xvEPA
United States
Environmental Protection
Agency
Development of a Fish Index of Biotic
Integrity to Assess the Condition of
West Virginia Streams:
Technical Support Document
RARE Sites
D Cold
A Cool
* Warm
^ Bus Ridge Mountains
Bj Central Appalachian Ridges and Valleys
^\ CentralAppalachians
| Western Allegheny Plateau
-------
EPA/600/R-06/010
February 16, 2006
Development of a Fish Index of Biotic Integrity to
Assess the Condition of West Virginia Streams:
Technical Support Document
David M. Walters
U.S. EPA Office of Research and Development
United States Environmental Protection Agency
Office of Research and Development
National Exposure Research Laboratory
26 West Martin Luther King Jr. Blvd.
Cincinnati, OH 45268
-------
NOTICE
The research described in this document has been funded by the United States Environmental
Protection Agency REMAP and RARE programs. It has been subjected to Agency peer and
administrative review and approved for publication as an EPA document.
Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
The correct citation for this document is:
Walters, D.M. 2006. Development of a Fish Index of Biotic Integrity to Assess the Condition of
West Virginia Streams: Technical Support Document. EPA/600/R-06/010. U.S. Environmental
Protection Agency, Cincinnati, Ohio.
11
-------
ACKNOWLEDGEMENTS
This report relied on the generous support of numerous individuals. Frank McCormick (U.S. Forest
Service) provided data analysis and technical support. Dan Cincotta (West Virginia DNR) led
crews to collect fish and other stream data. Alicia Shelton (SoBran Inc.) provided database support
and analysis of stream temperature data. Naomi Detenbeck (U.S. EPA) provided land cover data.
Lou Reynolds (U.S. EPA) and Karen Blocksom (U.S. EPA) provided reviews and helpful
comments to improve the report.
in
-------
Table of Contents
List of Figures v
List of Tables vii
List of Appendices viii
Introduction 9
Methods 10
Results 17
Discussion 31
References 32
Appendices 37
IV
-------
List of Figures
Figure 1. Site collection locales for West Virginia 1993-2002. RARE (Regional Applied
Research Effort) sites were sampled in 2001 and 2002. MAHA (Mid-Atlantic Highlands
Assessment) and MAIA (Mid-Atlantic Integrated Assessment) sites were sampled
between 1993-98 as part of the EMAP program 11
Figure 2. Intra- and inter-annual, and seasonal comparisons of native species richness.
Most of the 68 intra-annual revisits occurred within the same season (Spring, 1993-1994;
Summer, 1995-1998). In 1995 and 1996, 16 sites sampled in the spring were revisited in the
summer of 1995 and 1996. Five 1993-1994 spring sites were re-sampled during summer
1997-1998 as part of the Mid-Atlantic Integrated Assessment. Diagonal line represents 1:1
relationship 18
Figure 3. Cumulative distribution functions (CDFs) of index of biotic integrity (IBI) scores
for the Mid-Atlantic Highlands Assessment (MAHA) and the Mid-Atlantic Integrated
Assessment (MAIA). Indices were calculated using the IBI of McCormick et al. (2001) 19
Figure 4. Cumulative distribution functions (CDFs) of index of biotic integrity (IBI) scores
for the Mid-Atlantic Highlands Assessment (MAHA) and the Mid-Atlantic Integrated
Assessment (MAIA). Indices were calculated using the IBI of McCormick et al. (2001).
Region-wide results (solid lines) and state-specific results (dashed lines) are similar 19
Figure 5. Comparison of metric results from EMAP backpack and WVDNR parallel wire
electrofishing methods. The proportion of invertivore-piscivores and the proportion of
lithophils (clean gravel spawning individuals) show poor agreement 20
Figure 6. Responsiveness of IBI scores to measures of anthropogenic disturbance (MAIA
data, n = 43 sites plus 8 revisits). Disturbance was determined by (a.) classification of sites
using RBP and chemistry variables (McCormick et al. 2001), (b.) chemical classification
(Scott and Davis 2000), (c.) cumulative values for eleven disturbance criteria (see Methods
for details on disturbance calculations), (d.) condition class scores of Bryce et al. (1999).
Higher condition class scores correspond with higher levels of disturbance. Because of the
data requirements for calculation of condition class, fewer sites were classified 26
Figure 7. (A) Means of weekly maximum temperatures for all streams with temperature
loggers, (n = 84 sites total; n = 986 total weekly observations). Deployment and retrieval of
temperature loggers was staggered throughout the season, so observations for a given week
vary from 33-44 sites. Temperature data are from 2001 (41 sites) and 2002 (43 sites). Bars
represent a 95% confidence interval on the observations, and weeks designated with "a" are
significantly warmer than other weeks (p < 0.05). (B) Mean summer temperature. Means
are calculated from weekly maximum and minimum temperatures for the weeks of July 16-
Aug 12. Lines indicate temperature class designations for streams in Michigan (Wehrly et
al. 2003). Based on these criteria, n = 4 cold, 26 cool, and 54 warm streams 27
Figure 8. Response of index of biotic integrity (IBI) to WV reference condition
-------
classifications with sites plotted according to their temperature class 28
Figure 9. A. Cumulative distribution function (CDF) of index of biotic integrity scores for
West Virginia from EMAP surveys (n = 69). Points are coded for stream temperature
classification based on fish assemblage characteristics. B. CDF of index of biotic integrity
scores for West Virginia from EMAP surveys RARE project (n = 84) 29
Figure 10. Box plots of IB I and component metrics for cold (n = 3), cool (n = 26) and warm
(n = 54) streams from the RARE project. Metric acronyms are defined in Appendix 1 30
Figure 11. Plot of index of biotic integrity (IBI) and temperature for RARE sites.
Temperature was recorded in the field when fishes were sampled 31
VI
-------
List of Tables
Table 1. Disturbance classification thresholds for EMAP sites (1993-1998) 15
Table 2. Results of metric evaluation process. Metric definitions are provided in Appendix
I. Original metrics refer to those selected by McCormick et al. (2001) for the MAHAIBI 21
Table 3. Spearman rank correlation coefficients (r > 0.2; P <0.01) between fish assemblage
metric variables and chemical and physical habitat variables. Table of correlations offish
assemblage metrics with physical habitat variables is incorporated as a hyperlink to an Excel
file to conserve space 22
Table 4. Revised West Virginia Streams IBI metric scoring criteria 24
vn
-------
List of Appendices
Appendix I. Candidate metrics with descriptions, ranges, means and standard deviations.
Metrics in boldface represent original McCormick et al. (2001) variables (some names were
changed for analysis as indicated). Bold text designate revised metrics adopted here 34
Appendix II. Spearman's correlation coefficients (r > 0.2; p<0.05) showing response of WV
IBI score to chemistry and physical habitat variables sampled as part of EMAP (n = 69
sites). Stressor variable acronyms are defined in Appendix III 36
Appendix III. List if stressor variables compiled for WV stream sites 37
Vlll
-------
Introduction
The landscape, streams, and fish assemblages of the Mid-Atlantic region have endured a long
history of human impacts. Streams in the highlands have been subjected to stresses from acid
deposition, mining, logging, agriculture, and urban development (Raitz et al. 1984; Whitney 1994;
Jones et al. 1997). Agriculture and clear-cutting of highland and valley forests have exacerbated soil
erosion and sedimentation (USDA 1996). Active and abandoned coal mining resulted in mine
drainage that affected approximately 4,000 km of streams (Herlihy et al. 1990; USEPA 1995).
Extensive areas of the Ridge, Blue Ridge, and Appalachian Plateau ecoregions have poorly buffered
soils and steep slopes, rendering these highland streams particularly susceptible to acid precipitation
(Herlihy et al. 1993).
To accurately assess and manage the biotic integrity of aquatic ecosystems, a comprehensive
inventory of the biotic resources should be conducted, the current conditions of streams should be
determined, and the impacts of stressors (e.g., acid deposition, stream disturbances, mine drainage,
agricultural runoff, erosion, and domestic and industrial pollution) should be evaluated (Kazyak et
al. 1994). Although biomonitoring over the years has investigated many different assemblages for
use as indicators of water quality (e.g., Karr 1981, Lazorchak et al. 1998), recent USEPA guidance
documents recommend that fish and macroinvertebrate community analyses be adopted in state
water evaluation programs (US EPA 2002).
The West Virginia Division of Natural Resources (WVDNR) has conducted fishery studies in
wadeable streams since the early 1950s, focusing primarily on the status of game fish populations.
Survey data collected by WVDNR, including gamefish availability, standing crop, and recruitment,
are used to indirectly assess the ecological condition of stream resources, but West Virginia has not
initiated the use of fishes in their statewide assessment program.
Fish species exhibit diverse morphological, ecological, and behavioral adaptations to their natural
habitat and, thus, are particularly effective indicators of the condition of aquatic systems (Karr et al.
1986; Fausch et al. 1990; Simon and Lyons 1995). Human disturbance of streams and landscapes
alters key attributes of aquatic ecosystems: water quality, habitat structure, hydrological regime,
energy flow, and biological interactions (Karr and Dudley 1981). The index of biological integrity
(IBI) was developed to assess the condition of water bodies by direct evaluation of biological
attributes (Karr et al. 1986). The IBI is a composite index that integrates structural, ecological,
trophic, and reproductive attributes offish assemblages at multiple levels of organization (Fausch et
al. 1990). Originally developed for assessment of Midwestern U.S. warmwater streams, it has been
modified for use in other regions and waters (Simon and Lyons 1995; Lyons et al. 1996; Hughes et
al. 1998; McCormick et al. 2001). Several authors have argued that the IBI must be modified when
it is applied in different ecoregions (Fausch et al. 1984; Miller et al. 1988). In the Mid-Atlantic
region, researchers have developed IBIs for specific ecoregions (Scott and Hall 1997; Roth et al.
1998; Smogor and Angermeier 1999) or applied it to specific systems (Leonard and Orth 1986).
Over the last decade, a number of stream surveys and indicator development studies have been
conducted in the Mid-Atlantic Region. McCormick et al. (2001) developed a regional index of
biotic integrity for the assessment of Mid-Atlantic Highland wadeable streams. Data for the Mid-
Atlantic Highlands Assessment (MAHA) were collected from 1993-1996 as part of an
-------
Environmental Monitoring and Assessment Program (EMAP) study. In 1997-1998, the EMAP
study was expanded to include sites on the Coastal Plain of the eastern US and incorporate non-
wadeable streams as part of the Mid-Atlantic Integrated Assessment (MAIA). In 2000, USEPA
Region 3 initiated a Regional-EMAP (REMAP) and a Regional Applied Research Effort (RARE)
project in West Virginia (WV) in cooperation with the Office of Research and Development and
West Virginia Department of Natural Resources (Cincotta et al. 2001).
