more sensitive, but less exposed, faunistically richest instream habitat types. In addition, depositional habitats, such as pools, represent the most prevalent types of habitat across the Nation and may facilitate national and regional comparisons. Consequently, the structures of both types of communities are important to the determination of biological water-quality conditions (Kerans and others, 1992).
The types of instream habitats sampled, particularly the habitat identified as being faunistically richest, vary with the geographic location of the stream and the longitudinal position of the sampling reach along that stream. In wadeable streams and rivers, the richest habitats are probably found in coarse-grained, fast-flowing riffles (the riffle, main-channel, natural-bed habitat type in fig. 2), whereas fine-grained, organically rich pools offer the highest likelihood of exposure to sediment-borne co ntaminants (the pool, channel-margin, natural-bed habitat type in fig. 2). Larger, nonwadeable rivers, and sandy-bottomed, Coastal Plain streams probably do not have coarse-grained riffles but still contain fine-grained, organically rich pools. Consequently, an alternative richest habitat is identified and sampled in these systems. Such alternative habitats are chosen from the list presented in figure 2, in consultation with the regional biologist, liaison teams, and local biologists. (Details are provided in the section on Recommendations for Sampling Benthic Invertebrates.) Prior to collection of samples, each basic fixed site is visited to determine access, verify which of the 51 types of habitats are available for qualitative and semi-quantitative sampling, and establish the locations of the three sampling reaches. Most of these tasks should be accomplished during the on-site reconnaissance.
The appropriate season and hydrologic conditions for sampling are determined primarily by the life-history characteristics of the aquatic insects that dominate, at least numerically, in most riverine benthic invertebrate communities of North America. Various environmental factors influence insect life-history patterns and are considered in the selection of an appropriate sampling time (Hynes, 1970; Sweeney, 1984). Ideally, sampling occurs at a time of year when the majority of insects are at or near maturity and few species are in early instars or resting stages (for ex ample, eggs, pupae, or diapausing larvae). Early instars are problematic because they generally lack the morphological features necessary for identification to genus or species and may be difficult to collect because of their small size. The resting stages of most insects are either difficult to identify because there are no standardized taxonomic keys (for example, eggs) or difficult to collect because many pupae or diapausing larvae move into the hyporheos or streambanks where they are missed by stand ard collection procedures. Therefore, site-to-site differences in community development caused by differences in physical factors, such as temperature (Vannote and Sweeney, 1980; Ward and Stanford, 1982), dissolved oxygen (Nagell, 1981), and discharge (Patterson and Vannote, 1979; Wiggins and others, 1980), need to be considered in the selection of an appropriate sampling season and in the interpretation of biological data.
Water temperature is the primary physical factor directly influencing the rate of development of invertebrates and reproduction (Vannote and Sweeney, 1980). Development is commonly expressed as the cumulative number of degree days required to complete the aquatic portion of an insect's life cycle. Cumulative degree days are calculated as the sum of