Altered herb assemblages in fragments of the Brazilian Atlantic forest

Altered herb assemblages in fragments of the Brazilian Atlantic forest

Biological Conservation 191 (2015) 588–595 Contents lists available at ScienceDirect Biological Conservation journal homepage:

830KB Sizes 12 Downloads 116 Views

Biological Conservation 191 (2015) 588–595

Contents lists available at ScienceDirect

Biological Conservation journal homepage:

Altered herb assemblages in fragments of the Brazilian Atlantic forest Patrícia B. Lima a, Liliane F. Lima a, Bráulio A. Santos b, Marcelo Tabarelli c,⁎, Carmen S. Zickel d a

Programa de Pós-Graduação em Botânica, Universidade Federal Rural de Pernambuco, Recife PE 52171-900, Brazil Departamento de Sistemática e Ecologia, Universidade Federal da Paraíba, João Pessoa PB 58051-900, Brazil c Departamento de Botânica, Universidade Federal de Pernambuco, Recife PE 50670-901, Brazil d Departamento de Biologia, Área de Botânica, Universidade Federal Rural de Pernambuco, Recife PE 52171-900, Brazil b

a r t i c l e

i n f o

Article history: Received 17 April 2015 Received in revised form 3 August 2015 Accepted 5 August 2015 Available online xxxx Keywords: Disturbance Habitat fragmentation Herb species Human-modified landscape Physical condition Tropical forest

a b s t r a c t Understanding the response of tropical plant communities to human disturbance is critical for conserving biodiversity in a changing world. Here we examine the shifts experienced by understory herb assemblages while inhabiting small forest fragments in a fragmented Atlantic forest landscape to infer about community-level shifts imposed by either habitat loss or fragmentation. We established 100 25-m2 plots, placed randomly in 10 forest fragments and 10 forest interior patches, in which all herb species were recorded, litter accumulation, soil temperature and moisture were estimated. We recorded a total of 6027 herbs belonging to 134 species, with a predominance of ferns, grasses, aroids, sedges and calatheas. Forest fragments and forest interior exhibited similar densities of herbs: 64.4 ± 57.8 herbs/25 m2 vs. 56.1 ± 44.1, respectively. Species richness was reduced by a half in forest fragments at plot and habitat spatial scales. Fragments were particularly impoverished in terms of ferns, aroids and calatheas, but supported a subset of proliferating native herbs and indicator/exclusive species; i.e. a taxonomically and ecologically distinct herb flora. Fragments also supported less humid soils covered by a thicker litter layer and these attributes correlated to species distributions in both forest habitats. Our results suggest that habitat loss and fragmentation, particularly the establishment of illuminated and desiccated forest edges, result in the extirpation of particular ecological groups with a few species/ecological groups experiencing proliferation, such as light-demanding species. Collectively, these processes result in impoverished/altered assemblages at multiple spatial scales, potentially limiting the conservation services provided by humanmodified landscapes. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction Understory herbaceous plants, or ground herbs (sensu Poulsen, 1996), comprise an ecologically and taxonomically diverse group in both tropical and temperate forests (Costa et al., 2005; Gillian, 2007; Poulsen and Balslev, 1991; Whigham, 2004). In fact, this group includes a wide spectrum of life forms, or life-history strategies, including (1) ferns, monocots and dicots, (2) annual and perennial species, and (3) proto and facultative terrestrial, scandents and climbers (Cicuzza et al., 2013; Poulsen and Balslev, 1991). Even in tropical forests, in which tree species account for the majority of plant species diversity, understory herbs can represent 10–20% of local plant species richness (Cicuzza et al., 2013; Linares-Palomino et al., 2009 for a review). Herbs also contribute to nutrient uptake and cycling (Lalji and Singh, 1993; Pande, 2004), ecosystem productivity and are vital for the maintenance of many animal populations, such as herbivores, pollinators and frugivores that inhabit forest understorey, including ants, birds, and small mammals (Gillian, 2007; Richards, 1996; Whigham, 2004). Understory herbs are recognized to provide particular microhabitats (e.g. shaded ⁎ Corresponding author. E-mail address: [email protected] (M. Tabarelli). 0006-3207/© 2015 Elsevier Ltd. All rights reserved.

patches) and enhance habitat heterogeneity (Costa et al., 2005; Maraschin-Silva et al., 2009; Whigham, 2004); e.g. dense patches dominated by palmettos, calatheas and Zingiberaceae species (Cicuzza et al., 2013; Richards, 1996). In tropical forests, habitat loss and fragmentation have been considered major sources disturbance (Melo et al., 2013), with most studies on plant responses to these threats focusing on tree species, because of their remarkable role as ecosystem engineers that characterize forest ecosystems (see Laurance et al., 2002, 2011). Tree species may be negatively affected by (1) habitat desiccation imposed by edge-effects (e.g. increased wind turbulence and luminosity), (2) altered soil conditions (e.g. a thicker litter layer), (3) disruptions in plant–animal interactions such as herbivory, seed dispersal and pollination, (4) competition with light-demanding plant species (i.e. winner species sensu Tabarelli et al., 2012) and (5) increased incidence of diseases (Laurance et al., 2006b; Tabarelli et al., 2010b; Vasconcelos and Luizão, 2004). Particularly in the case of the Amazonian and Atlantic forests, tree species bearing large seeds with supra-annual reproduction, those depending on specialized pollination vectors and with a large stature may experience population collapse (Laurance et al., 2006a; Lopes et al., 2009; Oliveira et al., 2008), resulting in impoverished tree assemblages at multiple spatial scales (Laurance et al., 2006b; Lôbo et al., 2011). In contrast to