The primary purpose of this report is to document the development of a fish IBI for wadeable streams in
WV and to determine the applicability of the IBI in streams with different thermal regimes. The IBI
will be developed using fish data collected at EMAP sites from 1993-1998. The RARE sites are used
as an independent data set to test the robustness of the IBI across stream temperature regimes (i.e., cold,
cool, and warm water streams).
Methods
Study Area and Survey Design
Omernik (1987) and Woods et al. (1996) identified three primary ecoregions (Ridge and Valley,
Western Allegheny Plateau, and Central Appalachian Plateau) in West Virginia (a small segment of
the Blue Ridge lays in the eastern panhandle). The Ridge and Valley province consists of roughly
parallel northeast-southwest trending ridges and valleys that have a variety of widths, heights, and
geologic materials. The Western Allegheny Plateau is characterized by rounded hills separated by
narrow valleys. The Central Appalachian ecoregion is primarily a high, dissected, rugged plateau;
its terrain, cool climate, and infertile soils limit agriculture, resulting in a mostly forested land
cover. Extensive mixed mesophytic forests and mixed oak forests typically remain on the upland
terrain. Agriculture (dairy, livestock, and general farms) and residential developments are
concentrated in the valleys. Bituminous coal mines are common, and have caused the siltation and
acidification of streams.
Stream sites for the EMAP projects (hereafter referred to as "EMAP sites") were selected using a
randomized systematic design with a spatial component (Overton et al. 1991; Herlihy et al. 2000). The
sample population of streams in the region was delineated from digitized USGS topographic maps
(1:100,000 scale). Sample probabilities were set so that roughly equal numbers of first-, second-, and
third-order streams would be selected. Details on the sampling framework are provided by Davis and
Scott (2000) and McCormick et al. (2001). EMAP surveys included the Mid-Atlantic Highlands
Assessment (MAHA) sampled 1993-1996 and the Mid-Atlantic Integrated Assessment (MAIA)
sampled 1997-1998. Several streams in WV were also sampled under a Regional EMAP (REMAP)
study affiliated with the MAHA study, although habitat sampling was less extensive for the REMAP
sites. The total number of sites sampled in West Virginia under these programs was 96 with 12 revisits
(n = 108 samples).
RARE sites were selected using a stratified, random design. Sites were stratified by hydrologic
regime and land use (Detenbeck et al. 2004). Hydrologic regime was based on watershed storage
and main channel length, whereas land use was characterized as either high- or low-intensity.
High- and low- intensity designations were based on literature values for thresholds of land-use
activity at which clear degradation in biological or chemical condition was observed. The land-use
10
-------
based classes were developed for the predominant land uses (e.g., agricultural, urban/residential,
and mining activities) in each ecoregion.
Watersheds in the Potomac River basin were not included in the survey. A total of 119 sites were
selected and sampled in 2001-2002. Site selection processes are detailed in Detenbeck et al. (in
review). Stream sites surveyed in West Virginia for the EMAP/REMAP and RARE projects are shown
in Figure 1.
* MAHA/MAIA Sites
* RARE Sites
Figure 1. Site collection locales for West Virginia 1993-2002. RARE (Regional Applied Research
Effort) sites were sampled in 2001 and 2002. MAHA (Mid-Atlantic Highlands Assessment)/MAIA
(Mid-Atlantic Integrated Assessment) sites were sampled from 1993-98 as part of the EMAP program.
Integrating datasets: quantifying seasonal, regional and methodological biases — These analyses
draw on three large, complex datasets that differed in some degree in terms of sampling
methodology and scale. In an effort to minimize bias, the data sets were compared to identify
potential sources of seasonal, sample gear, and regional bias.
11
-------
Most MAHA data (1993 - 1996) were collected during the spring, whereas the majority of MAIA
data (1997 - 1998) were collected during the summer. During 1995 and 1996 MAHA crews made
29 revisits to 16 sites (region-wide) during the summer, in anticipation of the MAIA summer
sampling. Likewise, MAIA crews sampled 23 sites (region-wide) in both spring and summer 1997
and 1998. However, six samples collected in spring 1997 show much lower richness than samples
collected at the same sites again during summer 1997 and 1998 (i.e., richness > 6 species and
richness only 17-44% of subsequent samples). These samples suggest data quality problems may
exist and were treated as outliers and excluded from the seasonal analysis. Fish species richness
between seasons was compared using bivariate plots to identify any seasonal bias.
Fish sampling in the MAHA survey was originally restricted to wadeable streams with basin areas <
500 km2 in the upland ecoregions (Central Appalachian and Western Allegheny plateaus, Blue
Ridge and Ridge & Valleys) of the Mid-Atlantic Highlands. The MAIA survey was extended to the
Piedmont and Coastal Plain, neither of which is in West Virginia. McCormick et al. (2001) found
no significant differences in unscored metric values across ecoregions or basins for the nine metrics
in their IBI. They thus combined all site data for their analyses. RARE sites were also wadeable
streams draining < 500 km2. However, sampling was confined to West Virginia, so the spatial
extent of sampling was much smaller than the EMAP surveys. To test the assumption that the IBI
of McCormick et al. (2001) was as representative of the condition of streams in West Virginia as it
was of the Mid-Atlantic region, cumulative distribution functions (CDFs) of IBI scores of all
MAHA and MAIA sites were compared to CDFs of the scores from only the West Virginia sites for
those projects.
Fish collection methods varied between the EMAP and RARE surveys, so data from a subset of
sites sampled were compared using both techniques to determine if sampling method biased metric
values. As part of the 1997 MAIA project, WVDNR sampled 10 sites using both the parallel wire
and backpack electrofishing methods. Sampling events were at least two weeks apart. Raw metric
scores were calculated for each of the nine metrics and the MAHA IBI developed by McCormick et
al. (2001). Scores derived from the parallel wire and backpack methods were directly compared
using regression analysis to identify potential method bias.
Environmental Variables
Environmental variables were collected for EMAP and RARE sites at the basin and reach scales.
These data were collected to serve as stressor gradients for evaluating fish metrics and IBI
responsiveness. Environmental variables fall into four main categories: water chemistry, physical
habitat and riparian condition, basin and reach landscape characteristics, and temperature.
Water Chemistry — Water chemistry variables were derived from two primary sources, in-situ
measurements and water samples collected for laboratory analysis. In-situ measures included
dissolved oxygen (DO), pH, conductivity, and temperature. These data were collected using
standard field equipment and methods outlined in Plafkin et al. (1989). Water samples for
laboratory analysis were collected using standard EMAP protocols (Lazorchak et al. 1998).
Laboratory analysis of water samples followed standard US EPA methods (Davis and Scott 2000:
Cincotta et al. 2001). Analytes included nutrients, major anions and cations, alkalinity, suspended
solids, and heavy metals. Chemistry variables were collected at a single visit to each site.
12
-------
Physical Habitat and Riparian Condition — Stream habitat and riparian condition were measured
using methods provided by Kaufmann and Robison (1998). Major elements measured included
channel dimensions, channel gradient, channel substrate, habitat complexity and cover, riparian
vegetation, anthropogenic alterations, and channel riparian alterations. Measurements were taken
along the stream thalweg and 11 cross-sectional transects evenly spaced within the reach. This type
of habitat survey is laborious (requiring about 3 h per site) and was conducted at a subset of sites (n
= 67 EMAP sites and n = 104 RARE sites). A qualitative habitat survey (Barbour et al. 1999) was
also conducted at all EMAP sites and 104 RARE sites.
Basin and Reach Landscape Characteristics — Basin and reach landscape variables were compiled
from a variety of spatially extensive digital coverages. Basins were described in terms of
morphometry (e.g., size, shape, and topography) and land cover. EMAP site basins were
characterized using digital topographic coverages, Landsat Thematic Mapper (TM) data from 1991-
1993, and aerial photographs (Herlihy et al. 1998). Reach-scale attributes such as elevation and
gradient were also tabulated for survey sites. Similar watershed attributes were measured for
RARE sites. Variables, GIS databases, and methods are detailed in Detenbeck et al. (in review).
Land use for RARE basins were determined from the National Landcover Database (NLCD),
updated for surface mining.
Temperature — Temperature was collected once at EMAP and RARE sites at the time fishes were
sampled. Additionally, temperature loggers (StowAway® TidbiT®; Onset Corporation) were
deployed at RARE sites between late May and mid July and were retrieved in the fall (September-
October). Temperature was recorded hourly. Retrieved temperature data were plotted to identify
and remove any data that were recorded either prior to deployment or after removal from the
stream. Weekly maximum and minimum temperature was calculated for all sites. Weekly maxima
were averaged for all sites to demonstrate summer trends in temperature across the state. These
trends were used to identify the timing of peak summer temperatures. Presumably, this is the part
of the summer when fishes would be most stressed by high stream temperature. Henceforth, the
period is referred to as "summer". Summer temperature data were used to calculate mean weekly
temperature (the average of weekly maximum and minimum temperatures; Wehrly et al. 2003).
Based on these data streams were classified into three temperature categories (cold <19° C; cool 19-
22° C; and warm >22° C) (Wehrly et al. 2003).
Fish Assemblages
Fishes were sampled at EMAP sites using a combination of backpack electrofishing and seining
(McCormick and Hughes 1998). Fishes were sampled in a single pass and reach length was scaled
to 40 times mean stream width. Minimum and maximum lengths of 150 and 500 m, respectively,
were used. RARE sites were sampled using electric parallel wire technique (Holton and Sullivan
1954; Cincotta et al. 2001). Fish were sampled in a single pass over a reach of 160 m.
13
-------
Metric Selection and IBI Development
Database Management— The fish assemblage data collected in 1993-1996 as part of the EMAP
Mid-Atlantic Highlands Assessment (MAHA) were used to test fish assemblage metrics for
responsiveness to anthropogenic stressors (n = 45 sites plus 4 revisits). From those data, an IBI was
developed for the state of West Virginia. Data collected in 1997 and 1998 by the Mid-Atlantic
Integrated Assessment were then used to validate the metrics and IBI (n = 43 sites plus 8 revisits).
