-------
Habitat Variables
Suitability Indices
Ratio of spawning habitat
to summer habitat [area that
is less than 1 m deep and
vegetated (spring) divided
by total midsummer area] (V-)
Drop in water level during embryo
and fry stages (V^)
Percent of midsummer area with
emergent and/or submerged
aquatic vegetation or remains
of terrestrial plants (bottom
debris excluded) (V-,)
IDS during midsummer (V.)
Least suitable pH in spawning
habitat during embryo and
fry stages (V5)
Average length of frost-free
season (V)
Maximal weekly average
temperature (1 to 2 m
deep) (V7)
Area of backwaters, pools, or
other standing/sluggish
(less than 5 cm/sec) water
during summer, as a percent
of total area (V)
Stream gradient.(Vg)
Figure 4. A tree diagram for the northern pike riverine HSI model. Note
that habitat variables are not aggregated for separate life requisite
components (Inskip 1982).
IV-1-6
-------
CHAPTER IV-2
DIVERSITY INDICES AND MEASURES OF COMMUNITY STRUCTURE
Diversity is an attribute of biological community structure. The
concepts of richness and composition are commonly associated with
diversity. Species richness is simply the number of species, while
composition refers to the relative distribution of individuals among
the species, or evenness. Odum (1959) defined diversity indices as
mathematical expressions which describe the ratio between species and
individuals in a hiotic community. A major advantage of diversity
indices is that they permit the summarization of large amounts of data
about the numbers and kinds of organisms into a single numerical
description of community structure which is comprehensible and useful
to people not immediately familiar with the specific biota. Some
diversity indices are expressions of the number of taxa, usually
species, in the community. Whittaker (1964) referred to these formulas
as indices of "species diversity", i.e. the more species - the greater
the diversity. "Dominance diversity indices" (Whittaker, 1964)
incorporate the concepts of both richness and evenness; thus, diversity
increases as the number of species increases or as the individuals
become more evenly distributed between the species.
The response of bottom fauna to four types of pollution is represented
in Figure IV-2-1 (Keup 1966). Figure IV-2-1A shows that organic
pollutants generally decrease the number of species present while
increasing the numbers of surviving taxa, whereas toxic pollutants tend
to reduce both numbers and kinds of organisms (Figure IV-2-1B). In
general, the effect of all types of pollutant stress on community
structure is the loss of diversity. The value of diversity in natural
communities lies in the fact that the presence of many species insures
the likelihood of "redundancy of function" (Cairns et al. 1973). As
explained by Cairns and Dickson (1971), in a highly diverse community,
the constantly changing environment will probably affect only a small
portion of the complex bottom fauna community at any time. Because
there are many different kinds of organisms present, the role of those
eliminated as a result of natural environmental change will be filled
by other organisms. Thus the food cycle and the system as a whole
remain stable. On the other hand, natural environmental variation
might eliminate a significant portion of a community that has been
simplified by pollutant stress. With no organism available to fill the
vacated niche, the functional capacity of the unstable community may be
jeopardized. Generally, maintenance of diversity is important because
it enhances the stability of a system.
Diversity indices are commonly computed as one tool among many in the
analysis of aquatic (as well as terrestrial) communities. Some
prevalent reasons for measuring community diversity are listed below
(these purposes are by no means independent of each other):
o
To investigate community structure or functions
To establish its relationship to other community properties
such as productivity and stability
To establish its relationship to environmental conditions
IV-2-1
-------
DIRECTION OF FLOW
A
o;
UJ
ca
OJ
/ N
/ \
OJ
4->
i/>
03
UJ
o:
O)
-------
0 To compare communities
0 To evaluate the biotic health of the community
0 To assess the effects of pollutant discharges
0 To monitor water quality by biological rather than
physicochemical means
In analyses of freshwater aquatic communities, diversity studies generally
involve benthic macroinvertebrates or fish. Several advantages and
disadvantages have been given" for the study of these groups (Cairns and
Oickson 1971, Karr 1981), and are listed in Table IV-2-1. These two groups
are generally considered to be the most suitable organisms for evaluation
of community integrity. Whereas it might be desirable to investigate the
diversity of both fish and macroinvertebrates, the two groups generally are
not used in combination to calculate a single diversity index because of
differences in sampling selectivity and error.
DIVERSITY INDICES
Man^ indices of diversity have been developed. Some indices selected from
the literature are presented in Table IV-2-2, and the more common ones are
discussed below.
Species Diversity Indices
Of the expressions described as species diversity indices (equations 1
through 4 in Table IV-2-2, plus others), the Margalef formula is probably
the most popular. Once the sampling and identification is completed, it is
an easy matter to calculate the diversity index using the Margalef formula
by substituting the number of species(s) and the total number of
individuals (n) into the equation below.
The use of this formula, and others of the type, has some important
limitations. First, it is not independent of sample size. Menhinick
(19M) found that for sample sizes from 64 to 300 individuals the Margalef
diversity index varied from 3.05 to 14.74, respectively. In that study,
four- species diversity indices were evaluated for variation with sample
size and all were found unsatisfactory except for the equation referred to
as the Menhinick formula in Table IV-2-2. The second limitation of species
diversity indices is that, by definition, they do not consider the relative
abundance among species, and, therefore, rare species exert a high
contribution to the index value. To illustrate this limitation, Wi 1 hm
M972) calculated diversity by the Margalef and Menhinick formulas for
three hypothetical communities each containing five species and 100
individuals (see Table IV-2-3). Communities A, B, and C exhibit a wide
rancie of relative distribution of individuals between the five species.
Intuitively, community A is more diverse than community C, but the two
species diversity indices fail to express any difference.
IV-2-3
-------
TABLE IV-2-1. ADVANTAGES AND DISADVANTAGES OF USING MACRO INVERTEBRATES AND
FISH IN EVALUATION OF THE BIOTIC INTEGRITY OF FRESHWATER
AQUATIC COMMUNITIES (CAIRNS AND DICKSOM, 1971; KARR, 1981)
MACROINVERTEBRATES
Advantages
Fish that are highly valued by humans
are dependent on bottom fauna as a
food source.
Many species are extremely sensitive
to pollution and respond quickly to
it.
Bottom fauna usually have a complex
life cycle of a year or more, and if
at any time during their life cycle
environmental conditions are out.ide
their tolerance limits, they die.
Many have an attached or sessile mode
of life and are not subject to rapid
migrations, therefore they serve as
natural monitors of water quality.
Di sadvantages
They require specialized
taxononic expertise for
identification, which is also
time-consumi ng.
Background life-history
information is lacking for many
species and groups.
Results are difficult to
translate into values meaningful
to the general public.
FISH
Life history information is extensive
for most species.
Fish communities generally include a
range of species that represent a
variety of trophic levels (omnivores,
herbivores, insect!vores,
pianktivores, piscivores) and utilize
foods of both aquatic and terrestial
origin. Their position at the top of
the aquatic food web also helps
provide an integrated view of the
watershed environment.
Fish are relatively easy to identify.
Most samples can be sorted and
identified in the field, and then
released.
The general public can relate to
statements about conditions of the
fish community.
Both acute toxicity (missing taxa)
and stress effects (depressed growth
and reproductive success) can be
evaluated. Careful examination of
recruitment and growtn dynamics among
years can help pinpoint periods of
unusual stress.
0 L
Sampling fish communities is
selective in nature.
Fish are highly mobile. This
can cause sampling difficulties
and also creates situations of
preference and avoidance. Fish
also undergo movements on die!
and seasonal time scales.
There is a nigh requirement for
manpower and equipment for field
samp 1 ing.
IV-2-4
-------
TABLE IV-2-2. SUMMARY OF DIVERSITY INDICES
Descriptive Name
1. Simplest possible ratio of
species per individual
Formula
d =1
Reference
Wilhm, 1967
2. Gleason
d =
log n
Menhinick, 1964,
Gleason, 1922
3. Margalef
d =
s-1
I n n
Margalef, 1951
1956
4. Menhinick
d =
(n)
1/2
Menhinick, 1964
5. Mclntosh
n - (In..2)1/2
d =
n - (n!
Mclntosh, 1967
6. Simpson
d =
n (n-1
Simpson, 1949
7. Brillouin
H = (ij Hog n! - L log n.!j Brillouin, 1960
8. Shannon-Wiener
H = -L lp.log2p.J
Shannon and
Weaver, 1963;
Wiener, 1948
Approximate form of the
Shannon Index
. a - -I
Shannon Index using
biomass (weight) units
3^
- 'L
w . w .
Wilhm, 1968
IV--2-5
-------
TABLE IV-2-2. (Cont'd)
9. Hierarchical
Diversity Index
(HDD
HDI = H'(F)+HyH'GF(S)
Pielou, 1969,
1975
10. Hierarchical Trophic- HTDI = H ' (T. )+H ' (T9 )+H
Based D.I. (HTDI)
Ti
Osborne et al . ,
1980
11. Redundancy (r]
r =
d
d - d .
max mi n
Datten, 1962;
Wilhm, 1967
12. Equitability (e)
e =
Lloyd and
Ghelardi, 196^
13. Evenness (J,J' , v!
J =
max
Pielou 1969,
1975; Hurlbert,
1971
max
d -
v =
d - d .
max mi n
14. Number of moves (NM)
n (s
Rini
"ager, 1972
, r ^ j. • i * • T ^
15. Sequential Comparison Index
number of runs
number Of soecies
COT, JLnumber of taxa.
Cairns et al.
1968; Cairns S
Dickson, 1971;
Euikema et al.
1980
IV-2-6
-------
TABLE IV-2-2. (Cont'd)
KEY
H = d s H' =3= diversity index.
n = total number of individuals.
n. = number of individuals in species i.
s = total number of species.
ni
p. = probability of selecting an element of state i = —.
R. = rank of species i.
s' = the species required to produce the calculated d.i. value if
the individuals were distributed among the species accord-
ing to MacArthur's (1957, 1960) "broken-stick" model.
IV-2-7
-------
TABLE IV-2-3. DIVERSITY OF THREE HYPOTHETICAL COMMUNITIES EVALUATED BY THE
MARGALEF, MEMHINICK,
Community
A
B
C
nl
20
40
1
20
30
1
n3
20
15
1
n4
20
10
1
n5
20
5
96
AND SHANNON-WIENER INDICES
n
100
100
100
s
5
5
5
s-1
In n
0.87
0.87
0.87
n'/2
0.50
0.50
0.50
d
2.
1.
0.
32
67
12
Another shortcoming of species per individual formulas is that they are not
dimensionless , thus substitution of alternate variables for numbers - sucn
as biomass or energy flow - would produce values dependent on the arbitrary
choice of units.
The major advantage of using species diversity indices is the simplicity of
calculation; however, certain conditions for their proper use must be
considered. Since these formulas are dependent on sample size (except
possibly, the Menhinick equation), for intercommunity comparison the sample
sizes should be as nearly identical as possible. It must be kept in mind
that these expressions represent only the number of species and not any
expression of relative abundance. Finally, for use of variables other than
numbers, the units must be specified and kept consistent.
Dominance Diversity Indices
The most prominent dominance diversity index (equations 5 through 8 in
Table IV-2-2, plus others) is the Shannon-Wiener formula. This index is
used extensively in research projects, as is the Simpson equation. The
Shannon-Weiner diversity index evolved from information theory to the
functional equation shown below:
in which the ratio of' the number of individuals collected of species i
to the total number of individuals in the sample (n-j/n) estimates the
total population value (N-j/N), which is an approximation of the
probability of collecting an individual of species i (p-j). It should be
noted that the units of d using '1092 is the binary unit, or bit. Natural
logarithms or log-jg are sometimes substituted into the equation for
convenience, in which case different index values would be obtained, with
the units of nats or decits, respectively. The Shannon-Wiener diversity
index is calculated using base 10 logarithms, for two simple, hypothetical
samples in Example IV-2-1 (see statistical analysis section). A formula
for conversion between differently-based logarithms is given below:
Tog2Y- = 1 .443 In Y = 3.323 loglQY
The logarithm base and units should always be given when reporting data.
IV-2-8
-------
The dominance and species diversity indices discussed can be used to
measure the diversity of virtually any biological community (including
macroinvertebrates and fish), and their application is limited only by
sampling effectiveness. Wilhm and Dorris (1968) evaluated species
diversity of benthic macroinvertebrates using the Shannon-Wiener formula
and obtained values less than 1.0 in areas of heavy pollution, values from
1.0 to 3.0 in areas of moderate pollution, and values exceeding 3.0 in
clean water areas (values given are in decits).
Disadvantages of using the Shannon index (or others of the type) include
the considerable time, expense, and expertise involved in sampling,
sorting, and identification of samples. Calculation of the index value can
be mathematically tedious if done manually, but is greatly simplified if a
computer is available. Computer programs for computing d and r are
provided in the literature (Wilhm, 1970; Cairns and Dickson, 1971).
The Shannon-Wiener formula has a number of features which enhance its
usefulness. This index of diversity is much more independent of sample
size than the species diversity indices (Wilhm 1972). Since it
incorporates the concept of dominance diversity, the relative importance of
each species collected is expressed and the contribution of rare species to
diversity is low. This is illustrated by the d values calculated using the
Shannon equation for the three communities in Table IV-2-3. Also, the
Shannon formula is dimensionless, facilitating the measurement of biomass
diversity. Odum (1959) recognized that the structure of the biomass
pyramid held more ecological (trophic) significance than the numbers
pyramid because it takes many small individuals to equal the mass of one
large individual. The Shannon-Wiener equation can easily be modified to
accomodate any units of weight as shown below:
W," ,W-is
a = -I (-i)W-i)
Wilhn (1968) pointed out that use of this diversity index with units of
energy flow might be even more valuable to the study of community structure
and ^unction.
Hierarchical Diversity
Diversity indices, such as the Shannon-Wiener index, can be partitioned to
reflect the contribution made by different taxonomic and trophic levels.
Pielou (1975) suggested that a community showing more diversity at higher
taxonomic levels (e.g. genus and family) should be considered to be more
diverse than a community with the same number of species but congeneric or
cofanilial. Osborne et al (1980)questioned the ecological significance of
Pielou's suggestion, but investigated the use of the hierarchical diversity
index (HDI) shown below:
HDI = H'(F) + H'F(G) + H'FG(S)
in which H'(F) is the familial component of the total diversity, H'p(G)
is the generic component of the total diversity, and H'pg(S) is the
IV-2-9
-------
specific component of the total diversity. The equation used by Kaesler et
al . (1978) illustrates the calculation of the hierarchical components.
They used
o N. o fi N . o fi gij N.-
where a ,B,T, and <5 are weighting coefficients; subscripts 0, F, G, and S
represent order, family, genus, and species, respectively; o, f, and g
represent number of orders, families within orders, and genera within
families, respectively; N represents the number of individuals; and N-j
represents the number of individuals in the ith group. Osborne et al .
(1980) concluded that identification to the family level was sufficient to
detect intersite differences in that study, while the order level (Hughes,
1978) and generic level (Kaesler et al . , 1978) were sufficient in other
studies. Determination that identification to species or genus is
unnecessary for a particular study would reduce the time, expertise, and
expense required. A hierarchical diversity index would be of more
ecological value if it were based on trophic relationships rather than
taxonomy. Osoorne, et al . (1980) presented the following hierarchical
trophic diversity index (HTDI):
HTDI = H'(T]) + H'T1(T2) + H'T1T2(T3)
in which H'(T]) is the general trophic level component of the total
trophic diversity, H'y]^) is the functional group component of the
total trophic diversity, and ^'TlTZ^s) ^s ^e lowest taxonomic
unit component of the total trophic diversity. The classifications used in
the hierarchical trophic-based diversity index of Osborne et al . (1980) are
listed in Table IV-2-4A. Two classification systems were investigated by
Kaesler et al . (1978): the trophic classifications appear in Table IV-2-4B.
and the functional morphological classifications are shown in Table
IV-2-4C. All of these hierarchical diversity indices used benthic
macroinvertebrates as their group of study. Hierarchical diversity indices
based on trophic level and functional morphology are relatively new and
their utility will improve as more experience is gained. These indices are
of potentially great ecological value because of their functional (rather
than structural, e.g. taxonomic) approach to community analysis.
Evenness and Redundancy
When using dominance diversity indices, it is desirable to distinguish
between the two concepts of diversity incorporated into them, since it is
theoretically possible for a community with a few, evenly-represented
species to have the same index value as a community with many,
unevenly-represented species. For this reasons, relative diversity
expressions (equations 11 through 14 in Table IV-2-2, plus others) such as
eveness and redundancy are often used in conjunction with dominance
diversity indicies. Redundancy is an expression of the dominance of one or
more species and is inversely proportional to the wealth of species (Wilhm
and Dorris, 1968). To use the redundancy expression in conjunction with
the Shannon-Wiener index, the theoretical maximum diversity (dmax)
and minimum diversity (dm-jn) are calculated by the equations:
d = (-) Clog9n! - s
max n' c.
IV-2-10
-------
TABLE IV-2-4. FUNCTIONALLY-BASED HIERARCHICAL CLASSIFICATION SYSTEMS
A. Hierarchical trophic classification used for HTDI calculations
HTI
(Trophic level
Omni vore
Carni vore
Herbi vore
Detritivore
H
T2
(Functional group)
Filter Feeders
Collector-Gatherer-
Shredder-Engul fer
Engulfer-Shredder
Collector-Filterer-
Engulfer
Engulfer-Grazer
Engul fer-Collector-
Grazer
Engulfer
Piercer
Sera per-Col lector-Gatherer
Col lector-Gatherer-Shredder
Collector-FiIterer-Gatherer
Col lector-Gatherer
Col 1ector-Fi Iterer
Shredder
Shredder
Col lector-Gatherer
(Number of individuals)
Number of individuals of each
taxon within each functional
group.
Trophic classification of macrobenthic invertebrates. For any specific
application, not all possible combinations are likely to be realized.
Level of
Hierarchy Name
Subdivi sions
I Functional group
II Feeding mechanism
III Dependence
IV Food habit
V Species
shredders (vascular plant tissues)
collectors (detrital materials)
grazers (Aufwuchs)
predators
parasites
chewers and miners
filters (suspension feeders)
gatherers (sediment or deposit feeders
scrapers
chewers and suckers
swallowers and chewers
pi ercers
attachers
obiigate
facultati ve
herbi vory
detriti vory
carni vory
omni vory
number of individuals
IV-2-11
-------
TABLE IV-2-4 . FUNCTIONALLY-BASED HIERARCHICAL CLASSIFICATION SYSTEMS (Cont'd)
C. HBR (head, body, respiratory organ) classification of macrobenthic
invertebrates according to functional morphology: head position,
body shape, and respiratory organs.
Level of
Hierarchy
Name
Subdi vi si ons
I
II
Head position
category)
(feeding
Body shape (current
of stream)
III
Respiratory organs
(substratum)
IV
Species
hypognathous
prognathous
opi sthorhynchous
vestigial or other
flattened irregular
flattened oval
flattened elongate
compressed laterally
cyl i ndrical
elongate
short, compact
fusi form
i rregular
hemicylindrical or subtriangular
simple filamentous gills
compound filamentous gills
platelike gills
operculate gills
leaflike gills or organs
respiratory dish
respiratory tube
spiracular gills
caudal chamber
piastron
body integument
tracheal respiration
number of individuals
IV-2-12
-------
og2tn
s-D!
Then tne location of d between the theoretical extremes can be computed by
the redundancy formula: ^
r =
max
max" min
Table IV-2-5 illustrates the expression of redundancy.
TABLE IV-2-5. THE SHANNON-WIENER INDEX AND CORRESPONDING
REDUNDANCY VALUES FOR 11 HYPOTHETICAL
COMMUNITIES, (after Patten, 1962).
Communi
Species A
$9 . ...1
Sc 1
Sz 1
B
2
t
1
1
_
C
2
2
1
1
_
D
3
j
1
1
_
F
2
2
2
-
_
(N -
F
3
2
1
-
_
ties
6)
G
4
1
1
-
_
H
3
3
-
_
I
4
2
-
_
J
5
1
-
_
K
5
-
_
d(bits)2.53 2.25 1.93 1.79 1.61 1.^7 1.25 1.00 0.92 0.65 0.00
R 0.00 0.13 0.25 0.30 0.38 0.43 0.52 0.61 0.64 0.75 1.00
Expressions have
apportionment of
also been
individuals
developed to describe the evenness of
among species in a community. Evenness
measures have historically taken two forms. One is the ratio of diversity
to the maximum possible diversity, where dmax is defined as the
community in which all species are -equally distributed:
J' '
Where the logarithm is to the same base as used in the corresponding
diversity index calculation. However, log s is only an approximation of
Jmax
because all species
sampled. A measure of evenness
the
that
d-
v =
community generally will not be
does not depend on s is shown below:
min
- d
max rrn n
measure of evenness that the expression for redundancy
derived by the relationship r = 1-V; thus, redundancy may
also be thought of as a measure of the unevenness of apportionment of
individuals among species.
It was from this
(snown above) was
Secuential Comparison Index
The sequential companson index
index of diversity because of it
(non-academic) studies. The
estinatinn relative differences
SCI) is probably the most widely used
extensive worldwide use in industrial
SCI is a simplified, rapid method for
in Diological diversity and has been used
IV-2-13
-------
mainly for assessing the biological consequences of pollution. Use of the
SCI requires no taxonomic expertise on the part of the investigator.
Although it has been used with microorganisms, the SCI is predominately
used to evaluate diversity in benthic macroinvertebrate communities. The
collected specimens are randomly poured into a white enamel pan with
parallel lines drawn on the bottom. Only two specimens are compared at a
time. Comparisons are based on differences in shape, color, and size of
the organisms. If the imminent specimen is apparently the same as the
previous one, it is part of the same "run"; if it is not, it is part of a
new run. An easy way of recording runs is to use a series of X's and O's.
For example, the specimens shown in line one of Figure IV-2-2 would be
recorded, from left to right as X 0_ _X £ _X 0_ _X, or seven runs. The
specimens in line two would be tabulated by X X X _0 X X X. Sample two only
contains three runs and is obviously less dfverse. Ultimately, it will be
necessary to know the total number of taxa in the collection. This can
either be counted after determining the number of runs or determined
simultaneously by underlining the symbol of each new taxon as shown above.
Cairns, et al. (1971) described the following stepwise procedure for
calculating the Sequential Comparison Diversity Index:
1. Gently randomize specimens in a jar by swirling.
?. Pour specimens out on a lined white enamel pan.
3. Disperse clumps of specimens by pouring preservative or water on
clumps.
4. If the sample has fewer than 250 specimens, determine the number of
runs for entire sample and go to Step 12.
5. If sample has more than 250 specimens, determine the number of runs for
the first 50 specimens.
6. Calculate DI]_ where DI]_ = numbers of runs/50.
7. Plot DI} against the number of specimens examined as in Figure
IV-2-3.
8. Calculate the SCI for the next 50 specimens.
9. Determine the total number of runs for the 100 specimens examined.
10. Calculate a new DI^ for 100 specimens as in Step 5 and plot the value
obtained on the graph made in Step 7, where DIi = number of runs/100.
11. Repeat this procedure in increments of SO until the curve obtained
becomes asymptotic. At this point enough specimens have been examined
so that continued work will produce an insignificant change in the
final DI} value.
12. Calculate final DI-. where
DIi = number of runs
number of specimens
13. Record the number of different taxa observed in the entire sample. This
can be done after deriving the final 01} or simultaneously by simply
noting each new taxon as it is examined in the determination of runs.
IV-2-14
-------
2.
-I L
SO 100 ISO ZOO 250 300 3SO 400 450 SOO
NUMBER OF SPECIMENS
Figure IV-2-2. Determination of runs in SCI
technique (from Cairns and Dickson, 1971).
Figure IV-2-3. DI, and sample
size (from Cairns and Dickson,
1971).
DI.
1.0
0.9
C.8
0.7
C.6
0.5
0.4
0.3
0.2
Tf r
A
A = use line A to be 95%
confident the mean DI
1
is within 20% of true value
3
UT Ti TZ 13 14
Number of times to repeat SCI
examination on same sample
r—
i
I
! i
/ B = use line B to be 35%
I confident the mean DI.
1 is within 10% of true value
i l i i / i J I I ] ! 1 I ! f
<
ib
Figure IV-2-4. Confidence limits for DI, values (from Cairns and Dickson, 1971)
IV-2-15
-------
14. Determine from Figure IV-2-4 the number of times the SCI examination
must be repeated on the same sample to be 95 percent confident that the
mean DI^ is within a chosen percentage of the true value for DIi.
In most pollution work involving gross differences between sampling
areas, Line A of Figure IV-2-4 should be used. For example, suppose
DI]_ were 0.60. Using Line A of Figure IV-2-4 the SCI should be
performed twice to be 95 percent confident that the mean DIj is
within 20 percent of the true value.
15. After determining N, rerandomize the sample and repeat the SCI
examination on the same number of specimens as determined in Step
11. Repeat this procedure M - 1 times.
16. Calculate DIj by the following equation:
DIj = DI] x (number of taxa)
17. Calculate Dly by the following equation:
DIj = (DI] ) x (number of taxa)
18. Repeat the above procedure for each bottom fauna collection.
19. After determining the Dly for each bottom fauna collection at each
sampling station, there is a simple technique for determining if the
community structures of the bottom fauna as evaluated by the SCI
(Oly) value are significantly different within a station or between
stations. Calculate the 95 percent confidence intervals around each
Dly value. If the 95 percent confidence intervals do not overlap,
then the community structures of the bottom fauna as reflected by the
DIj values are significantly different. For example, suppose the
Dly value for Station 1 were 45 and for Station 2 were 28. In the
determination of Dly a decision was made to use Line A in Figure
IV-2-4, which means that the Dly is within 20 percent of the true
value 95 times out of 100. Therefore the 95 percent confidence
interval for the Dly value at Station 1 would be from 49.5 to 40.5,
or 10 percent of the Dly value on either side of the determined
Dly. Station 2 would have a 95 percent confidence interval for the
Dly value of from 30.8 to 25.2. The bottom fauna communities at the
two stations as evaluatd by the Dly index are significantly
different.
The SCI permits rapid evaluation of the diversity of benthic
macroinvertebrates. Some insight into the integrity of the bottom
community can be gained from Dly values. Cairns and Dickson (1971)
reported that healthy streams with high diversity and a balanced density
seem to have Dly values above 12.0, while polluted communities with
skewed population structures have given values for Dly of 8.0 or less,
and intermediate values have been found in semipolluted situations.
IV-2-16
-------
SPECIAL INDICES
Several expressions that are not diversity indices per _se_ but which
incorporate the concept of diversity have been formulated. These include
numerous biotic indices (Pantle and Buck, 1955; Beck, 1955; Beak, 1964;
Chutter 1971, Howmiller and Scott 1977, Hilsenhoff 1977, Winget and Mangum
1979), a composite index of "well-being" (Gammon 1976), and Karr's index
(Karr 1981). These indices are designed to evaluate the biotic integrity,
or health, of biological communities and ecosystems.
Biotic Indices
Beck (1955) developed a biotic index for evaluating the health of
streams using aquatic macroinvertebrates. In the equation
Biotic index = 2(n Class I) + (n Class II)
where n represents the number of macroinvertebrate species, more weight is
assigned to Class I organisms (those tolerant of little organic pollution)
than to Class II organisms (those tolerant of moderate organic pollution
but not of anaerobic conditions). A stream nearing septic conditions will
have a biotic index value of zero; whereas streams receiving moderate
amounts of organic wastes will have values from 1 to 6, and streams
receiving little or no waste will have values usually over 10 (Gaufin
1973).
The biotic index proposed by Hilsenhoff uses the arthropod community
(specifically insects, amphipods, and isopods) to evaluate the integrity of
aquatic ecosystems via the formula:
BI = 1C r\.a./n
where n^ is the total number of individuals of the ith species (or
genus), an- is the tolerance value assigned to that species (or genus),
and n is the total number of individuals in the sample (Hilsenhoff, 1977;
Hilsenhoff, 1982). Pollution tolerance values of zero to five are assigned
to species (or genera when species cannot be identified) on the basis of
previous field studies. A zero value is assigned to species found only in
unaltered streams of very high water quality, a value of 5 is assigned to
species known to occur in severely polluted or disturbed streams, and
intermediate values are assigned to species occurring in intermediate
situations. Calculation of this and other biotic indices are methods of
biologically assessing water quality.
Index of Hell-Being
Utilizing fish communities, Gammon developed a composite index of well-
being (IWB) as a t00^ f°r measuring the effect of various human
activities on aquatic communities (Gammon, 1976; Gammon and Reidy, 1981;
Gammon et al., 1981). This index was calculated by:
l^ = 0.5 In n + 0.5 In w + dno + dwt
in which n is the number of individuals captured per kilometer, w is the
weight in kilograms captured per km, d*np is the Shannon index based on
numbers, and dwt is the Shannon index based on weights. (The Shannon
index was calculated using natural logarithms).
IV-2-17
-------
Karr's Index of Biotic Integrity (IBI)
Karr (1981) presented a procedure for classifying water resources by
evaluating their biotic integrity using fish communities. Use of the
system involves three assumptions: (1) the fish sample is a balanced
representation of the fish community at the sample site; (2) the sample
site is representative of the larger geographic area of interest; and (3)
the scientist charged with data analysis and the final classification is a
trained', competent biologist with considerable familiarity with the local
fish fauna. For each of the twelve criteria listed in Table IV-2-6, the
evaluator subjectively assigns a minus (-), zero (0), or plus (+) value to
the sample. The grades are assigned numerical values - (-)=!, (0)=3, (+)=5
- which are summed over all twelve criteria to produce an index of
community quality. The sampled community is then placed in one of the
biotic integrity classes described in Table IV-2-7 based on numerical
boundaries such as those tentatively suggested by Karr (1981) and shown in
Table IV-2-8.
TABLE IV-2-6. PARAMETERS USED IN ASSESSMENT OF FISH
COMMUNITIES. (SEE ARTICLE TEXT FOR DISCUSSION.)
Species Composition and Richness
Number of Species
Presence of Intolerant Species
Species Richness and Composition of Darters
Species Richness and Composition of Suckers
Species Richness and Composition of Sunfish (except
Green Sunfish)
Proportion of Green Sunfish
Proportion on Hybrid Individuals
Ecological Factors
Number of Individuals in Sample
Proportion of Omnivores (Individuals)
Proportion of Insectivorous Cyprinids
Proportion of Top Carnivores
Proportion with Disease, Tumors, Fin Damage, and
Other Anomalies
BIOLOGICAL POLLUTION SURVEY DESIGN
The first step in planning any survey of water quality is to identify
specific objectives and clearly define what information is sought. For
instance, the objective of a use attainability analysis might be to
evaluate the water quality or degree of degradation of a body of water, in
general, in order to ascertain the accuracy of the current use designation.
