Development Of Index Of Biotic Integrated Expectations For The Ecoregions Of Indiana Volume 3 Northern Indiana Till Plain


         United States
         Environmental Protection
         Agency
Region 5
77 West Jackson Boulevard
Chicago, Illinois 60604
EPA905-R-S6-002
NTIS
September 1997
a
&EPA   Development of Index of
         Biotic Integrity Expedtions for
         the Ecoregions of Indiana
         III. Northern Indiana Till Plain

-------
                                                                            EPA 905/R-96/002
•\    Development of Index  of Biotic Integrity Expectations for  the Ecoregions

                      of Indiana:  III.  Northern Indiana Till Plain
                                        Thomas P. Simon
                               U.S. Environmental Protection Agency
                              Watershed and Non-Point Source  Branch
                                     77 West Jackson, WW-1CJ
                                       Chicago. IL  60604
                                    In  cooperation  with:
                          Indiana Department of Environmental Management
                                       Assessment Branch
                               100 North Senate Ave..  P.O. Box 6015
                                   Indianapolis,  IN 46206-6015
                                          July 1, 1997
                                    U.S. Environmental Protection Agency
                                    Region 5. Library (PL. 12J)
                                    77 West Jackson Boulevard, 12th Root
                                    Chicago, IL  60604-3590

-------
                                            NOTICE

Use of this document  is  intended  for the objective facilitation of information exchange between
the States and Federal Uater Pollution Control Biologists for which  it was  intended.   Mention
of trade names or commercial products does  not  constitute endorsement or recommendation for use.
When citing this document:

T.P. Simon.  1997.  Development of Index of Biotic Integrity expectations for the Ecoregions of
Indiana.  III.  Northern  Indiana Till  Plain.  U.S. Environmental Protection Agency,  Region  V, Water
Division. Watershed and Non-Point Source Branch. Chicago.  IL   EPA 905/R-96/002.
 If requesting copies of this document:

 U.S. Environmental Protection Agency
 Publication Distribution Center. ODD
 11027 Kenwood Road. Bldg. 5  - Dock 63
 Cincinnati, OH  45242

-------
                                             TABLE  OF  CONTENTS
        Section	                                  Page

        I.     List of  Figures                                                                n
        ii.    List of  Tables                                                                 iii
        111.   Executive Summary                                                              v
        iv.    Acknowledgements                                                               vi
        1.0  INTRODUCTION                                                                      1

           Definition of Reference Conditions                                                  2
           Criteria for Selecting Reference Sites                                              3
        2.0  STUDY AREA                                                                        3

           Physiographic Provinces                                                             5
           Ecoregions                                                                          5
           Drainage Features                                                                   7
           Historical St. Joseph River data                                                    8
        3.0  MATERIALS  AND  METHODS                                                           9

           Sampling                                                                            9
                Site Specific                                                                   9
                Habitat                                                                         9
                Community Analysis                                                             11
           Metrics                                                                            17
           Scoring Modifications                                                        '      53
         4.0   RESULTS AND DISCUSSION                                                           54

           St.  Joseph River Drainage                                                          54
           Biocntena Comparison of the Southern Michigan-Northern Indiana Til" Plain        54

         5.0  LITERATURE CITED                                                                  59

             APPENDICES

           A.  Tolerance,  trophic,  and  reproductive guilds classification for computing
              the Index of  Biotic  Integrity  for  Indiana taxa.
           B.  Site classification  percentages  based on  individual metric attributes.
           C.  Fish nomenclature changes  for  the  species of  fish occurring within the
               political  boundaries  of Indiana.
-"

-------
                                      LIST  OF  FIGURES

Figure
Number	:	                 Page

1      Map of Indiana showing the major and minor drainage basins
       (after USGS drainage maps).                                                     4

2      Map of Indiana showing the ecoregions designation
       of Omermk and Gallant (1988)                       '                             6

3      St. Joseph River drainage indicating the location of  sampled  locations
       during 1991.                                                                   10

4      Maximum species richness  lines  for  determining  trends  in  total
       number of species with increasing drainage area for the St. Joseph
       River drainage.                                                               20

5      Maximum species richness  lines  for  determining  trends  in  number
       of darter/madtom/sculpin  species with increasing  drainage area  for
       the St. Joseph River drainage.                                                 23

6      Maximum species richness  lines  for  determining  trends  in  the
       proportion of headwater species with increasing drainage  area for
       the St. Joseph River drainage.                                                 26

7      Maximum species richness  lines  for  determining  trends  in  number
       of sunfish species with increasing  drainage area for  the  St. Joseph
       River drainage.                                                               27

8      Maximum species richness  lines  for  determining  trends  in  number  of
       minnow species with increasing  drainage area for the  St.  Joseph  River
       drainage.                                                                     30

9      Maximum species richness  lines  for  determining  trends  in  number
       of sucker species with increasing drainage area for the St. Joseph
       River drain ge.                                                               31

10     Maximum species richness  lines  for  determining  trends in  number
       of sensitive species with increasing drainage area for the
       St. Joseph River drainage.                                                    35

11     Maximum species richness lines for  determining  trends in  the
       proportion of  tolerant species with  increasing  drainage area  for
       the St. Joseph River drainage.                                                 38

 12     Maximum  species  richness  lines for determining trends in the
       proportion of  ommvores with increasing drainage area for the
       St. Joseph River  drainage.                                                     39

-------
LIST OF FIGURES (CONTINUED)

Figure
Number Page

13 Maximum species richness lines for determining trends in the
proportion of insect!vores with increasing drainage area for the
St. Joseph River drainage. 42

14 Maximum species richness lines for determining trends in the
proportion of pioneer species with increasing drainage area for the
St. Joseph River drainage. 44

15 Maximum species richness lines for determining trends in the
proportion of carnivores with increasing drainage, area for the
St. Joseph River drainage. 45

16 Maximum species richness lines for determining trends in the
catch per unit effort with increasing drainage area for the
St. Joseph River drainage. 47

17 Maximum species richness lines for determining trends in the
proportion of simple lithophil species with increasing drainage
area for the St. Joseph River drainage. 50

18 Maximum species richness lines for determining trends in the
proportion of diseased, eroded fins, lesions, and tumors (DELT)
with increasing drainage area for the St. Joseph River drainage. 52
111

-------
LIST OF TABLES
Table
Number . Page

1 Attributes of Index of Biotic Integrity (IBI) classification,
total IBI scores, and integrity classes from Karr et al. (1986). 14

2 Index of Biotic Integrity metrics used to evaluate headwater streams
(< 20 miles2 drainage area) sites in the St. Joseph River drainage 15

3 Index of Biotic Integrity metrics used to evaluate wadable river
(> 20-1.000 miles2 drainage area) sites in the St. Joseph River drainage. 16

4 The distributional characteristics of Indiana darter (Etheostomatim).
madtom (Noturus). and sculpin (Cottus) species. 22

5 List of Indiana fish species considered to be headwater species for
Evaluating permanent habitat in streams (Smith. 1971). 24

6 List of Indiana sunfish species for evaluating quality pool habitats. 25

7 Distributional characteristics of Indiana sucker species (family
Catostomidae) in the St. Joseph River drainage. 29

8 List of Indiana fish species considered to be sensitive to a wide
variety of environmental disturbances including water quality and
habitat degradation. 34

9 List of Indiana fish species considered to be highly tolerant to
a wide variety of environmental disturbances including water
quality and habitat degradation for headwater and wadable river sites. 37

10 List of Indiana fish species considered to be ommvores. 39

11 List of Indiana fish species considered to be indicators of temporally
unavailable or stressed habitats (Lanmore and Smith 1953; Smith 1971) 43

12 List of Indiana species considered to be simple lithophilic
spawners. 49

13 Species list of taxa collected in the St Joseph River drainage. St. Joseph
Elkhart, and Little Elkhart River drairages. Indiana, during 1991. 15

14 Comparison of Michigan DEQ (1996) reference conditions derived from
Procedure 51 with reference conditions developed from Indiana's portion
of the Northern Till Plain. 58
IV

-------
EXECUTIVE SUMMARY

The Clean Water Act Amendments of 1987 suggest the development of biological criteria for
evaluating the quality of the nation's surface waters. The St. Joseph River drainage was
investigated in Indiana to determine water resource expectations for the Southern Michigan-
Northern Indiana Till Plain. A total of 104 sites were sampled in the St. Joseph River
drainage in order to develop and calibrate an Index of Biotic Integrity for use in this
region of Indiana. Based on anticipated variance within the St. Joseph River basin, sub-
drainages were established using natural divisions as recognized by Homoya et al (1985).

Three sub-drainages are recognized and include the major drainage units of the St. Joseph
River: St. Joseph River, Elkhart River, and Little Elkhart River drainages. Graphical
analysis of the data enabled the construction of maximum species richness lines for
calibrating the Index of Biotic Integrity for 12 metrics, as modified for application to
headwater and mid-sized wadable rivers. Metrics were primarily based on the previous works
of Karr (1981), Karr et al. (1986), Ohio EPA (1987), and Simon (1991). Metrics are similar
to those developed for the East Branch Little Calumet Division of the Lake Michigan
drainage. Central Corn Belt Plain. This includes the number of minnow species, sunfish
species, and a combination of sensitive benthic insect!vores, e.g. darters, madtoms, and
sculpins.

