  P  measured  at
Station 6.

     Table 3 provides information on amounts of  NHt-N  and  NOv-N  discharged
from two subwatersheds in the study are during 1975 tnrough 1978.   The amounts
of N03-N in drainage water appeared to be related to amounts  of  rainfall  in
the  watershed,   i.e.  losses  of NOZ-^J were highest in 1975 and 1977, the two

-------
                                     - 8 -


 years with highest rainfall.   However,  the  weighted mean NOT-N. concentrations
 in  drainage water appeared  to increase somewhat during the  course  of  the  pro-
 ject.  Losses of NCU-N were  relatively  high (average  of 12 and 8  kg N/ha   for
 stations  2  and 6,  respectively)  and likely reflect  the fact  that  much of the
 watershed is tile drained  and  soils  are maintained in high state  of fertility
 by  applications  of manure and  inorganic N fertilizers.  Losses  of NH*-N  were
 low throughout the study except  for  station 6  during 1978.   the  relatively
 high  N  loss obtained at  station  6  likely  resulted from septic tank effluents
 originating in Harlan during construction of the interceptor sewer  line.   Cal-
 culations  suggested that  14%  of the N  added to soils by natural  N  fixation as
 through manure or fertilizer applications appeared as NHt +  NCC  in  drainage
 water (Table 4).                                       *      j

   Table 4.   Calculated average losses and inputs of inorganic N for the Black
  	Creek watershed.	

                                                      Ave. inorg. N*
  	loss or gain	

   Inorg.  N losses:                                    leg x 10"^

    Measured  total loss                                      58.8
    From  precipitation runoff                                10.1
    From  septic  discharge
    From  applied or  fixed N                                  41.4**

  Inorg.  N inputs:

    Applied  and  fixed N                                     286.8***
    Infiltrated  precipitation                                34.9
    Total  inputs                                           321.7

  % of  total N inputs lost in drainage                       17.5%

  % of  applied and fixed N lost  in drainage	13.7%

  *Average of four years (1975 - 1978)

  **Calculated as the difference between total inorganic N loss and
    that derived  from precipitation runoff and septic discharge.

  ***Based on farmer information on fertilizer and manure applied
     and  average values for N fixation by legumes.

This finding suggests that improvements may be made in managing N additions to
soil to minimize the amounts which are present in drainage water.

     Analyses of water samples suggests that adoption of best management prac-
tices  by farmers has not led to a reduction in the discharge of  soluble forms
of N and P from  the watershed.  In fact, there is a suggestion  that losses  of
soluble  N  and P increased slightly as the conservation practices were imple-
mented.   In future projects some attention should be  given to adaption  of best
management  practices  which  minimize  the transport  of soluble nutrients from

-------
                                    -  9 -
soil to water.

     Table 5 presents the average proportions of total rainfall,  runoff,  and
sediment  transport  which  occurred in calendar quarters during the 1975-1978
measurement period.  Rainfall was nearly equally  distributed  throughout  the
quarters which

  Table  5.   Average proportions of  rainfall,  runoff, and  sediment and nutrient
            losses  in  the  BlacK CreeK watershed during  calendar quarters.
Parameter
Site Calendar quarter*
no. 	 	 	 ,_,
123

4
% of yearly total**
 Rainfall
2 & 6
20.8
33.6
26.8
18.7
Runoff
Sediment loss
Sediment N loss
Sediment P loss
Sol. inorg. P loss
NHj-N loss
NOj-N loss
2
6
2
6
2
6
2
6
2
6
2
6
2
6
46.9
47.6
34.2
25.4
22.7
• 23.4
32.1
28.4
52.9
59.3
50.6
67.1
44.2
42.7
34.1
30.8
54.1
65.7
67.2
• 64.3
60.3
46.0
25.4
16.4
35.6
15.9
36.6
31.6
3.8
4.1
2.6
2.2
1.6
2.9
2.8
4.6
3.6
5.0
2.3
1.7
2.6
2.9
15.2
17.5
9.2
6.7
8.5
9.4
4.8
21.1
18.1
19.3
11.5
15.4
16.7
22.7
  *1, Jan.-Mar.; 2, Apr.-June;  3, July-Sept.; 4, Oct.-Dec.
  **Average of  four years  (1975-1978)

the 2nd quarter (April - June)  having the highest proportion and the 4th quar-
ter  the lowest proportion.  Most of the runoff in the watershed occurred dur-
ing the first two quarters with the 1st quarter contributing during the second
quarter  (SS-65%  of  total)   and very low during the last two quarters of the
year.

     A high proportion of soluble inorganic P was transported during  the  1st
quarter  of  the year likely due to snowmelt runoff carrying soluble P leached
from plant material on the soil surface.  A substantial proportion of the sed-
iment bound P lost from the watershed was transported during the 2nd quater in
relation to the proportion of sediment which was transported  during  the  2nd

-------
                                    - 10 -


quater.   A  significant percentage of sediment-bound P was transported during
low flow conditions.  This suggests that nigh P solids originating from septic
tanxs  are  a  source  of  sediment  -  bound  P during low flow conditions in
ditches.

     More than 50% of NH+-N lost from the watershed was transported during the
1st  quarter_ indicating  that  snow melt may be a*major contributor of NHt-N.
Losses of NO^-N were concentrated during the first two quarters  of .the  year
vtoen runoff and percolation of water were high.  Nitrate in streams originates
largely from N applied as fertilizers and manure the previous year or mineral-
ized  from  organic  matter  after crop harvest in the fall.  There were low N
loses measured during the two quarters following fertilization during  a  par-
ticular  calendar year.  Losses of sediment - bound N were concentrated during
the 2nd quarter as was found for sediment and sediment - bound P loses.

     At station 2 (draining largely agricultural land)  more that  90%  of  the
total  P  lost  from  the  subwatershed  was transported as sediment - bound P
(Table 6).  However, at station 6 (receiving some septic effluent)  a  surpris-
ing  proportion (23%) of total P was transported as soluble forms of P.  These
findings  suggest  that  for  agricultural  land  the  reduction  in  sediment
discharge  can  have  a  marked effect upon minimizing total P loadings coming
from a watershed.

   Table 6.  Proportions of total P and N leaving the BlacK CreeK watershed
             transported as various nutrient forms.

   Form of nutrient             Site no.                    Site no.
   transported       	
                           %  of  total  P lost*          % of total N lost*

   Sol.  inorg. P           6.0           16.6            —          —

   Sol.  org.  P             3.0            6.3            —          —

   Sediment P           91.1           77.1            —          —

   NH+-N                  —             —             3.3         7.7

   NO~-N                  —             —            46.2        35.9

   Sol.  org.  N             —             —             5.8         8.7

   Sediment N             —             —            44.7        47.7


   *Average for four years (1975-1978).

For urban areas or farmland impacted by septic tanK. effluents attention should
also be directed toward reducing the amounts of soluble P lost from the area.

-------
                                    -'11 -


     About one - half of tne total N discharged from the Black Creek watershed
is  as  sediment  - bound N  and one - half is as NO^  (Table 6).  This finding
suggests that in any land areas best management practices must be directed  at
minimizing  runoff  and  leaching  of NOZ as well as reducing sediment losses.
Furthermore, modeling N discharge from agricultural watersheds will be  diffi-
cult  because of the requirement to describe:(i) sediment - bound N transport,
(ii) leaching of NCC-N to tile drains, and (iii) movement of  NO"  in  surface
runoff water.

     Flow in ditches in the Black Creetc watershed  originates  primarily  from
rain  storms  (event  related), however, base flow and snow melt significantly
contribute to total annual flow at certain times during  the  year.   Sediment
originates  primarily  from one to three large rainstorms (producing 2.5 cm of
runoff) which occur each year.  A relatively  low  proportion  of  transported
sediment originates for the numerous small storms which occur during the year.
However, many of the small storms occur when the soil surface  is  covered  by
vegetation and the erosion potential is low.

     Sediment - bound P transported to drainage ditches in the watershed  ori-
ginates  primarily from large rainfall events (Table 7) as was found for sedi-
ment.  Rainfall events of all sizes as well as snow melt are  major  contribu-
tors  to  the  soluble inorganic P leaving the watershed.  This finding likely
originates from the equilibrium wnich is established between P sorbed on  soil
particles  and  inorganic  P  in solution whenever rainfall or snow melt comes
into contact with soil.  This equilibrium is largely independent  of  sediment:
waters rates and the soluble inorganic P loads are a  function  of  volume  of
water  originating  in  the various flow producing events.  However, snow melt
runoff is apparently enriched in inorganic P leached from  plant  residues  on
the surface of the soil.

     Sediment - bound N in drainage water originates primarily from the  large
rainfall events responsible for most of the sediment transport.  Various sized
rainfall events and snow runoff transport NCU-N in  about  the  proportion  in
which  they  contribute  to  the total flow.  This finding suggests that NOZ-N
concentration in water present in ditches as a result of different  events  is
similar  and  the  amount transported is largely a foundation of the volume of
water originating form the various flow producing events.

     Water flowing from the watershed originates largely form surface  runoff,
however,  subsurface  and tile discharges into the ditches are responsible for
about 30% of total flow.  Almost  all  of  the  sediment  transported  in  the
watershed  originates  from  soil  erosion  and transport of soil particles in
runoff water.  Tile drainage water contained low concentrations  of  suspended
soils.

     Sediment - bound P originated primarily from surface runoff at both  sam-
pling stations, however, at station 6 a surprising proportion, 20% of sediment
- bound P,  was drained from septic tank  effluents..   Septic  tanKs  effluents
were  the  source  of  a  significant proportion (>30%) of soluble inorganic P
leaving the watershed, but surface runoff accounted for greater  than  55%  of
soluble  inorganic  P  in  runoff.   Surface runoff was the primary source for
soluble organic P in drainage water although septic tank effluent  contributed
a significant proportion in samples taken at station 6.

-------
                                    - 12 -
 .Table 7.   Proportions of water flow,  and  sediment and  nutrient losses from
            the Black CreeK watershed associated  with  different flow producing
            situations.
Parameter
•
Water flow (Runoff)

Sediment loss

Sol. inorg. P loss

NO~-N loss

Sediment N loss

Site
no.

2
6
2
4
2
6
2
6
2
6

BF

20
16
4
9
10
14
13
9
4
2
Type
SE
% of total
25
24
13
81
30
27
30
22
14
13
of flow*
LE
transported**
44
47
79
6
40
34
44
54
78
76

SM

11
13
4

20
25
13
15
4
9
  *BF, base  flow; SE, small  rainfall events resulting  in less than 2.5
  on of  runoff;  LE,  large  rainfall events  resulting  in more than 2.5 cm
  of runoff  SM,  snow melt  runoff.

  **Average  of four  years  (1975-1978).

     Ammonium N in drainage water originated largely in surface  runoff,  how-
ever,  both septic effluent and tile drainage water was a source for some NHt-
N.  Subsurface flow from tile  drainage  and  surface  runoff  were  equal  as
sources for NO^-Njsransported out of the watershed.  Best management practices
for minimizing NCL-N additions to surface water must be  designed  to  control
both  runoff and leaching losses of NO~-W.  Sediment - bound N originated pri-
marily in surface runoff.  Solids present in septic tames effluents were rela-
tively  low  in  total  N and the septic tank contributions were masked by the
large amounts of sediment - bound N present in surface runoff from cropland.

     Algae bioassay studies conducted to evaluate the  availability  of  P  in
suspended stream sediments of the Black Creek watershed established that about
20% of the total P and 30% of the inorganic P present would ultimately  become
available.   Therefore,  sediment  P  is the major source of algae available P
leaving the watershed (see Table 6)  and best management  practices  should  be
directed  at  reducing  sediment loss from the watershed if downstream effects
are to be minimized.

-------
                                    - 13 -


               IMPACT OF CROP SEQUENCE AND TILLAGE ON SOIL LOSS

                    By Jerry V.  Mannering and  Don Griffith^

     My purpose is to review some of our results from the  Black  Creek  study
regarding  conservation tillage.  Conservation tillage is certainly recognized
as an effective BMP in the Black Creek Project.  I will review with you some of
the  findings  regarding  conservation  tillage  from the Black Creek study and
also include other information not only from other  studies  in  Indiana,  but
surrounding  states as well regarding the impact of conservation tillage, pri-
marily on erosion control.

     One thing that you need to realize when you discuss conservation  tillage
is that soils are different and different soils  respond differently to various
forms of conservation tillage.  The north part of the watershed represents the
more  sloping area where runoff and erosion is more of a problem.  The central
section area is a transition area of the old beach ridges that has higher sand
content  in the surface soils, but also suffers from inadequate drainage.  The
south section is representative of the high clay, poorly drained soils  within
the  watershed.  There are good reasons why farmers have not flocked to a til-
lage system that leaves a large amount of residue on the surface in this  par-
ticular watershed and the reason is primarily because in much of the watershed
people have been fighting excessive water all their lives, and a  system  that
leaves a heavy mulch on the surface on poorly drained land aggrevates the wet-
ness problem.

     What we tried to do in the Black Creek study was to look at  conservation
tillage  on  a  range  of  soils  that  would well represent the makeup of the
watershed.  For example, we had a study located on  a  Haskins  loam  soil,  a
nearly  level  soil  that was influenced by sand cover from the old ridge.  We
tested conservation tillage on a Nappanee clay loam, an almost level soil  and
a  Hoytville silty clay, another almost level soil with less than 1% slope.  A
fourth test was located on a more sloping rolling area of Morley clay loam  on
a  4-4.1%  representing  the  upper  third  of  the  watershed.  Erosion under
equivalent rainfall would be 3, 4, or 5 times as much under  the  same  condi-
tions  on  the  rolling  part  of the watershed compared to the other 3 nearly
level areas.  The more level areas as have been fighting excess water  all  of
their  lives.   The reason the Black Creek watershed contains several drainage
channels is that farmers needed them to get rid of excess water so  that  they
could  make  this  an economic farming operation.  They cannot compete on this
kind of land without an effective subsurface drainage  system  and  they  need
adequate  outlets.   We  need to understand some of the drainage needs of this
area when we look at the overall impact of agriculture on water quality.

     In the Black Creek tillage tests discussed in this report wa were compar-
ing  the  influence of fall tillage on the erosion that occurs that subsequent
spring.  We compared 4 treatments? 1). we did nothing to the plots after  har-
vesting  (no-till),  2).  we  had  a  light fall discing, 3). a fall chiseling
around 8-9 inches deep,  and 4). a fall  moldboard  plowing  approximately  8-9
inches  deep.   With  the  moldboard  plowing you invert most of the residues,
trfiich has an influence on soil erosion, and on water quality.  The chisel plow
 1. Agronomists, Purdue University

-------
                                    - 14 -
on the other hand, depending on the amount of residues from  which  you  start
does  leave  appreciable  amounts of trash on the surface.  It also leaves the
surface in a quite roughened condition.  In some cases on some soils raoldboard
plowing  also  leaves the surface rough.  But with chisel tillage you have the
additional effects of the residue as well.  The light disking treatment, as we
used  it,  only  slightly  reduced the amount of residue cover on the surface.
Most conservationists agree that if we can Keep a sufficient amount of residue
on the surface we can do a tremendous job in reducing soil erosion.