P.B. Lima et al. / Biological Conservation 191 (2015) 588–595

the comprehensive picture provided for tree species, little information has been offered for herb species, particularly in tropical forests. Like tree species, many ecological groups of herb species may be negatively affected by conversion of natural landscapes into relictual landscapes dominated by edge-affected habitats, such as those herb species experiencing reduced seed germination and seedling survivorship and increased incidence of foliar diseases (Bruna, 1999; Bruna et al., 2005; Santos and Benítez-Malvido, 2012). The Brazilian Atlantic forest is one of the most important global biodiversity hotspots, which includes a diverse herb flora (Giulietti et al., 2005; Myers et al., 2000). Originally, it covered around 150 million ha, but recent estimations indicate that less than 16% of the forest remains (Ribeiro et al., 2009). In addition to being poorly protected (nature reserves only account for 1% of the original forest), the remaining forest cover is distributed in ca. 250 000 forest fragments, 80% of which are smaller than 50 ha and the average distance between fragments is ca. 1500 m (Ribeiro et al., 2009). Furthermore, almost half of the remaining vegetation is less than 100 m from the nearest edge (Ribeiro et al., 2009). Certainly, such human-modified landscapes offer an interesting opportunity to examine the potential effects of habitat loss and fragmentation on herb assemblages. Here we examine the shifts experienced by understory herb assemblages while inhabiting small forest fragments in a typical Atlantic forest landscape dominated by edge-affected habitats in order to infer about community-level shifts imposed by either habitat loss or fragmentation. Accordingly, herb assemblages inhabiting fragments and forest interior patches were described and compared in terms of structure (plant density and species richness), taxonomic/ecological composition and the presence of indicator species. We also investigated the influence of some patch metrics and soil conditions on these community-level attributes. We expected that herb assemblage shifts were congruent with


those exhibited by other plant groups such as trees; i.e. impoverished and taxonomic/ecologically distinct assemblages in response to edge creation (see Laurance et al., 2006a,b; Santos et al., 2008). 2. Materials and methods 2.1. Study landscape The study was carried out at Usina Serra Grande, owned by a large, private sugar company of the same name located in the state of Alagoas, northeastern Brazil (8°30′S, 35°50′W; Fig. 1). Information on the climate, soil, fauna and flora of this region is available in Santos et al. (2008). This landholding still retains ca. 9000 ha (9.2%) of the forest cover assigned to a unique biogeographic region of the Atlantic forest: the Pernambuco Center of Endemism (Santos et al., 2007). We selected a large (667 km2), severely fragmented landscape containing 109 forest fragments (ranging from 1.7 to 3500 ha), all of which are entirely surrounded by a uniform, stable and inhospitable matrix of sugarcane monoculture. Sugarcane cultivation at Serra Grande dates back to the 19th century, and provides a rare opportunity for Atlantic forest fragmentation studies. The old-growth forest interior areas of Coimbra, the largest fragment (3500 ha) remaining in the landscape, still retain a full complement of plant species typical of vast undisturbed tracts of Atlantic forest, such as large-seeded (Santos et al., 2008). However, populations of large-fruit eating birds and mammals such as guans, chachalacas, toucans, aracaris, cotingas and howler monkeys are declining or have already been extirpated locally and regionally due to overhunting and other human threats (see Chiarello, 1999; Galetti et al., 2006; Melo et al., 2006; Silva and Tabarelli, 2000). To our knowledge, there is no information on the ecology of herbs in the Serra Grande landscape.

Fig. 1. The Serra Grande landscape map indicating small forest fragments and the Coimbra forest (forest interior stands) in which herb assemblages were recorded.


P.B. Lima et al. / Biological Conservation 191 (2015) 588–595

2.2. Herb sampling To assess how herb assemblages change in forest fragments, we sampled herbaceous species in 10 small forest fragments (7.84– 83.63 ha) and the interior areas of Coimbra forest (our control site). Between October 2010 and March 2011 we established 100 5 m × 5 m plots in the landscape: 50 plots randomly drawn from a 175-ha grid located more than 200 m from the margins of Coimbra forest, and five plots randomly drawn from a 2-ha grid located at the geometric center of each of the 10 forest fragments. Plots were at least 50 m apart from each other regardless of their location in the landscape. In each plot, we collected and identified all herbs, i.e. all non-woody plants with green stem (Gonçalves and Lorenzi, 2011), which includes climbing herb and saprophyte, ferns, and excludes woody seedlings of trees and lianas (Inácio and Jarenkow, 2008; Silva et al., 2009). We defined an herbaceous individual as a non-woody plant with no connection to any other plant at the ground level (Maraschin-Silva et al., 2009). All botanical material was deposited in the Herbário Dárdano de Andrade Lima of the Instituto Agronômico de Pernambuco (IPA) (vouchers 88 948 to 89 103). Species identification followed APG III (2009) for angiosperms and Smith et al. (2006, 2008) for ferns. The scientific nomenclature was updated following the database of The Plant List (available at and the exotic herbs' presence was consulted on the site Flora do Brasil database (http:// 2.3. Litter, soil attributes and patch metrics To examine potential changes in litter accumulation in forest fragments we collected all dead organic matter on the soil surface contained in a collecting frame of 50 × 50 cm (adapted from Kleinpaul et al., 2005). Litter samples were deposited into paper bags for drying in an electric oven (50 °C) until they reached a constant weight (dry weight). The dry material was weighed on a balance with a maximum capacity of 5 kg and 1 g sensitivity (Fortes et al., 2008). To measure soil temperature we introduced a mercury thermometer 2 cm into the soil, at the same point of litter sampling (Lima-Ribeiro, 2008). To estimate soil moisture, after litter sampling, we collected the soil contained within a galvanized stainless steel sink of 10 × 10 × 5 cm. We deposited soil samples into polyethylene bags and weighed them on a balance with a 5 kg maximum capacity and 1 g sensitivity to calculate their wet weight. Then we transferred the soil to paper bags for drying in an electric oven (50 °C) until they reached a constant weight (dry weight) (Fortes et al., 2008). Soil moisture was calculated as the difference between the wet and dry weights divided by the dry weight and multiplied by 100 to be expressed as a percentage (EMBRAPA, 1997). All environmental variables were collected between 10 a.m. and 4 p.m. on days with similar climatic conditions. Finally, some patch metrics were adopted as explanatory variables for herb assemblages attributes. We refer to patch size, distance to the nearest forest edge, patch shape (sensu Laurance and Yensen, 1991) and connectivity (% of surrounding forest cover within a 1 km external buffer), which have been demonstrated to correlate with tree assemblage attributes in human-modified landscapes (Santos et al., 2008). As detailed by Santos et al. (2008), patch and landscape metrics for the Serra Grande landscape were quantified via GIS packages and vegetation-cover digital maps. 2.4. Data analyses To test for differences in herb density, litter accumulation, soil temperature and moisture between forest fragments and continuous forest we used Student t tests. Differences in species diversity (Shannon H′) between habitats were compared via a Hutchinson t test (Zar, 1996). A Mann–Whitney test was applied to examine differences in species richness between habitats. Generalized linear models (GLMs) were

