Fig 1.
a) Study sites in the northern region, corresponding roughly to Örebro county. Each red dot represents a study site consisting of one thinned production stand, one unthinned production stand and one Woodland Key Habitat stand. b) Study sites in the southern region, corresponding roughly to Jönköping county. Each red dot represents a study site consisting of one thinned production stand and one Woodland Key Habitat stand. Blue dots represent unpaired nature reserve stands. Green is forested land, teal is wetland, yellow is open (agricultural) land, and grey is urban land. Coordinates are in WGS84, with decimal degrees as units. Background map provided by [47].
Table 1.
Descriptive statistics for environmental factors and stand characteristics per forest type and region. The number of stands of each forest type per region is given in parentheses. WKH = Woodland Key Habitat.
Fig 2.
Stems measured within 5.5 m radius around each trap, summed for each forest type. Black circles represent median values. Corylus avellana was measured at base, not breast height. Note the logarithmic scale on the x-axis. A slight jitter has been applied to increase readability of overlapping points. Sample sizes are not equal for the four forest types (reserves: 3, Woodland Key Habitats (WKHs): 20, thinned stands: 20, unthinned stands: 10).
Fig 3.
Dead wood objects measured in a 10 m radius around each trap. Black circles represent median values. Decay stage 1 is newly dead wood, 2 and 3 are intermediate decay, and 4 is heavily decayed. Diameter is the average of the diameter at the two ends. Note the logarithmic scale on the x-axis. A slight jitter has been applied to increase readability of overlapping points and draw order has been randomized. Sample sizes are not equal for the four forest types (reserves: 3, Woodland Key Habitats (WKHs): 20, thinned stands: 20, unthinned stands: 10).
Table 2.
Species (spec.) and individuals (ind.) per species group and forest type. The number of stands per forest type is given in parentheses. WKH = Woodland Key Habitat.
Table 3.
Saproxylic alpha diversity and environmental factors.
GLMM results for all saproxylic species. For the categorical predictors “Forest type” and “Region”, Woodland Key Habitat (WKH) and Jönköping are the respective reference levels. The continuous predictors (Dead wood volume, Dead wood diversity, Canopy openness) have been unstandardized and all values have been back-transformed, giving odds ratios for the fixed effect estimates. The random factor “Stand” is nested within “Site”. Confidence intervals (CI) not overlapping 1 (indicating statistical significance) have been marked in bold.
Table 4.
Red-listed alpha diversity and environmental factors.
GLMM results for red-listed species. For the categorical predictors “Forest type” and “Region”, Woodland Key Habitat (WKH) and Jönköping are the respective reference levels. The continuous predictors (Dead wood volume, Dead wood diversity, Canopy openness) have been unstandardized and all values have been back-transformed, giving odds ratios for the fixed effect estimates. The random factor “Stand” is nested within “Site”. Confidence intervals (CI) not overlapping 1 (indicating statistical significance) have been marked in bold.
Fig 4.
NMDS of all traps, based on modified Gower dissimilarity of saproxylic species abundance data. Pairs of traps from the same stand have been connected with a line, and a hull has been drawn around all traps from the same forest type. K = 2, stress = 0.24. WKH = Woodland Key Habitat.
Table 5.
Multivariate dispersion (mean distances from centroid) for traps of each forest type with associated standard error (SE), from PERMDISP analysis based on modified Gower dissimilarity of saproxylic species abundance data. P-values below 0.05 in bold. WKH = Woodland Key Habitat.
Fig 5.
a-e) Species accumulation curves with 95% confidence intervals for each forest type and species group from the Jönköping sample. Samples are forest stands (data from two traps per stand pooled), extrapolated to twice the original sample size.
Fig 6.
a-e) Species accumulation curves with 95% confidence intervals for each forest type and species group from the Örebro sample. Samples are forest stands (data from two traps per stand pooled), extrapolated to twice the original sample size. In b), the inlay shows the curves for the Woodland Key Habitats and the thinned stands, with the unthinned stands removed for clarity.