     In this particular study we looKed in turn at the effect  of  tillage  on
soil  erosion,  the effect of crop sequence effects on soil erosion because we
tested the tillage  treatments  following  both  corn  and  soybeans  and  the
interaction effects between the type of crop and the method of tillage.  Tests
were made using simulated rainfall because we wanted to control the amount and
energy  of  rainfall  that was occurring so we could get a relative comparison
between treatments.  The storm applied was  an  intense  storm,  about  2  1/2
inches  per  hour one day followed 24 hours later by another 2 1/2-inch storm.
We measured the amount of runoff that occurred with a  water  stage  recorder.
Aliquot samples of runoff were taken every 5 minutes for determination of sed-
iment  load.  Again,  the  period  that  we  were  testing,  was  the  end  of
Wischmeier's  crop stage 3 (rough fallow for the plow, disk, and chisel)  and 4
(residue period for the no-till) .  Visual observations show  that  prior  crop
can  have  a  significant  effect on surface roughness, thus susceptibility to
erosion.  For example fall plowing high clay corn land  leaves  it  much  more
rough  and  cloddy  compared  to fall plowed soybean land.  Fall chiseled corn
land is also rougher than fall plowed soybean land.  Another  important  point
to  remember  when  evaluating the effect of tillage system on soil erosion is
row direction.  Any system that leaves marks or ridges such as chiseling  will
be  much more effective across than up and down slope.  A visual comparison of
light disking following corn verses light discing following soybeans show  the
former  to  be  cloddier and more resistant to erosion.  By the time the simu-
lated rain tests are made, the light disk  treatment  had  weathered  4  or  5
months  and  the easily transported soil particles had been removed by natural
runoff events.  Therefore when tests were made in  the  spring  percent  cover
following  corn  was  as  high on the disk treatment as the no-till treatment.
The surface roughness was similar on no-till plots  following  both  corn  and
beans, however, corn residues were much more plentiful than bean residues.  We
measured percent surface covered and the results are shown in table 1. for the
Hoytville soil.
Table 1. Surface Cover - Hoytville Silty Clay
%

No-till
Disk
Chisel
Plow
Soybeans
24
12
9
. 1
Corn
78
77
57
4
Soybeans/ corn
0.31
0.16
0.16
0.25
On the Hoytville silty clay loam following both soybeans and corn we are  set-
ting  increased  residue protection from conservation tillage compared to fall

-------
                                    - 15 -
moldboard plowing.  Total amounts of residue cover are much  higher  fallowing
corn  than  beans on all conservation tillage systems.  I don't believe that's
news to any of you, but it does indicate why at least one of the major reasons
why  land  following  soybeans is more erosive than land following corn.  Note
that only 31%, and 16%, respectively, as much cover occurs  after  beans  than
after corn on the no-till, disk, and chisel treatments.
     Effects of tillage on surface cover on the Morley clay loam is  given
table 2.
                                                                            in
Table 2. Surface Cover - Morley Clay Loam
%

No-till
Disk
Chisel
Plow
Soybeans
26
17
12
1
Corn
69
70
25
4
Soybeans/ corn
0.38
0.24
0.48
0.25
More residues are present following corn than following soybeans as  expected.
Residue  cover  following  beans  were 38%, 24% and 48% of the cover following
corn for the no-till, disk and chisel treatments.

     Soil losses on the Hoytville site are given in table 3.
Table 3. Soil Loss - Hoytville Silty Clay 0.8% Slope
t/ha
Disk
Chisel
Plow
Soybeans
7.8
6.9
9.3
5.3
Corn
1.1
.9
1.7
4.3
Soybeans/ co rn
7.1
7.3
5.6
1.2
Following soybeans soil loss from the no-till  and  the  disk  treatments  are
essentially  the same, but soil losses from chisel are a little higher.  Plow-
ing has the lowest soil loss.  On this level land the reason  that  soil  loss
from the chisel treatment is higher than on the plowing treatment soybean land
is that the chisel created a furrow for water to  flow  from  the  plot.   The
plowed treatment left the surface irregular and roughened and we had more sur-
face ponding.  Following corn there was sufficient residue left on the surface
of  the chisel plot to interrupt the furrow.  Following soybeans we didn't get
a whole lot of control form conservation  tillage.   Following  corn  we  did.
Soil  losses  were similar from no-till and disk, but remember they had amount
the same about of surface cover.  Chisel although  not  as  effective  as  the
other forms of conservation tillage still gave fairly good control compared to
plowing.  The relationship of soil loss following soybeans to  corn  is  quite
striking  seven  times  as much on the no-till and disk and 5 1/2 times on the
chisel.

-------
                                    - 16 -


     Table 4 contains soil loss data from the sloping Mbrley soil



Table 4. Soil Loss - Morley Clay Loam 4.0% Slope
t/ha
No-till
Disk
Chisel
Plow
Soybeans
13.4
12.4
30.3
40.9
Corn
2.4
2.5
15.0
21.8
Soybeans/ corn
5.6
5.0
2.0
1.9
Results from our original base plot data would indicate erosion was at least 3
times  more  serious  on the rolling (4%) land than it was on the nearly level
land in the watershed.  Positive influence from our  conservation  tillage  in
reducing  soil loss following soybeans occurs on the Morley site.  Soil losses
form the no-till and disk treatments were less than half these from the chisel
treatment  where tillage furrows were up and down the slope.  Still, soil loss
was 25% less from chisel than plow treatments.

     much more erosion control occurred from conservation tillage  after  corn
where soil loss from no-till and disk were about 10-12% of those from plowing,
and chisel losses were about 25% less than plowing.  Again rows  were  up  and
down  the slope.  A ratio of soil loss following beans and following corn com-
paring the conservation tillage system shows over 5 times as  much  soil  loss
under  the  no-till and disk treatments and 2 times as much for the chisel and
even on the plow treatment soil losses where he were almost 2  times  as  much
following  soybeans  than  following  corn.   The  point here being that as we
increase our soybean acreage we need to be much more careful about controlling
soil  erosion following that year of soybeans.  Remember, its not the year you
are growing soybeans, its the year after soybeans that is a problem.  Although
conservation  tillage is effective in reducing soil loss, it loses much of its
effect when you follow a crop like soybeans.  In addition you need to be  more
concerned with row direction with systems such as chisel tillage if you are to
do an effective job.

     The Black Creek results are just one of many we have about the  influence
of  conservation  tillage in reducing soil loss.  For example, tests conducted
at Coshocton, Ohio under natural rainstorms  show  no-till  to  be  especially
effective in reducing soil loss.  The data is given in table 5.   These results
were from a high intensity rainstorm.  Even though the no-tilled corn land was
on a slope of 21%, only a token amount of soil loss occurred.  This study also
demonstrated  the  value  of contouring in reducing soil erosion.  The results
demonstrate there is no question of the effectiveness of surface residues such
as you have in the no-till treatment in reducing soil erosion.

     The conclusions from our and other tillage studies, are as  follows:  1).
tillage  systems have a major effect on residue cover 2). prior crops signifi-
cantly effect residue cover 3). there is an interaction  between  the  tillage
system  and  crop  sequence  and  the amount of cover produced 4). an indirect
relationship exists between surface cover and soil erosion 5). systems such as
no-till or light disking that leave appreciable surface residue greatly reduce
soil erosion-it is not only just a small percentage its a major reduction  6).

-------
                               -  17 -
Table  5. Runoff  and  Soil  Loss on  Sloping -  and  Contour-Row Fields
         in Corn Watersheds  (Coshacton, OH).*
Tillage system and
row direction
Plow/clean till
(sloping rows)
Plow/clean till
(contour rows)
No-till
(contour rows)
%
Slope
6.6
5.8
20.7
Runoff as
% of rain
80
42
49
Soil loss
t/A
22.6
3.2
.03
 * About 5.3 inches of rain fall within a 7-hour period.
chisel  tillage  is only effective if residues are plentiful or surface  remains
rough so  its effect can vary from one soil type to another 7). row orientation
is very important on a system like chisel tillage on sloping lands 8). soybean
land  is much more erosive than corn land primarily because of the residue but
also because that soil tends to be looser.

     I  want to  say just a few things about the acceptance of conservation til-
lage systems by farmers and the adaptability of these systems to Indiana.  Don
Griffith has provided leadership in this portion of  the  study.  He  has  had
replicated  trials  out the last few years in which has first objective was to
determine which conservation tillage systems are adapted on the  primary  soil
types   in the watershed and the second objective was to have a high percentage
of the  fanners  in the watershed using  conservation  tillage  techniques.   We
haven't really succeeded in objective number 2.  Some of the reasons why are
discussed below.  Most certainly there has been an effort to get more  conser-
vation  tillage on the land through demonstrations and though research.  We've
had some successes and some failures.  But, we have learned from these Kind of
studies and results.

     Generally, we have  found  that  on  the  better  drained  soils  in  the
watershed,  even  no-till  systems will work providing pests (primarily weeds)
can be  adequately controlled with chemicals.  Table 6 contains  yield  results
from  the Morley clay loam soil, which is one of the more erosive soils in the
watershed.
Table 6. Morley Clay
Tillage system
No-till
Chisel
Plow
Loam - 1976
Harvest population
20,281
18,812
17,609

Yield
Corn
91.1
89.5
88.2

(Du/ac)
Soybeans
23.2
21.7
24.3
Although yields are not particularly high, both conservation  tillage  systems
were  competitive  with  plowing  for both corn and soybeans.  We can conclude
from these and other results that on these sloping, erosive lands conservation

-------
                                    - 18 -
tillage is competitive with plowing.  Additional results were obtained in 1979
and are given in table 7.
Table 7. BlacK Creek Tillage Trials - 1979 Morley - Blount Silt Loam	
Previous                Tillage                 Harvest                  Yield
  crop                  system                 population                bu/ac
Corn*
Soybeans
Plow
Chisel
No- till
Plow
Chisel
No-till
20,562
20,812
18,375**
21,812
22,062
20,562
114
112
100
134
136
134
*  Lodging was severe in all continuous corn plates, approximately
   due to corn root worms damage.
** Reduced stands in no-till continous corn resulted from poor seed
   cover due to wet soils at planting.
Where we were following corn no-till was not competitive with  plowing,  prob-
ably because of reduced stands.  When we were following soybeans, both no-till
and chisel were competitive with plowing.  Chisling was also competitive  with
plowing  where  corn  followed  corn.  Cn the poorly drained soils we have not
been getting the yields from conservation tillage that compare with  the  con-
ventional systems.

     Conclusions for these and other studies are as follows: 1).  fall  chisel
can replace moldboard for continuous corn or corn after beans without limiting
production where weeds can be controlled.  2). shallow tillage or no-till  for
continuous  corn  or  corn  after beans should not limit production on well or
moderately well drained soils where perennial weeds are not a serious problem.
3).  no-till  sod  planting  should not limit production compared to moldboard
wfiere perennial weeds -are not a  serious  problem.   4).  shallow  or  no-till
planting  is  liKely  to  be more successful on poorly drained soils when corn
follows soybeans or sod rather than corn.  5). perennial and herbicide  resis-
tant weeds are more liKely to limit soybean than corn yields with no plow til-
lage.  6). shallow or no-till planting compared to deeper tillage is liKely to
lead to more serious disease problems such as phytophthora root rot for soybe-
ans on poorly drained soils.  7). fanners are not liKely to adopt conservation
tillage  unless  success  can be demonstrated in their area.  8). some form of
conservation tillage that has been demonstrated to be adapted to soils  in  an
area  should  be  a  high priority BMP.  9). wnere pests are easily controlled
there should be little or no cost for the benefits gained in  erosion  control
and  water  quality.   10).  where perennial and resistant weeds are a problem
added herbicide costs and/or reduced yields may reduce  profit  by  $5-10/acre
for  corn.   Costs-in  terms of reduced yield or added chemicals, conservation
tillage for soybeans conclusions are not yet fully developed.

-------
                                 - 19 -

     We have put what we Know about tillage in Indiana into an extension  pub-
lication  AY-210  "Adaptability of Various Tillage-Planting Systems to Indiana
Soils." Vfe've got a big job to do as far as selling  conservation  tillage  in
areas  where  it is highly adapted.  I'm not concerned trying to get conserva-
tion tillage on all the land in the state of Indiana because we have  lots  of
land where its a high risK situation.  Poorly drained soils many times are not
well adapted to systems that leave lots of trash on the surface.   We're  con-
vinced  that  it certainly is tremendously effective in reducing soil erosion.
It can be used in combination with other conservation practices to  make  soil
erosion control more effective and even more complete, but we need to get more
farmers interested and more farmers convinced  that  it  will  worK  on  their
situation.

-------
                                    - 20 -
                                   ANSWERS
                              David B. Beasley
     Most hydrologic models attempt to model a physical system in which  there
is  an  input  of rainfall or some other driving meteorological variable which
interacts with the soil surface/ crop cover, and  subsurface  soil  layers  to
produce  runoff,  erosion,  chemical  washoff, subsurface drainage, etc.  This
concept is inherent to almost all watershed models that  use  a  deterministic
process.   The  general  hydrologic  and  erosional relationships that must be
addressed by a watershed model are presented in the Black CreeK  Final  Report
— Technical Volume (Lafce, 1977).

     Basically, large scale watershed models fit into two separate classifica-
tion  schemes.   They  are either long-term or event-oriented simulations.  In
addition, they either use distributed parameter or lumped parameter  concepts.
The  time  scale  used in watershed models is somewhat dependent upon what the
modeler or planner intends to do with the output data.  A long-term simulation
can  give  some  insights into overall loadings, net surface effects, etc.  An
event-oriented simulation uses a much shorter time increment and  attempts  to
describe,  in  detail, the storm-induced response of the hydrologic system and
any modifications that may have been made or planned.   For  non-point  source
studies,  the  event-oriented simulations attempt to describe those situations
in which the watershed is most active.

     The difference between lumped parameter and distributed parameter  model-
ing  methods and concepts is much harder to define.  Cne reason for this prob-
lem is the extent to which a particular model is either lumped or  distributed
can  be  quite variable.  Some models exhibit both lumped and distributed pro-
perties.  Essentially, a lumped  parameter  model  attempts  to  describe  the
overall system response using aggregated or lumped representations of physical
parameters.  In most cases, the lumped parameter loses its  physical  signifi-
cance  in  the process.  Another characteristic of lumping is the inability of
tne model to provide spatial output.  Cnce the system has been described,  the
output  point is fixed.  The distributed parameter model is, in effect, a sys-
tem of snail models (possibly lumped)  which provide the  ability  to  simulate
processes in a spatial and temporal sense.  Although much more physical signi-
ficance can be maintained, an assumption of unifonnity must be made within any
of  the small subdivisions.  The distributed concept also allows .for accessing
output from any or all points within the  modeled  area.   Two  direct  conse-
quences  of distributed parameter modeling are increased computer requirements
and costs.

     The ANSWERS (Areal Non-point Source Watershed Environment Response  Simu-
lation)  program  has  been  in  use for nearly four years.  In that time many
improvements in the actual model, as well as the operational  structure,  have
been  made.   ANSWERS  is  an  event-oriented, distributed parameter watershed
model.  The  original  ANSWERS  program  (Beasley,  1977)  was  based  on  the
 1. 1. Agricultural Engineering Department, Purdue University

-------
                                    -  21  -

distributed parameter watershed hydrology model developed by Huggins and Monke
(1966).   Channel flow, subsurface drainage, sediment detachment and transport
(i.e., erosion and deposition), and land use and management interactions  were
also included.

     Dr. Jack Burney, while a visiting professor at  Purdue,  rewrote  ANSWERS
into  the  basic  form used today.  The changes were, for all intents and pur-
poses, transparent to the user and had very little effect on the output of the
model.   However, the size of the simulation and the computer time required to
execute it were both reduced, resulting in a substantial savings in processing
costs.   Dr.  Burney  also  added  the  concept  of "shadow" channel elements.
Essentially, this concept allowed for every element to  be  considered  as  an
overland  flow  element with certain of these elements contributing their out-
flow directly to a companion or "shadow" channel element  (which  then  routed
the flow downstream).

     The early versions of ANSWERS utilized the GASP-IV simulation language as
the  basis of the modeling structure.  GASP-IV had many good features, such as
the ability to simultaneously solve  numerous  differential  equations  in  an
implicit  manner  and  the  ability  to  easily take care to the scheduling of
discrete occurrences within a continuous simulation.  However,  the  processor
time and space required by the many subroutines of GASP-IV led to its abandon-
ment in favor of a smaller, somewhat faster system of explicit solution  algo-
rithms.   The newer, FORTRAN-based version of ANSWERS is much more exportable,
since all of the routines required to run the model are internal  and  written
in FORTRAN, which almost every computer facility is capable of running.

