The relation of fungal communities to wood microclimate in a mountain spruce forest

Fungal Ecology 21 (2016) 1e9

Contents lists available at ScienceDirect

Fungal Ecology journal homepage: www.elsevier.com/locate/funeco

The relation of fungal communities to wood microclimate in a mountain spruce forest
clav Pouska a, *, Petr Macek b, Lucie Zíbarova
c Va
129, 16521 Praha 6 e Suchdol, Czech Republic Faculty of Forestry and Wood Sciences, Czech University of Life Sciences Prague, Kamýcka 

31, 37005 Cesk Faculty of Science, University of South Bohemia, Branisovska e Bud ejovice, Czech Republic c Resslova 26, 40001 Ústí nad Labem, Czech Republic a

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 7 June 2015 Received in revised form 22 December 2015 Accepted 20 January 2016 Available online xxx

Microclimatic conditions influence fungal growth, yet accurate descriptions of the relationships between the occurrence of fungi and microclimate (especially temperature) are lacking for dead wood in natural conditions. Here, we studied the occurrence of fungal fruit bodies on 2 m long segments of both standing and lying trunks of Norway spruce (Picea abies). The fungal assemblages were associated with properties of the segments related to the progression in wood decay, causes of tree death, and temperature and moisture conditions. Fluctuations in the temperature of wood decreased with increasing water content, and both water content and temperature stability increased with diameter and with the progression in wood decay. Red-listed species differed in their relations to both wood and microclimate parameters, which highlights the importance of the simultaneous presence of various wood types for the occurrence of rare and threatened species. © 2016 Elsevier Ltd and The British Mycological Society. All rights reserved.

Corresponding editor: Jacob HeilmannClausen Keywords: Ascomycetes Basidiomycetes Coarse woody debris Macrofungi Picea abies Temperature fluctuations Water content

1. Introduction The process of wood decay is reflected in the composition of fungal communities in individual lying trunks (Renvall, 1995; Høiland and Bendiksen, 1997; Heilmann-Clausen, 2001; Pouska et al., 2011; Rajala et al., 2011, 2012; Heilmann-Clausen et al., 2014). Species composition is also related to the decay rate of wood and is influenced by external microclimatic conditions (HeilmannClausen, 2001), as well as being interconnected with the way the tree died (Renvall, 1995; Pouska et al., 2011; Ottosson et al., 2015). In addition to the stage of decay, several other related characteristics have been demonstrated to be important for species composition on lying trunks (i.e. logs), such as the cover by bark
et al., 2012), contact with the ground (Renvall, 1995; Kubartova (Lindblad, 1998; Rajala et al., 2012) and moisture (Fukasawa et al., 2009; Rajala et al., 2011, 2012). The occurrence of fungal species is influenced by the properties of the wood they occupy, but fungal

* Corresponding author. E-mail address: [email protected] (V. Pouska). http://dx.doi.org/10.1016/j.funeco.2016.01.006 1754-5048/© 2016 Elsevier Ltd and The British Mycological Society. All rights reserved.

activity is in turn the main driver of decay, changing the chemical
 composition and structure of wood (e.g. Rypa cek, 1957; Stokland et al., 2012) and, under aerobic conditions, producing water, carbon dioxide and heat (Ryp
a cek, 1957; Boddy, 1983a,b; Bjurman and € , 2000). Wadso In Norway spruce, the size of logs is an important determinant of species composition (e.g. Lindblad, 1998; Rajala et al., 2011, 2012), and large logs are very important for the occurrence of certain species (Bader et al., 1995; Høiland and Bendiksen, 1997; Stokland and Larsson, 2011). Junninen and Komonen (2011) concluded that the threshold diameter critical for polypore species richness is 20e30 cm, above which the sporocarps of species demanding large-diameter trunks start to appear. Although some fungal species prefer the fine wood of spruce, most also form sporocarps on
r, 2005). It is unlikely that fine woody debris coarse wood (Allme (diameter 5e9 cm) can substitute coarse woody debris in managed spruce forests where red-listed species are concerned (Kruys and Jonsson, 1999). There are several explanations for the importance of trunk size on fungal community assembly. Larger trunks persist longer in

2

V. Pouska et al. / Fungal Ecology 21 (2016) 1e9

particular stages of decay (Harmon et al., 1986), which is important for many rare species that seem to depend on the slow decay process (Renvall, 1995), and for species that disperse mainly through spores, because these species have a greater chance of landing on a trunk in a suitable stage of decay due to a longer ‘window of opportunity’ (Stokland et al., 2012). Furthermore, the greater surface area of large trunks can collect more spores, which may be particularly important for rare species with low population €nsson et al., densities and spore deposition (Bader et al., 1995; Jo 2008). On the other hand, small trunks may not provide a sufficient amount of substrate for the sporocarp production of some species (Stokland and Kauserud, 2004). It is also hypothesised that increasing diameter influences species composition through the stabilisation of microclimatic conditions (Griffith and Boddy, 1991; Renvall, 1995). Microclimatic conditions include mainly variations in temperature and the content of water, oxygen and carbon dioxide in wood, which have all been reported to have an influence
cek, 1957; Boddy, 1983b,c; Harmon on the growth of fungi (Rypa € , 2000; Hendry et al., 2002), et al., 1986; Bjurman and Wadso resulting in various outcomes of interspecific interactions (Boddy, 2000; Carlsson et al., 2014). For example, competitive fungi are rather sensitive to environmental stress, and both drought and heat stress may decrease their competitiveness and hence cause shifts in the dynamics of competitive interactions (Crowther et al., 2014). Fluctuations in environmental conditions can facilitate species coexistence (e.g. Chesson and Huntly, 1997); for instance, fluctuating temperature can maintain the diversity of wood decay fungi in artificial communities (Toljander et al., 2006). On the other hand, excessive microclimatic fluctuations result in reduced fungal
 growth (e.g. Rypa cek, 1957; Jensen, 1969; Viitanen, 1997; Bjurman € , 2000). Therefore, it is important to investigate the and Wadso role of fluctuation intensity in wood of various sizes on the growth of fungi under natural conditions. Although morphological characteristics of trunks and characteristics influencing conditions in wood, such as solar radiation and canopy cover, have been inves€ssler et al., tigated in relation to fungal species composition (e.g. Ba 2010), studies on internal conditions in trunks and their relation to the occurrence of fungi are largely lacking. The scarcity of suitable wood in conventionally managed forests and the isolation of old-growth stands are the main causes for the rarity of many species of wood-inhabiting fungi (e.g. Bader et al., 1995; Stokland and Kauserud, 2004; Abrego and Salcedo, 2013;
n et al., 2013). Our study site in the Bohemian Forest is one Norde of the areas where a variety of dead wood is abundant, supporting a €ssler et al., 2012). high diversity of fungi (Ba In this study, we repeatedly investigated the occurrence of fungal sporocarps on 2 m long segments on both standing and lying trunks of Norway spruce (Picea abies), and related this to biotic and abiotic characteristics of the segments. Besides characteristics known to have an influence on fungal assemblages, we investigated microclimatic conditions within trunks as potential characteristics further explaining fungal occurrence. We hypothesised that the thickness of trunks influences the fungal species composition, namely through the variance of temperature and moisture conditions. We expected that large trunks would have more stable conditions than small ones, and such stability would likely support the occurrence of rare species, hence shaping the fungal species assembly. 2. Materials and methods 2.1. Study area The study was conducted in a mountain spruce forest in the  Bohemian Forest (Sumava) of the Czech Republic, on the northern