Separate metrics were developed for headwater streams (< 20 miles2) and wadable river (20-
1000 miles2) drainage area. Scoring criteria modifications were instituted when less than 50
individuals were collected from a sampling location. This affected the trophic composition,
tolerance, simple lithophil, and DELT proportional metrics. Stations with drainage areas
less than 20 miles2 used a metric which included darters, madtoms, and sculpins (all benthic
insectivores). These species are sensitive indicators of a high quality aquatic resource.
In reaches with drainage areas greater than 20 miles2 a metric evaluating only darter species
was used following the original IBI. The proportion of pioneer species was substituted for
the proportion of carnivores in small headwater streams. The number of sunfish species was
retained for both categories of stream sizes.

The distribution of IBI scores represented a normal curve for all St. Joseph River
subdrainages. The trend was towards increasing biological integrity with increasing drainage
area in all subdrainages. The only exception was the St. Joseph River mainstem which
possessed considerably better fish community structure and function characteristics in the
headwaters.

-------
ACKNOWLEDGEMENTS

The U.S. Environmental Protection Agency wishes to express their appreciation to Wayne
Davis. Valerie Jones, and Bom face Thayil, USEPA-Region V. Ambient Monitoring Section, and
John Winters. Denms Clark, and Lee Bridges, Indiana Department of Environmental Management
(IDEM) who enabled this project to be completed by managing and facilitating logistics and
sampling needs. Thomas Lauer, Fisheries Scientist, Indiana Department of Natural Resources
provided information from the Department's stream reports that enabled sampling at reference
sites. Field assistance was provided by Andrew Ellis, Doug Campbell, and Gregory
Nottingham, Steve Newhouse, James Stahl, and Stephen Wente. IDEM biologists. John Dustman,
Indiana University-Northwest, provided work space to process the large volume of samples.
Hydrologic unit and ecoregion maps were prepared by George Graettinger. USEPA. Water
Division, GIS Section. Numerous professional courtesies were provided by colleagues which
facilitated completion of this project: Shelby Gerking, Arizona State University. Chris
Yoder, Marc Smith, and Ed Rankin. Ohio EPA Historic records were provided by Susan Jewett.
National Museum of Natural History, Douglas Nelson and Gerald Smith, University of Michigan
Museum of Zoology: and William Eschmeyer, California Academy of Science. Special thanks to
Wayne Davis. Lee Bridges, and Steve Newhouse for constructive review comments on a previous
draft of the manuscript. The project manager, chief scientist, and author of this report
was Thomas P. Simon. Ph. D., Regional Biocnteria Coordinator.
Guest Reviewers:

Dr. Marc Oemke, Michigan Department of Environmental Quality
Mr. Marc Smith, Ohio Environmental Protection Agency
Dr. Lizhu Wang, Wisconsin Department of Natural Resources
Or Oenms Clark. Indiana Department of Environmental Management
Mr. Steve A. Newhouse, Indiana Department of Environmental Management
Mr. James Stahl. Indiana Department of Environmental Management
VI

-------
       Development  of  Index of  Biotic  Integrity Expectations for the
             Ecoregions of Indiana.  III. Northern  Indiana Till  Plain
 1.0 INTRODUCTION

 The term "biological  integrity" originated
 in the Water  Pollution Control Act
 Amendments  of 1972  (PL 92-500) and has
 likewise appeared in  subsequent versions
 (PL 95-217; PL 100-1). Karr and Dudley
 (1981) defined biological  integrity as.
 "the ability  of an  aquatic ecosystem to
 support and maintain  a balanced,
 integrated, adaptive  community of
 organisms having a  species composition,
 diversity,  and functional  organization
 comparable  to the best natural habitats
 within a region". The use  of a biological
 component to  evaluate the  ambient lotic
 aquatic community of  our nations surface
-waters has  been well  discussed elsewhere
 (Karr et al.  1986:  Ohio EPA 1987; Whittier
 et al. 1987;  Simon  et al.  1988; Davis
 1990; Fausch  et al. 1990;  Karr 1991).

 An assessment of the  St. Joseph River
 drainage enabled the  objective evaluation
 of specific metrics performance and
 evaluation  of reference conditions for the
 Northern Indiana Till Plain. The St.
 Joseph River  drainage has  impacts
 associated  with channelization and
 damming, agriculture, and  municipal and
 point source  dischargers.  The primary
 point sources are municipal facilities,
 chemical manufacturers,  and hydro-electric
 power generating  stations  distributed  in
 the main population centers of the basin.
 The affects of channelization and
 agriculture have  been well documented
 including thermal  increases  (Raney and
 Menzel  1969;  Brown  1976;  Brungs  and Jones
 1977; Hokanson and  Biesinger  1980: USEPA
 1980; McCormick et  al.  1981:  EPRI  1981);
 increased nutrient  and  allochthanous
 input,  and runoff and riparian  zone
 clearing.

 The  objective  of this study  was  to
 evaluate the  biological  integrity of
Indiana water resources based on "least
impacted" reference conditions for
establishing baseline conditions (Hughes
et al.  1986).   Least impacted reference
sites are representative of the watershed
under study and reflect the better sites
with minimum anthropogenic change.  Least
impacted is not synonymous with pristine.
Rather,  sites  are selected for their
representativeness of the area.   The  St.
Joseph  River drainage and the historical
Great Marsh have been dramatically changed
over the last  250 years with the draining
of the  wetland and the intensive ditching
projects that  completely changed the
landscape.   The following project  goals
were addressed during the Southern
Michigan-Northern Indiana Till  Plain
biological  criteria project-

o Develop biological  criteria for
  headwater,  mid-size,  and large river
  reaches using the Index of Biotic
  Integrity;

o Identify areas of least disturbance
  within the Northern Indiana Till
  Plain for establishing reference
  conditions;

o Develop maximum species richness  (MSR)
  lines from the reference database for
  each IBI metric as a log function of
  drainage area,

o Compare biocnteria to State of Michigan
  expectations for ecoregion

This technical report includes specific
Index of Biotic  Integrity criteria
including the development of metrics  and
maximum  species  richness lines, to
delineate areas  of least disturbance  in
the  Northern  Indiana Till Plain ecoregion.

-------
Indiana Ecoregion
Limited field collection has been
conducted in Indiana since the completion
of Gerking's distribution of Indiana
fishes. Less than 2% of Indiana's surface
waters had been assessed at the beginning
of this study.   Since limited information
was available for the selection of least
disturbed stations,  we attempted to sample
representative stream types of this region
in order to determine where least impacted
stream segments occurred.

Definition of Reference Conditions

In order to make accurate evaluations  of
the biological  condition of the region.
various baseline geological,  geographic,
and climatic differences need to be
assessed. The goal  is not to provide a
definition of pristine conditions, since
these types of conditions are either few
in number or nonexistent in heavily
populated states (Hughes et a/. 1982;
Whittier et al. 1987). Our expectations
are determined from the structural and
functional attainable natural conditions
of "least impacted" or reference
conditions. Assessment of these criteria
need to be modified nationally, since
regional differences can be attributed to
the expectations based on structure and
function that  determine the distribution
of fishes. The ecoregion concept  is useful
for clustering large  homogeneous  regions,
since  these areas are  influenced  by
different  physical  processes  (Omermk
1987).

 In order to  select  stations  for  sampling
it is  necessary  to  know  the  geographical
boundary of  the "ecoregions"  within the
State of Indiana.  A valid  ecoregion has
 boundaries where ecosystem variables  and
 patterns emerge (Hughes  et al.  1986).
 Omermk (1987) mapped the  ecoregions  of
 the  conterminous United States from maps
 of land-surface form, soil types.
 potential natural  vegetation, and land
 use.  Each ecoregion was then based on
areas of regional homogeneity. Ecoregions
became a very useful mechanism for
determining community complexity and for
establishing  boundaries associated with
various land forms.

Ecoregions provide a geographical
framework for determining the appropriate
response for streams of similar proportion
and complexity.  Reference conditions are
used for establishing the areas of "least
impact",  and will reveal the current
conditions of the surface waters of
Indiana  Once ecoregional expectations are
determined it is important to consider
that conditions  do not remain static.  On
the contrary, repeat monitoring and
sampling of stations, both reference and
site specific will need to be conducted in
order to document change over time and
further refine the IBI.

Reference conditions are not the same as
reference sites. Reference conditions are
the subtle patterns that emerge from the
regional database. Few if any nonimpacted
sites occur in North America, thus in
order to determine the extent of
degradation important attributes of stream
fish communities are analyzed to determine
the patterns of  "least impacted"
communities. The relevance of including
some sites which are not considered
pristine or "reference sites" is not
important because it is only the upper 5%
of the sites that determine the maximum
species richness  lines or 95th percentile
lines.

Because of  subregional  differences.
further demarcation  was  made by examining
the  role  of the basin or the watersheds
within ecoregions.  Fish  composition and
community structure is  determined, within
a natural  area  by the  availability of
water of appropriate quality and quantity
 to ensure existence, provide routes  of
 emigration, sustain growth,  and increase
 fitness through reproduction.  Likewise.