     The primary component relationships, although essentially intact from the
original  version  of  ANSWERS,  have been modified to the extent needed to fit
the newer model structure.  In addition, several new components have  been  or
are  in  the  process of being added.  These include: structural practices and
their effects on erosion, sediment movement and runoff water; lateral  ground-
water  movement  (interflow); channel erosion; nitrogen and phosphorus yields;
and several new statistical analyses of the  watershed  data  file.   Specific
component  relationships  are  described  in  detail  in the BlacK Creek Final
Report — Technical Volume (Lake, 1977).

     ANSWERS and the concepts behind  it  are  receiving  increasing  national
attention.   Presently, several organizations are either using or preparing to
use ANSWERS on their own planning projects.  The most notable of these is  the
Honey CreeK Watershed Project in Ohio where the U. S. Corps of Army Engineers,
in cooperation with both USDA and USEPA are modeling the Honey Creek  area  in
an  effort  to  determine the possible benefits of concentrated application of
conservation tillage practices.

                         Use of ANSWSERS in Planning

     ANSWERS was designed to simulate the hydrology, erosion response,  chemi-
cal  yield,  etc.  of  ungaged  agricultural  watersheds.  Due to the use of a
comprehensive descriptive data file and distributed parameters, the model  has
the ability to predict the consequences or benefits of land use and/or manage-
ment changes.  For these reasons, then, ANSWERS has  applicability  in  either
the planning or evaluation areas.

-------
                                    - 22 -

     To date, the validation effort has been aimed at the use of ANSWERS as an
evaluation  tool.  The success of this effort on several watersheds with vary-
ing land use, management, topography, climatic conditions has led to the  con-
clusion  that  ANSWERS should have an equally successful record as an _a priori
planning tool.

     Several examples of planning and/or evaluation programs using the ANSWERS
program  are  available.   Che program, a Special ACP project in Allen County,
Indiana, will be illustrated here to provide an  insight  into  a  methodology
developed  around  the ANSWERS model for planning and evaluating water quality
improvement programs.

     Data gathered on  soils,  hydrologic  and  erosional  response,  nutrient
yields,  etc.  as  part  of  the Black Creek study were available and directly
applicable.  In addition, the personnel in the Allen County Soil  Conservation
Service  (USDA-SCS)  and  Agricultural  Stabilization and Conservation Service
(USDA-ASCS) offices, along with the Allen County Soil and  Water  Conservation
District  (SWCD)  were familiar with and interested in using ANSWERS as a part
of the overall planning effort.

     In order to best utilize very limited monetary and personnel resources, a
planning  methodology  was  developed which would simplify watershed selection
and evaluation tasks.  The planning methodology was divided into four phases:

     1)   Establishment of a "baseline condition",

     2)   Planning of structural, tillage, and  management  changes  necessary
          for treating "critical areas",

     3)   Determination of water quality impacts  caused  by  "critical  area"
          treatment,

     4)   Apportioning cost sharing payments and credits on a cost  effective,
          priority basis.

                       Marie Delarme Watershed Example

     The Marie Delarme watershed in southeastern  Allen  County,  Indiana  has
been  identified  as  one  of 14 watersheds in the county with potential water
quality and erosion problems.  This predominately agricultural  area  is  1203
acres  (487 ha.) in size.  Of this area, 1043 acres (422 ha.)  is in row crops.
The  rest  of  the  watershed  is  designated  as  pastureland,  woodland,  or
homesites.   The  average  slope  in  the  watershed is 1.9 percent with local
slopes ranging from 1 to 6 percent.  Sixty percent of the watershed is  mapped
as  poorly drained silty clay loam soils (Blount, Crosby, and Hoytville).  The
remainder of the soils are the moderately permeable silt  loams  (Raskins  and
Rensselaer).

     The first phase in the planning methodology involved  the  setting  of  a
"baseline  condition".   In  order  to  eliminate  year  to year variations in
predicted benefits, the initial or "baseline" conditions were simulated assum-
ing  that  all of the tillable land (1043 acres)  was planted to conventionally
tilled corn.  Antecedent soil moisture was assumed at field capacity  and  the

-------
                                    - 23 -
 storm  used corresponded  to a  1.5 hour event with a return interval of slightly
 more than 8 years.  These conditions, when  applied  during  an  assumed  crop
 growth  stage   of  one   month after  planting, produce sediment yields at the
 outlet of the  watershed  which approximate the long-term average  annual  sedi-
 ment and particulate phosphorus yields for this land use pattern.

     Once the  "baseline" had  been simulated, a contour map which plotted local
 sediment yield or  deposition was produced.  This map depicted the "critical
 areas" or those areas where delivered sediment  yield  exceeded  one  ton  per
 acre.    With   the  map   in  hand, planners could determine which areas had the
 greatest problems.  In addition, the map could be used  for  gross  siting  of
 proposed BMPs (Best  Management  Practices).   The  location and size of the
 "critical" areas can, in many instances, determine the particular  BMPs  which
 the planners suggest for bettering the runoff water quality.

     After the planners  had several alternative control  strategies  in  mind,
 they   simulated each combination of BMPs using the same storm as the "baseline
 condition".  Although there is usually only one best solution,  there  may  be
 several  control strategies which reduce sediment or nutrients in the stream to
 acceptable levels.

     The final  phase of  the planning methodology involves the actual selection
 of  the   most  cost effective  alternatives.  Also, the setting of variable cost
 share  rates, determined by comparing individual or systems of BMPs to  overall
 watershed response can be performed.  Cost effectiveness, as defined here, is
 a function of  BMP cost divided by sediment yield reduction.  Once the  various
 BMPs  have  been  evaluated for cost effectiveness, they can be ranxed against
 each other and  the watershed  as a  whole.   Those  practices  which  are  more
 effective than the average  could be encouraged with higher than average cost
 shares,  while  the less effective measures could still be encouraged, but to  a
 lesser extent.

   •  Some of the alternatives looxed at in this particular example included:

     1)   Parallel Tiled Cutlet (PTO) terraces were installed where local sed-
          iment yields generally exceeded one ton per acre,

     2)   Chisel plowing was  instituted in all row cropped  areas  with  local
          sediment yields in excess of one half ton per acre,

     3)   Various combinations of PTO terraces and chisel  plowing.
Although  there are many more structural and tillage-based  BMPs   than  the  two
demonstrated, the comparisons were still  quite valid.  The most cost effective
practices were, quite logically,  the tillage-based BMPs.    However,  they  did
not reduce the sediment yield from the watershed to the extent  that structural
BMPs did.  The most effective strategy used a mix  of  the two BMPs with  struc-
tures in  the areas with the highest yields.   See Table 1 for sample results.

     The evaluation criteria, whether cost  effectiveness,   percentage  reduc-
tion,  or  actual   reduction, is  greatly  influenced by the assumptions used in
determining the "baseline condition".   Also,   rather  small changes  in  the
assumed input conditions (e.g., storm intensity or  volume,  soil moisture, sur-
face conditions) can produce much larger  changes in the output.   Although  the

-------
                                    - 24 -
storm  used in these simulations produced the average annual sediment yield in
northeastern Indiana, it might not be applicable in other parts of  the  coun-
try.

     The preceeding example described one methodology  in  which  ANSWERS  was
used  in  both the planning and evaluation roles.  The use of the sediment and
nutrient yield information produced along with  estimated  costs  for  various
treatment strategies can give the planner a very effective tool in vvorKing for
cost effective solutions to nonpoint source pollution problems.

-------
                                    - 25 -
             PRACTICAL USES OF THE ANSWERS MODEL IN BMP PLANNING:
                          AN ALLEN COUNTY EXPERIENCE

                                     .by

                                Daniel McCain*

     Other papers in this proceedings discuss the details of the worK done  in
water  quality  management  and  ongoing research of planning at the national,
regional or state level.  The focus of my presentation will be practical local
use of the ANSWERS computer model that is now setting priorities for conserva-
tion worK as related to water quality in Allen County.  At this time,  we  are
past the 5 year EPA funded BlacK CreeK (1972-1977) demonstrational project and
well into our second year of applying ASCS Special ACP Water Quality money.

     To be successful and get conservation on the land to improve water  qual-
ity,  we've  had  to involve people—not just agency personnel, but the people
that do the farming.  Ultimately, it is the farmers  wno  carry  out  national
objectives  for conservation.  To gain their cooperation, field people have to
bring them together on some mutual basis.  In the humid midwest  (corn  belt),
that mutual interest centers on drainage basins.

     In 1969, when I was assigned to worK in Allen  County,  the  emphasis  of
conservation  worn  in the county—by both the Soil Conservation Service (SCS)
and the Agricultural Stabilization and  Conservation  Service  (ASCS)—was  on
drainage.   More  than a decade later, this emphasis has shifted dramatically.
SCS and other agencies of the U.S. Department of  Agriculture  have  undergone
considerable  change  in how their appropriations are used, and their programs
have also changed.  In Allen County the difference has not been  due  entirely
to the earlier BlacK CreeK experience, but I'm certain that many local changes
occurred as a result of national trends interacting with our staff,  the  soil
and  water  conservation  district supervisors, and farmers during the 1970's.
We have found ourselves on the "cutting  edge"  with  our  local  adaption  of
nationally conceived "non-point source" concerns.

     After the worK in BlacK CreeK was complete, a search was made  for  other
problem  areas  in  the"  county.   On  a  critical  area  map,  targeted rural
watersheds were located that  had  the  worst  water  quality  problems.   The
results  was  a list of 13 small watersheds (1,000 to 3,000 acres)  we call the
"Dirty BaKer's Dozen." Using the ANSWERS model and  ranKing  these  watersheds
according  to gross soil loss per acre, the most important critical areas were
pinpointed.  In 1979 special funds were applied for and received  through  the
Agriculture  Conservation  Program (ACP)  for a water quality project.  Most of
the $75,000 in ACP funds received in 1979 has gone into one of the  13  target
watersheds,  a  1,645-acre area called the Brunson project.  In 1980, two more
critical watersheds were evaluated and the "Dirty BaKer's Dozen"  was reranKed.
This year $100,000 is spread into six (6)  of the thirteen (13)  watersheds.

     A primary reason for this special ACP funding was an innovative  approach
to  determine where planning is needed,  when a group of people come to us for
what they believe is simple-to-define technical assistance with drainage, they
  * District Conservationist, Soil Conservation Service, Ft. Wayne, IN.

-------
                                    - 26 -

don't realize that we're going to try our best to develop  their  appreciation
for  water  quality improvement, as well as solve their drainage problems.  We
are doubly fortunate in that the relationships that began emerging with  Black
Creek  in  1972  and  that  exist  among other locations decisionmaKing groups
reflects a strong commitment to attack problems head-on.

     Groups we are now working with in the Special AGP  Project  approach  are
receptive.   The  ANSWERS  model  is  a big reason why.  Planning with ANSWERS
involves the use of computer drawn maps such as the  "erosion  contours"  map.
This tool points out "hot-spots" or critical erosion areas identified by loca-
tion within the watershed.  A trained conservationist might wonder  why  these
areas  cannot  be located through field work instead of using printouts from a
computer model.  We can, but that's not the point.  A  computer-drawn  map  of
erosion  contours  in  the watershed gives a focus for meetings with groups of
fanners.  Talking with them about the map helps them become more receptive  to
learning  where  and  when  erosion  occurs  because of slope and concentrated
runoff.

     In meetings with the group, we emphasized the proximity of erodible lands
to  the drainage outlets.  When a conservation planner presents the group with
a few conclusions from this fact, farmers can readily visualize where  erosion
is  causing water quality problems.  The total sediment yields computer by the
model represent a net loss of topsoil through the "mouth" of the watershed.

     A 20- by 40-inch blowup  of  the  Universal  Soil  Loss  Equation  (USLE)
sliderule  calculator has been used before several groups.  The calculator has
not intimidated the groups:  by  using  such  visual  aids,  highly  technical
matters are comprehensible to fanners.  Our partnership with the groups has to
be educational-on both sides—to be effective.  Taking the group  on  a  field
trip to see opportunities and to recognize potential benefits also help.

     In conservation planning sessions with the groups-we have  tried  not  to
put  all our eggs in one basket, for example, with conservation tillage.  Suc-
cess with conservation tillage depends on many variables.  It was not possible
to  persuade  many farmers to convert to conservation tillage during the Black
Creek era simply because of these variables.  For example, climatic  variabil-
ity  over  a  4-year rotation might bring a wet spring, a dry spring, an early
spring, and a late spring.  Climate and other variables require the farmer  to
make daily decisions that can complicate his tillage plan.

     Not that I'm negative about conservation tillage—in fact  it's  the  one
practice  that  can  touch  every  acre—but  I've seen what can happen when a
farmer tries it without fully understanding it.  It may require him  to  adapt
his  equipment and make other changes.  Every spring day the farmer can face a
different set of weather conditions,  crop  prices,  operating  expenses,  and
other  things  which  he  has  no control but on which he must form decisions.
Therefore, it is important to offer the farmer conservation alternatives  that
won't  add  to  his burden of daily decisionmaking.  Seen positively, however,
conservation tillage is not a burden at all but an investment in wise  manage-
ment that pays off in savings in fuel, labor, time, water, and soil.

     Even so, practices such as terraces serve  as  permanent  "reminders"  of
conservation  on  the  landscape.  These practices can be very compatible with

-------
                                    - 27 -


 conservation tillage; most  important, however, they signal  a  "commitment  to
 conservation"  and become symbols of the group's progress in understanding and
 dealing with erosion and sedimentation.  And  if nature provides  a  disastrous
 spring for tillage, at least part of the conservation system will function.

     Permanent conservation practices require the farmer to make  fewer  deci-
 sions,  although  they  won't necessarily be any easier to make than decisions
 about tillage.  For example, if a  farmer  wants  to  construct  terraces  and
 waterways  and  needs financial help, he requests cost-sharing assistance from
 the ASCS office.  SCS provides an engineering plan and gives him a cost  esti-
 mate,  and  his  thoughts come down to a one time "yes or no" decision.  If he
 decides to go ahead with the practices, a  contractor  builds  them  and  they
 become  a  permanent  facility.   The  only questions remaining is whether the
 farmer will permanently maintain the practices.

     In all, then, there are three ways we can tackle cropland  erosion  prob-
 lems  related  to  water  quality.  First, we encourage changes in TILLAGE and
 planting techniques.  Second, we encourage CROP ROTATIONS that are  compatible
 with the farmer's present tillage system.  Or third, we suggest permanent land
 treatment practices such as TERRACES, which reduce slope length  and  increase
 temporary storage capacity  for runoff.  If the farmer selects any one of these
 changes—or some combination of them—improved water  quality  should  be  the
 result.

     When we targeted critical areas in the Krunson project we had to go after
 the  job  from  the top of  the hill down.  We didn't want to repeat an earlier
 experience in BlacK Creek;  that is, overselling the group on  what  they  were
 already  prepared  to  request—outlet development.  Also, in Black Creek more
 streambank protection than  necessary may have been installed  because  of  the
 farmers'   concern  about highly visible streambank erosion.  Black Creek find-
 ings showed that only about 7 percent of the sediment load entering the Maumee
 River  was caused by streambank erosion.  However, the fanners noticed stream-
 bank erosion more than they noticed sheet and rill erosion  on  sloping  crop-
 land.

     In planning with the groups and in orienting them to the kinds  of  prac-
 tices needed, ASCS and SCS  can provide farmers with cost sharing and technical
 help as far down the outlet as necessary to make a properly  functioning  pro-
 ject.   In  the  Brunson project, individual terrace outlets were safely taken
 down the watershed through  tile beside group grass waterways, and into a  pro-
 tected mutual open outlet.  All these practices helped as part of a protective
 scheme for the critical areas.

     It may also be necessary to study the  channel  far  enough  through  the
 critical   areas  to  find the unstable segments.   In Black Creek, we were most
 successful with a practice we call "training," that is,  putting  rock  riprap
 low  on  the channel banks  in unstable soils and  installing a 1 1/2- to 2-foot
drop structure to lessen channel grade.