Finally, data from WVDNR's RARE sites were used to calculate IBI scores from the 2000-2001
project years. Data from the MAHA and MAIA surveys had been entered into the EMAP database
management system and subjected to data entry quality assurance, including verification of species
identification based on museum vouchers. Data from the REMAP/RARE project were entered by
WVDNR and compiled using EMAP Surface Waters Information Management (SWIM) protocols.
Ecological Attributes for Metric Development — Fish ecological characteristics (e.g., spawning
guild and tolerance level) were originally compiled for the MAHA study (McCormick et al. 2001).
Species characterizations published in McCormick et al. (2001) were based largely on descriptions
in Jenkins and Burkhead (1994), with occasional reference to Trautman (1981) and Pflieger (1975).
Several fish species that were collected during the MAIA and RARE surveys were not included in
the original MAHA database or had not been completely or correctly characterized. New
characteristics were added to the file and re-evaluated the criteria, consistency and inclusion of
several original MAHA metrics (i.e., macro-omnivores, invertivore-piscivores, tolerant, intolerant,
benthic invertivore, benthic habitat, clean substrate spawner). Taxonomic data were updated,
erroneous species identifications were corrected, and native/alien designations were revised in both
MAIA and MAHA datasets. This file (FISHCHAR) is available electronically with the
supplementary material provided with this report.
Minimal Disturbance Criteria and Disturbance Scores — Expectations for biological metrics and
indices typically are based on conditions at minimally-disturbed locations (reference sites; Hughes
1995). Eleven measures of disturbance (i.e., stressors) were used to score both reference sites and
highly-disturbed sites (Table 1). Landscape (catchment) variables included: % catchment as
agriculture; % catchment as urban; and % catchment as agriculture, urban & mined. Water
chemistry variables included: chloride, ammonia, sulfate, acid neutralizing capacity, nitrate, total
nitrogen and total phosphorus. Disturbance designations were made for each disturbance indicator
with values above the appropriate criterion level. Sites were evaluated for each stressor and scored
a 1 for exceeding minimally disturbed criteria or a 3 for exceeding the highly disturbed criteria,
except acid neutralizing capacity, for which higher values are more desirable. If sites
characteristics were below the minimum disturbance criterion, they received a disturbance score of
0 for that variable. Cumulative scores based on the designation of disturbance classifications were
determined by the sum of scores for each disturbance variable. Sites with 0-1 points were classified
as "reference", sites with 2-6 points were classified as intermediate and sites with >6 points were
classified as highly disturbed. .
Candidate Metric Evaluation and Selection — In general, the positive metrics (those expected to
increase with better conditions) excluded introduced species, whereas negative metrics included all
species. Sixty-eight candidate metrics were developed in four categories: taxonomic, trophic,
reproductive and tolerance (variable names and descriptions are presented in Appendix I). The 13
14
-------
metrics that failed the range test and the two metrics that failed the signal to noise test in
McCormick et al. (2001) were not included. The candidate metrics list included "non-tolerant"
versions (with the tolerant species removed) for most of the positive metrics.
Table 1. Disturbance classification thresholds for EMAP sites (1993-1998).
Variable
Agriculture (% basin)
Agriculture + urban + mining (% basin)
Urban (% basin)
Chloride (ueq/1)
Ammonia (ueq/1)
Nitrate (ueq/1)
Total Nitrogen (ug/1)
Total Phosphorus (ug/1)
Sulfate (ueq/1)
Acid Neutralizing Capacity (ueq/1)
Rapid Habitat (mean score)
Highly
Disturbed
45
50
3.5
900
8
100
>750
150
1,000
<50
<12
Minimally
Disturbed
15
15
0.2
< 100
2
3
<750
<20
<400
>50
>16
Candidate metrics were screened with four successive "filters" following the approaches described
in Hughes et al. (1998) and McCormick et al. (2001). In the evaluation process, each metric was
examined for its scoring range, variability, responsiveness, and redundancy. Metrics were rejected
if they failed a range test (applied only to richness metrics; rejected if raw value ranges between 0
and 2 species) or a signal to noise test (ratio<3, where signal was the variance among sites and
noise was the variance among repeat visits (Kaufmann et al. 1999). The range test was only applied
to richness metrics.
To determine if the candidate metric was responsive to human disturbance, Spearman correlations
and bivariate plots (Hughes et al. 1998) were used to test the responsiveness of the remaining
candidate metrics to physical habitat structure and water quality (pH; sulfate concentration; total
nitrogen concentration; total phosphorus concentration; chloride concentration; percent sands and
fine substrate; relative bed substrate stability; density of large woody debris; fish cover; indices of
riparian and channel disturbance; and indices of channel, riparian, and watershed quality). Metrics
were plotted against two aggregate measures of human disturbance. First, metric sensitivity was
analyzed in streams classified into different disturbance classes (i.e., reference, mixed, nutrients,
and mine) based on stream chemistry (Herlihy 1990, Davis and Scott 2000). Second, metric
response was analyzed in streams classified using Bryce Condition Class, an index that uses
watershed and local stressors (e.g., in-stream sediment and habitat, basin forest cover, etc) to
evaluate human disturbance (Bryce et al. 1999).
15
-------
Scatterplots of the metric values versus stream size (logic watershed area), were visually assessed
with sites coded by membership in the "reference" sites, highly disturbed sites, or intermediate
disturbance sites (all other sites). These comparisons were used to determine if metrics were biased
for stream size across the disturbance categories. Metrics were retained for which a majority of
reference sites had better values than did the majority of disturbed sites. Pearson's Product Moment
correlation was used to test for redundancy among metrics. Only one metric out of each correlated
pair (r >/= 0.75) was retained. All statistical analyses were conducted in PC-SAS for Windows,
release 8.02 (SAS 2001).
Adjustments for watershed area — Some metrics were correlated with watershed size. These were
normalized for a watershed size of 100 km2 following the approach described by Urquhart (1982).
The regression equation of the metrics with watershed area (logio watershed area in km2) for the
reference sites was calculated. That reference regression equation was then applied to all sites, and
their residuals were calculated. Next, the expected value for reference data at a standardized
watershed area of 100 km2 was determined, and this constant was applied to residuals. This resulted
in all observations having non-negative values.
Metric andIBI scoring — Metric scoring followed McCormick et al. (2001). Metrics were scored
on a continuous scale from 0-10 based on the distribution of scores from sites in the calibration data
set. IBI scores were calculated by taking the sum of the nine metrics scores and multiplying the
sum by 1.11 to give index scores that ranged from 0-100.
Responsiveness of the IBI— The responsiveness of the IBI to stressors was evaluated by plotting it
against chemical and physical habitat variables as well as four aggregate measures that represented
general disturbance gradients among watersheds. Sites were classified as minimally-disturbed,
intermediate, and highly disturbed using a combination of Rapid Bioassessment Habitat (RBP)
variables (Barbour et al. 1999) and chemistry variables (see McCormick et al. (2001) for details on
classification). Streams were also classified into different disturbance classes (i.e., reference, mixed,
nutrients, and mine) based on stream chemistry (Herlihy 1990, Davis and Scott 2000). A third
aggregate measure was the continuous disturbance scores derived from 11 stressors listed in Table 1
(see Methods on Minimal Disturbance Criteria). Finally, Bryce Condition Class, an index that uses
watershed and local stressors (e.g., in-stream sediment and habitat, basin forest cover, etc) was
calculated to evaluate human disturbance (Bryce et al. 1999).
Relationship between IBI and stream temperature — Relationships between IBI and stream
temperature regime were assessed using three methods. First, box plots were used to illustrate the
distribution of cold, cool, and warm water sites among reference, intermediate, and disturbed sites.
Temperature was only recorded once at these EMAP sites, so thermal categories were assigned
based on the presence or absence of cool and coldwater taxa such as sculpins and salmonids.
Second, regression analysis was used to compare RARE site IBI scores with temperature recorded
in the field at the time fishes were sampled. Finally, RARE sites were categorized as cold, cool,
and warm based on continuous summer temperature data. Box plots were used to identify trends in
metrics and IBI scores among these temperature classes.
16
-------
Results
Comparing seasonal, regional, and methodological differences among datasets — Plots of species
richness at these seasonal calibration sites for spring versus summer showed little seasonal bias
(Figure 2). Likewise, a CDF plot for the population of upland streams in the spring (MAHA) and in
the summer (MAIA) were virtually identical (Figure 3). These analyses suggest that seasonal bias
is minimal at the site (Figure 2) or regional scale (Figure 3). These lines of evidence support the use
of West Virginia MAHA data for metric screening and West Virginia MAIA data for evaluating the
responsiveness of the IBI to disturbance.
The CDFs of all MAHA and MAIA sites along with the subset of MAHA and MAIA sites from
West Virginia sites for those projects were calculated to test the assumption that the IBI of
McCormick et al. (2001) was equally representative of stream condition at the state and regional
scales. Strong similarity among the plots suggests that the spectrum of stressors and fish
assemblage responses found in the region at large were represented in West Virginia (Figure 4). As
such, the approach of McCormick et al. (2001) was applied for metric and IBI calculation of the
RARE sites.
The comparison of raw metric scores from 10 sites sampled with the parallel wire and backpack
electrofishing methods showed strong agreement between methods for seven of the nine metrics
(Figure 5). The proportion of invertivore-piscivores (higher for the backpack method) and clean
gravel spawners (higher for the parallel wire method) differed between methods but the net effect of
these differences on the resulting IBI was negligible.
IBI Development and Testing
Metric selection — After excluding the 13 metrics from McCormick et al. (2001) that failed the
range test, no additional metrics were excluded based on this criterion. Two metrics (NTROPH,
PNEST) failed the signal to noise test. The results of the redundancy and responsiveness tests are
shown in Table 2. Metrics failed the responsiveness test if they did not show clear separation
between reference and disturbed sites. The results of this analysis produced a list of metrics
identical to that of McCormick et al. (2001). With the revisions to the species list and the
modifications of the assemblage characteristics, several metrics were revised, but not significantly
changed. All of these metrics were significantly correlated with some measures of water quality
and habitat (Table 3 and supplementary tables provided in electronic format). Many of the positive
richness metrics (those expected to increase with better conditions) were positively correlated with
chemical and habitat stressor variables. This was likely due to the positive relationship between
watershed area and many stressor variables (\..\DATA\WV RARE\SAS RESULTSVFISH v
HABITAT CORRELATION MATRIX.xls). In many cases, fish richness metrics increase as a
function of stream size.
