\
; f" Agency
United States Office of Research and EPA/600/R-00/108
Environmental Protection Development April 2001
Washington DC 20460
&EPA Comparisons of Boating and
Wading Methods Used to
Assess the Status of
tfjjb*^1
Flowing Waters
"
II
4 ^- "
'^MiJ*^*'
t
•:
s*
. t
'*
/%
V
\
* \
•-t*
^
»^»
^k
+ ^w
< * <
^k^l44 N
/%
/*
r%
***&•'•x
^/^
F
¥>^ v
vvr
w^'A-
\ W V ^- v
/^!» Sa_' ' \
5=5^.
-------
EPA/600/R-00/108
April 2001
Comparisons of Boating and Wading
Methods Used to Assess the
Status of Flowing Waters
Joseph E. Flotemersch and Susan M. Cormier
National Exposure Research Laboratory
Bradley C.Autrey
SoBran, Inc.
U.S. Environmental Protection Agency
National Exposure Research Laboratory
26 W. M.L.King Drive
Cincinnati, OH 45268
80% Recycled/Recyclable
Printed with vegetable-based ink on
paper that contains a minimum of
50% post-consumer fiber content
processed chlorine free
-------
This research described in this report has been funded wholly or in part
by the U.S. Environmental Protection Agency. This document has been pre-
pared at the EPA National Exposure Research Laboratory (Ecological Expo-
sure Research Division, Cincinnati, Ohio) under Contract 68-C6-0019.
This research was supported in part by an appointment to the Postgradu-
ate Research Participation Program administered by the Oak Ridge Institute
for Science and Education through an interagency agreement between the
U.S. DOE and the U.S. EPA.
Mention of trade names or commercial products does not constitute en-
dorsement or recommendation of use.
The correct citation for this document is:
Flotemersch, IE1., B.C. Autrey2, and S.M. Cormier1. 2000. Compari-
sons of Boating and Wading Methods Used to Assess the Status of Flowing
Waters U.S. Environmental Protection Agency, Cincinnati OH.
Section authors and current addresses are listed below.
Section 1: Joseph E. Flotemersch1
Section 2: Bradley C. Autrey2 and Joseph P. Schubauer-Berigan3
Section 3: Bradley C. Autrey2 and Joseph P. Schubauer-Berigan3
Section 4: Joseph E. Flotemersch1, Susanna DeCelles2, and
Bradley C. Autrey2
Section 5: Bradley C. Autrey2
Section 6: Joseph E. Flotemersch1
'U.S. Environmental Protection Agency, National Exposure Research Laboratory, 26 W. M. L. King
Drive, Cincinnati, OH 45268
2Sobran, Inc. c/o U.S. Environmental Protection Agency, National Exposure Research Laboratory, 26
W. M. L. King Drive, Cincinnati, OH 45268
3National Center for Environmental Assessment 26 W. M. L. King Drive, Cincinnati, OH 45268
-------
This document has been designed to provide an overview of the biologi-
cal, physical and chemical methods of selected stream biomonitoring and
assessment programs. It was written to satisfy the need to identify current
methods that exist for sampling large rivers. The primary focus of this
document is the boating methods used to assess flowing waters, but both boat-
based and wading methods are included. The target audiences are individuals
tasked:
1. to work with data generated from one or more of these programs;
2. to design or improve a bioassess- and monitoring program;
3. to conduct field work using methods (or based on methods) reviewed in
this text;
4. to conduct field comparisons among these methods to determine the
extent of their comparability and when each method is best employed.
This document is useful to these individuals in that it brings together
relatively obscure literature from a wide variety of sources and it presents
current and developing methods in a comprehensive context. These features
allow this document to serve as a guide for comparing the methods used by
various agencies for assessing large rivers.
Much of the included text has been largely adapted and modified from
the agency documents from which it was derived. This has been done pur-
posefully to reduce the risk of misinterpretation.
The primary focus of this document is the boating methods used to as-
sess flowing waters. However, both boat-based and wading methods are
included in this document for several reasons. First, most wading methods
-------
were developed before boating methods and boating methods are often deri-
vations of the wading methods that preceded them. Often, the methods used
while in boatable waters simply call for the wading methods to be used in
shallow areas (e.g., near the shore) or in the boat without any additional modi-
fications. The inclusion of the original (wading) method as well as the derived
(boating) method may also help illustrate how methods can be modified in
order to meet the specific requirements of a sampling agency. Another reason
that both sets are included is that it may be necessary to use both wading and
boating methods among sampling sites or within a single reach when a river
has varying depths. Finally, the inclusion of both sets of methods may help
agencies or individuals analyze data sets that were collected using both wad-
ing and boating methods.
The information regarding the boating and wading methods reviewed in
this document was derived from the available literature, the Internet, personal
experience and personal communications with research scientists from respective
agencies. Although some methods may have been modified or reduced since
their conception, methods are presented in their entirety so as to not diminish
their original intention. Where necessary, appendices are included to aid un-
derstanding of or differences among methodologies.
Methods employed by the reviewed bioassessment and monitoring pro-
grams varied greatly. Differences included, but were not limited to: overall
site selection (random, non-random), number and location of samples col-
lected within the selected site, index or sample period, stream length sampled,
time needed to execute methods in the field, time required to process samples
in the field, type of sample collected (qualitative, semi-quantitative, or quanti-
tative), equipment required to execute methods, expertise required to execute
methods successfully, and subjectiveness of method. These differences may
help individuals choose the methods appropriate to their sampling needs. Sum-
mary tables are included throughout the document that aid in understanding
the differences between the methods used by the various agencies.
-------
Section Page
1.1 Boating and Wading Methods 3
1.2 Overall Sampling Design of Reviewed Programs 4
1.2.1 USEPA-EMAP-SW Methods 4
1.2.2 USGS-NAWQA Methods 6
1.2.3 USEPA-RBP Methods 8
1.2.4 Ohio EPA Methods 9
1.2.5 MDNR-MBSS Methods 10
2.1 Development of Habitat Assessment Methods 12
2.1.1 USEPA-EMAP-SW 12
2.1.2 USGS-NAWQA 13
2.1.3 USEPA-RBP 13
2.1.4 Ohio EPA 13
2.1.5 MDNR-MBSS 13
2.2 USEPA-EMAP-SW Habitat Assessment Index 14
2.2.1 USEPA-EMAP-SW RHA Index 14
2.2.2 USEPA-EMAP-SW-PHab Assessment 15
2.2.3 Additional Habitat Parameters 15
2.3 USGS-NAWQA Habitat Assessment Protocol 15
-------
Section Page
2.3.1 Habitat Sampling Design 15
2.3.2 Basin Characterization 16
2.3.3 Segment Characterization 17
2.3.4 Reach Characterization 17
2.4 USEPA-RBP Habitat Assessment Index 18
2.5 Ohio ERA'S Qualitative Habitat Evaluation Index
(QHEI) 22
2.6 MDNR-MBSS Habitat Assessment Method 22
2.6.1 Qualitative Habitat Assessment 22
2.7 Differences and Similarities Between the Habitat
Assessment Methods 24
2.7.1 The USEPA-EMAP-SW RHA and the
USEPA-RBP Habitat Assessment Indices 25
2.7.2 The MDNR-MBSS Qualitative Habitat
Assessment Protocols and the Other
Programs 26
2.7.3 Ohio EPA QHEI and the Other Programs 27
2.7.4 USGS-NAWQA and the Other Programs 27
2.7.5 Broad Scale Differences Among the Habitat
Assessment Methods Used by the Five Reviewed
Programs 27
2.7.6 Local Scale Differences Among the Habitat
Assessment Methods Used by the Five Reviewed
Programs 28
2.7.7 Sampling Season 29
3.1 USEPA-EMAP-SW Water Chemistry Assessment 30
3.2 USGS-NAWQA Water Chemistry Assessment 34
3.2.1 Basic Fixed-Site Assessment 34
3.2.2 Intensive Fixed-Site Assessment 34
3.2.3 Water-Column Synoptic Studies 34
3.3 USEPA-RBP Water Quality Assessment 35
3.4 Ohio EPA Water Chemistry Assessment 35
3.4.1 Sample Types 36
3.4.2 Procedures for Collecting Grab Samples 36
3.4.3 Procedures for Collecting Composite
Sam pies 36
-------
Section Page
3.4.4 Parameters Requiring Special Collecting and
Handling Procedures 37
3.5 MDNR-MBSS Water Chemistry Assessment 37
3.6 Comparisons Between Programs 37
3.6.1 Sampling Methods 37
3.6.2 Analytes Sampled 38
4.1 USEPA-EMAP-SW Periphyton Assessment Program 39
4.1.1 Sample Collection 40
4.1.2 Sample Processing and Methods 40
4.2 USGS-NAWQA Algae Assessment Program 41
4.2.1 Sample Collection 41
4.2.1.1 Natural Substrates 41
4.2.1.1.1 Qualitative Multihabitat Periphyton Samples 42
4.2.1.1.2 Quantitative Targeted-Habitat Periphyton
Sam pies 42
4.2.1.2 Using Artificial Substrates to Collect Periphyton 44
4.2.1.3 Quantitative Phytoplankton Samples 44
4.2.2 Sample Processing and Methods 45
4.3 USEPA-RBP Periphyton Assessment Protocols 45
4.3.1 Sample Collection 46
4.3.1.1 Natural Substrates 46
4.3.1.2 Artificial Substrates 47
4.3.2 Methods for Semi-Quantitative Assessments of
Benthic Algal Biomass and Taxonomic Composition ....47
4.3.3 Periphyton Metrics 48
4.4 Indices 48
4.4.1 The Pollution Tolerance Index (PTI) 49
4.4.2 Percent Community Similarity (PSc) 49
4.4.3 The Autotrophic Index 50
4.5 Summaries of the Periphyton Assessment Programs
of the USEPA-EMAP-SW, USGS-NAWQA, and
USEPA-RBP 51
-------
Section Page
5.1 USEPA-EMAP-SW Macroinvertebrate Assessment 53
5.1.1 Wadeable Streams: Riffle/Run and Pool/Glide
Sampling 53
5.1.2 Beatable Streams 54
5.2 USGS-NAWQA Macroinvertebrate Assessment 55
5.2.1 Qualitative Multihabitat (QMH) Sampling Methods 55
5.2.2 Semi-Quantitative Targeted-Habitat Sampling
Methods 55
5.2.2.1 Wadeable Coarse-Grained Substrates 56
5.2.2.2 Beatable Coarse-Grained Substrates 57
5.2.2.3 Wadeable Fine-Grained Substrates 57
5.2.2.4 Beatable Fine-Grained Substrates 57
5.2.2.5 Woody Snags and Macrophytes 58
5.3 USEPA-RBP Macroinvertebrate Assessment 58
5.3.1 Single Habitat Approach 58
5.3.2 Multi-Habitat Approach 58
5.4 Ohio EPA Macroinvertebrate Assessment 59
5.4.1 Artificial Substrate 59
5.4.2 Natural Substrate 59
5.5 MDNR-MBSS Macroinvertebrate Assessment 60
5.5.1 Sampling Methods 60
5.6 Origin of Benthic Macroinvertebrate Indices 61
5.6.1 ThelCI 61
5.6.2 TheHBI 61
5.6.2.1 Scoring of the HBI 62
5.6.3 TheB-IBI 62
5.6.4 TheSBII 62
5.7 Indices and Metrics Used by the Programs for Analysis
of Benthic Macroinvertebrate Communities 63
5.7.1 USEPA-EMAP-SW Benthic Macroinvertebrate
Analysis 63
5.7.2 USEPA-RBP Benthic Macroinvertebrate Analysis 63
5.7.3 Ohio EPA Benthic Macroinvertebrate Analysis 65
-------
Section Page
5.7.4 MDNR-MBSS Benthic Macroinvertebrate Analysis 65
5.8 Comparison of Benthic Macroinvertebrate Indices 65
5.8.1 Indices 66
5.8.2 Sampling Locations 66
5.8.3 Sampling Equipment 67
6.1 USEPA-EMAP-SW Fish Data Collection Methods 68
6.1.1 Wadeable Streams 68
6.1.2 Beatable Streams 69
6.1.3 Data Recorded 69
6.2 USGS-NAWQA Fish Data Collection Methods 69
6.2.1 Wadeable Streams 71
6.2.2 Beatable Streams 71
6.2.3 Other Sampling Methods 72
6.2.4 Data Recorded 72
6.3 USEPA-RBP Fish Data Collection Methods 72
6.3.1 Wadeable Streams 73
6.3.2 Beatable Streams 73
6.3.3 Data Recorded 73
6.4 Ohio EPA Fish Data Collection Methods 73
6.4.1 Wadeable Streams 74
6.4.2 Beatable Streams 74
6.4.3 Data Recorded 74
6.5 MDNR-MBSS Fish Data Collection Methods 74
6.5.1 Wading Methods 74
6.5.2 Boating Methods 75
6.5.3 Data Collected 75
6.6 Origin And Development of the IBI and Modified IWB 75
6.7 Indices Used by the Programs to Interpret Fish Data 76
6.7.1 USEPA-EMAP-SW Fish Data Interpretation 76
6.7.2 USGS-NAWQA Fish Data Interpretation 78
6.7.3 USEPA-RBP Fish Data Interpretation 78
6.7.4 Ohio EPA Fish Data Interpretation 78
6.7.5 MDNR-MBSS Fish Data Interpretation 80
-------
Section Page
6.8 Comparison of the Fish Assessment Programs of the
USEPA-EMAP-SW, USGS-NAWQA, USEPA-RBP, Ohio
EPA, AND MDNR-MBSS 81
6.9 Conclusions Regarding Potential Comparisons of
Fish Data 82
A Descriptions of the Habitat Assessment
Parameters A-1
B Periphyton Metrics Listed in the USEPA-RBP B-1
C Benthic-IBI Metrics C-1
D ICI Metrics D-1
E Modified Index of Well-being (IWB) E-1
F Fish IBI Scoring F-1
G Fish IBI Metrics Used by USEPA-EMAP-SW,
Ohio EPA and the MDNR-MBSS Programs.
Metrics are Grouped by Association or Similarity... G-1
-------
Table Page
1-1. Contact Information for the Five Reviewed Programs 2
2-1. The Metrics And Scoring for the USEPA-EMAP-SW
RHA Index 14
2-2. Metrics and Scoring Used in the PHab Assessment 16
2-3. Additional Parameters Used For The USEPA-EMAP-
SW Protocols 17
2-4. The USGS-NAWQA Parameters Recorded for Basin
Characterization 18
2-5. The USGS-NAWQA Parameters Measured for
Segment Characterization 19
2-6. The USGS-NAWQA Parameters Measured for
Segment Characterization 20
2-7. The Metrics And Scoring Used in the USEPA-RBP'S
Habitat Assessment Index 22
2-8. The Metrics, Sub-metrics, and Scoring Ranges for the
Ohio EPA's QHEI 23
2-9. An Example of the Metric Scoring Method Used by the
QHEI 23
2-10. Habitat Quality Rankings Developed by the Ohio EPA
for QHEI Score Evaluation 24
2-11. Observations Recorded in Addition to the QHEI
Parameters 24
2-12. Metrics Used in the MDNR-MBSS Qualitative Habitat
Assessment Method 25
3-1. Water Chemistry/Water Quality Measurements Made
by USEPA-EMAP-SW, USGS-NAWQA, USEPA-RBP,
Ohio EPA and MDNR-MBSS in Conjunction with
Monitoring and Assessment 31
-------
Table Page
3-2. Dissolved Pesticides Analyzed by USGS-NAWQA in
Addition to Basic Fixed Site Analytes in Conducting
Intensive Fixed-site Assessment 35
3-3. Categories Available for Scoring the Estimated
Parameters of the USEPA-RBP'S Recommended
Water Quality Assessment 36
4-1. Proposed Periphyton Indicators of Stream Condition
and Associated Parameters 40
4-2. Analytical Methodologies: Stream Periphyton
Indicator 42
4-3. Microhabitats Used by the USGS-NAWQA Periphyton
Collection Protocol and Methods Used for the
Qualitative Survey 43
4-4. Summary Of RBP Collection Techniques for Periphyton
from Wadeable Streams 46
4-5. Scale Used to Score the Density of Microalgae in the
RBP Semi-quantitative Method 48
4-6. Diatom and Non-Diatom Metrics Summarized in the
RBP Manual 49
4-7. Methods Used by the Three Reviewed Programs for the
Collection and Assessment of Periphyton and
Phytoplankton Assemblages 50
5-1. Metrics Used In The Ohio EPA's ICI and Their Expected
Responses to Disturbance 61
5-2. Water Quality Levels Indicated by Different Ranges of
HBI Scores 62
5-3. Metrics Used for the CP B-IBI and the NCP B-IBI 63
5-4. Metrics Used in the USEPA's SBII and Their Expected
Responses To Disturbance 63
5-5. TheUSEPA's SBII Condition Categories and
Associated Score Ranges ?
5-6. Indices Used by the USEPA-EMAP-SW Protocols 64
5-7. Metrics Suggested by the USEPA-RBP 64
5-8. Comparison of Benthic Macroinvertebrate Indices,
Sampling Methods, Preferred Sampling Habitats, and
Preferred Sampling Seasons 66
-------
Table Page
6-1. Metrics in the Index of Biotic Integrity for the USEPA-
EMAP-SW Program 77
6-2. Metrics Recommended for Calculation by the USEPA-
RBP 79
6-3. Metrics Employed by the Ohio EPA with Expected
Response to Stress 80
6-4. Metrics Employed by MDNR-MBSS and Expected
Response to Stress 81
-------
Figure Page
1 -1. A Member of a Field Crew Ties a Red Flag in a
Tree in Order to Mark the End of a Transect at the
Proper Distance from the Previous Transect 5
2-1. A Member of a Field Crew Takes a Measurement of the
Canopy Density by Using a Densiometer 13
3-1. A Member of a Field Crew Fills a Cubitainer with Water
that will be Used in Water Chemistry Analysis 33
4-1. A Member of a Field Crew Dislodges Attached
Periphyton Using the EMAP-SWMethod 41
4-2. A Member of a Field Crew Dislodges Attached
Periphyton from its Substrate Using the USGS-NAWQA
Method with the SG-92 ?
4-3 A Member of a Field Crew Filters a Periphyton Sample
for Chlorophyll Analysis 45
5-1. A Modified Kick Net (Left) such as is used in the
USEPA-EMAP-SW Protocols and a D-frame Kick
Net (Right) such as is used in the USGS-NAWQA
Protocols 53
5-2. A Member of a Field Crew Processes an Benthic
Macroinvertebrate Sample before it is Transported to
the Laboratory for Analysis 56
5-3. A Member of a Field Crew Uses a Stiff-bristled Brush
to Remove the Attached Benthic Macroinvertebrates
from a Rock 57
5-4. A Hester-dendy Sampling Device Set in a River.
Note: This sampler was set in a more shallow area
for photographic purposes. Hester-Dendy samplers
are normally set approximately 1 meter below the
water's surface 60
-------
Figure Page
6-1. A Member of a Field Crew Uses a Backpack
Electroshocker to Sample the Fish in a Wadeable
Stream 69
6-2. A Member of a Field Crew Samples Fish Using the
Boat-based Electroshocking Technique For Beatable
Rivers 70
6-3. Before they are Released, the Fish are Identified,
Measured and Weighed and these Data are Recorded 70
-------
Programs reviewed in this document include the U.S. Environmental
Protection Agency's Environmental Monitoring and Assessment Program for
Surface Waters (USEPA-EMAP-SW), U.S. Geological Survey's National
Water-Quality Assessment program (USGS-NAWQA), U.S. Environmental
Protection Agency's Rapid Bioassessment Protocol (USEPA-RBP), Ohio
Environmental Protection Agency's flowing waters program (Ohio EPA), and
Maryland's Department of Natural Resources's Maryland Biological Stream
Survey program (MDNR-MBSS).
Numerous individuals at various agencies were contacted to aid in the
interpretation and understanding the methods reviewed. Their continued as-
sistance and cooperation is greatly appreciated.
Appreciation is especially extended to Marc Smith (Ohio EPA), James
Lazorchak (USEPA) and Marty Gurtz (USGS-NAWQA) for the review com-
ments they provided. These comments were important assets in the prepara-
tion of this document.
-------
AFDM Ash-Free Dry Mass
Al Autotrophic Index
APA Acid/Alkaline
Phosphatase Ac-
tivity
ANC Acid Neutralizing
Capacity
B-IBI Benthic Index of
Biotic Integrity
BEST Biomonitoring of
Environmental
Status and Trends
BOD Biological Oxygen
Demand
CC Conservation Com-
mission
COD Chemical Oxygen
Demand
CP Coastal Plains
DELT Deformities, Eroded
Fins, Lesions, and
External Tumors
DIG Dissolved Inorganic
Carbon
DO Dissolved Oxygen
DTH Depositional-
Targeted Habitat
DEP Department of
Environmental
Protection
DOC Dissolved Organic
Carbon
E. coli Escherichia coli
EAV Emergent Aquatic
Vegetation
EMAP Environmental
Moni-toring and
Assessment Pro-
gram
EPT Ephemeroptera,
Plecoptera,
and Trichoptera
CIS Geographic
Information Systems
H' Shannon-Weiner
Diversity Index
HBI Hilsenhoff Biotic
Index
IBI Index of Biotic
Integrity
ICI Invertebrate
Community Index
ID Identification
Iwb Index of Well-Being
LWD Large Woody Debris
MAH Mid-Atlantic High-
lands
MAIA Mid-Atlantic Inte-
grated Assess-
ment
-------
MANIA Monitoring and Non- RBP
Tidal Assessment
MBSS Maryland Biological RHA
Stream Survey
MDNR Maryland Depart- RTH
ment of Natural
Resources SAV
NAWQA National Water-
Quality Assessment SBII
program
NCR Non-Coastal Plains SG-92
NERL National Exposure SOC
Research Labora-
tory SW
Ohio EPA Ohio Environmental IDS
Protection Agency
PCA Principal Component TNDT
Analysis
PHab Physical Habitat TOC
PSc Percent Community TSS
Similarity
PTI Pollution Tolerance USEPA
Index
QHEI Qualitative Habitat
Evaluation Index USGS
QMH Qualitative Multi-
habitat
Rapid Bioassess-
ment Protocols
Rapid Habitat
Assessment
Richest-Targeted
Habitat
Submerged
Aquatic Vegetation
Stream Benthos
Integrity Index
Scum-Getter 92
Suspended Organic
Carbon
Surface Waters
Total Dissolved
Solids
Total Number of
Diatom Taxa
Total Organic Carbon
Total Suspended
Solids
United States Envi-
ronmental Protec-
tion Agency
United States Geo-
logical Survey
-------
by
Joseph E. Flotemersch
This document has been designed to
provide an overview of the biological, physi-
cal and chemical methods of selected stream
biomonitoring and assessment programs. The
target audiences are those individuals tasked
with working with the data generated from
one or more of these programs, yet unfamil-
iar with the basics of the sampling procedures
themselves. Other tasks that may be aided
by this document are the design or improve-
ment of a bioassessment and monitoring pro-
gram, conducting field work using methods
reviewed in this text, or conducting field com-
parisons among these methods to determine
the extent of their comparability and when
each method is best employed. However, this
document is not intended to serve as a sub-
stitute for the protocol manuals produced by
the respective agencies. Individuals intend-
ing on implementing any of these protocols
should, at a minimum, obtain a copy of the
agency's original protocol manual. It would
also be beneficial to these individuals to con-
tact the agencies in order to gain the insight
of the scientists who developed these proto-
cols or who utilize them on a regular basis.
Such contact could provide clarification or
modifications to the protocols of interest. Table
1-1 provides contact information for the five
agencies that are reviewed in this document.
The reviewed biomonitoring programs
differ not only in their methods for collecting
samples in the field but also their methods for
processing samples in the laboratory. While
the different laboratory methods may create
additional differences in the final data pro-
duced by the different agencies, these labora-
tory methods are outside the scope of this
document which will focus exclusively on the
field methods.
Much of the included text has been
largely adapted and modified from the agency
documents from which it was derived. This
has been done purposefully to reduce the risk
of misinterpretation.
Programs reviewed include the U. S. En-
vironmental Protection Agency's Environ-
mental Monitoring and Assessment Program
for Surface Waters (USEPA-EMAP-SW),
U.S. Geological Survey's National Water-
-------
Table 1-1. Contact Information for the Five Reviewed Programs
Biomonitoring
Program
Program Contact
General E-Mail
Contact and Web Sites
Publications
Contact
USEPA-EMAP-
SW
USGS-NAWQA
USEPA-RBP
Ohio EPA
JohnStoddard
U SEP A National Health and
Environmental Effects
ResearchLab/ORD
Western Ecology Division
Address:
200 S.W. 3 5th Street
CorvallisOR97333-4902
Telephone: 541-7544441
E-mail:
Stoddard@mail.cor.epa.gov
TomMuir
Coordinator, NAWQA
Address:
Mail Stop 3 660
1849 C Street, N.W.
Washington, D.C. 20240
Telephone: 703-648-5114
E-mail: tmuir@usgs.gov
Michael T. Barbour
Tetra Tech, Inc.
Ecological Sciences
Address:
10045 Red RunRoad,
Suite 110
OwingsMills,MD21117
Telephone: 410-356-8993
E-Mail:
Michael.Barbour@tetratech.com
Chris Yoder
Division of Surface Water/
Ecological Assessment Unit
Address:
4675 HomerOhio Lane
Groveport,OH43125
Telephone: 614-836-8778
Agency Mailing Address:
Lazarus Government Center
P.O. Box 1049
Columbus, OH 43216-1049
Agency Telephone: 614-644-2001
E-mail: emap@epa.gov
Web Site:
www.epa.gov/emap
Web Site:
www.water.usgs.
gov/nawqa/nawqa_
home.html
Web Site:
www.epa.gov/
owow/monitoring/rbp
E-Mail:
info-request@www.epa.
state.oh.us
Web Sites:
www.web.epa. ohio.gov
www.epa.state.oh.us
National Service Center
forEnvironmental
Publications
Address:
P.O.Box42419
Cincinnati, OH 45242-2419
Telephone: 800-490-9198
FaxNumber: 513-489-8695
U.S. Geological Survey
Earth Science and
Information Center
Open-File Reports Section
Address:
Box25286,MS517
Denver Federal Center
Denver, CO 80225
Telephone: 800-435-7627
800-872-6277
National Service Center
forEnvironmental
Publications
Address: P.O. Box 42419
Cincinnati, OH 45242-2419
Telephone: 800-490-9198
FaxNumber: 513489-8695
N/A
(continued)
-------
Table 1-1. Continued
Biomonitoring
Program
Program Contact
General E-Mail
Contact and Web Sites
Publications
Contact
MDNR-MBSS
Ann Smith
Monitoring and Nontidal
Assessment Program of the
Maryland Department of
Natural Resources
Telephone:410-260-8611
Web Sites:
www.dnr.state.
md.us/streams/ mbss/
mbss_methods.html
www.nt2.versar. com/
E-mail: asmith@dnr.state.md.us. mbss/mbss. html
Agency Mailing Address:
Tawes State Office Building
580 Taylor Avenue
Annapolis, MD 21401
Paul Miller
Tawes State Office Building,
C-2
MD Department of Natural
Resources
Address:
580 Taylor Avenue
Annapolis, MD 21401
Telephone: 410-260-8610
Email:
pmiller@dnr.state.md.us.
Quality Assessment program (USGS-
NAWQA), U.S. Environmental Protection
Agency's Rapid Bioassessment Protocol
(USEPA-RBP), Ohio Environmental Protec-
tion Agency's flowing waters program (Ohio
EPA), and Maryland's Department of Natu-
ral Resources's Maryland Biological Stream
Survey program (MDNR-MBSS). While the
USEPA-EMAP-SW, USGS-NAWQA and
USEPA-RBP programs are concerned with
assessing rivers on the National and Regional
levels, the Ohio EPA and MDNR-MBSS pro-
grams are concerned with assessing the riv-
ers in their respective states. These differences
in scale are reflected in the way each program
developed and currently implements their pro-
tocols.
The depth of flowing waters can be
roughly characterized as boatable or wade-
able. The methods used to assess the condi-
tion of these flowing waters may vary depend-
ing on their depth status. Because it is the goal
of this document to help individuals under-
stand the differences between the ways data
are collected, this document distinguishes
between boating and wading methods when
they occur.
The primary focus of this document is
the boating methods used to assess flowing
waters, however, both boating and wading
methods are included in this document for
several reasons. First, most wading methods
were developed before boating methods and
boating methods are often derivations of the
wading methods that preceded them. Often,
the methods used while in boatable waters
simply call for the wading methods to be used
in shallow areas (e.g., near the shore) or in
the boat without any additional modifications.
The inclusion of the original (wading) method
as well as the derived (boating) method may
also help illustrate how methods can be modi-
fied in order to meet the specific requirements
of a sampling agency. Another reason that
both sets are included is that it may be neces-
sary to use both wading and boating methods
among sampling sites or within a single reach
when a river has varying depths. Also, sepa-
rate protocols specifically tailored for either
boatable or wadeable streams are not avail-
able for all phases of all programs. Therefore,
-------
it is necessary to include the protocols that
are available even if they are not specified as
protocols for boatable streams. Finally, the
inclusion of both sets of methods may help
agencies or individuals analyze data sets
which were collected using both wading and
boating methods.
The USEPA has designated EMAP-SW
to develop the necessary monitoring tools that
can determine the current status, extent,
changes and trends in the condition of our
Nation's ecological resources on regional and
national scales (U.S. EPA 1998). The sam-
pling framework for this program consists of
40-km2 hexagons placed over a systematic tri-
angular grid of approximately 12,500 points
for the contiguous United States. The
program's national design states that approxi-
mately 800 lakes and 800 streams are chosen
from one quarter of the grid hexagons each
year, giving a four-year resampling cycle. The
field sampling sites are selected using statisti-
cal probability methods to ensure that robust
population inferences can be made and that
the sites represent the spatial distribution of
lakes and streams (Overton et al. 1991). Sites
are randomly selected by establishing size
strata, to ensure an adequate characterization
of larger lakes and streams.
The sampling period, or index period,
for USEPA-EMAP-SW varies with the loca-
tion and type of project being conducted. For
the Mid-Atlantic Integrated Assessment
(MAIA) project, a spring (April to June) in-
dex period was selected in 1993 and 1994. In
1997 and 1998, however, a summer (July to
September) index period was selected, which
coincided with the low flow period of streams
in this research area.
The elementary sampling unit used by
USEPA-EMAP-SW for biological, physical
and chemical data collection is a length of
stream 40 times the channel width. This length
was derived from pilot studies that indicated
this sampling approach was needed to collect
90% of the species in the stream reach. In
streams less than four meters wide, a length
of 150 m is used as a minimum sample reach
length. No maximum reach length was estab-
lished for boatable or wadeable streams.
Reaches are laid out so that 50% of the sur-
vey area is upstream, and 50% of the survey
area is downstream of the predetermined lati-
tude and longitude of the study site.
A designated sample reach is divided
into 10 subsections delineated by 11 transects
spanning the width of the stream and labeled
"A" through "K". The downstream endpoint
of the sample reach is transect "A". Transect
"B" is that point which is 1/10 (four channel
widths in big streams or 15 m in small streams)
of the designated stream length upstream from
the start point (transect A) [Figure 1-1 shows
a member of a field crew marking a transect
at the proper distance from the previous
transect.] When transect "B" is determined, a
roll of a die is used to determine the location
along the transect where sampling of certain
indicators will take place. Options are a
left(L), center(C), or right(R) sampling point.
After the first random selection (transect B),
sampling locations are assigned to each
transect, alternating in order as L, C, or R.
This process is repeated until the upstream
extent of the sample reach is located (transect
K).
-------
Figure 1 -1. A field crew memberties a flag in a tree to mark the a transect at the proper distance from
the previous transect.
Ecological indicators included in the
stream sampling program are physical habi-
tat, water chemistry, periphyton/phytoplank-
ton assemblages, sediment metabolism,
benthic macroinvertebrate assemblages,
aquatic vertebrate assemblages, fish tissue
contaminants, and sediment toxicity. This
document focuses on the water chemistry,
physical habitat, and assemblage indicators
only.
Physical habitat data are collected from
each stream reach. Stressor indicators derived
from the collected data are used to help ex-
plain or diagnose stream conditions relative
to various indicators. Important attributes of
physical habitat in streams are channel dimen-
sions, gradient, substrate characteristics, habi-
tat complexity and cover, riparian vegetation
cover and structure, disturbance due to hu-
man activity, and channel-riparian interaction
(Kaufmann 1993).
Water chemistry data are collected from
each stream in order to measure a variety of
physical and chemical analytes. Information
from these analyses is used to evaluate stream
condition with respect to stressors such as
acidic deposition (mine drainage), nutrient
enrichment, and other organic and inorganic
contaminants.
Periphyton samples are collected from
erosional and depositional habitats located at
each of the nine interior cross-sectional
transects (B through J). Four different types
of laboratory samples are prepared: 1) an ID/
enumeration sample to determine taxonomic
composition and relative abundances, 2) a
chlorophyll sample, 3) abiomass sample for
ash-free dry mass, and 4) an acid/alkaline
phosphatase activity sample. Benthic
macroinvertebrates are collected using a modi-
fied kick net. A kick net sample is collected
from each of the nine interior cross-sectional
transects (B through J) at the sampling point
(Left, Center, or Right) assigned when the
location of the sampling reach is determined.
Mussels and snails, within the kick net sample
points, are hand-collected. In boatable
-------
streams, drift nets are also used to collect
benthic macroinvertebrates.
Fish are sampled using a single-pass
electrofishing method covering the deter-
mined reach length. Each pass of the
electrofishing sampling has a duration of at
least 45 minutes but does not exceed three
hours. Herpetofauna observed in the course
of electrofishing for fish are collected and
identified to the species level.
The USEPA-EMAP-SW sampling
methods are detailed in Lazorchak et al.
(1998) for wadeable streams and Lazorchak
et al. (1999 draft version) for large rivers.
Boatable methods have been tested and re-
fined in a pilot study in Mid-Atlantic states
during 1997 and 1998 and Midwestern states
during 1999.
The objectives of the USGS-NAWQA
program are to: 1) describe current water-
quality conditions for a large part of the
Nation's freshwater streams, rivers, and aqui-
fers, 2) describe how water quality is chang-
ing over time, and 3) improve understanding
of the primary natural and human factors that
affect water-quality conditions (Fitzpatrick et
al. 1998). Investigations are performed on a
staggered time scale in 59 of the largest and
most significant hydrologic systems in the
country (Gilliom et al. 1995). Individual in-
vestigations are performed in study units and
consist of four to five years of intensive as-
sessment, which consists of a retrospective
analysis, occurrence and distribution assess-
ment, assessment of long-term trends and
changes, and case studies of sources, trans-
port, fate, and effects.
The USGS-NAWQA sampling design
is modified from an approach used by Frissel
et al. (1986) and includes four spatial scales:
basin., segment., reach., and microhabitat. Ba-
sins refer to entire stream systems. Segment?,
are streams bounded by confluences or chemi-
cal/ physical discontinuities. The reach scale
includes individual pools and riffles within
stream segments. Microhabitat data (e.g., ve-
locity, substrate type and depth) are collected
from the locations where invertebrate and al-
gal samples are taken. Basin and segment data
are collected using Geographic Information
Systems (GIS), topographic maps, and aerial
photographs but reach and microhabitat sam-
pling require site visits. Procedures for the col-
lection of reach data are described in later sec-
tions of this document. Procedures for col-
lecting microhabitat data are described in the
USGS-NAWQA protocols for the collection
of invertebrates (Cuffney et al. 1993a) and
algal samples (Porter et al. 1993).
Sampling sites are chosen to represent a
set of important environmental variables in the
Study Unit. Basic fixed sites are placed at or
near USGS gaging stations where continu-
ous discharge measurements are available.
Synoptic sites may be nongaged sites where
typically one-time measurements of a limited
number of characteristics are made with the
objective of answering a specific question.
The purpose of a synoptic site is to answer
questions regarding source, occurrence, or
spatial distribution. Only one sampling reach
is generally used to characterize a synoptic
site (Gilliom etal. 1995).
The location of each sampling reach is
usually related to a durable reference point
such as a stream gage or bridge pier that is
used to permanently define its location
(Meadoretal. 1993a). Sampling reaches are
located where instream and riparian habitat
conditions are representative of the local area
and support USGS-NAWQA study-unit ob-
-------
jectives. For example, sampling reaches
should be representative of a specific land use,
agricultural practice, or reference condition.
In order to meet these objectives, the sampling
reach may be located upstream, downstream,
or adjacent to the site location as long as the
water chemistry and hydrologic data collected
at the site accurately reflect conditions within
the sampling reach.
Sampling is conducted during low and
stable-flow periods, usually mid-June to early
October. These conditions increase the likeli-
hood that samples throughout the study unit
can be collected under similar flow conditions
(Gilliometal. 1995).
The primary determinant for the length
of the sampling reach is the presence of rep-
etitions of two geomorphic channel units, such
as a sequence of pool, riffle, pool, riffle
(Meadoretal. 1993b). Other determinants for
reach length are fish sampling considerations
(Meador et al. 1993a). Only those geomor-
phic channel units (riffle, run, and pool) that
cover more than 50% of the active channel
width are considered when determining the
length of the reach. If repetitions of geomor-
phic channel units are not present or are
present at intervals of greater than 1,000 m
(for example, in large rivers), the length of
the reach is determined to be 20 channel
widths based on the width of the channel at
the boundary of the reach. Theoretically, this
length will represent at least one complete
meander wavelength (Leopold and Wolman
1957). Regardless of the method used to es-
tablish the length of the sampling reach, the
minimum and maximum acceptable reach
lengths are 500 and 1,000 m, respectively, for
boatable sites; 150 and 300 m, respectively,
for wadeable sites; and 150 and 500 m, re-
spectively, for wadeable sites with stream
widths greater than 30 m. Typically, a single
sampling reach is established at each site,
however, three sampling reaches are estab-
lished at a subset of sites in order to assess
variability among sampling reaches.
Ecological indicators included in the
USGS-NAWQA stream sampling program
are water chemistry, tissue contaminants,
stream habitat, benthic and sestonic algal com-
munity samples, benthic invertebrate commu-
nities, and fish communities. This document
focuses on the water chemistry, physical habi-
tat, and community indicators only.
Stream habitat data are collected at each
sample site to relate habitat to other physical,
chemical, and biological factors to describe
water-quality conditions. Data collected at
each reach include measurements and obser-
vations of channel, bank, and riparian char-
acteristics (Meador et al. 1993b).
