Diel variability in counts of reef fishes and its implications for monitoring

Studies of reef fish assemblages in space rarely consider the effects of temporal variability on spatial comparisons, and when they do, usually examine timescales of months to years. The nature of fish monitoring surveys is such that particular locations may be surveyed at one time of day, and surveys designed to establish the degree of spatial variability in assemblages may be confounded if the order of sampling within treatments is not randomised with respect to time of day. In this study, we tested the degree of temporal variability in temperate reef fish counts at the same sites in New Zealand and Italy, within and between days. Repeated counts separated by months returned quite different assemblages, whereas counts separated by days did not. Although overall assemblage structure did not generally differ significantly with time of day, at some locations there was evidence of changes in the relative density of certain species in the afternoon relative to the morning. Care must be taken not to introduce systematic bias to spatial comparisons due to behavioural variability in fishes at differing times of day.