vvEPA
United States
Environmental Protection
Agency
EPA/600/R-08/143 | April 2010 | www.epa.gov/ncea
Causal Assessment of Biological
Impairment in the Bogue Homo River,
Mississippi Using the U.S. EPA's
Stressor Identification Methodology
National Center for Environmental Assessment
Office of Research and Development, Washington, DC 20460
-------
E PA/600/R-08/143
April 2010
Causal Assessment of Biological
Impairment in the Bogue Homo
River, Mississippi Using the
U.S. EPA's Stressor
Identification Methodology
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
NOTICE
The Mississippi Department of Environmental Quality and the U.S.
Environmental Protection Agency (U.S. EPA) through its Office of Research and
Development jointly prepared this report. It has been subject to the Agency's peer and
administrative review and has been approved for publication as an U.S. EPA document.
Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
ABSTRACT
This assessment of biological impairment in the Bogue Homo River, Mississippi is
taken from more than 700 court ordered assessments of the causes of impairments
requiring development of a total maximum daily load (TMDL). A TMDL is the calculation
of the maximum amount of a pollutant that a body of water can receive and still meet
water quality standards, and an allocation of that amount to the pollutant's sources. The
calculation of a TMDL is required for all waters that are listed as impaired, in
accordance with §303(d) of the Clean Water Act.
The Bogue Homo, a stream in southeast Mississippi, was initially listed as impaired
based upon an evaluation of information with no in situ measurements. Follow up
biological monitoring of benthic macroinvertebrates by the Mississippi Department of
Environmental Quality (MDEQ), confirmed that the stream was impaired. The MDEQ
chose to use the U.S. Environmental Protection Agency (U.S. EPA) stressor
identification (SI) process to aid in determining probable causes of biological impairment
in aquatic ecosystems. A conceptual diagram was developed to present the most
common sources of pollutants, causal pathways, proximate causes and specific
biological effects in the Bogue Homo. The candidate causes evaluated in this process
included decreased dissolved oxygen (DO) and altered food resources (organic matter),
unsuitable habitat, increased temperature, increased ionic strength, and/or increased
toxicity.
Data used to support the causal analysis process included benthic
macroinvertebrate-community metrics, sediment particle-size counts, various water
quality measurements, land use and land cover percentages, and other watershed
information. Tables were developed to compare observed parameter levels for the
impaired site to the 25th or 75th percentiles of a least-disturbed condition for the
bioregion (references sites selected on the basis of similar biological communities) or
site class (references sites selected on the basis of similar physical and chemical
characteristics), depending upon the parameter of concern. Biotic and abiotic
conditions from Bogue Homo were also compared to those of nearby unimpaired sites.
Lastly, scatter plots of biotic and abiotic site-class data were used to assess regional
stressor-response relationships. To add to the strength-of-evidence analysis, Bogue
ii
-------
Homo data were then compared to biological metrics of regional stressor-response
relationships that tended to be associated with different stressors.
This case study describes impairments, identifies candidate causes, evaluates
relationships between causes and biological response variables, and identifies the most
likely causes of impairment through a combined approach of elimination and strength of
evidence. The MDEQ identified altered food resources (which include organic
enrichment and nutrient enrichment and could lead to decreased DO) as probable
causes of impairment. Subsequent to this assessment, TMDLs were developed for the
applicable causes of impairment to Bogue Homo as identified through this causal
analysis process.
Preferred citation:
Hicks, M., K. Whittington, J. Thomas, J. Kurtz, A. Stewart, G. Suter II, and S.M. Cormier. 2010. Causal
Assessment of Biological Impairment in the Bogue Homo River, Mississippi Using the U.S. EPA's
Stressor Identification Methodology. U.S. Environmental Protection Agency, Office of Research and
Development, National Center for Environmental Assessment, Cincinnati, OH. EPA/600/R-08/143.
Cover photo:
MDEQ (Mississippi Department of Environmental Quality). 2005. The Bogue Homo River near Ovett.
Available at
http://www.deq.state.ms.us/Mdeq.nsf/pdf/TWB_BogueHomoLowDO&NutJun05/$File/PascagoulaRBBogu
eHomoLowDO&NutrientsJun05.pdf?OpenElement (accessed March 3, 2009).
iii
-------
TABLE OF CONTENTS
Page
LIST OF TABLES vii
LIST OF FIGURES viii
LIST OF ABBREVIATIONS ix
PREFACE x
AUTHORS, CONTRIBUTORS AND REVIEWERS xii
1. DEFINE THE CASE 1
1.1. REGULATORY CONTEXT FOR THE CASE 1
1.1.1. Mississippi's 1996 §303(d) List of Impaired Water Bodies 1
1.1.2. Process to Address 1996 §303(d) List of Impaired Water Bodies... 1
1.1.3. Regulatory Context of the Site 2
1.2. DESCRIPTION OF THE WATERSHED 2
1.3. POTENTIAL SOURCES OF STRESSORS 2
1.4. SPECIFIC BIOLOGICAL IMPAIRMENT 7
1.4.1. Site-Specific Comparisons (SSC) 7
1.4.2. Regional Comparisons 7
1.4.3. Beck's Biotic Index (Bl) 10
1.4.4. Number of Ephemeroptera, Plecoptera, and Trichoptera
(EPT) Taxa 12
1.4.5. Percent Ephemeroptera, Plecoptera, and Trichoptera (EPT)
(No Caenidae) 12
1.4.6. Percent Plecoptera 12
1.4.7. Percent Amphipoda 12
1.4.8. Percent Predators 12
1.4.9. Percent Sprawlers 13
2. LISTING THE CANDIDATE CAUSES 14
2.1. Mississippi's Standard LIST OF CAUSES OF IMPAIRMENT 14
2.2. PROCESS OF ELIMINATION 14
2.3. CONCEPTUAL MODEL OF CAUSAL PATHWAYS 16
3. EVALUATE DATA FROM THE CASE 19
3.1. SPATIAL/TEMPORAL CO-OCCURRENCE OF PROXIMATE
CAUSAL AGENTS AND BIOLOGICAL EFFECTS 19
iv
-------
TABLE OF CONTENTS cont.
Page
3.1.1. Decrease in Suitable Habitat 19
3.1.2. Altered Temperature Regime 21
3.1.3. Altered Food Resources and/or Decreased Dissolved Oxygen.... 22
3.1.4. Increase in Ionic Strength and/or Increase in Toxic Substances ..22
3.2. CAUSAL PATHWAY 22
3.2.1. Decrease in Suitable Habitat 22
3.2.2. Sources of Hydrologic Alteration 23
3.2.3. Sediment Sources 23
3.2.4. Altered Temperature Regime 23
3.2.5. Altered Food Resources and/or Decreased Dissolved Oxygen.... 23
3.2.5.1. Sources 23
3.2.5.2. Organic Matter 24
3.2.5.3. Nutrients 24
3.2.6. Increase in Ionic Strength and/or Increase in Toxic Substances ..24
4. EVALUATE DATA FROM ELSEWHERE 25
4.1. MECHANISTICALLY PLAUSIBLE CAUSE 25
4.1.1. Decrease in Suitable Habitat 25
4.1.2. Altered Temperature Regime 26
4.1.3. Altered Food Resources and/or Decreased Dissolved Oxygen.... 26
4.1.4. Increase in Ionic Strength and/or Increase in Toxic Substances ..26
4.2. STRESSOR-RESPONSE RELATIONSHIPS FROM OTHER FIELD
STUDIES 26
4.2.1. Scatter Plots 26
4.2.2. Box Plot Distributions 27
4.2.3. Decrease in Suitable Habitat 32
4.2.4. Altered Temperature Regime 32
4.2.5. Altered Food Resources and/or Decreased Dissolved Oxygen.... 32
4.2.6. Increase in Ionic Strength and/or Increase in Toxic Substances .. 34
5. COMPARISON OF CANDIDATE CAUSES 35
v
-------
TABLE OF CONTENTS cont.
Page
6. IDENTIFY THE PROBABLE CAUSE 37
6.1. PROBABLE PRIMARY CAUSE 37
6.1.1. Increased Organic and Nutrient Enrichment Altering Food
Resource and Leading to Low 37
6.2. PROBABLE SECONDARY CAUSES 37
6.2.1. Decrease in Suitable Habitat 37
6.3. LESS PROBABLE OR UNLIKELY CAUSES 38
6.3.1. Altered Thermal Regime 38
6.3.2. Increase in Ionic Strength and/or Increase in Toxic Substances .. 38
7. DISCUSSION AND HIGHLIGHTS 40
8. REFERENCES 44
APPENDIX A: DATA USED FOR CAUSAL ANALYSIS 47
APPENDIX B: COMMUNITY SYMPTOMOLOGY 49
APPENDIX C: PEARSON CORRELATION COEFFICIENT MATRIX 53
APPENDIX D: STRESSOR-RESPONSE RELATIONSHIPS EXPRESSED AS
SCATTER PLOTS OF BIOLOGICAL AND PHYSICAL/CHEMICAL DATA WITH
BOGUE HOMO PLOT HIGHLIGHTED 57
APPENDIX E: STRESSOR-RESPONSE RELATIONSHIPS EXPRESSED AS BOX
PLOTS OF BIOLOGICAL DATA 61
vi
-------
LIST OF TABLES
No. Title Page
1 LULC Data as Percentages for the Bogue Homo Watershed and Riparian
Zones 6
2 Biological Metrics, Bogue Homo and Comparison Conditions 11
3 MDEQ Standard List of Candidate Causes and More Related Processes 15
4 Water Quality Data, Bogue Homo and Comparison Conditions 20
5 Pearson Correlation Coefficients (r-values) of Biological Metrics and
Physical/Chemical Parameters Used for Scatter Plot Evaluation (n = 63) 28
6 Summary of Stressor-Response Relationships from Linear Correlation of
Various Biotic Variables Against Variables Associated with Candidate
Causes 29
7 Parameters Evaluated Using Box Plots 31
8 Summary of Stressor-Response Relationships from Box Plots of Various
Biotic Variables Based on Low and High Ranges of Variables Associated
with Candidate Causes 33
9 Strength of Evidence for Bogue Homo, Mississippi 36
vii
-------
LIST OF FIGURES
No. Title Page
1 Bogue Homo Watershed and Surrounding Area 3
2 LULC of the Bogue Homo Watershed 5
3 Site Classes, a Classification Based on Physical and Chemical
Characteristics 8
4 Bioregions, a Classification Based on Similar Biological Communities 9
5 Causal Pathways Related to Proximate Causes Analyzed in this Case 17
6 Conceptual Model for Bogue Homo 18
viii
-------
Bl
CADDIS
COD
CWA
DO
EPT
GIS
HBI
LD
LULC
M-BISQ
MDEQ
msl
NPDES
NTU
r
SI
SSC
TDS
TKN
TMDL
TOC
U.S. EPA
WQS
LIST OF ABBREVIATIONS
biotic index
Causal Analysis/Diagnosis Decision Information System
chemical oxygen demand
Clean Water Act
dissolved oxygen
Phylogenetic Orders Ephemeroptera, Plecoptera and Trichoptera
geographic information systems
Hilsenhoff Biotic Index
least disturbed
land use and land cover
Mississippi Benthic Index of Stream Quality
Mississippi Department of Environmental Quality
mean sea level
National Pollution Discharge Elimination System
nephelontric turbidity units
Pearson product moment correlation coefficient
stressor-identification
site-specific comparison
total dissolved solids
total Kjeldahl nitrogen
total maximum daily load
total organic carbon
U.S. Environmental Protection Agency
Water Quality Standard
ix
-------
PREFACE
This is a causal assessment of a biologically impaired river in the state of
Mississippi. The case was investigated by the Mississippi Department of Environmental
Quality (MDEQ) as a result of court ordered mandates to determine the causes and to
propose resolutions for more than 700 river segments that were listed as impaired on
the §303d list as required by the Clean Water Act. The causal assessment of the
Bogue Homo River was one of the first cases conducted by the MDEQ and resulted in a
determination of the impairment and completion of a total maximum daily load (TMDL)
as required by the court mandate. As may be expected, these early cases provided
opportunities to fine tune MDEQ's causal assessment process and the U.S.
Environmental Protection Agency (U.S. EPA) methodology which was revised in 2006
appearing in Web form at www.epa.gov/caddis. For the most part, the assessment
presented here is based on the earlier methods that were available at the time of the
assessment. It is presented here to illustrate how a causal assessment was performed
using the U.S. EPA (2000) Stressor Identification process and the lessons learned that
have since been incorporated into the Causal Analysis/Diagnosis Decision Information
System (CADDIS) Web site. The analysis was restructured from the original TMDL
(MDEQ, 2005) during a workshop at Canaan Valley, West Virginia in May of 2005 and
in subsequent discussions. The intention was to share and develop some of the
lessons learned. The sampling, analysis, and conclusions are those of researchers who
were employed by the MDEQ. Comments appearing in text boxes were prepared by
the U.S. EPA's National Center for Environmental Assessment (NCEA) except where
noted. NCEA provided editorial and formatting assistance to make the original MDEQ
report similar to four other case studies solicited as examples to include on through the
CADDIS Web site for other practitioners of causal assessment.
The Bogue Homo River case study is one of five causal analyses that were
completed prior to 2005 by states. These early cases determined the probable causes
of a biological impairment as required by the TMDL rule. Data for these cases were
limited. And yet, for the Bogue Homo, MDEQ developed evidence to show that some
causes co-occurred with the biological impairment, were a part of a larger causal chain
of events, occurred at sufficient levels known to cause the observed effects, and were
coherent with general ecological and scientific theory. Although available evidence was
not equivalent in amount or quality for all candidate causes, it was enough to identify
some probable causes and to suggest what additional, targeted data might improve the
confidence in the determination.
This case, as in all cases, could be improved, but represents the capabilities and
level of analysis that was available in 2005. Since then, additional analytical tools and
databases have become more readily available. States, tribes, and territories continue
to reduce the uncertainty of their analyses using the U.S. EPA's stressor-identification
process and CADDIS Web site. This and other case studies from the Canaan Valley
Workshop defined the impairment based on a biological index rather than more specific
impairments. This practice diminishes the ability to detect associations because
x
-------
summing the metrics into an index dampens the overall signal from individual metrics
and species that are responding differently to environmental conditions or stressors. In
the Bogue Homo River case, individual metrics of the index were analyzed after the
fact, at the workshop in 2005. Although the causal associations were not fully
developed, using the individual metrics allowed some causes to be identified and to
improve subsequent cases.
To address these and other issues, comment boxes have been inserted
throughout the Bogue Homo case study to supply important considerations or to
suggest other approaches that could strengthen the case. The analyses in the cases
cannot be modified as they are already a part of the MDEQ's public record. To make
this easier, the case studies are linked to relevant tools and guidance on the U.S. EPA
Web site: www.epa.gov/caddis.
In summary, the case study of the Bogue Homo River presents a very realistic
example of the difficulties of assigning specific causes to biological impairments. The
Bogue Homo River Case Study is a good example of several strategic techniques to
use for expediting causal analyses and TMDLs. Highlights include
1. Developing a list of commonly encountered causes in this region of the United
States and measurements that can be used to evaluate them.
2. Rationales for differentiating between deferred causes due to insufficient data or
the practical consideration, and elimination of causes based on logical
implausibility.
3. Evaluating spatial/temporal co-occurrence using two pieces of evidence from the
site and demonstrating a new type of evidence: spatial/temporal co-occurrence
using data from elsewhere.
4. Classification of field data prior to the development of stressor-response
relationships (termed stratification by MDEQ).
5. Using scatter plots to screen for potential stressor-response associations, and
box plots and regression plots to evaluate plausible stressor responses using
data from regional monitoring data.
6. The conclusions of the assessment were limited to a screening level assessment
by the low sampling density and lack of stream chemistry data and upstream
biological monitoring data.
Editor: Susan M. Cormier February 2010
xi
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
AUTHORS
Matthew Hicks
RMA, Inc.
Brandon, MS 39047
Jan Kurtz
U.S. EPA, Office of Research and Development, National Health and Environmental
Effects Research Laboratory
Gulf Breeze, FL 32561
Jeff Thomas
RMA, Inc.
Brandon, MS 39047
Kay Whittington
Mississippi Department of Environmental Quality
Surface Water Division
Jackson, MS 39204
CONTRIBUTING AUTHOR
Athur J. Stewart
Oak Ridge Associated Universities
Oak Ridge, TN 37830
Glenn Suter II
U.S. EPA, Office of Research and Development, National Center for Environmental
Assessment
Cincinnati, OH 45268
EDITOR
Susan M. Cormier, Ph.D.
U.S. EPA, Office of Research and Development, National Center for Environmental
Assessment
Cincinnati, OH 45268
xii
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS cont.
REVIEWERS
U.S. EPA REVIEWERS
Rick Ziegler
U.S. EPA, Office of Research and Development, National Center for Environmental
Assessment
Washington, DC 22202
Lester Yuan, Ph.D.
U.S. EPA, Office of Research and Development, National Center for Environmental
Assessment
Washington, DC 22202
Suzanne Lussier, Ph.D.
U.S. EPA, Office of Research and Development, National Health and Environmental
Effects Research Laboratory, Atlantic Ecology Division
Narragansett, Rl 02882
EXTERNAL REVIEWERS
Charles A. Menzie, Ph.D.
Exponent
1800 Diagonal Road, Suite 300
Alexandria, VA 22314
Kent W. Thornton, Ph.D.
FTN Associates, Ltd.
Little Rock, AR 72211
Mary D. Matlock, Ph.D., P.E., C.S.E.
University of Arkansas
Fayetteville, AR 72701
ACKNOWLEDGMENTS
We would like to thank Cris Broyles of IntelliTech Systems, Inc and Heidi Glick of
ECFlex, Inc. for their editorial services, the TSS staff of ECFlex, Inc. for their formatting
services, and Bette Zwayer and Ruth Durham of the U.S. EPA for shepherding this
document through editing, formatting, and clearance.
xiii
-------
1. DEFINE THE CASE
1.1. REGULATORY CONTEXT FOR THE CASE
The Clean Water Act (CWA) requires states to identify waters that are impaired.
