Combining observational and experimental data to estimate environmental and species drivers of fungal metacommunity dynamics

This dataset contains data from two studies that surveyed the presence or absence of wood-decay fruit bodies on a total of 1379 dead Norway spruce (Picea abies) logs. The logs were surveyed twice, 6 years apart. The study sites were composed of old-growth unmanaged forests where we studied naturally fallen logs. In Brattiken nature reserve (Sweden, 65°25"N, 16°06´E), 843 logs were inventoried in autumn of 1997 and 2003 (Edman & Jonsson, 2001; Berglund et al., 2005; Jönsson et al., 2008). In Rörstrand nature reserve (Finland, 60°27"N, 25°11´E), 536 logs were inventoried in autumn of 2002 and 2008 (Ovaskainen et al., 2010; Ottosson et al., 2014). Among a total of 210 fungal species, we chose 12 species that were present at both study sites in sufficient numbers and were expected to interact based on the literature. We used additional repeat visit data to account for imperfect detection (as Moor et al., 2021). Dead fruit bodies were excluded from the analysis. We included two log-level variables that are well-known to explain species occurrences: log decay stage and log diameter (cm). All environmental variables were transformed to mean zero and unit standard deviation (subtracting the mean and dividing each value by the standard deviation). We also included the decay stage variable squared, as some species are most frequent (in terms of occupancy, i.e.  number of presences divided by number of logs) in logs of intermediate decay.