Interpreting Illinois Fish-IBI Scores,

DRAFT: January 2005.

Illinois Environmental Protection Agency

Bureau

of Water

Surface Water Section

lEPA ATTACHMENTNO. Jl-.

---

2

Definition of biotic integrity

Biotic; integrity is the capability of supporting and maintaining a balanced, integrated,

adaptive community of organisms having a species composition, diversity, and functional

organization comparable to that of natural habitat of the region (Frey 1975; Karr and

Dudley 1981).

The essence of an.IBI score

An IBI score represents how much the biotic integrity (in terms of fish metrics) differs

at a site from a benchmark set of biological conditions (in terms of the same fish metrics)

that reflect a known level of biotic integrity. For Illinois fish IBIs, we define these

benchmark conditions--often called "referenceconditions"--as the biological conditions

expected in Illinois streams least disturbed by human impacts. Therefore, the degree to

which an IBI score deviates from the score that best represents the typical reference

conditions reflects the relative amount of human impact (i.e., loss of integrity) additional

to that already represented by the reference conditions.

How many different levels of biotic integrity can be distinguished?

An Illinois fish-IBI score can range from 0 to 60 in increments of one unit; however, sixty

different levels of biotic integrity cannot be distinguished. Earlier versions of Illinois fish

IBIs recognized five distinct integrity classes, each defined by a subrange of the

maximum possible range (12 - 60) of ISI scor.es (Karr et al. 1986; Table 1).

---

3

Table 1. Integrity classes, each represented by a subrange of Illinois fish-IBI

scores (from Karr eta!. 1986).

IBI

Integrity

Attributes

Score

Class

58 -: 60

Excellent

Comparable to the best situations without human disturbance; all regionally

expected species

for the habitat and stream size, including the most intolerant

forms,

are present with a full array of age (size) classes; balanced trophic

structure

48 - 52

Good

Species richness somewhat below expectation, especially due to the loss of the

most intolerant forms; some species are present with less than optimal abundances

or size distributions; trophic structure shows some signs of stress

40 - 44

Fair

Signs of additional deterioration include loss of intolerant forms, fewer species,

and highly skewed trophic

structure; older age classes of top predators may be

rare.

28 - 34

Poor

Dominated by omnivores, tolerant forms, and habitat generalists; few top

carnivores; growth rateS and condition factors

commonly depressed; hybrids and

diseased fish often present.

12 - 22

Very Poor

Few fish present, mostly introduced or tolerant forms; hybrids common; disease,

parasites, fin damage, and

other anomalies reqular.

The ability to distinguish among various levels of biotic integrity can be determined

objectively

by examining the precision of an IBI. For example, for an IBI that has a

maximum possible scoring range from 0 to 100 and a precision of 10/'0 (i.e.,:!: 5 points), one

could reasonably

expect to be able to distinguish among ten integrity classes. Several

studies have shown

that the among-year and within-year precision of fish IBIs can range

from about

5/'0 to 30/'0 (Steedman 1988; Fore et al. 1994: based on mean of three IBI

scores per site; Mebane et al. 2003), with a central tendency of about 1O/'0 (Lyons 1992;

Hughes

et al. 1998) to 15/'0 (Karr et al. 1987; Yoder and Rankin 1995).

We have

only preliminarily determined the precision of the new Illinois fish IBIs. Ideally,

determining

the precision of the Illinois fish IBIs requires repeated sampling (through

time and across space)

at a set of locations that reflect the entire range of possible IBI

scores and stream-sampling situations throughout Illinois. Additionally, to determine how

---

4

much IBI-score variability is attributable to factors other than human impact, the

repeated samples collected per location must occur over a period of time or interval of

space in which no meaningful change in the level of human impact .has occurred. Variability

in IBI scores attributable to among-year variation could be quantified by sampling each

site in at least two consecutive years. Variability in IBI scores attributable to within-year

variation could

be quantified by sampling each site at least twice in the same sampling

season,

but separated enough in time to limit the effect of the first sampling bout on

subsequent ones. Preliminary information from a limited set of repeatedly sampled Illinois

stream sites indicates about

171'0

precision (i.e.,

!

5 points) in the new fish IBIs (Holtrop

and

Dolan 2003). Based on this preliminary estimate-which is consistent with the results

of aforementioned studies-any Illinois IBI-score difference of ten or less should not be

interpreted as a meaningful difference in biotic integrity. Consequently, using an Illinois

fish

IBI, one can expect to be able to distinguish among, at most, six integrity classes.

Given that more analysis is needed to examine the variability of Illinois fish IBIs and to

test the ability to distinguish consistently among biotic-integrity classes, we

conservatively suggest five preliminary integrity classes.

integrity classes?

No definitive rules exist for dividing an IBI-score range into subranges that each

represents an integrity class. Illinois fish IBIs were originally divided into five uneven

classes with scoring gaps between each class (Table

1; Karr et al. 1986); however, Karr et

---

5

al. (1986) did not provide explicit rationales for establishing these scoring subranges. To

define integrity classes, some IBIs simply have been divided into equal subranges. For

example, Barbour

et al. (1996) divided into fourths the entire possible range of scores for

a multimetric macroinvertebrate index developed for Florida streams. The upper fourth

of the scoring range was considered "Very Good" integrity, the lower fourth was

considered "Very Poor", and the middle was divided into "Good" and "Poor" classes.