Non-tolerant versions of the clean substrate spawner and piscivore/invertivore metrics were selected
whereas McCormick et al. (2001) kept all species. The definition of non-tolerant species for the
cyprinid richness and benthic habitat species richness metrics (McCormick et al. 2001) was
modified by excluding all tolerant species rather than one or two species originally excluded.
17
-------
< 30 -
LLJ —
- CL 20 H
00
•. O
tUJ
w w
LU
or
0 -
• Intra-annual Revisits
o Inter-annual Revisits
° Spring v. Summer
o o a x a
'
a • • jr o
• o •x« o
o x
•X
• ^Q O
0T 0
A.
10 20
INDEX SITE VISIT
30
20
15 -
LU
or
5 -
0 -
• Spring 1998 vs. Summer Visit 1
O Spring 1998 vs. Summer Visit 2
r2=0.81
B.
10
SPRING 1998
15
20
Figure 2. A. Intra- and inter-annual, and seasonal comparisons of native species richness. Most of
the 68 intra-annual revisits occurred within the same season (Spring, 1993-1994; Summer, 1995-
1998). In 1995 and 1996, 16 sites sampled in the spring were revisited in the summer of 1995 and
1996. Five 1993-1994 spring sites were re-sampled during summer 1997-1998 as part of the Mid-
Atlantic Integrated Assessment (MAIA). Diagonal line represents 1:1 relationship. B. Intra-annual
comparisons of 23 sites sampled in 1998 for MAIA. The dashed line represents 1:1 relationship
and the solid line shows the linear regression for the dataset.
18
-------
01
CL
jo
3
E
3
O
100 -
80 -
60 -
40 -
20 -
0 -
20
40
60
80
100
IBI Score
Figure 3. Cumulative distribution functions (CDFs) of index of biotic integrity (IBI) scores for the
Mid-Atlantic Highlands Assessment (MAHA) and the Mid-Atlantic Integrated Assessment
(MAIA). Indices were calculated using the IBI of McCormick et al. (2001).
100 -
80 -
-------
20
16
12
8
i
LU
Q^ 0.8 -
5
y 0.4 -
— !
o:
Q_ 0.0
0.
0.
0.
NUMBER NON-TOLERANT * , ' '
BENTHIC SPECIES
.••
,-'
..-' •
*•'
•
,• •
• ,•'
I 8 12 16 2
PROP. OMNIVORES
.-•''
,..--'
f' m
• •-•"'.
.-•''
..-'
0.0 0.4 0.8
PROP. TOLERANTS
• t'
' .-•''' '
• *•'
.f' •
..*'''
8
4 -
0
0.4
0.2
0.0
0.
0.
0.
NUMBER OF INTOLERANT TAXA
.-•'
.•' 9
..•''
,.••'
• ,•' •
f'
.*
*'
4 8
PROP. NON-TOLERANT
INVERTIVORE-PISCIVORES
,.-'
,-'
.'.'""" *
3.0 0.2 0.4 C
PROP. COTTIDS .
,-'
,-'
_.- '
%^*''
...-''
16 '
12 '
8
44
0.8
0.4
0 0
.6
0.3
0.2
0.0
NUMBER NON-TOLERANT
CYPRINID SPECIES . .-•'
f-'
• • ,•'
.
• ,.«
f'f
^
.*
8 12 16
PROP.CLEAN GRAVEL • .-''
• .* .
•
,.-'
,.''
.-'
0.0 0.4 0.8
PROP. EXOTICS . ,.'''
,.-'
,•'•'
,--''
.,.-•''
,4'. .
0. 0. 0.
BACKPACK SHOCKER
Figure 5. Comparison of metric results from EMAP backpack and WVDNR parallel wire
electrofishing methods. The dashed line represents the 1:1 relationship. The proportion of non-
tolerant invertivore-piscivores and the proportion of non-tolerant clean gravel spawning individuals
show poor agreement compared with other metrics.
20
-------
Table 2. Results of metric evaluation process. Metric definitions are provided in Appendix I.
Original metrics refer to those selected by McCormick et al. (2001) for the MAHA IBI.
Failed responsiveness test Failed redundancy test
Original IBI
metrics
Retained metrics for WV
NUMNATSP
NFAM
NSDART
NSICTA
NSCENT_NONTOL
NSCATO2
NSCATO_NONTOL
NSBENT_HAB
NSCOLU
NSBENTJNV
NSBENT_INV_NONTOL
NSEXOT
NSTOLE
NSCLNSPWNR
NSCLNSPWNR_NONTOL
NREPROS
PCATO
PCATO_NONTOL
PCENT
PCENT_NONTOL
PBCST
PATNG
PNTGU
PCGBU
PCLNSPWNR
PCOLD2
PMICRO
PMICRO2
PBENTJNV
PCARN
PHERB
PINVERT
POMNI H
PCOLD
PBCLN
PTREPRO
NSP
NUMSPEC
NSCENT
NSCATO
PCYPR
PCYPTL
PINTOL
PPISC
PBENT_INV_NONTOL
PFISHBUG
PPISCINV
NSCYPR2 NSCYPR_NONTOL
NSBENT2 NSBENT_HAB_NONTOL
NSINTOL NSINTOL
PGRAVEL PCLNSPWNR_NONTOL
PCOTTID PCOTTID
PTOLE PTOLE
PMACRO PMACRO
PPISCINV2 PPISCINV_NONTOL
PEXOT PEXOT
21
-------
Table 3. Spearman rank correlation coefficients (r > 0.2; P <0.01) between EMAP fish assemblage metric variables and chemical and
physical habitat variables. The data are from the MAHA survey (n = 45 sites for chemical and land cover data; n = 26 sites for
physical habitat data. Full table of correlations offish assemblage metrics with physical habitat variables is incorporated as a
hyperlink to an Excel file to conserve space. Stressor acronym definitions are provided in Appendix III.
NUMNATSP
NATNSP
NUMSPEC
NSP
NSEXOT
NATIVFAM
NSCATO
NSCATO NONTOL
NSSUCK
NSCENT
NSCENT NONTOL
NSBASS
NSMINN
NSCYPR
NSCYPR NONTOL
NSCYPR2
NSDART
NSICTA
NSBENT2
NSBHAB
NSBENT HAB
NSBENT HAB NONTOL
NSCOLU
NSINTOL
NSSENS
NSTOLE
NTROPH
NSBENT INV
NSBENT INV NONTOL
NSCLNSPWNR
NSCLNSPWNR NONTOL
NREPROS
PCATO
PCATO NONTOL
PCENT
NO3
-0.26
-0.45
-0.45
-0.39
-0.38
NTL
-0.39
-0.37
-0.34
-0.35
PTL
-0.35
0.36
-0.26
-0.30
TSS
-0.39
-0.41
-0.42
-0.35
-0.31
-0.35
-0.58
-0.37
-0.34
-0.42
TURB
-0.42
-0.38
-0.38
AG
TOT
0.48
0.36
0.32
0.33
0.33
-0.39
-0.38
0.38
0.36
0.30
0.34
0.35
MINE
TOT
0.33
0.30
-0.26
URB
TOT
0.32
0.26
0.28
0.25
0.34
0.38
0.44
0.31
0.35
0.37
0.36
0.29
0.41
0.28
DIS-
TOT
0.47
0.36
0.31
0.28
0.33
-0.42
-0.37
0.38
0.36
0.33
0.28
0.41
0.36
PCT GF
-0.28
-0.31
-0.42
-0.27
-0.29
-0.38
PCT SA
0.29
0.28
0.29
0.28
0.31
0.36
0.34
PCT SAFN
-0.50
-0.30
0.39
XEMBE
D
0.36
0.27
0.31
0.39
-0.50
0.32
0.34
0.41
0.39
0.34
0.38
0.40
W1 HALL
0.42
0.38
0.33
0.43
0.31
0.39
0.42
0.33
0.34
0.40
0.41
0.45
0.40
0.39
0.43
22
-------
PCENT NONTOL
PCOTTID
PCYPR
PCYPR NONTOL
PATNG
PBCLN
PBCST
PCGBU
PNEST
PNTGU
PBENTSP
PGRAVEL
PCLNSPWNR
PCLNSPWNR NONTOL
PCOLD1
PCOLD2
PHIDE
PHIDE NONTOL
PEXOT
PNIS
PINTOL
PTOLE
PTREPRO
PBENT
PBENT INV
PBENT INV NONTOL
PCARN
PHERB
PINSE
PINVERT
PMACOMNI
PMACRO
PMICRO
PMICRO2
POMNI H
PPISC
PPISCINV
PPISCINV NONTOL
PPISCINV2
N03
-0.36
0.33
-0.30
-0.36
0.34
NTL
-0.32
PTL
0.29
0.36
0.39
TSS
0.37
-0.58
TURB
-0.41
AG
TOT
0.36
0.25
0.38
MINE
TOT
0.26
0.27
0.27
0.38
URB
TOT
0.34
-0.12
0.34
0.45
0.37
DIS-
TOT
0.39
0.37
0.26
PCT GF
-0.29
0.34
-0.33
0.37
0.37
-0.28
-0.31
-0.39
-0.32
PCT SA
0.32
0.31
0.27
PCT SAFN
0.34
0.29
0.35
0.32
XEMBE
D
0.33
0.31
0.39
W1 HALL
0.33
0.38
0.37
0.32
0.34
0.29
0.39
0.34
..\..\DATA\WV RARE\SAS RESULTSVFISH v HABITAT CORRELATION MATRIX.xls
23
-------
Metric andIBI Scoring — Raw scores for richness metrics (intolerant, cyprinid, and benthic habitat
species) increased with watershed size and required calibration. Raw scores for the proportion
metrics were unrelated to watershed size. These metrics were scored somewhat differently than
McCormick et al. (2001; Table 4). Positive metrics were scored based on the 50th percentile of
reference sites and 10th percentile of disturbed sites. Negative metrics were scored based on the 50th
percentile of reference site and 90th percentile of disturbed sites. The numbers values were rounded
to the nearest 10 to simplify scoring. For example, the 50th percentile of % Cottid individuals was
7% and was rounded to 10% for metric scoring. The scored metrics were plotted against watershed
size and observed no association with size for any metric. Scored metrics were summed and
calibrated to a range of 0 - 100, with the following exceptions. The IBI was set to 0 for sites with
watershed areas >2 km2 and < 10 individuals; the IBI was not calculated for sites with watershed
areas < 2 km2 and <10 individuals (McCormick et al. (2001).
