Biological Condition Gradient (BCG) A Model for Interpreting Anthropogenic Stress on the Aquatic Environment

*>EPA

United States
Environmental Protection
Agency

The Biological Condition Gradient (BCG)

A Model for Interpreting Anthropogenic
Stress on the Aquatic Environment

What is the BCG?

The Biological Condition Gradient model (BCG) is a conceptual, scientific framework for interpreting biological
response to anthropogenic stress. The framework is based on common patterns of biological response to stressors
that have been observed by aquatic scientists across the United States (USEPA 2016) (Figure 1). It supports consistent
interpretation of biological condition independent of the specific method used to collect data, the type of waterbody
being assessed, or the location of the waterbody. The framework is often used in biological assessments by
formalizing expert knowledge of biological conditions in quantitative models for specific aquatic systems. The models
consist of quantitative decision rules that are used to assign sites to a level of condition along a stress gradient.

An Example of the BCG Using Benthic Macroinvertehrates

Levels of Biological Condition

Natural structural, functional,
and taxonomic integrity is
preserved.

Structure & function similar to
natural community with some
additional taxa & biomass;
ecosystem level functions are
fully maintained.

Evident changes in structure
due to loss ofsome rare native
taxa; shifts in relative
abundance; ecosystem level
functions fully maintained.

Moderate changes in structure
due to replacement of some
sensitive ubiquitous taxa by more
tolerant taxa; ecosystem
functions largely maintained.

Sensitive taxa markedly
diminished; conspicuously
unbalanced distribution
of major taxonomic groups;
ecosystem function shows
reduced complexity &
redundancy.

Extreme changes in structure and
ecosystem function; wholesale
changes in taxonomic
composition; extreme alterations
from normal densities.

Mayflies

c
o
O

"ro
o
'~)
o
o
m

1 inch

rarie f ies

Bte

nsects

ISBs

ies

L c

Biackflies V

Caddisflies

Beetles

Stonefliel

- Mayflies

Caddisflies
Beetle

Midges
Crapeflies

Level of Exposure to Stressors

Watershed, habitat, flow regime
and water chemistry as naturally
occurs.

Chemistry, habitat, and/or flow
regime severely altered from
natural conditions.

Figure 1. Predicable, measurable changes in the benthic macroinvertebrate communities are observed in aquatic
ecosystems in response to stress (stress increases from left to right of the diagram). For example, in these benthic
macroinvertebrate samples from streams in Maine, taxa that are sensitive to stress (blue circle) typically disappear
as stress levels increase while more pollution tolerant taxa (green circle, moderately tolerant) persist and highly
tolerant taxa dominate at high stress levels (red circle). In some cases, such as in the assemblage shown midway on
the gradient, there may be an increase in the number of taxa or individuals with initial rise in stressor levels such as
nutrient pollution. Photographs courtesy of Maine Department of Environmental Protection.

-------
How can the BCG be used?

In conjunction with other environmental data and information, state
water quality management programs can use the BCG to:

•	Consistently determine and communicate current environmental
conditions relative to natural, undisturbed conditions.

•	Describe, in a refined manner, what environmental conditions are
attainable either through protection or restoration.

•	Provide feedback on how to achieve goals for protection or restoration
by tracking incremental changes in condition and assessing trends due
to management actions.

• Communicate what is biologically predicted to be gained, or lost, with
different management decisions.

...and, how has the BCG been used so far?

State water quality management programs have already used BCGs to:

•	Designate refined aquatic life uses along a gradient of stress, e.g. excellent, good, fair;

•	Develop biological criteria to measure attainment of the designated aquatic life use;

•	Inform Use Attainability Analysis;

•	Better understand the quality of reference sites and thus, more accurately define baseline
conditions;

•	Consistently describe current conditions and measure the cumulative impact of multiple stressors
on aquatic life;

•	Support adaptive management by tracking incremental changes in biological condition as remedial
actions, controls and best management practices are implemented;

•	Identify high quality waters for protection and provide an early warning signal of incremental
degradation;

•	Communicate to stakeholders the predicted impact of decisions on protection and management of
aquatic resources.

