Introduction

Anthropogenic influences on vegetation structure can substantially alter the communities of lichens (Will-Wolf et al. 2002; Goward and Spribille 2005; Martinez et al. 2006; Wolseley et al. 2006). This is because various lichen communities are susceptible to local species loss due to disturbance (Pharo et al. 2000; Uliczka and Angelstam 2000; Nordén and Appelqvist 2001). Additionally, lichen communities can also become dominated by species that are resistant to the effects of disruption or pollution (Jüriado et al. 2003; Humphrey et al. 2002; Will-Wolf et al. 2002; Brunialti and Giordani 2003; Nascimbene et al. 2007).

The conversion of native forest to tree plantations can influence lichen diversity (Humphrey et al. 2002; Wolseley et al. 2006). This is expected because community composition and diversity of epiphytic lichens can be determined by its host-trees characteristics (Jesberger and Sheard 1973; Lõhmus et al. 2007). Trees with bark that suffer shredding do not favor lichen colonization (Brodo 1973). Tree barks that produce resins and saps can also inhibit lichen growth (Topham 1977). The acidity and alkalinity of the bark can affect the lichen thallus, because pH can be critical for the reproduction of various species (Hale 1957; Brodo 1973). When forest environments are altered or transformed into different arboreal vegetation types, the availability of favorable host-trees can diminish, promoting changes in the lichen composition and decreases in species’ diversity (Sillet et al. 2000; Hilmo and Sastad 2001; Kantvilas and Jarman 2004; Pykälä 2004).

Another important aspect of lichen community structure is the lichen vertical distribution along the host trunk. Changes in temperature and humidity are key factors defining spatial distribution of lichens along tree trunks (Harris 1971; Lang et al. 1980). Tree plantations present, in general, low microhabitat heterogeneity in respect to light penetration, and vegetation cover due to poor canopy stratification. Therefore, conversion of native forests to tree plantations can modify physical characteristics that may alter lichen vertical distribution.

The Araucaria forest from southern Brazil is a subtropical rain forest that contains around 222 species of lichens, 200 of them occurring on tree trunks (Spielmann 2006). This forest has been extensively damaged or converted into plantations of exotic tree species, especially from the genera Pinus and Eucalyptus. The impact of these forest transformations on lichen communities is still unknown. In the present study we evaluate how the corticolous and foliose lichen community respond to vegetation type alterations such as the conversion of native forest to plantations of tree monocultures. Our specific aims are: (1) to determine if the richness, abundance, diversity and species composition of lichen communities differ between tree plantations and native forest stands; (2) to investigate if there are differences between vegetation types in the way lichen richness is distributed along trunk heights; (3) to determine possible differences in structural characteristics and/or lichen strategies that could explain lichen community distribution throughout vegetation types.

Methods

Study area

This study was undertaken at the National Forest of São Francisco de Paula (FLONA), southern Brazil (29° 02′ S and 50° 23′ W). The reserve lies over a basaltic basin formed during Mesozoic lava extrusion events. The region is located 912 m above see level, with an average temperature of 14.5°C and average rainfall of 2,252 mm year−1. The predominant vegetation type is Araucaria forest, characterized by the dominance of the species Araucaria angustifolia (Bert.) O. Ktze. (Araucariaceae, “Brazilian-pine”). The FLONA encompasses an area of 1,607 ha, which contains monoculture stands of tree species from the genera Araucaria, Pinus and Eucalyptus. This National Forest is classified as a Conservation Unit for Sustainable Use. Therefore, these stands differ from classical tree monoculture plantations in age (being older), in size (being smaller), and by surroundings (being of native forest). Additionally, disturbance of understory regeneration in these areas due to events of thinning occurred less frequently than in commercial tree plantations. All this “ecologically optimum management” provides an excellent scenario for the establishment of new lichen species in these tree plantation stands.

