Community structure and functional diversity of polypores ( Basidiomycota ) in the Atlantic Forest of Santa Catarina State , Brazil

Ecological studies have suggested that different groups of polypore species, acting as parasites and/ or saprophytes, degrade different types of woody substrates. These functional groups have different decay capabilities and hence different roles in ecosystems. The aim of this study was to describe the community (species composition and their functionality inferred on the basis of substrate preference) of wood-decaying polypores in


Introduction
Polypores constitute one of the main groups of wood-decay fungi.The diversity of wood-decay fungi of the Atlantic Forest has been approached by several authors (e.g., LOGUERCIO-LEITE et al., 2002;RYVARDEN;MEIJER, 2002;DRECHSLER-SANTOS et al., 2008).Recently, Baltazar and Gibertoni (2009) presented the most extensive literature list with 733 aphyllophoraceous species, of which more than 50% were polypores.
Despite their importance in the ecology of forests, there are few ecological studies about this group in the Atlantic Forest (e.g., COCKLE et al., 2012).Moreover, little is known about the functional diversity of this group of fungi in the Atlantic Forest.Ecological studies have suggested that different groups of polypore species, acting as parasites and/or saprophytes, degrade different types of woody substrates.These functional groups have distinct decay capabilities and perform different roles in ecosystems.Among the wood-decay fungi, Polyporaceae s.l., as well as some Corticiaceae s.l. and Hymenochaetaceae, have been considered well suited to study the key role of woody debris in old-growth forests (BADER et al., 1995).
Accordingly, Urcelay and Robledo (2004), on the basis of a study conducted in monospeciic Alnus acuminata Kunt (Betulaceae) forests of the Yungas in northwestern Argentina, proposed that it is possible to recognize functional groups of polypores according to the preferences of the species for particular types of substrate.The authors recognized groups of species that are directly related to different types of substrates (trunks and branches/twigs of trees living or dead, standing or on the ground and of different volumes) as well as different decay stages.
The polypore functional groups proposed by Urcelay and Robledo (2004) should be tested in a highly diverse ecosystem, such as the Atlantic Forest of Brazil.This domain is a diversiied mosaic, showing distinguishable structures and loristic composition, due to the differences in soil, topography and climate characteristics in the broad area of occurrence in Brazil.The plant species diversity estimated for the Atlantic Forest is around 20,000, of which 55% are trees (MITTERMEIER et al., 1999;CAPOBIANCO, 2001).This high diversity provides a large number of niches with different variables (humidity, temperature, volume and diameter of trunks and branches, stage of decomposition of wood, and physical and chemical structure of dead trees), which may influence the functional diversity and community structure of wooddecay fungi (RAYNER; TODD, 1979).
The aim of this study was to describe and characterize the community of wood-decay polypores of the Parque Nacional da Serra do Itajaí (PARNA-SI), in the Atlantic Forest of Santa Catarina State, Brazil, in terms of richness and abundance, and to infer their functionality using substrata/hosts.

Study area
The Parque Nacional da Serra do Itajaí (PARNA-SI) is located in the Itajaí Valley, in Santa Catarina State (Figure 1A), covering an area of 57,374 ha and with altitudes from 80 to 1039 m.a.s.l.The climate is temperate humid with warm summers (Cfa), and there is signii cant precipitation in all seasons (GAPLAN, 1986).The annual mean temperature ranges between 17 and 22°C, and the annual rainfall ranges between 1600 and 1800 mm, distributed between 120 and 140 days of rain during the year.The mean relative humidity is between 75 and 85% (KLEIN, 1979).This conservation area comprises a signii cant portion of the remnants of the Atlantic Forest domain, with patches of primary forests and others in advanced stages of regeneration (source: SOS Mata Atlântica and INPE Instituto Socioambiental).The study was conducted in the Parque Municipal Natural Nascentes do Garcia (extending from 27°01' to 27°06'S and from 49°06' to 49°10'W), Blumenau municipality, where two sample areas of 5000 m 2 (100×50 m) were plotted (Figure 1B).
The vegetation in the study area is classii ed as dense ombrophilous forest (GAPLAN, 1986).
All basidiomes of polypore species were recorded and all of the same species in the same substrate were recorded as one occurrence sampled.Substrates were identii ed to avoid recording the same specimens during different i eld expeditions.The following substrate data were recorded for each fungal occurrence: diameter and length (volume of the substrate was then calculated), and the condition [living trunk (LT); dead trunk (DT): standing or fallen logs and stumps; living branch (LB); dead branch (DB); and on the ground (OG)].
According to Urcelay and Robledo (2004), in the present study, it was assumed that there was a gradient in the diameter and volume of the logs and a gradient in the level of decay, ranging from the dead trunks to living in the soil, representing a decay succession.

