Comparison of sampling effort allocation strategies in a stratified random survey with multiple objectives

Stratified random survey is commonly used to estimate abundance indices of fish populations in multispecies survey, providing reliable data for stock assessment and fisheries management. In some cases, however, the sample size is relatively small because of the limitation of survey cost or other factors. The allocation methods of sampling efforts among strata in stratified random surveys with small sample size may need adjustment compared with traditional approaches. In this study, two sampling stations were allocated to each stratum first and then the remaining sampling units were allocated among strata using five traditional allocation methods. In order to distinguish them from traditional methods, we called them adjusted methods in this study. A simulation study was conducted to compare the performances of different allocation strategies of sampling efforts in a stratified random survey for estimating abundance indices of multiple target species. Relative estimation error (REE) and relative bias (RB) were used to measure the precision and accuracy of estimates of abundance indices under different allocation schemes of sampling efforts in the multispecies survey. The performances of different allocation schemes in estimating abundance indices varied greatly for different species over different seasons. The adjusted Neyman allocation scheme could significantly reduce the REE and RB of estimates of abundance index for single species survey. For multiple species surveys, the adjusted average-Neyman allocation method, the adjusted Yate allocation method, the adjusted proportional allocation method and current allocation method had relatively high accuracy and precision of estimates of abundance indices for four species in terms of the totalREE and totalRB. Though the adjusted average-Neyman allocation scheme did not always have the best performance, it was the optimal one considering the accuracy and precision of estimates of abundance indices for all species simultaneously. The allocation of sampling efforts among strata in stratified random surveys targeting for estimating abundance indices of multiple species should comprehensively consider the variance of abundance of different species in stratum and the seasonal changes. Keywords: Fishery-independent survey, Stratified random sampling, Computer simulation, Sampling efforts allocation