Sampling

The sampling was carried out in three stands of each vegetation type: native Araucaria forest (FO), Araucaria plantation (PA), Pine plantation (PP), and Eucalyptus plantation (PE); 12 stands total. Each stand was selected based on its availability and accessibility, but all stands were at least 1 ha in area and at least 100 m apart. For Eucalyptus plantations only three stands of minimum size of 1 ha were available. There was no standardization for stand distance from native forest but this vegetation type was reasonably spread around stands (Fig. 1). The three native Araucaria forest stands studied had been recorded as undisturbed forest until the end of the 19th century. Selective logging occurred in these stands from 1900 to 1950 and after that all three stands were left undisturbed. All plantation stands were established on old agricultural fields or pastures. The Araucaria plantations were between 44 and 56 years old, the Pine plantations between 35 and 38 years old and the Eucalyptus plantations between 15 and 31 years old. Different tree species were able to colonize naturally all stands studied from the beginning of their implementation, this management procedure allowed different types of host trees to be present in all plantations.

Fig. 1
figure 1

Map of the study area representing the locations of different vegetation types studied at the National Forest of São Francisco de Paula, southern Brazil

Full size image

In each replicate, ten host—trees were randomly sampled, with 120 host-trees in total. In order to choose host trees we randomized ten points in each 1 ha stand, and the nearest tree (planted or naturally established) that was at least 8 cm in DBH was chosen for lichen sampling. In each one of the stands, for each host-tree, the corticolous and foliose lichens were recorded from 30 to 150 cm above the ground by placing a rubber band around the trunk every 10 cm (Marcelli 1992). All the species that touched the rubber band were identified in the field or collected for posterior identification. Each rubber band was divided into 100 units of equal size, thus the number of rubber band units covered by lichens were used as a measurement of lichen abundance.

The lichen identification was carried out by observing anatomical sections of the thallus and fructifications using stereoscopic and optical microscopes. The external characteristics of the lobes like color and thallus aspect, width and length of the lobes, presence of pycnidia, and the aspect of the rhizines, ciliates, and apothecia were also analyzed. Coloration tests were used to determine the presence of the acid substances in the cortex and medulla. Identified samples were incorporated into the Herbarium Prof. Dr. Alarich Schultz (HAS) of the Zoobotanical Fundation of Rio Grande do Sul State, Brazil.

Each sampled host-tree was identified to species, and its bark type, bark pH, diameter at breast height (DBH), and tree height were recorded. The host-trees were identified in the field, or in the Laboratory with the help of specialists. For each tree species, the following bark types were identified: furrowed, fibrous, and smooth. The tree bark’s pH was determined in the field, in a clean space of the tree trunk, free of lichens and bryophytes using a digital pH measurer model PH—1,700—Instrutherm.

Data analysis

Species richness was considered as the total number of lichen species occurring in the 10 host-trees analyzed in each stand. The lichen abundance was calculated by adding up the number of rubber band units covered by lichens for the 13 height levels and 10 host-trees studied in each stand (13 height levels, 100 units per level, 10 host-plants).

We calculated two types of lichen diversity indexes, one based on cover that has the assumption that large individuals are more abundant and another index based on occurrence that has the assumption that the most abundant individuals are the ones present in the largest number of trees, independent of lichen size. The indexes were calculated using the Shannon–Wiener Diversity Index (Krebs 1999). For the diversity index based on cover, the relative abundance of each species was represented by the number of elastic units in which the species was recorded on each host-tree, divided by the total number of units in which the lichens were recorded. For the diversity index based on occurrence, the relative abundance of each species was represented by the number of host-trees on which the species was present in each stand, this number varied from 1 to 10. Lichen richness, abundance and diversity were compared among vegetation types with an ANCOVA analysis using the sum of the DBHs for each host-tree as a covariant variable to correct for possible differences in sampling effort between stands. Analyses were performed using the General Linear Model procedure of Systat 10 (2000) package.

To investigate possible changes in species similarity between lichen communities present in plantations and native forest stands a cluster analyses based on the relative Sørensen coefficient of dissimilarity as classification method, flexible beta = 0.25 was used as the clustering algorithm (McCune et al. 2002). To understand how the different lichen species influenced differences in community composition between vegetation types, a non-metric multidimensional scaling (NMS) was performed as ordination method. This analysis used the relative Sørensen coefficient of dissimilarity as classification method. The statistical analyses were performed using the PC-ORD 4.0 statistical program (McCune and Mefford 1999).

All lichen species were characterized as shade tolerant or light demanding species according Marcelli et al. (1998) plus 32 published articles on species taxonomic description, as well as based on the authors’ unpublished data. The lichen taxa were classified following their geographical distribution. The specimens listed as new records to science were referred to as having restricted distribution, whereas the others were classified as having tropical distribution (occurring only in tropical regions), temperate distribution (occurring only in temperate regions) and cosmopolitan distribution (spread in tropical and temperate regions).

