Full text of "Critical trends assessment Program 2002 report"

See other formats

333.91009773
If62c
2002

CRITICAL TRENDS
ASSESSMENT PROGRAM

2002 Report

ILLINOIS

NATURAL
RESOURCES

CRITICAL TRENDS ASSESSMENT PROGRAM
2002 REPORT

Illinois Department of Natural Resources

Office of Realty and Environmental Planning
One Natural Resources Way
Springfield, Illinois 62702

Office of Scientific Research and Analysis
Natural History Survey Division
607 East Peabody Drive
Champaign, Illinois 61820

July 2003

200
Printed by the authority of the State of Illinois

pata et
OH ate ny

pea hernih ake, ale ST ale Ra
a . - Pest
Oa aed tf. Wee <

, haan, Sioe
ee ci Ce od a

yay ay
ogre?
et ees)

viawion /. Gaus Alpe
jarvikd YA

+ | OSB Loi akauld

1 . oF

Sh at ER) idl k a ti ly

Other CTAP Publications

Critical Trends in Illinois Ecosystems

Illinois Land Cover, An Atlas, plus CD-ROM

Inventory of Ecologically Resource-Rich Areas in Illinois

Illinois Geographic Information System, CD-ROM of digital geospatial data

Regional assessments (geological, water, biological, and socio-economic resources)
for the following areas or watersheds:

Big Muddy River Lower Des Plaines River
Cache River Lower Sangamon River
Calumet Area Lower Rock River
Chicago River/Lake Shore Mackinaw River
Driftless Area Prairie Parklands

Du Page River Sinkhole Plain
Embarras River Spoon River

Fox River Sugar-Pecatonica Rivers
Illinois Big Rivers Thorn Creek

Illinois Headwaters Upper Des Plaines River
Illinois River Bluffs Upper Rock River
Kankakee River Upper Sangamon River
Kaskaskia River Vermilion River

Kinkaid Area Vermilion River (Illinois River Basin)
Kishwaukee River

All CTAP documents are available from the DNR Clearinghouse at (217) 782-7498 or TTY (217) 782-9175.
Selected publications are also available on the World Wide Web at http://dnr.state.il.us/orep/ctap.

For more information about CTAP. call (217) 524-0500 or e-mail at ctap2@dnrmail.state.il.us.

Equal opportunity to participate in programs of the Illinois Department of Natural Resources (IDNR) and those funded
by the U.S. Fish and Wildlife Service and other agencies is available to all individuals regardless of race, sex, national
origin, disability, age, religion or other non-merit factors. If you believe you have been discriminated against, contact
the funding source's civil rights office and/or the Equal Employment Opportunity Officer, IDNR, One Natural Resources
Way, Springfield, Ill. 62702-1271; 217/785-0067; TTY 217/782-9175. This information may be provided in an
alternative format if required. Contact the DNR Clearinghouse at 217/782-7498 for assistance.

oa tilt. tail

igual. semen tal ve
ce Shane Naar @ ines yw lasgmi-o 1 O7G BRE AC Ay o 4) = aM

AGW obit | oa aA no side te ;

ge RIAAIYS <wrasdeman » Niobe
ee Sr i a
Of Taiko so noigeT ogo Ag 1
. pre re 16
se abv dom ana su MEE ROOO-LBH TE ATEA-S08 5. BK
a ateailain ihc ee eee

Table of Contents

MSM s CARS UI taal errata teem ae care en cet e erc time sears ode Meee accieas fouuhan etek stutowence ute nesaase gocSsueceautpeauseee ses ]
Creating a Report Card for RiverWatch Stream Quality: Multi-Metric Biological Score © ................00.. 3
Creating a Report Card for RiverWatch Stream Quality: Multi-Metric Habitat Score .................... 16
Aquatic Insects Report: Biological and Habitat Condition of Illinois Streams .................... 28
What Are the Discarded Sites of CTAP Terrestrial Monitoring Telling Us About Illinois Habitats? ................ 38
Ornithological Report: The Depauperate Nature of the Average Illinois Bird Community:

AY CTAPR Study Aromel99 7620012 aNs CAGES Sey, aT OE She. HOPES, Bes DORIS... 48
POresh Wate Pall SOO IP Spree ZOU]. e secseccc_ . sie baee seas desascce ra vectdeceouccte mincundedwassedanbecdannsseecteted 61
Pidinie Watch 200 and ® 200204 070.5, RA A SRI, RR AOE TI, PI FDL Ba seeeaes 73
Riverwatch Data Summary” Results for! 20020. <M. RADE Ts SAE ROI wanes sccone 78

Botanical Report: Floristic Quality Assessment (FQA) as a Measure of the Naturalness
of thesGrasslandsrand Wetlands of Illingishony... 00. APO. PRP. RA... 85

Terrestrial Insect Report: The Importance of Leafhoppers (Hemiptera:Cicadellidae)

collected by the Critical Trends Assessment Program ..........sccscssscssesssesssseseseeseeseeseseeseseeees 97 OF

iii

* a VA te i ae

SEARO PO Ree Ry eee Or te a ere eee)

ismesinance. SRO08 taideH stasis

- eneeve a

snest inte soos ning?

eannnye beaten oieriyrvinnesy ‘soos bas 00
" t

eT Tal Lia as te ee eee eee ees ero oes er eh

Introduction

The Critical Trends Assessment Program was established in 1992 to inform policymakers about the state of
Illinois’ ecosystems. During the decade since then, CTAP has developed tools and programs to systematically
monitor changes in ecological conditions in Illinois. These programs not only help policymakers determine
the best course of action to protect our natural resources, but also provide information to state and local land
managers and the public as stewards of Illinois lands and waterways.

Each year CTAP produces a report that provides the most current data and analysis on Illinois ecosystems.
This year efforts have focused on making the information more accessible and understandable to the public
and other users. Ten papers written by CTAP and EcoWatch scientists have been compiled. The CTAP
scientists have just completed their first five-year cycle of monitoring streams, forests, grasslands and
wetlands across the state. EcoWatch volunteers have completed another year of stream, forest, and prairie
monitoring.

The first two papers are part of an effort to make all of the data that RiverWatch volunteers collect intelligible.
Virtually every piece of data entered on a RiverWatch datasheet is boiled down into two numbers: a
biological score and a habitat score. The biological score combines five measures of biological quality —
MBI, EPT, taxa richness, dominance, and percent worms — into a single index. The habitat score
incorporates habitat-related measures shown to be correlated with biological quality, including stream
substrate, surrounding land uses, silt, water odor, water appearance, canopy cover, and channel disturbance.
No longer will volunteers have to guess what their results mean. Their stream report card will tell them.

In the third paper, similar efforts were made with CTAP stream data. Overall quality rankings for each site
were generated by combining EPT, HBI, and Habitat, three measures of stream quality. Rankings for the 149
monitoring sites varied significantly across the state with streams in the Shawnee Hills showing the best HBI
scores and the highest proportion of excellent streams. In general, streams with meandering channels scored
higher, supporting 40% more EPT taxa on average than channelized streams. Larger streams generally scored
higher as well because of their more diverse habitats.

The fourth paper shows that, surprisingly, the sites we do not monitor often tell us as much as those we do
monitor. In identifying monitoring sites, CTAP scientists often reject many sites before finding one suitable
for monitoring. Grassland sites are most frequently rejected and forest sites least often rejected. One and a
half forest sites are evaluated for each one that is found to be suitable for monitoring, while it takes three
wetland sites and seven grassland sites to find a suitable monitoring site. Some sites do not meet the
monitoring criteria; generally they are two small or too degraded. For example, grasslands are frequently
regularly mowed monocultures. However, nearly a third of wetland and grassland sites no longer exist as that
habitat even though they were identified by the Land Cover Map or the National Wetland Inventory. Most
frequently they have been converted to row crops. It seems that many grassland or wetland sites move in and
out of cultivation depending on weather and economics.

The fifth paper presents results from the initial five year cycle of bird censusing at CTAP sites, with a focus
on habitat dependent, area dependent, threatened and endangered, and exotic species. Illinois forests seem to
be in the best condition compared to wetlands and grasslands, with a fair diversity of forest bird species.
Grasslands continue to be the most degraded habitat for birds. Almost one-third of grassland sites contain no
grassland-dependent species and four-fifths have no area-dependent birds. A few wetland-dependent birds are
relatively common in many Illinois wetlands (including Wood Duck, Mallard, and Willow Flycatcher),
although half of the wetland sites contain no wetland-dependent species. The Brown-headed Cowbird, a nest
parasite, was detected at a high percentage of sites across all habitats, ranging from 53% of wetland sites to

80% of forest sites. Overall, the data illustrate the degraded nature of Illinois habitats and the avian
communities they harbor.

The sixth paper presents the results from the 58 ForestWatch sites monitored by citizen scientists during last
year’s fall monitoring session and 42 sites monitored in the spring session. One-fourth of the oak-hickory
upland sites showed some degree of maple takeover. Half of the sites with flowering dogwoods reported
dogwood anthracnose during the 2001 monitoring period, an increase over previous years. More than two-
thirds of sites contain invasive shrubs, such as multi-flora rose or buckthorn. They cover 80 times the area
covered by disturbance-sensitive plants.

The seventh paper highlights the data from EcoWatch’s newest program, PrairieWatch. Of the 27 prairie sites
monitored in 2001 and 2002, 11 were reconstructions and eight were remnants (the remainder were
unspecified). While volunteers encountered a familiar excess of invasives and dearth of disturbance-sensitive
species, the prairie sites were healthier than the typical forest site or the typical Illinois grassland that is
dominated by introduced grasses. Disturbance-sensitive indicator plants covered one-fourth the area covered
by invasives. At 13 sites PrairieWatch volunteers counted 735 butterflies, half of them indicator species that
they had been trained to identify. The new Illinois Butterfly Site Index requires further testing to establish its
relationship to plant habitat quality.

The eighth paper summarizes the findings from the 224 sites that RiverWatchers monitored in 2002. Most
streams support high numbers of taxa that are somewhat tolerant of pollution, such as sowbugs and midges,
indicating some level of habitat degradation or pollution. However, pollution intolerant taxa manage to
survive in small pockets. Some watersheds are in better health than others. The Kaskaskia, for example,
scores relatively low on most stream indicators, while the Rock River scores fairly high.

The final two papers examined the usefulness of Floristic Quality Assessment and leafhopper species to
measure the quality of terrestrial ecosystems. FQA was found to be an excellent measure of the amount of
degradation an area had undergone; it was highly correlated with natural area grade. Wetlands generally
scored higher in floristic quality than grasslands and southern Illinois scored higher than other parts of the
state. Further work is needed to take the wealth of leafhopper and other insect data collected by CTAP
scientists and create indicators of ecosystem quality. Information on arthropod ecology, distribution, and
diversity can help to complete the picture of the quality of Illinois ecosystems.

Creating a Report Card for RiverWatch Stream Quality:
Multi-Metric Biological Score

David Baker

Introduction

EcoWatch citizen scientists collect and identify a sample of macroinvertebrates each spring during
RiverWatch monitoring. This information becomes the basis for calculating five different biological indicators
of stream quality—MBI, EPT richness, total taxa richness, dominance, and percent worms. RiverWatchers
also collect a wide variety of habitat and physical data that can be, although so far has not been, used to gauge
the quality of local streams. However, even for the biological indicators, it is not always clear what a
particular value means.

This paper and the one following it describe an effort to make all of the data that RiverWatch volunteers
collect intelligible. Virtually every piece of data entered on a RiverWatch datasheet is boiled down into two
numbers—a habitat score and a biological score. Each of these scores gets a percentile ranking from | to 100,
so that volunteers know whether their stream scored in the 5" percentile or the 99" percentile. Every site score
also will be rated as excellent, good, fair or poor. No longer will volunteers have to guess what their results
mean. Their stream report card will tell them.

Methods
Several steps were necessary to develop a biological report card for streams:

Put each indicator on a common scale,

Adjust for bias by using random sites weighted by natural division,

Assign weights to the indicators based on the strength of their relationships to watershed disturbance,
Combine the biological indicators into a single index or “biological score”,

Stratify the streams by natural division and stream width,

Define undisturbed or “reference” conditions for each natural division/width group,

Categorize biological scores into quality categories based on the reference conditions.

SO Ba soe

RiverWatch Biological Indicators

Macroinvertebrate Biotic Index (MBI) provides a weighted average of the pollution tolerance of
indicator organisms in a sample, measured on a scale from one to 11. Higher values indicate
more organic pollution, while lower values indicate less organic pollution.

Taxa Richness is the total number of taxa identified in a sample out of the 37 indicator taxa that
RiverWatchers are trained to identify. Generally, taxa richness increases with water quality and
habitat diversity.

EPT Taxa Richness is the number of Ephemeroptera (mayfly), Plecoptera (stonefly), and
Trichoptera (caddisfly) taxa present in a sample. EPT are most diverse in natural streams and
decline with increasing watershed disturbance.

Taxa Dominance measures the percentage of the three most common taxa compared to the
rest of the sample. Dominance by just a few taxa indicates lower stream quality. Generally, a
value greater than 80% is considered low aquatic diversity.

Percent Worms is the percent of the sample represented by Aquatic Worms and Bloodworm
Midges. A high percentage of these organisms indicates poor stream health.

In the first few steps a composite or multi-metric biological score was created. In the last few steps we
defined what the scores mean, in terms of stream quality. Like a teacher in a classroom, we determined the
curriculum (a set of indicators in this case), assigned weights to each subject, and developed the grading scale.

A few different statistics were employed to create the multi-metric biological index for stream quality—
means, standard deviations, correlations, and analysis of variance. The mean is merely the arithmetic average
while the standard deviation is a widely used measure of the dispersion of observations around the mean.
Typically, about two-thirds of observations are within one standard deviation of the mean and 95% are within
two standard deviations. In the analysis, the mean and standard deviation have been used to estimate what
percentile rank a particular observation holds in the distribution of biological scores.

Correlation coefficients are used to gauge the strength of relationships among variables. A coefficient close to
1.0 indicates a perfect 1:1 relationship, while a coefficient close to zero indicates no relationship. In the
analysis, the correlation coefficients were used to decide how to weight each biological indicator in the
composite biological index, based on how strongly each was correlated with measures of disturbance.

Table 1. Mean and standard deviation of biological indicators

Random Sites All Sites
mean stand dev mean stand dev

MBI 5.84 0.97 6.00 1.07
EPT 2.56 1.99 2.36 1.93
Taxa Richness 9.12 3.81 9.21 . 3.44
Dominance 0.79 0.15 0.80 0.14
% Worms 0.07 0.14 0.08 0.16
Bio Score 50 21 53 29

Table 2. Percentile ranks for biological indicators

Percentile Rank MBI EPT TaxaRichness Dominance % Worms
0 11.0 0.0 1 100% 100%
10 7.1 0.0 4 98.4% 24.3%
20 6.6 1.0 6 91.7% 18.3%
30 6.3 1.5 7 86.8% 14.0%
40 6.1 2.0 8 82.6% 10.2%
50 5.8 2.5 9 78.8% 6.7%
60 5.6 3.0 10 74.8% 3.2%
70 5.3 3.5 11 70.7% 0%
80 5.0 4.0 12 65.8% 0%
90 4.6 5.0 14 59.1% 0%
100 <2.6 27 217 <48.5% 0%

In scaling the indicators, sources of bias must also be addressed—no cheating allowed. Most RiverWatch sites
are selected by volunteers and may not be representative of statewide conditions. A subset of monitoring sites
has been randomly selected, and it is these sites that have been used to scale the indicators. EcoWatch
volunteers have collected 158 samples from 79 randonily selected sites.” Even these sites may not be truly
representative, since it has been more difficult to recruit volunteers in some areas than in others. For example,
47.5% of random sites are located in the Northeast Morainal division, but only 6.9% of streams are located
there. This additional source of bias has been addressed by weighting the mean and standard deviations of the
random sites by the distribution of streams within each natural division (Table 3).

Table 3. Distribution of monitoring sites by natural division

All Monitoring Sites Random Sites Distribution of

Natural Division* (1877 samples) (158 samples) Streams**
Driftless/Rock 7.1% 5.1% 5.1%
Northeast Morainal 27.8% 47.5% 6.9%
Grand Prairie 23.8% 22.2% 39.5%
Western Forest-Prairie 17.3% 13.3% 16.6%
Southern Till/Wabash 12.7% 8.9% 25.5%
Shawnee/Ozark 12.0% 3.2% 4.7%
Other 0.0% 0.0% 1.7%

“The 14 Natural Divisions have been consolidated into six groups. See discussion below.
“Based on number of sections (each generally one-mile square) that contains streams.

Weighting Indicators Based on Watershed Disturbance

Just as core classes are weighted more heavily than gym or health, the core biological indicators should be
weighted more heavily than secondary indicators. The best indicators of biological quality should score high
when a stream is relatively undisturbed (that is, in a natural state) and low when the stream is disturbed by
human development. The problem is to find an objective measure of watershed disturbance that can be used
to gauge the strength of the biological indicators. In a recent effort to recalibrate the Index of Biological
Integrity (IBI)—a multi-metric index based largely on fish data—the DNR and IEPA developed just such a
measure of disturbance for the 815 watersheds in the state. In fact, they developed two different versions of
the measure for evaluating disturbance to stream sites. The first reflects the disturbance in the watershed
where the site is located, while the second assesses the disturbance in the entire drainage above the site.

The two disturbance ratings take into account seven different measures of disturbance:

1. Proportion of undisturbed land (i.e. upland forest, wetland, bottomland forest)
2. Proportion of disturbed land (i.e. urban, agricultural, barren)

3. Proportion of strip-mined land

4. Maximum storage of impounded water

5. Maximum storage of impounded industrial, mining, or sewage wastewater

6. Number of hazardous point sources

7. Number of sewage point sources

+ Volunteers collected 181 total sample’ at the randomly-selected sites, but only the 158 that contained at least 25
organisms were included in the analysis, because low sample densities do not result in consistent scores.

Frequency
of te 7 Ss 53 8
|
|

_
o

fe)
60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 135 140
Drainage Disturbance Score

Figure 1. Distribution of disturbance scores
for the 317 watersheds monitored by RiverWatch

Each watershed receives a score of | to 20 for each measure, resulting in a total possible score of 7 to 140.
Actual scores range from 60.to a perfect score of 140, with an average of 110 (Fig. 1). Between 1996 and
2001, RiverWatch volunteers sampled streams in 317 of the 815 watersheds (nearly 40% of the total),
collecting from | to 47 samples in each.

Table 4 shows the relationships between the biological indicators and the two different versions of the
disturbance measure. Disturbance in the entire drainage seems to be more important to stream quality than the
disturbance in the local watershed, based on the strength of the correlations. All of the RiverWatch biological
indicators are significantly related to the drainage disturbance score. While the correlations are not spe
strong (.10 to .22), they arc all statistically significant.

Table 4. Correlations between biological indicators and disturbance scores

Biological Watershed Disturbance Drainage

Indicator — Score Disturbance Score Weights
MBI -.200 -.220 2
EPT -136 200 2
Taxa Richness 045 -106 1
Dominance -.053 -.125 1
Percent Worms -.138 -.143 ]

Note: Bold are statistically significant at the .05 level or better.

MBI and EPT are more strongly correlated to disturbance than are taxa richness, dominance and percent
worms. Therefore, the core indicators MBI and EPT are weighted double and the other indicators only once in
the multi-metric biological index.

Defining Stream Quality Levels

Stratification by Natural Division and Stream Width

Junior high students naturally perform better than elementary school students. Teachers do not compare howe a
student performs in their first grade class with a student in sixth grade; they only compare their first grader to
other first graders. Similarly, certain streams naturally perform better (or worse) on various measures based on
certain natural conditions and should only be compared to streams of a similar type. Illinois has been divided
into 14 natural divisions based on topography, glacial history, bedrock, soils, weather and distribution of
plants and animals. Tetra Tech, the consultant that is assisting the IEPA in improving its stream indicators,
categorized streams by natural divisions that were grouped into a handful of categories.”

Analysis of the RiverWatch data revealed that Illinois streams could be grouped into six categories based on
similarities in the way contiguous natural divisions scored on the various stream indicators (Table 5).°* For
example, streams in the Ozark, Shawnee, and Coastal Plain natural divisions in far southern Illinois generally

scored high on most of the biological indicators, while those in the Southern Till Plain and Wabash Border in
southeastern Illinois generally scored low.

Table 5. Grouping of natural divisions based on indicator quality

Disturbance Taxa
Rating Dominance} %Worms Richness
Driftless Hi Med hi Med Hi Hi
oar Coe a ear Pa ae Beata Fe hi me areeks hi
| NEMorainal —_| Morainal | Med low | low| Med | | Medhi _| hi

Grand Prairie Len ss Med
Upper MS/IL Bott. a Med Hi Med Hi

Wn Forest Prairie Med ii :
Mid. Miss. we a Med hi
Southern Till
Wabash bees ie
is

Ozark
Shawnee ff
Coastal Plain ut hi oe hi 0) Hi

In addition to differences characterized by natural division, streams can vary in quality based on the size of
the stream (Fig. 2). Generally, larger streams score higher on measures of biological integrity, largely because
they have more diverse habitats (Table 6). Only in the older, unglaciated areas like the Driftless or Shawnee/
Ozark does stream width not seem to be related to quality, according to the analysis of variance. Eight meters
or about 25 feet seems to be a critical cut off between large and small streams; so small streams are defined as
those under 25 feet in width and large streams those that are at least 25 feet in width. About 70% of the
streams monitored by volunteers are small and 30% large. All of the biological indicators show statistically
significant differences between small and large streams.

* Based on a seminar by Tetra Tech sponsored by IEPA..
** Two small natural divisions—lIllinois River and Mississippi River Sand Area, and Lower Mississippi Bottomlands—
had few or no sites and were left out of the analysis.

Mean of Taxa Richness

10.5 070

10 065
vo
ro]

95 A 060
8

9 8 055
[e)
iva}

85 S 050
c
bi

8 = 045

75
040
0 5 10 15 20 25 0 5 10 15 20 25
Stream Width (meters) Stream Width (meters)

Figure 2. Mean taxa richness and mean biological score by stream width

Table 6. Differences between small and large streams

small 1122 5.80
large, 467 5.47
EPT small 1122 2.30
large 467 3.57
Taxa small 1122 8.60.
Richness large 467 10.18
Dominance small 1122 82.1%
large 467 76.7%
% Worms small 1122 6.3%
large 467 4.2%
Biological Score small 1122 49.05
large 467 63.02

Note: Bold indicates statistically significant differences.

