DNA metabarcoding to quantify the ecological impact of forest decline on flying insect diversity in the Pyrenees
Résumé
Background: Forests suffer from an increase in frequency and severity of summer droughts and infestations of pathogens and insects.
Those factors cause high mortality of some keystone tree species (forest die-offs). Yet, how tree mortality and associated changes in forest
composition will affect local diversity and ecosystem functions remains unknown. Here, we aim at quantifying the impact of climate-
induced forest decline on biodiversity by measuring changes in the taxonomic structure of invertebrate communities along silver fir
(Abies alba) dieback and salvage logging gradients in the French Pyrenees. We examine patterns of variation in species diversity of flying
insect assemblages collected by Malaise traps deployed in 57 silver fir-dominated experimental plots (one Malaise trap per plot) in the
central and eastern Pyrenees. Sampling was carried out each month for over 4 months (May–August 2017). Samples were sequenced using
Illumina MiSeq and analyzed using the DAMe twin-tagging pipeline approach. Results: We obtained 224 bulk samples filled with a solu-
tion of monopropylene glycol plus ethanol. Despite high levels of DNA degradation detected in our samples, we found no major impact on
species detection, with more than 3500 operational taxonomic units (OTUs) in 18 different insect orders recovered. We found large species
temporal turnover (Jaccard Index: May–August = 0.35), as well as changes in community composition but no significant loss of species
diversity along the forest decline gradient. Significance: There is an urgent need to obtain detailed baseline data on species assemblages to quantify the impacts of climate change. Our study assessed biodiversity patterns on a scale and with a resolution that was previously
impossible and provides data essential for evaluating future biotic change. Our workflow coupling metabarcoding and Malaise trapping is simple to use and provides an affordable, reliable, and verifiable way of monitoring forest biodiversity at a large geographical scale.