These waters are published in the State's §303(d) List of Impaired Water Bodies. The
CWA also requires that a total maximum daily load (TMDL) be completed for each
waterbody pollutant combination on the §303(d) List. The U.S. Environmental
Protection Agency (U.S. EPA) was sued for not enforcing the TMDL requirements of the
CWA in the majority of states, including Mississippi. The lawsuit resulted in a
settlement that included a court-ordered 10-year schedule for TMDL development for all
water bodies on Mississippi's 1996 §303(d) List.
1.1.1. Mississippi's 1996 §303(d) List of Impaired Water Bodies
The 1996 §303(d) list consisted of over 700 bodies of water. Approximately 20%
of the listed waters were classified as impaired based on actual biological or chemical
monitoring data. The other 80% of the listed waters were based on anecdotal
information indicating potential impairment, but not on measured data. However, all
waters were part of the settlement, even those that were listed without data to support
actual impairment. The settlement set out a 10-year schedule: address water bodies
identified as impaired based on actual monitoring data by the end of the first 5 years,
and address water bodies identified as impaired, but not based on monitoring data, by
the end of the second 5 years.
1.1.2. Process to Address 1996 §303(d) List of Impaired Water Bodies
Mississippi Department of Environmental Quality (MDEQ) identified two tasks to
address before TMDL development for impaired waters. The first task was to revisit the
1996 list and confirm or refute the listing of the impaired waters, especially those where
monitoring data were lacking. Beginning in 2001, MDEQ implemented a statewide
monitoring strategy to collect biological,
physical, and chemical data from listed
waters. MDEQ then used those data to
determine, with a higher degree of certainty,
those waters that were impaired. The second
task was to determine the specific cause/s of
impairment of those waters that were
confirmed as impaired (see Comment 1). In
2003, MDEQ developed a causal analysis
process to identify specific causes of
impairment, so that actions could be taken to
improve water quality. Depending on the
identified causes of impairment, TMDL
development was often the action taken.
Comment 1. About the Comment Boxes.
At various points in this document, the
U.S. EPA editor provides comments. These
are not meant to indicate that the MDEQ
causal analysis is in error. The stressor-
identification (SI) process does not address
every possible option, nor does it provide
details on implementation, so there are many
opportunities for interpretation (U.S. EPA,
2000). The U.S. EPA encourages states
and tribes to improve and interpret the
methodology in ways that are appropriate to
their circumstances. Hence, the inserted
comments are meant to help other SI users
by indicating alternative approaches that
they might apply to their cases.
1
-------
1.1.3. Regulatory Context of the Site
Bogue Homo is one of 11 water bodies in Mississippi for which the causal
analysis process has been completed. The water use classification for Bogue Homo, as
established by the State of Mississippi in the Water Quality Criteria for Intrastate,
Interstate and Coastal Waters regulation, is Fish and Wildlife Support (MDEQ, 2003a).
The designated beneficial use for Bogue Homo is aquatic life support.
1.2. DESCRIPTION OF THE WATERSHED
Bogue Homo is a tributary to the Leaf River and its headwaters are near
Heidelberg in southeast Mississippi. It flows south to the Leaf River, a total distance of
approximately 86-km. Bogue Homo is slightly sinuous and lies within a watershed of
689-km2. Elevations of the stream range from around 150 m mean sea level (msl) near
its headwaters, to about 46 m msl at its confluence with Leaf River. The main stem of
Bogue Homo is impounded east of Laurel, Mississippi (Lake Bogue Homo). Bogue
Homo has several small feeder streams. The largest of these, Mill Creek, is also
impounded (Masonite Lake). The Bogue Homo watershed is mostly rural, dominated by
forest and pasture. There are several National Pollution Discharge Elimination System
(NPDES)-permitted facilities that discharge to the upper part of Bogue Homo. There is
no centralized wastewater collection and treatment service in the Bogue Homo
watershed. A map of the Bogue Homo watershed and some watershed characteristics
are shown in Figure 1.
Bogue Homo was on the state's 1996 §303(d) List of Impaired Water Bodies.
The entire watershed of Bogue Homo fell under the listing, and the evaluated causes
initially listed were pesticides, nutrients, and siltation. This listing was based primarily
on review of evaluated anecdotal and land use information (i.e., no in stream measured
data were available or used to list the site). In response to the court ordered mandate,
limited stream sampling was conducted throughout Mississippi. In stream
measurements from Bogue Homo (sampling station 487) in 2001 indicated impairment
based on the resulting Mississippi benthic index of stream quality (M-BISQ) score.
Based on this assessment, Bogue Homo remained on the §303(d) List of Impaired
Water Bodies. The previously listed causes—pesticides, nutrients, and siltation—were
removed, and "biological impairment" was identified as the basis for the listing. This
triggered the causal analysis process (MDEQ, 2001).
1.3. POTENTIAL SOU RCES OF STRESSORS
Potential sources of stressors were identified using data and information from
land use and land cover (LULC) characteristics of the Bogue Homo watershed and field
reconnaissance of the watershed. Additional watershed characteristics, such as
locations of NPDES discharges, septic tanks, and other miscellaneous sources of
pollution, were evaluated. Predominant land cover classes and uses in the Bogue
Homo watershed at the time of the study were forest (55.7% of the watershed area) and
2
-------
~
Rachuta
WW3.
Quitman
Raleigh
!.r
/
bylvarena
BayrSprings
" 1_H
w
Clarke
Bay Springs
Jasper
Smith
Simpson
Mize
-Heidelberg
Shubuta
Taylorsvme
sandersvii e
v/lount Olive
Laure
l:aure
^ Waynesboro
Wayne
ollms
Jones
w
Elllsville
m
Covington
Seminary
JeffersonxDavis'^
State'Line
This map produced by the Department
of Environmental Quality (MDEQ), Office of
Pollution Control, Surface Water Division,
Water Quality Assessment Branch, Data
Management Section on 25 October 2004.
The TMDL watershed boundary and TMDLWater
was produced by the MDEQ. All other map data
provided by MARIS.
Map Projection: Mississippi Transverse Mercator
The Mississippi Department of Environmental Quality
makes no warranties, expressed or implied, as to the
accuracy, completeness, currentness, reliability, or
suitability for any particular purpose, of the data u
contai ned on thi s m ap. -
MDEQ
Mississipp
^ LT"!/usHighway Bogue Homo River
53 Lake or pond IBI Watershed
County Boundary
Major River 9 ? 4 6 8 10
' I 1 I 1 I—=l MIlRS
—— Perennial Stream
^ IBI Station
Bogue Homo River IBI Watershed
FIGURE 1
Bogue Homo Watershed and Surrounding Area, In this study, the watershed
highlighted in grey is evaluated for biotic condition at site 487 using an index of biotic
integrity (IBI) based on invertebrates, termed the M-BISQ.
3
-------
pasture land (21.1% of the watershed area; see Figure 2). Percentage totals of LULC
for Bogue Homo are summarized in Table 1.
Anthropogenic land uses (urban cover, crops, pasture, and scrub/barren areas)
comprised 40.8% of the entire watershed upstream of the sampling station on Bogue
Homo. This value was 28.1 % for the 100-m buffer zone of the waterbody channel in the
watershed of Bogue Homo, upstream of the sampling station. However, the 100-m
buffer zone in a 1-km upstream of the sampling site had no urban land use, and only
11.6% of that land was composed of crops, pasture, scrub/barren land cover.
Results from field reconnaissance, geographic information systems
(GlS)-mapping analysis, and MDEQ file review revealed several more site-specific
candidates for potential sources, including several upstream NPDES discharges,
nonsewer areas, transportation corridors, an impoundment release, poultry and cattle
operations, and silviculture operations.
Land-use activities of note in the watershed include silviculture and a small
lumberyard. Excess nutrients, pesticides, and sediment can be delivered to nearby
streams from silviculture lands, especially when best management practices are not
used. Runoff from silviculture lands, agricultural lands and roads, as well as historic
landscape alteration and impoundments, can affect stream flow and result in increased
erosion, bank destabilization, and an excess deposition of sediment. Runoff from road
surfaces also can carry contaminants such as oil, gas, and metals associated with
automobile and road-maintenance activities (e.g., platinum, palladium, and zinc).
Poultry operators are permitted dischargers in the watershed and are a potential source
of nutrients, food additives, and organic matter. Urban and disturbed lands can produce
multiple physical and chemical stressors from residential, commercial, and industrial
areas in which different stressors are cumulative over time (temporal) and are
distributed further downstream in the watershed (spatial).
In summary, the following potential sources of stressors in the Bogue Homo
watershed were identified:
• Impoundment
• Nonsewer areas
• Lumberyard
• Silviculture practices
• Historic landscape alteration
• Pasture/grassland
• Upstream point source discharges
• Poultry operations
• Transportation/roadside ditches
4
-------
Clarke
Jasper
Smith
Simpson
Wayne"
Covington
Jones
JeffersprMDavis
FIGURE 2
LULC of the Bogue Homo Watershed
5
-------
TABLE 1
LULC Data as Percentages for the Bogue Homo Watershed and Riparian Zones
Land Cover
Entire Watershed
(EW)
100-m Buffer EW
100-m Buffer 1 -km
upstream of site
Acres
% Area
Acres
% Area
Acres
% Area
Urban
1,141
0.7%
206
0.5%
0
0%
Forest
94,992
55.7%
25,024
66.1%
149
88.4%
Cropland
2,846
1.7%
197
0.5%
0
0%
Pasture/Grassland
35,985
21.1%
4,445
11.7%
12
6.9%
Scrub/Barren
29,433
17.3%
5,851
15.4%
8
4.7%
Water
1,709
1.0%
628
1.7%
0
0%
Wetland
4,344
2.5%
1,533
4.0%
0
0%
Cloud/Shadow
0
0%
0
0%
0
0%
Total
170,450
100.0%
37,885
100.0%
169
100.0%
6
-------
1.4. SPECIFIC BIOLOGICAL IMPAIRMENT
Impairment of Bogue Homo was defined in terms of its M-BISQ score and
associated biological metric values, as compared to the least disturbed (LD) condition
for the region (see Appendix A). The specific impairment of selected M-BISQ metrics
was based on differences in metric scores between local and two types of regional
comparison sites (see Comment 2). The six site classes with Bogue Homo M-BISQ site
and comparison locations are seen in Figure 3. The six bioregions with Bogue Homo
M-BISQ site and comparison locations are seen in Figure 4.
1.4.1. Site-Specific Comparisons (SSC)
Observations from selected
site-specific comparison (SSC)
stations were used for
comparisons with the Bogue Homo
site. SSC stations were in close
proximity to the Bogue Homo
monitoring station (i.e., <30-km)
and were similar to Bogue Homo
with respect to physical, chemical,
and geological characteristics.
However, they were unimpaired,
based on a biological community
assessment.
1.4.2. Regional Comparisons
In addition to comparisons
at the site specific scale, two types
of regional comparisons were
made using reference sites with
similar physical and chemical characteristics (Site Class 6 comparison sites) and
another with reference sites with similar biological communities (East Bioregion
comparison sites). For modeling and assessment purposes, geographic strata of
Mississippi were identified based on natural variation in physical, chemical, and
biological stream characteristics, as well as on landscape variables such as soil type
and natural vegetative land cover. MDEQ sought to maximize inter-strata variability and
minimize intra-strata variability. Two stratification schemes for the state were identified
and used in the assessment process; one based on abiotic variability, termed Site
Classes, and the other based on biotic variability, termed Bioregions.
Abiotic variability refers to frequency and magnitude of changes in environmental
factors such as temperature, turbidity, and levels of nutrients. Biotic variability refers to
the frequency and magnitude of changes in the abundances of organisms in various
taxa. The process used to develop site class and bioregional stratification was
Comment 2. Comparison Sites.
The selection of reference or comparison sites based on
different selection factors focuses the evidence on different
parts of the causal pathway; that is, land use and stressor
related (site class) versus biological response related sets
of sites (bioregion). Both data sets are used to evaluate
stressor response from other field studies. When both
data sets support or weaken the cause, there is more
confidence than with just one line of evidence.
As an exploratory effort, two methods were used to
present the information: scatter plots with regression, and
box plots. Scatter plots have the advantage of showing all
the data for the candidate cause and the biological
response measure, along with the site of interest. The
disadvantage is that the complexity of the plot may be
overwhelming for communicating to a less experienced
audience. The box plots are simpler for communication
purposes, but do not provide a way to graphically display
both the value of the biological response and the
candidate cause from the site. For ease of
communication, it would probably be best to choose one
mode of communication.
7
-------
N
S
Legend
O Bogue Homo IBI Site
• Comparator Site
C3 Pascagoula River Basin
Vteterbody
Major Rivers
SITE CLASSES
Site Class 1
Site Class 2
Site Class 3
Site Class 4
Site Class 5
Site Class 6
This map produced by the Department of
Environmental Quality (MDEQ), Office of
Pollution Control, Surface Water Division,
Watershed Management Branch on
25 August 2004.
The Mississippi Department of Environmental
Quality makes no warranties, expressed or
implied, as to the accuracy, completeness,
currentness, reliability, or suitability for any
particular purpose, of the data contained on
this map.
o
25
50
100
Mississippi Department of
Environmental Quality
150
~ Miles
J
im
FIGURE 3
Site Classes, a Classification Based on Physical and Chemical Characteristics. The
Bogue Homo M-BISQ Site is in site class 6 and Comparison Locations reside in site
Classes 6 and 1.
8
-------
Legend
0 Bogue Homo IBI Site
• Comparator Site
C3 Pascagoula River Basin
\Afoterbody
Major Rivers
BIOREGIONS
q3 Black Be|t
C3 East
1 , Northwest
X Northeast
C3 West
, I c;: North Delta
This map produced by the Department of
Environmental Quality (MDEQ), Office of
Pollution Control, Surface Water Division,
Watershed Management Branch on
25 August 2004.
The Mississippi Department of Environmental
Quality makes no warranties, expressed or
implied, as to the accuracy, completeness,
currentness, reliability, or suitability for any
particular purpose, of the data contained on
this map.
Mississippi Department of
Environmental Quality
150
iMMrs
FIGURE 4
Bioregions, a Classification Based on Similar Biological Communities. The Bogue
Homo M-BISQ site and comparison locations reside in the east bioregion.
9
-------
described in detail in MDEQ (2003b). In general, site classes were areas of the state
containing streams with naturally similar physical, chemical, and land use
characteristics. Bioregions were areas of the state containing streams with naturally
similar benthic macroinvertebrate taxa and communities. They were identified by first
looking at similarity among Site Classes, then lumping or splitting Site Classes in order
to delineate regions with naturally similar benthic macroinvertebrate taxa composition.
Bogue Homo is located in Site Class 6 and the East Bioregion. For each site
class and bioregion, LD conditions have been defined for biological, habitat, sediment
particle size, water quality, and LULC data. In short, LD conditions were based on the
quartile distribution of the particular metric values from streams considered to be least
disturbed for each site class and bioregion. The upper quartile (75th percentile) of the
range of values from least disturbed streams was used to define LD biological metrics
that cause a decline in the M-BISQ score as their value increases (e.g., percent
Amphipoda). The lower quartile (25th percentile) of the range of values from least
disturbed streams was used to define LD conditions for biological metrics that increase
as the M-BISQ score increases (e.g., percent Plecoptera). The 50th percentile was
selected for some functional feeding groups and habitat preferences. For chemical
measures, the 25th percentile was selected for agents that increase as M-BISQ score
increases (e.g., dissolved oxygen [DO]). The 75th percentile was selected when an
increase in the agent was expected to cause a decrease in the M-BISQ (e.g.,
ammonia). The 50th percentile was selected for agents that were proportional (percent
silt).
Assessment of water quality status and subsequent placement of Bogue Homo
on the §303(d) List involved the comparison of the 2001 M-BISQ score for Bogue Homo
to the impairment threshold of the East Bioregion. The impairment threshold was
defined as the 25th percentile value of the range of M-BISQ scores from LD sites in the
East Bioregion. The M-BISQ score for Bogue Homo (50.07) was lower than the LD
impairment threshold value for the East Bioregion (61.35). This difference was great
enough to classify Bogue Homo as impaired.
More specifically, several biological metric
values from Bogue Homo were considerably
lower or higher than metric values for LD
conditions and site-specific comparison
stations (see Table 2). Those metrics are
described below and in Appendix B, with
reasons for considering them suggestive of
impairment based on the data in Table 2 (see
Comment 3).
1.4.3. Beck's Biotic Index (Bl)
This metric is based on individual
tolerance values for each benthic
macroinvertebrate taxon. It results in a total
Comment 3. Summary of Impairment.
Determining exactly what biological
changes have occurred at an impaired site
facilitates causal analysis. A clear sign of
the impaired biological community at
Bogue Homo is the low number of EPT
taxa. The number of EPT taxa (3) is less
than half of the regional least disturbed
conditions and only 19-30% of the local
reference (SSC) values. The related
parameters, percent EPT and percent
Plecoptera, are also clearly low. Another
notable attribute is the high percentage of
amphipods. Most SSC sites have none,
but the impaired site has 13%. Therefore,
assessors may choose to seek the
cause(s) of these particular shifts in the
benthic assemblage rather than all of the
differences (see Appendix B).
