Abstract
Environmental DNA (eDNA) metabarcoding has emerged as a promising approach to assess biodiversity and derive ecological status classes from water samples. However, a limitation of eDNA surveys is that detected DNA molecules may originate from other places or even dead organisms, distorting local biodiversity assessments. Environmental RNA (eRNA) metabarcoding has recently been proposed as a complementary tool for more localized assessments of the biological community. In this study, we evaluated the effectiveness of eDNA and eRNA metabarcoding for inferring the richness and species distribution patterns of vertebrates and invertebrates in a Central European lowland river. We collected water samples and analyzed them using a 12S marker for vertebrates and a COI marker for invertebrates. We detected 31 fish, 16 mammal, 10 bird and one lamprey species in the vertebrate dataset. While results were largely consistent, we detected a higher number of species when analysing eRNA (mean = 30.89) than eDNA (mean = 26.16). Also, eRNA detections had a stronger local signature than eDNA detections when compared against species distribution patterns from traditional fish monitoring data. For invertebrates, we detected 109 arthropod, 22 annelid, 12 rotiferan, eight molluscan and four cnidarian species. In contrast to the pattern of vertebrate richness, we detected a higher richness using eDNA (mean = 41.37) compared to eRNA (mean = 22.42). Our findings primarily show that eDNA and eRNA-based detections are comparable for vertebrate and invertebrate taxa. Biological replication was important for both template molecules studied. Signal detections for vertebrates were more localized for eRNA compared to eDNA. Overall, the advantages of the extra steps needed for eRNA analyses depend on the study question but both methods provide important data for biodiversity monitoring and research.