Distinct lichen community in riparian forests along an anthropogenic disturbance gradient in Southern Brazil

Abstract The riparian forest sustains an expressive richness and diversity of species and anthropogenic impacts in certain sites have caused changes in the structure of the communities. This study aimed to analyze the composition and structural parameters of the lichen community in riparian forests. The study was carried out in seven sites surrounded by different matrices: rural, urban and industrial. The lichens were mapped using the acetate method and the composition and phytosociological parameters were analyzed. A total of 208 species were identified. The riparian forest sites of the rural matrix differed from the others by the dominance of the morphological forms crustose and squamulose, as well as the predominance of species from humid and shaded environments. Phyllopsora lividocarpa, Phyllopsora parvifolia and Herpothallon minimum presented the highest importance values in the forest sites. The lichen community presented greater homogeneity in riparian forest sites of the urban-industrial matrix. Modifications in the species’ composition and structural parameters of the lichen community demonstrated a gradient of disturbances in the different matrices. Preservation actions of riparian forests are essential for the conservation of the species and landscape connectivity since they act as an important reservoir of biodiversity in sites of subtropical watersheds.


Introduction
Riparian forests maintain and preserve water courses, soil, and biodiversity (Ribeiro-Filho et al. 2009).These forest areas present high levels of biological diversity and productivity (Bennet & Simon 2004) and are responsible for housing highly adapted species in addition to generalist taxa (Gundersen et al. 2010).Anthropogenic disturbances in these sites cause habitat loss and changes in the composition of communities, which may lead to the extinction of species (Aragón et al. 2019).The knowledge of biodiversity, ecology and communities is important for the protection of forests.Assessing the quality of forest sites, such as riparian forests, as well as their continuity, human impact or number of threatened species is essential for the conservation of these ecosystems (Resl et al. 2018).Therefore, species composition, richness and diversity are considered good indicators of ecological continuity, reflecting the recent history of disturbance in forest areas (Aragón et al. 2019).
Epiphytic lichens are essential components in tropical forests due to their role in water and nutrient cycling (Benítez et al. 2018).These organisms are considered environmental indicators due to their capacity to demonstrate human and microclimatic disturbances (Chuquimarca et al. 2019;Käffer et al. 2021), helping evaluate ecosystem health (Bartholmess et al. 2004).Additionally, they are considered highly effective in diagnosing the quality of forest environments due to their sensitivity to the substrate (Tripp et al. 2019).Lichens can be used to assess the successional stage of forests (Koch et al. 2013) by demonstrating whether the ecosystem has remained unchanged over time (Aragón et al. 2019).Lücking et al. (2009) estimated approximately 7,000 lichen species for the tropical region and considered the knowledge of the species richness of tropical taxa and their taxonomy to be important to other research areas.The same author reports that the high richness in the tropics is attributed to different variables, including the absence of snow and the good conditions for photosynthesis.Recent studies in the South region of Brazil reported 107 new records for the state of Rio Grande do Sul (Aptroot et al. 2021(Aptroot et al. , 2022)).
In the last 25 years, forest areas had a net loss of 129 million ha, especially in developing tropical countries, where forests are used for agricultural purposes (Ripple et al. 2017).The Hydrographic Basin of Sinos River (HBSR) located in southern Brazil, is considered one of the most polluted in the country and is located in the Atlantic Forest Biome, which constitutes one of the 36 biodiversity hotspots in the world (<https://www.conservation.org>).The forest areas inserted in the HBSR region have been altered over the years, presenting different degrees of anthropogenic disturbance and causing direct changes in the ecosystem's dynamics.Specifically, past studies in this Basin found composition change and richness reduction in fern and lycophyte species, as well as in vascular epiphytes (Rocha-Uriartt et al. 2016), over a gradient of urbanization.The knowledge about the diversity and structure of the lichen community in the Atlantic Forest is of great importance for ecosystem conservation studies (Cáceres et al. 2016).Studies show that changes in the composition and richness of lichenized mycota in tropical forests are due to the influence of agricultural activities and urbanization (Chuquimarca et al. 2019;Koch et al. 2019).Accordingly, the determination of the composition, diversity and structure of the lichen community can be used as a diagnostic model of environmental quality in an urbanization gradient.On the other hand, the phytosociological approach of the lichen community as a tool in the analysis of forest areas in tropical and subtropical environments is still incipient (Leite et al. 2015;Käffer et al. 2015).Thus, we hypothesize that the structural parameters, diversity and composition of the lichen community are different in forest sites according to the matrices in which they are inserted, and we assume that there will be: (1) reduction of species richness in the rural-urban-industrial matrices; (2) modification in the structural parameters of the lichen community with a predominance of characteristic taxa in the urban-industrial matrices, and (3) homogenization of the lichen community in the most urbanized and industrialized environments.Therefore, the present study aims to: (1) verify possible differences in the composition and diversity between forest sites (2) analyze the composition of the lichen community occurring in different riparian forest sites, and (3) assess the phytosociological parameters of the corticicolous lichen community in riparian forest sites in different matrices in southern Brazil.