10
-------
TABLE 2
Biological Metrics, Bogue Homo, and Comparison Conditions
Difference from LD
condition
Least Disturbed Condition
East Bioregional Metrics
Bogue
Homo
East
Bioregion
Site Class 6
East
Bioregion
Site Class 6
percentile
used for LD
MBISQ
50.07
lower
lower
61.35
61.85
25th
Beck's Bl
18.00
comparable
lower
18.00
18.75
25th
HBI
4.68
higher
higher
4.34
4.35
75th
# Tanytarsini Taxa
1.00
lower
lower
2.00
1.99
25th
% Caenidae
0.00
lower
lower
0.43
0.42
75th
% EPT (No Caenidae)
3.65
lower
lower
9.10
12.51
25th
% dingers
40.10
lower
lower
51.20
55.17
50th
% Filterers
26.56
higher
higher
19.43
17.12
50th
Additional Biological Metrics
% Amphipoda
13.02
higher
higher
1.95
2.57
75th
# EPT taxa
3.00
lower
lower
6.95
7.56
25th
% Plecoptera
0.52
lower
lower
0.94
1.13
25th
% Predators
28.65
higher
higher
16.28
19.08
75th
% Sprawlers
26.56
higher
higher
19.57
15.39
75th
Site Specific Comparators
East Bioregional Metrics
Bogue
Homo
Black Creek
Bogue Homo
(downstream)
Big Creek
Oakahay
Creek
Yellow
Creek
MBISQ
50.07
72.24
64.65
72.31
66.01
82.46
Beck's Bl
18.00
28.00
23.00
16.00
22.00
34.00
HBI
4.68
4.46
4.41
3.98
4.45
3.81
# Tanytarsini Taxa
1.00
2.99
2.00
3.00
2.00
3.87
% Caenidae
0.00
0.00
0.00
0.00
0.00
0.40
% EPT (No Caenidae)
3.65
28.15
12.00
14.07
26.42
16.53
% dingers
40.10
54.20
57.33
73.37
51.22
65.73
% Filterers
26.56
28.99
39.11
50.75
29.27
55.24
Additional Biological Metrics
% Amphipoda
13.02
2.10
0.00
0.00
0.00
0.40
# EPT taxa
3.00
13.47
9.91
12.00
10.75
15.87
% Plecoptera
0.52
1.26
1.33
1.51
1.22
3.63
% Predators
28.65
6.30
12.00
7.54
5.28
14.52
% Sprawlers
26.56
9.66
10.67
9.05
10.57
7.66
Bolded metrics are those judged to be components of the impairment based on difference.
11
-------
tolerance value score for the sample, based on the individual tolerance values and
abundance of each taxon in the sample (Beck, 1954). A decrease in Beck's biotic index
(Bl) indicates an increase in stress. The Beck's Bl calculated from the sample taken
from Bogue Homo (18.0) is similar to the LD conditions for the East Bioregion (18.0)
and Site Class 6 (18.75). However, the Bogue Homo Beck's Bl value is lower than
most of the SSC station values.
1.4.4. Number of Ephemeroptera, Plecoptera, and Trichoptera (EPT) Taxa
This metric is the number of taxa representing three phylogenetic orders,
Ephemeroptera (mayflies), Plecoptera (stoneflies) and Trichoptera (caddis flies) (EPT).
In Bogue Homo, only 3.0 EPT taxa are found. This value is lower than the number of
EPT taxa found either in LD conditions (East: 6.95, Site Class 6: 7.56) or the SSC
stations (9.9-15.9).
1.4.5. Percent Ephemeroptera, Plecoptera, and Trichoptera (EPT) (No Caenidae)
This metric is the percentage of individuals representing the phylogenetic Orders
EPT, compared to the total number of individuals in the sample, excluding members of
the Family Caenidae (mayfly). The percentage of EPT (no Caenidae) from Bogue
Homo (3.65) is lower than LD condition percentages (East: 9.10, Site Class 6: 12.51)
and SSC station values (12-28).
1.4.6. Percent Plecoptera
This metric is the number of stoneflies (Plecoptera) compared to the total number
of individuals in the sample, expressed as a percentage. The percentage of Plecoptera
found in Bogue Homo is 0.52. This value is lower than LD conditions (East: 0.94, Site
Class 6: 1.13) and SSC stations (1.2-3.6).
1.4.7. Percent Amphipoda
This metric is the calculated percentage of amphipods (Order Amphipoda;
common name, scuds or side swimmers) relative to the total number of individuals in
the sample. In most small-to-medium-sized streams in Mississippi, amphipods
comprise a relatively small percentage of the total number of benthic
macroinvertebrates. The percentage of amphipods from Bogue Homo (13.0) is
substantially greater than the proportion of amphipods defined by LD conditions (East:
1.95, Site Class 6: 2.57) and from the samples of SSC stations (0-2).
1.4.8. Percent Predators
This metric is the percentage of organisms that are classified as "predators,"
relative to the total number of individuals in the sample. The term "predators" refers to a
type of functional feeding group classification (Merritt and Cummins, 1996). The
percentage of predators from the Bogue Homo sample (28.65) is greater than the
12
-------
percentage of predators in LD conditions (East: 16.28, Site Class 6: 19.08) and the SSC
stations (5.3-14.5).
1.4.9. Percent Sprawlers
This metric is the percentage of the total sample that is made up of organisms
that are classified as "sprawlers" (Merritt and Cummins, 1996). Sprawlers live on the
surfaces of leaves or fine sediments, and usually have body shapes or appendages
modified for staying on top of the substrate and maintaining respiratory surfaces free of
silt potentially giving them traits that make them more sediment tolerant. The
percentage of sprawlers from Bogue Homo (26.6) is greater than the corresponding
values for LD conditions (East: 19.57, Site Class 6: 15.39) and the SSC stations.
13
-------
2. LISTING THE CANDIDATE CAUSES
Developing the list of candidate causes involved consideration of sources of
stressors and the way in which they could lead to and cause the observed biological
impairment. It also included evaluation of LULC data for the Bogue Homo watershed at
various spatial scales; field observations during 2001; observations from a watershed
reconnaissance in 2004; and physical, chemical, and biological data from Bogue Homo.
Based on known linkages between the sources and proximate causes of impairment
information and ecological and watershed data, MDEQ cautiously eliminated unlikely
causes of impairment. In cases where data were lacking, the cause was not eliminated
unless there was overwhelming evidence to support the idea that the cause and/or
causal pathway in question could not contribute to the impairment. Data quality and
quantity were also considered when eliminating potential but unlikely causes of
impairment. Examples of reasons for eliminating a potential cause included lack of data
of sufficient quality, quantities that were not different than those of background or LD
conditions, and if the stressor or a source of the stressor was believed to be either
mechanistically implausible or absent from the watershed (see Comment 4).
2.1. MISSISSIPPI'S STANDARD LIST OF CAUSES OF IMPAIRMENT
MDEQ developed a standardized
list of causes of impairment commonly
encountered in streams and rivers
throughout Mississippi. The purpose of
creating this list was to avoid accidentally
omitting possible causes. The standard
list provided consistency to MDEQ's
causal analysis process on the impaired
waters from the §303(d) List of Impaired
Water Bodies. Although the list was most
likely not all-inclusive, it was
comprehensive enough for these
applications, as it was based on the
knowledge and experience of several
MDEQ scientists and engineers who have dealt with streams and water quality issues
for many years. The standard list of prospective causes of impairment developed for
Mississippi streams, and their associated monitoring indicators, is given in Table 3.
2.2. PROCESS OF ELIMINATION
Elimination of very unlikely stressors was recommended in the U.S. EPA
Stressor Identification Guidance Document (U.S. EPA, 2000) to keep the causal
analysis process from becoming unmanageable. To refine the stressor list early on,
very unlikely causes of impairment were eliminated. The elimination process involved
the use of LULC data, NPDES discharger source locations, aerial photography, and
Comment 4. Elimination Versus Deferment.
The Bogue Homo River assessment was
based on the original SI process (U.S. EPA,
2000). Updates to the process more clearly
distinguish elimination, a logical disproof of a
candidate cause, from deferment, a
postponement of assessment.
In the Bogue Homo assessment, lack of data of
sufficient quality required deferment. Evidence
that the candidate cause was at background
levels or was mechanistically implausible was
considered reasons for elimination. The
U.S. EPA recommends eliminating causes later
during analysis rather than when planning the
assessment to ensure that the rationale and
evidence for elimination is fully documented.
14
-------
TABLE 3
MDEQ Standard List of Candidate Causes and More Related Processes
Standard Candidate Causes
Related Processes
Oxygen Concentration and Oxygen
Demand
Alteration to photosynthesis/respiration balance
Decrease in oxygen and increase in oxygen demand
Temperature alteration
Alteration to natural flow regime
Increase in organic enrichment
Disruption of nutrient cycles
Organic Carbon
Increase in organic enrichment
Change in food source characteristics
Alteration to photosynthesis/respiration balance
Decrease in oxygen and increase in oxygen demand
Nutrients
Disruption of nutrient cycles
Alteration to photosynthesis/respiration balance
Decrease in oxygen and increase in oxygen demand
Change in food source characteristics
Temperature
Alteration to thermal regulation
Alteration to photosynthesis/respiration balance
Decrease in oxygen and increase in oxygen demand
Turbidity
Increase in amount and duration of suspended sediment
Alteration to natural flow regime
Decrease in suitable in-stream habitat
Alteration to channel morphology
Decrease in riparian vegetation
Sediment Particle Size
Increase in deposited sediment
Alteration to natural flow regime
Decrease in suitable in-stream habitat
Alteration to channel morphology
Habitat Evaluation
Alteration to natural flow regime
Decrease in suitable in-stream habitat
Decrease in riparian vegetation
Alteration to channel morphology
Increase in suspended and deposited sediment
Change in food source characteristics
Decrease in suitable floodplain habitat
Temperature alteration
Alteration to groundwater interaction
Conductivity, TDA and Chlorides
Increase ion concentrations
Alteration to natural freshwater/saltwater interaction
PH
Increase in alkalinity or decrease in hydrogen ion activity
Increase in acidity or hydrogen ion activity
None
Increase in toxic substance concentrations
15
-------
field reconnaissance information. Examples of reasons to eliminate a specific cause
during this process included: the causal pathway was believed to be mechanistically
implausible, or the causal agent and/or source for the cause did not occur within the
watershed (see Comment 5).
Based on the proximity of the
potential sources of stress to the Bogue
Homo sampling station and evaluation of
watershed characteristic data, only one
candidate cause was eliminated:
extremes of acidity or alkalinity. Change
in natural activity of hydrogen ions (pH) in
the water was used as the indicator for
this cause. Values of pH were at
background levels in Bogue Homo, due
to naturally acidic soils (a common,
statewide occurrence) or blackwater
streams, which are characterized in part
by a naturally low pH (i.e., pH 4.5-6.0).
For the purpose of developing the
final list of candidate causes, MDEQ
considered causes of two types:
proximate causes and intermediate
causes. Proximate causes are defined
as the immediate and effective cause(s)
of the biological impairment(s), such as low DO leading to asphyxiation. Intermediate
causes are those that are part of the causal pathway, but do not directly lead to the
immediate cause of impairment, such as a lack of riffles causing reduced aeration and
decreased DO. Figure 5 shows the final list of candidate causes of impairment for
Bogue Homo, with their link(s) to their respective proximate causes.
2.3. CONCEPTUAL MODEL OF CAUSAL PATHWAYS
After development of the candidate list of causes of impairment, a conceptual
model was developed that outlined the plausible relationships between potential
sources of stress, intermediate causes, proximate causes, and the biological response
variables. The conceptual model was used to further verify the list of candidate causes
through visual observation. In addition, it was used during subsequent evaluation of
candidate causes to identify probable causes of impairment. The conceptual model for
Bogue Homo is shown in Figure 6.
Comment 5. Listing Candidate Causes.
MDEQ's approach to eliminating candidate
causes has been effective because they were
very conservative when eliminating them. The
process of elimination involves analyzing types of
evidence during the planning phase. In essence,
the assessors listed a cause and performed a de
facto analysis with a type of evidence they
believed would refute the candidate cause,
namely spatial co-occurrence. Then, they
returned to listing and evaluating candidate
causes. In either approach, the evidence for
refuting a cause is collected, analyzed,
evaluated, and documented. The use of a
standard list of potential causes is an excellent
idea, but during planning, other case specific
causes should also be hypothesized. Also,
candidate causes should not be eliminated
without very strong proof. The list is better used
as a tool to prompt assessors to add candidate
causes to the list that are plausible, rather than to
eliminate them. Other lists may be found on the
Causal Analysis/Diagnosis Decision Information
System (CADDIS) Web site.
16
-------
Intermediates of Causal Pathway Proximate Candidate Causes
Increase in ion concentrations
Increase in toxic substance concentrations
Decrease in Dissolved Oxygen
Alteration to Thermal Regulation
Alteration to Food Source
Decrease in Suitable habitat
Increase in Ionic Strength
Increase in Toxic Substances
Alteration to photosynthesis/respiration balance
Decrease in oxygen and increase in oxygen demand
Increase in organic enrichment
Disruption of nutrient cycles
Change in food source characteristics
Alteration to natural flow regime
Decrease in suitable in-stream habitat
Alteration to channel morphology
Decrease in riparian vegetation
Increase in suspended and deposited sediment
Decrease in suitable floodplain habitat
Alteration to groundwater interaction
FIGURE 5
Causal Pathways Related to Proximate Causes Analyzed in this Case
17
-------
00
Roads/Bridges
Historic Landscape Alteration
Lumber Yard
Unsewered Areas
Silviculture
Point Sources
Impoundment
Pasture
Poultry
Increased salts
and minerals
Increase nutrients
(NN, TKN, TP)
Decrease
riparian vegetation
Increase organic matter
(TOC, BOD, COD)
Altered water
temperature input
Flow
alteration
Increase
solar input
Increase
bed scouring
algal bioi
Increase
bank erosion
Increase microbial
respiration
Decrease
allocthonous
organic matter
Increase
suspended and
deposited sediment
Increase in toxic
substances
% Clingers, % Sprawlers,
. % Predators ^
# EPT taxa,
% Plecoptera, HBI,
Vo Amphipoda, # Tanytarsini taxa
Increased Ionic
Strength
Alteration of food
Decrease in suitable
habitat
Decrease in
dissolved oxygen
Alteration to
thermal regulation
Sources of
Stress
Causal
Pathway
and
Intermediate
Causes
Proximate
Stressors
Effects
FIGURE 6
Conceptual Model for Bogue Homo
-------
3. EVALUATE DATA FROM THE CASE
Two types of evidence using data from the case were employed in the causal
analysis of the Bogue Homo:
1. Spatial/Temporal Co-occurrence of proximate causal agents and
biological effects.
2. Causal Pathway, which includes the occurrence of intermediate causal
agents that are components of the hypothetical causal pathway.
3.1. SPATIAL/TEMPORAL CO-OCCURRENCE OF PROXIMATE CAUSAL
AGENTS AND BIOLOGICAL EFFECTS
This type of evidence is an evaluation of the coincidence in space and time of the
impairment and the individual candidate causes (see Comment 6). The candidate
cause is not responsible for the impairment if it is not present at the impaired site. The
candidate cause is supported if it is present at the impaired site. MDEQ's method for
determining whether causal agents co-occurred with the impairment was to compare
water quality data from the impaired portion of the Bogue Homo to LD conditions and
SSC values from the same time period. Water quality data used were in the form of
indicator parameters of specific proximate causes. Comparison results used for
evaluation of spatial co-occurrence are presented in Table 4.
Comment 6. Spatial/temporal Co-occurrence from the Case and from Elsewhere
Spatial/temporal co-occurrence provides evidence that the biological effect was observed where and
when the cause was observed, and was not observed where and when the cause was absent. When
considering spatial co-occurrence from the case, the impairment and stressor at the site is compared to
sites within a close geographic proximity, ideally from a location within a few kilometers from the site.
Preferably, measurements of the proximate stressor and the effect are also simultaneously collected at a
site on the same day. Likewise, comparisons among sites should be made with data that were collected
within a reasonably similar time frame, using similar methods. Pairing data for location and date is a
matter of judgment, but generally stream reaches are compared to nearby or upstream reaches,
watersheds are compared to other watersheds, and multiple watersheds to other sets of watersheds.
For the Bogue Homo case, the MDEQ compared the Bogue Homo site to other nearby sites, SSCs. We
would consider this as evidence from the case. The MDEQ also compared the impaired site to LD
conditions from a larger geographical region. We would consider this as evidence from elsewhere,
another separate and important type of evidence. This is useful for increasing confidence that
co-occurrence seen with limited sampling at local sites is also reflected in a larger population of sites.