-------
Northern Indiana Till Plain
species-specific differences exist in
community structure which may not reveal
differences in current water quality but
may be determined by historical geomorphic
(Leopold et al. 1964) or zoogeographic
processes (Hocutt and Wiley 1986). Trends
in Indiana water quality were therefore
evaluated using a watershed approach
within an ecoregion framework.
Criteria for Selecting Reference Sites

Several procedures are available for
determining reference conditions. Larsen
et al. (1986) and Whittier et al. (1987)
chose sites after careful examination of
aerial photographs, watershed specific
information review, on-site
reconnaissance, and expert consultation.
This procedure requires that a limited
number of high-quality sites be sampled in
order to predict regional expectations.
The methods chosen for site selection were
based on the evaluation of Regional Water
Quality Planning Maps (USGS undated) that
identified known impact sources and
diffuse nonpoint sources that could
potentially influence a site. A balanced
distribution of sites within all parts of
the St. Joseph River drainage was
maintained against historic collections
sites (Jordan 1877; Gerking 1945; IDEM
1990). All sites were rigorously sampled
in order to get representative, distance
specific, quantitative estimates of
species richness and biomass. Maximum
species richness lines were then compiled
(see methods below), followed by
calculations of the Index of Biotic
Integrity values to reveal that stations
that were the "least impacted" stations
for the St. Joseph River drainage.

Reference sites are defined as the
stations which cumulatively define the
95th percentile line of the individual
metrics. Evaluation of habitat and other
physical parameters refined the final list
of reference sites. Sites that had habitat
or water quality deficiencies, but still
attained high index ratings would have
been removed from the final list. This
action was not required, since poor
habitat and water quality affected various
portions of the community resulting in a
lowered index score. These sites are not
pristine or undisturbed (few exist in
Indiana), but they do represent the best
conditions given the background activities
(i.e. anthropogenic impacts;
channelization; cultural eutrophication).

Sampling was conducted in all size classes
of river reaches in the St. Joseph,
Elkhart, Pigeon, and Little Elkhart Rivers
from the headwater (<20 mile2) to the
largest mainstem drainage area (ca. 1,000
mile2) in Indiana.
2.0 STUDY AREA

Indiana has an area of 36,291 square
miles, and drains the Ohio, the upper
Mississippi, and Great Lakes Regions
(Seaber et al. 1984). These three regions
were further subdivided into nine
subregions (Fig. 1), five of which drain
86% of the State (USGS 1990). The State of
Indiana lies within the limits of latitude
37° 46' 18" and 41° 45' 33" north, for an
extreme length of 275.5 miles in a north-
south direction; and between longitude 84°
47' 05" and 88° 05' 50" west with an
extreme width in an east-west direction of
142.1 miles.

The State has a maximum topographic relief
of about 900.9 ft, with elevations ranging
from about 300.3 ft above mean sea level
at the mouth of the Wabash River to
slightly more than 1,201.2 ft in Randolph
County in the east-central part of the
state.

This report considers only the St. Joseph
River drainage. The St. Joseph River

-------
 Indiana Ecoregions
   71
Great Lakes  Basin
Mississippi River
Ohio  River
Regional Boundary
Sub-Regional Boundary
Accounting Unit  Boundary
Cataloging  Unit  Boundary
County  Boundary
State  Boundary
Figure  1.   Map of Indiana showing Major and  Minor drainage basins  (from USGS data).
                                              4

-------
Northern Indiana Till Plain
drains an area of 4,285 mile2 (Hoggatt '
1975). It is contained within a single
ecoregion and drains 4.7% of Indiana. The
entire St. Joseph River watershed is
contained within the Northern Indiana Till
Plain ecoregion (Omermk and Gallant,
1988). The St. Joseph River is located in
northcentral and northeastern Indiana and
drains in a northeastern direction into
Michigan and then Lake Michigan. The St.
Joseph River is one of the largest
tributaries of Lake Michigan.
Physiographic Provinces

Fenneman (1946) divided the State into two
physiographic provinces based on the
maximum extent of glaciation. The
glaciated portion of the State contains
the Central Lowland province, which
includes the majority of the St. Joseph
River drainage, and the unglaciated
portion is termed the Interior Low
Plateaus province.

Schneider (1966) further divided Indiana
into three broad physiographic areas that
closely reflect the surface-water
characteristics of the State. The St.
Joseph River drains a portion of the
Northern Lake and Moraine Region.

The Northern Lake and Moraine Region
covers the northern one-fourth of the
State and is of variable relief. Its
characteristic deep peat deposits and
small lakes are restricted to the rugged,
terminal moraines. Numerous broad
lacustrine and outwash plains occur, often
marked by wide marshes (or marshes now
drained) broken by low sand ridges or
knolls. The northern section of the State
was covered during the most recent
Wisconsinan glacial event.

The last major glaciation event
dramatically altered northern Indiana
during the Wisconsinan period (14,000 to
22,000 years ago). As glaciers advanced
and retreated, the land surface was
dramatically altered as the landforms were
either scoured by advancing glacial ice o'r
the scoured materials were deposited by
retreating glaciers. Two distinct glacial
lobes are known to have advanced into
Indiana, from the northeast out of Lake
Erie and Saginaw Bay basins and from the
north from the Lake Michigan basin.
Ecoregions

Omernik and Gallant (1988) characterized
the attributes of ecoregions of the
midwestern states. Indiana has six
recognized ecoregions: Central Corn Belt
Plain, Huron-Erie Lake Plain, Southern
Michigan-Northern Indiana Till Plain
(referred to as Northern Indiana Till
Plain), Eastern Corn Belt'Plain, Interior-
Plateau, and Interior River Lowland (Fig.
2). The St. Joseph River basin drains the
Northern Indiana Till Plain ecoregion
(Omernik and Gallant (1988).

Northern Indiana Till Plain

Much of the ecoregion consists of
extensive crop and livestock production.
It is distinguished from other adjacent
ecoregions by the natural forest cover, a
high degree of urbanization, and extensive
quarrying. The broad, nearly flat to
rolling glaciated plain includes deeply
mantled glacial till and outwash. sandy
and gravelly beach ridges and flats, belts
of morainal hills, and bog kettle
depressions. Elevations range between a
few feet in the flatter portions of the
range to 600 ft along the shores of the
Great Lakes to over a 1000 ft on some
moraines. Streams are sluggish and are
bordered by riverine wetlands throughout
the 25,800 miles2 of the ecoregion.
Perennial streams include drainage ditches
and channelized streams with a density of

-------
Indiana  Ecoregions
              S/   '   •   v     r
              >  /Uy     .4
             -•fV   r  v
                                            V
Central Corn Belt  Plains


Eastern Corn Belt  Plains


Northern  Indiana  Till  Plains


Huron/Erie  Lake Plains


Interior Plateau


Interior River  Lowland
                      'J
 Figure  2.   Map  of Indiana showg  the  ecoregion  designation from  Omernik  and Gallant  (1988).

                                               6

-------
Northern Indiana Till Plain
one half mile per square mile. Lakes are,
common in some areas, however, many
depressions are filled with peat deposits
or dark mineral soils. Precipitation
occurs mainly during the growing season
and averages from 35 to 46 inches
annually.

Both perennial and intermittent streams
are common in the ecoregion. Intermittent
streams comprise between 10-15& of the
streams. Constructed drainage ditches and
channelized streams further assist in soil
drainage in flat, poorly drained areas.
Stream density is approximately one half
mile per square mile in the most typical
portions of the ecoregion.

The ecoregion is managed for cropland,
livestock, forest and woodland, and urban
use. Crops are mainly in corn, other feed
grains, soybeans, and hay for livestock.
Emphasis on livestock includes the dairy
cattle, but beef cattle, swine, sheep, and
poultry are also important. Approximately
25£ of the ecoregion is urbanized.

Most of the soils were developed under the
influence of deciduous forest vegetation.
The soils are derived from loamy glacial
drift and till. Hapludalfs and
Ochraqualf's are the dominant soil groups.
while, poorly drained soils include
Argiaquolls, Haplaquolls, Haplaquepts, and
Psammaquents. Udipsamments form on better
drained outwash plains. Medipnsts and
other organic soils can be found on
depressions.

The natural vegetation of the area
consists of diverse hardwood forests,
predominantly oak and hickory. However, a
significant amount of white oak. red oak.
black oak. bitternut hickory, shagbark
hickory, sugar maple, and beech exists.
This vegetation is not common in the mixed
forest to the north or on the poorly
drained lake plain to the south and east.
letter sites include red maple, white oak.
American elm, and basswoods. White ash,
red maple, quaking aspen, and black cherry
grow along rivers and stream corridors.
Drainage Features

Three major drainage units occur in the
Northern Indiana Till Plain of Indiana:
the St. Joseph River. Elkhart River.
Pigeon River, and the Little Elkhart River
drainages. Further mention of the St
Joseph River basin will refer only to that
contained in Indiana.
St. Joseph River

The St. Joseph River basin drains 1699
square miles in northern Indiana and 2586
square miles in southern Lower Michigan
(IDNR 1987). The Indiana portion of the •.
basin contains a unique combination of
natural lakes, wetlands streams with well-
sustained flows, and extensive sand and
gravel aquifer systems. During dry years
ground water contributes 70% of the stream
flow (Indiana Department of Natural
Resources 1980). The St. Joseph River
basin is characterized by more than 200
natural lakes, approximately 27,000
wetlands, and low-gradient streams develop
on outwash and till deposits. The St.
Joseph River drains 4.7% of the State. The
St. Joseph River begins near Hillsdale.
Michigan, and generally flows to the
southwest. In South Bend. Indiana, the
river turns sharply to the north, then
flows northwest until it enters Lake
Michigan near Benton Harbor, Michigan.
Approximately 41 miles (19%) of the St
Joseph mainstem is in Indiana. Average
channel slope is 2.5 feet per mile. Minor
tributaries include Christiana Creek.
Juday Creek. Baugo Creek, and Cobus Creek.
Average discharge for the St. Joseph River
during the period of record (1947-
present). at Elkhart. was 3.176 cfs Flow
ranged from a minimum daily discharge of