Study sites
The Hydrographic Basin of Sinos River (HBSR) is in the northeastern region of Rio Grande do Sul, Brazil.It covers an area of 3,746.68 km², encompassing 32 municipalities with a total population estimated at around 1,249,100 inhabitants (Sema 2019).The HBSR is inserted in the Atlantic Forest Biome.The basin's vegetation cover is reduced, although forest patches remain predominantly at the sources of the Sinos River and the streams that form it.The increasing urbanization associated with population growth caused several impacts on the HBSR, especially the decrease in the vegetation cover (Prosinos 2011).
The HBSR forest sites are characterized by several tropical species.The forest sites analyzed in this study correspond to two formations.Semideciduous Seasonal Forest, comprising the most extensive formation of the basin, with an amplitude of altitudinal distribution from 12 to 600 m, is characterized by generally less density and humidity with a less diverse and open understory.The Ombrophilous Forest occurs in an altitudinal amplitude ranging from 600 to 905 m and presents a humid habitat with dense understory, in addition to a herbaceous layer dominated by ferns and trunks covered by epiphytes.The families Myrtaceae, Meliaceae, Lauraceae, and Salicaceae predominate in number of individuals.This formation concentrates some of the largest and most preserved stretches of continuous forest in the basin, including the Caraá Environmental Protection Area, which houses one of the springs of the HBSR main river (Molz et al. 2016).The climate is classified as humid subtropical (cfa) with well-defined summers and winters, as well as rains throughout the year (Peel et al. 2007).

Selection of forest sites
The riparian forest sites (S1 to S7) were selected along the HBSR (from the source towards the mouth of the main river) in seven municipalities distributed into three different environmental matrices (rural, rural-urban, and urban-industrial).Levels of urbanization were considered for each matrix (Tab.1; Fig. 1).Each riparian forest site was selected based on its availability in the three different environmental matrices.All forests had an area of at least 1 ha and the phorophytes were 5 m to 15 m apart.

Sampling and identification
The lichen community was sampled using the acetate sheet method, which consists of placing five acetate sheets (20 × 20 cm) in sequence along the trunk from a height of 100 cm to 180 cm on the north and south sides of the trees (cardinal directions determined by a compass) (Fig. 2).The contour of the species' thalli was drawn with a pen (Käffer et al. 2015)    the species' coverage.For lichen community sampling, 10 phorophytes were randomly chosen in each riparian forest site, totaling 70 trees.The phorophytes were selected according to the following characteristics: diameter at breast height (DBH) ≥ 12.0 cm, similar trunk structure and no branches below 2.0 m from the ground (Tab.S1, available on supplementary material <https://doi.org/10.6084/m9.figshare.22626970.v1>).Lichen species that could not be identified in the field were collected for later identification in the Botany Laboratory of the Feevale University, RS, Brazil.To identify the species, stereoscopic and optical microscopes, chemical tests on the cortex, medulla and/or reproductive structures, taxonomic keys, and consultation of herbarium material were used, as well as confirmation by specialists in the lichen groups.The species classification followed Lücking et al. (2017) and the nomenclature was consulted in the Index Fungorum.More representative samples of the collected material were deposited in the Anchieta Herbarium (PACA118285 -118500), São Leopoldo, RS, with duplicates deposited in the didactic collection of the Botany Laboratory of the Feevale University.