Figure 3 shows the final stratification of streams by natural division and stream width. It shows that the
Northeast Morainal and Southern Till/Wabash streams scored the lowest on average, while the Driftless/Rock
and Shawnee/Ozark streams scored the highest. It also shows that wider streams, at least in several natural
divisions, also scored higher on average. It does not reveal, however, whether these differences are natural
differences or attributable to local stream conditions and disturbances. Nor does it tell us what a good stream
should score versus a poor stream.

Driftless/Rock Osmall

glarge

Northeast Morainal

Grand Prairie

Southern Till/Wabash

Western Foret ee ee

Shawnee/Ozark

0.1 0.2

ad
w

04 0.5 0.6 0.7 08

Biological Score

Figure 3. Average biological scores stratified by natural division and stream size

Defining Reference Conditions

What constitutes an “A” or excellent quality? Teachers often scale the grades on a particular term paper based
on the best papers, those prepared by the best students. When evaluating natural systems, like a stream,
scientists generally define exccllent quality based on reference conditions, that is sites that are the most
pristine. In Illinois, very few streams are in pristine condition, but those that are least disturbed can be
identified.

Illinois EPA used the Watershed Disturbance Measure to help identify potential reference sites for the IBI
along with a couple of additional “screens”. If any “undisturbed” site showed chemical contamination it was
excluded. The regional aquatic biologists were also asked to identify any sites that they knew to be locally
“disturbed”. The final list of reference sites included about 15% of the total pool of sites. Because of the lack
of both chemical data and professionals familiar with local sites, a slightly different approach was applied in
this study. The Drainage Disturbance Measure was used as the primary screen, but two other screens were
applied as well.

e The reference sites must score above the 75" percentile on the Drainage Disturbance Score.
e They must score above the 75" percentile on the Habitat Score.
e The few remaining sites that score below the 25" percentile on the Biological Score are excluded.

An excellent stream is then defined as a stream that scores above the average of the reference sites, a good
stream scores within one standard deviation of the average, a fair stream is within two standard deviations,
and a poor stream scores beyond two standard deviations (Table 7). Other scoring schemes are possible, but
this is one that is easily justified. The IBI similarly defines its highest category by the mean of the reference
sites, but defines its other categorics in a slightly different way. Figure 4 shows the distribution of reference
sites and random sites throughout the state, while Figure 5 shows the percentage of poor to excellent stream
quality grades for all of the 1,669 samples collected over the six-year period and for the 158 samples collected
from random sites.

* Reference sites

Figure 4. RiverWatch reference sites and random sites

All Samples

excellent

20%

Random Sites

excellent

good
16%

fair
20%

Figure 5. Distribution of stream quality grades, all samples and random sites

Table 7. Stream quality grades based on reference sites

Excellent

Greater than
mean of

biological scores to-1 SD

Driftless/Rock >66.5
Northeast Morainal-small >66.3
Northeast Morainal-large >78.7
Grand Prairie-small >73.9
Grand Prairie-large >86.5
Western Forest Prairie-small >65.1
Western Forest Prairie-large >70.9
Southern Till Plain/Wabash >69.6
Shawnee/Ozark >75.4
Average Pi ee.

Good Fair Poor
mean -1 to -2
SD <-2 SD
66.5 -48.1 48.1-29.7 <29.7
66.3-47.9 47.9-29.5 <29.5
78.7-60.5 60.5-42.3 <42.3
73.9-54.7 54.7-35.6 <35.6
86.5-73.3 73.3-60.2 <60.2
65.1-43.0 43.0-20.8 <20.8
70.9-57.0 57.0-43.1 <43.1]
69.6-48.0 48.0-26.3 <26.3
75.4-56.3 56.3-37.2 <37.2
72.2-52.5 52.5-32.8 <32.8

Figure 6 shows the distribution of stream quality across the natural division and width categories. About 45-
50% of the streams in the Shawnee/Ozark and Driftless/Rock are in excellent condition, while about 40% of
the streams monitored in the Southern Till Plain/Wabash and Northeast Morainal are poor in quality. Many of
the larger streams in the Grand Prairie and Western Forest Prairie also had fairly high quality.

DriftessRock

Northeast Moraind -
small

Northeast Moraind -
large

Grand Prairie -small

Grard Prarie- large

West Forest Prarie-
smal

West Forest Prarie-
large

Southem TillWabash

Shawnee/Ozark

Spoor far ®good excelent

Figure 6. Distribution of stream quality grades by natural division/width categories

It should be pointed out that the professional data discussed in the third paper show a smaller percentage of
excellent streams than the volunteer data. Volunteer-collected data generally do not provide as much
resolution in quality as professional data and particularly have difficulty distinguishing good from excellent
streams. An alternative to the grading scheme described above would be to define good streams as above the
mean of the reference sites and adjust the other categories accordingly. The resulting scale would be closer to
that used in the CTAP professional stream monitoring.

Conclusion

RiverWatchers.can now create a report card to see how their site scores on the five biological indicators and
the composite biological score. (This information will soon be available on the CTAP/EcoWatch web site.)
The report card not only shows them the percentile rank for each indicator, but also whether the stream
quality is excellent, good, fair or poor. Table 8 shows a sample report card from 1996 to 2001 for a site in the
Shawnee/Ozark Natural Division. Generally this site has scored high on most of the indicators, except for the
year 1997 when the site was monitored soon after severe flooding. Overall the quality of the site is excellent
and it is one of the reference sites for its natural division. Similar report cards can be compiled for larger
geographic units such as watersheds, ecosystem partnerships and ISIS Basins (Fig. 7 and 8). The next paper
completes the report card by adding a habitat score as a final element.

Table 8. Stream quality report card for RiverWatch site R1008401 on Clear Creek

(percentile scores)

Stream Group: Shawnee/Ozark Overall stream quality: excellent
Year MBI EPT Taxa Richness Dominance % Worms __ Biological Score

(raw scores)

Year Taxa Richness Dominance % Worms Biological Score

1996
1997
1998
1999
2000
2001
Average

— Poor
[| Fair
[__] Good
| | Excellent

Figure 7. Average biological score of ISIS basins. The numbers denote the number of
sites sampled over the period. Sites could have from one to six samples, and each
site's scores were averaged and weighted by natural division before the basin score
was calculated.

5G poi LAKES/CALUMET

LTL WAB/LOW WAB/SKILLET FK
6

Figure 8. Average biological score in 33 sub-basins. Numbers denote the
number of sites sampled over the period. Sites could have from one to six
samples, and each site’s scores were averaged and weighted by natural
division before the sub-basin score was calculated.

Creating a Report Card for RiverWatch Stream Quality:
Multi-Metric Habitat Score

David Baker

Introduction

What does it mean if the water in my stream is “milky”? Or if it smells “fishy”? Or if the stream is
channelized or lacks canopy cover? Volunteers are familiar with MBI, EPT and taxa richness as measures of
stream quality. These biological indicators are calculated when they fill out the RiverWatch data sheets or
submit their data electronically. However, every spring trained volunteers collect a wide variety of physical
data that can be used to gauge the quality of stream habitat. In addition to identifying a sample of
macroinvertebrates, RiverWatchers collect information on the stream’s substrate, turbidity, odor and color,
surrounding land uses, disturbances to the stream channel, canopy cover, amount of algae, water and air
temperature, and recent weather. Thus far none of this habitat-related data has been connected to the quality of
the stream in more than a very general way. This paper examines the relationship of the habitat data to the
biological indicators and to the multi-metric biological score described in the previous paper. The Habitat
Score described here completes the effort to create a report card of Illinois stream quality. Together the
Biological Score and Habitat Score use virtually every piece of information collected by RiverWatch citizen
scientists to rate stream quality.

Methods

Three statistical techniques were used to examine the relationships between stream habitat and biological
indicators—analysis of variance, correlation, and regression analysis. These enabled us to identify the habitat
characteristics that are the best indicators of stream quality and to combine these into a single multi-metric
index.

Analysis of variance (ANOVA) is a tool for determining if there are significant differences among groups on a
particular variable. In this case ANOVA was used to determine whether biological quality varies based on
habitat condition. For example, ANOVA showed that higher percentages of silt are associated with poorer
stream quality and that certain land uses are better for stream quality and others are worse. Using ANOVA,
most physical variables were given a score of one to four, with each higher value associated with higher
biological quality. For example, 0-25% silt was given a score of four, 26-50% a three, 51-75% a two, and 76-
100% a one. Table 1 provides the scoring scheme for all of the habitat measures.

Correlation coefficients were then calculated to gauge the strength of the relationships among variables. A
coefficient close to 1.0 indicates a perfect 1:1 relationship, whereas a coefficient close to zero indicates no
relationship. The coefficients showed which habitat characteristics are related in a statistically significant way
with biological indicators. Because of the large sample sizes (more than 1,600 stream samples), even
correlation coefficients lower than .10 could be statistically significant.

Finally, regression analysis was employed in the case of habitat measures which possess multiple
components, such as land uses or stream substrate. At most sites several different land uses were present or
dominant and, likewise, the stream substrates are generally composed of several different materials (boulders,
silt, etc.). Regression was used to combine these various components into a single metric. In addition, it was
used to create a multi-metric index using the strongest of the habitat indicators.

Table 1. Categorization scheme for habitat data

Category or score
Conanpen 4

Water appearance Foamy, reddish, Clear, milky, or
green, or other dark brown

Water odor Sewage, ores sy No odor
chlorine, rotten
eggs, or
petroleum

Tor Turbidity Heavy turbidity Medium Medium xii | | Slight big turbidity Sateen |

Canopy cover
> 75% 0%, 50-75% 26-50% 5-25%

Coarse substrate, Weighted score Weighted score Weighted score Weighted score
Weighted score = 2 x wt a =2-4 =5-7 >8
Cobble + Gravel +.

Bedrock,

where 0 = 0%, 1 = 1-5%,

2 = 6-25%, 3 = 26-50%,

4=51-75%, 5 = 76-100%

Silt/fine.substrate 16-100% 51-75% 26-50% 0-25%

Dominant land uses
Where.2=dominance of total score of total of -1 or -2 total of 0 to 2 total of 3 or more

| Habitat Measure Measure

forest, pasture, cropland, or <-2
grassland, 1=presence of

same,

-2=dominance of urban

land uses,

-1=presence of urban land

uses, or dominance of park

or low density residential

Other disturbances
Channelization, wastewater
discharge, pipe, upstream
dam

3 or more

eEehiraances 2 disturbances Only 1 disturbance | No disturbances

Stream Habitat Indicators

An examination of the data collected by volunteers during the past several years shows some very definite
relationships between the habitat data and biological indicators. In only a few cases are the relationships not
statistically significant or are spurious in some way. Because many different factors affect stream quality,
however, few habitat characteristics of the monitoring sites showed correlation coefficients greater than .30.
In combination, the physical data can explain about half of the variation in biological scores. Each of the
habitat variables and its usefulness as a stream indicator is discussed below.

Weather. Heavy rain in the two days before monitoring seems to adversely affect some of the biological
indicators (Table 2). By washing indicator organisms out of the stream bed, heavy rain reduces the number of
EPT taxa, increases the percentage of worms, and raises (worsens) the MBI. Less extreme weather events
seem to have very limited impact. Higher water and air temperatures also have a slight impact on stream
quality, although neither is significantly correlated with any of the indicators. The large differences between
surface and air temperatures, as occurs in the Driftless/Rock River’ area in northwest Illinois, may indicate
greater groundwater vs. surface recharge. Overall, the weather variables would not make good indicators of
stream quality, but data collected after heavy rain should be flagged and volunteers should be discouraged
from monitoring shortly after major rain events.

Table 2. Correlations between biological indicators and weather

Weather in Weather in Water Air
Last 24 Hours| Last 48 Hours | Temperature | Temperature

BioScore | _-.023, | 082 | 015] 017
M -omdOBTe., oli epnliRoes will vers. 033.0 eof ientOO5. aa

BI .087 :
EPT -.002
TaxaRichness| 029. | ~— 000s |. —-.025, [052s
Dominates” | ite” 029 Oe as
yworms [06s [04s | =042 | =043

Note: Bold are statistically significant at .05 level.

Water Appearance. Streams where the water is clear, milky, or dark brown in color seem to be higher in
quality, while those where the water is foamy, reddish, or greenish are lower in quality. The first three are
generally naturally occurring while the latter three indicate some kind of disturbance. While the statistical
relationship between water appearance and biological quality is not extremely strong—a correlation of only
.12 with total biological score—the relationship is consistent and statistically significant across all of the
biological indicators except for total taxa richness (Table 3).

Water Odor. Water odor is also related to stream quality (Table 3). Streams with no odor score best on most
indicators, while streams with a fishy smell are slightly impaired, and those with an odor of sewage, chlorine,
rotten eggs or petroleum are even more impaired. For example, streams with no odor average 5.65 for the
MBI, while those that are “fishy” average 6.10 and those that smell of sewage average 6.39.

* For purposes of this paper, the state’s natural divisions have been aggregated into six groups: Driftless/Rock River,
Northeast Morainal, Grand Prairie (including Upper Mississippi and Illinois River Bottomlands), Western Forest-Prairie
(including Middle Mississippi Border), Southern Till Plain/Wabash, and Shawnee/Ozark (also including Coastal Plain).
See previous paper.

Table 3. Correlations between biological indicators
and selected habitat variables

ee ee
| 13 | -.036_|
BPI eng 1 fecal _.c00) A 58 2h | 150 | 180
| TaxaRichness [043107 [| 180. [o0. caleGel
| Dominance | -.054_ | -.091 | -005 | -005_ | -.127
ie Worms —— | i — pad 9] ees) eT | -1032 |

Note: Bold are statistically significant at .05 level or better.

Turbidity. Turbidity has a statistically significant impact on MBI, EPT and worms, but seems unrelated to
dominance or taxa richness (Table 3). Clear streams are highest in quality, those with heavy turbidity are
worst in quality, and those with slight or medium turbidity are intermediate in quality. For example, MBI
averages 5.4 in clear streams and 6.0 in heavily turbid streams, while percent worms averages 3.8% in clear
streams and 8.0% in turbid streams.

Algal Growth. Streams with greater amounts of algae seem to be higher in quality (Table 3). This is
counterintuitive and may show that volunteers are not performing this procedure properly. RiverWatch
procedures ask volunteers to only look for filamentous algae, which bloom dramatically in nutrient enriched
waters, but they may be reporting all types of algae. Volunteers most frequently report algae in the rocky
bottomed streams in southern Illinois, which tend to be high in biological integrity. Volunteers should receive
further instruction on the procedures for measuring algal growth before this measure can be used as an
indicator of quality.

anopy Cover. The relationship of amount of canopy cover to stream quality is not a straightforward one.
RiverWatch sites with 1-50% canopy cover seem to perform better on the biological indicators, while those
with greater than 50% canopy or no canopy cover perform worse. A mix of areas where water is fully shaded,
some fully exposed to the sun and others receiving degrees of filtered light is optimal. Such conditions are
found in a mature forest.

Substrate. Streams with a rocky substrate, such as bedrock, boulders, cobble and gravel, tend to have more
EPT taxa, higher taxa richness, and fewer worms, while streams with higher amounts of silt score more
poorly on most biological indicators (Table 4). The amount of sand seems to be unrelated to stream quality in
the RiverWatch data. Cobble and silt are most strongly related to stream quality, with correlations of about
.30. The composition of the substrate varies across the natural divisions. For example, the Shawnee/Ozark has
the highest percentage of bedrock and boulders, while the Southern Till/Wabash, Northeast Morainal, and
Grand Prairie have the highest percentages of silt. Two different indicators can be created from the substrate
data—one using percent silt and the other a composite of the amounts of the various coarse substrates. A
regression analysis shows that a new variable composed of two times the amount of cobble plus the amount of
bedrock and gravel is more highly correlated to the biological scores than any other habitat characteristic
measured. '

Table 4. Correlations between olnetey indicators and stream substrate

Bedrock | Boulders Cobble Gravel Sand Coarse
Substrates*

SS | 277 | da | -021 | 266 | 329

oy ae Sar ee ST
or me ae eT TE | .202 | 135. | .026_ | -.224. [1266
eee ee ee
| Dominance | -.015 | -.063 | -092 | -.042, [023 | 041 | -.126 |

ae Worms [snr a9 ais [ae] os] as] a

Note: Bold are statistically significant at .05 level or better.
* Composite of 2 x cobble + 1 x (bedrock + gravel)

Stream Width, Depth, Velocity, and Discharge. Larger streams are associated with higher quality as measured
by most biological indicators (Table 5). These relationships are strongest for stream width and discharge, and
weaker for depth and velocity. Stream size is most highly correlated with EPT, showing a correlation of .233
with stream width. Streams that are more than 25 feet in width average 3.6 EPT taxa, while those under 25
feet average 2.3 taxa. Generally, wider streams have more varied macroinvertebrate habitats that support
greater EPT richness.

Table 5. Correlations between biological indicators and stream size

Note: Bold are statistically significant at .05 level or better.

Stream Channel Disturbances. Channelization clearly reduces stream quality as measured by MBI and EPT
(Table 6). Other types of stream channel disturbances such as the presence of pipes, wastewater treatment
discharge and, to a lesser extent, upstream dams all adversely affect stream quality as well. The more of the
disturbances present the more quality seems to be affected. Streams in the Northeast Morainal natural division
are the most channelized (23% of RiverWatch sites), most likely to have nearby water treatment plants (32%),
and most likely to have discharge pipes (34%). Streams in the Shawnee/Ozark and Driftless/Rock are
disturbed least by channelization (4%, 9%), pipes (13%, 12%), and water treatment (4%, <11%). Stream sites
with upstream dams are distributed fairly evenly across the state, with the fewest in the Southern Till/Wabash.
The number of these four disturbances present at a site is a fairly good stream quality indicator.

Habitat Sampled. As would be expected, streams with riffles are higher in quality than those where other
habitats are the primary habitat sampled, particularly sediment. For example, the MBI averages 5.62 at sites
where riffles are sampled compared to 6.03 where other habitats are sampled. However, the results are not
Statistically significant and the ANOVA demonstrates that by itself the habitats sampled would not be a good
stream quality indicator.

Table 6. Correlations between biological indicators and channel disturbances

Upstream | Wastewater No. of Channel
Dam Discharge Channelization|} Disturbances
BeBe lag fit ea ie eo

[Taxa Richness | _.014_| 056 | -033_ | 030 | .0s3__—
[Dominance | __.0%6 | 092 | 058 | 019 ~—~«Y~C
% Wom A

Note: Bold are statistically significant at .05 level or better.

Land Uses. Surrounding land uses definitely affect stream quality (Table 7 and Fig. 1). As expected, streams
where forest or grassland is dominant score well on the biological indicators. However, perhaps surprisingly,
agricultural land uses such as pasture and cropland are also associated with higher stream quality. Sites
dominated by urban land uses—sewage treatment, construction, landfills, golf courses, high density
residential, and commercial/industrial development—score more poorly on stream quality indicators. Streams
where parkland or low density residential is dominant are of intermediate quality. Only logging and mining
are not significantly correlated to the biological scores, but they are dominant at only a few sites.

Since volunteers may report multiple land uses to be dominant or present at a site, it is not straightforward to
use land use as an indicator. ANOVA was used to categorize sites by the number of positive and negative land
uses. The resulting composite variable was strongly related to stream quality, with a correlation of .29. Only
substrate is more strongly related to the biological indicators.

Table 7. Correlations between biological indicators and surrounding land uses

Forest Logging Golf Grass-
Course land

pease De eed ee ee
SL MRE eee ee wl
a ee eee
EAS EO AT
P“<o6 | 019 | 021 | -020| 113 | 050 | 070] -197
P% Worms | 077 | 013_| .061_| -032 | 109 | 051 | .096 | -.124 |

Commercial] Low Dens.
Industrial |Residential

Residential | Land Use

Cropland Sewage Landfill Construction
Treatment

[BioScore | 39 | -096 | ~083 | 003 | -006 | aan] -a0a|
| MBE 988s | 104 | 067-04 4a a 096 a
-epT______1s7_|_-.066_} -ut_{_.003_{ 086 _{_.437_{ 083 _{_1
-fzra Richness +089 __-083_{_=07t_{_017_{ ~os6_{ 122} st_{_
Dominance | ~081 i
ree Wwann [007 [tat] or? cone one see ts]

Forest
Pasture
Cropland
Grassland
Park land
Sewage treatment
Low density resicntal
High densityresidentia
Constuctin
Landfill

Commecid/irdustrial

Golfcourse

Biobgcal Score

Figure 1. Biological scores by dominant land use

Creating a Multi-metric Index

Once all of the measures of habitat condition were rescaled with scores from onc to four and the strength of
relationships examined, it was fairly straightforward to create a multi-metric index using regression analysis.
Figure 2 compares the correlations between each of the primary habitat variables and the composite biological
scorc. The weather variables and the sampled habitats have not been included, since they are not good stream
quality indicators. The stream size variables have also been excluded from the analysis, since stream width is
being uscd to categorize streams for purposes of the biological scores. Stepwise regression included all of the
remaining variables except for turbidity, which is the least correlated with biological score (Table 8). Its
cocfficicnt was not statistically significant in the regression equation, because other variables measure similar
factors (i.c. water appearance and silt).

Correlation coefficient

“ibidieal healt alsbiie bia Mase
oi MUA Bie ey!

Figure 2. Correlation between measures of habitat condition and biological score

Table 8. Regression of biological score and habitat characteristics
R = .495, R? = .245, F= 76.4

pitas cls be [ned fh relay vfs cach of thes fyracaigaleance |

Coarse subsrate | 76 «SSCA C*d SCOOT CS
Rinne Ons OT SETVINCS HP TIPIOVE HM QURIRNO Bee: Mey eke we. GOrount |
See RE Bae a ee

To simplify calculation of the composite habitat score the constant term was dropped and the following
weights used:

© coarse substrate 3.0
e land use 3.0
e water odor 1.5
e fine substrate (silt) 1.5
e canopy cover iTS
e channel disturbances 1.0
e water appearance 1.0

This results in a range in scores from 12.5 to 50 (12.5 if a site scores a one on all of the indicators and a 50 if
it receives a 4 for all of the habitat measures). Thus in equation form:

Habitat score =3 x (coarse substrate + land uses)
+ 1.5 x (canopy + silt + odor)
+ 1 x (appearance + channel disturbances)

Figure 3 shows that actual scores range from 20 to 50, with an average of about 39. The 10" percentile is a
score of 33 and the 90" percentile a score of 45. Habitat quality scores can then be given descriptive grades as
shown in Table 9. To assign a habitat quality grade to larger geographic units such as watersheds, ecosystem
partnerships, and ISIS Basins, individual site scores are averaged by the geographic unit and then categorized
(Fig. 4 and 5).