adopted to correlate assemblage attributes with forest patch metrics. These metrics were obtained according to the procedures described in Santos et al. (2008). As suggested for count data, such as plant density and species richness, we ran GLMs adopting a Poisson distribution with a log link function and corrected for overdispersion (Crawley, 2002). Species richness at habitat spatial scale was examined via species-individual accumulation curves after 1000 randomizations in the R software (v. 3.1.1). These curves were also used to infer about species turnover. To test the hypothesis that forest fragments were taxonomically distinct from the interior areas of Coimbra forest, we performed a non-metric multidimensional scaling (NMDS) ordination of all 100 plots using a Bray–Curtis dissimilarity matrix built from a species abundance by plot matrix (Krebs, 1989). Species abundance data were sqrttransformed and standardized (sensu Clarke and Gorley, 2005) in order to avoid any bias resulting from highly abundant species and differences in sample sizes (i.e. herb density per plot). To examine the effect of habitat type on the species similarity between plots, we ran an ANOSIM test with habitat type as a factor (Clarke and Gorley, 2005). Finally, we performed an indicator species analysis (sensu Dufrêne and Legendre, 1997) based on two groups of herb plots identified by both NMDS ordinations and ANOSIM tests: one consisting of forest interior plots and another of forest fragment plots. To assess the relationship between species composition and environmental variables we applied canonical correspondence analyses (CCAs) (ter Braak, 1988). The dependent matrix (species composition) was constructed considering only species with more than nine individuals in the entire dataset (Botrel et al., 2002), as species with lower densities contribute little to the ordination (Causton, 1988). The environmental matrix was built with data on, litter accumulation and soil temperature and moisture (ter Braak, 1988), resulting in CCAs with 62 species, 98 plots and three environmental variables. Finally, a Monte Carlo permutation test was performed to examine CCA correlations. Analyses were performed in Biostat 5.0 (Ayres et al., 2007), Primer 6.0 (Clarke and Gorley, 2005), JMP 8 (SAS Institute, Cary, NC) and MVSP 3.1 (Multivariate Statistical Package) (Kovach, 2007). 3. Results We recorded a total of 6027 herbs in the 100 25-m2 plots: 3221 in forest fragments and 2806 in the forest interior. These individuals were assigned to 134 species and 42 families (29 angiosperms and 13 ferns): Poaceae (17 spp.), Araceae (15), Cyperaceae (11) and Marantaceae (11) were the richest families in number of species (see Table A.1). At plot level, plant density varied from 64.4 ± 57.8 herbs/25 m2 in forest fragments (mean ± SD) to 56.1 ± 44.1 herbs/25 m2 in the forest interior, but this difference was not significant due to a great variance among plots. However, species richness dropped from 6.9 ± 3.0 species/25 m2 in the forest interior to 4.1 ± 2.2 species/25 m2 (U = 4.908; P b 0.0001), a reduction of 41% experienced by fragments. Species diversity experienced a similar reduction: 3.61 nats/individual in the forest interior vs. 2.73 nats/individual across forest fragments (Hutchinson t = 26.3; v = 6008.2; P b 0.05). At habitat level, forest fragments also exhibited reduced species richness with lower cross-plot species turnover (Fig. 2). As suggested by these curves, forest fragments tend to support half of the richness observed in forest interior at larger spatial scales. The 10 forest fragments analyzed varied from 7.8 ha to 83.6 ha in area, 1.2% to 27.8% in connectivity and had plots located 107 m to 301 m from the nearest edge. Despite this variation, plant density was not correlated to any patch metrics (Table 1). However, species richness correlated negatively with patch area, but positively with plot distance to the nearest edge and connectivity (Table 1). Habitats also differed in terms of species composition and the density of certain taxa. An NMDS ordination of herb plots based on species similarity (species composition plus density) resulted in two consistent

P.B. Lima et al. / Biological Conservation 191 (2015) 588–595

Fig. 2. Species–individual accumulation curves for herbaceous species recorded in 100 25-m2 plots in forest fragments and forest interior stands (Coimbra forest) in the Serra Grande landscape, northeast Brazil. Bar describes the standard deviation.

and clearly segregated clusters: one formed by the 10 interior areas of Coimbra forest and another formed by the small forest fragments (Fig. 3). In addition, an ANOSIM uncovered a significant correlation between habitat type and degree of taxonomic similarity between plots (R = 0.281; P = 0.001). Such taxonomic differentiation could have been caused by two aspects. First, about 68% of species (97 spp.) were exclusive to interior forest habitat, while 54% (68 spp.) were restricted to forest fragments; i.e. fragments were largely impoverished in terms ferns, aroids and calatheas. Second, patterns of species abundance were markedly different between habitats, with a higher level of species dominance in forest fragments; i.e. the most dominant species represented 13% of all individuals in forest interior (Spathiphyllum gardneri) but this specie achieved 0.6% in fragments (Fig. 4, Table A.1). Briefly, Parodiolyra micrantha, a native bamboo species, proliferated in forest fragments from 32 (1.15% of all plants recorded in forest interior) to 1028 individuals (32% of all plants recorded in the forest fragment). Finally, cross-habitat taxonomic differentiation was confirmed by an indicator species analysis, which underscored the occurrence of six indicator species in the floristically cohesive group of plots formed by fragments. In contrast, 13 indicator species were highlighted in the samples of forest interior plots (Table 2). In addition to taxonomic differentiation, herb assemblages also differed ecologically as ferns predominated among the species recorded exclusively in interior forest plots: 21 fern species (32%), 8 monocots (56%) and 37 dicot plant species (12%); while only 1 fern species (3%), 13 monocots (35%) and 23 dicot plant species (62%) were considered exclusive to forest fragments (see Table A.1).