BCG Success Stories - State Implementations of BCG Models

Minnesota, Alabama and California have incorporated the BCG as part of their biological assessment and
criteria programs to complement existing biological indices or other bioassessment tools and approaches. These
states have developed BCGs for their streams and assigned levels to help communicate reference-based
thresholds defined by either a biological index or a predictive taxa loss model. For example, reference streams in
Minnesota and Alabama typically scored at a BCG level considered "good", affirming the quality of the
reference sites and providing more detailed biological description of the biota that is being protected. There was
one region in Minnesota where reference streams scored lower than other regions due to both past and present
regionwide impacts from agriculture and human development - corresponding with a BCG level generally
considered "fair". The Minnesota BCG was used to more precisely define and communicate current conditions
and establish attainable goals for aquatic life in the different regions. Like Minnesota and Alabama, California
developed and compared the BCG levels of sites to independently derived bioassessment models. The scoring
aligned well, and California is able to communicate to the public the meaning of reference-based biological
expectations. These expectations will ultimately be used to inform levels of nutrients protective of biological
condition.

The BCG is a flexible framework
that can be applied to any
waterbodyand ecological region
and implemented by monitoring
programs that have different
level of technical capabilities
(e.g. one assemblage or more;
annual to quarterly sampling;
etc.). The more robust the
monitoring program, the more
confidence in the assessment.

-------
What are the components of a BCG?

The Biological Condition Gradient (BCG) has two key components:

• Attributes are measurable components of a biological system (Karr and Chu 1999)

Example: species composition, such as the number and proportion of sensitive and tolerant taxa

• Levels are the discrete levels of biological condition across a stressor-response curve. Example: Level 1
=natural; Level 6 = severely altered from natural

Attributes

The BCG framework depicts ecological condition in terms of measurable ecological characteristics, or attributes, of an
aquatic community in response to anthropogenic stress. The BCG attributes correspond to the characteristics used by
bioassessment programs to measure biological condition, determine restoration potential and track recovery. It is not
necessary to have information from all the attributes to develop a BCG. An expert panel recommends attributes that
correspond with what they consider to be the most ecologically important characteristics of the system and utilizing
what data is available for model development. To date, BCG models for streams, rivers and coral reefs have primarily
utilized attributes I - VI which characterize change in taxa in response to stress.

Biological and other ecological attributes used to characterize the BCG.

Attribute

Description

1. Historically documented,
long-lived, or regionally
endemic taxa

Taxa known to have been supported according to historical, museum, or archeological records, or taxa
with restricted distribution (occurring only in a locale as opposed to a region), often due to unique life
history requirements (e.g., sturgeon, American eel, pupfish, unionid mussel species).

II. Highly sensitive taxa

Taxa that are highly sensitive to pollution or anthropogenic disturbance. Tend to occur in low numbers,
and many taxa are specialists for habitats and food type. These are the first to disappear with disturbance
or pollution (e.g., most stoneflies, brook trout [in the east], brook lamprey).

III. Intermediate sensitive
and common taxa

Common taxa that are ubiquitous and abundant in relatively undisturbed conditions but are sensitive to
anthropogenic disturbance/pollution. They can be found at reduced density and richness in moderately
disturbed sites (e.g., many mayflies, many darter fish species).

IV. Taxa of intermediate
tolerance

Ubiquitous and common taxa that can be found under almost any condition, from unstressed to highly
stressed sites. They are broadly tolerant but can decline under extreme conditions (e.g., filter-feeding
caddisflies, many midges, many minnow species).

V. Highly tolerant taxa

Taxa that are of low abundance in undisturbed conditions but increase in abundance in disturbed sites.
Opportunistic species able to exploit resources in disturbed sites (e.g., tubificid worms, black bullhead).

