Evaluation of net and bottle sampling methods for zooplankton collection in reservoirs: an assessment using data from a nationwide reservoir monitoring program in Japan

Abstract Zooplankton play a key role in aquatic ecosystems and are commonly monitored in limnological studies. A variety of sampling techniques have been developed for this purpose, yet differences among methods can introduce biases in zooplankton community data. In Japan, the Ministry of Land, Infrastructure, Transport and Tourism (MLIT) conducts a long-term monitoring program known as the National Census on River Environments: Dam Reservoir Version (MLIT-NCRE:D), which includes routine surveys of aquatic organisms, such as zooplankton, in reservoirs across Japan. This program employs two main sampling techniques: net sampling and bottle sampling. However, sampling intensity and environmental conditions vary among survey sites. These methodological approaches and differences in sampling effort can introduce biases in zooplankton community data. In this study, we evaluated the effects of sampling method and sampling effort on zooplankton community assessments using data from the MLIT-NCRE:D program. Our analysis showed that while net sampling effectively captured a wide range of taxa, bottle sampling yielded more accurate estimates of abundance. Moreover, discrepancies between the two methods became more pronounced with increased sampling effort. These findings suggest that relying on a single sampling method may limit the accuracy of zooplankton community assessments in terms of both taxonomic composition and abundance. Given that the MLIT-NCRE:D program employs both sampling methods, we conclude that it provides comprehensive and ecologically meaningful monitoring data.