EPA/600/3-87/025
June 1987
THE OHIO
STREAM REGIONALIZATION PROJECT:
A COMPENDIUM OF RESULTS
by
Thomas R. Whittier1
David P. Larsen2
Robert M. Hughes1
Christina M. Rohm1
Alisa L. Gallant1
James M. Omernik2
Northrop Services, Inc.
200 SW 35th St.
Corvallis, OR 97333
2Environmental Research Laboratory -- Corvallis
U. S. Environmental Protection Agency
200 SW 35th St.
Corvallis, OR 97333
-------
DISCLAIMER
The Information 1n this document has been funded in
part by the United States Environmental Protection
Agency under USEPA 205(j) funds awarded to the Ohio
Environmental Protection Agency and USEPA contract no.
68-03-3124 to Northrop Services, Inc. It has been
subjected to the Agency's peer and administrative
review, and it has been approved for publication as an
EPA document. Mention of trade names or commercial
products does not constitute endorsement or
recommendation for use.
1i
-------
ABSTRACT
Regional patterns in terrestrial characteristics can be used
as a framework to monitor, assess and report the health of
aquatic ecosystems. In Ohio, five ecological regions were
delineated using combinations of spatial patterns in land-surface
form, land use, soil and potential natural vegetation. We
evaluated this framework by studying the water quality, physical
habitat, and fish and macroinvertebrate assemblages of 109
minimally impacted representative streams. Water quality and
fish assemblages showed clear regional differences. The highest
quality water and fish assemblages were consistently found in the
southeast ecoregion and the lowest quality in the northwest
ecoregion. We found no clear regional patterns in macro-
invertebrate assemblages and limited regional patterns in
physical habitat.
iii
-------
CONTENTS
Abstract 111
Figures v
Tables vi
Acknowledgments vii
1. Introduction 1
2. Ecoreglons of Ohio 3
3. Selection of Candidate Watersheds and
Sampling Sites 11
4. Field Sampling and Analyses 14
5. Data Analyses and Results 17
F1sh 18
Macrolnvertebrates 29
Water Chemistry 36
Physical Habitat 47
6. Interpretation 52
Aquatic Life Use Designations and Attainment. . . 52
Attainable Water Quality 57
Literature Cited 59
Appendix A 63
1v
-------
FIGURES
Number Page
1. The five ecoregions of Ohio 5
2. Dominant fish species, fraction of samples 21
3. Dominant fish species, relative abundances 22
4. Fish species richness 24
5. Index of Biotic Integrity 25
6. Index of Well-Being 27
7. Fish species diversity 28
8. Intolerant fish species 30
9. Macroinvertebrate richness and diversity 31
10. Aquatic insect taxa signature, dominant taxa 33
11. Aquatic insect indicator taxa, selected by TWINSPAN 35
12. Ionic strength measures of water quality 37
13. Nutrient measures of water quality 39
14. Metal concentrations 40
15. Temporal patterns of selected nutrients 42
16. Spatial patterns of total phosphorus in Ohio streams 44
17. Spatial patterns of conductivity in Ohio streams 45
18. Regional patterns in nutrient richness and ionic strength ... 46
19. Profiles of instream cover composition 50
20. Profiles of stream substrate composition 51
21. Comparison of fish species richness in an impacted stream
with regional reference sites 55
-------
TABLES
Number Page
1. Terrestrial characteristics of the five ecoregions
of Ohio 4
2. Distribution of sampling sites 13
3. Analytical methods, precision and accuracy goals, and
detection limits for chemical analyses 15
4. Sites and samples eliminated from analyses of fish data .... 19
5. Dominant species and percent of samples where species
> 10% of sample 20
6. Sites characterized by unusually high metal concentrations. . . 41
7. Ohio EPA biological criteria (fish) for determining
water quality use designations 53
8. Attainable fish assemblage attributes
for small streams in Ohio 56
vi
-------
ACKNOWLEDGMENTS
A project of this size involves the effort of many people.
We gratefully acknowledge their assistance. Jim Bland's insight
and perseverance was instrumental in initiating and funding this
study. The field and laboratory staff of the Ohio EPA provided
the data; Chris Yoder, Dan Dudley and Dave Altfater provided many
useful hours of discussion, planning and review of all phases of
the study. The computer support staff of ERL-Corval 1 is, Peggy
Walch, Joe Spitz, Margaret Spence, and Paula Jensen, performed
the often thankless tasks of data management, analyses and
document preparation. Sandra Henderson prepared the graphics for
many of the figures.
vii
-------
SECTION 1
INTRODUCTION
State environmental protection agencies are faced with the
responsibility of carrying out the mandates of the Clean Water
Act (CWA, PL 95-217) through a variety of regulations (Federal
Register 1983) promulgated by the United States Environmental
Protection Agency (USEPA). The emphasis has been on management
of water quality through establishment of chemical criteria that,
if met, presumably provide for the protection and propagation of
fish, shellfish, and wildlife. While this approach has improved
water quality, state agencies increasingly need to address the
goals of the CWA directly through setting biological objectives
or criteria for water quality management.
In particular, the Ohio Environmental Protection Agency
(OEPA) wished to determine the quality that was reasonably
attainable in streams throughout Ohio. The OEPA staff recognized
differences in streams in various parts of the state and also
wished to obtain information about streams that were not impacted
by point source discharges or substantial nonpoint sources.
Their stream monitoring program concentrated on waterways that
received municipal and industrial wastes; data were collected and
analyzed on a case by case basis.
At about the same time, scientists at the Environmental
Research Laboratory--Corval Us (ERL--Corvallis) were developing a
method to identify and characterize regional patterns in aquatic
ecosystems. The method delineated regions based on the
commonality of a variety of geographic characteristics; these
regions could be recognized by examining maps of land-surface
form, soil, land use, and potential natural vegetation. They
expected these ecoregions to provide a useful framework for
examining attainable quality for various kinds of water bodies.
Streams that occur within any particular region should reflect
the characteristics of the land they drain and therefore be
relatively similar to one another. Streams in different regions
should differ.
This regional framework seemed a useful approach by which
OEPA could obtain representative information to demonstrate what
might be attainable in the streams it sampled. Water quality
decision making could be improved by comparison of regional
expectations to specific sites to evaluate the extent and degree
of use impairment (or lack thereof). As a result, a memorandum
of understanding among Region V (USEPA), ERL-Corval1 is, and Ohio
-------
EPA established a program to identify a regional framework for
grading stream performance.
Briefly, the approach entailed delineating ecoregions in
Ohio; identifying watersheds and streams that were representative
of the ecoregions and were least disturbed by human activities;
sampling these streams for a variety of physical, chemical, and
biological characteristics; and determining the extent to which
patterns in these characteristics reflected ecoregional patterns.
The set of data obtained from the selected streams would form the
basis for quantitatively expressing attainable uses in the
various regions of Ohio. These attainable uses and other
statements herein are limited to small streams having watershed
areas of 10-300 mi2. They may not be appropriate for larger
streams and rivers. Also, because of the presence of row crop
agriculture throughout much of the state, most of the sites were
impacted to varying degrees by diffuse sources of pollution,
primarily agricultural pesticides and fertilizers, sediment, and
channel modifications.
Our purpose here is to present a general overview of this
project and its results. This report will be of interest to a
diverse group of readers: aquatic scientists, resource managers,
and regulatory personnel. This document describes: (1) the five
ecoregions in Ohio and the distinguishing features of each; (2)
the selection of candidate watersheds and study sites repre-
sentative of these ecoregions; (3) the field sampling and
analyses; (4) the results summarized in a way that makes them
useful as a reference data set indicating attainable quality.
The following references should be consulted for more detail
about various phases of this project: ecoregion concept develop-
ment (Bailey 1976, 1983, Hughes and Omernik 1981, Hughes et al.
1986, Larsen et al. 1986, Omernik et al. 1982, Omernik 1987, Rowe
and Sheard 1981, Warren 1979); regional reference sites (Hughes
et al . 1986); Ohio fish assemblages (Larsen et al. 1986); Ohio
water quality (Larsen and Dudley 1987); ecoregions in other
states (Hughes et al. 1987, Omernik and Gallant 1986, Rohm et al .
1987, Whittier et al . MS).
-------
SECTION 2
ECOREGIONS OF OHIO
We studied maps of land-surface form, land use, potential
natural vegetation, soil, surflclal geology, climate, and
hydrology for obvious regional patterns of homogeneity 1n
combinations of these factors. This analysis revealed five
ecoreglons (Table 1) defined by certain characteristics 1n land
use (Anderson 1970), land-surface form (Hammond 1970), potential
natural vegetation (Kuchler 1970), and soils (USDA 1957, USGS
1970). Cursory examination reveals a gradual transition in these
characteristics from northwest to southeast Ohio.
Through a map analysis and overlay process, we defined the
most typical portions of each ecoreglon as those areas where all
four of its characteristic components occurred 1n combination.
One region and some of the most typical areas of the ecoregions
are discontinuous. Lines distinguishing the major regions are
less precise than those defining the most typical areas and were
drawn to Include areas where most, but not all, of the character-
istics typifying a region occurred in combination. Such areas
were considered generally typical of their ecoreglons (Figure 1).
Descriptions of the ecoregions (below) provide a synopsis of
the watershed characteristics that affect aquatic ecosystems.
Each description is an amplification of the combination of
features that give the ecoreglon its Identity (Table 1) and
Includes the regionally Important human impacts, particularly
those that are diffuse or nonpoint in nature. We obtained this
Information from a variety of sources. The most helpful of these
was Land Resource Regions and Major Land Resource Areas of the
United States (Austin 1965, USOA 1981). We estimated watershed
sizes necessary to support perennial stream flow by interpolating
from USGS Water Resources Data and 1:250,000 scale maps.
Several land uses that Impact stream quality occur state-
wide. Much of the natural forest vegetation has been removed in
Ohio. Land clearing and soil compaction amplify the intensity
and frequency of flood and drought flows. General agriculture
near streams is a source of nutrients (fertilizer), toxic
chemicals (pesticides), and sediments from increased erosion.
These affect the chemistry and oxygen carrying capacity of the
water and erosion affects the physical habitat by Increasing
turbidity and sedimentation. Farmers and foresters often remove
the natural riparian vegetation thereby speeding runoff, reducing
shade (increasing water temperature and increasing photosynthetic
-------
u
a
1
g
1-4
I
01
£
§
ej
5
H
IB
0
1
"8
n
1
L
u r*
W 00
Q) ^H
•H M
t«.H
||
"o*"
10
85
•H M
|J
£8
u
tB «H
Jg
F^ {}
•H 0>
JJ U
W 4-»
(U (Q
•W tt)
»4 C
U-H
4JH
U T3
*" 3
o-q
gu
ii
5
£
O 3
•H 10
M 01
01 *J
4J to
CfH
M CL,
Western
Allegheny
Plateau
0
•H C
kl-H
10 IB
4-> fH
c a,
° 0)
ox
•H 18
U, U
u
B
M 10
O C
U-H
CfH
M Cu
01
4J 4J
WfH
01 C
-H -H
K 10
WfH
*» o,
c
O 01
3*
X J
01
in 0) IB
fHfH U
fH £ 0)
•H 18 -3
£ ^* O
£ -
4J 10 £
•HfH 4J
^H~3
I0£ IM
C 10 01
•H C -O -H
18 01 CfH
fH a to oi
a o •-< M
£
•H
£
o
U 10
M
IB
iH
&c
0)-H
M IB
•HO,
10
B
•H
IB
a
£
4J
10
10
C
•H
to
fH
a
^J
i
• 3 *•» 01
10 O UfH
rH M -H O
SOfH
1 O U
•H >,N-H
IM WO E
fH M 6 3
« 00,1
- 1 U
>• >. 1 -H
0) O IB -O E
o | o a i
u x -co
•H S 3 -H >|
E 3 01 B'H 01
3 OfH M OJ u a NO
r- ~
in o
fH
n <
•H Q 0)
•H t/1 O
O
o
•H 4J
£ 10
-v 01
go
U
•H - - if
4J ȣ J< IB
>i 01 U IB O
£rH 01 O
a a y c
O IB £ - IB
10 E 01 -H
fj _ . 4) .£
i t) u— o
w >,u c n
•a 10 oi a. 01 fH
01 01 Ai-HTJ IB
x M urn c a
•H O 3 3-H O,
E IM a 4JfH <
1
•o - »
- M 01£
.3328
&BSJS-
E M— X
"^01 * U
£ £ (0 £ O
U 4J T3 U fH
01 M O M E
tj) O O -H 0)
2 C 3 J3£
01
rH
a
•L
£ 10
U 01
01 V4
JSfi
£
III 4J
IB in
*•• 0)
5S
CJ **-»
o
r*
^ §s
IB -rt M
•H fH 4J Of
4J 19 IB fH
C M 4J £
t2Sf>-
-------
activity), decreasing cover and increasing erosion. These
effects are further increased by ditching, field tile drainage,
and channelizing and clearing stream beds. Livestock along or in
streams increase nutrient loads (manure), Increase turbidity and
sedimentation by eroding banks, and remove riparian vegetation by
browsing and trampling. Oil field operations may affect water
quality by spills of brine and petroleum, by seepage from sweet-
ening plant ponds and by sludge from drilling sites. Quarries
and gravel pits may increase turbidity and sedimentation in
nearby streams. Mining may severely degrade streams both
chemically (acidification) and physically.
Figure 1. The five ecoreglons of Ohio. Darker tones denote most
typical areas. I « Huron/Erie Lake Plain (HELP), II « Eastern
Corn Belt Plains (ECBP), III - Erie/Ontario Lake Plain (EOLP),
IV - Interior Plateau (IP), V « Western Allegheny Plateau (WAP).
Dots Indicate location of sampling sites.
-------
HURON/ERIE LAKE PLAIN
The Huron/Erie Lake Plain ecoregion 1s distinguished from
surrounding ecoreglons primarily by features related to poor soil
drainage. The ecoregion Is a nearly level, broad lake plain,
somewhat Interrupted by beach ridges and low moraines. Much of
the area was covered by forested wetlands 1n presettlement times.
The elevation is generally around 600 feet, rising to 800 feet on
some moraines. Local relief is generally only a few feet.
Typically in this ecoregion, mapped small streams and
drainage ditches are intermittent, while medium to larger streams
are perennial. Larger streams contained entirely within the
ecoregion drain as much as 400 to 500 mi2. The majority of
streams however, drain watersheds of less than 100 mi2 (larger
streams generally originate in adjacent ecoregions). The few
lakes and reservoirs are small, having surface areas of < 0.25
ml2. Average annual precipitation ranges from 31 to 35 inches,
fairly evenly distributed throughout the year and generally
adequate for crop production.
The extensive, nearly level plains and numerous depressions
in morainal areas have contributed to the formation of poorly
drained soils, primarily Aqualfs or Aquepts. Ochraqualfs and
Haplaquepts formed in lacustrine and glacial drift. Udipsamments
and Hapludalfs occur on beach ridges and other well drained
sites.