VI. Nonnative or intentionally
introduced

Any species not native to the ecosystem (e.g., Asiatic clam, zebra mussel, carp, European brown trout).
Additionally, there are fish native to one part of North America that have been introduced elsewhere.

VII. Organism condition

Anomalies of the organisms; indicators of individual health (e.g., deformities, lesions, tumors, disease).

VIII. Ecosystem function

Processes performed by ecosystems, including primary and secondary production; respiration; nutrient
cycling; decomposition. For example, shift of lakes and estuaries to phytoplankton production and
microbial decomposition under disturbance and eutrophication.

IX. Spatial/ temporal extent
of detrimental effects

The spatial and temporal extent of cumulative adverse effects of stressors; for example, groundwater
pumping in Kansas resulting in change in fish composition from fluvial dependent to sunfish.

X. Ecosystem connectance

Access or linkage (in space/time) to materials, locations, and conditions required for maintenance of
interacting populations of aquatic life; the opposite of fragmentation. For example, levees restrict
connections between flowing water and floodplain nutrient sinks (disrupt function); dams impede fish
migration, spawning.

-------
Levels

The BCG has been divided into six levels of biological condition in response to increasing levels of stress
(Figure 2). The six levels provide a flexible framework for a state to determine the number of levels that
can be implemented. The number of levels realized will be influenced by both natural and programmatic
reasons. For example, in a predominately forested perennial stream ecosystem supporting highly sensitive
taxa, it may be technically possible to discriminate 6 different levels of condition based on shifts from highly
sensitive to moderately sensitive to tolerant species. However, for stream systems that naturally
experience high variability in flow, ranging from no flow to scouring floods in a single stream channel,
species adapted to harsh conditions are expected in the natural, unstressed condition. In this scenario, it
may be difficult to recognize six levels of change in species composition when the highest level of the BCG
is defined by tolerant species. Additionally, some states may only be capable of discriminating three or
four levels along a gradient of stress depending upon the technical capabilities of their monitoring
program, whiie others might be capable of discerning six or more levels based on highly proficient
programs and robust data sets (U.S. EPA2013).

The Six Levels of the BCG

Levels of Biological Condition

Natural structural, functional,
and taxonomic integrity is
preserved.

Structure & function similar to
natural community with some
additional taxa & biomass;
ecosystem level functions are
fully maintained.

Evident changes in structure
due to loss ofsome rare native
taxa; shifts in relative
abundance; ecosystem level
functions fully maintained.

Moderate changes in structure
due to replacement of some
sensitive ubiquitous taxa by more
tolerant taxa; ecosystem
functions largely maintained.

Sensitive taxa markedly
diminished; conspicuously
unbalanced distribution
of major taxonomic groups;
ecosystem function shows
reduced complexity &
redundancy.

Extreme changes in structure and
ecosystem function; wholesale
changes in taxonomic
composition; extreme alterations
from normal densities.

Level of Exposure to Stressors

Watershed, habitat, flow regime
and water chemistry as naturally
occurs.

Chemistry, habitat, and/or flow
regime severely altered from
natural conditions.

Figure 2, The Biological Condition Gradient - a scientific framework to interpret biological response to increasing effects
of anthropogenic stress on aquatic ecosystems. Six levels of condition (Y axis) along a gradient of increasing stress (X axis)
ranging from naturally occurring to severely altered conditions are narratively described using biological information. In
this figure, the color gradient is a quick visual cue for condition level.

-------
How is a Quantitative BCG model developed?

A key step in development of a quantitative BCG model (Figure 3) is to convene a panel with expertise in taxonomy and
aquatic ecology. The panel is charged with calibrating the conceptual BCG with data for a specific aquatic system and
developing quantitative decision rules for assigning sites to BCG levels for that system using a combination of expert
elicitation and consensus.

BCG Calibration

Step 2

Analyze and prepare data

Identify

(expert
panel

t

Step 3
Convene expert panel

Step 4
Develop decision model

¦a

tm

ja

3

Step 5
Test and
review model:

adequate
performance?