Data analysis
For the analysis of the composition of the lichen community, all species recorded in the sampling area (acetate sheet method) were considered, as well as the collections carried out tree trunks located on the access trails to the forest sites (additional samples).
The Lichen Diversity Index (LDI) was calculated for the lichen communities in each riparian forest site.To calculate the LDI, the sum of the frequency of all species occurring on both sides of the trunk (N/S) was used for the 10 phorophytes analyzed in each forest site (Asta et al. 2002a adapted).To classify the diversity, the LDI scale proposed by Asta et al. (2002b) was used.This scale goes from 0 to 100, where the values of 0-20 = very low; 20.1-40 = low; 40.1-60 = moderate; 60.1-80 = high and 80.1-100 = very high.Cluster analysis was performed using Sørensen distance and group average (set to -0.25) as a clustering algorithm.Indicator species analysis and Monte-Carlo test were performed on the frequency (species occurring in more than two riparian forest sites) and abundance data of the lichen species (McCune et al. 2002).The analysis was performed in the program PC-Ord 6.08 (McCune & Mefford 2011).
The phytosociological parameters of the community (number of thalli, richness, frequency, coverage, and importance value) were calculated for each riparian forest site using only the data obtained by sampling the species using the acetate sheet method.Richness considered the species present on both sides (N and S) and the 10 analyzed phorophytes in each site.Frequency was calculated using the total number of occurrences of the species on both sides (N and S) and in the 10 analyzed phorophytes in each riparian forest site.Coverage (CA) was estimated by the sum of all the lichen thalli present on both sides (N and S) and the 10 sampled phorophytes for each riparian forest site.For the calculation of absolute coverage (CA), the acetate sheets were superimposed on a sheet with 100 squares of 2 cm each.The relative coverage (CR%) was calculated considering the value of the species' coverage divided by the total sum of the community's CA and multiplied by 100.The absolute frequency (FA) was estimated by the number of occurrences of each species and divided by the total number of phorophytes (10), while the relative frequency (FR%) was calculated considering the FA of each species divided by the total sum of the FA of the community and multiplied by 100.The Importance Value (IV) of each species was calculated considering the sum of the frequency and relative coverage data (Mueller-Dombois & Ellenberg 1974 adapted from lichen community).
To evaluate the relationship between phytosociological parameters of the lichen community (number of thalli, coverage, and importance value) and the different matrices (ruralurban-industrial), a one-way analysis of variance (ANOVA) was used.The analyses were performed using the software SPSS Statistics 2.0.
The lichen diversity index (LDI) ranged from 32.8 to 11.9, with low values recorded for rural matrix (S6 and S7) and a very low LDI for an urban-industrial matrix (S3) (Fig. 4).
In the cluster analysis, we verified a greater similarity between two sites from the urbanindustrial matrix and a closer relationship between three sites of the urban-industrial, rural-urban and rural matrices, while the other rural-urban site appeared distant from the other sites (Fig. 5).The two sites of the urban-industrial matrix were very similar, with more than 50% of the species of each forest site (S1 = 77.3% and S2 = 56.2%)belonging to the families Parmeliaceae, Graphidaceae and Physciaceae, which presented the largest number of representatives in this study.The forest sites of the urban-industrial, rural-urban e rural matrix were characterized by the presence of typical species

Phytosociological parameters
The riparian forest sites of the rural matrix differed from the others by the dominance of the morphological forms crustose with perithecia and squamulose.Nevertheless, the highest richness was recorded in the riparian forest site (S2) of the urban-industrial matrix, while the lowest richness was observed in the riparian forest site of the rural-urban matrix (S5), contrary to expectations.However, the greater richness in the urbanindustrial matrix is associated with a high incidence of species of the family Graphidaceae, especially of the genus Graphis, and Parmeliaceae of the genus Parmotrema, which are prevalent in urbanized areas and areas with more solar incidence.In a forest site of the rural-urban matrix (S5), the most representative genera were Porina, Leptogium and Ramalina.
The species with the highest importance value (IV) presents expressive frequency and coverage in riparian forest sites.In the sites of the urban-industrial matrix, five species presented  Differences were recorded between the forest sites of the urban-industrial and rural-urban matrices for the parameters number of thalli (ANOVA, F = 10.21,p < 0.001) and importance value (F = 10.40,p < 0.001); the same parameters were also different between the urban-industrial and urban matrices (number of thalli: F = 5.12, p < 0.001; importance value: F = 7.07, p < 0.001).For the forest sites of the rural matrix, significant differences were found for the parameters coverage (F = 9.92, p < 0.001) and importance value (F = 12.59, p < 0.001) in relation to forest sites of the rural-urban matrix.