-------
Indiana Ecoreaion
336 cfs, the 7 day. 10 year low flow of
818 cfs, to 18.600 cfs during maximum
discharge during the period of record
(An/in,  1989).
Elkhart River

The Elkhart River drainage is the major
southwestern segment of the St.  Joseph
River (comprising 699 miles2) mainly in
Noble and Elkhart Counties.  The  Elkhart
River has been impounded by numerous  low-
head dams upstream of the City of Elkhart.
and receives a substantial  amount of  Us
streamflow from groundwater  The major
tributary segments of the Elkhart River in
Indiana includes North Branch and South
Branch Elkhart Rivers. Solomons  Creek.
Yellow Creek, Stony Creek,  and Rock Run
Creek-Horn Ditch. The average discharge of
the Elkhart River near Goshen (Elkhart
County) during the period of record  (1931
to present) was 514 cfs. Discharge ranged
from a minimum daily low of 7.0  cfs.  81
cfs during the 7 day, 10 year low flow, to
6,180 cfs during maximum flow periods
(Arvin,  1989).
Little Elkhart River

The Little Elkhart River drainage is the
major central segment (draining 129 miles2)
which connects with the St  Joseph River
near Bristol. Elkhart County. The Little
Elkhart River is unaffected by man-made
regulation,  and receives a substantial
amount of  its streamflow from groundwater
The River  emanates southeast from Lagrange
County and flows northwest. Major
 tributary  segments of the  Little Elkhart
 include  Mathur  Ditch. Rowe-Eden Ditch,  and
 Emma  Creek.  The average discharge of  the
 Little  Elkhart  River  at Middlebury
 (Elkhart County)  was  100  4 cfs  during the
 period  of record (1971-1979. .new station
 1979-present).  Discharge  records  ranged
 from 32 cfs during minimum daily flow.  8  0
 cfs during  the 7  day,  10  year low flow,  to
 1.690  cfs during  maximum  flow periods
 (Arvin.  1989).
Pigeon River

The Pigeon River drainage  is  the  major
northern segment (draining 374 miles2)
connecting with the St  Joseph River  in
southwestern Michigan. The Pigeon River  is
unaffected by man-made regulation, and
receives a substantial amount of  its
streamflow from groundwa^er   The
headwater of the River is in Steuben
County and flow is northwest.  Major
tributary segments of the Pigeon River
include Fawn Creek. Fly Creek, and Turkey
Creek  The average discharge of Pigeon
River near Angola (Steuben County) was
78.9 cfs during the period of record
(1945-present).  Discharge records ranged '•
from 3.4 cfs during minimum daily flow,
5.8 cfs during the 7 day, 10 year low
flow, to 795 cfs during maximum flow
periods (Arvin.  1989).
Historical St. Joseph River Data

The St. Joseph River is the largest
Indiana tributary to Lake Michigan and
historically was stated to be one of
Indiana's highest quality resources   The
first  use of the St. Joseph River was as  a
commercial connection for fur traders
between the Great Lakes and the
Mississippi River   Portage between  the
Great  Lakes and  Mississippi River was
through the Great Marsh of the Kankakee
River. Since  then the St. Joseph River  has
been  intensively examined  including  its
 importance as a  trade  route  (Fatout  1985):
 hydrology (Bailey et  al.  1985:  Crompton et
 al.  1986; Hoggatt 1981):  surficial  geology
 and physiography (Indiana  DNR 1987:  Gray
 in preparation): Paleozoic and Quarternary
 geology (Gooding 1973:  IDNR 1987):
 drainage characteristics (IDNR  1987).

-------
                                                                       Northern Indiana Till Plain
stream-flow (Stewart 1983: Glatfelter
1984); groundwater flow (IDNR 1987);
limnology and chemistry of lakes (Scott.
1931; Eberly. 1959; Pearson, 1986: Bell
and Spacie 1978. 1979. 1988): nutrient and
sediment transport (IDEM,  unpublished data
from 1985-86).

The aquatic communities of the St. Joseph
River drainage have been correlated with
water quality (Simon and Newhouse, in
preparation). Various components of the
aquatic community of St Joseph River and
tributaries have been studied including
the microbial community (IDEM 1994).
plankton community (Mueller.  1959; IDEM
1994). macroinvertebrates  (Scott, et al.
1928. 1938; Stahl.  1959;  Newhouse.
unpublished data),  while the mussel fauna
was studied by Matters (1988).  The fish
community has also been well  studied in
lakes and streams including distribution
(Gerking 1945; Gulish.-1973); population
dynamics of stocked fish (Ledet 1989a;
Ricker. 1942a.b.c.d,e, 1955); and
fisheries potential of the St Joseph River
(Ledet. 1979. 1986. 1989.  1990). and
tributaries (Peterson. 1971. 1973.
1974a.b. 1979: Pearson, 1981: Peterson and
Ledet. 1982: Ledet, 1989.  1991a.b). Sport
fishes have been studied (Parks, 1949)
including Cisco  (Gulish, 1973. 1974),
bluegill (Ricker, 1942b),  and smallmouth
bass  (Ledet, 1989; Stefanavage. 1987).

The  St. Joseph River possesses a highly
diverse fish community. The earliest
records of Jordan (1877) suggest the  river
was  abundant with both food and non-game
species. The St. Joseph River received a
relatively small amount of collection
effort by early  ichthyologists. Jordan
(1890) collected at a  single site  on  the
St.  Joseph River at Mishawaka and  South
Bend.  Eigenmann  (1896) collected  from
Turkey Creek and several  lakes. Gerking
(1945) collected at approximately  30
localities in the watershed.
3.0 MATERIALS  AND   METHODS

Sampling

  Site Specific

In order to answer basin-specific
questions and to calibrate an IBI for
evaluating ecosystem health, a sufficient
number of samples were required from each
of the various drainages. A total of 104
locations (Fig. 3) were surveyed during
June through August 1991 in order to
compile the data needed to evaluate the
maximum species richness lines for
calibration of the Index of Biotic
Integrity.  Site and collection records are
maintained within the State of Indiana
files. Since the primary purpose of this
study was to evaluate the water quality of
Indiana using biological  methodology,  no
further evaluation of site specific data-,
(e.g.  site specific taxonomic species
lists) will be included other than an
overall taxa list for each sub-drainage.

To ensure repeat sampling at the exact
same site,  all locations are based on
latitude and longitude. Narrative
descriptions for mileage are from the
center ooint rather than the edge of the
nearest town since the boundaries of many
Indiana towns will change over the next
century. All sites were evaluated based on
drainage area, since this provides a
reliable quantification (Hughes et al.
1986) of stream size. As drainage area
increases fewer locations are available
for comparative analysis

  Habitat

The diversity of habitats sampled has a
major effect on data collection. A
"representative" sample always requires
that the entire range of riffle, run.
pool, and extra-channel habitat be
sampled, especially when large rivers are

-------
                        T
                        N
                                                       .S7 \II.OI  ,M/( UK, IA
Figure 3  St  Joseph River drainage indicating the location of sampled  locations
          during 1990 and 1991.

-------
Northern Indiana Till Plain
surveyed. Atypical samples result when
unrepresentative habitats are sampled
adjacent to the sampling site Species
richness near bridges or near the mouths
of tributaries entering large rivers,
lakes, or reservoirs are more likely to be
characteristic of large-order habitats
than the one under consideration (Fausch
et al. 1984)

A general site description of each
established sampling location was
conducted using the field observation
procedure of Ohio EPA (1989) and Rankin
(1989). The Quality Habitat Evaluation
Index (QHEI) takes into account important
attributes of the habitat which increases
heterogeneity. Scoring incorporates
information on substrate composition.
instream cover, channel morphology.
riparian zone and bank erosion, and pool
and riffle quality. Physical/chemical
parameters were recorded for each sample
site to assist in assessing the biological
data further: dissolved oxygen. pH.
temperature, and specific conductivity.
Equipment utilized for physical water
quality analysis was a Hydro!ab SVR2-SU
meter following the specifications of the
manufacturer.
Community Analysis

Sample Considerations

Only one electrofishing gear type is
needed at each location to collect a
representative sample (Jung and
Libosvarsky 1965. Ohio EPA 1989). A T&J
pulsed-DC generator capable of 300 volt
output. 1750 watts was mounted in a
Coleman Sport-canoe, floated in a Sport-
Yak, or attached to a long-line (see Ohio
EPA 1989 or USEPA 1988 for discussion of
gear) We collected by wading in shallow
riffles and runs, and floated through
pools and unwadeable habitat Sampling
included both shorelines in streams > 5 m
or followed a serpentine pattern on both
shores for streams < 5 m.

All fish encountered were collected at
each site. Adult and juvenile specimens
from each stream reach were identified to
species utilizing the taxonomic keys of
Gerking (1955). Trautman (1981). and
Becker (1983). Cypnmd taxonomy follows
Mayden (1989), changes in species
nomenclature are listed in Appendix C for
comparability with previous
investigations. The young-of-the-year fish
less than 20 mm in lengtn are not included
in Index of Biotic Integrity or composite
totals analysis. Early life stages exhibit
high initial mortality (Simon 1989) and
are difficult to collect with gear
designed for larger fish (Angermeier and
Karr 1986). Collection of fish from this
category will be retained for possible
future use in State water monitoring
programs (e.g. ichthyoplankton index (I2)).
Specimens greater than 20 mm TL are easily
collected using our gear. Juvenile
specimen survival at lengths > 20 mm TL
also show many species begin to function
in distinct trophic guilds and reflect
mature species attributes.