More recently, Barbour

et al. (1999) recommended, "Because the metrics are normalized.

to reference conditions and expectations for the stream classes, any decision on

subdivision should reflect the distribution of the scores for the reference sites." Several

IBI developments are consistent with this recommendation; they define integrity classes--

each

represented by a corresponding scoring subrange--that are based on particular

percentile values

of the distribution of IBI scores in a set of reference-condition

biological samples. For interpreting a macro

invertebrate IBI in Wyoming streams, Jessup

and Stribling (2002) used the 25

th

percentile of IBI scores of reference samples as the

threshold between "Good" and "Fair" integrity classes. The IBI subrange above the 25

th

_

percentile threshold was divided into two equal parts to indicate "Good" and "Very Good"

integrityclasses. The subrange below the 25

th

percentile was divided into three equal.

parts to yield "Fair", "Poor", and "Very Poor" classes. For a fish IBI (Davis and Scott

2000) and a macro invertebrate IBI (Klemm et al. 2003) in mid-Atlantic highland streams,

developers also used the 25

th

percentile of reference-condition scores as the threshold

between

"Good" and" Fair" integrity classes. They used the subrange below the 5

th

---

6

percentile (Davis and Scott 2000) or the pt percentile (Klemm et al. 2003) of reference-

condition scores to define a third integrity class, "Poor". Ohio EPA (1988; Yoder and

Rankin 1995) used the 25

th

percentile of reference-condition scores as the threshold

between

"Good" and "Fair" integrity classes; scores above the 75

th

percentile represented

an "Exceptional" class.

We use a similar approach

for defining biotic-integrity classes based on Illinois fish-IBI

scores. We

use the approximate 25

th

and 75

th

percentile values of IBI scores of least-

disturbed samples

to guide the delimiting of integrity classes. We divide the range of IBI

scores below the 25

th

percentile (i.e., IBI=45) of least-disturbed samples equally into

three integrity classes. The range above the 25

th

percentile is divided into two integrity

classes

at another threshold slightly above the 75

th

percentile value of IBI scores (Figure

1).

[[insert Figure 1 ]]

Interpreting an Illinois fish-IBI score

Simply defining the scoring subrange of each biotic-integrity class does not provide an

interpretation of the meaning of each class. Using IBI scores to guide sound resource-

management, conservation, and regulatory decisions requires knowledge and understanding

of the environmental setting reflected by an IBI score, and ultimately, of the degree to

which the reference conditions--on which all IBI scores are based--reflect an absolute

level

of biotic integrity. Because an Illinois IBI score simply reflects the

relative

amount

of deviation from the level of integrity expected in Illinois environmental settings least

---

7

disturbed by human impact, understanding the meaning of the IBI score that best

represents Illinois reference (i.e., least-disturbed) conditions provides the necessary basis

for interpreting all other scores.

Based

on the infroamtion used in this study, an Illinois fish-IBI score of 50 represents

the level of integrity that best reflects least-disturbed conditions in Illinois; 50 is the

median score for the set of least-disturbed samples used. However, this score does not

represent necessarily an exceptionally high, absolute level of biotic integrity. Rather, it

merely

represents the level of integrityoccuringin the chemical, physical, and biological

conditions considered as least-disturbed (by

human impacts) during the period in which the

fish samples used to develop the IBIs were collected: 1982 through 1998. Because no

undisturbed streams existed in Illinois during this period, these biological reference

conditions already reflect some human impact to the watersheds, streams, and fish

assemblages throughout

the state. Consequently, the degree to which an IBI score

deviates

below 50 reflects the relative amount of human impact

additional to

that already

represented

by the reference conditions.

Additional

to the need for understanding the absolute level of biotic integrity represented

by the reference conditions used to develop an IBI, valid interpretation of an IBI score

requires an understanding--albeit imperfect--of

the range of environmental conditions

that the index can possibly reflect.

---

8

If an Illinois-IBI score of 50 (i.e., reflecting the typical least-disturbed conditions) does

not necessarily reflect an exceptionally high, absolute level of biotic integrity, then where-

-along

the continuum of biotic integrity from O'Yo to 100'Yo--does an Illinois fish-IBI score

of 50 fit? We know of no quantitative, completely objective procedure to answer this

question, much like there is no definitive way to show how far a person is from. being 100'Yo

healthy or to determine the maximum allowable ambient concentration of cadmium that

ensures 100'Yo protection of aquatic life. Nonetheless, although we cannot perfectly

quantify biotic integrity from O'Yo to 100'Yo, we can use a quantitative measure of biotic

integrity (e.g., an IBI score) to distinguish consistently among various, less-extreme levels

of integrity.

We describe each integrity class in terms of the changes (Le., deviation from the

reference"conditions) in the fish metrics with increasing human impact (Table 2; Figures 5

and 6).

These changes are relative; providing limited information about the absolute level

of biotic integrity represented by an IBI score. To aid the interpretation of Illinois fish-

IBI scores, we also describe each integrity class in terms of direct measures of human

impact,

thus helping one understand where a relative measure of biotic integrity (an IBI

score) fits along an absolute scale of biotic integrity. The chemical and physical conditions

that help describe each integrity class are based on the limited set of watershed and site-

specific measures that we used to rate each fish sample for disturbance (Figures 2,3, and

4). The descriptions

in Table 2 do not address all of the possible ways that humans

impact

the integrity of streams and their watersheds (Karr and Dudley 1981; Karr et al.

---

9

1986; Yoder and Rankin 1998), but they do provide information integral to interpreting the

meaning of each of the five integrity classes. These disturbance measures serve as

explicit surrogates for the degree of naturalness (Angermeier 2000), a property on which

the meaning of biotic integrity directly depends (Frey 1975; Karr and Dudley 1981).

