A multimetric benthic macroinvertebrate index for the assessment of stream biotic integrity in Korea

Abstract

At a time when anthropogenic activities are increasingly disturbing the overall ecological integrity of freshwater ecosystems, monitoring of biological communities is central to assessing the health and function of streams. This study aimed to use a large nation-wide database to develop a multimetric index (the Korean Benthic macroinvertebrate Index of Biological Integrity—KB-IBI) applicable to the biological assessment of Korean streams. Reference and impaired conditions were determined based on watershed, chemical and physical criteria. Eight of an initial 34 candidate metrics were selected using a stepwise procedure that evaluated metric variability, redundancy, sensitivity and responsiveness to environmental gradients. The selected metrics were number of taxa, percent Ephemeroptera-Plecoptera-Trichoptera (EPT) individuals, percent of a dominant taxon, percent taxa abundance without Chironomidae, Shannon’s diversity index, percent gatherer individuals, ratio of filterers and scrapers, and the Korean saprobic index. Our multimetric index successfully distinguished reference from impaired conditions. A scoring system was established for each core metric using its quartile range and response to anthropogenic disturbances. The multimetric index was classified by aggregating the individual metric ..scores and the value range was quadrisected to provide a narrative criterion (Poor, Fair, Good and Excellent) to describe the biological integrity of the streams in the study. A validation procedure showed that the index is an effective method for evaluating stream conditions, and thus is appropriate for use in future studies measuring the long-term status of streams, and the effectiveness of restoration methods.

Figure 1

Spatial distribution of study sites assigned to each of the three status categories (reference, impaired, others), according to site classification criteria.

Figure 2

Comparison of core metric values between the reference and impaired site groups. Boxes represent interquartile ranges (25th–75th percentiles). Closed circles at the bottom and top of each box indicate the 5th and 95th percentiles, respectively. The mean (horizontal dotted line), median (horizontal solid line), and standard deviation (error bar) are shown in the box plots.

Figure 3

Comparison of the KB-IBI scores among reference, impaired and other sites. See Figure 2 for the description of the box plots.

Figure 4

Ordering of reference and impaired sites by the first two principle components (PC1 and PC2) based on eight core metrics.