(Bayerischer Plo €ckenstein; slope of the mountain Trojmezna 48
460 1900 N, 13
490 3700 E). The bedrock is coarse-grained granite. Average annual precipitation is 1100 mm; average annual air temperature is 3
C; average annual number of frost days is greater than 200; average dates of the first and last snow cover are 20th October and 10th May, respectively (Tolasz et al., 2007). However, July 2013 and the period from December 2013 to April 2014 were about 3
C warmer than the long-term normal from 1961 to 1990
n  ov, 1117 m a.s.l.). On the basis of our (meteorological station Chura measurements, average air temperature 15 cm above the soil surface from July to October 2013 was 10.3
C (range 4.1
C to 35.1
C; Supplementary data 1). Precipitation from 13th June to 21st October 2013 was 400 mm which corresponds to the long-term normal (Tolasz et al., 2007). The forest is disturbed old-growth dominated by P. abies trees that are the source of various types of dead wood in all decay stages. The majority of trees at the site were recruited between the years 1750 and 1870 (Svoboda et al., 2012). However, a bark beetle outbreak was exacerbated by a windstorm in 2007 and by 2008 most of the canopy trees were dead. Currently, living trees are present mainly as a variably tall and dense regeneration of P. abies and Sorbus aucuparia. In addition, the field-layer vegetation shades the studied trunks to a various degree, and is dominated by Athyrium distentifolium, Luzula sylvatica, Calamagrostis villosa, Avenella flexuosa and Vaccinium myrtillus. 2.2. Trunks and their characteristics Dead trunks of spruce (30 snags and 68 fallen logs) were selected so that a wide range of diameters and all available decay stages were represented. Two-metre long trunk segments were then established to record the occurrence of fungi together with their characteristics including microclimatic measurements. Characteristics of the segments are listed in Tables 1 and 2. Segments on standing snags had their lower edges just above the roots, and segments on lying logs were either in the middle of their length (in most cases), or sometimes in their thicker sections in order to include the largest diameters in logs (no segments included breaks). The causes of tree death (bark beetles, butt rot, competition, wind and unascertained) were also recorded for each trunk, and their probabilities were combined using fuzzy coding (for details see Pouska et al., 2011). Temperature characteristics (Table 2) were recorded, at 15 min intervals from 14th June to 21st October 2013, in 38 segments (i.e. trunks) in the sub-surface layer of wood (5 cm deep) using Pt100 resistance thermometers connected to data loggers (MINILOG-T6, 
Bude jovice, Czech Republic). Thermometers Fiedler AMS, Cesk e were mostly placed on the northern side of segments to avoid direct exposure to the sun. Holes for thermometers were drilled using a borer with a diameter of 10 mm; the remaining space around cables was filled with rubber and hole openings were sealed with silicone adhesive. 2.3. Sampling of fungi Inventories of fungal sporocarps on segments of trunks except branches were carried out from July to October 2013 (the majority of segments were visited at least four times although one snag and seven logs were visited three times) and in June 2014 (all segments visited). Dead sporocarps were omitted. Records (presence data) from all visits were pooled. We attempted to identify all sporocarps to species but we did not include Athelia spp. (due to unclear trophic modes), Galerina spp. (due to uncertain identification), mycorrhizal and mycoparasitic fungi, and some tiny ascomycetes (e.g. Actidium hysterioides) because they could be easily overlooked.

V. Pouska et al. / Fungal Ecology 21 (2016) 1e9

3

Table 1 Characteristics of 2 m long segments of all spruce trunks. Characteristic

Range and scale

Description

Elevation

1220 e1340 m Categorical 10e103 cm 1e3 (ordinal)

Approximate elevation of places with trunks

Trunk type Diameter Decay stage

Standing snag or lying log Diameter in the middle of segments Based on the stage of decay [1e5, details in Pouska et al. (2011)]; stage 1 was not present, stage 2 corresponds to new decay stage 1, and stages 4 and 5 are merged into stage 3. Thus: stage 1 e wood mostly hard (spike penetrates 1e2 cm into the wood), most of the bark still attached (though not necessarily for bark beetle-infested trunks); 2 e wood partly decayed on the surface or in the centre (spike penetrates 3e5 cm into the wood), large pieces of bark usually loosened or detached; 3 e most of the wood soft throughout. However, the central parts can remain hard, while the surface layers of the wood can be missing. Trunk is often covered by field-layer vegetation. There were no snags in decay stage 3 Surface 0e5 0 e wood surface under the bark or intact; 1 e small cracks, disintegration up to 6%; 2 e surface starting to fall off in a few places, disintegration (ordinal) disintegration around 10% of the segment surface (7e15%); 3 e 16e40% of the surface has fallen off; 4 e 41e75% of the surface gone; 5 e the whole surface layer missing for a long time Bark 0e100% Cover by bark; percentage cover of the segment surface Branches 5e1, 0 5 e most of the thinnest twigs still present; 4 e thinnest twigs mostly gone (at least half of branches still have branching of at least the 3rd order); 3 e most branches still with at least one fork (1st order); 2 e at least half of branch stubs present and still about 25 cm long; 1 e less than half of branch stubs present; 0 e the segment belongs to the lower part of the stem and branches were not present at the time of tree death Ground contact 0e4 0 e segments of snags or segments of logs that do not touch the ground; 1 e the segment touches the ground with less than ¼ of its length; (ordinal) 2 e ¼e½ of the segment on the ground; 3 e up to ¾ of the segment on the ground; 4 e the segment lies on the ground for more than ¾ of its length Lichens 0e50% Cover by lichens; percentage cover of the bare wood relative to the proportion of the bare wood on the segment surface Bryophytes 0e80% Cover by bryophytes; percentage cover of the segment surface Saplings 0e3 Saplings of spruce on a segment; 0 e none or only small seedlings, up to 5 cm high; 1 e saplings higher than 5 cm; 2 e at least 1 sapling (ordinal) between 26 cm and 1 m high; 3 e at least 1 sapling higher than 1 m Vegetation 0e85% Cover by vegetation; percentage cover of the segment by plants growing both around and on the trunk Contacting wood 0e2 (count) The number of other decaying trunks or stumps touching the segment Openness 37.9e80.3% The fraction of open sky unobstructed by vegetation or any other objects; based on hemispheric photographs taken 1.3 m above ground (photographs for the segments of snags were taken at the distance of 1 m on the southern side of snags) Water contenta 16e95.3% Volumetric water content in the surface of wood measured on the basis of electric resistance upon most visits (mean); tester WHT-860
Mezirí (Elbez, Velke cí, Czech Republic). Measurements were done at two random places in a segment each time, on a side that was not exposed to solar radiation a

Microclimatic.