3.1.1. Decrease in Suitable Habitat
Physical habitat was evaluated by visually assessing ten habitat parameters that
describe in stream habitat quality and quantity, channel form and stability, and riparian
condition. These visual assessments were calculated as scores for each of the ten
parameters, and a total habitat score was calculated as the sum of the ten scores. The
19
-------
TABLE 4
Water Quality Data, Bogue Homo and Comparison Conditions
Difference from LD
condition
Least Disturbed Condition
Chemical Parameters
Bogue Homo
East
Bioregion
Site Class 6
East
Bioregion
Site Class 6
percentile used for
LD
Turbidity
8.91
lower
lower
17.65
20.50
75th
Dissolved Oxygen (% saturation)
86.4
lower
lower
91.40
93.40
25th
Dissolved Oxygen (mg/l)
8.7
lower
lower
10.20
10.10
25th
PH
6.25
higher
lower
6.20
6.50
75th
Water Temperature (Celsius)
15.11
NA
NA
17.65
7.80
50th
Ammonia (mg/l)
0.3
higher
higher
0.22
0.20
75th
Nitrate-Nitrite (mg/l)
0.06
lower
lower
0.21
0.30
75th
Total Kjeldahl Nitrogen (mg/l)
0.79
higher
higher
0.64
0.70
75th
Total Phosphorus (mg/l)
0.04
lower
lower
0.07
0.10
75th
Chemical Oxygen Demand (mg/l)
36
higher
higher
21.00
20.80
75th
Total Organic Carbon (mg/l)
12
higher
higher
7.00
7.00
75th
Total Chlorides (mg/l)
30.5
higher
higher
6.00
6.40
75th
Total Dissolved Solids (mg/l)
98.8
higher
higher
27.30
29.90
75th
Physical Parameters
Total Habitat Score
125
lower
lower
155.00
151.30
25th
Instream Habitat Score
31
lower
lower
43.50
38.75
25th
Morphological Habitat Score
55
lower
lower
64.00
59.75
25th
Riparian Habitat Score
39
lower
lower
45.50
44.50
25th
% Silt/Clay
10
lower
lower
11.00
11.00
50th
% Sand
90
higher
higher
72.00
57.50
50th
% Hardpan Clay
0
comparable
comparable
0.00
0.00
50th
% Gravel
0
comparable
lower
0.00
9.50
50th
Site Specific Comparators
Chemical Parameters
Bogue Homo
Black Creek
Bogue Homo
(downstream)
Big Creek
Oakahay
Creek
Yellow Creek
Turbidity
8.91
8.00
10.00
9.00
18.00
8.00
Dissolved Oxygen (% saturation)
86.40
87.80
99.30
91.20
91.20
93.10
Dissolved Oxygen (mg/l)
8.70
9.05
11.01
9.60
9.55
9.90
PH
6.25
6.10
6.37
6.46
6.47
6.90
Water Temperature (Celsius)
15.11
14.03
10.70
13.04
13.29
13.50
Ammonia (mg/l)
0.30
0.21
0.10
0.10
0.10
0.21
Nitrate-Nitrite (mg/l)
0.06
0.12
0.14
0.50
0.35
0.17
Total Kjeldahl Nitrogen (mg/l)
0.79
0.85
0.68
0.48
0.68
0.53
Total Phosphorus (mg/l)
0.04
0.07
0.12
0.04
0.07
0.10
Chemical Oxygen Demand (mg/l)
36.00
19.00
29.00
10.00
22.00
19.00
Total Organic Carbon (mg/l)
12.00
9.00
10.00
3.00
9.00
6.00
Total Chlorides (mg/l)
30.50
10.70
22.80
8.20
6.40
50.20
Total Dissolved Solids (mg/l)
98.80
59.15
239.20
35.75
46.15
130.65
Physical Parameters
Total Habitat Score
125.00
169.00
126.00
116.00
137.00
184.00
Instream Habitat Score
31.00
52.00
44.00
33.00
53.00
57.00
Morphological Habitat Score
55.00
66.00
43.00
48.00
58.00
77.00
Riparian Habitat Score
39.00
51.00
39.00
35.00
26.00
50.00
% Silt/Clay
10.00
9.09
18.18
27.00
13.00
6.00
% Sand
90.00
57.58
46.46
66.00
68.00
83.00
% Hardpan Clay
0.00
0.00
0.00
0.00
0.00
11.00
% Gravel
0.00
33.33
35.35
7.00
19.00
0.00
Bolded parameters are those judged to possibly give information as to cause of impairment based on
difference. The 75th percentile was used for parameters that, as they increased, biological quality
decreased (e.g., ammonia). The 25th percentile was used for parameters that, as they increased, the
biological quality increased (e.g., dissolved oxygen).
20
-------
scale of the total habitat score was from 0 to 200, with increasing scores reflecting an
increase in habitat condition. The total habitat score from Bogue Homo (125) was lower
than most SSC station habitat scores (MDEQ, 2001). However, a monitoring station on
Bogue Homo downstream of the impaired site had an almost identical total habitat score
(126) as the impaired site, but was assessed as nonimpaired based on its M-BISQ
score. Although MDEQ reduces sampler subjectivity by training and through the use of
standard operating procedures, sampling variability is an inherent factor of any
qualitative assessment method and may be the reason for this lack of consistent
co-occurrence.
Suspended sediment was considered a component of habitat. However, turbidity
values from Bogue Homo do not suggest that elevated suspended sediment co-occur
with the impairment. Turbidity was low in 2001 (8.9 nephelometric turbidity units [NTU])
and is lower than LD condition values and most SSC station values in 2001, including
the downstream (nonimpaired) Bogue Homo site (10.0 NTU). Turbidity data from
Bogue Homo collected in 1997 were lower than or comparable to turbidity values for the
LD condition and the SSC stations. However, turbidity measurements can vary
dramatically with flow. Furthermore, base-flow conditions sometimes have low turbidity
levels, even in cases where suspended sediment has a negative effect on the biological
community. Measured levels of turbidity are therefore time-dependent. MDEQ
recommends collecting biological samples during base-flow conditions (i.e., not
immediately after an intense precipitation event, which can cause the water level to
fluctuate). To help ensure representative sampling, MDEQ sampled Bogue Homo only
during base-flow conditions (see Comment 7). Thus, the turbidity data may not
adequately characterize storm-related
suspended sediment problems, if such
problems exist.
Particle size-distribution data for
Bogue Homo substrates (10% silt/clay, 90%
sand) also did not suggest sediment as a major potential cause of impairment.
However, in the habitat assessment scoring procedure, the field crew observed, scored
and recorded the amount of sediment deposition as "not optimal."
In summary, low habitat quality and the
impairment of Bogue Homo do not
consistently co-occur, but this result is
uncertain (see Comment 8).
3.1.2. Altered Temperature Regime
Evaluation of co-occurrence of altered
temperature regime and biological effects was
problematic, because some measurements of
temperature were made in the winter, and the
collection methods did not address diel
Comment 7. Episodic Events.
A peer reviewer indicated that most
monitoring programs reflect base flow, rather
than episodic events. Historical water quality
might not indicate the importance of episodic
events in biological impairment.
Comment 8. Substrate Texture.
Percent sand at the impaired site (90%)
was higher than at any SSC site (46.5-83%)
and higher than the regional LC conditions
(57 and 72%). Percent gravel (0%) was
lower than all but one SSC site (0-35.3%)
and lower than the Site Class 6 LC
conditions (9.5%). Hence, if a sand
substrate with no admixture of gravel was
defined as a candidate cause, it could be
said to co-occur with the impairment. This is
an example of the use of specific causes
rather than the more inclusive candidate
cause of suitable habitat.
21
-------
variability. However, temperatures were higher in the impaired segment of Bogue
Homo than in SSC sites. The mechanism for increased temperature was not identified.
3.1.3. Altered Food Resources and/or Decreased Dissolved Oxygen
Concentrations of DO from Bogue Homo in 2001 were 8.7 mg/L (86.4%,
saturation). These values were lower than any of the SSC sites (9.0 to 11.0 mg/L and
87.8 to 99.3% saturation, respectively). Of the SSC stations, the nonimpaired Bogue
Homo downstream site had the highest levels of DO (11.0 mg/L, 99.3% saturation).
The chemical oxygen demand (COD) value (36 mg/L) for the impaired site, in 2001, was
greater than all SSC sites (COD range, 10-29 mg/L). In the summer of 1997, at the
Bogue Homo impaired site, DO was low (5.5 mg/L, 68.6% saturation) and COD levels
were elevated (29 mg/L) in the summer.
Food resource availability and composition can also have a significant direct and
indirect impact on stream metabolism, including primary and secondary production and
oxygen concentrations. No direct indicators of food resource availability or composition
are available. However, the low DO and high COD are suggestive of a different trophic
status at the impaired site.
3.1.4. Increase in Ionic Strength and/or Increase in Toxic Substances
Total dissolved solids (TDS) concentration at the impaired site are within the
range of SSC sites, and are less than half those of the unimpaired downstream site.
Total chlorides were greater at the impaired Bogue Homo site than at all but one of the
five SSC sites. Historical data indicate slightly elevated total chlorides at the impaired
Bogue Homo site, but not elevated TDS or specific conductance values. Water from the
impaired site was not tested for toxicity, and no historical data are available for toxic
substance concentrations at this site. Hence the impairment does not consistently
co-occur with these limited measures of chemical contamination. However, given the
paucity of measurements of potentially toxic chemicals, this candidate cause was not
scored.
3.2. CAUSAL PATHWAY
A causal pathway is the sequence of events that begins with the release or
production of a stressor from a source, and ends with a biological response. To
determine a complete causal pathway, MDEQ evaluated indirect causes, intermediate
steps, and the presence of sources that could account for the indirect causes leading to
the proximate causes.
3.2.1. Decrease in Suitable Habitat
The causal pathway to decreased habitat quality consists of two components:
hydrology and sediment.
22
-------
3.2.2. Sources of Hydrologic Alteration
There was little evidence to support an altered hydrological regime as a
component of the causal pathway to impairment of Bogue Homo. There was evidence
that the main channel was connected to the floodplain. Only minor erosion/bank
instability was noted in 2001 and during a reconnaissance visit in 2004. A large
impoundment (Lake Bogue Homo) is located above the impaired site, but the effects, if
any, of that impoundment on hydraulic conditions at the impaired site are unmeasured
and the lake is more than 40-km from the site.
3.2.3. Sediment Sources
Potential sources of alteration in sediment transport in the Bogue Homo include
land-disturbing activities that cause upland sediment runoff and bank failure, water level
fluctuations, and sediment barriers. Soils in the Bogue Homo watershed tend to be
naturally erodible. Watershed features that could increase inputs of sediment to the
impaired site include silviculture operations, direct access of livestock to the stream,
major drainage ditches along Highway 15, and commercial/residential development
outside of the cities of Laurel and Ellisville. Lake Bogue Homo traps sediment, but the
hydrologic effects of its release frequency and rate might increase bank and channel
erosion. Significant channel evolution can continue for many years after geomorphic,
hydrologic, or floodplain changes.
3.2.4. Altered Temperature Regime
Two potential sources of thermal alteration to Bogue Homo are identified: Lake
Bogue Homo, and decreased riparian canopy cover. Lake Bogue Homo is located
approximately 47-km upstream of the monitoring station on Bogue Homo. Therefore, it
is unlikely that Lake Bogue Homo has an effect on the thermal regime of the impaired
segment, but no data exist to evaluate this. Riparian canopy disturbance is observed
upstream of the impaired site and may allow for increased exposure of the stream to
solar radiation. The importance of this possibility, too, could not be determined with
available measurements.
3.2.5. Altered Food Resources and/or Decreased Dissolved Oxygen
Dissolved oxygen is depressed by decomposition of dissolved and particulate
organic matter and at night by algal respiration. Hence, dissolved and particulate
organic matter and nutrients that promote algal growth are components of the causal
pathway to low DO. In addition, they are components of the causal pathway leading to
altered food resources.
3.2.5.1. Sources.
Because predominant land uses in the watershed are not agricultural, agricultural
practices are probably not a primary source of excess nutrients or organic enrichment.
23
-------
However, some poultry and cattle operations are observed in the watershed.
Residential growth around the Laurel-Ellisville area, increased application of fertilizers in
residential areas, septic tank leachate, the few point sources located upstream of the
M-BISQ site, and Lake Bogue Homo itself are potential sources of elevated nutrients
and organic enrichment.
3.2.5.2. Organic Matter.
In 2001, the total organic carbon (TOC) level measured from Bogue Homo
(12 mg/L) is greater than the LD condition values and all SSC station values for TOC.
However, historical TOC values from Bogue Homo are not higher than comparison
conditions.
3.2.5.3. Nutrients.
Nutrient (nitrogen and phosphorus) concentrations from Bogue Homo during
2001 provided varying results in comparison to LD conditions and SSC station values,
depending on specific water quality indicators. Concentrations of total Kjeldahl nitrogen
(TKN) (0.8 mg/L) and ammonia (0.3 mg/L) were higher than LD condition values and all
SSC station values. However, nitrite-nitrate and total phosphorus values were low, in
comparison to the LD conditions and all SSC station values. In addition, all historical
nutrient data were lower than LD condition levels.
3.2.6. Increase in Ionic Strength and/or Increase in Toxic Substances
Causal pathway components regarding toxic substances were not considered
due to a lack of data. However, the sources noted for nutrients, above, could also
potentially contribute toxic substances to Bogue Homo River.
24
-------
4. EVALUATE DATA FROM ELSEWHERE
Two types of evidence using data from elsewhere were used in the causal
analysis of the Bogue Homo:
1. Mechanistically Plausible Cause, which is based on evidence from other
studies or basic biological principles, and
2. Stressor-Response Relationships from Other Field Studies, which is
based on a regional database.
4.1. MECHANISTICALLY PLAUSIBLE CAUSE
This type of evidence evaluates the plausibility that the effect resulted from the
cause, given what is known about the mechanisms involved in biotic and abiotic
interactions and the environment in which the interactions occur. This process involves
a logical evaluation of the ways in which a particular stressor could affect the biological
community, at the site where the impairment occurs. The effects can be direct (e.g.,
toxic) or indirect (e.g., food chain, energy regime) (see Comment 9). It does not
necessarily provide Bogue Homo specific evidence, rather it documents scientific
knowledge that the candidate cause could occur and result in the types of impairments
noted in the case.
4.1.1. Decrease in Suitable Habitat
Habitat alterations can affect
community composition, and habitat
alteration is considered to be one of the
major stressors on aquatic systems (Karr
etal.,1986). Poor-quality or
homogenous stream habitat reduces the
amount of cover, promotes instability,
and changes the sources and availability
of food. In addition to direct physical
habitat alteration, intermediate causal pathway components capable of contributing to
unsuitable habitat include increased suspended sediment, elevated turbidity, and
altered hydrologic processes. Increases in turbidity or suspended sediments can
interfere with oxygen uptake by clogging gills, reduce primary production by lowering
water clarity, alter trophic structure, and increase temperature. Each of these situations
can adversely impact benthic macroinvertebrate communities. Flow alteration,
especially decreased flow and channelization, detrimentally affect in-stream biota and/or
alter community composition (Hart and Finelli, 1999).
Comment 9. Not All Evidence is Equally
Informative.
A mechanistically plausible cause is a weak
argument unless there is actual evidence that a
mechanism occurs at the site or is impossible at
the site or in general. If all the candidate causes
are mechanistically plausible, it does not alter the
outcome of the assessment but simply
documents that none are implausible. A
statement to that effect can help shorten the
report and can be briefly included when giving
reasons for listing candidate causes.
25
-------
4.1.2. Altered Temperature Regime
The changes in abundance of taxa and functional groups that characterize the
impairment may be affected directly by natural or anthropogenic alterations of stream
temperature, because of differences in the temperature tolerances of aquatic
invertebrates, and indirectly by the effects of temperature on in-stream primary
production and decomposition (Vannote and Sweeney, 1980; Galli and Dubose, 1990).
4.1.3. Altered Food Resources and/or Decreased Dissolved Oxygen
Organic matter, other than leaf detritus, affects aquatic invertebrates in several
ways. Organic matter from inadequate waste treatment is a different energy source
than leaf biofilms that may change the relative abundances of species promoting
filterers and scrapers rather than shredders. Organic matter may promote algal and
bacterial growth and biomass which is also a different food source that can lead to a
change in the invertebrate assemblage. Furthermore, untreated organic matter or
decaying algae can cause episodic levels of low DO that directly affect aquatic
invertebrate abundances through asphyxiation.
4.1.4. Increase in Ionic Strength and/or Increase in Toxic Substances
Increased chemical concentrations and ionic strength can cause lethal and
sublethal toxic effects in benthic organisms, and can change community structure (see
Comment 10).
4.2. STRESSOR-RESPONSE RELATIONSHIPS FROM OTHER FIELD STUDIES
This type of evidence consists of
comparisons of Bogue Homo data to
regional stressor-response relationships.
Stressor-response relationships were
examined in two ways:
1. Scatter plots of biological metrics and
abiotic parameters from sites in Site
Class 6.
2. Box plot distribution of biological data
from the East Bioregion with low and
high ranges of abiotic values.
4.2.1. Scatter Plots
Comment 10. Mechanistic Specificity.
Since particular toxic substances are not
specified, it is not possible to be more specific
about mechanistic plausibility. For example,
since amphipods are relatively salt tolerant, the
ionic strength component of this candidate cause
is consistent with the observed impairment, which
includes increased amphipod abundance.
Amphipods are, however, sensitive to
cholinesterase-inhibiting pesticides, which are
also potentially part of this candidate cause.
Hence, without a more specifically defined cause,
we can go no further than to say that increased
chemical concentrations are a plausible
mechanism of benthic invertebrate community
impairment in the Bogue Homo.
Associations in other field studies between candidate causes and the effects that
characterize the impairment can support a causal relationship at the site. Because of
the potential for extraneous differences among sites, stressor-response relationships
were not derived for comparison to quantitative relationships at the site. Rather, linear
26
-------
correlations were used to determine the strength and sign of associations of biotic and
abiotic variables in Site Class 6.
Three steps were taken to identify specific stressor-response relationships that
are potentially indicative of causation. The first step was to derive the Pearson product
moment correlation coefficient (r) for combinations of biological metric values and
various physical and chemical values in Site Class 6. The total number of discrete
sampling events used to derive correlation coefficients and scatter plots was 63
(n = 63). The correlation matrix of all data is given in Appendix C. Most correlations
were weak, <0.25. Using a selection value of either 40 or 35 included the same
candidate causes. By choosing pairs that were >0.35, a manageable number of
associations could be examined (see Comment 11).
The second step was to plot the
data as scatter plots and the linear model
for those pairings of variables with a
positive or negative r-value of >0.35.
Scatter plots were visually examined for
the nature and distribution of data for
each pairing of variables. Scatter plots
whose correlation values appeared to be
heavily influenced by a few outlier data
that, if not included, would result in a
much smaller r-value, were not
considered for the third step.
The third step was to look at Bogue Homo biological values plotted against the
biological pairing of variables considered to be of positive or negative correlation and of
potential meaning. The consistency of study site values was compared to biotic and
abiotic variables from Site Class 6. The Pearson correlation coefficients of scatter plots
are shown in Table 5. A summary of the conclusions drawn based on comparison of
Bogue Homo values to select scatter plot curves is shown in Table 6, and the scatter
plots used for this evaluation, with plotted Bogue Homo values, are shown in
Appendix D.