Table 4. Revised West Virginia (WV) Streams IBI metric scoring criteria. UPPER = data
value expected for assemblage in good ecological condition; values >UPPER were given
metric score of 10. LOWER = data value expected for assemblage in poor condition; values <
LOWER were given metric score of 0. For negative metrics, values < UPPER were scored as
10; value > LWER were scored 0. Data values between LOWER and UPPER were rescaled to
a range of 0 to 10. Richness metrics were calibrated to watershed area (LWSAREA =
logio(watershed area (km2) +1)) based on data from the entire MAHA region (McCormick et
al. 2001). Scored metric correlations with WV IBI scores in right column.
Metric UPPER LOWER r
% Cottid Individuals 10 0 0.39
% Non-tolerant Clean Spawner 40 0 0.69
Individuals
% Macro-omnivore Individuals 0 20 0.29
(Negative)
% Tolerant Individuals 30 100 0.73
(Negative)
% Non-native Individuals 0 10 0.24
(Negative)
Intolerant Species 1+(3*LWSAREA) 0 0.74
Cyprinid Species 1+(4.5*LWSAREA) 0 0.55
% Non-tolerant Piscivore/Invertivore 50 0 0.18
Individuals
Non-tolerant Benthic Habitat Species 5.5*LWSAREA 0 0.69
24
-------
Responsiveness oflBI to disturbance gradients — The revised IBI was sensitive to four measures of
watershed condition. Minimally-disturbed sites had higher IBI scores than sites with intermediate
or high levels of disturbance (Figure 6a.). Sites with minimal chemical disturbance had higher IBI
scores than sites with mixed anthropogenic impacts, or sites with chemistry profiles indicating
agricultural or mining impacts (Figure 6b.). IBI was negatively correlated with disturbance score,
indicating that IBI declined with increasing disturbance (Figure 6c.). Likewise, the IBI showed
predictable declines in response to Bryce condition class. Those sites that showed the most
cumulative human impacts had the lowest scores (Figure 6d.). Disturbed sites tended to have more
variable IBI scores than reference sites (Figure 6b and 6d).
The West Virginia IBI was not correlated with watershed size. In general, reference sites had
higher scores than intermediate and highly disturbed sites (Bryce et al. 1999). The univariate
distributions of reference site IBI scores were very similar for all three methods of selecting
reference sites. The IBI was responsive to catchment and riparian disturbance, sedimentation and
nutrients (Appendix II).
IBI scoring Criteria — The approach described in McCormick et al. (2001) was used to set
narrative criteria based on the IBI. IBI scores exceeding the 75th percentile for the reference sites
(IBI>81) were classified as having excellent biotic integrity. Scores between the 75th and 25th
percentiles (70 < IBI < 81) were identified as having good biotic integrity. Scores between the 5th
and 25th percentiles (56 < IBI < 70) were described as being in fair condition and sites with scores
below the 5th percentile were judged to be poor condition.
Comparisons of IBI performance across thermal regimes
Temperature classification of RARE sites — Data for 84 of the 119 RARE streams (71%) were
analyzed. Sites were excluded from the analysis because data loggers were lost, failed, or returned
abnormally high temperatures (i.e., > 32° C) indicating that the instrument was not properly
submerged throughout the deployment. Plots of mean maximum temperature indicated a peak in
summer temperature in the weeks around July 23rd (Figure 7A). The weeks of July 16-Aug 13 were
selected to represent "summer" because this represented the longest continuous block of
significantly warmer weekly temperatures and because weekly data for all 84 sites were available
for this period. Based on these summer data, most sites were warm water (n = 54; Figure 7B).
Only four of the 84 sites met the criteria for cold water streams.
25
-------
100 -
80 -
£ 60 -
o
m 40 -
20 -
0 -
•
±
-^-
i
j
,
1
>
L
a
100 -
80 •
-------
"ro
CD
ro
(D
35
30
25
20
15
O
a a
a a
I L
J L
6-25 7-02 7-09 7-16 7-23 7-30 8-06 8-13 8-20 8-27 9-03 9-10
Date
15
RARE Sites
Figure 7. (A) Means of weekly maximum temperatures for all streams with temperature loggers, (n
= 84 sites total; n = 986 total weekly observations). Deployment and retrieval of temperature
loggers was staggered throughout the season, so observations for a given week vary from 33-44
sites. Temperature data are from 2001 (41 sites) and 2002 (43 sites). Bars represent a 95%
confidence interval on the observations and weeks designated with "a" are significantly warmer
than other weeks (p < 0.05). (B) Mean summer temperature. Means are calculated from weekly
maximum and minimum temperatures for the weeks of July 16- Aug 12. Lines indicate temperature
class designations for streams in Michigan (Wehrly et al. 2003). Based on these criteria, n = 4 cold,
26 cool, and 54 warm streams.
27
-------
100 -
80 -
g 60-
O
O
CO
m 40 -
20 -
0_
J ,_J
* ;
A
0 ;
I
i
O _j
\
r
3
1
r
• Coldwater
O Coolwater
A Warmwater
*
i
0
i
o
CJ3
Highly Disturbed Intermediate Min
imally Disturbed
Figure 8. Response of index of biotic integrity to WV disturbance condition classifications with
sites plotted according to their temperature class. Data are from EMAP surveys (n = 88 sites).
Relationships between IBI and stream temperature — The EMAP sites were categorized into cold,
cool, and warm water streams based on the published temperature preferences of resident fishes.
Based on these classifications, fewer reference sites were classified as warm water compared with
cold and cool water (Figure 8). None of the disturbed sites were cold water. Most of the disturbed
sites were warm, although cool sites were present in the intermediate and highly disturbed streams.
These trends were supported by the CDFs of IBI scores for these sites (Figure 9A). Cold water sites
consistently scored "fair" or better (i.e., IBI > 56), whereas cool and warm streams reflected a
broader range of IBI scores and had numerous sites in the "poor" category (i.e., IBI < 55). The
CDF of IBI scores for RARE sites differed from that of MAIA sites (Figure 9B). In general, RARE
sites scored higher and had a narrower range than MAIA sites. Cool sites showed a broad range of
scores similar to those at MAIA sites; however, warm sites all scored high with one exception.
Metric and IBI performance among temperature classes for RARE sites were directly compared
(Figure 10). The small sample of cold streams (n = 4) limited identification of trends at these sites.
One cold stream was not included in this analysis because no fish were collected at this site,
presumably due to acid mine drainage (stream pH = 3.9). The only two metrics that increased from
cool to warm streams were NSBEN_HAB_ NONTOL (number of nontolerant benthic habitat
specialists) and NSCYP_NONTOL (number of nontolerant cyprinids). None of the other metrics
showed a significant trend and the overall IBI score did not differ among temperature categories.
This result was corroborated by a regression analysis of IBI scores and field temperature (recorded
at the time offish sampling) (Figure 11). The plot illustrates no relationship between temperature
and IBI score (r2 = 0.01).
28
-------
Q
O
100 -
80 -
60 -
40 -
20 -
0 -
A
• COLDWATER
O COOLWATER
A WARMWATER
T3
o
o
CD
"0
O
X
LU
20
40
60
80
100
IBI
Q
O
100 H
80 -
60 -
40 -
20 -
0 -
B
• COLDWATER
O COOLWATER
A WARMWATER
40
60
80
100
IBI Score
Figure 9. A. Cumulative distribution function (CDF) of index of biotic integrity scores for West
Virginia from EMAP surveys (n = 69). Points are coded for stream temperature classification based
on fish assemblage characteristics. B. CDF of index of biotic integrity scores for West Virginia
RARE project (n = 84). Sites are coded for stream temperature based on data from temperature
loggers. Red vertical lines show categories for narrative IBI criteria.
29
-------
ffi 12-
a
d
fe °4
T
V
i
T
g 0,5-
O 010 -
I
±
I OB-
Q
Z UJ
£ I 06
Z 51 °4
O li
z o;
^ 9 n,
1 0-
• LJ
9 >
• i w °8-
T l|
til 06-
!E9
£ CD
JH z 04-
V ii
I r M 02-
-1- • o
• • £
•
T
i
i
•
t
i
I
T
1
•
Figure 10. Box plots of IB I and component metrics for cold (n = 3), cool (n = 26) and warm (n
54) RARE streams. Metric acronyms are defined in Appendix I.
30
-------
100 -,
90 -
80 -
or
o
o eo -
m
so -
40 -
30 -
20
= 0.01
10 15 20 25
FIELD TEMPERATURE (C)
30
35
Figure 11. Plot of IBI and temperature for RARE sites. Temperature was recorded in the field
when fishes were sampled.
Discussion
The recent MAHA, MAIA and RARE surveys have contributed a wealth of physical, biological and
chemical data for West Virginia streams. These separate studies provide independent datasets that
can be used to develop, test, and validate indices of biotic integrity or related hypotheses on
anthropogenic alteration of stream communities. However, comprehensive analysis of the datasets
was challenging due to differences in fish sampling methods. The analysis provided two lines of
evidence supporting the use offish data among datasets without normalizing for effects of temporal
variability or sampling methodology. Seasonal, intra-annual, and interannual variation within sites
was low. In addition, only marginal differences in selected metrics were observed between
backpack electrofishing and the parallel wire technique. Because these differences were minor, the
MAHA data were used to develop the IBI, the MAIA data was used to test IBI sensitivity to
disturbance, and the MAIA and RARE data were used to assess IBI applicability across thermal
regimes.
The IBI developed by McCormick et al. (2001), was used to demonstrate that IBI scores from
EMAP studies in West Virginia mirrored those from the larger Mid-Atlantic region. Because the
West Virginia sites showed a similar range of biotic condition to those in the Mid-Atlantic region,
11
-------
the metric selection methods of McCormick et al. (2001) were applied to the West Virginia MAHA
data. Not surprisingly, the same metrics selected for the larger region were found to be sensitive to
disturbance at the state-scale. Some metrics were modified slightly to improve their sensitivity to
disturbance (e.g., tolerant taxa excluded from positive metrics). Tests of the IBI using the MAIA
data showed that the IBI was sensitive to anthropogenic disturbance. IBI scores were negatively
correlated with a number of disturbance variables measuring cumulative impacts to streams. These
measures included a watershed disturbance class, disturbance score, condition class, and land use
classification.
McCormick et al. (2001) did not determine if their IBI was equally applicable in cold, cool and
warm water streams. A potential source of bias related to thermal regime is that cold water streams
naturally have lower diversity than warm water streams. Thus, low scores in cold water streams
would result from natural conditions rather than human disturbance. West Virginia IBI scores for
cold, cool, and warm water streams from the MAIA dataset were compared and showed no bias for
low scores in the cold and cool streams. Cold water streams were consistently classified as having
"fair" or better condition. Likewise, cool streams showed a normal distribution of IBI scores from
low to high.