These poorly drained soils support swampy elm/ash forest
vegetation. Major forest constituents are black ash (replaced by
white ash where drainage is slightly better), American elm and
red maple. Forest species also include silver maple, sycamore,
pin oak, swamp white oak, black tupelo and eastern cottonwood.
Only small remnants of these forests exist in this ecoregion.
Agriculture forms the economic base of this area. Corn,
winter wheat, soybeans, and hay are the principal crops, along
with sugar beets, field and seed beans, and a variety of canning
crops. Fruit and truck crops are grown on more coarse-textured
soils. Some farmland is in pasture and small woodlots.
Principal livestock include swine, poultry, and (near large
cities) dairy cattle.
Stream water quality in this ecoregion 1s affected primarily
by crop and livestock production practices. Channelization of
streams, construction of ditches, drainage of natural woodland
swamps and extensive removal of forests result 1n reduced
quantity and quality of habitat for stream biota throughout the
area.
-------
EASTERN CORN BELT PLAINS
The gently sloping glacial till plain comprising the Eastern
Corn Belt Plains ecoreglon 1s broken by hilly moraines, kames,
and outwash terraces. Elevations range from about 600 feet near
Lake Erie to greater than 1,100 feet. Local relief 1s generally
less than 50 feet except in some of the moraines in west central
Ohio, where relief reaches 200 feet. Valleys are typically
narrow and shallow.
Streams in the larger watersheds contained completely within
the ecoregion drain on the order of thousands of square miles,
e.g. the Great Miami River drains 3,500 mi2. Most mid-sized
streams are perennial; many smaller streams have been channel-
ized. Some larger streams are regulated to inhibit flooding.
There are relatively few reservoirs or natural lakes. Average
annual precipitation ranges from 34 to 38 inches, mostly
occurring during the growing season, and is generally adequate
for crop production.
This ecoregion supports hardwood vegetation characterized by
a predominance of American beech and sugar maple. White, black,
and northern red oaks, yellow-poplar, hickory, white ash and
black walnut accompany these in forested areas. On wetter sites
hardwood forests Include white, pin, and northern red oaks,
sweetgum and yellow-poplar as major constituents; shingle and
black oaks, and hickory may occur. Silver maple, cottonwood,
sycamore, pin oak, and elm occur along streams. Most of the
forests have been cleared.
Soils of this ecoregion were formed predominantly under
deciduous forests. They derived from calcareous glacial loam
till overlain by loess deposits in some southern portions. Many
of the soils are affected by poor Internal drainage. Hapludalfs
and Ochraqualfs formed on broad flat uplands and are the dominant
soil groups. Argiaquolls, Haplaquolls, and Medisaprists formed
in flats and depressions.
The ecoregion is almost entirely farmland. Approximately
75% of the area is In cropland; the remainder is permanent
pasture, small woodlots or urbanized. Corn and soybeans are the
principal crops. Feed grains and hay for livestock are also
grown. Truck and canning crops, and tobacco are grown locally.
Swine, beef and dairy cattle, and poultry are raised throughout
the area. There are numerous quarries and gravel pits.
Stream quality in the Eastern Corn Belt Plains 1s subject to
the influences of a highly agricultural economy. Quarries and
instream mining of gravel may impose local effects on stream
habitats.
-------
ERIE/ONTARIO LAKE PLAIN
The gently to strongly rolling dissected glacial plateau of
the Erie/Ontario Lake Plain ecoregion is underlain mostly by
sandstone and siltstone. Local relief varies from a few feet, in
gently rolling terrain, to as much as 200 or 300 feet in steeper
stream valleys. Local relief is greater here than in the two
northwestern and western Ohio ecoregions, but less than the
Western Allegheny Plateau. The elevation ranges from about 600
feet near Lake Erie to 1,200 feet on the uplands.
The majority of mapped streams in the ecoregion are
perennial and not deeply dissected. The larger watersheds
contained entirely within the ecoregion drain 400-600 mi2. There
are many lakes and reservoirs and some well developed wetlands.
Average annual precipitation is 35-40 inches, generally adequate
for crop production.
This ecoregion supports northern hardwood forest vegetation.
The predominant beech/maple/yellow birch forest in many parts is
similar to those in the Eastern Corn Belt Plains, while on wetter
soils, the forest composition more closely resembles the swamp
forests of the Huron/Erie Lake Plain. In mesophytic areas
associated trees are basswood, American elm, red maple, hemlock,
white ash, black cherry, white pine, northern red oak, balsam
fir, and white spruce. On moist sites American elm, black ash
and red maple dominate with constituents of silver maple, pin,
and swamp white oak, sycamore, tupelo and cottonwood.
Soils are mostly Udalfs and Aqualfs. Aquepts are
predominant along the Lake Erie shore. Soils are derived mainly
from glacial till and lacustrine sediments. Fragiudalfs,
Ochraqualfs, Fragiaqual f s, and Fragiudults formed on uplands.
Haplaquepts, Fragiaquepts, and Hapludalfs formed in lacustrine
sediments. Ochraqualfs formed in lower glacial till plains.
This ecoregion exhibits a mosaic of cropland, pasture,
livestock and poultry production, woodland and forest. Approx-
imately one fourth of the ecoregion provides pasture for cattle.
Cropland covers about one third and is Interspersed with pasture,
woodland, and forest. Cropland emphasis 1s on feed grains and
forage, principally hay and corn, for dairy cattle, which are
important near large cities. Cash crops include wheat, potatoes,
canning and truck crops, e.g. sweet corn, beans, cabbage, peas,
and onions, over most of the area, and vineyards, orchards, small
fruits along Lake Erie. Forests cover about one third of the
ecoregion, used primarily as woodlots, also for saw logs and
pulpwood. About 20% of the area 1s urbanized. There Is some oil
and gas drilling 1n the southern arm of the ecoregion.
Stream quality In this ecoregion is primarily affected by
agricultural practices. In urbanized areas streams are often
8
-------
channelized, regulated and used for industrial purposes.
Numerous gravel pits and quarries, and oil and gas drilling may
have local effects on nearby streams.
INTERIOR PLATEAU
Characteristics of the Interior Plateau ecoregion are
transitional between the adjacent Eastern Corn Belt Plains and
Western Allegheny Plateau ecoregions. This ecoregion includes
undulating to steep terrain formed from Illinoian glacial drift
materials. Elevations vary from about 500 feet near the Ohio
River, to more than 1,200 feet in hilly areas. Average annual
precipitation is about 42 inches, ordinarily adequate for crop
production, but less than half occurs during the growing season.
The majority of mapped streams in this ecoregion are
perennial. Streams in the largest watersheds contained
completely within the ecoregion drain from 200 to 500 mi2. The
ecoregion contains few lakes, most of which are constructed.
As with soils of the Western Allegheny Plateau, many of the
soils in the Interior Plateau formed in residuum of from
sedimentary rocks overlain by varying amounts of loess. Some
soils are derived from calcareous loam till materials with
localized mantlings of loess. Ochraqualfs formed on broad flat
uplands. Hapludalfs, 61 ossaqual f s, and Fragiaqualfs occur on
nearly level to gently sloping uplands. Hapludalfs occur on
steeper upland slope, Fragiudalfs on side slopes. Udifluvents
and Fluvaquents formed in silty alluvium on relatively narrow
floodplains. Hapludolls formed in shaly limestone materials.
Movement of moisture through a number of soils in this area is
impeded by claypans or fragipans. Limestone is present to within
5 feet of the surface over the eastern portion of the region.
This ecoregion supports mainly hardwood forest vegetation.
The ecoregion can be distinguished overall as having oak-hickory
forest vegetation. Other tree associates are yellow-poplar,
sugar maple, white and green ashes, sweetgum and black walnut.
Honeylocust is dominant on soils that formed in shaly limestone.
Riparian forests include silver maple, cottonwood, sycamore, pin
oak, elm and sweetgum.
Most of this ecoregion is farmland. About half the area is
cropland, primarily corn, soybeans, other feed grains, hay for
cattle and tobacco. The remaining farmland is in pasture. One
fifth of the area is forested. Influences on stream quality are
mainly from livestock and crop production. In cropland areas,
many rivers have been dammed for flood control.
-------
WESTERN ALLEGHENY PLATEAU
The Western Allegheny Plateau ecoregion, which consists of a
dissected plateau comprised of horizontally bedded sandstone,
siltstone, shale and limestone, 1s characterized by steeper, more
rugged terrain than neighboring ecoregions to the north and west.
Elevation ranges from about 600 feet along the Ohio river to over
1,300 feet 1n the higher ridges. Local relief 1s between 300 to
500 feet 1n most of the Ohio portion of the ecoregion.
Most streams 1n the larger watersheds contained entirely
within the ecoregion drain 200 to 400 mi2. However, some drain
> 650 mi2. Most mapped streams are perennial, and a few in
smaller agricultural valleys are channelized. The few natural
lakes are small, usually < 0.5 mi2. Reservoirs, though few,
outnumber natural lakes, and are generally < 5 mi2.
This ecoregion supports mixed mesophytic forest vegetation
1n which, characteristically, dominance 1s shared by greater
numbers of species than in the rest of Ohio. The composition of
this forest changes with moisture availability and soil
fertility. Major forest species are oak (white, black, northern
red, scarlet) and hickory (shagbark, bitternut, pignut, and
mockernut). Oak, blackgum, flowering dogwood, and pine
(Virginia, pitch, shortleaf) occur mainly on ridgetops and
shallower soils. Yellow-poplar, black walnut, red oak, and red
maple grow in more sheltered locations.
Soils of this ecoregion formed predominantly from unglac-
iated clay, shale, and siltstone, and include a capping of loess
in some areas. Soils are mainly Udalfs, Udults, and Ochrepts.
Hapludalfs formed In residuum from shale and siltstone.
Hapludults formed in residuum from acid siltstone, shale, and
sandstone. Dystrochrepts are common on steep slopes and ridges.
Land use in this ecoregion is limited by poor soils, steep
topography and high erosion hazard. Thus, most of the area is
forested and timber harvest 1s important. A large portion has
been strip-mined for coal. Less than 2Q% 1s cropland, which
occurs on valley floors, usually in alfalfa and small grains for
beef and dairy cattle. Fruit and vegetables are farmed on a
local scale. Urban growth continually Infringes on farmland
areas.
Stream water quality Is primarily affected by mining
operations. Numerous oil and gas fields also affect stream water
quality. Though agriculture accounts for a relatively small
proportion of land use, 1t occurs primarily in stream valleys,
resulting In loss of riparian vegetation and Increased
sedimentation.
10
-------
SECTION 3
SELECTION OF CANDIDATE WATERSHEDS AND STUDY SITES
First, we outlined all watersheds that fell completely
within the roost typical or generally typical portions of each
ecoregion. This prevented aggregation of characteristics from
different ecoregions or portions of ecoregions. We drew the
watershed outlines on a 1:500,000 scale topographic map of the
state and estimated the area of each watershed.
The smallest watersheds selected were those expected to have
permanent streams. In Ohio mean annual runoff is about 15 inches
(varying from about 10 inches in the northwest to about 20 inches
in the extreme northeast) with minor seasonal differences in
precipitation. Thus minimum watershed size was about 10 square
miles. The largest watersheds (100 to 300 mi2) were contained
entirely within a generally typical or most typical portion of an
ecoregion. Also, to better understand the stream-ecosystem
quality attainable in watersheds straddling two ecoregions, or
the most typical and generally typical portions of ecoregions, we
selected a few least-Impacted watersheds in these locations.
Using general information on point and nonpoint pollution
sources, we eliminated those watersheds and streams with a
relatively heavy human impact. The source materials that were
used for this sorting process included:
1. Maps of human population density and census data. These
provided a rough approximation of the sewered and unsewered
population in each watershed, as well as the number and
sizes of large cities where point source Impacts were more
likely.
2. Maps of land use, past and present strip mining, and streams
impacted by strip mines"! rnese maps helped reveal ETfe
watersheds that were heavily impacted by strip mining,
urbanization, and agriculture. Although land use maps were
not available for the western fifth of the state, the land
use there 1s generally limited to cropland or urban areas.
3. A watershed disturbance ranking compiled from the land use
and strip mining maps.For each watersned, we calculated a
disturoance ranking Dy multiplying the percent area in each
land use class by a disturbance value arbitrarily assigned
to each class. In the absence of an appropriate precedent,
we based the ranking on our estimate of relative probable
impact (for example, strip mining was assigned a value of
10; forest 0; cropland 4; industrial 7; and residential, 4).
In the Eastern Corn Belt Plains and the Huron/Erie Lake
Plain, watershed rankings differed little because of the
uniform influence of agricultural land use. In the other
three ecoregions, where there were great differences in
11
-------
agricultural land use, urbanization, forest cover, and strip
mining, the least- and most-impacted watersheds were obvious
from their rankings.
4. A list of important point sources occurring in the candidate
watersneimme unio LPA compiled tms list from tneir
files of known municipal and industrial point sources. This
list was purposely general to expedite plotting of point
sources with notations about their relative importance. By
using the receiving stream name and river mile index, we
plotted the point source Information on an overlay of the
candidate watersheds. In watersheds, ecoregions, or
portions of ecoregions where there were few point sources,
it was important to know the location and type of each point
source to avoid them or disregard them 1f they were
determined to be unimportant. On the other hand, in a
watershed littered with point sources, that knowledge alone
was sufficient to eliminate it.
Using this approach, we selected sets of least-impacted
candidate watersheds 1n the most typical and generally typical
areas of each ecoregion and in some areas that straddled regional
and most typical boundaries. We stress that these watersheds are
not pristine or undisturbed, but they represent the least-
impacted conditions in an area (by being outside the influence of
identifiable point and nonpoint sources), and they should there-
fore represent potentially attainable conditions from a regional
viewpoint.
FINAL SELECTION OF STUDY SITES
We based final selection on field examination of each of the
candidate watersheds and streams. Each candidate stream site,
and the watershed immediately upstream from the site, was photo-
graphed from altitudes approximately 2,000 and 5,000 feet above
the ground. These photos were used to: (1) assess typical
watershed and riparian characteristics in each region; (2) detect
significant disturbances not found by other means; (3) select
candidate sampling sites; and (4) provide visual aids for
briefings on the project. Finally, we inspected from the ground
each candidate site and two or three additional locations one to
two miles upstream or downstream to determine: (1) the
representative nature of the site; (2) the ease of access to the
water; and (3) the least-impacted sites. Factors examined
included the amount and age of stream channelization, amount and
size of riparian canopy, channel morphology, water volume, bottom
substrate size and heterogeneity, obvious color and odor
problems, and the amount of large woody debris in the channel.
Locations of the candidate watersheds, and the most typical
and generally typical portions of the aquatic ecoregions are
shown in Figure 1. Table 2 1s a regional summary of the number
of sites, their average watershed areas, and their type.
12
-------
t
n
s
•H
n
o>
e
•H
^H
H
IM
0
!
