United States
Environmental Protection
Agency
Environmental Monitoring
Systems Laboratory
Las Vegas NV 89114
Research and Development
EPA-600/S7-84-063 July 1984
SER& Project Summary
Macroinvertebrate Inventories
of the White River, Colorado
and Utah:
Significance of Annual,
Seasonal, and Spatial Variation
in the Design of Biomonitoring
Networks for Pollution
Detection
C.E. Hornig
An aquatic macroinvertebrate moni-
toring program is proposed for early
warning detection of toxic discharges
to streams in oil shale development
areas. Changes in stream biota are used
to signal the need for increasing levels
of chemical analyses to identify and
quantify toxic pollutants. This study
compiles invertebrate data taken during
three seasons (spring, summer, and
fall) and over five years (1976 to 1980)
from riffles along the White River in
Colorado and Utah. Spatial and temporal
variations in the biota are described
along with their implications for the
development of a monitoring system
that incorporates such comparative
surveys. In addition, the data provide
benthic biological information that is
generally comparable to previous studies
on the White River and that can be used
to expand the biological monitoring
data base before massive oil shale
development ensues.
This Project Summary was developed
by EPA's Environmental Monitoring
Systems Laboratory, Las Vegas, NV, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
Development of oil shale resources in
the western United States will increase
the potential for contamination of surface
water resources with a host of inorganic*
and organic toxicants. It is not feasible to
establish an "early warning" stream
monitoring network for the timely detec-
tion and location of toxic substances
strictly through chemical monitoring
because the long list of inorganic and
organic constituents identified as energy-
related waste and effluent components
makes extensive use of comprehensive
chemical analyses prohibitively expensive.
Therefore, in order to expand surveillance
of potentially affected streams, this study
incorporates both biological and chemical
monitoring. Changes in stream biota are
used to detect subtle impacts from low-
levels of toxicants released to the stream.
These changes signal the need for
chemical analyses to identify and quantify
the substances). This method of coupling
biological early-warning surveys with
follow-up chemical analyses permits
-------
efficient monitoring of entire stream
systems for toxic discharges.
Comparative surveys of faun will most
effectively detect the effects of pollution
when natural variation of the biota is well
documented. This assures that false
alarms will be minimized and, at the same
time, that the effects of toxicants will not
be mistaken for naturally occurring biotic
fluctuations. The full report on which this
summary is based records and synopsizes
five years of macroinvertebrate data from
the White River of Utah and Colorado.
Study Area
The study area consists of a 200-km
section of the White River in Utah and
Colorado (Figure 1). Sampling site
distributions provide good representation
of riffle environments found along the
river and ranging from clear, cold
headwaters with stable substrates and
rich invertebrate fauna to highly turbid
downstream reaches with unstable
debris-choked substrates. The White is
representative of larger streams flowing
out of the Rocky Mountains and across
semi-arid lands of the western United
States to the Colorado River.
The White River watershed is currently
relatively undisturbed by human activities,
but industrialization, in the form of oil
shale mining and processing, has recently
begun within the watershed and is
expected to expand greatly over the next
few years.
Methods
Survey Design and Sampling
Sites
A total of 74 invertebrate sample sets
from 27 White River collection sites and
11 different collection times were
processed (Figure 1, Table 1). These
collections represent from 3 to 15
replicate samples each, with the large
majority consisting of five replicates.
Collections were designed to depict three
basic features of natural fauna! variation:
(1) temporal (annual, seasonal, and short-
term) changes at specific collection sites;
(2) spatial (longitudinal) changes along
the White River on specific dates; and (3)
annual changes in the degree of similarity
between adjacent collection sites.
Annual variation in community com-
position was assessed using collections
taken during September and early
October for four consecutive years, from
1976 through 1979. Collections from
spring, summer, and fall of both 1978 and
1979 documented seasonal changes
within a given year. Invertebrate data
were also compared over shorter time
intervals (early versus late September
1976 and April versus May 1978) to
examine within-season changes. Sep-
tember 1979 sample sets taken from 16
separate sites provided documentation of
longitudinal changes in community
structure. Changes in degree of similarity
between adjacent sites were evaluated
using data obtained at sites located just
upstream and downstream from the
confluences of Piceance and Yellow
Creeks, two White River tributaries most
likely to be affected by oil shale develop-
ment (Figure 1).