Unlike several IBI developments cited in this report, we do not label each integrity class

(Figure 1; Table 2) with a term that reflects quality (e.g., Very Good, Good, Poor). By

avoiding"such value-laden terms, we emphasize that biotic integrity is an inherent,

quantifiable property of the environment (Frey 1975; Karr and Dudley 1981; Steedman and

Haider 1993; Angermeier and Karr 1994; Karr and Chu 1999; Angermeier 2000; National

Research Council 2001). Intrinsically, a measure of biotic integrity reflects very little

about how humans value a particular stream resource; determining the biotic integrity of a

stream differs distinctly from determining the resource value (to humans) of that stream.

However, because humans commonly value higher biotic integrity more than lower

integrity-as reflected in the objectives of the Clean Water Act-an IBI provides one of

several possible ways to help distinguish a highly valued resource from one of lesser value.

Aquatic Life Use?

Since 1986 the Illinois Environmental Protection Agency has been using fish-IBI scores to

help assess attainment of Aquatic Life Use in Illinois streams (Illinois EPA 1986; 2002).

Most recently, two IBI-score thresholds (based on previous versions of Illinois fish IBIs)

---

10

have been used to help determine one of three possible attainment outcomes: full support,

partial support,

or nonsupport (Figure 7). An understanding of the Clean Water Act goals

and objectives relevant

to aquatic life .can guide the selection of IBI-score thresholds for

assessing Aquatic Life Use. In this context, "full support" of Aquatic Life Use reasonably

could be

interpreted as meeting the Clean Water Act's interim aquatic-life goal (PL 92-

500:

Water Pollution Control Act Amendments of 1972). This interim goal is to achieve

water quality that provides, wherever attainable, for the protection and propagation of

fish, shellfish, and wildlife (Clean Water Act Section 101(a)(2)). Karr and Dudley (1981)

point

out some critical distinctions between this interim goal and the Clean Water Act's

ultimate

objective to restore and maintain the "...chemical, physical, and biological

integrity

of the Nation's waters..." Implicit in this objective is the maintenance of, at

least, a moderate level of biotic integrity and ultimately a restoration to high levels.

Consequently, a

more-farsighted interpretation of "full support" of Aquatic'Life Use could

be the level at which (at least) moderate to high biotic integrity exists. However, because

the levels of absolute biotic integrity that are actually attainable in Illinois streams

remain mostly undcoumented, it seems reasonable for now to define Aquatic Life Use

attainment as meeting the interim aquatic-life goal. Specifically, we set the IBI-score

threshold that helps define "full support" (of Aquatic Life Use) at a level of absolute

biotic integrity

that reflects minimal attainment of the interim goal.

Recently,

USEPA and natural-resource specialists from various states and tribes have

begun

to address Aquatic Life Use attainment and attainability by defining various levels

---

11

of biological condition along a gradient associated with levels of human impact. To date,

qualitative descriptions of biological

structure and function have been adopted to define

various levels of

this "biological condition gradient". In turn, these descriptions have been

helpful

in interpreting attainment thresholds for the interim aquatic-life goal and the

"biological integrity" objective of the Clean Water Act (see

www.epa.gov/emap/html/pubs/docs/groupdocs/symposia/symp2002/Davies.pdf).

We use a

similar approach

to interpret the levels of absolute biotic integrity represented by each

Illinois

fish-IBI integrity class (Table 2).

Based

on the Illinois-IBI integrity classes defined and described in this report (Figure 1;

Table 2), we suggest IBI= 40 as a threshold to help distinguish "full support" of Aquatic

Life Use from lesser levels

of attainment. In the context of the Clean Water Act's

interim aquatic-life goal, this threshold seems reasonable;

an IBI score of 40 represents

the 10

th

percentile of scores of least-disturbed samples. This score also represents a

level

of biotic integrity well above the 75

th

percentile of IBI scores at sites representing

the most-disturbed conditions evidenced in the dataset used to develop the new fish IBIs.

Whereas, these most-disturbed conditions do not necessarily represent the worst possible

conditions

in Illinois streams, the. limited set of disturbance measures do indicate levels of

human impact that have been documented to adversely affect aquatic life, thus decreasing

biotic integrity.

---

12

We think that explicit definition and description of the biological, chemical, and physical

conditions expected

to occur at various levels of biotic integrity can help clarify,

standardize, and improve the reliability of some of the subjectivity necessarily involved

with using

IBI scores to help assess attainment of Aquatic Life Use. Subjectivity, guided

by knowledge

and experience, is necessary for interpreting inherently uncertain measures

of the biological, chemical, and physical environment-measures on which natural-resource

professionals rely

to make decisions (Gregory and Keeney 2002). This subjectivity is not

exclusive

to interpreting and using biological measures such as an IBI score. For example,

considerable uncertainty exists

in using quantitative physicochemical measures as decision

criteria for use attainment, including concentration thresholds used as water-quality

standards (Thurston

et al. 1979; Clements and Kiffney 1994; Barnett and O'Hagan 1997;

Chapman et al. 1998; Hall and Giddings 2000; National Research Council 2001). Although

assessments

of designated uses, such as Aquatic Life Use, can never be

100'10

reliable,

they are continually improved by collaborative and iterative development of biological

indicators

in the context of the chemical and physical conditions to which aquatic life

responds (National Research Council 2001). Therefore, although direct measures of

aquatic life (e.g., an Illinois fish-IBI score) offer a more reliable way to assess attainment

of the Clean Water Act's aquatic-life goals and objectives than do traditional

physicochemical comparisons, these biological measures provide only limited information

if

used alone. Successful interpetation and use of biological indicators requires

corresponding information

on the physical and chemical settings in which aquatic organisms

---

13

live; in this way, biological measures, such as an Illinois fish-IBI score, complement rather

than replace the utility of more-traditional physicochemical measures.

