Sample Collection Sorting And Taxonomic Identification Of Benthic Macroinvertebrates

sample collection, sorting and taxonomic identification of benthic macroinvertebrates standard operating procedure 1. scop

Sample Collection, Sorting and Taxonomic Identification of Benthic
Macroinvertebrates
Standard Operating Procedure
1.
Scope and Applicability
The methods described herein are used in wadeable streams (1st through
5th order range). Application of this approach to large rivers is
beyond that scope of this method.
1.
General Considerations
Data Quality Objectives (DQOs) are quantitative and qualitative
statements developed to specify the quality of data needed to meet the
project’s needs.
1.
Index Period
The index period is the period of time that samples should be
collected to minimize seasonal variation. A single index period
provides a strong database enabling a wide range of management
objectives to be addressed. However, establishing index periods during
multiple seasons allows for a program to understand seasonal
variation. The index period for Region 8 is in late summer and early
fall.
The selection of the appropriate sampling period should be based on 3
factors that reflect efforts to1:
1.
minimize year-to-year variability resulting from natural events
(ie. drought, fire, etc),
2.
maximize gear efficiency, and
3.
maximize accessibility of targeted assemblage.
When monitoring for trends at a particular site, minimize seasonal
variation by sampling as close as possible to the same date each year.
2.
Site Selection
Site selection can either be “targeted” (focuses on potential
problems) or “probabilistic” (provides overall information). Random
selection of sites provides an unbiased assessment of the condition of
the waterbody, whereas site-specific (targeted) design provides
assessment of individual sites or stream reaches. Site-specific
locations must be similar enough to have similar biological
expectations for meaningful comparisons of impairment.
Riffle areas with cobble substrates are generally the most diverse and
productive habitat type, however these may not be representative of
the predominant type of habitat in the stream. The study design should
describe the variety of habitats, the rationale for representative
sampling, and the various sampling techniques that are appropriate for
the study.
3.
Sample Collection Methods
Quantitative sampling techniques sample a known area which allows for
the enumeration of organisms to determine population density,
diversity and abundance.
Semi-quantitative sampling methods are designed to collect a wide
variety of aquatic macroinvertebrates and determine diversity and
abundance.
4.
Checklist of Field Supplies
*
Macroinvertebrate sample bottles (wide-mouth)
*
Lab markers, external and internal labels, clear tape
*
Sampler (Modified Turtox kick net (500 µm mesh) with Dolphin
bucket (504µm) or Hess)
*
Wide-mouth 500 ml plastic jars
*
Field data sheets
*
95% ethanol
*
Camera
*
GPS Unit
*
500 µm sieve for washing and sorting out large objects
*
pencils
*
clipboard
*
hip boots or waders
*
container to store/transport samples
2.
Sample Collection Procedures
The following section describes sample collection procedures for
sampling benthic macroinvertebrates in wadeable streams. A schematic
overview of the BMI sampling procedure, including field and lab
methods is shown in Figure 1.
_s1041
_s1050







Figure 1. Schematic overview of Semi-Quantitative Sampling Protocol
for BMI Sampling.
1.
Semi-Quantitative Sampling Protocol
1.
Sampling Protocol Summary
The kick-net method is a semi-quantitative sampling technique designed
to collect a representative macroinvertebrate sample in riffle
habitats.2 Runs or other habitat may also be sampled.3 Where cobble
substrate is the predominant habitat, the single habitat approach
provides a representative sample of the stream reach. However, if the
cobble substrate represents less than 30% of the sampling reach, the
multi habitat approach should be taken.4
2.
Equipment
A Turtox modified bottom rectangular kick net (500 µm mesh) with a
Dolphin bucket (504 µm mesh) is used.
3.
Sample Collection
1.
Traveling Kick Net Method – Single Habitat Approach
The traveling kick net method disturbs the substrate while moving
diagonally and upstream to dislodge macroinvertebrates from their
habitat. As the macroinvertebrates are dislodged, they are swept into
the net by the current and collected in the Dolphin bucket. A
composite sample is collected from individual sampling spots in
riffles and runs of various velocities. Generally, a minimum of 2 m2
composited area is sampled.5
A 100 m reach representative of the characteristics of the sampled
stream should be selected. Whenever possible, the sampling area should
be at least 100 meters upstream from any road or bridge. The site
visit form should be completed prior to sampling to document site
descriptions, weather conditions, land use, in-stream attributes, GPS
coordinates and any other physical/chemical parameters. After
sampling, the macro habitats sampled section can be completed along
with any observations of aquatic flora and fauna.
A sample is collected by starting at the downstream end of the reach
and proceeding upstream. Using the kick net, several kicks are used to
sample at each individual sampling location within the reach. A kick
is a stationary sampling technique in which the net is positioned
downstream of the area sampled. Using the toe or heel of the foot, the
area in front of the net is disturbed, dislodging the upper layer of
gravel/cobble and scraping the underlying bed. Larger substrate should
be picked up and rubbed gently by hand to remove macroinvertebrates
into the net.
During each kick, pay attention to large material coming into the net.
If possible, clean off the material prior to it clogging the bucket
opening (ie. with your hands carefully rub off any macroinvertebrates
off of large stones, twigs, leaf litter, etc.). Inspect the large
material for macroinvertebrates prior to discarding it. After each
kick, wash the material collected by dipping the net into the running
water to wash the material into the Dolphin bucket.
Depending on the stream reach, a sample can be collected by starting
on the downstream corner of a riffle and ending on the far end
opposite of the side started on (see Fig. 2). If the thalweg is too
deep, alternatively, a zig-zag sampling pattern can be done (see Fig.
3).

