Variation in seagrass biomass estimates in low and high density settings: implications for the selection of sample size

Few seagrass biomass monitoring studies have considered the adequacy of monitoring intensity in their design. Power analysis is now widely used in ecological monitoring to determine sample size (replication) and the power (probability of not making a Type II error) of the monitoring design to detect change (effect size). We investigated seasonal variation of above-ground biomass of Zostera species at Woolooware Bay, Botany Bay, NSW and Ukerebagh Channel, Tweed River, NSW to show that seagrass biomass varies significantly between sites and seasonally. By conducting preliminary power analysis at each study site we found that our sampling design would only detect 70% change at Woolooware Bay, while <10% change would be detected at Ukerebagh Channel with the same intensity of sampling. We demonstrate the potential efficiency of harvesting as a means of estimating biomass in high biomass situations, where percentage cover may provide less discrimination between sampling sites.