---

Literature Cited

Angermeier, PL 2000. The natural imperative for biological conservation. Conservation

Biology 14:373-381.

Angermeier,

P.

L.

and J. R. Karr. 1994. Biological integrity versus biological diversity as

policy directives. BioScience 44:690-697.

Barbour, M. T., J. Gerritsen, G.

E.

Griffith, R. Frydenborg, E. McCarron, J. S. White, and

M.

L.

Bastian. 1996. A framework for biological criteria for Florida streams using

benthic

macroinvertebrates. Journal of the North American Benthological Society

15:185-211.

Barbour, M.T.,

J. Gerritsen, B. D. Snyder, and J. B. Stribling. 1999. Rapid bioassessment

protocols

for use in streams and wadeable rivers:

periphyt~n,

benthic

macro

invertebrates, and fish. Second Edition. EPA 841-B-99-002. USEPA, Office

of Water, Washington, DC.

Barnett,

V.

and

A.

O'Hagan. 1997. Setting environmental standards. The statistical

approach to handling uncertainty and variation. Chapman and Hall, London, UK.

Chapman, P. M.,

A.

Fairbrother, and D. Brown. 1998. A critical evaluation of safety

(uncertainty) factors for ecological risk assessment. Environmental Toxicology and

Chemistry 17:99-108.

Chessman,

B.C. and M. J. Royal. 2004. Bioassessment without reference sites: use of

environmental filters to predict natural assemblages of river macroinvertebrates.

Journal

of the North American Benthological Society 23:599-615.

Clements,

W. H. and P. K. Kiffney. 1994.. Assessing contaminant effects at higher levies of

biological organization. Environmental Toxicology and Chemistry 13:357-359.

Davis,

W. and J. Scott. 2000. Mid-Atlantic highlands stream assessment: technical

support document. EPA/903/B-00/004. USEPA Region 3, Office of Research and

Development,

Fort Meade, MD.

Fore,

L.

S., Karr, J. R., and Conquest,

L. L.

1994. Statistical properties of an Index of

Biological Integrity used to evaluate water resources. Canadian Journal of Fish and

Aquatic Sciences 51:1077-1087.

Frey,

D. G., 1975. Biological integrity of water-an historical approach. Pages 127-140

in

Ballentine, R. K. and

L.

J. Guarraia. coordinators. The integrity of water.

---

15

Proceedings of a symposium March 10-12,1975. U.S. Environmental Protection

Agency,

Office of Water and Hazardous Materials, Washington, DC.

Gregory, R. S. and R. L. Keeney. 2002. Making smarter environmental management

decisions. Journal

of the American Water Resources Association 38:1601-1612.

Hall,

L. H., Jr. and J. M. Giddings. 2000. The need for multiple lines of evidence for

predicting site-specific ecological effects. Human and Ecological Risk Assessment

6:679-710.

Holtrop,

A.M. and C.R. Dolan. 2003. Assessment of streams and watersheds in Illinois:

development

of a stream classification system and fish sampling protocols. Aquatic

Ecology Technical Report

03/15, Illinois Natural History Survey, Urbana, IL.

Hughes,

R.M. 1995. Defining acceptable biological status by comparing with reference

conditions. Pages 31-47

in

W.S. Davis and T.P. Simon. editors. Biological

assessment and criteria: tools for water resource planning and decision making.

Lewis Publishers, Boca Raton,

FL.

Hughes, R. M., P. R. Kaufman, A. T. Herlihy,

T.

M. Kincaid, L. Reynolds, and D. P. Larsen.

1998. A process

for developing and evaluating indices of fish assemblage integrity.

Canadian Journal

of Fish and Aquatic Sciences 55:1618-1631.

Illinois

Department of Natural Resources. 1996. Illinois land cover, an atlas. IDNR/EEA-

96/05, Illinois DNR, Springfield, IL.

Illinois Environmental Protection Agency. 1986. Illinois

water quality report 1984-1985.

IEPA/WPC/86-014. Springfield, IL.

Illinois Environmental Protection Agency. 2002. Illinois

water quality report 2002.

IEPA/BOW/02-006. Springfield, IL.

Jessup, B. K. and J. B. Stribling. 2002. Further evaluation of the Wyoming Stream

Integrity Index, considering quantitative and qualitative reference site criteria.

Report

to USEPA Region 8, Denver, CO by Tetra Tech, Owings Mills, MD.

Karr, J.R. and D. R. Dudley. 1981. Ecological perspective on water quality goals.

Environmental Management

5:55-68.

Karr, J. R., K. D. Fausch, P. L. Angermeier, P. R. Yant, and I. J. Schlosser. 1986. Assessing

biological integrity

in running waters: a method and its rationale. Illinois Natural

History Survey Special Publication 5, Urbana, IL.

---

16

Karr, J. R., Vant, P. R., Fausch, K. D., and Schlosser, I. J. 1987. Spatial and temporal

variability

of the Index of Biotic Integrity in three midwestern streams.

Transactions

of the American Fisheries Society 116:1-11.

Karr, J. R. and Chu,

E.

W. 1999. Restoring life in running waters: better biological

monitoring. Island Press, Washington,

DC.

Klemm, D.J. and ten others. 2003. Development and evaluation of a macroinvertebrate

biotic integrity index (MBII) for regionally assessing mid-Atlantic highland

streams. Environmental Management 31:656-669.

Lyons,

J. 1992. Using the index of biotic integrity (IBI) to measure environmental quality

in warmwater streams of Wisconsin. General Technical Report, NC-149, U.S.