Figure 2. Drawing of diagonal sampling path for traveling kick net
method.

Figure 3. Drawing of alternative sampling path for traveling kick net
method.
After flushing any clinging macroinvertebrates into the Dolphin
bucket, gently swing the net to remove as much of the water collected
as possible. Transfer the sample from the bucket to the sample
container(s) and preserve it in enough 95% ethanol to cover the whole
sample. Place a label inside of the sample with the site ID, date
collected and collector’s initials. Place another label with the same
information on the outside of the container. If needed, use clear
packing tape to ensure that the external label does not come off. If
more than one container is needed, make sure to indicate the container
number on both labels (ie. 1 of 2, 2 of 2). Record the same
information on the sample log sheet.
2.
Traveling Kick Net Method – Multiple Habitat Approach
Many streams vary in substrate, from cobble to sandy sediments. These
various habitats that are present in a stream reach should be taken
into account. As mentioned above, where cobble substrate is the
predominant habitat, the single habitat approach provides a
representative sample of the stream reach. However, if the cobble
substrate represents less than 30% of the sampling reach, the multi
habitat approach should be taken in which major habitats are sampled
in proportional representation within a sampling reach.6
All available instream habitats are sampled using a total of 20 kicks
are taken from all major habitat types in the reach. Each habitat
represented should be proportional to the amount of sampling done for
the reach (ie. if 50% of the habitat is sandy substrate, then 10 kicks
should be done in sandy substrate). Habitat types contributing less
than 5% of the stable habitat in the stream reach should not be
sampled. The remaining kicks should be allocated proportionately among
the predominant habitats.7
A 100 m reach representative of the characteristics of the sampled
stream should be selected. Whenever possible, the sampling area should
be at least 100 meters upstream from any road or bridge. The site
visit form should be completed prior to sampling to document site
descriptions, weather conditions, land use, in-stream attributes, GPS
coordinates and any other physical/chemical parameters. After
sampling, the macro habitats sampled section can be completed along
with any observations of aquatic flora and fauna.
A sample is collected by starting at the downstream end of the reach
and proceeding upstream. Using the kick net, a total of 20 kicks are
used over the length of the reach. A kick is a stationary sampling
technique in which the net is positioned downstream of the area
sampled. Using the toe or heel of the foot, the area in front of the
net is disturbed, dislodging the upper layer of gravel/cobble and
scraping the underlying bed. Larger substrate should be picked up and
rubbed gently by hand to remove macroinvertebrates into the net.
The kicks collected from multiple habitats are composited into a
single sample. During each kick, pay attention to large material
coming into the net. If possible, clean off the material prior to it
clogging the bucket opening (ie. with your hands carefully rub off any
macroinvertebrates off of large stones, twigs, leaf litter, etc.).
Inspect the large material for macroinvertebrates prior to discarding
it. After each kick, wash the material collected by dipping the net
into the running water to wash the material into the Dolphin bucket.
After flushing any clinging macroinvertebrates into the Dolphin
bucket, gently swing the net to remove as much of the water collected
as possible. Transfer the sample from the bucket to the sample
container(s) and preserve it in enough 95% ethanol to cover the whole
sample. Place a label inside of the sample with the site ID, date
collected and collector’s initials. Place another label with the same
information on the outside of the container. If needed, use clear
packing tape to ensure that the external label does not come off. If
more than one container is needed, make sure to indicate the container
number on both labels (ie. 1 of 2, 2 of 2). Record the same
information on the sample log sheet.
4.
Sample Quality Control
Samples must include the site ID, date collected, and collector’s name
on the inside and outside label. Chain-of-custody (COC) forms must
include the same information as the sample container labels.
After sampling has been completed at each site, all gear should be
rinsed thoroughly and inspected for any organisms or debris.
Field replicates are collected to measure total method error and
should represent 10% of the sites to evaluate precision or
repeatability of the sampling technique or the collection team. Field
replicates are either two or more samples collected side-by-side or
consecutively at the sampling site. Replicate samples should be taken
at reaches that are similar in depth, substrate, composition, and
gradient. Be careful to not contaminate one site with the other when
disturbing substrate (ie. do not disturb the duplicate site with
debris from the first site sampled). The difference between the
replicates represents total method error (ie. reproducibility of the
sampling technique, heterogeneity of the site, subsampling error, and
identification error).
Relative Percent Difference (RPD) is used to express precision:
RPD = [(x1 – x2)/{(x1 + x2)/2}] x 100
RPD = relative percent difference (%)
x1 and x2 = duplicate measurements of the same parameter
The results of replicated (ie. metric value) samples are usually
suggested to be <20% RPD. However, each project must state its
required replicate precision criteria based on those projects DQO’s.
2.
Laboratory Processing For Macroinvertebrate Samples
All samples should be recorded in a “Sample Log” upon receipt by
laboratory personnel (See Appendix). All information on the labels,
including the number of containers for each sample site, should be
recorded.
1.
Subsampling
Subsampling uses a representative portion of the field-collected
sample or analysis. Subsampling reduces the effort required for
identification of all macroinvertebrates collected in a sample and
sorts organisms from the sample matrix of detritus, sand and mud. A
fixed-count approach is preferred for subsampling. The subsample is
preserved separately from the remaining sample.
Thoroughly rinse the sample in a 500 µm sieve to remove preservative
and fine sediment. Large organic material that was not removed during
sampling should be rinsed, inspected for organisms, and then
discarded. When transferring the sample to the sieve and rinsing, care
should be taken to not lose any sample and to be gentle with the
macroinvertebrates (taking care during all processing steps aids in