Alternately, the analysis objective might be to determine the extent of
damage caused by a discharge or series of discharges. From such
information, the potential attainable use can be identified; judgments must
then be made regarding the benefits/costs of improving the degree of waste
treatment.
IV-2-18
-------
TABLE IV-2-7: BIOTIC INTEGRITY CLASSES USED IN ASSESSMENT OF FISH COMMUNITIES
ALONG WITH GENERAL DESCRIPTIONS OF THEIR ATTRIBUTES
Cl_as_s Attributes
Excellent Comparable to the best situations without Influence of
man; all regionally expected species for the habitat and
stream size, including the most intolerant forms, are
present with full array of age and sex classes; balanced
trophic structure.
Rood Species richness somewhat below expectation especially
due to loss of most intolerant forms; some species with
less than optimal abundances or size distribution;
trophic structure shows some signs of stress.
Fair Signs of additional deterioration include fewer
intolerant forms, more skewed trophic structure (e.g.,
increasing frequency of omnivores); older age classes of
top predators may be rare.
Poor Dominated by omnivores, pollution-tolerant forms, and
habitat general ists; few top carnivores; growth rates
and condition factors commonly depressed; hybrids and
diseased fish often present.
Very Poor Few fish present, mostly introduced or very tolerant
forms; hybrids common; disease, parasites, fin damage,
and other anomalies regular.
No Fish Repetitive sampling fails to turn up any fish.
TABLE IV-2-8: TENTATIVE RANGES ^OR THE BIOTIC
_ INTEGRITY CLASSES. _
Class Index Number
Excellent (E) 57-60
E-G 53-56
Good (G) 48-52
G-F 45-47
Fair (F) 39-44
F-P 36-38
Poor (P) 28-35
P-VP 24-27
Very Poor (VP) < 23
IV-2-19
-------
The next steps in planning the survey are to review all available reports
and records concerning the waste effluents and receiving waters, and to
make a field reconnaissance of the waterway, noting all sources of
pollution, tributaries, and uses made of the water.
Sampling Stations
There is no set number of sampling stations that will be sufficient to
monitor all types of waste discharges: however, some basic rules for a
sound survey design are listed below (Cairns and Dickson 1971). The
following describes an "upstream-dcwnstream" study. The reader should also
consult Section IV-"S on the reference reach approach to see an alternative
method.
1. Always have a reference station or stations above all possible
discharge points. Because the usual purpose of a survey is to
determine the damage that pollution causes to aquatic life, there must
be some basis for comparison between areas above and below the point or
points of discharge. In practice, it is usually advisable to have at
least two reference stations. One should be well upstream from the
discharge and one directly above the effluent discharge, but out of any
possible influence from the discharge.
2. Have a station directly below each discharge.
3. If the discharge does not completely mix on entering the waterway but
channels on one side, stations must be subdivided into left-bank,
midchannel, and right-bank substations. All data collected
biological, chemical, and physical - should be kept separate by
substations.
4. Have stations at various distances downstream from the last discharge
to determine the linear extent of damage to the river.
5. All sampling stations must be ecologically similar before the bottom
fauna communities found at each station can be compared. For example,
the stations should be similar with respect to bottom substrate (sand,
gravel, rock, or mud), depth, presence of riffles and pools, stream
width, flow velocity, and bank cover.
6. Biological sampling stations should be located close to those sampling
stations selected for chemical and physical analyses to assure the
correlation of findings.
7. Sampling stations for bottom fauna organisms should be located in an
area of the stream that is not influenced by atypical habitats, such as
those created by road bridges.
8. In order to make comparisons among sampling stations, it is essential
that all stations be sampled approximately at the same time. Not more
than 2 weeks should elapse between sampling at the first and last
stations.
IV-2-20
-------
For a long-term biological monitoring program, bottom organisms should be
collected at each station at least once during each of the annual seasons.
More frequent sampling may be necessary if water quality of anv riisrharop
changes or if spills occur. The most critical period
seasons.
any discharge
bottom fauna
changes or if spills occur. The most critical period for bottom fauna
organisms is usually during periods of high temperature and low flow of the
waterway. Therefore, if time and funds available limit the sampling
frequency, then at least one survey during this time will produce useful
information.
Sampling Equi pment
Commonly used devices for sampling benthic macroinvertebrate communities
include the Peterson dredge, the surber square foot sampler, aquatic bottom
nets, and artificial substrate samplers. Proper use of the first three
pieces of equipment requires that the operator exert the same amount of
effort at each station before comparisons can be made. This subjectivity
can cause error, but can be minimized by an experienced operator.
Artificial substrates standardize sampling to some extent by providing the
same type of habitat for colonization when placed in ecologically similar
conditions. A simple type of artificial substrate sampler is a wire basket
containing rocks and debris. Others consist of masonite plates or plastic
webs which can be floated or submerged. Additional advantages of
artificial substrate samplers are quickness and ease of use.
Fish sampling equipment includes electrofishing gear,
seine, purse seine), towed nets (otter trawl), gill
chemical toxicants (rotenone, antimycin). As
sampling effort must be put forth_ at each
equipment. Also, measures should be taken to
fish sampling.
Number of Samples
encircling gear (haul
nets, maze gear, and
discussed above, the same
station when using this
reduce the selectivity of
If comparisons are to be made between stations in a pollution survey, each
station must be sampled equally. Either an equal number of samples must be
taken at each station or an equal amount of time and effort must be
expended.
Organisms are not randomly distributed in nature, but tend to occur in
clusters. Because of this, it is necessary to take replicate samples in
orde<~ to obtain a composite sample that is representative of that station.
There is no "cookbook recipe" which defines the number of samples to take
in a given situation. Cairns and Dickson (1971) have found practical
than three artificial substrate samplers,
least three Surber square foot samples
samples required to describe the bottom
Naturally, increasing the number of
reliability of the data. The data of
samples taken at a given station are combined to form a pooled
It has been found that a plot of the pooled diversity index versus
experience to show that not less
3 to 10 dredge hauls, and at
represent the minimum number of
fauna of a particular station.
replicate samples increases the
replicate
sample.
IV-2-21
-------
cumulative sample units becomes asymptotic, and that once this asymptotic
diversity index value is found, little is gained by additional sampling.
Ideally, a base line study would be conducted to determine the optimum
number of samples for a pollution survey.
STATISTICAL ANALYSES
This section describes some of the statistical methods of comparing the
diversity indices calculated for different sampling stations.
Hutcheson's t-test
Hutcheson (1970) proposed a t-test for testing for difference between two
diversity indices:
Hl ' H2
t =
Hl ' H2
Where H}-H£ is simply the difference between the two diversity indices,
and
s
2 ]1/2
J
The variance of H may be approximated by:
I f1 log f1 - (If. log f.)/n
n
Where f-; is the frequency of occurrence
number of
associated
of species i and n
individuals in the sample. The degrees of
with the preceding t are approximated by:
^22 ,.22
is the
freedom
total
(df)
= (S
1
n.
n.
Convenient tables of filog2fi are provided by Lloyd, et al. (1968),
and t-distribution tables can be found in any statistics textbook (such as
Dixon and Massey, 1969; Zar, 1974; etc.). Example IV-2-1 demonstrates the
calculation of the Shannon-Wiener index (H) for two sets of hypothetical
sampling station data, and then tests for significant difference between
them using Hutcheson's t-test.
IV-2-22
-------
Example IV-2-1. Comparing Two Indices of Diversity (adapted from Zar
1974)7 ~~~
H0: The diversity index of station 1 is the same as the diversity
index of station 2.
HA: The diversity indices of stations 1 and 2 are not the same. The
level of significance (CL) = 0.05
Speci es
1
2
3
4
5
6
number of
individual s( i)
47
35
7
5
3
3
Station 1
percentaae( i)
47
35
7
5
3
3
ff log f .
78.5886
54.0424
5.9157
3.4949
1.4314
1.4314
i i
131.4078
83.4452
4.9994
2.4429
0.6830
0.6830
n. n
7T 1o9 n
-0.1541
-0.1596
-0.0808
-0.0651
-0.0457
-0.0457
100 100 144.9044 223.6613 -0.5510
Station 2
number of f , . . 2 f nf
Species individualsf f) percentagef i) M log fi ri I09 fi 7P
1 48
2 23
3 11
4 13
5 3
6 2
48
23
11
13
3
2
80.6996
31.3197
11. 4553
14.4813
1.4314
0.6021
135.6755
42.6489
11.9294
16.1313
0.6830
0.1813
-0.1530
-0.1468
-0.1054
-0.1152
-0.0457
-0.0340
100 100 139.9894 207.2494 -0.6001
IV-2-23
-------
HI = 0.5510 Hp = 0.6001
Sf; = 0.00136884 SJ] = 0.00112791
Hl H2
S = 0.0499
M1"H2
t = -0.98
df = 198.2 = 200
From a t-distribution table: t.
Q ntr?} ?00
Therefore, since the t value is not as great as the critical value for the
95 percent level of significance (CL = 0.05), the null hypothesis (H0) is
not rejected.
Analysis of Variance
Analysis of variance (ANOVA) can be used to test the null hypothesis that
all means are equal, e.g. H0: u^=U2 = .. .=U|<, where k is the number of
experimental qroups. "Single factor or "one-way" ANOVA is used to test
the effect of one factor (sampling site) on the variable in question
(diversity) in Example IV-2-2. Two-way ANOVA can be used for comparison of
spacial and temporal data.
In Example IV-2-2, each datum (X-j,-) represents a diversity index that
has been calculated for j replicate samples at each of i stations.
Also, x. represents the mean of station i, n^- represents the number of
replicates in sample i, and N(=£n.) represents the total number of indices
calculated in the survey.
After computing the mathematical summations, the ANOVA results are
typically summarized in a table as shown. The equality of means is
determined by the F test.
IV-2-24
-------
CL> groups df, error df = group MS
error MS
The critical value for this test is obtained from an F-distribution table
based on the degrees of freedom of both the numerator and denominator.
Since the computed F is at least as large as the critical value, H0 is
rejected, e.g. the diversity index means at all stations are not equal.
Example IV-2-2. A Single Factor Analysis of Variance (adapted from Zar
1974).
H0: Ul = u2 = u3 = u4 = u5
H^: The mean diversity indices of the five stations are not the same
Q = 0.05
Station 1
2. £2
3.32
3.64
3.46
2.91
3.10
Station
x .
i
n.
Station 2
3.96
4.08
3.79
3.71
4.36
4.24
1
3.21
6
Station 3
4.10
4.41
4.64
4.02
3.86
3.63
2 3
4.02 4.11
6 6
Station 4
4.63
4.21
4.35
4.88
4.37
4.01
4
4.41
6
Station 5
5.63
5.41
5.94
6.27
6.00
5.73
5
5.83
6
I x .
= U
n.
.
' J
19.25
/n. 61.76
I x
= '
24.14
97.12
/n = 580.84
.. = 129.49
24.67
101.43
c =
26.45
116.60
34.98
203.93
1
i = 583.21
II 12
= 558.92
total sum of squares = [ [ x^ -C
' J
1 J
24.29
IV-2-25
-------
groups sum of squares =
n.
- C = 21.92
error sum of squares = total ss - groups ss = 2.37
total degrees of freedom = N - 1 = 29
groups degrees of freedom = k - 1 = 4
error degrees of freedom = total df - groups df = 25
mean squared deviations from the mean (MS) = ss/df
groups MS = 21.92/4 = 5.48
error MS = 2.37/25 = 0.09
Summary of the Analysis of Variance
Source of Variation
SS
df
MS
total
groups
error
F = groups MS
error MS
24.29
21.92
2.37
5.480
0.095
29
4
25
57.68
5.480
0.095
F 0.05(1), 4, 25 = 2'76
Therefore, Reject H0 : u^u 2=u 3=u « = ur
Multiple Range Testing
The single factor analysis of variance tests whether or not all of the mean
diversity indices are the same, but gives no insight into the location of
the differences among stations. To determine between which stations the
equalities or inequalities lie, one must resort to multiple comparison
tests (also known as multiple range tests). The most commonly used methods
are the Student-Newman-Keul s test (Newman 1939, Keuls 1952) and the
Duncan's test (Duncan 1955).
Student-Newman-Keul s Test
Example IV-2-3 demonstrates the Student-Newman-Keul s (SNK) procedure for
the data presented in Example 2. Since the ANOVA in Example IV-2-2
rejected the null hypothesis that all means are equal, the SNK test may be
applied. First, the diversity index means are ranked in increasing order.
Then, pairwise differences ( xg-x/\ ) are tabulated as
IV-2-2 . The value of p is determined by the number of
of means being tested. Using the p value and the error
from the ANOVA, "studentized ranges," abbreviated ^.
shown in Example
means in the range
degrees of freedom
' are obtained
from a table of
calculated by:
q-distribution critical values. The standard error is
SE = (S2/n)1/2 = (error MS/n)1/2
If the k group sizes are not equal, a
For each comparison involving unequal n,
by:
slight modification is necessary.
the standard error is approximated
SE =
1/2
IV-2-26
-------
Example IV-2-3. Student-Newman-Keuls Multipie Range Test with Equal Sample
Sizes. This example utilizes the raw data and analysis of
variance presented in Example IV-2-2.
Ranks of sample means (i
Ranked sample means (x.)
1
3.21
2 •
4.02
3
4.11
4
4.41
5
5.83
SE = (error MS/n)1/? = (0.095/6)1/2 = 0.125
Comoa
(B vs.
5 vs
5 vs
5 vs
5 vs
4 V S
4 vs
4 vs
3 vs
3 vs
2 vs
MS on
A
1
• j.
. L
. w
t
. 1
, 2
» •-•
. 1
. 2
. 1
Di f Terence
5.83-3.21=2.62
5.83-4.02=1.81
".83-4.11=1.72
5.83-4.41=1.42
4.41-3.21=1.20
4. 41-^.02=0. 39
Do Not Test
4.11-3.21=0.90
Do Not Test
4.02-3.21=0.81
SE
0.
0.
0.
0.
0.
0.
0.
0,
126
125
126
126
126
126
126
126
q
20.79
14.37
13.65
11.27
9.52
3.10
7.14
6.43
P
5
4
3
2
4
3
3
2
in.
4.
3.
J *
2.
3.
2
3.
2.
05,24,p*
166
901
532
919
901
532
532
919
Concl usi
Reject
Reject
Reject
Reject
Reject
Accept
Reject
Reject
on
HO'
HO:
Ho:
H0:
H0:
HO:
HO:
u5=u1
u 5 = u £
U5=U3
U4 = u-j
U£ = U2
IJ3 = U1
U9 = U1
Since qn oc 0(- does not appear in the q-distribution table, qn n!- ?/, is used.
J.UO,£3,p U.uO,<-H,p
Overall conclusion: u-
-,
u9 = LU = u,
The q value is computed by:
= (XB - XA)/SE
If the computed
ther H
0
= u/\
q value is greater
is rejected.
than or equal to the critical value,
In Example 3, after accepting H0:u,i=U2 there is no need to test 4 vs.
3 or 3 vs. 2. The conclusions drawn in the example are that the community
at Station 1 has a si gni fi cantly - di fferent mean diversity index from all
other sampled communities; likewise, the Station 5 mean is different from
the others. However, the communities at Stations 2, 3, and 4 have
statistically equal diversity index means. These conclusions can be
visually represented by
different with a common
underlining the means that
line as shown below:
are not significantly
station 1234
meen diversity index 3.21 4.02 4.11 4.41
5
5.83
Conversely, any two means not underscored by the same line are
sigri ficantly different.
Duncan's Multiple Range Test
The theoretical basis of the Duncan's test is somewhat different from the
Stucent-Newman-Keul test, although the procedures and conclusions are quit
similar. Duncan's test makes use of the concept of Least Sigm'fican
IV-2-27
-------
Difference (LSD) which is related to the t-test.
discussed previously. The LSD is calculated by:
a form of which was
LSDa =
(2S2/n)1/2
where
is the
replications, and
freedom (MS and
variance). After
mean square for
t is the tabulated t
df for error are
determining p as in
error,
value
n
for
is the number of
the error degrees of
calculated in the analysis of
the SNK procedure, R values are
and
obtained from a table dependent on the level of significance, error df,
p. The shortest significant difference (SSD) is computed by the equation:
SSD = R(LSD)
Example IV-2-4 demonstrates Duncan's procedure for hypothetical data. As
before, the difference between means is calculated for every possible
pairwise comparison of means. This difference is then compared to the
corresponding SSD value and conclusions are drawn. If the difference is at
least as large as the SSD, then the null hypothesis - that the two means
are equal - is rejected; if the difference is less than SSD, H0 is
accepted. The results are visually represented as described for the SNK
test.
Example IV-2-4. Duncan's Multiple Range Test.
H0: Ul=u2=u3=u4
H^: The mean diversity indices of the four sampling stations are not
the same
d= 0.05
n = 4
Ranks of sample means (i)
Ranked sample means (x-j )
LSD
0.05
. 05
error MS = 0.078
1
5.3
= 0.447
2
5.7
3
5.9
error df=9
4
6.3
Compari son
(B vs.
4 vs.
4 vs.
3 vs.
3 vs.
2 vs.
mean di
visual
A )
1
3
1
2
1
versi
Di f ference
(XB - "
6.3-5.
6.3-5.
6.3-5.
5.9-5.
5.9-5.
5.7-5.
station
ty index
X{\ )
3=1. 0
7=0.6
9=0.4
3=0.6
7 = 0.2
3=0.4
1
5.3
representation
P
4
3
2
3
2
2
2
5.7
R
a
i
i
i
i
i
i
3
5.9
SSD
,df,p
.07
.04
.00
.04
.00
.00
4
6.3
=R(LSD)
0.
0.
0.
0.
0.
0.
48
46
45
46
45
45
Conclusion
reject
reject
accept
reject
accept
accept
Ho
Ho
HO
Ho
HO
HO
:U4=U]
:u4=u2
:u4=u3
:u3=Ul
:u3=u2
:U2=U1
IV-2-28
-------
COMMUNITY COMPARISON INDICES
Introduction
Whereas the statistical analyses discussed above can discern significant
differences between diversity indices calculated at two or more sampling
stations, community comparison indices have been developed to measure the
degree of similarity or dissimilarity between communities. These Indices can
detect spatial or temporal changes in community structure. Polluted
communities presumably will have different species occurrences and abundances
than relatively non-polluted communities, given that all other factors are
equal. Hence, community comparison indices can be used to assess the impact of
pollution on aquatic biological communities.
There are two basic types of community comparison indices: qualitative and
quantitative. Qualitative indices use binary data: in ecological studies, the
two possible attribute states are that a species is present or is not present
in the collection. This type of community similarity index is used when the
sampling data consists of species lists. Kaesler and Cairns (1972) considered
the jse of presence-absence data to be the only justifiable (and defensible)
approach when comparing a variety of organism groups (e.g. algae and aquatic
insects). Also, qualitative similarity coefficients are simple to calculate.
When data on species abundance are available, quantitative similarity Indices
can De used. Quantitative coefficients incorporate species abundance as well
as occurrence in their formulas, and thus, retain more information than
indices using binary data. An annotated list of community comparison indices
of both types appears in Table IV-2-9.
Qua 11 tative, Sirnijaunty^Jjdices
Although the terminology used in the literature varies considerably, the
qualitative similarity indices in Table IV-2-9 (1 - 6) are represented using
the symbolism of the 2X2 contingency table shown in Figure IV-2-5. In the form
of the contingency table shown, collections A and B are entities and all of
the species represented in a collection are the attributes of that entity.
Indices 1 through 4 in Table IV-2-9 are constrained between values of 0 and 1,
while equation 6 has a potential range of -1 to 1. The minimum value
represents two collections with no species in common and the maximum value
indicates structurally identical communities.
According to Boesch (1977), the Jaccard, Dice, and Ochiai coefficients are the
most attractive qualitative similarity measures for biological assessment
studies. The Jaccard coefficient (1) is superior for discriminating between
high'y similar collections. The Dice (2) and Ochiai (4) indices place more
emphasis on common attributes and are better at discriminating between highly
dissimilar collections (Clifford and Stehpenson, 1975; Boesch, 1977; Herricks
and Cairns, 1982). Thus, the nature of the data determines which index is most
suitable. The Jaccard coefficient has been widely used by some workers in
stream pollution investigations (Cairns and Kaesler, 1969; Cairns et al.,
1970. Cairns and Kaesler", 19/1; 'Kaesler at al., 1971; Kaesler and Cairns,
1972. Johnson and Brinkhurst, 1971; Foerster et al., 1974). Peters (1968) has
written BASIC computer programs for calculating Jaccard, Dice, and Ochiai
i ndi ces.
IV-2-29
-------
oo
LO
QJ
C
un
CTi
CM
in
QJ
O
C
QJ
i_
QJ
i_
QJ
cc
•o
i.
re
u
ro
O
c
ro
^
QJ
Isl
•O
C
re
o
to
in
cr.
CTi
u
Cfl
QJ
O
cri
QJ
O
m
OJ
o
ca
"O
c
re
>!
re
i.
CO
QJ
^
ro
.Q
•r-
Q.
•i—
E
LiJ
O
Q
O
CO
c\j
o
ii
r«H
-C
4-
re
SBH^-
CM
1
T3 re
4-
U
+
JO
"^
re
^j
F»H
I— 1
*"• ^
o
4-
re ro
^«*
X!
4-
re
i — i
CM U
~-s JD
F- 1 1
1 1 jQ
»— re
u
4-
re
M-^.
JD
4-
re
L-_J
Ei u
"^L_,
— . JD
•O 'i~
4. . —
I"* rt
«L« Jul
.— • «f—
. — . X
u
4- -
re re
^_« *r~
. — • X
•o • —
4- C
O •--
^-* E
.^v.
.0 t— j c
-f
re CM
o.
•— . i
J2
**— •—
x jo
•^
4- X
ro
•i- i
X
— • re
X
-j . —
c. — 3 :
— ro
jD -t-
•r- Q.
u— j
+
Ln
re •
•r- O
X
-— O ^H
i i n n n
^s
ro
S
O
0
>-
\—
$
C
— 2
z E
ro
2:
2:
o
o
u_
o
>-
o:
<
s
s
^3
CO
I
O-i
1
CM
1
>•
LU
— 1
=3
<
1—
E
0
o
u-
o
4-J
c
QJ
•^
o
•f—
u_
u-
QJ
E
re
z
QJ
>
*^
4->
0.
S_
CJ
1/1
QJ
O
QJ
O
O
•c
S.
re
o
(j
re
•
i— i
.^^
c
o
c
QJ
L-
o
u->
f
•r-
.*:
ISI
5
O
C
ro
.*:
QJ
N
O
"^--
X
QJ
•c
C
i— <
QJ
re
S
QJ
^~
O.
£
•r-
to
s_
a>
c
QJ
jc
o
•r—
S
T3
C
re
i — X
ro QJ
.* -o
O C
00 i—
«
OO
*— •
re
^
3
V)
4->
O
^-^
X
OJ
X3
C
>— 1
•f—
re
•r—
£:
!_)
O
.
«d-
X
eu
•o
c
t — 1
S-
flj
cn
ro
u.
.
uo
4->
C
QJ
.^
O
•?•
14-
>4-
QJ
0
O
c
o
.^
4-»
re
?"-
OJ
i_
s_
o
o
•M
c
•r~
o
Q.
.
<£>
O
•r-
4->
re
•p—
u
o
VI
VI
<
u-
o
4->
C
QJ
•^
O
•f—
14-
W-
QJ
O
CJ>
^—
r-~
re
•o
c
CD
^.
*->
c
QJ
0
U-
14-
QJ
O
O
>,
4->
•r—
L.
re
P—
*P»
E
*r-
OO
in
•^
4-J
^
3
O
1
>,
re
s_
CO
•
r^
>>
+J
C
re
•^-
E
»r—
V)
l/l
•^>
O
in 4->
•^ c
+-> QJ
S_ •!-
3 U
O •>-
1 >4-
>51*-
re QJ
s- o
CD O
>,
4?
cr
3
JS
o
c_>
*+-
o
>,
4-7
•^*
i-
ns
*^-
E
uo
QJ
CD
re
4->
C
QJ
O
QJ
Q.
IV-2-30
-------
Ol
01
ro
O)
Q.
"O
C
ro
H
ro
.C
JK:
C
Q_
en
LT)
Ol
ro
4->
to
s_
o
VD
CTl
C
!_
O
O
trt
O
CO
IO
CTi
re
_*:
O
O
to
fO
01
c
oo
JO
IX
fO
c
i
CVJ
J°
to
II
_d
ro
to
3:
!
X
I
i:
J3
rO
to
•9
ro
10
IX
I
-Q
•i—
X
/o
IX
ro
CO
OJ
OJ!
l
X
OJ
•o
c:
c
o
t/1
i_
ro
QJ
O.
"O
X
O)
T3
S_
o
CL
ro
I.
O
o
u~
o
GJ
T3
C
C
O
QJ
O
c
ro
+J
i/l
Qj c
s- o
!^ l^
O i.
co fa
01
4-> Q.
O C
s: CD
I T—
O OJ
S- O
Q- CJ
CO
IV-2-31
-------
TABLE IV-2-9 (continued)
Key: S = similarity between samples.
D = dissimilarity between samples.
a,b,c,d = (see Figure IV-2-5).
x. , x.K = number of individuals of species i at Station A or B.
ia ID
p. , p.. = relative abundance of species i at Station A or B.
i a ID
X , X, = total number of individuals at Station A or B.
a D
n = total number of different taxa.
X , X = Simpson diversity index for Station A or B.
a D
H , = Shannon-Wiener diversity index of Station A and B combined.
ab J
H = maximum possible value of H , .
max ab
H . = minimum possible value of H b.
*1a + *1b log xia * xib
_
ab " " Xa + Xb Xa + Xb
_ (Xa ^ Xb) log (Xa + Xb) - I x.a log x.a - \ x-b log
Hmax
H . =
min
X + X.
a b
IV-2-32
-------
COLLECTION A
CQ
O
t—M
t—
O
c
-------
The Fager coefficient (5) is simply a modification of the Ochiai index.
Because a correction factor is subtracted from the Ochiai index, the Fager
coefficient may range from slightly less than zero to slightly less than one;
this makes it less desirable. The Fager index has been used a great deal in
marine ecology.
Both the Sokal and Michener index (3) and the Point Correlation Coefficient
(6) include the double-absent term d. A number of authors (Kaesler and Cairns,
1972; Clifford and Stephenson, 1975; Boesch, 1977) have criticized the
approach of considering two collections similar on the basis of species being
absent from both.
Pinkham and Pearson (1976) illustrated the weaknesses of qualitative
comparison indices. The basic shortcoming is that two communities having
completely different species abundances but the same species occurrence will
produce the maximum index value, indicating that the two collections are
identical.
Quantitative Comparison Indices
Quantitative indices (7 - 12) consider species abundance in addition to mere
presence-absence. Incorporating species abundance precludes the over-emphasis
of rare species, which has been a criticism of the Jaccard coefficient
(Whittaker and Fairbanks, 1958). Quantitative measures are not as sensitive to
rare species as qualitative indices and emphasize dominant species to a
greater extent. Distance (11), information (9, 10), and correlation (12)
coefficients weight dominance even more than other quantitative indices.
Quantitative indices also avoid the loss of information involved in
considering only presence-absence data when species abundance data are
available. However, data transformations (e.g., to logarithms, roots, or
percentages) may be desirable or necessary for the use of some quantitive
comparison indices. Calculation of quantitative indices is more complicated
than qualitative coefficients, but can be facilitated by computer application.
The Bray-Curtis index (7) is one of the most widely used quantitive comparison
measures. Forms of this index have been referred to as "index of associaton"
(Whittaker, 1952), as "dominance affinity" (Sanders, 1960), and as "percentage
similarity of community" (Johnson and Brinkhurst, 1971; Pinkham and Pearson,
1976; Brock, 1977). The simplest and probably most commonly used form of the
Bray-Curtis index is the Percent Similarity equation:
Sab =2,min(pia,pib)
where the attributes have been standardized into a proportion or percent of
the total for that entity (collection). The shortcoming of the Percent
Similarity coefficient was illustrated by Pinkham and Pearson (1976) as shown
below.
IV-2-34
-------
TAXA
ABODE
Station A
Station B
40 20 10 10 10
20 10 5 5 5
In this hypothetical comparison, all species are
A as at Station B but their relative abundance
maximum similarity value of 1.0 is registered.
situation is germane to pollution assessment
difference between two sampling stations is the
eutrophication.
twice as abundant at Station
is identical; therefore, the
The authors felt that this
surveys in which the only
relative degree of cultural
In Table IV-2-9, the Bray-Curtis index is displayed as both a measure of
similarity and dissimilarity. Any community similarity index can be converted
to a dissimilarity measure by the simple equality:
D = 1 - S
Of course, values obtained by a dissimilarity expression are inversely related
to similarity values; they increase with decreasing similarity.
Pinkham and Pearson (1976) presented a community similarity index (8) that
would overcome the shortcomings of other indices (e.g. 1,3,7,12) that were
discussed in the article. Their similarity coefficient can be calculated
using either actual or relative (percent) species abundance, although they
suggested using actual abundance whenever possible. The authors also offered
a modified formula that includes a weighting factor for assigning more
significance to dominant species:
'ab
x
-------
the Morisita index ranges from zero for no resemblance to one for identical
collections. The Horn Index of Overlap is a manipulation of Shannon's
information theory equation that closely resembles the expression of community
redundancy developed by Margalef:
R = (K - H)/ (H - H . )
v max max mm
The observed value in Horn's index (Hab) is the Shannon index calculated for
the sum of the two collections being considered. The maximum diversity value
(Hmax) would occur if the two collections contained no species in common, and
the minimum diversity value (Hmin) would be attained if the two collections
contained the same species in the same proportions. It should be noted that
the equations given for Hab, Hmax, and Hmin in the key to Table IV-2-9 are
adapted from those given by Perkins (1983) since those appearing in the
original article (Horn, 1966) are apparently inconsistent with the Shannon
index. The Morisita and the Horn indices have been used in aquatic ecology
studies (Kohn, 1968; Bloom et al., 1972; Livingston, 1975; Heck, 1976).