Table 9. Habitat quality grades

| | Habitat score

[Excellent | 43-50 | _75.1-100%|
PFair | 365.39 | 25:1-50% |
Oa i a

20.0 25.0 30.0 35.0 40.0 45.0 50.0
Habitat Soore

Figure 3. Percentile rank of habitat scores

Conclusion

The multi-metric habitat score provides an excellent indicator of stream quality to supplement the various
biological indicators and the multi-mcetric biological score. It is significantly correlated to these biological
measures (correlation coefficient of ~.5). Several of the stream habitat measures that it includes can be
directly improved through management, restoration, or pollution control activities. The amount of canopy
cover can be affected, the natural channcl may be restored, nearby activities that affect stream odor and
appearance (and turbidity) may be reduced or buffered, land uses can be changed. Such changes should
improve overall stream quality and be reflected in both the habitat and biological scores.

Additional habitat characteristics will be added to the multi-metric habitat score in the future. Once the issues
related to properly measuring algal growth are resolved, this factor could be included. RiverWatch procedures
were recently revised to include anew measure of siltation to replace the embeddedness procedure, which
was not consistently implemented. Siltation could replace silt substrate in the index, leaving a single substrate
measure in the index. Turbidity may be added as well in the future, although it seems that other measures
capture the effects of turbidity.

The habitat score would be a more powerful indicator if it were to incorporate additional physical
characteristics of streams that are amenable to management and restoration. For example, channel sinuosity,
width of the riparian vegetation zone, and stream bank cover or stability all are known to affect the quality of
stream habitat. Volunteers could be trained to reliably measure cach of these characteristics.

Scores on the two multi-metric indices—habitat and biological—are tools that will track stream quality and

will help to gauge the success of activities to improve that quality. Together they take into account virtually
every picce of information that RiverWatch volunteers collect to create a report card of stream quality.

166
Fox & Des Plaines Rivers

Z| Excellent

Figure 4. Average Physical Score of ISIS basins. RiverWatch data, 1996-2001. The numbers denote the
number of sites sampled over the perlod. Sites could have from one to six samples, and each site’s scores
were averaged before the busin score was calculated.

\T LAKES/CALUMET

Figure 5. Average Physical Score in 33 sub-basins. RiverWatch data, 1996-200]. Numbers denote the
number of sites sampled over the period. Sites could have from one to six samples, and each site's scores
were averaged before the sub-basin score was calculated.

Aquatic Insects Report
Biological and Habitat Condition of Illinois Streams

R. Edward DeWalt
Introduction

Illinois Department of Energy and Natural Resources (1994) discussed several areas requiring additional
research related to understanding the condition of flowing water habitats in the state. One such area was the
need for long-term studies on aquatic insects. This is an assemblage for which long-term, quantitative
information is lacking in Illinois. Systematic works provided qualitative information on where species of
mayflies (Burks 1953), stoneflies (Frison 1935), and caddisflies (Ross 1944) were located and some
indication of the quality of habitat in which they existed. The CTAP professional aquatic entomologist has
been gathering quantitative data on mayflies, stoneflies, and caddisflies from randomly chosen streams since
1997. During the first five-year cycle of data collection, 149 sites were assessed. The objective of this first
phase was to assess the current condition and geographic trends in stream quality across the state and prepare
for investigation of long-term trends.

Methodology

Several parameters were measured to assess stream condition including water chemistry, habitat quality, EPT
taxonomic richness, and Hilsenhoff Biotic Index. Each of these is explained below. Snapshot values of several
water chemistry and physical attributes were collected at each site using a Solomat 520-C multiparameter
meter. These include water temperature, dissolved oxygen (reported here as percent saturation), pH, and
conductivity. The meter was calibrated each day of use for all parameters measured.

Measurement of habitat quality is important in estimating the potential for streams to support aquatic
communities. Habitat degradation from a variety of sources causes the most damage to aquatic systems (Karr
et al. 1986). CTAP uses a 12-point quality scoring scheme developed by the USEPA (Barbour et al. 1999 and
Plafkin et al. 1989) to measure habitat quality. Values range from 0 to 180, with greater values indicating
better habitat quality.

Stream conditions were also assessed using three orders of environmentally sensitive aquatic insects: the
Ephemeroptera (mayflies), Plecoptera (stoneflies), and Trichoptera (caddisflies) (collectively, EPT taxa).
These often contribute a major proportion of the abundance and species richness to the aquatic
macroinvertebrate assemblage found in streams. EPT taxa richness (number of unique taxa in a sample) is one
of the most efficient indices of stream condition. The history and usefulness of EPT taxa was recently
summarized by Lenat and Penrose (1996).

The Hilsenhoff Biotic Index, developed in Wisconsin, is a weighted average of the organic pollution tolerance
of aquatic insects. Most taxa in the region have been assigned tolerance values that ranged from 0 to 10
(Hilsenhoff 1987). A site value may range from 0 to 10 also, with higher values indicating greater tolerance or
poorer condition.

Regionalization of the results follows that of the 10 interior Illinois Streams Information System (ISIS)
basins. This will aid in the interpretation of spatial patterns in stream condition.

Tests for fit to a normal distribution for all data subsets (EPT, HBI, and Habitat versus 10 ISIS basins) found
only two of 30 subsets failing normality, both for the variable HBI. This level of non-normality was deemed
insufficient to warrant data transformation; hence, all statistical analyses were conducted using parametric
tests. A three factorial analysis of variance (ANOVA; SAS 1985) using ISIS basin, channel type (meandering

versus channelized), and stream width code (1-5, largest streams had higher integer value) was conducted to
find trends in the data. Each site was rated based on its position in percentile rankings of each ecological
indicator on a statewide basis. Percentile rankings and their corresponding quality ratings are presented in
Table 1.

Table 1. Percentile ranges and tentative quality ratings for stream ecological indicators

Percentile Ranking for Tentative Quality Rating
Ecological Indicators
290 Excellent
275 to <90 Good
250 to <75 Fair
230 to <50 Poor
<30 Very Poor

Overall quality rankings for each site were generated as a weighted average of EPT, HBI, and Habitat
percentiles using the following equation:

Overall “ile = (EPT “ile * 0.4) + (HBI “ile * 0.2) + (Habitat %ile * 0.4)

HBI is not as sensitive to degradation of stream condition as is EPT or Habitat, leading to the reduced
weighting for this variable (DeWalt, unpublished data). Qualitative ratings (excellent, good, etc.) were
constructed for each site based on the rating system developed for individual indicators (Table 1).

Results
From 1997 to 2001, 149 stream sites were sampled in 10 ISIS basins (Fig. 1, Table 2). The greatest percentage

of sites fell in the Kankakee/ Vermilion N/ Mackinaw river basins, while the Little Wabash drainage was
represented by the lowest percentage of sites.

Table 2. Number of CTAP randomly chosen stream sites sampled
during a 5-yr cycle beginning 1997

ISIS Basin 1997 1998 1999 2000 2001 Total %
Big Muddy/Saline/Cache 4 3 ] 4 ] 13 8.7
Embarras/Vermilion S 3 0 7 3 6 19 12.8
Fox/Des Piaines 2 4 2 0 4 12 8.1
Kankakee/Vermilion N/Mackinaw ] 8 5 7 3 24 16.1
Kaskaskia 4 6 ] | 5 17 11.4
La Moine 2 4 5 2 3 16 10.7
. Little Wabash 1 ] ] 2 ] 6 4.0
Rock 3 3 3 4 3 16 10.7
Sangamon 4 1 2 4 l 12 8.1
Spoon 4 1 3 3 3 14 9.4
Total 28 «31 30 30 30 149 100

e EPT Sampling Sites
Major Rivers
Major Watersheds
1- Rock River
2- Fox & Des Plaines Rivers
3- Kankakee, Vermilion & Mackinaw Rivers
4- Spoon River
5- Sangamon River
6- River
7- Kaskaskia River
8- Embarras & Vermilion Rivers
9- Little Wabash River
10- Big Muddy, Saline & Cache Rivers

0 20 40 60 Miles

Figure 1. Locations of CTAP stream sampling locations (1997-2001) and ISIS basin boundaries

Dissolved oxygen percent saturation (for 119 sites) was not significantly different across ISIS basins
(ANOVA, F=1.52, p=0.15, df=9). However, some basins might, with more data, prove to have higher daytime
saturation levels than others (Fig. 2). Conversely, conductivity varied significantly (F=2.87, p=0.0001, df=56),
with basin designation as the only significant factor (F=2.45, p=0.21, df=9). The Fox and Des Plaines basin
had a greater mean conductivity than any other basin (Fig. 3). Urbanization is a major factor increasing
conductivity beyond background levels, and much of this basin is heavily urbanized or otherwise rapidly
developing. There were no noticcable, significant pH trends across the state (F=1.15, p=0.30, df=57).

Streamside and in-stream habitat quality varied significantly across the state (F=3.13, p=0.0001, df=62). The
most important factor explaining habitat quality was channel type (F=75.6, p=0.0001, df=1), with meandering
streams scoring significantly higher (107.6 points) than channelized streams (70.9 points). A significant
interaction between channel type and basin also was discovered (F=2.8, p=0.006). Rock and Spoon basins
appcarcd not to have significant differences in habitat quality for the two channel types, whereas significant
differences cxisted in the other basins (Fig. 4). Channelized streams in the former scored higher, probably duc
to having higher quality riparian zoncs, a factor that heavily influences habitat quality ratings taken at each
site.

EPT spccies richness varied significantly across the state (F=2.68, p=0.0001, df=63), with channel type
accounting for the greatest variation (F=13.9, p=0.0003, df=1). Streams with meandering channels produced
an avcrage of 11.8 EPT (n=88), while channelized streams produced 7.] taxa (n=61). Stream width code was
also a significant factor (F=3.53, p=0.01, df=5) such that larger streams supported more EPT taxa (p=0.05)
(Fig. 5). Basin assignment was not a significant factor, although it was nearly so (F=1.9, p=0.06). Basin
assignment and stream width codc interacted significantly (F=1.8, p=0.02, df=4). EPT richness increased
strongly with increasing stream width codc in five of the 10 basins, while in the remaining five it
demonstrated little or no relationship (Fig. 6).

1”
120
Mm

Dissolved Oxygen % Saturation

Figure 2. CTAP stream sampling mean dissolved oxygen percentage saturation + standard error
for ISIS basins. Numbers in bars indicate sample size. No significant differences.

oOo

RW +
on
(ep) ee 2 = *s
= 7 2 a =
= mw z=
o : -_ fe
= a : TAN *
2 : . &
ov is |
“ Pac f Ss we o RS ye x »
pom Rina agin AP a ee a
e oe ie VS Be? 5pm
or we

Figure 3. CTAP stream sampling mean conductivity (uS/cm) + standard error for ISIS basins.
Numbers in bars indicate sample size. Horizontal bars indicate significant differences.

160
— Meandering

gp Uenndi ad
140

120
100
80
60
40

Habitat Quality Score

Figure 4. CTAP stream sampling mean habitat quality score + standard error for ISIS basins by
channel type. Numbers in bars indicate sample size. Note Rock and Spoon basin differences.

EPT Richness

Width Code 1 Width Code 2 Width Code 3 Width Code 4 Width Code 5

Figure 5. CTAP stream sampling mean EPT richness + standard error for five stream width codes (increasing integer
= increasing stream width). Numbers in bars indicate sample size. Horizontal bars indicate significant differences.

EPT Richness

30
| Widel gg Wide? | Wide? gWidol gg Wide
25 4
20 4
1s |
| | | , _f
| | a.
| wf
sf : ; r
BE: ll
i Fy 4 3
é 4
0
S >) @ < AS < a < &
& S < fe) S om iS fe) S
s are = e es Te e
& = > & w & aN 9 x
+ w
fe) ye

Figure 6. CTAP stream sampling mean EPT richness for ISIS basins and five stream
width codes (increasing integer = increasing stream width)

HBI scores did not vary significantly across the state (F=0.9, p=0.66, df=62) (Fig. 7). The Big Muddy ISIS
basin, an unnatural grouping of the Shawnce Hills and Coastal Plains natural divisions, produced the lowest
average HBI. The five streams from the Shawnee Hills subset were of the highest quality available in the
state. HBI there averaged 3.5 + 0.73 (mean + standard error) units.

Overall percentile rankings varied significantly across the state (F=3.1, p=0.0001, df=63), with channel type
being the most important factor (F=57.6, p=0.0001, df=1). Meandering streams scored 61.8%, while
channelized streams scored only 35.8%. Neither basin assignment nor width code alone were significant
factors (F=1.1, p=0.4, df=9 and F=2.1, p=0.09, df=4, respectively). However, a significant interaction with
basin and width code was noted (F=2.3, p=0.03, df=9). Statewide, overall percentiles increased with stream
size (Fig. 8), as they did in the Sangamon, La Moine, Embarras/Vermilion, Spoon, and Kankakee/Vermilion/
Mackinaw basins. However, the basins Big Muddy, Little Wabash, and Rock deviated from that trend. Figure
8 also suggests that the smallest streams were the most heavily degradcd in two of the largely agricultural
drainages (Sangamon, Kankakee/Vermilion/Mackinaw) and in the one mostly suburban drainage (Fox/Des
Plaines).

Discussion

Given that this sampling program was based on randomly selected sites, it is assumed that they are
representative of the state as a whole and that inference about the quality of other streams, and the frequency
in which they occur, may be drawn from this sample. Based on overall percentile scores, 45% of streams
sampled by this program were rated as in “poor” or “very poor” condition (Fig. 9). Some of the worst
offenders (Overall Percentile < 10%) in this grouping were Willow Creck (Fox/Des Plaines basin), Coal
Creek (Rock basin), South Branch of Crow Creek and Rock Creek (Kankakee/Vermilion/Mackinaw basin),
Bean Ditch (Embarras/Vermilion basin) and Pond Drainage Ditch (Little Wabash basin). These streams had
less than two EPT (two had none), were channelized and had no wooded riparian zone. The percentages of
fine sediment (sand, silt, and clay fractions combined) usually exceeded 80%, a trait promoted by heavy
erosion. These poorest-of-the-poor were not relegated to any one basin, but could be found in any, whether .

urbanized or agricultural.
ee
7 i e es Se Ss

* se s
& Pg & a @ SP
> “< S - ~

HBI Score

Figure 7. CTAP stream sampling mean HBI scores + standard error ISIS basins. Numbers in
bars indicate sample size. No significant differences.

Overall Percentile Score

OWidel mWide2 QWide3 mWided gWides

-
o

—]

| ee eee _ =: — TG
3m

60
wo .
wo
20 |
Rom so Ra ESS) oe s ot oF RK
a S e $ Ke oo oe oe et Nixa 2

oo Y ~ \
os > ee & ee ae .! ne 2

<o ye

¢
\
»
a

Figure 8. CTAP stream sampling mean overall percentile score by basin and stream
width codes (increasing integer = increasing stream width)

Excellent

Poor
28%

Figure 9. CTAP stream overall quality ratings and percentages of 149 streams ranking in categories

tw
wa

The chances of the program finding excellent quality streams was remote, but five (3%) were found that had
overall percentile scores 290% (Fig. 9). These included Shokokon Slough (La Moine basin), Sugar Creek
(Sangamon basin), La Moine River, and Gibbons and Threemile Creek (Big Muddy basin). These streams
supported in excess of 18 EPT taxa, had meandering courses, wide treed riparian zones, and produced some
of the lowest HBI index values in the state. The greatest proportion of streams with these characteristics can
be found in the Shawnee Hills subsection of the Big Muddy basin, but can also be found elsewhere in the
state. It is imperative that these best sites be found and characterized throughout the state, since they are the
key to determining the upper threshold for quality.

Channel type appeared to be the most important factor determining overall quality (and its components). Vast
improvements could be made if this one stream characteristic was focused on in policy and restoration
guidance given by state agencies, given that the landowners followed suit. Reestablishment and widening of
riparian zones (especially treed ones) would drastically reduce soil erosion, capture pollutants, reduce algal
blooms, and ameliorate water temperatures.

There are several projects to restore streams across the state including the IDNR sponsored Pilot Watershed
Program (Dodd 2000) and the restoration of natura] meanders and bank structure on Nippersink Creek in
McHenry County. The Nippersink Creek project, sponsored by the McHenry County Conservation District,
began only two years ago and can already demonstrate dramatically improved habitat quality. Biological
change may take longer, especially for aquatic insects, which are not as vagile as fish. They often require
longer time spans to reinvade restored streams (Barbour et al. 1999). This and other large-scale restorations in
the near future might require human aided reintroduction of the most sensitive, less vagile species to help
bridge geographic gaps between restored habitat and recolonization sources. Nonetheless, the modest success
of these projects gives us hope that improvement can be achieved.

Literature Cited
Barbour, M. T., J. Gerritsen, B. D. Snyder and J. B. Stribling. 1999. Rapid Bioassessment Protocols for Use in
Streams and Wadeable Rivers: Periphyton, Benthic Macroinvertebrates and Fish, Second Edition.

EPA 841-B-99-002. U. S. Environmental Protections Agency; Office of Water; Washington, D. C.

Burks, B. D. 1953. The mayflies, or Ephemeroptera, of Illinois. Illinois Natural History Survey Bulletin 26: 1-
216.

Dodd, H. R. 2000. Illinois Pilot Watershed Program. INHS Reports 364: 3.

Frison, T. H. 1935. The stoneflies, or Plecoptera, of Illinois. Illinois Natural History Survey Bulletin 20: 281-
467.

Hilsenhoff, W. L. 1987. An improved biotic index of organic stream pollution. Great Lakes Entomologist 20:
31-39.

Illinois Department of Energy and Natural Resources. 1994. The Changing Illinois Environment: Critical
Trends, Technical Report of the Critical Trends Assessment Project, Volume 3 Ecological Resources.
Illinois Department of Energy and Natural Resources, Springfield, IL, ILKENR/RE-EA-94/05.

Kart, 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.

28 p.

Lenat, D. R., D. L. Penrose. 1996. History of the EPT taxa richness metric. Bulletin North American
Benthological Society 13: 305-307.

Plafkin, J. L., M. T. Barbour, K. D. Porter, S. K. Gross, R. M. Hughes. 1989. Rapid Bioassessment Protocols
for use in Streams and Rivers: Benthic Macroinvertebrates and Fish. EPA 444/4-89-001. U.S.
Environmental Protections Agency; Office of Water; Washington, D. C.

Ross, H. H. 1944. The caddis flies, or Trichoptera, of Illinois. Illinois Natural History Survey Bulletin 23: 1-
326.

SAS. 1985. SAS User’s Guide: Statistics, ver. 5. SAS Institute Inc. 956 pp.

What are the discarded sites of CTAP terrestrial monitoring
telling us about Illinois habitats?

Rhetta Jack

Introduction

During the initial five years of monitoring (1997-2001), the Critical Trends Assessment Program (CTAP)
professional scientists have established and monitored 405 terrestrial sites for plants, birds, and insects. In the
process of establishing the monitoring sites, many sites and entire townships were discarded because they
failed to meet key requirements in the selection process. This report will cover the reasons why sites were
discarded and what this information may tell us about habitat loss, fragmentation, and the degradation of
Illinois ecosystems. Secondly, this report will investigate the role of the Illinois Land Cover Map and Atlas
(ILCMA) and the Illinois National Wetlands Inventory (INWI) in the site selection process.

CTAP sites were selected via a multi-randomized procedure and subject to monitoring criteria (Table 1).
Three main habitats are monitored statewide, which include forests, grasslands, and wetlands. The basic
monitoring unit for the CTAP professional monitoring is the township. Thirty townships with sites for each of
the three habitat types are monitored annually for a total of 90. For each of the three habitat types a full set of
townships were randomly ranked. Potential sites in forest and grassland townships were then picked and
randomly ranked using land cover data from the 1996 ILCMA. The ILCMA is based on data obtained from
satellite imagery 1991-1995. It delineates seven major land cover categories with 21 sub categories and
depicts what is covering the surface of the land (Illinois Department of Natural Resources 1995; 1996).

Table 1. Monitoring criteria

| Forests | Wetlands | rsstands

Monitoring Criteria trees in tract must average must be at least 50m x must be at least 50m x
at least 10 cm (4") dbh 10 m (500m*) in size 10 m (500m?) in size

stand must average <50% woody cover <50% woody cover
275% canopy cover (trees or shrubs) (trees or shrubs)

site must contain 75m 250% of wetland mowing frequency
radius of one forest plant species S3 times per year
community type (obligate or facultative)

230% plant cover if
open water present

Potential wetland sites were determined using the digital INWI. The INWI was generated from aerial
photography acquired from 1980-1987; most from 1983 (Suloway and Hubbell 1994; Illinois Department of
Natural Resources 1997). This database was not verified in the field and misses particular types of wetlands,.
such as small isolated wetlands (Levin et. al. 2002). Criteria used to identify potential wetland sampling sites
were based on wetland type and size. Specifically, wetlands suitable for CTAP monitoring are dominated by
emergent palustrine vegetation (i.e. rooted herbaceous hydrophytic vegetation such as sedges, rushes, forbs,
and grasses) and they are greater than two acres in size (Taft et al. 2002). There were 16,542 discreet
emergent wetlands larger than two acres known from within the State, totaling 166,256 acres (0.5% of
Illinois), with a mean size of 10.1 acres. These emergent wetlands were randomly ranked (1-indeterminate)
within each ranked township to establish sampling priority; field maps show their location (Molano-Flores
2002).

Townships are used in consecutive rank order and one primary site is monitored per township. Team members
visit the townships to assess the ranked sites in one of the three habitat types. The site must meet the
monitoring requirements (Table 1) at the designated center point, monitored by both botanists and
ornithologists. If the site does not meet those requirements then it is discarded and the next ranked site within
that township is evaluated, etc. Once all the ranked sites within a township are discarded, then the next
available township is assessed. Additionally, if a landowner denies permission or cannot be located with
reasonable effort, then that site is also discarded. A site becomes established for monitoring once the lowest
ranked acceptable site with landowner permission is found (Jack et al. 2002). It should be noted that
approximately 80% of all CTAP sites, used and unused, are on private property, which constitutes over 90%
of land in Illinois.