Water chemistry data are collected us-
ing three levels of sampling and analytical
intensity. These three levels are basic fixed-
site assessment, intensive fixed-site assess-
ment, and water column synoptic studies. The
basic fixed-site assessment assesses a suite of
analytes using continuous monitoring supple-
mented by fixed-interval and extreme-flow
sampling. Intensive fixed-site assessments
utilize a higher-frequency sampling scheme
and add pesticides to the analytes. Water-col-
umn synoptic studies are short-term investi-
gations specifically designed for a particular
study unit.
Benthic algal communities are charac-
terized by collecting qualitative and quantita-
tive periphyton samples at each sampling loca-
tion. In boatable streams, phytoplankton may
be collected from the water column to char-
acterize the sestonic algal community. Esti-
mates of algal biomass (i.e., chlorophyll con-
tent and ash-free dry mass) are also optional
-------
measures of water-quality conditions (Porter
etal. 1993).
Benthic invertebrates are characterized
to develop a list of taxa within the associated
stream reach and to determine the structure
of benthic invertebrate communities within
selected microhabitats of each reach. Benthic
macroinvertebrates are qualitatively collected
with a kick net, which may be supplemented
with seines, visual collections, grab samples,
and/or diver operated dome samplers if re-
quired by the stream's morphology. In addi-
tion, benthic invertebrates are collected semi-
quantitatively from a measurable area of natu-
ral substrate. When the natural substrate is
unsuitable for collection, artificial substrates
may be used (Cuffney et al. 1993a, b).
Fish communities are characterized in
order to relate fish community characteristics
to physical, chemical, and other biological
factors. A representative sample of the fish
community is collected using electrofishing
and/or seining, depending on the appropriate-
ness of each method for the particular sam-
pling site (Meador etal. 1993a). TheUSGS-
NAWQA sampling methods are detailed in
later sections.
The primary purpose of the USEPA-
RBP is to provide state and local water-qual-
ity monitoring agencies with a practical tech-
nical reference for conducting cost-effective
biological assessments of lotic systems
(Barbour et al. 1999). The methods included
are a synthesis of methods employed by vari-
ous state water resource agencies. Therefore,
the protocols do not contain a set sampling
design.
The USEPA-RBP methods state that for
assessment and monitoring, sites can either
be targeted sites, which are relevant to spe-
cial studies focusing on potential problems,
or random sites, which provide information
of the overall status or condition of the water-
shed, basin, or region. In a random or proba-
bilistic sampling regime, stream characteris-
tics may be highly dissimilar among the sites,
but will provide a more accurate assessment
of biological condition throughout the area
than targeted designs. Most studies conducted
by state water quality agencies for identifica-
tion of problems and sensitive waters are done
with a targeted design. Studies for aquatic life-
use determination can be done with a random
or targeted design (Barbour et al. 1999).
The recommended sampling season is
mid to late summer, when stream and river
flows are moderate to low, and less variable
than during other seasons. The USEPA-RBP
suggests that stream reach designations based
on a fixed or proportional distance method
are acceptable, and that decisions between the
two methods should be based on the results
of pilot studies (Barbour et al. 1999).
Suggested ecological indicators included
in the USEPA-RBP are measurements of
physicochemical parameters, as well as per-
iphyton, benthic macroinvertebrate and fish
communities (Barbour et al. 1999).
The habitat assessment protocols sug-
gested by the USEPA-RBP include 13
metrics. Three of the metrics are used only at
high gradient sites and three metrics are used
only at low gradient sites. Therefore, only ten
metrics are used at any one site. Each metric
is assigned a score that ranges from 0 to 20
points. Each metric is scored by matching
observations made of the entire sample seg-
ment with one of four established ranking
categories. Higher index scores are associated
with more pristine habitats.
-------
The recommended water sampling meth-
ods are intended to provide a brief and eas-
ily-obtained analysis of water chemistry that
can be completed in the field. The suggested
assessment includes four quantitative mea-
surements and four estimated measurements.
The four estimated parameters are each as-
signed to a scoring category.
The objectives of the recommended
RBP for periphyton assessment include as-
sessment of biomass, identification of species
and determination of the periphyton assem-
blages' biological condition. During periods
of stable stream flow, periphyton are collected
from all available microhabitats in the sam-
pling reach in the approximate proportion each
microhabitat occurs. Algal mats or other soft-
bodied algal forms can be collected from
deposit!onal areas. For chlorophyll analyses,
periphyton are scraped from fixed areas onto
a glass fiber filter. Periphyton can be sampled
by collecting from artificial substrates
(periphytometers) that are placed in aquatic
habitats and colonized over a period of time.
Semi-quantitative assessments of benthic al-
gal biomass and taxonomic composition can
be made rapidly with a viewing bucket
marked with a grid and biomass scoring sys-
tem.
The USEPA-RBP recommend benthic
macroinvertebrates be sampled using either a
single habitat or a multiple habitat approach.
In the single habitat approach, all riffle/run
areas within a 100-m representative reach are
candidates for sampling macroinvertebrates.
Cobble substrate is sampled where it is the
predominant habitat and alternative habitats
are sampled when cobble is not the dominant
substrate. Sampling begins at the downstream
end of the reach and proceeds upstream us-
ing a 1-m, 500-|im mesh kick net. The stream
is sampled two or three times at locations of
varying velocity in the riffle. In the multiple
habitat approach, all habitat types in a 100-m
representative reach are sampled in the ap-
proximate proportion in which they are rep-
resented in the reach. Sampling begins at the
downstream end of the reach and proceeds
upstream using a D-frame, 500-|im mesh dip
net. A total of 20 jabs or kicks are taken over
the length of the reach.
The methods suggested by the USEPA-
RBP for fish involves careful, standardized
field collection, species identification and enu-
meration, and analyses using aggregated bio-
logical attributes. The suggested fish collec-
tion procedure is a multi-habitat approach for
wadeable streams, which allows the sampling
of habitats in relative proportion to their local
availability. The USEPA-RBP endorses
electro-fishing as the most comprehensive and
effective single method for collecting stream
fishes. Protocols suggest that collect on efforts
begin at a shallow riffle, or other physical
barrier at the downstream limit of the sample
reach, and terminate at a similar barrier at the
upstream end of the reach.
In order to monitor the state's aquatic
resources, Ohio EPA uses an approach in
which each basin has the potential to be stud-
ied for one field season during a five-year
cycle. Each five-year study focuses inten-
sively on the biological, physical and chemi-
cal conditions found within the chosen study
basins. Study segments are identified based
on criteria such as their potential to be threat-
ened by current or projected local impacts or
their potential for harboring unique or critical
aquatic habitat and biota. The size of the
stream study segment is adjusted based on the
size of the stream and whether or not the
stream is boatable. In general, monitoring is
-------
based on approximately a 500-m segment if
the stream or river is boatable, a 150 to 200-
m segment if the stream or river is wadeable
or a headwater stream (<20 mi2 of drainage
area). Sampling is conducted during summer
low flow months (June 15 to October 15) and
the study areas are visited one to three times
during the field season. The number of visits
to a single study site depends on a variety of
factors. Typically, headwater sites or impacted
sites are sampled once in a field season and
wadeable and boatable sites are sampled twice
during a field season. The wadeable and boat-
able sites may be sampled three times in a field
season if resources permit (OEPA 1988).
Ecological indicators included in Ohio
EPA's stream sampling program include
physical habitat, water chemistry,
macroinvertebrate assemblages and fish as-
semblages.
The characterization of physical habitat
in Ohio streams has been addressed through
Ohio EPA's development of the Qualitative
Habitat Evaluation Index (QHEI). This index
was designed to provide an evaluation or es-
timate of habitat attributes that generally cor-
respond to those physical factors that affect
fish communities and other aquatic organisms.
Important attributes of the QHEI include sub-
strate, instream cover, channel morphology,
riparian and bank condition, pool and riffle
quality, and gradient (Rankin 1989).
Water-quality sampling and analysis are
conducted to provide data which can be used
to interpret the quality or condition of the
water under investigation. Collected samples
may be discrete or integrated grabs or com-
posites. Composite samples are preferred to
insure temporally representative samples. Dis-
crete grab samples and integrated grabs are
considered satisfactory under temporally uni-
form conditions (OEPA 1988). An additional
method used to monitor water quality are con-
tinuous monitors. The monitors are set in ar-
eas to be modeled and on an availability ba-
sis. They provide information on a river or
stream's temperature, pH, conductivity and
dissolved oxygen (DO) level.
Macroinvertebrates are primarily sampled
using Hester-Dendy artificial substrate samplers.
Samplers (n=5) are ideally placed in runs and
harvested after a six-week colonization period.
In addition, macroinvertebrates are sampled
qualitatively by kick-net sampling and/or hand-
picking natural substrates for a period of at least
30 minutes and then until no new taxa are ob-
served.
Fish are sampled in one, two or three
single electrofishing passes of each sampling
segment per season (OEPA 1988, 1989).
Each of these sampling methods is discussed
in greater detail during later sections.
The MDNR-MBSS approach is de-
signed to provide three years of full coverage
of the state's 18 basins that contain headwa-
ter, non-tidal, first, second, and third order
streams. Approximately 300, non-overlap-
ping, 75-m stream segments are sampled each
year. The streams are defined using 1:250,000
scale base maps and the segments are ran-
domly selected using a lattice sampling ap-
proach in which the segments are stratified
by year and basin. Within a stream order, the
number of segments sampled per basin is pro-
portional to the number of stream miles in the
basin. A predetermined number of segments
are selected from each basin and ranked in
order of selection. Extra segments are selected
as a contingency to the loss of sampling seg-
-------
ments as a result of field conditions. If a basin
contains a small number of sites, additional
segments are selected to increase sample size
(Rothetal. 1997a,b).
In each segment, seven components are
monitored. Five components, fish,
herpetofauna, macrophytes, mussels, and
habitat quality, are sampled in the summer
period (June 1 to September 30) and two com-
ponents, benthic invertebrates and water qual-
ity are sampled in the spring period (March 1
to May 1). Fish and habitat measurements are
taken during summer low flow conditions for
three reasons: 1) spawning migration offish
is minimal in the summer; 2) low flow condi-
tions are advantageous for electrofishing, and
3) low flow conditions provide an opportu-
nity to assess the area and type of habitat avail-
able to fish communities at a time when habi-
tat may be limiting. Benthic sampling is con-
ducted in the spring when, according to
Plafkin et al. (1989), macroinvertebrate as-
semblages are good indicators of ecosystem
health (Rothetal. 1997b).
The MBSS qualitative habitat assessment
method consists of 13 metrics. Each metric is
scored by matching observations made of the
sample segment to the one of four possible
ranking categories that describe possible con-
ditions. Each of the four ranking categories
has a range of possible scores. The method is
designed so that higher scores indicate more
pristine habitats. No total index score is com-
puted for the MDNR-MBSS habitat assess-
ment. In addition to the 13 qualitative habitat
assessment metrics, MDNR-MBSS makes an
additional six quantitative habitat assessment
measurements.
Chemical water samples are analyzed
following U. S. EPA's Handbook of Standard
Methods for Acid Deposition Studies (U.S.
EPA 1987). Parameters analyzed include/?//,
acid neutralizing capacity (ANC), conductiv-
ity, sulfate, nitrate, and dissolved organic car-
bon (DOC). These variables are believed to
describe basic water quality conditions with
an emphasis on changes related to acidic
deposit! on (Rothetal. 1997b).
Invertebrates are sampled using a "D"
net, sampling one-ft2 areas of all available
habitats, for a total area of 20 ft2 per 75-m
stream segment. Fish are sampled in two
electrofishing passes of each 75-m segment
(Roth et al. 1997b). Detailed descriptions of
the sampling methods are given in later sec-
tions.
-------
by
Bradley C. Autrey and Joseph P. Schubauer-Berigan
This section summarizes and evaluates
the habitat assessment protocols of five agen-
cies, USEPA-EMAP-SW, USGS-NAWQA,
USEPA-RBP, Ohio EPA, and MDNR-
MBSS. Itbegins with a description of the ori-
gin of the habitat indices most widely used
by these agencies. Then the habitat assess-
ment methods of each agency are summa-
rized. Finally, the methods are compared and
contrasted. The USGS-NAWQA and
MDNR-MBSS sections differ from the other
agencies' sections because USGS-NAWQA
and MBSS do not compute an index value
from the recorded metrics. Instead, many
metrics are used to determine whether rela-
tionships exist among the habitat variables or
if any relationships existbetween habitat vari-
ables and dependent variables such as fish,
invertebrate, or periphyton assemblages.
These relationships are then examined to de-
termine what they indicate about stream qual-
ity.
The methods used by the USEPA-
EMAP-SW, USGS-NAWQA, USEPA-RBP,
Ohio EPA, and MDNR-MBSS were each
developed to meet the objectives of their re-
spective programs. The way in which each
of these protocols was developed reflects the
differences and the similarities among these
agencies (e.g., their spatial scales and objec-
tives). Figure 2-1 shows a member of afield
crew making a physical habitat measurement.
The USEPA-EMAP-SW's habitat as-
sessment protocols were developed by
Kaufmann (1993) and Kaufman and Robison
(1998) for wadeable streams and Kaufmann
-------
Figure 2-1. A field crew member measures canopy density by using a densiometer.
(Lazorchak et al. 1999 draft version) for boat-
able rivers. Both sets of protocols use a ran-
domized, systematic spatial sampling design
which minimizes bias in the placement and
positioning of measurements (Lazorchak et
al. 1998, 1999 draft version).
2.1.2 USGS-NAWQA
The USGS-NAWQA habitat assessment
protocols were developed by Meador et al.
(1993b) and were revised by Fitzpatrick et
al. (1998). The stratification in USGS-
NAWQA's habitat sampling design is a modi-
fication of Frissell et al. (1986). In addition,
microhabitat assessment protocols were de-
veloped by Cuffney et al. (1993a) in conjunc-
tion with protocols for the collection of inver-
tebrates and by Porter et al. (1993) in con-
junction with protocols for the collection of
algae (Fitzpatrick et al. 1998). These micro-
habitat assessment protocols are not addressed
in this document.
2.1.3 USEPA-RBP
Barb our et al. (1999) state that the
USEPA-RBP methods for habitat assessment
are derived from the Wisconsin Stream Clas-
sification Guidelines (Ball 1982) and Meth-
ods of Evaluating Stream, Riparian, andBi-
otic Conditions (Platts et al. 1983).
2.1.4 Ohio EPA
The (QFtEI) which is currently used by
Ohio EPA was developed by Rankin (1989,
1991, 1995). The development of the index
was based on six broad metrics: substrate,
instream cover, channel morphology, riparian
and bank condition, pool and riffle quality,
and gradient. These metrics are used because
they have been shown to be correlated with
stream fish communities (Rankin 1989).
2.1.5 MDNR-MBSS
The MDNR-MBSS qualitative habitat
assessment methods were developed by
-------
Kazyak (1995). Initial development was
based on the USEPA-RBP (Barbour and
Stribling 1991) and Ohio EPA's QHEI
(OEPA 1988, Rankin 1989). Additional
metrics were included in order to meet the
specific objectives of MDNR-MBSS (Roth
etal. 1997b).
the Physical Habitat (PHab) assessment. In
addition to the RHA and the PHab, supple-
mental habitat parameters are measured which
enable a more complete stream characteriza-
tion. These separate sets of metrics are not
combined into a single habitat assessment
score (Kaufmann and Robison 1998).
1 ' The RHA index contains 12 metrics
The primary habitat assessment tech- (Table 2-1) which are defined in Appendix A
niques used by USEPA-EMAP-SW are the (Kaufmann and Robison 1998). Each metric
Rapid Habitat Assessment (RHA) index and is assigned a score that ranges from 0 to 20
Table 2-1. The Metrics
Metric
Instream cover
Epifaunal substrate
Velocity /depth regimes
Frequency of riffles
Channel alteration
Bank condition
Embeddedness
Channel flow status
Riparian vegetation zone
and Scoring For The USEPA-EMAP-SW RHA Index.
Description3
Amount and diversity of useable fish cover
Presence and size of riffles and amount of cobble substrate present
Variety of velocity /depth regimes
Frequency of riffles and the variety of habitat
Type and amount of channel alteration
Bank stability and erosion
Percentage of gravel, cobble, and boulders that are covered
by sediment
The degree to which water fills the channel
; Width of the riparian zone and the presence of human disturbances
Score
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-20
0-20
Sediment deposition
Bank vegetation protection
Grazing/disruptive pressure
Complete descriptions are j
Degree of bar development and effect of sedimentation on the
channel 0-20
Percentage of stream bank surfaces covered by vegetation 0-20
Degree of vegetative disruption by mowing or grazing on the banks 0-20
;iven in Appendix A.
-------
points. Scores for each metric are determined
by matching observations made of the entire
sample segment with one of four established
ranking categories. These ranking categories
each contain descriptions of the respective
metric and the observer chooses the category
with the characteristics that most closely
matches the observations. Each of the four
ranking categories has a range of possible
scores (e.g., Optimal 20-16; Sub-Optimal 15-
11; Marginal 10-6; Poor 5-0). The index is
designed so that higher scores indicate a more
pristine habitat. A maximum index score of
240 points is possible.
The PHab assessment is made up of four
metrics, each with a number of sub-metrics
(Lazorchak et al. 1998). Many of these sub-
metrics are based on quantitative field mea-
surements while others are based on ranked
categories of field measurements (Table 2-2).
All PHab metrics and sub-metrics are defined
in Appendix A. The measurements made from
the PHab assessment are not incorporated into
an overall score.
In addition to the RHA index and PHab
assessment metrics, USEPA-EMAP-SW pro-
tocols measure five supplemental habitat pa-
rameters. Two of the habitat parameters, gen-
eral assessment and local anecdotal informa-
tion., are text descriptions (Table 2-3). The
three remaining parameters are based on
ranked categories of field measurements and
classified lists of field observations (Table 2-
3). No scores are assigned to any of the pa-
rameters. Like the measurements for the
PHab, it is unclear how these measurements
are used in analysis. It is possible that the clas-
sified habitat information could be used to
ground truth GIS data layers, but that is not
directed by the protocols.
The goal of the USGS-NAWQA stream
habitat protocol (Meador et al. 1993b) is to
measure habitat characteristics that are essen-
tial in describing and interpreting water chem-
istry and biological conditions in the differ-
ent types of streams studied by USGS-
NAWQA. A basic overview of this sampling
program is contained in section 1.2 of this
document.
The USGS-NAWQA assesses habitat
conditions in four spatial scales, basin, seg-
ment, reach, and microhabitat (Fitzpatrick et
al. 1998). The basin serves as a fundamental
ecosystem unit and an important perspective
from which to understand the characteristics
of streams. A segment is a length of stream
that has relatively homogeneous physical,
chemical, and biological properties. A reach
is a sampling unit within the segment. Physi-
cal, chemical, and biological data are collected
from the reach. The microhabitat scale pro-
vides information on patterns of relations be-
tween biota and habitat with a fine-scale reso-
lution. Procedures for collection of microhabi-
tat data (e.g., velocity, substrate type, and
depth) are described in the USGS-NAWQA
protocols for the collection of invertebrate and
algal samples (Cuffney et al. 1993a; Porter et
al. 1993) and will not be described in this
document.
-------
Table 2-2. Metrics And Scoring Used In The PHab Assessment.
Metric Sub-metric
Channel/riparian cross-section
Scoring
Thalweg profile
Large woody debris (LWD) tally
Thalweg depth
Wetted width
Bar width
Soft/Small sediment
Side channel presence
Channel unit code
Pool form code
Total number of LWD
Meters
Meters
Meters
Present/absent
Present/absent
1 1 categories
Seven categories
Sum
Class of each LWD
Slope
Bearing
Substrate size class
Bank angle
Undercut distance
Wetted width
Bankfull channel width
Exposed mid-channel bar width
Incised height
Bankfull flow height
Canopy density
Dominant canopy vegetation
Areal cover class of large trees
Areal cover class of small trees
Dominant understory vegetation
Area cover of understory
Areal cover of ground cover
Type of instream fish cover
Areal cover offish cover
Presence of human influences
Discharge
12 categories
Meters/kilometer
0-360°
11 categories
0-90°
Meters
Meters
Meters
Meters
Estimated meters
Estimated meters
Percent
Five categories
Five categories
Five categories
Five categories
Five categories
Five categories
Eight categories
Five categories
Four categories
Velocity2 Meters/
second
aThe velocity-area method, timed filling method, and neutral buoyant object method are used for large, medium, and
small streams, respectively.
Basin and segment assessments for fixed
or synoptic sites are conducted using GIS,
topographic maps, or aerial photographs
(Tables 2-4, 2-5). Site visits are needed to
collect the data for reach and microhabitat
assessments. At a subset of fixed sites, reach
data are collected from multiple reaches and
during the base flow stage of different years
(Table 2-6).
Basin characterization consists of geo-
morphic descriptors derived from USGS 7.5-
minute topographic maps, climate and poten-
tial runoff characteristics, streamflow charac-
teristics, and land-cover data from thematic
maps. Climate descriptors used by USGS-
-------
Table 2-3. Additional Parameters Used For The USEPA-EMAP-S W Protocols.
Parameter
Sub-parameter
Scoring
Watershed activities and disturbances observed
Reach characteristics
Waterbody character
General assessment
Residential
Recreational
Agricultural
Industrial
Stream management
Vegetation cover type
Land use/type
Water clarity
Pristine
Appealing
Wildlife
Vegetation diversity
Forest age class
Seven categories3
Four categories3
Six categories3
Eight categories3
Eight categories3
Six categories1"
Four categories11
Four categories1"
Five categories0
Five categories0
Text
Text
Text
Local anecdotal information
None
Text
Categories are examples of typical disturbances and each is recorded as none, low, moderate, or high.
bEach category is recorded as rare (<5%), sparse (5-25%), moderate (25-75%), or extensive (>75%).
"Categories are ranks, one to five, with one being the least pristine/appealing and five being the most pristine/appealing.
NAWQA include precipitation, temperature,
evaporation, and runoff. At least three types
of streamflow characteristics of a basin are
useful: estimated peak flow, flood volume,
and seven-day low flow for given recurrence
intervals. Thematic maps of ecoregion, physi-
ographic province, geology, soils, land use,
and vegetation are also used to describe a
basin. The Basinsoft computer program
(Harvey and Eash 1996) has been developed
by the USGS to quantify basin characteris-
tics (Table 2-4).
eters measured within these categories are
given in Table 2-5.
The USGS-NAWQA protocols measure
segment characteristics in the categories of
gradient, sinuosity, and water-management
features (Fitzpatricket al. 1998). The param-
The selection of the sampling reach is
based on four criteria, stream width, stream
depth, geomorphology, and local habitat dis-
turbance. In general, the reach length is de-
termined by multiplying the mean wetted
channel width by 20. For boatable streams,
recommended minimum and maximum
stream lengths are 500 and 1,000 m, respec-
tively. The minimum and maximum reach
lengths for wadeable streams are 150 and 300
m, respectively. If possible, the reach should
contain at least two examples of two habitat
types from the categories of pools, runs, or
riffles. At the beginning of data collection, the
-------
Table 2-4. The USGS-NAWQA Parameters Recorded For Basin Characterization.
Parameter Description
Units
Drainage area
Average annual runoff
Average annual air temperature
Average annual precipitation
Average annual evaporation
Basin length
Minimum elevation
Maximum elevation
Basin relief ratio
Drainage shape
Stream length
Cumulative perennial stream
length
Drainage density
Drainage texture
Entire stream gradient
Estimated flow characteristics
Delineated area enclosed by a drainage divide km2
Average amount of water contributed through runoff cm
Average ambient air temperature °C
Average precipitation cm
Average surface evaporation cm
Length of entire basin km
Minimum elevation within the basin m
Maximum elevation within the basin m
The difference between maximum and minimum
elevation divided by basin length m/km
Drainage area divided by the square of the basin length km2/km2
The distance from the headwaters to the site km
The cumulative length of all perennial streams and canals
in the basin km
The cumulative perennial stream length divided by the
basin area km'1
The number of crenulations on the most crenulated contours/
contour line divided by the basin perimeter length km
Difference between elevations at 85 and 10% of the
stream length divided by the distance between those
points m/km
Estimated peak flow, flood volume, and seven-day
low flow
general condition of the reach is noted and channel, bank, and riparian characteristics.
11 equidistant transects are established These parameters are given in Table 2-6.
throughout the reach. The transects are estab- „ , ,.,,.,< ;,.,, ,,, ,.> , : : .
lished so that habitat character!sties are statis- ": '' • ' •: ' "• '/'" ' -:: :. : ''-J ! •
tically represented within the reach and ob- ';;^Hv/bS ! v; M ; £;;%/#
server bias is eliminated. The parameters mea- The index SUggested by the USEPA-
sured within the reach provide information on Rgp consists of 13 metrics (Barbour et al.
-------
Table 2-5. The USGS-NAWQA Parameters Measured For Segment Characterization.
Parameter Description
Units
Straight-line length of the segment km
Length of the main channel through the segment km
Elevation at upstream and downstream boundaries m
Curvilinear channel length divided by segment length km/km
Upstream elevation minus downstream elevation,
divided by segment length m/km
Type(s) of water management feature(s) likely to
influence segment habitat 21 categories3
Stream order Numerical
Sum of the orders for all upstream tributaries Numerical
Sum of the orders for tributaries contributing to
the next downstream segment Numerical
The average of three representative gradient
calculations based on a cross-sectional profile of
the segment valley. m/km
aThe categories of water management features are bridge, diversion, return flow, stp > 5 (more than 5 sewage
treatment plants), ips > 5 (more than 5 industrial point sources), impoundment, low-head dam, natural lake, bank
stabilizer, tile drain, none, channelized, feedlot, sewage treatment, gw inflow, hydropower, industrial, mining, storm
sewer, thermal, and other.
Segment length
Curvilinear channel length
Upstream and downstream
elevation
Sinuosity
Segment gradient
Water management feature
Strahler stream order
Link
Downstream link
Valley sideslope gradient
1999) (Table 2-7) (see Appendix A). Three
of the metrics, embeddedness, frequency of
riffles, and velocity/depth combinations., are
used only at high gradient sites, and three of
the metrics, pool substrate, pool variability,
and channel sinuosity., are used only at low
gradient sites. Therefore, only ten metrics are
used at any one site. Each metric is assigned
a score ranging from 0 to 20 points (Table 2-
7). The metrics bank stability, bank vegeta-
tion protection, and riparian vegetation zone
width, are assigned a score ranging from 0 to
10 points for each bank (0 to 20 points for both
banks combined). Each metric is scored by
matching observations made of the entire sample
segment with one of four established ranking
categories. The chosen categories should con-
tain the characteristics that most closely match
the observations. Each of the four ranking cat-
egories has a range of possible scores (e.g.,
Optimal 20-16; Sub-Optimal 15-11; Marginal
10-6; Poor 5-0). Higher index scores are asso-
ciated with more pristine habitats. The maximum
index score is 200.
-------
Table 2-6. USGS-NAWQA Parameters ForReach Characterization.
Parameter Description
Mean channel width
The average of three representative measurements of
wetted channel width
Curvilinear reach length Length of reach measured through channel
Distance between The reach length divided by ten
transects
Curvilinear distance from Distance along the channel from a reference location to
site to reach ends the upstream and reference downstream reach boundaries
Reach water-surface
gradient
Geomorphic channel
units
Difference between the water surface elevations at both
ends of the reach, divided by the reach length
The length of all riffles, runs, and pools that make up
more than 50% of channel width are recorded
For each of the 11 transects
Habitat type Whether the transect is located in a riffle, run, or pool
Wetted channel width
Width from the left edge of the water to the right edge of
the water, excluding bars, shelves, or islands
Bankfull channel width Width from the top edge of the left bank to the top edge
of the right bank
Channel features
Aspect
Canopy angles
Width of channel bars, shelves, or islands
Compass heading of downstream flow
Sum of the angles from the middle of the transect to the
visible horizons on the left and right banks, subtracted
from 180°
Riparian canopy closure The portion of the overhead view that includes
vegetation
Units
Forthe reach
Stage Water level at a fixed point
Instantaneous discharge Flow of the stream
Channel modification Any channel modification at the reach is noted
m
L/s
Seven
categories3 at
reach
m
m
m
m
m/m
m and type
Three
categories
m
m
m and type
0 to 360°
0 to 180°
0 to 100%
(continued)
aChoose from 1) natural -woody debris pile, 2) overhanging vegetation (terrestrial), 3) undercut banks, 4) boulders,
5) aquatic macrophytes, 6) manmade structure, 7) too turbid to determine, or 8) none.
-------
Table 2-6. Continued
Parameter
Description
Units
Dominant riparian
land use
Bank angle
Bank height
Bank substrate
Bank vegetative cover
Bank erosion
Habitat cover features
Depth
Velocity
Dominant bed substrate
Embeddedness
Silt present
Land use within an approximate 3 0-m distance from
the top bank
Angle formed by the bank at the stream bottom
Vertical distance from channel bed to the top of the
bank
Type of dominant bank substrate
Visual estimation of percentage of bank covered in
vegetation
Presence or absence of bank erosion at each end of
transect
Presence or absence of any mineral or organic matter
that produces shelter for aquatic organisms
Water depth from water surface to stream bed
Velocity at 60% depth when depth is less than 1 m, or
average velocity at 20 and 80% depth when depth is
more than 1 m.
Type of dominant bed substrate
The estimated portion five large substrate particles that
are surrounded or covered by fine-grained sediment
The presence or absence of significant amounts of silt
12 categories'1
Oto900c
m
Ten categories4
0 to 100%
Present/absent
Present/absent
in eight categories6
m
m/s
Ten categories4
0 to 100%
Present/absent
bChoose from 1) cropland, 2)pasture, 3)farmstead/barnyard, 4) silviculture, 5) urban residential/commercial, 6)
urban industrial,!) rural residential, 8) right-of-way, 9) grassland, 10) shrubs/woodland, 11) wetlands, or 12)
other.
0 Measurement may be greater than 90° if the bank is undercut.
dChoose from one of 1) smooth bedrock'concrete/hardpan , 2) silt/clay/marl/muck/organic detritus, 3) sand (0.063-2
mm), 4)fine/medium gravel (2-16 mm), 5) coarse gravel (16-32 mm), 6) very coarse gravel (32-64 mm), 7) small
cobble (64-128 mm), 8) large cobble (128-256 mm), 9) small boulder (256-512 mm), or 10) large boulder/irregular
bedrock/irregular hardpan/irregular artificial surface (>512 mm).
-------
Table 2-7. The Metrics and Scoring used in the
USEPA-RBP'S Habitat Assessment Index.
Metric
Scoring
Epifaunal substrate/
available cover
Channel alteration
Bank stability
Channel flow status
0-20
0-20
0-10 (perbank)
0-20
Riparian vegetative zone width 0-10 (perbank)
Sediment deposition 0-20
Bank vegetative protection 0-10 (perbank)
Velocity/Depth combinations -
(high gradient) 0-20
Frequency of riffles -
(high gradient)
0-20
Embeddedness - (high gradient) 0-20
Pool substrate - (low gradient) 0-20
Pool variability - (low gradient) 0-20
Channel Sinuosity - (low gradient) 0-20
The QHEI (Rankin 1989) consists of
seven metrics, six of which are made up of
two to four scored sub-metrics (Table 2-8).
Each sub-metric is further divided into sub-
categories which are used to determine the
sub-metric scores (Tables 2-8,2-9). To com-
pute a final score for the QHEI, the scores of
the sub-metrics are summed and the scores of
the seven metrics are summed. The maximum
score for the QHEI is 100 (Table 2-8). A habi-
tat quality ranking scheme has been produced
by Ohio EPA based on the overall QHEI
score (Table 2-10). According to Rankin
(1989), three metrics, pool quality, channel
quality., and substrate quality, are consistently
correlated with the fish IBI in Ohio. In con-
trast, riparian zone quality is found to be less
correlated with the fish IBI in Ohio (Rankin
1989). Because the scores among the metric
categories are different, the overall index score
is weighted to give different metrics varying
importance. The metrics substrate and
instream cover, by virtue of the way they are
designed, can have a maximum value greater
than 20 points. If, as a result of the field mea-
surements they are scored above 20 points,
the final scores must be truncated to 20. Nine
additional observations that are either not
scored or not used in the final cumulative scor-
ing, are recorded while performing a QHEI.
These additional observations are given in
Table 2-11.
The habitat assessment methods used by
MDNR-MBSS include a habitat assessment
protocol very similar to the USEPA-RBP's
habitat assessment protocol and the USEPA-
EMAP-SW RHA. It also includes a group of
nine, generally quantitative, additional mea-
surements that are similar to a number of those
performed for the USEPA-EMAP-SW PHab
(Table 2-2). Currently, no method exists for
incorporating these separate measurements
into a single habitat assessment score.
The MBSS qualitative habitat assessment
method (Roth et al. 1997b) consists of 13
-------
Table 2-8. The Metrics, Sub-metrics, and Scoring Ranges for the Ohio EPA'S QHEI.
Metric
Substrate
Instream cover
Channel Morphology
Riparian zone/bank erosion
Pool/Glide Quality
Riffle/run quality
Gradient (scaled by ft/mi)
QHEI Overall
Sub-metric
Type
Quality
Type
Amount
Sinuosity
Development
Channelization
Stability
Flood plain width
Flood plain quality
Bank erosion
Pool maximum depth
Current type
Pool morphology
Depth
Substrate stability
Embeddedness
Sub-metric scoring
range
Oto22
-7 to 4
OtolO
Itoll
Ito4
Ito7
Ito6
Ito3
Oto4
Oto3
Ito3
Oto6
-4 to 4
Oto2
Oto4
Oto2
-Ito2
2 to 10
Maximum metric
score
20a
20a
20
10
12
8
10
100
alf the sum of the sub-metric scores exceeds 20, the metric score is truncated to 20.
Table 2-9. An Example of the Metric Scoring Method used by the QHEI.
Composite metric Sub-metric Scoring categories
Scores
Riffle quality
Riffle/ run depth Generally, >10 cm deep, >50-cm maximum depth
Generally, >10 cmdeep, <50-cmmaximumdepth
Generally, 5-10 cmdeep
Generally, <5cmdeep
Riffle/run substrate Stable (e.g., cobble, boulder)
Moderately Stable (e.g., pea gravel)
Unstable (e.g., gravel, sand)
Embeddedness None
Moderate
Low
Extensive
4
3
2
1
2
1
0
2
1
0
-1
-------
Table 2-10. Habitat Quality Rankings Developed
by the Ohio EPA for QHEI Score
Evaluation.
Habitat quality ranking
QHEI score range
Very Poor
Poor
Fair
Good
Very Good
Excellent
Extraordinary
0-40
41-50
51-60
61-70
71-80
81-90
91-100
Table 2-11. Observations Recorded in Addition to
the QHEI Parameters.
Observation
How recorded
Additional comments/
pollution impacts
Sampling gear/
distance sampled
Water clarity
Water stage
Canopy
Gradient
Length, width, and
maximum depth at
sampling sites
Stream diagram:
cross sections
Stream map
Text
Type of fishing gear
used/length of
sampling reach
Clear, stained, or
turbid
Meters
Percent of sampling
site not shaded or
covered by woody
bank vegetation
Very low, low, low-
moderate, moderate,
moderate-high, high,
or very high
Meters
Two or three
drawings of the
stream cross section
Sketch of the entire
sampling section
metrics (Table 2-12, Appendix A). Each
metric is scored by matching observations
made of the sample segment to one of four
possible ranking categories that best de-
scribes observed conditions. Each of the
four ranking categories has a range of pos-
sible scores. The method is designed so that
higher scores indicate more pristine habi-
tats. Nine of the metrics are evaluated in
this fashion and assigned a score ranging
from 0 to 20 points. However, three of the
metrics, embeddedness, channel flow sta-
tus, and shading are given percentage
scores and one of the metrics, riparian
buffer, is given a score in meters (Table 2-
12). No total index score is computed for
the MDNR-MBSS habitat assessment. In
addition to the qualitative habitat assess-
ment metrics (Table 2-12), MDNR-MBSS
makes these quantitative habitat assessment
measurements:
• Maximum depth
• Stream gradient
• Wetted width
• Straight-line segment length
• Overbank flood height
• Discharge
The methods of the various agencies dif-
fer in the type, number, and scoring of metrics.
This section addresses these differences and
the similarities among the five methods.
-------
Table2-12. Metrics used in the MDNR-MBSS Qualitative Habitat Assessment Method.
Metric Description How scored
Instream habitat structure
Epifaunal substrate
Velocity/depth diversity
Pool/glide/eddy quality
Riffle quality
Channel alteration
Bank stability
Aesthetic rating
Remoteness rating
Embeddedness
Channel flow status
Shading
Riparianbuffer
Total
Perceived value of habitat based on its type and structure 0-20
Amount/variety of hard, stable substrates for benthic
invertebrates 0-20
Variety of velocity/depth regimes 0-20
Variety and spatial complexity of slow or still water habitat 0-20
Complexity and functional importance of riffle/run habitat 0-20
Degree and type of channel alteration 0-20
Presence of vegetation or other bank stabilizing material 0-20
Visual appeal of site, presence of human refuse, degree of
channelization, and vegetation disturbance 0-20
Presence of detectable human activity and accessability of site 0-20
Percentage of stream gravel, cobble, and boulder surface area
not surrounded by fine sediment 0-100%
Percentage of stream channel that has water 0-100%
Percentage of the site that is shaded 0-100%
Minimum width of vegetated buffer (50m maximum) meters
0-180
The USEPA-EMAP-SW RHA and the
USEPA-RBP indices are very similar in their
composition. Ten of the 12 RHA index
metrics are either very similar or directly com-
parable to USEPA-RBP metrics.