Fig. 3. NMDS ordination of the 100 25-m2 plots located in forest fragments and forest interior stands (Coimbra forest) in the Serra Grande landscape, northeast Brazil. Filled triangles = forest interior; open circles = small forest fragments.

Compared to forest interior, forest fragments exhibited increased litter accumulation (263.06 ± 97.8 g vs. 227.74 ± 90.7 g; t = 1.92; df = 98; P b 0.0285), reduced soil moisture (18.03% ± 7.33% vs. 23.9% ± 5.8%; t = −5.17; df = 98; P b 0.0001) and elevated soil temperature (23.71 °C ± 1.32 °C vs. 22.74 °C ± 0.59 °C; t = 4.71; df = 98; P b 0.0001). The CCA analysis revealed correlations between soil variables and the ordination of both plots and herbaceous plant species (Figs. 5 and 6). The first two axes explained ca. 74% of the variation (44.9% for axis 1 and 28.8% for axis 2, total inertia = 19.8). The eigenvalues for the two main axes scored 0.575 (axis 1) and 0.370 (axis 2), while the species–environment correlations were 0.835 and 0.726, for axis 1 and axis 2, respectively. These correlations proved to be significant (F = 2.1671; P = 0.003). The most important variables predicting plot ordination were soil temperature, soil humidity and soil litter cover, respectively (Fig. 5). More specifically, plots in forest interior correlated to the highest scores of soil humidity and the lowest scores of soil temperature. Fragment plots exhibited the inverse to this trend, but were also correlated to high scores of litter cover. A large subset of species (e.g. Desmodium axillare, Lepidagathis alopecuroidea, Ichnanthus tenuis and Oplismenus hirtellus) exhibited higher densities across fragment plots and were correlated positively to high scores of temperature and litter cover (Fig. 6). Conversely, species of Araceae, Marantaceae and ferns (e.g. Philodendron propinquum, Spathiphyllum cannifolium, Goeppertia brasiliensis, and Adiantum) were more abundant and frequent across the humid plots located in the forest interior. Just one herb species (Tradescantia zebrina) was considered an exotic species and it was recorded in forest fragments. 4. Discussion

Table 1 The influence of four patch metrics (edge distance, patch area, connectivity and patch shape) on plant density and species richness of herb plant assemblages across 10 Atlantic forest fragments in the Serra Grande landscape, northeast Brazil. Relationships were examined via GLMs. Assemblage attributes/patch metrics

L–R Chi-square

P value

Species richness Edge distance (m) Patch area (ha) Connectivity (% of surrounding forest cover) Patch shape

5.158 6.552 6.537 3.711

0.023 0.010 0.011 0.054

Plant density Edge distance (m) Patch area (ha) Connectivity (% of surrounding forest cover) Patch shape

0.764 0.915 0.132 0.464

0.3820 0.3388 0.7163 0.4957

Like other Atlantic forest landscapes, the Serra Grande landscape still supports a diverse ground herb flora, including a myriad of ferns, grasses, aroids, calatheas and palmetto species as already documented for other neotropical forest patches (Benítez-Malvido and Martínez-Ramos, 2003; Costa et al., 2005; Linares-Palomino et al., 2009; Ribeiro et al., 2010). However, our results suggest that although forest interior and forest fragments support similar herb assemblages in terms of plant density, assemblages differ in terms of species richness and taxonomic and ecological composition. In fact, forest fragments exhibited impoverished assemblages at plot scale, with species richness responding to patch metrics along the data range observed for patch area, connectivity and distance to nearest edge. In Serra Grande, assemblages were also impoverished at landscape scale, with higher levels of species dominance due to the proliferation of some native taxa; i.e. reduced beta diversity or species turnover.


P.B. Lima et al. / Biological Conservation 191 (2015) 588–595

Fig. 4. Species-rank curves for herbaceous plant species recorded in (A) forest interior stands, and (B) forest fragments in the Serra Grande landscape, northeast Brazil. Abbreviations for indicator species in both habitats are as follows: Pami, Parodiolyra micrantha; Hean, Heliconia angusta; Poa1, Poaceae 1; Cen2, Centrosema 2; Deax, Desmodium axillare; Ryce, Rhynchospora cephalotes; Spga, Spathiphyllum gardneri; Phfr, Philodendron fragrantissimum; Tri1, Triplophyllum 1; Mopl, Monotagma plurispicatum; Spca, Spathiphyllum cannifolium; Adi2, Adiantum 2; Hesp, Heliconia spathocircinata; Pha3, Pharus 3; Cyha, Cyperus haspan; Dan1, Danaea 1; Logu, Lomagramma guianensis; Trfo, Trimezia fosteriana; Icne, Ichnanthus nemoralis. *Indicator species of continuous forest; **Indicator species of small fragments.