     From the top of the watershed down,  an  opportunity  and   an  obligation
exists to explain technical  alternatives to the farmers.   The ANSWERS model  is
useful for these explanations  because  it  graphically  depicts  the  eroding
areas.   An overlay of the erosion contours map with an ownership map lets the

-------
                                    - 28 -
fanner see whether he has an erosion problem  that  requires  attention.   But
don't  tell him, "LOOK, you dirty farmer, you're causing all these water qual-
ity problems and you're going to have to  do  something  to  clean  them  up."
Instead,  approach  the group in a positive way by showing them the beneficial
things that they can do—both as a group and as individuals—to improve  water
quality and reduce the sedimentation on their neighbors' lands downstream.

     In some cases, the approach to land treatment in the Brunson project  was
turned  180  degrees  from the previous approach in Black Creek, where outlets
were usually developed first.  In the Brunson project, we started at  the  top
of  the watershed by securing commitments from fanners for cropland treatment.
Of the $75,000 allocated for the Dirty Baker's Dozen in  1979,  ASCS  approved
$60,000 for cost sharing of group parallel tile outlet (PTO) basin terraces in
the Brunson watershed.  To make the terraces work, waterways were  constructed
and outlets developed to handle the metered tile flow from the terraces.  Most
of the job was completed in 1979.  A second smaller group—SOUTHWEST  BRUNSON-
-with  some  of  the  same farmers, completed additional terraces and a mutual
waterway in the summer of 1980.

     The ANSWERS model could prove even  more  valuable  when  used  with  the
analysis of BMP's on farms and as a group.  The hypothetical watershed (figure
1) and reduction estimates (table 1) illustrate a means of using  the  ANSWERS
model  in planning.  Perhaps the time will come when the public will buy water
quality improvement with "dollars spent for tons saved."

                Table  1.  Effect of BMPs  in  reducing  sediment
                                Primary
                                  BMP
                       Level (Ave/Farm)
Group
A

Individuals
All Plus
Samuels
Cells
100
20
Application
Outlet
Livestock
exclus .
Initially
1350
1500
with BMP's
675
700
Reduction
50%
53%
 Sub
Group   Smith
  B     Fry

Sub
Group   Sharp
  C     Jones
        Green
        Gray
        Johnson
17
13
 6
14
 8
10
12
None
None
Waterway
Terraces
Tillage
Terraces
None
2000
   0
 700
1000
1200
 900
1200
2000
1000
 500
 400
 600
 400
1200
 0%
 0%
28%
60%
50%
55%
 0%
 Sub
Group   Jones
  D     Green

 Sub
Group   Gray
  E     Johnson
14
 8
10
12
Terraces
Tillage
Terraces
None
1000
1200
 900
1200
 400
 600
 400
1200
60%
50%
55%
 0%

-------
                        - 29 -
 SMITH
FRY
         V
                    JONES
          GREEN
              \
SAMUALS
GRAY
JOHNSON
             Figure 1.  »/P°thetical Watershed

-------
                                    - 30 -                     •-


                BIOLOGICAL PERSPECTIVE ON WATER QUALITY GOALS


                   By James R. Karr and Daniel R. Dudley ^


     Increased societal concern for tne state of tne  environment  is  clearly
manifest in tne proliferation of environmental legislation in tne past decade.
Early efforts directed towards pollution control focused on point  sources  of
pollution  because  of  tne ease witn which tney could be controlled and regu-
lated.  As tne magnitude of tne point source  problem  reduced,  tne  relative
contribution of nonpoint sources expanded.

     Concern for tne degradation  of  water  resource  quality  from  nonpoint
sources  has been a matter of special concern since passage of tne Water Qual-
ity Act Amendments of 1972 which called for the restoration and maintenance of
"the  chemical, physical, and biological integrity of the Nation's waters." In
the following pages, we evaluate the  progress  made  in  attainment  of  that
objective as a result of studies in tne BlacK CreeK watershed.

BlacK CreeK: The Central Question

     From its inception, the central question to be addressed by tne  nonpoint
control program in tne BlacK CreeK watershed was: Are traditional erosion con-
trol programs sufficient to not only reduce erosion but also  to  improve  the
quality of our water resources?  That is, is it possible to implement a volun-
tary program of erosion control in an agricultural watershed and thus  control
water  resource  problems  resulting from nonpoint pollution from agricultural
lands?  It is now clear that tne answer to that question is  a  distinct  no!!
That  is  not to say tnat some incremental improvement cannot be obtained witn
tne traditional approach.  Rather, attainment of tne objective of PL 92-500 is
not possible witn that conventional approach.  In tne following pages, we out-
line the primary deficiencies of that approach as well as suggest an  alterna-
tive  conceptual model which will produce a more comprehensive solution to the
nonpoint problem.

     ffowever, before we detail conclusions and results of the BlacK CreeK stu-
dies, several concepts should be clearly outlined.

Fishable and Swimmable

     The concept of "fishable and swimmable", first introduced in  Public  Law
92-500,  is a desirable, but ambiguous objective.  Fishable in tnis context is
often defined as maKing the stream useful to fishermen in capturing  sport  or
commercial  fish.   However, since many small streams contain too little water
to be used for swimming or to support a sport or commercial fishery, cney  are
often  discounted as not having any significance to tne fishable and swimmable
objective.  We feel that it is inappropriate to measure tne value of a  stream
reach  based  on tnis particular component of fisna'ble and swimmable criteria.
That quality must be more broadly defined than hoo< and line  locally  because

 1. Department of Ecology, Ethology and Evolution, University of Illinois and
    Division of Surveillance, Ohio Environmental Protection Agency

-------
                                    - 31 -


the importance of headwater streams to downstream reaches (in terms of produc-
tion  of  fishable  benefits  downstream)  is underemphasized in that context.
Although a headwater stream may never be fisnable, it is an integral component
of  the  watershed; its preservation is essential for downstream reaches to be
fishable and swimmable.  The biological integrity mandate of Public Law 92-500
depends  on  an  integrative  view  of the entire water resource system at the
watershed level rather than its consideration for local reaches of the stream.

Water Quality vs. The Quality of a Water Resource

     Planners, scientists, politicians and the general public commonly use the
phrase  "water  quality".   Almost  invariably, the use of that phrase implies
physical and chemical conditions of the water.  These include  such  items  as
temperature, dissolved oxygen, nutrient levels, and concentration of suspended
solids, heavy metals, and toxic chemicals.  The assumption is  generally  made
that  improvement in "water quality" will result in optimization of the widest
range of water uses by society,  (domestic,  industrial,  irrigation,  agricul-
tural, recreation, aesthetics).

     In addition, it is assumed that there is nothing else that  society  need
do  (or  can  do)  to improve the quality of water resources.  As we will show
below, both of these assumptions are false; their  continued  acceptance  will
result in progressive and continuing decay in water resources.

Loadings vs. Concentrations

     Commonly, the sole point of focus of efforts to model nonpoint  pollution
is  loadings (commonly annual loading, the total sediment or nutrient per unit
area exiting a watershed) .  It is clear that raa]or storm events play the  pri-
mary  role  in  determining  annual sediment or nutrient loading.  Tnese major
transitory events are especially important for  consideration  of  effects  on
downstream  areas,  particularly receiving waters downstream—natural laxes or
reservoirs.  However, it is also important to note that the. average conditions
expressed  in smaller runoff events, and even during base or low flow periods,
play a major role in governing the characteristics of stream communities  and,
thus,  the  biological integrity of a water resource.  More careful considera-
tion must be given in all monitoring and modelling  efforts  to  the  relative
merits of emphasizing concentration information throughout the year as opposed
to the loading information which is determined by a few transitory events.

     Both loading and concentration data during all flow  conditions  must  be
monitored and evaluated.  The relative emphasis on the two will vary depending
on the nature of the water resource problem under consideration.

Selection of Monitoring Sites

     The selection of monitoring locations plays a major role  in  determining
the reliability of water quality parameters arid water quality conditions meas-
ured for a watershed.  The selection of sanple sites  which  are  adjacent  to
bridges is often convenient and can be defended in many cases on those grounds
alone.  But, if there is some particular activity immediately upstream of that
site  which  significantly increases or decreases suspended  solids loads, con-
clusions about sediment and nutrient dynamics may  be  very  misleading.   For

-------
                                    - 32 -


example,  we  had one sample site with unusually high suspended solids concen-
trations.  We found that a couple  duc.
-------
                                  - 33 -
                             PROTECTIVE
                           ENVIRONMENT
                            (natural  areas)
  PRODUCTIVE
ENVIRONMENT
 (agriculture)
 COMPROMISE

ENVIRONMENT

 (classified
      channel)
                                URBAN-
                              INDUSTRIAL
                           ENVIRONMENT
          Figure 1.  Compartment model of the basic Kinds of environment
          required by mans partitioned  according  to ecosystem development
           and lifecycle resource criteria.  (Modified  from Cdura 1969).

 urban-industrial environments.  As will be discussed  below,   the  compartment
 model is useful  for addressing, in operational  terms, strategies of innovative
 soil and water conservation management.

 Stream Ecosystems

     An individual -stream or section of a stream  is not  an  isolated  system.
 Streams  and   rivers are  open ecosystans with dynamic imports and exports of
 nutrients,  energy, and  water  (Big.  2).   Major  changes  in tne  inputs  to
 upstream  (headwater)   areas  are   ultimately carried to and  affect downstream
 areas.  Further, some aquatic life, especially  fishes, may depend upon  migra-
 tion  to upstream or downstream areas for tne completion of their life cycles.
 The concept of the open ecosystem  has two important  management  implications.
 First, streams are subject to rapid and gross perturbations caused by land-use
 cnanges (urbanization,  intensive agriculture, etc.).  Second,  properly managed

-------
UPSTREAM
AND  LAND
SURFACE
           - 34.-
     TYPiCAL
     STREAM
     SECTION
DOWNSTREA
                                 I

"FR
-
— *s
"^
^

A >-"X 1 1 A T" 1 /*"*


   IMPORTS
NUTRIENTS
                              ENERGY
                                                 BIOTA
       EXPORTS
                 Figure 2.  Generalized flow diagram for aquatic ecosystem.

       land-use in watersheds can effectively and   rapidly  lessen  perturbations   in
       stream systems.

            A classification system developed by Horton (1945)  and modified by Kuehne
       (1962)  is  commonly  used by  aquatic  bioligists to discuss the progressive
       increase in stream size.  According to this  system, the smallest streams in  a
       watershed  are  first  order.   When two first-order streams join, they form a
       second order stream, when two second-order streams join,  they  form  a  third
       order stream; etc.  Ecological discussions of streams typically consider three
       size classes: the headwaters (1st and 3rd order),  intermediate-sized  rivers
       (4th  to  6th order), -and large rivers (7th and larger orders).  While this
       classification system is generally useful, note that stream order effects may
       vary  somewhat among watersheds.  For exanple, differences in size of upstream
       watershed or watershed topography may affect tne nature  of  the  stream-order
       pattern.

            Man alters streams by dredging new channels in poorly drained areas or  by
       modifying  existing  natural channels.  These man-engineered watercourses must
       be considered streams even though they  are  clearly  different  from  natural
       streams  in  many  respects (i.e., drainage  and flow characteristics, chemical
       and physical parameters, bottom type, etc.).  Important as these  differences
       are,  one  basic  ecological principle applies to both man-altered and natural
       streams; water, nutrients, and energy are exported to downstream areas.  Thus,
       man's  construction  of drainage ditches is  not separate form natural drainage
       patterns; rather, it is only an addition to  or a modification of  the  natural
       stream  network that profoundly effects water resources both locally and down-
       stream.

-------
                                   - 35 -
     We have been able to  identify what we feel   are  four  major  classes  of
variables  (Fig.  3)   which,  when  modified by man's activities, play primary
roles
          FLOW
          REGIME
          ENERGY
          SOURCE
          WATER
          QUALITY
          HABITAT
          STRUCTURE
BIOLOGICAL
INTEGRITY OF
AQUATIC   BIOTA
          Figure 3.   Primary veriables affecting  the structural and
            functional aspects of the biota of a headwater stream.
                     (Modified from Karr  and Dudley 1978)

in determining  the characteristics of  the  biota  of  running  water   (lotic)
ecosystems  (Karr  and  Dudley  1978).   These are water quality, flow  regime,
habitat structure, and energy source.

     Any project meant to address the mandates of PL 92-500 must address defi-
ciencies in all of these to insure the optimization of water resource quality.
Consideration of "water quality"  characteristics alone, as has  been  the  case
in  Blacx  Creek and many other NFS efforts, cannot be expected to produce the
water resource  quality which seems to be  the societal  objective  mandated  by
recent water resource legislation.

Flow Regime

     Fluctuating water levels are an itegral part of all stream ecosystems and
aquatic  organisms have evolved to compensate for changing flow regimes.  Even
areas decimated by catastrophic floods or droughts are  often quicKly  recolon-
ized.   But modifications of the  land surface with changing land use typically
results in flood peaKs and low-flow periods that are more  severe  as  well  as
more  frequent. Late summer low-flow periods may be extended while hydrograph
peaKs following runoff events  are often of shorter duration.

-------
                                    - 36 -
     High water periods are determined by the frequency,  occurrence, and   type
of  rainfall  event,  the timing of those rainfall  events,  and  such antecedent
conditions as soil moisture, time since tne last rain,  and  amount  and  type of
soil cover.  Flood events in natural watersheds tend  to have  a  dampened hydro-
graph, while those in modified watersheds tend to have  a  sharp   and  extreme
peaK.   Low flows in natural watersheds tend .to be  severe only  in  particularly
dry years, while low flow periods in modified watersheds  are  relatively   more
severe,  especially during late summer and early fall periods wnen rainfall is
at relatively lower levels in midwestern portions of  the  United States.

     When such flow events prevent seasonal migrations  of  fish or  interfere
with  egg  or  fry  development, irreversible catastrophic  changes may result.
Under the extreme condition of dewatering, the biota  may  be  lost  entirely.
Recognition of the significance of this problem has precipitated tne formation
of a special group within tne Office of Biological  Services of  the U.S.   Fish
and  Wildlife  Service.   This  group,  the  Cooperative  Instream  Flow Service
Group, is developing a detailed methodology for evaluating  flow   requirements
of  aquatic organisms.  Their primary objective is  the  development of  criteria
to allow assessment of the impact of altered stream-flow  on  habitat  charac-
teristics  and  the  use of an area by aquatic organisms  (StalnaKer and Arnett
1976).  They seex to identify the hydraulic conditions  necessary for a variety
of  groups  of  organisms, including different age  classes  of the  same species
when their requirements differ.  For example,  the probability distribution of
walleye  with respect to stream velocity varies among the age classes  and  with
reproductive state of fish (Fig. 4).
                    JD
                    03
                    -Q
                    O

co
d

o

0

(VJ
0

]\ f
\ i
"IV '
i TFRY
r! /
\
\
\
\
\
1 /
ArSPAWNING \
/'
f /
f*«
v\ /
i
i


-.-[ •N^!liv!NlLES..'^.DU LTS
                               1       2      3
                             Velocity (ft/sec)
        Figure 4.   Probability of use  curve  for  several age classes of
       Walleye.   (Adapted from unpublished  material of the Cooperative
        Instream Flow Service Group, with permission of C. StalnaKer.)

Fry are found in only  the  slowest water   while  juveniles,  and  especially
adults, utilize higher velocities.  Finally, spawning  fish require mucn higher
flow rates.  Modifications in a stream wnich destroy  areas  with  "spawning"

-------
                                    - 37 -
 velocities  may  nave  a  significant  negative effect on walleye reproduction
 although adult fish may not be directly affected.  These  efforts  to  examine
 the   flow  regimes  and  hydraulics of streams and their effects on biological
 integrity will maxe major contributions to the  management  of  running  water
 resources.

