Niche construction mediates climate effects on recovery of tundra heathlands after extreme event

Statistical analyses All data were analysed using the statistical environment R [42]. Linear mixed effects models were used to account for our nested design of explanatory variables with a mixture of fixed and random effects [43]. The continentality index for each study site was used as a proxy for the effect of climate along the coast-inland gradient and was calculated according to Rivas- Martinez et al 2011, showing a range of 21.8 to 30.1 indicating a oceanic/maritime to continen- tal climate [44]. The response to extreme event treatment with and without litter was similar for all variables, and data were pooled together for all analyses. We tested for main effects and interactions of the extreme event treatment and continentality index on species richness, mean biomass of all vascular plant species, mean E. nigrum biomass, and mean biomass of subordi- nate species in addition to changes in SLA, LDMC and LNC. Data from the bioassay and seed sowing experiment was also tested against the extreme event treatment and continentality index. Only data from 2019 were used for the statistical modelling since we were interested in studying the long-term effects of the treatment on the recovery of the vegetation along the cli- matic gradient. However, year was used as a predictor variable in the seed sowing experiment. Due to field work complications, data was missing from one of the sites (i.e. Snarby) for 2019 and therefore data from 2018 was used for this site. There are no statistical differences in the

biomass between the last three years for any of the sites (F.value = 1.5, p = 0.93, Fig 2c) and we therefore chose to use data from 2018 for the missing site. No statistics were run on data from seed bank and seed rain due to the low germination rate. Results A total of 21 vascular plant species were registered along the climatic gradient during the study (S1 Table), with a maximum of 7 vascular species per plot. Control plots remained constant both in species richness and frequency during the 10 years (Fig 2). Moreover, all sites showed similar mean biomass of vascular plant species in control plots, as measured in 2019, with values between 275–350 g/m2. Vegetation recovery We encountered no new species during the study. All species were present in both extreme event and control plots, except for Calluna vulgaris which was not registered in extreme event Fig 2. a) Mean species richness (2009–2019), b) mean frequency of vegetation presence (2009–2016) and c) total biomass (g/m2) (2017–2019) of all vascular plant species in control plots (Co) and extreme event (EE) plots registered throughout all study sites. The mean is presented with 95% confidence intervals. https://doi.org/10.1371/journal.pone.0245929.g002 PLOS ONE Niche construction effects on tundra after extreme event PLOS ONE | https://doi.org/10.1371/journal.pone.0245929 February 4, 2021 7 / 14