Methods

As part of the CTAP site evaluation process, team members record whether or not an individual site is
acceptable with the specific reasons. This information from 1997-2001 was used to develop a database for all
sites, and each site was classified to one of 10 main categories (Table 2). Several of the 10 main categories
concern habitat loss, fragmentation, and degradation. These categories are: no longer extant (i.e., habitat
destroyed), does not meet CTAP monitoring criteria (Table 1), township had no habitat, and too dangerous.
Grouping of the reasons for use or nonuse allows the data to be quantified and standardized across the three
habitat types.

Results

From 1997-2001, 405 or 90% of the base pool of 450 monitoring sites were selected and monitored, including
140 forests, 139 wetlands, and 126 grasslands. During this period, 16 forest and 46 wetland townships were
discarded because they contained no target habitat where sites could be picked. No grassland townships were
discarded for this reason.

For forests, 151 townships were assessed for sites. Eleven entire townships (7%) were discarded because no
designated site within that township met our monitoring criteria. Additionally, within assessed forest
townships, 277 potential sites were discarded because of the following categories: did not meet the
monitoring criteria, logistical reasons, were no longer extant, the landowner denied access, or dangerous
conditions existed (Fig. 1). Specific reasons are listed in Table 3.

A total of 203 townships were assessed for sites in wetland townships. Fifty-nine of those townships (29%)
had to be discarded because they contained no suitable wetlands for monitoring. Within the assessed
townships, 672 potential wetland sites had to be discarded mainly because they were no longer extant or did
not meet monitoring criteria (Fig. 2). Specific reasons are listed in Table 4.

Only 8 of 140 (6%) entire grassland townships assessed had to be discarded. However, within the assessed
grassland townships, 2,857 sites had to be discarded. Over half of those thrown out (58%) did not meet the
monitoring criteria and 35% were no longer extant (Fig. 3). The specific reasons are listed in Table 5.

Table 2. Ten main categories and subcategories

1. Sampled

site was used

2. No longer extant logged (forests only)
overgrazed (forest only)
drained (wetlands only)
bulldozed
developed

converted to row crops/monoculture

mowed too often
grown in by shrubs

too much canopy cover (W&G only)
too small

too degraded

not enough wetland vegetation (W only)
too little canopy (forests only)
overgrazed (wetlands and grasslands)
too wet (grassland only)

mowed yard

mowed roadside

golf course

cemetery

3. Did not meet
monitoring criteria

said no
unreasonable expectations

4. Landowner denies access

inaccessible
landowner not located
islands

inadequate time to assess

5. Logistics

meth lab
explosives

hazardous waste

water too deep

bull or aggressive cattle
dog
flooded

- 6. Too dangerous

used site with lower rank
quota of townships attained

7. Did not need

8. Extra bird site additional surveys for birds

no habitat for initial selection

9. Township had no habitat

10. Miscellaneous

ranks used in different year
data destroyed

assessed too late

unknown

Too dangerous
Landowner denial
No longer extant

29%
30%

Logistics

Does not meet criteria

0% 5% 10% 15% 20% 25% 30% 35% 40%

Percent of total

Figure 1. Forest sites discarded 1997-2001 (N= 274 sites).

Table 3. Reasons under the main categories for Forests

Categories Reasons Percent

Monitoring criteria too degraded
too little canopy cover
too small
grown in by shrubs

Logistics inaccessible
landowner not located
inadequate assessment time
No longer extant development
overgrazed
bulldozed
logged
TOW Crops
Landowner denial landowner denied access

Too dangerous methamphetamine lab
guard dog

Miscellaneous
Landowner denial

Too dangerous
Logistics

Does not meet criteria

No longer extant

0% 10% 20% 30% 40% 50% 60%

Percent of total

Figure 2. Wetland sites discarded 1997-2001 (N = 672 sites)

Too dangerous
Miscellaneous
Landowner denial
Logistics

No longer extant

Does not meet criteria

0% 10% 20% 30% 40% 50% 60%

Percent of total

Figure 3. Grassland sites discarded 1997-2001 (N = 2,857)

Table 4. Reasons under the main categories Wetlands

Categories Reasons Percent

No longer extant Tow crops
drained
developed
bulldozed

Monitoring Criteria too much canopy
too small
inadequate wetland vegetation
too degraded
overgrazed
grown in by shrubs
mowed

Logistics inaccessible
landowner not located
on islands
inadequate assessment time
site not located
permission too late

Too dangerous flooded
water too deep
hazardous waste

Landowner denial landowner denied access

Miscellaneous unknown
flash flood

Table 5. Reasons under main categories Grasslands

Categories Reasons Percent

Monitoring Criteria mowed/manicured
overgrazed
too small
too degraded
grown in by shrubs
too much canopy
too wet

No longer extant Tow crops
development
bulldozed

Logistics landowner not located
inaccessible
inadequate time to assess

Landowner denial landowner denied access

Miscellaneous data destroyed
assessed too late
point on map missing
rained out

Too dangerous flooded
explosives
bull/aggressive cattle
aggressive dog

Discussion

During the process of attaining the 405 monitoring sites, a total of 3,803 sites were discarded in all three
habitat types. The main reasons (88%) for sites to be unusable for CTAP monitoring are because they are no
longer representative of the target habitat due to destruction, degradation, and fragmentation. The primary two
categories representing these reasons are: no longer extant and did not meet monitoring criteria. Thirty-seven
percent of all sites discarded were no longer extant; Figure 4 shows the percentage of all the forest, grassland
and wetlands sites that were assessed and found to be destroyed. These sites were destroyed, or so altered that
they no longer functioned as the target habitat and natural recovery to functional habitat is unlikely. The
number one reason for this is conversion to row crop agriculture, accounting for 80% of all the no-longer-
extant sites. The second most common reason is development, accounting for 13%. Both of these reasons
should not be surprising since they show past and current land use changes in Illinois.

Of the discarded sites, 52% did not meet the monitoring criteria (Table 1). These sites are generally too
degraded by a variety of anthropogenic factors to accurately represent or function as the target habitat, at least
at the time of assessment. Some examples are overgrazing (removal of most groundcover or understory
vegetation), inadequate size, too much or not enough canopy cover, mowing more that three times a year,
partial development, and heavy logging (Table 2). One possible explanation for the high numbers of discarded
sites (over 50%) can be that CTAP monitoring criteria are overly selective and that good sites are discarded.
However, it should be pointed out that CTAP sites range from high quality to very poor, with most sites
somewhere in the mid to low range. As a whole, the sites can be viewed as average for their respective
habitats in the state. Finally, only 10% of the discarded sites were discarded because of reasons unrelated to
the habitat quality such as landowner denial and logistic problems.

40.0%
30.0%
20.0%
10.0%

0.0%

forest grassland wetland

Figure 4. Percentage of forest, grassland and wetland sites that were assessed but
discarded because the habitat was no longer extant

In addition to evaluating the status of Illinois habitats, the discarded CTAP monitoring sites data can be used
to evaluate the accuracy of the Land Cover Map and Atlas (ILCMA) and the Illinois National Wetlands
Inventory (INW]) in the site selection process. Of the three habitat types, forests sites are most intact, most
accuratcly depicted by the ILCMA, and have the fewest sites discarded (Fig. 1). Half of the discarded forest
sites were no longer intact or too degraded. In comparison, 85% of the discarded wetland sites and 93% of the
discarded grassland sites werc no longer intact or too degraded. Forest sites are well depicted on the ILCMA.
Only a handful of forest sites do not match the map, and those sites most likely were cleared in the last few
years. However, the ILCMA does not define the forest sites that have a full canopy but are disturbed
underneath the canopy by development and overgrazing. This is not in conflict with the stated limitations by
the authors of the ILCMA (Illinois Department of Natural Resources 1995; 1996).

In the case of grasslands the ILCMA is less useful for site selection for CTAP monitoring purposes; grassland
sites account for 75% of all the discarded sites. Part of the reason is that the ILCMA does not discern between
intact functional grasslands and regularly mowed monocultures. Ninety-three percent of discarded grassland
sites were thrown out because they do not meet the monitoring criteria or are no longer extant. Seventy
percent of the sites that do not mcct the monitoring criteria are mowed or manicured (e.g., yards, roadsides,
and golf courses). These differences only become apparent during the assessment of sites. In the case of no-
longer-cxtant sites, 85% of discarded grassland sites have been converted to row crop agriculture. This is due
to the conversion of pasture, hayficlds, and former CRP grounds to row crops. This is an ongoing process and
some of it has occurred since the publication of the ILCMA.

Finally, CTAP discarded sites data show that the INWI identifies wetlands where there are none and fails to
identify some existing ones. For example, a third of assessed wetland townships were discarded because none
of the picked wetlands were acceptable for CTAP. Half of the discarded wetland sites were no longer intact
and most (78%) had been converted to row crop agriculture. This is in line with other data concerning wetland
loss in Illinois and nationally (Havera and Suloway 1994). Some other reasons discarded wetlands were no
longer intact include draining, development, and bulldozing. Currently, it is not known whether or not these
differences between the sites and the depictions on the INWI were due to limitations of the INWI or if these
changes occurred subsequently to the INWI. Rarely is any detailed history of discarded wetland sites known.
However, in several instances landowners mentioned draining sites in recent years. Quite often new tile and
drainpipes are visible at the designated wetland site or heavy equipment is on site. A portion of the discarded
wetlands consists of wet areas in agricultural fields that are tilled in drier years. This evidence suggests some
of the sites were destroyed subsequent to the publishing of the INWI, nearly 20 years ago. Another portion of
assessed wetland sites, however, are not currently wetlands and appear to have been misidentified in the
INWI. Finally, a few sites have been discovered accidentally that are large wetlands and have not been picked
up on the maps. These do not include reclaimed or built wetlands since the INWI was published. Such
information would point to errors in the INWI, but the extent of these types of errors is difficult to ascertain as
assessment only occurs on selected sites. However, from 1997-2001 CTAP has assessed 1,060 wetlands

(6.4%) of 16,542 palustrine emergent wetlands in the state determined by the INWI. We believe that our
discarded site database is further evidence of the limitations associated with the INWI. A recent study of
wetlands in Lake County (Levin et al. 2002) has shown a similar pattern of missing or mis-identified wetlands
associated with INWI.

The discarded site database, in addition to the monitored site data, can provide insights into habitat loss,
habitat fragmentation and degradation of habitats in the state. The majority of sites that cannot be monitored
by CTAP scientists are either lost or so altered they no longer function as the habitat. Since assessment of
CTAP sites only occurs at the pre-designated sites picked using information derived from the ILCMA as well
as the INWI, the CTAP sites are subject to the limitations inherent to those products. However, the CTAP
discarded site data can provide information back to the ILCMA and INWI with on-the-ground assessment of
sites.

Literature Cited

Havera, S.P. and L.B. Suloway. 1994. Wetlands. in Illinois Department of Energy and Natural Resources. The
Changing Illinois Environment: Critical Trends. Technical Report of the Critical Trends Assessment
Project Volume 3: Ecological Resources, Springfield IL, ILENR/RE-EA-94/05(3).

Illinois Department of Natural Resources. 1995. Land Cover of Illinois Database. An electronic database.
http://www. inhs.uiuc.edu/igis/illinois/index.htm

Illinois Department of Natural Resources. 1996. Illinois land cover, an atlas. Illinois Department of Natural
Resources, Springfield, IL, IDNR/EEA-96/05.

Illinois Natural Resources Geospatial Data Clearinghouse. 1997. Illinois Wetlands Inventory database. An
electronic database. http://www. isgs.uiuc.edu/nsdihome/ISGSindex.html

Jack, R., S. Bailey, C. Carroll, and C: Dassler. 2002. Site evaluation, site selection, and documentation. in B.
Molano-Flores (ed.) Critical Trends Assessment Program Monitoring Protocols. Illinois Natural
History Survey, Office of the Chief Technical Report 2002-2, Champaign. 38 pp.

Levin, G.A., L. Suloway, A.E. Plocher, F.R. Hutto, J.J. Miner, C.A. Phillips, J. Agarwal, and Y. Lin. 2002.
Status and function of isolated wetlands in Illinois. Illinois Natural History Survey Special
Publication 23. 16 pp.

Molano-Flores, B. 2002. Critical Trends Assessment Program Monitoring Protocols. Illinois Natural History
Survey, Office of the Chief Technical Report 2002-2, Illinois Natural History Survey, Champaign. 38

pp-

Suloway, L. and M. Hubbell. 1994. Wetland Resources of Illinois: An analysis and Atlas. Illinois Natural
History Survey, Special Publication 15. 88 pp.

Taft, J., K. Robertson, S. Robinson, J. Brawn, C. Phillips, D. Niven, R.E. DeWalt, and L. Page. 2002. Habitat
criteria for study sites. in B. Molano-Flores (ed.) Critical Trends Assessment Program Monitoring
Protocols. Illinois Natural History Survey, Office of the Chief Technical Report 2002-2, Champaign.

38 pp.

Ornithological Report:
' The depauperate nature of the average Illinois bird community:
A CTAP study from 1997-2001

Steven Bailey and Rhetta Jack

Introduction

Habitat loss in Illinois since pre-settlement times has been well documented for all of Illinois’s major habitats
(Anderson 1970, Iverson et al. 1989, Suloway and Hubbell 1994). Fully greater than 99% of the original
prairie from pre-settlement times is now gone, and it is thought that only about 10% of pre-settlement
wetlands remain. At least one study conducted in Illinois looked at the drastic changes in habitats throughout
Illinois and the corresponding change in bird populations over a fifty-year period (Graber and Graber 1963).
In addition, the North American Breeding Bird Survey (BBS), a long-term (1966-2002), national, censusing
program has shown that many avian species, from a variety of habitats, are experiencing continued, long-term
declines, including population trends from Illinois (Robbins et al. 1989, Peterjohn et al. 1994, Pardieck and
Sauer 2000). However, the BBS program is not designed to determine causal factors of population change,
while the CTAP program is in a position to do so with the addition of botanical and GIS data and y anelyea
along with continued, long-term censusing.

The combination of the highly fragmented nature of Illinois’s habitats, along with large populations of the
parasitic Brown-headed Cowbird have been linked to these sometimes dramatic declines (Robinson et al.
1995). Most studies that have been conducted in Illinois to study how such factors are affecting the state’s
avifauna have concentrated their efforts on public landholdings, many of which are some of the larger or
higher quality tracts of remaining habitat left in the state (Herkert 1994, Thompson et al. 1995, Paine 1997,
Robinson et al. 1997). However, the great majority of Illinois land (>90%) is privately owned, and most
habitat patches on such lands are small, very fragmented, or isolated from larger, more conterminous tracts of
habitat. In this report, we will present the results from the initial five-year cycle of censusing done by CTAP
across Illinois. Data on habitat dependent (HD), area dependent (AD), threatened and endangered (T&E), and
exotic species will be presented for each of four habitats (forests, grasslands, wetlands, and shrub/scrublands).
A brief section will also deal with game/huntable species. Results from the initial five-year cycle of avian
censusing by the Critical Trends Assessment Program (CTAP), primarily on private land, further illustrate the
very degraded nature of Illinois habitats and the avian communities that they harbor.

Methods

From 1997-2001, CTAP omithologists censused 405 primary monitoring sites statewide (plus an additional 25
secondary, grassland sites). Of this total, 140 forest, 126 grassland (plus an additional 25 sites), and 139
wetland sites were censused from randomly selected townships in almost every county within IIlinois (Fig. 1).
Although study areas were located on a variety of public landholdings including city, county, state, and
federally owned properties, the overwhelming majority of sites were located on private landholdings.
Censuses were conducted from approximately the last week in May to the first few days of August.

For a complete description of CTAP avian and botanical protocols see Molano-Flores (2002). In general,
study areas vary from small woodlots and pastures to some of the largest tracts of grasslands and forests in the
state, as well as some of the better remaining examples of wetlands in the state. However, due to the
extremely fragmented and degraded nature of most CTAP study sites, the size of many grassland and wetland
study areas are towards the small end of acceptability. CTAP ornithologists attempt to place as many census
points in any given tract of land as temporal, topographical and logistical constraints will allow. Most sights
are only large enough for the placement of from one to three points, although some forest sites may contain as
many as 16 ten-minute point counts.

Bird Sites

e Forest
s Grassland \e
a Wetland

[___] Counties

10 0 1020 Miles
so

Figure 1. Locatian of monitoring sites across Illinois from 1997-2001

In wetlands an approximately 30 minute tape-playback of secretive, wetland bird species is played at each
census point. Grasslands and wetlands generally have many fewer census points than forest study sites due to
the small size of most sites. All birds detected are recone (unlimited distance), with distance and direction
noted for each individual.

As CTAP botanical protocols affect the size of many grassland sites, this in-turn has an effect on the species,
species richness, and density of grassland dependent birds found at the study areas. Some primary grassland
study sites are no more than 10-25 meter wide by 50-150 meter long grassy strips along roadsides or railroad
right of ways. In such instances, an additional, randomly chosen grassland of ten acres or more is picked
within the same township as the primary site. Similarities and/or differences between species make-up,
richness, and dominance, for the 126 primary grassland sites and the additional 25 sites where such additional
or “secondary” sites were needed, are discussed. In addition, sixty-five (65) additional wetland reference sites
were chosen within the randomly chosen townships from the wetlands identified on our maps as two acres or
greater in size. A large majority of these sites represent wetlands that were generally of much higher quality
as wetland bird habitat (i.e. the habitat generally contained more habitat dependent species), as compared to
the primary wetland site for that township. Detection rates are given for wetland dependent species for both
the primary sites and both primary and reference sites, to show how these mostly “higher quality” reference
sites compared with the randomly chosen “primary” sites.

For the purposes of this report, all species have been classified as habitat dependent species (HDS) and/or
area dependent species (ADS). Habitat dependent species (HDS) simply refers to those species which are
essentially tied to only one particular habitat type in which they occur (breed) in. For example, the Pied-
billed Grebe, a wetland dependent species (WDS), will virtually only be found in wetlands (with at least some
standing water), whereas the Red-winged Blackbird, which is not a WDS, will breed in wetlands, but can also
be found commonly breeding in grasslands. Area dependent (i.e. area sensitivity) species (ADS) refers to the
tolerance of a species to habitat fragmentation. That is, if a species is highly area sensitive, then it will require
a relatively large tract of its preferred habitat to breed in (i.e. will not generally use smaller, more fragmented
patches). Classifications of these ADS and HDS follow or are adapted slightly from Herkert (1993) and
Freemark and Collins (1992) grassland and forest bird species. Wetland Dependent Species (WDS) follow
Paine (1997), with additional WDS added by the authors for species not included in that study. In addition,
this report covers species richness (SR), or the number of bird species in a given area or habitat, for each of
the three habitats and the occurrence of state and federally threatened and endangered species (T&E) at CTAP
monitoring sites. Dominant species are also given for the former three categories for each habitat.

Results and Discussion

Forests

In forests, SR at any given site varied from 8-42 species. Again, it should be kept in mind that the number of
individual point counts varied from as little as a single census point to as many as sixteen, so to some degree
the number of species found reflects the size of the forest tract. However, this in turn is often a reflection on
the highly fragmented nature of Illinois’ forests. Sites ranged from a low of three (6 %) to 24 (50 %) forest
species out of a possible 48 forest HDS, both very low compared to areas of unbroken tracts of forest which
can be found in nearby Indiana and Missouri. However, both Robbins et al. (1989) and Freemark and Collins
(1992) have shown that small forest fragments (especially those less than 25 acres) in eastern North America
support few area sensitive or forest interior species. The most dominant species of these forest HDS at CTAP
sites were Blue Jay, Tufted Titmouse and Acadian Flycatcher. Such widespread and common species as
House Wren, Northern Cardinal, and Tufted Titmouse were the three most dominant species in forested sites
between 1997-2001. The former two species are also considered two of the least area sensitive woodland
species, with the latter only moderately sensitive. As has been shown in other studies, short distance migrants

and resident species are some of the more common forest species found in highly fragmented forest habitats
(Whitcomb et al. 1981, Ambuel and Temple 1983, Blake and Karr 1984, Hayden et al. 1985).

Concerning ADS (Table 1), 99 % of all forest sites had at least one moderately ADS, with a range of from 0-
13 (0%-87%) ADS. Dominant species in this category included the Tufted Titmouse, Red-eyed Vireo, and the
White-breasted Nuthatch. The nuthatch was present at 125 (89.2 %) of the 140 forest sites surveyed during
the period, up slightly from the 88.2 % noted at sites between 1997-2000 (Molano-Flores et al. 2002a). The
only forest species that had higher detection rates than the nuthatch were all species which exhibit low area
sensitivity (LAS), and include such common species as Downy Woodpecker (95.7%), Blue Jay (92.1%), and
Eastern Wood-pewee (92.1%). These three species were detected at 30 (100%), 29 (96.6%), and 29 (96.6%),
respectively, of the 30 forest sites monitored in 2001. For those most ADS, there was a range of from 0-5
(0%-39%) species at any given forest site, with fully 52 % (67) of sites where no highly ADS were detected.
Somewhat surprisingly, the Pileated Woodpecker, a relatively uncommon, statewide permanent resident, was
one of the three most dominant ADS species detected, at 29 (20.7%) sites, with the Ovenbird (26; 18.5%), and
Yellow-throated Vireo (47; 33.5%), being the two other most dominant ADS, both neotropical migrant
species. The very loud, ringing drum and call of the Pileated Woodpecker likely greatly facilitates the
detection of that species compared to virtually any other forest dwelling species. Two highly ADS which had
previously gone undetected in CTAP surveys were found during the 2001 monitoring, the Black-and-white
Warbler (at two sites), and the Hooded Warbler (at three sites). Detection rates for a large majority of both
HDS and ADS were up when compared with the 1997-2000 results (Molano-Flores et al. 2002a), with the
Wood Thrush (14.2%), Yellow-throated Vireo (9.0%), and Summer Tanager (6.9%), showing some of the
largest gains. These likely represent increases in the size and/or quality of CTAP sites monitored, rather than
any increase in the populations of these species. Unfortunately, the percentage of sites where Brown-headed
Cowbirds (a nest parasite) were detected also increased slightly, from the four year trend of 76.3% to the five-
year trend of 80.0%. The Brown Creeper, a statewide State Threatened (ST) species also was detected at an
increasing rate, having now been detected at 7 of 140 forest sites (5%), up slightly from the four-year trend of
3.6%.