 Water Quality

      In recent years most efforts to reverse the  degradation  in  quality  of
 water resources  have  focussed  on  the  physical and chemical properties of
 water.  Temperature, dissolved oxygen, concentrations of soluble and insoluble
 organics  and inorganics, heavy metals, and a wide variety of toxic substances
 are components of special interest.  They may affect biological  integrity  by
 directly  causing mortality or may shift'the balance among species as a result
 of subtle effects such as reduced reproductive rates or  changing  competitive
 ability.

      The importance of these factors on stream biota is widely  Known  (Warren
 1971, Hynes 1974).  Water quality factors which are of special concern include
 light, temperature, dissolved oxygen, suspended solids,  dissolved  ions,  and
 other materials.  These play critical roles in determining an area's suitabil-
 ity for aquatic organisms.  In addition to the average condition, extremes and
 their temporal pattern have important impacts on the biota.

      Each of these are of concern individually.  However, in  many  watersheds
 liKe  BlacK CreeK, human activities may precipi-tate problems of degradation of
 biotic interity because of the synergistic effect of  several  variables  (see
 discussion of algal blooms below).

 Habitat Structure

      The physical structure of the environment also  plays  a  major  role  in
 determining  the  number and Kinds of fishes and other organisms that can sur-
 vive  in a stream.  Channel geometry in natural watersheds typically includes a
 meandering  topography,  with  substrate  diversities  created by varying flow
 regimes length-wise along the stream channel  and  across  the  channel.    The
 result  is  substrate  sorting, the presence of pools and riffles,  erosion and
 deposition areas, and ultimately a dynamic  equilibrium  between  the  flowing
 water and its substrate.  Modified watersheds, on the other hand,  tend to have
very much reduced diversity in channel geometry;  they are often  straightened.
 Channel  maintenance activities commonly create a uniform substrate and reduce
depth diversity in the absence of pool and riffle  topography.   In  addition,
sedimentation increases due to a disequilibrium channel and/or because of ero-
 sion  from the land surface.

     Straight open channels in the presence of abundant  nutrients,   sunlight,
and  high  temperatures  creates ideal conditions for the choKing  algal blooms
which are such an obvious component of BlacK CreeK in late summer.    In  years
of  lower rainfall in late spring  and  early summer,  these algal  blooms develop
 in late May and  early June;  in years with  more substantial  rainfall during  the
early  summer,  the  algal blooms  are  curbed  by the  flushing  action of channel
 flow.

-------
                                    - 38 -
     These and other complex interations with the physical habitat of  streams
affect  the  biota  of the stream.  Bottom-dwelling invertebrates such as mol-
luscs (Harman 1972) and insects (Allan 1975)  seem to be especially affected  by
the  diversity  and  sorting  of  boctom  or  substrate types in an area (sand,
gravel, rocKs, etc.).  Substrate particle size  determines  the  size  of the
interstitial  spaces  which,  in  turn, affects the amount of water and oxygen
available to the bottom-dwelling community.    Adequate  interstital  space   is
also essential for the movement and feeding of aquatic invertebrates.  Fishes,
vtfuch use environments in a more three dimensional fashion, seem to respond  to
a  complex of structural features including substrate type, depth, and current
velocity (Gorman and Karr 1978).  Further, many fishes and some  invertebrates
require  places  of  concealment  (cover)   as  feeding locales or as places  to
escape predation.  General cover types  include  undercut  banks,  timber and
brush  snags,  and  aquatic vascular plants.   Without essential habitat struc-
ture, many forms of aquatic life are eliminated from streams.  However, as the
variety  of  habitat  conditions increases with the development of pools, rif-
fles, meandering topography, and the sorting  of substrate sizes, nabitat com-
plexity increases and supports a wider diversity of fishes (Fig. 5.).
 CD

TJ
.C

iZ
                     2.5
                    2.0
                     1.5-
                       1.0     2.0      3.0     4.0
                          Habitat   diversity
      Figure 5.   Relationship between habitat diversity and  fish species
                   diversity.  (From Gorman and  Karr  1978).

     In addition to the general  dependence  of  fish  community  structure  on
habitat characteristics, there is a more  subtle  significance to habitat struc-
ture.  Early in our BlacK Creex  efforts,  we noted major seasonal migration  in
the  watershed  (Karr  and  Gorman  1975).   In  addition, we noted less easily
explained movements which seemed to vary  in magnitude  among habitats  within
the  watershed.    This  stimulated a study of the movements  of fish  in several
stream reaches.   Fish were marxed  with   a  procedure  called cold  branding.
Silver  brands in shape of various letters were  supercooled  with liquid nitro-
gen and touched to the sides of  fish duplicating the  common  hot branding  used
to  marK  cattle  and  the livestock on open ranges.  We branded fish in three
major habitat types.  Three sampling stations were  selected  in the main  chan-
nel  of  BlacK  CreeK  in  areas  that nad  been   subjected to major channel

-------
                                   -  39 -

alterations  (Stations  12,  26,  and 29)  early in the  Black  CreeK  study.   The
second  major   habitat  was on tne Wann Drainage immediately east of tne BlacK
CreeK watershed (Station 13).   Although  there  has  been  no  recent  channel
modification worK  in this area, the stream reach had been modified perhaps ten
years earlier.   The lacK of disturbance over the years created a stream  which
had  begun  to   meander   in its channel base and in which dense vascular plant
populations provided cover. As reported earlier by Gorman  and  Karr  (1978),
this  section   of  stream  contained a richer fauna tnan that found in similar
reaches of the  BlacK CreeK watershed.   Our final study area was a  section  of
the  Wertz  Drain  where it traversed  the woodlot called Wertz Woods (Karr and
Gorman 1975),   This area has an especially rich fauna (Gorman and Karr 1978).
 1);
     Populations  in  higher  quality habitat are relatively more  secure   (Table
Table 1.  Recapture  rates,  habitat diversity and stream channel conditions
          at several  sites  where  fish were marxed by cold branding.
                  Channel and
                   Habitat
                  Conditions

                  Badly Disturbed

                  Disturbed, but
                   Recovering
   Wertz Drain     Relatively
   in Wertz Woods   Natural
 Habitat
Diversity*

  2.89
  3.05
  3.31
Number
of Fish
Marked

1,190
                  767
                  958
Percent of
   Fish
Recaptured
                                  15
                                  37
   *Data for June 1975 using the information-theoretic measure of diversity for  the
   composite of bottom, depth,  and  current velocity.  See Gorman and Karr 1978 for
   more detailed explanation of methods.
they are able to survive locally over  longer  periods.   Clearly, total emphasis
on  water  quality  in  the  physical-chemical  sense will not overcome habitat
structure deficiencies.  Further, we have provided  evidence in earlier reports
that  those areas with better quality  habitat also  have a beneficial effect on
water quality (Karr and Gorman  1975, Karr and Schlosser 1977,  1978,   Schlosser
and Karr 1980).

     In another study one of my (JRK)  graduate  students at the  University  of
Illinois  (P.  Angermeier,  pers.  commun.) has divided two sections of Jordon
CreeK in east-central Illinois  with 1/4"  mesh  hardware  cloth  supported  by
steel  posts.   Cn  one  side   of each section  all  cover features (e.g., logs,

-------
                                    - 40 -


limbs)  were removed from in or near the water.  On the other side, a  continu-
ous  series of similar objects was secured along the stream.  In. July and Sep-
tember, samples of the biomass of fish was 4.8 to 9.4 times  as  high  in  the
areas with structurally complex habitats.

     Further, the large fish, and  especially  the  top  predators  tended  to
select the structured habitat.  In this case we Know water quality is constant
in the structure and unstructured sides of the stream, yet the numbers of  the
fish  are  markedly different.  These improved habitat conditions seem to pro-
vide two things: habitat for small fish including a  diversity  of  substrates
for food organisms and hiding places (cover)  from which large fish can prey on
smaller species.  This again emphasizes the importance of habitat structure as
a determinant of biotic conditions in a stream.

     Note that, to a great extent, the hydraulics of flow  regimes  determines
the  physical  structure  of  stream  habitat,  and,  thus, the efforts of the
Instream Flow Group will clarify the problems of stream  management  for  both
flow regime and habitat structure.

Energy Source

     In stream ecosystems the form and source of the energy and nutrients  are
especially  important  in  determining  ecosystem characteristics.  The energy
contained in the chemical bonds of organic matter is the basic  energy  source
for  animals,  fungi, and many bacteria.  The process of breaxing the chemical
bonds to release energy an simpler compounds is  respiration.   Production  is
the  reverse process in which energy in the form of solar radiation and simple
compounds are converted into complex organic compounds.  Obviously, plants are
the  ma}or  producer  organisms  and  high production rates are dependent upon
abundant sunlight and essential nutrients.  The fundamental  energy  relation-
snip  can  be  expressed by the production (P) to respiration (R) ratio: P/R>1
when  production  exceeds  respiration  (autotrophy),  P/R<1  when  repiration
exceeds production (heterotrophy).  In streams, this basic energy flow charac-
teristic is sensitive to the organic loading from the terrestrial environment,
the  anount  of  sunlight and nutrients, the form or availability of nutrients
(simple compounds vs. complex organic compounds), and a number of  other  fac-
tors such as turbidity.

     Studies of the energetics of stream ecosystems (Cummins 1974) stress pro-
cess  oriented  attributes  such  as  production,  respiration,  energy  flow,
nutrient cycling, and trophic dynamics.  It is a  fundamental  postulate  tnat
many process oriented attributes of running water ecosystems change as streams
increase in size from headwaters to mouth.

     The transition from small headwater areas to major rivers is referred  to
as  the  stream  continuum.   Structural  and functional attributes of natural
stream ecosystems change along this continuum  (Table  2).   These  attributes
serve  as reference points to assess the status of the stream ecosystan in any
location.  If the ecosystem degradation resulting from these expectations,  it
may  be  due to ecosystem degradation resulting from man's activities.  At the
very least, it suggests that more detailed study is required.  The theoretical
foundations  for  these  "reference  points"  comes  to  a  great  extent from
forrested watersheds,  as  a  result,  it  may  be  necessary  to  develop  an

-------
Table 2.    General characteristics of running water ecosystems according to size of stream.  (Modified from cummins 1975).



size
Small
headwater
streams
(stream
order
1-3)


Medium
sized
streams
(4-6)



Large
rivers
(7-12)

Primary


source
Coarse particulate
organic matter
(CPOM) from the
terrestrial
environment

Little primary
production
Fine particulate
organic matter
(FPOM), mostly

Considerable
primary
production
FPOM from
upstream


Production


state 1
Heterotrophic

P/R <1





Autotrophic

P/R >1




HeterotropUic

P/R <1

Light and


regimes
Heavily
shaded

Stable
temperatures



Little
shading

Daily
temperature
variation high

Little shading

Stable
temperatures
Trophic status of dominant


Insects Fish
Shredders Invertivores

Collectors





Collectors Invertivores

Scrapers Piscivores
(grazers)




Planktonic Planktivores
collectors

                                                                                                                                     I
                                                                                                                                     .p-
1.    A stream is autotrophic if instream photosynthesis exceeds the respiratory requirement of organisms living in
     the area (i.e., P/R >1).   It is heterotrophic if import of organic material from upstream areas or the land
     surface is necessary,  (i.e., P/R <1).

-------
                                    - 42 -


alternate  foundation  for  raarKedly different terrestrial environments in the
dry nonforested regions of western Nortn Anerican (Minsnall 1978).

     Headwater streams in natural watersheds are usually neterotropnic.   That
is, they have production to reaspiration ratios (P/R) of less than 1.0 and are
dependent on food produced outside tne stream (allocthonous material).   Dense
tree  canopies shade tne headwaters so that instreara production is minor, gen-
erally from small populations of moss or periphytic algae (algae  attached  to
rocKs or other substrates).  One study in a New Hampshire watershed (deciduous
forest)  snowed that 99% of the energy requirements for tne biota of a  headwa-
ter  stream  was of allocthonous origin (Fisher and LiKens 1973).  A very dif-
ferent watershed in Oregon (coniferous forest)  demonstrated tne  same  general
pattern (Sedell et al. 1973).  In this situation the persistence of the biotic
community depends on a regular input of food (organic  matter)   from  external
sources.  The terrestrial environment supplies much of -the energy input in the
form of leaf liter shed in predictable seasonal  pattern  (fall  in  temperate
deciduous forest; dry season in tropical forest).

     The particle size of organic matter entering a stream is just  as  impor-
tant  to stream ecosystem functioning as the amount, type, or timing of energy
input.  In undisturbed headwater areas, the terrestrial  environment  produces
particulates  of relatively large size (such as leaves, twigs,  etc.) , referred
to as coarse paticulate organic matter (CPOM).    Bacteria  and   fungi  quicKly
colonize the CPOM and, as a result of their metabolic activity, speed the pro-
cess of fragmentation into small particles—fine  particulate  organic  matter
(FPQyi) .   (Any  organic  particle  less  than 1 millimeter in diameter is con-
sidered FPOM, regardless of its source.)  The  breaKdown  process  of  CPOM  is
accelerated  by benthic invertebrates, primarily aquatic insects, which ingest
and further fragment (or shred) the  CPOM.   Organisms  with  this  functional
capacity are called shredders.  Shredders utilize some of the energy contained
in the CPOM along with the rich growths of attached bacteria and  fungi.   But
most  of  the  CPOM  is  simply  converted to FPOM and is available for use by
another functional group of aquatic organisms called  collectors.   Collectors
either  filter  FPOM   from the water or gather it from the .sediments (Cummins
1973).  Because of structural adaptations, most  collector  organisms  utilize
FPOM  only  within  a  narrow size range (Cummins 1974), thus illustrating the
critical nature of particle size in stream ecosystems.  The  natural  associa-
tion  of  shredder  and  collector organisms in headwater streams results in a
highly efficient utilization of energy (organic matter) input.   Cummins (1975)
has  estimated  that  the biota processes about 80% of the particulate organic
matter (POM)  and 50% of the dissolved organic matter (DOM)  in natural first to
third order streams.

     Functional attributes are marKedly different  in  undisturbed  intermedi-
atesized rivers.  The stream becomes autotrophic (P/R>1) as the stream becomes
less shaded and algae and vascular plants increase in abundance.  CPOM  inputs
are reduced, resulting in decreased shredder abundance.  Incoming allocthonous
material is primarily FPOM from headwater areas and  a  variety  of  collector
organisms  are  common.   The autotrophic status of the stream  account for the
presence of a third functional group of aquatic macroinvertebrates.  These are
the  scraper  or  grazer  organisms that exploit periphytic algae and vascular
plants.  A few scrapers can always be found in natural headwater  streams  but
their abundance is severely limited by the low rate of primary  production.

-------
                                    - 43 -


     In large  rivers  (7tn and 12tn order) the  stream  again  becomes  hetero-
trophic due primarily to increased turbidities reducing light penetration and,
therefore, tne potential for photosynthesis.  Tne primary 'production that does
occur  is generated by pnytoplanxton (free-floating algae).  Free-floating col-
lectors (zooplanKton)  are  also  -present,  utilizing  tne  phytoplanKton  and
suspended  FPOM as food.  Collectors also predominate in tne sediments as FPOM
is tne major energy source.  Few scrapers or shredders occur in a large  river
environment.

     The fisn fauna also reflects the energy sources available  in  a  stream.
However,  fish can be more directly related to the value in human terms of the
water  resource (commercial and sport fish).  Cummins  (1975)  categorized  the
functional  attributes of fish communities according to the food habits of the
dominate fish.  Predominate food habits are somewhat different for  the  three
major  ecological areas of an undisturbed river system.  In headwater streams,
fishes that feed upon macroinverrtebrates (invetivores)  are dominant.   Inver-
tivores  along  with  piscivores  (fisn  that  consume  other  fish)  dominate
intermediate-si zed rivers.  Finally, in large rivers dominate members  of  the
fish  community  are  planKtivores (fishes feeding upon both pnytoplanKton and
zooplanKton).  Two additional categories are omnivores (consuming  both  plant
and  animal  matter in approximately equal portions)  and herbivores (consuming
primarily plan materials).  Qnnivores and herbivores are  rarely  dominant  in
natural running water systems.