Table 2 Temperature characteristics of segments (minimal and maximal temperature and parameters of the models of temperature trends) recorded in a subset of trunks (12 snags and 26 logs), values for other segments were predicted. Characteristic Recorded range (range of predicted values) Tmin Tmax f_ND b*M_ND c*V_ND c*V_DI

5.5 to 0.7
C (5.3 to 10.2) 23.5e46.8
C (21.3e46.2) 8.4e111.5 (0e115.3) 2.3e26.7 (0e23.2) 5.9e102.5 (5.2e100.3) 30.8e285.3 (0e231.1)

Description

Standard error of prediction

Minimal temperature recorded from June to October 2013

0.8

Maximal temperature recorded from June to October 2013

5.1

Frequency of night temperature decreases. A greater value expresses more changes Rate of change of the frequency of night temperature decreases with the change of their magnitude (parameter M). A greater value expresses fewer large changes Rate of change of the frequency of night temperature decreases with the change of their velocity (parameter V). A greater value expresses fewer fast changes Rate of change of the frequency of day temperature increases with the change of their velocity (V). A greater value expresses fewer fast changes

Information on trophic modes was obtained from Jülich (1984), Hansen and Knudsen (2000) and from the identification literature mentioned below. The species included in the study are listed in Table 3. Nomenclature follows Knudsen and Vesterholt (2012) for
n (2010) for corticioids, Ryvarden and agarics, Bernicchia and Gorjo Melo (2014) for polypores, and Index Fungorum (http://www. indexfungorum.org, accessed in 2015) for ascomycetes and heterobasidiomycetes. Red Lists of the Czech Republic (Holec and Beran, 2006) and Bavaria (Karasch and Hahn, 2010) were used to identify red-listed species. 2.4. Data processing Smoothing and pruning of temperature trends. Temperature

19.1 4.5 19.0 54.8

measurements were separated into day (6:00 to 17:59) and night (18:00 to 5:59) sequences. For each sequence, local minima and maxima were identified, and any smaller temperature change in the opposite direction occurring between two consecutive measurements of the same direction was omitted and such local minima and/or maxima were not considered. The parameters of selection to define a smaller change were set up for a matrix of magnitude (M) and velocity (V) values (magnitude range 0.1e5.0
C, and velocity of 0.0001e0.4
C min1; see later for the use of the velocity parameter). This algorithm for each selected pair of values of magnitude and velocity (within their respective ranges) was repeated until the function did not have any steps smaller than selected values of M or V. At each repetition, a set of new local maxima and minima was established, and the temperature trends

4

V. Pouska et al. / Fungal Ecology 21 (2016) 1e9

Table 3 Fungal species found on segments of 98 trunks of Picea abies on Trojmezn
a Mt., Bohemian Forest. Abbreviations are shown for species included in ordination analyses. Redlisted species are in bold. Species

Abbreviation

Number of segments Snags (30 in total)

Alutaceodontia alutacea Amyloxenasma grisellum ? Antrodia serialis Aphanobasidium pseudotsugae Armillaria spp. (rhizomorphs) Ascocoryne cylichnium Bertia moriformis Botryobasidium laeve Botryobasidium subcoronatum Botryobasidium vagum Calocera viscosa Chrysomphalina chrysophylla Cinereomyces lindbladii Coryne dubia (anamorph of Ascocoryne sarcoides) Dacrymyces stillatus Exidia pithya Exidia saccharina Fomitopsis pinicola Globulicium hiemale Gymnopilus penetrans Gymnopilus picreus Gymnopus androsaceus Heterobasidion parviporum Hymenochaete fuliginosa Hyphoderma argillaceum Hyphodontia pallidula Hypholoma dispersum Hypochnicium albostramineum ? Hypochnicium wakefieldiae Jaapia ochroleuca Lentinellus castoreus Lophium mytilinum Mucronella bresadolae Mycena aciculata Mycena maculata Mycena rubromarginata Mycena silvae-nigrae Mycena viridimarginata Oligoporus caesius Orbilia xanthostigma Peniophorella praetermissa Phellinus nigrolimitatus Phellinus viticola Pholiota astragalina Pholiota flammans Rigidoporus sanguinolentus Pseudohydnum gelatinosum Simocybe sumptuosa Stereum sanguinolentum Trechispora hymenocystis Trichaptum abietinum Tricholomopsis decora Tubulicrinis glebulosus ? Tubulicrinis subulatus Veluticeps abietina Xylodon asperus

AntSer AphPseu rhArm

1

BerMor

1

BotSub BotVag

DacSti ExPith FomPin

16 3 1 10

GymPic GyAndro HymFul

1

HyphD

Logs (68 in total) 1 1 5 7 2 1 10 1 3 3 1 1 1 1 21 2 10 1 1 5 3 2 2 2 2 3 1 1 2 1

1 MucBre MycMac MycRub MycSN MycVir OliCae OrbXan PePrae PheNig PheVit

2

1 3

PhySan

3 1 4 3 3 8 3 6 2 21 12 1 1 3 1 1

1 1 3 1

TriAbi

were considered uniform (i.e. either always decreasing or always increasing) between neighbouring local minima or maxima. Then, for each sequence of uniform temperature change a parameter of minimal velocity of the change was applied from the beginning of the function and if not fulfilled, then the respective step(s) was(were) not considered, and the sequence was thus truncated (i.e. function pruning). Sequences shorter than three time steps (intervals between temperature measurements) were not considered. Finally, each triplet (hereafter called event) of uniformly changing temperature time steps was counted and distinguished into night

1 1 1 1

temperature decreases or day temperature increases. Models of temperature trends. Counts of the events with different parameters of the magnitude of temperature changes as well as the velocity of such changes were plotted as a 3D surface. The following function was applied to fit the resulting area:

pffiffiffiffi N ¼f bMcV where f is the frequency of events (intercept); the coefficient b represents the speed of decrease in event frequency with