4.2.2. Box Plot Distributions
Bioregion-wide field data were used to evaluate exposure-response relationships
and determine whether they support or weaken evidence for candidate causes (see
Comment 12). The data within the bioregion data were confounded by other stressors;
therefore, it was judged that analysis of functional relationships, like linear correlations
applied to regional data, would not be appropriate. Rather, biological effects of extreme
high and low values of variables related to the causes were examined.
For specific abiotic variables, MDEQ identified low and high value using best
professional judgment. Low and high value ranges were identified by examining the
Comment 11. Selecting Evidence of Causal
Relationships.
Alternatively, the strongest correlations could
be selected for evaluation and may have
reflected the sensitivities of each metric to
particular physical or chemical characteristics.
Correlations are not used for hypothesis testing,
but as a descriptive statistic to select
associations for further analysis. For most
environmental data, the rank correlation is
usually more suitable (Spearman) rather than
Pearson parametric method.
27
-------
TABLE 5
Pearson Correlation Coefficients (r-values) of Biological Metrics and Physical/Chemical
Parameters Used for Scatter Plot Evaluation (n = 63)
Total Organic
Carbon (mg/L)
Chemical Oxygen
Demand (mg/L)
Total Nitrogen
(mg/L)
Total Phosphorus
(mg/L)
Total Chlorides
(mg/L)
Total Dissolved
Solids (mg/L)
Turbidity (NTU)
Silt Substrate
(% Total)
Gravel Substrate
(% Total)
M-BISQ Score
-0.58
-0.50
-0.46
-0.36
-0.24
-0.30
-0.57
-0.39
0.53
# Chironomidae Taxa
-0.24
-0.09
-0.25
-0.37
-0.48
-0.48
-0.09
-0.21
0.05
# Oligochaeta Taxa
0.17
0.28
0.32
0.09
0.03
0.02
0.21
0.37
-0.32
# EPT Taxa
-0.53
-0.52
-0.36
-0.20
-0.09
-0.19
-0.42
-0.31
0.36
# Filterer Taxa
-0.44
-0.33
-0.23
-0.26
-0.31
-0.41
-0.36
-0.35
0.52
# Plecoptera Taxa
-0.33
-0.38
-0.41
-0.09
0.18
-0.11
-0.32
-0.18
0.17
# Collector Taxa
-0.07
0.08
0.08
-0.13
-0.46
-0.40
0.10
-0.10
-0.11
# Diptera Taxa
-0.16
0.02
-0.18
-0.33
-0.50
-0.50
-0.03
-0.20
-0.01
% N on insects
0.44
0.41
0.49
0.30
0.06
0.08
0.44
0.44
-0.31
% Caenidae
-0.06
-0.15
-0.18
-0.06
0.46
0.46
-0.01
-0.10
-0.22
Beck's Biotic Index
-0.39
-0.38
-0.39
-0.30
-0.22
-0.33
-0.51
-0.32
0.30
Hilsenhoff Biotic Index
0.40
0.28
0.32
0.38
0.39
0.46
0.51
0.47
-0.59
Bold values are /"-values of associations used in the evaluation.
28
-------
TABLE 6
Summary of Stressor-Response Relationships from Linear Correlation of Various Biotic
Variables Against Variables Associated with Candidate Causes
Total Organic
Carbon (mg/L)
Chemical
Oxygen Demand
(mg/L)
Total Nitrogen
(mg/L)
Total
Phosphorus
(mg/L)
Total Chlorides
(mg/L)
Total Dissolved
Solids (mg/L)
Turbidity (NTU)
Silt Substrate (%
Total)
Gravel Substrate
(% Total)
M-BISQ
+ +
+ +
+
-
+
# Chironomidae taxa
+
# Oligochaeta taxa
# EPT taxa
+ +
# Filterer taxa
+
# Plecoptera taxa
+
# Collector taxa
+
# Dipteran taxa
0
% Noninsect individuals
0
+
-
% Caenidae individuals
-
Beck's Biotic Index
Hilsenhoff's Biotic Index
-
Summary
+
+ +
+
-
+
-
+
++ = strong evidence as cause
+ = week evidence as cause
0 = unclear
= weak evidence as not cause
= strong evidence as not cause
29
-------
Comment 12. Box Plots.
The use of box plots to create exposure-response relationships from field data and relate them to site
data might be clarified by further explanation. First, this use presumes a monotonically increasing or
decreasing relationship. If conditions are optimal at an intermediate level of the causal agent and sub-
optimal at the extremes, the technique is inappropriate. Second, box plots support a candidate cause if
the following are true:
1. Boxes do not overlap. If they do overlap, it suggests that they are not a significant cause in the
region because the levels of response do not greatly differ between extremely high and low
levels of the candidate cause.
2. The boxes are small. If the cause were important, one would expect a relatively consistent
response. This is a weaker criterion than 1.
3. The response at the impaired site falls on the box for the extreme level predicted by the causal
hypothesis. For example, the number of EPT taxa at the impaired site falls within the inter-
quartile range of EPT taxa for extremely high turbidity sites in the region, which supports
turbidity as a cause.
4. The level of the candidate cause at the impaired site falls within the appropriate regional
extreme range. That is, if the biological response at the impaired site corresponds to the levels
seen at an extreme level of the candidate cause (i.e., Criterion 3 is met), then the level of the
candidate cause should be extreme at the impaired site.
5. Criteria 1-4 are true for most, if not all, of the response metrics that define the impairment. That
is, if the candidate cause is responsible for the impairment, it should be associated with most or
all of its components.
MDEQ applies Criteria 1-3 in their analysis to obtain the results in Table 8, thus inspiring us to more
fully describe the use of box plots in causal analysis. The MDEQ analysis provides evidence to
strengthen or weaken candidate causes, but adding the other two criteria would provide more complete
evidence. For example, the number of EPT taxa and the percent crustacean/mollusk individuals are
clearly differentiated at extreme turbidity levels (1-2), and the levels of those biological metrics are as
expected for high turbidity levels (3), but the turbidity levels at the impaired site are not high (4). These
conditions suggest that turbidity is not the only cause of the impairment that is discussed by MDEQ in a
later section entitled "Identify the Probable Cause."
entire data set for each parameter and selecting ranges on the extreme ends. Among
the factors considered were: the amount of data in the range, and judgment as to
whether the range of values would be considered low or high based on existing water
quality criteria, literature, and professional judgment. For each abiotic parameter,
stations with data falling in the established low and high ranges were identified.
Biological metric data from these stations were displayed using two box plots, one
showing the range of biological metric data from stations in the high range and one
showing the range of biological metric data from stations in the low range. The pairings
of box plots for each biological metric were visually evaluated as to whether the metric
response coincided with the abiotic categorization. Evaluation of the box plots involved
noting whether the 25th and 75th percentile values overlapped, and if not, how close they
were. In addition, the spread of the inter-quartile range of individual boxes was
evaluated, with large inter-quartiles spreads being interpreted as high variability in
response, resulting in less confidence in that particular metric. Table 7 gives abiotic
parameters evaluated; low and high range values; number of data used for each range;
and corresponding biological metrics. The behaviors of sites in the bioregion were
compared to Bogue Homo data.
30
-------
TABLE 7
Parameters Evaluated Using Box Plots
Parameter
Low range
n value
High range
n value
Biological metrics with good
response
Total Organic Carbon (mg/l)
<1.0
53
> 10.0
52
% Tanytarsini individuals
% Non-Insect individuals
# EPT taxa
# Crustacean/Molluscan taxa
Chemical Oxygen Demand (mg/l)
< 10.0
298
> 30.0
47
# EPT taxa
% Non-Insect individuals
Dissolved Oxygen (% Saturation)
< 85%
64
100-103%
146
# Ephemeroptera taxa
% EPT individuals
Total Nitrogen (mg/l)
< 0.3
65
>2.0
44
M-BISQ
Beck's Biotic Index
% Tolerant individuals
% Intolerant individuals
% Filterer individuals
# Clinger taxa
# EPT taxa
# Plecoptera taxa
# Orthocladinae taxa
Total Phosphorus (mg/l)
<0.01
45
> 0.30
30
M-BISQ
Hilsenhoffs Biotic Index
% Tolerant individuals
% Intolerant individuals
% Tanytarsini individuals
% Oligochaeta individuals
# EPT taxa
# Orthocladinae taxa
% Filterer individuals
% Clinger individuals
Total Chlorides (mg/l)
< 3.0
99
> 50.0
20
Beck's Biotic Index
# Chironomidae taxa
Turbidity (NTU)
< 5
74
>70
33
M-BISQ
Hilsenhoffs Biotic Index
% Tolerant individuals
% Intolerant individuals
% Dipteran individuals
% Crustaceans/Molluscan individuals
% Collector individuals
% Filterer individuals
% Clinger individuals
# EPT taxa
Silt substrate (percent total)
< 5
162
>75
56
# Crustacean/Molluscan taxa
# Tanytarsini taxa
# Clinger taxa
Total Habitat Score
<70
54
> 160
67
M-BISQ
NCBI
% Tolerant individuals
% Intolerant individuals
# Total taxa
# Plecoptera taxa
% Filterer individuals
# Clinger taxa
31
-------
Biological metric data from Bogue Homo were compared to the suite of selected
box plots given above. Visual observation of the Bogue Homo values relative to the
ranges of high and low values, as
indicated by the box plots, was performed
to determine if the particular abiotic
variable may be indicative of biotic effects
that comprise the impairment (see
Comment 13). A summary of the
conclusions drawn based on comparison
of Bogue Homo values to select box plots
is shown in Table 8. All box plots,
including the plotted Bogue Homo values,
are shown in Appendix E.
4.2.3. Decrease in Suitable Habitat
There is supporting and weakening
evidence for a less suitable habitat as the
cause of biological impairment in the Bogue Homo. Based on scatter-plot and box plot
analyses, evidence was weak or inconclusive for a decrease in suitable habitat as a
potential stressor. Scatter plot analyses suggested suspended sediment (turbidity) and
percentage of silt substrate were not causative agents of impairment to Bogue Homo.
The only habitat related stressor-response correlation that suggested a potential cause
of impairment was percentage of gravel substrate. In contrast, box plot analyses
suggested that percentage of silt substrate was a potential causative agent. Also, there
was weak evidence for habitat degradation as a cause. This was based on the total
habitat assessment score, which includes many aspects of habitat quality. Box plot
analyses provided no evidence that suspended sediment was a cause of impairment.
4.2.4. Altered Temperature Regime
Biologic metrics related to temperature either did not correlate strongly with
abiotic parameters, or were correlated because of the presence of a few outlier data
points that "drove" the overall relationship. When the latter situation occurred, the
correlations were considered to be unreliable for making conclusions about potential
causes of stress. In some cases, too, data from Bogue Homo were not consistent with
the observed overall correlation gradient.
4.2.5. Altered Food Resources and/or Decreased Dissolved Oxygen
Decreased oxygen concentration, increased organic loading, and increased
nitrogen are suggested as potential stressors based on scatter plot and box plot
analyses. Only phosphorus concentration was not suggested as a causative agent.
Evidence was strongest with regard to chemical oxygen demand (scatter plot and box
plot) and dissolved oxygen concentration (box plot).
Comment 13. Specificity.
In addition to the evidence noted in
Section 4.2.3., none of the specific biological
responses that characterize the impairment -
decreased taxa richness of EPT, relative
abundance of EPT, relative abundance of EPT
taxa, or increased abundance of amphipods (the
latter also being represented by % noninsects)—
correlate strongly with a habitat variable. Also,
the most prominent habitat characteristic at the
impaired site, a high percentage of sand, is not
well correlated with any biological response.
Hence, at the bioregion scale, the specific
responses at the impaired site are not related to
habitat, and the distinguishing habitat
characteristic is not related to the measured
biological responses.
32
-------
TABLE 8
Summary of Stressor-Response Relationships from Box Plots of Various Biotic
Variables Based on Low and High Ranges of Variables Associated with Candidate
Causes
Total Organic Carbon (mg/l)
Chemical Oxygen Demand (mg/l)
Dissolved Oxygen (% Saturation)
Total Nitrogen (mg/l)
Total Phosphorus (mg/l)
Total Chlorides (mg/l)
Turbidity (NTU)
Silt substrate (% total)
Total Habitat Score
M-BISQ
-
0
0
+
# Total taxa
0
# Crustacean/Molluscan taxa
++
+
# Chironomidae taxa
-
# Orthocladinae taxa
0
0
# Tanytarsini taxa
++
# EPT taxa
++
+
+
+
# Ephemeroptera taxa
++
++
# Plecoptera taxa
-
++
# Clinger taxa
+
+
++
% Non-Insect individuals
+
+
% Crustaceans/Molluscan individuals
++
% Oligochaeta individuals
0
% Dipteran individuals
—
% Tanytarsini individuals
—
—
% EPT individuals
++
% Collector individuals
-
% Filterer individuals
—
—
—
-
% Clinger individuals
0
0
Beck's Biotic Index
—
—
HilsenhofFs Biotic Index
0
-
NCBI
+
% Tolerant individuals
—
—
—
—
% Intolerant individuals
-
-
-
0
Summary
+
++
++
-
-
—
-
++
+
++ = strong evidence as cause
+ = week evidence as cause
0 = unclear
= weak evidence as not cause
= strong evidence as not cause
33
-------
4.2.6. Increase in Ionic Strength and/or Increase in Toxic Substances
Total chlorides and total dissolved solids were evaluated. Scatter plot analyses
resulted in weak evidence for total chlorides as a potential cause; however, box plot
analyses resulted in strong evidence against total chlorides. In addition, scatter plot
analyses resulted in weak evidence against total dissolved solids as a cause (see
Comments 14 and 15).
Comment 14. Stressor-Response
Relationships for Laboratory Studies.
MDEQ did not use laboratory data as a type of
evidence from elsewhere. Laboratory test data
are available for suspended sediment,
temperature, dissolved oxygen, and total
dissolved solids. Like many others, MDEQ found
these data difficult to obtain. CADDIS has since
attempted to make this easier through links in
sections on stressor-response and for individual
stressors.
Comment 15. Symptoms.
The identification of assemblage or community
level characteristics would greatly benefit eco-
epidemiological investigations. The MDEQ's
attempt to apply that approach is illustrative (see
Appendix B). At this time, the U.S. EPA does not
recommend using community metrics levels as
symptomatic of specific stressors at this time
because indices and metrics respond to too
many stressors. However, this is an active area
of research (Relyea et al., 2000; Yuan, 2006).
34
-------
5. COMPARISON OF CANDIDATE CAUSES
This analysis provides a consistent and systematic approach for evaluating
whether available evidence supports, or does not support, each candidate cause.
Furthermore, for multiple causes, the process identifies the stressor(s) that are most
strongly supported by the evidence with a level of confidence assigned to each stressor
by the investigators. This is done according to the types of evidence adapted from the
U.S. EPA Guidance Document (U.S. EPA, 2000) and U.S. EPA Web site
(www.epa.gov/caddis.) A set of symbols were used to score the importance of various
causal relationships that describe the most prominent candidate causes.
After scores were assigned for each type of evidence from the case and from
types of evidence that use data from elsewhere, two additional types of evidence were
evaluated to score the overall pattern or tendency of the facts in the case. The two
types of evidence for comparing all the evidence for a candidate cause were
• Consistency of evidence, in which investigators evaluated the degree to
which assessments of relationships between impairment and stressors
were consistent, and
• Explanation of evidence, in which the investigators evaluated whether
inconsistencies can be explained by other information, in keeping with the
stressor hypothesis.
The strength-of-evidence table for Bogue Homo is shown in Table 9. Strength-of-
evidence analyses led us to conclude that multiple stressors may be contributing to
biological impairment in Bogue Homo. Most of the chemical and physical parameters
measured in Bogue Homo suggested possible causes; however, a few stood out as
being probable causes. These included altered food resources and low dissolved
oxygen associated with organic and nutrient enrichment. Decreased suitable habitat
was identified as an additional possible cause of impairment. However, evidence for
habitat alteration, such as sediment inputs or hydrologic regime alteration, was not very
strong. However, gravel substrates were absent and waters were turbid.
Increased ionic strength and/or increase in toxic substances were identified as
less-probable causes for impairment based on land use and the moderate impairment
at the site. However, little evidence existed for these factors and none for pesticides,
metals, or other potentially toxic compounds. Evidence that suggests that elevated
temperature contributed to the impairment was inconsistent and/or unsubstantiated.
35
-------
TABLE 9
Strength of Evidence for Bogue Homo, Mississippi
Type of Evidence
Unsuitable Habitat
Increased
Temperature
Altered Food
Resources (Organic
Matter) and/or
Decreased Dissolved
Oxygen
Increased Ionic
Strength and/or
Increased Toxicity
Evidence
Score
Evidence
Score
Evidence
Score
Evidence
Score
From the
Case
Spatial Co-
occurrence
Uncertain
0
Uncertain
0
Compatible
+
No
evidence
NE
Causal
Pathway
Evidence
for some
steps
+
Ambiguous
0
Evidence for
all steps
++
Evidence
for some
steps
+
From
Elsewhere
Mechanis-
tically
Plausible
Cause
A plausible
mechanism
exists
+
A plausible
mechanism
exists
+
A plausible
mechanism
exists
+
A plausible
mechanism
exists
+
Stressor-
Response,
Other Field
Studies
Unclear
0
Unclear
0
Qualitative
agreement
+
Unclear
0
Multiple
Types of
Evidence
Consistency
of Evidence
Some
consistent
+
No
evidence
NE
All
consistent
+++
Some
consistent
+
Explanation
of Evidence
Not
applicable
NA
Not
applicable
NA
Not
applicable
NA
Not
applicable
NA
-------
6. IDENTIFY THE PROBABLE CAUSE
6.1. PROBABLE PRIMARY CAUSE
6.1.1. Increased Organic and Nutrient Enrichment Altering Food Resource and
Leading to Low Dissolved Oxygen
COD and TOC levels were greater than LD conditions and greater than all of the
SSC streams in 2001, and were elevated in 1997 based on review of the historical data
at Ovett, Mississippi. Levels of ammonia and TKN at the impaired site were
consistently greater than the 75th percentile of the East Bioregion, Site Class 6
reference conditions, and most of the SSC sites (see Table 4). Concentrations of nitrite,
nitrate, and phosphorus were similar to those for the LD condition and at all of the SSC
sites. DO and DO percent saturation concentrations at the impaired site in 2001 were
lower than those in LD conditions and all SSC sites. Historical data also show that low
values of DO and DO percent saturation periodically occurred at the impaired site.
