Biological Sciences, Santa Barbara City College pipeline center for sustainability

Biology 130: Methods in Field Biology

Field Technique:  Invertebrate Survey

dragonfly mayfly


The following is excerpted from Benthic Macroinvertebrates in Freshwaters- Taxa Tolerance Values, Metrics, and Protocols

There are compelling reasons for the apparent popularity of freshwater macroinvertebrates in current biomonitoring practice; they offer a number of advantages:

  1. they are ubiquitous, so they are affected by perturbations in many different habitats,

  2. they are species rich, so the large number of species produces a range of responses,

  3. they are sedentary, so they stay put, which allows determination of the spatial extent of a perturbation,

  4. they are long-lived, which allows temporal changes in abundance and age structure to be followed, and

  5. they integrate conditions temporally, so like any biotic group, they provide evidence of conditions over long periods of time (the videotape referred to above).

To be fair, macroinvertebrates also have disadvantages, but these can be mostly overcome by proper experimental design. For example, macroinvertebrates do not respond to all impacts; the distribution and abundance of macroinvertebrates may be affected by factors in addition to the perturbation in question; and the distribution and abundance of macroinvertebrates vary seasonally.

There has been a melding of quantitative and qualitative approaches known as rapid assessment or rapid biomonitoring. These approaches are meant to provide an initial screening of water bodies for possible further investigation.


Metrics (or indices) allow the investigator to use meaningful indicator attributes in assessing the status of assemblages and communities in response to perturbation. For a metric to be useful, it must have the following technical attributes:

  1. ecologically relevant to the biological assemblage or community under study and to the specified program objectives;

  2. sensitive to stressors and provides a response that can be discriminated from natural variation.

Reference sites

Because species assemblages differ naturally among different regions (ecoregions) in North America and even between stream orders in the same ecoregion, many metrics require a reference site for each evaluation. The reference can be an unaffected reach in the same stream or in a neighbouring stream of the same order. Many of the indices in the protocols use 'tolerance scores' that were derived from large data bases of both published and unpublished studies of experts for all the major groups of taxa. Colonial taxa, like Porifera (sponges) and Bryozoa (moss animals), are not included in the scoring systems (Mackie, 2001).


Benthic macroinvertebrates are common inhabitants of lakes and streams where they are important in moving energy through food webs. The term "benthic" means "bottom-living", so these organisms usually inhabit bottom substrates for at least part of their life cycle; the prefix "macro" indicates that these organisms are retained by mesh sizes of ~200-500 mm (Rosenberg and Resh, 1993). The most diverse group of freshwater benthic macroinvertebrates is the aquatic insects, which account for ~70% of known species of major groups of aquatic macroinvertebrates in North America. More than 4000 species of aquatic insects and water mites have been reported from Canada. Thus, as a highly diverse group, benthic macroinvertebrates are excellent candidates for studies of changes in biodiversity. However, benthic macroinvertebrates can be difficult to work with unless the proper study design is used (Rosenberg and Resh, 1993). For example:

  1. quantitative sampling is difficult because the contagious (i.e. clumped or patchy) distribution of benthic macroinvertebrates requires large numbers of samples to achieve reasonable precision in estimating population abundance. The resulting processing and identification requirements for samples can be costly and time consuming. An alternative would be to use rapid assessment procedures;

  2. the distribution and abundance of benthic macroinvertebrates are affected by a large number of natural factors, which have to be accounted for to determine changes in biodiversity; and

  3. some groups of benthic macroinvertebrates are taxonomically difficult, although the development of new and improved keys is a high priority in research.

The collection of benthic macroinvertebrates from lakes and streams is usually a straightforward procedure using standard equipment. However, the removal of organisms from background material can be tedious and time-consuming unless available labor-saving strategies are used (see below) and the identification of organisms to the species level, when possible, requires substantial training and skill. The processing of samples can be successfully accomplished by non-specialists, but the involvement of systematists is recommended for species-level identifications. Data-analysis procedures are standard, and can be done by anyone trained in elementary statistics.