     Our results in BlacK CreeK indicate major disturbances  in  these  energy
source  (functional)   dynamics.  Many of the modified channel areas seem to be
autotrophic rather than heterotrophic.(Table 3).because of  the  abundance  of
sunlight  and nutrients.  The abundant algal blooms alter the organic load and
habitat characteristics of the stream.   Research in needed to  determine  what
level  of  autotropny  can  be  tolerated  without a distruption in biological
integrity.

     The trophic status of the  aquatic  invertebrate  community  has  changed
(Karr  and  Dudley  1978)  in  response  to a variety of factors.  The organic
matter processing efficiency in the disturbed  headwater  system  is  modified
thus increasing organic loading to downstream areas.

     The trophic status of the migrant fishes has shifted from  priscivore  to
omnivore  because  of  declining  water  quality and  stream habitat structure.
This has increased the populations of the  less  desirable  fish  species  and
decreased  the  nunber of top predators that act as a natural population checK
on other species.  (See Karr Dudley 1978 for more detailed documentation using
the Maumee River watershed).

     In summary,  then, we find that the onslaught  of  human  effects  on  the
biotic  integrity of the water resource system of BlacK CreeK is affected by a
diversity of factors and not  just  water  quality  in  the  physical-chemical
sense.  Briefly,  several Key points are reiterated here:

     1.  Allocthounous organic matter  inputs:   FPOM   input  from  sewage  and
stormwater  runoff is substantial  as  evidenced  by nigh bacterial contamination
(Dudley and Karr  1979).   This change  along with the modification in  form  and
content  of  CPOM discussed  earlier results  in  major  structural  and functional

-------
                                    - 44 -
Table 3.  General characteristics (relative) of natural (Cummins 1974)  and modified




          (Karr and Dudley 1978)  headwater streams in eastern United States.
WATER QUALITY
    Natural
  Light and temperature   Heavily shaded




                          Stable temperatures
  Dissolved Oxygen







  Suspended Solids




  Concentration







  Dissolved ions







FLOW REGIME




  Flood events
  Low flows
Relatively stable
Low to very low
Generally low
Damped hydrograph
   Modified




Open to sunlight




Very high summer temperature






Highly variable






Highly variable
High especially for P and N











Hydrograph peaks sharp and




 severe
Moderately severe only   Moderately severe each year




 in dry years             in late summer and early fall;




                          extremely severe in dry years.
HABITAT STRUCTURE
  Pools and Riffles       Channel topography and   Reduced and/or destroyed




  Meandering Topography    substrate diversity      by channel maintenance




                           in equilibrium with      activities




                           stream hydraulics
  Sedimentation
Minor except in a few




 unstable bank areas
Major problem with sediment




 source from land and from




 unstable banks; sedimentation




 decreases habitat diversity




 and directly abrades organisms

-------
                                    - 45 -
Table 3.   (Continued)







ENERGETICS




  Particulate organic




   matter size and




   source
  Production (trophic)




   state
Predominantly coarse     Less coarse and more fine




 particulate organic      particulate organic matter -




 matter - from forested   from agricultural and domestic




 terrestrial environ-     sewage




 ment
Little primary pro-




 duction




Heterotrophic; P/R <1
Algal blooms common




Autotrophic; P/R  <1
  Trophic Status of Dominant




    Insects               Shredders,  collectors




    Fishes                Invertivores
    Migrant fishes
Top predators
Scrapers, collectors




Invertivores but forced to




 select a broader range of




 food types







Mostly filter feeders




 and/or omnivores

-------
                                    -  46  -

changes in the stream ecosystem.

     2.  Nutrient availablility: Concentrations of simple nutrient forms (P04;
NO-j,  NH4)  do  not  limit  algal populations.  In addition, inputs of complex
organic compounds associated with CPOM ars not effectively processed.

     3.  Sunlignt availability; All of unshaded  stream  channels  results  in
high  solar  energy  input.  Coupled with available nutrients (#2 above),  this
results in buildup in algal populations (CPOM) wnich  are  either  subject  to
slow decay in the headwaters or are washed downstream in large quantities  dur-
ing high flows.  These algal blooms add to the organic  load  of  the  aquatic
system  and  change  the  physical  characteristics  of the stream environment
(reducing current velocities, covering natural substrates, etc.).

     4.  Temperature and dissolved oxygen imbalance: Seasonal and  daily  pat-
terns  of temperature and dissolved oxygen are exaggerated and poorly buffered
from environmental influences (weather extremes, organic loading, etc.)

     5.  Stream habitat characteristics:  The diversity and stability  of  high
quality  stream  habitat  is  low  (Gorman  and  Karr 1978).  The ditching and
drainage efforts prevalent in many agricultural  watersheds  prepetuates  this
problem.

     6.   Seasonal low flows: The loss of natural vegetation and  installation
of  complex  drainage networks results in rapid runoff instead of slow release
of excess water.  As a result, extreme low flows  during  dry  periods,  espe-
cially  in  late  summer  and  early fall,place, considerable stress on aquatic
ecosystems,

     7.  Changes ir\ Insect and fish communities; These and other shifts  in the
4  primary variables (individually and in the aggregate)  cause major snifts in
the benthic insect faunas as well  as   the   fish  communities.   In  addition,
because  of  the  effect of these changes on the use of headwaters as spawning
and nursery areas, the fish of larger  downstream areas are also affected (Karr
and Dudley 1978).

     Clearly, more than just "water quality" conditions must be  addressed  if
the "Fishable and swimmable" objectives of  PL 92-500 are to be attained.

BMP's vs. Best Management Systems

     In early development of the BlacK Cree.K study the  list  of  conservation
practices  for improvement of water quality was limited to the erosion control
practices used by the  Soil  Conservation  Service.   Slowly,  this  list   was
reduced  to  a  subset  thought to have some value in improving water quality.
The disadvantage of this approach is that a number of other potential  activi-
ties which may result in improvements  in  the quality of water resource are not
considered.  Further, the potential benefits of an integrated networK. of  ero-
sion  control  practices  to  reduce erosion, coupled with practices which may
only benefit water quality, may be greater  than  the  erosion,  coupled,  with
practices alone.  That possibility has not  been adequately explored in earlier
studies, including BlacK CreeK.  The time is right for more effective examina-
tion of careful application of an expanded  list of BMPs into

-------
                                    - 47 -
Best Management Systems

     The following questions must be routinely asked: What will be the  effect
of  juxtapositon  of several practices?  How will they affect the widest range
of water resource characteristics, not just now will they affect erosion  con-
trol  on the land or water quality?  We must more regularly examine the impact
of nonpoint activities with and without varieties of management  alternatives.
What  are  the impacts of these on biological integrity?  It is important that
that assessment include both local and downstream areas, as well  as  upstream
areas.  A further advantage of planning for integrated best management systems
is that they may allow society to capitalize on the benefits to water  quality
which  may  accrue  from the presence of integrated biotic communities.  After
all this is the fundamental principle behind the effective action  of  primary
sewage-treatment  facilities.  With this philosophy, we expect that the dollar
cost to society may be lower per incremental improvement in the quality of our
water resources.

Innovative Management to Restore Biological Integrity

     We now address  the  specific  measures  that  would  improve  biological
integrity  in  streams  and rivers of predominantly agricultural basins of the
eastern United States.  Cur recommendations are designed specifically for  the
BlacK  CreeK  watershed,  but  the  applicability  of these recommendations is
broader.

     The foundation of innovative management is Cdum's compartmentalized model
of  environments  required  by  man  (Fig.  1).   Man clearly needs productive
environments (i.e., agriculture)  and much of the Midwest needs to  be  devoted
to  agricultural  production.   However, protective environments that preserve
biological integrity are also needed in all ecosystems to insure their contin-
ued functioning.  If midwestern rivers, liKe the Maumee, are to be included in
the national mandate for biological integrity, then we believe it is necessary
to  incorporate the sound management of type 3 environments within those river
ecosystems.  The type 3 environment represents a compromise between productive
and  protective uses.  Traditional soil and water conservation programs stress
the productive environment as is demonstrated  by  the  record  of  goals  and
accomplishments of the BlacK CreeK project (Morrison 1977).  Soil conservation
practices applied to the land have water quality benefits but they are only  a
part  of  a  system  of  practices required for the sound management of stream
ecosystems.

     At least two systems of land management (Fig. 6) might be applied to  the
BlacK CreeK watershed in an effort to optimize  production  (agriculture)   and
protection (stream ecosystem integrity).  The central feature of both alterna-
tives is the designation of selected areas as type 3 environments  where  pro-
tective  land  use  receives  priority over the most productive uses.  Farming
need not be eliminated from these  areas  but  alternative  to  the  presently
intensive  agriculture  must be found.  Possible alternatives include rotation
with limited row crops/conservation tillage systems, improved pasture  manage-
ment  with  the elimination of woodlot grazing, and permanent vegetative cover
on erosive slopes.  Under such managanent the average soil loss from  cropland
would be below the maximum tolerable loss for preserving the soil resource.

-------
                Table 4.  A generalized management system to improve the biological integrity of Black Creek

                          and the anticipated impact  on agricultural production within the watershed.
         GOAL

Water Quality

  reduction in sediment and

  nutrients
Flow regime

  less extreme fluctuations

  in stream discharge
Habitat structure

  improvements in stream

  habitat for fish and

  other aquatic life
          RECOMMENDED PRACTICES

Traditional practices, especially conservation

tillage, terraces, grass waterways, filter

strips along stream channels, animal waste

management plans, and soil fertility testing

and management plans

Augmenting low flows through storage and later

release of storm runoff and/or pumping ground

water during dry periods.  Conserv.ition practices

listed under water quality help in reducing peak

stream discharge


Stream renovation (IS) practices instead of large

scale slreambank protection (channelization).

Preserve natural habitat features (pools, riffles

meandering, cover, substrate size sorting,  etc.),

to the maximum extent possible
      IMPACT ON
      PRODUCTION

Production reduced slightly

by conservation tillage

on some soils and the loss of

cropland for filter strips.
Minimal impact on production

through augments ting low flows
The hydraulic improvements of

channelization are only slightly

greater than improvements under

renovation practices (18).

Agricultural production would

not be affected by appreciably
00
I

-------
            Table 4.   (Continued)
          GOAL
Energy source




  energy relationships




  capable of maintaining




  community structure




  and function
                                             RECOMMENDED  PRACTICES
The management of a forested riparian environ-




ment that insures inputs of (TOM and a reduction




in solar radiation.  {Additional water quality




benefits such as improved temperature and




dissolved oxygen and the trapping of sediment




and nutrients are predicted under sum management




(ll)|.  An initial 'stocking' of the stream wtl'.i '




CPOM and aquatic invertebrates may be considered.
                                                                                             IMPACT ON




                                                                                             PRODUCTION
greater flood damages.  In




Black Creek impaired tile




drainage outlets are uncommon,




meaning stream renovation




would have little Impact through




the impairment of subsurface




drainage







Loss of some croplant




adjacent to streams
                                                                                                                                   v£>



                                                                                                                                    I

-------
                                                                                                             I
                                                                                                            Ul
                                                                                                            o
                                                                                                             I
                  (a)
Figure 6.  Black Creek watershed divided into type 1 (unshaded)  and type 3 shaded)  environments.
Type  1  environments  are productive and accomodate intensive agriculture,  'type 3 environments
represent a compromise between productive and protective qualities and  function in preserve bio-
logical integrity.  Conservation practices in type 1 environments address all four primary vari-
ables influencing biological integrity.   See text and Table 4  for further explanation.

-------
                                    -  51  -

     Fig. 6a is probably tne best alternative for Black CreeK -because it  will
likely  nave  less  widespread  effect  on drainage than tne otner alternative
(Fig. 6b) .  However,  6b  mignt  be  tne  best  choice  in  otner  watersneds.
Clearly,  a  wide diversity of intermediate alternatives could be developed to
satisfy local needs.   An  intensive  researcn  program  is  necessary  before
informed decisions can be made on optimum management programs.

     The  important concepts nere is that the land  and  its  associated  biota
play  a  primary role in regulating water quality.  In type 3 environments tne
management strategy is to effect  improvements  in  tne  four  variables  tnat
influence  biological  integrity of BlacK CreeK.  Practices aimed at improving
water quality must be implanented in botn type 1 and type 3 environments.  Tne
recommended practices for improving flow regime, habitat structure, and energy
source are limited in application to the areas designated as type  3  environ-
ments  (Fig.  6).   It is important to note that every watershed is unique and
that the practices and impacts can vary considerably among watersheds, as they
do  when  planners  select  practices  for erosion reduction.  We realize land
managed in this manner may not  always  be  economically  competitive  in  the
current  agricultural  system.  Potential mechanisms to solve this problem are
now enumerated.

Implementation Mechanisms

     The purpose of this paper is not  to  analyze  incentive  programs  which
might  speed implementation of the philosophy outlined above.  However, we can
make some general comments on incentives  in  hopes  of  stimulating  detailed
analysis of their costs and benefits.

     The objective of these and other incentives is  to  maKe  less  intensive
farming  on  type 3 environments competitive with farming operations in type 1
environments while preserving some of  the  other  environmental  benefits  of
these  areas.   This can be accomplished by subsidies underwritten by society,
the principal benefactor.

Classified Streams

     The principle involved in setting aside  areas  for  protection  is  well
established.   Unique  natural  areas  or historical sites have long been pro-
tected from further development to enhance their long-term value  to  society.
Periodically federal agencies iinplenent set-aside programs to ta«e land out of
production or to conserve soil resources.   A  system  of  classified  streams
should be developed to reduce local erosion and its effect on downstream water
resources.  Additional benefits form such programs might derive from increased
availability of local recreational resources (Karr and Scnlosser 1978).  Since
headwaters play an especially important role in determining  resource  quality
throughout  watersheds  (Karr  and  Dudley  1978), efforts to benefit soil and
water resources might emphasize a classified headwater approach.

Green TicKet

     Hie basic outline of the "green ticKet" program (Lake 1978) is to provide
economic  incentives  to  the farmer (or other land user)  through governmental
programs.  These incentives must  improve  the  profitability  of  a  farm  in

-------
                                    - 52 -


exchange  for  installation and maintenance of needed conservation measures on
the land A sliding scale of incentives might exist to yield  greater  benefits
to  a  farmer  on  areas identified as more critical.  For example, areas that
might be part of  a  larger  classified  headwater  area  might  yield  higher
economic  gain to the landowner than a patchworK of areas yielding lower bene-
fit to society.  We can even visualize groups of farmers exerting pressure  on
neighbors  to develop a classified headwater program on their marginal land in
the name of soil and water conservation benefit to society and economic  bene-
fit to them as individuals.  Such programs should be encouraged on areas iden-
tified as locations where treatment of the smallest possible are {or at lowest
economic  cost)  will yield the greatest benefit to society.  Under these cir-
cumstances, land holders might be eligible to collect extra ASCS or ACP  bene-
fits,  to  pay  lower  rates  on crop insurance, or to lower interest rates in
federal loan programs.

     Many other incentive programs could and should  be  sought.   These  must
protect  tne economic stake of the agricultural community and also produce the
greatest benefit to society as a whole.

Institutional Approach to Implementation

     Finally, our experience in the BlacK Creex project  has  yielded  insight
into  some  of the strengths and weaxnesses of present institutional programs.
Traditional  soil  and  water  conservation  programs  fail  to  manage  water
resources  effectively  because they have emphasized soil resources, drainage,
production and, to a lesser extent, water quality.  They  do  not  manage  the
energy  source,  habitat  characteristics, or flow regimes of streams with the
"biological integrity" mandate in mind.  iMany have incorrectly assumed that if
water  pollution declines, habitat quality in a broad sense will be optimized.
hhile traditional programs may have reduced pollution  from  cropland  runoff,
they  have sacrificed natural energy source characteristics, flow regimes, and
high quality steam habitat.  How frequently, for example,  nave  SCS  planners
asKed,  with  biological  integrity  in mind, "How will implementation of this
plan impact energy and nutrient supplies, flow regimes, and habitat quality in
local  and downstream areas?" Clearly, the result cannot always be to preserve
the biota, but without consideration of the  question,  we  will  continue  to
degrade components of our biological environment.