However, they were not low enough to violate the state's Water Quality Standard
(WQS) criterion. .Biological metric values from Bogue Homo that supported increased
nutrients, decreased DO, and organic enrichment as stressors included percent and
number of EPT taxa; percent Plecoptera (low DO, increased nutrients and organic
enrichment); and percent Amphipoda (decreased oxygen and organic enrichment).
Comparison of Bogue Homo data to regional stressor-response relationships also
suggested elevated nutrients, low DO, and organic enrichment as possible factors.
Decreased levels of DO and/or altered food resources, in association with their
intermediate pathway of organic and nutrient enrichment, were indicated as the most
likely causes of biological impairment based on the weight of evidence. This included
the presence of potential sources (poultry and cattle operations, residential growth in
the Laurel-Ellisville area, septic tank leachate, Lake Bogue Homo, and point sources)
and elevated levels of nutrients (specifically ammonia and TKN).
6.2. PROBABLE SECONDARY CAUSES
6.2.1. Decrease in Suitable Habitat
The habitat quality score for the Bogue Homo impaired site was lower than the
LD conditions for the East Bioregion and Site Class 6 sites, and lower than the habitat
scores for most of the SSC stations (see Table 4). Especially in comparison to the SSC
sites, the lower overall score was influenced by the low in-stream habitat score.
Interestingly, the total habitat score at the downstream Bogue Homo site was nearly
identical to that of the impaired site, but the downstream site was not impaired. This
result suggests that habitat quality in Bogue Homo did not contribute much to the
impairment. However, the downstream site had a higher in-stream habitat score and a
significant amount of gravel substrate, as well as reduced levels of some other
stressors, which would contribute to the better biological rating.
37
-------
The comparison of Bogue Homo data to regional stressor-response relationships
resulted in weak or inconclusive evidence for decrease in suitable habitat as a potential
stressor. Potential sources of habitat degradation in the watershed included land-
disturbing activities such as clearing for silviculture; watershed construction activities
that cause upland sediment runoff and stream bank failure; livestock access to streams;
extensive drainage ditches along Highway 15; increased commercial and residential
development; naturally erodible soils; in-stream sources of sediment; releases of water
from Lake Bogue Homo; and historic alteration of the landscape. After reviewing all
available evidence, MDEQ concluded that a decrease in suitable habitat is plausible as
a secondary cause of biological impairment.
Sediment and hydrologic characteristics typically associated with reduced habitat
suitability were considered to be less likely to contribute to the impairment (see
Comment 16). Turbidity measurements for Bogue Homo did not suggest suspended
sediment as a potential cause of impairment. Neither the 2001 data (see Table 4) nor
the historical data indicated that turbidity was elevated. In the habitat assessment
scoring procedure, the field crew observed, scored, and recorded the amount of
sediment deposition as suboptimal. However, few signs of other sediment or hydrologic
alteration effects (i.e., high erosion/bank instability, channel alteration, rapid water level
fluctuation, incision, and/or low flow) were
noted in 2001 or during the reconnaissance in
2004. The downstream comparative site on
Bogue Homo had a similar sediment
deposition score and a higher level of
turbidity, but was not impaired.
6.3. LESS PROBABLE OR UNLIKELY CAUSES
6.3.1. Altered Thermal Regime
Water temperature data from M-BISQ at the two Bogue Homo sites, plus limited
historical data, did not point to temperature as a likely stressor. Similarly, biological
data did not suggest elevated water temperature as a cause of impairment. Because of
lack of co-occurrence (similar temperature levels at downstream site in 2001 with
nonimpaired biology), lack of regional stressor-response relationships, and lack of a
causal pathway (few potential sources), water temperature was not considered a
stressor of Bogue Homo.
6.3.2. Increase in Ionic Strength and/or Increase in Toxic Substances
Concentrations of TDS and chlorides at the impaired site were elevated,
compared to LD levels and to those at most of the SSC sites in 2001 (see Table 4).
However, even the elevated values were far below state WQS. The downstream Bogue
Homo SSC site had chloride levels similar to the impaired site, and a much higher TDS
concentration (239 mg/L) than the impaired site, in 2001, with no biological impairment.
Neither TDS nor chlorides were associated with impairments in the regional data sets.
Comment 16. Suitable Habitat.
In sandy bottom streams where gravel
substrates do not naturally occur, woody
debris can provide more stable substrates.
No data were available to evaluate the
relative amount of woody debris.
38
-------
Potential sources of TDS and chlorides in Bogue Homo included upstream point
sources, agriculture operations, sand/gravel mining, soil disturbance from clear-cutting,
and some possible petroleum activities. However, no definitive sources were identified.
Because co-occurrence was inconsistent and obvious sources were limited, MDEQ
concluded that increased ionic strength was not a major stressor of Bogue Homo.
MDEQ found no evidence for or against toxics, oil/grease, or soaps/surfactants
as stressors. Concentrations of these pollutants cannot be evaluated until adequate
data on these compounds in the water and/or sediments are collected. Lack of
identified toxic releases or pollution incidents, and the abundance of benthic organisms
at the sampling site, suggest minimal or no impact from toxic substances. Sources for
toxics are present in the watershed as similarly identified for ionic strength; however,
these sources are relatively few and generally far removed from the study site. As a
result, MDEQ felt that increased toxics probably were not a likely cause of biological
impairment.
39
-------
7. DISCUSSION AND HIGHLIGHTS
Application of the causal analysis process to Bogue Homo allowed MDEQ to
determine the most probable cause(s) of biological impairment in the §303(d)-listed
reach. This process identified organic and nutrient enrichment leading to an altered
food resource and low dissolved oxygen as probable causes and a decrease in suitable
in-stream habitat as a possible cause. Following the causal analysis process, MDEQ
developed TMDLs of Bogue Homo for organic matter and nutrients loadings.
Through the development of this case study, MDEQ later improved and
strengthened their routine causal analysis process. The conceptual model provided in
this report has been revised to better depict potential links between sources of
pollutants and biological response indicators, including a more comprehensive and
explicit distinction between proximate causes and their associated intermediate causal
pathways and indicators. Additionally, through better understanding of the intended
causal analysis process, MDEQ modified the strength-of-evidence analyses and the
associated scoring tables.
As illustrated in this revised case study, emphasis now is placed on ensuring that
data (and their associated analyses) developed for one type of evidence, are used to
support only one type of evidence. The purpose of this effort is to avoid inappropriately
overemphasizing a prospective cause by "double counting" data and/or their analyses.
Careful consideration is given to determine which type of evidence a type or group of
data and associated analyses reveals. To highlight and disseminate only pertinent data
and information used in strength-of-evidence analyses, MDEQ no longer reports types
of evidence that are not used due to lack of data or to inapplicable data.
Two types of screening methods were identified during the analysis that might
strengthen the overall causal analysis process in the future. One screening method
involves refining the causal analysis process to be more time-efficient, and to allow the
user to quickly evaluate data to identify causal pathways that need intensive data
collection to fill data gaps. This screening method would be helpful when users are not
comfortable with the amount or quality of data (spatial coverage, number of samples,
temporality) available for the TMDL process. The other screening method involves
visiting multiple stream reaches throughout the watershed to better determine the
spatial scale of the impairment. This method allows the investigators to focus on a
reduced set of more-likely causes. This application would be used when data are
limited in spatial and temporal scale, but differs from the data collection effort noted
previously in that the spatial-assessment effort is much less. In general, the method
would involve rapid assessment of multiple stream reaches using benthic community
sampling, water quality sampling, and physical stressor evaluation. A field screening
technique would be used to collect, identify, and collate benthic data, resulting in
immediate benthic assemblage assessment and no samples to process in the lab.
While on site, physical and chemical measurements would be made and water quality
grab samples would be collected as resources allow.
40
-------
The causal analysis procedure used by MDEQ was based on methods outlined
in the U.S. EPA Stressor Identification Guidance Document (U.S. EPA, 2000). MDEQ
developed a standard process of data consideration to support or refute candidate
causes. The process entails prioritizing prospective causes based largely on weight of
evidence, with the end result usually involving placing each of the candidate causes in
one of four possible categories: Probable Primary Cause/s, Probable Secondary
Cause/s, Less Probable Cause/s, or Unlikely Cause/s.
MDEQ has an extensive §303(d) List of Impaired Waters. For many of these
bodies of water, the causes of impairment are unknown. For most, classification as
impaired was based on a single sample of benthic macroinvertebrates. In addition, a
lawsuit involving these waters has resulted in a strict deadline for completing the
TMDLs. Because of the large number of impaired waters in need of stressor
identification, the similarity of amount and type of data to be used for causal analysis
associated with each site, and the short amount of time available to perform the
analyses, MDEQ has developed several techniques for making the analysis faster,
more routine, and more consistent. These are described below.
A "standard" list of potential stressors of Mississippi waters was initially
developed. This list was prepared following round-table discussions involving MDEQ
scientific and water quality staff. The list was framed in the context of five major groups
of environmental factors that are known to affect biological community structure or
function: chemical processes, physical processes, hydrological processes, biological
interactions, and energy regimes.
Within each group, related causes of impairment that might stress aquatic biota
in streams and rivers were identified. Only indicators that MDEQ uses in its surface
water quality monitoring program strategy were included. This list became the
"universal list of causes of impairment" that was used to start all causal analysis efforts
in Mississippi. Use of a "standard list" resulted in a more consistent, timely evaluation of
data for determining probable causes of impairment.
MDEQ used a work-team approach to perform causal analyses. Attempts were
made to engage multi-disciplinary staff as members of the work teams for each
waterbody. To achieve causal analysis process results of the highest possible quality,
MDEQ formed and engaged a multi-disciplinary team composed of biologists,
engineers, water quality scientists, and geographers. The intent was to incorporate
expertise in physical, hydrological, chemical, and biological processes that were
applicable to aquatic ecosystems in Mississippi. The use of this team was intended to
reduce individual bias, increase the breadth of experience and knowledge, and increase
objectivity in an inherently subjective process. This multidisciplinary approach hopefully
strengthened the causal analysis and subsequent TMDL process by increasing the
breadth of exploration of potential causes and effects, given the implicit complexity of
aquatic ecosystems and factors that are involved in changes to biological community
structure and function. GIS data were included in the SI process, and information and
analysis tools were used throughout the process. MDEQ explored, refined and gained
41
-------
needed experience in the causal analysis process by taking on the first impaired
waterbody as a pilot project, and engaging the entire team in the pilot causal analysis
process. This proved to be very useful in ensuring consistent and adequate
performance as well as covering many issues at the start of the process that might
otherwise have been missed.
MDEQ relied heavily on the quality of data to add strength to the TMDL process.
Scientifically sound and robust biological data collection methods and sample analysis
techniques were developed before the causal analysis process. This greatly improved
confidence in conclusions drawn about the data. The ability to use comprehensive
biological assemblage data has been especially advantageous in light of the small
number of chemical and physical samples available for most impaired sites. Much of
the strength of the evidence process employed by MDEQ was based on stressor-
response relationships, which are very informative due to the high quality of biological
data. MDEQ also used various spatial stratification schemes for the state, depending
on modeling of chemical, physical, and biological data. These strata were developed as
part of the development of the state's biological monitoring and assessment program.
Presently, MDEQ is exploring additional methods for diagnosing causes of
impairment using benthic community data. These include the use of stressor-specific
tolerance values, stressor-specific models of community taxonomic composition, and
conditional probability analysis. As causes are verified through the SI process, the
results of this characterization must be summarized and confidence level assessed, in
either quantitative or qualitative terms. The level of confidence is stated primarily
through identification of uncertainties for each cause. In many cases, especially if the
causal inferences are based mostly on the strength-of-evidence analysis, uncertainties
cannot be quantified, because the stressor identification conclusion is not based on a
single source of uncertainty, but rather, on multiple types of evidence sometimes
supported by more than one line of evidence. In such cases, the degree of uncertainty
for the decision is characterized qualitatively as the best estimate of causation with an
accompanying disclosure of uncertainties and evidential proof. The final decision, in
terms of determining whether the causal evidence presented is adequate to justify
applicable management actions (e.g., TMDL, restoration strategies) to address the
suspected cause and/or prioritize remediation actions for multiple causes, then
becomes the responsibility of the regulatory or administrative authority.
In order to increase confidence in the evaluation of data in these SI analyses, it is
recommended that
• M-BISQ should be recalibrated using more data, which will increase the
accuracy of LD condition values and decrease the confidence intervals
associated with individual metrics and M-BISQ scores.
• Sampling performance should be continually maintained and strengthened
through training, measurement, evaluation of sampler performance, and
implementing corrective actions when necessary.
42
-------
• More physical/chemical samples should be collected at study sites to
increase the probability that samples are representative of actual stream
conditions, and effort should be invested to capture diel, seasonal, and
flow-dependent fluctuations where applicable.
• Quantitative data regarding habitat, hydrologic, and geomorphologic
characteristics should be collected.
43
-------
8. REFERENCES
Barbour, M.T., J. Gerritsen, B.D. Snyder and J.B. Stribling. 1999. Rapid
Bioassessment Protocols for Use in Streams and Wadeable Rivers: Periphyton, Benthic
Macroinvertebrates, and Fish. 2nd ed. U.S. Environmental Protection Agency, Office of
Water, Washington, DC. EPA/841/B-99/002.
Beck, W.M. 1954. Studies in stream pollution biology. Q. J. Fl. Acad. Sci. 17:211-227.
Cummins, K.W., M.A. Wilzbach, D.M. Gates, J.B. Perry and W.B. Taliaferro. 1989.
Shredders and riparian vegetation. Bioscience. 39(1):24-30.
Davis, W.S. and T.P. Simon, Eds. 1995. Biological Assessment and Criteria. Tools for
Water Resource Planning and Decision Making. Lewis Publishers, Boca Raton, FL.
DeShon, J.E. 1995. Development and application of the invertebrate community index
(ICI). In: Biological Assessment and Criteria: Tools for Water Resource Planning and
Decision Making, W.S. Davis and T.P. Simon, Eds. Lewis Publishers, Boca Raton, FL.
p. 217-243.
Fore, L.S., J.R. Karr and R.W. Wisseman. 1996. Assessing invertebrate responses to
human activities: Evaluating alternative approaches. J. N. Am. Benthol. Soc.
15(2):212-231.
Galli, J. and R. Dubose. 1990. Water temperature and freshwater stream biota: An
overview. In: Thermal Impacts Associated with Urbanization and Stormwater
Management Best Management Practices, Appendix C. Metropolitan Washington
Council of Governments, Washington, DC.
Hart, D.D. and C.M. Finelli. 1999. Physical-biological coupling in streams: The
pervasive effects of flow on benthic organisms. Ann. Rev. Ecol. Syst. 30:363-395.
Hayslip, G.A. 1993. EPA Region 10 In-Stream Biological Monitoring Handbook (for
Wadeable Streams in the Pacific Northwest). U.S. Environmental Protection Agency-
Region 10, Environmental Services Division, Seattle, WA. EPA/910/9-92/013.
Hilsenhoff, W.L. 1987. An improved biotic index of organic stream pollution. The Great
Lakes. Entomologist. 20(1):31-39.
Karr, J.R., K.D. Fausch, P.L. Angermeier, P.R. Yant and L.J. Schlosser. 1986.
Assessing biological integrity in running waters: A method and its rationale. Special
Publication 5. Illinois Natural History Survey.
Kerans, B.L. and J.R. Karr. 1994. A benthic index of biotic integrity (B-IBI) for rivers of
the Tennessee Valley. Ecol. Appl. 4:768-785.
44
-------
MDEQ (Mississippi Department of Environmental Quality). 1997. Mississippi Land
Cover Project. Prepared by the Space Remote Sensing Center, Stennis Space Center,
Stoneville, Mississippi, for Mississippi Department of Environmental Quality, Jackson,
MS. (For further information contact Mr. Gary Hennington [601-961-5183]).
MDEQ (Mississippi Department of Environmental Quality). 1999. Mississippi
Department of Environmental Quality Office of Pollution Control Laboratory Standard
Operating Procedures. Mississippi Department of Environmental Quality, Jackson, MS.
MDEQ (Mississippi Department of Environmental Quality). 2001. Quality Assurance
Project Plan for 303(d) List Assessment and Calibration of the Index of Biological
Integrity for Wadeable Streams in Mississippi. Mississippi Department of Environmental
Quality, Jackson, MS. (For further information contact Randy Reed [601-961-5158]).
MDEQ (Mississippi Department of Environmental Quality). 2003a. Mississippi
Commission on Environmental Quality regulation WPC-2: Water Qulaity Criteria for
Intratate, Intestate and Coastal Waters. Mississippi Department of Environmental
Quality, Jackson, MS. Available at
http://www.deq.state.ms.us/newweb/MDEQRegulations.nsf/RN/WPC-2.
MDEQ (Mississippi Department of Environmental Quality). 2003b. Development and
Application of the Mississippi Benthic Index of Stream Quality (M-BISQ). June 30.