Despite reduced scores of species richness across multiple spatial scales, forest fragments exhibited taxonomically and ecologically distinct assemblages, with several overrepresented families and a small pool of “exclusive species” and native proliferating species. It is noteworthy that there has been an almost complete extirpation of fern species across forest fragments, with a disproportional increment in the relative contribution of dicots among fragment exclusive species, even though the Serra Grande landscape supports a diverse fern flora with 22 species (16.4% of the whole herb flora). Habitats also differed in terms of edaphic conditions with fragments supporting more desiccated and harsh conditions (i.e. thicker litter cover). The abundances of several species were correlated with soil conditions, suggesting a causal mechanism for such altered assemblages recorded across forest fragments. There are no more than a handful of papers reporting how tropical herb communities respond to habitat loss and fragmentation, although this process has been relatively well investigated for temperate forests (see Gillian, 2007; Honnay et al., 2005; Kolk and Naaf, 2015; Whigham, 2004). In the case of tropical forests, our findings offer additional evidence for the establishment of impoverished herb assemblages

at local scale with some native taxa apparently benefiting from the establishment of edge-affected habitats (Benítez-Malvido and Martínez-Ramos, 2003; Tomimatsu et al., 2011). However, we also offer evidence for more pervasive shifts in herb communities such as (1) reduced species richness, (2) elimination of particular plant groups (e.g. shade-tolerant ferns), (3) proliferation of native species, (4) a negligible role played by exotic or weedy species, which contrasts to disturbed forests in India (Rasingam and Parthasarathy, 2009), and (5) assemblage shifts correlated with the modification of soil attributes and patch metrics, not only at plot, but also at habitat scale (see Digiovinazzo et al., 2010). The altered assemblages documented in the fragments of Serra Grande represent a comprehensive description of tropical herb assemblages, and offer an informative reference about potential responses exhibited as natural landscapes are converted into human-modified landscapes. It is true that some understory herb species bear a set of traits conferring superior competitive and colonization ability such as: effective seed dispersal, phenological plasticity, efficient absorption of water and nutrients via superficial and dense root systems, higher

Table 2 The herb indicator species (sensu Dufrêne and Legendre, 1997) recorded in forest fragments and forest interior stands with their respective botanical family in the Serra Grande landscape, northeast Brazil. Indicator species



Indicator value (IV)ª

P valueª

Parodiolyra micrantha Heliconia angusta Poaceae sp. Centrosema sp. Desmodium axillare Rhynchospora cephalotes Spathiphyllum gardneri Philodendron fragrantissimum Triplophyllum sp. Monotagma plurispicatum Spathiphyllum cannifolium Adiantum sp. Heliconia spathocircinata Pharus sp. Cyperus haspan Danaea sp. Lomagramma guianensis Trimezia fosteriana Ichnanthus nemoralis

Poaceae Heliconiaceae Poaceae Fabaceae Fabaceae Cyperaceae Araceae Araceae Tectariaceae Marantaceae Araceae Pteridaceae Heliconiaceae Poaceae Cyperaceae Marattiaceae Dryopteridaceae Iridaceae Poaceae

Small fragments Small fragments Small fragments Small fragments Small fragments Small fragments Interior Interior Interior Interior Interior Interior Interior Interior Interior Interior Interior Interior Interior

32.97 15.17 15.04 12 12 12 45.65 37.8 30 26 23.79 18 18 18 16 12 12 12 10

0.001 0.035 0.042 0.024 0.024 0.031 0.001 0.001 0.001 0.001 0.001 0.004 0.001 0.002 0.007 0.032 0.034 0.023 0.049


Significant values for the Indicator Species Analysis according to Dufrêne and Legendre (1997).

P.B. Lima et al. / Biological Conservation 191 (2015) 588–595

Fig. 5. Ordination of herb assemblage plots via a Canonical Correspondence Analysis (CCA) considering two forest habitats (Open triangles = small forest fragments, Filled triangles = forest interior stands) and three environmental variables (MOIS = soil moisture, TEMP = soil temperature, LITTER = litter accumulation) in the Serra Grande landscape, northeast Brazil.

photosynthetic efficiency and vegetative reproduction (Maraschin-Silva et al., 2009; Richards, 1996; Vieira and Pessoa, 2001). On the other hand, several species have been documented to exhibit (1) limited dispersal and colonization ability (Honnay et al., 2005; Petit et al., 2004; Uriarte et al., 2010; Verheyen et al., 2003), (2) habitat specialization (e.g. soil type, microtopography, light environment) with a patchy distribution (Costa et al., 2005; Honnay et al., 2005; Tuomisto et al., 2003, 2012; Whigham, 2004), (3) increased sensitivity to physical edge-effects, particularly to habitat desiccation or reduced water availability (Bruna, 1999, 2002; Paciencia and Prado, 2005), and (4) sensitivity to competition with weeds, increased litterfall, herbivory and diseases (Aguiar and Tabarelli, 2010; Henle et al., 2004; Leal et al., 2014). Thus, habitat loss and fragmentation leading to the establishment of human-modified landscapes may result in (1) altered availability of regeneration niche, (2) altered mortality and fecundity and (3) increased vulnerability to demographic and environmental stochasticity (Hobbs and Yates, 2003). Although we did not explicitly examine the mechanisms leading to shifts in herb assemblages in the Serra Grande landscape, we can offer some evidence for potential drivers and affected plant groups. Firstly, a potential mechanism may be the elimination of particular habitats and their plant specialists, since habitat loss is not a random process, with agriculturally valuable soil/lands being constantly targeted (Tabarelli et al., 2010a,b). In the Serra Grande landscape, humid habitats associated with creeks and springs are missing in forest fragments, considering that shaded habitats are associated with old-growth forest stands (Turner, 1996). In fact, forest fragments in this landscape are dominated by edge-affected habitats with clear impacts on the structure of tree assemblages (Oliveira et al., 2004; Santos et al., 2008; Tabarelli et al., 2010a). It is likely that habitat loss and fragmentation greatly reduce the availability of humid microhabitats and diversity of soil types, while edge-affected habitats offer suitable conditions for lightloving and disturbance-adapted species. This rationale is also consistent with the collapse of ferns across forest fragments, since the majority can be considered shade-demanding species that depend on humid habitats in undisturbed forest patchesin the Atlantic and other tropical forests (Costa et al., 2005); e.g. Blechnum, Danaea, Polybotrya, Lomagramma and Trichomanes species (Barros et al., 2006; Kluge and Kessler, 2011; Kozera et al., 2009). Reduced availability of humid microhabitats may also be the case for the lack of some aroids, palmettos and calatheas species across the Serra Grande forest fragments (see Cicuzza et al., 2013).


