Estimating the abundance of banded kokopu (galaxias fasciatus gray) in small streams by nocturnal counts under spotlight illumination

The abundance of banded kokopu (Galaxias fasciatus Gray) in small streams has usually been determined by the labour intensive and invasive method of electric fishing. Recently, nocturnal counts under spotlight illumination have been used to determine presence or absence and relative abundance of banded kokopu, but the proportion of the population seen was unknown. We compared 20 spotlight counts of banded kokopu in approximately 20 m reaches in streams in the North Island, New Zealand, to population estimates determined by removal electric fishing in the same reaches. Spotlight counts were related to population estimates over a range of densities, and on average, spotlight counts were 64% of the population estimates. Though we tried to separate age-0 fish from older fish visually in the spotlight counts, the size frequency distribution of the fish caught by electric fishing showed that the visual separation was not reliable. In addition, visual counts were generally inefficient for age-0 fish (40-70 mm total length), as only about 40% were observed. Banded kokopu were also recorded in streams using time-lapse video recordings with a camera sensitive to low light levels. Diel activity showed two major peaks, one in the early morning from 0400 h to 0900 h, and the other in the afternoon and evening from 1300 h to 1900 h. Fish were less disturbed by the observer.s approach after dark than during the day, so we suggest that from dusk to about 2200 h is the best time for visual counts of banded kokopu by spotlight in summer months