4J
5
•H
M
4J
n
Mj
0
%
CM
«
iH
jQ
2
e >i
0 M
w-4 «
n OITJ
o u c
U U 3
U 0 0
S*
•^
C&3
M 4) «
4) J= 4>
4J 0>il
« t) id
4) i-H i— 1
S^H eu
««
h
0 3
-H id
u at
0) •(•>
V «
C^H
H 04
o
•H C
1-4 -H
« id
4J r+
C 04
^«
OIX
•H fl
U J
U
JJ
CH
M 41 W
4) 0 C
4J -H
M C Id
StHI-t
O O4
U
v e
-H-H
U «
U rH
^>04
c
O 0)
MX
3 «
XJ
rg
t^
CM r* «-i «H m
o •*>
f-«
II
M
*J
•H O
w «
^ r» ^c eo i1 IM
rH -H 001
id >-i t-i r-
4J
g
f>
in
O VO O V •» P4
in m I
tH »-l «-t
in
00
VO
TT in O Ot * rH
t-i m ^r i
9>
r^
9t
«v in ^ in in IH
i-i T-I r- vo i
00
^>
r» to IH »-i
O 01 >H
in >» i
o
»H
O PM
— -H
c
- rH SN -
E-* * M
M £ u « n
« — -H -O 4)
*j a c N
•H i-l S 3 -H
w * frT o n
u m
UJ J W^ MH
•W •« ^1 w
O PU --I EH 0)
K rH 0 JS
Ij EH « \ W C
41 M EH M id 4)
.Q 4J 4> £ 4) C -H D>
g M C *J « -O C
3 O 4) C 10 4) 4) id
2£(90 » £ £ K
13
-------
SECTION 4
FIELD SAMPLING AND ANALYSES
The OEPA sampled water quality characteristics approx-
imately monthly from July 1983--November 1984, including:
temperature, conductivity, alkalinity, hardness, pH, dissolved
oxygen, nitrite, nitrate, ammonia-nitrogen, Kjeldahl nitrogen,
total phosphorus, total organic carbon, total residue, magnesium,
calcium, cadmium, chromium, nickel, zinc, iron, copper, and lead.
Subsurface samples were taken midstream, directly collected in
1-liter disposable polyethylene cubitainers or using a plastic
bucket and then pouring into the cubitainers. Nitrogen and
phosphorus were preserved with 2 ml concentrated nitric acid;
alkalinity and conductivity samples were iced. Samples for total
organic carbon were collected in acid-rinsed 25-ml polypropylene
vials, preserved with mercuric chloride and shipped to ERL-
Corvallis for analysis. All samples were analyzed according to
methods summarized in Table 3.
Quality assurance procedures for all samples included:
field blanks and duplicates collected at a 5 percent frequency;
and replicates, spiked samples and blanks analyzed at a 5 to 10
percent frequency for laboratory analyses (OEPA 1984a). Samples
not meeting precision and accuracy goals specified in Table 3
were rejected.
The OEPA sampled half the sites during each of two
consecutive summers (1983, 1984) for physical habitat, fish and
macroinvertebrate assemblages. We selected sites to be sampled
each summer to represent each region and to prevent spatial bias
in the results. The physical habitat (percent canopy; mapped
gradient; dominant riparian vegetation; percent cobble, gravel,
sand, silt, muck, bedrock; percent instream cover provided by
macrophytes, stumps, brush, snags, undercut banks; mean, maximum,
and minimum depth and width of riffles, runs, and pools) was
assessed twice, once during midsummer when crews sampled the fish
assemblages and again during macroinvertebrate sampling.
The OEPA sampled for fish at each site two to three times at
approximately one-month intervals from late July to October.
Rivers too deep to wade were sampled with a boat-mounted electro-
fisher. Small, shallow creeks were sampled with a backpack
electrofisher and the other streams were sampled with a towed
electrofisher. Whenever possible the towed unit was used because
of its greater effectiveness. Sites sampled by boat, towed unit
14
-------
I—I
IB
U
"E
U
U
l-l
0
•H r-
in a\
•H rH
0 ^
01 <
u a.
aw
in
in —
•C TJ
4J 3
Q) 4J
E v)
rH CO
10 -H
Ufi
•rl 4J
4J
rH -H
IB
CTJ
IB C)
in
"o
in
r*^ fl)
M (A
10 S*
£ rH
E 10
3 C
W IB
•
m
01
rH
J3
H
•4
Q
£
fM
^
U
IB
O
U
B
O
•rH
in
u
^
IX
"H
4J
fl)
rH
J5
10
•H
tl
IB
>
X^_ X^
i i
0 0
in rH
in o
rH rH
+ 1 +1
rfP tfP
O O
rH rH
+i +i
B
5 B
4J 0
IB -H
4j o.
•H *-l
H O
in
TJ «Q
C <
4J
iO U
E -H
4J O
3 4J
< <
;„
4J
•rH
B E
"H 3
rH -rl
10 O
.* rH
rH IQ
< U
•J
i
O
rH
in
rH
•H
ac
o
+ 1
i^
0)
N
^
rH
IB
I
Ul
0
U
•H
IH
fl)
g
Q
3
S
U
•H
E
10
Q\
14
O
rH
iO
4J
0
H
k,
B
t4
10
U
E
W
O
rH
in
•f 1
df
m
-rl
01
Oi
TJ
IH
m
01
O
4J
m
4J
|Q
0)
§
^
4J
•H
^
•H
4J
O
3
T3
B
8
•j j j ,j
o o o o
£ £ £ £
m
in o o in
o o IN o
o o o o
rH rH rH rH
+ 1 +1 +1 +1
df Of Of
0) 0) O
O, B TJ B IH
a o IB t?> 4->
O VJ fl) IB -H
CJ M J £ 2
tJ J J J
U O O O
o IN in o
rH O O
O -H
•H O
4-1
u a
1 1
K at
IH
§0) 01
N 4J
•H >1 (0
•3 ie o oi
IB E--H fi
U < 4J 0.
IB
TJTJ NTI
4) 0) -H 0)
4J 4J IH 4J
IB 10 01 IB
E E rH E
4J 4J CO 4->
3 3 01 3
< < 2 <
rH
JB
IB
•O
rH
0)
0) 4) IB US
4J 4J -H
10-H CrH
IH M O 10
4J 4J g 4J
•H-H g O
2 2 5 H
J
i
in
0
O
0
rH
•H
00
o
*l
B
O
•rH
4J
U
3
TJ
0)
K
TJ
U
U
•rH
M
U
a
T»
0)
4J
c
4J
3
rH
IB
o
w
3
^
^
a
V)
0
£
10
Tl
IH
•g
10
4J
U)
TJ
B
U)
4J
B
0)
c
0)
IH
HI
iO
I
VH
U)
0)
(0 -H
A) IH
IB 10
U
•H IB
rH
a c
3 fl)
•O 0)
D<4J
B 0)rH
o .5 o)
E >
iO 01 0)
U J
B B
O 0) E
"H IH O
4J 01 -H
iO UH 4J
•H >*H O
>"H fl)
0) TJ 4J
•O (U
MH Q
TJ 0
IB 01 3
TJ 0) E
C IH-H
IB O> B
4J 01 -H
WTJ £
rH
-------
or backpack were fished for a distance of approximately 500, 300
or 200 meters, respectively, with the availability of micro-
habitats determining the actual distance fished. All captured
fish were identified, counted and weighed in the field. Selected
specimens were preserved to confirm field identifications. They
sampled macroinvertebrates once in the summer with modified
Hester-Dendy multiple plate samplers and kick net transects in
riffles or areas of greatest habitat diversity where riffles were
absent (Pollard and Kinney 1979). Samples were preserved in
formalin in the field, identified later to the lowest taxonomic
level that keys permitted, and quantified by number and size per
time and area sampled.
16
-------
SECTION 5
DATA ANALYSES AND RESULTS
The results are organized by sections on fish, macroin-
vertebrates, water quality, and physical habitat. They are
summarized primarily graphically because visual displays are
easiest to grasp. We used box plots to summarize many of the
results (Reckhow 1980). Box plots contain sample sizes, medians,
ranges, and 10tn, 25tn, 75tn and 90tn percentlles. Box plots have
the advantage over other methods of presentation, such as means,
standard deviations and standard errors, because they do not
assume a particular data distribution. They provide more
information about the data distribution, such as central
tendencies, outliers and skewness.
Our purpose 1s to show how the regional framework 1s useful
for examining and characterizing the range of conditions in a
geographic area and for distinguishing major differences among
regions. This is an important basic step toward determining
attainable water quality. We do not present statistical tests to
determine whether the characteristics of one region differ
significantly from those of another. For many variables, two of
the regions have little or no overlap, making statistical tests
of difference unnecessary. The other three regions show varying
degrees of overlap. From a management viewpoint It may be
appropriate to combine data for some variables and regions.
However, the regions are geographically distinguishable and for
some variables this distinction might be Important, particularly
1f stream characteristics are to be related to land management
practices.
Data from sites in the most typical and generally typical areas
have been combined for each region. The differences among
regional characteristics in Ohio are not as distinct as in some
parts of the country (e.g. the mountainous West), so that the
transitional nature of the generally typical areas 1s subtle.
Although the data may display similar values among the regions,
the factors that cause them to be similar might be different,
e.g. two very different types of soils may produce similar
nutrient chemical conditions in streams.
17
-------
FISH
We present a variety of measures to characterize and summar-
ize the fish assemblages, Including dominance, species richness,
species diversity, Karr's index of biotic Integrity (Karr 1981,
Fausch et al. 1984), a composite index or index of well being
(Gammon 1980), and the fraction of the fish community that is
tolerant of turbidity, sedimentation and low dissolved oxygen.
After consulting with OEPA biologists, we omitted several
sites and samples from the analyses. We excluded six sites
because, subsequent to the design and sampling phases, watershed
and stream characteristics were discovered that Indicated these
sites were not representative of least-Impacted conditions. In
addition, we excluded 17 individual samples (out of 312
remaining) due to Identifiable sampling problems (Table 4). We
calculated all indices for each sample. For each index, we
examined both the maximum and median scores at each site. Mostly
this report uses site maxima. For species richness and percent
intolerant individuals, we also combined all fish caught at each
site for a composite sample. Hybrids are not counted as
"species" in the following indices: IBI, species richness, and
percent intolerant individuals. For composite Index, species
diversity, and site composite species richness, hybrids are
treated as "species." Sites on regional boundaries have been
excluded from most analyses.
Dominant Species by Ecoregions
We used two approaches to describe the dominant fish species
for each ecoregion. First, we listed the species that were among
the numerically dominant in any sample (> 10% of sample). For
each region, we calculated the fraction of samples in which each
of those species were dominant (Table 5). Histograms of the same
data show ecoregional differences in the dominance of these 23
species (Figure 2).
The second approach to dominant species is a regional
signature based on relative abundance of species. We selected
all species that had a mean relative abundance of > 2% for any
region. For each region, we constructed box plots of relative
abundances of only those species that occurred in at least half
of that region's samples (Figure 3).
Species Richness
Species richness (the total number of different species
sampled) 1s a simple and direct measure of the variety of species
present in a community. In general, species richness in small-
to medium-sized streams increases as the size of a stream
increases. Because of this relationship and because of the range
of stream sizes examined in this project, box plots of species
richness by ecoregion contain variability attributed to that
18
-------
Table 4. Sites and samples eliminated from analyses of fish data.
The following sites were dropped from the analyses.
SRP CODE STREAM RATIONALE
NGTS14B Bad Creek
WMG116 Mad River
WMT61A Honey Creek
EMT32C Potter Creek
EMT29 Pymatuning Creek
SGT107 still Fork Sandy
channelized/deep trough
atypical/cold water stream
no obvious reason, but low
species and individual counts
marsh stream, 0-2 species
swamp stream/black water, deep
dam backwater
The following samples were dropped from the analyses.
SRP CODE
STREAM
SAMPLE ft
RATIONALE
NMT66B Zielke Ditch
NMT72B Cries Ditch
WGT118 North Fork Paint
WMG114 Honey Creek
EGT38 Sugar Creek
SMG102 Sunfish Creek
SMT1 Pine Creek
SMT4 Federal Creek
SMT5 Wolf Creek
SMT5B W. Br. Wolf Creek
SMT5C S. Br. Wolf Creek
IP109 Stonelick Creek
BIW77S E. Fork Little
Miami
3 different site/closer
to unsewered community
3 dry stream
1 boat sampling
4 late in year
3 high flow
2 & 3 after fish kill(s)
3 late, cold/rainy
1 high flow
1 storm event
2 leaf fall covered water
2 data sheet missing
2 boat sample/deep water
2 & 3 stream drying/low
species counts
2 & 3 boat shocker only
low species counts
19
-------
I
•b
I
•s
I
2L
e
£
I
L
5
L>
O
Lt
4J
C
^
E £
« f
« ?
• -i
'3
0
Li
4
4J
tl
U
C £
tl TO
Ij
Li
tu
C
o
a
EC
^
. M
V
•t
a. *o
^~
tt
n o.
* §
n *>
•o
.'--
•^ ti
*-t
VI
Lt
t>
00
j
fsl
rJ
to
^
•— *
g
C
c
•H
e
M
0
c
c
• 3
CO
f>
Z
—
\
c
E
tnose
I
CO
00
o*
*
^
1
c
VI
0
c
1
00
oo
*
c
c
E
M
0
c
1
CO
o
«n
^
^
o
c
«
o
c
4J
c
CO
CO
ao
n
ri
w'
Li
tl
C
•vl
f
M
•8
CX
LI
«J
tn
*
o
**>
rt
L>
J5
1
*H tl
N C
L. O
C V)
0
o
o*
•*
?7
JO
J=
U
w
Li
CM
^
^
Jtf
2
a
CO
CO
•~*
o
r-i
4>
i
TJ
W
LI
C
tl
S
m
,,1
m
ri
ri
*~*
O
Li
•-H tl
* C
U O
4J »-*
C V)
t)
U
m
•9
r-l
U
tl
°
0
«o
M
CO
^
V)
**
Li
W
3
S
JC
V)
**-»
§
M
c
b
ao
oo
*^
C^
ri
S
X
vt
•o
Ll
1
•9
ri
n
ri
%— -
1
U
tl
t>
Li
?n
*£>
r>
LI
t>
c
M
•o
g.
"C
I/I
o
ID
rt
rt
—
2
o
2
g.
V)
|
CO
Li
£
tl
c
o
vt
1
i
o
CO
CO
—
£>
£
C
V
t>
o
er
,^
—
^
ri
Li
V
C
J=
VI
•o
Li
cn
^
o>
w
M
IV
JO
M
U
O
DC
to
rn
•"•
^
J=
VI
1
V)
c
£
2
Li
S
Li
•o
>'
1
»A
r-*
^*
o
c
w
c
tl
w
o
PC
^
ri
•~*
o
tt
n
o
•o
c
-o
rs
2
CO
s
U
£}
.n
^
~*
M
~
^
g
ic
•o
s.
to
o
tn
1
e
J
*j
^
o
00
J=
M
C
jj
«
00
j
is*
£
M
00
1
CQ
.,
0
r*
—
|
C
»*-»
J.
o
w
VI
1
oc
0
o*
0^
"-
1
JC
VI
«
u>
™
2
«
iJ
U
•3
w
1
o
-4
o*
^
^
s
Li
•3
TJ
S
S
r-.
