Heterogenous distribution of kākahi (freshwater mussel Echyridella menziesii) environmental DNA in 5 New Zealand lakes of differing size and geomorphology

The abundance of freshwater mussels is declining globally. To better understand and possibly reverse this trend, we need more information on the occurrence and distribution of freshwater mussels. Using environmental DNA (eDNA) to detect species is a sensitive and cost-effective approach that can complement conventional monitoring methods. The quantity and distribution of eDNA in a waterbody determines how many samples researchers need to collect for a successful survey. There has been limited research on appropriate sampling strategies for detecting freshwater mussel eDNA in lakes. The aim of this study was to compare the occurrence and concentration eDNA from the New Zealand freshwater mussel or kākahi (Echyridella menziesii Gray, 1843) with density data obtained from divers and to use statistical approach to determine the optimal number of samples required when establishing a monitoring program. We took surface-water samples for eDNA detection from 5 lakes of different size and depth. At the same locations, divers measured kākahi density using traditional transects. For eDNA, we amplified the mitochondrial cytochrome oxidase 1 gene using droplet digital polymerase chain reaction. All 5 lakes had substantial populations of kākahi and high eDNA occupancy (79%). We detected kākahi eDNA at both shallow and deep sites. Detectability (presence/absence) of kākahi eDNA varied between 61 and 96% and was generally higher in shallow small lakes which is likely related to the lake’s geomorphology. Our models indicated that sampling 4 sites in triplicate is sufficient to detect kākahi eDNA if occupancy is high (>80%), but the number of sites sampled needs to be increased in lakes where there is no prior knowledge on kākahi populations. The results of this study provide guidance that can be used to inform development of sampling strategies for kākahi monitoring in lakes.