Department of Agriculture, Forest Service, North Central Forest Experiment

Station, St. Pciul, MN.

Mebane, C. A.,

T.

R.Maret, and R. M. Hughes. 2003. An index of biological integrity (IBI)

for Pacific Northwest rivers. Transactions of the American Fisheries Society

132:239-261.

National Research Council. 2001. Assessing

the TMDL approach to water quality

management. National Academy Press, Washington,

DC.

.

Ohio Environmental Protection Agency. 1988. Biological

criteria for the protection of

aquatic life:

Volume II. Users manual for biologiccil field assessment of Ohio

surface waters. Ohio Environmental Protection Agency, Division of Water Quality

Monitoring and Assessment,

Surface Water Section, Columbus, OH.

Steedman, R. J. 1988. Modification and assessment of an Index of Biotic Integrity to

quantify stream quality in southern Ontario. Canadian Journal of Fish and Aquatic

Sciences 45: 492-501.

Steedman,

R. J. 1994. Ecosystem health as a management goal. Journal of the American

Benthological Society 13:605-610.

Steedman,

R. J. and W. Haider. 1993. Applying notions of ecological integrity. Pages 47-

61

in

S. Woodley, J. Kay, and G. Francis, editors, Ecological integrity and the

management of ecosystems. St. Lucie Press, Delray Beach, FL.

Suter, G. W. 1993. A critique of ecosystem health concepts and indexes. Environmental

Toxicology and Chemistry 12:1533-1539.

---

17

Suter, G. W. 1993. A critique of ecosystem health concepts and indexes. Environmental

Toxicology

and Chemistry 12:1533-1539.

Thurston,

R.

V.,

R. C. Russo, C. M. Fetterolf, Jr.,

T.

A. Edsall, and Y. M. Barber, Jr. editors.

1979. A review

of the EPA red book: Quality criteria for water. Water Quality

Section, American Fisheries Society, Bethesda, MD.

Wang,

L., J. Lyons, P. Kanehl, and R. Gatti. 1997. Influence of watershed land use on

habitat quality and biotic integrity in Wisconsin streams. Fisheries 22:6-12.

Wang,

L., J. Lyons, P. Kanehl, R. Bannerman, and E. Emmons. 2000. Watershed

urbanization

and changes in fish communities in southeastern Wisconsin streams.

Journal

of the American Water Resources Association 36:1173-1189.

Yoder,

C. O. and Rankin, E.

T.

1995. Biological criteria program development and

implementation in Ohio. Pages 109-144

in

W. S. Davis and T. P. Simon. editors.

Biological Assessment

and Criteria: Tools for Water Resource Planning and Decision

Making.

CRC Press, Inc., Boca Raton, FL.

Yoder, C. O. and E.

T.

Rankin. 1998. The role of biological indicators in a state water

quality management process. Environmental Monitoring

and Assessment 51:61-88.

---

60

55

50

45

40

W

~

35

o

30

0

en

25

--

CO 20

15

10

5

O.

§

0

Integrity Class 1

..........T......................o....................

.

o

8

Integrity Class

2

8

o

0

...........1.................

.................0.......................,.............

o

.

§

~ntegrity

,C/a!S

3

8

...........g..................=r...........

t

..........

1

...........H....

Integrity Class

4

8

o

.

Y

0

o

............................................................9

1

80.

.

Integrity Class

5

.

§

N= 164

128

64

60

LEAST

INTERMEDIATE MOST(Phys.) MOST(Chem.)

DISTURBANCE

Figure

1.

Boxplots of Ilinois fish-IBI scores observed in least-, intermediate, or most-disturbed conditions. "MOST(C em.)" represents

samples most "chemic

".

,

~l'Id

sedimel'lt physicochemical param"

.)" represents samples

most "physically" disturbed, based on various watershed measures and on site-scale measures of in-stream physical habitat. Samples rated

most-disturbed for both chemical and physical measures are excluded from these boxplots. Five classes of biotic integrity are indicated by

IBI~score

subranges. For each boxplot, the rectangle ("box") represents the range of values from the 25

th

to 75

th

percentiles; the central

half of the observed values occurs in this range. A horizontal line in each "box" indicates the 50th percentile. Vertical lines extending from

each box indicate a range from the 10

th

to 90

th

percentiles, and points beyond this range represent single observations.

---

IBI SUBRANGE

56-60

FOREST & 46-55

WETLAND 31-45

16-30

0-15

o-r==

m_-r

-,

I

crm---1

I

I

10.0 00

oco-i

I

I

Q)

0

<D

@------j

I

I

' 0

00

o

I

I

I

I

I

o

o

00 0

o

cjG-;0

c([}!n

00

&-m aD

0

0

l----~IOO

I

I

o

I

I

56-60

46-55

MINING 31-45

.

16-30

0-15

56-60

46-55

URBAN 31-45

16-30

0-15

f------l

0

<j})

0

<lDo 00

~

I

I

o

o

o

181

56-60: N= 11

46-55: N= 49

31-45: N= 66

16-30: N= 40

0-15: N= 10

56-60

46-55

AGRIC. 31-45

16-30

0-15

o I

I

r

u~.

Q)

000

CDI

I

I

f-----rn

0

o 0

0 000

I

I

I

om

00

o (DOl

I

I

~

0

0

0

o-l

I

,

I

0

0.00

0.25

0.50

0.75

1.00

Figure 2. Boxplots of selected watershed land-use/cover measures used to rate fish samples for level of human impact

(disturbance).