Many different study designs can be
employed to assimilate a reference
database. Two broad categories of random
and nonrandom designs can enable sampling
localities to be targeted for specific
habitat types or provide a representative
picture of the area under consideration.
We used a nonrandom study design since our
intentions were to avoid known point and
non-point sources of pollution.

An additional consideration for choosing a
non-random designs include spatial
coverage. Numerous studies (Ohio EPA.
1989) have utilized a reference site
approach that compares the very best an
area has to offer against typical or
representative portions of the region
Additional study is needed to evaluate
11

-------
 Indiana Ecore"ion
site variation and seasonal trends causing
sites to have to be repeat sampled. Ohio
has had a significant advantage in
determining where "least impacted" or
reference sites exist because of the
extensive work of Trautman (1981), and
other ichthyologists before him.
Unfortunately, the historic record for
Indiana begins in the late 1800's  and ends
at approxiamtely 1945 with the published
work of Shelby Gerking.  Little sampling  of
the fish community of Indiana  has  been
completed since this time  Based on
information presented in the  National
305(b)  report  to Congress.  Indiana had
less than 2% of the surface waters
assessed prior to 1990.  We initiated  this
project to determine where these "least
impacted" sites occurred and  assimilated a
database to address immediate  data needs
for biocnteria development. We suggest
that the criteria presented in this
document is a  "first attempt"  to evaluate
Indiana surface waters.

During 1991. drought conditions prevailed
for the Southern Michigan-Northern Indiana
Till Plain  Further research  is needed to
evaluate the response of the  criteria
under differing water cycles.

The length of stream reach sampled is an
important consideration. Karr et a].
(1986) recommended in larger streams  to
select several contiguous riffle-pool
sequences rather than relying on  a
standard length. When electrofishing
equipment was employed  in larger rivers
d  e. >  1.000 mi2),  samples were taken in
units of 0.5  to  1.0 km  (Gammon et  al.
1981)  The  length of the sample reach was
long enough to  include  all major habitat
types. Distances of  11  to  15  stream widths
were generally  adequate to sample  two
cycles  of  habitat  (Leopold et al.   1964).
Ohio EPA (1989)  suggested  that after  150-
200 m  of stream length  no  significant
 increases  in  the IBI  are observed, however
 species  richness may still  increase  until
 250-300  m.  The additional  increase in
 effort  is  not justified by the assessment
 capability  of the  index so the minimum
 distance of 15 times  the mean  stream width
 was  adopted.  Additional  site information
 (e.g. photographs-  latitude and longitude)
 were recorded on the  data  sheet.

 Selecting the appropriate  time  of  year  for
 sampling is critical. Karr  et al   (1986)
 found that periods of low-to moderate
 stream flow are preferred and the
 relatively variable flow conditions.of
early spring  and late autumn/winter should
be avoided. Species richness tends to be
higher later  in summer due to the presence
of young-of-the-year of rare species, but
this can be avoided if data analysis does
not incorporate young-of-the-year species.
Samples of limited area may be less
variable in early summer than comparable
samples taken later in the year  Each site
was sampled for a single pass on both
shorelines for nonwadeable locations.
Sample Site Selection

Fish sample sites were selected based upon
several factors

1)  Choosing stream reaches not affected
     by point source dischargers that were
     typical of the region under study.

2)  Stream  use issues (i.e. municipal
     treatment works, non-point source.
     nutrient reduction).

3)  Location of  physical  stream features
    (e.g. dams,  changes  in geology.
    changes in  stream order,  presence  of
    stream  confluence,  etc.):

 4)  Location of non-point sources  of
      pollution (e.g.  urban areas  or
      obvious farm runoff);
                                              12

-------
Northern Indiana Tilt Plain
5) Variations in habitat suitability for
fish;

6) Atypical habitat not representative of
River reach or basin.

Whenever possible, sites were located
upstream from pollution sources and
adjacent tributaries (Gammon 1973).
Stations were selected to include natural
areas, parks (Federal, State. County, and
Local), exceptional designated streams,
and from historical sampling locations
whenever avaiTable.

When non-impacted areas were not present.
"least impacted" areas were selected based
on the above criteria. Sites were chosen
which indicated recovery from
channelization or potential non-point
source areas, and which had a suitable
riparian buffer on the shoreline. When a
series of point source'dischargers were
located on a river, every effort was made
to sample upstream of the discharger or to
search for areas of recovery between
dischargers (Krumholz 1946).

When impoundments or other physical
habitat alterations had been imposed on a
river, sampling was conducted in the
tailwaters of a dam (area immediately
downstream). Tailwaters possess the
greatest resemblance of the lotic habitat
The serial discontinuity concepts of
Stanford et al. (1988) predicts that the
thermal character of a stream below a dam
will be "reset" toward that typical of the
stream reaches above the dam. In areas
where sampling could not be accomplished
downstream of the physical structure due
to lack of access, stream tributary
segments were located upstream of the dam
away from the immediate influence of the
pooled portion. Likewise, bridges were
sampled on the upstream side, away from
the immediate vicinity of the structure
and latent bridge construction effects. If
downstream sampling was conducted because
of better habitat considerations, sampling
was terminated at least 50 m downstream of
the bridge.

Fish from each location were identified to
species and enumerated. Smaller and more
difficult to identify taxa were preserved
for later examination and identification
in the laboratory. All fish were examined
for the presence of gross external
anomalies. Incidence of these anomalies
was defined as the presence of externally
visible morphological anomalies (i.e.
deformities, erosion, lesions/ulcers)
Specific anomalies include: anchor worms:
leeches; pugheadedness: fin rot: Aeromonas
(causes ulcers, lesions, and skin growth.
and formation of pus-producing surface
lesions accompanied by scale erosion):
dropsy (puffy body); swollen eyes; fungus:
ich; curved spine: and swollen-bleeding
mandible or opercle. Incidence is
expressed as percent of anomalous fish
among all fish collected. Incidence of
occurrence was computed for each species
at each station.

Hybrid species encountered in the field
(e.g. hybrid centrarchids. cyprinids) were
recorded on the data sheet, and when
possible, potential parental combinations
recorded.
Index of Biotic Integrity

The ambient environmental condition was
evaluated using the Index of Biotic
Integrity (Karr 1981; Karr et a '. 1986)
This index relies on multiple parameters
(termed "metrics") based on community
concepts, to evaluate a complex biotic
system. It incorporates professional
judgement in a systematic and sound
manner, but sets quantitative criteria
that enables determination of a continuum
between very poor and excellent based on
species richness and composition, trophic
and reproductive constituents, and fish
13

-------
 Indiana Ecoregion
Table  1. Attributes of Index of Biotic Integrity (IBI) classification, total IBI scores,
           and integrity classes from Karr et  a?.  (1986).
Total IBI
score
       Integrity
         Class
Attributes
58-60
  J-52
40-44
28-34
12-22
Excellent     Comparable to the best situation without human disturbance: all
              regionally expected species for the habitat and stream sue.
              including the most intolerant forms, are present with a full
              array of age (size) classes: balance trophic structure

Good          Species richness somewhat below expectations,  especially due to
              the loss of the most intolerant forms,  some soecies are present
              with less than optimal  abundances or size distributions,  trophic
              structure shows some signs of stress

Fair          Signs of additional  deterioration include loss  of intolerant
              forms,  fewer species,  highly skewed trophic structure (e.g.
              increasing frequency of ommvores and other tolerant species):
              older age classes of top predators  may  be rare.

Poor          Dominated by ommvores.  tolerant forms,  and habitat generalists:
              few top carnivores:  growth rates and condition  factors commonly
              depressed: hybrids and diseased fish often present.

Very Poor     Few fish present, mostly introduced or  tolerant forms; hybrids
              common: disease, parasites,  fin damage,  and other anomalies
              regular.

No Fish       Repeated sampling finds no fish.
abundance and condition.  The twelve
original Index of Biotic  Integrity metrics
reflect insights from several perspectives
and cumulatively are responsive to changes
of relatively small magnitude, as well  as
broad ranges of environmental degradation.

Since the metrics are differentially
sensitive to various perturbations (e.g.
siltation or toxic chemicals), as well  as
various degrees or levels of change within
the  range of integrity, conditions at a
site can be determined with  considerable
                                   accuracy. The  interpretation of the  index
                                   scoring  is  provided  in six narrative
                                   categories  which  have been tested  in the
                                   midwestern  United States  (Karr  1981.  Table
                                    Several  of the metrics  are  drainage size
                                    dependent and require calibration to
                                    determine numerical  scores  (Tables 2-3).
                                    Drainage size effects were  determined by
                                    evaluating trends in species or
                                    proportions of individuals  with increasing
                                    (log adjusted) drainage area  The
                                              14

-------
                                                                      Northern Indiana Till Plain
Table 2. Index of Biotic Integrity metrics used to evaluate headwater streams
         (<20 miles2 drainage  area)  sites  In  the  St.  Joseph River drainage.
Metric
Category
Metric
                                           Scoring Classification
                                                      3               1
Species
Composition
Trophic
Composition
Fish
Condition
Total  Number of Species

Number Darter/Madtom/Sculpin Spp

% Headwater Species

Number of Minnow Species

Number Sensitive Species

% Tolerant Individuals

% Ommvore Individuals1
< 20 square miles

% Insectivores  Individuals1
< 20 square miles

% Pioneer Species Individuals1

Catch per Unit  Effort1

% Simple Lithophil  Individuals1

% DELT Individuals1
Vanes with drainage area (Fig. 4)