Fig. 6. Ordination of 62 herbaceous plant species inhabiting forest fragments and forest interior stands via a Canonical Correspondence Analysis considering three environmental variables (Serra Grande landscape, northeast Brazil). Each arrow indicates the direction of maximum change of each environmental variables across the diagram, and its length is proportionate to the rate of change in this direction (i.e. environmental variables with longer arrows are more closely related to the pattern of species distribution shown in the ordination diagram). Environmental variables: MOIS = soil moisture, TEMP = soil temperature, LITTER = litter accumulation. Plant species: 1 = Hygrophila costata, 2 = Lepidagathis alopecuroidea, 3 = Ruellia paniculata, 4 = Heteropsis linearis, 5 = Heteropsis sp. 1, 6 = Philodendron fragrantissimum, 7 = Philodendron propinquum, 8 = Philodendron sp. 1, 9 = Rhodospatha oblongata, 10 = Spathiphyllum cannifolium, 11 = Spathiphyllum gardneri, 12 = Aechmea mertensii, 13 = Tradescantia zebrina, 14 = Costus spicatus, 15 = Cyperus haspan, 16 = Rhynchospora cephalotes, 17 = Rhynchospora comata, 18 = Scleria latifolia, 19 = Scleria gaertneri, 20 = Dioscorea leptostachya, 21 = Centrosema sp. 2, 22 = Desmodium axillare, 23 = Heliconia angusta, 24 = Heliconia psittacorum, 25 = Heliconia spathocircinata, 26 = Trimezia sp. 1, 27 = Goeppertia brasiliensis, 28 = Goeppertia effusa, 29 = Ctenanthe sp. 2, 30 = Ischnosiphon gracilis, 31 = Ischnosiphon longiflorus, 32 = Monotagma plurispicatum, 33 = Bertolonia marmorata, 34 = Liparis nervosa, 35 = Oeceoclades maculata, 36 = Passiflora sp. 1, 37 = Ichnanthus leiocarpus, 38 = Ichnanthus nemoralis, 39 = Ichnanthus tenuis, 40 = Olyra latifolia, 41 = Olyra sp.1, 42 = Oplismenus hirtellus, 43 = Parodiolyra micranta, 44 = Parodiolyra sp. 1, 45 = Pharus sp. 3, 46 = Poaceae 1, 47 = Poaceae 2, 48 = Poaceae 3, 49 = Poaceae 4, 50 = Coccocypselum cordifolium, 51 = Monocots 01, 52 = Ctenitis sp. 1, 53 = Mickelia guianensis, 54 = Polybotrya cylindrica, 55 = Polybotrya sp. 1, 56 = Trichomanes pinnatum, 57 = Danaea sp. 1, 58 = Danaea sp. 2, 59 = Adiantum sp. 1, 60 = Adiantum sp. 2, 61 = Triplophyllum sp. 1, 62 = Thelypteris macrophylla.

Note that shade-tolerant herbs have been identified as disturbancesensitive species and a typical component of old-growth forest patches in both temperate (see Gillian, 2007; Kolk and Naaf, 2015) and tropical forest biotas (see Fontoura et al., 2006; Tuomisto et al., 2002, 2003); i.e. old-growth flora as also recognized among tree species (see Tabarelli et al., 2008). It is also consistent with (1) a fragmentation-related proliferation of taxa that are considered indicators of disturbed habitats with increased light levels; e.g. Olyra latifolia, P. micrantha and Lygodium venustum (Oliveira and Longhi-Wagner, 2001; Prado, 2005); and (2) the presence of more desiccated and disturbed soils (i.e. increased litter fall) across forest edges and small fragments as previously documented (Didham and Lawton, 1999; Kapos, 1989; Lima-Ribeiro, 2008; Siqueira et al., 2004), which correlated with the distributions and abundances of many species in our study landscape. As a common element of forest edges and small forest fragments across the Atlantic forest region, the dwarf bamboo P. micrantha represents a classic example of a native winner species (sensu Tabarelli et al., 2012), proliferating in edgeaffected habitats. According to this rationale it is reasonable to propose that those herbs recorded in the fragments of Serra Grande landscape are light-demanding species associated with treefall gaps in forest interior patches, while being able to benefit from edge-affected habitats as soon as forest edges are established.


