![[logo] US EPA](http://www.epa.gov/epafiles/images/logo_epaseal.gif){kind=logo}
Sample Processing
Sample processing can range from a cursory examination and documentation of presence/absence to a quantitative measurement of abundance and distribution among the samples and will likely include subsampling, particularly for algae and macroinvertebrates. The level of and process for subsampling should be tested to ensure that the sample is representative of the waterbody segment being assessed. King and Richardson (2002) found that fixed-area subsamples were generally less efficient than fixed counts, with 200- and 300-individual fixed counts resulting in significantly greater assemblage-environment than 10% fixed areas. At the more rigorous levels, the number of organisms needed for an adequate subsample is specified (Cao et al. 2002; Hughes et al. 2002). Cao and Hawkins (2005) found that results of simulations imply that those bioassessment methods that rely on estimates of taxa richness derived from small fixed-count subsamples (e.g., 100-300 individuals) may significantly underestimate true biological impairment. A systematic treatment of samples is required, regardless of the level of subsampling to ensure the greatest extent of accuracy and precision. The ultimate determination of biological condition is dependent upon the appropriateness of sample processing. In the laboratory, as in the field, a strong QA/QC program is desired to ensure that 1) sample sorting procedures are being followed and no organisms are missed in the sample, and 2) the taxonomy is consistent and accurate. This is done by performing QC checks on both the sample sorting and organism identification steps, and retaining a voucher collection of the samples and reference collection of the taxa found in the samples. Having samples re-sorted by independent laboratories and organisms re-identified by independent taxonomists are critical for assuring data integrity for assessments (Stribling et al. 2008).
Some Frequently Asked Questions
Question: How does the level of macroinvertebrate subsampling affect the results of taxa richness metrics?
Answer: Analyses (data from New Jersey and Ohio) support the conclusion that the rate of taxa richness loss increases at subsample sizes of less than 500 organisms. However, the potentially negative influence of taxa richness loss can be offset if metrics are standardized to the 95th percentile of the subsample and transformed to metric scores as a component of an index. We found that sites in New Jersey were ranked similarly in biological condition regardless of subsample size when metric scores were compared versus taxa loss alone. Small subsample sizes may not impact condition assessments; however, causal analyses may be compromised because the characterization of the “true” benthic community is not known. Based on Ohio analyses, variability around an index score will be less at higher subsample sizes (See Figure below).
Figure (above). Comparison of recalculated Ohio ICI values based on 200, 400, 600, and 800 organism subsamples to Ohio ICI values based on whole samples. Scatterplots (with best fit line and 95 percent confidence intervals o the slope) of subsample-derived index scores on whole sample index scores. Subsample sizes are shown above each plot.
References
Cao, Y. and C.P. Hawkins. 2005. Simulating biological impairment to evaluate the accuracy of ecological indicators. Journal of Applied Ecology 42:954-965.
Cao, Y., D.P. Larsen, R.M. Hughes, P.L. Angermeier, and T.M. Patton. 2002. Sampling effort affects multivariate comparisons of stream assemblages. Journal of the North American Benthological Society 21:701-714.
Hughes, R.M., P.R. Kaufmann, A.T. Herlihy, S.S. Intelmann, S.C. Corbett, M.C. Arbogast, and R.C. Hjort. 2002. Electrofishing distance needed to estimate fish species richness in raftable Oregon rivers. North American Journal of Fisheries Management 22:529-540.
King, R.S. and C.J. Richardson. 2002. Evaluating subsampling approaches and macroinvertebrate taxonomic resolution for wetland bioassessment. Journal of the North AmericanBenthological Society 21:150-171.