These ten metrics are:
• epifaunal substrate
• velocity/depth regimes
• frequency of riffles
• channel alteration
• bank condition or stability
• embeddedness
• channel flow status
• riparian vegetation zone
• sediment deposit!on
-------
• bank vegetation protection
The RHA index has two metrics,
instream cover and grazing/disruptive pres-
sure, that are not included in the USEPA-RBP
index and the USEPA-RBP index has three
metrics, channel sinuosity, pool variability,
and pool substrate, that are not used by the
RHA index. The criteria used to evaluate the
two metrics, instream cover and epifaunalsub-
strate, by the RHA index are combined into
one metric, epifaunal substrate, by the
USEPA-RBP index. Whereas all 12 of the
RHA index metrics are scored for every
sample stream segment, only ten of the 13
USEPA-RBP index metrics are scored for a
sample segment. Three of the USEPA-RBP
metrics, embeddedness, frequency of riffles,
and velocity/depth combinations, are used
only at high gradient sites, and three of the
USEPA-RBP metrics, pool substrate, pool
variability, and channel sinuosity, are used
only at low gradient sites. Finally, one major
difference between the USEPA-RBP index
and the overall USEPA-EMAP-SW habitat
assessment methods is that the USEPA-
EMAP-SW habitat assessment method in-
cludes two additional components, the PHab
and additional assessment parameters (Tables
2-2,2-3). These additional elements provide
quantitative measurements of parameters such
as channel sinuosity and discharge that are
qualitatively assessed by the USEPA-RBP
index.
Maryland's MBSS qualitative habitat
assessment protocols were partially derived
from the USEPA-RBP index and are, there-
fore, similar to both the RHA and USEPA-
RBP indices (Table 2-12). The MDNR-
MBSS qualitative habitat assessment proto-
cols have seven metrics, epifaunal substrate,
velocity/depth diversity, channel alteration,
bank stability, embeddedness, channel flow
status, and riparian buffer, with similar or
identical evaluation criteria to USEPA-RBP
metrics. Six metrics, instream cover, pool/
glide/eddy quality, riffle quality, shading, aes-
thetic rating, and remoteness rating, are in-
cluded in the MDNR-MBSS qualitative habi-
tat assessment protocols, but not in the
USEPA-RBP index. Also, the USEPA-RBP
index contains six metrics, pool substrate,
pool variability, frequency of riffles, sediment
deposition, bank vegetation protection, and
channel sinuosity, that are not used in the
MDNR-MBSS qualitative habitat assessment
protocols. As with the RHA index, the
MDNR-MBSS qualitative habitat assessment
separates the evaluation criteria used in
USEPA-RBP epifaunal substrate metric into
two metrics, instream cover and epifaunal
substrate, and all of the metrics are scored for
every stream segment, regardless of the gra-
dient level. Unlike the RHA and the USEPA-
RBP, which only evaluate the riparian buffer
to 18 m on each bank, the MDNR-MBSS
qualitative protocols measure the riparian zone
to a distance of 50 m on each bank. The
MDNR-MBSS protocols, like those of the
USEPA-EMAP-SW, make a number of ad-
ditional quantitative measurements of the
stream segment physical features (Section
2.6.1) as well as categorizing the adjacent land
use. The data from the two components of
the MDNR-MBSS protocols are not incor-
porated into an overall habitat score.
MBSS is unique in that it is the only pro-
gram that identifies instream submerged
aquatic vegetation (SAV), emergent aquatic
-------
vegetation (EAV), and riparian vegetation
to species (USEPA-EMAP-SW uses veg-
etation categories and the Ohio EPA QHEI
only addresses vegetation in terms of per-
cent cover). Aquatic plants are also not
sampled concomitantly with the standard
Ohio EPA stream habitat and biotic assess-
ment sampling.
The Ohio EPA QHEI is the most
unique of the indices reviewed. Substantial
differences exist between the scoring sys-
tem and metric definitions in the QHEI and
in the other four indices. The scoring cat-
egories of the QHEI metrics are not
grouped like the other indices, but rather
individual scores are assigned to numerous
scoring categories which are part of metrics
or sub-metrics (Table 2-8). Each metric and
sub-metric is uniquely designed and con-
sists of varying numbers of scoring catego-
ries. The individual scoring categories range
in the number of points assigned to each
category and are, therefore, not equally
weighted. Some of the QHEI metrics can
have total scores greater than the maximum
scores permitted for those metrics. If the
total exceeds the maximum score for the
metric, the score is truncated to the maxi-
mum score value. The QHEI is similar to
the USEPA-RBP, RHA methods, and the
MDNR-MBSS assessment methods in that
it qualitatively assesses some of the major
features of stream structure related to the
quality of stream habitat. These structural
features include substrate, instream cover,
physical channel features, and flow regime.
Unlike the other protocols, the QHEI has
established habitat quality ranking stan-
dards based upon index scores.
One of the primary differences between
the methods used to characterize habitat for
the USGS-NAWQA and those used by the
other four agencies is that NAWQA has ex-
tensive characterization of the habitat on four
spatial scales, basin, segment, reach, and mi-
crohabitat (Tables 2-4, 2-5, and 2-6). The
protocols for USGS-NAWQA are unique also
because there is no formal index score calcu-
lated. The program instead focuses on the use
of repeatable, quantitative data in order to pro-
duce nationally-consistent stream quality
evaluations and the use of additional qualita-
tive data for the generation of qualitative in-
dices where applicable (Fitzpatrick et al.
1998).
Contrasting the assessment methods
used by USEPA-RBP and USEPA-EMAP-
SW and those used by Ohio EPA and
MDNR-MBSS reveals a number of differ-
ences between these sampling methods. Dif-
ferences exist at the broad scale in dealing with
study site identification and assessment of the
status of the aquatic resources. Also, differ-
ences exist at the local scale in the methods
used to collect data. At the broad scale, iden-
tification of the MDNR-MBSS and USEPA-
EMAP-SW sampling sites is accomplished
using statistically-based sampling designs.
However, no statistical designs are used by
Ohio EPA or USEPA-RBP to identify the
-------
study segments. In its first nine-year cycle,
USGS-NAWQAused a common sampling
design for 59 of the most environmentally
significant watersheds in the nation. It uses
a design on four spatial scales, but is not
statistically based.
The USEPA-EMAP-SW sampling
framework consists of hexagons placed
over a grid map of the contiguous United
States with 12,500 points. Using statistical
probability methods, approximately 800
lakes and 800 streams are chosen from 25%
of the grid hexagons each year. Therefore,
this method has a four-year sampling cycle
(Overton et al. 1991). In order to ensure an
adequate characterization of larger lakes and
streams, sites are randomly selected from
established size strata.
MDNR-MBSS uses a similar approach
which is designed to provide full coverage
of the state's 18 drainage basins over a pe-
riod of three years. Approximately 300 non-
overlapping stream segments are randomly
selected using a lattice sampling method and
are sampled each year. Within a stream or-
der, the number of segments sampled per
basin is proportional to the number of stream
miles in the basin.
In contrast to the USEPA-EMAP-SW
and MDNR-MBSS methods, Ohio EPA
uses a five-year cycle to monitor Ohio's
aquatic resources. Each year of the five-year
cycle focuses intensively on the biological,
chemical, and physical habitat data found
within a chosen basin. Study sites are iden-
tified based on criteria such as the potential
to be threatened by local impacts or their
potential for harboring unique or critical
aquatic habitat or biota. Unlike the method
used by the Ohio EPA, the methods used
by USEPA-EMAP-SW and MDNR-
MBSS, allow robust population inferences
to be made and ensure that the sites represent
the spatial distribution of lakes and streams
within the study areas.
At the local scale, a number of differ-
ences exist between the sampling methods
used by the reviewed programs. The sampling
reach length for the USEPA-EMAP-SW as-
sessment is generally 40 times the stream
channel width and in the USGS-NAWQA
sampling method, the reach length is gener-
ally 20 times the stream channel width. In
contrast, the USEPA-RBP, Ohio EPA and
MDNR-MBSS procedures use fixed sam-
pling reach lengths. USEPA-RBP and
MDNR-MBSS uses a sampling reach of 75
m for wadeable streams. The sampling reach
length for Ohio EPA is generally a 500-m
segment if the stream is boatable or a 150 to
200-m segment if it is a wadeable stream.
Quantitative thalweg profile measure-
ments are made using the USEPA-EMAP-
SW and MDNR-MBSS protocols. Quantita-
tive measurements of reach average and maxi-
mum depth, and pool/glide/riffle/run length,
width, and depth are made using the Ohio
EPA method. Between 100 and 150 indi-
vidual thalweg profile measurements are
made along the sample reach using the
USEPA-EMAP-SW protocol, as opposed to
3, (one each at 0- 25, 50, and 75 m along the
sample segment), for the MBSS index and
11 sets of thalweg measurements per sample
-------
reach using the USGS-NAWQA protocol.
Clearly, the sampling density for quantitative
measurements is much greater for the
USEPA-EMAP-SW index than for the other
programs' indices. Also, depending on the
index used, the specific habitat and location
sampled, the assessment made by the USEPA-
EMAP-SW may be based on a larger seg-
ment of the stream than the assessments made
by the other programs.
have on the scoring of metrics associated with
all of the assessment methods. For instance,
life history traits such as fish spawning and
insect emergence or changes in stream flow
associated with seasonal or short term patterns
of precipitation, can dramatically influence the
presence or absence of organisms and affect
other estimates and evaluations based on the
timing of single measurements of physical and
chemical parameters.
Sampling season is an important factor
to consider because of the influence it can
-------
by
Bradley C. Autrey and Joseph P. Schubauer-Berigan
This section summarizes and evaluates
the surface water column chemistry assess-
ment methods for USEPA-EMAP-SW,
USGS-NAWQA, USEPA-RBP, Ohio EPA,
and MDNR-MBSS. The basic objective of
surface water column chemistry assessment
is to characterize surface water quality by
measuring a suite of analytes. Water chemis-
try data are measurements of chemical con-
centrations and physical properties of
stream water. Because each program has a
unique set of objectives, each suite of analytes
is also unique. A summary of the analytes
used by the five reviewed programs is pre-
sented in Table 3-1. Figure 3-1 shows a mem-
ber of a field crew filling a cubitainer with a
water sample that will be used in water chem-
istry analysis.
In addition to surface water column
samples, the USEPA-EMAP-SW and USGS-
NAWQA programs have additional protocols
which are used to analyze the quality of
ground water and use bed sediment and tis-
sue analyses to further assess surface water
quality. These additional analyses are impor-
tant for the programs' understanding of water
quality and an integral part of their water qual-
ity assessment programs. However, only sur-
face water column sampling and analyses are
addressed in this document.
The objectives of the USEPA-EMAP-
SW water chemistry protocols are to deter-
mine the acidity/alkalinity of the water, to
characterize the trophic condition of the
stream, to ascertain the presence or absence
of chemical stressors, and to classify the wa-
ter chemistry type. At each sampling reach,
water chemistry measurements are made in
situ and water samples are collected for labo-
ratory analysis (Table 3-1). One 4-L
cubitainer and two 60 ml syringes are filled
from a flowing portion of the stream, labeled,
and stored in a cooler with ice. These samples
are shipped to the analysis laboratory within
24 hours of collection (Herlihy 1998).
The in situ measurements include spe-
cific conductance, dissolved oxygen, and tern-
-------
Table 3-1. Water Chemistry /Water Quality Measurements made by USEPA-EMAP-SW, USGS-NAWQA,
USEPA-RBP, Ohio EPA and MDNR-MBSS in Conjunction with Monitoring and Assessment3
Analytes
USEPA-
EMAP-SW
USEPA-
RBP
USGS-
NAWQA1'
Ohio
EPAC
MDNR-
MBSS
Physical analytes
Color L
Conductivity/Specific conductance F
Dissolved oxygen (DO) F
Residue (total, filtered, non-filtered)
Stream type
Temperature (C) F
Total dissolved solids (TDS)
Total suspended solids (TSS) L
Turbidity L
Water odors
Demand analytes
Biological OxygenDemand (BOD)
Chemical oxygen demand (COD)
Nutrient analytes
Acid neutralizing capacity (ANC) L
Alkalinity L
Bicarbonate L
Carbonate L
Chlorine, residual
Dissolved inorganic carbon (DIG) L
Nitrogen as ammonia L
Nitrogen as nitrate (NO3) L
Nitrogen as nitrite (NO2)
Nitrogen as nitrate-nitrite NO3-NO2
Nitrogen, total L
pH L
Silica L
Sulfate L
Phosphorus, ortho
Phosphorus, total L
Phosphorus, total dissolved
Organic analytes
Dissolved organic carbon (DOC) L
Suspended organic carbon (SOC)
Total organic carbon (TOC)
F
F
Fd
F
Fd
F,L
F
F
L
F,L
F
F
L
L
L
L
L
F,L
L
L
L
L
L
L
L
L
F,L
F
L
L
L
L
L
L
L
F,L
L
L
L
F,L
F
L
L
L,F
L
L
(continued)
a L indicates analysis takes place in the laboratory, and F indicates analysis takes place in the field.
b These are the analytes used in USGS-NAWQA's basic fixed-site analysis.
0 These analytes were derived from those taken to assess stream quality in Ohio EPA (1995).
d These are estimated measurements.
-------
Table 3-1". Continued
Analytes
USEPA-
EMAP-SW
USEPA-
RBP
USGS-
NAWQAb
Ohio
EPAC
MDNR-
MBSS
Organic waste analytes
Water surface oils
Oil and grease
Phenolics, total
Metal analytes
Aluminum, total/dissolved
Aluminum, inorganic monomeric
Aluminum, PCV reactive
Arsenic
Barium
Beryllium
Boron
Cadmium
Calcium
Chromium
Cobalt
Copper
Hardness
Iron
Lead
Lithium
Magnesium
Manganese
Mercury
Molybdenum
Nickel
Selenium
Silver
Strontium
Vanadium
Zinc
Bacteria analytes
E. coli
Fecal conforms
Fecal streptococci
Ionic analytes
Anion Deficit (C-A)
Anions, estimated organic
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
(continued)
a L indicates analysis takes place in the laboratory, and F indicates analysis takes place in the field.
b These are the analytes used in USGS-NAWQA's basic fixed-site analysis.
0 These analytes were derived from those taken to assess stream quality in Ohio EPA 1995.
d These are estimated measurements.
-------
Table 3-1". Continued
Analytes
Anions, sum
Cations, base sum
Cations, sum
Chloride
Fluoride
Ionic strength
Potassium
Sodium, total
USEPA- USEPA-
EMAP-SW RBP
L
L
L
L
L
L
L
USGS-
NAWQAb
L
L
L
L
Ohio
EPAC
L
L
L
MDNR-
MBSS
Radio-chemicals
Gross alpha
Gross beta
Radium-226
Tritium
Uranium
L
L
L
L
L
a L indicates analysis takes place in the laboratory, and F indicates analysis takes place in the field.
b These are the analytes used in USGS-NAWQA's basic fixed-site analysis.
0 These analytes were derived from those taken to assess stream quality in Ohio EPA 1995.
Figure 3-1. A member field crew member fills a cubitainer with water that will be used in water
chemistry analysis.
-------
perature. The samples from the two 60 ml
syringes are used to measure pH, dissolved
inorganic carbon (DIC), and monomeric alu-
minum species. The bulk 4-L sample is used
to measure the major ions, nutrients, total
iron, total manganese, turbidity, and color
(Herlihy 1998).
The USGS-NAWQA program has three
basic levels of water chemistry analyses, ba-
sic fixed-site assessment, intensive fixed-site
assessment, and water column synoptic stud-
ies. The intensity of sampling and the analytes
measured differ among these three levels.
concentrations, high and low flows and con-
centrations that occur less often during the
two-year sampling period have a small chance
of being sampled. All samples are flow
weighted and cross-sectionally integrated by
standard USGS methods. Complete descrip-
tions of sample collection and processing
methods are provided by Shelton (1994).
Each time a basic-fixed site is sampled,
field measurements (e.g., water temperature,
pH, conductivity, DO) are made, and samples
are submitted to the laboratory for analyses
of a national target list of suspended sediments,
dissolved solids, major ions and metals, nu-
trients, and dissolved and suspended organic
carbon. These analytes (Table 3-1) are selec-
tively augmented in some study units as re-
quired to meet specific local needs (Gilliom
etal. 1995).
Data from basic fixed-site sampling are
used for assessing temperature, specific con-
ductance, suspended sediment, major ions and
metals, nutrients, and organic carbon. The
sampling strategy at each basic-fixed site con-
sists of three types of sampling activities, con-
tinuous monitoring, fixed-interval sampling,
and extreme-flow sampling, each of which is
conducted for at least two years.
Continuous monitoring is conducted by
automated gaging stations for the entire sam-
pling period. Fixed-interval sampling is the
collection of samples at regular time intervals
for laboratory analyses. The minimum and
most common sampling frequency is monthly
during the minimum two-year period of op-
eration. Extreme flow sampling usually con-
sists of four to eight supplemental samples per
year. Al though fixed-interval sampling pro-
vides data for the most common flows and
Intensive fixed-site assessments are con-
ducted for one year and are the same as basic
fixed-site assessments except for more fre-
quent sampling and the addition of dissolved-
pesticide analyses (Table 3-2). The goal of
intensive fixed-site sampling is to accurately
assess the dissolved pesticides in the stream
through relatively high-frequency sampling at
a few carefully chosen sites during key peri-
ods (Gilliom etal. 1995).
Water-column synoptic studies are short-
term investigations designed to address wa-
ter-quality issues specific to a study unit or
region (two to three study units). Every wa-
ter-column synoptic study is custom designed
to provide more specific water-quality infor-
-------
Table 3-2. Dissolved Pesticides Analyzed by USGS-NAWQA in Addition to Basic Fixed Site Analytes in
Conducting Intensive Fixed-Site Assessment.
Category3
Pesticides
Amides
Carbamates
Alachlor, Metolachlor, Napropamide, Pronamide, PropachlorPropanil
Aldicarb, Aldicarb sulfone b, Aldicarb sulfoxide b, Butylate, Carbaryl,
Carbofuran, 3-Hydroxyb, EPTC, Methiocarb, Methomyl, Molinate, Oxamyl,
Pebulate, Propham, Propoxur, Thiobencarb, Trillate
2,4-D (acid), Dichlorprop (2,4-DP), 2,4-DB, MCPA, MCPB, Silvex (2,4,5-TP),
2,4,5-T,Triclopyr
Benflurahn, Ethaflurarin, Oryzalin, Pendimethalin, Trifluralin
Chlorothalonil, Dacthal (DCPA), p,p'-DDE, Dichlobenil, Dieldrin, alpha-
HCHb,gamma-HCH
Azinphos-methyl, Chlorpyrifos, Diazinon, Disulfoton, Ethoprop, Fonofos,
Malathion, Methyl parathion, Parathion, Phorate, Terbufos
cis-Permethrin
Atrazine, desethylb, Cyanazine, Metribuzin, Prometon, Simazine
Bromacil, Terbacil
Fenuron, Diuron, Fluometuron, Linuron, Neburon, Tebuthiuron
Acifluorfen, Bentazon, Bromoxynil, Chloramben, Clopyralid, Dicamba, 2,6-
Diethylaniline b, Dinoseb, DNOC, 1-Napthol b, Norflurazon, Picloram,
Propargite
a Some of the analytes listed may be deleted or qualified depending on method performance for ambient
samples.
b Degradation products
Chlorophenoxy herbicides
Dinitroanilines
Organochlorides
Organophosphates
Pyrethroids
Triazine herbicides
Uracils
Ureas
Miscellaneous
mation than fixed-site data. Most water-col-
umn synoptic studies are conducted in the
second and third years of the three-year in-
tensive data-collection phase. This is done
after initial results from the first year of sam-
pling can be combined with existing data to
guide the study design (Gilliom et al. 1995).
estimated parameters are each assigned to a
category. The categories for these parameters
are given in Table 3-3 (Barbour et al. 1999).
The objective of the USEPA-RBP is to
recommend water sampling methods which
will provide a brief and easily-obtained analy-
sis of water chemistry. The protocols recom-
mend a water-quality assessment that can be
made entirely in the field. The suggested as-
sessment includes four quantitative measure-
ments, temperature, dissolved oxygen, pH,
and conductivity and four estimated measure-
ments, stream type, water odors, water sur-
face oils, and turbidity (Table 3-1). The four
The objective of the Ohio EPA water
sampling guidelines is to provide data which
can be used to interpret the quality or condi-
tion of the stream being sampled. The analytes
measured by the Ohio EPA are given in Table
3-1. Because water quality characteristics are
not uniform between sites, the Ohio EPA con-
siders the mixing conditions of the stream
when designing a sampling regime. The Ohio
EPA makes a series of conductivity and tem-
perature measurements to check the mixing
conditions in the stream and those mixing
conditions determine the types of samples that
will be taken (OEPA 1988).
-------
Table 3-3. Categories Available for Scoring the
Estimated Parameters of the USEPA-
RBP'S Recommended Water Quality
Assessment
Parameter
Categories
Stream type
Water odors
Water surface oils
Turbidity3
Coldwater
Warmwater
Normal
Sewage
Petroleum
Chemical
None
Other (with notation)
Slick
Sheen
Globs
Flecks
None
Clear
Slightly turbid
Turbid
Opaque
a In addition to the given categories, the color of
the water is also noted for this parameter.
The Ohio EPA uses two primary types
of samples, grabs and composites. Grab
samples are individual samples gathered over
a period of time not exceeding 15 minutes. If
a stream is evenly mixed, the grab samples
can be integrated. Integrated grab samples can
be either horizontally integrated samples or
vertically integrated samples. The horizontally
integrated samples are mixtures of grab
samples gathered from different points across
the width of the stream and vertically inte-
gratedsamples are mixtures of grab samples
gathered from different depths of the stream.
Composite samples are mixtures of dis-
creet samples taken at equal time intervals.
These samples allow variable water quality
characteristics to be averaged over a period
of time. The length of time is determined by
factors such as the intended use of the data
and the specific characteristics of the stream
being sampled (OEPA 1988).
Before grab samples are taken, the mix-
ing condition of the stream is determined. If
the mixing condition cannot be determined,
samples are taken near the stream sample
where the velocity and turbulence are the
greatest. If the stream is very wide/deep or if
it is incompletely mixed, integrated grab
samples must be taken.
The individual collecting the water
sample should wade into the stream or, if col-
lecting from a bridge, use a bucket and a rope.
The collecting bucket should be rinsed with
ambient water. Water is collected while fac-
ing upstream and from the top 40% of the
water column. Enough water is collected to
fill two one-quart cubitainers and a one-gal-
lon cubitainer. Before the cubitainers are filled,
they are expanded and rinsed with a small
amount of the sample. After they are filled,
they are labeled, excess air is removed, and
they are stored at 4° C until preserved. Samples
are preserved by adding an ampule of sulfu-
ric acid, nitric acid and sodium hydroxide
(OEPA 1988).
Composite samples are taken from a
single point in the stream and can be collected
-------
with automatic samplers or manually. Auto-
matic samplers are preferred because they can
increase the frequency and regularity of the
samples taken. Samples can either be col-
lected directly into a composite jar or collected
as aliquots. If collected as aliquots, samples
are mixed in a compositor that has been rinsed
with stream water and transferred into
cubitainers. If it is not possible to set an auto-
matic sampler, manual samples are taken.
Manual samples are collected using the same
basic procedure as grab samples. The samples
are collected in aliquots that are the propor-
tion of the total sample needed. For example,
if 1,000 ml are being collected in eight
aliquots, each aliquot should be 125 ml
(OEPA 1988).
L bottle for all analytes except pH. A water
sample for pH is collected in a syringe so that
air bubbles can be expunged. Samples are
stored on ice and shipped to the analysis labo-
ratory within 48 hours. Chemical analyses are
conducted as described in the Handbook of
Methods for Acid Deposition (U.S. EPA
1987). The exception is that the sample for
ANC analysis, is reduced in volume to 40 ml
for easier handling (Roth et al. 1997b).
During the summer, in-situ measure-
ments are made ofDO,pH, temperature, and
conductivity. These additional measurements
are made in order to further characterize wa-
ter quality conditions that may influence bio-
logical communities. These measurements are
taken at an undisturbed portion of the stream
using calibrated electrode probes (Roth et al.
1997b).
When sampling water for bacteria analy-
sis, a sample is collected in four one-ounce
bottles containing sodium thiosulfate crystals
and topped with foil-lined screw caps. When
sampling water to test for oil and grease, a
sample is collected in a 1-L widemouth glass
j ar with a Teflon or aluminum foil lined screw
cap. When sampling near an area that may
exceed limits for acidity/alkalinity within a
given time period, measurements should not
come from composite samples (OEPA 1988).
During the spring, water samples are
collected from each site and analyzed forpH,
ANC, conductivity, sulfate, nitrate, and DOC.
At each site, a grab sample is collected in a 1-
Of the five programs reviewed, all ex-
cept USEPA-RBP collect water samples for
laboratory analyses in addition to making
water-chemistry measurements in the field.
The USEPA-RBP recommends field mea-
surements of eight parameters only and no
laboratory analyses. This allows the USEPA-
RBP to meet its obj ective of suggesting meth-
ods for the rapid assessment of stream qual-
ity. The USGS-NAWQA program uses au-
tomatic samplers at gaging stations. There-
fore, that program is able to take a large num-
ber of samples over fixed increments of time.
The remaining programs rely heavily on
samples gathered during a small number of
visits to the field. Based on sampling meth-
-------
ods and including pesticide analysis, the
USGS-NAWQA program conducts the most
thorough evaluation of water chemistry.
Of the 60 total analytes measured, only
four, conductivity, DO.pH, and temperature,
are common to all five programs.
The Ohio EPA and USGS-NAWQA
monitoring programs measure more contami-
nants than the other programs. Ohio EPA
monitors bacteria (i.e., fecal coliforms and
fecal strep) and USGS-NAWQA monitors for
the presence of a suite of pesticides. The
USEPA-EMAP-SW measures a large num-
ber of analytes, including several ionic
analytes not measured by other programs. The
MDNR-MBSS and the USEPA-RBP each
measure only eight analytes. Measuring a
small number of analytes allows these pro-
grams to quickly, if not thoroughly, assess the
chemical and physical properties of the
streamwater.
-------
by
Joseph Flotemersch, Susanna DeCelles, and Bradley C. Autrey
The term periphyton refers to the proto-
zoa, fungi, bacteria, mosses, and algae that
are attached to or are in close proximity to the
substrata of an aquatic system. However, pe-
riphyton surveys that are used to assess stream
quality deal primarily with microscopic algae
(microalgae) assemblages (Rosen 1995). Pe-
riphyton are useful indicators of stream qual-
ity because they reproduce rapidly, have short
life cycles and their assemblages are there-
fore very responsive to disturbances. In addi-
tion, most periphyton taxa can be identified
to species by experienced phycologists, and
tolerance or sensitivity to specific changes in
environmental condition are known for many
species (Rott 1991, Dixit et al. 1992).
Phytoplankton are microalgae that are
buoyantly suspended in the water column of
aquatic systems. They are passively trans-
ported by currents and turbulent mixing, and
reflect water quality conditions of the water
mass in which they occur (Clesceri et al.
1989). Phytoplankton are especially valuable
as indicators of water quality when large ar-
eas are assessed, when resources are limited,
or when phytoplankton are an important part
of the ecosystem being studied.
Diatoms are a type of microalgae that
are often the focus of phytoplankton and pe-
riphyton assessments. They are useful indi-
cators of biological condition because they are
found in all aquatic habitats.
The USEPA-EMAP-SW program de-
fines periphyton as algae, fungi, bacteria, pro-
tozoa, and associated organic matter affiliated
with the channel substrates (Hill 1998). Per-
iphyton are useful indicators of environmen-
tal conditions because they respond rapidly
and are sensitive to a number of anthropo-
genic disturbances, including habitat destruc-
tion and contamination by nutrients, metals,
-------
herbicides, hydrocarbons, and acids. Periphy-
ton indices of stream condition are being de-
veloped based on the composite indices for
biotic integrity, ecological sustainability, and
trophic condition (Hill 1999). The composite
indices will be calculated from measured or
derived indices that include species richness,
species diversity, cell density, ash free dry
mass (AFDM), chlorophyll content, and en-
zyme activity acid/alkaline phosphatase ac-
tivity (APA), which individually indicate eco-
logical condition in streams. The metrics as-
sociated with the periphyton indicators are
summarized in Table 4-1 (Hill 1998).
At each stream reach, composite index
samples are collected from erosional and
depositional habitats located at each of the
nine interior transects (transects B through J;
See Section 1.2.1). Samples are collected from
the sampling point assigned (left, center, or
Table 4-1. USEPA-EMAP Proposed
Periphyton Indicators Of Stream Condition And
Associated Parameters
Indicator and
Description
Associated
Parameters
Species composition
Cell density (cells/cm2)
Chlorophyll (ug./cm2)
Standing stock
(mgAFDM/cm2)
Phosphatase activity
(mmol/gAFDM)
Species diversity,
evenness, auteco-
logical indices
Abundance
Standing crop,
productivity, trophic
status, autotrophic
index
Productivity,
trophic status
Community activity
(function)
right; section 1.1) during the layout of the
reach. In erosional habitats, a sample of rock
or wood substrate is removed from the stream.
Attached periphyton are dislodged from a 12-
cm2 area on the upper surface of the substrate
with a stiff-bristled toothbrush for 30 seconds.
Figure 4-1 shows a member of a field crew
dislodging periphyton using the EMAP tech-
nique. Dislodged periphyton are then washed
into a 500-ml bottle using stream water. In
depositional habitats., a 12-cm2 area of soft
sediment is defined and the top 1 cm from
that area is vacuumed into a 60-ml syringe.
The erosional habitat samples from the nine
transects are compiled into an erosional habi-
tat composite index sample and the deposi-
tional habitat samples from the nine transects
are compiled into a depositional habitat com-
posite index sample (Hill 1998).
Four different types of laboratory
samples are prepared from each of the two
composite index samples, an ID/enumeration
sample, a chlorophyll sample, a biomass
sample, and an acid/alkaline phosphatase ac-
tivity (APA) sample.
ID/enumeration samples are used to de-
termine taxonomic composition and relative
abundances. These samples are preserved in
10% formalin. Chlorophyll samples are pre-
pared by filtering a 25-ml aliquot of each com-
posite index sample through a 0.4 to 0.6-|im
glass fiber filter. Biomass samples are used
for AFDM analysis. The preparation of fil-
ters for biomass samples is the same as for
chlorophyll samples except that the filters have
been combusted, desiccated, rehydrated, dried
and weighed. The APA samples are used to
measure enzymatic activity. They are pre-
pared by freezing 50-ml subsamples of each
-------
Figure 4-1. A member of afield crew dislodges attached periphyton using the EMAP-SW method.
composite index sample. Analytical method-
ologies are summarized in Table 4-2 (Hill
1998).
4.2 USGS-NAWQA
Algae Assessment
Program
Benthic algae and phytoplankton com-
munities are characterized in the USGS-
NAWQA program as part of an integrated
physical, chemical, and biological assessment
of the Nation's water quality (Porter et al.
1993).
4.2.1 Sample Collection
Periphyton may be collected from natu-
ral substrates by scraping, brushing, siphon-
ing, or other methods appropriate to each
microhabitat. Porter et al. (1993) describe
methods for collecting periphyton from mi-
crohabitats. The collection of phytoplankton
samples, or the use of artificial substrates for
collecting periphyton samples, are listed as
options for collection efforts in large boatable
streams and rivers to meet specific program
objectives. Estimates of algal biomass (chlo-
rophyll content and ash-free dry mass) are
optional measures that may be useful for in-
terpreting water-quality conditions. The char-
acter of periphyton microhabitats in the sam-
pling reach determines the types of sampling
devices and methods used for collecting rep-
resentative algal samples. Relevant site infor-
mation, sampling information, and microhabi-
tat characteristics are recorded on data sheets.
Table 4-2 list the measurements made during
the USGS-NAWQA periphyton and phy-
toplankton analyses.
4.2.1.1 Natural Substrates
Periphyton samples are collected from
the surfaces of natural substrates in relation
to the presence of microhabitats in the sam-
pling reach and the selection of habitats for
benthic invertebrate sampling (Section 5.3).
Sampling is conducted at locations chosen to
-------
Table 4-2. USEPA-EMAP Analytical Methodologies used for Periphyton
Summary of Methods
Sample Type
and Measurement
Expected Range
and/or Units
References
ID/Enumeration
Species composition,
Relative density
Chlorphyll:
Chlorophyll a
Bio mass
AFDM
APA
Enzymatic activity
species/ sample,
cells/ml, or cells/cm2
1 to 100 ug/cm2
mg/cm2
mmol/g, AFDM
mmol/cm2
Quantitative sample collected and
preserved; Soft algae analysis by
Palmer cell counts (200 organisms)
using either strip count or random
field technique; Diatom analysis
using permanent slides mounted in
Naphrax (500 frustules) using a strip
count.
Quantitative filtration; Extraction of
filter into acetone; Analysis by
spectrophotometry (monochromatic)
Quantitative filtration; Gravimetric
analysis
Spectrophotometric determination
Weitzel(1979);
APHA(1991)
APHA10200
H-2; APHA
(1991)
APHA(1991)
Sayleretal.
(1979)
represent combinations of natural and anthro-
pogenic factors that are important in influenc-
ing the water quality at local, regional, and
national scales (Porter et al. 1993). An over-
view of the sampling design can be found in
section 1.2. Each sampling reach is charac-
terized using a combination of qualitative and
quantitative periphyton samples.
(QMH) periphyton sample is prepared by
compositing collections of periphyton from
all instream microhabitat types present in the
sampling reach (Porter et al. 1993). The pos-
sible microhabitats that are targeted by the
QMH sampling are listed in Table 4-3.
Qualitative periphyton samples are col-
lected to document taxa richness in all avail-
able periphyton microhabitats present in the
sampling reach. This qualitative multihabitat
The goal of quantitative periphyton
sample collection is to measure relative abun-
dance and density of taxonomically-represen-
tative periphyton within: (1) a richest-targeted
habitat (RTH), which supports the taxonomi-
cally richest assemblage of organisms within
-------
Table 4-3. Microhabitats Used By The USGS-NAWQA Periphyton Collection Protocol And Methods
Used For The Qualitative Survey
Microhabitat
Description
Collection Methods
Epilithic
Epidendric
Epiphytic
Epipelic
Epipsammic
Submerged rocks, bedrock or
other hard surfaces
Submerged tree limbs, roots
or other wood surfaces
Submerged plants or macroalgae
Fine streambed sediments
Coarse streambed sediments
(e.g., sand)
Rocks are removed from the water. The attached
algal material is removed by hand or scraped into a
sample container. Bedrock may be sampled using a
PVC pipe sampler.
Woody material is removed from the water. The algal
material is removed by hand or scraped into a sample
container.
The plant or macroalgal material is removed from the
water. The attached algae is scraped or brushed into
a sample container. The liquid contents are
squeezed from algal mats or aquatic vascular plants
into the same sample container.
The top 5-10 mm of pigmented fine sediment is
collected using a disposable pipette and bulb, a
similar suction device, or a spoon or scoop.
The top 5-10 mm of pigmented coarse sediments are
collected using a disposable pipette and bulb, a
similar suction device, or a spoon or scoop.
a sampling reach, and (2) a depositional-tar-
geted habitat (DTK), where organisms are
likely to be exposed to sediment-borne con-
taminants for extended periods of time. Typi-
cal RTH areas include riffles in shallow,
coarse-grained, high-gradient streams or
woody snag habitats in sandy-bottomed
coastal streams. For the RTH portion of the
quantitative collection, periphyton are nor-
mally collected from five locations within the
sampling reach. At each location, periphyton
samples are taken from five representative
substrates (25 total samples). When available,
epilithic (see Table 4-3) samples are taken. If
epilithic substrates are not available, then
epidendric samples are taken. If there are no
epilithic or epidendric substrates, then epi-
phytic samples are taken. The SG-92 sam-
pling device is used to quantify the size of the
sampled area. The SG-92 is a syringe barrel
fitted with a rubber o-ring on one end. The
end with the rubber o-ring is placed flat on
the substrate surface so that a seal is formed.
A periphyton brush is then placed through the
syringe barrel and used to dislodge the at-
tached periphyton from the surface of the sub-
strate. The sample area is then washed with a
squirt bottle and the dislodged periphyton are
rinsed into the sample collection container.
Figure 4-2 shows a member of a field crew
using a SG-92 and a brush to dislodge per-
iphyton from a substrate. If the substrate sur-
face is irregular so that the rubber o-ring can-
not form a seal, the periphyton can be brushed
from the entire substrate and the entire sub-
strate is then fitted with aluminum foil. The
-------
Figure 4-2. A member of a field crew dislodges attached periphyton from its substrate using the USGS-
NAWQA method with the SG-92.
substrate is discarded and the foil is returned
to the laboratory so that the surface area of
the substrate can be determined. If bedrock is
to be sampled, then a PVC pipe sampler is
used. The periphyton from all 25 samples are
composited into the same sample collection
jar.
An example of a DTH area is an organi-
cally-rich deposit!onal area such as a pool. If
epilithic or epidendric (see Table 4-3) sub-
strates are available in the DTH area, then
periphyton should be collected in the same
manner as they are collected from the RTH
areas. However, if these substrates are not
present, then epipelic or epipsammic micro-
habitats should be sampled. In order to sample
epipsammic or epipelic habitats, the top half
of a disposable 47-mm plastic petri dish is
gently pushed into the streambed sediment.
Then, a small sheet of Plexiglas or spatula is
slipped under the petri dish top so that the sedi-
ment is trapped inside. The contents are then
rinsed into a sample jar. Because the volume
of the petri dish top can be measured, then
the sample can be quantified. Five sediment
samples are taken for the entire reach. All
DTH samples are composited into a single
sample jar.
The quantitative periphyton samples
should be obtained prior to collecting quali-
tative algae and benthic invertebrate samples
unless there are sufficient personnel and space
within the sampling reach to ensure that the
two sampling activities do not interfere with
one another (Porter etal. 1993).
4.2.1.2 Using Artificial
Substrates to Collect
Periphyton
When natural substrates cannot be
sampled because of inaccessibility of the mi-
crohabitats, cost of sample collection, or
safety issues associated with the collection of
representative samples, artificial substrates can
be used in sampling reaches These limitations
are more likely to occur in large rivers and
should be duly considered when designing a
-------
sampling program for this type of system.
Samples obtained from artificial substrates
may have reduced heterogeneity compared to
those obtained from natural substrates but can
be used to compare water quality among
streams with disparate periphyton microhabi-
tats. However, data from artificial substrates
cannot be compared with data from natural
substrates. If artificial substrates are used for
one or more stream reaches in a basin, it is
recommended that they be used at all sites so
that meaningful water-quality interpretations
can be made. The advantages and limitations
of artificial substrates are discussed in Porter
etal.(1993).