CD
t)
U
3
V)
1
o
a>
^
•*"
c
JC
VI
C
8.
ri M M CM n r*
r* o O co oo ao
m n r- < »n *
c
O. JC M U W
«-* VI tl « 4J
3 -r« V C Li
O «n M T* * (
VI C * J= T3
3 T? VI
•O VI >
« >. c 8
-• c c o .2
•J 0) C E C
O Lr O O «*
I O *~> O DC
-» »n »n ri
O OO 00 CO
r* oo O O
^. ^
M Li
» b O C
*-> t) JC -*
1- J^ t) JC
•O 3 L-
V) C
c' •> v «n
X ••* — * vi
O JC O O
-» 5 (5 oe
o n
«— ' r-i
^
C^ '*4
•-« JC
J "
So
o.
9 5
>- to
20
-------
100
80
60
40
20
S3
1 bluntnose minnow
2 creek chub
3 green sunfish
4 blackstripe topminnow
5 fathead minnow
6 yellow bullhead
7 johnny darter
8 white sucker
9 rockbass
10 rosefin shiner
11 greenside darter
12 bluegill
13 mottled sculpin
14 common shiner
15 rainbow darter
16 fantail darter
17 spotfin shiner
18 longear sunfish
19 sand shiner
20 golden redhorse
21 emerald shiner
22 central stoneroller
23 striped shiner
123 7 8 10 11 17 18 20 22 23
Eastern Corn Belt Plains
100
eo
60
I"
20
100
80
60
40
20
1 23456789
23
lish tp»ci«$
Huron/Erie Lake Plain
123 7 B 9 It 121314 IS 16
Erie/Ontario Lake Plain
2223
100
80
60
40
20
0
R
A
%
''{
\
'•3
1
:;
'•/.
i R
1 1
!
ill
13 10 16 2021
Interior Plateau
too
80
60
40
20
1 2 17161920212223
Western Allegheny Plateau
Figure 2. Dominant fish species. Fraction of samples 1n each
region 1n which these species were dominant ( > 10% of sample).
21
-------
z
<
•
§
a!
IU
- <
O
£
S I
u
a
IU
S S
) •ouepunge
.c u a> o
i.kL. in L. 3 V> L, t.
Ol O> Ol ••- Ol l/< I- O> Ol
c *-> c i/> t*. *J en O c c
•»" k, •>- I/I C L. O .C — O
frre.C'e 3 m .c 13 jr *->
int^m^ i/»x> OIIAI/I
•— IA c U
•oxc-o*1- mv.— k. T3«—
•— o*^Oi>*-n>ro-^OiCOliO
^ ^ I** 4^ C^JO O> IQ f Ol & k>
O-'-OOSUCCfc.'— U C
E (O a. o.<— o o 01
Of k. I/I IX) ^ k> ^" ^ C O> VI W
E »- *J 0) 01 •— c o.^J "- «
J3 U
(0
S
e
O
e U
a E
u
H
- to
- 2
-------
source. Therefore, we regressed species richness on watershed
area (as a measure of stream size, see Hughes and Omernlk 1983),
then examined residuals from this relationship for regional
patterns.
We regressed the maximum number of species found at each
site (from two or three samples) against log^o watershed area
(Figure 4a). Box plots by ecoreglon of the residuals from this
relationship reveal a pattern that will reoccur for all of the
other Indices (Figure 4b). For watersheds of similar size,
species richness tends to be lowest 1n the Huron/Erie Lake Plain
and highest 1n the Western Allegheny Plateau with Intermediate
values 1n the other regions. Greatest variation occurs in the
Eastern Corn Belt Plains.
Karr's Index of Blotlc Integrity
We introduce the IBI by quoting from Karr (1981):
Accurate assessment of biotic integrity requires a
methodology that Integrates responses of blotic
communities through an examination of patterns and
processes from population to ecosystem levels. One
approach is to develop an array of biological metrics
like the leading economic indicators used in
econometric analyses. The Index of Biotic Integrity
(IBI) uses data from collections of entire fish
communities and summarizes them as 12 ecological
characteristics, or metrics, which can be classified
into three categories: species richness and
composition, trophic composition, and fish abundance
and condition. Values of each metric are compared to
values expected at sites of similar stream size and
regional location with minimal human influences....
The metrics chosen for this analysis are measur-
able attributes of the community that are correlated
with blotic integrity, which 1s not directly measur-
able. Each of the metrics is of Interest for the
information that it conveys about the overall structure
and function of the stream community. In addition,
each is of Interest because it reflects something about
biotic integrity. The values of the metrics are
functions of the underlying blotic integrity; biotlc
integrity is not a function of the metrics.
The IBI scores were calculated as outlined by Fausch et al.
(1984). Some of these scores may be slightly reduced by
excluding exotic species (common carp and goldfish) from species
counts. We used a restricted 11st of Intolerant species. Karr et
al. (1986) suggested that intolerant species be limited to 10-15%
of the species present In the area, i.e., 12 of 99 species
(Appendix A). Only those species most sensitive to habitat
23
-------
40
? 34
V—'
CO
UJ 28
O
LLJ
Q. o9
CO "
16
LL
10
1 2
0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
Iog10 WATERSHED AREA (mi*)
16 T
A
D
I °~
12 -
e -
— 90 1
7' W
- 75 1 _/
£ < A •
— Median 1 9
» ai 0 -
Relative
Sample Size -e -
-12
D
J
"TV
j i
i
>
n .
~~ e
TJ
J
, c
,-1
1
9
J~
HELP ECBP EOLP IP WAP
ECOREGION
Figure 4. Fish species richness, (a) Regression of maximum fish
species at each site v_s_ logjo watershed area: 1 « Huron/Erie
Lake Plain, 2 = Eastern Corn Belt Plains, 3 « Erie/Ontario Lake
Plain, 4 = Interior Plateau, 5 * Western Allegheny Plateau.
(b) Boxplots of the residuals of the species richness regression
by region.
24
-------
degradation, yet are fairly widespread, were used. Anomaly
scores are based on the fraction of fishes observed with
anomalies, excluding blackspot. To examine possible relation-
ships among blackspot, incidences of other anomalies, and IBI
scores, we performed correlations of the fraction of individuals
with blackspot versus the fraction of all other anomalies, and
with the IBI scores both statewide and within ecoregions. No
clear pattern was discerned. Blackspot incidence was not
correlated with IBI scores or other anomalies. Other anomalies
showed a slight negative correlation (r = -.26) with IBI scores
(as expected). High incidence of blackspot (> 10%) was used to
reduce scores where the fraction of other anomalies was near
cutoff points. Refinements of these and other metrics and
scoring criteria are being developed by OEPA as part of their
ongoing monitoring program.
Generally, highest IBI values occurred at sites in the
Western Allegheny Plateau, while lowest values occurred at sites
in the Huron/Erie Lake Plain. In fact, the lowest values in the
former exceeded the highest values in the latter. Values in the
transitional ecoregions were intermediate and, especially 1n the
Eastern Corn Belt Plains and Erie/Ontario Lake Plain, displayed
wide ranges (Figure 5).
r '
I5oH
LU
sr
V) 5 '
46-
5
L _
38-
• 34-
o
o
a
L30
o
i
HELP ECBP EOLP IP
ECOREGION
WAP
Figure 5. Boxplots of the maximum site Index of Biotic Integrity
scores by region. Qualitative evaluations from Karr (1981).
25
-------
Gammon's Index of Well-Being
Gammon's Index of Well Being (IWB) incorporates two widely
used indices: diversity and abundance. Gammon (1976, 1980)
reported that it appears to reflect environmental quality more
satisfactorily than previously developed community indices.
Whereas the IBI was developed for small streams, the IWB was
designed for, and usually applied to, large streams and rivers.
It is calculated (by OEPA) as:
IWB = 0.5 In N + 0.5 In W + H'(no.) + H'(wt.)
where N
= number of fish captured per kilometer for boat
samples, or per 0.3 kilometer for towed
generator and backpack electrofisher
W = weight in kg per distance as above;
H'(no.) = Shannon diversity based on numbers;
H'(wt.) = Shannon diversity based on weight.
The different weightings used above result from differences in
sampling efficiency of the equipment.
We found a slight (r* = 0.19) but statistically significant
(P < 0.0001) linear relationship between IWB and log^n watershed
area (Figure 6a). Residuals display the expected regional
pattern, although variability is quite high (Figure 6b).
Negative values occur 1n the Huron/Erie Lake Plain, and positive
values in the Western Allegheny Plateau. Because of the very low
correlation between IWB and watershed area, It seems appropriate
to refer directly to the regional patterns 1n IWB (Figure 6c).
Species Diversity
Species diversity is a commonly used community index that
combines species richness and equitability, the relative
abundance of species. Several calculating methods are available.
OEPA formerly Included Shannon diversity among its analyses, so
the same index is used here.
Species diversity is linearly related to log^o watershed
area (r? = 0.36, p < 0.0001); a stronger relationship probably
exists if the four circled values are excluded as apparent
outliers (Figure 7a). Residuals display the previously noted
ecoregional patterns, negative values in
Plain and positive values In the Western
(Figure 7b). Relatively low values also
Plateau.
the Huron/Erie Lake
Allegheny Plateau
occur in the Interior
26
-------
12.0
11.0
LLJ
§10.0
o
i9'0
8.0
7.0
0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
Iog10 WATERSHED AREA CmJ2)
2.0
1.5
1.0
•1.0-
-1.5 -
-2.0
HELP ECBP EOLP IP WAP
ECOREGION
12.0-1
11.0 -
|£ 10.0-
O
O
eo
CD 9.0 -
8.0-
7.0-
A
HELP ECBP EOLP IP WAP
ECOREGION
Figure 6. Index of Well-Being (IWB). (a) Regression of maximum
IWB score at each site vs logiQ watershed area: 1 « Huron/Erie
Lake Plain, 2 « EasternTorn Belt Plains, 3 « Erie/Ontario Lake
Plain, 4 * Interior Plateau, 5 « Western Allegheny Plateau.
(b) Boxplots of residuals of IWB regression by region.
(c) Boxplots of actual IWB scores by region.
27
-------
3.0
1.4
0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
logio WATERSHED AREA Cmi3
1.0 n
0.5-
CO
Q 0.0 -
GO
LU
tr
-0.5-
-1.0
HELP ECBP EOLP IP
ECOREGION
WAP
Figure 7. Fish species diversity, (a) Regression of maximum
fish species diversity (H1) at each site vs^logjo watershed area:
1 * Huron/Erie Lake Plain, 2 « Eastern Corn Belt Plains,
3 « Erie/Ontario Lake Plain, 4 « Interior Plateau, 5 « Western
Allegheny Plateau. Apparent outlier values circled, (b) Boxplots
of residuals of fish species diversity regression by region.
28
-------
Intolerant Species and Individuals
Other measures of the health or integrity of a stream
include the number of species and fraction of Individuals in a
fish community that are generally intolerant of environmental
degradation. We consider intolerant species those that tend to
require high dissolved oxygen levels and low levels of turbidity
and siltation. Information regarding tolerance is generally
available from books that summarize fish distributions and
habitat by state. We used Becker (1983), Carlander (1969, 1977),
Lee et al. (1980), Pflieger (1975), Smith (1979), and Trautman
(1981) to establish the list of species tolerances summarized in
Appendix A.
As seen with the other measures, there is a significant
linear relationship between the number of intolerant species and
watershed area (r2 = 0.34, p < 0.0001) (Figure 8a). Note that
the number of intolerant species in the Huron/Erie Lake Plain
appears to increase little or not at all with stream size.
Residuals display the typical pattern seen for the other
variables, negative 1n the Huron/Erie Lake Plain, generally
positive in the Western Allegheny Plateau, with a fairly wide
range of values for some ecoregions (Figure 8b). There is no
relationship between the fraction of intolerant individuals and
stream size so these are displayed directly (Figure 8c).
MACROINYERTEBRATES
We present only data collected by the kick net method for
macrolnvertebrates. Preliminary analyses of data collected using
Hester-Dendy multiple-plate samplers did not produce a clearer
regional pattern than the kick net data. Not all organisms
collected could be identified to species. Preliminary plots of
the percent of individuals and of taxa keyed to each level
indicated no regional differences in the level of taxonomic
resolution. For this report, "taxa" refers to the lowest level
of taxonomic identification achieved for each specimen and in
this sense is analogous to "species".
Richness and Diversity
The total numbers of different macroinvertebrate taxa
collected at each site (taxa richness) do not show distinct
differences among ecoregions (Figure 9a). The regional medians
range from 27 in the Huron/Erie Lake Plain to 31 In the
Erie/Ontario Lake Plain. The total range is from a low of 13 in
the Western Allegheny Plateau to a high of 51 in the Erie/Ontario
Lake Plain. Shannon diversity values (H1) for the taxa collected
at each site also do not differ among the ecoregions (Figure 9b).
The medians range from 3.06 in the Interior Plateau to 3.39 in
29
-------
r^ 18
OT 15
UJ
O 12 -
UJ '* 1
CL
00 9
r-
z
6
3
0
oc
LU
5 5
43 243 2-^14 3
B 5 32 3^f^ S 2
3 3 32 2>-^2 S 3 2
E 3 >^^^ 23 5
11
0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
login WATERSHED AREA (mi*)
'10
80-1
5 60-
SPECIES
X
£2 40-
LL
t-
£L
UJ
_J
P 20-
=
J
)
]
-r
O
u
T
lA i
r
0
(
V
8.0 -
6.0 -
4.0 -
A 3 2'°"
!
e
••*.
B o -
to
UJ
C-2.0-
-4.0 -
V
. -R n
1
111
D
. . -
R 1 Ju
A 11
T
o
..
V
T
T
-L.
J L
o
--8-iT-
HELP ECBP EOLP IP WAP
ECOREGION
HELP ECBP EOLP IP WAP
ECOREGION
Figure 8. Intolerant fish species, (a) Regression of maximum
number of Intolerant fish species at each site vs login watershed
area: 1 « Huron/Erie Lake Plain, 2 * Eastern Ccfrn Belt Plains,
3 = Erie/Ontario Lake Plain, 4 - Interior Plateau, 5 - Western
Allegheny Plateau, (b) Boxplots of residuals of Intolerant fish
species regression by region, (c) Boxplots of maximum percent of
intolerant fish species each site by region.
30
-------
50-
40-
X
t-
0 30.
CC
UJ
CD
20.