Data analysis
Estimated richness was calculated with ACE and CHAO1 estimators, with one individual-based frequency sample (all polypore species frequency records across all sampling plots and dates), using the software EstimateS.These estimators were selected because they are nonparametric estimators that do not assume the distribution type of the dataset, do not i t a particular model and are based on abundance data (Version 9.0, viceroy.eeb.uconn.edu/estimates).
The relative frequency of each polypore species in each substrate condition was calculated as , where "n xi " is the number of occurrences of each polypore species "x" in each substrate condition "i" (viz., LT, DT, LB, DB, and OG) and where "Ax" is abundance of polypore species "x" (across all sampling plots and dates).The species richness in each substrate condition was also calculated.
Cluster analyses were performed to classify polypore species into groups as recommended by Goodall (1980).This method is based on distances between species and combines the species computed from their relative frequency in different substrate conditions (Fr xi ) and the average volume of the substrates on which they occurred.The mean Euclidean distance was used as a measure of similarity between species and clustering was based on the average linkage method using the Infostat statistical package (DI RIENZO et al., 2012).Species with two or more occurrences were included in the analysis (URCELAY; ROBLEDO, 2004).
The sampling effort was tested using complementarity (dissimilarity), an empirical measure of the degree of species uniqueness between different samples.This empirical test for complementarity was calculated as samples.This empirical test for complementarity , where "Cjk" is the complementarity between samples (or plots) "j" and "k", "Ujk" is the total number of taxa encountered in only one plot, and "Sjk" is the total richness of both plots combined (COLWELL; CODDINGTON, 1994).Sampling effort for species richness is most efi cient when complementarity between samples (Cjk) is about 0.5 (LODGE et al., 2004).
Community structure and functional diversity of polypores

Community structure (abundance and richness of species)
Community structure based on species abundance showed (Figure 3 Abundance and species richness increased with decay stage of the substrate.Among 152 specimens collected, six (four species) were recorded on LT, 68 (22 species) on DT, 71 (40 species) on DB and seven (i ve species) on the ground (OG) (Figure 4; Table 1).

Community structure (functional groups)
On the basis of the relative frequency of each species on the different log conditions and the mean diameter and volume of the log on which the basidiomes were found, five main groups of polypore species were identiied at 2.30 dendrogram distance of the cluster analysis (Figure 5).Phylloporia spathulata (Hook.)Ryvarden (with relative frequency of 1.00, OG) and Phylloporia chrysita (Berk.)Ryvarden (with relative frequency of 1.00 on LT of extremely small volume) formed two functional groups, Groups 1 and 2, respectively.
Group 3 was formed by Ganoderma australe (Fr.)Pat., Phellinus detonsus (Fr.)Ryvarden and Fuscoporia wahlbergii.These species are able to cause decay and develop basidiomes in standing living trunks, but can also grow as saprophytes and then develop basidiomes on dead trunks.Groups 4 and 5 consisted essentially of saprophytic species on decaying dead wood, and they were differentiated by the relative frequency and the average size of the substrates.Group 4 was formed by seven species that grow preferentially on dead trunks of large size; the largest volumes observed were in this group.Group 5 comprised 18 species that grew exclusively on dead branches of small volume.Within Group 5, two species.Species with two or more occurrences were included in the analysis.