4.2.1.3 Quantitative
Phytoplankton Samples
Phytoplankton are more reflective of
conditions in the open water column than pe-
riphyton which are truly benthic indicators and
represent conditions at the sediment/substrata-
water interface. Quantitative phytoplankton
samples are obtained by collecting a repre-
sentative whole-water sample. A sample vol-
ume of 1 L is sufficient for samples collected
from productive, nutrient-enriched rivers as
indicated by water color, but a larger sample
volume is required for samples collected from
unproductive, low-nutrient rivers as indicated
by water transparency. Phytoplankton
samples taken in conjunction with water-
chemistry sampling are taken with a depth-
integrating sampler. Alternatively, quantitative
phytoplankton samples can be collected with
a water-sampling bottle or with a pump. If
chlorophyll is not to be determined, the entire
sample is preserved with buffered formalin.
For chlorophyll determinations, an
unpreserved subsample is withdrawn from the
phytoplankton sample, and the aliquot is fil-
tered onto a glass-fiber filter. The filtered
subsample volume should be sufficient to en-
sure that adequate algal biomass is retained
on the filter. Filters are then wrapped in alu-
minum foil, placed into a sample bottle or
container, and immediately stored on dry ice
(Porter et al. 1993). Figure 4-3 shows a mem-
ber of a field crew filtering a phytoplankton
sample for chlorophyll analysis.
Figure 4-3. Afield crew member filters a periphyton sample for chlorophyll analysis.
-------
Algal samples are labeled in the field.
Optional algal samples for the determination
of chlorophyll concentrations or ash-free dry
mass are processed in the field, placed on dry
ice, and submitted for analyses. Both the pe-
riphyton and phytoplankton samples can be
used for the chromatographic-fluorometric
and spectrophotometric analyses of
chlorophyl a and chlorophyl b. The periphy-
ton samples can additionally be used for the
determination of biomass through both dry
weight and ash weight analyses. Samples for
the identification and enumeration of algal taxa
are preserved with buffered formalin and
shipped to a laboratory for analysis (Porter et
al. 1993).
The USEPA-RBP recognizes benthic
algae as primary producers that integrate
physical and chemical disturbances to the
stream reach and that are sensitive indicators
of environmental conditions (Barbour et al.
1999). The objectives of theRBP forperiphy-
ton assessment include, but are not limited to:
1) assessment of biomass, 2) identification of
species, and 3) determination of the periphy-
ton assemblages' biological condition. The
methods endorsed by the RBP are a compos-
ite of the techniques used in Kentucky, Mon-
tana, and Oklahoma (Kentucky DEP 1993,
Bahls 1993, Oklahoma CC 1993). Periphy-
ton assemblages serve as good biological in-
dicators because they generally exhibit high
species richness and respond rapidly to ex-
posure but also recover quickly when the in-
sult is removed. In addition, most periphyton
taxa can be identified to species by experi-
enced biologists, and tolerance values to spe-
cific environmental conditions are known for
many species (Rott 1991; Dixit et al. 1992).
Diatoms are particularly useful indicators of
biological condition because they are found
in all lotic systems.
Three basic periphyton collection tech-
niques for wadeable streams are reviewed and
Table 4-4. Summary of RBP Collection
Techniques for Periphyton from Wadeable Streams
Substrate Type
Collection Technique
Hard removable substrate
gravel, pebbles, cobble, Remove representative
and woody debris
substrates from water;
brush or scrape
representative area of
algae from surface and
rinse into sample jar.
Soft removable substrate
mosses, macroalgae, Place a portion of plant
vascular plants, root
wads
into a sample container
with water, shake
vigorously; remove
plant.
Large non-removable
substrates
boulders, bedrock,
logs, trees, roots
Place PVC pipe with a
neoprene collar at one
end on the substrate so
that the collar is sealed
against the substrate.
Dislodge algae in the
collar with a brush or
scraper and retrieve
them with a pipette.
-------
summarized in Table 4-4 (Plafkin et al. 1989;
Barbouretal. 1999).
For an accurate assessment of the assem-
blage, samples should be collected during
periods of stable stream flow. High flows can
scour the stream bed and flush the periphy-
ton downstream.
Peterson and Stevenson (1990) recom-
mend a three-week delay following high,
bottom-scouring stream flows to allow
recolonization and succession to a mature
periphyton community (Plafkin et al. 1989;
Barbouretal. 1999).
The collection procedures have been
adapted from Kentucky and Montana proto-
cols (Kentucky DEP 1993, Bahls 1993). Pe-
riphyton should be collected from all avail-
able microhabitats in the sampling reach.
Composite qualitative samples are collected
from microhabitats in the approximate pro-
portion each microhabitat occurs. Both riffles
and pools are sampled if available. Algal mats
or other soft-bodied algal forms can be col-
lected from depositional areas with forceps, a
suction bulb and disposable pipette, a spoon
or an eyedropper.
All samples should be placed in water-
tight, unbreakable, wide-mouthed containers.
A 4-oz (125-ml) sample is usually sufficient
for analysis (Bahls 1993). Lugol's solution
(potassium iodide), buffered 4% formalin,
ethanol or other preservatives may be used to
preserve samples.
For chlorophyll analyses, periphyton are
scraped from fixed areas onto a glass fiber
filter. Filters are wrapped in foil and frozen
for transportation to the laboratory (Plafkin et
al. 1989; Barbouretal. 1999).
Periphyton can be sampled by collect-
ing from artificial substrates that are placed in
aquatic habitats and colonized over a period
of time. This procedure is particularly useful
in boatable streams, rivers with no riffle ar-
eas, wetlands, or the littoral zones of lentic
environments. Both natural and artificial tech-
niques are useful in monitoring and assessing
waterbody conditions, and have correspond-
ing advantages and disadvantages (Stevenson
and Lowe 1986, Aloi 1990).
The methods summarized here are a
composite of those specified by Kentucky
(Kentucky DEP 1993), Florida (Florida DEP
1996), and Oklahoma (Oklahoma CC 1993).
The RBP endorses the use of periphytometers.
Periphytometers are sampling devices that can
either be deployed as floating or benthic. They
are fitted with glass slides, glass rods, clay
tiles, plexiglass plates, or similar substrates and
deployed at the sampling location for two to four
weeks. A minimum of three periphytometers are
placed at each site to account for spatial vari-
ability, depending upon the research design and
hypothesis being tested. Samples can be
composited or analyzed individually. After the
incubation period, slides are collected and
subsampled for chlorophyll a and taxonomic
analysis. Storage containers for chlorophyll
a are filled with deionized water and those
for taxonomic analysis are filled with ambi-
ent water. Microslides for taxonomic analy-
sis are scraped and samples are preserved.
Samples should be stored in a dark refrigera-
tor until they are processed. Microslides for
chlorophyll analysis should be scraped and
rinsed with deionized water onto a glass-fi-
ber filter. Filters with captured algal cells are
wrapped in foil and frozen to await extrac-
tion and analysis (Plafkin et al. 1989; Barbour
etal. 1999).
-------
Semi-quantitative assessments of
benthic algal biomass and taxonomic compo-
sition can be made rapidly with a viewing
bucket marked with a grid and biomass scor-
ing system (Stevenson and Lowe 1986). The
advantage of using this technique is that it
enables rapid assessment of algal biomass
over large areas. This technique is a survey
of the natural substrate that does not require
laboratory processing, and may be an alter-
native screening technique to other RBP
methods (Plafkin et al. 1989; Harbour et al.
1999).
At least three transects across the habi-
tat are established. Riffles or runs in which
benthic algal accumulation is readily observed
and easily characterized are preferred loca-
tions for establishing the transects. Three lo-
cations are selected objectively on each
transect. Algae in each location are charac-
terized by observing the stream through the
bottom of the viewing bucket and counting
the number of dots covered by macroalgae.
The maximum length of the macroalgae is
measured and recorded. If two types of
macroalgae are present, information for each
type of macroalgae is measured and recorded
separately. While viewing the same area, the
number of dots under which substrate occur
that are of a suitable size for microalgae ac-
cumulation is recorded. The type of
microalgae (usually diatoms and blue-green
algae) is determined and the density under
each dot estimated using the scale in Table 4-
5. The density of algae on the substrate is
characterized by calculating the average per-
cent cover of the habitat by each type of
macroalgae, the maximum length of each type
of macroalgae, the mean density of each type
of microalgae on suitable substrates, and the
maximum density of each type of microalgae
on suitable substrates (Plafkin et al. 1989;
Barbouretal. 1999).
The periphyton metrics summarized in
the RBP manual are in use by several states
(Kentucky DEP 1993, Bahls 1993, Flordia
DEP 1996) (Table 4-6). Two metrics are mea-
surements of taxa richness (total taxa and
Shannon diversity); these are estimated from
the count of taxa encountered in a target num-
ber of cells (500 cells). If the cell counts vary
Table 4-5. Scale Used to Score the Density of
Microalgae in the RBP Semi-quantitative Method
Microalgal Density
Score
Substrate rough with no evidence of 0
microalgae
Substrate slimy, but no accumulation of 0.5
microalgae is evident
A thin layer of microalgae is evident 1
2
Accumulation of microalgal layer
fromO.5-1 mm thick is evident
Accumulation of microalgal layer
from 1-5 mm thick is evident
Accumulation of microalgal layer
from 5-20 mm thick is evident
Accumulation of microalgal layer
from >20 mm thick is evident
3
-------
Table 4-6. Diatom and Non-diatom Metrics
Summarized in the RBP Manual
Diatom Metrics
Non-diatom Metrics
Total number of diatom
taxa(TNDT)
Shannon diversity (for
diatoms)
Percent community
similarity (PSc) of
diatoms
Pollution tolerance
index for diatoms
Percent sensitive
diatoms
Percent motile diatoms
Taxa richness of non-
diatoms
Indicator non-diatom
taxa
Relative abundance
of all taxa
Number of Divisions
represented all taxa
Chlorophyll a
Ash-free dry-mass
(AFDM)
Percent Achnanthes
minutissima
See Appendix B for details.
by more than 20% from 500, then it may be
necessary to adjust the taxa richness estimate
with a rarification formula (Barbour and
Gerritsen 1996). Periphyton metrics are de-
scribed in Appendix B.
The amount of pollution present can shift
the structure of the natural community of dia-
toms (Patrick 1963,1964; Patrick etal. 1954;
Patrick and Hohn 1956; andHohn 1959). The
methods of water quality assessment using
diatoms can be classified into three main
types. The first method is the saprobic sys-
tem and its derivatives in which diatom as-
semblages are characterized by their tolerance
to organic pollution (Kolkwitz and Marsson
1908, Liebmann 1962, Sladecek 1973). A
second method is based on the classification
of diatoms according to their sensitivity to all
types of pollution (Fjerdingstad 1950, 1960;
Coste 1974). Fjerdingstad (1950, 1960)
classified diatom species according to their
ability to withstand varying amounts of pol-
lution and then described communities in
terms of dominant and associated species. A
third category of methods is based on the di-
versity of diatom communities. These meth-
ods include plotting the number of species
against the number of individuals per species
(Patrick 1964) as well as calculating diver-
sity indices (review by Archibald 1972).
An example of a water-quality assess-
ment method based on the pollution tolerance
of diatom assemblages is the Pollution Toler-
ance Index (PTI), which is used by the Ken-
tucky Department of Environmental Protec-
tion (DEP). ThePTIismostsimilartolhatofLange-
Bertalot (1979) and resembles the Ffilsenhoff Bi-
otic Index (HBI) for macroinvertebrates (Hilsenhoff
1987). Lange-Bertalot distinguished three catego-
ries of diatoms according to their tolerance to
pollution, with the mosttoleranttaxabeing assigned
a value of 1 (e.g., Nitzschiapalea, Gomphonema
parvulum) and sensitive taxabeing assigned a value
of 3. For the PTI, Lange-Bertalot's categories
were expanded to four. Therefore the result-
ing PTI diatom pollution tolerance values
range from 1 (most tolerant) to 4 (most sensi-
tive). The formula used to calculate PTI is:
PTI=
N
Where n is the number of cells counted
i
for species i, t. is the tolerance value of spe-
cies /' (1-4), and Nis the total number of cells
counted. Tolerance values have been gener-
ated from several sources, including Lowe
-------
(1974), Patrick and Reimer (1966, 1975),
Patrick (1977), Lange-Bertalot (1979), Descy
(1979), Sabater et al. (1988), Bahls et al.
(1992), and Oklahoma Conservation Com-
mission (1993).
An example of a water-quality assess-
ment method based on the diversity of dia-
tom assemblages is percent community simi-
larity (PSc) by Whittaker (1952). ThePSc was
chosen for use in diatom bioassessments be-
cause it shows community similarities based
on relative abundances, and in doing so, gives
more weight to dominant taxa than to rare
ones. PSc should only be used when compar-
ing a study site to a control site, or when con-
ducting multivariate cluster analysis. If the
emphasis is comparing a study site to a re-
gional reference condition (i.e., a composite
of sites), PSc should not be used. PSc values
range from 0 (no similarity) to 100%.
The formula for calculating percent com-
munity similarity is:
s
PSc=100-0.5^ja1-b1
1=1
Where a. is the percentage of species /
in sample A, and b. is the percentage of spe-
cies/'in sample B.
macroinvertebrates) as well as non-living de-
tritus. AFDM values are used in conjunction
with chlorophyll a as a means of determining
the trophic status of streams through the use
of the Autotrophic Index (Al). The formula
used to calculate the Al is:
AI = AFDM(mg/m2)/
Chlorophyll a (mg/m2).
High Al values (>200) indicate the com-
munity is dominated by heterotrophic organ-
isms, and can indicate poor water quality
(Weber 1973, Weitzel 1979, Matthews et al.
1980). This index should be used with dis-
cretion, because non-living organic detritus
can artificially inflate the AFDM value.
The USEPA-RBP (Barbour et al. 1999)
recommends that the Al be modified to:
Al = Chlorophyll a (mg/m2)/
AFDM (mg/m2)
In this form, the index is positively re-
lated to the autotrophic proportion of the as-
semblage instead of the heterotrophic propor-
tion. Also, the modified index would have
better statistical properties as a proportion or
percent (normally, chlorophyll a/AFDM val-
ues are approximately 0.1%) than the origi-
nal index.
Because periphyton are found on or in
close proximity to the substrate, Ash Free Dry
Mass (AFDM) values are used as tools to as-
sess their assemblages. AFDM is used as an
estimate of total organic material accumulated
on the substrate. This organic material in-
cludes all living organisms (periphyton and
Because they do not evaluate periphy-
ton/phytoplankton in their assessments of
stream quality, no methods for the Ohio EPA
-------
or MDNR-MBSS are reported in this section.
Table 4-7 summarizes the assessment meth-
ods used by the USEPA-EMAP-SW, USGS-
NAWQA and USEPA-RBP. The USEPA-
EMAP-SW program assesses algal assem-
blages using a quantitative method to sample
erosional or depositional habitats. They use
the periphyton samples for four types of
analyses: ID/enumeration, chlorophyll, bio-
mass, andAPA.
The USGS-NAWQA program uses
both qualitative and quantitative methods to
sample natural substrates. In addition, artifi-
cial substrates and samples from the water
column can be used to further quantify the
conditions of the periphyton and phytoplank-
ton assemblages.
The USEPA-RBP recommends the
qualitative collection of periphyton from natu-
ral substrates as well as a quantitative assess-
ment from artificial substrates. In addition, the
USEPA-RBP suggests a rapid semi-quanti-
tative method for assessing the macroalgae.
Table 4-7. Methods used by the Three Reviewed Programs for the Collection and Assessment of
Periphyton and Phytoplankton Assemblages
Methods
USEPA-
EMAP-SW
USGS-
NAWQA
USEPA-
RBP
Collection methods
Periphyton from natural substrates - quantitative
Periphyton from natural substrates - qualitative
Periphyton from artificial substrates
Periphyton from natural substrates - semi-quantitative
Phytoplankton
Analysis methods
ID/enumeration
Chlorophyll
AFDM
APA
X
X
X
X
X
X
X
X1
X
X
X
X
X
X
X
X
X
'This method is an option for the USGS-NAWQA program, but it is not typically used (Gurtz, personal
communication 1999).
-------
by
Bradley C. Autrey
This section compares the benthic
macroinvertebrate sampling methods from the
three federal programs, the USEPA-EMAP-
SW, the USGS-NAWQA program, and the
USEPA-RBP, as well as the two state pro-
grams, the Ohio EPA and MDNR-MBSS.
The differences among the methods from these
five programs reflect their regional differ-
ences, the divergent ecological interests in
sampling benthic macroinvertebrates, and the
various habitats sampled.
Most water quality agencies that rou-
tinely collect water quality data study benthic
macroinvertebrates (Southerland and Stribling
1995). Several factors contribute to the high
utilization of benthic macroinvertebrates as
indicators of stream condition:
• benthic macroinvertebrates are present
in a variety of habitats,
• sampling is relatively easy to conduct
and it has a limited detrimental effect on
the resident biota,
• benthic macroinvertebrates are relatively
sedentary,
• benthic macroinvertebrates are sensi-
tive to a wide range of chemical stressors,
• assemblages are often made up of spe-
cies that have a broad range of pollution
tolerances,
• the response of benthic macroinverte-
brates to physical and chemical stressors
has been widely described and
• many states have background benthic
macroinvertebrate data.
Combined, these factors allow for the
cumulative chemical and physical attributes
-------
of aquatic ecosystems to be effectively as-
sessed through the evaluation of their benthic
macroinvertebrate assemblages (BEST 1996).
5.1 USEPA-EMAP-SW
Macroinvertebrate
Assessment
The USEPA-EMAP-SW benthic
macroinvertebrate protocol (Klemm et al.
1998) is used to evaluate the overall condi-
tion of and detect the relative stress levels in
wadeable and boatable streams. Sampling pro-
tocols for wadeable streams are based on the
USEPA-RBP III - Benthic Macroinvertebrates
(Plafkin et al. 1989) with the modification of
a one person kick net procedure developed
for the USGS-NAWQA program (Cuffney
et al. 1993a) replacing USEPA-RBP's origi-
nal two-man kick net procedure. In boatable
streams, benthic macroinvertebrates are
sampled with drift nets in addition to the modi-
fied kick net procedure. Figure 5-1 shows a
modified kick net.
5.1.1 Wadeable Streams:
Riffle/Run and Pool/
Glide Sampling
When sampling riffle/run habitats in
wadeable streams, a 595/600 urn modified
kick net is used to collect organisms at the
nine interior transects, at either the left, right,
or middle points of each transect as deter-
mined by the role of a die (see section 1.1).
The sampler is held securely on the stream
bottom while kicking the substrate vigorously
for 20 seconds in an area of about 0.5 m2 in
Figure 5-1. A modified kick net (left) such as is used in the USEPA-EMAP-SW protocols and a D-frame
kick net (right) such as is used in the USGS-NAWQA protocols.
-------
front of the net. Heavy organisms (such as
mussels and snails), in the sample area are
hand-picked and placed into the net. At the
end of the 20-second period, with the net still
being held in place, any organisms found on
rocks in the delimited area are placed in the
net. The net contents are then rinsed into the
riffle bucket that is half filled with water. All
riffle samples are combined into a single com-
posite riffle bucket.
When sampling pool/glide habitats in
wadeable streams, a 595/600 jim modified
kick net is used to collect samples at the inte-
rior transects where very slow water is present.
Heavy organisms on the stream bottom are
hand picked and placed into the net. A 0.5 m2
area of substrate is disturbed by vigorous kick-
ing. A 20-second sample is collected by drag-
ging the net repeatedly through the disturbed
area just above the bottom while vigorously
kicking. The net is kept moving in order to
prevent collected organisms from escaping.
After 20 seconds, organisms found on loose
rocks in the sample area are placed into the
net. Net contents are placed into the pool
bucket that is half filled with water. All pool
samples are combined into a single compos-
ite pool bucket. If there is too little water to
use the kick net, the substrate is stirred with
gloved hands and a US Standard #30 sieve is
used to collect the organism from the water
for 20 seconds in the same way the net was
used in larger pools.
The contents of the riffle and pool buck-
ets are individually poured through a US
Standard #30 sieve (Figure 5-2). The buck-
Figure 5-2. A fieldcrew member processes a benthic macroinvertebrate sample before it is transported
to the laboratory for analysis.
-------
ets are rinsed with stream water in order to
ensure that all organisms are evacuated. Large
objects are rinsed with stream water and dis-
carded. The sieve is thoroughly rinsed and its
contents are washed into ajar that is labeled
with sampling information and designated as
"riffle" or "pool". In order to preserve the
sampled organisms, 95% ethanol is added to
eachjaruntil afmal concentration of atleast
70% is obtained. Eachjar is capped and sealed
until the samples are analyzed (Klemm et al.
1998).
In boatable streams, kick net sampling
is conducted the same way as in wadeable
streams with the exceptions that all 11
transects are sampled, instead of 9, and all
samples are combined into a single compos-
ite sample, instead of separate composite
samples for riffle/run habitats and pool/glide
habitats. Also, in boatable streams, benthic
macroinvertebrates are additionally sampled
using drift nets. Each drift net consists of a
nylon or nylon monofilament bag (595-600
|im) that is 1 m in length at the closed end.
The open end is 30.48 cm X 45.72 cm. At
each sampling location, two drift nets are set
at the downstream end of a sample reach
(transect A). If possible, one drift net is set
about 25 cm from the bottom substrate and
one drift net is set about 10 cm below the sur-
face of the water. In systems with stronger
currents, both nets may be set 10 cm below
the surface of the water. Nets can be set with
stainless steel rods, but are usually deployed
using two floating drift net assembly devices
(Wildco 15-D10), one of which may be out-
fitted with a deep-deep drift attachment
(Wildco 15-D12).
Drift nets are set for three to four hours
and only in streams with currents greater than
0.05 m/s. Once the drift nets are set in the
stream, the water velocity at each net open-
ing is measured and recorded. After the nets
have been set for three to four hours, the wa-
ter velocity is again measured at each net open-
ing and recorded. The nets are then removed
from the stream and the samples are combined
and sieved using a sieving bucket (595 |im-
mesh/standard #30). After being cleared of
macroinvertebrates, large debris from the
sample is discarded. The composite sample
is then transferred to a collection j ar and pre-
served with 95% ethanol.
The results of the drift net benthic
macroinvertebrate collection are reported per
unit of time and flow (Allan and Russek 1985,
Klemm etal. 1998).
USGS-NAWQA utilizes several types
of sampling equipment and techniques for the
collection of benthic macroinvertebrates. The
proper type of sampling equipment and tech-
nique depends on the morphology of the
stream or river being sampled as well as the
objectives of the study (Cuffney et al. 1993 a,
1993b).
The purpose of qualitative multihabitat
sampling is to obtain the most complete list
of invertebrate taxa possible during approxi-
mately one hour of sampling. This is accom-
plished by sampling as many habitat types
within the sampling reach as is possible with
approximately equal intensity. The primary
-------
sampling device used in wadeable streams is
a D-frame kick net equipped with a 210- urn
mesh net. Kicking, dipping, or sweeping
motions, as appropriate, are used to collect
samples from the substrate. Figure 5-1 shows
a D-frame kick net.
Visual detection and seines are used to
collect firmly attached and highly-motile in-
vertebrates, respectively. Visual collection in-
volves manually collecting large rocks, coarse
debris, or other substrates and visually locat-
ing and removing any associated organisms.
This method is useful for collecting sessile or-
ganisms and organisms that burrow into hard
substrates. Figure 5-3 shows a member of a
field crew brushing attached benthic
macroinvertebrates from a rock into a sieve.
Seining with a 3.2 jam mesh can be used to
collect larger, highly motile organisms, such
as amphipods, decapods and freshwater
prawns.
The choice of collection methods for
QMH samples from boatable habitats depends
upon the depth of the water, current velocity,
and bed material. Grab samplers are suitable
for sand or fine gravel substrates in moder-
ate-current conditions and waters of medium
depths. Shipek and Van Veen samplers are
useful in extremely deep and fast rivers with
sand or fine gravel bottoms. A diver-operated
dome sampler is used in deep rivers when the
bed material is composed of large gravel,
cobble, boulder, or bedrock (Cuffney et al.
1993a).
5.2.2 Semi-Quantitative
Targeted-Habitat
Sampling Methods
The purpose of semi-quantitative tar-
geted-habitat sampling is to obtain represen-
tative samples of benthic invertebrate commu-
Figure 5-3. A field crew member uses a stiff-bristled brush to remove the attached benthic
macroinvertebrates from a rock.
-------
nities from two instream habitat types: 1) a
habitat supporting the most taxonomically
diverse community of benthic invertebrates
(richest-targeted habitat, RTH), usually a fast-
flowing, coarse-grained riffle; and 2) a fine-
grained, organically rich depositional habitat
(depositional-targeted habitat, DTK), usually
a pool. Semi-quantitative sampling methods
usually characterize the structure of inverte-
brate communities in terms of the relative
abundances of each taxon. The type of sam-
pler used to collect a semi-quantitative sample
depends upon the depth, velocity, and sub-
strate within the instream habitat that is to be
sampled. Artificial substrates are used in situ-
ations where natural substrates cannot be
sampled due to inaccessibility of the habitat,
cost of sample collection, or safety concerns.
Under certain conditions, such as a large, deep
river with cobble, boulder, or bedrock sub-
strate, artificial substrates may offer the only
viable means of obtaining benthic
macroinvertebrate samples.
All nets and screens used in the collec-
tion of semi-quantitative samples have a mesh
size of 425 jim. Samples are washed, sieved,
and split in the field to reduce the bulk of the
composite sample to less than 0.75 L. Samples
collected and processed in this manner are
preserved in 10% formalin (Cuffney et al.
1993 a).
Disturbance-removal sampling tech-
niques are the most appropriate method for
sampling wadeable coarse-grained substrates
with current velocities greater that 5 cm/s.
These techniques involve defining a specific
area, disturbing the substrate within that area
to dislodge invertebrates into a sampler, and
then removing the larger substrate elements
to acquire any specimens that are adhering
tightly to the rocks. Hess samplers, Surber
samplers, stovepipe corers, and box samplers
are examples of samplers that can be used in
these situations (Cuffney et al. 1993a).
Coarse substrates in boatable streams
(water deeper than approximately 0.50-0.75
m) cannot be effectively measured using most
disturbance-removal type samplers. A diver-
operated dome sampler, artificial substrates,
and stovepipe samplers (for water less than
0.75 m deep) can be used in these situations.
Nets with 425- |im mesh are used in each
case, to catch organisms dislodged or sus-
pended in the sampler (Cuffney et al. 1993a).
Grab samplers are appropriate for sam-
pling in shallow, fine-grained riffles or pools.
All screening on the grab should have mesh
openings of 425 jim or smaller (Cuffney et
al. 1993a).
Grab samplers can be used from boats
to obtain samples from deep rivers with fine-
grained substrates. A hand or power winch is
recommended for sampling in deep waters or
using weighted grab samplers. All screening
on the grab sampler should have mesh open-
ings of 425 |im or smaller (Cuffney et al.
1993a).
When snags are used in the semi-quan-
titative RTH portion of the macroinvertebrate
-------
survey, they are sampled by removing sec-
tions of tree limbs with a saw or lopping shears
and collecting the attached invertebrates by
hand picking and brushing the limb's surface
and cavities. The loss of motile or loosely at-
tached organisms can be minimized by plac-
ing a net downstream from the limb to catch
dislodged organisms. The lengths and diam-
eters of the sampled snags are recorded in
order to estimate the surface areas.
When macrophyte beds are used in the
semi-quantitative RTH portion of the
macroinvertebrate survey, they can be
sampled with disturbance-removal samplers.
Net samplers can be used if there is sufficient
current to wash the dislodged plant and ani-
mal material into the net. A knife or trowel
can be used to dislodge the plant material from
the substrate. Stovepipe samplers may prove
more effective and should be used when the
macrophytes are too tall to allow use of a
dredge. The macrophytes that are removed
should be inspected carefully for invertebrates
that are attached and for those that burrow
into stems (Cuffney et al. 1993a).
The current USEPA-RBP methods
(Plafkinetal. 1989; Harbour etal. 1999) em-
phasize the sampling of a single habitat in
wadeable streams, preferably those having
riffles/runs, because macroinvertebrate diver-
sity and abundance are usually highest in these
habitats. When some streams lack the riffle/
run habitats, a method suitable to sampling a
variety of habitats is desired. The proposed
multi-habitat sampling approach is designed
to sample maj or habitats in proportional rep-
resentation within a sampling reach.
A 100-m reach that is representative of
the stream is selected. All riffle/run areas within
the 100-m reach are candidates for sampling
macroinvertebrates because macroinvertebrate
diversity and abundance are usually highest
in cobble substrate. Where cobble substrate
is the predominant habitat, this sampling ap-
proach provides a representative sample of
the stream reach. In cases where cobble sub-
strate represents less than 30% of the sam-
pling reach, alternative habitats (such as snags,
vegetated banks, submerged macrophytes,
and sand) will need to be sampled.
Sampling begins at the downstream end
of the reach and proceeds upstream. Using a
1-m, 500-|im mesh kick net, the stream is
sampled two or three times at locations of
various velocity in the riffle. A kick in the
single habitat approach is a stationary sam-
pling accomplished by positioning the net and
disturbing 1 m2 upstream of the net. Large
substrate particles are gathered and the at-
tached organisms are removed. The sample
is then transferred to sample containers and
preserved in 95% ethanol (Barbour et al.
1999).
A 100-m reach that is representative of
the stream is selected. Different types of habitat
are to be sampled in the approximate propor-
tion in which they are represented in the reach.
Sampling begins at the downstream end of
the reach and proceeds upstream. A total of
20 jabs or kicks are taken over the length of
the reach. A jab consists of forcefully thrust-
ing the net into the habitat for 0.5 m. A kick
-------
in the multi-habitat approach is a stationary
sampling accomplished by positioning the D-
frame, 500 um mesh dip net and disturbing
the substrate for a distance of 0.5 m upstream
of the net. The jabs or kicks collected from
the multiple habitats are combined to obtain a
single homogeneous sample. The sample is
transferred to sample containers and preserved
in 95% ethanol (Barbour et al. 1999).
5.4 Ohio EPA
Macroinvertebrate
Assessment
Assessments of the ambient
macroinvertebrate community by the Ohio
EPA (OEPA 1988, 1989) consists of two
types: 1) intensive surveys of stream or river
reaches using multiple sites in upstream to
downstream longitudinal or synoptic sub-ba-
sin configurations, and 2) multiple-year sam-
pling at a specified fixed station on a stream
or river. Sampling sites are located based on
the characteristics of the stream or river, and
in accordance with the survey objectives.
5.4.1 Artificial
Substrate
The primary sampling equipment used
for quantitative sampling is the modified
Hester-Dendy artificial substrate sampler. It
is constructed of 0.125-inch tempered hard-
board cut into three in2 plates and 1.0 in2 spac-
ers. A total of eight plates and twelve spacers
are used for each sampler. Plates and spacers
are placed on a 0.25-inch eyebolt so there are
three single spaces, three double spaces, and
one triple space between the plates. The total
surface area of the sampler, excluding the
eyebolt, is 145.6 in2 (approximately 1.0 ft2).
Figure 5-4 shows a Hester-Dendy sampler in
place at a sampling location.
Figure 5-4. A Hester-Dendy sampling device placed in a river. Note: This sampler was set in a more
shallow area for photographic purposes. Hester-Dendy samplers are normally set approximately 1 meter
below the water's surface.
-------
Before the samplers are placed in
streams, they are tied to concrete construc-
tion blocks in order to anchor them in place.
Whenever possible, samplers are placed in
runs rather than in pools or riffles, so that a
steady flow of water is running through the
sampler and an attempt is made to place all
samplers in habitats that are as similar to each
other as possible. At each sampling site, a set
of five artificial substrate samplers are ex-
posed for a six-week period, usually between
June 15 and September 30.
Retrieval of the samplers is accom-
plished by separating them from the concrete
block and placing them in one-quart plastic
containers while still submersed. Enough
formalin is added to each container to ap-
proximate a 10% solution (OEPA 1989).
minutes. Once the 30 minute minimum sam-
pling time has been met, sampling is contin-
ued until no new taxa are collected.
In addition, quantitative samples of
macroinvertebrates inhabiting the natural sub-
strates can also be optionally collected. This
is accomplished by using a Surber square-foot
sampler, with # 30-mesh netting, and a hand
cultivator with two-inch tines. Standing on the
downstream side of the sampler, the collector
works the substrate using the hand cultivator.
For large rocks, a brush can be used. Three
to five Surber samplers are taken at each site
(OEPA 1989).
For the purpose of metric development,
qualitative samples of macroinvertebrates in-
habiting the natural substrates are also col-
lected at the same time that the artificial sub-
strate sampler is retrieved. All available habi-
tat types are sampled and voucher specimens
are retained for laboratory identification. In
shallow waters, forceps and a triangular ring
frame with a US Standard #30-mesh (595-
600 jam) dip net are used. Grab samplers can
be used in deep waters. The qualitative sam-
pling continues until, as determined by gross
examination, no new taxa are taken.
When only qualitative samples are col-
lected, an attempt is made to sample a riffle,
run, margin, and pool habitat at each station.
Stations should be sampled in order, moving
from upstream to downstream, to detect any
changes between sites. Sample areas should
be physically similar among the different sites.
Collections are made for a minimum of 30
For this program benthic macroinverte-
brates are collected to provide a qualitative
description of the community composition at
each sampling site (Janicki et al. 1993). Sam-
pling is conducted in the spring index period
(between March 1 and May 1) in wadeable
streams (Roth et al. 1997b).
A 600-|im mesh D net is used to collect
organisms from habitats with the highest prob-
able taxonomic diversity; thus, riffle areas are
preferred, because macroinvertebrate abun-
dance and diversity are usually highest in riffle
areas. Other habitat types include rootwads,
woody debris, leaf packs, macrophytes and
undercut banks. A variety of techniques are
used for collection, such as kicking, jabbing,
and gently rubbing hard surfaces by hand to
dislodge organisms. Each jab covers one ft2.
For every 75-m segment, 20 sites are sampled.
-------
Combined substrates from each segment are
preserved in 70% ethanol (Roth et al. 1997b).
The four primary benthic macroinverte-
brate indices used by these programs to de-
termine water quality conditions are the In-
vertebrate Community Index (ICI), the
Hilsenhoff Biotic Index (HBI), the Benthic
Index of Biotic Integrity (B-IBI), and
Ephemeroptera, Plecoptera, and Trichoptera
(EPT) richness. EPT richness is a simple in-
dex (Lenat 1987) that incorporates three or-
ders of macroinvertebrates which are gener-
ally intolerant to poor water conditions. Also
reviewed is the Stream Benthos Integrity In-
dex (SBII), which is currently under devel-
opment by the National Exposure Research
Laboratory (NERL) for the USEPA-EMAP-
SW.
Table5-l. Metrics used in the Ohio EPA's ICI
and Their Expected Responses to Disturbance
Metric
Expected response
to disturbance
Total number of taxa Decrease
Total number of Ephemeroptera Decrease
taxa
Total number of Trichoptera taxa Decrease
Total number of Dipteran taxa Increase
Percent Ephemeroptera Decrease
composition
Percent Trichoptera composition Decrease
Percent Tanytarsini midge Increase
composition
Percent other Dipteran and Increase
non-insect composition
Percent tolerant organisms Increase
Total number of qualitative EPT Decrease
taxa
Development of the ICI was a result of
the 1983-84 Ohio Stream Regionalization
Project, a cooperative pilot venture between
Ohio EPA and USEPA/ERL-Corvallis
(Whittier et al. 1987). It is now the primary
tool used by the Ohio EPA for measuring the
condition of macroinvertebrate communities
(DeShon 1995). Table 5-1 shows the metrics
included in the Ohio EPA's ICI and their ex-
pected responses to disturbances. These ten
metrics are scored and summed to obtain an
ICI value.
The USEPA-EMAP-SW, USEPA-RBP,
and MDNR-MBSS use the HBI. Hilsenhoff
(1977) refined the index first proposed by
Chutter (1972) in developing the HBI. Resh
and Jackson (1993) found the HBI to be an
effective measurement discriminating be-
tween impaired and unimpaired sites in Cali-
fornia. A North Carolina study found that both
the EPT and the HBI are good indicators of
stream water quality (Wallace et al. 1996).
The HBI attempts to summarize the overall
pollution tolerance of the benthic
macroinvertebrate community. Its value is
calculated using the following formula:
Where n is the number of individuals in
each taxon, a is the tolerance value assigned
-------
to that taxon and N is the total number of in-
dividuals in the sample. Tolerance values for
individual taxa are listed in Hilsenhoff (1988).
Tolerant organisms are those frequently as-
sociated with gross organic contamination and
are generally capable of thriving under anaero-
bic conditions (given a score of 4 or 5). Fac-
ultative organisms are those having a wide
range of tolerance that frequently are associ-
ated with moderate levels of organic contami-
nation (given a score of 2 or 3). Intolerant
organisms are those that are usually not found
associated with organic contaminants and are
generally intolerant of even moderate reduc-
tions in dissolved oxygen (given a score of 0
or 1). Organisms not listed in Hilsenhoff
(1988) are given a value of 5, unless avail-
able information suggests otherwise.
An HBI value is calculated using the
pollution tolerance values for the represented
taxa (Hilsenhoff 1988) and the equation given
in section 5.6.2. The resulting value can be
used as an indicator of water quality. The
water quality categories indicated by the re-
spective HBI scores are given in Table 5-2.
The MDNR-MB S S developed two ver-
sions of the Benthic Index of Biotic Integrity
(B-IBI) for the Monitoring and Non-Tidal
Assessment (MANTA) Division of the
Table 5-2. Water-Quality Levels Indicated by
Different Ranges of HBI Scores.
Range of HBI Scores Indicated Water Quality
0.00-3.75
3.76-4.25
4.25-5.00
5.01-5.75
5.76-6.50
6.51-7.25
7.26-10.00
Excellent
Very Good
Good
Fair
Fairly Poor
Poor
Very Poor
MDNR. One version is for the coastal plains
(CP) region of Maryland and the other is for
the non-coastal plains (NCP) region (Table
5-3). These indices were modeled after Karr
et al.'s (1986) Index of Biotic Integrity (IBI).
While the IBI was developed to estimate the
condition of an aquatic ecosystem based on
its fish community, the B-IBIs will allow the
MDNR to more accurately assess the condi-
tion of its streams by surveying their benthic
macroinvertebrates (Roth et al. 1997b). Defi-
nitions of metrics used in the B-IBI and scor-
ing parameters may be found in Appendix C.