Samples are usually subsampled to save processing time because either the samples are excessively large or there are large numbers of them. Thus, the sample is quantitatively reduced, the invertebrates from a known portion of the sample are counted, and these counts are extrapolated back to the entire sample. Samples need to be homogeneous, so large organisms or pieces of debris should be removed prior to subsampling. Several methods are available; for example, the volumetric method of Wrona et al., the weight-based method of Sebastien et al., and the spatial (sample-splitting) method of Marchant have proven reliable. In the Marchant method, a predetermined number of invertebrates (100, 200, or 300) is removed from a box subdivided into 100 cells and then counting stops. No matter what method is used, precision and accuracy need to be assessed initially by comparing selected subsamples to the total sample. Subsampling error should be estimated in at least 10% of the samples being processed by sorting another subsample of equal size (Environment Canada and Department of Fisheries and Oceans, 1993); allowable deviations between counts should be set at a reasonable but consistent level.

Rare species may be missed by subsampling, which is an important concern in biodiversity studies. Thus, rare species deserve special consideration when selecting a subsampling method; alternatively, avoid subsampling completely.

Subsampling 100 animals is sufficient for rapid bioassessments. The results of the ANOVAs and associated power calculations revealed only modest gains in the ability to distinguish lakes using subsamples of 200 or 300 animals. The only exceptions to this conclusion are studies that use indices based on richness measures and rare taxa, where larger counts are necessary to adequately census rare individuals.

Identification of Specimens

The sweep-net collections of adult insects along shorelines are valuable to the identification of immature aquatic forms. Another, more time-consuming endeavor is to establish rearing programs to provide associated stages. The taxonomy of aquatic insects is based mainly on the adult form, although the immatures are the forms most frequently collected in aquatic sampling. If the adult, cast pupal skin, and cast last larval skin are available for holometabolous insects (i.e. those with complete metamorphosis such as midge flies), or the adult and a series of cast nymphal skins are available for hemimetabolous insects (i.e. those with incomplete metamorphosis such as mayflies), then the immature forms can often be identified by working backward from the adult. Merritt and Cummins (1996) review field-based and laboratory rearing methods for major insect groups.


Use of kick nets

A study in Nova Scotia used a standardized 5 minute kick and sweep technique. Starting at a depth of 1-metre, the kicker would slowly walk towards the shore and back out to the same spot kicking up the substrate. If time permitted, another transect was completed immediately adjacent to the first and the process was repeated until 5 minutes were up. A second kick and sweep replicate was also completed, this would start immediately adjacent to the last transect of the first kick and sweep.

Following behind the kicker, the sweeper would gather the sample into the net. A seine net with a mesh of 280μm size was used to insure that most macroinvertebrates would be captured. The sample was then emptied into a large bucket and water was used to rinse the net off completely into the bucket. In order to avoid adding any extra sample into the collected substrate, water was poured over the back of the net.

To avoid biased picking of the more easily visible large individuals, the collectors focused on looking for movement in the water, picking out anything that caught the eye. The collected organisms were placed into a container of 95% ethanol (which was subsequently diluted by some lake water sucked up into the droppers with the organisms) for preservation and transportation to the lab. If a picked subsample contained a large amount of plant material, a portion of this was also taken. The remaining water and material from the picked sample was returned to the lake.


Biological Indices (Kirsch, 1999; Mandaville, 1999)

We will use the Modified Family Biotic Index

Family Biotic Index (FBI, Metric 2- RBP II) The Biotic Index was originally developed by Hilsenhoff (1982) to provide a single ‘tolerance value’ which is the average of the tolerance values of all species within the benthic arthropod community. The Biotic Index was subsequently modified to the family-level with tolerance values ranging from 0 (very intolerant) to 10 (highly tolerant) based on their tolerance to organic pollution.

Modified Family Biotic Index, RBP II (Plafkin et al ., 1989) Tolerance values (Table-2) range from 0 to 10 for families and increase as water quality decreases. The index was developed by Hilsenhoff (Hilsenhoff, 1988) to summarize the various tolerances of the benthic arthropod community with a single value. The Modified Family Biotic Index (FBI) was developed to detect organic pollution and is based on the original species-level index (BI) of Hilsenhoff (Table-3, Chapter V of this report). Tolerance values for each family were developed by weighting species according to their relative abundance in the State of Wisconsin.

