The SOP requirements for subsampling and taxonomic resolution were strictly adhered to, 

 deviating only when the quality of the specimen was lacking due to missing body parts needed for 

 identification. When organisms were too immature to confidently take to the taxonomic level 

 outlined in the SOP, they were more conservatively identified. 



Following is a description of the subsampling procedure: Each sample was rinsed in a 0.5 

 mm sieve to remove preservative. The washed sample was then transferred to an appropriate size 

 invertebrate sorting tray marked into square quadrants. Water was added to the tray to allow 

 complete dispersion of the sample and even distribution of the organisms. Quadrants were 

 randomly selected and organisms removed from each quadrant until the total number of organisms 

 fell within the range of 270 to 330 (±10% of 300 organisms), or until there were no more 

 invertebrates to remove, whichever occurred first. Any organism lying over a line separated by 

 two quadrants was considered to be in the quadrant containing its head. 



Data Analysis 



Community structure, fianction and sensitivity to impact were characterized for each 

 subsample using a battery of metrics developed by Montana DEQ for streams in the Plains 

 Ecoregions of the state (Bukantis 1997). Two approaches were employed in the analysis of data 

 for this report. The first approach relied on an ecoregional reference and scoring criteria; metric 

 values were compared to the established Plains Ecoregions reference values (Table 1). Values and 

 scoring criteria were derived from data from the Plains ecoregions and revised by McGuire in his 

 review of 1995. All metrics used by McGuire were used in this analysis. The ecoregional 

 reference approach allows comparison of these sites to plains sites elsewhere in the state. 



In the second analysis, an internal reference (Table 2) was established for these streams; a 

 reference value for each metric was established for all sites based on the performance of that 

 metric at all sites studied. The best value, if appropriate for the analysis, was chosen as the point 

 of comparison for each metric used. Tentative scoring criteria for the internal reference were 

 devised from an analysis of the ranges of metric values over a data set gleaned from other sources. 

 Data from 1992, 1993 and 1994 surveys of Plains Ecoregions reference streams (McGuire 1994a, 

 1994b, 1995) provided fifteen cases, while data from a 1995 survey conducted by Montana DEQ, 

 as yet unpublished, provided nine more cases. The total of twenty-four cases is not a large 

 database from which to establish scoring criteria for streams throughout the Plains Ecoregions; 

 however, a wide range of biotic health was manifest in the twenty-four sites, and it was 

 considered a usefijl starting point for the establishment of tentative scoring criteria for internal 

 references for bioassessments of plains streams. Enlargement of the Plains Ecoregions database, a 

 process already underway, will add reliability to this effort. 



For both analyses, actual metric values were compared to the reference values to obtain 

 metric scores (Table 4 for the ecoregional reference approach and Table 5 for the internal 

 reference approach). Total metric scores were obtained by summing scores for all metrics, and an 

 impairment classification and a use support category for each site was derived fi"om this total 

 score. 