Although the Wood Thrush and Red-eyed Vireo, only moderately ADS, have been detected at 57.8% and
70.7% of all CTAP forest sites between 1997-2001 (Molano-Flores et al. 2002a), many sites that seem
appropriate for these species have either contained few individuals of one or the other species, or none at all.
Both species are a common component of a large variety of forested habitats in Illinois including both
bottomland and upland forests, and the Red-eyed Vireo is often one of the most common species (Graber and
Graber 1963, Graber, et al. 1971, 1985). Both species are frequent hosts of the Brown-headed Cowbird in the
state as well (Graber et al. 1971, 1985, Robinson et al. 1995, Robinson et al. 1997). Studies in varying size
woodlots in Illinois have shown that the productivity of species such as the Red-eyed Vireo and Wood Thrush
"is likely not good enough (due to predation and parasitism) to achieve replacement from local breeding areas,
and such species are only “holding on” in these areas due to likely recruitment from outside areas from as far
away as 100 miles or more in southern Indiana and Missouri (Brawn and Robinson 1994, Brawn and
Robinson 1996). Localized extinctions or near extinctions of some populations of Red-eyed Vireos, Wood
Thrushes and likely other species such as tanagers and some species of wood warblers may become
increasingly common, and initial observations and census data by the CTAP ornithologists are beginning to
bear this out at several of the small, highly fragmented forest sites common in this study.

Grasslands
In grasslands, SR was similar to forested sites, with sites varying from 5-46 species. However, primary
grassland sites averaged fewer census points per site and species diversity in grasslands is much less than in

forests. Grassland associated species such as the Red-winged Blackbird were the dominant species found at
CTAP grassland sites, followed by the Common Yellowthroat and Song Sparrow (Table 2), also species

Table 1. Forest habitat dependent species detection rates
Total number of forests monitored from 1997-2001= 140

Neotropical migrants:

Statewide distribution

Species # sites detected % sites detected
Eastern Wood-pewee 129 92.1%
Great-crested Flycatcher 115 82.1%
Yellow-billed Cuckoo 105 75.0 %
Red-eyed Vireo 99 70.7 %
Wood Thrush 81 57.8 %
Scarlet Tanager 77 55.0 %
Yellow-throated Vireo 47 33.5%
Ovenbird 26 18.5%
American Redstart 6 4.2%
Cerulean Warbler 2 14%
Black-and-white Warbler 7) 14%
Mostly southern '/, of state, but present statewide

Species # sites detected % sites detected
Kentucky Warbler 48 34.2 %
Northern Parula 41 29.2%
Lousiana Waterthrush 27 19.2%
Prothonotary Warbler 10 7.1%
Worm-eating Warbler 9 6.4%
Hooded Warbler 3 2.1%
Mainly present only in north or south '/, of state

Species . #sites detected % sites detected
Summer Tanager 30 21.4% (S)
Yellow-throated Warbler 9 6.4 % (S)
Veery 3 2.1 % (N)
#Pine Warbler 2 1.4 % (S)
# = only found in large stands of mature pines

Short-distance migrant or resident:

Species # sites detected % sites detected
Downy Woodpecker 134 95.7%
Blue Jay 129 92.1%
White-breasted Nuthatch 125 89.2 %
House Wren 68 48.5%
Black-capped Chickadee 67 47.8%
Red-headed Woodpecker 61 43.5%
Carolina Wren 59 42.1% — mostly southern IL
Carolina Chickadee 51 36.4 %
Pileated Woodpecker 29 20.7 %
Cooper’s Hawk 13 9.2%
*Red-shouldered Hawk 7 5.0 % — mostly southern IL
*Brown Creeper 7 5.0 %
Brown-headed Cowbird 112 80.0 %

* = State Threatened Species

Table 2. Grassland HDS detection rates 1997-2001 (includes 126 primary sites and 25 additional
sites for 151 total). (Area dependency from Herkert et al. 1993)

Species # sites detected % sites detected Area dependency
Eastern Meadowlark 94 62.2 moderate
Dickcissel 68 45.0 low
Grasshopper Sparrow 31 20.5 moderate
Vesper Sparrow (N) 18 11.9 low
Savannah Sparrow (N) 15 9.9 high
Bobolink (N) 15 9.9 high
Sedge Wren 12 7.9 moderate
Henslow’s Sparrow** 8 sv high
Western Meadowlark (NW) 5 3.3 moderate
Northern Harrier** 2 iS high
Upland Sandpiper** ] 0.6 high
Short-eared Owl** 0 0.0 high
Grassland associated species

Red-winged Blackbird 130 86.0 low
Common Yellowthroat 120 79.4 low
Song Sparrow 113 74.8 low
Field Sparrow 99 65.5 low
Loggerhead Shrike* 4 2.6 ?
Brown-headed Cowbird 94 62.2 low

(N) = distribution in northern '/, of state only

(NW) = distribution mainly in northwestern '/, of state
* = State Threatened Species

** = State Endangered Species

? = likely moderate to high

common to other habitats in Illinois (wetland, shrub and edge areas). Of those species that are grassland
HDS, the Eastern Meadowlark was by far the most common and widespread at CTAP monitoring sites,
followed by the Dickcissel, Sedge Wren and Savannah Sparrow which were much less common and
widespread. The Bobolink, historically a characteristic and common-to-abundant bird of the Grand Prairie
Region of Illinois was the dominant species at only one site. Thirty-four (30%) of the 126 primary grassland
sites contained no grassland HDS. Of the ADS species, four species are considered moderately area
dependent. Sites ranged from 0-4 (0%-100%) of these species, but with only one site having all four
moderately ADS. Fifty-one (40.5%) sites had no moderately ADS. Of the six possible grassland species that
are considered highly ADS in the state, sites ranged from having only 0-2 (0%-33%) of these species, with the
dominant species in this category being the Savannah Sparrow, followed by the Bobolink and the State
Endangered (SE) Henslow’s Sparrow. A large majority of the 126 primary CTAP grassland sites, 105 (83%),
contained no ADS. Illustrating the small and highly fragmented nature of most Illinois grasslands, the four
most common grassland species detected at CTAP sites all only show either low or moderate area sensitivity
(Table 2).

Due to botanical protocols in choosing sites, many of the CTAP grassland sites are only small and/or narrow

strips of grassland habitat, which are not acting as functioning ecosystems for grassland bird species. For this
reason, in sites of less than ten acres, another (secondary) site is chosen for CTAP omithologists (only) to

monitor. In these larger areas, a significant trend has been noted. In 15 (60%) of the 25 additional or
secondary sites that were chosen, at least one moderately ADS was noted where the primary site had no
moderately (or highly) ADS. In addition, 9 (36%) of these 25 secondary sites also contained at least one
highly ADS when the primary site had none, while in none of the remaining 16 sites did the primary site have
any highly ADS. In 4 of the 9 instances, these were state threatened and endangered species.

When both primary and secondary sites are considered (for a total of 151 sites), not much changes in a list of
commonality, with the most common species remaining the Eastern Meadowlark, found at 94 (62.2%) of the
151 sites. Dickcissel remained common, being detected at 68 (45%) sites, and Grasshopper Sparrow being
the next most common HDS with 31 (20.5%) sites registering this species. Although the Brown-headed
Cowbird parasitizes grassland species much less commonly than forest species in Illinois (Robinson et al
2000), it was recorded at 94 (62.2%) sites, though many of these were as fly-overs. Only ten (7.9%) of 126
primary grassland sites contained T&E species. Nine sites had a single T&E species and one had two T&E
species. Other than the SE Henslow’s Sparrow, recorded at 8 (5.2%) of the 151 sites, only two other SE
grassland HDS were recorded. Two (1.3%) Northern Harriers (SE) and one (0.6%) Upland Sandpiper (SE)
was noted in the combined 151 primary and secondary grassland sites. However, the grassland associated
Loggerhead Shrike (ST) was also detected at four (2.6%) sites as well.

Wetlands

SR in wetland sites was comparable to both forest and grassland sites with a range of 6-40 species. Of the
four most dominant species found at CTAP wetland sites, none were wetland HDS (Table 3). The percentage
of wetland sites (49.7% or 69 of 139 sites) with wetland HDS remained relatively the same between 1997-
2001 as compared to the four-year trend (1997-2000) of 46.7% (51 of 109 sites) (Molano-Flores et al. 2002a).
However, as with forest sites, detection rates for a large majority of the wetland species was somewhat to
substantially higher than the four-year trend (Table 3). This table shows twenty-four wetland HDS, five
wetland associated species, and the parasitic Brown-headed Cowbird, along with a comparison of the four-
year (1997-2000) trend with the first full five-year cycle (1997-2001), as well as sixty-five additional, mostly
high quality reference sites.

There are more T&E bird species possible in wetland habitats than in all other habitats combined in IIlinois
(18 wetland, 5 grassland, 4 forest). However, there were only slightly more T&E species noted at CTAP
wetlands, compared to grassland or forest sites, with only 13 (9.3%) of 139 primary wetland sites recording
T&E species. Ten sites had a single T&E species, two had 2 T&E species, and one site had three.
Interestingly, for 12 T&E species found at wetland sites, reference sites and primary sites each had six species
where the majority of T&E species found were at those respective types of sites (Table 3). Of note though
were King Rail (SE) which was found only at five reference sites and Yellow-headed Blackbird (SE) found at
only two reference sites. However, the only sites where American Bitterns (2), Snowy Egrets (1), and Yellow-
crowned Night-Herons (1) were recorded were at primary wetland sites.

Although wetland HDS do not appear to exhibit any area dependency, the number and quality (including and
especially water depth) of other wetlands in a landscape seem to be just as important in attracting HDS. Of
the 139 primary wetland sites, the range of HDS has only been 0-8 (0%-20%), by far the lowest of the three
study habitat types. Unlike grasslands, wetlands have a relatively long list of HDS (Table 3). However, fully
50% (70) of the 139 primary wetland sites had no WDS. Of those that did, there were a few clear dominants,
including Wood Duck, Mallard, and Willow Flycatcher. Canada Goose would likely replace Willow
Flycatcher, but due to their early breeding season, many adults and young have left their breeding sites before
CTAP censuses begin. CTAP wetlands are dominated by wetland associated species (Table 3), with the five
main wetland associated species all having detection rates higher than the any wetland HDS. The extent to
which Brown-headed Cowbirds use wetland habitats has been little studied. Cowbirds were noted at only 74
(53.2 %) of the primary wetland sites, the lowest percentage of the three main habitat types.

Table 3. Wetland habitat dependent species (HDS) detection rates

Species

Great Blue Heron

Wood Duck

Mallard

Green Heron

Willow Flycatcher

Canada Goose

Swamp Sparrow (N)

**Little Blue Heron (S)

Marsh Wren (N)

*Least Bittern

American Coot
**Black-crowned Night-Heron
Virginia Rail (N)

*Pied-billed Grebe

*Common Moorhen

Sora (N)

** American Bittern (N)

** Sandhill Crane (N)
**Yellow-crowned Night-Heron (S)
**Snowy Egret (S)

**King Rail

**Yellow-headed Blackbird (N)
Common Snipe (N)

#Mute Swan (N)

Red-winged Blackbird
Common Yellowthroat
Song Sparrow

Indigo Bunting
Killdeer

Brown-headed Cowbird

# sites
59
42
41
33
24
19
15

SCOCOOCO COF RRP NNNNNW SUD ~

_
No
_

119
118
118

1997-2001 (139 sites)

% sites
42.4%
30.2%
29.4%
23.7%
17.2%
13.6%
10.7%
5.0%
4.3%
3.5%
2.8%
2.1%
1.4%
1.4%
1.4%
1.4%
1.4%
0.7%
0.7%
0.7%
0.0%
0.0%
0.0%
0.0%

87.0%
85.6%
84.8%
84.8%
43.8%

53.2%

N = population mainly/entirely in northern '/, of state
S =population mainly/entirely in southern '/, of state

= statewide distribution
* = State Threatened
** = State Endangered
= Introduced/feral species

# sites

3]
16
19
17
15

SCOCOOCOOCOOOCO OF NN YE NN SUNY ON

1997-2000 (109 sites)
% sites

28.4%
14.7%
17.4%
15.6%
13.8%
6.4%
8.3%
0.9%
4.6%
3.7%
1.8%
1.8%
0.9%
1.8%
1.8%
0.9%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%

1997-2001 (+ ref.)

# sites

9]
63
59
57
37
31
31

—_
SON UK KS NNN ND AIA DHW ~1 CO

186
175
176
167

104

% sites
44.6%
30.8%
28.9%
27.9%
18.1%
15.1%
15.1%
3.9%
8.3%
2.9%
3.4%
3.4%
2.9%
2.4%
2.4%
2.4%
0.9%
1.0%
0.4%
0.4%
2.4%
0.9%
0.0%
0.0%

91.1%
85.6%
86.2%
81.8%
47.0%

50.9%

Shrub/scrubland

Shrub/scrubland habitats are not studied in the CTAP program, per se, although shrub/scrubland bird species
are often noted in censuses of all three habitats, especially wetlands and grasslands. Most grasslands have a
shrub component to at least a small portion of the site. One of the most common of shrubland species in the
state, the Brown Thrasher, was detected at 35 (23.1%) of all (151) grassland sites. Three shrubland species,
all neotropical migrants, which were detected with some frequency at CTAP grassland sites included the Blue
Grosbeak found at nine (5.9%), Yellow-breasted Chat noted at 21 (13.9%), and the Orchard Oriole detected at
25 (16.5%) of all 151 grassland sites. The Bell’s Vireo, a species declining fairly precipitously over much of
its range (Pardieck and Sauer 2000), in part due to heavy Brown-headed Cowbird parasitism, was detected
with some regularity at both grassland and wetland CTAP study sites. This species was noted at 13 (8.6%) of
151 grassland sites and 12 (8.6%) of 139 primary wetland sites.

Exotic/Introduced Species

Exotic bird species are not the concer that many non-native plant species are to habitats throughout the state.
However, CTAP censuses show that the two dominant species, European Starling and House Sparrow, remain
relatively common in most Illinois habitats, and still present a problem to cavity nesting bird species (Table
4). This will likely only increase, especially as forests continue to become even more fragmented. As of yet,
Mute Swans do not present the potential problem that they might present if they were to spread and become
more common outside of their northeastern Illinois stronghold. They have yet to be detected at one CTAP
site. Other species which have not been detected at a CTAP site, include the Monk Parakeet, which is
currently only established in a relatively few urban/surburban areas in (mainly) Cook County and the Eurasian
Collared-Dove. The latter species, although yet to be detected on CTAP censuses, will likely become at least
a fairly common species throughout the state in the not too distant future. With the breadth of CTAP
monitoring, this program should track the spread and colonization of this species across the state very nicely.

Game/huntable species

Table 5 presents detection rates for some of the avian species most sought after by the hunting community, but
also see Table 3. Due to the timing of CTAP censuses, several game species are under-represented in the
table, and so detection rates are somewhat-to-substantially lower than true population levels would otherwise
indicate. They are given here for reference value only. In the case of the American Crow, most detections in
all habitats usually represent calling birds heard at some distance (often greater than 300 meters) or are birds
flying over the habitat and not actually utilizing that habitat. Pheasant and Northern Bobwhite are also heard
at some distance, again many times greater than 300 meters from the census point where they are registered.

Table 4. Introduced/exotic species detection rates

Forests (140 sites) Grasslands (151 sites) Wetlands (139 sites)
Species # sites % sites # sites % sites # sites % sites
Mute Swan 0 0.0%
Ring-necked Pheasant 21 13.9% 14 10.7%
Gray Partridge 0 0.0% 0 0.0%
Rock Dove 2 1.4% 15 9.9% 14 10.7%
European Starling 18 12.8% 94 62.2% 55 39.5%
House Finch 20 14.2% 28 18.5% 20 14.3%
House Sparrow 10 , 7.1% 55 36.4% 24 17.2%
Eurasian Tree Sparrow 0 0.0% 5 3.3% 2 1.4%

Table 5. Game/huntable species (* = does not apply)

Forests (140) Grasslands (151) Wetlands (139)
Species # sites % sites # sites % sites # sites % sites
American Crow 113 80.7 103 68.2 87 62.5
Mourning Dove 83 59.2 109 1PM 85 61.1
Northern Bobwhite 37 26.4 66 43.7 37 26.6
Wild Turkey 30 21.4 3 1.9 4 2.8
Ring-necked Pheasant 4 _ 21 13.9 14 10.7
Gray Partridge * i 0 0 0 0

Wild Turkey are also likely found in a much larger percentage of CTAP sites than detection rates suggest, but
due to this species wary nature many birds likely go unnoticed. Of interest to at least some hunters, the
virtual lack of detection of the Gray Partridge could mean that this species may be on its way out in IIlinois.
Other anecdotal evidence points to this, as very few sightings of this species have been made in recent years
in their former range within the state. Again, the breeding season of the resident population of Canada Geese
is virtually over by the time most CTAP censuses are begun. Rails, snipe and American Woodcock are all
under-detected due to their secretive nature, timing of breeding season, or other issues of conspicuousness or
a combination of all of these factors.

Conclusions

As with the four-year trends, results from the first five-year cycle show that highly ADS/HDS in all of
Illinois’s habitats have low to very low detection rates at CTAP study sites. This is a direct reflection on both
the small size and fragmented nature of all of Illinois’ habitats. Grassland sites continue to be the most
depauparate, with few T&E species noted other than the Henslow’s Sparrow (which is currently experiencing
a population boom in Illinois), almost as few ADS, and even smaller numbers of HDS. Data collected by
CTAP biologists show that this habitat is in the most serious need of restoration, and if some species of
grassland birds are to continue to exist in the state, the quality and size of sites needs to improve dramatically
in the years to come. The nature of the additional grassland sites that are chosen when primary sites are not
large enough show that size is one of the main considerations grassland species use when determining what

_ sites to attempt to breed in. Many of these extra CTAP sites are pastureland, hayfields, or set-aside lands, and
almost all contain a higher avian species diversity, including the presence of both HDS and ADS, than
randomly chosen primary sites. As botanical data show the poor quality and presence of exotic species at
grassland sites (Molano-Flores et al. 2002b), the CTAP program also shows the higher importance to
grassland birds of plant structure, rather than native or otherwise “higher quality” of plant species present at a
site.

Data collected on habitat and area sensitive species at CTAP sites show Illinois forests to be in the best
condition, at least for birds, of the three habitats. Even many of the smallest tracts still contain at least a
remnant number and fair diversity of typical forest bird species. Although a few species such as the Cerulean
(in trouble over much of its range), Hooded, and Black-and-white Warblers are detected at extremely low
rates, most other declining neotropical migrant species and short-distance migrants are still occurring in
Illinois forests in relatively good numbers. It will be interesting to see what kinds of long-term trends are
noted for species like Wood Thrush and Red-eyed Vireo, which currently are still being detected at most sites,
but which could have a precipitous decline (due to fragmentation, predation and cowbird parasitism) if large
areas of intact forests in neighboring states begin to become more and more fragmented. Although Brown-
headed Cowbird detection rates remain high, it is likely that they are actually even higher at CTAP sites than

the data indicate, with virtually 100% of all forest sites not only having high detection rates but also likely
having high cowbird parasitism rates (see Robinson et al. 1995 and Robinson et al. 1997). Likely the only
reason many CTAP sites have yet to detect cowbirds is the fact that many sites have just one or two census
points. With more points and/or repeat visits, cowbirds would likely be found at all sites. Cowbird/host ratios
at many of the larger CTAP sites will likely provide even more evidence to back up findings of very high
cowbird/host ratios (and high parasitism rates) found by other studies in Illinois at higher quality sites. Using
GPS and GIS data, we should be able to determine what size of a habitat patch may be needed to lower
cowbird/host ratios in a larger habitat matrix of forest (or grassland) habitat.

Data collected at Illinois wetlands over the first five-year cycle show that there are a few HDS that are still
relatively common in many Illinois wetlands, especially at sites with at least some standing water throughout
much of the breeding season. Water depth, or lack thereof, is one of the primary reasons that many CTAP
wetland sites lack WDS of birds. This is also likely the main reason for the large number of wetlands that
lack even one wetland HDS. Quality reference sites will become more important to give us a good
comparison to judge just how poor many of the CTAP sites appear to be, both in quality and species diversity,
especially in judging the fate of T&E species that are currently found very infrequently at CTAP sites.

Although the CTAP study was not set up to study all habitats within the state, some inferences will be able to
be made regarding a few other habitats where bird use is concerned, especially shrubland habitats. Bird data
from the three main habitat types will also be able to show general trends in the increase, decrease, spread, or
lack thereof, of almost-all of Illinois’ introduced/exotic avian species. Although relatively few threatened and
endangered species are detected at CTAP study sites, this program likely documents more T&E species than
any other single program or agency within the state, and should provide a good indication as some species
rebound, and a early warning system for species (like the Cerulean Warbler) that continue to decline.

Literature Cited

Ambuel, B., and S.A. Temple. 1983. Area-dependent changes in the bird communities and vegetation of
southern Wisconsin forests. Ecology 64:1057-1068.

Anderson, R.C. 1970. Prairies in the prairie state. Transactions of the Illinois State Academy of Sciences
63:214-221.

Blake, J.G., and J.R. Karr. 1984. Species composition of bird communities and the conservation benefit of
large versus small forests. Biological Conservation 30:173-187.

Brawn, J.D., and S.K. Robinson. 1996. Source-sink population dynamics may complicate the interpretation
of long-term census data. Ecology 77:3-12.

Brawn, J.D., and S.K. Robinson. 1994. Forest birds in Illinois: Changes in abundances and breeding ecology.
Erigenia 13:109-116.

Freemark, K., and B. Collins. 1992. Landscape ecology of birds breeding in temperate forest fragments. Pp.
443-454 in J.M. Hagan and D.W. Johnson, eds., Ecology and Conservation of Neotropical Migrant
Landbirds. Smithsonian Institution Press, Washington, D.C. 609 pp.

Graber, R.R., and J.W. Graber. 1963. A comparative study of bird populations in Illinois, 1906-1909 and
1956-1958. Illinois Natural History Survey Bulletin 28(3):383-528.

Graber, R.R., J.W. Graber, and E.L. Kirk. 1971. Illinois birds: Turdidae. Illinois Natural History Survey
Biological Notes 75. 44 pp.

Graber, J.W., R.R. Graber, and E.L. Kirk. 1985. Illinois birds: vireos. Illinois Natural History Survey
Biological Notes 124. 38 pp.

Hayden, T.J., R.L. Clawson, and J. Faaborg. 1985. Estimates of minimum area requirements for Missouri
forest birds. Transactions, Missouri Academy of Science 19:11-22.

