1Department of Aquatic Biosciences, Tokyo University of Fisheries, Konan, Tokyo 108-8477, 2Japan Sea-Farming Association, Chiyoda, Tokyo 101-0047, Japan
The effects of sample size and marking rate on Petersen estimates of population size were evaluated by a mark-recapture experiment. We released 1,200 striped jack juveniles into a 150t rearing tank, of which 5% were left-abdomen fin-clipped and 10% right-abdomen fin-clipped. We made 27 population estimates based on 18 samples and nine daily totals of the samples, which were taken from the tank twice a day during nine days. The error of the estimate was 1,200±600 fish when the mark rate for the population was 10% for any sampling fraction. When the mark rate was 5% a large sampling fraction, more than 10%, was required to keep the error within ±600 fish. The assumptions of the Petersen method were examined for the data. No mortality was observed during the experiment in the closed population. The binomial sampling was confirmed that all but one sampling met the assumption. The relationship between the sample size and the precision of the population estimate was also examined, given various tag rates. In order to obtain precise estimates, it is essential to ensure that the assumptions are satisfied. Larger sample sizes are required at smaller mark rates.