LAUSR

Abstract Macrozoobenthos survey programs often tend to be costly and time-consuming, which call for the development of cost-effective sampling designs. In this study, ordinary kriging interpolation was used to estimate the values at unsurveyed locations based on macrozoobenthos survey data in tidal flat as the ‘true’ values for this simulation study. Three potential survey designs, including simple random sampling (SRS), stratified random sampling (StRS) and cluster sampling (CS), were compared in estimating abundance indices of three macrozoobenthic species and species diversity index. A computer simulation study was developed to evaluate if sampling efforts could be reduced while maintaining data quality for quantifying the survey objectives. In general, StRS performed the best in estimating the target indices over different seasons, followed by CS and SRS. Sampling efforts in the three sampling designs could be reduced while relatively high precision and accuracy of estimates could still be achieved. This study suggests that cost and negative impacts of survey on the tidal flat ecosystem can be substantially reduced if proper studies can be done to optimize the survey design based on computer simulation study. Such a post-survey simulation study could improve survey designs and aid to acquire the optimal sampling efforts to achieve the most important survey goals. Although applied to one system, the framework developed in this study can be also applicable to other survey programs.