Species traits and ecosystem characteristics affect species detection by eDNA metabarcoding in lake fish communities

Abstract Although environmental DNA (eDNA) metabarcoding is an exceptionally useful and powerful tool for monitoring biodiversity, little is known about whether the traits of organisms and their ecosystem characteristics affect eDNA metabarcoding performance. Nationwide surveys can provide more detailed insights, yet such studies have rarely been conducted. In order to evaluate eDNA metabarcoding performance in broad‐scale monitoring, we examined the effects of species ecological/biological traits and ecosystem characteristics on species detection rates and the implications for community analysis. In addition, we tested the effects of sample mixing and transportation methods, including cooling and freezing, on eDNA metabarcoding. On a nationwide scale, we conducted eDNA metabarcoding for fish communities in 18 Japanese lakes. By comparing species records, we observed that certain traits, including body size, body shape, saltwater tolerance and habitat preference, influenced eDNA detection. In addition, the proportion of species detected decreased significantly with an increase in lake surface area owing to ecosystem size effect on species detection. We conclude that species traits, including habitat preference, body size and ecosystem size, should be considered when assessing the eDNA metabarcoding performance in broad‐scale monitoring.