Aquatic macroinvertebrates are widely used indicators of freshwater ecosystem health, but sample processing is labour-intensive, with morphology-based identifications requiring specialist taxonomic skills. This study evaluated a DNA-based workflow for monitoring aquatic macroinvertebrate communities by comparing identifications from traditional morphology to those from metabarcoding. The study aimed to improve the accuracy, efficiency and scalability of sample processing, providing an example of how genomic datasets can be incorporated into routine long-term monitoring frameworks.
Over a three-year period, macroinvertebrates were sampled from eight sites across four streams within Kakadu National Park, (Northern Territory), with five replicate samples collected per site. Samples were processed using morphology-based species-level identification and eDNA metabarcoding. The latter applied to two matrices, one containing organisms and their associated detritus, the other, hand-sorted macroinvertebrate samples. Amplicon sequencing targeted a COI and an16S rRNA marker, with bioinformatics performed using QIIME 2 and combination of BOLD and a custom reference database.
Preliminary results indicated morphology, matrix metabarcoding and sorted metabarcoding samples recovered similar patterns of macroinvertebrate assemblages among sites. Significant RELATE correlations (ρ = 0.404–0.573, p ≤ 0.001) and strong Procrustes concordance indicated that both metabarcoded approaches captured the principal ecological gradients observed in morphology-derived data.
Overall, the concordance in site-level assemblage patterns indicated that metabarcoding could replace morphological assessments within the Office of the Supervising Scientist’s long-term macroinvertebrate monitoring program in Kakadu streams. The similarity between genomic-derived data from matrix and hand-sorted macroinvertebrate samples provides promise in diminishing the workload associated with sorting. In concert, this project provides guidance on the selection and validation of markers, sample-processing methods, and regional reference databases required to establish a consistent and defensible protocol for ecological assessment.