Format: ORAL

Fia Bengtsson1,2, Nils Cronberg1, Johan Ekroos1,3,4, Jose A. L. Villegas5, Abu B. Siddique5, Per Stenberg5

1 Biodiversity unit, Lund University 2 NINA , Norwegian Institute for Nature Research 3 Department of Agricultural Sciences, Plant Production Sciences, University of Helsinki, Helsinki, Finland 4 Helsinki Institute of Sustainability Science, HELSUS, University of Helsinki, 00014 Helsinki, Finland 5 Department of Ecology and Environmental Sciences, Umeå University

Bryophytes are physiologically highly dependent on climatic conditions, making them useful as early indicators of ongoing changes in biological communities. However, monitoring bryophytes is highly work-intensive, rarely carried out over extensive periods and even less frequent at temporal scales that allow detection of changes in phenology. eDNA approaches are surging in biodiversity monitoring, but bryophyte applications are still in their infancy. We explore a unique eDNA dataset (from the project Swe-BITS) based on airborne particles originally collected from Kiruna, northern Sweden, by FOI (Swedish Defence Research Agency) to detect radioactive downfall. Biological material, that had accumulated on glass fibre filters exchanged weekly during 35 years, was analysed for total DNA and matched against genetic data. As much as ten percent of the generated reads originated from wind-dispersed bryophyte spores. We selected 16 taxonomic groups based on information about local presence, available genomic data for discrimination and level of phylogenetic isolation. We found substantial changes in onset of spore dispersal in spring throughout the time series, with most bryophyte taxa advancing their spore dispersal a whole month, extending the entire season a month or more. The end-of-season timing remained largely the same. Phenological shifts may predate population changes, and both occurrence and phenology data from eDNA may inform conservation planning and efforts. At present, the taxonomic precision of the method is limited by the genomic resources available, but the possiblilities are steadily increasing. The proportional nature of the data, with varying relative abundances of different kinds of airborne biological matter between sampling events, makes the analyses computationally challenging. We conclude that eDNA from air samples is a promising and potentially cost-efficient way to monitor changes in bryophyte communities.