Prepared by Tetra Tech, Inc., Owings Mills, MD, for the Mississippi Department of
Environmental Quality, Office of Pollution Control, Jackson, MS. (For further
information on this document, contact Randy Reed [601-961-5158]). Available at
http://www.deq.state. ms.us/mdeq.nsf/pdf/WMB_fullM_BISQReport/$File/303dlBI_FINAL
_Report_070903_Report_and_Append.PDF?OpenElement.
MDEQ (Mississippi Department of Environmental Quality). 2005. Total Maximum Daily
Load: Biological Impairment Due to Organic Enrichment/Low Dissolved Oxygen and
Nutrients, The Bogue Homo River Pascagoula Basin Jones County Mississippi.
Available at
http://www.deq. state. ms.us/mdeq.nsf/pdf/TWB_BogueHomoLowDO&NutJun05/$File/Pa
scagoulaRBBogueHomoLowDO&NutrientsJun05.pdf?OpenElement.
Merritt, R.W. and K.W. Cummins. 1996. An Introduction to the Aquatic Insects of North
America. 3rd ed. Kendall/Hunt Publishing Co., Dubuque, IA.
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 Regulations and Standards, Washington, DC.
EPA/440/4-89/001.
Relyea, C.D., G.W. Minshall and R.J. Danehy. 2000. Stream Insects as Bioindicators
of Fine Sediment. Presented at Watershed Management 2000 Conference, June
21-24, Fort Collins, CO. Water Environment Federation, Alexandria, VA.
45
-------
Resh, V.H., R.H. Norris and M.T. Barbour. 1995. Design and implementation of rapid
bioassessment approaches for water resources monitoring using benthic
macroinvertebrates. Aust. J. Ecol. 20:108-121.
Shackelford, B. 1988. Rapid Bioassessments of Lotic Macroinvertebrate Communities:
Biocriteria Development. Arkansas Department of Pollution Control and Ecology, Little
Rock, AR. WQ88-00-0.
Smith, E.P. and J.R. Voshell, Jr. 1997. Studies of Benthic Macroinvertebrates and Fish
in Streams within EPA Region 3 for Development of Biological Indicators of Ecological
Condition. Virginia Polytechnic Institute and State University, Blacksburg, VA.
U.S. EPA. 2000. Stressor Identification Guidance Document. Office of Water,
Washington, DC. EPA/822/B-00/025.
Vannote, R.L. and B.W. Sweeney. 1980. Geographic analysis of thermal equilibria - a
conceptual-model for evaluating the effect of natural and modified thermal regimes on
aquatic insect communities. Am. Natural. 115(5):667-695.
Yuan, L.L. 2006. Estimation and Application of Macroinvertebrate Tolerance Values
(Final). U.S. Environmental Protection Agency, Washington, DC. EPA/600/P-04/116F.
46
-------
APPENDIX A
DATA USED FOR CAUSAL ANALYSIS
All available water quality data, other environmental data, and information for the
Bogue Homo watershed were gathered and reviewed. These data were compiled from
various MDEQ databases and the U.S. EPA Legacy STORET database. Data used in
the causal analysis included benthic macroinvertebrate assemblage metrics, taxonomic
information, qualitative habitat assessment scores, sediment particle-size percentages,
measurements of various water quality parameters, LULC percentages, and other
miscellaneous watershed information. The majority of data were collected during the
winter of 2001 as part of a statewide monitoring and assessment strategy for the
purpose of assessing the 1996 §303(d) List of Impaired Water Bodies. Some water
quality data evaluated were collected in 1997, as part of MDEQ's Ambient Monitoring
Program.
A.1. 2001 STATEWIDE MONITORING AND ASSESSMENT STRATEGY
In 2001, MDEQ implemented a statewide monitoring and assessment strategy
that involved collection of data on biological and habitat condition, sediment particle
size, various water quality characteristics, remotely sensed land use, and land cover
percentages. All data collected were used in a multi-step process to develop a
regionally calibrated Index of Biological Integrity, which MDEQ refers to as the M-BISQ.
The M-BISQ process allowed each sampling site to be scored based on selected
biological metric values, and to be compared to least disturbed conditions. The least
disturbed condition was defined by combining certain metric values from streams in
each particular region considered to be least disturbed, calculating overall M-BISQ
scores, and using the 25th percentile value of the range of all least disturbed scores for
each particular region. Individual site scores were then compared to the least disturbed
condition for their region. If the score fell below the least disturbed condition, the site
was considered impaired.
A.2. BIOLOGICAL DATA
Collection and processing of benthic macroinvertebrate samples and benthic
macroinvertebrate taxonomy were performed using Standard Operating Procedures
outlined in MDEQ (2001). Over 60 different biological metrics describing various
characteristics of the macroinvertebrate communities were derived from the taxonomic
data. A suite of regional specific metrics was used to calculate an overall M-BISQ score
according to methods outlined in the M-BISQ QAPP (MDEQ, 2001). Biological data
used for analyses included individual metric values and the overall M-BISQ score.
47
-------
A.3. HABITAT DATA
Quality of the physical habitat was assessed using a visually based scoring
procedure in which ten habitat parameters were rated on a continuous scale. The
scores for these ten parameters were summed to calculate the overall habitat score
(MDEQ, 2001). Scores for individual habitat parameters also were summed in three
subcategories describing the stream environment: in-stream habitat conditions, stream
morphology characteristics, and riparian habitat condition.
A.4. SEDIMENT PARTICLE SIZE DATA
Sediment particle size was measured using a method based on the 100-particle
Wolman pebble count method (MDEQ, 2001). Data are presented as the percent of
silt/clay, sand, gravel, cobble, boulder, and/or hardpan clay, relative to the total number
of particles.
A.5. WATER QUALITY DATA
Various physical and chemical parameters were measured using U.S. EPA-
approved methods, both in the field and via laboratory analyses (MDEQ, 1999, 2001).
MDEQ water-quality criteria and/or historic target thresholds were used for
comparisons.
A.6. WATERSHED CHARACTERIZATION DATA
Land use and land cover percentages were calculated using GIS based data
layers developed as part of the Mississippi Land Cover Project (MDEQ, 1997). The
land-use and land-cover percentages were calculated for the entire watershed area, a
100 m buffered area around the channel for the entire watershed, and a 100 m buffered
area around the channel for an area defined by a 1-km radius from the monitoring
station. Other watershed characterization data include census information; forestry and
agricultural statistics; soil survey results; permitted facilities database; and watershed
investigations.
48
-------
APPENDIX B
COMMUNITY SYMPTOMOLOGY
Many attempts have been made to interpret benthic macroinvertebrate
community responses as symptomatic of specific stressors. The MDEQ's attempt to
use seven M-BISQ metrics and additional metrics in that way is presented below and
summarized in Table B-1.
These interpretations are based on MDEQ's synthesis of the following
publications: Barbour et al. (1999), Cummins et al. (1989), DeShon (1995), Fore et al.
(1996), Hayslip (1993), Hilsenhoff (1987), Kerans and Karr (1994), Plafkin et al. (1989),
Resh et al. (1995), Shackelford (1988) and Smith and Voshell (1997).
B.1. BECK'S BIOTIC INDEX (Bl)
A decrease in this metric is thought to indicate an increase in stress. However,
this metric does not necessarily discriminate among types of stress. Bogue Homo's
Beck's Bl value gives little information as to the specific cause of impairment.
B.2. HILSENHOFF BIOTIC INDEX (HBI)
Bogue Homo's Hilsenhoff Biotic Index (HBI) value gives little information as to
the specific cause of impairment. However, this index was originally developed as an
indicator of organic enrichment (Hilsenhoff, 1987).
B.3. NUMBER OF TANYTARSINI TAXA
In general, members of the Tribe Tanytarsini are pollution sensitive. Therefore, a
decrease in their numbers may suggest an increase in perturbation that affects niche
space, habitat, and/or food source for this particular group of taxa. It may also reflect
other stressors such as increased nutrients, alteration of sediment regime, and
decreased oxygen. It is difficult to pinpoint a specific stressor as the cause of change in
this metric. The number of Tanytarsini taxa from the Bogue Homo sample gives little
information as to the specific cause of impairment, as the differences seen are small.
B.4. PERCENT CAENIDAE
Many mayfly taxa are considered to be intolerant to pollution and stress.
Members of the family Caenidae are considered to be generally more tolerant than
other mayflies to many stressors, as indicated by the higher tolerance values derived for
these taxa. An increase in the Percent Caenidae metric suggests an increase in stress,
in particular slight to moderate nutrient enrichment, reduction in levels of DO, and/or
decrease in habitat availability and/or complexity. The percent of Caenidae from the
Bogue Homo sample gives little information as to the specific cause/s of impairment.
49
-------
TABLE B-1
Summary of Biological Metrics from Bogue Homo and Potential Associated Causes
Disturbed
Habitat
(including
sediment and
hydrology
Decreased
Dissolved
Oxygen
Altered
Food
Resource
Increased
Temperature
No
Information
Beck's Biotic
Index (Bl)
X
Hilsenhoff Biotic
Index (HBI)
X
Number of
Tanytarsini Taxa
X
Percent
Caenidae
X
Percent EPT (no
Caenidae
X
X
X
X
X
Percent dingers
X
Percent Filterers
X
Percent
Amphipoda
X
Number of EPT
Taxa
X
X
X
X
X
Percent
Plecoptera
X
X
X
X
X
Percent
Predators
X
Percent
Sprawler
X (hydrology)
50
-------
B.5. PERCENT EPHEMEROPTERA, PLECOPTERA, AND TRICHOPTERA (EPT)
(NO CAENIDAE)
Mayflies, stoneflies, and caddisflies are generally considered, as a group, to be
intolerant of pollution and stress. Caenidae taxa were subtracted because this family of
mayflies is considered to be more tolerant to various stressors as compared to most
other EPT. Thus, it could have skewed this metric. A decrease in percent EPT
generally is viewed as evidence for an increase in stressors. Many specific stressors,
such as decrease in ambient DO concentration; increase in temperature; decrease in
stable, suitable and/or diverse habitats; and increase in sediment deposition have been
shown to result in elimination of EPT taxa. In addition, alterations to natural food
resources, increase in nutrient concentrations, increase in toxic substances, and
increase in organic enrichment have been found to alter the community structure and
composition of EPT. Elucidation of a specific cause based on a decrease in this metric
is difficult because EPT respond to many stressors.
B.6. PERCENT CLINGERS
Clinging organisms rely on structure and physical habitat, so a decrease in this
metric suggests a decrease in available structure and habitat. The relatively low dinger
abundance suggests that a decrease in stable, suitable, and diverse habitats may be a
cause of impairment. However, due to the slight differences seen, the evidence is not
strong.
B.7. PERCENT FILTERERS
The percentage of filter-feeding organisms in moderate-sized streams typically is
on the order of 20-50%. The natural variability of feeding group metrics is not well
understood. Thus, only extreme variances from LD conditions and levels at comparison
sites should be considered. Changes in the relative abundance of filterers suggest that
an alteration of food resources may be a contributing cause of impairment in Bogue
Homo; however, the evidence is not consistent or strong.
B.8. PERCENT AMPHIPODA
Streams characterized by a large number of Amphipods can be the result of
multiple factors, including decreased predation (fewer predators or increased cover for
predator avoidance) and a shift in food resource. Most often, the food resource of
amphipods is detritus material—fine particulate organic matter or coarse particulate
organic matter. A larger number of amphipods could indicate a relative increase in the
abundance of detritus that they can use as food. Amphipods can withstand elevated
levels of chloride. Thus, elevated salinity may be a contributing factor to impairment of
Bogue Homo.
51
-------
B.9. NUMBER OF EPT TAXA
Most mayflies, stoneflies, and caddisflies are considered to be relatively
intolerant of pollution and stress. Low taxa richness could suggest a decrease in stable,
suitable, and diverse habitats, a decrease in DO concentration, increased temperature,
and/or altered food resource. However, elucidation of a specific cause in a decrease in
this metric is difficult.
B.10. PERCENT PLECOPTERA
As a group, stonefly taxa generally are thought to be relatively intolerant of
pollution, compared to most other families of benthic macroinvertebrates. Stoneflies
depend on swift, cool waters for high concentrations of DO and cool temperatures.
Along a gradient of stress, stoneflies are usually the first to drop out of the community
make-up. Thus, a decrease in this metric suggests possible reductions in DO or water
velocity, or increases in temperature. The percent of Plecoptera from the Bogue Homo
sample suggests that there has been a decrease in stable, suitable, and diverse
habitats; a decrease in DO concentration; altered food resources; and/or increased
temperatures. However, the differences seen are small; therefore, evidence is weak.
B.11. PERCENT PREDATORS
The response of this metric to stress is variable. In general, however, the
percent of predators within a benthic macroinvertebrate community from a small to
medium sized least degraded stream is between 10 and 20% (Davis and Simon, 1995).
This relatively high abundance suggests a possible altered food resource as a
contributing cause of impairment of Bogue Homo.
B.12. PERCENT SPRAWLERS
An increase in the proportion of sprawlers suggests a decrease in water velocity
(i.e., an increase in the amount of stagnant area and depositional zone of a stream).
Their high relative abundance in Bogue Homo suggests a decrease in water velocity in
many habitats found there.
52
-------
APPENDIX C
PEARSON CORRELATION COEFFICIENT MATRIX
53
-------
TABLE C-1
Pearson Correlation Coefficient Matrix of Biological Metrics with Physical and Chemical Indicators
M-BISQ Score
Shannon Diversity Index
% Amphipoda
% Bivalves
% Chironomidae
% Coleoptera
% Crustaceans and Molluscans
% Dipterans
% Gastropoda
% Isopods
% Non-insects
% Odonata
% Oligocaheta
Ratio of Orthocladinae to Chironomidae
% Plecoptera
% Tanytarsini
Ratio of Tanytarsini to Chironomidae
% Trichoptera
% Caenidae
% EPT (No Caendiae)
% Ephemeroptera No Caenidae
Turbidity (log, NTU)
-0.57
-0.06
0.35
0.42
-0.26
0.23
0.46
-0.19
0.13
0.30
0.44
-0.18
0.20
0.10
0.09
-0.28
-0.35
-0.39
-0.01
-0.40
-0.35
Dissolved Oxygen (percent saturation)
0.16
-0.11
0.10
-0.22
-0.15
-0.04
-0.02
-0.14
0.02
-0.04
-0.08
0.04
-0.16
-0.03
0.22
-0.09
-0.04
0.11
0.14
0.25
0.19
PH
0.02
-0.25
0.11
-0.06
-0.27
-0.02
0.09
-0.12
0.03
0.19
0.05
-0.08
-0.07
0.11
0.18
0.01
0.10
0.02
0.35
-0.01
-0.09
Total Nitrogen (log+1, mg/l)
-0.46
-0.07
0.34
0.36
-0.30
0.35
0.45
-0.31
0.29
0.28
0.49
0.04
0.34
-0.02
-0.18
-0.20
-0.28
-0.27
-0.18
-0.33
-0.22
Total Phosphorus (log+2, mg/l)
-0.36
-0.13
0.25
0.27
-0.34
0.27
0.30
-0.28
0.14
0.15
0.30
0.09
0.15
0.24
-0.02
-0.23
-0.26
-0.14
-0.06
-0.14
-0.10
Chemical Oxygen Demand (log, mg/l)
-0.50
-0.15
0.38
0.28
-0.23
0.10
0.38
-0.10
0.27
0.10
0.41
-0.25
0.31
0.09
-0.09
-0.28
-0.36
-0.29
-0.15
-0.36
-0.28
Total Organic Carbon (log, mg/l)
-0.58
-0.19
0.37
0.36
-0.21
0.14
0.45
-0.10
0.27
0.26
0.44
-0.15
0.21
0.12
-0.05
-0.23
-0.32
-0.47
-0.06
-0.48
-0.36
Total Chlorides (log, mg/l)
-0.24
-0.36
0.04
0.02
-0.24
-0.07
0.03
0.02
-0.09
0.00
0.06
-0.03
0.11
0.21
0.00
-0.02
0.10
-0.24
0.46
-0.26
-0.21
Total Dissolved Solids (log, mg/l)
-0.30
-0.44
0.05
0.15
-0.30
0.01
0.05
0.03
0.07
-0.09
0.08
-0.04
0.11
0.23
0.01
-0.05
0.05
-0.29
0.46
-0.32
-0.27
Total Habitat Score
0.27
0.28
-0.03
0.22
-0.06
0.14
0.02
-0.19
0.09
-0.09
-0.03
-0.01
-0.13
-0.06
0.19
-0.02
0.05
0.18
-0.12
0.28
0.21
Instream Habitat Score
0.34
0.26
-0.09
0.14
0.02
0.10
-0.02
-0.13
0.08
-0.04
-0.08
-0.04
-0.18
-0.06
0.05
0.07
0.16
0.25
-0.06
0.27
0.21
Morphological Habitat Score
0.16
0.19
-0.06
0.30
-0.20
0.12
0.04
-0.19
0.01
-0.03
0.03
-0.09
0.00
-0.05
0.19
-0.06
0.04
0.14
-0.10
0.23
0.16
Riparian Habitat Score
0.17
0.23
0.06
0.10
0.03
0.11
0.03
-0.14
0.12
-0.15
-0.02
0.10
-0.14
-0.04
0.22
-0.04
-0.05
0.07
-0.13
0.20
0.15
% silt/clay
-0.39
-0.10
0.05
0.38
-0.29
0.08
0.37
-0.25
0.09
0.51
0.44
-0.05
0.42
0.15
-0.16
-0.25
-0.23
-0.17
0.10
-0.32
-0.27
% sand
-0.11
0.06
0.13
-0.19
0.32
0.04
-0.12
0.26
-0.07
-0.28
-0.14
0.10
-0.13
-0.14
-0.24
0.18
0.04
-0.15
0.08
-0.25
-0.15
% hardpan clay
0.19
-0.07
-0.06
0.09
-0.12
-0.11
-0.07
0.03
-0.15
-0.07
-0.12
-0.10
-0.19
-0.03
0.65
-0.04
0.08
0.14
0.01
0.20
-0.05
% gravel
0.53
0.07
-0.20
-0.23
-0.03
-0.08
-0.26
-0.08
0.06
-0.21
-0.31
-0.01
-0.28
0.02
0.15
0.10
0.23
0.39
-0.22
0.65
0.59
-------
TABLE C-1 cont.