Discussion
The differences in structural parameters and species composition associated with homogenization of the lichenized community in forest sites inserted in urban-industrial matrices, together with the predominance of taxa characteristic of these regions, corroborated our initial hypothesis.
In this study, the species of the families Parmeliaceae (Parmotrema), Physciaceae (Physcia) and Graphidaceae, with Graphis being the predominant genus, were dominant in the riparian forests of the urban-industrial matrix.These families are characterized as cosmopolitan (Galloway 2008;Thell et al. 2012) and dominant in tropical regions (Rivas-Plata et al. 2012).
In the forest areas of rural-urban and rural matrices, the species of Phyllopsora, Malmidea, Porina and Herpothallon stood out with their higher importance values, causing differences in the composition of the lichen community, especially in relation to the forest areas of the urban-industrial matrix.Species of the genus Phyllopsora prefer partially shaded habitats and are more restricted to riparian forests, occurring on tree trunks of tropical lowland forests (Timdal 2008).Malmidea and Porina are characteristic of tropical regions, and species of the family Porinaceae are dominant in the lichen community of tropical forests (Kalb et al. 2011).Species of the genus Herpothallon are common in shaded tropical forests, growing on wet bark or bryophytes (Aptroot et al. 2009).Biological characteristics of the species and the forest structure of riparian forest sites, especially concerning humidity and shading, may be related to the predominance of these taxa.As for the higher representativeness of the abovementioned families, they predominated in the studies related to lichenized mycota in forest and/or urban environments for subtropical regions (Käffer et al. 2015;Koch et al. 2016;Lucheta et al. 2018Lucheta et al. , 2019)).
Lichens are affected by deforestation and forest exploitation and consequently, the conversion of forest sites by tree extraction changes microclimatic conditions, such as luminosity and humidity (Aragón et al. 2019), especially by changes in canopy coverage, and affects the composition of lichen species (Benítez et al. 2018;Soto-Medina et al. 2019).According to Lakatos et al. (2006), some crustose lichens, mainly in tropical forests, have adaptations to repel water because they have a dense layer of flattened hyphae or projections of hyphae from the medulla forming a hydrophobic layer.Thus, this layer could help them tolerate excessive humidity in certain locations.The riparian forests present heterogeneity in the composition and structuring of the species (Ribeiro-   et al. 2009), and many environmental aspects are changing globally due to human activities, including changes in climatic variables, which directly affect the lichen community.Studies with vascular epiphytes in the HBSR area have verified changes in richness and diversity of the riparian forest sites in different matrices and have emphasized the importance of their conservation and preservation (Rocha-Uriartt et al. 2016).In this study, we found that the preservation actions associated with agropastoral practices were decisive for the conservation of habitats, such as in the S4 forest site of the rural-urban matrix, in which rare species were found, including Herpothallon pustulatum (crustose) and the genera Ricasolia and Sticta, which are known to occur in more preserved environments (Käffer & Martins 2014;López et al. 2016;Lehnen et al. 2017).
The decrease in the richness of lichen species on isolated trees in squares and parks inserted in rural sites towards urban-industrial areas of the HBSR was also observed by Lucheta et al. (2018) and differentiation in the environmental quality of forest areas of the HBSR for lichen communities were related especially to luminosity and altitude (Käffer et al. 2021).Furthermore, structural parameters and morphological characteristics can certainly be applied as indicators in the analysis of the effects of land use on riparian ecosystems (Chuquimarca et al. 2019).
In this study, changes in species composition, associated with changes in landscape structure, demonstrated the gradient of disturbances in riparian forest sites in different matrices.Thus, the maintenance of the subtropical riparian forests preserves the heterogeneity in the composition of the species of the HBSR.
Modifications in the structure of riparian forest sites (fragmentation) associated with microclimate change contributed to the results.Strategies for conservation and preservation of riparian forest sites are essential for the balance and maintenance of the biota in general, but especially of rare species, and improve the connectivity of the landscape with organisms, acting as important reservoirs of biodiversity, such as in sites of subtropical watersheds.

Figure 2 -
Figure 2 -Schematic representation of the method used to sample the lichen community in the phorophytes analyzed in the riparian forest sites of the hydrographic basin.The rectangles represent the acetate sheets (20 × 20 cm) arranged at a height of 100 cm to 180 cm on both sides of the trunk (north and south).

Figure 5 -
Figure 5 -Cluster analysis considering the composition of species in the riparian forest sites of the Hydrographic Basin of Sinos River, southern Brazil.(Urb/Ind = Urban-Industrial; Rur/Urb = Rural-Urban; Rur = Rural).

Filho
and later used to calculate

Table 1 -
Location of forest sites (from the source to the mouth) located in the Hydrographic Basin of Sinos River.Brazil.