P.B. Lima et al. / Biological Conservation 191 (2015) 588–595

In synthesis, it is likely that habitat loss and the creation of forest edges have reduced habitat availability and/or habitat quality for shade-tolerant species and habitat specialists, causing either complete population extirpation or inferior population performance across remaining edge-affected habitats, although it is still not clear how population viability is affected. We are most likely referring to “forest species” or the old-growth flora, which approach the classical k-strategist syndrome as already argued for tree species (Tabarelli et al., 2008); i.e. perennial and long-lived organisms with reduced investments in sexual reproduction. Shifts in herb assemblages examined here reinforce the notion that habitat loss and fragmentation, particularly the establishment of illuminated and desiccated forest edges, result in the extirpation of particular ecological groups (i.e. non-random extinctions) with few species or ecological groups experiencing proliferation such as light-demanding species (Santos et al., 2012; Tabarelli et al., 2010a,b). Collectively, these processes result in impoverished assemblages and biotic homogenization at multiple spatial scales (Lôbo et al., 2011), potentially limiting the conservation services provided by human-modified landscapes. Further studies should investigate the basic mechanisms affecting the vital rates and viability of populations, since it is essential to provide any management guideline of biodiversity persistence, mostly on this irreplaceable tropical forest. Acknowledgments We thank the Usina Serra Grande for research facilities and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) (304598/ 2011-9) for graduate scholarships to PBL and LFL and research grants to MT, BAS and CSZ. We are also grateful to Conservation International (CIBrazil) and Centro de Pesquisas Ambientais do Nordeste (CEPAN) for logistical support. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. References Aguiar, A.V., Tabarelli, M., 2010. Edge effects and seedling bank depletion: the role played by the early successional palm Attalea oleifera (Arecaceae) in the Atlantic forest. Biotropica 42, 158–166. APG III, 2009. An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants. Bot. J. Linn. Soc. 161, 105–202. 1111/j.1095-8339.2009.00996.x. Ayres, M., Ayres Júnior, M., Ayres, D.L., Santos, A.A., 2007. BIOESTAT — Aplicações estatísticas nas áreas das Ciências Bio-Médicas. Ong Mamiraua, Belém, PA. Barros, I.C.L., Santiago, A.C.P., Pereira, A.F.N., Pietrobom, M.R., 2006. Pteridófitas. In: Pôrto, K.C., Almeida-Corez, J.S., Tabarelli, M. (Eds.), Diversidade biológica e conservação da Floresta Atlântica ao Norte do Rio São Francisco. Ministério do Meio Ambiente, Brasília, pp. 147–171. Benítez-Malvido, J., Martínez-Ramos, M., 2003. Impact of forest fragmentation on understory plant species richness in Amazonia. Conserv. Biol. 17, 389–400. http://dx.doi. org/10.1046/j.1523-1739.2003.01120.x. Botrel, R.T., Oliveira-Filho, A.T., Rodrigues, L.A., Curi, N., 2002. Influência do solo e topografia sobre as variações da composição florística e estrutura da comunidade arbóreo-arbustiva de uma floresta estacional semidecidual em Ingaí, MG. Rev. Bras. Bot. 25, 195–213. Bruna, E.M., 1999. Seed germination in rainforest fragments. Nature 402, 139. http://dx. Bruna, E.M., 2002. Effects of forest fragmentation on Heliconia acuminata seedling recruitment in central Amazonia. Oecologia 132, 235–243. s00442-002-0956-y. Bruna, E.M., Vasconcelos, H.L., Heredia, S., 2005. The effect of habitat fragmentation on communities of mutualists: a test with Amazonian ants and their host plants. Biol. Conserv. 124, 209–216. Causton, D.R., 1988. An Introduction to Vegetation Analysis Principles and Interpretation. Unwin Hyman, London. Chiarello, A.G., 1999. Effects of fragmentation of the Atlantic forest on mammal communities in south-eastern Brazil. Biol. Conserv. 89, 71–82. S0006-3207(98)00130-X. Cicuzza, D., Kromer, T., Poulsen, A.D., Abrahamczyk, S., Delhotal, T., Piedra, H.M., Kessler, M., 2013. A transcontinental comparison of the diversity and composition of tropical

forest understory herb assemblages. Biodivers. Conserv. 22, 755–772. http://dx.doi. org/10.1007/s10531-013-0447-y. Clarke, K.R., Gorley, R.N., 2005. PRIMER v6: User Manual/Tutorial. PRIMER-E, Plymouth. Costa, F.R.C., Magnusson, W.E., Luizao, R.C., 2005. Mesoscale distribution patterns of Amazonian understory herbs in relation to topography, soil and watersheds. J. Ecol. 93, 863–878. Crawley, M.J., 2002. Statistical Computing. An Introduction to Data Analysis Using S Plus. Wiley, J, New York. Didham, R.K., Lawton, J.H., 1999. Edge structure determines the magnitude of changes in microclimate and vegetation structure in tropical forest fragments. Biotropica 31, 17–30. Digiovinazzo, P., Ficetola, G.F., Bottoni, L., Andreis, C., Padoa-Schioppa, E., 2010. Ecological thresholds in herb communities for the management of suburban fragmented forests. For. Ecol. Manag. 259, 343–349. Dufrêne, M., Legendre, P., 1997. Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecol. Monogr. 67, 645–666. 1890/0012-9615. EMBRAPA, 1997. Manual de Métodos de Análise de Solos. segunda ed. CNPS, Rio de Janeiro. Fontoura, S.B., Ganade, G., Larocca, J., 2006. Changes in plant community diversity and composition across an edge between Araucaria forest and pasture in South Brazil. Rev. Bras. Bot. 29, 79–91. Fortes, F.O., Lúcio, A.D., Storck, L., 2008. Plano amostral para coleta de serapilheira na Floresta Ombrófila Mista do Rio Grande do Sul. Ciênc. Rural 38, 2512–2518. http:// Galetti, M., Donatti, C.I., Pires, A.S., Guimarães Jr., P.R., Jordano, P., 2006. Seed survival and dispersal of an endemic Atlantic forest palm: the combined effects of defaunation and forest fragmentation. Bot. J. Linn. Soc. 151, 141–149. 1095-8339.2006.00529.x. Gillian, F.S., 2007. The ecological significance of the herbaceous layer in Temperate forest ecosystems. BioScience 57, 845–858. Giulietti, A.M., Harley, R.M., Queiroz, L.P., Wanderley, M.G.L.M., Van Den Berg, C., 2005. Biodiversity and conservation of plants in Brazil. Conserv. Biol. 19, 632–639. http:// Gonçalves, E.G., Lorenzi, H., 2011. Morfologia Vegetal — organografia e dicionário ilustrado de morfologia das plantas vasculares. segunda ed. Instituto Plantarum de Estudos da Flora, São Paulo. Henle, K., Davies, K.F., Kleyer, M., Margules, C., Settele, J., 2004. Predictors of species sensitivity to fragmentation. Biodivers. Conserv. 13, 207–251. B:BIOC.0000004319.91643.9e. Hobbs, R.J., Yates, C.J., 2003. Impacts of ecosystem fragmentation on plant populations: generalising the idiosyncratic. Aust. J. Bot. 51, 471–488. BT03037. Honnay, O., Jacquemyn, H., Bossuyt, B., Hermy, M., 2005. Forest fragmentation effects on patch occupancy and population viability of herbaceous plant species. New Phytol. 166, 723–736. Inácio, C.D., Jarenkow, J.A., 2008. Relações entre a estrutura da sinúsia herbácea terrícola e a cobertura do dossel em floresta estacional no Sul do Brasil. Rev. Bras. Bot. 31, 41–51. Kapos, V., 1989. Effects of isolation on the water status of forest patches in the Brazilian Amazon. J. Trop. Ecol. 5, 173–185. Kleinpaul, I.S., Schumacher, M.V., Brun, E.J., Kleinpaul, J.J., 2005. Suficiência amostral para coletas de serapilheira acumulada sobre o solo em Pinus elliottii Engelnm, Eucalyptus sp. em floresta estacional decidual. Rev. Árvore 29, 965–972. 1590/S0100-67622005000600016. Kluge, J., Kessler, M., 2011. Influence of niche characteristics and forest type on fern species richness, abundance and plant size along an elevational gradient in Costa Rica. Plant Ecol. 212, 1109–1121. Kolk, J., Naaf, T., 2015. Herb layer extinction debt in highly fragmented temperate forests — completely paid after 160 years? Biol. Conserv. 182, 164–172. 1016/j.biocon.2014.12.004. Kovach, W.L., 2007. MVSP — A MultiVariate Statistical Package for Windows, ver. 3.1. Kovach Computing Services, Pentraeth, Wales, U.K. Kozera, C., Rodrigues, R.R., Dittrich, V.A.O., 2009. Composição florística do sub-bosque de uma Floresta Ombrófila Densa Montana, Morretes, PR, Brasil. Floresta 39, 323–334. Krebs, C.J., 1989. Ecological Methodology. Harper Collins Publ. Lalji, S., Singh, J.S., 1993. Importance of short-lived components of a dry tropical forest for biomass production and nutrient cycling. J. Veg. Sci. 4, 681–686. Laurance, W.F., Yensen, E., 1991. Predicting the impacts of edge effects in fragmented habitats. Biol. Conserv. 55, 77–92. Laurance, W.F., Lovejoy, L.E., Vasconcelos, H.L., Bruna, E.M., Didham, R.K., Stouffer, P.C., Gascon, C., Bierregaard Jr., R.O., Laurance, S.G., Sampaio, E., 2002. Ecosystem decay of Amazonian forest fragments: a 22-year investigation. Conserv. Biol. 16, 605–618. Laurance, W.F., Nascimento, H.E.M., Laurance, S.G., Andrade, A., Ribeiro, J., Giraldo, J.P., Lovejoy, T.E., Condit, R., Chave, J., Harms, K.E., D'Angelo, S., 2006a. Rapid decay of tree-community composition in Amazonian forest fragments. Proc. Natl. Acad. Sci. U. S. A. 103, 19010–19014. Laurance, W.F., Nascimento, H.E.M., Laurance, S.G., Andrade, A.C., Fearnside, P.M., Ribeiro, J.E.L., Capretz, R.L., 2006b. Rain forest fragmentation and the proliferation of successional trees. Ecology 87, 469–482. Laurance, W.F., Camargo, J.L.C., Luizão, R.C.C., Laurance, S.G., Pimm, S.L., Bruna, E.M., Stouffer, P.C., Williamson, G.B., Benítez-Malvido, J., Vasconcelos, H.L., Van Houtan, K.S., Zartman, C.E., Boyle, S.A., Didham, R.K., Andrade, A., Lovejoy, T.E., 2011. The fate of Amazonian forest fragments: a 32-year investigation. Biol. Conserv. 144, 56–67.