If two entities (i.e. communities) are thought of as points in an
n-dimensional space whose dimensions are determined by their attributes (i.e.
species occurrence and abundance ), then the linear distance between the two
points in the hyperspace can be construed as a measure of dissimilarity
between the two entities. The two distance formulas shown in Table IV-2-9
(11) are simply forms of the familiar geometrical distance formula,
which has been expanded to accomodate n dimensions. Sokal (1961) divided the
distance by n to produce a mean squared difference, which he felt was an
appropriate measure of taxonomic distance. Values computed by the distance
formulas may range from zero for identical collections to infinity; the
greater the distance the less similar the two comunities are. Because the
difference in species abundance is squared in the numerator, the distance
formulas are heavily influenced by abundant species and may over-emphasize
dominance. The similarity of disparate communities with low species
abundances may be overstated, while the resemblance of generally similar
communities with a few disproportionately nigh species abundances may be
understated. To avoid indicating misleading resemblance, it may be necessary
to transform data (e.g. to squared or cubed roots) before computing taxonomic
distance.
The Product-Moment Correlation Coefficient (12) is a popular resemblance
measure that ranges from -I (completely dissimilar) to +1 (entirely similar).
Several undersirable characteristics of this measure have been cited (Sneath
and Sokal, 1973; Clifford and Stephenson, 1975; Boesch, 1977). Deceptive
resemblance values can result from outstandingly high species abundances or
the presence of many species absences, and non-identical communities can
register perfect correlation scores. Pinkhair and Pearson (1976) demonstrated
how the Product-Moment Correlation Coefficient, like the Percent Community
Similarity Index, indicates maximum similarity for two communities having the
same relative species composition but different actual species abundances.
-2-36
-------
Experimental Evaluation of Comparison Indices
Brock (1977) compared the Percent Community Similarity Index (7) and the
Pinkham and Pearson Similarity Index (8) for their ability to detect changes
in the zooplankton community of Lake Lyndon B. Johnson, Texas, due to a
thermal effluent. For this study, the Pinkham and Pearson index was
considered too sensitive to rare species and not sensitive enough to dominant
forms, whereas the Percent Similarity coefficient was more responsive to
variation in dominant species and relationships between dominant and
semi-dominant forms. Linking dominance to function, the author concluded that
the later index may better indicate structural-funcitonal similarity between
communities.
Perkins (1983) evalutaed the responsiveness of eight diversity indices and
five community comparison indices to increasing copper concentrations. The
indices were calculated for bioassays conducted using benthic
macroinvertebrates and artificial streams. The indices evaluated by Perkins
correspond to equations presented in Tables IV-2-2 and IV-2-9 except: Perkins
tested the Bray-Curtis dissimilarity index; Perkins' Biosim index is Pinkham
and Pearson's index, and the distance formula tested by Perkins (not included
in this report) is shown below.
D -
The results of the study appear
are presented for comparison.
x. -x.,
T3 ib
Xia+xib
1/2
in Figure IV-2-6; the diversity index results
The diversity indices did not clearly demonstrate the perturbation caused by
increasing copper concentrations. The Shannon and Brillouin formulas
increased initially, in spite of a decreasing number of species, because of
increasing evenness of species distribution. Other than the increasing
diversity indicated at the lower copper concentrations, these two indices
refected perturbation effectively by decreasing rapidly with increasing
poVutant concentration. The Mclntosh, Simpson, and Pielou (evenness) indices
(not shown for 28 days in Figure IV-2-6) resembled the trends demonstrated by
the Shannon and Brillouin formulas albeit less dramatically. Because the
results obtained for those three indices were less pronounced, they were more
difficult to interpret than the Shannon and Brillouin findings.
The community comparison indices were found to be good indicators of the
perturbation of macroinvertebrate communities caused by copper pollution.
Although the Bray-Curtis index was considered the most accurate after 14 days,
all of the comparison indices tested effectively reflected community response
after 28 days (see Figure IV-2-6). Note that by definition the Biosim,
Monsita, and Percent Community Similarity indices decrease as similarity
dec-eases, while the Distance and Bray-Curtis dissimilarity indices increase.
It has frequently been suggested that it may be desirable to apply several
indices in a pollution assessment study (Peters, 1968; Brock, 1977; Perkins,
1983).
IV-2-37
-------
OS" 10 15
Log Cu*"lfig/!)
l=Shannon
2=Brillouin
3=P1elou
4=Simpson
5=McIntosh
6=Menhinick
7=Species(xlO)
8=Equitabi1 it.y
(b)
81-
<5
S 5:
a
I 4I
'0
05 " "iO
LogOT fyug/D
(c)
i.o
91
3
l=Distance
2=Bray-Curtis
3=% Similarity
4=Morisita
5=Biosim
2.0
Log
d)
Figure IV-2-6.
Evaluation of diversity indices and community comparison indices
using bioassay data: a,c=after 14 days; b,d=after 28 days (from
Perkins, 1983).
IV-2-38
-------
Numerical Classification or Cluster Analysis
A common use of similarity indices is in numerical clasification of biological
communities. Numerical classification, or cluster analysis, is a technique
for grouping similar entities on the basis of the rsemblance of their
attributes. In instances where subjective classification of communities is
not clear-cut, cluster analysis allows incorporation of large amounts of
attribute data into an objective classification procedure. Kaesler and Cairns
(1972) outlined five steps involved in normal cluster analysis. First, a
community similarity index is chosen based on pre-determined criteria and
objectives. Second, a matri-x of similarity coefficients is generated by
pairwise comparison of all possible combinations of stations. The third step
is the actual clustering based on the resemblance coefficients. A number of
clustering procedures are discussed in the literature (Williams, 1971; Sneath
and Sokal, 1973; Hartigan, 1975; Boesch, 1977). In the fourth step, the
clustered stations are graphically displayed in a dendogram. Because
multi-dimensional resemblance patterns are displayed in two dimensions and
because the similarity coefficients are averaged, a significant amount of
distortion can occur. For this reason, a distortion measure should be
evaluated and presented as the fifth step in the cluster analysis. The
Cophenetic Correlation Coefficient (Sokal and Rohlf, 1962) is a popular metric
of display accuracy. An additional step in any cluster analysis application
should be interpretation of the numerical classification results since the
technique is designed to simplify complex data and not to produce ecological
interpretation.
SUMMARY
The ability of a water resource to sustain a balanced biotic community is one
of tne best indicators of its potential for beneficial use. This ability is
essential to the community's health. Although several papers have criticized
the jse of diversity indices (Hurlbert,1971; Peet,1975; Godfrey,1978), Cairns
(1977) stated that "the diversity index is probably the best single means of
assessing biological integrity in freshwater streams and rivers". Cairns
concluded that no single method will adequately assess biological integrity,
but rather its quantification requires a mix of assessment methods suited for
a specific site and problem. The index of diversity is an integral part of
that mix. Community comparison indices are also useful in assessing the
biological health of aquatic systems. By measuring the simiarity (or
dissimilarity) between sampling stations, community comparison indices
indicate relative impairment of the aquatic resource.
IV-2-39
-------
CHAPTER IV-3
RECOVERY INDEX
It is important to examine the ability of an ecosystem to recover from
displacement due to pollutional stress in order to evaluate the potential
uses of a water body. Cairns (1975) developed an index which gives an
indication of the ability of the system to recover after displacement. The
factors and rating system for each factor are:
(a) Existence of nearby epicenters (e.g., for rivers these might be
tributaries) for providing organisms to reinvade a damaged system.
Rating System : l=poor, 2=moderate, 3=good
(b) Transportability or mobility of disseminules (the disseminules might be
spores, eggs, larvae, flying adults which might lay eggs, or other stages
in the life history of an organism which permit it to move to a new area).
Rating System : l=poor, 2-moderate, 3=good
(c) Condition of the habitat following pollutional stress (including
physical habitat and chemical quality).
Rating system : l=poor, 2=moderate, 3=good
(d) Presence of residual toxicants following pollutional stress.
Rating System : l=large amounts, ?=rnoderate amounts, 3=none
(e) Chemical-physical environmental quality after pollutional stress.
Rating System : l=in severe disequilibrium, ?=partially restored,
3=norma1
(f) Management or organizational capabilities for control of damaged area.
Rating system : l=none, 2-some, 3=strong enforcement possible.
Using the characteristics listed above, and their respective rating
systems, a recovery index can be developed. The equation for the recovery
index follows:
Recovery Index = a xbxcxdxexf
400+ = chances of rapid recovery excellent
55-399 = chances of rapid recovery fair to good
less than 55 = chances of rapid recovery poor
This index and the rating system was developed by Cairns based on his
experience with the Clinch River. For a full description of the rationale
for the rating factor, the reader should refer to Cairns (1975).
IV-3
-------
CHAPTER IV-4
INTOLERANT SPECIES ANALYSIS
NICHE CONCEPT
The ecological niche of a species is its position and role in the biological
community. Hutchinson (1957) described niche as a multidimensional space, or
hypervolume, that is delineated by the species' environmental requirements and
tolerances. Physical, chemical, and biological conditions and relationships
constitute the dimensions of the hypervolume, and the magnitude of each dimen-
sion is defined by the upper and lower limits of each environmental variable
within which a species can persist. If any one of the variables is outside of
this range the organism will die, regardless of other environmental conditions.
TOLERANCE
The "Law of Toleration" proposed by Shelford (1911) is illustrated in Figure
IV-4-1. For each species and environmental variable there is a range in the
variable intensity over which the organism functions at or near its optimum
level. Outside the maximum and minimum extremes of the optimum range there are
zones of physiological stress, and, beyond, there are zones of intolerance in
which the functions of the organism are inhibited. The upper and lower toler-
ance limits (also called incipient lethal levels) are intensity levels of the
environmental variable that will eventually cause the death of a stated frac-
tion of test organisms, usually 50 percent.
VARIABILITY OF TOLERANCE
The tolerance of an organism for a lethal condition is dependent on its gene-
tic constitution - both its species and its individual genetic makeup - and
its early and recent environmental history (Warren 1971). Acclimation has a
marked effect on the tolerance of environmental factors such as temperature,
dissolved oxygen, and some toxic substances (see Figure IV-4-2). Tolerance is
also a function of the developmental stage of the organism and it may change
with age throughout the life of the animal. Because of this variability, no
two organisms have exactly the same tolerance for a lethal condition and toler-
ance limits must be expressed in terms of an "average" organism.
INTERACTIONS INFLUENCING TOXICITY
An organism's tolerance for a particular lethal agent is dependent not only on
its own characteristics but also on the environmental conditions. The inter-
actions between lethal and nonlethal factors are well documented and are ad-
dressed elsewhere in this handbook (Chapters II-5 and III-2). Briefly, these
nonlethal effects include:
IV-4-1
-------
Lower limit of tolerance
Upper limit of tolerance •
Zone of
physiological
stress
Zone of
physiological
stress
Low
Figure IV-4-1.
GRADIENT-
Law of toleration in relation to distribution and
level--often a normal curve (modified by Kendeigh
Shelford (1911)).
-»-High
copulation
(1974) from
10 - 20 2
Acclimation temperature (C)
Figure IV-4-2. The zones of tolerance of brown bullheads (Ictalurus nebulosus
and chum salmon (Oncorhynchus keta) as delimited by incipient
lethal temperature and influenced by acclimation temperature
(after Brett 1956).
IV-4-2
-------
Hardness. Increasing hardness decreases the effect of toxic metals on aqua-
tic organisms by forming less-toxic complexes.
pH. The dissociation of weak acids and bases is controlled by pH and either
the molecular or ionic form may be more toxic.
Alkalinity and Acidity. These modify pH by constituting the buffering capa-
city of the system.
Temperature. Increasing temperature enhances the effect of toxicants by in-
creasing the rates of metabolic processes.
Dissolved Oxygen. Decreasing dissolved oxygen concentration augments the
exposure and absorption of toxicants by increasing the necessary irriga-
tion rate of respiratory organs.
When two or more lethal agents are present, several types of interactions are
possible: synergistic, additive, antagonistic, or no interaction.
INTOLERANT SPECIES ANALYSIS
The tolerance ranges for environmental variables differ widely between spe-
cies. Thus, the range of conditions under which an organism can survive (its
niche) is broader for some species than it is for others. Fish species with
narrow tolerance ranges are relatively sensitive to degradation of water qual-
ity and other habitat modifications, and their populations decline or disap-
pear under those circumstances before more tolerant organisms are affected. In
general, intolerant species can be identified and used in evaluating environ-
mental quality. The presence of typically intolerant species in a fish sam-
pling survey indicates that the site has relatively high quality; while the
absence of intolerant species that, it is judged, would be there if the envi-
ronment was unaltered indicates that the habitat is degraded.
LISTS OF INTOLERANT FISH SPECIES
While the tolerance limits of a fish species for a particular environmental
factor can be defined relatively precisely by toxicity bioassays, its degree
of tolerance may vary considerably over the range of physical, chemical, and
biological variables that may be encountered in the environment. The variables
that are the object of intolerant species analysis are intentionally left
vague in order to accommodate the variety of situations precipitated by man's
activities. A species may be intolerant of alterations in water quality or in
habitat structure, such as those listed below.
Water Quality Changes Habitat Alterations
increased turbidity substrate disruption
increased siltation cover removal
increased water temperature changes in velocity and discharge
increased dissolved solids removal of instream and streamside
organic enrichment vegetation
lowered dissolved oxygen water level fluctuation
impoundment and channelization
blockage or hinderance of migration
IV-4-3
-------
Many species can be identified that are relatively intolerant of anthropogenic
alterations of the aquatic environment compared to other fish, Appendix C con-
tains a list of fish species, nationally, which are relatively intolerant to
one or more of the environmental changes shown above. The information in Appen-
dix C is based on literature sources (Wallen 1951; Trautman 1957; Garlander
1969, 1977; Scott and Grossman 1973; Pflieger 1975; Moyle 1976; Timbol and
Maciolek 1978; Smith 1979; Muncy et al. 1979; Lee et al. 1980; Morrow 1980;
Johnson and Finley 1980; U.S. EPA 1980; Karr 1981; Haines 1981; and Ball 1982)
and on the professional judgment of State and University biologists.
The darters and sculpins are listed only by genus in Appendix C. Identifica-
tion of those taxa to species would have been inconvenient (together, Ammoj^
crypta, Ethegstoma, Percina, and Cottus contain 150 species in the United
States) and largely unnecessary because, with a few possible exceptions, all
of the species of darters and sculpins can be considered intolerant. Karr
(1981) recognized the johnny darter (Etneostoma nigrum) as the most tolerant
darter species in Illinois and Ball (1982) did" not categorize the johnny
darter as an intolerant forage fish. Other darter species that appear to be
relatively more tolerant of turbidity, silt, and detritus than others in their
genus are listed below:
mud darter Etheostoma asprigene
bluntnose darter E . chjongsomum
slough darter F'. gracile
cypress darter ET proe1iare
orangethroat darter F. spectabiTe
swamp darter F. TusTfprme""
river darter Fercina shumardi
The list in Appendix C is intended to be used by knowledgeable biologists as a
rough guide to the relatively intolerant fish species in their state. Site-
specific editing is left to persons familiar with the local fish fauna and en-
vironmental conditions. Local editing of the provided data should produce a
workable list for intolerant species analyses of the streams in that area.
IV-4-4
-------
OMNIVORE-CARNIVORE (TROPHIC STRUCTURE) ANALYSIS
INTRODUCTION
Water pollution problems nearly always involve changes in the pathways by
which aquatic populations obtain energy and materials (Warren 1971). These
changes lead to differential success of constituent populations which affects
the composition of the aquatic community. Anthropogenic introduction of or-
ganic substances or mineral nutrients directly increases the energy and ma-
terial resources of the system, but other pollution problems - such as pH or
temperature changes, toxic materials, low dissolved oxygen, turbidity, silta-
tion, et cetera - also lead to changes in trophic pathways. Thus, the health
of a system can be evaluated through a study of its trophic structure. The
following material concentrates on stream and river systems. Lakes will have
different structural aspects.
TROPHIC STRUCTURE
The ecosystem has been described as the entire complex of interacting physi-
cochemical and biological activities operating in a relatively self-supporting
community (Reid and Wood 1976). The biological operations of an ecosystem can
be viewed as a series of compartments which are described by three general cat-
egories: producers, consumers, and decomposers. The producers include all auto-
trophic plants and bacteria (both photosynthetic and chemosynthetic) which, by
definition, are capable of synthesizing organic matter from inorganic sub-
strates. The consumers are heterotrophic organisms that feed on other organ-
isms, and are typically divided into herbivores and carnivores. Herbivores
(primary consumers) feed principally on living plants while carnivores (sec-
ondary, tertiary, and quarternary consumers) feed principally on animals that
they kill. Another type of consumer, ,the omnivore, feeds nearly equally on
plants and animals, and occupies two or more trophic levels. The decomposers
include all organisms that release enzymes which break down dead organisms.
Food chains are sometimes used to simply represent feeding relationships be-
tween trophic levels (e.g., plant > herbivore > carnivore). Ecosystems common-
ly contain three to five links in their food chains. Diagramming all of the
pathways of energy and material transfer in a community entails many inter-
connecting food chains, forming a complex food web.
The concept of trophic structure, first formally discussed by Lindeman (1942),
is a method of dealing with the pathways of energy and material transfer which
focuses on functional compartments without considering the specific feeding
relationships. The pathways between functional compartments are illustrated in
Figure IV-5-1. Trophic structure is commonly represented by trophic or ecolog-
ical pyramids. An ecological pyramid is a diagramatic representation of the r-
elationships between trophic levels arranged with the producers making up the
base and the terminal or top carnivore at the apex. An ecological pyramid may
IV-5-1
-------
HERBIVORES .
CARN1VORFS
1C.)- (C.)-(C,)
PRODUCERS
DECOMPOSERS
Figure IV-5-1. Trophic pathways of an ecosystem (after Reid and Wood 1976)
S-5
, C.-1.5
C.-11
-
C.-37
P-809
kcal/m'
(•1
S-5060
C,-383
C,-3368
P-20,810
Figure IV-5-2.
kcal/m'/YEAR
Ecological pyramids for Silver Springs, Florida, indicating (a)
biomass and (b) productivity. P=producers; C=consumers; S=sapro-
phytes or heterotrophs (after Odum 1957).
IV-5-2
-------
represent the number of Individuals that compose each trophic level, or, of
more ecological significance, the biomass or productivity of each level (Fig-
ure IV-5-2). Because energy transfer between trophic levels is less than 100 p-
ercent efficient the pyramid of productivity must always be regular in shape,
while pyramids of numbers and biomass may be partially inverted in some in-
stances (Richardson 1977).
TROPHIC STRUCTURE OF FISH COMMUNITIES
Fish communities generally include a range of species that represent a variety
of trophic levels. The trophic classification system shown below was used in
the assessment of fish fauna of the Illinois and Maumee River basins (Karr and
Dudley 1978, Karr et al. 1983).
(1) Invertivore - food predominantly (>75%) invertebrates.
(2) Invertivore/Piscivore - food a mixture of invertebrates and fish; rela-
tive proportions often a function of age.
(3) Planktivore - food dominated by microorganisms extracted from the water
column.
(4) Omnivore - two or more major (>2b% each) food types consumed.
(5) Herbivore - feed mostly by scraping algae and diatoms from rocks, and
other stream substrates.
(6) Piscivore - feed on other fish.
Schlosser (1981, 1982a, 1982b) used the trophic structure of fish communities
to investigate differences in Illinois stream ecosystems. His categorization
scheme appears in Table 1.
In addition to representing a range of trophic levels, fish utilize foods of
both aquatic and terrestrial origin, and occupy a position at the top of the
aquatic food web in relation to plants and invertebrates. These facts enhance
the ability of fish communities to provide an integrative view of the water-
shed environment (Karr 1981).
BIOLOGICAL HEALTH
Degradation of water quality and habitat affects the availability of many food
resources, resulting in changes in the structure and functions, and, thus, the
health of the aquatic community. Structural characteristics include the num-
bers and kinds of species and the number of individuals per species. These
parameters can be evaluated relatively quickly via compilation of species
lists, calculation of diversity indices, and identification of indicator spe-
cies. The importance of evaluating the impact of pollution on community func-
tions - such as production, respiration, energy flow, degradation, nutrient
cycling, and other rate processes - is becoming increasingly evident, and,
ideally, any study of community health should include both structural and
functional assessment. However, use of functional methods has been hindered
because they are often expensive, time-consuming, and not well understood.
IV-5-3
-------
TABLE IV-5-1.
TROPHIC GUILDS USED BY SCHLOSSER (1981, 1982A, 1982B)
TO CATEGORIZE FISH SPECIES
Herbivore - detritivores (HD)
Omnivores (OMN)
Generalized Insectivores (GI)
Surface and Water Column
Insectivores (SWI)
Benthic Insectivores (BI)
Insectivore - Piscivores (IP)
HD species fed almost
toms or detritus.
entirely on dia-
OMN species consumed plant and animal
material. They differed from GI species
in that, subjectively, greater than 25
percent of their diet was composed of
plant or detritus material.
GI species fed on a range of animal and
plant material including terrestrial
and aquatic insects, algae, and small
fish. Subjectively, less than 25 per-
cent of their diet was plant material.
SWI species fed
or terrestrial
surface.
on water column drift
insects at the water
BI species fed predominantly
ture forms of benthic insects.
on imma-
IP species fed on aquatic invertebrates
and small fish. Their diets ranged from
predominantly fish to predominantly in-
vertebrates.
IV-5-4
-------
Examining the trophic structure of a community can provide insight into its
production and consumption dynamics. A trophic-structure approach to the study
of the functional processes of stream ecosystems has been proposed by Cummins
arid his colleagues (Cummins 1974, 1975; Vannote et al. 1980). Their concept
assumes that a continuous gradient of physical conditions in a stream, from
its headwaters to its mouth, will illicit a series of consistent and predict-
able responses within the constituent populations. The River Continuum Concept
identifies structural and functional attributes that will occur at different
reaches of natural (unperturbed) stream ecosystems. These attributes (sum-
marized in Table IV-5-2) can serve as a reference for comparison to measured
stream data. Measured data which are commensurate with those predicted by the
river continuum model indicate that the studied system is unperturbed, while
disagreement between actual and expected data indicates that modification of
the ecosystem has occurred (Karr and Dudley 1978).
EVALJATIQN OF BIOLOGICAL HEALTH USING FISH TROPHIC STRUCTURE
Karr (1981) developed a system for assessing biotic integrity using fish com-
munities, which is discussed in Chapter IV-2: Diversity Indices. Three em-
pirical trophic metrics are incorporated into Karr's index of biotic integrity
(IBI). They are:
(1) the proportion of individuals that are omnivores,
(2) the proportion of insectivorous individuals of the Cyprinidae family,
and
(3) the presence of top carnivore populations.
Karr (1981) observed that the proportion of omnivores in a community increases
as the quality of the aquatic environment declines. Nearly all major consumer
species are omnivorous to a degree (Darnell 1961), so populations are con-
sidered to be truly omnivorous only if they feed on plants and animals in
nearly equal amounts or indiscriminately (Kendeigh 1974). Recall that Karr and
Schlnsser used 25 percent of plant material ingested as the level for distin-
guisiing between omnivores and other trophic guilds. Presumably, changes in
the food base due to pollutional stress allow the euryphagic omnivores to be-
come dominant because their opportunistic foraging ecology makes them more suc-
cessful than more specific feeders. Omnivores are often virtually absent from
unmodified streams. Even in moderately - altered streams omnivorous species
usually constitute a minor portion of the community. For this reason, the bi-
ologist responsible for assessment must be familiar with the local fish fauna
and aquatic habitats in order to be able to interpret subtle disproportions in
trop'iic structure. In general, Karr (1981) has found samples with fewer than
20 percent of individuals as omnivores to be representative of good environ-
mental quality, while those with greater than 45 percent omnivores represent
badl./ degraded sites.
Karr (1981) reported that a strong inverse correlation exists between the abun-
dance of insectivorous cyprinids and omnivores. Thus, communities containing a
large proportion of insectivorous members of the minnow family (>45%) tends to
indicate relatively high environmental quality.
IV-5-5
-------
o
•o
co
o
a;
o
0
i/> i.
i. CP
CU N
Q. «3
re t-
w 01
o •—•
co
in
0)
C
re
Q.
U IT
C O
O -P
•P O
.* QJ
C r—
re r—
i— O
Q- O
ght and
temperature
regimes
1/1
T5 JD E
' re 4->
Q. — < (/I
a.
o
s_
•4->
o
S-
OJ
^->
(U
a> -r-
r- -O
*j re
+•> JC
a;
S-
>, 3 C
r- 4J O
*r™ re »»~
re s_ -(->
•o 01 re
CD
•o
re
3
4J
re
r— O.
-DE
re 01
4-= -W
co
iS
"o
CD
(/I
»
LO
a:
^~
CL.
o
0.
I
s
3
(J
a.
o
a;
V
-
es ^_
CVJ
i
LO
I
UJ
_J
CD
Primary
energy
source
a>
4->
re a)
r— S- .C
3
O 4->
.^ ^J £ ,_ ^->
4-j re O re c
i_ e j- •!- a»
re <+-. u E
at o -u c
•r- -— in O
CD cs: a; s-
in re o i- -r~
s- cno.. s- >
re i- o cu c
o o— •»-> a>
o
s_
re
-r- 0
CJ
,
re o> i—
•— 4J *J
3 -M m
O «3 O
-.EE
s_ u -
re T- ,—
Q-C s
re o
01 OIQ.
c v. u.
•r- O •
O)
•— C
-C2 O
re -i-
S- >, *J
o> i- u
"O (0 3
••- E T3
in -r- O
C !_ S_
O O. O-
CJ
FROM from
upstream
(C
d) o>
S- N
•P-i-
co en
a> E
+-> in re
re E a> 4.
r— t3 4)
re re s-
CO
m
PO
3T3 1 —
>r- OJ OJ «3
-O N 1- I
0) -t- -P ^p
s • — •
*
in - —
OJ S- CM
CTXD ,— i
S- > 1
re-r— i~^.
_J s- -—
¥
S-
h
g
subd
re
o
Q-CSJ
&'
re QJ
c
01 .C
E O)
re 3
a> i«i
-------
Faucsh et al. (unpublished manuscript) investigated the regional applicability
of the IBI. Results from the two least disturbed watersheds in the study --
the Embaras River, Illinois and the Red River, Kentucky — confirmed the fixed
scoring criteria proposed by Karr (1981) for omnivores and insectivorous cypri-
nids. At most of the undisturbed sites in each stream, omnivores constituted
20 percent or less of all individuals and at least 45 percent of individuals
were insectivorous cyprinids.
The presence of viable, vigorous populations of top carnivores is another in-
dicator of a relatively healthy, trophically diverse community used in Karr's
index. As described earlier, top carnivores constitute the peak of the eco-
logical pyramid, and, therefore, occupy the highest trophic level in that par-
ticular community. Degradation of environmental quality causes top carnivore
populations to decline and disappear. Theoretically, since top carnivore pop-
ulations are supported (directly or indirectly) by all of the other (lower)
trophic levels, they serve as a natural monitor of the overall health of the
community. Because of their position atop the food chain, terminal carnivores
are most vulnerable to detrimental effects of biomagnified toxicants. Also,
predation by top carnivores keeps the populations of forage and rough fish in
check, thereby functioning to maintain biotic integrity. As always, it is as-
sumed that the project biologist will use considerable personal knowledge of
local ichthyology and ecology in adjusting expectations of top carnivore spe-
cies to stream size. The top carnivore populations must be evaluated in rela-
tion to what would be there if the habitat were not modified. Defining the
baseline is a major problem in any study of pollutional stress. In determining
the baseline community, the biologist may rely on the faunas of similar, unal-
tered habitats in the area, literature information, and personal experience --
remembering the concepts of the river continuum model.
The results of research conducted throughout the midwest tend to support the
theoretical basis of the omnivore and top carnivore metric approaches to as-
sessing biotic integrity (Larimore and Smith 1963, Cross and Collins 1975,
Menzel and Fierstine 1976, Karr and Dudley 1978, Schlosser 1982a, Karr et al.
1983). Fausch et al. (unpublished manuscript) evaluated five watersheds in
Illinois, Michigan, Kentucky, Nebraska, and North and South Dakota using the
IBI, and found that scores accurately reflected watershed and stream condi-
tions.
However, experts in the field recognize that the omnivore - top carnivore anal-
ysis may not be applicable in every situation on a nationwide basis. Reser-
vations over use of this approach seem to be based on three variables.
(1) Type of pollutional stress - e.g., the trophic metrics proposed by Karr
(1981) were largely derived from agricultural watersheds in which sedi-
mentation and nutrient enrichment are the predominant forms of anthro-
pogenic stress; other pollution problems such as toxic waste discharge
could conceivably have a different impact on fish trophic structure.
IV-5-7
-------
(2) Type of aquatic habitat - e.g., headwater streams, large rivers, and
flowing swamps represent very different environments which are charac-
terized by a variety of trophic pathways and food sources.
(3) Type of ambient fish fauna - e.g., no or very tolerant top carnivores
might be present naturally, or no or very intolerant omnivores.
LIST OF OMNIVOkES AND TOP CARNIVORES
Examples of resident omnivore and top carnivore fish species are listed nation-
ally in Appendices B-l and B-2, respectively. These tables were compiled based
on information found in the literature (Morita, 1963; Carlander, 1969, 1977;
Pflieger, 1975; Moyle, 1976; Timbol and Maciolek, 1978; Smith, 1979; Morrow,
1980; Lee et al.,1980; Karr et al., 1983). The purpose of the lists is to
provide a framework for assessing omnivore and top carnivore populations.
However, because of the geographic variability in feeding habits, the gaps in
available foraging data, and the dynamic nature of range boundaries, some
members of the list may not occupy the specified trophic compartment in a
particular area, while other species that belong on the list may have been
overlooked. The list is intended to be used by knowledgeable biologists who
are capable of adding and deleting species where necessary to produce a list
which is appropriate for the particular area of study.