The Stream Benthos Integrity Index
(SBII) was developed by the NERL for
Table5-3. Metrics used for the CP B-IBI and
the NCP B-IBI
Metric CPB-IBI
Total Number of Taxa
Number of EPT Taxa
Number of Ephemeroptera
Taxa
Number of Dipteran Taxa
Percent Ephemeroptera
X
X
X
NCPB-BI
X
X
X
X
X
Percent Tanytarsini of
Chironomidae
X
Percent Tanytarsini
Number of Intolerant Taxa
Percent Tolerant Individuals
Beck's Biotic Index X
Number of Scraper Taxa X
Percent Collectors
Percent Clingers X
X
X
X
X
-------
USEPA-EMAP-SW, specifically for the Mid-
Atlantic Highlands (MAH) region. The SBII
is a multimetric index developed using a
stepwise process to evaluate candidate metrics
and best professional judgement for final se-
lection of metrics. Seven metrics were selected
for inclusion in the SBII (Table 5-4), with the
score of each metric ranging from 0 to 1 on a
continuous scale. Scoring of metrics is based
on the fraction of the "best attainable value"
observed at a site, where the "best attainable
value" is established using the 95th (metrics
that decrease in response to stress) or 5th
(metrics that increase in response to stress)
percentile of the overall distribution of each
metric. Two of the metrics are adjusted for
watershed size prior to scoring. The SBII
ranges from 0 to 7, with 3 condition catego-
ries and 2 transition ranges (Table 5-5), based
on a power analysis.
Table5-5. The USEPA's SBII Condition
Categories and Associated Score Ranges
Condition
Range of Scores
Good
Good-Fair transition
Fair
Fair-Poor transition
Poor
5 to 7
>4.5to<5
2.5to4.5
>2to<2.5
Oto2
lytical techniques used by USGS-NAWQA
are not presently available and are, there-
fore, not included in this section.
The USEPA-EMAP-SW protocols utilize
three indices to analyze the metrics gathered from
the survey of benthic macroinvertebrates and are
currently developing a fourth index (Table 5-
6). Together, these indices allow the USEPAto
thoroughly evaluate the relative health of its riv-
ers and streams (Klemm et al. 1998).
This section contains the metrics and
indices used by the programs to analyze
benthic macroinvertebrate data. The ana-
Table 5-4. Metrics used in the USEPA' s SBII
and Their Expected Responses to Disturbance
Metric
Expected Response
to Stress
Number of taxa Decrease
Number of EPTtaxa Decrease
% Intolerant taxa Decrease
%Plecopterataxa Decrease
Hilsenhoff Biotic Index Increase
% Oligochaetes and leeches Increase
% Chironomid taxa Increase
In addition to the metrics in Table 5-7,
the USEPA-RBP also suggests the calcula-
tion of the HBI (section 5.6.4) which weighs
the relative abundances of taxa with their tol-
erances to pollution (Barbour et al. 1999).
Ohio EPA evaluates benthic community
fitness using the Invertebrate Community In-
-------
Table 5-6. Indices used by the USEPA-EMAP-SW Protocols
Index Definition
Expected Response
to Perturbation
Percent EPT Number of individuals in each order of Ephemeroptera (mayflies), Decrease
Plecoptera (stoneflies), and Trichoptera (caddisflies) in each
sample divided by the total number of individuals in the sample
Shannon Incorporates both richness and evenness in a measure of general Decrease
Diversity Index diversity and composition; H' = © (N log - n log n))/N, where n is
the total number of individuals of /'* species, N is the total number
of individuals, and © is 3.321928 which converts base 10 log to
base2 log. H'ranges fromO to 3.321928 logN
Hilsenhoff Uses relative abundance weighted by pollution tolerances to Increase
Biotic Index evaluatewater quality. HBI = ((n x a)/N), where n is the total number
of individuals in the /'* taxon, a is the tolerance value assigned to
that taxon, and N is the total number of individuals in the sample.
Stream Benthos Integrates lOmacroinvertebrate population or community metricsinto Increase
Integrity Index* a single biological integrity index score using specimens that have
been identified to genera and/or species levels of identification.
'Currently under development by USEPA-EMAP-SW.
Table 5-7. Metrics Recommended by the USEPA-RBP
Metric
Definition
Expected Response
to Perturbation
Total number of taxa
Number of EPT taxa
Measures the overall variety of the macroinvertebrate
assemblage
Sum of the number of taxa in the insect orders
Decrease
Decrease
Percent dominant taxon
Ratio of Hydropsychidae/
Trichoptera
Ratio scrapers/
(scrapers+filterers)
% shredders
Ephemeroptera, Plecoptera, and Trichoptera
Measures the dominance of the single most abundant Increase
taxon
Number of individuals in Hydropsychidae family Increase
divided by the number of individuals in class
Trichoptera
Number of individual scrapers divided by the sum Decrease
of the number of individual scrapers and filterers
Relative abundance of the shredder functional feeding Decrease
group
-------
dex (ICI). The ICI consists of 10 structural
community metrics, each with four scoring
categories of 6, 4, 2, and 0. The point system
evaluates a sample against a database of 247
relatively undisturbed reference sites through-
out Ohio. Each metric was visually examined
to determine if any relationship existed with
drainage area. When it was decided if a di-
rect, inverse, or no relationship existed, the
appropriate 95% line was estimated and the
area beneath quadrisected as determined by
the distribution of the reference points. Some
percent abundance and taxa richness catego-
ries were not quadrisected, since the data
points showed a tendency to clump at or near
zero. In these situations, a quadripartite
method was used in which one of the four
scoring categories included zero values only,
and the remaining scoring categories were
delineated by an equal division of the refer-
ence data points.
Six points are given if a metric has a value
comparable to those of exceptional stream
communities; 4 points, if comparable to typi-
cal good communities; 2 points, if slightly de-
viating from the expected range of good val-
ues, and 0 points for metric values strongly
deviating from the expected range of good
values. The summation of the individual met-
ric scores results in the ICI value (OEPA
1989). Definitions of metrics and justification
for inclusion in the ICI can be found in Ap-
pendix D.
B-IBI (section 5.1.3) to characterize the
benthic community status. The B-IBI consists
of seven metrics for the CP region, and nine
metrics for the NCP region. The point sys-
tem evaluates a sample against a database of
37 reference sites in Maryland. For each sam-
pling location, metrics are developed and
scores (1, 3, or 5) assigned according to the
thresholds (10th, 50th, or 90th percentiles, re-
spectively) established during the indicator
development process. Raw index scores for
the CP and NCP indices ranged from 7 to 35
and 9 to 45, respectively. These scores were
adjusted to a common scale ranging from 1
to 5, to be consistent with the MDNR-MB SS
fish IBI. On this scale, a score of 4-5 indi-
cates good stream quality, 3-3.9 indicates fair
stream quality, 2-2.9 indicates poor stream
quality, and 1-1.9 indicates very poor stream
quality (Roth et al. 1997b; Stribling et al.
1998).
The MDNR-MB SS calculates the EPT
(section 5.1), the HBI (section 5.1.4), and the
Primarily, programs that conduct benthic
macroinvertebrate surveys have the objective
to assess the overall quality of the studied
stream based on its benthic macroinvertebrate
community. Also, most programs have simi-
larities in their preferred methods for conduct-
ing the surveys. For example, all programs
sample within a defined length of stream, all
programs use multimetric indices in the analy-
ses of macroinvertebrate data, and all programs
compare the index values from individual
sites to reference conditions. However, be-
cause each program has its own subset of
objectives which reflect the needs of the re-
gion it serves, each program has its own sub-
-------
set of methods to meet those obj ectives (Table
5-8).
The USEPA-EMAP-SW uses three in-
dices, EPT, Shannon Diversity (H'), andHBI,
and is currently developing a fourth index,
the SBII. USEPA-RBP suggests using two
indices, EPT and HBI. MDNR uses EPT,
HBI, and B-IBI. Ohio EPA uses the ICI.
USGS-NAWQA does not provide methods
on the calculation of indices from its field data.
The method used to select sampling lo-
cations varies between programs. Programs
frequently choose sites in order to assess a
specific area such as previously studied tar-
get areas or point sources. However, the
EMAP protocols use randomly chosen sites
in order to make a regional assessment of
stream quality. Also, there are differences in
the habitat type in which benthic samples are
Table 5-8. Comparison of Benthic Macroinvertebrate Indices, Sampling Methods, Preferred Sampling
Habitats, and Preferred Sampling Seasons
USEPA-
EMAP-SW
USGS-
NAWQA
USEPA-
RBP
Ohio
EPA
MDNR
MBSS
Indices
EPT X
HBI X
SEE X
B-IBI
ICI
Shannon Diversity (H') X
Sampling Methods
D-Net
Dip Net
Kick Net
Modified Kick Net X
DriftNet X
Hester-Dendy
Slack Sampler
Habitat Types
Riffle Areas X
Pool Areas X
Run Areas
Seasons
Spring Sampling
Summer Sampling X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
-------
taken. MDNR samples only in riffle areas,
Ohio EPA samples primarily in runs, USGS-
NAWQA samples in riffles and pools,
USEPA-RBP suggests sampling in riffle and
run areas, and USEPA-EMAP-SW samples
in riffles, runs, and pools.
The USEPA-EMAP-SW uses a 595-pm
modified kick net sampler and 595-jim drift
nets, USEPA-RBP suggests using a 500-|im
kick net and 500-|im dip net, USGS-NAWQA
uses a 210-jom dip net for qualitative sampling
and a 425-|om sieve for semi-quantitative sam-
pling, MDNR uses a 600-|im mesh D net, and
Ohio EPA uses a Hester-Dendy for quantita-
tive sampling and 600-|im dip nets for quali-
tative sampling. The mesh size used for sam-
pling is not consistent between programs and
this may influence sample content. The vari-
ous methods used to sample benthic
macroinvertebrates from substrate result in
characteristic sampling differences among the
five programs. Ohio EPA uses both natural
and artificial substrate samplers, while USGS-
NAWQA, MDNR, USEPA-EMAP-SW,
and USEPA-RBP use a natural substrate sam-
pler. Using an artificial substrate sampler is a
quantitative method that allows obj ective sam-
pling to take place in areas that are difficult to
reach. However, sampling with artificial sub-
strate takes more time and personnel than
does natural substrate sampling. Also, an ar-
tificial substrate sampler may selectively
sample certain taxa and misrepresent the rela-
tive abundance of taxa in the natural sub-
strates. Natural substrate sampling takes less
time and personnel than does artificial sub-
strate sampling, but it is less quantitative.
-------
by
Joseph E. Flotemersch
The principal methods used by the five
reviewed assessment programs to survey fish
communities are electro-fishing or electro-
fishing in conjunction with seines or nets. The
differences between the programs lie in how
sites are selected, the length of the sample
reach, the amount of time spent sampling,
how the seines or nets are implemented, and
how the data are analyzed. The dissimilari-
ties among the programs' methods are a re-
sult of the differences between the programs'
regions as well as the differences between the
programs' objectives.
gram for assessment is the sampling reach. It
has a length of 40 times the channel width
with a minimum length of 150 m. No maxi-
mum length has been specified.
Currently, both wadeable and boatable
streams are being sampled. However, the only
methods that have been fully documented are
those addressing wadeable systems (Lazorchak
et al. 1998). Methods for boatable systems are
currently being piloted. These methods will be
discussed in this document, but they should be
viewed as pilot methods.
Data collection occurs at randomly se-
lected sites within a designated region (see
section 1.1). Fish are sampled during a sum-
mer index period (July to September), which
coincides with the low flow period of streams
in the research areas. The elementary sam-
pling unit used by USEPA-EMAP-SW pro-
The USEPA-EMAP-SW design utilizes
a single-pass electro-fishing method covering
the determined reach length. In wadeable
streams, block nets are placed at the down-
stream and upstream limits of the sampling
reach when the sample reach is a large con-
tinuous pool. An attempt is made to thor-
oughly fish the entire segment, sampling all
available cover and habitat structures while a
consistent effort is applied over the entire pass.
Sampling is continued for at least 45 minutes
-------
and should not exceed three hours. Seining
may be used in conjunction with electro-fish-
ing to ensure sampling of those species which
may otherwise be under represented by an
electro-fishing survey alone (e.g., darters,
sculpins, benthic cyprinids). Seines may also
be used as block or kick nets to selectively
isolate sections of the stream being electro-
fished (e.g., snags, riffles, cut-banks), in sites
where streams are too deep for electro-fish-
ing to be conducted safely, or in turbid, simple,
soft-bottomed streams where seining is more
effective. Figure 6-1 shows a member of a
field crew using backpack shocker to electro-
fish a wadeable stream.
6.1.2 Beatable Streams
In boatable systems, the stream reach is
fished with a boat-based electro-fishing unit
(Figure 6-2). Electro-fishing begins at the fur-
thest upstream section and proceeds down-
stream until the entire stream reach has been
covered. If the width of a stream requires that
sample reaches exceed 5 km, members of the
pilot field crews have suggested that electro-
fishing the entire reach may not be logistically
wise. In these situations, options include trun-
cating the reach or sampling every other
transect.
6.1.3 Data Recorded
Captured fish are identified in the field,
if possible, and counted. Sport fish and very
large specimens are identified, measured and
released (Figure 6-3). For other species, the
maximum and minimum lengths are recorded.
A voucher sub-sample of 25 individuals from
each species is identified and preserved in
approximately 20% formalin. Additional
specimens (above the 25 voucher) are counted
and released (McCormick and Hughes 1998).
6.2 USGS-NAWQA Fish
Data Collection Methods
The objective of the USGS-NAWQA
characterization offish community structure
Figure 6-1. A field crew member uses a backpack electro-shockerto sample fish in a wadeable stream.
-------
Figure 6-2. A memberof a field crew samples fish using the boat-based electroshocking technique for
beatable rivers.
Figure 6-3. Before being released, the fish are identified, measured and weighed and these data are
recorded.
-------
is to relate fish community traits to physical,
chemical, and other biological factors as part
of an integrated assessment of the nation's
water-quality conditions. Protocols have been
published for wadeable and boatable streams
and both will be discussed in this document
(Meadoretal. 1993 a).
Sampling sites (eitherfixed or synoptic}
are chosen to represent the set of environmen-
tal conditions considered important for con-
trolling water quality in the study unit. Fixed
sites are located at or near USGS gaging sta-
tions where continuous discharge measure-
ments are available. Three sampling reaches
are used to represent environmental conditions
associated with each fixed site. Synoptic sites
are non-gaged sites where one-time samples
of a limited number of physical and chemical
characteristics are measured. Only one sam-
pling reach is generally used to characterize a
synoptic site. The purpose of a synoptic site
is to answer questions regarding source, oc-
currence, or spatial distribution.
Sampling is conducted during low and
stable-flow periods (usually mid-June to early
October). These conditions increase the like-
lihood that samples throughout the study unit
can be collected under similar flow conditions.
The primary determinant of sampling
reach length is geomorphology. An attempt
is made to include at least two types of geo-
morphic channel units in the sampling reach.
Where this is not possible, reaches are cho-
sen that include one meander wavelength,
based on 20 times the distance of the channel
width. The minimum and maximum lengths
of sampling reaches in boatable streams are
500 m and 1,000 m, respectively. The mini-
mum and maximum lengths of sampling
reaches for fish sampling in wadeable streams
is 150 m and 300 m, respectively. These pa-
rameters were set to ensure the efficient col-
lection of representative fish samples.
Wadeable streams are sampled with
backpack (Figure 6-1) or towed electro-fish-
ing gear and, in contrast to other programs,
use a double-pass approach to sampling rather
than a single-pass approach. Backpack
electro-fishing is used in relatively small, shal-
low headwater streams, whereas towed
electro-fishing is employed in relatively wide,
wadeable streams with deep pools. Sampling
is conducted in an upstream direction. All
captured fish are placed immediately in either
a holding box or live well for future process-
ing. After the first pass is completed and all
fish are processed, a second pass is conducted
in the same manner, and usually in the same
area, as the first pass. In order to avoid sam-
pling the same individuals twice, no fish are
released until the second pass is completed.
Following electro-fishing, seining is
used to collect small-sized individuals,
thereby allowing for a more representative
sample to be taken. The seine configuration
and method employed are dependent on the
geomorphic channel units present and the
degree of complexity of the habitat features
within a sampling reach (Meador et. al.
1993a).
Boatable streams are sampled using
electro-fishing boats (Figure 6-2). Sampling
is conducted downstream, from the upstream
boundary of the sampling reach along the
shoreline. This is to allow the fish to swim
into the approaching electrical field. The boat
is operated at a speed equal to or slightly
greater than water velocity. Sampling is con-
ducted in two passes, one for each shoreline.
-------
Beatable streams can also be sampled
using the beach seine in wadeable shoreline
areas. Three samples should be taken from
accessible parts of the upper, lower, and
middle sections of the boatable sampling
reach. The fish from the three seine hauls are
combined and processed before release.
Other sampling methods are used to ob-
tain a representative sample of the fish com-
munity when electro-fishing and seining is not
effective (e.g., in water with extremely-low
conductivity). In situations where seining may
be ineffective because a sampling reach con-
tains a large number of woody snags, debris,
or other obstructions, gill nets and hoop nets
may be used to collect a representative sample
offish. Gill nets capture fish by entangling
them in a fabric mesh that is not actively
moved by man or machine. They require one
trip for deployment, one trip for collection,
and have the potential to be vandalized. Gill
netting can kill fish, therefore, it must not be
conducted in areas where endangered or
threatened species may be present. The net
should be set in the late afternoon and remain
in the water for several hours, but no longer
than 24 hours. The number offish collected
in the gill net is not linearly related to the du-
ration of the set (Hubert 1983), so the exact
duration of the set should depend on flow
conditions and the presence of drifting debris.
Hoop nets capture fish by trapping them
in an enclosed mesh trap. Unlike gill nets, fish
caught by hoop netting can be released with
little or no harm. The duration of the set should
depend on the flow conditions and the pres-
ence of drifting debris. To harvest, the hoop
net is raised at the cod end and the fish are
removed. Two hoop nets are set within the
sampling reach.
Regardless of the sampling method, a
representative sample is taken to provide in-
formation on the presence and relative abun-
dance of the species which represent the fish
community inhabiting the sampled stream. An
attempt is made to identify all fish in the field
to the species level. If there is uncertainty re-
garding the identification of specimens, rep-
resentative samples are preserved in formal-
dehyde for later identification in the labora-
tory (Meador et al. 1993 a).
The USEPA-RBP methods were de-
signed to provide guidance on cost-effective
approaches to problem identification and trend
assessment of our nation's resources. The
methods suggested by the USEPA-RBP for
fish involves careful, standardized field col-
lection, species identification and enumera-
tion, and analyses using aggregated biologi-
cal attributes. Data provided by the fish
USEPA-RBP can serve to assess use attain-
ment, develop biological criteria, prioritize
sites for further evaluation, and assess status
and trends offish assemblage. The suggested
fish collection procedure is a multi-habitat
approach for wadeable streams, which allows
the sampling of habitats in relative proportion
totheirlocal availability (Barbouretal. 1999).
The USEPA-RBP states that for assess-
ment and monitoring, sites can either be tar-
geted, i.e., relevant to special studies that fo-
cus on potential problems, or random, which
provides information of the overall status or
condition of the watershed, basin, or region.
In a random or probabilistic sampling regime,
-------
stream characteristics may be highly dissimi-
lar among the sites, but will provide a more
accurate assessment of biological condition
throughout the area than targeted design. Most
studies conducted by state water quality agen-
cies for identification of problems and sensi-
tive waters are done with a targeted design.
The recommended sampling season is
mid to late summer, when stream and river
flows are moderate to low, and less variable
than during other seasons. The USEPA-RBP
suggest that the stream length to be sampled
can be either a fixed or a proportional dis-
tance, with the selection based on the results
of pilot studies.
The USEPA-RBP endorses electro-fish-
ing as the most comprehensive and effective
single method for collecting stream fishes.
Protocols suggest that collection efforts be-
gin at a shallow riffle, or other physical bar-
rier at the downstream limit of the sample
reach, and terminate at a similar barrier at the
upstream end of the reach. Each sample
should contain riffle, run, and pool habitats,
when available. It is further suggested that if
a reach contains a bridge or a road crossing,
sufficient sampling be conducted upstream of
the structure to minimize the hydrological ef-
fects on the overall quality of the habitat.
The suggested sampling scheme for
wadeable streams uses a two-person crew that
electro-fishes in an upstream direction using
a bank-to-bank sweeping technique that maxi-
mizes coverage area. All wadeable habitats
within the reach should be sampled in a single
pass which terminates at an upstream barrier.
Fish are held in buckets for subsequent iden-
tification.
The USEPA-RBP state that a propor-
tional-distance designation may be desirable
in order to allow for variation in reach length
based on stream width (e.g., 40 times wetted
width). If a proportional distance approach is
used in large streams, electro-fishing should
be limited to a maximum distance of 500 m
or a maximum time of three hours per sam-
pling site (Klemm et al. 1993).
Field identifications of collected fish are
acceptable; however, voucher specimens pre-
served in a formalin solution must be retained
for laboratory verification, particularly if there
is any doubt about the correct identity of the
specimens. Because the collection methods
used are not consistently effective for young-
of-the-year fish and because their inclusion
may seasonally skew bio-assessment results,
it is suggested that fish less than 20 mm in
total length not be identified or included in
standard samples (Barbour et al. 1999).
The selection of fish sampling sites is
based upon several factors including, but not
limited to: 1) location of point source discharg-
ers; 2) stream use designation evaluation is-
sues; 3) location of physical habitat features;
4) location of non-point sources of pollution;
5) variations in macro-habitat; and 6) prox-
imity to ecoregion boundaries. Ohio EPA
methods for boatable and wadeable streams
have been published (OEPA 1988) and both
will be discussed in this document.
Fish sampling generally takes place be-
tween mid-June and mid-October. The total
-------
time a site is fished varies depending on the
current, number offish being collected, and
amount and type of cover within a zone. How-
ever, an Ohio EPA review of electro-fishing
samples suggest at least 1300-1500 seconds
should be spent boat electro-fishing a 0.5 km
stream segment (Ohio EPA 1989).
The principal method used by Ohio EPA
to obtain fish relative abundance and distri-
bution data is pulsed direct current electro-
fishing. Boatable sites are electro-fished for
500 m and wadeable sites are electro-fished
for 150-200 m. Each site is electro-fished two
or three times during the sampling season
(Ohio EPA 1988).
Wadeable streams are sampled with
backpack (Figure 6-1), sportyak or longline
electro-fishing methods developed by Ohio
EPA.
Boatable sites are sampled using electro-
fishing methods based on the work of
Gammon (1973, 1976) and the experience of
the Ohio EPA.
Captured fish are identified in the field
with laboratory vouchers required for any new
locality records, new species, and those speci-
mens that cannot be field identified. The col-
lection techniques used are not consistently
effective for fish less than 15-20 mm in length,
therefore, identification and inclusion in the
sample are not recommended.
assess the status of biological resources in
Maryland's non-tidal streams and determine
the extent to which acidic deposition has af-
fected or may be affecting critical biological
resources in the state. The MDNR-MBSS
targets streams of 3rd order and less. The In-
dex of Biological Integrity (IB I) for fish that
was derived and utilized by the state of Mary-
land compares the condition of biological as-
semblages to that of a regional reference rep-
resenting conditions minimally influenced by
anthropogenic disturbance.
Sample sites were selected in a probabi-
listic manner using a multi-stratification de-
sign. This geographic stratification facilitated
the effective use of a limited number of crews.
Two basins were randomly selected, without
replacement, from each region for each sam-
pling year. One randomly selected basin in
each region was to be visited twice to quan-
tify between-year variability in the response
variables.
The MDNR-MBSS samples a fixed
stream length of 75 m during the summer in-
dex period. Sites are sampled using a double-
pass electro-fishing methodology. In general,
a single electro-fishing unit is used when the
segment width is less than ten meters and two
or more units are used for greater widths.
Block nets are placed at each end of the seg-
ment and direct current backpack electro-fish-
ing units (Figure 6-1) are used to sample the
entire segment. An attempt is made to
throughly fish each segment, sampling all
available cover and habitat structures through-
out the segment. A consistent effort is applied
over the two passes.
The Maryland Biological Stream Survey
(MBSS) is a statewide monitoring survey to
For each pass, all non-game species are
weighed together for an aggregate biomass
-------
measurement and all game species are
weighed in aggregate to the nearest 10 g. For
each pass, up to 50 individuals of each
gamefish species (i.e., trout, bass, walleye,
pike, chain pickerel, or striped bass) are mea-
sured for total length (Figure 6-3). For both
passes, up to 100 fish of each species are ex-
amined for visible external pathology or
anomalies. This sampling approach allows for
the computation of several metrics useful in
calculating a biological index and in produc-
ing estimates offish species abundance.
Also, supplemental electro-fishing is
conducted at non-random sites in which only
the presence of each fish species is recorded.
This provides auxiliary qualitative informa-
tion on fish distributions. For the supplemen-
tal samples, the sampling effort is based on a
minimum of 600 seconds or double the
elapsed time since the last new species was
recorded.
Because boatable streams do not fall
within the framework of the program's ob-
jectives, the MDNR-MBSS does not provide
methods for boatable streams.
Captured fish are identified to species,
if possible, and counted. Any individuals
which cannot be identified to species are re-
tained for laboratory confirmation.
After the processing of the fish collec-
tion is completed in the field, voucher speci-
mens are retained for each species not previ-
ously collected in the drainage basin and the
remaining fish are released. All voucher speci-
mens and fish retained for positive identifica-
tion in the laboratory are examined and veri-
fied (Roth etal. 1997b).
The IBI was first developed by Karr
(1981) for use in small warm water streams
in central Illinois and Indiana, and further re-
fined by Karr et al. (1986). The original ver-
sion had 12 metrics that reflected fish species
richness and composition, number and abun-
dance of indicator species, trophic organiza-
tion and function, reproductive behavior, fish
abundance, and condition of individual fish.
Each metric received a score of five points if
it had a value similar to that expected for a
fish community characteristic of a system with
little human influence, a score of one point if
it had a value similar to that expected for a
fish community that departs significantly from
the reference condition, and a score of three
points if it had an intermediate value.
The original version of the IBI quickly
became popular. As it became more widely
used, different versions were developed for
different regions and different ecosystems.
These new versions also had multi-metric
structures, but differed from the original ver-
sion in the number, identity, and scoring of
metrics (Simon and Lyons 1995). Some ver-
sions developed for streams and rivers retain
many of the original IBI metrics, with metrics
usually being modified as a part of an effort
to compensate for insensitivities to environ-
mental degradation in a particular geographic
area or type of stream. Similarly, the metrics
used in versions of the IBI developed for other
types of ecosystems, such as estuaries, im-
poundments, and natural lakes, usually bear
a limited resemblance to those of the original
version yet retain its overall structure (Simon
and Lyons 1995).
-------
The multi-metric indices currently used
by USEPA-EMAP-SW, Ohio EPA, and
MDNR-MBSS have all followed this same
chain of development; all contain some metrics
with origins in Karr et al.'s 1986IBI. In gen-
eral, selection of which metrics to drop,
modify or add have been determined by first
developing a list of candidate metrics (vari-
ous attributes of the fish assemblages) and then
statistically determining which formulations
were effective in discriminating between ref-
erence sites and sites known to be degraded.
The Index ofWell-Being (Iwb), or com-
posite index, was developed by Gammon
(1976) to evaluate the response of riverine fish
communities to environmental stress. This
index was tested using data from the Wabash
River in Indiana (Gammon 1976; Gammon
et al. 1981) and subsequently from other riv-
ers in Indiana, Ohio (Yoder et al. 1981;
Gammon 1980), and Oregon (Hughes and
Gammon 1987). Some investigators have
modified the original Iwb for specific appli-
cations.
The Iwb incorporates four measures of
fish communities that have traditionally been
used separately; numbers of individuals, bio-
mass, the Shannon diversity index (H') based
on numbers offish, and the Shannon diver-
sity index (H'} based on weights of fish
(OEPA 1989). The computational formulas
for the Iwb and Shannon index are provided
in Appendix E.
USEPA-RBP, Ohio EPA, and MDNR-
MBSS programs endorse or have developed
versions of the IBI (Karr 1981) for use in their
respective waters. The IBI includes discrete
measurements of assemblage attributes, or
metrics based on species composition, trophic
composition, abundance, and condition
(Davis 1995). In addition to the IBI, Ohio
EPA subjects data to a modified version of
the Index of Weil-Being (Iwb). The Iwb is
based on structural attributes of the fish com-
munity whereas the IBI additionally incorpo-
rates functional characteristics. Their use in
combination is suggested by Ohio EPA
(1988) to provide a rigorous evaluation of
overall fish community condition. The
USGS-NAWQA program does collect infor-
mation on aquatic vertebrates, but specific
methods used to interpret data were not avail-
able as of the completion of this document.
The USGS-NAWQA program does not rely
on a single index approach such as the IBI;
rather, a combination of multivariate and
multimetric approaches to data analysis are
used to examine factors affecting biological
water-quality characteristics. Indices that have
been locally or regionally calibrated to refer-
ence conditions are used at the study-unit level
where required data are available.
There are two primary indices utilized
by these assessment programs to interpret
collected fish data. The USEPA-EMAP-SW,
The goal of the USEPA-EMAP-SW
program is to monitor the condition of the
Nation's ecological resources, to evaluate the
success of current policies and programs, and
to identify emerging problems before they
become widespread or irreversible (Gurtz and
Muirl994).
The USEPA-EMAP-SW program is in
the process of developing an IBI for wade-
able streams in the MAH region of the United
-------
States. The USEPA-EMAP-SW MAH ver-
sion of the IBI is being developed by exam-
ining the responses offish community metrics
to physical, chemical, habitat and landscape
indicators of catchment disturbance.
Univariate and multivariate analyses of rela-
tionships among fish community metrics,
habitat integrity and anthropogenic distur-
bance are being used to develop this index.
Table 6-1 lists the metrics proposed for inclu-
sion in the index.
USEPA scientists developed their IBI by
randomly selecting sampling sites in the des-
ignated study area, collecting field measure-
ments, and then analyzing the resulting data,
with respect to candidate metrics, in order to
establish expectations for minimally degraded
streams. Reference values were derived from
sites scoring in the upper 15% of all sites
sampled. Individual sites were therefore com-
pared to this reference condition rather than
upstream, or similar stream, individual "ref-
erence sites" selected as being minimally im-
pacted, as is commonly practiced, by best pro-
fessional judgement. This IBI is being devel-
oped for wadeable systems and its metrics are
not adjusted for watershed size. This is prob-
ably a reflection of the size of the watersheds
of the study area (most are less than 500 km2),
the fact that these were predominantly upland
systems, and the historical biogeography of
the fish fauna.
The 16 metrics of the MAH IBI will be
scored continuously from 0-10 and the result-
ing IBI scores converted from a range of 0-
160 to a range of 0-100%. No information is
currently available concerning the develop-
ment of an IBI for boatable systems
(McCormick and Hughes 1998).
The initial steps in deriving an IBI score
for a wadeable location involves the collec-
tion and identification of samples and enter-
Table 6-1. Metrics in the Index of Biotic
Integrity forthe USEPA-EMAP-SW Program.
Metric
Expected response
to stress
Native species richness Decrease
Native family richness Decrease
Sensitive species richness Decrease
Tolerant individuals Increase
Benthic species richness Decrease
Water column species richness Decrease
Alien individuals Increase
Number of trophic guilds Decrease
Percent top carnivores Decrease
Percent invertivore individuals Decrease
Percent herbivores Decrease
Percent omnivore individuals Increase
Number of specialized reproduc- Decrease
live strategies
Proportion of gravel spawning Decrease
species
Proportion tolerant substrate Increase
spawners
Total abundance Decrease
ing collected information into a database.
Once this process is complete, species-spe-
cific information relevant to the metrics is
determined. This information is obtained from
a list that contains the taxa occurring in the
waters of the study area as well as designa-
tions for use in IBI metrics (Appendix F).
Parameters assigned to individual species in-
clude tolerance, trophic status, habitat prefer-
ence, reproductive strategy, and watersheds
-------
to which the species is native. Totals are de-
rived and metrics are scored and summed.
Streams with an IBI value of >85% are used
as the reference condition, scores between 70-
85% are acceptable., streams with IBI values
between 50-70% are marginally impaired.,
and IBI scores below 50% are highly im-
paired.
Protocols for the interpretation of fish
data collected from boatable sites have yet to
be developed (McCormick and Hughes
1998).
Table 6-2. Metrics Recommended for Calculation
bytheUSEPA-RBP
The methods used by the USGS-
NAWQA program to interpret information
collected on aquatic vertebrates program is not
available as of the completion of this docu-
ment (Meador et al. 1993a). The USGS-
NAWQA program does not rely on a single
index approach such as the IBI; rather, a com-
bination of multivariate and multimetric ap-
proaches to data analysis are used to examine
factors affecting biological water-quality char-
acteristics. Indices that have been locally or
regionally calibrated to reference conditions
are used at the study-unit level where required
data are available.
Metric
Expected Response
to stress
Total number of fish species Decrease
Number and identity of darter Decrease
species
Number and identity of sunfish Decrease
species
Number and identity of sucker Decrease
species
Number and identity of intolerant Decrease
species
Proportion of individuals as green Increase
sunfish
Proportion of individuals as Increase
omnivores
Proportion of individuals as Decrease
insectivorous cyprinids
Proportion of individuals as top Decrease
carnivores
Number of individuals in sample Decrease
Proportion of individuals as Increase
hybrids
Proportion of individuals with Increase
disease, tumors, fin damage, and
skeletal anomalies
The USEPA-RBP endorses the techni-
cal framework of the multi-metric Index of
IBI developed by Karr (1981) for the assess-
ment offish assemblages. The 12 metrics in-
cluded in Karr's (1981) original IBI are in
Table 6-2.
Although the USEPA-RBP recom-
mends the framework of Karr's (1981) IBI,
they also recommend that some modifications
may be needed to adjust for the regional dif-
ferences between surveys. The protocols fur-
ther state that the IBI "serves as an integrated
analysis because individual metrics may dif-
fer in their relative sensitivity to various lev-
els of biological condition" (Barbour et al.
1999). Calculation and interpretation of the
IBI involves a sequence of activities includ-
ing, fish sample collection, data tabulation,
-------
regional modification and calibration of
metrics, and determination of expected val-
ues (Barbour et al. 1999). Once this process
is complete, species-specific information rel-
evant to the metrics can be assigned.
For each sampling location, metrics are
developed and scores (1, 3, or 5) are assigned
according to the thresholds established dur-
ing the indicator development process. The
final IBI score is the sum of all metric scores
(Barbour etal. 1999).
Table 6-3. Metrics Employed by the Ohio EPA
with Expected Response to Stress.
The Ohio EPA assessment program was
designed to support all state agency surface
water programs. Ohio EPA has used measur-
able characteristics of instream fish since
1980. The principal measures of overall fish
community health used by the Ohio EPA are
the Iwb, developed by Gammon (1976) and
modified by Ohio EPA, and the IBI devel-
oped by Karr (1981).
The IBI utilized by Ohio EPA contains
12 metrics specifically tailored to Ohio sur-
face waters and Ohio EPA sampling meth-
ods. The IBI metrics used by the Ohio EPA
to evaluate wading sites (Table 6-3; Appen-
dix F) closely approximate those proposed by
Karr (1981) and refined by Fausch et al.
(1984) and Karr et al. (1986). Substantial
modifications were necessary for the IBI
metrics used for the boat sites and headwater
sites. These changes were made in recogni-
tion of the different sampling efficiency and
selectivity of the boat methods and the differ-
ent faunal characters of larger streams and riv-
ers and headwater areas. However, these
modifications were made in keeping with the
guidance given by Karr et al. (1986). Three
basic divisions are made; wading sites, boat
Metric
Expected response
to stress
Total number of species1 (a,b,c) Decrease
Number of darter species Decrease
(a2,b)/Percent round-bodied
suckers3 (c)
Number of headwater species (a)/ Decrease
Number of sunfish species (b,c)
Number of minnow species (a)/ Decrease
Number of sucker species (b,c)
Number of sensitive species (a)/ Decrease
Number of intolerant species (b,c)
Percent tolerant species (a,b,c) Increase
Percent omnivores (a,b,c) Increase
Percent insectivorous species Decrease
(a,b,c)
Percent pioneering species (a)/ Decrease
Percent top carnivores (b,c)
Number of individuals4 (a,b,c) Decrease
Number of simple lithophilic Decrease
species (a)/Percent simple lithophils
(b,c)
Percent DELT anomalies5 (a,b,c) Increase
^Headwater sites, drainage areas less than 20 mi2.,
sampled with wadeable methods.
bWading sites, sites sampled with wadeable
methods.
°Boat sites, these sites are sampled with boat
methods.
'Excludes exotic species.
Includes sculpins.
Includes suckers in the genera Hypentelium,
Moxostoma, Minytrema, and Erimyzon', excludes
white sucker (Catostomus commersoni).
"Excludes species designated as tolerant, hybrids,
and exotics.
Includes deformities, eroded fins, lesions, and
external tumors (DELT).
-------
sites, and headwater sites. Generally, wading
sites are those having a drainage area of less
than 300 mi2 but greater than 20 mi2. Boat
sites include streams and rivers that are too
deep and large to sample effectively with
wading methods. Boat sites generally exceed
100-300 mi2 in drainage area. Headwaters
sites are defined as sampling locations with
drainage areas less than 20 mi2.
The value of each metric is compared to
the value expected at a reference site located
in a similar geographic region where human
influence has been minimal. Ratings of 5, 3,
or 1 are assigned to each metric according to
whether its value approximates (5), somewhat
deviates from (3), or strongly deviates from
(1) the value expected at a reference site. The
maximum IBI score possible is 60 and the
minimum is 12. Reference site scores are
grouped by ecoregion (Omernik 1987) and
used to statistically generate region specific
use attainment criteria (OEPA 1988).
The Iwb used by the Ohio EPA is a
modified version of that developed by
Gammon (1976). The Iwb is based on struc-
tural attributes of the fish community. Four
measures offish communities that tradition-
ally have been used separately are: numbers
of individuals, biomass, and the Shannon di-
versity index (//') based on numbers and
weights offish.
The modified Iwb retains the same com-
putational formula as the conventional Iwb
developed by Gammon (1976). The differ-
ence is that highly tolerant species, exotic spe-
cies, and hybrids are eliminated from the num-
bers and biomass components of the Iwb.
However, tolerant and exotic species are in-
cluded in the two Shannon index calculations.