V. Pouska et al. / Fungal Ecology 21 (2016) 1e9

increasing parameter magnitude (M); and coefficient c represents the speed of decrease in event frequency with increasing parameter velocity (V). In other words, the greater is a value (b or c), the smaller is the frequency (or probability) of more drastic (long or fast) temperature changes. Statistical fits of each model were tested (the significance of each regression coefficient as well as of the whole regression equation), and only significant fits (p < 0.05) were considered. Night temperature decreases (ND) and day temperature increases (DI) were the only temperature trends used, and a combination of function terms (e.g. b*M) with temperature trends defined the temperature characteristics (Table 2). 2.5. Statistical analyses The parameters of the above mentioned function (listed among temperature characteristics in Table 2) were subsequently used in statistical analyses. Relations of temperature characteristics to trunk type, diameter class (<30 cm ‘thin’, 30 cm ‘thick’), decay stage, vegetation and water content (converted to three categories according to frequency) were analysed using Generalized Linear Mixed Models (GLMM, Supplementary data 1). Differences in water content between snags and logs, diameter classes and decay stages were tested using the t-test or KruskaleWallis test. Temperature measurements were done on a subset of segments (12 snags and 26 logs). Linear models were constructed on the basis of characteristics (Table 1) available for all segments to obtain missing values of temperature characteristics (Table 2). Parametric transformations (no transformation, log transformation, squareroot transformation and square transformation) were considered to maximise the linearity of relations of predictors with individual response variables. Final models (Supplementary data 2) were constructed using a step-wise selection procedure from all predictors (Table 1) based on AIC value (function step). The function predict was applied on the final models to obtain missing values. To estimate the reliability of models, standard errors of predictions (standard deviation of residuals of predicted values) were calculated on the basis of the jackknife method (Efron and Tibshirani, 1993). All statistical analyses except ordination analyses (see below) were done using the program R, version 3.1.1 (R core team, 2014). The species composition of wood-decaying fungi on segments and relationships to the characteristics of segments were analysed using unconstrained and constrained ordination methods in Can oco 5 (ter Braak and Smilauer, 2012). Because the species composition on segments was highly variable, we expected that most species would have unimodal relations to gradients in the data, and decided to use methods based on the model of unimodal species response, i.e. Detrended Correspondence Analysis (DCA) and Canonical Correspondence Analysis (CCA), which was confirmed by the length of axes in DCA. Because standardisation by species (as well as by samples) is inherent in DCA/CCA, the effect of species with low frequencies might introduce high levels of noise into the results. To avoid this, we omitted species with less than three occurrences and also 6 segments without sporocarps. Finally, 25 species (185 records) on 86 segments (27 snags and 59 logs) were used in the analyses. Characteristics listed in Tables 1 and 2 and causes of tree death were used as predictors of fungal species composition on segments. Ordination analyses were done in the same manner as by Pouska et al. (2011). The cover characteristics (lichens, bryophytes and vegetation) originally recorded as percentages were log (x þ 1) transformed. The unconstrained ordination (DCA) was calculated, i.e. the ordination based solely on species composition, and we examined the correlation of individual characteristics with these unconstrained ordination axes. The Kendall's rank correlations of

5

characteristics with the DCA axes (CaseR scores) were calculated. Studied trunks were scattered on the slope in groups that differed in sample size and in combinations of diameter and decay stages. Since such groups could not be used as blocks in analyses, elevation, which had a significant marginal effect in CCA (p ¼ 0.011, 2.1% of variation explained), was set as the covariable in tests of simple partial effects, in forward selection and in variation partitioning. In CCA, two complementary strategies of selecting important predictors were used. First, characteristics were selected using forward selection. This approach selects a set of good and sufficient predictors so that no other predictor is able to significantly improve the fit. Because some of the (biologically interesting) predictors might have been excluded simply due to their correlation with other predictors, we also calculated the simple partial effects of individual characteristics. The simple partial effect is here the effect of a single predictor, independent of the others, after accounting for the effect of elevation. Variation partitioning (Borcard et al., 1992)  with adjusted explained variation (see Smilauer and Leps, 2014) was applied to separate the variability explained by the three groups of variables, i.e. descriptive (Table 1 except elevation and water content), microclimatic (water content and Table 2), and cause of tree death. The cause of tree death has five categories, which corresponds to four degrees of freedom because the last category is collinear since they sum up to one for each sample. Therefore, using forward selection, we reduced the number of variables in each other group so that it did not exceed four: bark, bryophytes, ground contact and surface disintegration were among the descriptive characteristics; water content, c*V_DI and f_ND were among the microclimatic characteristics. The significance was tested using Monte-Carlo permutation tests with 999 permutations. CCAs were calculated on the same data set as DCA, and in addition, simple partial effects and forward selection were calculated for segments of logs only, 21 species (139 records) on 59 segments. 3. Results The most abundant fungal species on segments were Dacrymyces stillatus, Phellinus nigrolimitatus and Fomitopsis pinicola (Table 3). In total, 220 records were made of 56 species on 98 segments. Eight red-listed species were recorded, two of them (P. nigrolimitatus and Phellinus viticola) were fairly abundant, Gymnopilus picreus occurred five times, Hymenochaete fuliginosa three times, but the other four species occurred only once (Table 3). Results of GLMM (details in Supplementary data 1) showed that Tmin was lower in thin segments and decreased with lower water content; Tmax was higher in thin segments and increased with lower water content. Both f_ND and b*M_ND were higher in snags and increased with decay, i.e. there were more but smaller temperature decreases in snags as compared to logs as well as in more decayed snags and logs. The parameter c*V_ND was higher in snags than in logs and also in more decayed trunks, which indicates that temperature decreased more slowly in snags and in more decayed trunks (both snags and logs). The parameter c*V_DI was higher in snags than in logs. In other words temperature increased more slowly in snags. Water content was higher in logs than in snags, higher in thick than in thin logs though not in snags, and it increased with the progression in wood decay, being lower in weakly decayed logs than in moderately and strongly decayed logs (Table 4). The result of DCA (Fig. 1) and correlations of characteristics with DCA axes (Table 5) show that the first axis corresponded mainly to the advancing decay of wood, and also to water content. Species with sporocarps occurring on wood in the early stage of decay (and with lower water content) were e.g. Exidia pithya and Trichaptum

6

V. Pouska et al. / Fungal Ecology 21 (2016) 1e9

Table 4 Results of tests [t-test (T), KruskaleWallis (H)] of differences in water content between groups of segments, with corresponding means and group sizes. Significant p-values are in bold. Mean water content (n) Segment groups Trunk type (snag, log) Diameter class (thin, thick) Snag diameter class (thin, thick) Log diameter class (thin, thick) Decay stage (1, 2, 3) Snag decay stage (1, 2) Log decay stage (1, 2, 3) a e b

Group 1 40.95 (30) 49.79 (40) 40.99 (14) 54.58 (26) 49.1 (38)a 45.44 (19) 52.76 (19)a

Test statistics Group 2 65.12 (68) 63.19 (58) 44.26 (16) 71.64 (42) 57.68 (41)a 33.18 (11) 66.66 (30)b

Group 3

T/H 4.85 2.66 0.42 3.08 15.79 1.6 10.8

75.04 (19)b 75.04 (19)b

p 0.001 0.009 0.680 0.003 0.001 0.121 0.005

denote significant difference between respective groups for KruskaleWallis test.