IV-5-8
-------
CHAPTER IV-6
REFERENCE SITES
Introduction
The goal of this section is to suggest an objective, ecological
approach that should aid States in determining the ecological potential of
priority aquatic ecosystems, evaluating and refining standards,
prioritizing ecosystems for improvements, and comprehensively evaluating
the ecological quality of aquatic ecosystems. The objectives of this
section are to demonstrate the need for regional reference sites and to
demonstrate how they can be determined. To do this the need for some type
of control or reference sites will be discussed and alternate types will be
outlined, the concept of ecological regions and methods for determining
them will be described, aspects that should be considered when selecting
reference sites will be listed, and the limitations of the regionalization
method will be discussed.
Although correlation between a disturbance and the resulting
functional or structural disorder can stimulate considerable insight, the
disorder that results from disturbing a water body can be demonstrated
scientifically only by comparing it with control or reference sites. To
scientifically test for functional or structural disorder, data must be
collected when the disturbances are present and when the disturbances are
absent but everything else is the same. Disorders that are unique to the
disturbed areas must be related to the disturbances but separated from
natural variability. This requires carefully selected reference sites, but
it is difficult or impossible to find pristine control or reference sites
in most of the conterminous United States. Also, it is unlikely that
pristine reference sites would be appropriate for most disturbed sites
because they would differ in ways besides the distrubance, as will be
discussed later.
The most commonly used reference sites are upstream and downstream of
the recovery zone of a point source. However, these sites provide little
value where diffuse pollution is a problem, where channel modifications are
extensive, where point sources occur all along the stream, where the
stream's morphology or flow changes considerably among sites, or where
various combinations of these disturbances occur. Hughes et al. (1983)
suggest a different approach, which reduces the problems of upstream-
downstream reference sites. Their approach is based on first determining
large, relatively-homogeneous, ecological regions (areas with similar
land-surface form, climate, vegetation, etc.) followed by selection of a
series of reference sites within each region. These sites could possibly
serve as references for a number of polluted sites on a number of streams
thereby economizing on and simplifying concurrent or future studies. A
modification of Hughes et al.'s approach has been tested on two polluted
streams in Montana (Hughes MS) and the approach is being rigorously tested
on 110 sites in Ohio (Omernik and Hughes 1983).
IV-6-1
-------
The logical basis for Omernik and Hughes' aoproach was developed from
alley (1976), Green (1979), Hall et al. (1978), and arren (1979). Their
logic fits well with the proposed water quality standards regulation
(Federal Register 1982) that suggests grouping of streams wherever
possible. Bailey stressed that heterogeneous lands, such as those managed
by the U.S. Forest Service, must be hierarchically classified by their
capabilities. He added that classification should be objective,
synthesized from present mapped knowledge, and based on the spatial
relationships of several environmental characteristics rather than on one
characteristic or on the similarity of the characteristics alone.
One of Green's ten principles for optimizing environmental assessments
is that wherever there are broad environmental patterns, the area should be
broken into relatively homogeneous subareas. Clearly, this principle
applies to most States. Hall et al. found that studies that incorporate
several variously-impacted sites were more useful than separate intensive
studies of one or two sites and more practical than long-term pre- and
post- impact studies.
Warren proposed that a watershed/stream classification should
integrate climate, topography, substrate, biota, and culture at all levels,
as opposed to considering them separately. He also stated that the
integration and classification should be hierarchical and be determined
from the potentials of the lands and waters of interest, rather than from
their present conditions. Streams within Warren's proposed classification
would have increasingly similar ecological potentials as one moved down
through the hierarchy to ever smaller watersheds or ecological regions.
The Concept of Ecological Regions
The ecological potential of a reference or disturbed site is
considered to be the range of ecological conditions present in a number of
typical, but relatively-undisturbed sites within an ecological region.
Such relatively-undisturbed sites, can be found even in the channelized
streams of the Midwest Corn Belt (Marsh and Luey 1982). One should not
suppose that such sites represent pristine or undisturbed controls, only
that they are the best that exist given the prevalent land use patterns in
an ecological region. Because of the major economic and political strains
required, we do not believe that resource managers or even knowledgeable
and concerned citizens will change those general land use patterns much.
But such persons will need to know the best conditions they can expect in a
water body in order to decide whether the economic and noneconomic benefits
of a particular water body standard are worth their economic and
noneconomic costs. To make such determinations rationally, the reference
sites must also be typical of a region. That is, their watersheds must
wholly reflect the predominant climate, land-surface form, soil, potential
natural vegetation, land use, and other environmental characteristics
defining that region, and the site itself must contain no anomalous
feature. For example, a cobble-bottomed stream in an entirely forested,
highly dissected watershed would not be typical of the sand and
gravel-bottomed streams in the agricultural prairies of the Midwest, nor
could it be a useful predictor of such an agricultural stream's ecological
potential, even though such a watershed and stream might be found in such a
region.
IV-6-2
-------
Although all aquatic ecosystems differ to some degree, the basis of
ecological regions is that there also is considerable similarity among
aquatic ecosystem characteristics and that these similarities occur in
definable geographic patterns. Also, the variabilities in the present and
potential conditions of the chemical and physical environment and the biota
are believed to be less within an area than among different areas. For
example, streams in the Appalachian Mountains, are more similar to each
other than to those in the Corn Belt or those on the Atlantic Coastal
Plain. It is assumed that streams acquire their similarities from
similarities in their watersheds and that streams draining watersheds with
similar characteristics will be more similar to each other than to those
draining watersheds with dissimilar characteristics. Thus, an ecological
region is defined as a large area where the homogeneity in climate,
land-surface form, soil, vegetation, land use, and other environmental
characteristics is sufficient to produce relative homogeneity in stream
ecosystems.
The concept of an ecological region is an out-growth of the work of
vegetation ecologists, climatologists, physiographers, and soil
taxonomists, all of whom have sought to display national patterns by
mapping classes of individual environmental characteristics (USDI
Geological Survey 1970). James (1952) discusses the value of integrating
or regionalizing such environmental characteristics and Warren (1979)
provides an excellent rationale for classifying ecological regions, but
Bailey's ecoregion map (1976) cones the closest to actually doing so.
However, Bailey's map incorporates a hierarchical approach, concentrating
on an individual environmental characteristic at each level, and does not
yet incorporate land-surface form or land use. Hughes and Omernik (1981b)
agree with Warren that it is most useful to integrate these features at
every level in the hierarchy of ecological regions. Such an approach
facilitates the mapping of ecological regions at a national, state, or
county level with increasing resolution (but decreasing generality) at each
lower level.
Ecological regions should improve States' abilities to manage aquatic
ecosystems in at least four ways (Hughes and Omernik 1981b): (1) They
should provide ecologically-meaningful management units. Such units allow
objective and logical synthesis of existing data from ecologically-similar
aquatic ecosystems and, using that synthesis, extrapolation to other
unstudied ecosystems in the same ecological region. (2) They should
provide an objective, ecological basis to refine use classifications and to
evaluate the attainment of uses for aquatic ecosystems. This is because
they provide an ecological basis for determining typical and potential
states of aquatic ecosystems located in similar watersheds. (3) They
should provide an objective ecological basis to prioritize aquatic
ecosystems for improvements or for attainability analyses. Given knowledge
of the typical and potential conditions of aquatic ecosystems in the
separate ecological regions of a State, that State can rationally determine
what to expect from improvements and thereby know where it will get the
greatest ecological returns for its investments. (4) They should simplify
setting site-specific criteria on site-specific biota, as allowed by the
proposed water quality regulation. Rather than set separate criteria for a
large number of sites at enormous expense, a State could use criteria
obtained from a series of sites that typify potential conditions in each
ecological region of that state or neighboring states.
IV-6-3
-------
The process of selecting reference sites can be broken into two major
phases with most of the work done in an office. First, the ecological
regions, and most-typical area(s) of interest are determined. Second,
various sizes of candidate watersheds and reaches are evaluated for
typicalness and level of disturbance in order to select reference sites.
Determining Ecological Regions
There are several methods for determining ecological regions.
Trautman (1981) suggested that one factor, physiography, could be used to
determine patterns of stream types and fish assemblages in Ohio. Lotspeich
and Platts (1982) believed regions should be determined from two factors,
climate and geology. Bailey (1976) used three factors, climate, soil, and
potential natural vegetation, in his ecoregion map of the United States but
suggested adding land-surface form and lithology if smaller ecoregions are
mapped. Warren (1979) proposed that five factors, climate, topography,
substrate, biota and culture, should all be incorporated in watershed
classification. Hughes and Omernik (19Slb), Omernik et al. (1982), and
Omernik and Hughes (1983) overlaid maps of land-surface form, soil
suborders, land use, and potential natural vegetation in studies of the
Corn Belt and Ohio, but suggest using precipitation, temperature, and
lithology if major differences in these factors are suspected. Lotspeich
and Platts, Bailey, and Warren all emphasized the use of hierarchical
ecoregions, moving from broad national regions thousands of square
kilometers in size to small watersheds a few square kilometers in area. A
much different approach to determining ecological regions is the stream
habitat classification of Pflieger et al. (1981). They used cluster
analysis of fish collections from throughout Missouri to group localities
having similar fish faunas. Where States have computerized fish collection
data from a thousand or more sites, cluster analysis is a useful approach,
however only a handful of States have such data.
Because of the diversity of methods for determining ecological
regions, the limited testing of their applicability to aquatic ecosystems,
and the limited number of large computerized data files, States are
encouraged to select a method that allows the greatest potential for later
modification. The method of Hughes and Omernik requires no prior
collection data and appears to allow more modification than the others.
The greater number of characteristics used to determine regions increases
the opportunity that those regions will have a variety of uses by several
agencies and greater value in predicting impacts of managment actions.
Therefore, their method is outlined by the following steps:
1. Select, the area and aquatic characteristics of interest. In many cases
the area of interest will be a State, but wherever major environmental
characteristics or watersheds do not coincide with state borders, States
may find it useful and economical to work cooperatively and incorporate
portions of neighboring States. Aquatic characteristics of interest may
include fish and macro-invertebrate assemblages and various aspects of
the chemical and physical environment affecting those assemblages.
IV-fi-4
-------
Select broad environmental characteristics most likely to control the
aquatic characteristics of interest. Environmental characteristics to
consider are climate (especially mean annual precipitation and summer
an.d winter temperature extremes), land-surface form (types of plains,
hills, or mountains), surficial geology (types of soil parent material),
soils (whether wet or dry, hot or cold, shallow or deep, or low or high
in nutrients), potential natural vegetation (grassland, shrubland, or
forestland, and dominant species), major river basins (especially
important in unglaciated areas for limiting fish and mollusk
distribution), and land use (especially cropland, grazing land, forest,
or various mixes of these). National maps of most of these
characteristics are available in USDI-Geological Survey (1970), but,
often, larger-scale State maps can be obtained from State agencies or
university departments.
Examine maps of the selected environmental characteristics for classes
of characteristics that occur in regional patterns. When original maps
differ in scale or when finer resolution is required, a mechanical
enlarger/reducer, photocopy machine, photo-enlarger, or slide projector
can be used to produce maps of the desired scale. Select those classes
of characteristics that best represent tentative ecological regions.
For example, is the predominant class of land-surface form flat plains
or high hills; is the predominant potential natural vegetation oak
forest or ash forest? List the predominant class of all the
characteristics considered for each tentative ecological region.
Overlay the selected environmental characteristics mapped at the same
scale and outline the most-typical areas in each tentative ecological
region. The maps are examined in combination on a light table and lines
are drawn on a sheet of clear plastic or transparent paper (e.g.
albanene). Most-typical areas are those areas in each tentative
ecological region where all the predominant classes of environmental
characteristics in that region are present. These can be considered as
most-typical areas because they contain all the classes of
characteristics that will be used to determine that ecological region.
For example, if the predominant classes of land use, potential natural
vegetation, and land-surface form in an ecological region are cropland,
grassland, and plains, respectively, only the portion of that region
where cropland, grassland, and plains all occur together would be
most-typical. This overlay approach and some of the environmental
characteristics are similar to those used by McHarg (1969) in his
examination of the values of various land uses in the Potomac River
Basin.
Determine which environmental characteristics best distinguish between
regions. Where the major characteristics abruptly differ at the same
place (e.g. hilly forestlands vs. prairie croplands) this is easily
done, but where there are gradual transitions (e.g. from flat to smooth
and irregular plains with decreasing amounts of croplands and increasing
forestlands) it is more difficult and the boundries are less precise.
At. one boundary the distinguising characteristic may be land-surface
form and surficial geology, at another it may be land use or a river
IV-6-5
-------
basin divide. Thus, this boundary determination is a subjective - not a
mechanical or McHargian - process and it requires considerable judgment
and knowledge of the key environmental characteristics along the
tentative boundary. See Figure IV-7-1 for an example of a final
product. Fianlly, the regional lines are transferred to a base map of
the area of interest. On a State level, most of this work should be
done using map scales of 1:500,000 to 1:7,SOO,000. The base map should
then be circulated among knowledgeable professionals to evaluate the
significance of the ecological regions as drawn.
For cases where top-priority aquatic ecosystems are anomalies, or where
the State is interested in only a few sites, it may be more appropriate to
use a slightly different approach based only on the watershed
characteristics of the sites in question. For such cases, rather than
analyze the entire State, researchers can determine the climate,
land-surface form, soils, potential natural vegetation, land use, river
basin, etc. of the watershed upstream of the site of interest. The same
classes of characteristics elsewhere in the State or neighboring States can
then be determined from maps. The rest of the regionalization process is
the same as described above. The major difference in this approach is
that, because of the spatially-narrower objective, fewer ecological regions
will be determined, consequently, the product would have only local
applicat ion.
Determining Candidate References Reaches
The most-typical areas are considered the most-logical places to
locate reference reaches for several reasons: (1) Such areas should
contain a narrower range of land use or disturbance potentials compared to
the entire region or other regions. Hence, there should be a narrower
range of aquatic ecosystem conditions in these most-typical areas compared
to the entire region or other regions. (2) Such areas are more likely to
be free of major anomalies that might produce undisturbed sites that are
also atypical, such as an entirely forested, mountainous watershed in a
region typified by shruhlands and plains. (3) Such areas can potentially
represent the greatest number of streams in the ecological region because
they drain watersheds having all the predominant classes of environmental
characteristics that were used to identify the region. (4) Such areas best
represent the prevailing land use of the ecological region and the best
background conditions likely. For example, there is little likelihood of
transforming an area dominated by rangeland into forest!and, therefore, the
predominant land use in the watershed of a reference reach in such an area
should be grazing.
For the above reasons, if watersheds of reference or benchmark reaches
are to have the broadest possible applicability, they should fall entirely
within the most-typical areas of ecological regions. Thus, the size of the
most-typical area will determine the maximum size of such watersheds. The
smallest watersheds should include the smallest intermittent or permanent
streams and ponds that support spawning or rearing or valued populations.
Valued populations may include sport, commercial, rare, threatened,
endangered, forage, or intolerant species of any phylum.
IV-6-6
-------
Ref 1 ni ng jthe Number^ of Canji date Reference Reaches
Regardless of how candidates for reference watersheds are determined
there are several important aspects to consider when selecting reference
reaches:
1. Human Disturbances. Obviously, watersheds that contain dense human
populations, concentrations of mines or industry, several or important
point sources, or major and atypical problems with diffuse pollution
(e.g. acidification, soil erosion, overgrazing, mine wastes, landslides)
should be eliminated from consideration as reference watersheds.
Intentional stocking of sport fishes and incidental releases of aquarium
and bait organisms have extended the ranges of many aquatic species. If
these introductions are only local, knowledge of such populations should
be considered when selecting least-disturbed watersheds because
introduced stocks of species are one of the most detrimental changes
that humans initiate in aquatic ecosystems. Where human disturbances
are mapped this step should be done for the entire State.
2. Size: Because of the gradual change in many stream characteristics from
headwaters to rivers (Vannote et al. 1980), plus application of
MacArthur and Wilson's (1967) theory of island biogeography to lakes
(Rarbour and Brown 1974), it is important to consider the size of the
reference reaches when they are to be compared with a priority water
body. Although stream order (Strahler 1957) has often been used by
biologists to approximate stream size, Hughes and Omernik (1981a, 1983)
give several reasons why watershed area and mean annual discharge are
preferable measures. Limnologists typically use surface area and volume
to estimate lake size. Although regional differences make any
generalizations difficult, the stream order of priority and reference
reaches should not differ by more than one order in most cases and the
watershed areas usually should differ by less than one order of
magnitude.
3. Surface water hydrology. While determining size, the researcher should
also briefly examine the types of the watersheds, streams, or lakes for
anomalies. Large scale topographic maps will usually reveal whether the
streams are effluent or influent, i.e., whether the net movement of
water if from the streams to the ground water or the reverse. The same
maps reveal drainage lakes, lake type (kettle, solution, oxbow, etc.),
amount of ditching or canalization, and drainage pattern (dendritic,
trellis, aimless, etc.).
4. Refugia. Parks, monuments, wildlife refuges, natural areas, preserves,
state and federal forests, and woodlots are often indicated on large
scale topographic maps and locations of others can be obtained from
state agencies charged with their administration. Such refugia are
often excellent places to locate reference sites and reference
watersheds.
TV-6-7
-------
5. Groundwater hydrology. Reports from the State water resource agency and
the State office of the U.S. Geological Survey reveal whether lakes are
influent or effluent. The direction of water movement in lakes is
extremely important in determining their nutrient balance, causes of
eutrophication, and possible results of lake restoration efforts. For
example, in shallow effluent lakes with small watersheds the major
source of nutrients is the atmosphere and hence uncontrollable.
6. Runoff per unit area. This is extremely important in estimating stream
size. The summarized runoff data are published in U.S. Geological
Survey reports for each State. These data can be used to estimate
isolines of runoff per unit area or existing runoff maps produced by
State water resource agencies can be used. For a national example, see
USDI - Geological Survey (1970).
7. Water chemistry. These data can be used to estimate background or
typical conditions. Most are not summarized, but they can be located
using NAWDEX and are available from computerized data bases such as
WATSTORE and STORET and from State water reports of the U.S. Geological
Survey and State water resource agencies.
8. Geoclimatic history. The historical geomorphology and climate determine
the basin divides and historical connections among water bodies and
basins. The absence of such connections and the locations of basin
divides and major gradient changes determine centers of origin or
endemism. Regionally, continental glaciation, ocean subsidence, and
pluvial flooding, and locally, stream capture, canals, and headwater
flooding all provided passages across apparent barriers that allowed
range extension, and, in large part, determine the present ranges of
primary freshwater fish and mollusks. This information is usually
available from university geology departments and often from the state
geologist.
9. Known zoogeographic patterns. These are best revealed by maps in books
and articles on the biota of the state, e.g. Smith (1983), Trautman
(1981), or Pflieger (1975). Such patterns may also be predicted by
present river basins where the basin divides are substantial and the
river mouths distant.
After considering the broad watershed and regional aspects of the
candidate watersheds, the highly-degraded or unusual watersheds should be
easily rejected. Candidate reaches can then be selected and ranked or
clustered by expected level of disturbance. At this level of resolution,
the researcher should study air photo mosaics and large-scale (1:24,000-
1:250,000) maps of the candidate reaches. Stream gradient, distance from
other refugia, barriers (falls, dams) between reference reaches and other
refugia, distance from the major receiving water, number of mines, and
buildings, amount of channelization, and presence of established monitoring
or gaging sites should all be considered. The list of candidate reaches
should be distributed to other professionals to query them about their
knowledge of disturbance levels, previous or concurrent studies, fish
stocking schedules, fish catch per unit effort, spawning or hatching
pulses, valued species, etc.
IV-6-8
-------
Selecting Actual Reference Sites
All the preceding research can, and should, be done in an office. It
is then useful to view and photograph the reduced number of candidate
reaches from the air. A small wing-over airplane flying 300-1500 meters
above the ground is ideal for this or recent stereo pairs of air photos can
suffice. The candidate reach should be examined at several access points
to assess typical and least-disturbed conditions, i.e., the absence of farm
yards, feed lots, livestock grazing, irrigation diversions, row crops,
channelization, mines, housing developments, clearcuts, or other small
scale disturbances should be rejected, though the candidate reaches may be
moved upstream of them. The main reasons for this aerial view are to
determine what the candidate watersheds and reaches typically look like, to
characterize relatively undisturbed conditions, and to help select actual
reference sites. The photographs are also useful as visual aids in
briefings and public meetings. This phase is not essential if the chief
state ecologist has developed this knowledge of present conditions through
years of experience statewide.
Finally, the remaining candidate reaches can be assessed and ranked
for disturbance from the ground. Three to four candidate reference sites
in each reach should be examined for typical natural features, least-
disturbed channel and riparian characteristics, and ease of access. The
concept of typicalness of natural features is similar to that of
typicalness of watershed features; for example, riffle-pool morphology and
swift current would not be typical of coastal plain or swamp streams and
such anomolous sites should not be included as reference sites.
One of the best indicators of least-disturbed sites is extensive,
old, riparian forest (see Section II-6). Another is relatively-high
heterogeneity in channel width and depth (shallow riffles, deep pools,
runs, secondary channels, flooded backwaters, sand bars, etc.). Abundant
large woody debris (snags, root wads, log jams, brush piles), coarse bottom
substrate (gravel, cobble, boulders), overhanging vegetation, undercut
banks, and aquatic vascular macrophytes and additional substrate
heterogeneity and concealment for biota. Relatively high discharges;
clear, colorless, and odorless waters; visually-abundant diatom, insect,
and fish assemblages; and the presence of beavers and piscivorous birds
also indicate relatively-undisturbed sites.
In order to confidently ascertain whether a designated biotic use of
a priority aquatic ecosystem is attainable it is necessary to (1) clearly
define that use in objective, measurable, biotic conditions and (2) examine
those conditions in at least three least-disturbed reference sites. We
have described a process to locate and rank a number of least-disturbed
reference sites. However, there are several limitations to that approach.
To date this process has only been tested on streams with watersheds less
than 1600 km?. Major lakes and rivers can be examined in the same
manner, but a multistate or national analysis will be needed and greater
allowances for variability in the level of disturbance and the degree of
typicalness may be necessary because large ecosystems encompass more
variability, they are more likely to receive major point sources, and they
are rarer to begin with.
IV-6-9
-------
Where priority aquatic ecosystems are unique it will be more
difficult to find reference sites. For example, if the priority system is
a forested watershed with a high-gradient stream in Iowa, where such a
system is rare, it would be necessary to seek reference sites in
neighboring States. Where a stream passes through extremely dissimilar
ecological regions, reference streams should do likewise. For example, the
Yampa River of Northwestern Colorado passes from spruce-forested mountains
through sagebrush tablelands and should not be compared with a river that
flows through only one of those regions.
Stream reaches above barriers, such as the falls on the Cumberland
River or the relatively steep gradients of the Watauga River at the North
Carolina-Tennessee border, should not be compared with those below because
few purely aquatic species have passed those historical barriers. Streams
that had glacial or pluvial connections (such as the Susquehanna and James
Rivers) may have more species in common than neighboring rivers of either,
the neighboring rivers have similar environmental conditions. Gilbert
(1980) provides a clear discussion of these possible zoogeographic
anomalies using examples from the eastern United States. Decisions about
reference sites must also take such knowledge into consideration.
Finally, ecological regions and reference sites as described herein
are believed most useful for making comparisons between broad assemblage-
level patterns or patterns between widely-ranging and common species of
importance, not between the presence or absence of specific uncommon or
localized species viewed separately. That is, multivariate approaches such
as ordination and classification or biotic indices such as K.arr's (1981)
are most applicable and researchers should not expect to discriminate among
sites that vary only slightly.
Summa ry
The final product of this approach is a map like that of Figure
IV-7-1. Data from the reference sites in each ecological region can be
compared with those from disturbed sites in that region. For aquatic
ecosystems that cross boundaries between ecological regions, state
ecologists ought to examine data from the reference sites in those
respective regions. Comparisons should he limited to ecosystems of similar
size.
Rather than an ad hoc, best - biological judgment approach, a
regionalization approach as described provides a rational, objective means
to compare similarities and differences over large areas. The regions
provide ecologically-meaningful management units and they would help in the
organization and interpretation of State water quality and NPS reports.
Data from the reference sites provide an objective, ecological basis to
refine use classifications and, when compared with more disturbed sites, to
evaluate the attainment of uses. Knowledge of potential conditions in a
region provides an objective, ecological basis to predict effects of land
use changes and pollution controls, to prioritize aquatic ecosystems for
improvements, and to set site-specific criteria. Regular monitoring of the
reference sites and comparisons with historical information will provide a
useful assessment of temporal changes, not only in those aquatic
ecosystems, but in the ecological regions that they model.
IV-6-10
-------
I NORTHWEST FLAT PLAINS
H WESTERN ROLLING PLAINS
m NE and SW IRREGULAR
Iff DISSECTED SOUTHEAST
Most Typical Areas
Generally Typical Areas
• Study Watersheds
IV-6-11
-------
0 SECTION V: INTERPRETATION
-------
CHAPTER V
INTERPRETATION
INTRODUCTION
There are many use classifications which might be assigned to a water body,
such as navigation, recreation, water supply or the protection of aquatic
life. These need not be mutually exclusive. The water body survey as discussed
in this manual is concerned only with aquatic life uses and the protection of
aquatic life in a water body.
The water body survey may also be referred to as a use attainability analysis.
The objectives in conducting a water body survey are to identify:
1. What aquatic protection uses are currently being achieved in the water
body,
2. What the causes are of any impairment to attaining the designated aqua-
tic protection uses, and
3. What the aquatic protection uses are that could be attained, based on
the physical, chemical and biological characteristics of the water body.
The types of analyses that might be employed to address these three points are
summarized in Table V-l. Most of these are discussed in detail elsewhere in
this manual.
CURRENT AQUATIC PROTECTION USES
The actual aquatic protection use of a water body is defined by the resident
biota. The prevailing chemical and physical attributes will determine what
biota may be present, but little need be known of these attributes to describe
current uses. The raw findings of a biological survey may be subjected to
various measurments and assessments, as discussed in Chapters IV-2, IV-4, and
IV-5. After performing a biological inventory, omnivore-carnivore analysis,
and intolerant species analysis, and calculating a diversity index and other
indices of biological health, one should be able adequately to describe the
condition of the aquatic life in the water body.
It will be helpful to digress at this juncture briefly to discuss water body
use classification systems and their relationship to the water body survey.
Classification systems vary widely from state to state. Some consist of as few
as three broad categories, while others include a number of more sharply-
defined categories. Also, the use classes may be based on geography, salinity,
recreation, navigation, water supply (municipal, agricultural, or industrial),
or aquatic life. Often an aquatic protection use must be categorized as either
V-l
-------
TABLE V-l. SUMMARY OF TYPICAL WATER BODY EVALUATIONS (from EPA,1983, Water Quality Standards Handbook).
PHYSICAL EVALUATIONS
CHEMICAL EVALUATIONS
BIOLOGICAL EVALUATIONS
" Instream Characteristics
- size (mean width/depth)
- flow/velocity
- total volume
- reaeration rates
- gradlent/pools/rlff les
- temperature
- suspended solids
- sedimentation
- channel modifications
- channel stahllIty
" Substrate composition and
characteristics
0 Channel debris
0 Sludge deposits
0 Riparian characteristics
0 Downstream characteristics
* dissolved oxygen
0 toxicants
0 nutrients
- nitrogen
- phosphorus
" sediment oxygen demand
° salinity
" hardness
0 alkalinity
0 PH
0 dissolved solids
Biological Inventory (Existing Use
Analysis)
- fish
- macroinvertebrates
- microinvertebrates
- phytoplankton
- macrophytes
Biological Condition/Health Analysis
- Diversity Indices
- HSI Models
- Tissue Analyses
- Recovery Index
- Intolerant Species Analysis
- Omn1vore-Carnivore Analysis
Biological Potential Analysis
- Reference Reach Comparison
V-2
-------
a warmwater or coldwater fishery. Clearly, little information is required
to place a water body into one of these two categories. Far more
information may be gathered in a water body survey than is needed to
assign a classification, based on existing classes, but the additional
data may be necessary to evaluate management alternatives and refine use
classification systems for the protection of aquatic life in the water
body.
Since there may not be a spectrum of aquatic protection use categories
available against which to compare the findings of the biological survey;
and since the objective of the survey is to compare existing uses with
designated uses, and existing uses with potential uses, as seen in the
three points listed above, the investigators may need to develop their own
system of ranking the biological health of a water body (whether
qualitative or quantitative) in order to satisfy the intent of the water
body survey. Implicit to the water body survey is the development of
management strategies or alternatives which might result in enhancement of
the biological health of the water body. To do this it would be necessary
to distinguish the predicted results of one strategy from another, where
the strategies are defined in terms of aquatic life. The existing State
use classifications will probably not be helpful at this stage, for one
may very well be seeking to define use levels within an existing use
category, rather than describing a shift from one use classification to
another. To conclude, it may be helpful to develop an internal use
classification system to serve as a yardstick during the course of the
water body survey, which may later be referenced to the legally
constituted use categories of the state. Sample scales of aquatic life
classes are presented in Table V-? and v_3.
CAUSFS OF IMPAIRMENT OF AQUATIC PROTECTION USES
If the biological evaluations indicate that the biological health of the
system is impaired relative to a "healthy" or least disturbed control
station or reference aquatic ecosystem (e.g., as determined by reference
reach comparisons), then the physical and chemical evaluations can be used
to pinpoint the causes of that impairment. Figure V-l shows some of the
physical and chemical parameters that may be affected by various causes of
change in a water body. The analysis of such parameters will help clarify
the magnitude of impairments to attaining other uses, and will also be
important to the third step in which potential uses are examined.
ATTAINABLE AQUATIC PROTECTION USES
The third element to be considered is the assessment of potential uses of
the water body. This assessment would be based on the findings of the
physical, chemical and biological information which has been gathered, but
additional study may also be necessary. Procedures which might be
particularly helpful in this stage include the Habitat Suitability Index
Models of the Fish and Wildlife Service, that may indicate which fish
species could potentially occupy a given habitat; and the Recovery Index
of Cairns et al. (1977) which estimates the ability of a system to
recover following stress. A reference reach comparison will be
particularly important. In addition to establishing a comparative
V-3
-------
TABLE V-2. BIOLOGICAL HEALTH CLASSES WHICH COULD BE USED
IN WATER BODY ASSESSMENT (Modified from Karr, 1981)
Class Attributes
Excellent Comparable to the best situations unaltered by man; all re-
gionally expected species for the habitat and stream size,
including the most intolerant forms, are present with full
array of age and sex clases; balanced trophic structure.