Discussion
This work was based on the occurrence of species through the presence of their basidiomes.We conducted the sampling over four seasons to consider the seasonal variability and to include as many species as possible.Our results and conclusions are conservative, since we underestimated the community by not considering all specimens that not developed basidiomes.However, presence-absence of basidiomes is a method accepted for ecological works (URCELAY; ROBLEDO, 2004).
Species richness observed (58) was near 50% of that estimated (ACE = 114.1 and CHAO1 = 122.2; Figure 1).This represents less than 16% observed and 31% estimated with ACE and 33% estimated with CHAO1 of the ca.365 polypore species reported by Baltazar and Gibertoni (2009) for the Atlantic Forest domain.Sampling effort was enough as suggested by the complementary index (LODGE et al., 2004); abundance and richness (Figure 4; Table 1) values would certainly be higher with more sampling effort (additional plots).However, we believe that dominance (Figure 3) and functional structures (Figure 5) would not change substantially.From a practical point of view, the sample design of this study was enough to get a preliminary overview of the community structure.Species richness and abundance increased as the decay stage increased (Figure 4).This is in accordance with previous results (URCELAY; ROBLEDO, 2004) and the theoretical frame of decay succession (RAYNER; TODD, 1979).
Regarding the functional structure of the community, our results agree with those of Urcelay and Robledo (2004).We could identify two additional functional groups (Groups 1 and 2), both constituted by Phylloporia species, considered rare for the study area (Figure 3).Phylloporia species are parasites of roots in the ground, living branches and living stems of creepers (DECOCK et al., 2013).This genus is not present in the Alnus acumniata mountain forests where polypore functional groups were originally proposed (ROBLEDO et al., 2003;URCELAY;ROBLEDO, 2004).Although these groups are constituted by one single species each one, in the Atlantic Forest, there are more species that would have the same substrate preference and that consequently would integrate these groups.The same group of Phylloporia spathulata could include Amauroderma schomburgkii (Mont.& Berk.)Torrend, Amauroderma sprucei (Pat.)Torrend, Amauroderma sp. and Coltricia barbata Ryvarden & de Meijer, also recorded in the study area, and usually considered parasitic on roots of living trees (FURTADO, 1981;RYVARDEN, 2004).Groups 3-5 identiied in this study agree with those proposed by Urcelay and Robledo (2004), however with a different species composition.Another interesting aspect, in accordance with those of Urcelay and Robledo (2004), is that each group was composed of at least one dominant, few subordinate and several rare species.Our results suggest that communities of polypores would have this dominance/functional structure regardless of the fungal species arrangement and the structural complexity of the ecosystem (i.e.diversity and abundance of woody substrates).
Ecological studies describing the community structure of the polypores, as presented here, are extremely important, in particular in regard to the current conservation issues that threaten the forest remnants.Particular substrates and/or key functional species could be identiied, and then could be considered when conservation decisions are taken.

FIGURE 1 :
FIGURE 1: Study area: (A) PARNA Serra do Itajaí in Santa Catarina State, with study area outlined (the square represents the Parque Natural Nascentes do Garcia enlarged in B); (B) Parque Natural Nascentes do Garcia showing the positions of plots.

FIGURE 2 :
FIGURE 2: Observed S(est) and estimated richens by CHAO1 and ACE estimators.Estimators show different values.

FIGURE 3 :
FIGURE 3: Abundance of species of the PARNA Serra do Itajaí: D. dominant species; S. subordinate species and R. rare species.speciesand R. rare species.

FIGURE 5 :
FIGURE 5: Cluster analysis of different polypore species of the PARNA Serra do Itajaí based on relative frequency of the species in each substrate condition and the mean volume of the substrates where basidiomes were found.Arrows indicate dominant species.Species with two or more occurrences were included in the analysis.