In unpolluted streams the FBI was higher than the BI, suggesting lower water quality, and in polluted streams it was lower, suggesting higher water quality. These results occurred because the more intolerant genera and species in each family predominate in clean streams, whereas the more tolerant genera and species predominate in polluted streams. Thus the FBI usually indicates greater pollution of clean streams by overestimating BI values and usually indicates less pollution in polluted streams by underestimating BI values. The FBI is intended only for use as a rapid field procedure. It should not be substituted for the BI; it is less accurate and can more frequently lead to erroneous conclusions about water quality (Hilsenhoff, 1988).

The family-level index has been modified for the RBP II to include organisms other than just arthropods using the genus and species-level tolerance values adopted by the State of New York (Bode et al., 1991, 1996, 2002; cf., Table-4, Appendix A of this report). Although the FBI may be applicable for toxic pollutants, it has only been evaluated for organic pollutants. The formula for calculating the Family Biotic Index is:


Family Biotic Index Water Quality Degree of Organic Pollution
0.00-3.75 Excellent Organic pollution unlikely
3.76-4.25 Very good Possible slight organic pollution
4.26-5.00 Good Some organic pollution probable
5.01-5.75 Fair Fairly substantial pollution likely
5.76-6.50 Fairly poor Substantial pollution likely
6.51-7.25 Poor Very substantial pollution likely
7.26-10.00 Very poor Severe organic pollution likely

benthic invert tolerance

benthic invert tolerance

There are many other indices that can be used depending on the question:

Simpson’s Diversity Index (D) Diversity within the benthic macroinvertebrate community was described using the Simpson’s diversity index (“D”), which was calculated as:

simpson diversity equation

where “pi ” is the proportion of individuals in the “i th ” taxon of the community and “s” is the total number of taxa in the community. This index places relatively little weight on rare species and more weight on common species (Krebs, 1994). Its values range from 0, indicating a low level of diversity, to a maximum of 1-1/s.

Shannon-Wiener Diversity Index (H) Used by the Gerritsen et al (1998), the Shannon-Wiener Diversity index (H) is commonly used to calculate aquatic and terrestrial biodiversity. This index was calculated as:

shannon weiner diversity equation

where “pi ” is the proportion of individuals in the “i th ” taxon of the community and “s” is the total number of taxa in the community. As the number and distribution of taxa (biotic diversity) within the community increases, so does the value of “H” (Gerritsen et al., 1998).

Biological Monitoring Working Party (BMWP) The Biological Monitoring Working Party score (BMWP) provides single values, at the family level, representative of the organisms’ tolerance to pollution. The greater their tolerance towards pollution, the lower the BMWP score. To reflect conditions within North America, Mackie (2001) has modified this index. BMWP was calculated by adding the individual scores of all families, and order Oligochaeta (Friedrich et al., 1996), represented within the community ( cf. Table-6, Appendix B this report).

Average Score Per Taxon (ASPT) The Average Score Per Taxon (ASPT) represents the average tolerance score of all taxa within the community, and was calculated by dividing the BMWP by the number of families represented in the sample (Friedrich et al ., 1996). From this value, the water quality of each lake was assessed (Mackie, 2001; cf. Table-5, Appendix B this report).

Taxa Richness (TR, Metric 1- RBP II) Taxa Richness (TR) indicates the health of the community through its’ diversity, and increases with increasing habitat diversity, suitability, and water quality (Plafkin et al ., 1989). TR equals the total number of taxa represented within the sample. The healthier the community is, the greater the number of taxa found within that community.

EPT Index (Metric 6- RBP II) The Ephemeroptera, Plecoptera, and Trichoptera (EPT) index displays the taxa richness within the insect groups which are considered to be sensitive to pollution, and therefore should increase with increasing water quality. Initially developed for species-level identifications, this index is valid for use at the family-level (Plafkin et al., 1989). The EPT index is equal to the total number of families represented within these three orders in the sample.

Ratio of EPT and Chironomidae (EPT/C, Metric 4- RBP II) The abundance of EPT and Chironomidae indicates the balance of the community, since EPT are considered to be more sensitive and Chironomidae less sensitive to environmental stress (Plafkin et al., 1989). A community considered to be in good biotic condition will display an even distribution among these four groups, while communities with disproportionately high numbers of Chironomidae may indicate environmental stress (Plafkin et al., 1989). The EPT/C index is calculated by dividing the sum of the total number of individuals classified as Ephemeroptera, Plecoptera, and Trichoptera by the total number of individuals classified as Chironomidae.

