Evaluating the efficacy of sample collection approaches and DNA metabarcoding for identifying the diversity of plants utilized by nectivorous bats

In this study, we evaluated the efficacy of sample collection approaches and DNA metabarcoding to identify plants utilized by nectivorous bats. Samples included guano collected from beneath bat roosts and pollen-swabs from bat fur, both of which were subjected to DNA metabarcoding and visual identification of pollen (microscopy) to measure plant diversity. Our objectives were to determine whether DNA metabarcoding could detect likely food plants of nectivorous bats, whether sample types would produce different estimates of plant diversity, and to compare results of DNA metabarcoding to visual identification. Visual identification found that 99% of pollen was from Agave, which is thought to be the bats' main food source. The dominant taxon found by metabarcoding was also Agavoideae, but a broader diversity of plant species was also detected, many of which are likely "by-catch" from the broader environment. Metabarcoding outcomes differed between sample types, likely because pollen-swabs measured the plant species visited by bats and guano samples measured all items consumed in the bat's diet, even those that were not pollen or nectar. Overall, metabarcoding is a powerful, high-throughput tool to understand bat ecology and species interactions, but careful analysis of results is necessary to derive accurate ecological conclusions.