the identification process). Sometimes, the sample will need to soak
in water to hydrate the macroinvertebrates and ensure that all the
alcohol is removed from the sample. If more than one container was
used, the contents of all the containers should be composited and
homogenized.
After rinsing, gently spread the sample evenly across a gridded pan.
Add just enough water to the pan to make sure that the organisms do
not dry out. However, if too much water is added, the organisms will
float in and out of the grids, making it hard to target the grid.
Grids are selected randomly using dice or a random numbers table. All
organisms are picked from the randomly selected grid. Any organism
that is lying over the line separating two grids is considered within
the grid that its head is in. If the head cannot be determined, the
organism is considered to be in the grid containing most of its body.
Pick randomly selected grids one at a time until the desired number is
reached (subsample of 300). Pick each grid thoroughly until all of the
organisms are picked. Even if the total count will go over 300,
continue picking the grid until finished. Organisms picked can be
transferred from the gridded tray to a Petri dish with 70% ethanol,
making sure to not include any algae or other vegetation in the
sub-sample. If there is vegetation, carefully use tweezers to separate
the vegetation from the organism.
The total number of grids picked should be noted on the laboratory
picking sheet. Return the leftover sample to the container and add a
label that says “sample residue” along with the original label.
Transfer the picked organisms form the Petri dish to a film
canister/vial with 70% ethanol. Label the outside and inside with site
ID, date collected, total count and initials of the subsampler (see
label example in Appendix).
1.
Count vs. Non-Count Specimens
1.
Include but do not count
1.
S mall critters - Include small organisms,
but do not count small organisms that are too small
for the taxonomist to be able to identify to genus.
For example, in the photos to the r ight
(this page and previous page), the insect on the far
right is an example of an organism to include but
not count in the total 300 minimum organism count.
The organism in the left hand side of the pictures
is about 2 cm long. If you are in doubt about
whether or not to count an organism, ask the
taxonomist or be safe and do not count it. A
reference vial with examples of “small” animals will
be provided to each sorter to help determine what
“small” is.
2.
D amaged critters – Include, but do not count
body parts not including at least the head AND
thorax. This means that you include but do not count
detached heads, half bodies, or partially decomposed
bodies of organisms, as shown in the image to the
right.
3.
Other rules - Include, but do not count:
1.
• eggs
2.
• exuvia
3.
• aquatic insect pupa – most often Diptera,
1.
Trichoptera
4.
• empty mollusk shells
1.
• surface dwellers such as Collembola
2.
(Springtails), Gerridae
3.
and Veliidae (striders)
5.
• worm fragments lacking the anterior
6.
(head) end (see picture – the anterior end
7.
lacks any eyespots or suckers)
8.
• adult aquatic insects (excluding
9.
coleoptera & hemiptera)
10.
• terrestrial insects
11.
• fish, amphibians
2.
DO NOT include or count: plant parts (eg. stems, seeds,
etc) or algae
2.
Subsampling Quality Control
Ten percent of the samples in each study/project should be examined by
a qualified co-worker. The grids chosen and tray used for sorting and
any organisms missed by the picker should be examined. If more than
10% of the subsample (ie. for a subsample count of 300, 30 organisms)
is found in remaining grids that have been “picked” training will need
to be done again. Samples will be checked until the qualified
co-worker decides that training is complete.
All laboratory equipment used for subsampling will be thoroughly
rinsed, and picked free of organisms or debris after each sample.
2.
Taxonomy
Identification should be to the lowest practical taxonomic level
(LPTL). Typically, this is to the genus level or subfamily for the
Chironomidae (level 1 taxonomic resolution). Some identification keys
for the region allows for identification to the species level.
Genus/species level of identification provides more accurate
ecological and environmental information and impairment sensitivity
than identification to family level. However, specimen condition (ie.
damage, early instar, poor slide mount) may only allow for
identification at more coarse levels (tribe, subfamily, family, etc).
Most organisms are identified to genus level using a dissecting
microscope. Midges (Chironomidae) are mounted on slides and identified
using a compound microscope.
The identity, stage and count should be recorded on the Laboratory
Bench Sheet (See Appendix). The taxonomist’s initials should also be
recorded on the sheet.
Archived samples are placed in vials with 70% ethanol and a label
indicating the site ID, date collected, taxa, stage and taxonomist’s
initials.
1.
Count vs. Non-count Specimens
All specimens in the sub-sample should be identified and counted by
the taxonomist. However, there are some exceptions that are described
as non-counts, which include:
*
Empty mollusk shells
*
Worm fragments lacking the anterior end
*
Body parts NOT INCLUDING at least the head and thorax
*
List of taxa to reject from benthic analyses:
*
Cladocera
*
Copepoda
*
Branchiura
*
Non-benthic insects such as Gerridae, Hydrometridae,
Notonectidae, Collembola and Gyrinidae (adults)
*
Nematoda
For more clarification see the section on count vs. non-count
specimens in the sub-sampling section above.
2.
Quality Control for Taxonomy
Accuracy can be determined using any of the following:
*
Museum-based type material;
*
The most current and accepted taxonomic literature; or
*
A reference collection, verified by an independent taxonomic
specialist.
Precision is evaluated by direct comparison of results (list of taxa
and number of individuals) of a randomly-selected sample that is
processed by 2 taxonomists or laboratories. Precision is quantified
for both specimen enumeration and taxonomic identification for each of
the QC samples.
Reference collection specimens should be properly labeled, preserved
and stored in the laboratory for future use. Verification of the
reference collection should be recorded. Information on completed
samples will be recorded in the “sample log”. Additionally, a library
of basic taxonomic literature should be maintained in the taxonomic
laboratory. Taxonomists should participate in periodic training on
specific taxonomic groups to ensure accurate identification.
3.
Appendix: Field and Laboratory Forms
1.
Field Forms
1.
Site Visit Form