The RARE dataset was also analyzed for relationships between IBI score and stream temperature.
This analysis was hampered by two limitations of the RARE dataset. First, the number of cold
water streams was too low (n = 4) to draw any conclusions for these streams. Second, IBI scores
for RARE sites were skewed toward high scores, so the gradient of biotic condition was shorter
than that observed for MAHA and MAIA sites. Given these shortcomings, warm sites did tend to
score high, but that the scores of cool sites were distributed evenly from "poor" to "good". Direct
comparisons of cool and warm streams showed no difference between most metrics or cumulative
IBI scores. In addition, IBI scores from RARE sites were unrelated to stream temperature recorded
at the time offish sampling.
References
Barbour, M. T., J. Gerritsen, B. D. Snyder, and J. B. Stribling. 1999. Rapid Bioassessment Protocols
for Use in Wadeable Streams and Rivers: Periphyton, Benthic Macroinvertebrates, and Fish,
Second Edition. EPA 841-B-99-002, Office of Water, Washington, D.C.
Bryce, S. A., D. P. Larsen, R. M. Hughes, and P. R. Kaufmann. 1999. Assessing relative risks to
aquatic ecosystems: a Mid-Appalachian case study. Journal of the American Water Resources
Association 35:23-36.
Cincotta, D., F. Fulk, andN. E. Detenbeck. 2001. A Small Watershed Characterization,
Classification, and Assessment for West Virginia Utilizing EMAP Design and Tools. Quality
Assurance/Quality Control Plan. U.S. Environmental Protection Agency, National Exposure
Research Laboratory, Cincinnati, OH.
32
-------
Davis, W. S., and M. Scott. 2000. Mid-Atlantic Highlands Streams Assessment: technical support
document. EPA/903/B-00/004. U. S. Environmental Protection Agency. Ft. Meade, Maryland.
Detenbeck, N. E., L. A. Jagger, S. L. Stark, M. A. Starry. WV REMAP Final Report: Watershed
Classification Framework for the State of West Virginia. Washington, DC, 600/R-03/141.
Fausch, K. D., J. R. Karr, and P. R. Yant. 1984. Regional applications of an index of biotic integrity
based on stream fish communities. Transactions of the American Fisheries Society 113:39-55.
Fausch, K. D., J. Lyons, J. R. Karr, and P. L. Angermeier. 1990. Fish communities as indicators of
environmental degradation. Pages 123-144 in S. M. Adams, editor. Biological indicators of stress
in fish. American Fisheries Society, Symposium 8, Bethesda, Maryland.
Herlihy, A. T., P. R. Kaufmann, and M. E. Mitch. 1990. Regional estimates of acidic mine drainage
impact on streams in the Mid-Atlantic and Southeastern United States. Water, Air and Soil
Pollution 50:91-107.
Herlihy, A. T., P. R. Kaufmann, M. R. Church, P. J. Wigington, Jr., J. R. Webb, and M. J. Sale.
1993. The effects of acidic deposition on streams in the Appalachian Mountain and Piedmont
region of the mid-Atlantic United States. Water Resources Research 29:2687-2703.
Herlihy, A. T., J. C. Stoddard, and C. B. Johnson. 1998. The relationship between stream
chemistry and watershed land use in the mid-Atlantic region, U.S. Water, Air and Soil Pollution
105:377-386.
Herlihy, A. T., D. P. Larsen, S. G. Paulsen, N. S. Urquhart, and B. J. Rosenbaum. 2000. Designing a
spatially balanced, randomized site selection process for regional stream surveys: the EMAP Mid-
Atlantic pilot study. Environmental Monitoring and Assessment 63:95-113.
Holton, G.D., and C. R. Sullivan. 1954. West Virginia's electrical fish-collecting methods.
Progressive Fish Culturist 16(1): 10-18.
Hughes, R. M. 1995. Defining acceptable biological status by comparing with reference conditions.
Pp. 31-47 in W. S. Davis and T. P. Simon (eds.) Biological assessment and criteria: tools for water
resource planning and decision making. CRC Press. Boca Raton, Florida.
Hughes, R. M., P. R. Kaufmann, A. T. Herlihy, T. M. Kincaid, L. Reynolds, and D. P. Larsen. 1998.
A process for developing and evaluating indices offish assemblage integrity. Canadian Journal of
Fisheries and Aquatic Sciences 55:1618-1631.
Jenkins, R. E. & N. M. Burkhead. 1994. Freshwater Fishes of Virginia. American Fisheries Society,
Bethesda, Maryland.
33
-------
Jones, K.B., K.H. Ritters, J.D. Wickham, R.D. Tankersly, R.V. O'Niell, DJ. Chaloud, E.R. Smith
and A.C. Neale. 1997. An Ecological Assessment of the United States Mid-Atlantic Region: A
Landscape Atlas. U.S. Environmental Protection Agency, Washington, D.C. EPA/600/R-97/130.
Karr, J. R. 1981. Assessment of biotic integrity using fish communities. Fisheries 6(6):21-27.
Karr, J. R., and D. R. Dudley. 1981. Ecological perspectives on water quality goals. Environmental
Management 5:55-68.
Karr, J. R., K. D. Fausch, P. L. Angermeier, P.R. Yant, and IJ. Schlosser. 1986. Assessing
Biological Integrity in Running Waters: a Method and its Rationale. Illinois Natural History Survey
Special Publications No. 5, Champaign, 111.
Kaufmann, P. R. and E. G. Robison. 1998. Physical Habitat Characterization. Pp. 77-118 in J. M.
Lazorchak, D. J. Klemm, and D. V. Peck, editors. Environmental Monitoring and Assessment
Program - Surface Waters: Field Operations and Methods for Measuring the Ecological Condition
of Wadeable Streams. EPA/620/R-94/004F.
Kaufmann, P.R., P. Levine, E.G. Robison, C. Seeliger, and D.V. Peck. 1999. Quantifying physical
habitat in wadeable streams. U.S. Environmental Protection Agency, Research Triangle Park, NC.
EPA620/R-99/003.
Kazyak, P.F. 1994. Maryland Biological Stream Survey Sampling Manual. Prepared by Versar,
Inc., Columbia, MD, for Maryland Department of Natural Resources, Chesapeake Bay Research
and Monitoring Division.
Lazorchak, J. M., D. J. Klemm, and D. V. Peck, editors. 1998. Environmental Monitoring and
Assessment Program - Surface Waters: Field Operations and Methods for Measuring the Ecological
Condition of Wadeable Streams. EPA/620/R-94/004F. U.S. Environmental Protection Agency,
Washington, D.C.
Leonard, P. M., and D. J. Orth. 1986. Application and testing of an index of biotic integrity in
small, coolwater streams. Transactions of the American Fisheries Society 115:401-414.
Lyons, J., L. Wang, and T. D. Simonson. 1996. Development and validation of an index of biotic
integrity for coldwater streams in Wisconsin. North American Journal of Fisheries Management
16:241-256.
McCormick, F. H. and R.M.Hughes. 1998. Aquatic vertebrate indicator. 1998. Pp. 159-180 in J.
M. Lazorchak, D. J. Klemm, and D. V. Peck, editors. Environmental Monitoring and Assessment
Program - Surface Waters: Field Operations and Methods for Measuring the Ecological Condition
of Wadeable Streams. EPA/620/R-94/004F.
34
-------
McCormick, F. H., R. M. Hughes, P. R. Kaufmann, D. V. Peck, J. L. Stoddard, and A. T. Herlihy.
2001. Development of an index of biotic integrity for the Mid-Atlantic Highlands region.
Transactions of the American Fisheries Society 130:857-877.
Miller, D. L., P.M. Leonard, R.M. Hughes, J.R. Karr, P.B. Moyle, L.H. Schrader, B.A. Thompson,
R.A. Daniels, K.D. Fausch, G.A. Fitzhugh, J.R. Gammon, D.B. Halliwell, P.L. Angermeier & DJ.
Orth. 1988.1988. Regional applications of an Index of Biotic Integrity for use in water resource
management. Fisheries 13(5): 12-20.
Omernik, J. M. 1987. Ecoregions of the conterminous United States. Annals of the Association of
American Geographers 77:118-125.
Overton, W. S., D. L. Stevens, and D. White. 1991. Design Report for EMAP, Environmental
Monitoring and Assessment Program. EPA/600/3-91/053. U.S. Environmental Protection Agency,
Corvallis, Oregon.
Pflieger, W. L. 1975. The Fishes of Missouri. Missouri Department of Conservation. Jefferson City,
Missouri.
Plafkin, J. L., M. T. Barbour, K. D. Porter, S. K. Gross, and R. M. Hughes. 1989. Rapid
Bioassessment Protocols for Use in Streams and Rivers: Benthic Macroinvertebrates and Fish. U.S.
EPA Office of Water, EPA/440/4-89/001, Washington, D.C.
Raitz, K.B., Richard Ulack, and T. R. Leinbach. 1984. Appalachia: A Regional Geography Land,
People, and Development. University of Kentucky Press, Lexington, KY.
Roth, N., M. Southerland, J. Chaillou, R. Klauda, P. Kazyak, S. Stranko, S. Weisberg, L. Hall Jr.,
and R. Morgan III. 1998. Maryland Biological Stream Survey: development of a fish index of biotic
integrity. Environmental Monitoring and Assessment 51:89-106.
SAS Institute, Cary, NC. Software applications. 2001.
Scott, M. and L. Hall. 1997. Fish assemblages as indicators of environmental degradation in
Maryland coastal plain streams. Transactions of the American Fisheries Society 126:349-360.
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,
editors. Biological Assessment and Criteria: Tools for Water Resource Planning and Decision
Making. CRC Press, Boca Raton, Florida, USA.
Smogor, R. A., and P. L. Angermeier. 1999. Relations between fish metrics and measures of
anthropogenic disturbance in three IBI regions in Virginia. Pages 585-610 in T. P. Simon, editor.
35
-------
Assessing the Sustainability and Biological Integrity of Water Resources Using Fish Communities.
CRC Press, New York, New York.
Trautman, M. B. 1981. The Fishes of Ohio. Ohio State University Press. Columbus, Ohio.
Urquhart, N. S. 1982. Adjustment in covariance when one factor affects the covariate. Biometrics
38:651-660.
United States Department of Agriculture (USD A). 1996. America's Private Land, a Geography of
Hope. USD A, National Resources Conservation Service.