This modification eliminates the undesired
effect caused by the high abundance of toler-
ant species, but retains their desired influence
on the Shannon indices. Computational for-
mulas for the index of well being and the
Shannon diversity index are in Appendix E.
Maryland scientists began their develop-
ment of an IBI by first establishing expecta-
tions for minimally degraded streams and then
comparing the ability of candidate metrics to
discriminate between these reference sites and
sites known to be degraded. The resulting IBI
consists of eight metrics (Table 6-4), each of
which quantitatively describe attributes of the
biological community. Each of the metrics
used has an expected direction of change in
response to anthropogenic stress. For each
sampling location, metrics are developed and
scores (1, 3, or 5) assigned according to the
thresholds established during the indicator
development process. The final IBI score is
the mean the metric scores. No IBI score is
assigned to sites having watershed area less
than 300 acres (Roth et al. 1997b).
The initial steps in deriving an IBI score
for a location involves collecting, identifying,
and entering collected information into a da-
tabase. Once this process is complete, spe-
cies specific information relevant to the
metrics can be assigned. This information is
obtained from a Maryland fish species list that
contains designations for use in IBI metrics
(Appendix F). Parameters assigned to indi-
vidual species included tolerance, trophic sta-
tus, native or non-native status by watershed,
if the species was considered benthic, and if
the species was a lithophilic spawner. Totals
are derived and metrics scored as in Appen-
dix E. The metrics used by the MDNR-MBSS
for their IBI are given in Table 6-4 (Roth et
al. 1997c; Striblingetal. 1998).
-------
Table 6-4. Metrics Employed by MDNR-MBSS
and Expected Response to Stress.
Metric
Expected response
to stress
Number of native species Decrease
Number of benthic species Decrease
Percent of tolerant individuals Increase
Percent abundance of dominant Increase
species
Percent generalist, omnivores, Increase
and invertivores
Number of individuals perm2 Decrease
Biomass (g per m2) Decrease
Percent lithophilic spawners Decrease
Percent insectivores Decrease
Site selection - The method used to de-
termine the location of the sampling sites var-
ies among the five programs discussed in this
document. For the USEPA-EMAP-SW and
MDNR-MBSS, sampling sites are randomly
selected. The USGS-NAWQA usually uti-
lizes fixed sampling sites. Ohio EPA selects
its sites based on site-specific and regional is-
sues. The USEPA-RBP states that for assess-
ment and monitoring, sites can either be "tar-
geted", i.e., relevant to special studies that
focus on potential problems, or "random",
which provides information of the overall sta-
tus or condition of the watershed, basin, or
region.
Sampling season/Index period - All five
programs reviewed either use or endorse the
use of a summer index period. The general
consensus for this is that this period coincides
with the low and stable flow period; these
conditions increasing the likelihood that
samples throughout the study will be collected
under similar flow conditions.
Stream distance sampled/sampling
reach - The method used to determine the
stream length to be sampled at a chosen site
varies among the selected programs. The
USEPA-EMAP-SW program uses a stream
length that is 40 times the wetted width or
150m, whichever is greater. The reach length
sampled by the USGS-NAWQA program
includes two types of geomorphic channel
units or 20 times the channel width if repeti-
tive geomorphic channel units are not present.
Acceptable ranges for wadeable streams is
150 to 300 m where the acceptable range for
boatable stream is 500 to 1000 m. Ohio EPA
samples 150 to 200 m in wadeable streams
and 500 m in boatable streams. MDNR-
MBSS uses a fixed stream length of 75 m.
The USEPA-RBP manual suggests that ei-
ther a fixed-distance method or a proportional-
distance method of determining reach length
would be acceptable, but final decisions
should be based on the goals of the study as
well as results of pilot studies conducted in
the study area.
Sampling method - All of the programs
reviewed in this document use electro-fish-
ing, either alone or in conjunction with other
sampling gear, to assess fish populations.
Ohio EPA uses electro-fishing exclusively in
-------
both wadeable and beatable streams. Each
stream length is sampled in either 2 or 3 passes
per sampling season with the electro-fishing gear.
The USEPA-EMAP-SW and USGS-
NAWQA use electro-fishing methods with the
assistance of additional gear, principally seines.
The two programs differ, however, in that the
USEPA-EMAP-SW program electro-fishes one
bank of the designated stream length in one pass
whereas the USGS-NAWQA program uses a
double-pass sampling scheme to sample both
banks on the same day. The MDNR-MBSS
also uses a double-pass electro-fishing method
to sample both banks on the same day in addi-
tion to incorporating the use of block nets to
delimit the reach if necessary. The use of seines
to delimit a stream reach is also occasionally
employed by the USEPA-EMAP-SW program.
The USEPA-RBP endorses a single pass
electro-fishing method supplemented with sein-
ing and further suggests the use of block nets to
delimit the reach if necessary.
Measure offish community health - Many
of the metrics used in the regionally-developed
IBIs overlap between the programs. Among the
three programs that have published IBIs, the
number of metrics employed varies. The
USEPA-EMAP-SW IBI contains 16 metrics,
the Ohio EPA IBI contains 12 metrics, and the
MDNR-MBSS IBI contains 8 metrics. Within
programs, some metrics vary depending upon
the size of the stream sampled (Ohio EPA) or
upon its location (MDNR-MBSS).
In addition to its own IBI, the Ohio EPA
also uses a modified version of Gammon's
(1976) Iwb. This index incorporates measure-
ments concerning the structure of the fish com-
munity.
All sampled sites are scored against an es-
tablished set of criteria. The USEPA-EMAP-
SW program compares sampled sites to expec-
tations for minimally degraded streams. Mini-
mally impacted values were derived from sites
scoring in the upper 15% of all sites sampled.
Individual sites are therefore compared to a
reference condition rather than values derived
from minimally impacted reference sites. The
USGS-NAWQA, USEPA-RBP, Ohio EPA,
and MDNR-MBSS programs either use or
suggest the use of reference sites. This in-
volves comparing sampled sites to the value
expected at a reference site located in a simi-
lar geographic region where human influence
has been minimal.
Different researchers and programs may
have different reasons for conducting
bioassessments and these differences do not
necessarily require the same level or type of ef-
fort in sample collection, taxonomic identifica-
tion, or data analysis (Gurtz and Muir 1994).
However, different methods of sampling and
analysis may yield comparable data for certain
objectives despite differences in effort (Barbour
et al. 1999). As an example, we can compare
the conclusions drawn by different programs
conducting research in the same areas. A pilot
field study comparing some of the methods of
three of the reviewed programs (USEPA-
EMAP-SW, USGS-NAWQA, Ohio EPA) con-
currently in large river systems was conducted
in the summer of 1999. Such studies will yield
useful information about methods employed,
especially in reference to the effectiveness of
compared methods in detecting differences
when they exist and not detecting differences
when they do not exist. Such comparisons
would also be beneficial to cost and benefit
analyses of methodologies.
-------
Allan, J.D., and E. Russek. 1985. The
quantification of stream drift. Canadian
Journal of Aquatic Science42(2):210-215.
Allison, L.N., J.G. Hnath, and W.G. Yoder.
1977. Manual of common diseases,
parasites, and anomalies ofMichigan fishes.
Michigan Department of Natural Re-
sources, Lansing, MI. Fish Management
Report No. 8.
Aloi, I.E. 1990. A critical review of recent
freshwater periphyton field methods.
Canadian Journal of Fisheries and Aquatic
Science 47:656-670.
APHA. 1991. Standard methods for the
examination of water and wastewater. 18th
edition. American Public Health Associa-
tion, Washington, D.C.
Archibald, R.E.M. 1972. Diversity in some
South African diatom associations and its
relation to water quality. Water Research
6:1229-1238.
Bahls, L.L., R Burkantis, and S. Tralles. 1992.
Benchmark biology of Montana reference
streams. Department of Health and
Environmental Science, Helena, MT.
Bahls, L.L. 1993. Periphyton bioassessment
methods for Montana streams. Montana
Water Quality Bureau, Department of
Health and Environmental Science, Helena,
MT.
Ball, J. 1982. Stream classification guidelines for
Wisconsin. WisconsinDepartmentofNatural
Resources Technical Bulletin, Wisconsin
Department ofNatural Resources, Madison,
WI.
Balon,E.K. 1975.Reproductiveguildsoffishes: A
proposal and definition. Journal of the
Fisheries Research Board of Canada 32:821-
864.
Barbour,M.T.,andJ.B. Stribling. 1991. Use of
habitat assessmentinevaluatingthebiological
integrity of stream communities. Pages 25-3 8
in G. Gibson (ed.). Biological criteria:
Research and regulation, proceedings of a
symposium, 12-13 December 1990, Arling-
ton, Virginia. Office of Water, U.S. EPA,
Washington, D.C., EPA-440-5-91-005.
Barbour, M.T., and J. Gerritsen. 1996.
Subsamplingofbenthicsamples:Adefenseof
the fixed-count method. Journal of the North
AmericanBenthological Society 15(3):386-
391.
Barbour, M.T., J. Gerritsen, B.D. Snyder, and J.B.
Stribling. 1999.Rapidbioassessment proto-
cols for use in streams and wadeable rivers:
-------
Periphyton, benthic macroinvertebrates, and
fish, second edition. EPA 841-B-99-002.U.
S. Environmental Protection Agency; Office
ofWater; Washington, D.C.
Baumann, P.C., W.D. Smith, and W.K.
Parland. 1987. Tumor frequencies and
contaminant concentrations in brown
bullhead from an industrialized river and
recreational lake. Transactions of the
American Fisheries Society 116(l):79-86.
Becker, G.C. 1983. Fishes of Wisconsin.
University ofWisconsinPress, Madison, WI.
Berkman, H. E., and C. F. Rabeni. 1987. Effects
of siltation on stream fish communities.
EnvironmentalBiology ofFishes 18:285-294.
Berra, T.M., and R. Au. 1981. Indices of
teratological fishes from Cedar Fork Creek,
Ohio. Ohio Journal of Science 81(5):225.
BiomonitoringofEnvironmental Status and Trends
(BEST) Program Staff. 1996. Summary
report from a workshop on selection oftier 1
bioassessmentmethods.NationalBiological
Service Information and Technology Report
7.
Chutter, F .M. 1972. An empirical biotic index of
the quality of water in South African streams
andrivers. Water Research 6:19-30.
Clesceri, L. S., A.E. Greenberg, andRR. Trussell
(eds.). 1989. Standard methods for the
examination of waster and wastewater, 17th
ed.: American Public Health Association,
American Water Works Association, and
Water Pollution Control Federation, Wash-
ington, D.C.
Cooper, J.M., and J.L. Wilhm. 1975. Spatial
and temporal variability in productivity,
species diversity, and pigment diversity of
periphyton in a stream receiving domestic
and oil refinery effluents. Southwestern
Naturalist 19:413-428.
Coste, M. 1974. A biological method for the
assessment of the quality of the Main River
based on benthic and periphytic diatom
communities. CommissionEuropean Com-
munity Health Protection Direction: Koblenz
1975.
Cuffney, T., M.E. Gurtz, and M.R. Meader.
1993a. Methods for collecting benthic
invertebrate samples as part of the National
Water Quality AssessmentProgram. U.S.
Geologic Survey Open File Report 93 -406.
Cuffney, T., M.E. Gurtz, and M.R. Meader.
1993b. Guidelines for processing and
quality assurance of benthic invertebrate
samples collected as part of the National
Water-Quality AssessmentProgram. U.S.
Geologic Survey Open File Report 93 -407.
Davis, W.S. 1995. Biological assessment and
criteria: Building on the past. Pages 15-29 in
W.S. Davis and T.P. Simon, editors.
Biological assessment and criteria: Tools for
water resource planning and decision
making. Lewis Publishers, BocaRaton, LA.
Descy, J.P. 1979. A new approach to water
quality estimation using diatoms. Nova
Hedwigia 64:305-323.
DeShon, J. 1995. Development and application
of the invertebrate community index (ICI).
Pages 217-243 in W.S. Davis, and T.P.
Simon (eds.). Biological assessment and
criteria: Tools for water resource planning
and decision making. Lewis Publishers,
BocaRaton, LA.
Dixit, S.S., J.P. Smol, J.C. Kingston, andD.F.
Charles. 1992. Diatoms: Powerful indica-
tors of environmental change. Environmen-
tal Science and Technology 26:23-33.
-------
Fausch, D.D., J.R. Karr, andP.R. Yant. 1984.
Regional application of an index of biotic
integrity based on stream fish communities.
Transactions of the American Fisheries
Society 113:39-55.
Fitzpatrick, F. A., I.R. Waite, P.J. D'Arconte,
M.R. Meador, M.A. Maupin, and M.E.
Gurtz. 1998. Revised methods for charac-
terizing stream habitat in the National
Water-Quality Assessment Program. U.S.
Geological Survey Water Resources
Investigations. Report 98-4052.
Fjerdingstad,E. 1950. The microfauna of the river
Molleaa; with special reference to therelation
of the benthic algae to pollution. Folia
LimnologicaScandinavica,No. 5.
Fjerdingstad, E., 1960. Forurening afvaandlob
biologiskbedomtNordiskHygieniskTidskrift
41:149-196.
FloridaDepartment of Environmental Protec-
tion (FLDEP). 1996. Standard operating
procedures for biological assessment.
Florida Department of Environmental
Protection, Biology Section.
Frissell, C. A, W. J. Liss, C.E. Warren, andM.D.
Hurley. 1986. Ahierarchicalframeworkfor
stream habitat classification-Viewing streams
in watershed context: Environmental Manage-
ment 10:199-214.
Gammon, J.R. 1973. The effect of thermal
inputs on the populations of fish and
macroinvertebrates in the Wabash River.
Purdue University Water Resources Center
Technical Report 32.
Gammon, J.R. 1976. The fish populations of the
middle 340 km of the Wabash River.
Purdue University Water Resources Center
Technical Report 86.
Gammon, J.R. 1980. The use of community
parameters derived from electrofishing
catches of river fish as indicators of
environmental quality, in seminar on water
quality tradeoffs. U.S. Environmental
Protection Agency, Washington, D.C.
EPA-905/9-80-009.
Gammon, J.R., A. Spacie, J.L. Hamelink, and
R.L. Kesler. 1981. Role of electrofishing in
assessing environmental quality ofthe Wabash
River. Pages 307-324 in J.M. Bates and C.I
Weber (eds.). Ecological assessment of
effluentimpacts on communities of indigenous
aquatic organisms. American Society for
Testing and Materials, Philadelphia, PA.
Special Technical Publication 730.
Gilliom, R.J., W.M. Alley, and M.E. Gurtz.
1995. Design of the National Water-
Quality Assessment Program — Occur-
rence and distribution of water-quality
conditions: U. S. Geological Survey Circular
111.
Gurtz, M.E., and Thomas A. Muir (eds.). 1994.
Report of the interagency biological methods
workshop. USGS Open-File report 94-490,
Raleigh, NC.
Harvey, C.A., and D.A Eash. 1996.
Description, instructions, and verification
for Basinsoft, a computer program to
quantify drainage-basin characteristics:
U.S. Geological Survey Water-Resources
Investigations Report 95-4287.
Herlihy, A.T. 1998. Water chemistry. Pages
57-66 in J. Lazorchak, D.J. Klemm, and
D. V. Peck (eds.). Environmental monitor-
ing and assessment program-surface wa-
ters: Field operations and methods for
measuring the ecological condition of
wadeable streams. U.S. Environmental
Protection Agency, Office of Research and
-------
Development, Cincinnati, OH. EPA/620/
R-94/004F.
Hill,B. 1998. Periphyton. Pages 199-132 in J.
Lazorchak, DJ. Klemm, and D.V. Peck
(eds.). Environmental monitoring and
assessment program-surface waters: Field
operations and methods for measuring the
ecological condition of wadeable streams.
U.S. Environmental Protection Agency,
Office of Research and Development,
Cincinnati, OH. EPA/620/R-94/004F.
Hill,B. 1999. Quality assurance project plan for
support of the periphyton and sediment
metabolism indicators: U. S. Environmental
Protection Agency, Cincinnati, OH. EPA
Contract Number 68-C6-0019: Work
Assignment 2-08.
Hilsenhoff, W.L. 1977. Use of arthropods to
evaluate water quality of streams. Technical
BulletinNo. 100, Wisconsin Department of
Natural Resources, Madison, WI.
Hilsenhoff, W.L. 1987. An improved biotic
index of organic stream pollution. Great
LakesEntomologist20:31-39.
Hilsenhoff, W.L. 1988. Rapid field assessment
of organic pollution with a family level biotic
index. Journal of the North American
Benthological Society 7(l):65-68.
Hohn, M.H. 1959. The use of diatom
populations as a measure of water quality in
selected areas of Galveston and Chocolate
Bay, Texas. Publications of the Institute of
Marine Science 6:206-212.
Hubert, W.A. 1983. Passive capture tech-
niques. Pages 95-122 in L.A.Nielsen and
D.L. Johnson (eds). Fisheries techniques.
American Fisheries Society, Bethesda,
MD.
Hughes, R.M., and J.R. Gammon. 1987.
Longitudinal changes in fish assemblages
and water quality in the Willamette River,
Oregon. Transactions of the American
Fisheries Society 116(2): 196-209.
Janicki, A.D., D. Wade, H. Hemibuch, P.
Wilson, P. Jacobson, andP. Kazyak. 1993.
Maryland biological stream survey design
report. Maryland Department of Natural
Resources, Chesapeake Bay and Research
and Monitoring Division, Linthicum, MD.
CBRM-AD-93-1.
Jenkins, R.E., and N.M. Burkhead. 1993.
Freshwater fishes of Virginia. American
Fisheries Society, Bethesda, MD.
Jones, J.R., B.H. Tracy, J.L. Sebaugh, D.H.
Hazelwood,andM.M. Smart. 1981. Biotic
index tested for ability to assess water
quality of Missouri Ozark streams.
Transactions of the American Fisheries
Society 110(5):627-637.
Karr, J.R. 1981. Assessment of biotic integrity
usingfish communities. Fisheries 66:21-27.
Karr, J.R., K.D. Fausch, P.L. Angermeier, P.R
Yant, andl.J. Schlosser. 1986. Assessing
biological integrity in running waters: A
method and its rationale. Illinois Natural
History Survey Special Publication 5.
Kaufmann,P.R. (ed.). 1993. Physical habitat.
Pages 59-69 /«R.M. Hughes (ed.). Stream
indicator and design workshop. U.S.
Environmental Protection Agency, Corvallis,
OR.EPA/600/R-93/138.
Kaufmann,P.R., and E.G. Robison. Physical
habitat characterization. 1998. Pages 77-
118 in J.M. Lazorchak, D.J. Klemm, and
D. V. Peck (eds.). Environmental monitor-
ing and assessment program surface waters:
-------
Fieldoperationsandmethodsformeasuring
the ecological condition of wadeable
streams. Office of Research and Develop-
ment, U.S. Environmental Protection
Agency, Cincinnati, OH. EPA/620/R-94/
004F.
Kazyak, P.P. 1995. Maryland biological stream
survey manual. Maryland Department of
Natural Resources, Chesapeake Bay Re-
search andMonitoringDivision.
Kentucky Department of Environmental Pro-
tection (KDEP). 1993. Methods for
assessing biological integrity of surface
waters. Kentucky Department of Environ-
mental Protection, Division of Water,
Frankfort, KY.
Klemm, D.J., P.A. Lewis, F. Fulk, and J.M.
Lazorchak. 1990. Macroinvertebrate field
and laboratory methods for evaluating the
biological integrity of surface waters. U.S.
Environmental Protection Agency, Envi-
ronmental Monitoring and Support Labora-
tory, Cincinnati, OH. EPA-600-4-90-030.
Klemm, D. J., Q. J. Stober, and J.M. Lazorchak.
1993. Fish field and laboratory methods for
evaluatingthe biological integrity of surface
waters. Environmental Monitoring and
Support Laboratory, U. S. Environmental
Protection Agency, Cincinnati, OH. EPA/
600/R-92/111.
Klemm, D.J., J.M. Lazorchak, andP.A. Lewis.
1998. Benthicmacroinvertebrates. Pages
147-160 in J.M. Lazorchak, D.J. Klemm,
and D. V. Peck (eds.). Environmental
monitoring and assessment program surface
waters: Field operations and methods for
measuring the ecological condition of
wadeable streams. Office of Research and
Development, U. S. Environmental Protec-
tion Agency, Cincinnati, OH. EPA/620/R-
94/004F.
Kolkwitz, R, andM. Marsson. 1908. Okologie
der pflanzlichen saprobien. Festschrift zur
Feier des 25 Jahrigen Bestehens der
Deutschen Botanischen Gesellschaft,
Berichte der Deutschen Botanischen
Gesellschaft26A:505-519.
Kuehne, R.A., andR.W. Barbour. 1983. The
American Darters. University Press of
Kentucky, Lexington, KY.
Lange-Bertalot, H. 1979. Pollution tolerance as
a criterion for water quality estimation.
NovaHedwigia64:285-304.
Lazorchak, J.M., B.H. Hill, and D.J. Klemm
(eds.). 1999. Environmental monitoring and
assessment program surface waters: Field
operations and methods for measuring the
ecological condition of large rivers (Draft
version). Office of Research and Develop-
ment, U.S. Environmental Protection
Agency, Cincinnati, OH.
Lazorchak, J.M., D.J. Klemm, andD. V. Peck.
1998. Environmental monitoring and as-
sessment program surface waters: Field
operations and methods for measuring the
ecological condition of wadeable streams.
Office of Research and Development, U. S.
Environmental Protection Agency, Cincin-
nati, OH. EPA/620/R-94/004F.
Leidy, R.A., andP.L. Fielder. 1985. Human
disturbance and patterns offish species
diversity in the San Francisco Bay drainage,
California. Biological Conservation 33:247-
267.
Lemmon, P.E. 1957. A new instrument for
measuring forest overstory density. Journal
of Forestry 55:667-669.
-------
Leopold, L.B., andM.G. Wolman. 1957. River
channel patterns: Braided, meandering, and
straight. United States Geological Survey
Professional Paper 282-B.
Lenat, D.R. 1987. Water quality assessment
using a new qualitative collection method for
freshwater benthic macroinvertebrates.
North Carolina DEM Technical Report,
Raleigh, NC.
Liebmann, H. 1962. Handbuch der frischwasseer-
undabwasseerbiologie, Volume 1. (2. Aufl.)
Munchen: R. Oldenbourg.
Linsley, R.K., M.A. Kohler, and J.L.H.
Paulhus. 1982. Hydrology for engineers.
McGraw-Hill Book Co. New York, NY.
Lowe, R.L. 1974. Environmental requirements
and pollution tolerance of freshwater
diatoms. U.S. Environmental Protection
Agency, Environmental Monitoring Series,
Cincinnati, OH.
Matthews, R.A., P.P. Kondratieff, and A.L.
Buikema, Jr. 1980. Afield verification of the
use and of the autotrophic index in
monitoring stress effects. Bulletin of
Environmental Contamination and Toxicol-
ogy 25:226-233.
McCormick, F.H., and R.M. Hughes. 1998.
Aquatic vertebrates. Pages 161-182 in
J.M. Lazorchak, DJ. Klemm, and D. V.
Peck (eds.). Environmental monitoring and
assessment program surface waters: Field
operations and methods for measuring the
ecological condition of wadeable streams.
Office of Research and Development, U. S.
Environmental Protection Agency, Cincin-
nati, OH. EPA/620/R-94/004F.
Meador, M.R., C.R. Hupp, T. Cuffney, and
M.E. Gurtz. 1993 a. Methods for sampling
fish communities as part of the National
Water Quality AssessmentProgram. U.S.
Geologic Survey Open File Report 93 -104.
Meador, M.R., T. Cuffney, and M.E. Gurtz.
1993b. Methods for characterizing stream
habitat as part of the National Water Quality
Assessment Program. U.S. Geologic
Survey Open File Report 93 -408.
Mills,H.B., W.C. Starrett, andF.C.Bellrose.
1966. Man's effect on the fish and wildlife of
the Illinois River. Illinois Natural History
Survey Biological Notes 57.
Ohio Environmental Protection Agency (OEPA).
1988. Biological criteria for the protect on
of aquatic life: Volumes I-II. OEPA,
Division of Water Quality Monitoring and
Assessment, Surface Water Section,
Columbus, OH.
Ohio Environmental Protection Agency (OEPA).
1989. Biological criteria for the protect on
of aquatic life: Volume III. Standardized
biological field sampling and laboratory
methods for assessing fish and
macroinvertebrate communities. Division of
Water Quality Monitoring and Assessment,
Surface Water Section, Columbus, OH.
Oklahoma Conservation Commission (OCC).
1993. Development of rapid bioassessment
protocols for Oklahoma utilizing character-
istics of the diatom community. Oklahoma
Conservation Commission, Oklahoma City,
OK.
Omernik, J.M. 1987. Ecoregions of the
conterminous United States. Annals of the
Association of American Geographers
77:118-125.
Overton, W.S.,D. White, andD.L. Stevens, Jr.
1991. Design report for EMAP, the
environmental monitoring and assessment
program. U. S. Environmental Protection
-------
Agency, Office of Research and Develop-
ment, Washington, D.C. EPA-600-3-91-
053.
Page,L.M. 1983. Handbook of Darters. TFH
Publication, Inc., Neptune City, NJ.
Palmer, C.M. 1969. A composite rating of algae
toleratingorganicpollution. Journal ofPhycology
5:78-82.
Palmer, C.M. 1977. Algae and water pollution.
U.S. Environmental Protection Agency,
Cincinnati, OH. EPA-600/9-77-036.
Patrick, R. 1963. The structure of diatom
communitiesundervarying ecological condi-
tions. Conference on the problems of
environmental control on the morphology of
fossil and recent protobionta. Annals of the
New YorkAcademy of Sciences 108(2):359-
365.
Patrick, R. 1964. A discussion of natural and
abnormal diatom communities. Pages 185-
204 /'«D.F. Jackson (ed.) Algae and man.
Plenum Press, NY.
Patrick, R. 1977. Ecology of freshwater
diatoms. Pages 284-332 in D. Werner
(ed.). The biology of diatoms. Botanical
monographs Volume 13. University of
California Press, Berkeley, CA.
Patrick, R., M.H. Hohn, and J.H. Wallace.
1954. A new method for determining
the pattern of diatom flora. Notulae
Naturae of the Academy of Natural
Sciences of Philadelphia, No. 259.
Patrick, R., and M.H. Hohn. 1956. The
diatometer - a method for indicating the
conditions of aquatic life. Proceedings
American Petroleum Institute, Section 3,
Refining, Addresses, and Reports,
36(3):332-339.
Patrick, R., and C.W. Reimer. 1966. The
diatoms of the United States, exclusive of
Alaska and Hawaii. Monograph No. 13.
Academy of Natural Sciences of Phila-
delphia, Philadelphia, PA.
Patrick, R., and C.W. Reimer. 1975. The
Diatoms of the United States. Volume 2,
Part 1. Monograph No. 13. Academy of
Natural Sciences of Philadelphia, Phila-
delphia, PA.
Peterson, C.G., andR.J. Stevenson. 1990.
Post-spate development of epilithic algal
communities in different current environ-
ments. Canadian Journal of Botany
68:2092-2102.
Pflieger, W.L. 1975. The fishes of Missouri.
Missouri Department of Conservation,
Jefferson City, MO.
Plafkin, J.L., M.T. Barbour, K.D. Porter, S.K.
Gross, and R.M. Hughes. 1989. Rapid
bioassessment protocols for use in rivers
and streams: Benthicmacroinvertebrates
and fish. U.S. EPA, Office of Water
Regulations and Standards, Washington,
D.C. EPA-440-4-89-001.
Platts, W.S., W.F. Megahan, and G.W.
Minshall. 1983. Methods for evaluating
stream, riparian, andbiotic conditions.
USDA, USFWS, Ogden, UT. General
Techni cal Report INT-138.
Porter, S.G., T.F. Cuffney, M.E. Gurtz, and
M.R. Meador. 1993. Methods for
collecting algal samples as part of the
national water-quality assessment pro-
gram: U.S. Geological Survey Open-File
Report 93-409.
Post, G. 1983. Textbook of fish health. TFH
Publications, Inc., Neptune City, NJ.
-------
Prescott, G.W. 1969. The algae: A review.
HoughtonMfflin Company, Boston, MA.
Rankin, E.T. 1989. The qualitative habitat
evaluation index (QHEI): Rational, methods
and application. Ecological Assessment
Section, Division ofWater Quality Planning
and Assessment, OhioEPA.
Rankin, E.T. 1991. The use of the qualitative
habitat evaluation index for use attainability
studies in streams and rivers in Ohio. In
George Gibson, (ed.). Biological Criteria:
Research and regulation. Office of Water,
U.S. EPA, Washington, D.C. EPA440/5-
91-005.
Rankin, E.T. 1995. Habitat indices in water
resource quality assessments. Pages 181-
208 //I W.S. Davis and T.P. Simon (eds.).
Biological assessment criteria: Tools for
water resource planning and decision
making. Lewis Publishers, Boca Raton, FL.
Resh, V.H., and J.K. Jackson. 1993. Rapid
assessment approaches to biomonitoring
using benthic macroinvertebrates. Pages
195-233 in D.M. Rosenberg and V.H.
Resh (eds.). Freshwater biomonitoring and
benthic macroinvertebrates. Chapman and
Hall, NY.
Robison, E.G., and R.L. Beschta. 1990.
Characteristics of coarse woody debris for
several coastal streams of southeast Alaska,
USA. Canadian Journal of Fisheries and
Aquatic Sciences 47:1684-163-93.
Robison, E.G., and P.R. Kaufmann. 1994.
Evaluating two objective techniques to
define pools in small streams. Pages 659-
668 in R. A. Marston and V. A. Hasfurther
(eds.) Effects of human induced changes on
hydrologic systems. Summer Symposium
Proceedings, American Water Resources
Association. June 26-29, 1994, Jackson
Hole, WY.
Rosen, B.H 1995. Use of periphyton in the
development of biocriteria. Pages 209-215
in W.S. Davis and T.P. Simon (eds.).
Biological assessment and criteria: Tools for
water resource planning and decision
making. Lewis Publishers, Boca Raton, FL.
Roth, N.E., J.C. Chaillou, and M. Gaughan.
1997a. Guide to using 1995 Maryland
biological stream survey data. Maryland
Department of Natural Resources, Chesa-
peake Bay and Watershed Programs,
Monitoring and non-tidal assessment
division, Annapolis, MD. CBWP-
MANTA- EA-97-3.
Roth, N.E., M.T. Southerland, J.C. Chaillou,
J.H. Volstad, S.B. Weisberg, H.T. Wilson,
D.G. Heimbuch, and J.C. Seibel. 1997b.
Maryland biological stream survey: Eco-
logical status of non-tidal streams in six
basins sampled in 1995. Maryland
Department of Natural Resources, Chesa-
peake Bay and Watershed Programs,
Monitoring and Non-tidal Assessment,
Annapolis, MD. CBWP-MANTA-EA-
97-2.
Roth,N.E., M.T. Southerland, S.B. Weisberg,
J.C. Chaillou, P.F. Kazyak, S.A. Stranko,
R.J. Klauda, L.W. Hall, Jr, and R.P.
Morgan, Jr. 1997c. Development of an
index of biotic integrity for fishes in
Maryland streams. Maryland Department
of Natural Resources, Chesapeake Bay
and Watershed Programs, Monitoring and
Non-tidal Assessment, Annapolis, MD.
Rott, E. 1991. Methodological aspects and
perspectives in the use of periphyton for
monitoring and protecting rivers. InE.A.
Whitton, E. Rott, and G. Friedrich (eds.).
-------
Use of algae for monitoring rivers. Institut
fur Botanik, University of Innsbruck,
Austria.
Sabater, S., F. Sabater, and J. Armengol.
1988. Relationships between diatom
assemblages and physicochemical vari-
ables in the River Ter (NE Spain).
Internationale Revue der Gesamten
Hydrobiologie 73:171-179.
Sayler,G.S.,M.Pusiss,andM. Silver. 1979.
Alkaline phosphatase assay for freshwa-
ter sediments: Application to perturbed
sediment systems. Applied and Environ-
mental Microbiology 38:922-927.
Shelton, L.R. 1994. Field guide for collecting
and processing stream-water samples for
the National Water Quality Assessment
Program. USGS Open-File Report 94-
455.
Simon, T.P., and J. Lyons. 1995. Application
of the index of biotic integrity to evaluate
water resource integrity in freshwater
ecosystems. Pages 245-262 in W.S.
Davis and T.P. Simon (eds.). Biological
assessment criteria: Tools for water
resource planning and desision making.
Lewis Publishers, Boca Raton, FL.
Sladecek, V. 1973. System of water quality
from the biological point of view, Arch.
Hydrobiologie Beih. 7, Ergebn. Limnol.
7.
Smith, P.W. 1979. The fishes of Illinois.
University of Illinois Press, UrbanalL.
Southerland,M.T., and J.B, Stribling. 1995.
Status of biological criteria development
and implementation. Pages 81-96 in
W.S. Davis and T.P. Simon (eds.).
Biological assessment criteria: Tools for
water resource planning and desision
making. Lewis Publishers, Boca Raton,
FL.
Stack, B.R. 1989. Factors influencing pool
morphology in Oregon coastal streams.
M.S. Thesis, Oregon State University.
Stevenson, RJ. 1984. Epilithic and epipelic
diatoms in the Sandusky River, with
emphasis on species diversity and water
pollution. Hydrobiologia 114:161-174.
Stevenson, R.J., and R.L. Lowe. 1986.
Sampling and interpretation of algal patterns
for water quality assessments. Pages 118-
149 in E.G. Isom (ed.). Rationale for
sampling and interpretation of ecological
data in the assessment of freshwater
ecosystems. American Society of Testing
and Materials. ASTM STP 894.
Stribling, J.B., B.K. Jessup, and J.S. White.
1998. Development of abenthic index of
biotic integrity for Maryland streams.
Maryland Department of Natural Re-
sources, Chesapeake Bay and Watershed
Programs, Monitoring and Non-tidal
Assessment, Annapolis, MD. CBWP-EA-
98-3.
Trautman, M.B. 1981. The fishes of Ohio. Ohio
State University Press, Columbus, OH.
U. S. Environmental Protection Agency (EPA).
1987. Handbook of methods for acid
deposition studies: Laboratory analysis for
surface waster chemistry. EPA/600/4-87/
026.
U. S. Environmental Protection Agency (EPA).
1998. Environmental monitoring and as-
sessment program (EMAP-SW): Research
plan 1997. U. S. Environmental Protection
Agency, Washington, D.C. EPA/620/R-
98/002.
-------
Wallace, J.B., J.W. Grubaugh, and M.R.
Whiles. 1996. Biotic indices and stream
ecosystem processes: Results from an
experimental study. Ecological Applica-
tions 6:140-151.
Weber, C.I. (ed). 1973. Biological field and
laboratory methods for measuring the
quality of surface waters and effluents. U. S.
Environmental Protection Agency, Office of
Research and Development, Cincinnati,
OH. EPA 670-4-73-001.
Weitzel, R.L. (ed.). 1979. Methods and
measurements of periphyton communities:
A review. American Society for Testing and
Materials, Philadelphia, PA. Special Tech-
nical Publication 690.
Whittaker, R.H. 1952. A study of summer
foliage insect communities in the Great
Smokey Mountains. Ecological Mono-
graphs 22:6.
Whittaker, R.H., and C.W.Fairbanks. 1958.
A study of plankton copepod communi-
ties in the Columbia basin, southeastern
Washington. Ecology 39:46-65.
Whittier, T.R., D.P. Larsen, R.M. Hughes,
C.M. Rohm, A.L. Gallant, and J.M.
Omernik. 1987. The Ohio stream
regionalization proj ect: A compendium of
results. U.S. Environmental Protection
Agency, Environmental Research Labora-
tory, Corvallis, OR. EPA 600-3-87-025.
Wilhm, J.L., and T.C. Dorris. 1968.
Biological parameters for water quality
criteria. Bioscience 477-481.
Yoder, C.O., P. Albeit, and M.A. Smith.
1981. The distribution and abundance of
fishes in the mainstem Scioto River as
affected by pollutant loadings. Ohio EPA
Technical Report 81/3. Columbus, OH.
-------
The habitat assessment index being de-
veloped by USEPA-EMAP-SW currently
contains three distinct indices: 1) the Rapid
Habitat Assessment (RHA) index; 2) the
Physical Habitat Assessment (PHab) index;
and 3) the Streams/Rivers Assessment (SRA)
index. Short descriptions of the individual
assessment metric comprising these indices
are given below (Kaufmann and Robison
1998).
The USEPA-EMAP-SW RHA index is
very similar to both the MDNR-MBSS and
USEPA-RBP indices. The 12 metrics used
in the RHA index are described below. Each
ranking category has a range of possible
scores associated with it (i.e., Optimal 20 to
16, Sub-Optimal 15 to 11, Marginal 10 to 5,
Poor 5 to 0) based on an assessment of the
entire sample segment. A total maximum in-
dex score of 240 is possible. Unlike the QHEI,
no negative metric scores are used and no
habitat-ranking scheme has been produced.
1) Instream Cover (Fish) - Scores are
based on the amount and diversity of useable
fish cover types observed across the entire
sampling segment. The highest scores are
given to areas having more than a 50% mix
of boulders, cobble submerged logs, under-
cut banks, or other stable habitat and judged
to have adequate amount of habitat. The low-
est scores are given to areas with less than
10% of these cover types and that obviously
lack an adequate amount of habitat. Scored 0
to 20.
2) Epifaunal Substrate - Scores are
based on assessing the entire sampling seg-
ment for the presence and size of riffles and
the amount of cobble substrate present. The
highest scores are given to areas that have
well-developed riffles and runs and streams
-------
with an abundance of cobble. The lowest
scores are given to areas in which riffles and
runs are almost non-existent and that lack
cobble substrate. Scored 0 to 20.
3) Velocity/Depth Regimes - Scoring of
this metric is based on the variety and veloc-
ity of velocity/depth regimes found within the
stream sample segment. Streams with the four
velocity regimes, slow-deep, slow-shallow,
fast-deep, and fast-shallow, are scored the
highest and those that are dominated by one
velocity/depth regime (usually slow-deep) are
scored the lowest. Scored 0 to 20.
4) Frequency of Riffles - Scores for this
metric are based on the frequency and occur-
rence of riffles and the variety of habitat found
within the stream sample segment. Streams
with frequent riffles and diverse habitat are
scored the highest. Streams with poor habitat
and low frequency of well-developed riffles
are scored the lowest. Scored 0 to 20.