2.
C OC
3.
Label Example
SiteID:_LP1________________
Stream: La Plata River________
Collection Date: 03/23/09_____
Sampler: Kick Net Single Habitat
Collector(s): EGM, CRA______
Container: 1 of 2_____________
2.
Laboratory Forms
1.
Sample Log

2.
Laboratory Bench Sheet
1.
BMI Subsampling

2.
BMI Identification

3.
Label Example
Project ID: SUIT Biannual_____
SiteID:_LP1________________
Collection Date: 03/23/09_____
Subsampler: PH_____________
Count: 306_________________
1.
Appendix B: BMI Taxa ID


2.
Appendix D: Metric Table


1 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, Second Edition. EPA
841-B-99-002. U.S. Environmental Protection Agency; Office of Water;
Washington, D.C.
2 Barbour et al. 1999.
3 MDEQ. 2006. Sample collection, sorting, and taxonomic identification
of benthic macroinvertebrates. Standard operation procedure
WQPBWQM-009, revision no. 2. Water Quality Planning Bureau, Montana
Department of Environmental Quality, Helena, Montana. (Available from:
http://www.deq.state.mt.us/wqinfo/QAProgram/WQPBWQM-009rev2_final_web.pdf)
4 Barbour et al. 1999.
5 Barbour et al. 1999.
6 Barbour et al. 1999.
7 Mid-Atlantic Coastal Streams Workgroup (MACS). 1996. Standard
operating procedures and technical basis: Macroinvertebrate collection
and habitat assessment for low-gradient nontidal streams. Delaware
Department of Natural Resources and Environmental Conservation, Dover,
Delaware.