Varies with drainage area (Fig. 5)

> 50*          25- <50       < 25 (Fig.  6)

Vanes with drainage area (Fig  8)

Varies with drainage area (Fig. 10)

<33*          33-66*        >66*  (Fig.11)
<33*
33-66*
>66* (Fig.  12)
>60*          30-60*        <30  (Fig. 13)

< 33*         33-66*        >66*  (Fig. 14)

Vanes with drainage area (Fig.  16)

>45*          15-45*        <15*  (Fig. 17)

<0.1*         0.1-1.3*   >1.3* (Fig  18)
  Special scoring procedures  are required  when  less than 50  individual  fish are collected
                                              15

-------
Indiana Ecoreqion
Table 3. Index of Biotic Integrity metrics used to evaluate wadable river
         (>20 -  <1.000 miles2  drainage area) sites  in the St. Joseph River drainage.
Metric
Category
Metric
       Scoring Classification
                  3              1
Species
Composition
Trophic
Composition
Fish
Condition
Total Number of Species

Number of Darter Species

Number of Sunfish Species

Number of Sucker Species

Number of Sensitive Species

% Tolerant Individuals

% Ommvore1  Individuals
< 1.000 square miles

% Insectivores1  Individuals
< 1.000 square miles

% Carnivores1  Individuals


Catch per Unit Effort

% Simple Lithophils Individuals

% DELT ' Individuals
Vanes with drainage area (Fig  4)

> 3           2          < 2   (Fig. 5)

Vanes with drainage area (Fig 7)

Vanes with drainage area (Fig. 9)

Vanes with drainage area (Fig. 10)

<33*          33-66*        >66*  (Fig.  11)
<33*


>60*

>16*
33-66*


30-60*

8-16*
>66* (Fig. 12)


<30* (Fig. 13)

<8*  (Fig. 15)
Vanes with drainage area (Fig.  16)

>45*           15-45*         <15*  (Fig.  17)

1 3* (Fig  18)
  Special scoring procedures are required  when  less  than  100  individual fish are collected.
 ecoregion  approach  developed by USEPA-
 Corvallis.  Oregon,  was  utilized to compare
 "least impacted"  zones  within  the region
 (Omermk 1987).  Ohio EPA (1987). modified
 several of the original 12 metrics in
 order to make them more sensitive to
 environmental effects based on their
                                    experiences in Ohio and to  account for
                                    stream and river size,  faunal  differences.
                                    and sampling gear selectivity. The current
                                    study utilizes the experiences of the Ohio
                                    EPA and Karr et al. (1986)  in developing
                                    an IBI for  Indiana streams  and rivers in
                                    the Northern  Indiana Till Plain
                                              16

-------
Northern Indiana Till Plain
Metrics

In general, the metrics utilized for the
current study are those developed by the
State of Ohio (Ohio EPA 1989) for analysis
of surface water designated use-
attainment. This includes modification of
several of the original Index of Biotic
Integrity metrics as proposed by Karr
(1981).

Although the methodology and application
of the ecoregional expectations are
similar in approach to Ohio and much of
the information below is taken directly
from the Otiio document (Ohio EPA 1989). a
significant difference exists between the
Indiana and Ohio reference conditions.
This difference exists in how the metric
expectations are developed. In Ohio, the
ecoregional reference stations were
combined into a single data set for the
entire State, and later modifications were
developed for a single ecoregion.

In Indiana, "least impacted" conditions
are being developed on a regional basis,
with a priori recognition of basin
differences within ecoregion. based on the
natural division classification of Homoya
et al. (1985). Further evaluation at the
completion of the of the study will
determine if differential metric treatment
is warranted for basin specific or larger
scale criteria development.

The Index of Biotic Integrity is sensitive
to differences in collection effort and
gear type. In order to account for these
inherent biases, separate expectations are
developed for each of the two stream
classification types utilized in the
current study. Headwater stream sites (<
20 miles2) were primarily sampled for 50-
100 m using wading techniques. These sites
were sampled using a long-line
configuration usually off bridges, while
larger wadable rivers (> 20-1000 miles2)
were sampled using the sport-yak
configuration. This technique requires a
sampling distance of 100-300 m and wading
in all available habitats.

Below is an explanation of each of the
twelve metrics utilized for the
calibration of the Indiana Index of Biotic
Integrity for the Northern Indiana Till
Plain. Due to inherent differences at
approximately 20 miles2 drainage area.
different metrics were necessary to
evaluate both headwater (<20 miles2
drainage area) and wadable rivers (>20-
1000 miles2 drainage area) No differences
were observed between the ecoregions and
subbasins for most metrics. This was
anticipated due to the limitations of the
gear type chosen and that large rivers
tend to be integrators of the upstream
drainage area. Maximum species richness
lines were drawn following the procedure
of Fausch et al. (1984) and Ohio EPA
(1987). Scatter plot data diagrams of
individual metrics were first evaluated
for basin specific patterns. The
trisection method was used to depict the
maximum species richness lines. This
requires the uppermost line to be drawn so
that 95£ of the data area lies beneath.
The other two lines are then drawn so the
remainder of the area beneath the 95th
percent!le li-ne is divided into three
equivalent areas In situations where no
significant deviation in relationship was
observed within the three basin segments.
the segments were pooled to reflect an
ecoregional consensus. Likewise, if no
relationship with increasing drainage area
was observed, the maximum species richness
lines either leveled off at the point
where no additional increases were
exhibited or horizontal plots were
delineated indicating no increase with
drainage area.

Differentiation between headwater and
wadable stream and river sites are
indicated on the graphs by a vertical
dashed line on the appropriate metrics
k
17

-------
 Indiana Ecoregion
This relationship was determined by
searching for tnmodal patterns  in the
basin specific data set plots.  The tails
of distribution of the data are not
significant. However the point where the
data differentiates into two distinct
peaks suggest that the transition between
headwater and wadable streams is at 20
miles2  (% headwater  taxa)  and  between
wadable and large rivers at 1.000 miles2 (
large river individuals).   Finally,  a
comparison was made between criteria
established for the ecoregion between
Michigan and Indiana
                                              18

-------
Northern Indiana Till Plain
Metric 1. Total Number of Fish Species (Headwater and Wadable Sites)
Impetus

This metric-is utilized for all of the
stream classification types used for
calibrating the Indiana Index of Biotic
Integrity. Unlike the Ohio metric, exotic
species are included in the total number
of taxa. The premise behind this metric is
based on the observation that the number
of fish species increases directly with
environmental complexity and quality of
the aquatic resource (Karr 1981: Karr et
al. 1986). Although the number of exotic
or introduced species may be indicative of
a loss of integrity (Karr et al. 1986:
Ohio EPA 1989). the differences between
lower levels of biotic integrity
resolution may be due to colonization of
habitats by pioneer or tolerant taxa which
tend to incorporate exotic species.

This single metric is considered to be one
of the most powerful metrics in resolving
water resource issues since a direct
correlation exists between high quality
resources and the numbers of species for
warmwater assemblages (Ohio EPA 1987:
Davis and Lubin 1989: Plafkin et al 1989:
Simon, 1991). As total number of species
increases, species become more
specialized and have narrower niche
breadths, numerous higher level
interactions occur and presumably enable
greater efficiency in resource
utilization. The delimitation between
headwater and wadable Indiana streams in
the Central Corn Belt Plain ecoregion was
made primarily on the data from this
metric. Headwater and wadable streams are
differentiated at 20 miles2 drainage area.
Headwater and Wading Sites

The number of species is strongly
correlated with drainage area at
headwater, wadable stream, and river sites
up to ca. 1,000 miles2. Determining the
Index of Biotic Integrity scoring criteria
for this metric did not require the
recognition of watersheds. Comparison of
maximum species richness lines for the
appropriate basin and drainage area did
not reveal any significant differences
between ecoregion or subwatershed (Fig. 4,
headwater and wading sites).
19

-------
                                      Northern Indiana Till Plain
O   Elkhart
                                                 St.       +   Lake     +   Pigeon
                                                 Joseph       Michigan
rv>
CD


ID
O
S
$
1
5


40

30

20

10
	 1 	 1 — | 1 1 1 1 l| 	 1 	 1 	 1 — 1 I I ll| 	 1 r r i ••ill i i i i i i i i | — r i — i—
O 4-*'!
"
,-* " .
i. ^C& 5 A A •
*-' A . .-'*'
-'''' 0 +°*.--''°
: A>-'A' ftr'0*" 3 *'•
• ** ^ * * «• ^^** 2v A
* C * - * * n A
A* *» •" "* "
0 ' 	 "
O.1 1 10 1OO 1OOO
                                            DRAINAGE AREA  (SQ. Ml)
                   Figure 4. Maximum specie: richness lines for determining trends in total
                           number of species with increasing drainage area for the St. Joseph
                           River drainage

-------
Northern Indiana Till Plain
Metric 2. Number of Darter/Madtom/Sculpin Species (Headwater < 20 miles2)
Number of Darter Species (Wadable Rivers > 20- <1,000 miles2)
Impetus

Karr et al. (1986) indicated that the
presence of members of the tribe
Etheostomatim are indicative of a quality
resource Darters require high dissolved
oxygen concentrations, are intolerant of
toxicants and siltation, and thrive over
clean substrates.

