PERFORMANCE OF THE MACROINVERTEBRATE COMMUNITY INDEX - EFFECTS OF SAMPLING METHOD, SAMPLE REPLICATION, WATER DEPTH, CURRENT VELOCITY, AND SUBSTRATUM ON INDEX VALUES

The influences of sampling method, water depth, current velocity, and substratum on two macroinvertebrate-based biotic indices were investig ated based upon 523 samples from 55 stony riffle sites on 23 New Zeala nd streams. A single hand-net sample estimated the site Macroinvertebr ate Community Index (MCI) within +/- 15% and four replicates yielded /- 10%. Between 8 and 10 replicate Surber samples produced +/- 10% pre cision. Quantitative MCI (QMCI) values were more variable, with 10 or 11 replicate Surber samples required for +/- 10% precision. Two proced ures for detection of statistically significant differences between pa ired MCI or QMCI values are described. The detectable difference metho d (equal sample sizes) is preferred for statistical reasons but a t-te st method can be used for unequal sample sizes. MCI and QMCI were rela tively independent of depth, velocity, and substratum within the sampl ed ranges of these variables. This is a major advantage for the assess ment of water pollution or enrichment using these indices. However, to avoid possible complications brought about by extreme values, samplin g within the following ranges of these variables is suggested: depth 0 .1-0.4 m, velocity 0.2-1.2 m s-1, and substrate 60-140 mm median diame ter.