Herkert, J.R., R.E. Szafoni, V.M. Kleen, and J.E. Schwegman. 1993. Habitat establishment, enhancement and
management for forest and grassland birds in Illinois. Division of Natural Heritage, Illinois
Department of Conservation, Natural Heritage Technical Publication # 1, Springfield, Illinois. 20 pp.

Herkert, J.R. 1994. Breeding bird communities of midwesterm prairie fragments: The effects of prescribed
burning and habitat-area. Natural Areas Journal 14:128-135.

Iverson, L.R., R.L. Oliver, D.P. Tucker, P.G. Risser, C.D. Burnett, and R.G. Rayburn. The forest resources of
Illinois: an atlas and analysis of spatial and temporal trends. Illinois Natural History Survey Special
Publication 11. 181pp.

pp.

Molano-F lores, B., R. Jack, and S. Bailey. 2002a. Ornithological Report: Habitat use and area sensitivity of
birds in a highly fragmented Illinois landscape. in Critical Trends Assessment 2001 Report, Illinois
Department of Natural Resources, Springfield, IL. 64 pp.

Molano-Flores, B., J. Ellis, C. Carroll, and G. Spyreas. 2002b. Botanical Report: Tracking non-native
species in Illinois. in Critical Trends Assessment 2001 Report, Illinois Department of Natural
Resources, Springfield, IL. 64 pp.

Paine, C.R. 1997. Abundance and nesting productivity of wetland-dependent birds in northeastern Illinois.
Final Project Report to the U.S. Fish & Wildlife Service. 78pp.

Pardieck, K.L., and J.R. Sauer. 2000. The 1995-1999 summary of the North American breeding bird survey.
Bird Populations 5:30-48.

Peterjohn, B.G., J.R. Sauer, and W.A. Link. 1994. The 1992 and 1993 summary of the North American
breeding bird survey. Bird Populations 2:46-61.

Robbins, C.S., J.R. Sauer, R.S. Greenberg, and S. Droege. 1989. Population declines in North American
birds that migrate to the neotropics. Proceedings of the National Academy of Science 86:7658-7662.

Robinson, S.K., F.R. Thompson III, T. M. Donovan, D. R. Whitehead, and J. Faaborg. Regional forest
fragmentation and the nesting success of migratory birds. Science 267:1987-1990.

Robinson, S.K., J.D. Brawn, and J.P. Hoover. 1997. Effectiveness of small nature preserves for breeding

birds. Pp 154-188 in Conservation in highly fragmented landscapes (M.W. Schwartz, ed.). Chapman
and Hall, New York, N.Y.

Robinson, S. K., J. P. Hoover, and J. Herkert. 2000. Cowbird parasitism in a fragmented landscape: effects of
tract size, habitat, and abundance of hosts. Pages 280-297 in J. N. M. Smith, T. L. Cook, S. I.
Rothstein, S. K. Robinson, and S. G. Sealy, editors. Ecology and management of cowbirds.
University of Texas Press, Austin, Texas, USA.

Suloway, L., and M. Hubbell. 1994. Wetland resources of Illinois: an analysis and atlas. Illinois Natural
History Survey Special Publication 15. 88 pp.

Thompson III, F.R., S.K. Robinson, D.R. Whitehead, and J.D. Brawn. 1996. Management of central
hardwood landscapes for the conservation of migratory birds. Pp 117-143 in Management of
midwestern landscapes for the conservation of neotropical migratory birds (F.R. Thompson III ed.).
1995 December 5; Detroit, MI. Gen. Tech. Rep. NC-187. St. Paul, MN: U.S. Dept. of Agriculture,
Forest Service, North Central Forest Experiment Station. 207 pp.

Whitcomb, R.F., C.S. Robbins, J.F. Lynch, B.L. Whitcomb, M.K. Klimkiewicz, and D. Bystrak. 1981.
Effects of forest fragmentation on avifauna of the eastern deciduous forest. Pages 123-205 in R.L.
Burgess and D.M. Sharpe, editors. Forest island dynamics in man-dominated landscapes. Springer-
Verlag, New York, Ney York, USA.

ForestWatch Fall 2001 - Spring 2002
Matt Buffington
Fall Tree Survey

Data from 58 sites were submitted in fall 2001, 22 of which had never been monitored before. A total of 157
sites have been monitored since spring 1998. The tree composition of the forests from 2001 was similar to
other years. As shown in Table 1, oaks and hickories, the species that historically dominated the majority of
Illinois forests, contributed heavily to the total basal area of trees (51% of total basal area) and were fairly
abundant (27% of total abundance). This is in contrast to many understory species that contribute relatively
little to the total basal area as compared to their contribution to total abundance. For instance, flowering
dogwood, persimmon, sassafras, hawthorn and ironwood combined for only 3% of the total basal area but
11% of the total abundance. Persimmon is somewhat common in the southern third of the state and was tenth
in abundance and sixteenth in importance. This was the result of two sites that are heavily dominated by this
species. In 1999, the most recent year many of these sites were previously monitored, persimmon was tenth in
abundance and thirteenth in importance, again as a direct result of these same two sites.

Table 1. Importance values of trees from ForestWatch sites

Tree Taxa Relative Abundance Relative Basal Area Importance Value
White Oak 4.8 18.5 23.3
Hickory spp. 12.3 10.3 22.6
Sugar Maple 10.2 13 17.5
Ash spp. 6.9 dol, 14.6
Black Cherry 9.1 43 13.4
Red Oak 2.3 48S bi
Slippery Elm TES: 3.6 10.9
American Elm 6.5 3:3 9.8
Hackberry 3.6 2.9 6.5
Black Oak 2.4 3.1 Ses,
other 65 taxa 34.5 30.3 64.8

Thirty-one of the 58 sites monitored in 2001 also had tree data collected in fall 1999. There were a few sites
that had some relatively large differences in the abundance and basal area values, but on the whole the aver-
age number of trees and average basal area for each site was similar between years.

The average diameter of all the trees monitored was 20.7 cm, or slightly over eight inches. Nearly 80% of the
trees were <20 cm (Fig. 1). This is a rather small diameter and indicates that Illinois forests are mainly second
growth with large amounts of moderately sized trees. This can be contrasted with the tree height data col-
lected in the spring from these same sites. In the spring, 15 of the tallest trees are measured for height and
diameter. The average diameter of these tall trees was 34.2 cm (13.6 inches). Although the tallest trees are not
necessarily the largest in diameter, they are certainly among the largest in the forest.

# trees perha

500
(@)
E ¢€
EF oimbige acne eile: abs! &
Oy 72 MMB VS -S
at dincin £2, wneslnguillt wick Oeeenteh
Oo ooo eo N
idl ON PN = RP i”

diameter class

Figure 1. Number of trees by diameter class

Fall Shrub Survey

Forty of the 58 sites (69%) had onc or morc invasive shrub taxa present. Multiflora rose was the most com-
mon invasive shrub in terms of the number of sites in which it was recorded, while shrub honcysuckles
averaged the greatest number of stems (Table 2). The ratio of invasive shrub stems to the total amount of
shrub stems indicates the degree to which these species are dominating a forest. Among the sites with shrubs,
the average ratio of invasive shrubs to total shrubs was 63%, with nearly half of the sites 290% (Fig. 2 and 3).
The average ratio has been around 50% or greater every year of monitoring. Considering ForestWatch looks
for approximately 18 invasive shrub species (including 12 shrub honeysuckle and two buckthorn species) and
there are over 250 different shrub species in Illinois, the data indicate that a relatively small number of
invasive e shrub taxa are a major problem in forests.

Thirty sites at which shrubs were recorded in 2001 also had shrub data in 1999. Although the average number
of stems for shrub honeysuckle, buckthorn, multiflora rose, and gooseberry was greater in 2001, the differ-
ence was not significant. The number of sites with gooseberry and Japanese honeysuckle in the shrub survey
can be comparcd to the values from the quadrat survey in the spring. These two spccies were added to the
spring survey because they often do not reach one meter in height, the minimum hcight required for shrubs to
be included in the fall survey, and thus may be undercounted in the fall shrub layer survey. The other five
invasive taxa are more likely to be at Icast one meter tall. There were 42 sites with fall and spring data.

Table 2. Amount of shrubs and vines recorded at ForestWatch sites

Taxa # Sites Average # Stems (m?/ha)
Honeysuckle Shrub 18 564
Buckthorn 13 524
Highbush Cranberry 3 7
Autumn Olive 2 21
Multiflora Rose 29 422
Gooseberry 17 414
Other Shrubs 40 479
Japanese Honeysuckle 3 29
Other Vines 37 es 263

Ratio to Native Shrubs

Number of Stems

edie aiae ag
cedanmaat |

ania

5
a,28
ae
233°
on Ss
-2@e@@

ForestWatch, Fall 2001 Data

Figure 2. Invas

# of sites
°

0
x = x x 3S
w [@) w oO oO
Ni Ole On “|
v To) ro) nn A
N wo ~~

Figure 3. Invasive shrub ratios

Nine sites reported gooseberry in both fall and spring, while three sites reported it just in the fall and another
three reported it just in the spring. Additional monitoring will determine if gooseberry is being undercounted
in the fall shrub survey. There was a bigger difference with Japanese honeysuckle. Seven sites had this vine in
the spring and onc different site reported it in the fall. Although this vine docs grow vertically up canopy trees
and can overwhelm entire arcas within a forest, it is often seen growing along the ground, covering extensive
arcas of the forest floor and understory trees and shrubs. This probably accounts for the increasc in the reports
during the spring survey comparcd to the fall shrub survey.

Fall Dogwood Anthracnose Survey

Dogwood anthracnose continues to be a problem in many forests where flowering dogwoods are found.
Fifteen sites contained flowcring dogwoods and eight of these reported dogwood anthracnose (Fig. 4). This
percentage is higher than the roughly 30% that has been reported from ForestWatch sites in previous years.
Whether this reflects a spread of the disease or inconsistency in identifying the disease is uncertain. The
majority of the sites monitored in 2001 that were also monitored in 1999 did not have flowering dogwoods.
However, there were some yearly differences in reporting for some sites. At some sites, flowering dogwoods
may be reported in onc year but not another. Year-to-year differences in the recording of flowering dogwood
trees and saplings are not unexpected. Slight changes in the placement of the transect lines can result in one or
more trees being recorded in onc ycar and not another. What is most important is properly identifying and
reporting the disease when it is present and tracking its spread over time. This information has only just begun
to be gathered as morc sites are revisited each year.

Fall Maple Takeover Analysis

Invasive shrubs and species-specific discases are forces altering the composition of Illinois forests. Another
factor that is affecting the makeup of forests is changes in management. Many forests, particularly oak-
hickory uplands, are changing since the removal of fire from the landscape. The result is an increase in fire
intolerant but shade tolerant species, especially those associated with mesic-uplands like sugar maple. Nine of
the 37 oak-hickory uplands showed some degree of maple takeover. This was determined by looking at the
number of trees in cach size class for oaks, mapics, and hickories. Sites that had a relatively higher ratio of
maples as compared to oaks and hickorics in the smaller diameter classes may be expericncing maple take-
over. Some sites are borderline, such as a site from Kane County (Fig. 5). Other sites exhibit much greater
evidence of takeover, as in Figure 6 which is a site from McDonough County. Finally, other sites have a major
problem, as evidenced in Figure 7, a site from Coles County.

e Dogwood, no antnracose
4 Dogwood. signs of anthracnose

Figure 4. Presence of dogwood anthracnose

(" Sugar Maple

#trees per
@ Oak

hectare

B tett wen eas er oe Et €

CSTE ON cee eS Pa ee

DiS tig (Sse Sa” SoS

Y neaieve Maw 89 cheSualese My. <0

We cQyOwuenica ie iit
SIN (CO oS ip

diameter class

Figure 5. Example of a site that may exhibit maple takeover in the future

250 -
200 -

QO Sugar Maple
@ Oak
ti Hickory

# trees per 150-
hectare 400 -

50 ;
0

30-40 cm
50-60 cm
>60 cm

E
3S
=)
“
s)
=

40-50 cm}

diameter class

Figure 6. Example of a site that currently shows signs of maple takeover

—=Sugar Maple
# trees per Oak
hectare
it Hickory

Ee
G
Yo)
Vv

5-10.cm
10-20 cm
20-30 cm
30-40 cm
40-50 cm
50-60 bi
>60 cm

diameter class

Figure 7. Example of a site that clearly has a major problem with maple takeover

Fall Correlations

Geography plays an important role in plant distribution and other ecological factors. This is especially true in
Illinois which spans roughly 400 miles from north to south, has a very large urban areca in the north, extensive
agriculture in the central region, and morc forest and rolling topography in the south. Therefore it is useful to
look at the data in terms of where in the state the sites are located. In vegetative studics, analyses are often
based on dividing the state into thirds and cxamining differences among these geographical regions (Fig. 8).
Data from the fall 2001 and spring 2002 monitoring cycle came from 22 sites in the northern zonc, 22 in the

central, and 14 in the south.

Figure 8. The three biogeographical regions used in the analysis

Several indices were significantly correlated to the zone in which sites were located as shown in Table 3. The
correlation between zone and the shrub data generally relate to the large amount of buckthorn found in the
northern part of the state. Over 99% of the buckthorn was in the north zone. Honeysuckle and multiflora rose
are relatively more common statewide but even these species tend to occur more often in the northern and
central zones. The opposite is truc of the presence of flowering dogwood and therefore anthracnose. Flower-
ing dogwood is found primarily in the southern zone and reaches the northern limits of its range in the central
zone.

Spring Ground Cover Survey

Forty-two sites were monitored in spring 2002; an additional four sites were flooded and could not be sur-
veyed. The results of the ground cover survey were similar to other years—sensitive species were uncommon
and when invasive species were present, they covered large areas (Table 4, Fig. 9).

Only six sites had disturbance-sensitive taxa in the quadrats and only two different taxa were recorded. This is
in contrast to 24 sites that recorded invasive taxa in the quadrat survey. Total cover of invasive taxa was seven
times greater than disturbance-sensitive and common native combined (Table 5). In order to compare invasive
cover to other years, gooseberry and Japanese honcysuckle cover must be removed as these taxa were not
previously included in the ground layer survey. If these two specics are ignored, only 15 sites had invasive
taxa in the quadrats and the amount of cover drops from 721.09 to 350.08 m’/ha. The latter value is similar to
that reported in spring 2000, the last time many of these sites were monitored. Obviously, these two species,
especially Japanese honeysuckle which spreads horizontally and can cover large amounts of arca, contribute
heavily to the amount of invasive species cover.

Table 3. Correlations of various ForestWatch indices *

tree up/bottom tree tree mean
richness abundance basal area tree DBH
.009 416 .699 .084 .188 .023
-013 .657 488 811 435 413
-004 .038 -039 .760 .239 439
oe a eee et oe
.929 543 .490 .798
-000 -040 895 .690 .190 555
.148 001 .793 -000
mean DBH -.149 -.323 -.086 -.754 .682 1.0
.265 .013 520 -000 -000

*First line is Pearson correlation coefficient and second is significance value, or P-value.
Values in bold are significant.

Many plants grow in widely scattered patches, making it unlikely that many species will occur in any of the
15 0.25 m? quadrats. It is more likely to encounter these occasional species during the presence-absence
survey which covers 1500 m?, whether the plant is patchy or not. If a species is recorded during the presence-
absence survey but not the quadrat survey, this indicates the species has a rather patchy distribution. However,
if a species is found during the presence-absence survey and in the quadrats, this strongly suggests the species
is more uniformly distributed or has a greater abundance than many other plants. Garlic mustard and Japanese
honeysuckle are good examples of this, as is bleeding hearts to a lesser extent. In addition to being recorded in
many presence-absence sections, garlic mustard and Japanese honeysuckle were recorded in several of the
quadrats on the sites in which they were detected. In contrast, six sensitive species were detected in the
presence-absence survey but not in the quadrats. Even during the presence-absence survey, most of the
disturbance-sensitive species are not recorded frequently. They were found at 15 sites during this survey
compared to 33 sites for invasive species.

The data continue to point to the simple fact that when invasive species are in a forest, they tend to dominate.
The best management is to prevent them from becoming established by removing them when they are first
detected. Once they are established, invasive species can be quite difficult to eradicate. Just keeping garlic
mustard out of a site is critical as this is the most prevalent non-native species encountered at ForestWatch
sites. It continues to cover about three times as much area as sensitive and common species combined.

Mean Sensitive Plant Cover *

Mean Invasive Plant Cover *

Figure 9. Mean invasive a

Table 4. Frequency and cover of indicator taxa

Detection by site
Disturbance-sensitive Species | (# of sites, from quads)

Detection by site Ground Cover
(# of sites, from pres abs) | (average m?/ha)

0
0
0

Blue cohosh

White trillium (all species)
Doll’s eyes

Large-flowered bellwort
Bleeding hearts (both species)
Maidenhair fern

Virginia spiderwort

Hepatica (both varieties)

—_
ooor$o
I o
[eo e} —

Common Native Species

— —
oOoWOrrK DOWN or omoo°cn[e

Virginia bluebells 0.78
Wild columbine 0
Blue phlox 8.29
Red trillium 8.60
Blue-eyed Mary 5.89
Wild geranium 27.60
Swamp buttercup 39.15
Sensitive fern 0
Invasive Species aera RES

Garlic mustard 16 16 308.22
Dame’s rocket 1 1 0.39
Moneywort 0 2 0
Ground ivy 3 8 41.47
Missouri Gooseberry 13 25 146.19
Japanese Honeysuckle 8 10 233.65

Table 5. Cover of general plant categories

Spring Downed Wood Survey

Downed wood provides important habitat to many organisms and is part of the normal functioning of a forest.
Presence of downed wood in a wide range of diameter classes is an indication that the forest is mature. The
majority of the downed wood was in the smaller diameter classes (Fig 10). The ratio of downed wood by size
class in 2002 was very similar to the results from 1999-2001. It is unlikely this will change very quickly as
the type of downed wood reflects the type of forests being monitored. Because Illinois forests are relatively
young and there are a relatively smal] number of very large trees present in most forests, a large majority of
the downed wood will likcly continuc to be the smaller classes, Iess than 30 cm. Only until the current large
trees grow in diameter then fall over will there likely be a change in the size of downed wood being reported.

[ 510m
= 10.1-20 cm

- Figure 10. Distribution of downed wood by diameter

Spring Human Use Survey

Volunteers look for eight gencral human uses while monitoring, including looking for trash and trails, and
noticing cut tree stumps and evidence of grazing. Strewn garbage continues to be the most common human
use observed in forests, and was reported at almost half of the sites. Tree stumps were also common, being
located at 17 sites. Three types of human use were correlated with each other. Many of the sites that had
hiking trails were being used by people other than the monitoring group while the surveys were being com-
pleted. Another. interesting but probably not too surprising correlation was that collected garbage was posi-
tively correlated with the presence of vehicle trails and human structures.

Spring Tree Height Survey

The average height of the tallest trees on ForestWatch sites was 21.7 m (range 10.5-30.7; S.D.=4.43). These
trees had an average diameter of 34.2 cm (range 15.0-47.2; S.D.=8.24). This was very similar to that of spring
2000, the other year many of these sites were previously monitored. It is also similar to the values scen in
2001 in which a completely different set of sites was monitored. This gives evidence that the tallest trees in
many Illinois forests are about 35 cm DBH and 22 m tall on average.

PrairieWatch 2001 and 2002

Matt Buffington

Between 2001 and 2002, 27 PrairieWatch sites were monitored, 16 of which were monitored for the first time.
Of the 27 sites from the past two years, 11 sites were reconstructions, eight were remnants, and the remainder
were unspecified. Forty-six total sites have been monitored since 1999.

Ground Cover Survey

Three different plant groups were examined during the ground cover survey using 0.25 m? quadrats: distur-
bance-sensitive, common native, and invasive species. Disturbance-sensitive species were not widespread on
any of the sites. They averaged less than 250 m? per hectare cover, which is less than a fourth of the area
covered by invasive plants (Table 1). Common natives were encountered at 23 of 27 sites and covered the
most area of the three plant categories. Ninety-three percent of the common native ground cover is from the
three dominant prairie grasses, big and little bluestem and Indian grass. This is not surprising as these grasses
are often dominant in Illinois prairies. All of the common native species were found among all the sites, but
disturbance-sensitive species cream wild indigo, green milkweed, and the gentians were not recorded at any
sites nor were the invasive species daylily, teasel, Autumn olive, and black locust.

Table 1. Cover of indicator taxa

Average cover Detection
(m” per ha) (% of sites, n=27)

Detection
(% of quadrats, n=540)

Plant Category

Sensitive
Common
Invasive

53%

Table 2 shows the amount of cover for three major plant categories, plus bare ground and litter. Grasses and
forbs were the main plant types observed in the quadrats, while there was very little woody plant cover. The
relative scarcity of woody plants is promising as normally functioning prairies typically have only a small
amount of woody cover. There appears to be some balance between bare ground and plant litter. Bare ground
typically increases with an increase in fire frequency while litter decreases. The presence of litter and bare
ground were inversely proportional, although not significantly so (Table 3).

Table 2. General cover types

Table 3. Correlations of various cover values

[es [ex [om | wr | nw [omass rons | wooo] wane [orren] Sars |

0.03
0.87 | 0.15

0.90 | 0.76 res aacteseel tee acai crane Whasnae hee
reo beborabtendet heel ered hook tajeeons| ae >
Sat a sph cae mem aaET paar” Soy eo
0.57 | 0.88 | 0.42 | 0.29 | 0.37 | . sda Pony’ fe <> ro
PP | tos | toe | tae ae Le one | |
0.09 | 0.90 | 0.30 | 0.22 | 0.38 | 0.00 hi aa eae
“eal APSR eA as Po aed
0.09 | 031 | 027 | 0.06 | 038 | 0.55 | 0.10
SS ae
0.78 | 038 | 0.14 | 032 | 019 | 036 | 020 | 018
-0.05 |-0.03 | 0.16 | -0.03 |-0.22 | 1.00
0.72 | 0.76 | 0.67 | 0.06 | 0.81 | 0.86 | 0.42 | 0.87 | 0.27] .
-0.01 |-0.48 |-0.02 | 0.34 | 0.12 |-0.07
PB tibrn ey peat -0.30 |-0.07 |-0.30 | 0.06 |-0.07 | 0.13 [-0.12| 0.03 |0.07
0.93 | 0.04 | 0.33 | 0.83 | 0.32 | 0.85 0.83

Twenty-seven sites were used for all indices except IBSI, which had 13 sites.

