THE EFFECTS OF SAMPLING FREQUENCY ON ESTIMATES OF RECRUITMENT OF THE DOMINO DAMSELFISH DASCYLLUS-ALBISELLA GILL

Sampling frequency and methods can potentially affect estimates of demographic rates in population studies. To determine the effects of various sampling protocols on estimating recruitment rate, the rate of larval settlement by the coral reef damselfish Dascyllus albisella Gill to an experimental grid of coral heads was monitored over several intervals by either removing newly settled larvae (recruits) at each census, or not. By sampling every 5 h during two 24-h periods, it was found that most recruits (75% on Day 1 and 81% on Day 2) settled between dusk and dawn. Coral heads that already supported conspecific juvenile groups attracted more recruits than empty coral heads during both 24-h periods, although this was statistically significant on Day 2 only. Subsequent tagging of recruits from 1 to 19 days after settlement revealed that movement between coral heads was negligible after the 1st day postsettlement, but apparent mortality was high over the same period. The settlement rate of larvae to the coral grid during a 19-day period was estimated by employing three protocols of census frequency and recruit removal. This suggested that frequent sampling with recruit removal would most closely estimate true settlement rate for D. albisella and probably other species, while less frequent censuses without recruit removal may provide closer estimates of the size of the recruit cohort that will enter the juvenile population. Considerable variation exists in the sampling protocols used by researchers studying settlement rate in damselfishes and I caution against directly comparing results among these studies.