Each boxplot represents the distribution of disturbance-measure values for fish samples (count: N) in which the

IBI score is within each of five subranges (i.e., integrity classes; also see Figure 1). The GIS watershed polygons used for

analyses were created before this IBI project; we did not create a watershed for each IBI site; therefore, only fish samples at

sites located near the most-downstream point of each watershed are considered in these boxplots. Please see Figure 1 for

further explanation of the boxplots.

.

---

19

IBI SUBRANGE

o

N= 11

N= 49

N= 66

N= 40

N= 10

o

181

56-60:

46-55:

31-45:

16-30:

0-15:

o

o

o

o

I

~

rTI

I

~o

~

~aD

[}-m

~o

.~

D

()

I

ED

56-60

INDUS. or 46-5

OIL-EXTR.31-45

PT.SOURCE 16-30

0-15

56-60

SEWAGE 46'-55

PT.SOU RCE 31-45

16-30

0-15

0.00

0.05

0.10

0.15

0.20

NUMBER PER SQUARE KILOMETER

Figure 3. Boxplots of densities of selected point-source measures used to rate fish samples fqr level of human impact (disturbance) at

the watershed scale. Each boxplot represents the distribution of disturbance-measure values for fish samples (count= N) in which the

IBI score is within each of five subranges (i.e., integrity classes; also see Figure 1). Only fish samples at sites located near the most-

downstream point of each watershed are considered in these boxplots. "INDUS. or OIL EXTR... " are major industrial discharges,

industrial landfills, oil wells, or salt wells. "SEWAGE... " are major sewage discharges or sewage-waste landfills. To improve visual

resolution at lower values, the x-axis is truncated at a density near 0.20. Please see Figure 1 for further explanation of the boxplots.

---

20

:0

Ot---j

I

I

!

00000oo

I

I

I

I:

'00 000 00 00 '"

o

0'

I

I

__=J

'0 oo¢

N= 155

N=

82

N= 21

N= 141

N= 25

N= 141

N= 155

N=

82

N= 21

<;>

N= 25

o

o

o

o

0

o

o

o

00

0

Ii:

. I I 0

0'

I

:

:

0

I

:

I

0

I

H

I

I

I0

i

I

I

I000

i

I

I

'0

00 0

0

?

I

I 0

0

i

I

I

00

i

I

I-----<

0

0

ii, I I

00

c?

I

~

I

I

I

:

:"

;0

0>--1

I

I

:

¢ 000 I

I

I

I

'00000

00 0

0

00 I

: I

I

I: I

600

0 COQX><D

"

I

I :

aD OOCDO=

: CD

?

~I

I

I

I; 0 oro 0

0

:CD

<,

I

!

I

,¢, 00

0 00

0

i

f------1

I

I :

I 0

<}

~

I

':

I

:

I

:

0

?"

0 0

0

I

'I

:

0

60

,

I

:

'

I

I

06

0:

:

:

I

I

f-----i

:

(

I

II

i

I

IBI SUBRANGE

56-60

%

CHEM. 46-55

EX C E ED. 31 -45

IN WATE R 16-30

0-15

56-60

%

CH EM. 46-55

EX

C

E ED. 31 -45

IN SED. 16-30

0-15

56-60

%SUBSTR. 46-55

DIVERSITY 31-45

16-30

0-15

56-60

46-55

% FIN E S 3 1-45

16-30

0-15

56-60

46-55

%

RUN 31-45

16-30

0-15

o

25

50

75

100

Figure 4. Boxplots of selected measures used to rate each fish sample for level of human impact (disturbance) at the site, at the time of

the fish sample. Each boxplot represents the distribution of disturbance-measure values for fish samples (count=

N)in

which the IBI

score is within each of five subranges (Le., integrity classes; also see Figure 1). "%CHEM. EXCEED. IN WATER" and "'7'oCHEM. EXCEED.

IN SED." are the percent of a selected set of water or sediment parameters, respectively, at extreme levels (please see footnote in

Table 2 for further explanation). Number of fish samples in each integrity class is identical for "%SUBSTR. DIVERSITY", "'7'oFINES",and

"'7'oRUN",which are in-stream measures reflecting the composition of the stream bottom and the channel morphology. Please see Figure 1

for further explanation of the boxplots.

---

21

j------I

0

o

I

-

I

I

f------l

C---

0

I

~

I

0

o

0

,-

o

I

0

I

J----1

I

I

I

I

I

o

00-

I

0

I

I

I

I

I

II

II

I

I

I0

I0

oo

o

o

o

[------

I

o

0

I

II

I

1II

I

o

I

o

I

II

o

I

II

0

II

,

00

o

I

-1

o

I

0

I

I

1-;----

I

1-----

__

I

I

I

I

I

I

[----:-u_l

u

UI

I

I

III

I

o

o

o

o

I

I

I

I

I

-1

I

o

o

o

o

IBI SUBRANGE

56-60

FISH 46-55

SPECIES 31-45

16-30

0-15

56-60

MINNOW 46-55

SPECIES 31-45

16-30

0-15

56-60

SUCKER 46-55

SPECIES 31-45

16-30

0-15

56-60

SUNFISH 46-55

SPECIES 31-45

16-30

0-15

56-60

BENTHIC 46-55

INVERTVR. 31-45

SPECIES 16-30

0-15

56-60

INTOLRNT.46-55

SPECIES 31-45

16-30

0-15

o

1

-2

3

4

METRIC SCORE

5

6

Figure 5. Boxplots of scores of taxa-richness IBI metrics. Each boxplot represents the distribution of metric-score values of fish

samples

in which the IBI score is within each of five subranges (i.e., integrity classes; also see Figure 1). Twenty-five (N=25) fish samples

scored

in the highest-integrity class (IBI=56-60). For IBI=46-55, N= 141; for IBI=31-45, N=155; for IBI=16-30, N=82; and for IBI=O-

15, N=21.