Table 5 Kendall's rank correlations (tau) between characteristics and the first two DCA axes, and simple partial effects (explained variability) of characteristics used in CCA. Characteristic

DCA e 1st axis

DCA e 2nd axis

CCA

Elevation Trunk type (snag/log) Diameter Decay stage Surface disintegration Bark Branches Ground contact Lichens Bryophytes Saplings Vegetation Contacting wood Openness Water content Tmin Tmax f_ND b*M_ND c*V_ND c*V_DI

0.136 0.422*** 0.112 0.602*** 0.616*** 0.499*** 0.015 0.448*** 0.264*** 0.453*** 0.412*** 0.305*** 0.147 0.190* 0.249*** 0.034 0.109 0.024 0.029 0.171* 0.152*

0.013 0.013 0.031 0.079 0.131 0.102 0.005 0.091 0.092 0.008 0.111 0.100 0.149 0.027 0.267*** 0.188* 0.111 0.057 0.095 0.029 0.085

Covariable 3.1** 1.5 4.4** 5.4** 5.9** 1.3 3.7** 2.5** 5.0** 2.8** 3.1** 1.0 1.1 3.8** 1.2 1.9* 2.0* 2.0* 1.4 2.0*

***p < 0.001; **0.001  p < 0.01; *0.01  p < 0.05.

Fig. 1. (A) DCA ordination diagram showing fungal species assemblages on 2 m long segments of spruce trunks. The first (horizontal) and second (vertical) axes are shown, together explaining 15.3% of the variability in species data. Full species names are given in Table 3. (B) Ex-post projection of characteristics recorded for segments significantly correlated with these axes. The arrow direction of Tmin corresponds to higher (less extreme) values.

abietinum, and species occurring in the late stage of decay were, for example, Mucronella bresadolae and Mycena maculata. The following characteristics of segments were significantly correlated with the first axis in DCA: trunk type (snag/log), decay stage, surface disintegration, bark, ground contact, lichens, bryophytes, saplings, vegetation, openness, water content, c*V_ND and c*V_DI;

the second axis was correlated with water content and Tmin. The third axis was correlated with bark, lichens, saplings and c*V_DI, and the fourth axis was correlated only with elevation. The following characteristics had significant simple partial effects in CCA: bark, surface disintegration, bryophytes, decay stage, water content, ground contact, trunk type, vegetation, saplings, lichens, f_ND, b*M_ND, c*V_DI and Tmax (Table 5). The analysis for logs (excluding snags) showed significant simple partial effects of these characteristics (order according to the amount of explained variability): bark, surface disintegration, bryophytes, decay stage, vegetation, water content, lichens, saplings, ground contact, Tmin and Tmax. Forward selection in CCA identified bark, water content, bryophytes, surface disintegration, lichens and f_ND (in this order) as the best complementary predictors. Since surface disintegration was strongly positively correlated with decay stage (Supplementary data 1), the latter was not selected. Fig. 2 shows that especially sporocarps of E. pithya, F. pinicola and T. abietinum occurred on wood with greater bark cover. P. nigrolimitatus (redlisted) occurred frequently on wood with greater bryophyte cover; Physisporinus sanguinolentus, M. bresadolae and Mycena viridimarginata preferred wood with a disintegrated surface. G. picreus (red-listed) occurred on wood with higher water content and also with greater bryophyte cover. Only H. fuliginosa and P. viticola (both red-listed) were relatively frequent on drier wood with a higher f_ND. None of the species strongly preferred wood with higher

V. Pouska et al. / Fungal Ecology 21 (2016) 1e9

7

4. Discussion

Fig. 2. CCA ordination diagram showing the relations of fungal species to descriptive and microclimatic characteristics selected by the forward selection. The first and second axes are shown, and together explain 10.4% of variability in species data.

lichen cover, but Botryobasidium subcoronatum had a slightly positive affinity. Peniophorella praetermissa and D. stillatus were less frequent on wood with higher water content or bryophyte cover. Forward selection in the analysis for logs only identified these characteristics: bark, bryophytes, vegetation, surface disintegration and b*M_ND; relations of species to bark, bryophytes and surface disintegration (Supplementary data 3) were consistent with relations in the analysis for all segments (Fig. 2), although water content, lichens and f_ND were not selected in this case. Instead, vegetation and b*M_ND were included. According to other axes (not shown), Mycena silvae-nigrae and M. viridimarginata were relatively frequent on logs with higher vegetation cover, and Orbilia xanthostigma, B. subcoronatum and Mycena rubromarginata on logs with a higher b*M_ND. Variation partitioning (Fig. 3) showed significant partial effects of all three groups of variables, i.e. descriptive ones related to wood decay (bark, bryophytes, ground contact, surface disintegration; p ¼ 0.001), cause of tree death (p ¼ 0.003), and microclimatic (water content, c*V_DI, f_ND; p ¼ 0.004). Descriptive characteristics explained the largest amount of variation in species composition (9.9%), with 6.1% being unique for them. The cause of tree death explained 5.1% of the variation (uniquely explained 2.7%). Microclimatic characteristics explained 4.6% (uniquely explained 1.8%). In total, 14.3% of the variation (25.6% if unadjusted) was explained by these three groups of variables.

Fig. 3. Venn diagram of variation partitioning among the four descriptive characteristics, cause of tree death and the three microclimatic characteristics. The parts of variation that were explained exclusively by one group of characteristics or by more groups together are shown. Asterisks denote the significance of partial effects (p < 0.01).