Yes

Calibrated BCG model with
quantitative decision rules for
assigning sample sites to BCG levels

Decision rules are logic statements that experts use to assign sites to BCG levels, starting with
narrative statements such as "If Plecoptera richness is high in this stream, based on stream size,
substrate and flow, then biological condition is ciose to natural, unimpacted conditions. I would assign
this stream to a BCG level 2." An expert typically articulates 2 or more logic statements in making a
BCG level assignment. Through an iterative process of expert elicitation and metric testing, a
consensus set of quantitative decision rules for each BCG level are developed by the expert panel. The
set of decision rules constitutes a BCG quantitative model for an aquatic ecosystem.

-------
What expertise is needed on the panel?

The BCG consensus approach asks the experts to make judgments on the ecological significance of changes in the
aquatic biota and come to consensus on a set of quantitative decision rules for assigning sites to BCG levels. For
this approach to be credible and valid, the panel should:

•	be comprised of experts with a wide and deep breadth of knowledge and expertise across the
biodiversity of the region and not be constrained to a single agency to minimize internal bias. At
a minimum, expertise in taxonomy and aquatic ecology specific for the waterbody and the region
is required.

•	have enough cumulative experience to understand what historic communities looked like and what
happened to communities prior to initiation of the CWA to avoid defining BCG levels 1 and 2 based
on existing degraded conditions.

•	include expertise in different and widely accepted methodological and analytical approaches for
development of a robust and broadly applicable mode

What makes an effective panel?

For the panel to successfully construct a quantitative BCG model:

• the panel's first task is to develop a detailed narrative description of physical characteristics that define pristine,
or unstressed, condition for the study area. This establishes an agreed upon natural standard for evaluating
site data.

• the data sets used in model development should represent the complete range of conditions that
exist within the region for which the BCG is being developed.

• it is essential that the expert logic in developing the decision rules be fully documented so the
reasoning underlying the rules is transparent
and can be easily implemented in
the future.

• panel discussion is facilitated to fully
and fairly engage all experts and not
allow any individual to dominate the
discussion or dismiss the opinion and
logic of others.

• the range of variability among experts
around a site score is examined and
included in model documentation so
that that the strengths and limitations
of the quantitative model can be
clearly understood, and the model
appropriately applied.

-------
BCG Model Development

Site Data Expert Judgment BCG Level Assignments

BCG
Attribute

Species

Common

Name

Count

brook silverside

gizzard shad

river carpsucker

quillback

III

highfin

carpsucker

III

northern hog
sucker

silver redhorse

III

black redhorse

golden redhorse

III

shorthead
redhorse

spotfin shiner

common carp

redfin shiner

hornyhead chub

sand shiner

• Taxa Life
Histories
Field Ecological

Experienced^knowledge

Assign BCG Level
Relative to
Natural

What
Biological
Change
Would Lead to
a Higher or
Lower BCG
Level?

Panelist

BCG
Level

Rationale

The presence /diversity of sucker and
Centrarchids species (intermediate
sensitive) are consistent with my
experience in this size stream and my
expectation for BCG level 3

There should be more
intermediate sensitive species for
this size stream relative to what 1
expect for BCG Level 3 streams.

Expected more diversity in general;
and expected more minnow diversity
(8-10 in BCG Level 3 streams of this
size vs. the 5 collected), caveat: some
might have been missed due to
method (boat sampling vs. wadeable
sampling).

Expected more benthic species
particularly with the presence of
coarse substrates and low percent
fines.

There is a lack of benthic species given
the presence of a firm (coarse)
substrate compared to our definition
of BCG Level 3 streams.

Good sucker species diversity
comparable to other BCG Level 3
streams of this size, but expecting
more minnow and darter species.

