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

allowed us to account for microclimatic variations at site level, e.g. temperature, snow arrival and meltdown, and extreme winter warming events such as frost in the middle of winter due to snow cover absence (Table 1). Freezing days are the sum of days per study year where the mean daily temperature at ground level was below -1˚C, which indicates an absence of protec- tive snow cover and hence vegetation is likely exposed to freezing temperatures. Precipitation data was gathered from publicly available maps and online databases [31, 32]. Soil pH was measured in blocks 1,6 and 10 at all sites during late August 2016 with the use of pH-indicator strips (Merck 109531 Color Card MColorpHast pH-indicator strips, pH 0–6). Vegetation sampling Between 2009 and 2016, we recorded the frequency of vegetation presence of all vascular plant species in late August, by visually detecting the presence/absence of each species within 16 sub- plots (0.0125 m2 each) per plot with a wooden frame of 50cm x 50cm (i.e. the frequency of occurrence of each species) [33]. Between 2017 and 2019, the point intercept method was applied to estimate vascular plant biomass where the abundance of vascular plant species was measured as the total number of hits on 20 pins within the plot [34]. All vascular plant species present within the plot but not hit were also accounted for giving them an abundance value of 0.1. Prior to statistical analyses, the point intercept estimates were converted to biomass esti- mates [35]. Pin ends touching bare soil, moss and lichen were also registered and represent a

measure of their cover. Plant traits To assess the impacts of vegetation recovery on ecosystem properties induced by the extreme event, we gathered information on three traits known to affect components of the carbon and/ or nitrogen cycles at the leaf, plant and ecosystem level [36, 37]. The species registered in all plots were assigned values for specific leaf area (SLA [mm2 mg-1]), leaf dry matter content (LDMC [mg g-1]), and leaf nitrogen content (LNC [mg g-1]) according to the LEDA [38] and TRY [39] databases. Trait values selected for each species were based on location, all species were assigned values belonging to northern Norway or Sweden. The aggregated values of leaf traits were calculated according to Garnier et al (2004), where the relative contribution of each species to the total biomass of the community was multiplied by the species trait value [36]. Seed bank study In the centre of all 144 plots, a seed bank sample was collected in spring 2009 using a 10-cm long metal cylinder (2.5 cm in diameter) placed down at soil level below the litter. The soil cores were kept in plastic bags and frozen at -18˚C within 8-h after collection. After 3 months, samples were thawed in dark chambers at 2˚C before they were crumbled onto petri dishes (8.7 cm in diameter) with a moist filter paper. All samples were placed in a phytotron chamber that simulated the optimal conditions for most northern alpine plants, that is, 72% relative humidity and 24h of simulated daylight (Philips TLD 489 fluorescent tubes) with 12h at 18˚C

and 12h at 12˚C. All samples were sprayed with water every third day. Emerging seedlings were classified as dicotyledons or monocotyledons and removed from the petri dish. When no more seedlings emerged (about 12 weeks after the experiment was set up), all samples were given a new stratification period for 4 weeks in a dark chamber at 2˚C. After the stratification period, samples were placed in chambers with optimal growing conditions and seedling emer- gence was again registered for another 12 weeks. PLOS ONE Niche construction effects on tundra after extreme event PLOS ONE | https://doi.org/10.1371/journal.pone.0245929 February 4, 2021 5 / 14