Lichen species richness was compared between heights and vegetation types, using height, vegetation type, and the vegetation type x height interaction as independent variables. The analysis was performed using the GLM—General Linear Model procedures of SYSTAT 10.

To investigate possible differences between vegetation types in the availability of proper trees for lichen fixation, average measurements of tree DBH and bark pH, as well as measurements of host-tree richness were compared among vegetation types, by ANOVA, using the LSD a posteriori test. The availability of host tree bark types for lichen fixation was compared among vegetation types using a Chi-square test.

Results

Seventy-eight lichenized fungi, from 18 genera, were recorded. Among them, 81% were colonized by chlorophyceans and 19% by cyanobacteria (Table 1). In total, 27 species were found in the native Araucaria forests, 49 species in the Araucaria plantations, 27 species in the Pine plantations, and 30 species in the Eucalyptus plantations.

Table 1 Composition and occurrence of lichen species in four vegetation types in south Brazil
Full size table

Species occurrence

There were differences in species distribution among vegetation types. Of the 44 species that were specific of a single vegetation type, 11 occurred in the native Araucaria forest, 15 occurred only in the Araucaria plantations, 5 only in the Pine plantations and 10 only in the Eucalyptus plantations. Four species occurred in all vegetation types: Heterodermia leucomela (L.) Poelt, Heterodermia obscurata (Nyl.) Trevis, Leptogium azureum (Sw.) Mont. and Parmotrema rampoddense (Nyl.) Hale (Table 1). Around 15% of the species found in the native Araucaria forest occurred also in the tree plantations: Of those species 10 (83.3%) were found in both forest and plantations. One (8.3%) was found in Araucaria forest and Pine plantation and 8.3% in the Eucalyptus plantations. About 25% of those species that were found in the Araucaria plantations were also found in the Pine and Eucalyptus plantations.

Richness, diversity, and abundance

Lichen richness, abundance and diversity tended to be greater in the Araucaria plantations (Fig. 2a–d). However, this pattern was only significant for lichen richness (Table 2, Fig. 2a). The sampling effort (sums of tree DBH) did not show any significant effect on the lichen richness and diversity (Table 2).

Fig. 2
figure 2

Lichen species richness (a) Cover abundance (b), Shannon’s Diversity Index based on lichen cover (c) and Shannon’s Diversity index based on lichen occurrence in host-trees (d). Vegetation types represent: FO native Araucaria forest, PA Araucaria plantation, PP Pine plantation and PE Eucalyptus plantation

Full size image
Table 2 F values and probability (P) for richness, abundance, diversity based on cover, and diversity based on occurrence of lichen species
Full size table

Species composition

Cluster analysis yielded three distinct groups revealing differences in species composition among vegetation types. The vegetation type which showed the most distinct species composition was the Araucaria forest (Fig. 3a). The NMS (non-metric multidimensional scalling) ordination revealed that Araucaria forests formed a distinct group strongly related to shade tolerant species. The analysis produced 6 axes and the percentage of variance explained by the first two axes added up 62.9, 43.5% in the first axis and 19.4% in the second. Axis 1 showed positive correlation with shade tolerant species characteristics of native Araucaria forest such as Phyllopsora confusa Swinsc & Krog, Lobaria tenuis Vain and species of the genus Sticta. The second axis showed negative correlation with light demanding species such as species of the genus Heterodermia and family Parmeliaceae (Fig. 3b).

Fig. 3
figure 3

Cluster of similarity in community composition based on the relative Sørensen coefficient of dissimilarity (a); and ordination of community composition using the Non-Metric Multidimensional Scaling (NMS) method based on the relative Sørensen coefficient of dissimilarity (b). Vegetation types represent: FO native Araucaria forest, PA Araucaria plantation, PP Pine plantation and PE Eucalyptus plantation

Full size image

Distribution of shade and light demanding species

The contingency table showed that there were differences among vegetation types in relation to the abundance of shade and light demanding species. In the native Araucaria forest there was a greater observed frequency of shade tolerant lichens in relation to the expected frequency, however, in the Pine and Eucalyptus vegetation type the inverse was true, a greater percentage of light demanding lichens was encountered (χ 2 = 16.4, df = 3, P = 0.001, Fig. 4). In the Araucaria plantations, however, there was no difference in the occurrence of these two lichen groups. There was no significant relationship between the lichen species’ geographical distribution and the vegetation type they occurred (χ 2 = 0.38, df = 3, P > 0.05), thus, species with wide or restrict distributions occurred in all vegetation types studied.