10
a
I
Jl
e
y
-
i
I
e
-
e
y
t
5.0-
C^ 4.0-
I
n w
| O
SPECIES DIVER:
ro u
b o
• i
T I
h _ .,,, , ro
b
II A
n 1
e •
n
A
e
y
•
i
1
e
1
1
HELP ECBP EOLP IP WAP
ECOREGION
HELP ECBP EOLP IP WAP
ECOREGION
CO
tr
UJ
g
O
tr
UJ
CD
9-,
8-
6-
5-
II
V
HELP ECBP EOLP IP
ECOREGION
WAP
1.0-1
0.75'
CO
0.50-
cc
UJ
D
CC
0 0.25
A
HELP ECBP EOLP IP
ECOREGION
WAP
Figure 9. Macro-invertebrate richness and diversity, (a) Taxa
richness of macroinvertebrates by region, (b) Taxa diversity
(H ) by region, (c) Macroinvertebrate order richness by region
(d) Macroinvertebrate order diversity by region.
31
-------
the Western Allegheny Plateau, and both extreme values fall in
the Western Allegheny Plateau (1.47 to 4.54).
Slight differences exist among ecoregions when aquatic
macroinvertebrate richness and diversity are calculated at the
order level. Medians of the total numbers of orders represented
span from 5.0 in the Eastern Corn Belt Plains ecoregion to 6.5 in
the Interior Plateau (Figure 9c). These differences seem to
result primarily from the Megalopteran and Plecopteran taxa which
are present at a greater proportion of sites in the Interior
Plateau than the other ecoregions and marginally represented or
absent in the Eastern Corn Belt Plains and Huron/Erie Lake Plain.
The latter two ecoregions also tend to have fewer Odonate taxa.
Note that the median richness in the Huron/ Erie Lake Plain is
also the maximum value in this ecoregion. Median insect order
diversities span from 0.431 in the Huron/ Erie Lake Plain to
0.561 in the Interior Plateau (Figure 9d). The lower range, 25th
percentile and the median of the order diversity values are
notably lower in the Huron/Erie Lake Plain ecoregion, apparently
as a result of both moderately low richness and high dominance of
several Ephemeropteran and Dipteran taxa.
Insect Taxa Dominance by Ecoregion
A "taxa signature" or plot of the relative abundances of
various insect taxa that could be expected in each ecoregion was
constructed as follows: First, for each ecoregion, the 10 most
numerically abundant taxa at each site were listed. Then, those
taxa among the 10 most abundant at four or more sites In any eco-
region were placed on the master list. Box plots of the relative
abundances of these 24 taxa were prepared for each ecoregion.
Some differences appear among the ecoregions in signature
taxa presence, abundance, and degree of dominance (Figure 10).
Cheumatopsyche sp. appears to be the dominant Trichopteran in all
ecoregions.Tt~Ts practically the only Trichopteran 1n the
Huron/Erie Lake Plain and Interior Plateau with the exception of
some Hydropsyche depravada and Hydroptila sp. in the former and
Chimarra obscura" and symplfitopsycne bifTSTa in the latter. In the
remaining three ecoregtons anfive Trichopteran signature taxa
are represented. The proportional abundances of Isonychia sp.
and Baeti s sp. are also fairly high at many sites in a11
ecoregions.~H'eptageni a sp. and Stenacron sp. also appear
frequently in moderate concentrations m tffe Huron/Erie Lake
Plain but decline in Importance in the other ecoregions, except
for Stenacron sp. in the Erie/Ontario Lake Plain. Proportional
abundances of STenonema vicari urn and Caeni s sp. Increase as one
passes from the Huron/trie Lake Plain, througTf the other
ecoregions, to the Western Allegheny Plateau.
Polypedilurn convictum is the most consistently abundant
Dipteran across ecoregions and appears to be practically the only
32
-------
NlVld 3MV1 OIUV1NO/3IH3
S 5 8 g 2 •
AN3HD311V NU31S3M
«o
c
en
»b3id03103
0.
**
E
3
*^
E
«/>
<•>
,
S
£•
o
L.
^
to
f
f
u
e-l
«
">»
c
c
*
E
0*
Of
t-
^
o.
£
IA
0.
o
E
3
£
IA
3
4-*
C
•^
O
3
0.
O
U
o
in
«
*
KI
i.
Q.
t>
0)
£
U
IA
o.
O
i-
z
o.
*rt
•A
Q.
•5
c
a/
u
u
E
to
"O
•o
**-
£
VI
o
£
0.
to
E
3
f
>
C
o
3
^5
0)
^
^_
a.
r.
Ck,
^
^j
o.
o
X
E
3
C
f-
u
3
T3
Oi
^
,__
0
a.
CO
ID
£
^
C
VI
a.
vt
3
O
O
.^
£
U
O
o
^.
4-»
IA
0V
CL
vt
^,
**
ca
CL
VI
VI
*
>,
c
ID
4-*
•a
to
o
o
CL
ID
C
O*
O)
ID
**
X
Vt
3
3
•«-
3
k.
4^
C
*
O
0>
£
oc
^
r
«A
C
O
k.
u
c
»/>
vt
c
o>
u
IA
L.
•^
^
3
VI
k.
*
>^
C
ID
h-
CSJ
E
^.
t—
£
U
O.
E
01
c
o
c
4-*
«/>
Jrf
•^
i.
£
VI
3
C
0>
a.
o*
a.
fn
E
3
L
U
*>
E
0>
c
o
c
CO
VI
VI
E
c
0>
4-*
to
«
0.
vt
V-
c
u «
•^ x
O"4->
< C
o E
•-< o
3CM
CO
*-tewn**ft«>r^oo
S 5 8 S 2 °
NIVId 3XV1 3lb3/NOUnH
So a o e e
« « n fw ••
CNIVld 1138 NbOO NU31SV3
33
-------
one present in the Interior Plateau ecoregion. Cladotanytarsus
sp. and Tanytarsus glabrescens show greater dominance 1n tne
Huron/Erie Lake Plain than in any other ecoregion. Cricotopus
bicinctus appears in small numbers in the Huron/Erie LaJce Plain
and Erie/Ontario Lake Plain, while Stichtochironomus sp. and
Simul i urn s. p. appear in moderate numbers only in tne laTter.
Abundances~of Rheotanytarsus exiguus appear greatest in the
Eastern Corn Belt Plains, Erie/Ontario Lake Plain and Western
Allegheny Plateau. Coleopterans appear abundant with fairly high
degrees of dominance in all ecoregions, particularly the Eastern
Corn Belt Plains and Interior Plateau.
Indicator Taxa
The Cornell Ecology Program TWINSPAN (Gauch 1982) was used
to hierarchically divide the samples (sites) into clusters based
on similarities in taxa composition. This program also
identifies characteristic taxa of each cluster. Rather than
identifying the numerically dominant taxa, it chooses the
(indicator) taxa that distinguish each cluster from the others.
Because the geographic pattern of sites grouped by TWINSPAN
resembles the ecoregion classification, these taxa are of inte-
rest. We prepared box plots of the relative abundances of the
indicator taxa, by insect order, for each ecoregion (Figure 11).
The number of indicator taxa present with any degree of
dominance increases from 10 in the Huron/Erie Lake Plain to 17 in
the Western Allegheny Plateau. Hemiptera are not important in
distinguishing clusters except in the latter ecoregion where
Rhagovelia s p. and Microvelia sp. are present at some sites. The
TricnopteranTcomponent ranges from none in the Huron/Erie Lake
Plain to moderate numbers of Symphitopsyche bifida in the other
ecoregions; Hydropsyche dicantha becomes equally important in the
southern two ecoregions.me Megalopteran Corydalus cornutus
also helps to distinguish these two ecoregions from the others.
Isonychia sp., Baeti s sp., and Stenacron sp. are fairly abundant,
and differ proportionaTTy in each ecoregion"; they are the only
Ephemeropteran indicators in the Huron/Erie Lake Plain.
Stenonema vicarium is an Ephemeropteran indicator in the eastern
two ecoregions, wnile £. pulchellurn replaces it in the Eastern
Corn Belt Plains and Interior Plateau; both species are observed
in the Western Allegheny Plateau ecoregion.
For the Dipterans, Polypedilum convictum occurs in moderate
abundances in all ecoregions.It is combined with Tanytarsus
gl abrescens in the Huron/Erie Lake Plain, and with RheotanyTafrsus
exiguus in the Erie/Ontario Lake Plain. P. convictum 1s combined
with both other Dipterans in the Eastern Corn Beit Plains and
Western Allegheny Plateau, and neither in the Interior Plateau.
The Coleopteran Stenelmis sp. Is the most dominant indicator at
sites in the Erie/untario Lake Plain, Eastern Corn Belt Plains,
and Interior Plateau.
34
-------
VU31d03103
VtiBidld
Q-
Ol
IO
10 I-
f— O>
01 O
at L
O.
Vt
^
•-
o
0)
^
QC
Q.
•»-
f-
o
t.
y
•r-
z
£
C
U
f-
•o
0*
U
>1
0.
0
k.
^
^
X
^
&
£
U
o.
0
.^
f
Q.
E
t/>
a
c
t.
o
U
Vt
a
^
L.
O
O
e
0>
JC
U
*—
3
O.
ID
E
c
o
c
ftj
^1
t/>
E
9
t.
IP
U
>
to
E
c
o
c
0)
to
BL
^
c
o
VI
«
r
L
VH31dlO -)j
•r
VB3idOH3H3Hd3 |_i
VU31dO1V03W
VBSidOHOIBl
VB idlNSH
£ 1 ! • j r-
~ ' L
£ H
I — T
1 — v
O ^^
« »C
*
n »
-
X 0.
IO f>
L. 3
o t-
ie o>
U JC
•r- 4->
•o
C >.
U 01
£ «n o
C Ol
IB ^_ ,_
Ol
« U ««
« *» 10
10 X
n = *
U
Z - o>
2 «-« =
e> 01
• gr
- ^-s
10
^
n
(SI
NtVld 3MV1 31U3/NObnH
SNIVId J.13B NUOD NU31SV3
35
-------
WATER CHEMISTRY
The results 1n this water chemistry section are grouped Into
three categories: ionic strength (conductivity, alkalinity,
total hardness, calcium, and magnesium); nutrients (nitrate,
nitrite, ammonia, Kjeldahl nitrogen, total phosphorus, and total
organic carbon); and selected metals (copper, iron, and lead).
Patterns 1n water chemistry are displayed in four ways.
First, we selected the median value over the 16-month sampling
interval to represent a site and used those site medians to
develop regional box plots. The median value on each box plot is,
therefore, a median of medians.
A second way presents selected nutrients at selected sites
in the Eastern Corn Belt Plains by graphs of changes through the
16-month sampling interval. These time series graphs illustrate
the range in values found in one region. Sites exhibiting
patterns of low nutrient concentration occur in watersheds that
may differ naturally or anthropogem'cally from those watersheds
where high levels occur. We will develop this theme further in
the interpretation section. We chose nutrients for these
displays because of national efforts to minimize nutrients in
water bodies.
A third method displays data by mapping the spatial distrib-
ution of water chemistry values. This provides a synoptic pic-
ture of the spatial similarity of site chemistry; it indicates
how well ecoregional patterns correspond with water chemistry
patterns. If other spatial patterns stand out, they may suggest
causal mechanisms controlling the pattern.
Finally, we used principal components analysis (PCA) to
Identify components that would express patterns in nutrient
richness and ionic strength. PCA is a technique that extracts
from correlated data one or a few variables (principal
components) that can account for most of the variability in a set
of multivariate correlated data (SAS 1985). We examined the
correspondence between spatial patterns in these variables and
the ecoregions.
Box Plots of Water Chemistry
Ionic Strength
Streams in the'Huron/Erie Lake Plain and Eastern Corn Belt
Plains tend to exhibit higher concentrations of chemical
constituents in this category than do streams in the other three
regions (Figure 12). Although there is a considerable overlap in
values among adjoining regions, the same general pattern occurs
among all the variables.
36
-------
1200
1000-
u
5
800
O 600'
O
O 400
200
600
500
01 400
£
CO
CO 300
HI
Q
OC 200
I
100
60
50
I 40
Z
i 30
LLJ
O 20
10
HELP ECBP EOLP IP
ECOREGION
WAP
HELP ECBP EOLP IP
ECOREGION
WAP
HELP ECBP EOLP IP
ECOREGION
WAP
r- 300-1
• 150
'°°
so
O
t- 0
140
120
= 100-
^
Ob
£ eo
s
O 40
20
0
~75
Relative
Sample Size
HELP ECBP EOLP IP
ECOREGION
WAP
HELP ECBP EOLP IP
ECOREGION
WAP
Figure 12. Selected 1on1c strength measures of water quality by
region. Boxplots of site medians of 16 monthly samples.
37
-------
Nutrients
Concentrations and variability of nitrogenous compounds tend
to be highest in the Huron/Erie Lake Plain and lowest in the
Western Allegheny Plateau (Figure 13). In some cases, extremely
high values occur, e.g. nitrate values at some sites in the
Huron/Erie Lake Plain approach the 10 mg/1 public water supply
standard. In the same region, some values are quite low,
suggesting substantially different kinds of management practices
or natural differences in the respective watersheds.
Total phosphorus exhibits a similar pattern to that of the
nitrogen compounds. Total phosphorus values at most sites in the
Western Allegheny Plateau were at or below analytically
detectable limits of the method used. The pattern for total
organic carbon differed from that seen for the other nutrient
variables such that values in the Erie/Ontario Lake Plain were
similar to those of the Huron/Erie Lake Plain; lowest values
occurred in the Western Allegheny Plateau.
Iron, Copper, and Lead
A similar pattern to that seen for nutrients is also seen
for these metals (Figure 14). Highest values and greatest vari-
ability occur in the Huron/Erie Lake Plain and the Eastern Corn
Belt Plains; lowest values are in the Western Allegheny Plateau.
A few sites are characterized by a consistent pattern of high
values throughout the sampling season; medians are shown in Table
6. There might be unusual watershed activity or local perturb-
ations associated with these sites relative to other sites.
Temporal Patterns in Water Chemistry Variables
We selected several sites in the Eastern Corn Belt Plains to
illustrate the seasonal patterns in the nutrient concentration in
these streams and to indicate the extremes that can occur within
ecoregions, even at minimally impacted sites. It is likely that
there are identifiable management activities that differ among
these watersheds. An understanding of these activities might be
a useful guide for improving chemical water quality, but not
necessarily biotic integrity (Karr et al. 1985).
Large differences between total phosphorus and Kjeldahl
nitrogen occur among these watersheds (Figure 15a, 15b), but no
consistent seasonal pattern is evident in all. Also, within
sites, large differences 1n concentration occur over relatively
short periods of time; however, these are not so large as to mask
the clear differences among high and low sites. The variability
within sites 1s not unexpected as nutrient levels often fluctuate
with changes 1n stream flow associated with storm events.
In contrast to these measures, nitrate seems to display a
definite seasonal pattern, with the exception of site WGT 61A;
while relatively low values occur in the late summer-early fall,
38
-------
0.5-1
0.4.
0.3-
W
tc
o
Q.
W
O 0.2
O.
0.1
a>
8-1
6-
UJ
O
s «•
ai
i-
(T
uflu
klr-1 LaJ 5 U-
HELP ECBP EOLP IP WAP
ECOREGION
HELP ECBP EOLP IP WAP
ECOREGION
2.0-1
o>
1 1.5
UJ
O
O
E 1.0
X
0.5
0.16-1
£0.12-
ai
O
§ O.OB-
t-
z
iu
E 0.04
z
HELP ECBP EOLP IP WAP
ECOREGION
HELP ECBP EOLP IP WAP
ECOREGION
0.35-
E 0.30-
^ 0.25-
Z
UJ
g 0.20-
a.