The first line is the Pearson correlation coefficient and the second is the level of significance, or P-value.
Numbers in bold are significant correlations.

DS=disturbance-sensitive

CN=common native

IH=invasive herbaceous

IW=invasive woody

NN=all invasive species combined

SAPS=number of saplings per hectare.

There were several interesting correlations among the various cover types, three of which were significant—
as grass cover increased, forb cover and the number of saplings decreased; and the butterfly index (discussed
later) was positively correlated with the presence of invasive woody plants. In some prairies, grasses can
become dominant at the expense of forb cover, perhaps as a result of regular spring burns, which often favor
warm-season grasses like big and little bluestem. However, based on the data available it is difficult to
determine if prescribed burning is affecting the relationship between forbs and grasses at Prairie Watch sites.
Variations in fire frequency, timing, and intensity greatly affect what plants will be present in a prairie. This
type of fire information is not currently available for many PrairieWatch sites. In addition, there are other
environmental factors that affect the results of burning. For instance, burning in years of low precipitation is
more damaging to many plant species than burning in years of average and above average rainfall. It must
also be noted that a decrease in forb cover does not necessarily mean a decrease in richness. The effects of

burning on richness are varied and likely depend on the interaction of numerous biotic and abiotic factors.
Isolating one factor as the cause for changes in another is a daunting and often futile task.

The two other significant correlations require some explanation. Because most prairie grasses respond favor-
ably to fire while tree saplings do not, it was not surprising that there was an inverse and significant relation-
ship between these two plant groups. The correlation between the butterfly index and invasive woody cover
must be viewed with much caution. There were only 13 sites that had IBSI values and of these only three had
any invasive woody taxa. Thus there is not much data to make a strong argument that this is an actual rela-
tionship. Further discussion of the butterfly index occurs later in this report.

Since nine out of 11 disturbance-sensitive taxa are forbs, it was not surprising that forb and disturbance-
sensitive taxa cover were positively correlated. Another intuitive correlation was the positive relationship
between plant litter and the amount of cover from invasive woody plants. Litter and woody plants generally
increase in the absence of fire so this relationship is not unexpected. Interestingly, cover of disturbance-
sensitive taxa and woody plants was positively correlated. There is no clear explanation for this as distur-
bance-sensitive species are adapted to survive fires.

It must be noted that correlations only provide a small amount of information. For PrairieWatch, one of the
primary interests is the change in condition over time, which cannot be surmised just through these correla-
tions. Is there a change in the average amount of plant litter across all sites? Are disturbance-sensitive species
becoming less common statewide?

Shrub Survey

Shrubs and saplings were recorded from 40% of the sites but only a small number of shrub and sapling stems
were recorded overall. Prairie willow was the only native shrub recorded, and then only one plant from one
site. Eight different sites had invasive shrubs present but only one had more than one species. The total
number of invasive shrub stems was quite small except for the one site with multiple species. A total of 27
shrub stems were recorded from seven of the sites while the other site had 86 shrub stems, primarily buck-
thorn and honeysuckle. Sixty-five total saplings were recorded from seven sites. The number of saplings
ranged from two to 28 among the sites, or 200-2,800 saplings per hectare. These results are promising but
certainly cannot be used to depict the condition of prairies statewide. Many grasslands have a significant
problem with invasive shrubs and saplings.

Large Tree Survey

The sampling area for large trees (2 5cm diameter at breast height) was the maximum width of 50 m for most
sites but six sites were narrower than 50 m so the sampling area was therefore less than 50 m wide. Only five
of 27 sites recorded large trees in the sampling area. Two of these had a fairly large number of trees, 63 and
75 trees each. Eighty-seven percent of these were <30cm in diameter. The other three sites had only one,
three, or six trees each, and eight of these were <30cm.

Prairie Size

The size of the prairies being monitored was quite variable. The average area of the prairies was slightly
under eight hectares, 77,683 m?, and ranged from 1,893 to 700,000 m’. The amount of edge was also variable,
averaging 822 meters and ranging from 124 to 1,865 meters of edge. Two sites did not have perimeter data
and if these are ignored, the average area was 73,663 m’. Unfortunately, the minimum amount of edge re-
quired to meet this amount of area is 962 meters, which is actually greater than the value determined from the
data. This indicates there is a problem with the estimates of area and/or perimeter. Efforts will be made to
address this problem in order to provide accurate data concerning edge and perimeter. This is important

because an increase in the amount of edge implies an increasing amount of stress. Edges are often composed
of species that will invade a prairie if left alone. The more edge, the more work that is needed to maintain the
prairie. In addition, the prairie plants become more isolated from each other with more edge.

Land Use Survey

Surrounding land use and conditions directly along prairie borders play a big role in determining what is
happening at a site and what type of management may be required. Sites surrounded by savanna or forest face
much different challenges compared to sites surrounded by residential areas and cropland. Ten of the 14
possible land cover types were recorded from Prairie Watch sites in 2001 and 2002. Seventy-five percent of
the records were of four types: forest, cropland, residential, and other. Some of the more common “other”
types were fields, quarry, and orchard. Many sites recorded a border type other than those provided, including
a mowed area or fire break, trail, and ravine. There were also many instances (26%) where there was no
border, meaning there was an abrupt change from the surrounding land use to the prairie. Of the actual border
types, a paved surface was the most common (17%) followed closely by hedgerow (13%).

Butterfly Survey

Butterflies were monitored at 13 sites between 2001 and 2002. Monitoring at seven sites was completed
during the official butterfly period (June 15 - August 1) while the other sites were completed during the plant
survey later in the year (August 15 - October 1). This is acceptable as many of the sites were monitored for
the first time and there was not enough time to register and establish a site before August. Much of the data
collected reflect new volunteers testing their butterfly identification skills.

A total 735 butterflies were recorded from 13 sites. Average detection rates can be determined for 11 of these
sites (two did not include the amount of area surveyed). Approximately 10.2 hectares were monitored in 9.58
hours in which 683 butterflies were detected. This translates to roughly 7.0 butterflies/ha/hour. Over half
(313) of the butterflies were recorded as “Other”, and thus were not one of the indicator species. The most
common indicator was eastern-tailed blue followed closely by the pearl crescent (Table 4). Forty-nine of the
51 wood nymphs were recorded from one site. The three species most closely associated with high quality
grasslands, silver-bordered, meadow, and regal fritillaries, were not observed at any of the sites nor were the
dainty sulphur, American copper, or variegated fritillary.

PrairieWatch is attempting to quantify the condition of the habitat by using a new, experimental index based
on the butterfly species present, their abundance, and their overall environmental significance, which relates
to the butterflies’ preferred habitat and their commonness. This index is termed the Illinois Butterfly Site
Index (IBSI). Although this is an experimental effort it is conceptually based on standard diversity indices
used in other ecological studies. The index is designed such that an increase in the IBSI value implies greater
habitat quality. As this is a unique and thus far untested index, interpreting the data should be done with
considerable caution and IBS] values can only be compared to other IBSI values. Table 5 shows the IBSI
results from the 13 sites with butterfly data.

It is not certain if sites with high IBSI values are necessarily of greater quality. The site with the greatest IBSI
value, for instance, did not have any disturbance-sensitive or common native plants in the quadrats but did
have invasive plants. In contrast, the site with the second lowest IBSI value had above average disturbance-
sensitive and common native cover and a very small amount of invasive herbaceous plants. These two sites
are from the northeastern part of the state and both were monitored around the second week of September,
although they were monitored in different years. In addition to different years of monitoring, there are various
environmental conditions that can affect butterfly monitoring. For example, the site with an IBSI value of
eight was monitored on a day with 90% cloud cover and moderate winds, conditions in which most butterflies
are inactive. PrairieWatch will continue to experiment with this index to determine how well it applies to
actual conditions at the site.

Table 4. Butterflies recorded from PrairieWatch sites, 2001-2002

Taxa # Individuals

Eastern-tailed Blue 86
Pearl Crescent 84
Wood Nymph 51
Black Swallowtail 45
Monarch 28
Great-spangled Fritillary 21
Buckeye 12
Clouded Sulphur
Dogface Sulphur

Little Sulphur

Painted Lady

Aphrodite Fritillary
American Painted Lady
American Copper
Dainty Sulphur

Meadow Fritillary

Regal Fritillary
Silver-bordered Fritillary
Variegated Fritillary
Other

WwW
—

Table 5. IBSI values for PrairieWatch sites, 2001-2002

PrairieWatch SiteID # Species Total Abundance IBSI
P0209706 8 46 263.04
P0412301 6 54 216.37
P0211101 5 59 191.38
P0412302 7 31 183.56
P0209705 4 44 131.46
P0611701 5 18 96.64
P0611702 3 39 90.79
P0204304 4 21 82.92
P0802901 4 19 80.75
P0204301 3 16 54.00
P0204303 4 7 43.84
P0108501 3 4 30.00
P0204302 1 4 8.00

Abundance includes butterflies recorded as “Other” and thus is not included
in the IBSI calculation.

RiverWatch Data Summary Results for 2002

Alice Brandon

Introduction

The Illinois RiverWatch Program is the stream-monitoring component of the Illinois EcoWatch Network
(EW), a volunteer monitoring initiative coordinated through the Illinois Department of Natural Resources
(IDNR). RiverWatch seeks to document long-term trends in stream health as reflected by biological monitor-
ing of 37 benthic macroinvertebrate indicator taxa, presence or absence of macroinvertebrates of “special
interest,” and a habitat survey that documents long-term changes to stream segments over time.

This report summarizes statewide results for the 2002 monitoring season. Data from 1996 to 1999 along with
data from other EW programs was recently published in the report Critical Trends in Illinois Ecosystems.
RiverWatch data from 2001 was also used in the most recent CTAP Annual Report (winter 2001).

Volunteers monitor both volunteer selected and randomly selected stream sites with an emphasis on increas-
ing the number of random sites. Data from all sites was combined for this report, as RiverWatch does not have
an adequate number of random sites at the watershed level. Readers interested in more in-depth analyses may
access the RiverWatch database for years 1995-2002 at http://dnr.state.il.us/orep/ ecowatch/, and IDNR
Critical Trends Assessment Reports at http://dnr.state.il.us/orep/ctap/

RiverWatch Metrics

The RiverWatch program uses multiple metrics to provide quantitative information on stream health. Metrics
are based upon those of professional biologists and were developed with assistance from biologists with the
Illinois Natural History Survey and Illinois Environmental Protection Agency. For more information please
refer to the RiverWatch Stream Monitoring Manual, revised 5" Edition.

Why Multiple Metrics?

The metrics used by RiverWatch are useful indicators because they contribute unique information about
stream quality and are also correlated with one another. For example, the “% of EPT taxa” is significantly and
negatively correlated with the MBI. This supports the hypothesis that as the MBI scores increase and stream
quality decreases so does the number of EPT taxa present (pollution intolerant taxa). While correlations
confirm the relationships of the stream indicators to one another, results do not always fit the model. A
watershed may score well on one indicator and poorly on another. For example, the Spoon watershed typi-
cally scores well on RiverWatch metrics such as the % of EPT taxa but according to state biologists it also has
below average scores for native fish species and stream habitat quality (see Critical Trends in Illinois Ecosys-
tems). This is why it is useful to use multiple metrics and track various indicator taxa, which may highlight
different stream health concerns.

Other Data Collected

The presence or absence of certain native and invasive species provides RiverWatch with additional informa-
tion on stream condition. Tracking invasive species such as zebra mussel, Chinese mystery snail, and Asiatic
clam helps biologists track the distribution and migration of these species throughout the state. Native species
such as fingernail clam and native mussels were once abundant species that are quickly disappearing from
Illinois streams. 7

Table 1. Metrics utilized by RiverWatch to measure stream quality or health

ea How it is calculated : What it indicates
Taxa Total number of taxa identified in the sample. As taxa richness increases; generally so
Richness does water quality.

% EPT Number of mayflies, stoneflies & caddisflies Streams with a high number of these

taxa divided by the number of organisms pollution intolerant taxa are considered
sampled & multiplied by 100. to be in good health.

Number of aquatic worm and bloodworm

midge taxa divided by the number of organ-
*MBI = Macroinvertebrate Biotic Index; note that taxa dominance was not used in this report.

Streams with a high number of these
pollution tolerant taxa are considered to
be in poor health.

isms sampled & multiplied by 100.

Number of organisms from the three most
common taxa collected divided by the number
of organisms sampled.

Dominance by just a few taxa indicates
low stream quality. Generally, a value >
80% indicates low aquatic biodiversity.

Taxon’s total tolerance value divided the
number of organisms sampled. It provides a
weighted average of the pollution tolerance of
the organisms collected.

The MBI was developed to detect
organic pollution such as sewage.
A score of:

< 6.0 = good water quality

6.0 to 7.5 = fair water quality

7.6 to 8.9 = poor water quality
> 9.0 = very poor water quality

Volunteers also collect habitat information such as the stream bottom substrate type(s), stream width, and
stream discharge (a measure of the volume of water passing the stream site). This information documents how
a stream segment changes over time and helps to explain the number and type of macroinvertebrates col-
lected. For example, swiftly moving streams with rocky bottoms provide better habitat for most EPT taxa than
do slow moving, muddy bottom streams.

Taxa and Sample Abundance Counts

In 2002, the most prevalent taxa statewide were midge, hydropsychid caddisfly, sowbug, scud and black fly
(Fig. 1). These taxa are pollution tolerant and inhabit a wide range of stream types.

The armored mayfly, snipe fly and saddle case caddisfly were rarely collected (present < 2% of sites) and are
the least common taxa detected this year. The saddle case caddisfly is especially sensitive and inhabits high
quality streams that have thus far escaped degradation. The snipe fly, while also fairly intolerant to pollution,
is at the end of its range in Illinois and is restricted to the northern one-third of the state.

In previous years, RiverWatch has been concerned that volunteers may not be collecting an adequate
macroinvertebrate sample size. In most Illinois streams, volunteers should easily collect 50
macroinvertebrates. The program prefers samples with a minimum of 100 organisms as larger samples better
reflect the macroinvertebrate diversity located at any given stream site. Collecting fewer organisms increases
the likelihood many taxa at the stream will not be collected. This may ultimately affect taxa richness mea-
sures. In 2002, on average, volunteers collected 96 macroinvertebrates after sub-sampling with samples
ranging from 1 to 472. Twenty-three percent of the monitored sites sampled less than 50 macroinvertebrates,
34% sampled 51 to 100 macroinvertebrates and 43% sampled more than 100 macroinvertebrates.

Hydropsychid
caddisfly Blackfly

Mean Abundance

Macroinvertebrate Taxa

Figure 1. Mean abundance of the most common taxa identified statewide 2002 (N=234)

Taxa Richness, Percent Composition of Indicator Organisms & MBI

The number of taxa identified ranged from | to 21. Twenty-eight percent of the monitored sites identified five
or fewer indicator taxa, 55% had six to 10 taxa, and 29% had 11 or more taxa.

EPT taxa comprised, on average, 22% of the organisms sampled while worm taxa made up less than 5% of
the organisms. However, voluntecrs are most likely to sample riffle habitats, the preferred habitat for EPT
taxa, rather than undercut banks and sediment where worm (aquatic worm and bloodworm midge) taxa
predominate. The most common taxa were neither EPT nor worm taxa, but sowbugs and midges.

The average MBI score for monitored streams was 5.61. The poorest MBI score was 11.0 and the best score
was 3.52. Less than 6% of sites have MBI scores above 7.0. According to the MBI few Illinois streams
monitored by RiverWatch arc in poor health.

Macroinvertebrates of Special Interest

Fingernail clams, a native species, are the most common taxa of special interest reported and occurred at 29%
of the monitored sites. Voluntcers reported no Chinese mystery snails or Zebra musscls this year. This most
likely indicates these invasive species have not yet spread from larger rivers to the smaller streams monitored
by volunteers. Volunteers noted the presence of native mussels at 12% of the sites and Asiatic clams at 10%
of the sites (Table 2). These results are consistent with previous years, indicating native mussels are occurring
at the same levels as reported previously.

Number of Monitored Sites

RiverWatch volunteers monitored 234 sites statewide in 2002. Another 18 could not be monitored due to
reported flooding or low water. The number of monitored sites is 33% lower in comparison to previous years
(for example, 349 sites were monitored in 2001). The decrease is likely attributable to the budget cuts experi-
enced by EcoWatch in June / July 2002 during the height of monitoring. Numbers are fairly encouraging
considcring the timing of the cuts.

Table 2. Presence of macroinvertebrates of special interest across all monitored sites

Organism Type Occurrence

Native mussels Native species 28 sites (12%)
Fingernail clams Native species 67 sites (29%)
Zebra mussels Invasive species 0 site (0%)
Asian clams Invasive species 23 sites (10%)
Chinese mystery snail Invasive species 0 sites (0%)
Rusty crayfish Invasive species 20 sites (8%)

RiverWatch uses watershed designations based on the Illinois Streams Information System database, which
puts all Illinois streams into ten major watersheds. The Fox watershed has the most monitored sites (78) and
the Little Wabash watershed the least (0) (Fig. 2), a reflection of the large volunteer base in the Chicago
metropolitan area. Most watersheds have a minimum of 15 monitored sites.

Stream Data by Watershed

The Embarras, Sangamon and Kaskaskia watersheds score the lowest across all RiverWatch metrics while the
Rock and Kankakee score the highest. The La Moine, Spoon, Big Muddy and Fox watersheds score some-
where in between these two groups (Table 3). This contradicts previous results because typically the Embarras
scores much higher across multiple RiverWatch metrics and is a prime example of the importance of long-
term data.

Mean taxa richness at the watershed level ranges from 11 to 7 with the Kankakee having the highest taxa
_ richness and the Embarras having the lowest (Table 3). This suggests the Kankakee is in better stream health
than most other watersheds and is consistent with results in 2001.

The Kaskaskia and La Moine have the lowest mean % of EPT taxa. These organisms are less prevalent in
streams with poor stream health. At the other end of the spectrum, the Rock, Kankakee, and Spoon watersheds
have the highest mean % of EPT taxa. The Sangamon and Embarras have the highest mean % worm (aquatic

Table 3. Averages for RiverWatch indices separated by watershed for the 2002 monitoring season

Watershed Sites Taxa Organisms % EPT | % Worm Rocky | Silt/sandy
(#) richness sampled taxa taxa substrates| substrates
ripen eines

| Big Muddy _| Be crm Vt

alent eereenl aan nena eae

= ae Se ee ee
earikabes wa) octudSpr ff Seainsc ob Loe fs ROB ENBO) cf aGIND [i GARTEN) AAMNIGS & OS) ABO |
|Keskaskia © | 21 | 6.1 | 8 | 100(89)m |" “12 eotenng Dane Pere
[ha Maine yenfs, 829 fei 5.7a: boon oeformbd ROBY are el 3ney] oo! WG collecting Ms atequase |
pRaok:vertetadic s2Goie| aard.9p JnodGuanbtx MOMEAD ola ieBd shol sid Bil yaa RARONO May Wd
pSangamonstr (is < Woe sfo5:9s jeofe eR opp 6S M4s rifles PC oantins adOrgey sansaiak Y
[spon [8 | 2] 8 | Gs) [ep af OP
Mean Macroinvertebrate Biotic Index (MBI), mean taxa richness, mean organisms sampled with means from 200] in parentheses, and

the mean percentage (%) of Ephemeroptera, Plecoptera and Trichoptera (EPT), bloodworm midge and aquatic worm (Worm) in the
samples.

Number sites monitored

2)
od
°
°
a

@ m n x x = = >] (?,)
o 3 2 2 = 2 3 ° =
Ss 8 zl) = = oF & ra}
c 4 > ov

5
a nace | = 3
< “o ro) ry) =

Figure 2. Number of sites monitored by watershed in 2002

worm and bloodworm midge) taxa and the Rock and the Big Muddy the lowest. This suggests pollution
tolerant organisms are more common in the Sangamon, Kaskaskia and Embarras in comparison to other
watersheds. The Embarras and Kaskaskia also have streams with high percentages of silt substrate types,
which is the habitat preferred by worms. The Big Muddy, Kankakee and the Rock have a high percentage of
streams with rocky bottom substrate types, which are all well above the statewide average of 56% (Table 3).
The parameter % substrate siltation coverage (SSC) measures the percentage of the stream bottom covered in
fine silty particles. High SSC may adverscly impact macroinvertebrates as it decreases the amount of avail-
able habitat. This can cause sensitive EPT taxa to decline or be extirpated from a stream. The Kaskaskia
watershed has an average SSC of 60.8, which is 50% higher than the statewide SSC of 40.7. This watershed
also has the lowest percentage of EPT present.

The Embarras watershed appears to change from having relatively high stream quality to having low quality.
However, there is a 30% drop in the number of monitored sites from 2001. It is likely this decrease skewed
the data rather than there being a drastic change in stream quality over a one-year time period.

Most watersheds have high mean sample abundances of sowbug, scud, hydropsychid caddisfly and midge
(Table 4). The Kankakee, Rock and Embarras are the only watersheds without sowbug in the list of the top
five most common taxa. The Kaskaskia watershed has an unusually high abundance of midges. This
watershed’s streams are primarily muddy-bottom, where midge taxa survive well. The Rock and Embarras are
the only watersheds where mayfly taxa mean abundances are high. Hydropsychid caddisfly has the highest
mean sample abundance for any EPT taxa, and is also the most pollution tolerant EPT taxa.

Table 4. The five most common indicator taxa for each watershed

a i Raa.
(24 sites) (16.4) (13.4) (8.2) (6.9) (5.5)
eras) | ay | a | en
78 sites 28.6 15.9 12.6 9.5 8.6
15 sites 22.2 8.6 6.6 4.8 4.3

(21 sites) (31.4) (15.8) (12.5) (10.3) (6.9)

(27 sites) (26.0) (19.0) (17.1) (17.0) (7.1)
esi | asa | cies
(10 sites) (21.2) (11.8) (5.1) (4.8)
Gey | aon | ‘asg | Uso
Statewide Midge Sowbug Scud Caddisfly Black Fly

N= mean abundance across all samples; 'Hydropsychid Caddisfly. *Swimming Mayfly, *Broadwinged Damselfly. Note: Three
additional sites were added to the database after the data analysis was performed. Therefore, 236 sites have data this year.

