Zooplankton plays an essential role in marine pelagic ecosystems. They are generally classified into species based on their morphology. The morphological classification requires expertise, resulting in issues with the identification of cryptic species and immature stages of zooplankton, which are difficult to identify morphologically. In this study, we propose a metabarcoding method using high-throughput sequencing to show the diversity and community structure of zooplankton, independent of morphological classification. First, we select an appropriate molecular marker for metabarcoding analysis of copepods, which are highly diverse and abundant. A bioinformatics protocol was also developed to understand copepod communities using large amounts of sequence data. The proposed method was then applied to zooplankton communities in the Pacific area to show large-scale patterns of copepod community structure and species diversity. The metabarcoding approach was applied to a dietary analysis of fish larvae and biomonitoring of zooplankton, contributing to the understanding of food web structures and changes in marine ecosystems.