% 0.15-
•
<
S
< 0.05.
n.
. -i-
HELP ECBP EOLP IP
ECOREGION
WAP
12-
10-
O
m
E
o
o
z
o
(E
O
_)
<
o
HELP ECBP EOLP IP
ECOREGION
WAP
Figure 13. Selected nutrient measures of water quality by
region. Boxplots of site medians of 16 monthly samples.
39
-------
10-
O)
3 6
^^
or
LU
Q.
a.
O 4
O *
2-
A
HELP ECBP EOLP IP WAP
ECOREGION
ID-,
8-
O
2-
JL.
HELP ECBP EOLP IP WAP
ECOREGION
o>
6-
7-
6-
5-
4-
II
HELP ECBP EOLP IP WAP
ECOREGION
Figure 14. Selected metal concentrations by region. Boxplots of
site medians of 16 monthly samples (12 for copper).
40
-------
relatively high values characterize the remainder of the year
(Figure 15c). For ammonia the pattern shows that sites with low
concentrations are characterized by values consistently at the
detection limits of the analytical procedures used, while other
sites display large increases from this background level with
apparently no consistent seasonal pattern (Figure 15d).
Although the sites chosen to display temporal patterns
represent the extremes for one region, the general patterns hold
true for sites in other regions. In the Huron/Erie Lake Plain
patterns and extremes are similar to those just described, while
for the other three regions differences among extremes are not so
great. This is also reflected in the box plots of median values
over the entire 16-month interval (Figure 13).
Table 6. Sites characterized by unusually high
metal concentrations (median values).
Iron Copper Lead
Site (SRP Code) mg/1 ug/1 ug/1
Huron/Erie Lake Plain
Little Auglaize (NGT7BB) 2.7 6.5
Brush Creek (NGT75) 2.9
Black Creek (NGT93) 7.0
Powell Creek (NMT67) 6.5
Lost Creek (NMT68B) 7.6
Eastern Corn Belt Plains
Mill Creek (WGT117) 9.1
Slate Creek (WGT62B) 2.6 9.4 10.0
Eagle Creek (WMG49B) 7.0
Mad River (WMG116) 10.0
41
-------
•o
o
2
_l CO
32
o
«
1.
fl) r— -tJ
•0 ••-
« •—
O "» 0>
C C >
o •»- o>
O <0 r-
C 41
O) +-> CT>
•^ i— O
J- 41 t-
c -o
C !-•- •
I-
0) ••->••->.-
4-> U) ^ O)
4->
10 UJ -~ Ol
0. X) f—
O>^-
r- JT +J ^>
I. • C
O C M O
o.-^^ e
E 0> E
O) «l > <
t- Q> 01
1-1 T> ».
O) O •
O) ** -C V*
t- U Q-r—
3 oi ui a>
O)r— O >
f aj x: O)
LL. in 0.1—
Cl/Buu) SfldOHdSOHd IVlOi
o co to * e>j
(I/BUJ) 31VU1IN
42
-------
Spatial Patterns In Selected Water Chemistry Variables
Spatial patterns in water chemistry characteristics can be
displayed through color- or shade-coded dot maps. These maps are
constructed by dividing the range of site values (medians in this
case) into Intervals that split the sites Into relatively evenly
sized groups. The number of intervals depends on the size of the
data set and the level of resolution desired. We found four to
nine intervals provided good resolution.
Representative of nutrient patterns, the total phosphorus map
shows the consistently low values in streams throughout the
Western Allegheny Plateau (with one exception) and the generally
high values in the Huron/Erie Lake Plain (Figure 16). It also
illustrates the high variability among sites in the Eastern Corn
Belt Plains and the incidence of sites with low values scattered
throughout this ecoregion. Representative of 1on1c strength
patterns, the conductivity map indicates lowest values in the
streams of the Western Allegheny Plateau and the Interior Plateau
and highest values in the Huron/Erie Lake Plain and the Eastern
Corn Belt Plains (Figure 17). The map also shows some spatial
segregation of values 1n regions with higher variability, as 1s
seen in the patterns in the Erie/Ontario Lake Plain. Here,
lowest values occur at sites in the northeast and higher values
are found in central and southwestern areas of the ecoregion.
Principal Components Analysis (PCA)
The minimum correlation among the ionic strength variables
is between conductivity and alkalinity (r = 0.73); for all other
pairs of variables, r 1s 0.80 or higher. As a result, PCA axis I
(PCA I) accounts for a high proportion (90%) of the total
variability in this multivariate data set; each variable is
almost equally loaded on Ionic strength PCA I.
The correlations among the nutrient richness variables are
not as high as those for ionic strength because nutrients fluct-
uate more independently of each other. This lower correlation is
also reflected in the lower fraction of multivariate variability
accounted for by the principal axes 1n the nutrient analysis: PCA
I accounts for 64% of the variability; adding PCA II increased
that to 78% of the variability. The lower correlations are
reflected in the PCA I and PCA II loadings; not all variables are
loaded as similarly on PCA I as they are for the ionic strength
variables on the ionic strength PCA I. Total phosphorus,
Kjeldahl nitrogen, and nitrite are equally and most heavily
loaded on PCA I, while nitrate, ammonia, and total organic carbon
are also equally but less heavily loaded on PCA I. PCA II
reflects a dominance of total organic carbon with a lesser
influence of nitrates. PCA I seems to express the overall
quality of nutrient richness of the stream water, I.e. the
combination of nitrogen, phosphorus, and carbon compounds.
43
-------
I O
K
rfftk. ^^
: .•/£ °V °8 e. _a
•> o • • ^. °^^ •
TOTAL
PHOSPHORUS
0.050
-------
CONDUCTIVITY
350< 0^500
500<©5600
600< •<700
Figure 17. Spatial patterns of conductivity 1n Ohio streams.
Values are site medians of 16 monthly samples.
45
-------
The results of the two PCAs are summarized as a graph of
nutrient richness PCA I vs. ionic strength PCA I (Figure 18).
This graph displays a clear relationship between the water
quality of the sites and their ecoregions. Sites in the Western
Allegheny Plateau are concentrated in the area of low nutrient
richness and low to intermediate ionic strength. Sites in the
Interior Plateau group closely with intermediate values for ionic
strength and nutrient richness while sites for the Erie/Ontario
Lake Plain are slightly more scattered and encompass those of the
Interior Plateau. Sites in the Huron/Erie Lake Plain and the
Eastern Corn Belt Plains are similar along the ionic strength
axis but separate somewhat along the nutrient richness axis.
5-
4-
3 -
< 2 -
0
a.
T 1 -
IIC STRENGl
J. o
1 1
£.
O
-2-
-3-
-4-
X *N
° N °
\ Rio o 1 ^^— — — -.^.^
/~V °s*-QoS^ /* /
/ 1 • » *v O ^*S
l'a \/ >TT^ xx? . ^.--^
ID ^^ S \
[— ^^fa ^^ «
1 °D D I * T '
1 D ,D D /
1 D Dl D / • HURON/ERIE LAKE PLAIN
\ "^B } *t. / ° EASTERN CORN BELT PLAINS
» l~5s * ERIE/ONTARIO LAKE PLAIN
\XD / A INTERIOR PLATEAU
""" D WESTERN ALLEGHENY PLATEAU
i i i i i i i i i i i
-2
-1
123456
NUTRIENT RICHNESS [PCA 1)
Figure 18. Regional patterns in nutrient richness and ionic
strength variables Indicated by principal component axis I scores
for each. Areas enclosed indicate hypothesized attainable water
quality for each region.
46
-------
Groupings that define water quality typical of each eco-
region are Indicated by the enclosed areas on Figure 18. Each
region can be distinguished from the others based on the combin-
ation of nutrient richness and 1on1c strength. These areas are
delineated subjectively to Indicate the general regional
dlstlnctlveness of water quality. While not all sites fit this
pattern uniformly, especially 1n the Eastern Corn Belt Plains,
the ecoregional patterns are evident. The Huron/Erie Lake Plain
and the Western Allegheny Plateau differ considerably from each
other with the other three regions having Intermediate water
quality indicative of their transitional nature.
PHYSICAL HABITAT
The physical habitat measures were not collected
consistently enough to allow a thorough analysis. Here, we
present a narrative summary of the trends discernible 1n each
region and a brief overview of cover and substrate patterns.
Huron/Erie Lake Plain
About one third of the streams sampled in this ecoregion
were intermittent. Currents were never fast, distributed among
none, Interstitial and moderate. Only one site had clear water;
the majority were turbid and some were stained. Most reaches
were > 50% pools and development was poor more often than good.
Pools had primarily sandy/silty bottoms (70%). About 70% of the
riffle substrates were gravel and sand, most of the remainder
were silt and clay. Several sites had cobble substrate.
About 75% of the sites had been channelized and were in
various states of recovery. Bank slopes were generally moderate
to steep; the majority were eroding moderately to severely.
There was 10-60% instream cover at most sites, consisting mainly
of logs and trees. Canopy openness varied widely (10-100%) but
within this ecoregion canopy tended to be less open than in other
ecoregions, except the Eastern Corn Belt Plains. Emergent
vegetation was observed at half the sites. Where buffer
vegetation occurred, it generally consisted of 3-10 m of shrubs
and immature beech/oak/maple growth. Land use was agriculture at
every site (mostly rowcrops with some pasture). About half the
sites also had residential areas, forest and open vegetation.
Eastern Corn Belt Plains
All but one of the streams sampled in this ecoregion had
continuous flows, generally with slow to moderate current. About
half of the sites were turbid with silt and diatoms the most
frequent sources. About 70% of the reaches were 1n pools;
development was good more often than poor. Pools had very
diverse substrates, ranging from boulder to silt and clay at each
47
-------
site. About 60% of the riffle substrates were composed of cobble
and gravel, the remainder were generally divided among larger and
smaller particle sizes.
Bank slopes varied from gentle to steep and generally had
little or no evidence of erosion. Only a few sites had been
channelized. Instream cover was comprised equally of undercut
banks, rocks and boulders, and logs and trees. There was a wide
range of canopy cover, but generally it was less open than in
other ecoregions, except the Huron/Erie Lake Plain. Emergent
vegetation was noted at about half the sites. Buffers of 15 to
30 m were generally present, often consisting of mature trees or
shrubs and grasses. Every site had agricultural land use
(rowcrops or pasture). Approximately two thirds of the sites
also had some forest or open vegetation.
Erie/Ontario Lake Plain
Streams in this ecoregion were generally continuous (several
were intermittent). Most had moderate flows, a few had slow- or
no-flows. About half the sites were turbid. Generally 10-40* of
the reaches were in riffles with development good at half of the
sites and poor at the remainder. Pool substrates ranged from
silt and clay to boulder at every site. Riffle substrates were
about 90% cobble and gravel.
Few sites had evidence of channelization and banks varied
from gently sloping to steep and stable. A quarter of the sites
had moderate to severe erosion. Sites in this ecoregion tended
to have the least instream cover (10-40%). The range of canopy
openness was wide but generally more open than in the Huron/Erie
Lake Plain and Eastern Corn Belt Plains ecoregions. A fair
amount of emergent vegetation, principally grasses, occurred at
most sites. The predominant land use was forest with some open
vegetation, and agriculture at several sites.
Interior Plateau
The streams sampled in this ecoregion had continuous flows
with moderate currents. The water was generally turbid. There
were equal numbers of reaches that were all pool and reaches that
were 20% riffle; development was good. Pool substrates were fine
gravel and sand. Where riffles were present their substrates
were mostly cobble mixed with some boulders and coarse gravel.
Stream channels were natural and banks varied from gradual
to steep. Generally there was little to moderate erosion. The
instream cover was 20-80% and unlike other ecoregions 1t
consisted mostly of rocks and boulders. There was a wide range
of canopy openness, but the canopy was generally more open than
in the Huron/Erie Lake Plain and Eastern Corn Belt Plains
ecoregions. Emergent waterwillow occurred at a few sites.
48
-------
Buffers strips of mature trees, 10 to 30 m wide, were found at
most sites. All sites were agricultural (mixed pasture and
rowcrops), and most had some forest.
Western Allegheny Plateau
Continuous flows were characteristic of streams in this
ecoregion. Currents were moderate in about two thirds of the
streams, the remainder divided equally into fast and slow
classes. Clarity was highly variable. Generally 2-40% of the
reaches were in riffle and the majority of sites had good
development, the rest divided equally between excellent and poor.
Pools had diverse substrates with sand and fine gravel most
dominant. Riffle substrates were about 75% cobble and gravel;
the remaining sites split between larger and smaller particle
sizes.
The channels of all sites were natural. Most banks had
moderate to steep slopes. About 50% of the sites had moderate to
severe erosion. The site's in this ecoregion had the greatest
instream cover (40-70%). Canopy openness was variable, but
generally more open than in the Huron/Erie Lake Plain and Western
Corn Belt Plains ecoregions. There was noticeable algae on rocks
at many sites and emergent waterwillow at others. Land use at
most sites was half agricultural (50% pasture, 50% rowcrops) and
half forest (with some open vegetation).
Instream Cover and Substrate Composition
Total instream cover (Figure 19) was generally greater in
the Western Allegheny Plateau (median = 57%) than in the other
regions (medians: 22-33%). The portion of cover provided by
undercut banks did not vary among regions (medians: 1-7%). Cover
provided by rocks and boulders was greatest in the Interior
Plateau, median = 20% (2-10% in the others). Cover by logs and
trees was greatest in the Huron/Erie Lake Plain (median = 20%)
and the Western Allegheny Plateau (median = 15%).
Pool substrate profiles (Figure 20) are quite similar among
regions except for the Huron/Erie Lake Plain which generally has
smaller particle size substrates. Riffle substrate profiles show
the same overall pattern of small particle sizes in the north-
western part of the state and larger sizes in the south.
49
-------
1 total % of instream cover
2 undercut bank
3 rocks and boulders
4 logs and trees
100-j 'OP-
2 I
< 80- - 5 80-
t , . o-
g _ 5
< 60-
_i
UJ
E
UJ 40-
^
Z
o
E
§ 20-
0
1
e
"•"^
m 60*
E
8 <0-
1 z
E
T T . S 20-
PI CO
1 1 ^
UJ
1
C
i r
f
J
^ It
n
. u
L'JL'J
1234
234
100-
z
< 80-
_J
CL
UJ
^ 60-
O
E 40-
2
i 2°'
UJ
E
UI
0
1100-
r
I
1
80-
3
. <
1 *
1 5 60-
"
I 2
-------
aivbisans
•CD-
•O-
•D-
aivkusens iood
-T
v>
O
CL
E
O
u
O)
u
I.
Ol
NIVId 3MV1 OIUV1NO/3IU3
dOIU3lNI
nV31V1d AN3HO3T1V NU31S3M
ID
I-
3ivuisans
•D 01
M o> €1 O>
aivdisans iood
k. •— ifl »- W ^ ** ^ *^
J,a O)
a> •—
o t.