(excerpt- complete data not shown) Median BCG Score - 3-

Figure 4. How does the panel develop numeric decision rules? Over the course of data evaluation, the expert panel develops a set of
narrative and quantitative decision rules that define each BCG level. As a first step, the panel defines undisturbed and/or minimally
disturbed conditions (geophysical, chemical, landform, etc) for the waterbody of interest and the region in which it is located. This serves
as the shared benchmark from which the panel evaluates biological data and assigns sites to BCG levels. Data from sites representing a
range of conditions, from no or low to high levels of stress, are then provided to the panel. The experts are asked to assign each site into
one of the BCG levels based on its biological data including taxa lists, tolerance values, cumulative abundance data and graphs. Facilitators
elicit the reasoning and metrics used by each expert in their ratings which are used to derive the rules for classifying BCG levels. This figure
shows an excerpt from the recording of individual expert logic applied in assessing a stream site in Illinois using fish assemblage data, only
a portion of the datasheet is shown for illustration. A plus or minus indicates a site's data does not fit 100% with an expert's expectation
for a level because there are indications in the data that a site may be closer to an adjacent level in quality - either the level above (+) or
below (-). This information assists the panel in articulating what they judge as a significant change that would result in a level change.

Through an iterative process combining individual expert elicitation and development of consensus, a set of narrative rules are
proposed for each BCG level. Quantitative measures, or metrics, are tested for how well they replicate expert judgement and
discriminate between levels (Figure 5). Following their initial development, a preliminary set of quantitative decision rules are
tested by the panel with new data sets to ensure that sites are consistently evaluated and replicate their expert judgement.

-------
BCG Level 3: Evident changes in structure of the biotic community and minimal changes in ecosystem
function. Some changes in structure due to loss of some rare native taxa; shifts in relative abundance
of taxa but intermediate sensitive taxa are common and abundant; ecosystem functions are fully
maintained through redundant attributes of the system.

Coral Reef Benthic Narrative Rule

Coral Reef Benthic Quantitative Rule

Coral cover is moderately high

LPI coral cover > 20 (10-30) %

Coral are moderately diverse

¥

LPI coral species > 4 (3-5) species

Sensitive Coral Species are represented

LPI Attribute II, III, + IV species > (1-3) species

Cover is mostly live and healthy organiOTs

Bare substrate and algal turf with sediment
< 30 (40-20) % >

Live cover of Orbicella is relatively high ^

Live cover of Orbicella > 20 (15-25) % f

LPI Attll,lll,IVtaxa>2

% Live Star Coral (Orbicella) > 20%
%liveOrbicellacover z07o

4

BCG Level

3 4 5
BCG Level

LPI coraltaxa>4species

4

BCG Level

4 5
BCG Level

Figure 5: Metrics are tested that translate a narrative rule into a quantitative measure. If a metric does not discriminate a BCG
level from another, the metric is not used. This example highlights two BCG level 3 narrative rules for coral cover (top) and the
testing of the metrics for discriminating between BCG levels 3 and 4 (bottom) for coral reef communities off the coast of
Puerto Rico and U.S.V.I. Box plots illustrate the median, 25th and 75th percentiles (box) and outlier values (whiskers) for a
group of data. The red line represents the central decision threshold for BCG level 3 that replicates expert panel consensus.
Level of confidence, or model performance, is calculated but not visually shown here.

-------
What happens when there are no pristine or minimally stressed conditions?

Historical data and information can be used to develop a detailed narrative description or, if possible, narrative or
semi-quantitative decision rules, for BCG level 1 when there are no or few pristine, or minimally stressed,
conditions remaining in the region for which a BCG is to be developed. When this situation occurs, the quantitative
model is developed only for those BCG levels that can be empirically defined. A detailed, narrative description of
BCG level 1, and sometimes BCG level 2, can be used to more accurately communicate to the public current
conditions by providing context and, in some places, serve as a narrative guide for restoration of sites assessed as
close to the higher BCG level conditions.

How does the BCG model work?

A site is evaluated based on the decision rules for each level starting with the highest level of the model (Figure 6).
The evaluation cascades (e.g., filters) to the next level until a BCG level is assigned.