5) Channel Alteration - Scoring of this
metric is based on the type and amount of
channel alteration and disruption found within
the stream sample segment. Streams with no
channelization or dredging present are scored
the highest and those that are dominated (more
than 80% of the reach) by channelization and
disruption are scored the lowest. Scored 0 to
20.
6) Bank Condition (Bank Erosion) -
Scores for this metric are based on evidence
of bank stability and erosion. Streams with
stable banks and showing little evidence of
erosion or bank failure are scored the high-
est. Streams that have unstable banks, banks
with many eroded areas, and banks showing
60 to 100% evidence of erosional scarring are
scored the lowest. Scored 0 to 20.
7) Embeddedness - Scoring for this met-
ric is based on the percentage of stream gravel,
cobble, and boulder particle surface area that
is surrounded by fine sediment or flocculent
materials. High scores are given for areas with
low embeddedness (0 to 25% surrounded) and
low scores are given to areas with high
embeddedness (more than 75% surrounded).
Scored 0 to 20.
8) Channel Flow Status - Scores for this
metric are based on the degree to which wa-
ter fills the channel and the amount of exposed
substrate that occurs within the channel.
Streams in which the water reaches the base
of both banks and a very small proportion of
the channel substrate is exposed are scored
the highest. Streams that have little water in
the channel, most of which is in standing
pools, are scored the lowest. Scored 0 to 20.
9) Riparian Vegetation Zone Width (Least
Buffered Side) - Scores for this metric are based
on the width of the riparian zone and the pres-
ence or absence of human disturbances.
Streams with a riparian zone width of more
than 18m and no evidence of impacts from
human activities are scored the highest.
Streams with a riparian zone width of less
than 6 m and evidence of human activities
are scored the lowest. Scored 0 to 20.
10) Sediment Deposition - Scores for this
metric are based on the degree of bar devel-
opment and the extent that the stream chan-
nel is affected by sedimentation within the
stream sample segment. Streams with little or
no bar enlargement and those where less than
5% of the stream bottom is affected by sedi-
ment deposition are scored the highest.
Streams with heavy deposits of fine sediment,
increased bar development, and more than
50% of the bottom changing frequently due
to sedimentation are scored the lowest. Scored
0 to 20.
11) Bank Vegetative Protection - Scores
for this metric are based on the percentage of
-------
the stream bank surfaces that are covered by
vegetation. Streams that have more than 90%
of their bank surfaces covered by vegetation
are scored the highest. Streams that have less
than 50% of their bank surfaces covered by
vegetation are scored the lowest. Scored 0 to
20.
12) Grazing or Other Disruptive Pres-
sure - Scores for this metric are based on the
degree of vegetative disruption by mowing
or grazing on the banks of the stream. Stream
banks that are minimally disturbed are scored
the highest. Streams with banks that have very
disturbed vegetation (vegetation removed to
an average of < 2") are scored the lowest.
Scored 0 to 20.
The PHab has four primary metrics, each
of which is made up of a varying number of
sub-metrics. Many of these sub-metrics are
based on direct numerical measurements made
in the field and are therefore quantitative rather
than qualitative. Some of the PHab metrics
are based on ranked categories of field mea-
surements. The goal of the PHab sampling
design is to assess habitat and other stream
conditions over the sampling reach. No over-
all composite score is produced by this index.
la-g) Thahv eg Profile - The thalweg pro-
file is a longitudinal survey of the sub-metrics:
Maximum Depth, Wetted Width, Bar Width,
Soft/Small Sediment Presence, Channel or
Pool Type, Pool Forming Element, and Side
Channel Presence. The thalweg measure-
ments (except wetted width) are generally
taken at 100 to 150 equally spaced points (10
to 15 intervals between each of 11 channel
cross-section sampling stations) along the
centerline of the stream between the two ends
of the sample reach. Thalweg wetted width
is measured at 21 equally spaced intervals (at
each of 11 channel cross-section sampling
stations and a station mid-way between cross-
section sampling stations). Spacing of the thal-
weg measurements is based on the channel
width. The samples are taken at 1 m, 1.5 m or
0.01 times reach length, for channel widths
of less than 2.5 m, 2.5 to 3.5 m, and more
than 3.5 m, respectively. Sampling is designed
to resolve deep areas and habitat units that
range from 1/3 to /^the channel width. Sam-
pling proceeds upstream along the middle of
the channel. Data from the thalweg profile is
intended to allow the calculation of indices
of residual pool volume, stream size, channel
complexity, and the relative proportions of
habitat types such as riffles and pools.
la) Thalweg Profile, Maximum Depth -
The greatest depth in the channel is measured
to the nearest cm, at each of the 100 incre-
ments of length upstream along the mid-chan-
nel line. The thalweg maximum depth is not
necessarily the mid-channel line.
Ib) Thahv eg Profile, WettedWidth - The
thalweg wetted width is the width between
the left and right wetted boundaries (the point
at which substrate particles are no longer sur-
rounded by free water). It is measured across
and over bars. Widths are measured to the
nearest 0.1 m for widths up to 3 m and to the
nearest 5% of the width if the width is greater
than 3 m. They are usually only measured at
21 sample stations. However, if a higher reso-
lution is needed, thalweg wetted widths can
be taken at all 100 to 150 sample stations.
Ic) Thalweg Profile, Bar Width - Bars
are defined by PHab as channel features be-
-------
low the bankfull mark that are dry during
baseflow conditions. Islands are features that
are dry even during bankfull conditions. If a
mid-channel feature is as high as the surround-
ing flood plain, it is treated as an island. When
present, bar widths are determined at each
thalweg.
Id) Thahv eg Profile, Soft/Small Sediment
Presence - When the rod or staff is used to
make the thalweg depth measurement, it is
also used to determine the presence or absence
of small, loose, soft sediments at each of the
thalweg sampling stations. Small/soft sedi-
ments are defined by PHab as fine gravel,
sand, silt, clay, or muck.
le) Thalweg Profile, Channel or Pool
Type - A channel unit scale habitat classifica-
tion is used to visually determine and classify
channel or pool features into one of 12 pos-
sible categories at each of the thalweg sam-
pling stations. These categories include: glide,
riffle, rapid, cascade, falls, dry channel, or
one of five pool types. The feature should be
at least as long as the channel is wide if it is to
be included.
If) Thalweg Profile, Pool Forming Ele-
ment - When present, pools are classified us-
ing seven categories, based on the element
from which the pool is formed (e.g., boulder,
large woody debris, etc.).
Ig) Thahv eg Profile, Side Channel Pres-
ence - The presence of side channels is noted
at each of the thalweg sampling stations.
Notes about their point of convergence and
divergence with the main channel are taken.
2) Woody Debris - The large woody de-
bris (LWD) measurement used by PHab is a
simplified version of Robison and Beschta's
(1990) method. It provides quantitative esti-
mates of the number, size, total volume and
distribution of wood in the stream reach.
LWD is defined by PHab as woody material
with a small end diameter of at least 10 cm
and a length of at least 1.5 m. All pieces of
LWD in (partially or fully) or spanning the
active channel (flood channel up to bankfull)
are tallied for the area between each sampling
cross section. The tallies are assigned to sepa-
rate categories based on: 1) location in the
channel (above or in), 2) length (1.5 to 5 m, 5
to 15 m, or more than 15m) and 3) large end
diameter (more than 0.8 m, 0.8 to 0.6 m, 0.6
to 0.3 m, 0.3 to less than 0.1 m). When length
is evaluated, only the part with a diameter
more than 10 cm is included. Each piece of
LWD is counted as one tally entry and the
whole piece is included even if part of it is
outside the bankfull channel. The LWD is
assigned to the sampling cross section con-
taining the large end.
3a-c) Channel and Riparian Cross-Sec-
tions - Three primary classes of measurements
are performed at the 11 channel cross section
stations: 1) quantitative measurements of
channel cross-section dimensions, bank char-
acteristics and stream channel gradient, sinu-
osity, and riparian cover; 2) visual estimates
of substrate size class and embeddedness, ar-
eal cover class and type of riparian vegeta-
tion in canopy, mid-layer and ground cover,
areal cover class offish concealment features,
aquatic macrophytes, and filamentous algae;
and 3) recorded observations of human dis-
turbances and their proximity to the channel.
3 a) Channel and Riparian Cross-Sec-
tions, Quantitative Measurements - The cross-
sectional dimensions, bankfull width, wetted
width and bar width are measured as described
above for the thalweg profile stations. The
channel bankfull height is estimated as the
height of the bankfull flow above the water
level. The channel incised height is estimated
-------
as the height from the water surface to the
first terrace of the flood plain (the area at or
above the bankfull height). The slope or gra-
dient, determined using a clinometer, and the
bearing, determined using a compass, are
measured between the cross section stations.
Supplemental measurements are taken in situ-
ations where the direct line of sight between
stations is obscured. Estimates of residual pool
depth and volumes may be made possible, by
applying methods described by Stack (1989)
and Robison and Kaufmann (1994), to the
slope and the thalweg depth and width mea-
surements. Channel sinuosity can be com-
puted using the bearing and distance measure-
ments. Riparian canopy cover over the stream
is quantified using a Convex Spherical
Densiometer (Lemmon 1957). Four readings
(one in each direction while standing in the
center of the stream) are taken at each of the
11 cross section stations. Two bank side read-
ings (one on each bank) are also taken at each
site. These measurements are made with the
observer's back to the stream.
3b) Channel and Riparian Cross-Sec-
tions., Visual Estimates - Substrate size class
and embeddedness are evaluated at five
equally spaced points centered between the
wetted channel width boundaries, at each of
the 11 channel cross section stations. Water
depth and distance from the left bank is also
determined at each sampling point. The sub-
strate at each point is visually inspected and
classified into one of 11 categories based on
size or origin. For particles larger than sand,
the average embeddedness in a 10 cm circle
is estimated. Observations are made to esti-
mate areal cover class and type of riparian
vegetation in canopy (more than 5 m high),
mid-layer or understory (0.5 to 5 m high), and
ground cover (less than 5 m high). A portion
of the riparian zone from the shoreline to a
distance of 10 m on either side of the bank
and 5 m up and down stream (10 m X 10m
area on each bank) is assessed at each of the
11 channel cross-section stations. For each 10
m X 10 m area, and for the canopy and un-
derstory cover categories, the percent total
cover (expressed as one of four possible cat-
egories: 1 = Sparse, <1%; 2 = moderate, 10
to 40%; 3 = heavy, 40 to 75%, or 4 = very
heavy, >75%) comprised by each of five
broad vegetation types is noted. The percent
total cover is also estimated for each bank area,
using the same classification for big and small
trees in the canopy, woody and non-woody
vegetation in the understory; and woody, non-
woody, and barren categories in the ground
cover layer. Using the classification scheme
outlined above, the percent total areal cover
of seven kinds offish concealment features
(e.g., aquatic macrophytes, filamentous algae,
woody debris, etc) is estimated for the area 5
m up and down stream at each of the 11 chan-
nel cross section stations.
3c) Channel and Riparian Cross-Sec-
tions, Recorded Observations - The presence
and proximity of 11 categories of human in-
fluence in the riparian and stream areas 5 m
up and down stream at each of the 11 chan-
nel cross section stations, is noted.
4) Discharge - Discharge is measured at
one location in each sample segment by one
of four methods: 1) velocity-area (Linsley et
al. 1982), 2) portable weir, 3) calibrated
bucket, or 4) time of movement of a neutrally
buoyant object. The velocity-area method is
preferred in streams large enough to use a
water velocity meter. Using this approach, the
water velocity at a depth of 0.6 of the total
depth, at each of 15 to 20 points, equally
spaced across the stream width, is measured.
In smaller streams one of the other methods
may need to be used. Discharge is measured
at the point, where the water chemistry
samples are taken.
-------
The USEPA-EMAP-SW SRA index is
based on approximately 5 metrics (or com-
ponents), depending on how they are
grouped. Two of the components, General
Assessment and Local Anecdotal Information
are written descriptions. The remaining SRA
metrics are based on ranked categories of field
measurements and classified lists of field ob-
servations. No scores are assigned to any of
the metrics. Like the measurements for the
PHab, it is unclear how these measurements
will be utilized in an analysis scheme and no
overall index score for the SRA is available.
1) Watershed Activities and Distur-
bances - Watershed activities are broken into
five major types: residential, recreational,
agricultural, industrial, and stream manage-
ment. Listed under each of these activity cat-
egories is are examples of typical disturbances
associated with each activity. The presence
or absence of each disturbance is noted and
the intensity of each disturbance ranked into
one of three categories, low, moderate, or
high.
2a-c) Reach Characteristics - Three
major categories: vegetation cover type, land
use, and water clarity are used to describe and
classify the character of the stream sampling
reach.
2a) Reach Characteristics, Vegetation
Cover - The vegetative cover observed at the
sample reach is noted and classified into one
of five possible categories '.forest, shrub, wet-
land, bare ground, or macrophytes. During
this process, each vegetation cover type is
ranked, based on the percent of the reach it
comprises (i.e., rare <5%, sparse 5 to 25%,
moderate 25 to 75%, and extensive >75%).
2b) Reach Characteristics, Land Use/
Type - The land use/type observed at the
sample reach is noted and classified into one
of four possible categories: agriculture row
crop, agriculture grazing, logging, or devel-
opment. During this process, land use/type is
ranked, based on the percent of the reach it
comprises (i.e., rare <5%, sparse 5 to 25%,
moderate 25 to 75%, and extensive >75%).
2c) Reach Characteristics, Water Clar-
ity - The type of water clarity observed at the
site is ranked into one of four categories: clear,
murky, highly turbid, or storm influenced.
3a-b) Waterbody Character - Two cat-
egories, disturbance impact and aesthetic
quality, are used to assess the waterbody char-
acter at each sample reach.
3 a) Waterbody Character, Disturbance
Impact - The waterbody character at each
sample reach is assessed for the degree of dis-
turbance impact observed. This metric is
ranked from 1 (highly disturbed) to 5 (pris-
tine).
3b) Waterbody Character, Aesthetic
Quality - The waterbody character at each
sample reach is assessed for it's aesthetic qual-
ity. This metric is ranked from 1 (unappeal-
ing) to 5 (appealing).
4) General Assessment - A general as-
sessment is conducted for stream reach by
taking notes on the wildlife, vegetation diver-
sity, and forest age class (0 to 25, 25 to 75,
>75 yrs) observed at the site.
5) Local Anecdotal Information - Local
anecdotal information for the study reach is
described.
The USEPA-RBP index is very similar
to both MDNR-MBSS and USEPA-EMAP-
-------
SW RHA indices. A short description of each
of the 13 metrics that comprise the USEPA-
RBP habitat assessment index are listed be-
low (Barbour et al. 1999). Three of the
metrics, embeddedness, frequency of riffles.,
and velocity/depth combinations., are only
used at high gradient sites, and three of the
metrics, pool substrate, pool variability, and
channel sinuosity, are only used at low gradi-
ent sites. As a result, only ten metrics total are
used at any one site. Each ranking category
has a range of possible scores associated with
it (i.e., Optimal 20 to 16, Sub-Optimal 15 to
11, Marginal 10 to 5, Poor 5 to 0) based on
an assessment of the entire sample segment.
All of the metrics have a maximum score of
20 points. The metrics bank stability, bank
vegetation protection, and riparian vegetation
zone width, have maximum scores of 10
points for each bank (maximum 20 points to-
tal). A total maximum index score of 200
points is possible.
1) Epifaunal Substrate and Available
Cover - Used to assess the relative quality of
natural structures in the stream as sites for use
as refugia, feeding, and reproduction. Scores
are based on the amount and diversity of sub-
strate for epifaunal colonization and fish cover
observed across the entire sampling segment.
The highest scores are given to areas having
more than a 70% (in high gradient streams)
or more than 50% (in low gradient streams)
mix of favorable, stable, substrates and cover
types such as submerged logs/snags, under-
cut banks, cobble, or other stable habitat and
at a stage to allow full colonization. The low-
est scores are given to areas with less than
20% (in high gradient streams) or less than
10% (in low gradient streams) of these cover
types and that obviously lack an adequate or
stable habitat. Scored 0 to 20.
2) Velocity/Depth Combinations (High
Gradient) - This metric is only used for high-
gradient streams. Scoring of this metric is
based on the variety of velocity of velocity/
depth regimes found within the stream sample
segment. Streams with the four velocity re-
gimes, slow-deep, s\ow-shallow, fast-deep,
and fast-shallow, are scored the highest and
those that are dominated by one velocity/
depth regime (usually slow-deep) are scored
the lowest. Scored 0 to 20.
3) Pool Substrate Characterization
(Low Gradient) - This metric is only used for
low-gradient streams. It is used to assess the
type and condition of substrates found in
pools. Scoring for this metric is based on the
presence of particular substrate types, root
mats, and submerged aquatic vegetation
(SAV). Generally, an area with diverse sub-
strates support a more diverse array of organ-
isms as compared to areas with uniform sub-
strates. Scores are high for areas exhibiting
the presence of mixed substrates, gravel and
firm sand, root mats, and SAV. Scores are low
for areas with hard-pan clay or bedrock and
no SAV. Scored 0 to 20.
4) Pool Variability (Low Gradient) - This
metric is only used for low-gradient streams.
It rates the overall mixture of pool types found
in streams by size and depth. Scoring of this
metric is based on the variety of basic pool
types found within the stream sample seg-
ment. Streams that have all four pool types,
large-deep, large-shallow, small-deep, and
small-shallow, are scored the highest and
those that are dominated by one pool type
(usually small-shallow) or that lack pools, are
scored the lowest. Scored 0 to 20.
5) Frequency of Riffles or Bends (High
gradient) - This metric is only used for high-
gradient streams. Scores for this metric are
based on the frequency or occurrence of riffles
and the variety of habitat found within the
-------
stream sample segment. Streams with frequent
riffles and diverse habitat are scored the high-
est. Streams with poor habitat and a low fre-
quency of well-developed riffles are scored
the lowest. Scored 0 to 20.
6) Channel Alteration - Is used to assess
the impact of large scale changes on the shape
of the stream channel. Scoring of this metric
is based on the type and amount of channel
alteration and disruption found within the
stream sample segment. Streams with no
channelization or dredging present are scored
the highest and those that are dominated
(>80% of the reach) by channelization and
disruption are scored the lowest. Scored 0 to
20.
7) Bank Stability (Condition of Banks) -
Scores for this metric are based on evidence
of bank stability and erosion. Eroded banks
indicate a problem of sediment movement and
deposition, and suggest a scarcity of cover and
increased organic input to streams. Streams
with stable banks and showing little evidence
of erosion or bank failure (<5% affected) are
scored the highest. Streams that have unstable
banks, banks with many eroded areas, and
banks showing 60 to 100% evidence of ero-
sional scarring, are scored the lowest. Scored
0 to 10 for each bank, 0 to 20 total.
8) Embeddedness (High Gradient) -
This metric is only used for high-gradient
streams. It is used to assess the extent to which
stream substrates are buried by silt, sand or
mud. Scoring for this metric is based on the
percentage of stream gravel, cobble, and boul-
der particle surface area that is surrounded by
fine sediment. Scores are high for areas of low
embeddedness (0 to 25% surrounded) and
low for areas with high embeddedness (>75).
Scored 0 to 20.
9) Channel Flow Status - Scores for this
metric are based on the degree to which wa-
ter fills the channel and the amount of exposed
substrate that occurs within the channel.
Streams in which the water reaches the base
of both banks and a very small proportion of
the channel substrate is exposed are scored
the highest. Streams that have little water in
the channel, most of which are standing pools,
are scored the lowest. Scored 0 to 20.
10) Riparian Vegetation Zone Width
(Least Buffered Side) - Scores for this metric
are based on the width of the riparian zone
and the presence or absence of human distur-
bances. Streams with a riparian zone width
of more thanlS m and no evidence impacts
from human activities are scored the highest.
Streams with a riparian zone width of less
than 6 m and evidence of human activities
are scored the lowest. Scored 0 to 10 for each
bank, 0 to 20 total.
11) Sediment Deposition - Is used to as-
sess the impact of sedimentation on the stream
bottom and pools. Scores for this metric are
based on the degree of bar development and
the extent that the stream channel is affected
by sedimentation within the stream sample
segment. Streams with little or no bar enlarge-
ment and those where less than 5% (for high-
gradient streams) or less than 20% (for low-
gradient streams) of the stream bottom is af-
fected by sediment deposition are scored the
highest. Streams with heavy deposits of fine
sediment, increased bar development, and
more than 50% (for high-gradient streams) or
more than 80% (for low gradient streams) of
the stream bottom changing frequently due
to sedimentation, are scored the lowest.
Scored 0 to 20.
12) Bank Vegetative Protection - This
metric supplies information on the ability of
the bank to resist erosion as well as some ad-
ditional information on the potential for nu-
-------
trient uptake by plants, the control of instream
scouring, and stream shading. Scores for this
metric are based on the percentage of the
stream bank surfaces that are covered by veg-
etation. Streams that have more than 90% of
the bank surfaces covered by vegetation, par-
ticularly native vegetation, with little evidence
of grazing or mowing are scored the highest.
Streams that have less than 50% of their bank
surfaces covered by vegetation, disruption of
streamside vegetation is very high, and veg-
etation has been removed to an average height
of less than 5 cm, are scored the lowest.
Scored 0 to 10 for each bank, 0 to 20 total.
13) Channel Sinuosity (Low Gradient)
- This metric is only used for low-gradient
streams. Scores for this metric are based on
degree of meandering or sinuosity that occurs
over the channel length. It is used for streams
in which distinct riffles are uncommon.
Streams in which the bends in the channel
increases its length by three to four times are
scored the highest. Streams with straight
channels are scored the lowest. Channel braid-
ing is considered normal in coastal plains and
low-lying areas so this parameter is not easily
ranked in these areas. Scored 0 to 20.
Listed below is a short description of
each of the seven metrics that comprise the
QHEI (Rankin, 1989). Six of the metrics are
based on two or four scored sub-metrics. Each
sub-metric is further divided into scored cat-
egories which are matched with field obser-
vations to produce the scores. The Gradient
metric is the only metric that does not contain
a sub-metric. To compute a final overall score
for the QHEI, the scores of the sub-metrics
are summed and then the scores of the com-
posite metrics are summed. The maximum
score for the composite metrics range from 8
to 20. The maximum total score of the QHEI
index is 100.
la-b) Substrate (Type and Quality) -
Scores are based on evaluation of two
submetrics, substrate type and substrate qual-
ity. The submetric substrate type includes
identification and diversity of the substrate
types present. The submetric substrate qual-
ity includes determining the origin of the
benthic material (parent material), the extent
of silt cover, and embeddedness at the sample
site. Scored a maximum of 20.
la) Substrate, Type - The type of sub-
strate observed in the sample segment is se-
lected from a list often scored categories. The
scores range from 0 for artificial substrate to
10 for boulder/slabs. The two most common
substrates at the sample site are identified from
the list. A single category is selected twice if
it predominates (more than 75-80% of the
bottom area or clearly is the most function-
ally dominant type). The total number of sub-
strate types (more than four = 2 points, or four
or fewer = 0 points) is used to evaluate sub-
strate diversity. Substrate types must comprise
more than 5% of the sampling area to be in-
cluded. Any substrate types observed but not
included in the scored categories are recorded.
Scored 0 to 21.
Ib) Substrate, Quality - The type of par-
ent material observed in the sample segment
is selected from a list of seven scored catego-
ries. The scores range from -2 for coal fines
to 1 for limestone or tills. All of the categories
of parent materials observed at the sample site
are identified from the list. The extent of silt
cover observed at the sample segment is
evaluated using four scored categories that
-------
range from silt heavy, (nearly all of the stream
bottom covered with a deep layer of silt; -2
points) to silt free; (substrates exceptionally
clean; 1 point). Silt cover is defined as a sub-
strate being covered by more than one inch
of silt. The extent of embeddedness observed
at the sample segment is evaluated using four
scored categories that range from extensive,
more than 75% of the sample area (-2 points)
to none (1 point). Substrates are considered
embedded if more than 50% of the surface of
the substrate is embedded in fine material and
the substrate cannot be easily dislodged. Natu-
rally sandy streams are not included, but
streams embedded by sand as a result of hu-
man activities are included. Scored -5 to 3.
2a-b) Instream Cover (Type and
Amount) - Scores are based on evaluation of
two submetrics, cover type and cover amount.
Scoring the submetric instream cover type
entails identifying the cover types present.
Scoring the submetric instream cover amount
entails estimating the amount or extent of the
useable cover at the sample site. (Limited to a
maximum 20 points)
2a) Instream Cover, Type - All the cover
types observed in the sample segment are se-
lected from a list of nine scored categories.
All of the categories are scored 1 point each
except the deep pool category, which is scored
2 points. Cover types must comprise more
than 5% of the sampling area to be included.
Cover types in areas of the stream with insuf-
ficient depth (usually <25 cm) to make them
useful are not scored. The undercut banks and
rootwad categories are not selected unless
undercut banks occur without rootwads are a
major category. Scored 0 to 10.
2b) Instream Cover, Amount - The ex-
tent of the instream cover at the sample seg-
ment is estimated using four scored catego-
ries that range from extensive (more than 75%
of the sample area, 11 points) to nearly ab-
sent (less than 5% of the sample area or when
no large patch of cover exists any where in
the sampling area, 1 point). If the estimated
amount of cover falls between two catego-
ries, then both categories are chosen and the
scores averaged. Scored 1 to 11.
3a-d) Channel Morphology - Scores are
based on the evaluation of four submetrics, chan-
nel sinuosity, development, channelization, and
stability. These submetrics were chosen to em-
phasize facets of the stream channel that are
related to the creation and stability of stream
habitat. Scoring channel sinuosity entails es-
timating the degree to which the channel me-
anders. Scoring channel development entails
evaluating the presence and quality of riffle/
pool habitat at the sample site. Scoring chan-
nel channelization entails evaluating the pres-
ence and status of man-made channel modi-
fications at the sample site. Scoring channel
stability entails estimating the degree channel
bank stability. Scored a maximum of 20
points.
3 a) Channel Morphology, Sinuosity -
The degree of the channel sinuosity of the
sample segment is estimated using four scored
categories. Scoring of the categories is based
on the number of outside bends, how well
these bends are defined, and the development
of deep outside areas and shallow inside ar-
eas. Scores for this submetric range from 4
points for two or three well-defined outside
bends with deep outside areas and shallow
inside areas, to 1 point for a straight channel.
Scored 1 to 4.
3b) Channel Morphology, Development
- The presence and quality of riffle/pool habi-
tat at the sample site is evaluated using four
categories, ranging in score from excellent (7
-------
points) to poor (1 point), based on the defini-
tion and development of quality riffle/pool
habitat. Higher scores are associated with ar-
eas that have distinct examples of deep pools
that vary in depth, deep riffles and runs, and
riffles with larger substrate (gravel, rubble or
boulders). Lower scores are given to areas that
are predominantly glides; that lack riffles, ar-
eas that have shallow riffles and pools, and
that have riffles with sand and fine gravel sub-
strates. Scored 1 to 7.
3c) Channel Morphology, Channelization
- Evaluation of the presence and status of man-
made channel modifications at the sample site
is based on the presence and recovery status of
man-made channel modifications. Sites are
classified into four possible categories: none
(6 points), recovered (4 points), recovering
(3 points), or recent/no recovery (1 point). The
specific modification is also classified into one
of nine un-scored categories. Scored 1 to 6.
3d) Channel Morphology, Stability - The
degree channel bank stability is classified into
one of three categories, high (3 points), me-
dium (2 points) or low (1 point), based on the
quantity of bedload; signs bank erosion or
effects of wide water level fluctuations; or the
presence of false banks. Artificially stable
(e.g., concrete) stream channels receive a high
score, even though they generally have a nega-
tive impact on fish for reasons other than sta-
bility. More stable channels tend to have stable
riffles and pools, little bedload, and banks with
little or no erosion. Scored 1 to 3.
4a-c) Riparian Zone - Scores are based
on evaluation of three submetrics, (riparian
zone width, quality and bank erosion). These
submetrics were chosen to emphasize the
quality of the riparian zone buffer and the flood
plain vegetation. Scoring for all three
submetrics is accomplished by scoring both
banks of the stream and then averaging the
scores to get an overall score for the each sub-
metric. For each sub-metric, only one category
(for each bank) should be selected unless con-
ditions are considered intermediate between
two categories. In these instances the two cat-
egories are identified and the scores averaged.
Scoring riparian zone width entails estimat-
ing the width of the stream side vegetation.
Scoring riparian zone quality entails identi-
fying the predominant type of floodplain land
use or habitat along the banks of the site. Scor-
ing riparian zone bank erosion entails evalu-
ating the degree of bank alteration at the site.
Scored a maximum of 10 points.
4a) Riparian Zone, Width - This sub-
metric is defined as the width of the riparian
vegetation. Width estimates are only made for
forest, shrub, swamp and old field vegetation.
Weedy urban and industrial lots are not in-
cluded. Estimates are classified into five
scored categories: wide (more than 50 m, 4
points), moderate (10-50 m, 3 points), nar-
row (5-10 m, 2 points), very narrow (5-10 m,
2 points), and none (0 points). Scores for both
the left and right banks are averaged. Scored
Oto4.
4b) Riparian Zone, Quality - The pre-
dominant type of land use or habitat observed
along each bank of the site floodplain is se-
lected is assigned to one of eight scored cat-
egories. The floodplain is the either the area
immediately outside the riparian zone or
greater than 100 ft from the stream (which-
ever is wider). Scores associated with the cat-
egories range from 0 points for open pasture/
row crop, urban/industrial, and mining/con-
struction, to 3 points for forest/swamp. The
score for both banks are averaged to provide
an overall estimate of riparian zone quality
for the site. Scored 0 to 3.
4c) Riparian Zone, Bank Erosion - Ri-
parian zone bank erosion is assessed using
-------
the Stream Bank Soil Alteration Ratings from
Platts et al. (1983). Bank erosion is classified
into one of three scored categories, none/little
(3 points), moderate (2 points), or heavy/se-
vere (1 point). The ranking categories are
based on the percentage of the stream bank
that is unstable, eroding, broken down or false
(Platts et al. 1983). Both the left and right
banks are scored and the scores averaged.
Scored 1 to 3.
5a-c) Pool/Glide Quality - Scores are
based on evaluation of three submetrics, maxi-
mum depth, current type, and morphology.
These submetrics were chosen because they
are related to the quality of pool/glide habi-
tats. Scoring maximum depth entails estimat-
ing the maximum depth of the pool. Scoring
current type entails evaluating the types and
diversity of water current velocities found at
the site. Scoring morphology entails assess-
ing the ratio of pool width to riffle width ob-
served at the sample site. Scored a maximum
of 12 points.
5 a) Pool/Glide Quality: Maximum Depth
- The observed pool habitats are classified by
maximum depth into five scored categories
(>1 m, 6 points; 0.7-1 m, 4 points; 0.4-0.7 m,
2 points; <0.4 m, 1 points; and <0.2 m, 0
points). Pools and glides with maximum
depths less than 20 cm are considered to have
lost their function. Scored 0 to 6.
5b) Pool/Glide Quality: Current Type -
Based on observed water flow patterns and
other characteristics such as waves and water
borne objects, the Pool/glide current types
present at the site are classified into seven
scored categories (Fast, Moderate, Slow and
Eddies all are scored 1 point; Torrential and
Interstitial, -1 point; and Intermittent, -2
points). All of the categories observed at a site
are scored and then summed to provide an
overall sub-metric score. Scored -2 to 4.
5c) Pool/Glide Quality: Morphology -
Based on the ratio of pool width to riffle width
observed at the sample site, the pool/glide
morphology is classified into one of three
scored categories: Wide, pool width>riffle
width (2 points); Equal, pool width=riffle
width (1 point); and Narrow, pool width
-------
unstable (gravel or sand, 0 points). Scored
Oto2.
6c) Riffle/Run Quality, Embeddedness -
The extent of embeddedness of the sample
segment is evaluated using four scored cat-
egories that range from extensive (more than
75% of the sample area, -1 points) to none (2
points). Substrates are considered embedded
if more than 50% of the surface of the sub-
strate is embedded in fine material and the
substrate can not be easily dislodged. Scored
-Ito2.
7) Gradient - Scores are assigned to the
sites based on the local stream gradient cal-
culated using a 7.5 topographic map. The gra-
dient is calculated by measuring the stream
length between first contour lines up and
down stream of the sample site and dividing
the distance by the contour interval. If the
contour lines are too close together, a mini-
mum distance of one mile should be used.
Judgement may need to be exercised in areas
containing features such as waterfalls and
impoundments. Scores increase as the gradi-
ent increases to a maximum of 10 points for a
gradient of 9.9 to 13.1 feet per mile, after which
the scores decline with increasing gradient.
The lowest score is assigned sites that have
gradients in excess of 65.6 ft per mile (2
points). Scored a maximum of 10 points.
Miscellaneous Measurements Made -
Other measurements made in the course of
completing an Ohio EPA QHEI include: 1)
classification of channel morphology/modi-
fications; 2) percent composition of pool, riffle
and run features in the stream reach; 3) the
gear distance, water clarity and water stage,
during each of three electroshocking passes;
4) an aesthetic rating of the stream reach; 5)
the percentage of canopy opening above the
stream reach; 6) a ranking of the stream gra-
dient (high, low, or moderate); 7) quantita-
tive measurements of stream reach average
width and average and maximum depth; 8)
quantitative measurements of pool/glide/riffle/
run length, width and depth; and 9) notes on
the representativeness of the reach with re-
gard to the stream and pollution impacts over-
all. These measurements/observations are not
scored or used in the final QHEI scoring.
Listed below is a short description of
each of the 13 metrics that comprise the
MDNR-MBSS QHA index (Roth et al.
1997b). Only 9 of the 13 metrics are scored.
Each scored metric has a maximum score of
20 points. The index is still under develop-
ment and no total index score been devised.
1) Instream Habitat - Scoring of this
metric is based on the perceived value of the
habitat to the fish community. Sites that dis-
play a variety of habitat types, particle sizes,
and hypsographic complexity are assigned
higher scores only where flows are sufficient
for fish to utilize these habitats. Sites lacking
these qualities are assigned low scores. The
presence of ferric hydroxide does not cause a
lower score unless precipitates have changed
the gross physical nature of the substrate. Zero
scores are assigned to segments where none
of the habitat is usable by fish. Scored 0 to
20.
-------
2) Epifaunal Substrate - The rating of
this metric is based on the amount and vari-
ety of hard, stable substrates available for use
by benthic invertebrates. The presence of fea-
tures that inhibit colonization such as floccu-
lent materials, fine sediments, and unstable
substrates will reduce the scores assigned to
segments. Scored 0 to 20.
3) Velocity/Depth Diversity - Scoring of
this metric is based on the variety of velocity/
depth regimes found within the stream seg-
ment. Low gradient streams are usually scored
lower. Scored 0 to 20.
4) Pool/Glide/Eddy/Quality - Scoring of
this metric is based on the variety and spatial
complexity of slow or still water habitat within
the sample segment. These habitats may in-
clude larger eddies in high gradient streams.
Higher scores are assigned to segments that
provide cover for fish (e.g., undercut banks
or woody debris). Scored 0 to 20.
5) Riffle/Run Quality - Scores for this
metric are based on the complexity and func-
tional importance of riffle/run habitat. Higher
scores are assigned to segments dominated by
deep riffle/run areas, stable substrates and a
variety of current velocities. Scored 0 to 20.
6) Channel Alteration - Scores for this
metric are based on the degree and type of
alteration of the stream channel. Some of the
types alterations included are: concrete chan-
nels, artificial embankments, obvious straight-
ening of the natural channel, rip-rap, or re-
cent bar development. The type, placement
and extent of bar development is used as an
indicator of the degree of flow fluctuation and
substrate stability. Greater bar development
or a higher percentage of artificial armoring
(e.g., rip-rap or concrete) of the steam bank
results in lower scoring. Scored 0 to 20.
7) Bank Stability - Scoring of this metric
is based on the presence of riparian vegeta-
tion or other bank stabilizing material. The
scoring is explicitly based on a ranking of the
bank stability, the degree of erosional scar-
ring, the potential for erosion caused by flood
conditions and the degree of bank sloping.
The presence of steep slopes alone, does not
result in the segment being scored low. Scored
0 to 20.
8) Embeddedness - Scoring for this met-
ric is the percentage of stream gravel, cobble,
and boulder particle surface area that is sur-
rounded by fine sediment or flocculent mate-
rials.
9) Channel Flow Status - Scoring for this
metric is the percentage of stream channel,
minus exposed substrates and landforms, that
has water.
10) Riparian Buffer - Scored as the mini-
mum width of vegetated buffer (50 m maxi-
mum). Cultivated fields containing any bare
soil are not considered riparian buffers. For
segments which have variable buffer widths
or receive direct delivery of storm runoff or
sediments, the narrowest buffer in the segment
is scored (e.g., 0 m if parking-lot runoff en-
ters the stream directly), even though a por-
tion of the segment may have a well devel-
oped buffer. If the riparian zone on one side
slopes away from the stream and there is no
direct runoff delivery point, the score should
be based on the opposite bank. The dominant
buffer zone is classified into one of five cat-
egones,forest, old field, emergent vegetation,
mowed lawn, tall grass, or logged are a, and
the dominant adjacent land cover into one of
10 categories, bare soil, railroad, paved road,
parking-lot/industrial/commercial, gravel
road, dirt road, pasture, orchard, cropland,
or housing.
-------
11) Shading - Scoring for this metric is
the percentage of segment that is shaded. Both
the extent (total area) and the duration (day
length) of shading is considered in scoring
shading (e.g., full and dense shading all day
in summer is 100% and full exposure all day
in the summer is 0%).
12) Aesthetic Rating- Score is based on
the visual appeal of the site, the presence of
human refuse, and the degree of
channelization and riparian vegetation distur-
bance. Segments in essentially a natural state,
with no human refuse and that have a visu-
ally outstanding character are scored the high-
est. Scored 0 to 20.
13) Remoteness - Scoring is based on
presence of detectable human activity and the
difficulty in accessing the segment. The high-
est scores are given to streams that are diffi-
cult to access, are more than 0.25 miles from
the nearest road, and that show little or no
evidence of human activity. Segments which
are immediately adjacent to roadside access
or have an unnatural and/or unpleasant view,
smell, or sound are noted, are scored the low-
est. Scored 0 to 20.