Field Methods and Sample
Processing
All invertebrate samples were obtained
in riffle areas using the Standardized
Traveling Kick Method. Formalin-preser-
ved samples were thoroughly rinsed in
the laboratory, and debris and organisms
Federal Oil Shale
Lease Tracts
Miles
Kilometers
Figure 1. Approximate locations of biological sampling stations in the White River. Utah and Colorado. 1976to 1980 (numbers indicate stream \
reaches, as shown in Table 1).
-------
were then separated from gravel. Most
groups of invertebrates were identified to
the lowest taxonomic level possible from
available literature.
Results
White River macrobenthic collection
sites were grouped for presentation
purposes by river section. Rangely,
Colorado, separates the downstream and
middle reaches, and Meeker, Colorado,
separates the middle and upstream
reaches (Figure 1 and Table 1). Typically,
differences exist in terms of relative
abundance rather than of presence or
absence. Important sources of natural
variation that should be considered for
the design of biomonitoring surveys
include annual variation, seasonal
changes, short-term temporal variation,
and fauna! changes along the river.
Incorporation of
Macroinvertebrate Data into an
Integrated Monitoring Program
In the absence of man-induced distur-
bances or natural phenomena that
substantially alter stream characteristics,
changes in macroinvertebrate commun-
ity composition in a downstream progres-
sion are normally gradual. Consequently,
abrupt changes in the biota between
adjacent upstream/downstream sites
may be an indication that significant
water quality changes have occurred,
thereby signaling the need to incorporate
additional elements into the monitoring
program.
The degree of community change will
determine the level of monitoring required
to identify and quantify causative agents
(Table 2). If differences between upstream
and downstream communities fall within
the range of natural variability, as
established by baseline sampling, the
communities are judged not to have been
affected, and level 1 monitoring is
continued. If level 1 monitoring reveals
between-site differences that exceed
natural variability, a water quality impact
is suspected, and level 2 monitoring is
initiated to identify causative agents. If
level 1 monitoring reveals complete or
nearly complete downstream alterations
in the biota, level 3 monitoring is
implemented.
Criteria for the degree of community
change needed to trigger a monitoring
decision are derived by close inspection
of predevelopment baseline data, includ-
ing variations in these data. Large
community changes will alert the aquatic
biologist to recent toxic discharges. In
such cases, the biologist will easily be
able to further pinpoint the location of
recent toxic discharge(s) by collecting
downstream from the reference site until
changes are first noted.
Because the purpose of the biomonitor-
ing approach outlined is to detect
sources of complex mixtures of toxic
pollutants, it is important that information
on fauna! changes be maximized. Thus,
the data are analyzed at the species level
rather than to reduce data into generalized
indices, such as diversity measures or
through use of clustering techniques.
The additional effort required to carefully
identify and compare more common
species and groups from adjacent stream
sites is minimal in relation to the total
Table 1. Seasons and Years of Macroinvertebrate Collections Processed from White River
Sampling Sites
Year and Season of Collections
Fall2 Fall Spr.3 Sum. Fall Spr. Sum. Fall Spr.
Collection Site' 1976 1977 1978 1978 1978 1979 1979 1979 1980
Downstream Reaches
20 X
30 XXX
40 X
48 X
50 X
52 X
55
60
Middle Reaches
90
92
100
110
112
114
115
120
130
140
160
170
XX X
XX X
X
X X
X X
X X X X X
X X X X X
X
X
X
X
X
X
X
X
X
X
X
X
X
X X
X X
X
XXX
XXX
X
X
X
X
X
X
Upstream Reaches
180
190
220
230
290
300
350
X
X
X
X
X
X X
X
X
X
X
'See Figure 1 for location of collection sites.
'First column represents early September 1976 and second column represents late September
1976.
3First column represents April 1978 and second column represents May 1978.
Table 2. Suggestions for Using Community Change to Establish Monitoring Requirements
Level
Degree of
Community
Change
Degree of Monitoring
1 Within natural
variation
2 Exceeds natural
variation
Complete or nearly
complete changes
Continue macroinvertebrate community monitoring
schedule
Conduct water column monitoring for conventional
parameters (e.g.. conductivity, dissolved oxygen.
temperature. pHj. Conduct sediment and/or tissue
analyses for suspected pollutants {e.g., priority pollutant
scan).