Good Fish and macroinvertebrate species richness somewhat less
than the best expected situation, especially due to loss of
most intolerant forms; some species with less than optimal
abundances or size distribution (fish); trophic structure
shows some signs of stress.
Fair Fewer intolerant forms of fish and macroinvertebrates are
present. Trophic structure of the fish community is more
skewed toward an increasing frequency of omnivores; older
age classes of top carnivores may be rare.
Poor Fish community is dominated by omnivores; pollution-toler-
ant forms and habitat generalists; few top carnivores;
growth rates and condition factors commonly depressed; hy-
brids and diseased fish may be present. Tolerant macroinver-
tebrates are often abundant.
Very Poor Few fish present, mostly introduced or very tolerant forms;
hybrids common; disease, parasites, fin damage, and other
anomalies regular. Only tolerant forms of macroinverte-
brates are present.
Extremely Poor
No fish,
life.
very tolerant macroinvertebrates, or no aquatic
V-4
-------
Table V-3; Aquatic Life Survey Rating System (EPA, 1983 Draft)
A reach that is rated a five has;
-A fish community that is well balanced among the different levels
of the food chain.
-An age structure for the most species that is stable, neither
progressive (leading to an increase in population) or regressive
(leading to a decrease in population),
-A sensitive sport fish species or species of special concern always
present.
-Habitat which will support all fish species at every stage of their
life cycle.
-Individuals that are reaching their potential for growth.
-Fewer individuals of each species.
-All available niches filled.
A reach that is rated a four has:
-Many of the above characteristics but some of them are not
exhibited to the full potential. For example, the reach has a well
balanced fish community; the age structure is good, sensitive
species are present; but the fish are not up to their full growth
potential and may be present in higher numbers; an aspect of the
habitat is less than perfect (i.e. occasional high temperatures
that do not have an acute effect on the fish); and not all food
organisms are available or they are available in fewer numbers.
A reach that is a three has:
-A community is not well balanced, one or two tropic levels
dominate.
-The age structure for many species is not stable, exhibiting
regressive or progressive charisteristics.
-Total number of fish is high, but individuals are small.
-A sensitive species may be present, but is not flourishing.
-OtT~er less sensitive species make up the majority of the biomass.
-Anadromous sport fish infrequently use these water as a migration
route.
A reach that is rated a two h_as;
-Few sensitive sport fish are present, nonsport fish species are
more common than sport fish species.
-Snecies are more common than abundant.
-Age structures may be very unstable for any species.
-The composition of the fish population and dominate species is very
changeble.
-Anedromous fish rarely use these waters as a migration route.
-A small percent of the reach provides sport fish habitat.
A reach that is a one has:
-The ability to support only nonsport fish. A occasional snort fish
may be found as a transient.
A reach that is rated a zero has:
-No ability to support a fish of any sort, an occasional fish may be
found as a transient.
V-5
-------
SOURCE OF MODIFICATION
ry
LLJ
f—
LU
s:
«c
cc.
i
i_
4-1 13
C E
OJ —
Sk
(Q •^-«
OJ -—
t. ,
r— O*. i-
i- ^
V> O V1
I
I
T
1
I
I
I
I
I
0
I
I
I
D
I
• — -c
•c c
U- C 13
•4- 13
0 •— 0.
CO. —
3 O 3
OC S- Co.
u o
P— t*- -^- --^
re W *4— 4^ (/)
i. O O 13 <1J
3 C 1-4 •—
•M
33 O) tn
U *J C C 31-
•^ VI 13 C "C OJ
W 13 ^D 13 CO.
cr- o. u x: — is
**. — rs <_) • — Q-
C
I
I I
I I
I
I
I I
I I I
I I
D
I 1 I
I I
I I
I
I I
I
I
r
I I
I I D
000
I I
C
I
I I I I
I I
I
C
D
4-)
X
a;
i —
I
I
I
I
I
I
I
I
I
0
I
I
I
•c
c
13
01
= g1
5 =
•^* "3
C i—
•i~ O.
Lu 0
1-
i — 4->
13 U
*j a>
01 . —
z: LU
C
I
I
I
I
I
I
I
I
T
i
I
0
D
D
0
I
&
3
V
O
u
4^
&>
a-
I
I
I
I
I
I
0
I
I
I
I
I
0
D
D
C-0
T3
C
'C
C
O
1.
1—1
0
I
I
I
*
I
I
I
I
I
I
I
I
I
0
D
D
0
^.
C
"C
c
T
4~>
C
^
a_
c
i
i
i
i
i
i
i
i
i
i
i
D
0
0
D
I
I
Meat Produc
Production
rushing
4- O
•O 0;
C f4 Ol
13 -r- £
>-, — — 1
s- *J _.
-f- i_ -C
'_)
W
I/)
fl
a.
C
I
I
I
I
I
I
1
I
I
I
I
I
I
r
I
0
0
0
I
T
Figure V-l. Potential Effects of Some Sources of Alteration on Stream Parameters;
0 = decrease, I - increase, C = change.
V-6
-------
baseline community, defining a reference reach can also provide insight to the
aquatic life that could potentially occur if the sources of impairment were
mitigated.
The analysis of all information that has been assembled may lead to the defini-
tion of alternative strategies for the management of the water body at hand.
Each such strategy corresponds to a unique level of protection of aquatic
life, or aquatic life protection use. If it is determined that an array of
uses are attainable, further analysis which is beyond the scope of the water
body survey would be required to select a management program for the water
body.
A number of factors which contribute to the health of the aquatic life will
have been evaluated during the course of the water body survey. These may be
divided into two groups: those which can be controlled or manipulated, and
those which cannot. The factors which cannot be regulated may be attributable
to natural phenemona or may be attributable to irrevocable anthropogenic
(cultural) activities. The potential for enhancing the aquatic life of a
water body essentially lies in those factors over which some control may be
exerted.
Whether or not a factor can be controlled may itself be a subject of contro-
versy for there may be a number of economic judgments or institutional consid-
erations which are implicit to a definition of control. For example, there are
many cases in the West where a wastewater discharge may be the only flow to
what would otherwise be an intermittent stream. If water rights have been es-
tablished for that discharge then the discharge cannot be diverted elsewhere,
applied to the land for example, in order to reduce the pollutant load to the
stream. If a stream does not support an anadromous fishery because of dams and
diversions which have been built for water supply and recreational purposes,
it is unlikely that a concensus could be reached to restore the fishery by re-
moving the physical barriers - the dams - which impede the migration of fish.
However, it may be practical to build fish ladders and by-passes to allow
upstream and downstream migration. In a practical sense these dams represent
anthropogenic activity which cannot be reversed. A third example might be a
situation in which dredging to remove toxic sediments in a river may pose a
much greater threat to aquatic life than to do nothing. In doing nothing the
toxics may remain in the sediment in a biologically-unavailable form, whereas
dredging might resuspend the toxic fraction, making it biologically available
and also facilitating wider distribution in the water body.
The points touched upon above are presented to suggest some of the phenomena
which may be of importance in a water body survey, and to suggest the need to
recognize whether or not they may realistically be manipulated. Those which
cannot be manipulated essentially define the limits of the highest potential
use that might be realized in the water body. Those that can be manipulated
define the levels of improvement that are attainable, ranging from the current
aquatic life uses to those that are possible within the limitations imposed by
factors that cannot be manipulated.
V-7
-------
SECTION VI: REFERENCES
-------
CHAPTER VI
REFERENCES
CHAPTER II-l: FLOW ASSESSMENTS
Bovee, K., 1982. A Guide to Stream Habitat Analysis Using the Instream Flow
Incremental Methodology, FWS/OBS-82/26. U.S. Fish and Wildlife Service, Fort
Collins, CO.
Hilgert, P., 1982. Evaluation of Instream Flow Methodologies for Fisheries in
Nebraska. Nebraska Game & Park Commission Technical Bulletin No. 10, Lincoln,
NB.
Tennant, D.L., 1976. Instream Flow Regimens for Fish, Wildlife, Recreation and
Related Environmental Resources, pp. 359-373. In J.F. Osborn, and C.H. Allman,
eds. Proceedings of the Symposium and Specialty Conference in Instream Flow
Needs. Vol. II, American Fisheries Society, Bethesda, MD.
CHAPTER II-2: SUSPENDED SOLIDS AND SEDIMENTATION
Atchinson, G.J., and B.W. Menzel, 1979. Sensitivity of Warmwater Fish
Populations to Suspended Solids and Sediments. In Muncey, R.J. et al. "Effects
of Suspended Solids and Sediment on Reproduction and Early Life of Warmwater
Fishes." U.S. EPA, Corvallis, OR, EPA/600/3-79-049.
Benson, N.G., and B.C. Cowell, 1967. The Environmental and Plankton Diversity
in Missouri River Reservoirs, pp. 358-373. In Reservoir Fishery Resources
Symposium. Reservoir Comm., Southern Div., Am. Fish. Soc., Bethesda, MD.
Butler, J.L., 1963. Temperature Relations in Shallow Turbid Ponds. Proc. Okla.
Acad. Sci. 43:90.
Cairns, J. Jr., 1968. Suspended Solids Standards for the Protection of Aquatic
Organisms. Eng. Bull. Purdue University 129:16.
Chew, R.L., 1969. Investigation of Early Life History of Largemouth Bass in
Florida. Florida Game and Fish Comm. Proj. Rept. F-024-R-02. Tallahassee, FL.
Ellis, M.M., 1969. Erosion Salt as a Factor in Aquatic Environments. Ecology
17:29.
European Inland Fisheries Advisory Committee, 1964. Water Quality Critria for
European Freshwater Fish: Report on Finely Divided Solids and Inland
Fisheries. EIFAC Tech. Paper(l) 21 pp.
Iwamoto, R.N., E.O. Salo, M.A. Madeq, R.L. Comas and R. Rulifson, 1978.
Sediment and Water Quality: A Review of the Literature Including a Suggested
Approach for Water Quality Criteria With Summary of Workshop and Conclusions.
EPA 910/9-78-048.
-------
Swingle, H.S., 1956. Appraisal of Methods of Fish Population Study Part IV:
Determination of Balance in Farm Fish Ponds. Trans. N. Am. Wild. Conf. 21:298.
Trautman, M.G., 1957. The Fishes of Ohio. Ohio State Univ. Press, Columbus.
683 pp.
U.S. EPA. 1976. Quality Criteria for Water. U.S. EPA, Washington, D.C. U.S.
Government Printing Office, 055-001-01099.
CHAPTER II-3: POOLS, RIFFLES AND SUBSTRATE COMPOSITION
Edwards, E.A., et al., 1982. Habitat Suitability Index Models: Black Crappie.
U.S. Fish and Wildlife Service, Ft. Collins, CO. FWS/OBS-82/10.6.
Edwards, E.A., et al., 1982. Habitat Suitability Index Models: White Crappie.
U.S. Fish and Wildlife Service, Ft. Collins, CO. FWS/OBS-82/10.7.
Hickman, T. and R.F. Raleigh, 1982. Habitat Suitability Index Models:
Cutthroat Trout. U.S. Fish and Wildlife Service, Ft. Collins, CO.
FWS/OBS-82/10.5.
Hynes, H.B.N., 1970. The Ecology of Running Waters. University of Toronto
Press, Toronto.
Lagler, Karl F., et al., 1977. Ichthyology. John Wiley & Sons, NY. 506 pp.
La Gorce, J. (editor), 1939. The Book of Fishes. National Geographic Society,
Washington, D.C. 367 pp.
McMahon, T.E., 1982. Habitat Suitability Index Models: Creek Chub. U.S. Fish
and Wildlife Service, Ft. Collins, CO. FWS/OBS-82/10.4.
McMahon, T.E. and J.W. Terrell, 1982. Habitat Suitability Index Models:
Channel Catfish. U.S. Fish and Wildlife Service, Ft. Collins, CO.
FWS/OBS-82/10.2.
Migdalski, Edward C. and G.S. Fichter, 1976. The Fresh and Salt Water Fishes
of the World. Alfred A. Knopf, NY. 316 pp.
Odum, E.P., 1971. Fundamentals of Ecology. W.B. Saunders Co. 574 pp.
Stalnaker, C.B. and O.L. Arnette (editor), 1976, Methodologies for the
Determination of Stream Resource Flow Requirements: An Assessment. U.S.
Fish and Wildlife Service, FWS/OBS-76/03.
Stuber, Robert J., et al., 1982. Habitat Suitability Index Models: Bluegill.
U.S. Fish and Wildlife Service, Ft. Collins, CO. FWS/ OBS-82/10.8.
Whitton, B.A., (editor), 1975. River Ecology. University of California Press.
724 pp.
VI-2
-------
CHAPTER II-4: CHANNEL CHARACTERISTICS AND EFFECTS OF CHANNELIZATION
Arner, D.H., et al. 1976. Effects of Channelization on the Luxapalila River on
Fish, Aquatic Invertebrates, Water Quality, and Furbearers. U.S. Fish and
Wildlife Service, Washington, D.C. FWS/DBS-76/08.
Barclay, J.S., 1980. Impact of Stream Alterations on Riparian Communities in
Southcentral Oklahoma. U.S. Fish and Wildlife Service, Albuquerque, NM.
FWS/OBS-80/17.
Brown, S., et al., 1979. Structure and Function of Riparian Wetlands. In
Strategies for Protection and Management of Floodplain Wetlands and Other
Riparian Ecosystems, Johnson, R.R., and McCormick, J.F. (editors), U.S. Dept.
of Agriculture, Washington, D.C., Tech. Rept. WO-12, pp. 17-32.
Bulkley, R.V., 1975. A Study of the Effects of Stream Channelization and Bank
Stabilization on Warm Water Sport Fish in Iowa: Subproject No. 1. Inventory of
Major Stream Alterations in Iowa. U.S. Fish and Wildlife Service, Washington,
D.C. FWS/OBS-76/11.
Bulkley, R.V., et al. 1976. Warmwater Stream Alteration in Iowa: Extent,
Effects on Habitat, Fish, and Fish Food, and Evaluation of Stream Improvement
Structures (Summary Report). U.S. Fish and Wildlife Service, Washington, D.C.,
FWS/OBS-76/16.
Cairns, J., Jr., et al., 1976. The Recovery of Damaged Streams. Assoc. SE
Biol. Bull., 13:79.
Chow, V.T., 1959. Open Channel Hydraulics. McGraw-Hill Book Co., NY. 680 pp.
Chutter, F.M., 1969. The Effects of Silt and Sand on the Invertebrate Fauna of
Streams and Rivers. Hydrobiologia, 34:57.
Cummins, K.W., 1973. Trophic Relations of Aquatic Insects. Ann. Rev. Entomol.,
18:183.
Cummins, K.W., 1974. Structure and Function of Stream Ecosystems. Bioscience,
24:631.
Cummins, K.W., 1975. Ecology of Running Waters: Theory and Practice. In Proc.
Sandusky River Basin Symposium, in Baker, D.B., et al., (editors) Heidelburg
College, Tiffin, OH.
Cummins, K.W., and G.H. Lauff, 1969. The Influence of Substrate Particle Size
on the Microdistribution of Stream Macrobenthos. Hydrobiologia, 34:145.
Etnier, D.A., 1972. Effect of Annual Rechanneling on Stream Population. Trans.
Amer. Fish. Soc., 101:372.
Frederickson, L.H., 1979. Floral and Faunal Changes in Lowland Hardwood
Forests in Missouri Resulting from Channelization, Drainage, and Impoundment.
U.S. Fish and Wildlife Service, Washington, D.C. FWS/OBS-78/91.
VI-3
-------
Gammon, J.R., 1979. The Effects of Inorganic Sediment on Stream Biota. Water
Poll. Con. Res. Series, 108050 DWC 12/70, U.S. EPA, Washington, D.C.
Gorman, OJ., and Karr, J.R., 1978. Habitat Structure and Stream Fish
Communities. Ecology, 59:507.
Griswold, B.L., et al., 1978. Some Effects of Stream Channelization on Fish
Populations, Macroinvertebrates, and Fishing in Ohio and Indiana. U.S. Fish
and Wildlife Service, Columbia, MO, FWS/OBS-77/46.
Huggins, D.G., and R.E. Moss, 1975. Fish Population Structure in Altered and
Unaltered Areas of a Small Kansas USA Stream. Trans. Kansas Acad. Sci., 77:18.
Huish, M.T., and G.B. Pardue, 1978. Ecological Studies of One Channelized and
Two Unchannelized Swamp Streams in North Carolina. U.S. Fish and Wildlife
Service, Washington, D.C. FWS/OBS-78/85.
Hynes, H.B.N., 1970. The Ecology of Running Waters. Univ. of Toronto Press,
Toronto, 555 pp.
Karr, J.R., and I.J. Schlosser, 1977. Impact of Nearstream Vegetation and
Stream Morphology in Water Quality and Stream Biota. U.S. EPA, Athens, GA,
Ecol. Res. Series, EPA-600/3-77-097.
King, D.L., and R.C. Ball, 1967. Comparative Energetics of a Polluted Stream.
Limnol. Oceanog., 12:27.
King, L.R., 1973. Comparison of the Distribution of Minnows and Darters
Collected in 1947 and 1972 in Boone County, Iowa. Proc. Iowa Acad. Sci., 80:
133.
King, L.R., and K.D. Carlander, 1976. A Study of the Effects of Stream
Channelization and Bank Stabilization on Warmwater Sport Fish in Iowa:
Subproject No. 3. Some Effects of Short-Reach Channelization on Fishes and
Fish Food Organisms in Central Iowa Warmwater Streams. U.S. Fish and Wildlife
Service, Washington, D.C. FWS/OBS-76/13.
Lavandier, R., and Caplancef, J., 1975. Effects of Variations in Dissolved
Oxygen on the Benthic Invertebrates of a Stream in the Pyreenees. Ann. Limnol.
11.
Leopold, L.B., et al., 1964. Fluvial Processes in Geomorphology. W.H. Freeman
and Co., San Francisco, CA.
Leopold, L.B., and W.B. Langbein, 1966. River Meanders. Scientific American
214:60.
Lund, J., 1976. Evaluation of Stream Channelization and Mitigation of the
Fishery Resources of the St. Regis River, Montana. U.S. Fish and Wildlife
Service, Washington, D.C. FWS/OBS-76-07.
Maki, T.E., et al., 1980. Effects of Stream Channelization on Bottomland and
Swamp Forest Ecosystems. Univ. of North Carolina, Chapel Hill, NC,
UNC-WRRI-80-147.
VI-4
-------
Marzolf, G.R., 1978. The Potential Effects of Clearing and Snagging on Stream
Ecosystems. U.S. Fish and Wildlife Service, Washington, D.C. FWS/OBS-78-14.
Meehan, W.R., 1971. Effects of Gravel Cleaning on Bottom Organisms in the
Southern Alaska Streams. Prog. Fish-Cult., 33:107.
Minshall, G.W., and P.V. Winger, 1968. The Effect of Reduction in Stream Flow
on Invertebrate Drift. Ecology, 49:580.
Minshall, J.W. and J.N. Minshall, 1977. Microdistribution of Benthic
Invertebrates in a Rocky Mountain Stream. Hydrobiologia, 53:231.
Montalbano, F., et al., 1979. The Kissimmee River Channelization: A
Preliminary Evaluation of Fish and Wildlife Mitigation Measures. In Proc. of
the Mitigation Symp., Colorado State Univ., Ft. Collins, CO, pp. 508-515.
Morris, L.A., et al., 1968. Effects of Main Stream Impoundments and
Channelization Upon the Limnology of the Missouri River, Nebraska. Trans.
Amer. Fish. Soc., 97:380.
Nebeker, A.V., 1971. Effect of Temperature at Different Altitudes on the
Emergence of Aquatic Insects from a Single Stream. Jour. Kansas. Entomol.
Soc., 44:26.
O'Rear, C.W., Jr., 1975. The Effects of Stream Channelization on the
Distribution of Nutrients and Metals. East Carolina Univ., Greenville, NC,
UNC-WRRI-75-108.
Parrish, J.D., et al., 1978. Stream Channel Modification in Hawaii. Part D:
Summary Report. U.S. Fish and Wildlife Service, Columbia, MO FWS/OBS-78/19.
Pfleiger, W.L., 1975. The Fishes of Missouri. Missouri Dept. Conserv.,
Jefferson City, MO.
Possardt, E.E., et al., 1976. Channelization Assessment, White River, Vermont:
Remote Sensing, Benthos, and Wildlife. U.S. Fish and Wildlife Service,
Washington, D.C. FWS/OBS-76/07.
Schmal, R.N., and D.F. Sanders, 1978. Effects of Stream Channelization on
Aquatic Macroinvertebrates, Buena Vista Marsh, Portage County, WI. U.S. Fish
and Wildlife Service, Washington, D.C. FWS/OBS-78/92.
Simpson, P.W., et al., 1982. Manual of Stream Channelization Impacts on Fish
and Wildlife. U.S. Fish and Wildlife Service, Kearneysville, WV FWS/OBS-82/24.
Swenson, W.A., et al., 1976. Effects of Red Clay Turbidity on the Aquatic
Environment. In Best Management Practices for Non-Point Source Pollution
Control Seminar, U.S. EPA, Chicago, IL, EPA 905/9-76-005.
Tebo, L.B., 1955. Effects of Siltation, Resulting from Improper Logging, on
the Bottom Fauna of a Small Trout Stream in the Southern Appalachians. Prog.
Fish-Cult. 17:64.
VI-5
-------
Vannote, et al., 1980. The River Continuum Concept. Can. Jour. Fish. Aquat.
Sci., 37:130.
Wallen, E.I., 1951. The Direct Effect of Turbidity on Fishes. Oklahoma A&M,
Stillwater, OK, Biol. Series No. 2, 48:1.
Walton, O.E., Jr., 1977. The Effects of Density, Sediment Size, and Velocity
on Drift of Acroneuria abnormis (Plecoptera). OIKOS, 28:291.
Wharton, C.H., and M.M. Brinson, 1977. Characteristics of Southeastern River
Systems. In Stategies for Protection and Management of Floodplain Wetlands and
Other Riparian Ecosystems, Johnson, R.R. and J.F. McCormick (editors),
U.S.D.A., Washington, D.C., Tech. Report WO-12, pp. 32-40.
Whitaker, G.A., et al., 1979. Channel Modification and Macroinvertebrate
Diversity in Small Streams. Wat Res. Bull., 15:874.
Williams, D.C., and J.H. Muncie, 1978. Substrate Size Selection by Stream
Invertebrates and the Influence of Sand, Limnol. Oceanog. 73:1030.
Winger, P.V., et al., 1976. Evaluation Study of Channelization and Mitigation
Structures in Crow Creek, Franklin County, Tennessee and Jackson County,
Alabama. U.S. Soil Conservation Service, Nashville, TN.
Wolf, J., et al., 1972. Comparison of Benthic Organisms in Semi-Natural and
Channelized Portions of the Missouri River. Proc. S.D. Acad. Sci., 51:160.
Yang, C.T., 1972. Unit Stream Power and Sediment Transport. A.S.C.E., Jour.
Hydraulics Div., 98:1805.
Zimmer, D.W., 1977. Observations of Invertebrate Drift in the Skunk River,
Iowa. Proc. Iowa Acad. Sci., 82:175.
Zimmer, D.W., and R.W. Bachman, 1976. A Study of the Effects of Stream
Channelization and Bank Stabilization on Warmwater Sport Fish in Iowa:
Subproject No. 4. The Effects of Long Reach Channelization on Habitat and
Invertebrate Drift in Some Iowa Streams. U.S. Fish and Wildlife Service,
Washington, D.C. FWS/OBS-76/14.
Zimmer, D.W., and R.W. Bachman, 1978. Channelization and Invertebrate Drift in
Some Iowa Streams. Water Res. Bull. 14:868.
CHAPTER II-5: TEMPERATURE
Brungs, W.A. and Jones, B.R., 1977. Temperature Criteria for Freshwater Fish:
Protocol and Procedures, U.S. EPA, Duluth, EPA-600/ 3-77-061.
Butler, J.N., 1964. Ionic Equilibrium, A Mathematical Approach, Addison-
Wesley, Reading, MA.
Carlander, K.D., Handbook of Freshwater Fishery Biology, Vols. I (1969) and II
(1977). Iowa State University Press, Ames, Iowa.
VI-6
-------
Cherry, D. and Cairns, C., 1982. Biological Monitoring, Part V - Preference
and Avoidance Studies, Water Research, 16:263.
Hokanson, K., 1977. Temperature Requirements of Some Percids and Adaptations
to the Seasonal Temperature Cycle, J. Fish. Res. Board Can., 34:1524-1550.
Karr, J.R. and Schlosser, 1978. I.J., Water Resources and the Land Water Inter-
face, Science 201: 229-234.
Klein, 1., 1962. River Pollution, II. Causes and Effects, Butterworths, London.
Machenthun, K.M., 1969. The Practice of Water Pollution Biology, U.S. DOI,
Federal Water Pollution Control Agency, U.S.G.P.O., Washington, DC.
Metcalf and Eddy, Inc., 1972. Wastewater Engineering, McGraw-Hill.
Morrow, J.E., 1980. The Freshwater Fishes of Alaska, Alaska Northwest
Publishing Company, Anchorage.
Scott, W., and Grossman, E., 1973. Freshwater Fishes of Canada, Fish. Res.
Board Can., Bulletin 184.
Stumm, W. and Morgan, 1970. J. Aquatic Chemistry, Wiley-Interscience, New York.
Warren, C.E., 1971. Biology and Water Pollution Control, W.B. Saunders
Company, Philadelphia.
CHAPTER II-6: RIPARIAN EVALUATIONS
Behnke, A.C., et al., 1979. Biological Basis for Assessing Impacts of Channel
Modification: Invertebrate Production, Drift and Fish Feeding in Southeastern
Blackwater River. Environmental Resources Center, Rep. 06-79. Georgia Inst.
Techn., Atlanta.
Behnke, R.J., 1979. Values and Protection of Riparian Ecosystems. In The
Mitigation Symposium: A National Workshop on Mitigating Losses of Fish and
Wildlife Habitats. Gustav A. Sandon, Tech. Coordinator, U.S.D.A., Rocky Mt.
For. and Rng. Exp. Stn., Ogden, UT, Gen. Tech. Rept., RM-65 p. 164-167.
Bolen, E.G., 1982. Playas, Irrigation and Wildlife in West Texas.
Transactions, North American Wildlife Conference.
Brinson, M.M., B.L. Swift, R.C. Plantico and J.S. Barclay, 1981. Riparian
Ecosystems: Their Ecology and Status. U.S. Fish and Wildlife Service
FWS/OBS-81/17.
Campbell, C.J., 1970. Ecological Implication of Riparian Vegetation
Management. J. Soil Water Conserv. 25:49.
Grouse, M.R. and R.R. Kindschy, 1981. A Method for Predicting Riparian
Vegetation Potential. Presented at Symposium on Acquisition and Utilization of
Aquatic Habitat Inventory Information. Portland, OR.
VI-7
-------
Cowardin, L.M., et al., 1979. Classification of Wetlands and Deepwater
Habitats of the United States. U.S. Fish and Wildlife Service, Washington,
D.C. FWS/OBS-79/31.
Council of Environmental Quality, 1978. Our Nation's Wetlands. An Interagency
Task Force Report. U.S. Government. Printing Office, Washington, D.C.
(041-011-000045-9).
Greeson, P.E., et al., editors, 1979. Wetland Function and Values: The State
of Our Understanding. American Water Resources Association, Minneapolis, MM.
Hawkins, C.P., M.L. Murphy and N.H. Anderson, 1982. Effects of Canopy,
Substrate Composition and Gradient on the Structure of Macroinvertebrate
Communities in Cascade Range Streams of Oregon. Ecology 63:1840.
Johnson, R.R. and D.A. Jones, 1977. Importance, Preservation and Management of
Riparian Habitat: A Symposium. U.S.D.A. For. Serv., Gen. Tech. Rep. RM-43. Ft.
Collins, Co.
Johnson, R.R. and J.F. McCormik, 1978. Strategies for Protection and
Management of Floodplain Wetlands and Other Riparian Ecosystems. U.S.D.A. For.
Serv., Gen. Tech. Rep. WO-12, Washington, D.C.
Karr, J.R. and I.J. Schlosser, 1977. Impact of Vegetation and Stream
Morphology on Water Quality and Stream Biota. U.S. EPA Cincinnati, Ohio EPA/
3-77-097.
Karr, J.R. and I.J. Schlosser, 1978. Water Resources and the Land-Water
Interface. Science 201:229.
Lotspeich, F.B., 1980. Watershed as the Basic Ecosystem: This Conceptual
Framework Provides a Basis for a Natural Classification System. Water
Resources Bulletin, American Water Resources Association, 16(4):581.
Moring, J.R., 1975. Fisheries Research Report No. 9, Oregon Dept. of Fish and
Wildlife, Corvallis.
Mueller-Dombois, D. and H. Ellenberg, 1974. Aims and Methods of Vegetation
Ecology. John Wiley and Sons, NY.
Peterson, R.C. and K.W. Cummins, 1974. Leaf Processing in a Woodland Stream.
Freshwater Biology 4:343.
Platts, W.S., 1982. Livestock and Riparian-Fishery Interactions: What are the
Facts? Trans. No. Amer. Wildlife Conf. (47), Portland, OR.
Ross, S.T. and J.A. Baker, 1983. The Response of Fishes to Periodic Spring
Floods in a Southeastern Stream. The American Midland Naturalist 109:1.
Schlosser, I.J,, 1982. Fish Community Structure and Function Along Two Habitat
Gradients in a Headwater Stream. Ecological Monographs 52:395.
VI-8
-------
Sedell, J., et al., 1975. The Processing of Conifer and Hardwood Leaves in Two
Coniferous Forest Streams. I. Weight Loss and Associated Invertebrates. Verh.
des. Inter. Vereins. Limn. 19:1617.
Sharpe, W.E., 1975. Timber Management Influences on Aquatic Ecosystems and
Recommendations for Future Research. Water Res. Bui. 11:546.
U.S. EPA, 1976. Forest Harvest, Residue Treatment, Reforestation and
Protection of Water Quality. U.S. EPA, Washington, D.C. EPA 910/9-76-020.
Van der Valk, A.6., C.B. Davis, J.L. Baker and C.F. Beer, 1980. Natural
Freshwater Wetlands as Nitrogen and Phosphorus Traps for Land Runoff p.
457-467. In Wetland Functions and Values: The State of Our Understanding, P.E.
Greeson, et al. (editors) Amer. Water Res. Asso. Minneapolis, MN.