Fig. 4
figure 4

Frequency of shade tolerant and light demanding lichen species, in four vegetation types: FO native Araucaria forest, PA Araucaria plantation, PP Pine plantation and PE Eucalyptus plantation

Full size image

In the native Araucaria forest, shade tolerant genera such us Sticta, Phyllopsora, and Coenogonium occurred in 66.7% of all trees sampled. The abundance of these shade tolerant genera was also very high in the Araucaria forest, the only vegetation type where Phyllopsora, and Coenogonium were encountered (Table 1). Leptogium is also another genus characteristic of shady and humid environments, however, its species are very adaptable to different vegetation types (Wolseley 1991), especially Leptogium azureum (Sw.) Mont., which was present in the four vegetation types studied. Like most Brazilian vegetation were Parmeliaceae is by far the most important family (Marcelli 1998), all tree plantations support a great number of species from this light-demanding group. This family was quite frequent in our samples and was responsible for 44.8% of the species recorded in all vegetation types. However, only 2.6% of these Parmeliaceae species were reported in the native Araucaria forest at the height levels sampled. The most common genus of this family, Parmotrema encompasses 33.3% of the lichen species sampled in plantations.

Vertical distribution of species richness

In all vegetation types, lichen species richness was significantly larger at the highest trunk height measurements (Table 3, Fig. 5). Richness was consistently greater in Araucaria plantations followed by native Araucaria forests. Nevertheless, the interaction among height and vegetation types was not significant (Table 3, Fig. 5).

Table 3 ANOVA table for lichen richness
Full size table
Fig. 5
figure 5

Distribution of lichen richness in relation to trunk height in four vegetation types at the National Forest of São Francisco de Paula, southern Brazil: FO native Araucaria forest, PA Araucaria plantation, PP Pine plantation and PE Eucalyptus plantation

Full size image

Host-tree composition

In terms of composition and occurrence of the 120 host-trees sampled, the following species presented the highest frequency of individuals: Araucaria angustifolia (Bert.) O. Ktze (23.3%), followed by Pinus spp. (20%) and Eucalyptus spp. (19.2%). Tree species dominance differed among vegetation types. In the native Araucaria forest host-tree dominance was low (Fig. 6a), however, 73.3% of the trees sampled in Araucaria plantations were A. angustifolia trees (Fig. 6b). In the Pine plantations, the Pinus species stood out, representing 80% of the analyzed host-trees (Fig. 6c), and in the Eucalyptus plantation, the Eucalyptus species represented 76.7% of the sampled host-trees (Fig. 6d).

Fig. 6
figure 6

Host-tree frequency distribution sampled in four vegetation types at the National Forest of São Francisco de Paula, southern Brazil: FO native Araucaria forest (a), PA Araucaria plantation (b), PP Pine plantation (c) and PE Eucalyptus plantation (d)

Full size image

There were differences among vegetation types for various host-tree characteristics such us DBH, bark pH and bark structure (Table 4). In terms of bark structure, there were significant differences in their occurrence among vegetation types (χ 2 = 24.51, df = 6, P < 0.01). Fissured barks occurred at a higher frequency than expected in the native Araucaria forest (15.8%) and in the Pine plantation (11.7%), being less frequent in the Araucaria and Eucalyptus plantations. There was a higher frequency of smooth bark than expected in the Eucalyptus plantation (1.7%), compared to the other vegetation types. Furrowed barks were more frequently found in the Araucaria (20.8%) and Eucalyptus (18.3%) plantations while in the Pine plantation and in the native Araucaria forest, this frequency was lower than expected. For all tree species sampled DBH varied between 8 and 63.7 cm. DBH of sampled host trees was significantly higher in the Pine plantations (F = 12.65, df = 3,8, P < 0.05). For all vegetation types bark pH varied between 5.9 and 8.7. Eucalyptus species presented the lowest values of bark pH (Table 4), and host-tree bark pH was on average, lower in the Eucalyptus plantation (F = 18.62, df = 3,8, P < 0.05).