Life history information for all of the 28
Indiana species indicates darters are
insectivorous, habitat specialists, and
sensitive to physical and chemical
environmental disturbances (Page 1983;
Kuehne and Barbour 1983). Darters are
excellent indicators of a quality
resource, and are generally found in
riffle habitats.
Ictaluridae (madtoms: genus NoturusKTable
4). The sculpins and madtoms are benthic
insect!vores and functionally occupy the
same type of niche as darters. Their
inclusion enables a greater degree of
sensitivity in evaluating streams that
naturally have significantly fewer darter
species. By adding madtoms and sculpins
this metric does not asymptote with
increased drainage area for headwater
sites (<20 mi1es2)(Fig. 5). The number of
darter, madtom and sculpin species was
found to increase with increasing drainage
area for each of the four watersheds. No '
differences in ecoregion or watershed
expectations were observed between sites
of the Northern Indiana Till Plain.
Headwater and Wadable Streams and Rivers

The darters include the genera:
Ammocrvpta. Crystal 1 aria. Etheostoma. and
Percina. Of the 28 species recorded from
Indiana, six are commonly found throughout
the State and are not restricted to a
particular stream size (Gerking 1945).
Thirteen of these 28 species are confined
to the Ohio River basin; none of the
species are restricted to the Mississippi
River-basin; and a single species occurs
only in the Great Lakes drainage.

For sites having drainage areas less than
20 miles2, this metric also includes
members of the family Cottidae and
Wadable River Sites

Madtoms and sculpins are more difficult to
collect with increasing drainage area.
since madtoms are typically nocturnal in
their habits. The expected number of
sculpin and madtom species declines in the
St. Joseph River drainage with increasing
drainage area. Thus, only the number of
darter species are included in cumulative
scoring for drainage areas greater than 20
miles2 due to sampling bias and the patchy
distribution of sculpins and madtoms in
wadeable rivers (>20-1000 mi2). This
conforms with the original IBI and is
consistent with Karr's original intentions
(Karr 1981).
21

-------
 Indiana Ecoreqion
 Table  4. The distributional  characteristics of Indiana darter
          (Etheostomatini), madtom (Noturus).  and soil pin  (Cottus)  species
Species
                                       Distribution  In  Indiana Drainages
Statewide
 Ohio
River
Great
Lakes
Mississippi
River
Ammocrvpta  pellucida
A. clara
Crvstallaria asprella
Etheostoma  asprigene
£. blenmoides
£. caeruleum
£  camurum
£. cMorosoma
£. exile
£. flabellare


£. maculatum
£. mlcroperca1
£. mqrum
£. spectablle
£. squamiceps
£. tippecanoe
£. vanatum
£. zonale
Perclna  caprodes
P. copelandi
£. evides
£. maculata
P. phoxocephala

P. shumardi
P. vigil
Noturus  eleutheru^
N  flavus

N. exilis
M. nnurus
M. nocturnus
M. stiqmosus
Cottus bairdi
L. carolinae
C^ cognatus
       X
       X
       X

       X

       X
1 Restricted to northern  portions of these drainages.
                                           22

-------
               Northern Indiana Till Plain
o   Elkhart
                          St.        +   Lake      +   Pigeon
                          Joseph        Michigan
vRTER SPECIES
0
s °
fe
cc
>
5
o
0
	 1 — i i i 1 1 n | 	 1 — i i i 1 1 1 1 1 i i i i • • 1 1( r i i i 1 1 •• i i ii

A jftrt* + &4Q 4^ Q & '
*'**' 3
* ^ * *


.1 1 1O 1OO 1OOO
                      DRAINAGE AREA  (SQ. Ml)
Figure 5. Maximum species richness  lines  for determining trends in number
        of darter/madtom/sculpin  species with increasing drainage area for
        the St. Joseph River drainage.

-------
Indiana Ecoregion
Metric  3. Percent Individuals  as Headwater Species  (Headwater  <20  miles2)
            Number  of Sunfish Species  (Wadable  Sites >20 -  <1,000 miles2)
Impetus

This metric  followed Karr (1981)  and  Karr
et al.  (1986) by including the number of
sunfish species (family Centrarchidae) and
excluding  the black basses (Micropterus
spp). Unlike the Ohio metric,  the redear
sunfish Lepomis microlophus  is included
because it is native to Indiana (Table 6).
Hybrid  sunfish are not included in this
metric  following Ohio EPA (1989).

This metric  is an important  measure of
pool habitat quality.  It includes all
members of the sunfish genera  Ambloplites
(rock bass), Centrarchus (round sunfish),
Lepomis (sunfish), and Pomoxis (crappies),
as well as,  the ecological equivalent
Elassomatidae (Elassoma zonatum).  Sunfish
normally occupy slower moving water which
may act as "sinks" for the accumulation of
toxins and siltation. This metric measures
degradation of rock substrates (i.e.
gravel  and boulder) and instream cover
(Pflieger 1975: Trautman 1981), and the
associated aquatic macroinvertebrate
community which are an important food
resource for sunfish  (Forbes and
Richardson 1920:  Becker 1983). Sunfish are
important components  of the aquatic
community since  they  are wide  ranging, and
distributed in most  streams and rivers of
 Indiana. They are also  very susceptible to
sampling  using electrofishing  gear. Karr
et  al.  (1986) found  sunfish to occupy the
 intermediate to upper ends  of sensitivity
 of  the index of biotic integrity  (IBI).

 Headwater Streams

 Pool habitat is a limiting  factor in many
 headwater streams  This prohibits sunfish
 colonization by because of their deep-
 bodied morphology   I replaced the number
 of sunfish species with the proportion
Table  5. List of Indiana  fish
           species considered  to be
           headwater  species for
           evaluating  permanent
           habitat in  streams
           (Smith,  1971).

Common Name   Scientific  Name

Least brook lamprey   Lampetra aepvptera

American brook  lamprey L- appendix

Redside dace          Clinostomus elongatus

                     Rhimchthys atratulus

So Redbelly dace     Phoxinus ervthroaaster

Brook stickleback     Qulaea  inconstans

Fantail darter        Etheostoma  flabellare

Mottled sculpin      Cottus  bairdi

Banded sculpin        £.  cognatus
Blacknose  dace
                                             24

-------
Northern Indiana Till Plain
of headwater species at sites with
drainage areas less than 20 miles2 (Ohio
EPA. 1987). Nine headwater species were
defined by Ohio EPA (1987) and their
presence indicates permanent habitat with
low environmental stress (Table 5). The
presence of headwater species does not
show a trend with increased drainage area
(Fig. 6).

Wadatrte Streams and Rivers

Sunfish colonization is limited by the
amount of pool habitat in many river
reaches. This metric did not show any
difference in scoring based on sub-basin.
The number of sunfish species increased
with increasing drainage, area (Fig. 7).
Table 6. List of Indiana sunfish
species for evaluating
quality pool habitat.

Common Name Scientific Name
Rock bass

Flier

Green sunfish

Pumpkinseed

Warmouth

Orangespotted sunfish L- h
Ambloplites rupestns

Centrarchus
macrooterus
Bluegill

Longear sunfish

Redear sunfish

Spotted sunfish

Bantam sunfish

White crappie

Black crappie
L.macrochirus

L. megalotis

L.microlophus

L. punctatus

L. symmetncus

Pomoxis annularis

L. mgromaculatus
Banded pygmy sunfish Elassoma zonatum
25

-------
       Northern Indiana Till Plain
O   Elkhart
St.       *   Lake     +
Joseph        Michigan
                                               Pigeon
% HEADWATER SPECIES
100
75
50
25
0
0
	 1 	 1 1 1 1 1 ll| 	 1 	 1 	 1 1 O ll| 	 I -T" T"l 1 1 ll| 1 1 1 lllll| 1 1 -1
A A
: A + :
A A A
: A * 5 '•
, A A
• ---- ....--- - -j, 	 M-.ruTjy. -
: , - A* :
AAA o 3 . -
A AA + 0
!••••••••••• ••••••••• •^••a--*-- -"••"•**•• *• ---r- ----•••••••
A A AA A
S A A 0 A ^
. A A^ A
* A*. ».,,,*. ^ -A,. rfno-<3fcftfloi • -no--' •*,.,"
1 1 1O 1OO 1OOO
               DRAINAGE AREA  (SO. Ml)
Figure 6. Maximum species richness .lines for determining trends in the
        proportion of headwater species with increasing drainage area  for
        the St. Joseph River drainage.

