Qualitative and quantitative methods to
process benthic macroinvertebrate (BMI)
samples have been developed and tested
by the U.S. Geological Survey?s National
Water Quality Laboratory Biological Group.
The qualitative processing method is
based on visually sorting a sample for up to
2 hours. Sorting focuses on attaining organisms
that are likely to result in taxonomic
identifications to lower taxonomic levels (for
example, Genus or Species). Immature and
damaged organisms are also sorted when
they are likely to result in unique determinations.
The sorted sample remnant is
scanned briefly by a second person to determine
if obvious taxa were missed.
The quantitative processing method is
based on a fixed-count approach that targets
some minimum count, such as 100 or
300 organisms. Organisms are sorted from
randomly selected 5.1- by 5.1-centimeter
parts of a gridded subsampling frame. The
sorted remnant from each sample is resorted
by a second individual for at least 10
percent of the original sort time. A large-rare
organism search is performed on the unsorted
remnant to sort BMI taxa that were
not likely represented in the sorted grids.
After either qualitatively or quantitatively
sorting the sample, BMIs are identified by
using one of three different types of taxonomic
assessment. The Standard Taxonomic
Assessment is comparable to the
U.S. Environmental Protection Agency
Rapid Bioassessment Protocol III and typically
provides Genus- or Species-level taxonomic
resolution. The Rapid Taxonomic Assessment
is comparable to the U.S. Environmental
Protection Agency Rapid Bioassessment
Protocol II and provides Familylevel
and higher taxonomic resolution. The
Custom Taxonomic Assessment provides
Species-level resolution whenever possible
for groups identified to higher taxonomic
levels by using the Standard Taxonomic
Assessment. The consistent use of standardized
designations and notes facilitates
the interpretation of BMI data within and
among water-quality studies. Taxonomic
identifications are quality assured by verifying
all referenced taxa and randomly reviewing
10 percent of the taxonomic identifications
performed weekly by Biological
Group taxonomists. Taxonomic errors discovered
during this review are corrected.
BMI data are reviewed for accuracy and
completeness prior to release. BMI data are
released phylogenetically in spreadsheet
format and unprocessed abundances are
corrected for laboratory and field subsampling
when necessary.