The occurrence of sporocarps of different species of wooddecaying fungi was influenced by moisture and temperature conditions in trunks. Furthermore, it was related to several characteristics expressing the process of wood decay and the cause of tree death, as has also been shown in earlier studies (Renvall, 1995; Høiland and Bendiksen, 1997; Heilmann-Clausen, 2001; Pouska et al., 2011). Similarly to our study, Fukasawa et al. (2009) found a significant effect of decay stage, moisture and type of beech trunks on the composition of the fungal community. Also in spruce logs, the community composition of (metabolically active) fungi was shown to have a relationship to wood density, diameter and moisture (Rajala et al., 2011). Yet, to the best of our knowledge, no study has documented a relationship of the wood-decaying fungal community composition to temperature under natural conditions. Most characteristics of trunks were correlated (most of them are also influenced by the activity of fungi). Our results indicate that temperature fluctuations are fewer at greater water content, and both water content and temperature stability increase with the
 progression in the decay of wood (see also Rypa cek, 1957; Harmon et al., 1986; Renvall, 1995; Fukasawa et al., 2009; Forrester et al., 2012; Rajala et al., 2012). Furthermore, our measurements indicate that trunks with larger diameter buffered temperature changes in their surface layer (measurement depth 5 cm). This implies higher temperature stability in thick trunks as compared to thinner ones (Harmon et al., 1986; Stokland et al., 2012). Similarly, thick logs were wetter, and, as Harmon et al. (1986) reported, moisture is more stable deeper in logs. In addition to increasing water holding capacity with progressing decay, moisture is generally influenced by a combination of factors including trunk type, bark, lichen and bryophyte cover, ground contact and shading, due to differences in water absorption from soil and from precipitation and differences in drying (e.g. Harmon et al., 1986; Griffith and Boddy, 1991; Harmon and Sexton, 1995; Renvall, 1995). In addition, a considerable amount of water in dead wood originates from respiration processes (Ryp
a cek, 1957; Harmon et al., 1986; Stokland et al., 2012) and this may stabilize moisture especially in large trunks (cf. Boddy, 1983a). Wood-decaying fungi differ in their tolerance to heat (Carlsson et al., 2012) and many fungi from the temperate zone cease to
cek, 1957; grow if temperature increases above 35
C (e.g. Rypa Boddy, 1983c). The temperature in some thinner trunk segments exceeded 40
C during hot sunny days. Temperature can also exceed 40
C in the sun-exposed surface of thicker trunks (see Graham, 1925 or Savely, 1939 for temperatures in sun-exposed logs), but was lower in our case because trunk sides with thermometers were not exposed to intense solar radiation; furthermore, temperature maxima were lower deeper in thick segments (Pouska and Macek, unpublished results). Temperature in thin unshaded trunks was sometimes too far from the optima of many species, and may have even exceeded their growth limits (32e44
C, Humphrey and Siggers, 1933). Considering that the thinnest trunks were also prone to desiccation, they may have indeed been unfavourable for some species of fungi. Extreme high temperatures occur infrequently and we do not expect that this would eliminate species even in the thinnest trunks (less stresstolerant species could survive in an inactive state). Temporal high temperature may change competitive balance between species (Carlsson et al., 2014), and even if mycelia of species less tolerant to stress are present in wood, their physiological constraints and decreased competitiveness can hamper their ability to fruit. Heilmann-Clausen (2001) concluded that the second ordination axis in DCA (i.e. the second major variation in species composition) was, in addition to decay rate, related to the amount of

8

V. Pouska et al. / Fungal Ecology 21 (2016) 1e9

microclimatic fluctuation induced by wind exposure, distance to forest edge and sun exposure. Our results are in concert with this observation, as two characteristics expressing microclimate inside trunks, i.e. water content and minimal temperature, were correlated with the second most important community gradient (Table 5); furthermore, in CCA, temperature fluctuation frequency (f_ND) and again water content clearly corresponded to the second ordination axis. Temperature stability (further emphasized by a negative correlation of minimal and maximal temperatures; Supplementary data 1) is hence an important factor related to fungal species composition in dead wood. Environmental characteristics with a potential to influence microclimatic conditions, such as vegetation and openness, were correlated only with the most important community gradient (the first DCA axis). While this contradicts the results of Heilmann-Clausen (2001), it agrees with those of Lindblad (1998). Such a discrepancy might be explained by the different woody species involved, as our and Lindblad's studies were performed on spruce, while Heilmann-Clausen studied fungal communities on beech, and also by different site conditions. In all three studies (the present study; Lindblad, 1998; HeilmannClausen, 2001), ground contact was correlated with the most important community gradient. In addition, two characteristics of the velocity of temperature changes (c*V_ND and c*V_DI) and also water content, were correlated with the first axis in DCA, meaning that all these characteristics correspond to the progression in wood decay and to the main gradient in species composition. The constrained analysis of relationships among environmental characteristics and fungal species (CCA) revealed significant effects of all three groups of characteristics (descriptive, cause of tree death and microclimatic) on the community assembly. Descriptive characteristics explained the largest part of the variability and confirmed that characteristics related to the process of wood decay play a crucial role in fungal community composition. Variation partitioning showed the importance of cause of tree death for the occurrence of fungal sporocarps. Although 2 m long segments host only some parts of the communities that can be found along whole logs (Pouska et al., 2011), these species assemblages are still associated with the causes of tree death. Finally, variation partitioning showed that although microclimatic characteristics uniquely explained the smallest part of the variation, their effect on species composition was still highly significant. In line with earlier observations of different colonisers on log parts with different moisture (Hudson, 1968, p. 845), it is likely that microclimate modifies ways of succession in otherwise similar trunks, which may bring additional heterogeneity to pathways influenced by primary colonisers. Still, further studies would be needed to elucidate the influence of microclimate on succession under natural conditions. Surprisingly, the diameter of segments did not have a significant relationship to species composition, either when tested for snags and logs together, or for logs only. Although this is inconsistent with relationships previously observed in natural logs of spruce (Høiland and Bendiksen, 1997; Lindblad, 1998; Pouska et al., 2011) and in 4 m long cut logs (Edman et al., 2004), we attribute this discrepancy to the complex design in our study (the selection of trunks of different diameters combined with different decay stages, e.g. trunks of similar diameter yet with several other contrasting properties). For instance, the ranges of values in thin/thick trunks as compared to the range in all trunks were 94/100% in water content, 100/50% in Tmin, 73/80% in Tmax, 100/97% in f_ND, 87/92% in b*M_ND, 100/95% in c*V_ND, 100/95% in c*V_DI, 100/98% in bark, 100/70% in lichens, 94/100% in bryophytes, 100/94% in vegetation, 96/97% in openness; and 100/100% in all the following characteristics: elevation, trunk type (snag or log), decay stage, surface disintegration, ground contact and saplings. We, therefore, assume