CHAPTER III-l: WATER QUALITY INDICES
Brown, R.M., et al., 1970. "A Water Quality Index - Do We Dare?" Water and
Sewage Works, p. 339.
Dinius, S.H., 1972. "Social Accounting System for Evaluating Water Resources"
Water Resources Res. 8(5):1159.
Harkins, R.D., 1974. "An Objective Water Quality Index" Jour. Water Poll.
Cont. Fed. 46(3):588.
Kendall, M., 1975. Rank Correlation Methods, Charles Griffen and Co., London.
U.S. EPA, 1978. "Water Quality Indices: A Survey of Indices Used in the United
States," U.S. EPA, Washington, D.C., 600/4-78-005.
CHAPTER III-2: HARDNESS, ALKALINITY, pH AND SALINITY
Andrew, R.W., et al., 1977. Effects of Inorganic Complexing on the Toxicity of
Copper to Daphnia magna. Water Research, 11: 309.
Calamari, D. and Marchetti, R., 1975. Predicted and Observed Acute Toxicity of
Copper and Ammonia to Rainbow Trout (Salmo gairdneri Rich.). Progress in Water
Technology, 7: 569.
Calamari, D., et al., 1980. Influence of Water Hardness on Cadmium Toxicity to
Salmo gairdneri Rich. Water Research. 14: 1421.
Carroll, J.J., et al., 1979. Influences of Hardness Constituents on the Acute
Toxicity of Cadmium to Brook Trout (Salvelnus fontinalis). Bulletin of Environ-
mental Contamination and Toxicology, 22: 575.
European Inland Fisheries Advisory Commission. 1969. Water Quality Criteria
for European Freshwater Fish - Extreme pH Values and Inland Fisheries. Water
Research, 3: 593.
VI-9
-------
Graham, M.S. and Wood, C.M., 1981. Toxicity of Environmental Acids to the
Rainbow Trout: Interactions of Water Hardness, Acid Type, and Exercise.
Canadian Journal of Zoology, 59: 1518.
Haines, T.A., 1981. Acid Precipitation and its Consequences for Aquatic
Ecosystems: A Review. Transactions of the American Fisheries Society, 110:669.
Haranath, V.B., et al., 1978. Effect of Exposure to Altered pH Media on Tissue
Proteolysis and Nitrogenous End Products in a Freshwater Fish Tilapia
mossambica (Peters). Indian Journal of Experimental Biology, 16: 1088.
Hillaby, B.A., and Randall, D.J., 1979. Acute Ammonia Toxicity and Ammonia
Excretion in Rainbow Trout (Sal mo gairdneri). Journal of the Fisheries
Research Board of Canada 36:621.
Kinka_de M.L., and Erdman, H.E., 1975. The Influence of Hardness Components
(Ca and Mg ) in Water on the Uptake and Concentration of Cadmium in a
Simulated Freshwater Ecosystem. Environmental Research, 10: 308.
Lloyd, R., 1965. Factors that Affect the Tolerance of Fish to Heavy Metal
Poisoning, In: Biological Problems in Water Pollution^ 3rd Seminar, U.S.
Department of Health Education and Welfare, pp. 181-187.
Maetz, J. and Bornancin M., 1975, referenced in Calamari, et al., 1980.
Mount, D.I.,1973. Chronic Effect of Low pH on Fathead Minnow Survival, Growth,
and Reproduction. Water Research, 7: 987.
Pagenkopf, G.K., et al., 1974. Effect of Complexation on Toxicity of Copper to
Fish. Journal of the Fisheries Research Board of Canada, 31: 462-465.
Peterson, R.H., et al.,1980. Inhibition of Atlantic Salmon Hatching at Low pH.
Canadian Journal of Fisheries and Aquatic Sciences, 37:370.
Reid, G.K., 1961. Ecology of Inland Waters and Estuaries, D. Van Nostrand
Company, New York.
Sawyer, C.N. and McCarty, P.L., 1978. Chemistry for Environmental Engineering,
McGraw-Hill Book Company, New York.
Shaw, T.L. and Brown, V.M., 1974. The Toxicity of Some Forms of Copper to
Rainbow Trout. Water Research, 8: 377-392.
Stiff, M.J., 1971. Copper/Bicarbonate Equilibria in Solutions of Bicarbonate
Ions at Concentrations Similar to those Found in Natural Waters. Water
Research, 5: 171-176.
Thurston, R.V., et al., 1974, referenced in U.S. EPA, 1976.
U.S. EPA, 1976. Quality Criteria for Water, U.S. EPA, Washington, D.C.
Warren, C.E. 1971. Biology and Water Pollution Control, W.B. Saunders Company,
Philadelphia, Pennsylvania.
VI-10
-------
CHAPTER IV-1: HABITAT SUITABILITY INDICES
Inskip, P.O., 1982. Habitat Suitability Index Models: Northern pike, U.S. Fish
and Wildlife Service, Ft. Collins, CO, FWS/OBS-82/10.17.
McMahon, I.E. and J.W. Terrell, 1982. Habitat Suitability Index Models:
Channel Catfish. U.S. Fish and Wildlife Service, Ft. Collins, CO,
FWS/OBS-82/10.2.
Terrell, J.W., et al., 1982. Habitat Suitability Index Models: Appendix A.
Guidelines for Riverine and Lacustrine Applications of Fish HSI Models With
the Habitat Evaluation Procedures, U.S. Fish and Wildlife Service, Ft.
Collins, CO, FWS/OBS-82/10.A.
CHAPTER IV-2: DIVERSITY INDICES AND MEASURES OF COMMUNITY STRUCTURE
Beak, T.W., 1964. Biotic Index of Polluted Streams and Its Relationship to
Fisheries. Second International Conference on Water Pollution Research, Tokyo,
Japan.
Beck, W.M. Jr., 1955. Suggested Method for Reporting Biotic Data. Sewage Ind.
Wastes, 27:1193.
Bloom, S.A., et al., 1972. Animal-Sediment Relations and Community Analysis of
a Florida Estuary. Marine Biology, 13:43.
Boesch, D.F., 1957. Application of Numerical Classification in Ecological
Investigations of Water Pollution. EPA-600/3-77-033, U.S. EPA, Corvallis.
Bray, J.R. and Curtis, J.T., 1957. An Ordination of the Upland Forest
Communities of Southern Wisconsin. Ecological Monographs, 27:325.
Brillouin, L., 1960. Science and Information Theory. 2nd ed. Academic Press
Inc. NY.
Brock, D.A., 1977. Comparison of Community Similarity Indexes. Journal Water
Pollution Control Federation, 49:2488.
Buikema, A.L. Jr., 1980. Pollution Assessment: A Training Manual. UNESCO, U.S.
MAB Handbook No. 1. Washington, D.C.
Cairns, J. Jr., et al., 1968. The Sequential Comparison Index - A Simplified
Method for Non-Biologists to Estimate Relative Differences in Biological
Diversity in Stream Pollution Studies. Jour. Water Poll. Control Fed., 40:1607.
Cairns, J.R., Jr. and K.L. Dickson, 1969. Cluster Analysis of Potomac River
Survey Stations Based on Protozoan Presence-Absence Data. Hydrobiologia,
34:3-4, 414-432.
Cairns, J. Jr., et al., 1970. Occurrence and Distribution of Diatoms and Other
Algae in the Upper Potomac River. Notulae Naturae Acad. Nat. Sci.
Philadelphia, 436:1.
VI-11
-------
Cairns, J. Jr. and K.L. Dickson, 1971. A Simple Method for the Biological
Assessment of the Effects of Waste Discharges on Aquatic Bottom-Dwelling
Organisms. Jour. Water Poll. Control Fed., 43:755.
Cairns, J., Jr. and R.L. Kaesler, 1971. Cluster Analysis of Fish in a Portion
of the Upper Potomac River. Trans. American Fishery Society, 100:750.
Cairns, J. Jr., et al., 1973. Rapid Biological Monitoring Systems for
Determining Aquatic Community Structure in Receiving Systems. In Biological
Methods for the Assessment of Water Quality, (J. Cairns, Jr. and K.L. Dickson,
editors) American Society for Testing and Materials, STP 528, p. 148.
Cairns, J.R., Jr., 1977. Quantification of Biological Integrity. In The
Integrity of Water (R.K. Ballentine and L.J. Guarraia, editors) U.S.
Government Printing Office, Washington, D.C.
Chutter, F.M., 1972. An Empirical Biotic Index of the Quality of Water in
South African Streams and Rivers. Water Resources, 6:19.
Clifford, H.T. and W. Stephenson, 1975. An Introduction to Numerical
Classification. Academic Press, New York.
Czekanowski, J., 1913. Zarys Metod Statystycznych. Die Grundzuge der
Statischen Metoden, Warsaw.
Dixon, W.J. and F.J. Massey, Jr., 1969. Introduction to Statistical Analysis,
3rd ed. McGraw-Hill, NY.
Duncan, D.B., 1955. Multiple Range and Multiple F Tests, Biometrics, 11:1.
Fager, E.W., 1972. Diversity: A Sampling Study. Amer. Natur., 106:293.
Foerster, J.W., et al., 1974. Thermal Effects on the Connecticut River:
Phycology and Chemistry. Journal Water Pollution Control Federation, 46:2138.
Gammon, J.R., 1976. The Fish Populations of the Middle 340 km of the Wabash
River. Technical Report No. 86, Purdue University Water Resources Research
Center, West Lafayette, IN, pp. 1-48.
Gammon, J.R. and J.M. Reidy, 1981. The Role of Tributaries During an Episode
of Low Dissolved Oxygen in the Wabash River, IN. In AFS Warmwater Streams
Symposium. American Fish Society, Bethesda, MD.
Gammon, J.R., et al., 1981. Role of Electrofishing in Assessing Environmental
Quality of the Wabash River. In Ecological Assessments of Effluent Impacts on
Communities of Indigenous Aquatic Organisms (J.M. Bates and C.I Weber,
editors) Am. Soc. Testing and Materials, STP 730, Philadelphia, PA.
Gaufin, A.R., 1973. Use of Aquatic Invertebrates in the Assessment of Water
Quality. In Biological Methods for the Assessment of Water Quality, (J.
Cairns, Jr. and K.L. Dickson, editors) Am. Soc. for Testing and Materials, STP
528, Philadelphia, PA.
VI-12
-------
Gleason, H.A., 1922. On the Relation Between Species and Area. Ecology, 3:158.
Godfrey, P.O., 1928. Diversity as a Measure of Benthic Macroinvertebrate
Community Response to Water Pollution. Hydrobiologia, 57:111.
Hartigan, J.A., 1975. Clustering Algorithms. Wiley-Interscience, NY.
Heck, K.L. Jr., 1976. Community Structure and the Effects of Pollution in
Sea-Grass Meadows and Adjacent Habitats. Marine Biology, 35:345.
Herricks, E.E. and J. Cairns Jr., 1982. Biological Monitoring. Part III:
Receiving System Methodology Based on Community Structure. Water Research,
16:141.
Hilsenhoff, W.L., 1977. Use of Arthropods to Evaluate Water Quality of
Streams. WI Dept. Nat. Resour. Tech. Bull. No. 100.
, 1982. Using a Biotic Index to Evaluate Water Quality in
Streams. WI Dept. Nat. Resour. Tech. Bull No. 132.
Horn, H.S., 1966. Measurement of "Overlap" in Compaative Ecological Studies.
American Naturalist, 100:419.
Howmiller, R.P. and M.A. Scott, 1977. An Environmental Index Based on Relative
Abundance of Oligochaete Species. Jour. Water Poll. Control Fed. 49:809.
Hughes, B.D., 1978. The Influence of Factors Other than Pollution on the Value
of Shannon's Diversity Index, for Benthic Macroinvertebrates in Streams, Water
Res., 92:359.
Hurlbert, S.H., 1971. The Nonconcept of Species Diversity: A Critique and
Alternative Parameters. Ecology, 52:577.
Hutcheson, K., 1970. A Test for Comparing Diversities Based on the Shannon
Formula. Jour. Theoret. Biol. 29:151.
Jaccard, P., 1912. The Distribution of Flora in an Alpine Zone. New Phytol.,
11:37.
Johnson, M.G. and R.O. Brinkhurst, 1971. Associations and Species Diversity in
Benthic Macroinvrtebrates of Bay of Quninte and Lake Ontario. Jour. Fish. Res.
Bd. Canada, 28:1683.
Kaesler, R.L., et al., 1971. Cluster Analysis of Non-Insect
Macro-Invertebrates of the Upper Potomac River. Hydrobiologia, 37:173.
, 1978. Use of Indices of Diversity and Hierarchical
Diversity in Stream Surveys. In Biological Data in Water Pollution Assessment;
Quantitative and Statistical Analyses (K.L. Dickson, et al., editors). Am.
Soc. Testing and Materials, STP 652, Philadelphia, PA.
Kaesler, R.L. and J. Cairns, Jr., 1972. Cluster Analysis of Data from
Limnological Surveys of the Upper Potomac River. Am. Midland Naturalist, 88:56.
VI-13
-------
Keuls, M., 1952. The Use of the "Studentized Range" in Connection with an
Analysis of Variance. Euphytica, 1:112.
Keup, L.E., 1966. Stream Biology for Assessing Sewage Treatment Plant
Efficiency. Water and Sewage Works, 113:411.
Kohn, A.J., 1968. Microhabitats, Abundance and Food of Conus on Atoll Reefs in
the Maldive and Chagos Islands. Ecology, 49:1046.
Livingston, R.J., 1975. Impact of Kraft Pulp-Mill Effluents on Estuarine
Coastal Fishes in Apalachee Bay, FL. Marine Biology, 32:19.
Lloyd, M.J., et al., 1968. On the Calculation of Information-Theoretical
Measures of Diversity. Amer. Midland Natur., 79:257.
Lloyd, M., and R.J. Ghelardi, 1964. A Table for Calculating the "Equitability"
Component of Species Diversity. Jour. Anim. Ecol., 33:217.
MacAuthur, R.H., 1957. On the Relative Abundance of Bird Species. Proc. Nat.
Acad. Sci. Washington, D.C. 43:293.
, 1960. On the Relative Abundance of Species. Am. Natur., 94:25.
Margalef, R., 1951. Diversidad de Especies en las Communidades Naturales. Pub.
Inst. Biol. Apl. (Barcelona) 9:5.
, 1956. Information of Diversidad Especifica en las Communidades
de Organismos. Invest. Pesq., 3:99.
, 1958. Information on Theory in Ecology. English translation by
W. Hall in Yearbook of the Society for General Systems Research, 3:36.
Mclntosh, R.P., 1967. An Index of Diversity and the Relation of Certain
Concepts to Diversity. Ecology, 48:392.
Menhinick, E.F., 1964. A Comparison of Some Species-Individuals Diversity
Indices Applied to Samples of Field Insects. Ecology, 45:859.
Morisita, M., 1959. Measuring of Interspecific Association and Similarity
Between Communities. Memoirs Faculty Sci., Kyushu Univ. Sen. E. Biol., 3:65.
Newman, D., 1939. The Distribution of Range in Samples from a Normal
Population, Expressed in Terms of an Independent Estimate of Standard
Deviation. Biometrika, 31, 20.
Ochiai, A., 1957. Zoogeographical Studies on the Soleoid Fishes Found in Japan
and its Neighboring Regions - II. Bull. Japan. Soc. Sci. Fisheries, 22:526.
Odum, E.P., 1959. Fundamentals of Ecology, 2nd ed. W.B. Sanders Co.,
Philadelphia, PA.
Osborne, L.L., et al., 1980. Use of Hierarchical Diversity Indices in Lotic
Community Analysis. Jour. Appl. Ecol., 17:567.
VI-14
-------
Pantle, R., and H. Buck, 1955. Die Biologische Uberwachund der Gewasser und
die Darstellung der Ergebnisse. Gas und Wasserfach, 96:604.
Patten, B.C., 1962. Species Diversity in Net Phytoplankton of Raritan Bay.
Jour. Mar. Res., 20:57.
Peet, R.K., 1975. Relative Diversity Indices. Ecology, 56:496.
Perkins, J.D., 1983. Bioassay Evaluation of Diversity and Community Comparison
Indexes. Jour. Water Poll. Con. Fed., 55:522.
Peters, J.A., 1968. A Computer Program for Calculating Degree of
Biogeographical Resemblence Between Areas. Systematic Zoology, 17:64.
Pielou, E.C., 1969. An Introduction to Mathematical Ecology.
Wiley-Interscience, NY, 286.
, 1975. Ecological Diversity. Wiley-Interscience, NY, 165.
Pinkham, C.F.A. and J.G. Pearson, 1976. Applications of a New Coefficient of
Similarity to Pollution Surveys. Jour. Water Poll. Con. Fed., 48:717.
Sanders, H.L. 1960. Benthic Studies in Buzzards Bay III. The Structure of the
Soft-Bottom Community, Limnology and Oceanography 5:138.
Shannon, C.E., and W. Weaver, 1963. The Mathematical Theory of Communication.
University of Illinois Press, Urbana, IL.
Sneath, P.M.A. and R.R. Sokal, 1973. Numerical Taxonomy. The Principles and
Practice of Numerical Classification. Freeman, San Francisco.
Simpson, E.H., 1949. Measurement of Diversity. Nature, 163:68.
Sokal, R.R., 1961. Distance as a Measure of Taxonomic Similarity. Systematic
Zoology, 10:70.
Sokal, R.R., and C.D. Michener, 1958. A Statistical Method for Evaluating
Systematic Relationships. Univ. Kansas Sci. Bull., 38:1409.
Sokal, R.R. and F.J. Rohlf, 1962. The Comparison of Dendograms by Objective
Methods. Taxon, 11:33.
Whittaker, R.H., 1952. A Study of Summer Foliage Insect Communities in the
Great Smoky Mountains. Ecological Monographs, 22: 6.
, 1964. Dominance and Diversity in Land Plant Communities.
Science, 147:250.
Whittaker, R.H. and C.W. Fairbanks, 1958. A Study of Plankton and Copepod
Communities in the Columbia Basin, Southeastern Washington. Ecology, 39:46.
Wiener, N., 1948. Cybernetics. John Wiley & Sons, Inc., NY, 194 p.
VI-15
-------
Wilhm, J.L., 1967. Comparison of Some Diversity Indices Applied to Populations
of Benthic Macroinvertebrates in a Stream Receiving Organic Wastes. Jour.
Water Poll. Control. Fed., 39:1673.
, 1968. Use of Biomass Units in Shannon's Formula. Ecology, 49:153.
, 1970. Range of Diversity Index in Benthic Macroinvertebrate
Populations. Jour. Water Poll. Con. Fed., 42:R221.
, 1972. Graphic and Mathematical Analyses of Biotic Communities in
Polluted Streams. Ann. Rev. Entomology, 17:223.
Wilhm, J.L. and T.C. Dorris, 1968. Biological Parameters for Water Quality
Criteria. Bioscience, 18:477.
Williams, W.T., 1971. Principles of Clustering. Ann. Rev. Ecol. Syst., 2:303.
Winget, R.N. and F.A. Mangum, 1979. Biotic Condition Index: Integrated
Biological, Physical, and Chemical Stream Parameters for Management. U.S.
Forest Service Intermountain Reg., p. 1-51.
Zar, J.H.» 1974. Biostatistical Analysis, Prentice-Hall, Inc., Englewood
Cliffs, NJ, 620 p.
CHAPTER IV-3: RECOVERY INDEX
Cairns, J. Jr., 1975. Biological Integrity-A Quantitative Determination. In
U.S. EPA, The Integrity of Water (R.K. Ballentine and L.J. Guarraia, editors).
U.S. Government Printing Office, Washington, D.C., 055-001-01068-1.
CHAPTER IV-4: INTOLERANT SPECIES ANALYSIS
Ball, J., 1982. Stream Classification Guidelines for Wisconsin. Technical
Bulletin (Draft), Wisconsin Department of Natural Resources, Madison.
Brett, J.R., 1956. Some Principles in the Thermal Requirements of Fishes.
Quarterly Review of Biology 31:75.
Carlander, K.D., 1969 and 1977. Handbook of Freshwater Fishery Biology, Vols.
I and II, Iowa State University Press, Ames, Iowa.
Haines, T.A., 1981. Acidic Precipitation and Its Consequences for Aquatic Eco-
systems: A Review. Trans. Amer. Fish. Soc., 110: 669.
Hutchinson, G.E., 1957. Concluding Remarks in Population Studies: Animal
Ecology and Demography, Cold Spring Harbor Symposia on Quantitative Biology,
22:415.
Johnson, W.W., and Finley, M.T., 1980. Handbook of Acute Toxicity of Chemicals
to Fish and Aquatic Invertebrates. U.S. FWS, Washington, D.C., Resource
Publication 137.
VI-16
-------
Karr, J.R., 1981. Assessment of Biotic Integrity Using Fish Communities.
Fisheries 6:21.
Kendigh, S.C., 1974. Ecology with Special Reference to Animals and Man.
Prentice-Hall, Inc., Englewood Cliffs, NJ.
Lee, D.S., et al., 1980. Atlas of North American Freshwater Fishes. N.C.,
State Mus. Nat. Hist,, Raleigh, NC.
Morrow, J.E., 1980. The Freshwater Fishes of Alaska. Alaska Northwest
Publishing Co., Anchorage, AK.
Moyle, P.B., 1976. Inland Fishes of California. University of California
Press, Berkeley, CA.
Muncy, R.J., et al., 1979. Effects of Suspended Solids and Sediment on
Reproduction and Early Life of Warmwater Fishes: A Review. U.S. EPA,
Corvallis, OR, EPA-600/3-79-042.
Pflieger, W.L., 1975. The Fishes of Missouri. Missouri Dept. Conserv.,
Jefferson City, MO.
Robins, C.R., et al., 1980. A List of Common and Scientific Names of Fishes
from the United States and Canada, 4th ed., AFS Special Publ. No. 12,
Bethesda, MD.
Scott, W.B., and Grossman, E.J., 1973. Freshwater Fishes of Canada. Fisheries
Research Board of Canada, Bull. 184.
Shelford, V.E., 1911. Ecological Succession. Biol. Bull. 21: 127-151, 22:1.
Smith, P.M., 1979. The Fishes of Illinois. University of Illinois Press,
Urbana, IL.
Timbol, A.S., and Maciolek, J.A., 1978. Stream Channel Modification in Hawaii.
Part A: Statewide Inventory of Streams, Habitat Factors, and Associated Biota.
U.S. FWS, Columbia, MO, FWS/OBS-78/16.
Trautman, M.B., 1957. The Fishes of Ohio. Ohio State University Press,
Columbus, OH.
U.S. EPA, 1980. Ambient Water Quality Criteria for Aldrin/Dieldrin, Chlordane,
DDT, Endosulfan, Endrin, Heptachlor, Lindane, PCBs, Toxaphene, Cyanide,
Arsenic, Cadmium, Chromium, Copper, Lead, Mercury, Nickel, Selenium, Silver,
and Zinc. U.S. EPA, Washington, D.C., EPA 440/5-80-
Vannote, R.L., et al., 1980. The River Continuum Concept. Can. Jour. Fish.
Aquat. Sci., 37:130.
Wallen, E.I., 1951. The Direct Effect of Turbidity on Fishes. Oklahoma A&M
College, Stillwater, OK, Biol. Series No. 2, 48:1.
Warren, C.E., 1971. Biology and Water Pollution Control. W.B. Saunders Co.,
Philadelphia, PA.
VI-17
-------
CHAPTER IV-5: OMNIVORE-CARNIVORE ANALYSIS
Cairns, J., Jr., 1977. Quantification of Biological Integrity. In the
Integrity of Water (R.K. Ballentine and L.J. Guarraia, editors). U.S.
Government Printing Office, Washington, O.C., 055-001-01068-1.
Carlander, K.D., 1969. Handbook of Freshwater Fishery Biology, Vol. I. Iowa
State University Press, Ames, IA.
, 1977, Handbook of Freshwater Fishery Biology, Vol. II. Iowa
State University Press, Ames, IA.
Cross, F.B. and J.T. Collins, 1975. Fishes in Kansas (R.F. Johnson, editor).
University of Kansas Publ., Museum of Nat. Hist., Lawrence, KS.
Cummins, K.W., 1974. Structure and Function of Stream Ecosystems. BioScience
24:631.
, 1975. The Ecology of Running Waters: Theory and Practice. In
Proc. Sandusky River Basin Symp. (D.B. Baker, et al. editors). International
Joint com. on the Great Lakes, Heidelburg College, Tiffin, OH.
Darnell, R.M. 1961. Trophic Spectrum of an Estuarine Community, Based on
Studies of Lake Ponchartrain, Louisiana. Ecology 42: 553.
Fausch, K.D., et al., 1982. Regional Application of an Index of Biotic
Integrity Based on Stream Fish Communities, Submission to Trans. Amer. Fish.
Soc.
Karr, J.R., 1981. Assesment of Biotic Integrity Using Fish Communities.
Fisheries 6:21.
Karr, J.R., et al., 1983. Habitat Preservation for Midwest Stream Fishes:
Principles and Guidelines, U.S. EPA, Corvallis, OR, EPA-600/3-83-006.
Karr, J.R. and D.R. Dudley, 1978. Biological Integrity of a Headwater Streams:
Evidence of Degradation, Prospects for Recovery. In Environmental Impact of
Land Use on Water Quality: Final Report on the Black Creek Project
(Supplemental Comments) (J. Morrison, editor), U.S. EPA, Chicago, IL,
EPA-905/9-77-007-D, pp. 3-25.
Kendeigh, S.C., 1974. Ecology with Special Reference to Animals and Man.
Prentice-Hall, Inc., Englewood Cliffs, NO.
Kuehne, R.A., 1962. A Classification of Streams, Illustrated by Fish
Distribution in an Eastern Kentucky Creek. Ecology 43: 608.
Larimore, W.R., and P.M. Smith, 1963. The Fishes of Champaign County, Illinois
as Affected by 60 Years of Stream Changes. 111. Nat. Hist. Sur. Bull. 28:299.
Lee, D.S., et al., 1980. Atlas of North American Freshwater Fishes. N.C. State
Mus. Nat. Hist., Raleigh, NC.
Lindeman, R.L., 1942. The Trophic-Dynamic Aspect of Ecology. Ecology 23.
VI-18
-------
Menzel, B.W., and H.L. Fierstine, 1976. A Study of the Effects of Stream
Channelization and Bank Stabilization on Warmwater Sport Fish in Iowa. No. 5:
Effects of Long-Reach Stream Channelization on Distribution and Abundance of
Fishes, U.S. Fish and Wildlife Service, Columbia, MO, FWS/OBS-76-15.
Morita, C.M., 1953. Freshwater Fishing in Hawaii. Div. of Fish and Game, Dept.
Land Nat. Res., Honolulu, HI.
Morrow, J.E., 1980. The Freshwater Fishes of Alaska. Alaska Northwest
Publishing Co., Anchorage, AK.
Moyle, P.B., 1976. Inland Fishes of California. University of California
Press, Berkeley.
Odum, H.T., 1957. Trophic Structure and Productivity of Silver Springs, FL,
Ecol. Monogr. 27:55.
Pflieger, W.L., 1975. The Fishes of Missouri. Missouri Dept. Conserv.,
Jefferson City, MO.
Reid, G.K., and R.D. Wood, 1976. Ecology of Inland Waters and Estuaries, 2nd
Ed., D. Van Nostrand Co., NY.
Richardson, J.L., 1977. Dimensions in Ecology. Williams & Wilkins, Baltimore,
MD.
Robins, C.R., et al., 1980. A List of Common and Scientific Names of Fishes
from the United States and Canada, 4th ed., Special Pub!. No. 12, American
Fisheries Soc., Bethesda, MD.
Schlosser, I.J, 1981. Effects of Perturbations by Agricultural Land Use on
Structure and Function of Stream Ecosystems. Ph.D. dissertation, University of
Illinois, Champaign - Urbana, IL.
, 1982a. Trophic Structure, Reproductive Success, and Growth
Rate o? Fishes Tn a Natural and Modified Headwater Stream. Can. Jour. Fish
Aquat. Sci. 39:968.
, 1982b. Fish Community Structure and Function Along Two Habitat
Gradients in aHeadwater Stream. Ecol. Monog. 52: 395.
Scott, W.B., and E.J. Grossman, 1973. Freshwater Fishes of Canada. Fisheries
Research Board of Canada, Bull. 184.
Shelford, V.E., 1911. Ecological Succession. Biol. Bull. 21:127, 22:1.
Smith, P.W., 1979. The Fishes of Illinois. University of Illinois Press,
Urbana, IL.
Timbol, A.S. and J.A. Maciolek, 1978. Stream Channel Modification in Hawaii.
Part A: Statewide Inventory of Streams, Habitat Factors, and Asociated Biota.
U.S. Fish and Wildlife Service, Columbia, MO, FWS/OBS-78/16.
VI-19
-------
Trautman, M.B., 1957. The Fishes of Ohio. Ohio State University Press,
Columbus, OH.
U.S. EPA, 1980. Ambient Water Quality Criteria (several volumes) for
Aldrin/Dieldrin, Chlordane, DDT, Encosulfan, Endrin, Heptachlor, Lindane,
PCBs, Toxaphene, Cyanide, Arsenic, Cadmium, Chromium, Copper, Lead, Mercury,
Nickel, Selenium, Silver, and Zinc. U.S. EPA, Washington, D.C., EPA 440/5-80.
Wallen, E.I., 1951. The Direct Effect of Turbidity on Fishes. Oklahoma A&M
College, Stillwater, OK, Biol. Series No. 2, 48:1.
Warren, C.E., 1971. Biology and Water Pollution Control. W.B. Saunders,
Philadelphia, PA.
CHAPTER IV-6: REFERENCE REACH COMPARISON
Bailey, R.G., 1976. Ecoregions of the United States. U.S.D.A.-Forest Service.
Intermtn. Reg. Ogden, UT.
Barbour, C.D. and J.H. Brown, 1974. Fish Species Diversity in Lakes. Am. Nat.
108:473.
Federal Register, 1982. Proposed Water Quality Standards and Public Meetings.
47(210):49234.
Gilbert, C.R., 1980. Zoogeographic Factors in Relation to Biological
Monitoring of Fish. In Biological Monitoring of Fish (C.H. Hocutt and J.R.
Stauffer, Jr., editors). D.C. Hath Co., Lexington, MA, p. 309-355.