Beck's Bl
Hilsenhoff Biotic Index
North Carolina Biotic Index
% Dominant Taxa
Ratio of Baetidae to Ephemeroptera
Ratio of Hydropsychidae to EPT
Ratio of Hydropsychidae to Trichoptera
% Intolerant
% Tolerant
# Total taxa
# EPT taxa
# Ephemeroptera taxa
# Plecoptera taxa
# Trichoptera taxa
# Diptera taxa
# Chironomidae taxa
# Orthocladinae taxa
# Tanytarsini taxa
# Coleoptera taxa
# Crustacean and Molluscan taxa
# Oligochaeta taxa
Turbidity (log, NTU)
-0.51
0.51
0.32
0.02
0.03
-0.13
-0.15
-0.38
0.44
-0.13
-0.42
-0.28
-0.32
-0.41
-0.03
-0.09
-0.24
-0.30
-0.02
0.33
0.21
Dissolved Oxygen (percent saturation)
0.17
-0.14
-0.22
0.10
0.07
0.12
0.17
0.14
-0.14
-0.03
0.23
0.17
0.40
0.05
-0.04
-0.08
0.29
-0.03
-0.05
-0.15
-0.19
PH
-0.20
0.25
0.06
0.28
-0.08
0.14
0.20
-0.20
0.15
-0.12
0.09
0.01
0.33
-0.04
-0.22
-0.23
-0.01
-0.06
0.02
-0.05
0.00
Total Nitrogen (log+1, mg/l)
-0.39
0.32
0.33
0.07
-0.23
-0.28
-0.22
-0.36
0.26
-0.06
-0.36
-0.23
-0.41
-0.28
-0.18
-0.25
-0.39
-0.37
0.25
0.41
0.32
Total Phosphorus (log+2, mg/l)
-0.30
0.38
0.34
0.10
-0.12
-0.14
-0.10
-0.27
0.37
-0.16
-0.20
-0.21
-0.09
-0.15
-0.33
-0.37
-0.31
-0.39
0.08
0.25
0.09
Chemical Oxygen Demand (log, mg/l)
-0.38
0.28
0.30
0.11
-0.18
-0.21
-0.18
-0.27
0.28
-0.22
-0.52
-0.41
-0.38
-0.43
0.02
-0.09
-0.19
-0.38
-0.12
0.31
0.28
Total Organic Carbon (log, mg/l)
-0.39
0.40
0.30
0.15
-0.19
-0.36
-0.39
-0.42
0.29
-0.27
-0.53
-0.43
-0.33
-0.46
-0.16
-0.24
-0.23
-0.42
-0.09
0.41
0.17
Total Chlorides (log, mg/l)
-0.22
0.39
0.28
0.32
-0.17
-0.24
-0.15
-0.34
0.40
-0.35
-0.09
-0.18
0.18
-0.10
-0.50
-0.48
-0.34
-0.35
-0.10
0.08
0.03
Total Dissolved Solids (log, mg/l)
-0.33
0.46
0.36
0.39
-0.16
-0.23
-0.12
-0.42
0.48
-0.42
-0.19
-0.25
0.11
-0.21
-0.50
-0.48
-0.25
-0.36
-0.13
0.08
0.02
Total Habitat Score
0.41
-0.26
-0.19
-0.27
-0.26
-0.19
-0.19
0.31
-0.15
0.23
0.27
0.24
0.16
0.22
0.10
0.08
0.03
0.14
0.20
0.00
-0.02
Instream Habitat Score
0.42
-0.30
-0.20
-0.25
-0.23
-0.15
-0.14
0.29
-0.21
0.25
0.25
0.23
0.13
0.21
0.16
0.14
0.05
0.26
0.28
-0.05
-0.05
Morphological Habitat Score
0.22
-0.15
-0.18
-0.21
-0.23
-0.20
-0.17
0.17
-0.03
0.11
0.22
0.20
0.13
0.18
-0.08
-0.11
-0.17
-0.06
0.23
0.00
0.08
Riparian Habitat Score
0.38
-0.19
-0.09
-0.22
-0.17
-0.13
-0.15
0.31
-0.13
0.20
0.19
0.16
0.12
0.15
0.16
0.17
0.20
0.15
0.00
0.05
-0.08
% silt/clay
-0.32
0.47
0.39
0.08
-0.13
-0.20
-0.24
-0.25
0.50
-0.11
-0.31
-0.28
-0.18
-0.26
-0.20
-0.21
-0.26
-0.28
0.06
0.32
0.37
% sand
0.05
0.03
0.11
-0.05
0.12
0.03
-0.03
-0.18
-0.11
0.11
0.00
-0.01
-0.11
0.08
0.21
0.18
0.08
0.12
-0.13
-0.06
-0.11
% hardpan clay
0.04
-0.03
-0.18
0.09
-0.07
0.10
0.09
-0.08
-0.11
-0.03
0.10
0.03
0.29
-0.01
-0.06
-0.08
0.16
-0.03
0.01
-0.04
-0.01
% gravel
0.30
-0.59
-0.58
-0.06
0.05
0.20
0.32
0.58
-0.40
0.02
0.36
0.38
0.17
0.24
-0.01
0.05
0.14
0.19
0.15
-0.32
-0.32
-------
TABLE C-1 cont.
# Collector taxa
# Filterer taxa
# Predator taxa
# Scraper taxa
# Shredder taxa
% Collectors
% Filterers
% Predators
% Scrapers
% Shredders
# Burrower taxa
# Climber taxa
# Clingertaxa
# Sprawler taxa
# Swimmer taxa
% Burrowers
% Climbers
% dingers
% Sprawlers
% Swimmers
Turbidity (log, NTU)
0.10
-0.36
-0.26
-0.24
0.11
0.34
-0.28
-0.04
-0.23
-0.17
0.23
-0.04
-0.43
-0.07
-0.05
0.21
-0.08
-0.50
-0.04
-0.04
Dissolved Oxygen (percent saturation)
0.01
0.12
-0.06
-0.03
0.02
-0.13
0.04
0.02
0.17
0.11
-0.01
-0.22
0.20
-0.02
-0.13
-0.42
-0.18
0.22
-0.02
0.08
PH
-0.14
-0.14
-0.18
-0.02
0.11
-0.10
0.26
-0.20
0.01
-0.22
-0.15
-0.08
-0.03
-0.19
-0.11
-0.33
-0.08
0.09
-0.02
-0.06
Total Nitrogen (log+1, mg/l)
0.08
-0.23
-0.07
0.22
-0.10
0.21
-0.22
-0.08
-0.09
-0.18
0.01
-0.03
-0.26
-0.10
0.12
0.02
0.00
-0.33
-0.32
0.00
Total Phosphorus (log+2, mg/l)
-0.13
-0.26
-0.01
-0.09
0.02
0.10
-0.14
-0.01
0.06
-0.22
-0.07
0.09
-0.18
-0.16
0.13
-0.02
0.08
-0.25
-0.07
-0.01
Chemical Oxygen Demand (log, mg/l)
0.08
-0.33
-0.26
-0.07
-0.02
0.32
-0.18
-0.11
-0.18
-0.10
0.11
-0.17
-0.40
0.02
-0.12
0.09
-0.09
-0.39
-0.03
-0.13
Total Organic Carbon (log, mg/l)
-0.07
-0.44
-0.22
-0.24
-0.03
0.20
-0.26
0.02
-0.14
-0.06
0.02
-0.16
-0.51
-0.08
-0.05
-0.01
-0.01
-0.42
0.05
-0.23
Total Chlorides (log, mg/l)
-0.46
-0.31
-0.04
-0.30
-0.15
-0.06
0.19
-0.05
0.01
-0.24
-0.32
-0.17
-0.20
-0.32
-0.03
-0.29
-0.10
0.00
0.26
-0.28
Total Dissolved Solids (log, mg/l)
-0.40
-0.41
-0.20
-0.35
-0.08
-0.05
0.22
-0.12
0.00
-0.27
-0.26
-0.21
-0.29
-0.33
-0.13
-0.29
-0.13
0.00
0.19
-0.31
Total Habitat Score
0.17
0.38
0.07
0.18
0.01
-0.03
-0.02
0.13
0.09
-0.16
-0.15
-0.18
0.35
0.20
0.16
0.04
-0.13
0.05
-0.09
0.12
Instream Habitat Score
0.21
0.43
0.06
0.34
0.03
-0.09
0.03
-0.03
0.16
-0.09
-0.10
-0.15
0.40
0.23
0.09
0.01
-0.12
0.12
-0.06
0.03
Morphological Habitat Score
0.03
0.16
0.01
0.05
0.02
-0.08
0.06
0.12
0.02
-0.23
-0.17
-0.22
0.17
0.09
0.12
-0.06
-0.16
0.05
-0.11
0.08
Riparian Habitat Score
0.18
0.34
0.10
0.08
-0.02
0.07
-0.12
0.23
0.06
-0.07
-0.09
-0.08
0.31
0.17
0.17
0.13
-0.03
-0.04
-0.04
0.18
% silt/clay
-0.10
-0.35
0.01
0.02
0.05
0.18
-0.21
-0.04
-0.14
-0.18
0.04
0.18
-0.31
-0.18
0.01
-0.05
0.20
-0.41
0.08
-0.13
% sand
0.15
-0.02
0.10
-0.09
-0.03
0.14
-0.07
0.03
-0.11
0.25
0.10
-0.02
-0.03
0.21
-0.07
0.19
0.01
0.02
0.09
-0.08
% hardpan clay
0.06
-0.07
-0.10
-0.24
0.23
-0.24
0.21
0.12
-0.11
-0.06
0.00
-0.01
-0.02
-0.02
-0.06
-0.12
-0.08
0.26
-0.15
-0.12
% gravel
-0.11
0.52
-0.10
0.24
-0.15
-0.28
0.26
-0.09
0.39
-0.12
-0.17
-0.22
0.45
-0.04
0.16
-0.13
-0.23
0.36
-0.15
0.33
Each correlation is based on 63 observations.
-------
APPENDIX D
STRESSOR-RESPONSE RELATIONSHIPS EXPRESSED AS SCATTER PLOTS OF
BIOLOGICAL AND PHYSICAL/CHEMICAL DATA WITH BOGUE HOMO PLOT
HIGHLIGHTED
57
-------
2 —
0.0
FIGURE D-1
Stressor-Response Relationships, Shown as Scatter plots: Decrease in Suitable Habitat
0.5 1.0 1.5
Turbidity (NTU, log)
100
90
80
70
60
50
40
30
20
0.0
0.5 1.0 1.5
Turbidity (NTU, Log)
2.0
0.5 1.0 1.5
Turbidity (NTU, log)
10 20 30 40 50 60 70 80 90 100
% Silt-clay Substrate
0°oo°
D.5 1.0 1.5
Turbidity (NTU, log)
0.5 1.0 1.5
Turbidity (NTU, log)
100
l l
1 1
1 1 1 1
90
-
80
-t °
1 0
0
o
70 J
!_ C© ° 0
_
""¦"SfcQOOO
CO
o fi'-SL
£D
60'
2.
-
2
Q) O
O
50
- «o
40
- o°° o
30
-o
O
20
1 1
1 1
1 1 1 1
20 30 40 50 60 70 80 90 100
% Silt-clay Substrate
0 10 20 30 40 50 60 70 80 90 100
% Silt-clay Substrate
20 30 40 50
% Gravel Substrate
20 30 40 50
% Gravel Substrate
Bogue Homo is indicated as a red symbol in each plot.
-------
FIGURE D-2
Stressor-Response Relationships, Shown as Scatter plots: Altered Food Resources (Organic Matter) and/or Decreased
Dissolved Oxygen
100
90
a
70
U)
CQ
bU
50
40
30
20
~i 1 1 1 1 1 1 r
_i i i i i i i i_
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9
Chemical Oxygen Demand (mg/l, log)
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9
Chemical Oxygen Demand (mg/l, log)
100
I I
80
i i
90
"
70
-
80
60
°
O ° ° °o
J3
O
o 70
O >\o0000lI,o a
8 50
-
^!o0oo o ° ° _
°
° °Sl'o0 °0CfcpO0
20
I I
n
°^l 00 n I
0.5 1.0
Total Nitrogen (mg/l, log +1)
1.5
0.0
0.5 1.0
Total Nitrogen (mg/l, log +1)
1.5
0.0 0.5 1.0 1.5
Total Phosphorus (mg/l, log +2)
100
90
a
70
U)
CQ
bU
50
40
30
20
i
I I I I
O
_
i
°
8 ©o
L
o 0 o o o °
o g o
° 8 8
a
o
a
O O
-
° 0
-
° °
-
"
I I I I
0.0 0.2 0.4 0.6 0.8 1.0
Total Organic Carbon (mg/l, log)
6r
1.2
0L
0.0 0.5 1.0
Total Nitrogen (mg/l, log +1)
0.2 0.4 0.6 0.8
Total Organic Carbon (mg/l,
0.0 0.5 1.0 1.5
Total Phosphorus (mg/l, log +2)
Bogue Homo is indicated as a red symbol in each plot.
-------
FIGURE D-3
Stressor-Response Relationships, Shown as Scatter plots: Increase in Ionic Strength and/or Increase in Toxic Substances
Total Chlorides (mg/l, log)
Total Chlorides (mg/l, log)
40
30
b
Q_
20
Z
LU
6
Total Dissolved Solids (mg/l, log)
30
s
20
c
Q.
Q
Total Dissolved Solids (mg/l, log)
Bogue Homo is indicated as a red symbol in each plot.
-------
APPENDIX E
STRESSOR-RESPONSE RELATIONSHIPS EXPRESSED AS BOX PLOTS OF
BIOLOGICAL DATA
61
-------
FIGURE E-1
100
90
80
2 70
I 60
g 50
m 40
5 30
20
10
0
Stressor-Response Relationships Expressed as Box Plots: Decrease in Suitable Habitat
1
Tprbidity
low high
Turbidity
low high
Turbidity
Turbidity
I
low high
1
Turbidity
low high
100
90
w 80
70
<"
2> 60
2 50
| 40
£ 30
^ 20
10
0
90
c 80
ro
o
« 70
| 60
c 50
03
£ 40
03
0 on
0 30
-2
1 20
^ 10
oN
o
low high low high
Bogue Homo is represented by the black dot in each plot.
—i
Turbidity
100
90
80
70
60
50
40
30
20
10
0
1
Turbidity
'T
low high
low high
Turbidity
100
90
80
70
(/)
o
60
(D
50
o
O
40
30
20
10
0
100
90
80
70
(/)
60
(i)
O)
c
50
O
40
30
20
10
0
1
Turbidity
low high
1
Turbidity
low high
-------
FIGURE E-1 cont
CD
CO
100
90
80
70
60
2?
0)
50
iZ
40
5s
30
20
10
0
Total Habitatj Score
low high
Total Habitat Score
low high
/o Silt
100
90
(/) 80
E
.« 70
§) 60
° 50
2 40
0)
£ 30
^ 20
10
0
100
90
80
2 70
O
W 60
O
Total Habitat Score
CO
CQ
50
40
30
20
10
0
15
10
ro 5
(O
3
6
low high
Total HabitatrScore
•hT
low high
% Silt
low 9 low high
Bogue Homo is represented by the black dot in each plot.
Total Habitat Score
30
Total Habita
Score
2 20
low high
low high
Total Habitat Score
low high
Total Habitat Score
low high
low high
-------
FIGURE E-2
Stressor-Response Relationships Expressed as Box Plots: Altered Food Resources and/or Decreased Dissolved Oxygen
CD
Total Organic Carbon
Total Organic Carbon
Total Organic Carbon
low high
Chemical Oxygen Demand
Dissolved Oxygen (% Sat.)
Dissolved Oxygen (% Sat.)
ro 10
70
low high
Total Nitrogen
Total Nitrogen
Total Nitrogen
« 70
£ 40
Total Organic Carbon
low high
Chemical Oxygen Demand
Q)
o
low high
80
60
50
40
Total Nitrogen
low high
low high
Bogue Homo is represented by the black dot in each plot.
-------
FIGURE E-2 cont.
100
90
80
70
2?
o
W 60
O
CO
CQ
50
40
30
20
10
0
30
* 20
ro
10
1
Total Nitrogen
low high
low high
Total Nitrogen
Total Phosphorus
CO 60
50
40
o 30
20
10
9 -
o 7"
o 6 -
co
a=
o
.c
5 -
4 -
3 -
2 -
1 -
0 —
1
Total Nitrogen
low high
Total Phosphorus
£ 50
low high
Total Phosphorus
T
12
10
¦c
O
60
40
o 20
¦c
O
w
Total Nitrogen
low high
Total Phosphorus
i=!=i
low high
1
Total Phosphorus
Total Nitrogen
40
20
low high
Tota Phosohorus
low high
Total Phosphorus
low high
low high
Bogue Homo is represented by the black dot in each plot.
-------
FIGURE E-2 cont.
Total Phosphorus
Total Phosphorus
Total Phosphorus
>• 30
a
Bogue
low high low high
Homo is represented by the black dot in each plot.
low high
CD
CD
-------
FIGURE E-3
Stressor-Response Relationships Expressed as Box Plots: Increase in Ionic Strength and/or Increase
Toxic Substances
Tota Chlorides
Tota Chlorides
low high
low high
Bogue Homo is represented by the black dot in each plot.
------- |