Summary

Most streams have high numbers of taxa tolerant to pollution such as sowbugs and midges. In addition, results
for taxa richness, % EPT taxa, % worm taxa, and the MBI suggest streams of intermediate quality predomi-
nate the monitored sites. Most Illinois streams have experienced some level of habitat degradation or pollu-
tion. This is not surprising considering agriculture or urbanization cover most of the state with little remnant
natural areas intact. However, pollution intolerant taxa manage to survive in small pockets. These streams also
tend to have high mean taxa richness and % EPT taxa. Bloodworm midges and aquatic worms, while fairly
common, did not dominate at most streams.

Results by watershed indicate that some watersheds are in better health than others. However, the Kankakee,
Spoon and Sangamon watersheds have relatively low sample sizes (below 20 sites) making it somewhat
difficult to be certain if the data reflect actual stream conditions or variability in volunteer selected sites.
While RiverWatch has an adequate number of randomly selected sites at the statewide level, it is still working
to increase the number of randomly selected sites at the watershed level.

Watersheds with high % EPT also tended to have streams with rocky bottom substrates and low SSC. In 2002,
the Kankakee and Rock watersheds have better than average stream quality while the Embarras and
Kaskaskia watersheds have lower than average stream quality. The results for the Embarras are somewhat
surprising, since in previous years it had high stream quality (see summary report 2001). One possible expla-
nation is a 30% decrease in the number of streams monitored from 2001 to 2002. The other watershed results
are consistent with previous years’ findings.

Botanical Report

Floristic Quality Assessment (FQA) as a Measure of the Naturalness
of the Grasslands and Wetlands of Illinois

Greg Spyreas, Connie Carroll, James Ellis, and Brenda Molano-Flores

Introduction

Throughout the United States, biological systems have been extensively degraded by human activities
(Whitney 1994). In Illinois, less than 0.01% of the original grasslands and less than 0.07% of the original
wetlands remain in an undegraded, natural state (IDNR 1994). The difficulty of measuring the “naturalness”
or “natural integrity” of the few remaining habitats across our highly disturbed landscape has long been a
problem for plant ecologists. How does a field botanist confer years of experience and knowledge about a
natural area and its plants in a brief and meaningful way to laypersons? For example, the State’s more than
3,200 plant species differ tremendously in their predilection for natural versus disturbed habitats. How do we
objectively designate which habitats are biologically valuable as natural areas and which are less valuable? At
a time when the willingness and enthusiasm for re-creation and restoration of natural areas in our region is
increasing, how can we gauge and explain the ecological success of habitat restorations? These questions
highlight the need for standardized, scientifically meaningful, botanical information that can be easily inter-
preted by everyone. Unfortunately, to date, the most commonly used indicators of ecologically valuable native
habitats (e.g. species diversity, the presence of rare species) are often insufficient and unpredictable.

Recently, Floristic Quality Assessment (FQA) has become a popular method to measure habitat integrity, or
more specifically, vegetative natural area quality. National, state, and local agencies that are both public and
private (e.g. IDNR, IDOT, The Nature Conservancy, Kane County Forest Preserve) charged with managing
large natural areas now rely heavily upon these measures.

FQA scores are comprised of two measures: the value C, the average Coefficient of Conservatism; and the
value J (I= CVN [N = the number of plant species at a site]). Coefficients of Conservatism (CC) are set
integers (0-10) assigned to each vascular plant in Illinois (Taft et al. 1997). Highly conservative plants
(CC=10, 9, 8) only thrive in high-quality, intact, natural areas, and non-conservative plants (CC=0, 1, 2) are
usually common (i.e. horsetail, ragweed) (Taft et al. 1997). Non-native plants are assigned values of zero
(CC=0). An area with a high floristic quality score would be expected to have many species with high Coeffi-
cients of Conservatism (5-10). For example, we would expect an undisturbed old-growth forest in Illinois to
score approximately 4 - 5 for its C, and 35-45 for its 7. On the other hand, a weedy lawn would score at
around 1-2 for its C, and 3-7 for its /.

Despite the increasing usage of FQA, no studies have rigorously tested its effectiveness in gauging natural
area quality (Nichols 1999; Traina 2001; Mushet et al. 2002; Rooney and Rogers 2002). Using the uniquely
extensive, statewide dataset collected through the Critical Trends Assessment Program (CTAP), we set out to
determine if FQA is a precise measure of natural area quality. We also sought to see how other measures, such
as diversity and the presence of non-native species, correlate with other measures of natural area quality. By
comparing a site’s disturbance grade with its floristic quality we attempted to determine how well the FQA
measures correlate with perceived naturalness. Additionally, we also used FQA to illuminate general trends in
the State’s floristic quality. For example, which habitat types or regions of the state score higher in FQA
measures than others? Or, how does the presence of introduced species affect a natural area’s FQA score?

Methods

Site data was gathered as part of the Illinois Critical Trends Assessment Program (CTAP) (IDNR 2001). Sites
were randomly selected in forests, wetlands, and grasslands using satellite based geographic information
systems vegetation coverage. Over four years (1997-2000) a total of 108 palustrine emergent wetlands and 97
grasslands of many different community types across the state were sampled following CTAP protocols
(Carroll et al. 2002) (Fig. 1). Forest data was not used in this study in order to simplify the analysis. Sites
were sampled beginning in southern Illinois and finishing in northern Illinois, where wetlands were visited
from June through July, and grasslands were visited from July through the last week of August. For geo-
graphic comparisons, the state was roughly divided into thirds (i.e. North, Central, and South) (Fig. 1).

At each site, twenty 0.25 m’ permanent quadrats along a 41-m transect were used to estimate ground cover.
Along this transect, woody stems <1 meter tall and <Scm dbh were counted in a 4 by 41 meter plot. Trees > 5
cm dbh were also tallied in a 50 by 41 meter plot incorporating the transect. Using historical and current land
usage information gathered from landowners we ranked the amount of human disturbance each site had
incurred, and designated this disturbance as its “Grade”. Grade A sites are the most natural and have received
little to no noticeable human disturbance. Grades B, C, and D sites are increasingly more degraded by distur-
bances. Finally, grade E sites are so disturbed that they have none of their original plant community or native
vegetation left (e.g. planted pasture, planted hayfields, roadsides).

Results and Discussion

A number of trends emerged from the data. The majority of CTAP sites were of grades C, D, or E, and there-
fore very few high quality sites were encountered. Overall, we found FQA to be an excellent measure of the
amount of degradation an area had undergone, that is, FQA measures were highly correlated with natural area
quality (grade) (Table 1, Figs. 2 and 3). But, it should be noted that because we sampled relatively few high
quality natural areas, the sensitivity of FQA at separating small differences between our best natural areas is
still uncertain. The number of species (often called species richness or diversity) that an area contained was
not a consistent measure of natural area quality (Table 1). Despite the high correlation between grade and
FQA, some community types (sand prairie, natural ponds) scored idiosyncratically higher or lower when
compared to the pooled wetlands and grasslands.

Table 1. Spearman rank correlations for floristic quality
measures of Illinois grasslands and wetlands

Number Number of Number of Shannon Introduced
of species introduced native diversity cover
species species

-0.34**

-0.62**

** P< 0.0001, * P<0.001

Empty cells indicate invalid correlations because one value is used in the calculation of the other.
N = 205.

“Grade” is a measure of disturbance to a site from human uses.

“Shannon diversity” is a measure of diversity.

“Introduced cover” is the percentage of a site that is dominated by introduced species.
Differences are significant if P < 0.05.

e Grass anc Sample Sites 3 7 4 Se on tRe |

e Wetlard Samole Site« ; be a ann —*
— Counties Bes aa . ju!” 8 | - 5
(tap Sections : is e } -
e es ' ).
— central \, pe a OL a a er ec)
north len 2 ' e y
: 4 = a ,
() south ar J | ar a
+ eo ees
Uy . (SS Ses
= ® » e ae
{ i | 5
USAT Yo ea
aan ot
\ |
. £ een
= 5 rt r—
1 10 20 Miles an I wat
( — »
] eerie | si
Ly Si :
se ais
4

Pore

Figure 1. Location of wetland and grassland CTAP sites sampled from 1997-2000

Mean site indices for selected grassland types

Floristic Quality
Index

Mean C*10
Number of species
Shannon Diversity
Index*10

Percent Non-native
20 species cover

> O04 @

Prorte Pasture Aband. pasture Havtield Old treld Stupamne 9 Waldhte plant

Figure 2. Mean site measurements for natural and anthropogenic grasslands grouped by community type.
All communities shown besides Prairie are grade E. Some site measurements were multiplied or divided by
10 to facilitate comparison on one graph.

For the state as a whole we found that the morc an area was dominated by non-native species the lower its
FQA scores (Table 1). Additionally, wetlands on average scored higher in floristic quality (wetlands I = 8.14
per site, C = 2.24 per site, grasslands I = 4.65 per site, C = 1.00 per site) despite having fewer species (wet-
lands species per site 14.4, grasslands species per site 18.7) than grasslands. This is probably because wet-
lands were far less dominated by non-native species than grasslands (Table 2). Additionally, the southern third
of Illinois scored higher in FQA compared to the rest of the state (Fig. 4), which we believe appropriately
reflects its distinct topography, soils, climate, and land use (i.e. the Shawnce National Forest, IDNR 2001).

When assessing the conservation valuc of any area, many critical factors such as wildlife habitat, fauna, soil
microbia, genctic heritage, hydrologic function, aesthetics, etc., are not measured by FQA. There is no single
value that will ever encapsulate the dense interactions and complexities of worth in a natural area. However,
we found that FQA is gencrally a simple, objective, and repeatable measure of habitat degradation that
foregoes individual judgments. Further studies of FQA are still warranted, especially focusing on higher
quality sites. Additionally, as other authors have pointed out FQA should be used in concert with other factors
of importance such as sizc, presence of rare specics, rarity of community type, and proximity to other valu-
able sites, when determining natural arca quality (Taft et al. 1997). However, we have gencrally found FQA
measures to be useful in providing casily understandable information that assists us in our evaluations of
ecosystems and their biological health.

Mean Site indices for selected wetland types

Floristic Quality Index 4
Mean C*10

Number of Species
Shannon Diversity
Index*10

Percent C over of

| Non-native species

3
> ouode

Salve Meimdow Wet Prame Wet Restoration Wet Phitield Wat Old Field Wet pasture

Figure 3. Mean site measurements for selected wetlands grouped by community type. All communities shown
besides Wet Prairie and Sedge Meadow are grade E. Some site measurements were multiplied or divided by
10 to facilitate ease of comparison on one graph.

Table 2. T-test comparing site measures for Illinois grasslands and wetlands

Grasslands Wetlands

I <0.0001
Cc <0.0001

Number of Species <0.0022
Introduced cover <0.0001

Number of species = 205.
“Introduced cover” is the percentage of a site that is dominated by introduced species.
Differences are significant if P < 0.05.

Oo 20
= 6 a
~~ =
o B © 15
ie 5
> A c
= fe)
10
s S
2@ 2 2
a7) 2 05
5 3
xo '
oO
O 00
a 70
&
D
oO 5x
6 3}
Oo
o 4
" 3
o@ 0 a
Cc wo
= mo)
2 oO 2
o 5 2
@
3 ° 10
a E
om oO - ae)
North Central South North Central South

Region within Illinois

Figure 4. Regional comparisons for wetland and grassland sites (+ 1 Standard Error) ANOVA: N= 205.

I: p< 0.036, C: p<0.018, Species Richness p< 0.543, Introduced Cover p<0.001. Different letters indicate
significant differences between groups.

Literature Cited
Carroll, C., C. Dassler, J. Ellis, G. Spyreas, J.B. Taft, and K. Robertson. 2002. Plant Sampling Protocols. in B.
Molano-Flores, editor. Critical Trend Assessment Program Monitoring Protocols. Office of the Chief

Technical Report 2002-2, Illinois Natural History Survey, Champaign. 38pp.

IDNR. 1994. The changing Illinois environment: critical trends. Vol. 3 Technical report. Illinois Department
of Energy and Natural Resources, Springfield, IL ILENR/RE-94/05.

IDNR. 2001. Critical trends in Illinois ecosystems. Illinois Department of Natural Resources, Office of realty
and environmental planning, Office of scientific research and analysis, Springfield, IL.

Mushet, D. M., N. H. Euliss, and T. L. Shaffer. 2002. Floristic quality assessment of one natural and three
restored wetland complexes in North Dakota, USA. Wetlands 22:126-138.

Nichols, S. A. 1999. Floristic Quality Assessment of Wisconsin lake plant communities with example applica-
tions. Lake and Reservoir Management 15:133-141.

Rooney, T. P., and D. A. Rogers. 2002. The modified floristic quality index. Natural Areas Journal 22: 340-
344.

Taft, J. B., G. Wilhelm, D. Ladd, and L. A. Masters. 1997. Floristic Quality Assessment for vegetation in
Illinois, a method for assessing vegetation integrity. Erigenia 15:1-24 + Appendix.

Traina, J. 2001. Vascular flora of Van Horn Woods, Plainfield Township, Will County, IL. Transactions of the
Illinois State Academy of Science 94:139-149.

Whitney, G. 1994. From coastal wilderness to fruited plain: a history of environmental change in temperate
North America, 1500 to the present. Cambridge University Press, NY, NY.

Terrestrial Insect Report

The Importance of Leafhoppers (Hemiptera:Cicadellidae) collected by the
Critical Trends Assessment Program

Adam Wallner

Leafhoppers arc insccts that belong to the infraorder Auchenorryncha in the order Hemiptera. They are
recognizcd by their picrcing-sucking mouthparts, which they use to fecd on a wide variety of vascular plant
specics, including grasses, scdgcs, broad-leafed woody and herbaccous plants of many familics, conifers, as
well as fungi (Dictrich 2000). Hemiptera is the fifth most speciose order of insccts, after bectles, flies, wasps,
and moths (McKamcy 1999). The suborder Homoptera, particularly, the infraorder Auchenorryncha (sensu
lato = Hemiptera: Homoptera + Heteroptera), comprises most of these, with about 50,000 known spccies. As
part of the Critical Trends Assessment Project (CTAP) from 1997 — 2001, data on these insccts have been
collected. In this report, information on three species of Auchenorryncha will be presented. These species
have been sclected because they can be considered either a new state record or they have economical/ccologi-
cal importance.

From the Auchenormhycha spccics sampled, Lebradea flavovirens (Gillette and Baker), a leafhopper from the
family Cicadellidac, was the only state record (Fig. 1). One individual was recorded from a wetland site in
Lee County (Fig. 2). Lebradea flavovirens is an exotic species, naturally occurring in Finland, Siberia,
Kamchatka, Kurile Islands, Sakhaim, Korcan Peninsula, Maritime Territory — Neartic region, and considered
threatened in Finland (Ossiannilsson 1983). They have been observed to feed on Calamagrosits spp., or reed
grass, and are therefore found in dry as well as in marshy habitats (Vilbaste 1980). Native reed grass has been
documented in several counties throughout Illinois, as well as distributed in patches where L. flavovirens was
collected (Mohlenbrock and Ladd 1978, Mohlenbrock 1986). Therefore, it seems likely that additional
collecting will show L. flavovirens distribution coincides in large part with that of reed grass species through-
out Illinois.

Figure 1. Lateral view of Lebradea flavovirens

Ecologically important Auchenorryncha observed in the CTAP samples, is the leafhopper species Evacanthus
nigramericanus (Hemiptera: Cicadellidac) Hamilton (Fig. 3). This species is recognized by its black colora-
tion, as well as reddish-brown spots located on its face (DeLong 1948, Hamilton 1983). Only three individu-
als were found in the CTAP forest samples — two collected in Kankakee County, 1999, and one collected in
Bureau County, 2000 (Fig. 2). Many species of Evacanthus are grass-feeders in the Korean Peninsula (Kwon
1983); in England Evacanthus interruptus has been recorded from hops, and the overwintering eggs may be
laid in the cracks of dead wood (Massce 1943); on some Japanese mountains, E. interruptus is reported to
feed on the aster plants (Ishihara 1953); and Evacanthus acuminatus is reported to inhabit and feed on sphag-
nous spruce wood and rich swampy wood (Linnavouri 1952). Evacanthus nigramericanus has been found on
herbivorous vegetation in moist woodlands of Illinois, particularly feeding on closely related fern species
(DeLong 1948). Additionally, this insect resides in primary forest habitats that have little disturbance. Since
this species is only found in relatively pristine habitats, it may be a useful biological indicator of the health of
Illinois forest habitats and could be used to infer conservation management decisions on private as well as
public land.

Athysanus argentarius Metcalf (Hemiptera: Cicadellidac) (Fig. 4), a grass-feeding lcafhopper species, is both
an ecological and economically important auchenorrhychan. This leafhopper was introduced to North
America from Europe and has been observed in southern Ontario and in western Canada (Chiykowski 1979).
Athysanus argentarius is a phytophagous feeder of grasses, grasslike herbs, rushes, sedges, and various
herbaceous plants. This Icafhopper is ubiquitous in many wetland and grassland CTAP samples (Fig. 2). In
addition, this leafhopper species has'also been found in fields containing brome grass (Bromus inermis).
Brome grass is a native species from Europe and has been introduced into the Midwest where it occurs in

Figure 3. Lateral view of Evacanthus nigramericanus

Figure 4. Lateral view of Athysanus argentarius

pastures and at the edges of moist woodlands (Mohlenbrock 1986). Thus, high numbers of A. argentarius
found in the CTAP sites may suggest poor habitat quality.

Economical importance of this leafhopper has shown that they are potential transmitters of aster yellow virus
in several commercial crops, such as celery and barley species (Chiykowski 1979). It should be pointed out
that the primary vectors for this disease are leafhoppers from the genus Macrosteles (e.g., M. quadrilineatus
[= M. fascifrons] and M. phytoplasma) (Chaput and Sears 1998; Heu et al. 2002). The aster yellow virus is
thought to be native to eastern Oregon, Washington, and Idaho (Lenzen and Hutchison, 2002, Oregon State
University Extension Services 2002); however, it has been introduced into the Midwest and is of concern to
commercial vegetable (i.e., carrots) growers there. In the Hermiston, Oregon area, lcafhoppers enter potatoes
when surrounding vegetation desiccates. Symptoms of infected plants are rolled up tip leaves, development of
an off-green or yellowish cast, acrial tubers form, and interveinal Icaf tissue dics. Further research is needed
to examine if A. argentarius is a vector of commercial crops in Illinois.

As previously stated in CTAP Terrestrial Arthropods Reports (Dietrich and Biyal 1998 and 1999 unpub-
lished), CTAP provides a unique opportunity to compile invaluable new data on Illinois terrestrial arthropod

ecology, distribution, and diversity. By incorporating this type of data, a close-to-complete picture of the
quality of ecosystems in Illinois can be obtained.

References
Chaput, J. and M. Sears. 1998. The Aster leafhopper and aster yellows. Ministry of Agriculture and Food,
Division of Agriculture and Rural, Ontario, Canada. Factsheet No. 98-071.

Chiykowski, L. N. 1979. Athysanus argentarius, an introduced European leafhopper,
as a vector of aster yellows in North America. Can. J. of Pathol. 1: 37 — 41.

DeLong, D. M. 1948. The leafhoppers or Cicadellidae of Illinois (Eurymelinae-Balcluthinae). Bull. Ill. Nat.
Hist. Surv. 24: 97-376.

Dietrich, C. H. and M. Biyal .1998. Critical Trends Assessment Project (CTAP): Report on Terrestrial
Arthropods, 1997. Illinois Natural History Survey, Center for Biodiversity. Unpublished Report.

Dietrich, C. H. and M. Biyal .1999. Critical Trends Assessment Project (CTAP): Report on Terrestrial
Arthropods, 1998. Illinois Natural History Survey, Center for Biodiversity. Unpublished Report.

Dietrich, C. H. 2000. Leafhopper FAQ’s.
http://www. inhs.uiuc.edu/~dietrich/IfhFAQ.html [Accessed: 11/16/02]

Hamilton, K. G. A. 1983. Introduced and native leafhoppers common to the old and new worlds (Rhynchota:
Homoptera: Cicadellidae). Can. Ent. 115: 473 - 511.

Heu, R.A., R.T. Hamasaki, B.R. Kumashiro, and S.K. Fukuda. 2002. Aster leafhopper. Plant Pest Control
Branch, Division of Plant Industry, Hawaii Department of Agriculture, Hawaii. New Pest Advisory
No. 02-01.

Ishihara, T. 1953. A tentative checklist of the superfamily Cicadelloidea of Japan (Homoptera). Sci. Rep.
Matsuyama Agr. Coll. 11: 1 —72.

Kwon, Y. J. 1983. Classification of leafhoppers of the subfamily Cicadellinae from Korea (Homoptera:
Cicadellidae). Korean J. Ent. 13: 15-25.

Lenzen, B. and W.D. Hutchison. 2002. Aster leafhopper. Minnesota Extension Service, University of Minne-

sota. http://www.vegedge.umn.edu/vegpest/colecrop/aster.htm [Accessed: 11/16/02]

Linnavuori, R. 1952. Studies on the ecology and phenology of the leafhoppers
(Homoptera) of Raisio (s. w. Finland). Ann. Soc. Zool. Zool. Bot. Fennicae “Vanamo” 14: 1-31.

Massee. 1943. Evacanthus interruptus egg. Ent. Mon. Mag. 79: 270.

McKamey, S. H. 1999. Biodiversity of tropical Homoptera with the first
data from Africa. Amer. Ent. 48: 213 — 222.

Mohlenbrock, R. H. 1986. Guide to the vascular flora of Illinois. Revised and
enlarged edition. Southern Illinois University Press, Carbondale, Illinois.

Mohlenbrock, R. H., and D. M. Ladd. 1978. Distribution of Illinois vascular plants.
Southern Illinois University Press, Carbondale, Illinois.

Oregon State University Extension Services. 2002. Plant Disease Control: Potato (Solanum tuberosum )-

Aster Yellows (Late-breaking Virus). http://plant-disease.orst.edu/disease/cfm?RecordID=881

[Accessed 11/25/02].
Ossiannilsson, F. 1983. The Auchenorrhyncha (Homoptera) of Fennoscandia
and Denmark. In Fauna Entomologica Scandinavica, pp. 868 — 870. Scandinavian Scientific Press

Ltd., Vinderup, Denmark.

Vilbaste, J. 1980. On the Homoptera-Cicadinea of Kamchatka. Annls Zool. 35:
367-418.

Yeop E. H., and Y. J. Kwon. 1994. Systematic and biogeographic studies on the
subfamily Cicadellinae from Korea (Homoptera: Cicadellidae). Ins. Koreana 11: 99 -159.

Natural History Survey
Library

3 0112 06733546

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Critical trends assessment Program 2002 report"

See other formats