Green, R.H., 1979. Sampling Design and Sampling Methods for Environmental
Biologists. John Wiley and Sons, NY.
Hall, J.D., et al., 1978. An Improved Design for Assessing Impacts of
Watershed Practices on Small Streams. Verh. Interna. Verein. Limnol. 20:1359.
Hughes, R.M., Effects of Mining Wastes on Two Stream Ecosystems: Demonstration
of an Approach for Estimating Ecological Integrity and Attainable Uses.
Hughes, R.M., et al., 1982. An Approach for Determining Biological Integrity
in Flowing Waters. In Place Resource Inventories: Principles and Practices
(T .8. Brann, L.O. House IV, and H.G. Lund, editors). Soc. Am. Foresters,
Bethesda, MD.
Hughes, R.M. and J.M. Qmernik, 1981a. Use and Misuse of the Terms Watershed
and Stream Order. In The Warmwater Streams (L.A, Krumholz, editor). Symposium
Am. Fish. Soc., Bethesda, MD.
, 1981b. A Proposed Approach to Determine
Regional Patterns i"n Aquatic Ecosystems. In Acquisition and Utilization of
Aquatic Habitat Inventory Information (N.B. Armantrout, editor). Am. Fish.
Soc., Bethesda, MD.
VI-20
-------
, 1983. An Alternative for Characterizing Stream
Size.In Dynamics of Lotic Ecosystems (T.D. Fontain III and S.M. Bartell,
editors). Ann Arbor Science, Ann Arbor, MI.
Karr, J.R., 1981. Assesment of Biotic Integrity Using Fish Communties.
Fisheries 6:21.
Lotspeich, F.B. and W.S. Platts, 1982. An Integrated Land-Aquatic
Classification System. N. Amer. J. Fish. Mgmt. 2:138.
MacArthur, R.H. and E.O. Wilson, 1967. The Theory of Island Biogeography,
Princeton Univ. Press., Princeton, NJ.
Marsh, P.C. and J.E. Luey, 1982. Oases for Aquatic Life in Agricultural
Watersheds. Fisheries 7:16.
Omernik, J.M. and R.M. Hughes, 1983. An Approach for Defining Regional
Patterns of Aquatic Ecosystems and Attainable Stream Quality in Ohio. Progress
Report. U.S. EPA, Corvallis, OR.
Pfleiger, W.L., M.A. Schene, Jr., and P.S. Haverland. 1981. Techniques for the
Classification of Stream Habitats With Examples of Their Application in
Defining the Stream Habitats of Missouri. In Acquisition and Utilization of
Aquatic Habitat Inventory Information (N.B. Armantrout, editor). Am. Fish.
Soc., Bethesda, MD.
Strahler, A.N., 1957. Quantitative Analysis of Watershed Geomorphology. Trans.
Am. Geophys. Union 38:913.
Trautman, M.G., 1981. The Fishes of Ohio. Ohio State Univ. Press.
U.S.D.I.-Geological Survey, 1970. The National Atlas of the United States of
America. U.S. Government Printing Office, Washington, D.C.
Vannote, R.L., et al.s 1980. The River Continuum Concept. Can. J. Fish. Aquat.
Sci. 37:130.
Warren, C.E., 1979. Toward Classification and Rationale for Watershed
Management and Stream Protection. EPA-600/3-79-059. NTIS Springfield, VA.
VI-21
-------
" APPENDIX A-l:
SAMPLE HABITAT SUITABILITY INDEX
(Channel Catfish)
-------
Biological Services Program
FWS/OBS-82/10.2
FEBRUARY 1982
HABITAT SUITABILITY INDEX MODELS:
CHANNEL CATFISH
Fish and Wildlife Service
U.S. Department of the Interior
-------
FWS/OBS-82/10.2
February 1982
HABITAT SUITABILITY INDEX MODELS: CHANNEL CATFISH
by
Thomas E. McMahon
and
James W. Terrell
Habitat Evaluation Procedures Group
Western Energy and Land Use Team
U.S. Fish and Wildlife Service
Drake Creekside Building One
2625 Redwing Road
Fort Collins, Colorado 80526
Western Energy and Land Use Team
Office of Biological Services
Fish and Wildlife Service
U.S. Department of the Interior
Washington, D.C. 20240
-------
PREFACE
The habitat use information and Habitat Suitability Index (HSI) models
presented in this document are an aid for impact assessment and habitat man-
agement activities. Literature concerning a species' habitat requirements and
preferences is reviewed and then synthesized into HSI models, which are scaled
to produce an index between 0 (unsuitable habitat) and 1 (optimal habitat).
Assumptions used to transform habitat use information into these mathematical
models are noted, and guidelines for model application are described. Any
models found in the literature which may also be used to calculate an HSI are
cited, and simplified HSI models, based on what the authors believe to be the
most important habitat characteristics for this species, are presented.
Use of the models presented in this publication for impact assessment
requires the setting of clear study objectives and may require modification of
the models to meet those objectives. Methods for reducing model complexity
and recommended measurement techniques for model variables are presented in
Appendix A.
The HSI models presented herein are complex hypotheses of species-habitat
relationships, not statements of proven cause and effect relationships.
Results of model performance tests, when available, are referenced; however,
models tnat have demonstrated reliability in specific situations may prove
unreliable in others. For this reason, the FWS encourages model users to
convey comments and suggestions that may help us increase the utility and
effectiveness of this habitat-based approach to fish and wildlife planning.
Please send comments to:
Habitat Evaluation Procedures Group
Western Energy and Land Use Team
U.S. Fish and Wildlife Service
2625 Redwing Road
Ft. Collins, CO 80526
-------
CONTENTS
Page
PREFACE : iii
ACKNOWLEDGEMENTS vi
HABITAT USE INFORMATION I
General 1
Age, Growth, and Food 1
Reproduction 1
Specific Habitat Requirements 1
HABITAT SUITABILITY INDEX (HSI) MODELS 4
Model Applicability • 4
Model Description - Riverine 5
Model Description - Lacustrine 8
Suitability Index (SI) Graphs for
Model Variables , 9
Riverine Model 15
Lacustrine Model 17
Interpreting Model Outputs 22
ADDITIONAL HABITAT MODELS 24
Model 1 24
Model 2 25
Model 3 25
REFERENCES CITED 25
-------
CHANNEL CATFISH (Ictalurus punctatus)
HABITAT USE INFORMATION
General
The native range of channel catfish (lc^J_u_rus punctatus) extends from
the southern portions of the Canadian prairie provinces south to the Gulf
states, west to the Rocky Mountains, and east to the Appalachian Mountains
(Trautman 1957; Miller 1966; Scott and Crossman 1973). They have been widely
introcuced outside this range and occur in essentially all of the Pacific and
Atlantic drainages in the 48 contiguous states (Moore 1968; Scott and Crossman
1973). The greatest abundance of channel catfish generally occurs in the open
(unleveed) floodplains of the Mississippi and Missouri River drainages (Walden
1964).
Age, Growth, and Food
Age at maturity in channel catfish is variable. Catfish from southern
areas with longer growing seasons mature earlier and at smaller sizes than
those from northern areas (Davis and Posey 1958; Scott and Crossman 1973).
Southern catfish mature at age V or less (Scott and Crossman 1973; Pflieger
1975) while northern catfish mature at age VI or greater for males and at age
VIII or greater for females (Starostka and Nelson 1974).
Young-of-the-year (age 0) catfish feed predominantly on plankton and
aquatic insects (Bailey and Harrison 1948; Walburg 1975). Adults are oppor-
tunistic feeders with an extremely varied diet, including terrestrial and
aquatic insects, detrital and plant material, crayfish, and molluscs (Bailey
and Harrison 1948; Miller 1966; Starostka and Nelson 1974). Fish may form a
major part of the diet of catfish > 50 cm in length (Starostka and Nelson
1974). Channel catfish diets in rivers and reservoirs do not appear to be
significantly different (see Bailey and Harrison 1948; Starostka and Nelson
1974). Feeding is done by both vision and chemosenses (Davis 1959) and occurs
primarily at night (Pflieger 1975). Bottom feeding is more characteristic but
food 's also taken throughout the water column (Scott and Crossman 1973).
Additional information on the composition of adult and juvenile diets is
provided in Leidy and Jenkins (1977).
Reproduction
Channel catfish spawn in late spring and early summer (generally late May
through mid-July) when temperatures reach about 21° C (Clemens and Sneed 1957;
Marzolf 1957; Pflieger 1975). Spawning requirements appear to be a major
factor in determining habitat suitability for channel catfish (Clemens and
Sneed 1957). Spawning is greatly inhibited if suitable nesting cover is
unavailable (Marzolf 1957).
Specific Habitat Requirements
Channel catfish populations occur over a broad range of environmental
conditions (Sigler and Miller 1963; Scott and Crossman 1973). Optimum riverine
-------
habitat is characterized by warm temperatures (Clemens and Sneed 1957; Andrews
et al. 1972; Biesinger et al. 1979). and a diversity of velocities, depths, and
structural features that provide cover and food (Bailey and Harrison 1948).
Optimum lacustrine habitat is characterized by large surface area, warm temper-
atures, high productivity, low to moderate turbidity, and abundant cover
(Davis 1959; Pflieger 1975).
Fry, juvenile, and adult channel catfish concentrate in the warmest
sections of rivers and reservoirs (Ziebell 1973; Stauffer et al . 1975; McCall
1977). They strongly seek cover, but quantitative data on cover requirements
of channel catfish in rivers and reservoirs are not available. Debris, logs,
cavities, boulders, and cutbanks in lakes and in low velocity (< 15 cm/sec)
areas of deep pools and backwaters of rivers will provide cover for channel
catfish (Bailey and Harrison 1948). Cover consisting of boulders and debris
in deep water is important as overwintering habitat (Miller 1966; Jester 1971;
Cross and Collins 1975). Deep pools and littoral areas (£ 5- m deep) with
> 40% suitable cover are assumed to be optimum. Turbidities > 25 ppm but
< 100 ppm may somewhat moderate the need for fixed cover (Bryan et al. 1975).
Riffle and run areas with rubble substrate and pools (< 15 cm/sec) and
areas with debris and aquatic vegetation are conditions associated with high
production of aquatic insects (Hynes 1970) consumed by channel catfish ir,
rivers (Bailey and Harrison 1948). Channel catfish are most abundant in river
sections with a diversity of velocities and structural features. Therefore, it
is assumed that a riverine habitat with 40-60% pools would be optimum for
providing riffle habitat for food production and feeding and pool habitat for
spawning and resting cover (Bailey and Harrison 1948). It also is assumed
that at least 20% of lake or reservoir surface area should consist of littoral
areas (< 5 m deep) to provide adequate area for spawning, fry and juvenile
rearing, and feeding habitat for channel catfish.
High standing crops of warmwater fishes are associated with total
dissolved solids (TDS) levels of 100 to 350 ppm for reservoirs in which the
concentrations of carbonate-bicarbonate exceed those of sulfate-chloride
(Jenkins 1976). It is assumed that high standing crops of channel catfish in
lakes or reservoirs will, on the average, correspond to this TDS level.
Turbidity in rivers and -reservoirs and reservoir size are other factors
that may influence habitat suitability for channel catfish populations.
Channel catfish are abundant in rivers and reservoirs with varying levels of
turbidity and siltation (Cross and Collins 1975). However, low to moderate
turbidities (< 100 ppm) are probably optimal for both survival and growth
(Finnell and Jenkins 1954; Buck 1956; Marzolf 1957). Larger reservoirs
(> 200 ha) are probably more suitable reservoir habitat for channel catfish
populations because survival and growth are better than in smaller reservoirs
(Finnell and Jenkins 1954; Marzolf 1957). Other factors that may affect
reservoir habitat suitability for channel catfish are mean depth, storage
ratio (SR), and length of agricultural growing season. Jenkins (1974) found
that high mean depths were negatively correlated with standing crop of channel
catfish. Mean depths are an inverse correlate of shoreline development (Ryder
et al. 1974), thus higher mean depths may mean less littoral area would be
available. Jenkins (1976) also reported that standing crops of catfishes
(Ictaluridae) peaked at an SR of 0.75. Standing crops of channel catfish were
-------
postively correlated to growing season length (Jenkins 1970). However, harvest
of channel catfish reported in reservoirs was not correlated with growing
season length (Jenkins and Morais 1971-).
Dissolved oxygen (DO) levels of 5 mg/1 are adequate for growth and
survival of channel catfish, but D.O. levels of > 7 mg/1 are optimum (Andrews
et al, 1973; Carlson et al. 1974). Dissolved oxygen levels < 3 mg/1 retard
growth (Simco and Cross 1966), and feeding is reduced at D.O. levels < 5 mg/1
(Randolph and Clemens 1976).
Adjjj_t. Adults in rivers are found in large, deep pools with cover. They
move to riffles and runs at night to feed (McCammon 1956; Davis 1959; Pflieger
1971; 1975). Adults in reservoirs and lakes favor reefs and deep, protected
areas with rocky substrates or other cover. They often move to the shoreline
or tributaries at night to feed (Davis 1959; Jester 1971; Scott and Grossman
1973).
The optimal temperature range for growth of adult channel catfish is
26-29c C (Shrable et al. 1969; Chen 1976j. Growth is poor at temperatures
< 21° C (McCammon and LaFaunce 1961; Mack!in and Soule 1964; Andrews and
Stickrey 1972) and ceases at < 18° C (Starostka and Nelson 1974). An upper
lethal temperature of 33.5° C has been reported for catfish acclimated at
25° C (Carlander 1969).
Adult channel catfish were most abundant in habitats with salinities
< 1.7 ppt in Louisiana, although they occurred in areas with salinities up to
11.4 ppt (Perry 1973). Salinities < 8 ppt are tolerated with little or no
effect, but growth slows above this level and does not occur at salinities
> 11 ppt (Perry and Avault 1968).
Embryo. Dark and secluded areas are required for nesting (Marzolf 1957).
Males build and guard nests in cavities, burrows, under rocks, and in other
protected sites (Davis 1959; Pflieger 1975). Nests in large impoundments
generally occur among rubble and boulders along protected shorelines at depths
of about 2-4 m (Jester 1971). Catfish in large rivers are likely to move into
shallow, flooded areas to spawn (Bryan et al. 1975). Lawler (1960) reported
that spawning in Utah Lake, Utah, was concentrated in sections of the lake
with abundant spawning sites of rocky outcrops, trees, and crevices. The male
catfish fans embryos for water exchange and guards the nest from predators
(Miller 1966; Minckley 1973). Embryos can develop in the temperature range of
15.5 to 29.5° C, with the optimum about 27° C (Brown 1942; Clemens and Sneed
1957). They do not develop at temperatures < 15.5° C (Brown 1942). Embryos
hatch in 6-7 days at 27° C (Clemens and Sneed 1957).
Laboratory studies indicate that embryos three days old and older can
tolerate salinities up to 16 ppt until hatching, when tolerance drops to 8 ppt
(Alien and Avault 1970). However, 2 ppt salinity is the highest level in
which successful spawning in ponds has been observed (Perry 1973). Embryo
survival and production in reservoirs will probably be high in areas that are
not subject to disturbance by heavy wave action or rapid water drawdown.
F_ry. The optimal temperature range for growth of channel catfish fry is
29-30° C (West 1966). Some growth does occur down to temperatures of 18° C
(Starcstka and Nelson 1974), but growth generally is poor in cool waters with
average summer temperatures < 21° C '(McCammon and LaFaunce 1961; MaclOin and
-------
Soule 1964; Andrews et al. 1972) and in areas with short agricultural growing
seasons (Starostka and Nelson 1974). Upper incipient lethal levels for fry
are about 35-38° C, depending on acclimation temperature (Moss and Scott 1961;
Allen and Strawn 1968). Optimum salinities for fry range from 0-5 ppt;
salinities £ 10 ppt are marginal as growth is greatly reduced (Allen and
Avault 1970).
Fry habitat suitability in reservoirs is related to flushing rate of
reservoirs in midsummer. Walburg (1971) found abundance and survival of fry
greatly decreased at flushing rates < 6 days in July and August.
Channel catfish fry have strong shelter-seeking tendencies (Brown et al.
1970), and cover availability will be important in determining habitat suit-
ability. Newly hatched fry remain in the nest for 7-8 days (Marzolf 1957) and
then disperse to shallow water areas with cover (Cross and Collins 1975). Fry
are commonly found aggregated near cover in protected, slow-flowing (velocity
< 15 cm/sec) areas of rocky riffles, debris-covered gravel, or sand bars in
clear streams (Davis 1959; Cross and Collins 1975), and in very shallow
(< 0.5 m) mud or sand substrate edges of flowing channels along turbid rivers
and bayous (Bryan et al. 1975). Dense aquatic vegetation generally does not
provide optimum cover because predation on fry by centrarchids is high under
these conditions, especially in clear water (Marzolf 1957; Cross and Collins
1975). Fry overwinter under boulders in riffles (Miller 1966) or move to
cover in deeper water (Cross and Collins 1975).
Juvenile. Optimal habitat for juveniles is assumed to be similar to that
for fry. The temperature range most suitable for juvenile growth 1s reported
to be 28-30° C (Andrews et al. 1972; Andrews and Stickney 1972). Upper lethal
temperatures are assumed to be similar to those for fry.
HABITAT SUITABILITY INDEX (HSI) MODELS
Model Applicability
Geographic area. The model is applicable throughout the 48 conterminous
States. The standard of comparison for each individual variable suitability
index is the optimum value of the variable that occurs anywhere within the 48
conterminous States. Therefore, the model will never provide an HSI of 1.0
when applied to water bodies in the Northern States where temperature-related
variables do not reach the optimum values for channel catfish found in the
Southern States.
Season. The model provides a rating for a water body based on its ability
to support a self-sustaining population of channel catfish through all seasons
of the year.
Cover types. The model is applicable in riverine, lacustrine, palustrine,
and estuarine habitats, as described by Cowardin et al. (1979).
Minimum habitat area. Minimum habitat area is defined as the minimum
area of contiguous suitable habitat that is required for a species to succes-
fully live and reproduce. No attempt has been made to establish a minimum
-------
habitat size for channel catfish, although this species is most abundant in
larger water bodies.
V.e_ni.t!c/tJ'°.lJ_eye-!- The acceptable output of these models is an index
between 0 and 1 which the authors believe has a positive relationship to
carrying capacity. In order to verify that the model output was acceptable,
sample data sets were developed for calculating HSI's from the models..
verification are
" s.
The sample data sets and their relationship to model verif
discussed in greater detail following the presentation of the model
Model Description
It is assumed that channel catfish habitat quality is based primarily on
their food, cover, water quality, and reproduction requirements. Variables
that have been shown to have an impact on the growth, survival, distribution,
abundance, or other measure of well-being of channel catfish are placed in the
appropriate component and a component rating derived from the individual
variable suitability he important for rating
the food component because if cover is available, fish would be more likely to
occupy an area and utilize the food resources. Substrate (V*) is included
because stream production potential of aquatic insects (consumed directly by
both cnannel catfish and their prey species) is'related to amount and type of
substrate.
Cover component. Percent pools (Vj) is included because channel catfish
utilize pools as cover. Percent cover (V2) is an Index of all types of
objects, Including logs and debris, used for cover in rivers. Average current
velocity in cover areas (Vi,) is important because the usable habitat near a
cover object decreases if cover objects are surrounded by high velocities.
Water quality component. The water quality component is limited to
temperature, oxygen, turbidity, and salinity measurements. These parameters
have been shown to effect growth or survival, or have been correlated with
changes in standing crop. Variables related to temperature, oxygen, and
salinity are assumed to be limiting when they approach lethal levels. Toxic
substances are not considered.
-------
Habitat Variables
% cover (V2)
Substrate type (V\)
Life Requisites
Food (CF),
% pools
% cover (V2)
Average current velocity (Vlt)
Cover (Cc).
Temperature (adult) (Vs)
Temperature (fry) (V12)
Temperature (juvenile)
Dissolved oxygen (V,)
Turbidity (V7)
Salinity (adult) (V9)
Salinity (fry, juvenile)
/
Length of agricultural growing season (Vt)
Water quality
% pools
% cover (Va)
Dissolved oxygen (V,)
Temperature (embryo) (V10)
Salinity (embryo) (V
Reproduction (CR)'
Figure 1. Tree diagram illustrating relationship of habitat variables
and life requisites in the riverine model for the channel catfish.
Dashed lines indicate optional variables in the model.
-------
Habitat Variables
Life Requisites
% cover (V2)
% littoral area (V3) -
Total dissolved solids (Vj6)
Food (CF),
% cover (Va)
littoral area (V,)
Cover (Cc),
Temperature (adult) (Vs)
Temperature (fry) (V12)
Temperature (juvenile) (Vifc)
Dissolved oxygen (Vt)
Turbidity (V,)
Salinity (adult) (V,)
Salinity (fry, juvenile) (V13)
Length of agricultural growing season (Vt)
Water quality (CWQ)
% cover (V,)
Dissolved oxygen (V,)
Temperature (embryo) (Vi0)
Salinity (embryo)
Reproduction (CR)'
Storage ratio (V1S)
Flushing rate (Vi7)
Other (CQT)
Figure 2. Tree diagram illustrating relationship of habitat variables
and life requisites in the lacustrine model for the channel catfish.
Dashed lines indicate optional variables in the model.
-------
Reproduction component. Percent pools (Vj) is in the reproductive compo-
nent because channel catfish spawn in low velocity areas in rivers. Percent
cover (V2) is in this component since channel catfish require cover for
spawning. If minimum dissolved oxygen (DO) levels within pools and backwaters
during midsummer (V,) are adequate, they should be adequate during spawning,
which occurs earlier in the year. Db levels measured during spawning and
embryo development could be substituted for V,. Two additional variables,
average water temperatures within pools and backwaters during spawning and
embryo development (V10) and maximum salinity during spawning and embryo
development (VM) are included' because these water quality conditions affect
embryo survival and development.
Mode_1_ Description - Lacustrine
Food component. Percent cover (V2) is included since it is assumed that
if cover is available, channel catfish would be more likely to utilize an area
for feeding. Percent littoral area (Vj) is included because littoral areas
generally produce the greatest amount of food and feeding habitat for catfish.
Total dissolved solids (TDS) (Vn) is included because adult channel catfish
eat fish, and fish production in lakes and reservoirs is correlated with TDS.
Cover component. Percent cover (V2) is included since channel catfish
strongly seek structural features of logs, debris, brush, and other objects
for shelter. Percent littoral area (V3) is included because all life stage
predominantly utilize cover found in littoral areas of a lake.
Water quality component. Refer to riverine model description.
Reproduction component. Percent cover (V2) is included since catfish
build nests in dark and secluded areas; spawning is not observed if suitable
cover is unavailable. Percent littoral area (V,) is included since catfish
spawning is concentrated along the shoreline. DO (V,), temperature (Vxo) and
salinity (Vn) are included because these water quality parameters affect
embryo survival and development.
Other component. For reservoirs, storage ratio (V1S) and maximum flushing
rate when fry are present (V17) are included in this component because storage
ratio may affect standing crop and the flushing of fry from a reservoir outlet
can reduce the abundance of fry.
-------
Suitability Index (SI) Graphs for Model Variables
This section contains suitability index graphs for the 18 variables
described above, and equations for combining selected variables into a species
HS! using the component approach. Variables pertain to a riverine (R) habitat,
lacustrine (L) habitat, or both (R, L).
Habitat Variable
R (VJ
Percent pools during
average summer flow.
Suitability Graph
R,L (V2) Percent cover (logs,
boulders, cavities,
brush, debris, or
standing timber) during
summer within pools,
backwater areas, and
littoral areas.
X
01
00
0.4-
0.2-
0.0
10 20 30 40 50
-------
(V3) Percent littoral area
during summer.
1.0
0.0
25 50 75 100
(V,,) Food production potential
in river by substrate type
present during average
summer flow.
A) Rubble dominant in
riffle-runs with some
gravel and/or boulders
present; fines (silt
and sand) not common
aquatic vegetation
abundant (> 30%) in
pool areas.
B) Rubble, gravel,
boulders, and fines
occur in nearly equal
amounts in riffle-run
areas; aquatic vegeta-
tion is 10-30% in
pool areas.
C) Some rubble and gravel
present, but fines or
boulders are dominant;
aquatic vegetation is
scarce (< 10%) in pool
areas.
D) Fines or bedrock are
the dominant bottom
material. Little or
no aquatic vegetation
or rubble present.
1.0
X
OJ
TD
C
00
0.8 -
0.6 .
0.4 -
0.2 -
0.0
B C
Class
10
-------
R,L (V,) Average midsummer water
temperature within
pools, backwaters, or
littoral areas (Adult).
1.0
I 0.8
$ 0.6
•r—
3
0.2 -
0.0
10
20 30
°C
40
R,L (Vt) Length of agricultural
growing season (frost-
free days).
Note: This variable
is optional.
1.0
I 0.8
»—i
£ 0.6
•u
3
0.4
0.2
0.0
125
Days
250
R,L (V7) Maximum monthly average
turbidity during summer.
0.0
100
11
-------
R,L
(V.)
R,L
(V.)
Average minimum dissolved
oxygen levels within
pools, backwaters, or
littoral areas during
midsummer.
1.0
Maximum salinity
during summer
(Adult).
00
0.8 -
0.6 -
3 0.2 J
0.0
1.0
I 0.8 -J
c
$ 0.6 .,
•t~
1 0.4
•M
•I™
^ 0.2 „
0.0
4 6
mg/1
10
ppt
R,L (V10) Average water
temperatures within
pools, backwaters,
and littoral areas
during spawning and
embryo development
(Embryo).
•o
1.0
0.8 .
0.6 -
0.4 4
0.2 H
0.0
10
20
°C
12
-------
R,L
Maximum salinity
auring spawning
and embryo development
(Embryo).
1.0
0.0
10
ppt
20
R,L (Viz) Average midsummer water
temperature within pools, x
backwaters, or littoral ^ Q.8 -
areas (Fry).
oo
40
R.,L
Maximum salinity
during summer
(Fry, Juvenile).
5 6 78 9 10
ppt
13
-------
/Wf«j'ji: ifii'J-,umnn:r w.jtL
temperature within
pools, backwaters, or
littoral areas
(Juvenile).
X
o>
•? 0.8J
>»
£ 0.6 -
•r-
3 0.4 -
" 0.2 _
0.0
10 20 30 40
°C
Storage ratio.
Monthly average IDS
(total dissolved
solids) during
summer.
1000
14
-------
(VJ7) Maximum reservoir
flushing rate while
fry present (Fry).
(V,,)
Riverine Model
Average current velocity
in cover areas during
average summer flow.
X
01
T3
C
1.0
0.8
0.6
0.4
0.2
0.0
10 20 30 40
cm/ sec
50
These equations utilize the life requisite approach and consist of four
components: food, cover, water quality, and reproduction.
Food (Cp).
CF =
V2 + V.
15
-------
Cover (Cc).
Cc = (Vx x V2 x V,,)173
Water Qua! ity (C./n).
CWQ
2(V6 * V12 + Vlh)
-- 3 + V7 + 2(V.) + V, * V13
If Vs, V12, V,,, V,, V,, or VM is < 0.4, then CWQ equals the lowest
of the following: Vs, V12, Vllt, V,, V,, Vj,, or the above equation.
Note: If temperature data are unavailable, 2(V6) (length of agricul-
tural growing season) may be substituted for the term
2(V5 + V12 + V.J
i in the above equation
Reproduction (CR).
CR = (V, x V,1 x V,2 x V1CJ x Vxl)1/8
If V,, Vjo, or.Vu is < 0.4, then CR equals the lowest of the
following: Mt , V10, Vu, or the above equation.
HSI determination .
21/6
HSI = (CF x Cc x CWQ2 x CR2) , or
WQ
If Cwo or CR is < 0.4, then the HSI equals the lowest of the
, following: CWQ, CR, or the above equation.
16
-------
Sources of data and assumptions made in developing the suitability Indices
are presented in Table 1.
Sample data sets using riverine HSI model are listed in Table 2.
Lacustrine Model
This model utilizes the life requisite approach and consists of five
components: food, cover, water quality, reproduction, and other.
Food (CF).
V, + V, + V14
CF = 3
Cover (C-).
Cc = (V, x V,)1/2
Water Quality (CWQ).
C * same as in Riverine HSI Model
Reproduction (CR).
CR = (V x V, x V,2 x V102 x VM)1/8
If V,, V,,, or V,, is < 0.4, then CR equals the lowest of the
following: Vlt V,,, Vn, or the above equation.
Other (CQT).
V17
17
-------
Table 1. Data sources and assumptions for channel catfish suitability indices.
Variable and source
Assumption
Vs Bailey and Harrison 1948
Bailey and Harrison 1948
Marzolf 1957
Cross and Collins 1975
Bailey and Harrison 1948
Marzolf 1957
Cross and Collins 1975
Vk Bailey and Harrison 1948
Vj Clemens and Sneed 1957
West 1966
Shrable et al. 1969
Starostka and Nelson 1974
Biesinger et al. 1979
V, Jenkins 1970
V7 Finnell and Jenkins 1954
Buck 1956
Marzolf 1957
V, Moss and Scott 1961
Andrews et al. 1973
Carlson et al. 1974
Randolph and Clemens 1976
V, Perry and Avault 1968
Perry 1973
Optimum conditions for a diversity of
velocities, depths, and structural
features for channel catfish will be
found when there are approximately equal
amounts of pools and riffles.
The strong preference of all life stages
of channel catfish for cover indicates
that some cover must be present for
optimum conditions to occur.
Lakes with small littoral area will pro-
vide less area for cover and food pro-
duction for channel catfish and are there-
fore less suitable.
The amount and type of substrate or the
amount of aquatic vegetation associated
with high production of aquatic insects
(used as food by channel catfish and
channel catfish prey species) is optimum.
Temperatures at the warmest time of year
must reach levels that permit growth in
order for habitat to be suitable. Optimum
temperatures are those when maximum growth
occurs.
Growing seasons that are correlated with
high standing crops are optimum.
High turbidity levels are associated with
reduced standing crops and therefore are
less suitable.
Lethal levels of dissolved oxygen are
unsuitable. DO levels that reduce feeding
are suboptimal.
Salinity levels where adults are most
abundant are optimum. Any salinity
level at which adults have been
reported has some suitabilty.