•O
O VI
o
s. ex
3
01 t.
•t- O
00000 00000
* « «
-------
SECTION 6
INTERPRETATION
AQUATIC LIFE USE DESIGNATIONS AND ATTAINMENT
The objective of the Clean Water Act (CWA, PL 95-217) is to
restore and maintain the physical, chemical, and biological
integrity of the Nation's waters. State and Federal water
quality standards have been established to help meet that
objective (Federal Register 1983). These standards serve a dual
function: they establish water quality goals and serve as a
regulatory basis for establishing treatment controls. Water
quality standards consist of the combination of designated uses
for water bodies and water quality criteria that, if met,
presumptively protect those uses. States are required to specify
appropriate uses to be achieved and protected, as well as to
adopt criteria to protect those uses. The criteria are numerical
values or may be narrative where numerical criteria cannot be
established or need to be supplemented.
States have traditionally designated uses in a qualitative
way. This is particularly evident for uses that pertain to
aquatic life. In some cases, "aquatic life" is the designated
use. In others, the aquatic life use might be identified as warm
or cold water fishery, or salmonid passage. What is generally
lacking are specific or quantitative measures that characterize
the use and provide a test of whether the water body actually
supports that use. Therefore, although the Act requires
assurance of protection and propagation of a balanced indigenous
population of fish, shellfish, and wildlife, states rarely
specify quantitative measures to meet this objective. There are
a few exceptions. For example, OEPA (19845) specifies four
aquatic life categories, with qualitative and quantitative
statements about fish species composition, species richness and
diversity, and numerical abundance of fish and macroinvertebrates
used as measures that delineate those uses (Table 7). Wisconsin
has developed a similar system that establishes five categories
of aquatic Hfe use, and establishes quantitative assessments of
various ecological characteristics that determine whether a
particular use category has been met.
One of the difficulties in determining whether uses are
being attained and therefore whether criteria are appropriate is
the lack of reference data describing what to expect.
Quantitative expression of aquatic life uses is needed to serve
as a benchmark for site specific or regional assessments. Partly
this has not been done because aquatic ecosystems display such
variability and because there has been no conceptual framework
52
-------
4J
C
i
•H
4J
JJ
IB
IB
in
c
o
•H
4-1
IB
I
M
01
4J —
•H <
ki a
*••
"« 4J
U O
O ki
•H 0)
O 4J
•H IB
A *
< C
p4 IB
U 01
O U
•H
X«U
O O
r^
0)
l-l
A
iB
H
I
i
I
I
i
I
13
8
S
H
hi
Ii)
E
I
bl
g
1
1
1
1
|
1
1
1
1
I
1
1
, 3
^
S
ff
H
UJ
w
E
i
i
i
i
i
i
i
i
£ «-*
Is
£ *-l
U
M
= fH
•H VI
10 VI
* r~t
O
t_l
£ M
1> VJ I
O in 5
o IB 5
o
^
«
C l-l
o —
•H V) pC
4J V) 5
a 10 u
o .-< _
o u
X
bl
£
C
O >•
-tn u
4-1 V) O
W M O>
3 id 01
« u
w
VI
01
u
01
o.
VI
•s
u
s.
X
01 kl
c
U 01
M n
0 0
VI
01
U 5
01 0
Of*
in
•o ^
V 01
u
01 ki 01
Oi O O
X C
oi tJ id
c-o
01 01 C
E « 3
O Q A
I/) 10 10
tu
O
c
O M
•H 01
id u
-> 01
o a
o «
V)
VI-O
« 01
iH O
3 Oi
« X
3 0)
14
O
tH
10
C
o
•H rH
4J n
O. 3
0) VI
O 3
X C
U 3
01
0>
10
V) <-l
•H e
U 01
01 VI
a w
I/) 10
t-i
U
c
01
n
3
E
01
ki 0)
o u
c
u id
CT>
01 C
W 3
£> J3
< 10
C
0>
in
V
a.
4J
c
10
•o
c
^
01
•ft in
U 01
VI U
1/1 in
IM
o
in
i
"I
3 —
O
*J V
o in
in r~
C 1 1
C mo
•-« ^ r-
o
U A V
~~ f^
in in
r-' r^
S A V
>. —
t-i in
rH •
id »
c
O 1
D.
Oi X
O t» —
X-H
^< 0)
ft f4
in mQ
u OX
> ||
O O--I
m ^>
in
01
•H
U
W
U 01
C I-*
id «
u c
01 .1
o o
l-i -a
o>
c
••-1
c
M C
01 •<-!
•H O*
uS
M tP *J
C «
4J •-< C
C W -H
10 10 E
U. 0) O
0> ki T)
«H O
O C O
o
1 l/l
01 - 01
u -OH
o >. o> o
01 kl M 0>
u 01 01 O.
X D' in
O* in c
C-H iO T3
-H 0> W
*J *4 14 ki 0)
84J id X ki
id a: u D.
in u?
kl
01
•o
o
c
kl
s
o
in •
«-H in
" ?
4J r;
V) 4J
i i
u> O*
c
kl -H
O ~H
a
in E
J^
•O n
c
10 ki
0>
kl X
01 4-1
E O
ft) 14
•O O
«4J
*J
W >«
3 X<
E «
«^
01 >.
rH Id
A E
id
•H 01
« D<
t> C
id id
o>
kl *.
•0 VI
0)
10 i-H
u a
id • E
•O vi id
M VI
•M «
-H r-l DI
— O C
•«• ki X
Id VI
•O ^l-H
C 3MJ
id u O
**» IJ 4J
» ki O
rM id 01
- O.-H
-1 01
W HI M
01 X id
O C
O> H C
01 O
4J *O
id oi ^«
U kl^H
Id -H
C ki
in H E
C VI -H
O 01 ki
•H -O O.
4J
•ft fl] TJ
•a 2 oi
c w
O O IB
O 4J A
id XI
53
-------
within which to partition that variability. A regional
classification system, such as the one developed at
ERL-Corvallis, can provide the needed framework for States to
establish attainable aquatic life uses and to protect those uses.
The design of this project called for selecting stream sites
that were in least-disturbed watersheds representative of the
different ecoregions of Ohio. Therefore, the data we obtained
should provide a good picture of reasonably attainable conditions
in streams of similar size throughout Ohio. Box plots of the
data by ecoregion indicate the central tendencies and range of
conditions found within each region. It might be useful to
consider as goals the range of conditions attained by the best
50% of the reference sites. For example, the IBI values in the
top 50% of the range for an ecoregion might be those specified as
attainable (see Figure 5).
The figures presented throughout the results section are
useful guides as a comparative data base. They can be contin-
ually expanded as monitoring programs obtain more data from
least-disturbed sites. Some of these data can come from
comprehensive water quality surveys when data are collected from
unimpacted sites upstream of sites suspected to be impacted.
This regional framework allows comparison of the quality of any
site relative to the least impacted ones. It could show,
relative to the reference situations, the effectiveness of
treatment actions for improving stream conditions. Reference
sites could serve in evaluating the quality of upstream control
and downstream recovery sites (Hughes 1985; Hughes et al. 1986).
We assessed fish species richness in Little Yellow and
Yellow Creeks as an example of how the regional reference data
can be combined with the standard upstream/downstream analyses to
display the relative magnitude of impacts. The Yellow Creek
Basin is a small watershed in the Huron/Erie Lake Plain. The
headwater of Little Yellow Creek is the effluent from the Leipsic
Wastewater Treatment Plant (WWTP). Little Yellow Creek runs
about 6.4 mi. before joining Yellow Creek, a somewhat larger
stream. Land use in the Yellow Creek Basin is primarily row crop
agriculture, typical of this ecoregion. Both streams have been
extensively channelized, and riparian vegetation 1s sparse,
although some areas have relatively good riparian cover. A
comprehensive water quality survey on those streams was conducted
in 1981, including fish sampling at six sites in August and
September (Ohio EPA, 1982).
The Yellow Creek fish species richness data are plotted with
data from the Huron/Erie Lake Plain ecoregion (Figure 21). Note
that the watershed size of Little Yellow Creek is smaller than
any of the reference sites, while watershed sizes for Yellow
Creek are within the range of the reference sites. Extrapolation
of the species richness-watershed area regression suggests that
54
-------
28-,
24 -
CO
CO
UJ
z
X
16-
DC
co 12
UJ
O
UJ
0.
CO
8-
4 -
0-
©
A"
Regional Reference Sites
Line of Best Fit
Yellow Creek
Site Upstream from
Impacted Tributary
GT
©
©
Confluence of
Little Yellow and
Yellow Creeks
A Little Yellow Creek
Sampling Sites
Q Yellow Creek
Sampling Sites
© Regional Reference
Sites
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
Iog10 WATERSHED AREA (km2)
2.6
Figure 21. Comparison of fish species richness 1n an Impacted
stream with data from the regional reference sites 1n the
Huron/Erie Lake Plain.
species richness of Little Yellow Creek falls within the range of
values expected for watersheds of this size 1n this ecoregion,
but might be at the low end. The species richness at the
uppermost site on Yellow Creek is also within the expected range
for sites in this region. However, the lower two sites display
definite Impact. The key point here is not only have species
richness values decreased (not major decreases of themselves) but
that they are noticeably lower than the minimally-impacted
streams of the same size in this region.
This effect 1s probably due to the combined effects of
wastewater originating from the Leipsic WWTP and degraded
physical habitat (Ohio EPA, 1982). Physical habitat conditions
at the downstream Yellow Creek sites are more degraded than at
the upstream sites, and the intermediate site Is of poorest
quality, reflected in the lowest species richness relative to
regional expectations. Attainment of expected species richness
in this basin probably requires the restoration of more natural
physical habitat conditions, as well as continued or Increased
treatment of Leipsic wastewater.
55
-------
Although we used species richness 1n this example, any of
several measures of environmental health or a combination could
be used to show how particular sites or basins match expect-
ations. Note that It would have been Inappropriate to expect
species richness 1n the Yellow Creek Basin to have attained
values characteristic of the Western Allegheny Plateau, at least
under current land management practices (compare with Figure 4).
Ohio's current aquatic life use designations, although
relatively detailed, do not Incorporate regional differences 1n
stream potentials, except through the professional judgement of
the technical staff of OEPA. Data collected during this project
could be used to quantify regional differences. The largest
differences In fish communities occur between the Huron/Erie Lake
Plain and the Western Allegheny Plateau. These regions also
differ most geographically: flat plains vs. rolling hills; row
crop agriculture vs. mixed forest and cropland. Smaller
differences occur among the other regions. Therefore, 1t might
be reasonable to Identify three sets of reference conditions for
fish that reflect expectations 1n the Huron/Erie Lake Plain, the
Western Allegheny Plateau, and In the remaining transitional
regions. It 1s worth noting that several of these regions extend
Into bordering states, so data from least Impacted sites In those
states would be useful additions to the current reference set.
We present an example In Table 8 of how quantitative
criteria might be used within a regional framework to determine
whether existing use designations are likely to be met. The form
is consistent with Ohio's current practice of Identifying
Table 8. Attainable fish assemblage attributes for small streams in Ohio.
Values are based on 50th percentiles of this study and subject to
further refinement by Ohio EPA.
Eastern Erie/ western
Huron/Erie Corn Belt Ontario Interior Allegheny
Lake Plain Plains Lake Plain Plateau Plateau
Species Richness*
Number of Intolerant
10-20
1-2
13-30
2-10
13-30
2-10
13-30
2-10
16-35
4-14
Species*
Percent Intolerant > 5 > 25 > 30 > 35 > 25
Individuals
Index of Biotic > 32 > 44 > 44 > 42 > 49
Integrity
Index of Well-Being > 8.2 > 9.3 > 9.0 > 9.0 > 9.7
aSite specific values must be determined from maximum species richness and maximum
number of intolerant species lines (Figures 4, 8) for watershed area of site.
56
-------
objective criteria for uses, but here the expectations are all
quantitative and specified regionally. This regional perspective
does not preclude Ohio EPA from maintaining Its current set of
use designations and criteria; they simply provide a set of
reasonable expectations (which may be further refined as OEPA
gathers more data) for streams in the various regions of the
state.
ATTAINABLE WATER QUALITY
Identifying attainable water quality as a function of
ecoregions is a straightforward way to specify realistic goals.
Recall that we selected sites in minimally impacted watersheds in
each ecoregion. For example, in the Huron/Erie Lake Plain, this
minimal impact means row crops (corn and soybeans), a fringe
riparian forest at best, tile drainage, and heavy fertilizer
applications. It is apparent that the water quality in this
region is both enriched and of relatively high ionic strength (a
function of natural land type and land use) compared with other
regions in Ohio. Thus, attainable water quality, as observed at
these minimally impacted sites, 1s one way to express realistic
expectations for this region (without major changes in land
management). It certainly represents what has been attained
under existing conditions.
Knowledge of regionally attainable water quality can be used
to the advantage of water quality managers and the public.
Assessing streams affected directly by point sources, feedlots,
mining, or other harmful activities compared to regional water
quality goals provides a way to demonstrate the degree of impact.
This regional assessment approach does not force unrealistic
requirements in areas where the water quality standards are
unlikely to be met. It is a way to place water quality goals
into an environmental perspective, eliminating unrealistic
expectations.
As described for the biological component, attainable water
quality can be summarized as univariate box plots or maps that
display values from regional reference sites. Regionally
attainable water quality can be defined by selecting a range of
values representative of the highest quality achieved in that
region. The water quality achieved by 50% of the sites with
highest overall quality might be designated as the goal. For
example, attainable phosphorus levels in the Western Allegheny
Plateau would be < 0.05 mg/1, but in the Huron/Erie Lake Plain a
more realistic goal would be < 0.15 mg/1 (see Figure 13). Graphs
that display these goals could then be compared with data from
other sites to show the degree and extent to which goals are met.
Examination of the variability within a particular region
can provide insight into land management practices that would
57
-------
minimize water quality degradation. For example, the
considerable variation among sites in the Eastern Corn Belt
Plains might be used to advantage. These watersheds occur in
areas of the same land type; however, there might be some
localized anomalies that explain high or low values. For
example, although the soil in the region is generally of one
type, sites with particularly high water quality values might be
located in watersheds with a soil not mapped on the small-scale
maps used to delineate regions. A more likely scenario is that
the watersheds characterized by the high values are managed
differently from those with the low values. There might be less
riparian forest along the streams characterized by the high
values, or an abundance of feedlots in the watershed, or tillage
and fertilizer application patterns might differ. Where great
within-region variation in water quality is observed, explanation
of the variation should lead to watershed management procedures
that minimize water quality problems originating from diffuse
sources. This does not suggest drastic changes in land use
patterns, but merely educates the manager about how current use
might be modified to result in overall water quality benefits.