USEPA. 1995. Stream fisheries impacted by acid mine drainage in Maryland, Ohio, Pennsylvania,
Virginia and West Virginia. http://www. epa.gov/reg3giss/datalib/metamd95. txt.
USEPA. 2002. Consolidated Assessment and Listing Methodology: Toward a Compendium of
Best Practices. Office of Wetlands, Oceans, and Watersheds. Washington, DC.
http://www.epa.gov/owow/monitoring/calm/calm_contents.pdf.
Wehrly, K. E., M. J. Wiley, and P. W. Seelbach. 2003. Classifying regional variation in thermal
regime based on stream fish community patterns. Transactions of the American Fisheries Society
132(l):18-38.
Whitney, G. C. 1994. From Coastal Wilderness to Fruited Plain. Cambridge University Press.
Cambridge.
Woods, A.J., J.M. Omernik, D.D. Brown, and C.W. Kiilsgaard. 1996. Level III and IV Ecoregions
of Pennsylvania and the Blue Ridge Mountains, the Ridge and Valley, and Central Appalachians of
Virginia, West Virginia, and Maryland. EPA/600/R-96/077. U.S. Environmental Protection
Agency, National Health and Environmental Effects Research Laboratory, Corvallis, OR. 50p.
36
-------
Appendix I. Candidate metrics with descriptions, ranges, means and standard deviations. Metrics
in boldface represent original McCormick et al. (2001) variables (some names were changed for
analysis as indicated). Italics designate revised metrics adopted here.
Metric
NSP
NUMSPEC
NUMNATSP
NFAM
NSCENT NONTOL
NSCENT
NSCATO
NSCATO2
NSCATO NONTOL
NSCYPR
NSCYPR NONTOL
NSDART
NSICTA
NSBENT HAB
NSBENT2
NSBENT HAB NONTOL
NSCOLU
NSBENT INV
NSBENT INV NONTOL
NSINTOL
NSSENS
NSTOLE
NSCLNSPWNR
NSCLNSPWNR NONTOL
NSEXOT
NTROPH
NREPROS
PCATO
PCATO NONTOL
PCENT
PCENT NONTOL
PCOTTID
PCYPR
PCYPR NONTOL
PCYPTL
PATNG
PBCLN
Description
No. of species including unknowns
No. of species
No. of native species
No. of native families
No. of non-tol. Centrarchids
No. of Centrarchids
No. of sucker species
No. of sucker species
No white suckers
No. of non-tolerant sucker species
No. of cyprinid species
No. of non-tolerant cyprinid species
No. of darter species
No. of ictalurid species
No. of benthic habitat specialist species
No. of benthic habitat species
excluding blacknose dace
No. of non-tolerant benthic habitat
specialists
No. of pelagic species
No. of benthic invertivore species
No. of non-tolerant benthic invertivore
species
No. of sensitive species (larger list)
No. of sensitive species (original
McCormick et al. list = NSINTOL)
No. of tolerant species
No. of clean gravel spawning species
No. of non-tolerant NSCLNSPWNR
No. of non-indigenous species
No. of trophic guilds
No. of reproductive guilds
Prop, of sucker indiv.
Prop, of non-tolerant sucker indiv.
Prop, of Centrarchids
Prop, of non-tolerant Centrarchids
Prop, ofsculpins
Prop, of cyprinids
Prop, of non-tolerant cyprinids
Prop, of tolerant cyprinids
Prop, of attacher non-guarders
Prop, of clean gravel broadcast
Range
33
32
32
7
4
6
4
3
3
12
9
11
3
19
16
16
14
20
18
11
6
9
15
15
4
6
4
0.37
0.37
0.90
0.33
0.72
1.00
0.67
1.00
0.53
0.60
MEAN
10.70
10.60
9.97
3.54
1.22
1.70
1.10
0.63
0.73
4.42
2.75
1.96
0.19
4.99
3.70
3.09
4.99
4.82
3.46
3.03
1.18
3.69
3.90
3.24
0.52
3.79
2.66
0.05
0.03
0.08
0.05
0.06
0.61
0.22
0.43
0.08
0.09
STD
7.68
7.61
7.38
1.68
1.40
1.83
1.05
0.71
0.88
2.91
2.36
2.40
0.53
3.94
3.37
127
3.91
4.11
3.62
2.94
1.58
2.17
3.29
3.22
0.80
1.61
1.37
0.07
0.06
0.16
0.07
0.13
0.30
0.19
0.34
0.12
0.12
37
-------
Metric
Description
Range
MEAN
STD
spawners
PBCST
PCGBU
PNEST
PNTGU
PGRAVEL
PCLNSPWNR
PCLNSPWNR NONTOL
PCOLD1
PCOLD2
PNIS
PEXOT
PINTOL
PTOLE
PTREPRO
PBENT INV
PBENT INV NONTOL
PCARN
PHERB
PINVERT
PMACOMNI
PMACRO
PMICRO
PMICRO2
POMNI H
PPISC
PFISHBUG
PPISCINV
PPISCINV2
PPISCINV NONTOL
NUMFISH
Prop, of broadcast spawners (tolerant)
Prop, of clean gravel buriers
Prop, of nest building indiv.
Prop, of nestguarders
Prop, of gravel spawners
Prop, of clean gravel spawners
Prop, ofnon-tol. clean gravel spawners
Prop, of coldwater indiv
Prop, of coolwater indiv
Prop, of non-indigenous species
(original McCormick et al list =
PEXOT)
Prop, of non-indigenous species
(revised list)
Prop, of sensitive indiv.
Prop, of tolerant indiv.
Prop, of tolerant reproductive indiv.
Prop, of benthic invertivores
Prop, of non-tolerant benthic
invertivores
Prop, of carnivores
Prop, of herbivores
Prop, of all invertivores
Prop, of macro-omnivores (original
McCormick et al list = PMACRO)
Prop, of macro-omnivores (revised list)
Prop, of micro-omnivores
Prop, of micro-omnivores (revised list)
Prop, of omniv. + herbiv.
Prop, of piscivores
Prop, of invertivore-piscivores
Prop, of invertivore-piscivores (revised
list)
Prop, of invertivore piscivores
excluding creek chub
Prop, ofnon-tol. Invertivore-piscivores
Number of indiv.
0.13
1.00
0.34
1.00
1.00
1.00
0.87
0.87
0.87
0.60
0.60
1.00
LOO
1.00
1.00
0.81
1.00
0.50
0.95
0.35
0.35
1.00
1.00
1.00
0.17
1.00
1.00
0.87
0.87
1718
0.01
0.35
0.07
0.39
0.59
0.44
0.26
0.05
0.05
0.05
0.05
0.22
0.51
0.40
0.45
0.24
0.27
0.07
0.40
0.03
0.03
0.22
0.22
0.31
0.01
0.26
0.28
0.10
0.09
236.25
0.03
0.27
0.09
0.24
0.23
0.26
0.25
0.18
0.18
0.12
0.12
0.26
0.34
0.24
0.25
0.20
0.26
0.10
0.28
0.06
0.07
0.23
0.23
0.24
0.03
0.26
0.25
0.18
0.18
271.85
38
-------
Appendix II. Spearman's correlation coefficients (r > 0.2; p<0.05) showing response of WV IBI
score to chemistry, physical habitat, and land cover variables sampled as part of EMAP (n = 69
sites). Stressor variable acronyms are defined in Appendix III.
Stressor Variable Spearman's r
AG_TOT -0.2
DISTOT -0.2
FOR_TOT 0.2
MINE_TOT -0.26
NONRES -0.33
TOT_RD 0.29
HOUSINGDENSJCM -0.48
_TOT -0.25
ALTD -0.22
DOC -0.31
MN -0.39
NH4 -0.21
NIL -0.2
PCT_FN -0.31
W1_HALL 0.26
QR1 -0.25
QRDIST1 -0.35
QRPHALT2 -0.25
39
-------
Appendix III. List of stressor variables compiled for WV stream sites.
Acronym
Description
ACID CLS
AG TOT
ALKCALC
ALTD
AMD_CLS
AMDCLS
AMDCLS2
ANC
AREA_WS
AREASUM
AREASUMC
AREAWSHA
ASPCTDEG
bank_veg
BAR TOT
CA
chan_alt
chan_fls
chan_sin
CHL
CHL_M2
CHL_MASS
CL
cnd_bank
COS
COND
CONDTION
CROWS D
DIC
DIST_CLS
DISTOT
DOC
ECOREG
ECOREGL4
ELEV
elevmax
ELEVMEAN
ELEVMIN
embedded
Acid Dep. Condition (ANC)
% watershed - agricultural lands
Calculated Alkalinity (|ieq/L)
Total Dissolved aluminum (|ig/L)
Acid Mine Drainage Condition (SO4, ANC)
AMD Classification
Lumped AMD Classification
Gran Acid Neutralizing Capacity (ueq/L)
Watershed area digitized from maps
Residual Pool Vert Profile Area (m2/reach)
Residual Pool Vert Profile Area (m2/chan.)
Watershed Area in Hectares
Est. aspect of watershed longest dim.
bank protective vegetation score
% watershed - barren land
Calcium (|ieq/L)
lack of channel alteration score
channel flow status score
channel sinuosity score
Amount of Chlorophyll a (mg)
Chlorophyll a (mg)/m2 of Stream Bed
Ratio of Choro-a(mg):Periphyton AFDM(g)
Chloride (|ieq/L)
condition of banks score
Calculated Carbonate (|ieq/L)
Specific Conductance (|iS/cm)
Sandys Site Condition Class (l=good)
Straight line valley length of reach (m)
Dissolved Inorganic Carbon (mg/L)
Overall Disturbance Class
Sum of land use(URB_TOT+AG_TOT+MINE_TOT)
Dissolved Organic Carbon (mg/L)
Omernik Rev. Ecoregion ID
Omernik Level 4 Ecoregion ID (1996 ver.)