ETO Index The Ephemeroptera, Trichoptera, and Odonata (ETO) index represents the taxa richness of these groups (Gerritsen et al., 1998). The ETO index is equal to the total number of families represented within these three orders in the sample. Ephemeroptera, Trichoptera and Odonata are considered to be sensitive to pollution. This index has no reference, but provides a comparison of the abundance of these groups within one study site over time.

Percent Contribution of Dominant Family (%DF, Metric 5- RBP II) The Percent Contribution of Dominant Family or percent dominance (%DF) equals the abundance of the numerically dominant family relative to the total number of organisms in the sample. This index indicates the present state of the community balance at the family level. For example, a community dominated by relatively few families would have a high %DF value, thus indicating the community is under the influence of environmental stress (Plafkin et al., 1989).

Community Loss Index (CLI, Metric 7- RBP II) The Community Loss Index (CLI) measures the loss of benthic taxa in a study site with respect to a reference site. Values range from 0 to “infinity” and increase as the degree of dissimilarity between the sites increases (Plafkin et al ., 1989). CLI was calculated as:

community loss equation

where “a” is the number of taxa common to both sites, “d” is the total number of taxa present in the reference site, and “e” is the total number of taxa present in the study site. In this study, CLI was determined by comparing the total number of taxa present in each study lake (“e”) to the number of taxa present in each site of the reference lake (“d”). This was done to account for the variation that occurs under natural conditions.

Percent Similarity Comparisons (PSC- RBP II) Several of the aforementioned indices (TR, FBI, scr/f-c, EPT, EPT/C, % DF, CLI, and shredders/total) are used by the RBP-II to assess the biological condition of study sites (Plafkin et al., 1989). Referring to this, the indices were given scores depending on whether the actual values indicated the community to be non-impaired (score = 6), moderately impaired (score = 3), or severely impaired (score = 0) by pollution. The biological condition scores for each site were summed and divided by the total indices score for the reference site, e.g. for each index the reference site would receive a score of 6, to provide a percent comparison between the reference and study sites. Among the indices used, scr/f-c regularly produced no measurable values due to the absence of scrapers in the reference lake (Dollar Lake), i.e. the dividend had a value of 0 and the index was not computable. In the sites where scr/f-c did not produce a numeric value, this index was not included in the calculation of percent comparison. For FBI, scr/f-c, and EPT/C, the “average” indices for each study lake were compared to the indices for each site of the reference lake, while the TR and EPT values, based on total representation within each study lake, were compared to the indices for each site of the reference lake. Actual % DF values for each site of the study lakes were used and not compared to the reference lake. As well, actual CLI values for each study lake (total) were used, since comparison to the reference lake was incorporated into its calculation. In addition, the ratio of shredders/total was not used since no plant material was collected to determine the presence of organisms classified as shredders. Finally, the percent comparisons were averaged to provide an overall percent comparison of each lake to the reference lake.

Percent Model Affinity (PMA) Percent Model Affinity (PMA) is used to compare how similar a study site is with respect to a model non-impacted community, and is based on the percent abundance of seven major macroinvertebrate groups (Novak and Bode, 1992). From this, the biological effects of pollutants on an existing community can be measured. For the kick samples obtained for this study, the model non-impacted community consisted of 40% Ephemeroptera, 5% Plecoptera, 10% Trichoptera, 10% Coleoptera, 20% Chironomidae, 5% Oligochaeta, and 10% Others (Novak and Bode, 1992). The percent contributions for each of the seven groups at each site (summing 100) of the study lakes were determined and compared with those of the model community. PMA was calculated by summing the lesser of the two values (actual and model values) for each site of the study lakes, and used to assess water quality.

Additional Indices In accordance with the recommendations of the Ontario Ministry of the Environment (Somers et al., 1998), the following indices were also included in the present study. These indices were calculated as their proportional abundance relative to the total number of organisms in the sample: % Oligochaetes , % Amphipods , % EPT , % Insects , % Non-Dipteran Insects , %Dipteran Insects , % Gastropods , and % Pelecypods .