Finally, a regional framework and reference data also can be
used to help monitor the quality of water statewide by
facilitating the design of efficient sampling strategies. Areas
known to be similar to each other can be represented by samples
drawn from relatively few representative sites, whereas more
variable areas will require greater sampling effort. Unique
situations should be examined individually. Moreover, knowledge
of regionally attainable water quality gives an Important
perspective for applying what is learned from site-specific
monitoring programs. Large negative deviations from regional
reference data suggest greater potential for significant
Improvement. Sites with minor deviations probably have achieved
what is reasonably attainable and further expenditure would
result in minimal benefits. Large positive deviations suggest
streams worthy of special protection.
58
-------
LITERATURE CITED
Anderson, J. R. 1970. Major land uses. Map (scale 1:7,500,000).
Plates 157-159. ^n: The National Atlas of the United
States of American. U.S. Geological Survey. U.S.
Government Printing Office. Washington, D.C.
Austin, M. E. 1965, revised 1972. Land Resource Regions and Major
Land Resource Areas of the United States. Agriculture
Handbook 296. USDA Soil Conservation Service. Washington,
D.C.
Bailey, R. G. 1976. Ecoreglons of the United States. Map
(scale 1:7,500,000). U.S. Department of Agriculture, Forest
Service. Intermountaln Region. Ogden, Utah.
1983. Delineation of ecosystem regions. Environ.
Manage. 7:365-373.
Becker, G. C. 1983. Fishes of Wisconsin. The University of
Wisconsin Press, Madison. 1052 pp.
Carlander, K. D. 1969. Handbook of Freshwater Fishery Biology.
Volume I. The Iowa State University Press, Ames. 752 pp.
1977. Handbook of Freshwater Fishery Biology.
voiume II. The Iowa State University Press, Ames. 431 pp.
Fausch, K. D., J. R. Karr, and P. R. Yant. 1984. Regional
application of an index of biotic Integrity based on stream
fish communities. Trans. Am. Fish. Soc. 113:39-55.
Federal Register. 1983. Water Quality Standards Regulation.
48:51399-51413.
Gammon, J. R. 1976. The Fish Populations of the Middle 340 km of
the Wabash River. Tech. Report No. 86. Purdue Univ. Water
Resources Research Center. West Lafayette, Indiana. 73 pp.
1980. The use of community parameters derived from
eiectrofishing catches of river fish as Indicators of
environmental quality, p. 335-363 In: Seminar on Water
Quality Tradeoffs. EPA-905/9-80-OD?. U.S.Environmental
Protection Agency. Washington D.C.
Gauch, H. Jr. 1982. Multlvariate Analysis 1n Community Ecology.
Cambridge University Press. New York. 298 pp.
59
-------
Hammond, E. H. 1970. Classes of land-surface form. Map (scale
1:7,500,000). Plates 61-63. In^: The National Atlas of the
United States of America. U.S. Geological Survey. U.S.
Government Printing Office. Washington, D.C.
Hughes, R. M. 1985. Use of watershed characteristics to select
control streams for estimating effects of metal mining
wastes on extensively disturbed streams. Environ. Manage.
9:253-262.
Hughes, R. M., and J. M. Omernik. 1981. A proposed approach to
determine regional patterns in aquatic ecosystems, p. 92-
102. In: N. B. Armantrout (ed.). Acquisition and
Utilization of Aquatic Habitat Inventory Information. Am
Fish. Soc. Bethesda, MD.
1983. An alternative for characterizing stream
size. p. 87-101. In: T. D. Fontaine III and S. M. Bartell
(eds.), Dynamics of~Lotic Ecosystems. Ann Arbor Science,
Ann Arbor, Michigan.
Hughes, R. M., D. P. Larsen, and J. M. Omernik. 1986. Regional
reference sites: A method for assessing stream potentials.
Environ. Manage. 10:629-635.
Hughes, R. M., E. Rexstad, and C. E. Bond. 1987. The relation-
ship of aquatic ecoregions, river basins, and physiographic
provinces to the ichthyogeographic regions of Oregon. Copeia
1987:423-432.
Karr, J. R. 1981. Assessment of biotic integrity using fish
communities. Fisheries 6:21-27.
Karr, J. R., R. C. Heidinger, and E. H. Helmer. 1985. Effects of
chlorine and ammonia from wastewater treatment facilities on
biotic integrity. J. Wat. Poll. Con. Fed. 57:912-915.
Karr, J. R., K. D. Fausch, P. L. Angermeier, P. R. Yant, I. J.
Schlosser. 1986. Assessing biological integrity in running
waters: A method and its rationale. 111. Nat. Hist. Surv.
Spec. Pub. No. 5. Champaign, IL. 28 pp.
Kuchler, A. W. 1970. Potential natural vegetation. Map (scale
1:7,500,000). Plates 89-91. In: The National Atlas of the
United States of America. U. ST Geological Survey. U. S.
Government Printing Office. Washington, D.C.
Larsen, D. P. and D. R. Dudley. 1987. A regional approach for
assessing attainable water quality: Ohio as a case study.
J. Soil Water Cons. (In Press).
60
-------
Larsen, D. P., J. M. Omernik, R. M. Hughes, C. M. Rohm, T. R.
Whittier, A. J. Kinney, A. L. Gallant, and D. R. Dudley.
1986. Correspondence between spatial patterns in fish
assemblages in Ohio streams and aquatic ecoreglons.
Environ. Manage. 10:815-828.
Lee, D. S., C. R. Gilbert, C. H. Hocutt, R. E. Jenkins, D. E.
McAllister, and J. R. Stauffer, Jr. 1980. Atlas of North
American Freshwater Fishes. North Carolina State Museum of
Natural History, Raleigh. 854 pp.
Ohio Environmental Protection Agency. 1982. Comprehensive Water
Quality Report for Yellow Creek and Brush Creek, Maumee
River Basin, Putnam County, Ohio. Ohio Environmental
Protection Agency, Division of Wastewater Pollution Control,
Columbus, OH.
Ohio Environmental Protection Agency. 1984a. Manual of Ohio EPA
Surveillance Methods and Quality Assurance Practices. 3rd
Revision. Columbus, Ohio.
Ohio Environmental Protection Agency. 1984b. Biological Field
Evaluations Group Methods Study and Water Quality Criteria
Justication. (Draft). Columbus Ohio.
Omernik, J. M. 1987. Ecoreglons of the conterminous United
States. Ann. Assoc. Amer. Geogr. 77:118-125.
Omernik, J. M., and A. L. Gallant. 1986. Ecoreglons of the
Pacific Northwest (with map). EPA/600/3-86/033.
U.S.Environmental Protection Agency, Corvallis, OR. 39 pp.
Omernik, J. M., M. A. Shirazi, and R. M. Hughes. 1982. A synoptic
approach for regionalizing aquatic ecosystems, p. 199-218.
In: T. B. Brann, L. 0. House IV, and H. G. Lund (ed.). In-
pTace Resource Inventories: Principles and Practices. Soc.
Am. Forest. Bethesda, MD.
Pflieger, W. L. 1975. The Fishes of Missouri. Missouri Dept.
of Conservation, Jefferson City. 343 pp.
Pollard, J. E., and W. L. Kinney. 1979. Assessment of
Macroinvertebrate Monitoring Techniques 1n an Energy
Development Area. EPA-600/7-79-163. U. S. Environmental
Protection Agency. Las Vegas, NV. 26 pp.
Reckhow, K. H. 1980. Techniques for exploring and presenting
data applied to lake phosphorus concentration. Can. J.
Fish. Aquat. Sci. 37:290-294.
61
-------
Rohm, C. M., J. W. Giese and C. C. Bennett. 1987. Evaluation of
an aquatic ecoregion classification of streams in Arkansas.
J. Fresh. Ecol . 4:127-140.
Rowe, 0. S., and J. W. Sheard. 1981. Ecological land classif-
ication: A survey approach. Environ. Manage. 5:451-464.
SAS Institute Inc. 1985. SAS Users Guide: Statistics. 1985 ed.
SAS Institute, Gary, NC. 956 pp.
Smith, P. W. 1979. The Fishes of Illinois. University of
Illinois Press, Urbana. 314 pp.
Trautman, M. B. 1981. The Fishes of Ohio. Ohio State
University Press, Columbus, Ohio. 782 pp.
U. S. Department of Agriculture Soil Conservation Service. 1981.
Land Resource Regions and Major Land Resource Areas of the
United States. U. S. Department of Agriculture Handbook
296. Washington, D.C. 156 pp. and map.
U. S. Department of Agriculture Soils Surveys of the States of
the North Central Region of the United States. North
Central Regional Publication No. 76, Bulletin 544.
Agricultural experimental station, University of Wisconsin,
Madison. 192 pp. and map.
U. S. Environmental Protection Agency. 1979. Methods for
Chemical Analysis of Water and Wastes. EPA 600/4-79-020.
U. S. Environmental Protection Agency, Cincinnati, OH.
U. S. Geological Survey. 1970. Distribution of principal kinds of
soils: Orders, suborders, and great groups. Map (scale
1:7,500,000). Plates 85-86. In: The National Atlas of the
United States of America. U. ST Geological Survey. U. S.
Printing Office. Washington, D.C.
Warren, C. E. 1979. Toward Classification and Rationale for
Watershed Management and Stream Protection. EPA 600/3-79-
059. U. S. Environmental Protection Agency, Corvallis, OR.
143 pp.
62
-------
Appendix A:
TOLERANCE AND TROPHIC LEVELS OF 99 OHIO FISH SPECIES.
These are the trophic levels and tolerances used for
this study by ERL-C and are not necessarily Ohio EPA's
final designations. The * Indicates the most
Intolerant species as used to calculate IBI scores.
SPECIES
NORTHERN BROOK LAMPREY
OHIO LAMPREY
LEAST BROOK LAMPREY
AMERICAN BROOK LAMPREY
LONGNOSE GAR
GIZZARD SHAD
BROWN TROUT
RAINBOW TROUT
CENTRAL MUDMINNOW
.GRASS PICKEREL
NORTHERN PIKE
MUSKELLUNGE
BIGMOUTH BUFFALO
BLACK BUFFALO
QUILLBACK
RIVER CARPSUCKER
HIGHFIN CARPSUCKER
SILVER REDHORSE
BLACK REDHORSE
GOLDEN REDHORSE
SHORTHEAD REDHORSE
RIVER REDHORSE
NORTHERN HOGSUCKER
TROPHIC LEVEL
OMNIYORE
PISCIVORE
OMNIVORE
OMNIVORE
PISCIVORE
OMNIVORE
PISCIVORE
INSECTIVORE
INSECTIVORE
PISCIVORE
PISCIVORE
PISCIVORE
OMNIVORE
OMNIVORE
OMNIVORE
OMNIVORE
OMNIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
TOLERANCE
INTOLERANT
INTOLERANT
MNTOLERANT
INTOLERANT
MODERATELY TOLERANT
TOLERANT
*INTOLERANT
MNTOLERANT
MODERATELY TOLERANT
INTOLERANT
INTOLERANT
INTOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
INTOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
INTOLERANT
MNTOLERANT
63
-------
SPECIES
TROPHIC LEVEL
TOLERANCE
WHITE SUCKER
SPOTTED SUCKER
CREEK CHUBSUCKER
COMMON CARP
GOLDFISH
GOLDEN SHINER
HORNYHEAD CHUB
RIVER CHUB
BIGEYE CHUB
GRAVEL CHUB
BLACKNOSE DACE
CREEK CHUB
TONGUETIED MINNOW
SUCKERMOUTH MINNOW
SOUTHERN REDBELLY DACE
REDSIDE DACE
EMERALD SHINER
SILVER SHINER
ROSYFACE SHINER
REDFIN SHINER
ROSEFIN SHINER
STRIPED SHINER
COMMON SHINER
BIGEYE SHINER
STEELCOLOR SHINER
SPOTFIN SHINER
INSECTIVORE
INSECTIVORE
INSECTIYORE
OMNIVORE
OMNIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
HERBIVORE
INSECTIVORE
PISCIVORE
INSECTIVORE
INSECTIVORE
HERBIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
TOLERANT
MODERATELY TOLERANT
MNTOLERANT
TOLERANT
TOLERANT
TOLERANT
MNTOLERANT
MNTOLERANT
MNTOLERANT
INTOLERANT
MODERATELY TOLERANT
TOLERANT
INTOLERANT
MODERATELY TOLERANT
MNTOLERANT
MNTOLERANT
MODERATELY TOLERANT
INTOLERANT
MNTOLERANT
TOLERANT
INTOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
INTOLERANT
MODERATELY TOLERANT
TOLERANT
64
-------
SPECIES
TROPHIC LEVEL
TOLERANCE
SAND SHINER
MIMIC SHINER
GHOST SHINER
SILVERJAW MINNOW
MISSISSIPPI SILVERY MINNOW
FATHEAD MINNOW
BLUNTNOSE MINNOW
CENTRAL STONEROLLER
CHANNEL CATFISH
YELLOW BULLHEAD
BROWN BULLHEAD
BLACK BULLHEAD
FLATHEAD CATFISH
STONECAT
BRINDLED MADTOM
TADPOLE MADTOM
BLACKSTRIPE TOPMINNOW
TROUTPERCH
BROOK SILVERSIDE
WHITE BASS
WHITE CRAPPIE
BLACK CRAPPIE
ROCKBASS
SMALLMOUTH BASS
SPOTTED BASS
LARGEMOUTH BASS
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
HERBIVORE
OMNIVORE
OMNIVORE
HERBIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
PISCIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
PISCIVORE
PISCIVORE
PISCIVORE
PISCIVORE
PISCIVORE
PISCIVORE
PISCIVORE
MODERATELY TOLERANT
INTOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
INTOLERANT
TOLERANT
TOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
TOLERANT
MODERATELY TOLERANT
MNTOLERANT
INTOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
INTOLERANT
MODERATELY TOLERANT
TOLERANT
MODERATELY TOLERANT
INTOLERANT
INTOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
65
-------
SPECIES
TROPHIC LEVEL
TOLERANCE
WARMOUTH
GREEN SUNFISH
BLUEGILL
ORANGESPOTTED SUNFISH
LONGEAR SUNFISH
PUMPKINSEED
SAUGER
WALLEYE
YELLOW PERCH
DUSKY DARTER
BLACKSIDE DARTER
SLENDERHEAD DARTER
LOGPERCH
EASTERN SAND DARTER
JOHNNY DARTER
GREENSIDE DARTER
BANDED DARTER
VARIEGATE DARTER
RAINBOW DARTER
ORANGETHROAT DARTER
FANTAIL DARTER
FRESHWATER DRUM
MOTTLED SCULPIN
BROOK STICKLEBACK
PISCIVORE
PISCIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
PISCIVORE
PISCIVORE
PISCIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
INSECTIVORE
MODERATELY TOLERANT
TOLERANT
TOLERANT
TOLERANT
MNTOLERANT
INTOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
INTOLERANT
MODERATELY TOLERANT
INTOLERANT
INTOLERANT
MNTOLERANT
MODERATELY TOLERANT
INTOLERANT
INTOLERANT
INTOLERANT
INTOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
MODERATELY TOLERANT
MNTOLERANT
INTOLERANT
66
------- |