Est. elevation of stream index site (m)
Highest watershed elevation (m)
Mean Watershed Elevation (m)
Min Watershed Elevation (m)
gravel not buried by fines score
40
-------
epif_sub
EXOT_CLS
FE
FEN_SECT
FISH_D
FOR TOT
frq_riff
grazing
HCO3
HOUDENKM
in_cover
K
LRBS_TST
LSUB_DMM
LTEST
LTROFF_M
LWD_CLS
LWSKM2
MG
MINE_TOT
MN
NA
NH4
NO3
NONRES
NRP
NTL
NTL_CLS
NTROPH
NUTRCLS
PCAN C
PCAN D
PCAN E
PCAN M
PCAN_N
PCTBIGR
PCT BL
PCT CA
PCT CB
PCT_FA
PCT_FAST
PCT FN
PCT GC
epifaunal substrate score
Condition Class based on nonnative fish
Total Iron (mg/L)
Fenneman physiographic section designation
Reach Length (m) — as the fish swims
% watershed - forest
riffle frequency score
vegetative grazing disturbance score
Calculated Bicarbonate (|ieq/L)
Housing unit density (housing/km2)
instream cover score
Potassium (|ieq/L)
Logio[Relative Bed Stability] - Fast estimate
Substrate-Mean Logi0(Diameter Class mm)
Logio[Erodible Substr Dia.(mm)]-Fast estimate
Approx. meters of annual runoff
LWD Condition (XFC_LWD)
Logic watershed area (km2)
Magnesium (|ieq/L)
% watershed - mines/quarries/gravel pits
Total Manganese (mg/L)
Sodium (|ieq/L)
Ammonium (|ieq/L)
Nitrate (|ieq/L)
% watershed - non-residential urban lands
Number of residual pools in reach
Total Nitrogen (|ig/L)
Nutrient Condition (NTL)
Number of trophic guilds
Nutrient Classification
Riparian Canopy Coniferous (Fraction of reach)
Riparian Canopy Deciduous (Fraction of reach)
Rip Canopy Broadlf evrgrn (Fraction of reach)
Rip Canopy Mix Conif-Decid (Fraction of reach)
Rip Canopy Absent (Fraction of reach)
Substrate >= Coarse Gravel (>16 mm) (%)
Substrate Boulders - 250-4000 mm (%)
Cascade (% of reach)
Substrate Cobbles - 64-250 mm (%)
Falls (% of reach)
Fast Water Habitat (% riffle & faster)
Substrate Fines - Silt/Clay/Muck (%)
Substrate Coarse Gravel - 16-64 mm (%)
41
-------
PCT_GF Substrate Fine Gravel - 2-16 mm (°/
PCT_GL Glide (% of reach)
PCT_HP Substrate Hardpan -- (%)
PCT_OM Substrate Organic Detritus -- (%)
PCT_ORG Substrate Wood or Detritus -- (%)
PCT_POOL Pools -- All Types (% of reach)
PCT_RI Riffle (% of reach)
PCT_SA Substrate Sand -- .06-2 mm (%)
PCT_SAFN Substrate Sand & Fines - <2 mm (°/i
PCT_SFGF Substrate <= Fine Gravel (<= 16 mm) (%)
PCT_SLOW Slow Water Habitat (% Glide & Pool)
PCTCHARP % of channel length that forms residual pools
PCTCHASD % of channel length with sediments present
PCTRCHRP Residual pool length proportion (% reach)
PCTRSED Thalweg Sediment (<16mm) Pres.(% length of Thalweg)
PFC_ALG Filamentous Algae Presence (% reach)
PFC_ALL Any Types Fish Cover Present (% reach)
PFC_AQM Aq. Macrophytes Presence (% reach)
PFC_BIG LWD,RCK,OHB or HUM Fish Cover Pres (% reach)
PFC_BRS Brush & Small Debris Presence (% reach)
PFC_LWD LWD Presence (% reach)
PFC_NAT Any Natural Fish Cover Present (% reach)
PFC_OHV Overhang. Veg. Presence (% reach)
PFC_RCK Boulders Presence (% reach)
PFCJJCB Undercut Bank Presence (% reach)
PHSTVL Closed System pH
PMTD_C Rip MidLayer Coniferous (Fraction reach)
PMTD_D Rip MidLayer Deciduous (Fraction reach)
PMID_E Rip MidLayer broadlf evrgrn (Fraction reach)
PMID_M Rip MidLayer Mix Con-Decid (Fraction reach)
PMID_N Rip MidLayer Absent (Fraction of reach)
pool_sub pool substrate characterization score
pooljvar pool variability score
PRECIP_M Approx. annual precipitation (m)
PROJECT EMAP or REMAP
PTL Total Phosphorous (|ig/L)
PTL_CLS Nutrient Condition (PTL)
QR1 Riparian Quality Index
QRDIST1 Riparian Disturbance Index
QRPHALT2 Riparian Habitat Condition
QRVegl Riparian Vegetation Index
QRVeg2 Riparian Vegetation Index 2 (Understory layer)
RBPMEAN mean of all nonmissing scores
42
-------
RBPSUM
RD DEN
REACHLEN
REF
REFIAN
REFPHIL
REF SANDY
RIP_CLS
ripa_veg
ROUGHNES
RP100
RP100C
RPGT100
RPGT50
RPGT75
RPMXAR
RPMXDEP
RPMXLEN
RPMXVOL
RPMXWID
RPXDEP
RPXLEN
RPXVOL
RPXWID
SECTNAME
SED_CLS
sedi_dep
SINU
SIO2
SITEJD
SLOPE
SLOPMEAN
SO4
SOBC
STATE
STRAHLER
TEMPSTRM
TOLERNT9
TOT RD
TOTPLEN
TOTPLENC
TOTPVOL
TOTPVOLC
sum of all nonmissing scores
Road density (m/ha)
Length of sample reach (m)
RBP Habitat/Chem Reference Site (Y/N)
Ian NABS Paper Ref Site (N>100) (Y/N)
Ref Site by Chem/RBP/Quant. Phab.
Ref Site by Chem/RBP/Phab/Condtion
Riparian Condition (QRPHALT2)
width of riparian vegetation zone score
Terrain Roughness (unitless)
Mean Residual Depth (m2/100m)
Mean residual area per 100 m of chan.
Residual Pools >100cm deep (number/reach)
Residual Pools >50cm deep (number/reach)
Residual Pools >75cm deep (number/reach)
Max. RP profile area in reach (m2/pool)
Maximum residual depth in reach (cm)
Max. residual pool length in reach (m/pool)
Max volume of any pool in reach (m3)
Max residual width of any pool in reach (m)
Mean RP depth in reach (cm/pool)
Mean length of residual pools (m/pool)
Mean residual pool volume (m3/pool)
Mean residual width of reach (m)
Section name on Fenneman (1946) map
Excess Sediment Condition (LRBS_BW5)
lack of sediment deposition
Channel Sinuosity (m/m)
Silica (mg/L)
Site identification code
Approx. slope (HI_TO_LO / WSLTH)
Mean Watershed Slope (%)
Sulfate (ueq/L)
Sum of Base Cations (ueq/L)
Site State Location
Stream Order (Strahler)
Stream temperature (C)
Final IBIPTOLE Metric Score
m road in watershed (1992 TIGER files)
Total residual pool length (m/reach)
Total residual pool length (m/chan.)
Total residual pool volume (m3/reach)
Total residual pool volume (m3/chan.)
43
-------
TOTSDLEN
TOTSDLNC
TRASHED
TSS
TURB
URB TOT
velocity
VISIT_NO
W1_HAG
W1_HALL
W1JTNOAG
W1H CROP
W1H_LOG
W1H_MINE
W1H_PSTR
W1H_PVMT
W1H_ROAD
WETL_TOT
WS_AREA
WS COND
WSDISTRB
XC
XCDENBK
XCDENMID
XCEMBED
XCL
XCM
XCMG
XCMGW
XCMW
XCS
XDEPTH
XEMBED
XFC_ALG
XFC_ALL
XFC_AQM
XFC BIG
XFC_BRS
XFC_LWD
XFC_NAT
XFC OHV
XFC_RCK
XFC UCB
Total RP length with sediment (m/reach)
Total RP length with sediment (m/chan.)
Highly Disturbed Site by Chem/RBP/Sandy
Total Suspended Solids (mg/L)
Turbidity (NTU)
% watershed - urban lands
presence of velocity/depth regimes score
Visit Number
Rip Dist—Sum Agric Types (ProxWt Pres)
Rip Dist-Sum All Types (ProxWt Pres)
Rip Dist--Sum NonAg Types (ProxWt Pres)
Rip Dist-Row Crop (ProxWt Pres)
Rip Dist—Logging Activity (ProxWt Pres)
Rip Dist—Mining Activity (ProxWt Pres)
Rip Dist—Pasture/Hayfield (ProxWt Pres)
Rip Dist-Pavement (ProxWt Pres)
Rip Dist-Road/Railroad (ProxWt Pres)
% watershed wetlands
Watershed area (km2)
Watershed Condition Class (Colleens)
Human Disturbance Level in Watershed
Riparian Veg Canopy Cover
Mean Bank Canopy Density (%)
Mean Mid-channel Canopy Density (%)
Mean Embeddedness—Channel only (%)
Riparian Canopy > 0.3m DBH (Cover)
Rip Veg Canopy+Mid Layer Cover
Rip Veg Canopy+Mid+Ground Cover
Rip Veg Canopy+Mid+Ground Woody Cover
Rip Veg Canopy+Mid Layer Woody Cover
Riparian Canopy <= 0.3m DBH (Cover)
Thalweg Mean Depth (cm)
Mean Embeddedness—Channel+Margin (%)
Fish Cvr-Filamentous Algae (Areal Prop)
Fish Cvr-All Types (Sum Areal Prop)
Fish Cvr-Aq. Macrophytes (Areal Prop)
Fish Cvr-LWD,RCK,UCBorHUM(Sum Area Prop)
Fish Cvr-Brush&Small Debris (Areal Prop)
Fish Cvr-Large Woody Debris (Areal Prop)
Fish Cvr-Natural Types (Sum Areal Prop)
Fish Cvr-Overhang Veg (Areal Prop)
Fish Cvr-Boulders (Areal Prop)
Fish Cvr-Undercut Banks (Areal Prop)
44
-------
XINC_H Channel Incision Ht.-Mean (m)
XSLOPE Channel Slope -- reach mean (%)
XUN Undercut Distance—Mean (m)
XWD_RAT Mean Width/Depth Ratio (m/m)
XWIDTH Wetted Width - Mean (m)
XWXD Mean Width*Depth Product (m2)
YEAR Sample Year
ZN Dissolved Zinc (mg/L)
45
-------
vvEPA
United States
Environmental Protection
Agency
Office of Research and Development
National Exposure Research Laboratory
Cincinnati, OH 45268
Official Business
Penalty for Private Use
$300
EPA/600/R-06/010
February 16, 2006
PRESORTED STANDARD
POSTAGE & PAID
EPA PERMIT NO. G-35
------- |