that trunk thickness is not a sufficiently narrow characteristic within our dataset that would itself drive species composition, because several other characteristics seem to be driving it. On the other hand, as water content, minimal and maximal temperatures changed with diameter, these characteristics may, thus, partly reflect its effect. We, therefore, conclude that the diameter per se was not significantly related to species assemblages on the segments studied here, but its effect may be noticeable through more stable temperatures and greater moisture in thicker trunks. Incidentally, three red-listed species that were included in ordination analyses belong to the family Hymenochaetaceae. Despite their relatedness, P. nigrolimitatus seems to prefer quite different wood than the other two species, H. fuliginosa and P. viticola. While P. nigrolimitatus was more frequently found on more decomposed wood with greater cover of bryophytes, both H. fuliginosa and P. viticola did not show such a preference, and their incidence on trunk segments with frequent temperature fluctuations that were negatively correlated with moisture corresponds to the ability of these species to produce sporocarps on relatively drier wood. The fourth red-listed species, G. picreus, is distant in phylogeny from the previously mentioned species and preferred very wet logs with abundant bryophytes. The red-list status of these species differs; P. nigrolimitatus and H. fuliginosa are red-listed both in the Czech Republic and in Bavaria, whereas P. viticola is red-listed only in Bavaria and is fairly common in the Bohemian Forest €ssler et al., 2012). G. picreus is also red-listed only in Bavaria. The (Ba remaining four red-listed species (Table 3) were found only rarely. Similarly to the associations of red-listed species with different causes of tree death (Pouska et al., 2011), their different relations to microclimate support the commonly accepted idea that the heterogeneity of dead wood is important for the diversity of fungi and for the survival of rare and threatened species (e.g. Stokland et al., 2012; Abrego and Salcedo, 2013). Our results agree with general hypotheses on the development of fungal communities in decaying wood, which may, in addition to the process of wood decomposition, also result from competitive interactions and microclimatic conditions (Heilmann-Clausen, 2001; Crowther et al., 2014). The significance of drought stress has been emphasised in structuring fungal communities (Crowther et al., 2014), and accordingly, in our study water content was the best explanatory variable among the characteristics of the microclimate inside trunks. Acknowledgments
and M. Pouskov
We thank M. Hegedüs, L. Kubistova a for assistance with data collection, M. Pavlí cko for accommodation, H. Ostrow for the identification of corticioids, J. Leps for help with 
sek and P. Smilauer model formulation, J. Kuba for advice on data analysis, D. W. Hardekopf for revising the English, and reviewers for comments. This study was supported by the Czech Science Foundation (13-23647P). PM was supported by the CzechPolar grant MSMT LM2010009. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.funeco.2016.01.006. References Abrego, N., Salcedo, I., 2013. Variety of woody debris as the factor influencing woodinhabiting fungal richness and assemblages: is it a question of quantity or quality? For. Ecol. Manag. 291, 377e385.
r, J., 2005. Fungal Communities in Branch Litter of Norway Spruce: Dead Allme Wood Dynamics, Species Detection and Substrate Preferences. Dissertation.

V. Pouska et al. / Fungal Ecology 21 (2016) 1e9 Swedish University of Agricultural Sciences, Uppsala. Bader, P., Jansson, S., Jonsson, B.G., 1995. Wood-inhabiting fungi and substratum decline in selectively logged boreal spruce forests. Biol. Conserv. 72, 355e362. €ssler, C., Müller, J., Dziock, F., Brandl, R., 2010. Effects of resource availability and Ba climate on the diversity of wood-decaying fungi. J. Ecol. 98, 822e832. €ssler, C., Müller, J., Svoboda, M., Lepsova
, A., Hahn, C., Holzer, H., Pouska, V., 2012. Ba Diversity of wood-decaying fungi under different disturbance regimes e a case study from spruce mountain forests. Biodivers. Conserv. 21, 33e49.
n, S.P., 2010. Corticiaceae s.l., Fungi Europaei 12. Edizioni CanBernicchia, A., Gorjo dusso, Alassio. € , L., 2000. Microcalorimetric measurements of metabolic activity Bjurman, J., Wadso of six decay fungi on spruce wood as a function of temperature. Mycologia 92, 23e28. Boddy, L., 1983a. Microclimate and moisture dynamics of wood decomposing in terrestrial ecosystems. Soil Biol. Biochem. 15, 149e157. Boddy, L., 1983b. Carbon dioxide release from decomposing wood: effect of water content and temperature. Soil Biol. Biochem. 15, 501e510. Boddy, L., 1983c. Effect of temperature and water potential on growth rate of woodrotting basidiomycetes. Trans. Br. Mycol. Soc. 80, 141e149. Boddy, L., 2000. Interspecific combative interactions between wood-decaying basidiomycetes. FEMS Microbiol. Ecol. 31, 185e194. Borcard, D., Legendre, P., Drapeau, P., 1992. Partialling out the spatial component of ecological variation. Ecology 73, 1045e1055.  ter Braak, C.J.F., Smilauer, P., 2012. Canoco Reference Manual and User's Guide: Software for Ordination, Version 5.0. Microcomputer Power, Ithaca. Carlsson, F., Edman, M., Holm, S., Eriksson, A.-M., Jonsson, B.-G., 2012. Increased heat resistance in mycelia from wood fungi prevalent in forests characterized by fire: a possible adaptation to forest fire. Fungal Biol. 116, 1025e1031. Carlsson, F., Edman, M., Holm, S., Jonsson, B.-G., 2014. Effect of heat on interspecific competition in saprotrophic wood fungi. Fungal Ecol. 11, 100e106. Chesson, P., Huntly, N., 1997. The roles of harsh and fluctuating conditions in the dynamics of ecological communities. Am. Nat. 150, 519e553. Crowther, T.W., Maynard, D.S., Crowther, T.R., Peccia, J., Smith, J.R., Bradford, M.A., 2014. Untangling the fungal niche: the trait-based approach. Front. Microbiol. 5, 579. Edman, M., Kruys, N., Jonsson, B.G., 2004. Local dispersal sources strongly affect colonization patterns of wood-decaying fungi on spruce logs. Ecol. Appl. 14, 893e901. Efron, B., Tibshirani, J., 1993. An Introduction to Bootstrap. Chapman and Hall, New York. Forrester, J.A., Mladenoff, D.J., Gower, S.T., Stoffel, J.L., 2012. Interactions of temperature and moisture with respiration from coarse woody debris in experimental forest canopy gaps. For. Ecol. Manag. 265, 124e132. Fukasawa, Y., Osono, T., Takeda, H., 2009. Dynamics of physicochemical properties and occurrence of fungal fruit bodies during decomposition of coarse woody debris of Fagus crenata. J. For. Res. 14, 20e29. Graham, S.A., 1925. The felled tree trunk as an ecological unit. Ecology 6, 397e411. Griffith, G.S., Boddy, L., 1991. Fungal decomposition of attached angiosperm twigs. II. Moisture relations of twigs of ash (Fraxinus excelsior L.). New Phytol. 117, 251e257. Hansen, L., Knudsen, H. (Eds.), 2000. Nordic Macromycetes. Ascomycetes, vol. 1. Nordsvamp, Copenhagen. Harmon, M.E., Franklin, J.F., Swanson, F.J., Sollins, P., Gregory, S.V., Lattin, J.D., Anderson, N.H., Cline, S.P., Aumen, N.G., Sedell, J.R., Lienkaemper, G.W., Cromack Jr., K., Cummins, K.W., 1986. Ecology of coarse woody debris in temperate ecosystems. Adv. Ecol. Res. 15, 133e302. Harmon, M.E., Sexton, J., 1995. Water balance of conifer logs in early stages of decomposition. Plant Soil 172, 141e152. Heilmann-Clausen, J., 2001. A gradient analysis of communities of macrofungi and slime moulds on decaying beech logs. Mycol. Res. 105, 575e596. Heilmann-Clausen, J., Aude, E., van Dort, K., Christensen, M., Piltaver, A.,