     A case could be made for confining soil and water conservation  districts
and  the Soil Conservation Service (SCS) to their traditional roles of curbing
soil erosion and its associated water pollution.  After closely observing  the
activities  of  these  agencies  in  the BlacK Creetc project and elsewhere, we
firmly believe that districts and the SCS should  expand  their  roles.   This
belief  is  based  on  a  cooperative relationship with farmers.  We seriously
doubt whether any other federal,  state,  or  local  agency  could  match  the
already existing cooperative relationships among districts, SCS and fanners.

     Thus, the districts and SCS appear to be the  best  equipped  agents  for
implanenting water resource improvenent plans in agricultural areas.  However,
an increased role by these  agencies  carries  increased  responsibilities  to
society.   The  only way to satisfactorily meet those responsibilities will be
to expand the training base of SCS employees, or to seeK more regular partici-
pation  in  development  of  plans  by personnel familiar with the disciplines

-------
                                    - 53 -


involved with the biological integrity of water resources.  Planners and field
technicians need to be trained in tne ecological principles that are the basis
of understanding and recognizing sensitive aquatic resources.  Otner  existing
agencies,  such  as Cooperative Extension Service and the special short-course
facilities of many universities, could fill this educational gap.

     Achieving clean water goals  will  depend  on  well-organized  and  well-
conceived plans for control of non-point sources.  For success in agricultural
areas, district and SCS activities must be integrated with  the  stated  goals
for  resource  utilization throughout an area (i.e., a river basin).  Rational
decisions must be made with public input on such issues as the  desired  level
and  type  of  urbanization, agricultural production and water-resource value.
Cnce these decisions are made and incorporated into the general framework of a
208  (or other) plan, district and SCS programs must center on implementing the
needed practices in areas where the greatest overall benefits will accrue.

     Typically, district and SCS contact is with people who voluntarily  apply
for  soil  and water conservation practices.  Servicing this need has been and
will continue to be useful in several respects, but effective soil  and  water
resource  management  requires  that  action  be taken quicKly in the critical
areas of a watershed.  (Many would argue that this is an old policy.  However,
we  emphasize  that  the method of identifying critical areas will differ with
the expansion from soil resources to biological integrity.)  Shifting  district
and  SCS  emphasis to these critical areas will require innovation, especially
in educating the farming community and worKing cooperatively  with  landowners
in  critical  areas,  regardless  of whether or not they voluntarily apply for
assistance.

     In conclusion, we believe a prerequisite for the effective management  of
land  use and water resources is a basic understanding of biological integrity
by those individuals and groups closely associated with the soil  conservation
movement.   The history of soil and water conservation in this country reveals
the strong emphasis on exclusively farm-oriented programs.   The ~Black  Creek
project  exemplifies the traditional approach, and its shortcomings in improv-
ing and maintaining biological integrity are the same as or similar  to  those
of other traditional projects.

     The soil  conservation  movement  has  always  been  dedicated  to  total
resource  conservation,  but  the  demand  for  food  and fiber has led to the
emphasis on productive landscapes.  However, man will successfully manage  the
earth's  resources only if he modifies the environment in ways compatible with
ecological principles.  There is a clear  need  to  conserve  less  productive
landscapes  that  function  to protect the environment, its resources, and its
vital biological processes.  Society, and  especially  the  soil  conservation
movement  as stewards of the land, has an obligation to establish and maintain
such landscapes.  The national goal of restoring "the physical,  chemical  and
biological integrity" of water resources rests on our ability to percieve this
basic ecological tenet and take innovative action in seeding solutions.   take
innovative action in seeding solutions.

-------
                                    - 54 -


Summary

     The central assumption of the BlacK CreeK study is that traditional  ero-
sion control programs are sufficient to insure nigh quality water resources in
agricultural areas.  We have tried to outline the inadequacies of that assump-
tion,  especially  as  they  relate to the goal of attaining biotic integrity.
The declining biotic integrity of  our  water  resources  over  the  past  two
decades  is clearly not totally due to water quality (the effects of physical-
chemical factors) degradation.  Improvement in many  of  the  aspects  of  the
quality  of  our  water  resources  in  a  much broader context than physical-
chemical characteristics.  Other deficiencies in  nonpoint  pollution  control
programs are discussed and a new approach to the problem is outlined.

Acknowledgements

     Financial support for this study came, in part, from  U.S.  Environmental
Protection  Agency Grant #0005103 to the Allen County Soil and Water Conserva-
tion District.  J. LaKe, D. McCain, J. Morrison, L.   Page,  I.  Schlosser,  D.
Sharp, P. W. Smith, L. Toth and R. Warner and several anonymous reviewers made
helpful comments on an earlier draft of the manuscript.

References Cited

     Allan, J. D. 1975.  The distributional ecology and diversity  of  benthic
insects in Cement CreeK, Colorado.  Ecology.  56: 1040-1053.

     Ballentine, R. K. and L. J.   Guarraie  (eds.)   1975.    The  integrity  of
water: A Symposium. USEPA.

     Cummins, K. W. 1973.  Trophic relations of aquatic  insects.   Ann.  Rev.
Ent." 18: 183-206.

     Cummins, K. W. 1974.  Structure and function of stream ecosystems.  BioS-
cience.  24: 631-641.

     Cummins, K. W. 1975.  The ecology of running waters:  theory and practice.
In  Proc.  SandusKy River Basin Symp., Inter. Ref.  Group Great LaKes Pollution
from Land Use Activities.

     Dudley, D.  R. and J. R. Karr 1978.   Reconciling streambanK  erosion  con-
trol  with  water quality goals.   In J.  LaKe and J.  Morrison (eds.) .  Environ-
mental impact _o_f land use on water quality: Final report _on  the  BlacK  CreeK
Project.   (Supplementary  Comments).   U.S.  Environmental Protection Agency,
Chicago, IL.  EPA-905/9-77-007-D.  pp. 101-106.

     Fisher, S.  G. and G. E. LiKens 1973.   Energy  flow  in  Bear  BrooK,  New
Hampshire:  an  integrative  approach  to  stream ecosystem metabolism.  Scol.
Monogr.  43: 421-439.

     Gorman, 0.  T. and J. R. Karr 1978.   Habitat  structure  and  stream  fish
communities.  Ecology.  59: 507-515.

-------
                                    - 55 -
     Barman, W. 1972.  Bentnic substrates: tneir effect on   fresh  water  mul-
losca.  Ecology 53: 271-277.

     Horton, R. E. 1945.  Erosional development of streams and tneir  drainage
basins;  nydropnysical  approacn to quantitative morphology.  Bull. Geol. Soc.
Amer.  56: 275-370,

     Haynes, H. B. N. 1974.  Tne Biology of Polluted  Waters.   Univ.  Toronto
Press, Toronto. 202 pp.

     Karr, J. R. and 0. T. Gorman 1975.  Effects  of  land   treatment  on  tne
aquatic  environment.   In Non-point source pollution seminar.  U. S. Environ-
mental Protection Agency, Chicago, IL EPA-905/9-75-007.  pp. 120-150.

     Karr, J. R. and I. J. Schlosser 1977.  Impact  _of  nearstream  vegetation
and  stream morphology and water quaility and stream biota.  U. S. Environmen-
tal Protection Agency, Atnens, GA. EPA-600/3-77-097.  91 pp.

     Karr, J. R. and I. J. Schlosser 1978.  Water resources and the  landwater
interface.  Science.  201: 229-234.

     Karr, J. R. and D. R. Dudley 1978.  Biological integrity of  a  headwater
stream:  evidence  of  degradation, prospects for recovery.  In J. LaKe and J.
Morrison (eds.).  Environmental impact of land use  _on  water  quality;  Final
Report  on  the Blacx CreeK Project.  (Supplemental Comments).  U. S. Environ-
mental Protection Agency, Chicago, IL.  EPA-905/9-77-007-D.  pp. 3-25.

     Kuehne, R. A. 1962.  A classification of streams illustrated by fish dis-
tribution in an eastern Kentucxy creeK.  Ecology.  43: 608-614.

     LaKe, J. 1978.  Text of speech presented to Purdue Nbnpoint Source Pollu-
tion  Committee,  Stewart  Center,  Purdue  University,  West  Lafayette,  IN,
December 1, 1978.   Published by the National Association of Conservation  Dis-
tricts. 5 pp.

     Minshall, G.  W. 1978.  Autotrophy in stream ecosystems.  BioScience.  28:
767-771.

     Morrison, J.  1977.  Environmental impact o£_ land use  _on  water  quality:
final  report  _on  the 31acK CreeK Prooject - technical report.  EPA 905-9-77-
007B, p.  237-250.

     Nunnally, N.  R.  1978.  Stream renovation: An alternative  to  channeliza-
tion.  Environ. Manage.  2:  403-411.

     Cdum, E. P.  1969.   The strategy of ecosystem development.   Science.  164:
262-270.   Schlosser, I. J. and J.  R.  Karr 1980.   Determinants of Water Quality
in Agricultural Watersheds.   Water Resources Center, University  of  Illinois,
Urbana, IL.  Water Resources Center Report No. 147,  75 pp.

     Sedell,  J. R., F.  J.  TrisKa,  J.  D.  Hall, N.  H.  Anderson, and J.  H.  Lyford
1973.   Sources  and  fates  of organic  inputs  in coniferous forest streams.
Cont. 66,  Coniferous Forest Biome,  IBP,  Oregon State Univ.  23  pp.  Cited  in

-------
                                    - 56 -


Cummins 1974.  BioScience.

     StalnaKer, C. 3, and J. L. Arnett 1976.  Metholologies for the determina-
tion  _of  stream resource flow requiremets: An Assessment.  Utah State Univer-
sity, Logan.

     Warren, C. E. 1971. Biology and Water Pollution Control.  W. B. Saunders,
Philadelphia. 434 pp.

     Westman, W. E. 1978.  Measuring the inertia and resilience o£ ecosystems.
BioScience.  28: 705-710.

     Woodwell, G. M. 1975.  Biological Integrity - 1975.  In R. K.  Ballentine
and  L. J. Guarraia (eds.) The Integrity of Water, U. S. Environmental Protec-
tion Agency.  Washington. 230 pp.

-------
                                    - 57 -
                 BlacK Creek Implications: Present and Future
                                      by
                                 L.F.  Huggins-1-

     Previous speaKers have presented findings of the Black Creek Project as
 they relate to specific subject areas.  I will attempt to distill from these
 specific results some general project conclusions and to focus on their
 national implications.

 WHAT IS WATER QUALITY?

     It has been recognized for some time that agriculture, because of the
 large land mass involved, does contribute to problems associated with non-
 point source pollution.  However, when any effort is made to devise control
 programs an immediate difficulty is encountered.  We have never really come to
 grips with the difficult issue of defining what water quality standards we are
 trying to achieve, except as very broad goals delineated in PL 92-500.

     Dr. James Karr's presentation emphasized the importance of considering
 streams such as BlacK Creek as breeding waters for the Maumee basin.  For such
 streams, in addition to providing acceptable chemical concentrations, it is
 vital that habitat and stream structure be preserved.  On the other hand,
 satisfactory lake conditions also require concern about total annual chemical
 yields from contributing catchments.  These examples emphasize the necessity
 of viewing NPS pollution from a broad perspective rather than just localized
 conditions.

     Another factor which complicates establishment of NPS water pollution
 standards is that, except for irrigated areas, the bulk of such pollution is
 storm induced.  Thus, the critical low-flow,  high concentration standards
 established for point source pollution are not relevant to NPS stream stan-
 dards.  While concentration levels cannot be ignored, habitat maintenance and
 annual yields into receiving lakes are usually more critical factors.  Furth-
 ermore, the storm induced nature of this type pollution introduces a stochas-
 tic element that.complicates issues.

     It is also necessary to recognize that standards developed for different
goals than water quality may be complementary to improved water quality, but
 should not be expected to suffice.  Specifically, attainment of tolerable soil
 loss levels, developed in conjunction with the Universal Soil Loss Equation as
 a yardstick for preserving long-term productivity, does not assure that a
 satisfactory level of water quality will be achieved.

     While the above discussion delineates some of the difficulties with
developing meaningful NPS water quality standards, the necessity of progress
 in this area cannot be ignored.  This discussion is meant to emphasize that it
 is essential for these standards to give due  consideration to many diverse
perspectives.   Much of the valid criticism of current public assistance pro-
grams, whether they deal  with water quality or other areas,  results from
 selection of singular objectives undertaken with too narrow a perspective of
 1. Professor, Dept. of Agric.  Engineering,  Pardue  University, W.  Lafayette,
    IN

-------
                                    - 58 -


society's overall needs.

MANAGEMENT BMPs

     NPS pollution is insidious and difficult to control because, by defini-
tion, it is that pollution which arises from dispersed and poorly identifiable
locations.  Because agriculture is the dominate use of total land mass in most
arable regions of the country, its contributions of chemicals and sediment to
streams and laxes is often large, even when loadings per unit area may be low.

     The importance of management-type BMPs, in contrast to more traditional
structural measures originally designed for erosion control, to attaining
improved water quality must be recognized.  These measures, such a residue and
tillage management, have the distinct advantage of directly protecting large
areas.  While it is certainly true that wide differences occur in the pollu-
tion contribution from individual areas, this ability to "treat" 100 percent
of a field's surface does contribute to overall effectiveness.  Furthermore,
these BMPs can often be applied for very low capital costs.

     The lack of increased utilization of management BMPs is a consequence of
several practical difficulties rather than any lacK of effectiveness.  Diffi-
culties with management BMPs include: 1) a lacx of public visibility, 2)  a
lacK of permanence (they can be abandoned quickly without any capital
expense), and 3) public cost-sharing is difficult to administer because effec-
tive performance is often dependent upon a time-critical application which can
only be verified for a brief interval.  Despite these difficulties, the
overall cost effectiveness of management BMPs requires that we develop innova-
tive ways to overcome the problems and obtain more widespread application of
them.

   .  It is difficult to overemphasize the importance of maintaining a NPS con-
trol program that is built upon voluntary participation encouraged by publicly
supported incentives.  The success of any program will critically depend  upon
the positive cooperation of a large number of individuals.  Furthermore,  the
effectiveness of almost all BMPs is strongly influenced by a landowner's
management decisions*  Maximum effectiveness will be achieved only if these
individuals understand the purpose of and actively support the program which
resulted in the BMP installation.

     A voluntary NPS program, if properly conceived and administered, can be
particularly effective with the agricultural community.  Historically, this
group has demonstrated a concern about their environment and a willingness to
cooperate with their neighbors to improve general conditions.  These attitudes
and the publicly funded agencies established to provide technical and finan-
cial assistance to the agricultural community should be utilized to the max-
imum degree possible.

     There are at least three general characteristics which are vital to  the
success of a voluntary NPS control program.  Briefly, they must be 1) effec-
tive locally, 2) flexible, and 3) applied on a priority basis.  Each point
requires further elaboration.

-------
                                    - 59 -
     First, recommended NFS control measures must be adapted to local condi-
 tions,  i.e. they must be botn effective and economically viable.  No one Knows
more intimately the unique soil and topographic conditions of a land parcel
 than the person wno annually tills the soil and harvests its crop.  The credi-
 bility  of  the entire program is lost, and its shortcomings widely publicized,
when inappropriate measures are recommended.  This requirement suggests the
need to provide field personnel with improved analytical tools that can be
 used to accurately show the farmer what benefits will be obtained from alter-
native  BMPs and which management decisions are critical to its successful
operation.

     Secondly, any national program must recognize the need to preserve the
maximum level of local flexibility.  Conditions vary greatly from farm to farm
as well as regionally.  It is seldom true that one certain practice is greatly
superior in all respects to certain other practices.  Furthermore, successful
farmers must base management decisions on numerous, non-related factors rather
than a  single consideration such as water quality improvement.  Letting an
individual choose between multiple BMPs of roughly equal effectiveness or with
correspondingly different cost sharing permits other, personally important,
factors to be considered and greatly improves the palitability of the entire
program.

