Metabarcoding a Metacommunity: detecting change in a wetland wilderness

摘要: Abstract The complexity and natural variability of ecosystems present a challenge for reliable detection change due to anthropogenic influences. This issue is exacerbated by necessary trade-offs that reduce the quality resolution survey data assessments at large-scales. Peace-Athabasca Delta (PAD) large inland wetland complex in northern Alberta, Canada. Despite its geographic isolation, PAD threatened encroachment oil sands mining Athabasca watershed, hydroelectric dams Peace watershed. Methods capable reliably detecting changes ecosystem health are needed evaluate manage risks. Between 2011 2016, aquatic macroinvertebrates were sampled across gradient flood frequency, applying both microscope-based morphological identification, DNA metabarcoding. Using multi-species occupancy models, we demonstrate metabarcoding detected much broader range taxa more per sample compared traditional was essential identifying significant responses thermal regimes. We show family-level masks high variation among genera, first time, quantify bias barcoding primers on probability community. Interestingly, patterns community assembly near random, suggesting strong role stochasticity dynamics metacommunity. seriously compromises effective monitoring local scales, but also reflects resilience hydrological variability. Nevertheless, simulations showed greater efficiency metabarcoding, particularly finer taxonomic resolution, provided statistical power detect landscape scale.