---

o

0I

I

~

o

I II

I I

I

I

I

I000

I

0

IBI

SUBRANGE

56-60

PROP. 46-55

SPECLST. 31-45

BENTHIC 16-30

INVERlVRS. 0-15

o

o

I

I

22

56-60

PROP. 46-55

GENRLST. 31-45

FEEDERS 16-30

0-15

o

_u

~

o

CI

I

I

I

~

I

I

I

'

o

I

I

'0

f-----j

I

I

56-60

PROP. 46-55

MINERAL 31-45

SPAWNERS 16-30

0-15

56-60

PROP. 46-55

TOLERANT 31-45

SPECIES 16-30

0-15

I

II

o

0I

-

o

I I

[] I

I

I

II

I

0

I

I0

I

I

I

I

o

I

I

o

0

0

I------j

I

I

I

I

I

f-------J

0

o 1

234

METRIC SCORE

5

6

Figure 6. Boxplots of scores of proportional IBI metrics. Each boxplot represents the distribution of metric-score values of fish

samples

in which the IBI score is within each of five subranges (Le., integrity classes; also see Figure 1). Twenty-five (N=25) fish samples

scored

in the highest-integrity class: IBI=56-60. For IBI=46-55, N= 141; for IBI=31-45, N=155; for IBI=16-30, N=82; and for IBI=O-

15, N=21.

---

23

Table 2. Classes of biotic integrity for Illinois fish IBis. Descriptions are based on conditions depicted in Figures 2 -

6;

numbers of samples are provided in those figures. "Typical" conditions are defined by the interquartile range (Le., from the

25

th

to 75

th

percentile values) of disturbance-measure values observed nearest in time, as possible, to the collection of the

fish sample. Selected additional description addresses less-"typical" observations.

Subrange

56-60

Biotic-

Integrity

Class

1

Description of Typical Biological, Physical

,

2

Conditions

Biotic integrity is higher than that expected in Illinois streams that reflect the typical reference (Le., least-

disturbed) conditions. as currently defined. The number

of native fish species is greater than that in streams

reflecting the current, typical reference conditions primarily due to presence

of intolerant species. Reproductive

and trophic functional structure appear balanced. The typical physical and chemical conditions have been

evidenced as follows.

Watershed conditions include absence

of major sewage point sources and sewage-waste landfills. Industrial

point sources, oil-extraction point sources, or industrial landfills occur at

0--0.001

per square kilometer. Strip

mining (post

1949)

is absent (but

2

of

11

sites had

1.8%

and

3.8%).

Other watershed-scale impacts include

22--42%

agricultural and

0.3--0.7%

urban land use, in the stream corridor

(240

meters centered on stream

channel). From

13

to

49%

forest or wetland are present in the stream corridor.

At the site scale,

0--14%

of physicochemical parameters in water and

0--19%

of physicochemical parameters in

sediment are at extremes. Channel mor.phology consists of

41--74%

riffle or pool channel units, and

16--38%

of the stream bottom consists of particles smaller than fine gravel, except for streams with naturally higher

amounts

of sand or clay [[provide regional exceptions here?]]. Diversity of substrate particle-size classes is

2.18-3.62 (29-64%

of maximum observed; reciprocal of Simpson's D diversity index).

---

24

46 -55

2

Biotic integrity is similar to that expected in Illinois streams that reflect the typical reference conditions, as

currently defined. Relative to condtions in Integrity Class

1,

the number of native fish species is reduced

primarily due to loss

of some intolerant species. Reduced abundances of mineral-substrate spawners indicate

slight imbalance

in reproductive functional structure. Trophic fiunctional structure appears balanced. The

typical physical and chemical conditions have been evidenced as follows.

Watershed conditions include absence

of major sewage point sources and sewage-waste landfills. Industrial

point sources, oil-extraction point sources, or industrial landfills occur at

0--0.007

per square kilometer. Strip

mining (post

1949)

is absent (but

5

of

49

sites had

0.04--6.5%)

Other watershed-scale impacts include

39--

71 %

agricultural and

0.2-2.1 %

urban land use, in the strea"m corridor

(240

meters centered on stream

channel). From

6

to

31 %

forest or wetland are present in the stream corridor.

At the site scale,

0--14%

of physicochemical parameters in water and

0--19%

of physicochemical parameters in

sediment are at extremes. Channel morphology consists of

15--63%

riffle or pool channel units, and

15--46%

of the stream bottom consists of particles smaller than fine gravel, except for streams with naturally higher

amounts

of sand or clay. Diversity of substrate particle-size classes is

2.00--3.00 (24--49%

of maximum

observed; reciprocal of Simpson's D diversitv index).

31 - 45

3

Biotic integrity is lower than that expected in Illinois streams that reflect the typical reference conditions, as

currently defined. Number of native fish species is reduced from reference conditions primarily due to further

loss of intolerant species, but also due to loss of sucker species and benthic-invertivore species. Reduced

abundances of specialist benthic invertivores and increased abundances of generalist feeders indicate slight

to

moderate imbalance in trophic functional structure. Further reduction in abundances of mineral-substrate

spawners indicates moderate imbalance in reproductive functional structure. The typical physical and chemical

conditions have been evidenced as follows.

Watershed conditions include absence of major sewage point sources, sewage-waste landfills, industrial point

sources, oil-extraction point sources, and industrial landfills. Strip,mining (post

1949)

is absent (but

6

of

66

sites had

0.01--7.7%).

