The Xiamen Symposium on Marine Environmental Sciences (XMAS 2027)

High-resolution eDNA detection is important for characterizing harmful algal diversity, tracing biogeographic patterns and improving HAB risk assessment. Here, we present a long-read metabarcoding approach targeting ~1.4 kb fragment that spans the complete ITS1–5.8S–ITS2 region and the partial LSU region (D1–D2). By combining the high variability of ITS with the phylogenetic signal of LSU, this marker enables the resolution of harmful algal taxa at the species level and, where reference data permit, at the subspecies or lineage level. The fragment was amplified using the forward primer EukSSU, 5′-TAGGTGAACCTGCRGAAGGAT-3′, and the reverse primer D2C, 5′-CCTTGGTCCGTGTTTCAAGA-3′. Amplicons were primarily sequenced using PacBio HiFi to generate highly accurate circular consensus reads, and the same marker can also be adapted to Oxford Nanopore sequencing when suitable consensus-generation and error-correction workflows are applied. We evaluated this approach using ~1,000 seawater eDNA samples collected from broadly distributed sites along the Chinese coast. The primer set showed good amplification performance across multiple HAB-related lineages, including Amphidomataceae, Kareniaceae, Dinophysis and Alexandrium among dinoflagellates, Chattonellales among raphidophytes, and Pseudo-nitzschia and Nitzschia among diatoms. The resulting dataset resolved many closely related taxa and revealed fine-scale spatial patterns. To facilitate data exploration, we further developed an online platform (A temporary link: http://www.biolmic.dsmynas.com:3210/) that allows users to submit ITS, LSU or ITS-LSU sequences and retrieve their occurrence records and distribution patterns within the dataset. By integrating primer design, long-read sequencing, taxonomic assignment and web-based data retrieval, this approach provides a practical tool for HAB molecular monitoring, reference-library development, cryptic diversity detection, ecological lineage identification, habitat suitability modelling and future risk prediction under environmental change.