How does the BCG model work? Like a cascade...

Example: coldwater sample from sit where watershed size is < 10 mi2 and
brook trout are native*

Does the sample
meet ALL BCG
Level 1 criteria?

NO

Does the sample
meet ALL BCG
Level 2 criteria?

NO

~

Does the sample
meet ALL BCG
Level 3 criteria?

NO

~

# Total taxa < 4

Sensitive taxa (Att I + II) - present
Native brook trout - present
% Sensitive taxa (Att II + III) > 50%
% Sensitive individuals (Att II + III) > 60%
% Tolerant (Att V + Va + Via) individuals < 5%
Non-native salmonids (Att VI) - absent

YES

~

Assigned to
BCG LEVEL 1

1# Totaltaxa < 8

1 Sensitive taxa (Att II + III) - present
1 Native brook trout - present
1 % Sensitive taxa (Att II + III) > 40%

1 % Native brook trout: total salmonid individuals > 40%
1 % Tolerant non-salmonid (Att V + Va + Via) individuals < 10%

YES

~

Assigned to
BCG LEVEL 2

# Total individuals < 20
Sensitive taxa (Att I + II) - present
Salmonids - present

% Sensitive taxa (Att II + II) and salmonid taxa > 25% YES
% Sensitive taxa (Att II + II) and salmonid individuals > 20%
% Non-native trout: total salmonid individuals < 70%
% Tolerant (Att V + Va + Via) individuals < 40%

~

Assigned to
BCG LEVEL 3

And so on...

* In some situations, alternate rules had to be developed. For example, more taxa naturally occur in large vs. small streams, so total taxa
richness rules were adjusted for watershed size. Some rules also had to be adjusted for streams in which brook trout are not native.

Figure 6. Biological data from a site are analyzed for a match with the quantitative decision rules for a BCG level.

-------
What constitutes the final BCG model?

The final model is a set of quantitative decision rules for assigning sites to BCG levels. The rules are supported by
expert knowledge and narrative descriptions for each level (Figure 7, coral reef fish assemblage rule for BCG level
3) and includes a method for entering site data (e.g. R code, database, or spreadsheet) to routinely assign sites to
the appropriate BCG level.

Definition: Evident changes in structure of the biotic community and minimal changes in
ecosystem function— Some changes in structure due to loss of some rare native taxa; shifts in
relative abundance of taxa, but intermediate sensitive taxa are common and abundant; ecosystem
functions are fully maintained through redundant attributes of the system

Conceptual
Model

Level

Attributes

CO
fl>

Physical structure: Moderate to high rugosity, moderate reef built above bedrock, some irregular cover for fish
habitat, water slightly turbid, low sediment, floes or film on substrate.

d)
_l

O
O

m

a>

Corals: Moderate coral diversity; large old colonies (Orbicella) with some tissue loss; varied population structure
(usually old colonies, few middle aged & some recruitment); Acropora thickets maybe present; rare species ab-
sent.

Sponges: Autotrophic species present but highly sensitive species missing.

a
E

Gorgonians: Gorgonians more abundant than in level 1.

*
p

Condition: Disease and tumor prevalence slightly above background level, more colonies have irregular tissue loss.

a.
a

Fish: Noticeable decline of large apex predators, groupers, snappers, etc. Small reef fish more abundant.

<

Vertebrates: Large, long-lived species locally extirpated (turtles, eels).

a

o
o

Other Invertebrates: Diadema, lobster, small crustaceans & polychaetes less abundant than level 1, large sensitive
anemones species missing.

o

Algae/plants: Crustose coralline algae present but less, turf algae present and longer, more fleshy algae present.

