Development of Reference Station Scoring Criteria for Biomonitoring with Benthic Macroinvertebrates in the Piedmont Region of Georgia

As biocriteria become increasingly accepted by states for evaluating stream designated uses, the use of biomonitoring data for determining stream impairments should be carefully designed, applied, and interpreted based on a solid understanding of the local and regional stream conditions as well as the characteristics of the biological communities and how the communities respond to anthropogenic disturbance. A key to interpreting biomonitoring data involves the development of reference stations as a benchmark to determine degrees of impairment and to evaluate the overall biotic integrity of a stream. The two most important steps for analyzing biological data is the selection of appropriate metrics and selection of appropriate reference streams.Appropriate metrics for biomonitoring are well documented in the literature and have been studied for years (Plafkin et al., 1989; Kerans and Karr, 1994; Barbour et al. 1997), but very little research has been dedicated to the definition of appropriate reference stations and the methods for the selection and use of reference station data even though this is a critical step to the analysis of the results. This paper examines the methods for selecting an appropriate reference station, presents the development of a reference station database, and presents a standard approach for scoring benthic macroinvertebrates in the piedmont of Georgia.A total of 15 reference stations and 53 study station were examined from extensive biomonitoring conducted over the past five years in metro Atlanta. Select reference stations were sampled on multiple occasions. Study stations represent a variety of different landuses and degrees of urbanization, from some of Atlanta's most urbanized watersheds to suburban watersheds with less development. Study stations represent a gradient of development as determined from examination of watershed landuse and existing point sources. Reference station were chosen in various watersheds in the piedmont region of Georgia and represent least disturbed rural watersheds. Both reference and study stations have been sampled to represent different watershed sizes, seasons, drainage basins.Benthic macroinvertebrates were sampled at each monitoring station following qualitative techniques of the draft standard operating procedures (SOP) for the Georgia Bioassessment Protocols (GBP) (GA DNR, 2000). The GBP is a multi-habitat approach that was modified slightly to maximize efficiency of fieldwork and analysis while providing the data necessary to complete the GBP assessment. This modification is consistent with EPA's RBPs (Plafkin et al., 1989) and involved making a composite of samples collected from the various habitats for analysis and data evaluation. The composite sample, referred to as a “multi-habitat sample,” was derived from the habitat types present at a station.To maintain uniform collection, the equipment used, collection methods, site length (or area), and unit effort were comparable among stations. The major habitat types (undercut banks, rocks, vegetation, sand, riffles, runs, and pools) at each site, as well as the proportion of each habitat type sampled, were recorded in the field log book or on log sheets.The benthic macroinvertebrate samples were identified to the lowest taxonomic level practical, and the results were used to compute six community, population, and functional metrics following the GBP (GA DNR, 2000). Each metric or index represents a slightly different component of community structure and/or function and provides a measure of biotic integrity.The result of the study indicate substantial variability among reference stations that were sampled. Several sources of variability were examined including seasonal, watershed size, and drainage basin, but none could be used to explain the observed variation. Because of the variability in the reference station data, biotic integrity scores for the study stations were highly variable and did not correlate well with observations by field scientists, fish biotic integrity scores, habitat scores, or landuse patterns.This study presents the use of a reference station database for scoring benthic macroinvertebrate data. The use of multiple reference stations provided a more robust measure of reference condition and produced more consistent scores that were better correlated with observations in the field and other measures of stream conditions (i.e. fish biotic integrity, landuse, and habitat scores).The use of reference stations is an extremely important step for both benthic macroinvertebrate and fish community biomonitoring. In many areas in the United States, finding good reference stations is very difficult since most watersheds have varying levels of anthropogenic disturbance. As states begin to implement biomonitoring programs, a robust measure of reference station condition is needed to accurately represent stream water quality conditions and to consistently interpret results among different watersheds.