Table 4 Host-tree characteristics and occurrence in four vegetation types at the National Forest of São Francisco de Paula
Full size table

Discussion

Various lichen species of native Araucaria forest did not find difficulties to establish in Araucaria, Pine, and Eucalyptus plantations. This pattern was probably due to the management procedures applied in these plantations. All monoculture stands were surrounded by native forest, had smaller size and were allowed to last longer than commercially managed plantations. These factors improved the chances of species migration from the forests to monoculture stands (Sillet et al. 2000; Pykälä 2004). These findings highlight the importance of maintaining stands that are “ecologically managed” in commercial tree plantations (Fonseca et al. 2009). Such stands could function as corridors promoting establishment, reproduction and lichen species translocation between forest areas.

Conversion of native forest to plantations, nonetheless, altered lichen community composition. The greatest occurrence of shade tolerant species was recorded in the native Araucaria forest while an increase in light demanding lichen species was found in tree plantations. These results may be explained by the well-established fact that disturbance has a negative effect on shade species from forest understorey whereas canopy species tend to increase in abundance. This pattern has been shown for many organisms including bryophytes and insects (Lesica et al. 1991; Kantvilas and Jarman 2004; Gunnarsson et al. 2004; Martinez et al. 2006; Lõhmus et al. 2007), and has been widely registered for lichens from neotropical regions (Lücking 1999; Sipman 1997, 2006; Komposch and Hafellner 2000, 2003; Cáceres et al. 2008; Rivas Plata et al. 2008). Our results show that few light demanding species of Parmeliaceae were reported for the native Araucaria forest at the height levels sampled. However, species of this family are dominant in the canopy of humid tropical forests (Sipman and Harris 1989; Fleig and Grüninger 2008). Indeed, various light demanding species were encountered on branches that fell from the native Araucaria forest canopy (Käffer unpublished data). These results suggest that this native Araucaria forest may be able to maintain a greater diversity of light demanding lichen species at its upper strata and that these species may have been able to establish themselves in the understory of plantations due to the greater light penetration therein. This study emphasizes that converting native forests into exotic tree plantations may not affect light demanding lichen species but can cause the loss of species typical to shady and humid environments. This pattern occurs even when plantations are surrounded by native forest and less intensively managed.

Lichen community structure in terms of species richness, and diversity was also affected by the conversion of native forest to monoculture plantations. The greatest diversity of lichens was encountered in the Araucaria plantations when compared to all other vegetation types studied. The more advanced age of these stands, compared to the exotic tree plantations, may partially explain these results. Nonetheless, Araucaria plantations were younger than native Araucaria forest stands, and still sustained greater lichen diversity in its understorey. In Araucaria plantations, we found the largest frequency of A. angustifolia a native host-tree. Araucaria has a wide range of lichen species adapted to it and is an excellent host-tree for lichen establishment, because the fibrous structure of its bark is able to retain water thus increasing bark humidity levels. Studies undertaken in woody vegetation confirm that alterations in lichen community structure could be attributed to host-tree composition and their bark characteristics (Hawksworth 1975; Hale 1983; Ferry and Lodge 1996). Therefore, a greater diversity of lichens can be preserved in the landscape, if plantations of exotic Pine and Eucalyptus trees are replaced by plantations of the native species A. angustifolia.

Lichen richness increased with trunk height at the same rate for all vegetation types. Changes in trunk height may cause microclimate changes in humidity and light that in general modify lichen species composition and its vertical distribution (Harris 1972; Lang et al. 1980; Hilmo and Sastad 2001). With increases in trunk height, light becomes increasingly available and new species became better able to colonize tree trunks. These results demonstrate that light can be a limiting factor for lichen establishment in the understorey of all vegetation types studied.

Our results show that the conversion of rain forests to monoculture plantations of exotic tree species decreases lichen species diversity and alters lichen community composition. Even when the best options of management for conserving biodiversity are applied, exotic tree plantations suffer from losses of lichen species that reach up to 42% of the species available for colonization. We conclude that the use of native tree species with bark structure favorable to lichen establishment, such as A. angustifolia, would be the best option for the maintenance of lichen diversity in commercially altered landscapes. The large scale implementation of native tree vegetation type may not only lead to increases in lichen diversity but can also favor the great variety of organisms that use lichens for food and shelter.