r, T., Odor, Veerkamp, M., Walleyn, R., Siller, I., Standova P., 2014. Communities of wood-inhabiting bryophytes and fungi on dead beech logs in Europe e reflecting substrate quality or shaped by climate and forest conditions? J. Biogeogr. 41, 2269e2282. Hendry, S.J., Boddy, L., Lonsdale, D., 2002. Abiotic variables effect differential expression of latent infections in beech (Fagus sylvatica). New Phytol. 155, 449e460.  
Holec, J., Beran, M. (Eds.), 2006. Cervený seznam hub (makromycet u) Cesk e republiky [Red List of fungi (macromycetes) of the Czech Republic]. Príroda, vol. 24, pp. 1e282 (In Czech with English summary). Høiland, K., Bendiksen, E., 1997. Biodiversity of wood-inhabiting fungi in a boreal

9

coniferous forest in Sør-Trøndelag County, Central Norway. Nord. J. Bot. 16, 643e659. Hudson, H.J., 1968. The ecology of fungi on plant remains above the soil. New Phytol. 67, 837e874. Humphrey, C.J., Siggers, P.V., 1933. Temperature relations of wood-destroying fungi. J. Agric. Res. 47, 997e1008. Jensen, K.F., 1969. Effect of constant and fluctuating temperature on growth of four wood-decaying fungi. Phytopathology 59, 645e647. €nsson, M.T., Edman, M., Jonsson, B.G., 2008. Colonization and extinction patterns Jo of wood-decaying fungi in a boreal old-growth Picea abies forest. J. Ecol. 96, 1065e1075. €tterpilze, Gallertpilze und Bauchpilze. In: Gams, H. Jülich, W., 1984. Die Nichtbla (Ed.), Kleine Kryptogamenflora. Band IIb/1. Gustav Fisher Verlag, Jena. Junninen, K., Komonen, A., 2011. Conservation ecology of boreal polypores: a review. Biol. Conserv. 144, 11e20. Karasch, P., Hahn, C., 2010. Rote Liste gef€ ahrdeter Großpilze Bayerns. Bayerisches Landesamt für Umwelt, Augsburg. Knudsen, H., Vesterholt, J. (Eds.), 2012. Funga Nordica, Agaricoid, Boletoid, Clavaroid, Cyphelloid and Gastroid Genera. Nordsvamp, Copenhagen. Kruys, N., Jonsson, B.G., 1999. Fine woody debris is important for species richness on logs in managed boreal spruce forests of northern Sweden. Can. J. For. Res. 29, 1295e1299. Kubartov
a, A., Ottosson, E., Dahlberg, A., Stenlid, J., 2012. Patterns of fungal communities among and within decaying logs, revealed by 454 sequencing. Mol. Ecol. 21, 4514e4532. Lindblad, I., 1998. Wood-inhabiting fungi on fallen logs of Norway spruce: relations to forest management and substrate quality. Nord. J. Bot. 18, 243e255.
n, J., Penttil€ Norde a, R., Siitonen, J., Tomppo, E., Ovaskainen, O., 2013. Specialist species of wood-inhabiting fungi struggle while generalists thrive in fragmented boreal forests. J. Ecol. 101, 701e712.
, A., Edman, M., Jo €nsson, M., Lindhe, A., Stenlid, J., Ottosson, E., Kubartova Dahlberg, A., 2015. Diverse ecological roles within fungal communities in decomposing logs of Picea abies. FEMS Microbiol. Ecol. 91 fiv012.
, A., 2011. How do log characteristics inPouska, V., Leps, J., Svoboda, M., Lepsova fluence the occurrence of wood fungi in a mountain spruce forest? Fungal Ecol. 4, 201e209. R core team, 2014. R: a Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna. URL: http://www.R-project.org. €kip€ €, R., Pennanen, T., 2011. RNA reveals a Rajala, T., Peltoniemi, M., Hantula, J., Ma aa succession of active fungi during the decay of Norway spruce logs. Fungal Ecol. 4, 437e448. €a €, R., 2012. Fungal community dyRajala, T., Peltoniemi, M., Pennanen, T., M€ akipa namics in relation to substrate quality of decaying Norway spruce (Picea abies [L.] Karst.) logs in boreal forests. FEMS Microbiol. Ecol. 81, 494e505. Renvall, P., 1995. Community structure and dynamics of wood-rotting Basidiomycetes on decomposing conifer trunks in northern Finland. Karstenia 35, 1e51. 

akaRypa cek, V., 1957. Biologie drevokazných hub. Nakladatelství Ceskoslovensk e d, Praha. demie ve Ryvarden, L., Melo, I., 2014. Poroid fungi of Europe. Synop. Fungorum 31, 1e455. Savely Jr., H.E., 1939. Ecological relations of certain animals in dead pine and oak logs. Ecol. Monogr. 9, 321e385.  Smilauer, P., Leps, J., 2014. Multivariate Analysis of Ecological Data Using Canoco 5. Cambridge University Press, Cambridge. Stokland, J., Kauserud, H., 2004. Phellinus nigrolimitatus e a wood-decomposing fungus highly influenced by forestry. For. Ecol. Manag. 187, 333e343. Stokland, J.N., Larsson, K.-H., 2011. Legacies from natural forest dynamics: different effects of forest management on wood-inhabiting fungi in pine and spruce forest. For. Ecol. Manag. 261, 1707e1721. Stokland, J.N., Siitonen, J., Jonsson, B.G., 2012. Biodiversity in Dead Wood. Cambridge University Press, New York. Svoboda, M., Janda, P., Nagel, T.A., Fraver, S., Rejzek, J., Ba ce, R., 2012. Disturbance history of an old-growth sub-alpine Picea abies stand in the Bohemian Forest, Czech Republic. J. Veg. Sci. 23, 86e97. Tolasz, R., Míkov
a, T., Valeri
anov
a, A., Vo zenílek, V. (Eds.), 2007. Climate Atlas of Czechia. Czech Hydrometeorological Institute, Praha Olomouc. €gberg, N.O.S., 2006. Environmental Toljander, Y.K., Lindahl, B.D., Holmer, L., Ho fluctuations facilitate species co-existence and increase decomposition in communities of wood decay fungi. Oecologia 148, 625e631. Viitanen, H.A., 1997. Modelling the time factor in the development of mould fungi e the effect of critical humidity and temperature conditions on pine and spruce sapwood. Holzforschung 51, 6e14.