Narrative
Decision Rules
for Fish
Assemblage

BCG Level 3

Total Taxa

Richness moderate to high

Number of all sensitive taxa

Small to moderate proportion of richness

Total biomass (kg/km2)

Fish biomass moderate to high

Piscivores

Presence of some snappers and other piscivores

Parrotfish

Large body parrotfish present

Damselfish

Damsels do not dominate catch

Groupers

Groupers present

Reef habitat rule

More stringent

Hardbottom habitat rule

Less stringent

Quantitative
Decisions Rules for
Fish Assemblage

BCG Level 3

Total Taxa

nt_total > 15 (10-20) (nt = # of taxa)

Number of all sensitive taxa

nt_att23 > 6 (4-8)

Total biomass (kg/km2)

bio total > 35,000 (30,000-40,000) (bio = biomass)

Piscivores

pb_SP + pb_LP > 0 (pb =% biomass)

Parrotfish

nt_Parrot2 > 1 (0-2)

Damselfish

pd_damsels < 25% (20-30) (pd = % density)

Groupers

nt_Grouper > 0

Reef habitat rule*

best 6 of 7 rules

Hardbottom habitat rule*

best 5 of 7 rules

Figure 7. The BCG process is intended to explicitly link expert knowledge with statistical analysis in development of a
quantitative BCG model. This example shows the sequential linkage between the conceptual level description, narrative
and then numeric rules for assigning a coral reef fish community to BCG level 3. *The rules are adjusted for substrate.

-------
BCG Success Stories - Applications of the BCG

In 2018, a BCG for benthic macroinvertebrates in Puget Sound Lowlands and Willamette Valley streams was
developed by an expert panel that included scientists from King and Snohomish Counties, Oregon Department of
Environmental Quality| Washington Department of Ecology, USEPA, and USGS, as well as regional taxonomists.
Regional stream macroinvertebrate data (taxa lists with 500 individuals per sample) were used to develop the BCG
model. The quantitative model was developed for BCG levels 2 through 6. There were no pristine conditions remaining
in the study area. However, BCG level 1 was narratively defined using historic data and expert knowledge. Once the
rules were defined by the experts and consensus achieved, the rules were automated in Multiple Attribute Decision
Models using logic and set theory. The model replicated expert panel's decisions with over 96% accuracy. The states
and counties are moving forward to use the BCG to improve communication with stakeholders, enhance their existing
monitoring and assessment programs, and set restoration targets and priorities.

The state of Connecticut provides an example of
how the BCG can be used to communicate with
the public. As part of Connecticut's 2018
monitoring and assessment report, the state
provides a web site that maps the BCG
categories based on rules developed from a
New England expert panel. The map illustrates
the BCG levels for fish and/or macroinvertebrate
assemblages and illustrates the location of sites
with minimal stress, moderate stress, or major
stress (From State of CT environmental web site:
h ttps://ctdeepwatermonitoring. github. io/
BCGMap/).

More tools and examples will become available
as states integrate the BCG into their
water quality management programs.

v 1 V r t,,J

^ Macroinvertebrate Data
| Fish Data

BCG Value

0 1 or 2 (Minimal Stress)
^ 3 or 4 (Moderate Stress)
0) 5 or 6 (Major Stress)

Interested in more information?

Please contact Susan Jackson, Office of Water (Mail Code 4304T), Environmental Protection
Agency, 1200 Pennsylvania Avenue NW., Washington, DC 20460or by email at
jackson.susan k@epa. gov

EPA Office of Water
Washington DC 20460
EPA-822-F-21-001
February 2021

References:

U.S. EPA. 2013. Biological Assessment Program Review: Assessing Level of Technical Rigor to Support Water Quality Management.
EPA 820-R-13-001. Office of Science and Technology, Washington, DC 20460.

U.S. EPA. 2016. A Practitioner's Guide to the Biological Condition Gradient: A Framework to Describe Incremental Change in Aquatic
Ecosystems. EPA 842-R-16-001. Office of Science and Technology, Washington, DC 20460.

Karr, J.R. and Chu, E.W. 1999. Restoring Life in Running Waters: Better Biological Monitoring. Island Press, Washington DC.

-------