18
-------
Table 1. (concluded)
Variable and source
Assumption
Vlo Brown 1942
Clemens and Sneed 1957
Perry and Avault 1968
Perry 1973
'12
V,:
v,,
'It
McCammon and LaFaunce 1961
Moss and Scott 1961
Macklin and Soule 1964
West 1966
Allen and Strawn 1968
Andrews 1972
Starostka and Nelson 1974
Allen and Avault 1970
Andrews et al. 1972
Andrews and Stickney 1972
Jenkins 1976
V,, Jenkins 1976
V17 Walburg 1971
Miller 1966
Scott and Grossman 1973
Cross and Collins 1975
Optimum temperatures are those which
result in optimum growth. Temperatures
that result in death or no growth are
unsuitable.
Salinity levels at which spawning has
been observed are suitable.
Optimum temperatures for fry are those
when growth 1s best. Temperatures that
result in no growth or death are unsuit-
able.
Salinities that do not reduce growth
of fry and juveniles are optimum.
Salinities that greatly reduce growth
are unsuitable.
Temperatures at which growth of juveniles
is best are optimum. Temperatures that
result in no growth or death are unsuit-
able.
Storage ratios correlated with maximum
standing crops are optimum; those cor-
related with lower standing crops are
suboptimum.
Total dissolved solids (TDS) levels cor-
related with high standing crops of warm-
water fish are optimum; those correlated
with lower standing crops are suboptimum.
The data used to develop this graph are
primarily from southeastern reservoirs.
Flushing rates correlated with reduced
levels of fry abundance are suboptimal.
High velocities near cover objects will
decrease the amount of usable habitat
around the objects and are thus
considered suboptimum.
19
-------
Table 2. Sample data sets using riverine HSI model.
Variable
% pools Vt
% cover V2
Substrate for V,,
food production
Temperature-Adult
(° C) V,
Growing season V8
Turbidity (-ppm) V7
Dissolved oxygen
(mg/1) V,
Sal im'ty-adul t
(ppt) V,
Temperature- Embryos
(°C) VIQ
Sal inity- Embryo
(ppt) VM
Temperature-Fry
(° C) V12
Sal ini ty-Fry/
Juvenile (ppt) Vaj
Temperature-
Juvenile (° C) Vllt
Velocity Vlt
Data set 1 Data set 2
Data SI Data SI
60 1.0 90 0.6
50 1.0 10 0.4
silt- 0.7 silt- 0.5
gravel sand
28 - 1.0 32 0.4
180 0.8
50 1.0 210 0.5
4.5 0.6 4.0 0.5
< 1 1.0 < 1 1.0
25 0.8 21.5 0.5
< 1 1.0 < 1 1.0
26.5 0.8 32 0.7
< 1 1.0 < 1 1.0
29 1.0 32 0.7
15 1.0 5 1.0
Data
Data
15
5
sand
22
-
160
4.0
< 1
28.5
< 1
23
< 1
22
30
set 3
SI
0.5
0.2
0.2
0.5
-
0.8
0.5
1.0
0.5
1.0
0.5
1.0
0.5
0.3
20
-------
Table 2. (concluded) -i
Variable
Component SI
CF =
CC =
CWQ =
CR =
HSI =
Data set 1
Data SI
0.85
1.00
0.87
0.86
0.88
Data set 2
Data SI
0.45
0.62
0.40*
0.58
0.40*
Data set 3
Data SI
0.20
0.31
0.69
0.47
0.43
*Note: CWQ < 0.4; therefore, HSI = CWQ in Data Set 2.
21
-------
HSI determination.
HSI = (CF x Cc x CWQ> x CR2 x CQT)1/7 , or
If CWQ or CR is < 0.4, then the HSI equals the lowest of the
following: CWQ, Cn, or the above equation.
Sample data sets using lacustrine HSI model are listed in Table 3.
Interpreting Model Outputs
The proper interpretation of the HSI produced by the models is one of
comparison. If two water bodies have large differences in HSI's, then the one
with the higher HSI should be able to support more catfish than the water body
with the lower HSI, given that the model assumptions have not been violated.
The actual differences in HSI that indicate a true difference in carrying
capacity are unknown .and likely to be high. We have aggregated a large number
of variables into a single index with little or no quantitative information on
how the variables interact to effect carrying capacity. The probability that
we have made an error in our assumptions on variable interactions is high.
However, we believe the model is a reasonable hypothesis of how the selected
variables interact to determine carrying capacity.
Before using the model, any available statistical models, such as those
described under model 3 in the next section, should be examined to determine
if they better meet the goals of model application. Statistical models are
likely to be more accurate in predicting the value of a dependent variable,
such as standing crop, from habitat related variables than the HSI models
described above. A statistical model is especially useful when the habitat
variables in the data set used to derive the model have values similar to the
proposed model application site. The HSI models described above may be most
useful when habitat conditions are dissimilar to the statistical model data
set or it is important to evaluate changes in variables not included in the
statistical model.
The sample data sets consist of different variable values (and their
corresponding SI score), which although not actual field measurements, are
thought to represent realistic conditions that could occur in various channel
catfish riverine or lacustrine habitats. We believe the HSI's calculated from
the data reflect what carrying capacity trends would be in riverine or lacus-
trine habitats with the characteristics listed in the respective data sets.
22
-------
Table 3. Sample data sets using lacustrine HSI model.
Variable
% cover V2
% littoral area V3
Temperature-Adult
(° C) V$
Growing season Vt
Turbidity V7
Dissolved oxygen V,
Salinity-Adult
(ppt) V,
Temperature-Embryo
(° C) V10
Salinity- Embryo
(ppt) VM
Temperature-Fry
(° C) Vlt
Salinity-Fry/
Juvenile (ppt) Vj,
Temperature-
Juvefiile (° C) V,»
Storage ratio VJS
TDS (ppm) V,.
Data
Data
50
40
26
180
175
4.5
< 1
25
< 1
26.5
< 1
29
1.5
200
set 1
SI
1.0
1.0
1.0
0.8
0.7
0.6
1.0
0.8
1.0
0.8
1.0
1.0
0.9
1.0
Data
Data
10
20
20
-
210
4.5
< 1
21.5
< 1
32
< 1
32
.3
300
set 2
SI
0.4
0.7
0.3
-
0.5
0.6
1.0
0.5
1.0
0.7
1.0
0.7
0.7
1.0
Data
Data
5
70
33
-
250
2.5
< 1
28
< 1
23
< 1
22
0.8
600
set 3
SI
0.2
0.6
0.2
-
0.3
0.2
1.0
0.5
1.0
0.5
1.0
0.5
1.0
0.6
Flushing rate
while fry
present (days) V17 15 1.0 4 0.4 11 1.0
23
-------
Table 3. (concluded)
Variable
Component SI
CF '
CC '
cwq =
CR =
Cnr =
OT
HSI =
Data set 1
Data SI
1.00
1.00
0.82
0.83
0.95
0.89
Data set 2
Data SI
0.70 '
0.52
0.30*
0.56
0.55
0.30*
Data set 3
Data SI
0.47
0.33
0.20*
0.20
1.00
0.20*
*Note: CWQ < 0.4; therefore, HS! = CWQ in Data Sets 2 and 3.
ADDITIONAL HABITAT MODELS
Model 1
Optimal riverine habitat for channel catfish is characterized by the
following conditions, assuming water quality is adequate: warm, stable water
temperatures (summer temperatures of 25-31° C); an approximate 40-60% area of
deep pools; and .abundant cover in the form of logs, boulders, cavities, and
debris (> 40% of pool area).
number of above criteria present
Hoi = •—•
24
-------
Model 2
Optimal lacustrine habitat for channel catfish is characterized by the
following conditions, assuming water quality is adequate: warm, stable water
temperatures (summer temperatures of 25-30° C); large surface area (> 500 ha);
moderate to high fertility (IDS 100-350 ppm); clear to moderate turbidities
(< 100 JTU); and abundant cover (> 40?o in areas < 5 m deep).
number of above criteria present
= 5
Model_J
Use the reservoir standing crop regression equations for catfishes pre-
sented by Aggus and Morais (1979) to predict standing crop, then divide the
predicted standing crop by the highest standing crop value used to develop the
regression equation, in order to obtain an HSI.
REFERENCES CITED
Aggus, L. R., and D. I. Morais. 1979. Habitat suitability index equations
for reservoirs based on standing crop of fish. Nat!. Reservoir Res.
Program. Rept. to U.S. Fish Wild!. Serv., Hab. Eval. Proj., Ft. Collins,
CO. 120 pp.
Allen, K. 0., and J. W. Avault. 1970. The effect of salinity on growth of
channel catfish. Proc. Southeastern Assoc. Game and Fish Commissioners
23:319-331.
Allen, K. 0., and K. Strawn. 1968. Heat tolerance of channel catfish,
Ictalurus punctatus. Proc. Southeastern Assoc. Game and Fish
Commissioners 21:399-411.
Andrews, J. W., and R. R. Stickney. 1972. Interactions of feeding rates and
environmental temperature on g.rowth, food conversion and body composition
of channel catfish. Trans. Am. Fish. Soc. 101(1):94-99.
Andrews, J. W., L. H. Knight, and T. Murai. 1972. Temperature requirements
for high density rearing of channel catfish from fingerlings to market
size. Prog. Fish-Cult. 34:240-242.
Andrews, J. W., T. Murai, and G. Gibbons. 1973. The influence of dissolved
oxygen on the growth of channel catfish. Trans. Am. Fish. Soc.
102(4):835-838.
25
-------
Bailey, R. M., and H. M. Harrison, Jr. 1948. Food habits of the southern
channel catfish (Icta_lury_s lacustris p_u_nc:tartu_s) in the Des Moines River,
Iowa. Trans. Am. Fish. Soc.'75:110-138.
Biesinger, K. E., R. B. Brown, C. R. Bernick, G. A. Flittner, and K. E. F.
Hokanson. 1979. A national compendium of freshwater fish and water
temperature data. Vol. I. U.S. Environ. Protection Agency Rep., Environ.
Res. Lab., Duluth, Minn. 207 pp.
Brown, B. E., 1. Inman, and A. Jearld, Jr. 1970. Schooling and shelter
seeking tendencies in fingerling catfish behavior. Trans. Am. Fish. Soc.
99(3):540-S45.
Brown, L. 1942. Propagation of the spotted channel catfish (Ictalurus
lacustris punctatus). Trans. Kansas Acad. Sci . 45:311-314.
Bryan, C. F., F. M. Truesdale, and D. S. Sabins. 1975. A limnological survey
of the Atchafalaya Basin, annual report. Louisiana Coop. Fish. Res.
Unit, Baton Rouge. 203 pp.
Buck, H. D. 1956. Effects of turbidity on fish and fishing. Trans. N. Am.
Wildl. Conf. 21:249-261.
Carlander, K. C. 1969. Channel catfish. Pages 538-554 jjn Handbook of fresh-
water fishes of the United States and Canada, exclusive of the
Perciformes. Iowa State Univ. Press, Ames. 752 pp.
Carlson, A. R., R. E. Siefert, and L. J. Herman. 1974. Effects of lowered
dissolved oxygen concentrations on channel catfish (Ictalurus punctatus)
embryos and larvae. Trans. Am. Fish. Soc. 103(3):623-626.
Chen, T. H. 1976. Cage culture of channel catfish in a heated effluent from
a power plant, Thomas Hill reservoir. Ph.D. Dissertation, Univ. Missouri,
Columbia. 98 pp.
Clemens, H. P., and K. E. Sneed. 1957. Spawning behavior of channel catfish,
Ictalurus punctatus. U.S. Fish Wildl. Serv. Spec. Sci. Rep.-Fish. 219.
11 PP-
Cowardin,, L. M., V. Carter, F. C. Golet, and E. T. LaRoe. 1979. Classifica-
tion of wetlands and deepwater habitats of the United States. U.S.D.I.
Fish and Wildlife Service. FWS/OBS-79/31. 103 pp.
Cross, F. B., and J. T. Collins. 1975. Fishes in Kansas. Univ. Kansas Mus.
Nat. Hist. Publ. Educ. Ser. 3. 180 pp.
Davis, J. 1959. Management of channel catfish in Kansas. Univ. Kansas Mus.
Nat. Hist. Misc. Publ. 21. 56 pp.
Davis, J. T., and L. E. Posey, Jr. 1958. Length at maturity of channel
catfish (Icta/lurys .la.c_u_s_tjris) in Louisiana. Proc. Southeastern Assoc.
Game and Fish. Commissioners 21:72-74.
26
-------
Flnnell, J. C., and R. M. Jenkins. 1954. Growth of channel catfish in
Oklahoma waters: 1954 revision. Oklahoma Fish. Res. Lab., Rept. 41.
37 pp. (Cited in Miller 1966.)
Hynes, H. B. N. 1970. The ecology of running waters. Univ. Toronto Press,
Canada. 555 pp.
Jenkins, R. M. 1970. The influence of engineering deiiyn and operation and
other environmental factors on reservoir fis-her> resources. Water
Resources Bull. 6(1):110-119.
Jenkins, R. M. 1974. Reservoir management prognosis: migraines or miracles.
Proc. Southeastern Assoc. Game and Fish Commissioners 27:374-385.
Jenkins, R. M. 1976. Prediction of fish production in Oklahoma reservoirs on
the basis of environmental variables. Ann. Oklahoma Acad. Sci. 5:11-20.
Jenkins, R. M., and D. I. Morais. 1971. Reservoir sport fishing effort and
harvest in relation to environmental variables. Pages 371-384 in G. E.
Hall, ed. Reservoir fisheries and limnology. Am. Fish. Soc.Spec
Publ. 8.
Jester, D. B. 1971. Effects of commercial fishing, species introductions,
and drawdown control on fish populations in Elephant Butte Reservoir, New
Mexico. Pages 265-285 jn G. E. Hall, ed. Reservoir fisheries and
limnology. Am. Fish. Soc. Spec. Publ. 8.
Lawler, R. E. I960. Investiqations of the channel catfish of Utah Lake. Utah
State Dept. Fish Game. Inform. Bull. 60-8. 69 pp.
Leidy, G. R., and R. M. Jenkins. 1977. The development of fishery compart-
ments and population rate coefficients for use in reservoir ecosystem
modeling. Contract Report Y-77-1, prepared for Office, Chief of
Engineers, U.S. Army, Washington, D.C. 72 pp.
Macklin, R., and S. Soule. 1964. Feasibility of establishing a warmwater
fish hatchery. Calif. Fish .Game, Inland Fish. Admin. Rept. 64-14.
13 pp. (Cited in Miller 1966.)
Marzolf, R. C. 1957. The reproduction of channel catfish in Missouri ponds.
J. Wildl. Manage. 21(l):22-28.
McCall, T. C. 1977. Movement of channel catfish, Ictalurus punctatus. in
Cholla Lake, Arizona, as determined by ultrasonic tracking. Western
Assoc. Game Fish. Comm. 57:359-366.
McCammon, G. W. 1956. A tagging experiment with channel catfish (Ictalurus
punctatus) in the lower Colorado River. Calif. Fish Game 42(4):323-335.
McCammon, G. W., and D. A. LaFaunce. 1961. Mortality rates and movement in
the channel catfish population of the Sacramento Valley. Calif. Fish
Game 47(l):5-26.
27
-------
Killer, E. E. 1966. Channel catfish. Pages 440-463 _1_n A. Calhoun, ed.
Inland fisheries management. Calif. Fish Game Res. Agency, Sacramento.
546 pp.
Minckley, W. L. 1973. Fishes of Arizona. Arizona Fish Game Publ., Phoenix.
293 pp.
Moore, G. A. 1968. Vertebrates of the United States. McGraw-Hill, New York.
Moss, D. D., and D. C. Scott. 1961. Dissolved oxygen requirements of three
species of fish. Trans. Am. Fish. Soc. 90(4):377-393.
Perry, W. G. 1973. Notes on the spawning of blue and channel catfish in
brackish water ponds. Prog. Fish-Cult. 35(3):164-166.
Perry, W. G., and J. W. Avault. 1968. Preliminary experiments on the culture
of blue, channel, and white catfish in brackish water ponds. Proc.
Southeastern Assoc. Game and Fish Commissioners 22:396-406.
Pflieger, W. L. 1971. A distributional study of Missouri fishes. Univ.
Kansas Mus. Nat. Hist. Publ. 20(3) :225-570.
Pflieger, W. L. 1975. Fishes of Missouri. Missouri Dept. Conserv. Publ.,
Columbia. 343 pp.
Randolph, K. N., and H. P. Clemens. 1976. Some factors influencing the
feeding behavior of channel catfish in culture ponds. Trans. Am. Fish.
Soc. 105(6):718-724.
Ryder, R. A. 1965. A method for estimating the potential fish production of
north-temperate lakes. Trans. Am. Fish. Soc. 94(3):214-218.
Ryder, R. A., S. R. Kerr, K. H. Loftus, and H. A. Regier. 1974. The morpho-
edaphic index, a fish yield estimator - review and evaluation. J. Fish.
Res. Board Can. 31(5):663-688.
Scott, W. B., and E. J. Crossman. 1973. Freshwater fishes of Canada. Fish.
Res. Board Can. Bull. 184. 966 pp.
Shrable, J. B., 0. W. Tiemeier, and C. W. Deyoe. 1969. Effects of temperature
on rate of digestion by channel catfish. Prog. Fish-Cult. 31(3):131-138.
Sigler, W. F., and R. R. Miller.- 1963. Fishes of Utah. Utah Fish Game, Salt
Lake City. 203 pp.
Simco, D. A., and F. B. Cross. 1966. Factors affecting growth and production
of channel catfish, Ictalurus punctatus. Univ. Kansas Mus. Nat. Hist.
Publ. 17(4):191-256.
Starostka, V. J., and W. R. Nelson. 1974. Channel catfish in Lake Dane. U.S.
Fish Wild!. Serrv. Tech. Pap. 81. 13 pp.
28
-------
Stauffer, J. R., Jr., K. L. Dickson, J. Cairns, Jr., W. F. Calhoun, M. T.
Masnik, and R. H. Myers. 1975. Summer distribution of fish species in
the vicinity of a thermal discharge, New River, Virginia. Arch.
Hydrobiol. 76(3):287-301.
Trautman, M. B. 1957. Fishes of Ohio. Ohio State Univ. Press. 683 pp.
Walburg, C. H. 1971, Loss of young fish in reservoir discharge and year-class
survival, Lewis and Clark Lake, Missouri River. Pages 441-448 HI G. E.
Hall, ed. Reservoir fisheries and limnology. Am. Fish. Soc. Spec.
Publ. 8.
Walbirg, C. H. 1975. Food of young-of-year channel catfish in Lewis and
Clark Lake, a Missouri River reservoir. Am. Midi. Nat. 93(1):218-221.
Walden, H. T. 1964. Familiar freshwater fishes of America. Harper and Row,
New York. 324 pp.
West, B. W. 1966. Growth, food conversion, food consumption and survival at
various temperatures of the channel catfish, Ictalurus punctaUjs
(Raf inesque). M.S. Thesis. Univ. Arkansas, Fayettevil le. (Cited~ln
Shrable et al. 1969.)
ZiebeVl, C. 1973. Ultrasonic transmitters for tracking channel catfish.
Prog. Fish-Cult. 35(l):28-32.
-------
APPENDIX B-l. NATIONAL LIST OF OMNIVORE FISH SPECIES.
Common name
Gizzard shad
Threadfin shad
Central mudminnow
Eastern mudminnow
Mexican tetra
Longfin dace
Goldfish
Grass carp
Common carp
Si Tverjaw minnow
Alvord chub
Utah chub
Tui chub
Blue chub
Sonora chub
Yaqui chub
Speckled chub
Blotched chub
California roach
Virgin spinedace
Hardhead
Bluehead chub
Golden shiner
White shiner
Common shiner
Bigmouth shiner
Blacknose shiner
Spottail shiner
Swallowtail shiner
Sand shiner
Skygazer shiner
Mimic shiner
Black side dace
Northern redbelly dace
Southern redbelly dace
Bluntnose minnow
Fathead minnow
Blacknose dace
Speckled dace
Redside shiner
Creek chub
River carpsucker
Quill back
Highfin carpsucker
Utah sucker
Longnose sucker
Bluehead sucker
Owens sucker
Flannelmouth sucker
Largescale sucker
Sacramento sucker
Latin name
Dorosoma cepedianum
Dorosoma petenense
Umbra limi
Umbra pygmaea
Astyanax tetra
Agosia chrysogaster
Carassius auratus
Ctenopharyngodon idell a
Cyprinus carpio
Ericymba buccata
Gil a alvordensis
Gil a atravia
Gil a bicolor
Gil a coerulea
Gila ditaenia
Gil a purpurea
Hybopsis aestivalis
Hybops is ins ignis
Lavinia symmetricus
Lepidomeda mollispinis
Mylopharodon conocephalus
Nocomis leptocephalus
Notemigonus crysoleucas
Notropis albeolus
Notropis cornutus
Notropis dorsal is
Notropis heterolepis
Notropis hudsonius
Notropis procne
Notropis stramineus
Notropis uranoscopus
Notropis volucellus
Phoxinus cumberlandensis
Phoxinus eos
Phoxinus erythrogaster
Pimephales notatus
Pimephales promelas
Rhlnichthys atratulus
Rhinichthys osculus
Richardsonius balteatus
Semotilus atromaculatus
Carpi odes carpio
Carpiodes cyprinus
Carpiodes velifer
Catostomus ardens
Catostomus catostomus
Catostomus discobolus
Catostomus fumeiventris
Catostomus latipinnis
Catostomus macrocheilus
Catostomus occidental is
-------
Mountain sucker Catostomus platyrhyncus
Rio grande sucker Catostomus plebeius
Tahoe sucker Catostomus tahoensis
Blue sucker Cycleptus elongatus
Smallmouth buffalo Ictiobus bubalus
Black buffalo Ictiobus niger
Oriental weatherfish Misgurnus anguillicaudatus
Snail bullhead Ictalurus brunneus
Black bullhead Ictalurus melas
Yellow bullhead Ictalurus natalis
Flat bullhead Icalurus platycephalus
Channel catfish Ictalurus punctatus
Walking catfish Clarias batrachus
Chinese catfish Clarias fuscus
Desert pupfish Cyprinodon macularius
Sheepshead minnow Cyprinodon variegatus
Plains killifish Fundulus zebrinus
Porthole livebearer Poeciliopsis gracilis
Gil a topminnow Poeciliopsis occidental is
Pinfish Lagodon rhomboides
Black acara Cichlasoma bimaculatum
Rio grande perch Cichlasoma cyanoguttatum
Firemouth Cichlasoma meeki
Jewel fish Hemichromis bimaculatus
Mozambique tilapia Tilapia mossambica
Redbelly tilapia Tilapia zilli
Shiner perch Cymatogaster aggregate
-------
APPENDIX B-2. NATIONAL LIST OF TOP CARNIVORE FISH SPECIES.
Common name
Bull shark
Alligator gar
Spotted gar
Longnose gar
Florida gar
Shortnose gar
Bowfi n
Machete
Ladyfish
Tarpon
Skipjack herring
Hickory shad
Pink salmon
Chum salmon
Coho salmon
Sockeye salmon
Chinook salmon
Golden trout
Arizona trout
Cutthroat trout
Rainbow trout
Atlantic salmon
Brown trout
Arctic char
Bull trout
Brook trout
Dolly varden
Lake trout
Inconnu
Redfin pickerel
Grass pickerel
Northern pike
Muskellunge
Chain pickerel
Sacramento squawfish
Colorado squawfish
Northern squawfish
Umpqua squawfish
Flathead catfish
Burbot
Fat snook
Tarpon snook
Snook
White bass
Striped bass
Yellow bass
Rock bass
Roanoke bass
Redeye bass
Small mouth bass
Suwanee bass
Latin name
Carcharhinus leucas
Atractosteus spatula
Lepisosteus oculatus
Lepisosteus osseus
Lepisosteus platyrhincus
Lepisosteus platostomus
Ami a calva
Elops affinis
Elops saurus
Megalops atlanticus
Alosa chrysochloris
Alosa mediocris
Oncorhynchus gorbuscha
Oncorhynchus keta
Oncorhynchus kisutch
Oncorhynchus nerka
Oncorhynchus tshawytscha
Salmo aguabonita
Salmo apache
Salmo clarki
Salmo gairdneri
Salmo salar
Salmo trutta
Salvelinus alpinus
Salvelinus confluentus
Salvelinus fontinalis
Salvelinus malma
Salvelinus namaycush
Stenodus leucichthys
Esox americanus americanus
Esox americanus vermiculatus
Esox lucius
Esox masquinongy
Esox niger
Ptychocheilus grandis
Ptychocheilus lucius
Ptychocheilus oregonensis
Ptychocheilus umpquae
Pylodictis olivaris
Lota lota
Centropomus parallelus
Centropomus pectinatus
Centropomus undecimalis
Morone chrysops
Morone saxatilis
Morone mississippiensis
Ambloplites rupestris
Ambloplites cavifrons
Micropterus coosae
Micropterus dolomieui
Micropterus notius
-------
Spotted bass Micropterus punctulatus
Largemouth bass Micropterus salmoides
Guadalupe bass Micropterus treculi
White crappie Pomoxis annularis
Black crappie Pomoxis nigromaculatus
Yellow perch Perca flavescens
Sauger Stizostedion canadense
Walleye Stizostedion vitreum
Gray snapper Lutjanus griseus
Freshwater drum Aplodinotus grunniens
Spotted seatrout Cynoscion nebulosus
Red drum Sciaenops ocellatus
Goldeye Hiodon alosoides
White catfish Ictalurus catus
Blue catfish Ictalurus furcatus
Tucunare Cichla ocellaris
Snakehead Channa striata
-------
APPENDIX C. NATIONAL LIST OF INTOLERANT FISH SPECIES.
Common name
Cisco
Arctic cisco
Lake whitefish
Bloater
Kiyi
Bering cisco
Broad whitefish
Humpback whitefish
Shortnose cisco
Least cisco
Shortjaw cisco
Pink salmon
Chum salmon
Coho salmon
Sockeye salmon
Chinook salmon
Pygmy whitefish
Round whitefish
Mountain whitefish
Golden trout
Arizona trout
Cutthroat trout
Rainbow trout
Atlantic salmon
Brown trout
Arctic char
Bull trout
Brook trout
Dolly varden
Lake trout
Inconnu
Arctic grayling
Largescale stoneroller
Redside dace
Cut lips minnow
Bigeye chub
River chub
Pallid shiner
Pugnose shiner
Rosefin shiner
Bigeye shiner
Pugnose minnow
Whitetail shiner
Blackchin shiner
Blacknose shiner
Spottail shiner
Sailfin shiner
Tennessee shiner
Yellowfin shiner
Ozark minnow
Ozark shiner
Latin name
Coregonus artedii
Coregonus autumnal is
Coregonus clupeaformis
Coregonus hoyi
Coregonus kiyi
Coregonus laurettae
Coregonus nasus
Coregonus pidschian
Coregonus reighardi
Coregonus sardinella
Coregonus zenithicus
Oncorhynchus gorbuscha
Oncorhynchus keta
Oncorhynchus kisutch
Oncorhynchus nerka
Oncorhnchus tshawytscha
Prosopium coulteri
Prosopium cylindraceum
Prosopium williamsoni
Salmo aguabonita
Salmo apache
Salmo clarki
Salmo gairdneri
Salmo salar
Salmo trutta
Salvelinus alpinus
Salvelinus confluentus
Salvelinus fontinalis
Salvelinus malma
Salvelinus namaycush
Stenodus leucichthys
Thymallus arcticus
Campostoma oligolepis
Clinostomus elongatus
Exoglossum maxillingua
Hybobsis amblops
Nocomis micropogon
Notropis amnis
Notropis anogenus
Notropis ardens
Notropis boops
Noropis emiliae
Notropis galacturus
Notropis heterodon
Notropis heterolepis
Noropis hudsonius
Notropis hypselopterus
Notropis leuciodus
Notropis lutipinnis
Notropis nubilus
Notropis ozarcanus
-------
Silver shiner
Duskystripe shiner
Rosyface shiner
Safron shiner
Flagfin shiner
Telescope shiner
Topeka shiner
Mimic shiner
Steelcolor shiner
Coosa shiner
Bleeding shiner
Bandfin shiner
Blackside dace
Northern redbelly dace
Southern redbelly dace
Blacknose dace
Pearl dace
Alabama nog sucker
Northern hog sucker
Roanoke hog sucker
Spotted sucker
Silver redhorse
River redhorse
Black jumprock
Gray redhorse
Black redhorse
Rustyside sucker
Greater jumprock
Blacktail redhorse
Torrent sucker
Striped jumprock
Greater redhorse
Ozark madtom
Elegant madtom
Mountain madtom
Slender madtom
Stonecat
Black madtom
Least madtom
Margined madtom
Speckled madtom
Brindled madtom
Frecklebelly madtorn
Brown madtom
Roanoke bass
Ozark rock bass
Rock bass
Longear sunfish
Darters
Darters
Darters
Sculpts
O'opu alamoo (goby)
O'opu nopili (goby)
O'opu nakea (goby)
Notropis photogenis
Notropis pilsbryi
Notropis rubellus
Notropis rubricroceus
Notropis signipinnis
Notropis telescopus
Notropis topeka
Notropis volucellus
Notropis whipplei
Notropis xaenocephalus
Notropis zonatus
Notropis zonistius
Phoxinus cumberlandensis
Phoxinus eos
Phoxinus erythrogaster
Rhinichthys atratulus
Semotilus margarita
Hypentelium etowanuni
Hypentelium nigricans
Hypentelium roanokense
Minytrema melanops
Moxostoma anisurum
Moxostoma carinatum
Moxostoma cervinum
Moxostoma congestum
Moxoatoma duquesnei
Moxostoma hamiltoni
Moxostoma lachneri
Moxostoma poecilurum
Moxostoma rhothoecum
Moxostoma rupiscartes
Moxostoma valenciennesi
Noturus albater
Noturus elegans
Noturus eleutherus
Noturus exilis
Noturus flavus
Noturus funebris
Noturus hildebrandi
Noturus insignis
Noturus leptacanthus
Noturus miurus
Noturus munitus
Noturus phaeus
Ambloplites cavifrons
Ambloplites constellatus
Ambloplites rupestris
Lepomis megalotis
Ammocrypta sp.
Etheostoma sp.
Percina sp.
Cottus sp.
Lentipes concolor
Sicydium stimpsoni
Awaous stamineus
*U.b. GOV:,UNMENT PUINTING OFFTCL:
!98j-0-<<27-65;t/271
-------