     Finally, and perhaps most difficult to achieve because of political con-
siderations and possible charges of favoritism, is the need to recognize that
a given BMP will not be equally effective at improving water quality when it
is applied to different locations.  A cost-effective program requires the del-
ineation of priority areas and incentives that are at least partially depen-
dent upon water quality benefits expected from each individual situation.  To
assure  that such a program can be fairly administered will require innovative
new ideas from the public institutions and additional technical tools for
helping local agencies objectively assess water quality benefits on a site-
specific basis.

DETERMINING BMP EFFECTIVENESS

     There is an abundance of evidence that national water quality goals will
be attained only if due consideration is given to controlling  NPS as well as
point sources of pollution.  Furthermore, after a mucn needed  and rather mas-
sive effort at controlling point source discharges from waste  water treatment
plants, we are in a period of diminishing impact per dollar invested.   Despite
these considerations,  only a toKen effort has been undertaken  to control NPS
pollution.

     The neglect of NPS pollution  control results from a combination of fac-
tors, but two of the most important ones are:  1)  the difficulty of quantifying
sources due to the complex nature  of the diverse forms of such pollution and
2)  the uncertain effectiveness of  the general  control approach,  using  BMPs,
currently proposed.   Solid evidence is critically needed concerning  overall
benefits that can reasonably be expected from  a specific control  program.  To
date, only a  few projects such as  BlacK Creex,  the seven Model  Implementation
Projects and  some thirteen "experimental"  Rural  Clean Water Projects have been
funded  to demonstrate  what might be attainable  on a  national basis.

-------
                                    - 60 -
                             Monitoring = Truth?

     One fundamental misconception concerning water quality held by tne vast
majority of persons, within both the scientific and informed lay communities,
is that true conditions and control effectiveness can be determined only by
field monitoring.  The perception, though often not explicitly stated, is that
bottles of water should be collected and subjected to sophisticated laboratory
analyses to determine what is present so that quality conclusions can be
drawn.

     The primary problem of monitoring is not associated with the laboratory
analysis of a collected sample, although there are still significant difficul-
ties with certain chemical constituents.  Rather, it is with determining the
source of pollutants present in the sample, assessing the true significance of
individual component levels (the standards issue raised above)  and determining
impacts of proposed treatments on pollutant yields.  It must be concluded that
many unKnowns associated with NFS pollution cannot be effectively resolved by
monitoring.  Unfortunately, the pervasive misconceptions about monitoring have
governed all publicly funded efforts to evaluate NFS control measures and have
significantly slowed real progress toward development of programs with proven
effectiveness.

         >            Monitoring—Strengths and WeaKnesses

     No economically feasible monitoring program can be devised which is capa-
ble of establishing cause-effect relationships between NFS pollution and con-
trol measures on a watershed scale, even for a watershed as small as the 20
sq.mi. area of BlacK Create.  Especially on a short-term basis.  This situation
prevails because of the storm-induced nature of NFS pollution, seasonal varia-
tions in weather patterns and the uncontrolled nature of the many factors
which profoundly influence levels of such pollution.

     In view of the situation just described, of what utility are field moni-
toring efforts related to NFS pollution?  Tney can, especially when directed
toward biological community determinations and habitat evaluation, determine
overall water quality conditions of a watershed.  Furthermore, when restricted
to field sized areas with a single land use, monitoring can quantify the bene-
fits of individual control measures.  Such information is vital to the
development of the only viable alternative tool for assessing and controlling
NFS pollution, simulation models.

     In summary, monitoring programs are expensive.  They are slow to produce
meaningful results.  They cannot establish cause-effect on a watershed scale.
Finally, comprehensive monitoring of small, single practice areas is critical
to the successful development of any methodology to assess NFS pollution and
design control programs.  Furthermore, the supply of such information is woe-
fully short.

                     Simulation—Strengths and Weaknesses

     There is certainly no shortage of models available wnich purport to simu-
late at least some phase of water pollution problems.  The basic shortcomings
of all currently available models can be summarized by criticizing them as

-------
                                    - 61 -
incomplete and inaccurate.  Models are incomplete because they do not account
for all of tne many factors involved in something as complex as NFS pollution.
It is also a valid criticism that many models do not even accurately simulate
the limited number of processes which they claim to include.  The "bottom
line" is that many of the currently "available" models are not very good.

     All models are not created as equals!  While this truth should be self-
evident, the poor performance demonstrated by some crude models has seriously
undermined the credibility of the entire simulation approach.  It is certainly
true that all models include approximations of the real processes they are
trying to simulate.  However, the adequacy of these approximations must be
judged on the basis of the requirements for each particular application.
Thus, a given model may be completely unsatisfactory for some applications,
but quite satisfactory for another.  While this complicates the selection tasK
for the model user, it must be recognized that there is simply no such thing
as a single best model.

     The credibility of a modeling approach suffers unfairly relative to moni-
toring because shortcomings of any model are so obvious.  Relationships used
by a model are explicit and clearly documented in a manual or its computer
program implementation.  Thus, the omission or crude approximation of one or
more component processes intuitively thought to be significant for a particu-
lar application raises doubt concerning the adequacy of that model.  To vary-
ing degrees, such issues can be raised with all models.

     In contrast, unwarrented faith is commonly granted numbers reported by
monitoring studies.  Seldom are these results published with sufficient infor-
mation to permit a rigorous evaluation of the overall uncertainty in the
values.  Issues such as the timeliness of sample collection during changing
flow conditions, physical conditions of the sampler intaKe, obstructions to
normal stream flow conditions, etc. are almost never published (in the
interest of brevity or a general lack of understanding concerning their impact
on the result??).  Even laboratory analyses of water samples are subject to
serious discrepancies.  For example, as,< two independent laboratories to
determine levels of sediment bound nutrients in identical samples.  Better
yet, attempt to split a single sediment laden sample and send both subsamples
through the same laboratory.  A comprehensive evaluation of the overall uncer-
tainty associated with NFS physical/chemical water quality samples would'show
that errors in excess of 50 percent are not uncommon.

     The advantages of simulation studies are: 1)  they are inexpensive, at
least in comparison with meaningful monitoring programs; 2)  they produce
results much more quicKly than monitoring and 3)  they can analyze hypothetical
situations that can be used in a planning effort.   This latter advantage can
be especially significant to the success of a voluntary program as was illus-
trated earlier by Dan McCain.   The opportunity to  measure improvements rela-
tive to a common baseline rather than current conditions permits allowance for
previous responsible stewardship.  This avoids giving the greatest public
reward to individuals that have been tne worst environmental offenders prior
to the announcement of the newest program to control  such abuses.   The inabil-
ity of past government programs to proceed in this manner has encouraged poor
citizenship and resulted in public disrespect for  many programs.

-------
                                    - 62 -
     If one's purpose is to determine water quality impacts of a specific pat-
tern of applied BMPs, the advantage of simulating nypotnetical conditions is
also significant.  By simulating storm patterns of particular or long-term
significance, confounding factors as construction activities concurrent witn a
monitoring program and unusual weather patterns during the period of record
can be eliminated.

CLASSES OF MODELS

     The difficulty of selecting the most appropriate model for a given appli-
cation was alluded to above.  While space does not permit the development of a
recommended procedure to follow in making such a selection, it is important to
recognize the existence of two major model classes.

     The first class of models should be called basin or regional scale
models.  These models are designed to analyze general or trend data over very
large geographic regions that encompass several states.  In their most
comprehensive form, they will simulate macro-economic conditions as well as
water quality phenomena.  These models are designed to assess overall or aver-
age conditions and predict trends.  Such models can assist with establishment
of national water quality goals and program levels.

     The second class of models might be called implementation scale models.
This is meant to infer that they are designed to be used to assist planners
and engineers with selection and locating of individual BMPs.  Just as
Knowledge that the average depth of a stream might be two feet deep is of no
value to a non-swimmer deciding whether to wade across, results from basin
scale models are useless for implementation planning.  Cnce an overall program
scope has been established, an entirely different Kind of model is required to
assist with implementing the program.

     An implementation model must be able to assess impacts of the unique com-
bination of features present in the vicinity where individual BMPs might be
located.  To obtain the voluntary participation of local landowners, control
measures that are effective and compatible with conditions on that parcel of
land must be offered.-  Just because a given practice might result in a satis-
factory average level of pollution control from an entire basin does not mean
it will be effective everywhere in the region.  Participation will be forth-
coming when the landowner understands how suggested controls will worK on
his/her land and is convinced that its installation will thereby contribute to
the overall societal goal of improving water quality.  Thus an implementation
scale model must be very site specific.

     It is possible, though not always the best approach, to use an implemen- '
tation scale model as a part of a basin-wide study.  In order to be feasible,
this approach generally requires applying the more detailed model on a sta-
tistically representative number of subwatersheds within the basin.

SUMMARY AND CONCLUSIONS

     Some of the BlacK CreeK experiences which have national policy implica-
tions for NPS control programs have been discussed.  They can be summarized as
follows:

-------
                                    - 63 -
     The establishment of quantitative NFS pollution standards, while very
difficult, is urgently needed.  Such standards must reflect a broad range of
considerations and should vary with different regions of the country.

     While the effectiveness of individual BMPs will depend upon local condi-
tions, there has generally been inadequate utilization of management-oriented
practices in deference to structural measures.  While management practices
suffer from lacK of public visibility and are difficult to administer, their
relatively low capital costs and the effectiveness generally attainable due to
the areal extent of treatment warrants mucn effort to overcome these problems
and increase their utilization.

     The effectiveness of any NFS control program is intimately dependent upon
daily management decisions made by individual landowners.  Therefore, a pro-
gram based upon voluntary participation is not only politically desirable, but
potentially much more effective than a regulatory approach.  To be successful,
voluntary programs must be designed to permit the use of measures which are
locally adapted, offer the participant multiple alternatives and equitably
distribute public and private costs.

     Obtaining solid evidence concerning the effectiveness of NFS control
measures is admittedly difficult.  However, unwarranted reliance has been
placed on monitoring prograns.  It is impossible to determine watershed scale
cause-effect relationships by monitoring, especially within a time frame of 3
to 5 years.

     Despite the many shortcomings of current simulation models, they clearly
offer the best available technology for analyzing NFS pollution problems and
planning control programs.  However, the selection of a particular model is
difficult because it must be dependent upon the types of pollution which are
of prime importance.

     Two fundamentally different classes of models are required for national
program develop and for implementation planning.   The ANSWERS model,  discussed
by David Beasley, is .an example of an implementation scale simulation model.

-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing}
1. REPORT NO. 2.
EPA-905 79-81-004
4. TITLE AND SUBTITLE
Summary of the Black Creek Project (Progress
Report through 1980 Project Year Based 01
in Washington, D. C. , Feb. 1980 Chicago, i:
3. RECIP
5. REPO
Report) Jui
i Seminars B.PERR
LI., Mar. 1980
7. AuTHORts) James B. Morrison, Darrell felson, Jerry V. s. PERF<
Mannering, Don Griffith, David B. Beasley, Daniel
McCain, James R. Karr, Daniel R. Dudley & L. F. Huggins
9. PERFORMING ORGANIZATION NAME AND ADDRESS 10. PRO
Allen County Soil and Water Conservation Districts A42]
Executive Park, Suite 103
2010 Inwood Drive
Fort Wayne, Indiana 46805
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
536 South Clark Street
Chicago, Illinois 60605
11. CON
G-SOI
13. TYP
Progi
14. SPOI
USEPj

RT DATE
16 1981
DRMING ORGANIZATION CODE
3RM1NG ORGANIZATION REPORT NO.
GRAM ELEMENT NO.
32A
TRACT/GRANT NO.
D5335
E OF REPORT AND PERIOD COVERED
ress- Report Feb & Mar 198C
^SORING AGENCY CODE
*.
15. SUPPLEMENTARY NOTES
Ralph G. Christensen, Section 108 (a) Program Coordinator
Carl D. Wilson, EPA Project Officer
16. ABSTRACT
This is a progress report of the Black Creek sediment control project. This report
discusses the details the work done in water quality management and ongoing
research fo planning at the national, regional or state level. The Black Creek
project exemplifies the traditional approach, and it shortcomings in improving
and maintaining biological integrity are the same as or similar to those of other
traditional projects.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
Water quality
Tillage
Leaching
Soil erosin
Non-point source
Agricultural watershed
13. DISTRIBUTION STATEMENT
Document is available to the public through
She Technical Information Service,
Sprin£fipM. VA ?21fi1
b. IDENTIFIERS/OPEN ENOE

19. SECURITY CLASS {This
None
20. SECURITY CLASS (This
Wnr>£>
ED TERMS c. COSATI Field/Group

Report} 21. NO. OF PAGES
66
page} 22. PRICE
EPA Form 2220-1 (9-73)

-------
                                                        INSTRUCTIONS

    1.   REPORT NUMBER
        Insert the EPA report number as it appears on the cover of the publication.

    2.   LEAVE BLANK                                         .  ^

    3.   RECIPIENTS ACCESSION NUMBER
        Reserved for use by each report recipient.

    4.   TITLE AND SUBTITLE
        Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently.  Set subtitle, if used, in smaller
        type or otherwise subordinate it to main title. When a report is prepared in more than one volume, repeat the primary title, add volume
        number and include subtitle for the specific title.

    5.   R~EPORT DATE
        Each report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g., date of issue, date of
        approval, date of preparation, etc.).

    6.   PERFORMING ORGANIZATION CODE
        Leave blank.

    7.   AUTHOR(S)
        Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.).  List author's affiliation if it differs from the performing organi-
        zation.

    8.   PERFORMING ORGANIZATION REPORT NUMBER
        Insert if performing organization wishes to assign this number.

    9.   PERFORMING ORGANIZATION NAME AND ADDRESS
        Give name, street, city, state, and ZIP code. List no more than two levels of an organizational hirearchy.

    10.  PROGRAM ELEMENT NUMBER
        Use the program element number under which the report was prepared. Subordinate numbers may be included in parentheses.

    11.  CONTRACT/GRANT NUMBER
        Insert contract or grant number under which report was prepared.

    12.  SPONSORING AGENCY NAME AND ADDRESS
        Include ZIP code.

    13.  TYPE OF REPORT AND PERIOD COVERED
        Indicate interim final, etc., and if applicable, dates covered.

    14.  SPONSORING AGENCY CODE
        Leave blank.

    15.  SUPPLEMENTARY NOTES
        Enter information not included elsewhere but useful, such as:  Prepared in cooperation with, Translation of, Presented at conference of,
        To be published in, Supersedes, Supplements, etc.

    16.  ABSTRACT
        Include a brief (200 words or less) factual summary of the most significant information contained in the report. If the report contains a
        significant bibliography or literature survey, mention it here.

    17.  KEY WORDS AND DOCUMENT ANALYSIS
        (a) DESCRIPTORS • Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify  the major
        concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.

        (b) IDENTIFIERS AND OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
        ended terms written in descriptor form for those subjects for which no descriptor exists.

        (c) COSATI FIELD GROUP - Field and group assignments are to be taken from the 1965 COSATI Subject Category List. Since the ma-
        jority of documents are multidisciplinary in nature, the Primary Field/Group assignment(s) will be specific discipline, area of human
        endeavor, or type of physical object. The application(s) will be cross-referenced with secondary  Field/Group assignments that will follow
        the primary posting(s).

    18.  DISTRIBUTION STATEMENT
        Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited."  Cite any availability to
        the public, with address and price.

    19. & 20. SECURITY CLASSIFICATION
        DO NOT submit classified reports to the National Technical Information service.

    21.  NUMBER OF PAGES                                                                                          i
        Insert the total number of pages, including this one and unnumbered pages, but exclude distribution list, if any.

    22.  PRICE
        Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (9-73) (Reverse)

-------