Other watershed-scale impacts include

36--70%

agricultural and

0.2--1.5%

urban land

use,

in the stream corridor

(240

meters centered on stream channel). From

5

to

34%

forest or wetland are

present

in the stream corridor.

At the site scale,

0--29%

of physicochemical parameters in water and

0--19%

of physicochemical parameters in

sediment are at extremes. Channel morphology consists of

6--56%

riffle or pool channel units, and

22--74%

of

the stream bottom consists

of particles smaller than fine gravel, except for streams with naturally higher

amounts of sand or clay. Diversity

of substrate particle-size classes is

1.57--2.77 (14--43%

of maximum

observed; reciprocal

of Simpson's D diversity index).

---

25

16 - 30

4

Biotic integrity is much lower than that expected in Illinois streams that reflect the typical reference conditions,

as currently defined. Number of native species is reduced further from reference condtions due to near-

complete loss of intolerant species and further pronounced loss

of sucker species and benthic-invertivore

species. Imbalance

of fish-community structure is evidenced as indiscriminate loss of species across major

families (minnows, suckers, sunfish). Further reductions

in abundances of specialist benthic invertivores and

mineral-substrate spawners indicate moderate to extreme imbalance

in trophic and reproductive functional

structure. The typical physical and chemical conditions have been evidenced as follows.

Watershed conditions include 0--0.002 major sewage point sources or sewage-waste landfills and 0--0.20

industrial point sources, oil-extraction point sources, or industrial landfills per square kilometer. Strip mining

(post 1949)

is absent (but 5 of 40 sites had 0.4--47%). Other watershed-scale impacts include 37--66%

agricultural and 0.2--3.6% urban land use,

in the stream corridor (240 meters centered on stream channel).

From 10 to 34% forest or wetland are present

in the stream corridor.

At the site scale, 0--25%

of physicochemical parameters in water and 0--43% of physicochemical parameters

in sediment are at extremes. Channel morphology consists of 2--44% riffle or pool channel units, and 24--73%

of the stream bottom consists of particles smaller than fine gravel, except for streams with naturally higher

amounts

of sand or clay. Diversity of substrate particle-size classes is 1.37--2.34 (9--33% of maximum

observed; reciprocal

of Simpson's D diversity index).

0-15

5

Biotic integrity is much lower than that expected in Illinois streams that reflect the typical reference conditions,

as currently defined. Number

of native species is reduced further due to pronounced, indiscriminate loss of

species across major families (minnows, suckers, sunfish) with a concurrent increase in the proportion of

tolerant species. Intolerant species are absent; benthic-invertivore species are nearly absent. Pronounced

reductions

in abundances of specialist benthic invertivores and mineral-substrate spawners indicate extreme

imbalance

in trophic and reproductive functional structure. The typical physical and chemical conditions have

been evidenced as follows.

Watershed conditions include 0--0.012 major sewage point sources or sewage-waste landfills and 0--0.010

industrial point sources, oil-extraction point sources, or industrial landfills per square kilometer. Strip mining

(post 1949)

is absent (but 2 of 10 sites had 5% and 15%). Other watershed-scale impacts include 1--25%

agricultural and 0--40% urban land use,

in the stream corridor (240 meters centered on stream channel). From

35 to 52% forest or wetland are present

in the stream corridor.

At the site scale, 29--57%

of physicochemical parameters in water and 19--81 % of physicochemical

parameters

in sediment are at extremes. Channel morphology consists of 5--52% riffle or pool channel units,

and 11--62%

of the stream bottom consists of particles smaller than fine gravel, except for streams with

naturally higher amounts

of sand or clay. Diversity of substrate particle-size classes is 1.30--2.95 (7--48% of

maximum observed; reciprocal of Simpson's D diversity index).

---

26

1 The watershed conditions of each fish sample (collected during 1982-1998) are represented by measures of land use/cover based on satellite

imagery of conditions in 1991-1995, supplemented by existing statewide spatial databases (Illinois Department of Natural Resources 1996). The

GIS watershed polygons used for analyses were created before this 181 project; we did not delineate a watershed for each 181 site. Thus, only fish

samples at sites located near the most-downstream point of each watershed were considered

in the statistical summaries of watershed-scale

measures. Few samples were available to depict watershed conditions for the highest (N=11) and lowest (N=10) integrity classes; therefore, for

these classes, the "typical" conditions described are merely best approximations.

"Physical" conditions at the site scale are represented by a single sample of in-stream physical habitat that was collected nearest

in time as

possible to each fish sample. All selected physical-habitat samples were collected during the "summer", base-flow period (approximately June

through mid-October)

2 Physicochemical conditions at sites are represented by the percent of parameters for which concentrations were extreme. For most parameters,

extreme concentrations were those higher than the 85

th

percentile

value

in samples co-collected nearest in time as possible with fish, throughout

the state during 1982 - 1998. For percent dissolved-oxygen saturation and pH,

values

outside the 15

th

_85

th

percentile range were considered

"extreme". The single water and single sediment sample collected nearest

in time as possible to the fish sample were selected to represent the

physicochemical conditions at the site at that time. All selected physicochemical samples were collected during the "summer" , base-flow period

(approximately June through mid-October). Water parameters were: total ammonia, nitrate/nitrite, total phosphorus, boron, percent dissolved-

oxygen saturation, pH, conductivity, and total suspended solids. Sediment parameters were: total phosphorus, Kjeldahl nitrogen, arsenic,

cadmium, chromium, copper, iron, lead, manganese, mercury, zinc, percent

volatile

solids, total polychlorinated biphenyls, total DDT, dieldrin, and

total chlordane.