-------
                                                                   Northern  Indiana Till Plain
                                                                                    o.
                                                                                    o>
                                                                                  %l \ 2\« :
% 'o j
i\ \<3 ', i
^M!
^ * * 1 '
^ ^*
X
• • • — i —

0
o
o



o
o


o
*~
r-
                                                          i
                                                          <
                                                          S
                                                          <
                                                                                 c oo


                                                                                 C O)

                                                                                 "~ 4->
                                                                                 I/)


                                                                                 c o
                                                                                    Si
                                                                                 •i-  CJ
                                                                                 oi  a>    ra
                                                                   e  ^3 <-
                                                                   r-  tO Q)
                                                                   X    >
                                                                   IT3 '•«— ••—
                                                                  z:  o ce.
                                                                  i—

                                                                   OJ

                                                                   3
                    O
                                                                O
                             HSIdNDS  dO
                                       27

-------
 Indiana Ecoregion
Metric 4.  Number of Minnow  Species  (Headwater Sites < 20 miles2)
             Number of Sucker  Species  (Wadable Sites  (>20-< 1000 miles2)
Impetus

The original  Index of Biotic  Integrity
metrics included the number of sucker
species (Karr 1981.  Karr  et al   1986).
Suckers represent a  major component of the
Indiana fish  fauna since  their total
biomass usually  ranks  them among the
highest biomass  contributors  in the
community.  Most  sucker species are
intolerant  to habitat  and water quality
degradation (Phillips  and Underhill 1971;
Karr et a/.  1986:  Trautman 1981; Becker
1983) and this results in metric
sensitivity at the higher end of
environmental quality.  Suckers, due to
their long  life  span (10-20 years),
provide a long-term  assessment of past
environmental conditions. Of  the 19
species historically found in Indiana.
Lagochila lacera is  considered extinct.
seven species are widely  distributed
throughout the State (Table 7).  Extant
sucker genera include; Cvcleptus.
           Catostomus . Erimyzon.
Hvpentelium. Ictiobus.   Minytrema. and
Hoxostoma.
Headwater Sites

The number of minnow species is
substituted for the number of sucker
species in headwater sites (Fig.  8).   The
number of sucker species decreases  rapidly
with declining drainage area (Fig.  9).
While few different sucker species  have
been observed at locations with drainage
areas less than 20 miles2.  The number of
minnow species generally correlates with
 increased environmental quality. This
metric includes
members able to represent a wide variety
of biological integrity.  Species such as
the hornyhead chub (Nocomis biguttatus)
and rosyface shiner (Notropis rubellus)
are examples of minnow species which can
occur in high quality headwater streams
Minnow species represent  both ends  of the
biological  integrity continuum.  A direct
relationship exists between increasing
number of minnow species  and increasing
drainage area (Fig.  8).
Wadable River Sites

The number of sucker species,  with  the
exception of Catostomus  commersom.
Ictiobus and Carpi odes,  represent
sensitive species intolerant to thermal,
siltation. and toxins stresses.  The
redhorses are particularly important
indicator organisms m rivers.  The  most
sensitive suckers include members of  the
genera Cvcleptus. Hvpentelium.  Moxostoma.
Minytrema. and Erimyzon.  These species  are
effectively sampled with electrofishing
gear and comprise a significant component
of riverine fish faunas.  Their feeding  and
reproductive requirements are sensitive to
turbidity and marginal to poor water
quality  The number of species were not
significantly different among the four
watersheds  (Fig. 9)
                                              28

-------
                                                                   northern Indiana Till Plain
Table 7.  Distribution
characteristics of Indiana  sucker
species  (family Catostomidae).

                           Large  Rare
Species         Statewide  Rivers  Taxa

Cvcleptus elonoatus         X      X

Carpi odes carpio        X   X

£. cyprinus             X

£. velifer                  X      X

Catostomus catostomus              X

Catostomus commersoni  X

Erimvzon  oblongus      X


Hvpentelium  nlgricans       X      X

Ictiobus  bubal us       X     X

1. cyprinellus             X      X

i. niaer                          X

Laaochila lacera      EXTINCT

Mlnytrema melanops     X

Moxostoma anisurum     X     X

M. carinatum               X       X

M. duquesnei          X     X

M. erythrurum         X     X

M macrolepidotum     X     X

M. valenciennesi            X       X
                                            29

-------
                                                                              o
                                                                              "1
                                                                              It
             Northern Indiana Till Plain
     o   Elkhart   A   St.        +   Lake     +   Pigeon

                      Joseph        Michigan

tu
u
LU
a
co
\
i

u.
O
oc
rp
>
]
2


2O




15

TO


5




O
O
i i i i i i 1 1 1 I i i i i I 1 1| i i i i i i ii| i i ' i i i i i| i i r—
.
-
O

^'•'
.'-'' 5 •
A ,'** „„--'"" -
A .-'' O -f * .---'
A A4»* A O A ,-*-*" A
A A'J^ OJ>'''C& 0 3 -
£± <^xTA ^V^- *tf2k A Q OA »«•»-••""**""'
A -*tf\ - *^r *A A A ^S^^A * * * ^^" O O .
jfir ***** ^*A_ * ^Sfr- A* AfT */uC^ O O A A ' A •


*i *j "**
.1 1 1O 1OO 1OOO
                  DRAINAGE AREA (SQ. Ml)
Figure 8. Maximum species richness lines for determining trends in number of

        minnow species with increasing drainage area for the St  Joseph
        drainage.                                            K

-------
Northern Indiana Till Plain
0 ElKhart A St. * Lake * Pigeon
Joseph Michigan
NUMBER OF SUCKER SPECIES
^0 01 0
— , r l i i ni| • r • ' ' ' " 1
A,-
,.-•:''' 5 !..-'
.,--"" ...i9""6'"
A ,'*' AA,.0-b
-------
Indiana Ecoregion
Metric 5. Number of Sensitive Species (Headwater and Wadable Sites)
Impetus

The number of sensitive species metric
distinguishes between streams of highest
quality. Designation of too many species
as intolerant will prevent this metric
from discriminating among the highest
quality resources. However. Karr (1981)
and Karr et al. (1986) calibrated
expectations based on watershed scales and
not regional or state scales. Only species
that are highly intolerant to a variety of
disturbances were included in this metric
so it will respond to diverse types of
perturbations (Table 8; see Appendix A for-
species-specific information).

The number of intolerant taxa is a
modification of the original index as
developed by Ohio EPA (1987). The number
of intolerant taxa, proposed by Karr
(1981), is not synonymous with the
sensitive species metric. The metric
includes moderately intolerant species
when sampling at headwater sites. This
combination is called sensitive species
since few intolerant taxa are expected.
The moderately intolerant species meet
most of the established criteria of Ohio
EPA (1987). An absence of these species
would indicate a severe anthropogenic
stress or loss of habitat.

The criteria for determining intolerance
is based on the numerical and graphical
analysis of Ohio's regional data base.
Gerking's (1945) documentation of
historical changes in the distribution of
Indiana species, and supplemental
information from regional ichthyofaunal
texts (Pflieger 1975: Smith 1979: Trautman
1981, Becker 1983; Burr and Warren 1986).
Intolerant taxa are those which decline
with decreasing environmental quality and
disappear, as viable populations, when the
aquatic environment degrades to the "fair"
category (Karr et aJ. 1986). The
intolerant species list was divided into
three categories, all are included in this
metric for scoring:

1). common intolerant species (I)
species that are intolerant, but are
widely distributed in the best
streams in Indiana:

2). uncommon or geographically restricted
species (S): species that are
infrequently captured or that have
restricted ranges;

3). rare or possibly extirpated species
IR1:intolerant species that are
rarely captured or that lack recent
status data.

Commonly occurring intolerant species made
up 5-10? of the common species in Indiana.
however represent 35-40? of the entire
statewide list. This was a recoraiended
guideline of KaT (1981) and Karr et al.
(1986). Although the addition of species
designated as uncommon or rare sensitive
species (categories 2 and 3). inflates the
number of intolerant species above the 10?
guideline, nowhere in the State do all of
the species coexist at the same time In
order to evaluate streams in the headwater
and wadable site categories, only the
sensitive species metric will be used
until further resolution is possible with
the addition of adjacent eccregion
sampling. Until more sampling is completed
or improvements in water quality warrant
it. the sensitive species metric (Ohio EPA
1987) will be used for all headwater
streams and wadable river sites in
Indiana.
32
-------
Northern Indiana Till Plain
Headwater and Wadable Streams and Rivers

The number of intolerant species increases
with drainage area among headwater and
wading sites (Fig. 10). Intolerant taxa
are scarce in headwaters of the Northern
Indiana Till Plain and increase at larger
wading sites. In order to provide
meaningful stream reach comparisons in
Indiana, the sensitive species metric is
currently retained until further
evaluation can be completed.
33
-------
Indiana Ecoregion
Table 8. List of Indiana fish species considered tc be sensitive to a
wide variety of environmental disturbances including water quality and
habitat degradation.
Common Name
Scientific Name
Common Name
Scientific Name
Ohio lamprey Ichthvomyzon bdellium
Northern brk lamprey 1. fossor
Least brook lamprey Lampetra aepvptera
American brk lamprey L appendix
Paddlefish
Goldeye
Mooneye

Redside dace
Streamline chub
Gravel chub
Speckled chub
Bigeye chub
Pallid shiner
Rosefin shiner
Hornyhead chub
River chub
Pugnose shiner
Popeye shiner
Bigeye shiner
Ironcolor shiner
Blackchin shiner
Blacknose shiner
Sand shiner
Silver shiner
Rosyface shiner
Weed shiner
Mimic shiner
Pugnose minnow
Longnose dace
Blue sucker
Highfin carpsucker
Northern hogsucker
Silver redhorse
River redhorse
Black redhorse
Golden redhorse
Shorthead redhorse
Greater redhorse
Polyodon spathula
Hiodon alosoides
H. terqisus

Clinostomus elongatus
Erimvstax dissimilis
£. x-punctata
Extrarlus aestivalis
Hvbopsis amblops
H. amms
Lvthrurus ardens
Nocomi s biguttatus
N. micropogon
Notropis anogenus
M- ariommus
N. boops
N- chalvbaeus
N. heterodon
M- heterolepis
N. ludibundis
N. photoqenis
N. rubellus

N. volucellus
Opsopoeodus emiliae
Rhinichthvs cataractae
Cvcleptus elongatus
Carpiodes velifer
Hvpentelium nigricans
Moxostoma anisurum
M. carinatum
M. duquesnei
M ervthurum
M. macro!epidotum
M. valenciennesi
Mountain madtom
Slender madtom
Stonecat
Brindled madtom
Freckled madtom
Northern madtom
Southern cavefish
Southern cavefish
Northern studfish
Starhead topminnow