Conduct level 2 plus water column analyses for suspected
pollutants and bioassays with toxicant source (e.g.,
effluent, stream water, leachate, etc.).
-------
expenditure needed for the proper
monitoring of streams. As the aquatic
biologist becomes familiar with stream
conditions and the biology and ecology of
resident invertebrates, he will also
become more efficient and accurate at
making appropriate comparisons.
Information on the sensitivities of
many aquatic invertebrates, including
many species of immature insects, to
various toxic metals and certain organic
compounds (e.g., chlorinated hydrocar-
bons) is available in the literature.
Disappearance or reduction in density of
known sensitive species from a particular
site would provide an immediate clue that
toxic chemicals may be entering the
system. Chemical analysis of sediments
and stationary invertebrates may reveal
the presence of complex mixtures of
organic compounds that are below
detectable levels in the water or which
are caused by intermittent discharges
and would consequently be missed by
water-column sampling.
Although considerable information is
available on the organic and elemental
components of energy-related wastes
and effluents, information on the toxicity
to fresh water aquatic invertebrates of
complex mixtures of organic compounds
associated with these wastes is limited.
Information on the toxicity of complex
mixtures of wastes and effluents to
resident stream organisms could be
obtained from toxicity tests conducted in
the field and laboratory. Information
derived in this manner would aid consid-
erably in relating changes in community
composition to toxic wastes for purposes
of interpreting biomonitoring data.
Conclusions
• Although annual (year-to-year) varia-
tions in White River biota can be
substantial, these changes are gene-
rally consistent between adjacent
collection sites with similar habitats.
Thus, between-site faunal comparisons
offer good reliability for detecting
impacts originating in a stream reach
bracketed by adjacent upstream-
downstream sties. With this consider-
ation, incorporating the biomonitoring
survey method into a western stream
monitoring program, as illustrated, is
feasible.
• Short-term variability found in White
River biota dictates that when several
sites are sampled, they must be
sampled as close to the same date as
possible.
• Criteria for site selection must include
the degree of longitudinal (site-to-site)
variability of stream biota being
surveyed and locations of areas of
highest potential for toxic introductions.
Stream biota in the middle reaches of
the White River (between Rangely and
Meeker, Colorado) change only gradu-
ally. Consequently, collection sites
need only be established in the vicinity
of potential pollutant sources (e.g.,
disposal piles, tributaries, subsurface
seeps).
• Sampling frequency will depend on
the importance placed on early detec-
tion of water quality deterioration.
However, seasonal progression of
stream invertebrate communities
dictates a minimum of one sampling
each spring, summer, and fall. Rates of
recolonization of affected substrates
by drift from unaffected areas upstream
may, however, dictate more frequent
sampling (e.g., monthly or every two
months).
• Measurement of conventional water
quality parameters and chemical
analyses of selected invertebrates and
sediments may aid identification and
assessment of toxicants discharged
intermittently or which are below
detectable levels in water-column
analysis. These measurements also
assist identification of non-toxic fac-
tors that may cause community shifts
(e.g., low dissolved oxygen or tempera-
ture shifts).
• The stream biota survey approach is
suggested for detecting and assessing
water quality changes in western
lands as they open to development.
Continued experience with its use in
such a context will result in technique
refinement, thereby increasing effici-
ency of the approach.
• If field surveys indicate that toxic
pollutants have substantially altered
macroinvertebrate communities, sup-
plemental field and laboratory toxicity
tests are suggested to assess the
sensitivities of common White River
species exposed to suspected pollu-
tants. This information facilitates cor-
rect interpretation of survey results for
detection and assessment of pollutant
effects.
C. E. Hornig is with Department of Biological Sciences, University of Nevada-Las
Vegas, Las Vegas. NV89154.
Wesley L. Kinnev is the EPA Project Officer (see below).
The complete report, entitled "Macroinvertebrate Inventories of the White River,
Colorado and Utah: Significance of Annual, Seasonal, and Spatial Variation in
the Design of Biomonitoring Networks for Pollution Detection," (Order No. PB
84-198 936; Cost: $16.00, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Las Vegas, NV 89114
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
PS
------- |