Some indices use feeding methods as a way to categorize diversity

Feeding measures

Feeding measures or trophic dynamics encompass functional feeding groups and provide information on the balance of feeding strategies (food acquisition and morphology) in the benthic assemblage. Examples involve the feeding orientation of scrapers, shredders, gatherers, filterers, and predators. Trophic dynamics (food types) are also included here and include the relative abundance of herbivores, carnivores, omnivores, and detritivores. Without relatively stable food dynamics, an imbalance in functional feeding groups will result, reflecting stressed conditions. Trophic metrics are surrogates of complex processes such as trophic interaction, production, and food source availability (Karr et al. 1986, Cummins et al. 1989, Plafkin et al. 1989). Specialized feeders, such as scrapers, piercers, and shredders, are the more sensitive organisms and are thought to be well represented in healthy streams. Generalists, such as collectors and filterers, have a broader range of acceptable food materials than specialists (Cummins and Klug 1979), and thus are more tolerant to pollution that might alter availability of certain food. However, filter feeders are also thought to be sensitive in low-gradient streams (Wallace et al. 1977). The usefulness of functional feeding measures for benthic macroinvertebrates has not been well demonstrated. Difficulties with the proper assignment to functional feeding groups has contributed to the inability to consider these reliable metrics (Karr and Chu 1997).

The main categories of functional feeding groups include (Feeding Strategies and Pollution Tolerance of Macroinvertebrates):

  • Shredders Chew on intact or large pieces (>1 mm) of plant material.
    • Examples: giant stoneflies, Northern caddisflies.
    • Found in: leaf packs, water-logged wood, headwater streams and areas with a high percentage of canopy cover.
  • Scrapers/grazers Scrape off and consume thin layer of algae growing on solid substrates in shallower waters.
    • Examples: snails, flatheaded mayflies, water pennies
    • Found in: more open areas with enough sunlight to support algal growth; rocks in open-canopied areas, mid-stream reaches.
  • Collectors (collector/filterers and collector/gatherers) Consume very small pieces of detritus (<1 mm)
    • Examples: common netspinner caddisflies, back flies, brush-legged mayflies, mussels
    • Found in: rocks and mud; common in all reaches, but make up larger proportion in lower reaches where sediment collects
  • Predators Feed on living animals; may swallow smaller prey whole, tear pieces out of larger prey, or suck out body fluids
    • Examples: predaceous diving beetles, dragonfly Iarvae, common stoneflies
    • Found in: all habitat types, in smaller proportion relative to other feeding groups

Ratio of Scraper and Filtering Collector Functional Feeding Groups (scr/f-c, Metric 3- RBP II) The Scraper and Filtering Collector index (scr/f-c) is calculated by dividing the total number of individuals classified as scrapers by the total number of individuals classified as filtering collectors within the sample. This index is independent of taxonomy, since some families may represent several functional feeding groups (Plafkin et al., 1989). When compared to a reference site, shifts in the dominance of a particular feeding group corresponds to the abundance of a particular food source, which reflects a specific type of impact on the community (Plafkin et al., 1989).

Ratio of Shredder Functional Feeding Group and Total Number of Individuals Collected- CPOM Sample (shredders/total, Metric 8- RBP II) Also based on the Functional Feeding Group Concept, the abundance of the Shredder Functional Group relative to the abundance of all other Functional Groups allows evaluation of potential impairment as indicated by the CPOM-based Shredder community. Shredders are sensitive to riparian zone impacts and are particularly good indicators of toxic effects when the toxicants involved are readily adsorbed to the CPOM and either affect microbial communities colonizing the CPOM or the Shredders directly. The degree of toxicant effects on Shredders versus Filterers depends on the nature of the toxicants and the organic particle adsorption efficiency. Generally, as the size of the particle decreases, the adsorption efficiency increases as a function of the increased surface to volume ratio. Because water-borne toxicants are readily adsorbed to FPOM, toxicants of a terrestrial source (e.g., pesticides, herbicides) accumulate on CPOM prior to leaf fall thus having a substantial effect on Shredders. The focus of this approach is on a comparison to the reference community which should have a reasonable representation of Shredders as dictated by seasonality, region, and climate. This al lows for an examination of Shredder or Collector “relative” abundance as indicators of toxicity (Plafkin et al., 1989).


 Copyright Notice and Credits
Revised 26 January, 2015
Bio 130 Student Information Page Our Raccoon will bring you home Biological Sciences Home Page