Miscellaneous Measurements Made -
Other miscellaneous measurements made in
the course of completing an MDNR-MBSS
habitat assessment include: la, b) thalweg
depth and velocity at 0,25, 50 and 75 m along
the sample segment; 2) wetted width; 3) maxi-
mum stream depth; 4) overbank flood height;
5) categorization of adjacent land use (11 cat-
egories); 6) categorization of stream charac-
ter (26 categories); 7) number of woody de-
bris; 8) number of rootwads; and 9) flow (Lat
Loc, depth, velocity).
-------
TNDT is an estimate of diatom species
richness. High species richness is assumed for
unimpacted sites and species richness is ex-
pected to decrease with increasing pollution.
Slight levels of nutrient enrichment, however,
may increase species richness in headwater
or naturally unproductive, nutrient-poor
streams (Bahls et al. 1992).
non Diversity value to the Maximum Shan-
non Diversity value (David Beeson; S.M.
Stoller Corporation, personal communica-
tion). Species diversity, despite the contro-
versy surrounding it, has historically been used
with success as an indicator of organic (sew-
age) pollution (Wilhm and Dorris 1968, We-
ber 1973, Cooper and Wilhm 1975). Bahls et
al. (1992) uses Shannon diversity because of
its sensitivity to water quality changes, and
Stevenson (1984) suggests that changes in
species diversity, rather than the diversity
value, may be useful indicators of changes in
water quality.
The Shannon Index is affected by both
the number of species in a sample and the dis-
tribution of individuals among those species
(Klemm et al. 1990). Because species rich-
ness and evenness may vary independently,
under certain conditions, Shannon diversity
values can be misleading (e.g., when the to-
tal number of taxa is less than 10). Assess-
ments for low-richness samples can be im-
proved by comparing the assemblage Shan-
The PSc index, discussed by Whittaker
(1952), was used by Whittaker and Fairbanks
(1958) to compare planktonic copepod com-
munities. It was chosen for use in diatom
bioassessments because it shows community
similarities based on relative abundances, and
therefore gives more weight to dominant taxa
than to rare ones. PS only applies to com-
-------
parison to a control site, or to multivariate
cluster analysis. If emphasis is comparison to
regional reference condition (i.e., a compos-
ite of sites), PSc will not be useful. PSc values
range from 0 (no similarity) to 100% (identi-
cal).
The formula for calculating PSc is:
The formula used to calculate PTI is:
where a. = the percentage of species / in
sam pie A and b. = the percentage of spe-
cies /in sample B.
The pollution tolerance index (PTI) used
by Kentucky DEP is most similar to that of
Lange-Bertalot (1979) and resembles the
Hilsenhoff biotic index for macroinvertebrates
(Hilsenhoff 1987). Lange-Bertalot distin-
guished three categories of diatoms accord-
ing to their tolerance to increased pollution,
with species assigned a value of 1 for most
tolerant taxa (e.g., Nitzschia palea or
Gomphonemaparvulum) to 3 for relatively
sensitive species. For the PTI, Lange-
Bertalot's list has been adapted to four cat-
egories to differentiate a large moderately tol-
erant group of species (similar to his splitting
of category 2 diatoms into 2a and 2b); the
Kentucky DEP diatom pollution tolerance
values range from one (most tolerant) to four
(most sensitive). Tolerance values have been
generated from several sources, including
Lowe (1974), Patrick and Reimer (1966,
1975),Patrick(1977),Lange-Bertalot(1979),
Descy (1979), Sabateretal. (1988),Bahlset
al. (1992), and Oklahoma Conservation Com-
mission (1993).
AT
where r\. = number of cells counted for
species i, t = tolerance value of species
/ (1,2,or 3), and N = total number of cells
counted.
The percent sensitive diatoms metric is
the sum of the relative abundances of all in-
tolerant species. This metric is especially im-
portant in smaller-order streams where pri-
mary productivity may be naturally low, caus-
ing the other metrics to underestimate water
quality.
The percent motile diatoms is a siltation
index, as the relative abundance ofNavicula
+ Nitzschia + Surriella. This metric is espe-
cially important in smaller-order streams
where primary productivity may be naturally
low, causing the other metrics to underesti-
mate water quality.
This species is a cosmopolitan diatom
that has a very broad ecological amplitude. It
is an attached diatom and often the first spe-
cies to pioneer a recently scoured site, some-
times to the exclusion of all other algae. A.
minutissima is also frequently dominant in
streams subjected to acid mine drainage (e.g.,
-------
Silver Bow Creek, Montana) and to other
chemical insults. The percent abundance of
A. minutissuma has been found to be directly
proportional to the time that has elapsed since
the last scouring flow or episode of toxic pol-
lution. For use in bioassessment, the quartiles
of this metric from a population of sites has
been used to establish judgement criteria (e.g.,
0-25% = no disturbance, 25-50% = minor
disturbance, 50-75% = moderate disturbance,
and 75-100% = severe disturbance). Least-
impaired streams in Montana may contain up
to 50% A. minutissima (Loren Bahls, retired
phycologist and Chief of Nonpoint Section
of the Montana Department of Environmen-
tal Quality, personal communication).
TableB-l. IndicatorTaxa(TakenFromKentucky
DEP1993).
In general, an inverse relationship exists
between the number of soft algae present and
impairment. Extremely low taxa richness of
non-diatoms indicates the possible occurrence
of a toxicity problem (e.g., acid mine drain-
age), while high taxa richness suggests clean
water. However, extremely high taxa richness
in low-order streams may indicate a minor
degree of nutrient enrichment, while low taxa
richness may be natural in low-order streams
with low nutrient inputs.
Taxa
Indicator Condition
Occur atapH of 7 orbelow.
Occur at a pH of 7 or above.
Have a growth requirement
for organic nitrogen; often
associated with wastewater
treatment plant effluents.
Tolerate elevated chloride
concentrations (including
brackish water forms).
Characteristic of water with
high nutrient concentra-
tions.
Morphological changes are
an indication of physio-
logical stress often found in
association with toxic
materials (e.g., metals).
All taxa that cause water to
taste and/or smell noxious;
taxa will be identified in
streams used for domestic
water supplies.
of indicator taxa is recorded and used in con-
junction with other data to determine water
quality impairment.
Acidophilictaxa
Alkaliphilictaxa
Heterotrophic taxa
Halophilictaxa
Eutrophic taxa
Aberrant diatoms
Taste and odor taxa
Certain taxa are good indicators of pol-
lution. Autecological information on these
indicator taxa is available in published refer-
ences (Palmer 1969, 1977; Prescott 1969;
Lowe 1974; and Patrick and Reimer 1966,
1975). Indicator categories are provided in
Table B-l. Presence and relative abundance
The relative abundances of all taxa can
be calculated from counting a pre-determined
number of cells or, relative abundance of each
taxon (diatoms are combined under the head-
ing Bacillariophyceae) can be estimated as
follows:
-------
Rare Present in <25% of the exam-
ined fields and only 1 unit per
field
Common Present in 25-75% of the
examined fields and 2-10
units per field
Abundant Present in >75% of the exam-
ined fields and > 10 units per
field.
Representatives from several phyla of
algae are common from sites with good wa-
ter quality. The number of phyla represented
is reported as an indicator of diversity.
Benthic chlorophyll a values are used
as an estimate of algal biomass. Chlorophyll
a values can be extremely variable because
of the patchiness of periphyton distribution.
Therefore, assessments are based on a mean
of three or more replicate samples. These val-
ues are used to compare biomass accrual at
the same station overtime or between stations
during the same sampling period. High chlo-
rophyll a values may indicate nutrient enrich-
ment, while low values may either indicate
low nutrient availability, toxicity, or low-light
availability because of shading, sedimenta-
tion, or high turbidity. Chlorophyll a values
are used only in support of other analyses.
Benthic AFDM values are used as an
estimate of total organic material accumulated
on the artificial substrate. This organic mate-
rial includes all living organisms (algae, bac-
teria, fungi, protozoa, and macroinvertebrates)
as well as non-living detritus. Ash-free dry-
mass values have been used in conjunction
with chlorophyll a as a means of determining
the trophic status (autotrophic vs. het-
erotrophic) of streams. The Autotrophic In-
dex (AI) is calculated as follows:
AI = AFDM (mg/m2)/Chlorophyll a
(mg/m2).
High AI values (>200) indicate the com-
munity is dominated by heterotrophic organ-
isms, and extremely high values indicate poor
water quality (Weber 1973; Weitzel 1979;
Matthews et al. 1980). This index should be
used with discretion, as non-living organic
detritus can artificially inflate the AFDW
value.
The USEPA RBP (Barbour et al. 1999)
recommends that the AI be modified as chl/
AFDM. The index is then positively related
to the autotrophic proportion of the assem-
blage and not the heterotrophic component.
Also, the index will have better statistical
properties as a proportion or percent (chl/
AFDM is usually about 0.1% of the assem-
blage by mass) than in the original form as
AFDM/chl.
-------
Scoring Schemeforthe B-IBI
Metrics Used in the Coastal Plain B-IBI
Total number of taxa
Number of EPT taxa
Percent Ephemeroptera
Percent Tanytarsini of Chironomidae
Beck's Biotic Index
Number of scraper taxa
Percent clingers
Metrics Used in the Non-Coastal Plain B-IBI
Total number of taxa
Number of EPT taxa
Number of Ephemeroptera taxa
Number of Diptera taxa
Percent Ephemeroptera
Percent Tanytarsini
Number of intolerant taxa
Percent tolerant
Percent collectors
5
>24
>6
>11.4
>13.0
>12
>4
>62.1
5
>22
>12
>4
>9
>20.3
>4.8
>8
<11.8
>31
3
11-24
3-6
2.0-11.4
XXO-13.0
4-12
14
38.7-62.1
3
16-22
5-12
24
6-9
5.7-20.3
XX04.8
3-8
11.848
13.5-31.0
1
<11
<5
<2.0
0.0
<4
<1
<38.7
1
<16
<5
<2
<6
<5.7
0.0
<5
>48
<13.5
1 Total number of taxa - Measures the
overall variety of the macroinvertebrate as-
semblage. Expected to decrease with increas-
ing perturbation.
2 Number of EPT taxa - Number of
taxa in the insect orders Ephemeroptera (may-
flies), Plecoptera (stoneflies), and Trichoptera
(caddisflies). Expected to decrease with in-
creasing perturbation.
3. Percent Ephemeroptera - Percent
mayfly nymphs in the sample. Expected to
decrease with increasing perturbation.
4. Percent Tanytarsini of Chironomidae -
Percent of chironomids in the tribe Tanytarsini.
Expected to decrease with increasingperturbation.
5. Beck's Biotic Index - Weighted sum
of intolerant taxa (= 2 x number of Class 1
taxa + number of Class 2 taxa; where Class 1
taxa have tolerance values of 0 and 1, Class 2
-------
taxa have values from 2 to 4). Expected to de-
crease with increasing perturbation.
6. Number of scraper taxa - Number of
taxa that scrape food from substrate. Expected
to decrease with increasing perturbation.
1. Percent clingers - Percent of sample
primarily adapted for inhabiting flowing water,
as in riffles. Expected to decrease with increas-
ing perturbation.
1. Total number of taxa - Measures the
overall variety of the macroinvertebrate assem-
blage. Expected to decrease with increasing
perturbation.
2. Number of EPT taxa - Number of taxa
in the insect orders Ephemeroptera (mayflies),
Plecoptera (stoneflies), and Trichoptera
(caddisflies). Expected to decrease with increas-
ing perturbation.
3 Number of Ephemeroptera taxa -
Number of mayfly taxa. Expected to decrease
with increasing perturbation.
4. Number of Diptera taxa - Number
of "true" fly taxa (includes midges). Expected
to decrease with increasing perturbation.
5. Percent Ephemeroptera - Percent
mayfly nymphs in the sample. Expected to
decrease with increasing perturbation.
6. Percent Tanytarsini - Percent of
Tanytarsini midges to total fauna. Expected
to decrease with increasing perturbation.
1. Number of intolerant taxa - Num-
ber of taxa considered to be sensitive to per-
turbation (Hilsenhoff values 0-3). Expected
to decrease with increasing perturbation.
8. Percent tolerant individuals - Per-
cent of sample considered tolerant of pertur-
bation (Hilsenhoff values 7-10). Expected to
increase with increasing perturbation.
9. Percent collectors - Percent of sample
that feeds on detrital deposits or loose surface
films. Expected to decrease with increasing
perturbation.
-------
1. Total Number of Taxa - Taxa rich-
ness has historically been a key component
in most all evaluations of macroinvertebrate
integrity. Healthy, stable biological commu-
nities have high species richness and diver-
sity. Expected to decrease with increasing
perturbation.
2. Total Number of Mayfly Taxa - May-
flies are an important component of an undis-
turbed stream macroinvertebrate fauna. They are
pollution sensitive and are often the first to
disappear with the onset of perturbation. Ex-
pected to decrease with increasing perturba-
tion.
3. Total Number of Caddisfly Taxa -
Caddisflies are often a predominant component
of the macroinvertebrate fauna in larger, rela-
tively unimpacted Ohio streams and rivers.
Though tending to be slightly more pollution
tolerant than mayflies, they display a wide
range of tolerances among types. Few can
tolerate heavy pollution stress, and are there-
fore good indicators of environmental condi-
tions. Expected to decrease with increasing
perturbation.
4. Total Number of Dipteran Taxa - Of
all maj or aquatic invertebrate groups, dipterans,
especially midges of the family
Chironomidae, have the greatest faunal diver-
sity and display the greatest range of pollu-
tion tolerances. Under heavy pollution stress,
they can often be the only insect collected.
Larval taxonomy has improved greatly for the
group and clear patterns of organism assem-
blages have become distinct under water qual-
ity conditions ranging from the pristine to the
heavily organic and toxic. Expected to de-
crease with increasing perturbation.
5. Percent Mayflies - The percent abun-
dance of mayflies in a sample can react
strongly and rapidly to often minor environ-
mental disturbances. Mayfly abundance is
reduced considerably under slight impact and
is essentially non-existent under severe im-
pact. Expected to decrease with increasing
perturbation.
6. Percent Caddisflies - Percent abun-
dance of caddisflies is strongly related to
stream size. Optimal habitat and availability
of appropriate food type seem to be the main
considerations for large populations of
caddisflies. Because of their general position
as an intermediately pollution-tolerant group
between mayflies and dipterans, and because
they disappear rapidly under environmental
-------
stress, zero scores are restricted to those sites
draining areas less than 600 square miles where
no caddisflies are collected. At sites draining
areas greater than 600 square miles, appropriate
habitat conditions are much more likely to ex-
ist, and caddisflies should be present in at least
minimal numbers. Expected to decrease with in-
creasing perturbation.
7. Percent Tribe Tanytarsini Midges -
Tanytarsini midges are a tribe of the chirono-
mid subfamily Chironomidae. The larvae are
generally burrowers or clingers, and many spe-
cies build cases out of sand, silt, and/or detritus.
Many species feed on microorganisms and de-
tritus through filtering and gathering though a
few are scrapers. Eleven genera and up to 140
species occur in North America, though only 8
genera and 21 distinct taxa have been collected
in Ohio. They appear to be relatively pollution
sensitive and often disappear or decline under
even minor pollution stress. Expected to decrease
with increasing perturbation.
8. Percent Other Dipterans and Non-
insects - Community percentage of all dipter-
ans (excluding the midge tribe Tanytarsini) and
other non-insect invertebrates, such as aquatic
worms, flatworms, scuds, aquatic sow bugs,
freshwater hydras, and snails. This metric is one
of two negative metrics of the ICI. Taxa are those
that generally tend to become predominant un-
der adverse water quality conditions. Expected
to increase with increasing perturbation.
9. Percent Tolerant Organisms - Those
organisms that appear to be extremely pollution
tolerant and tend to predominate in cases of se-
vere perturbation. This is a negative metric. List
of pollution-tolerant organisms used:
• Aquatic segmented worms: Oligochaeta
• Midges: Pseclrotanypusdyari, Cricotopus
bicinctus, Cricotopus sylvestris,
Nanocladius
• distinctus, Chironomus, Dicrotendipes
simpsoni, Gryptotendipesbarbipes,
Parachironomus
• hirtalus, Polypedilum fallax,
Polypedilum illinoense
• Limpets: Ferrissia
• PondSnails: Physella
Expected to increase with increasing per-
turbation.
10. Total Number of Qualitative
Ephemeroptera, Plecoptera, and Trichoptera
Taxa - Generated by the qualitative sample taken
in conjunction with the artificial substrate sam-
pling. Affected by the kinds of natural sub-
stances available in the sampling area, the
metric is a measurement of habitat quality.
Expected to decrease with increasing pertur-
bation.
-------
Iwb = 0.5 InN + 0.5 In B +#(no.) +#(wt.)
where:
N = relative numbers of all species ex-
cluding species designated highly tol-
erant
B = relative weights of all species ex-
cluding species designated highly tol-
erant
//(no.) = Shannon diversity index based
on numbers.
//(wt.) = Shannon diversity index based
on weight.
//=-(ni)/Nloge(ni)/N
where:
n = relative numbers or weight of the ith
species
N = total number or weight of the sample
Relative abundance (number and weight)
data are derived from pulsed D.C. electro-fish-
ing catches where sampling effort is based on
a per kilometer basis for boat methods and on
a 0.3 kilometer basis for wading methods
(OEPA 1988).
-------
1. Native species richness - Modified
from Karr's (1981) Species Richness. Native
species richness is a classic measure of
biodiversity with focus on natives. This is
important where introductions are common.
2. Native family richness - Replaces
Karr's (1981) Darter, Sunfish, and Sucker
Richness. A measure of biodiversity at the
family level of organization. Useful for as-
sessing the degree to which the reach sup-
ports families typically represented by only a
single species, and therefore whose losses
mean the loss of entire families from the as-
semblage.
3. Sensitive species richness - Modified
from Karr's (1981) Intolerant Species Rich-
ness. Species likely to be the first to disap-
pear following anthropogenic disturbance and
the last to recover following restoration. Most
useful at discriminating among reaches with
higher quality assemblages.
4. Percent tolerant individuals - Modi-
fied from Karr's (1981) percent Green Sunfish.
Evaluates the tendency of one or more weedy
species to dominate the assemblage. Typically
highly disturbed sites are numerically domi-
nated by tolerant species. In the Appalachians,
the blacknose dace and creek chub are prime
examples. However, these taxa may naturally
dominate very small streams. Calculated as:
1-(proportion of tolerant individuals in
excess of 10%).
5. Benthic species richness - Modified
from Karr's (1981) Darter Species Richness.
Measures quality of habitat (substrate) for
small bottom dwelling species; includes dart-
ers, sculpins, benthic minnows (e.g., dace,
lamprey).
6. Water column species richness -
Modified from Karr's (1981) Sunfish Species
Richness. Measures quality of water column
(especially pools) for stronger swimming spe-
cies that feed largely on drifting prey; includes
sunfish, many minnows, salmonids.
7. Percent alien individuals - This is a
measure of the degree to which the site is con-
taminated by biological pollution. Also, they
represent a direct disturbance themselves as a
result of predation and competition with spe-
-------
cies that are not adapted to coexisting with
them; includes common carp, brown trout,
rainbow trout, many sunfishes, and bass.
8. Number of trophic guilds - Mea-
sures niche diversity in streams.
9. Percent top carnivore (invertivore-
piscivore) individuals - Modified from Karr's
(1981) percent Carnivore; includes species
that are piscivores or invertivore-piscivores as
adults (bass, pike, several sunfishes, eel). Es-
timates the ability of the food chain to sup-
port fish that prey largely on other fish, verte-
brates, or large macrobenthos. Calculated as:
proportion of top carnivores/expected
value of 10%.
10. Invertivore individuals - Measures
the capacity of the food base to support the
major trophic group of fishes in most streams.
Prey includes both insects and other inverte-
brates. Calculated as:
proportion of invertivores/expected value
of 50%.
11. Percent herbivores - This metric
includes herbivorous scrapers and
phytoplanktivores. These species disappear
when sediment decreases food quality. Cal-
culated as:
1 - (proportion of herbivores in excess
of 10%).
12. Percent omnivore individuals - A
measure of the dominance of trophic guilds
by individuals that can eat either plant or ani-
mal materials. These are trophic generalists
with at least 25% of its diet as animals and at
least 25% is plants. Ecomorphology (mouth
gape, dentition, pharyngeal teeth, gut length)
also suggest dietary niche. Calculated as:
1 - (proportion of omnivores in excess
of 20%).
13. Number of specialized reproduc-
tive strategies - Replaces Karr's (1981) per-
cent hybrids. The number of different repro-
ductive strategies represented in the assem-
blage not to include generalist or broadcast
spawners. A measure of niche diversity in
streams, it evaluates the degree to which the
reach supports a variety of reproductive strat-
egies.
14. Proportion of gravel spawning
species - Replaces percent Simple Lithophils
metric of some authors. Comprised of some
representatives of Baton's (1975) Lithophilic
A.1,A.2, .1 and B.2 species.
15. Proportion of tolerant substrate
spawners - They may spawn over gravel,
vegetation, detritus, sand or silt or construct a
nest, guard it against predation and maintain
it, fanning or otherwise manipulating the eggs
to remove silt or increase flow over the nest.
Eggs are demersal and/or adhesive. Calcu-
lated as:
1 - (proportion of tolerant reproductive
individuals in excess of 10%.
16. Total abundance - The number of
individuals collected at the site. Low abun-
dance may result from toxic or extremely oli-
gotrophic waters. Calculated as:
number of individuals/expected value of
500.
1. Total Number of Indigenous Fish
Species - This metric is used with all three
versions of the IBI. Exotic species are not in-
-------
eluded. This metric is based on the well-docu-
mented observation that the number of indig-
enous fish species in a given size stream or
river will decline with increasing environmen-
tal disturbance. (Karr 1981; Karr et al. 1986).
Thus, the number of fish species metric is
expected to give an indication of environmen-
tal quality throughout the range from excep-
tional to poor. Exotic (i.e., introduced) spe-
cies present in a system through stocking or
inadvertent releases do not provide an accu-
rate assessment of overall integrity and their
abundance may even indicate a loss of integ-
rity (Karr etal. 1986).
2. Number of Darter Species (Wad-
ing, Headwaters), Proportion of Round-
bodied Catostomidae (Boat Method) - The
darter species metric is reflective of good water
quality conditions (Karr et al. 1986). None of
the species in this group have been found to
thrive in degraded stream conditions. Eleven
of the 22 Ohio species have been found to be
highly intolerant of degraded conditions based
on the Ohio EPA intolerance criteria. Life
history data on this group show darters to be
insectivorous, habitat specialists, and sensi-
tive to physical and chemical environmental
disturbances (Kuehne and Barbour 1983).
These factors make darter species reliable in-
dicators of good water quality and habitat
conditions.
3. Number of Sunfish Species (Wad-
ing, Boat), Proportion of Headwaters Spe-
cies (Headwaters) - This metric follows Karr
(1981) and Karr et al. (1986) by including
the number of sunfish species (Centrachidae)
collected at a site, excluding the black basses
(Micropterus spp.). The redear sunfish
(Lepomis microlophus) is not included be-
cause, in Ohio, it is introduced and only lo-
cally distributed. Hybrid sunfish are also ex-
cluded from this metric.
4. Number of Sucker Species (Wad-
ing, Boat), Number of Minnow Species
(Headwaters) - All species in the family
Catostomidae are included in this metric.
Suckers represent a major component of the
Ohio fish fauna with their total biomass in
many samples surpassing that of all other spe-
cies combined. The general intolerance of
most sucker species to habitat and water qual-
ity degradation (Karr 1981; Trautman 1981;
Becker 1983; Karr et al. 1986) results in a
metric with a sensitivity at the high end of
environmental quality. In addition the rela-
tively long life spans of many sucker species
(10-20 years) (Becker 1983) provides a long-
term assessment of past and prevailing envi-
ronmental conditions. Of the 19 species still
present in Ohio (one is extinct), seven are
widely distributed throughout the state.
5. Number of Intolerant Species
(Wading, Boat), Number of Sensitive Spe-
cies (Headwaters) - The number of intoler-
ant species metric is designed to distinguish
streams of the highest quality. As a result, the
sensitivity of this metric is at the highest end
of biotic integrity. Designation of too many
species as intolerant will prevent this metric
from discrimination among the highest qual-
ity streams. Only species that are highly in-
tolerant to a variety of disturbances were in-
cluded in this metric so that it will respond to
diverse types of perturbations; species intol-
erant to one type of disturbance, but not an-
other were not included.
6. Percent Abundance of Tolerant
Species (Replacing Karr's % Green Sun-
fish) - This metric is a modification of one of
Karr's (1981) original IBI metrics, the per-
centage of the fish community comprised by
green sunfish {Lepomis cyanellus). This met-
ric was designed to detect a decline in stream
quality from fair to poor. The green sunfish is
-------
a species that is often present in moderate
numbers in many Midwest streams and can
become a predominant component of the com-
munity in areas with degraded habitat and/or
water quality. This ability to survive and re-
produce in disturbed environments makes this
species sensitive to changes in environmental
quality in severely impacted areas. Although
green sunfish are one of the most widely dis-
tributed and numerically abundant fish spe-
cies found in the Midwest, they show a de-
cided preference towards smaller sized and
low gradient streams. This limits their utility
in assessing impacts in larger streams and riv-
ers. Karr et al. (1986) suggested that other
species could be substituted for the green sun-
fish if they respond in a similar manner. Sev-
eral species meeting this criterion were in-
cluded to give this metric an improved sensi-
tivity for the range of stream and river sizes
encountered in Ohio. Because individual spe-
cies have habitat requirements that are keyed
to stream size, composition of the tolerant
species metric shifts with drainage area and
this metric remains useful among small, me-
dium, and large streams and rivers.
7. Percent Omnivores - The Ohio EPA
definition of an omnivorous species follows
Karr (1981) and Karr et al. (1986) with two
important distinctions added. Specialized fil-
ter-feeding species which technically are
omnivorous are not included. Specialist filter
feeders are represented in Ohio by the paddle-
fish (Polyodon spathuld) and brook lamprey
ammocoetes. These species are generally sen-
sitive to environmental degradation. Since the
omnivore metric is designed to measure in-
creasing levels of environmental degradation
due to a disruption of the food base it is not
appropriate to include these sensitive, filter
feeding species in this metric. This metric was
further restricted to those species that did not
show feeding specialization and were re-
ported primarily as omnivores in all studies
reviewed. This removes such species as chan-
nel catfish (Ictaluruspunctatus) which may
or may not feed as an omnivore under differ-
ent environmental conditions.
8. Proportion of Insectivores (All) -
This metric is designed to be sensitive over
the middle range of biotic integrity. A low
abundance of insectivorous species can reflect
a degradation to the insect food base of a
stream (Karr et al. 1986). As disturbance in-
creases, the diversity of benthic insects de-
creases, production becomes more variable,
and the community often becomes predomi-
nated by a few taxa( Jones etal. 1981). Thus,
specialist feeders such as specialist insecti-
vores will decrease and be replaced by gen-
eral! st feeders such as omnivores. This repre-
sents a modification from Karr et al. (1986)
using insectivorous Cyprinids alone.
9. Top Carnivores (Wading, Boat),
Proportion of Pioneering Species (Head-
waters) - Karr (1981) developed the top car-
nivore metric to measure community integ-
rity in the upper functional levels of the fish
community. And Karr (1981) and Karr et al.
(1986) were followed in designating a spe-
cies as a top carnivore. Species which feed
primarily on other vertebrates or crayfish are
included in this metric. As with the omnivore
metric, species which display feeding plas-
ticity are excluded (e.g., channel catfish).
10. Number of Individuals in a Sample
(All) - This metric assesses population abun-
dance as the number of individuals per unit
of sampling effort. This metric is most sensi-
tive at the low to middle end of biotic integ-
rity when polluted sites yield fewer individu-
als (Karr et al. 1986). In such cases, the nor-
mal trophic relationships are disturbed enough
-------
to either have severe effects on fish produc-
tion or directly reduce fish abundance through
toxic effects. As integrity increases, total abun-
dance increases and becomes more variable
with natural factors such as ionic concentra-
tion, temperature, and amount of energy
reaching the stream surface. However, cer-
tain perturbations, such as channelization with
canopy removal, can lead to increases in the
abundance of fishes, especially tolerant spe-
cies, (e.g., bluntnose minnow). Thus, inclu-
sion of these species may obscure negative
environmental change. To decrease the vari-
ability in the scoring of this metric, it excludes
species designated as tolerant.
11. Proportion of Individuals as
Simple Lithophilic Spawners - This metric
was designed as a replacement metric for the
proportion of individuals as hybrids. In Ohio
streams, the hybrid metric was not a consis-
tent indication of water quality. Hybrids have
been observed to occur in high quality Ohio
streams (e.g., minnow hybrids), can arise from
sensitive parent species (e.g., longear sunfish),
are often times absent from headwaters
streams and severely impacted streams, and
they can be difficult to identify. Although the
frequency of hybridization has often been
associated with habitat degradation this did
not appear consistently enough in the Ohio
EPA data base to distinguish this type of im-
pact.
12. Proportion of Individuals with
Deformities, Eroded Fins, Lesions, and
Tumors (BELT) (replaces Karr's % dis-
eased individuals) - This metric keys in on
the health of individual fish within a commu-
nity using the percent occurrence of external
anomalies and corresponds to the percentage
of diseased fish in Karr's (1981) original IBI.
Studies of wild fish populations have revealed
that these and other anomalies are either ab-
sent or occur at very low rates at reference
sites, but reach higher percentages at impacted
sites (Mills et al. 1966; Berra and Au 1981;
Baumann et al. 1987). Common causes of
DELT anomalies are described in Allison et
al. (1977), Post (1983) and Ohio EPA (1988)
and include the effects of bacterial, viral, fun-
gal, and parasitic infections, neoplastic dis-
eases, and chemicals. An increase in the fre-
quency of occurrence of these anomalies is
generally an indication of stress and environ-
mental degradation which may be caused by
chemical pollutants, overcrowding, improper
diet, excessive siltation, and other distur-
bances. Blackspot is not included because the
presence and varying degrees of infection may
be natural and not related to environmental
degradation (Allison et al. 1977; Berra and
Au 1981). Also, analysis of Ohio data has
shown no clear relationship between black
spot and stream degradation (Wittier et al.
1987). Other parasites are also excluded due
to the lack of a consistent relationship with
environmental degradation although their ef-
fects can resemble and lead to tumors, defor-
mities, and lesions. Prior to using this metric,
Ohio EPA (1987a) should be referred for con-
sistent data-recording procedures and as a ref-
erence for specific anomalies included in each
category.
The metrics used in the IBI represent
various attributes of the fish assemblage in-
dicative of ecological quality, so that differ-
ences in metric values reflect important dif-
ferences in stream conditions.
1. Number of native species - The con-
cept of species richness has been used exten-
sively to assess the quality of ecological sys-
-------
terns. In most cases, the number offish spe-
cies supported by streams of a given size in a
given region decreases with environmental
degradation (Karr et al. 1986). The reduction
in number of species may be as a result of
reduced diversity of habitats, the loss of spe-
cies that are sensitive to pollutants, or other
human-induced impacts. Introduced species
are not included in this metric because the
presence of these species may result in a
higher species number than would naturally
be found in a given stream. In addition, the
species richness value for a site in which spe-
cies have been introduced would not reflect
the lowered richness that may result from
human disturbance at the site. Leidy and
Fiedler (1985) found that species richness in-
creased at sites with moderate human distur-
bance mostly due to the addition of introduced
species. There are some potential exceptions
to this rule. For example, minimally disturbed
coldwater systems, dominated by salmonids
and sculpin, tend to have low number of spe-
cies.
2. Number of benthic species - Benthic
fish species are sensitive to degradation of stream
benthic habitats because of the their specific re-
quirements for reproducing and feeding on the
stream bottom (Page 1983). Benthic habitats are
degraded by channelization, siltation, and reduc-
tion of dissolved oxygen and are often degraded
in streams with watersheds that contain a great
deal of impervious surface. Berkman and
Rabeni (1987) documented reduced abundance
of benthic insectivores in streams with increased
amounts of silt in riffles. Benthic specialists in-
cluded in this metric are darter, sculpin, madtom,
and lamprey species.
3. Percent tolerant individuals - Intol-
erant species are among the first to be affected
by perturbations (Jenkins and Burkhead 1993,
Pflieger 1975, Smith 1979, Trautman 1981).
As specific habitats required by habitat spe-
cialists are degraded, the relative abundance
of tolerant, habitat generalists becomes
greater.
4. Percent abundance of the dominant
species - The contribution of the dominant
(tolerant) taxa to the fish community is likely
to increase as the amount and extent of deg-
radation increases. As intolerant species be-
come less abundant, tolerant species increase
in relative abundance in degraded streams and
may become the dominant taxa (Karr et al.
1986). This metric was calculated as the per-
cent contribution of the single dominant fish
species to the total number of individuals at a
site.
5. Percent of individuals as general-
ists, omnivores, or invertivores - The domi-
nance of general! st feeders increases as spe-
cific food sources become less reliable, i.e.,
when degraded conditions reduce the abun-
dance of particular prey items. An opportu-
nistic foraging strategy makes generalists
more successful than specialized foragers be-
cause they are better suited to a shifting food
base in the presence of degraded conditions
than are more specialized feeders (Karr et al.
1986).
6. Percent of individuals as insecti-
vores - This metric takes into account the re-
sponse of fishes to impacts on lower trophic
levels. Fewer insectivorous fishes are col-
lected in degraded streams probably due to
decreases in the supply of preferred insects,
reflecting degraded chemical or habitat qual-
ity (Karr etal. 1986).
7. Abundance (number of individu-
als) per square meter - Degraded streams
are generally expected to yield fewer individu-
als than less severely impacted streams.
Streams of similar size with greater heteroge-
-------
neity of habitat generally contain larger num-
bers of individuals than streams with homo-
geneous habitat as a result of anthropogenic
impact on the stream. In addition, streams with
degraded chemical or habitat tend to support
only tolerant species of fishes are likely to
have depressed overall numbers of fishes. One
notable exception is elevated abundance in
the presence of excess nutrients, particularly
of tolerant species.
8. Biomass per square meter - The bio-
mass that a stream can accommodate is a func-
tion of the quantity and quality of available
stream habitat. As with abundance, the biom-
ass in a stream is expected to be lower in de-
graded streams compared to higher quality
streams. In general, more and larger fishes are
expected in higher quality streams. Larger
individuals of a species may be indicative of
longevity of the individuals. Long lived indi-
viduals indicate that the streams may have a
history of good stream quality.
9. Percent of individuals as lithophilic
spawners - Lithophilic spawners (Balon
1975) utilize rocks, rubble, or gravel substrates
for egg deposition. Because they require clean
spawning substrates and may use interstitial
spaces, lithophils are particularly susceptible
to siltation. Since silt is likely the most com-
mon stream pollutant in the state of Maryland,
this metric may be useful in identifying streams
that are degraded with substantial silt loads.
-------
MDNR-MBSS Method for Deriving IBI Scores for the State Data Sets
Coastal Plain Metrics
Number of native species
Number of benthic species
Percent tolerant individuals
Percent abundance of dominant species
Percent generalists, omnivores, and
invertivores
Number of individuals per square meter
Biomass(gperm2)
Percent lithophilic spawners
1
Criteria vary with
stream size*
Criteria vary with
stream size*
More than 80
More than 78
More than 99
Less than 0.47
Less than 5.1
0
3
80to31
78to31
99 to 88
0.47 to 0.62
5.1 to 9.6
0 to 0.6
5
Less than 3 1
Less than 3 1
Less than 88
More than 0.62
More than 9. 6
Mo re than 0.6
Non-Coastal Plain Metrics
Number of native species
Number of benthic species
Percent tolerant individuals
Percent abundance of dominant species
Percent generalists, omnivores, and
invertivores
Number insectivores
Number of individuals per m2
Percent lithophilic spawners
Criteria vary with
stream size*
Criteria vary with
stream size*
More than 82
More than 78
More than 95
Less than 5
Less than 0.22
Less than 6
82 to 50
78to51
95 to 59
5 to 33
0.22 to 0.63
6 to 32
Less than 50
Less than 5 1
Less than59
More than 3 3
More than 0.63
More than 3 2
*Metrics were adjusted for watershed area as follows: adjusted value = observed value/expected value,
where expected value = m x log (watershed area in acres)+b. Values of m and b are:
Coastal Plain
Slope (ml Intercept (b)
Non-Coastal Plain
Slope (ml Intercept (bl
Number of native species
Number of benthic species
5.2142
1.4478
-7.7258
-2.5532
6.3258
0.9016
-12.7351
-1.2345
-------
Scoring Criteria For Adjusted Metrics
Coastal Plain
Number of native species-adjusted value
Number of benthic species-adjusted value
Non-Coastal Plain
Number of native species-adjusted value
Number of benthic species-adiusted value
1
MorethanO.74
Less than 0.70
Less than 0.47
Less than 0.44
3
0.74 to 1.05
0.70 to 0.99
0.47 to 0.77
0.44 to 0.82
5
More than 1.05
Less than 0.70
More than 0.77
More than 0.82
-------
1
2
3
4
5
6
7
8
9
Alternative IBI USEPA-
Metrics EMAP-SW
# Species
# Native fish species X
# Native families X
# Darter species
# Darter and sculpin species
# Benthic species X
# Sunfish species
# Headwater species
% Headwater species
# Sucker species
# Minnow species
# Intolerant species
# Sensitive species X
% Tolerant species X
% Omnivores X
% Generalists, omnivores,
invertivores
% Insectivores
% Insectivorous species X
Ohio Ohio EPA
EPA Headwater
X
X X
X
X
X
X
X
X
X
X
X
X X
X X
X X
MDNR-MBSS MDNR-MBSS
Coastal Non-Tidal Plains
X
X X
X X
X X
X
X
X X
(continued)
-------
Appendix G (continued)
AlternativelBI USEPA- Ohio OhioEPA MDNR-MBSS MDNR-MBSS
Metrics EMAP-SW EPA Headwater Coastal Non-Tidal Plains
10 % Top carnivores X X
% Pioneering species X X
11 # Individuals XXX X
Density of individuals
% Abundance of dominant
species
Biomass X
12 % Simple Lithophils X
# Simple Lithophilic species X
% Silt-intolerant spawners X
Proportion of gravel X
spawning species
Proportion of tolerant X
substrate spawners
13 % Diseased individuals XX X
14 # Alien Individuals X
15 # Trophic Guilds X
16 % Herbivores X
X
X
X
X
17 # Specialized Reproductive X
strategies
------- |