P.B. Lima et al. / Biological Conservation 191 (2015) 588–595 Leal, I.R., Wirth, R., Tabarelli, M., 2014. The multiple impacts of leaf-cutting ants and their novel ecological role in human-modified Neotropical forests. Biotropica 46, 516–528. Lima-Ribeiro, M.S.L., 2008. Efeitos de borda sobre a vegetação e estruturação populacional em fragmentos de Cerradão no Sudoeste Goiano, Brasil. Acta Bot. Bras. 22, 535–545. Linares-Palomino, R., Cardona, V., Hennig, E., Hensen, I., Hoffmann, D., Lendzion, J., Soto, D., Herzog, S., Kessler, M., 2009. Non-woody life-form contribution to vascular plant species richness in a tropical American forest. Plant Ecol. 201, 87–99. http://dx.doi. org/10.1007/s11258-008-9505-z. Lôbo, D., Leão, T., Melo, F.P.L., Santos, A.M.M., Tabarelli, M., 2011. Forest fragmentation drives Atlantic forest of northeastern Brazil to biotic homogenization. Divers. Distrib. 17, 287–296. Lopes, A.V., Girão, L.C., Santos, B.A., Peres, C.A., Tabarelli, M., 2009. Long-term erosion of tree reproductive trait diversity in edge-dominated Atlantic forest fragments. Biol. Conserv. 142, 1154–1165. Maraschin-Silva, F., Scherer, A., Baptista, L.R.M., 2009. Diversidade e estrutura do componente herbáceo-subarbustivo em vegetação secundária de floresta Atlântica no sul do Brasil. Rev. Bras. Biosci. 7, 53–65. Melo, F.P.L., Dirzo, R., Tabarelli, M., 2006. Biased seed rain in forest edges: evidence from the Brazilian Atlantic forest. Biol. Conserv. 132, 50–60. biocon.2006.03.015. Melo, F.P.L., Arroyo-Rodríguez, V., Fahrig, L., Martínez-Ramos, M., Tabarelli, M., 2013. On the hope for biodiversity-friendly tropical landscapes. Trends Ecol. Evol. 28, 462–468. Myers, N., Mittermeier, R.A., Mittermeier, C.G., Fonseca, G.A.B., Kent, J., 2000. Biodiversity hotspots for conservation priorities. Nature 403, 853-845. 35002501. Oliveira, R.P., Longhi-Wagner, H.M., 2001. Poaceae. In: Wanderley, M.G.L., Shepherd, G.J., Giulietti, A.M. (Eds.), Flora fanerogâmica do estado de São Paulo vol. I. Instituto de Botânica, São Paulo. Oliveira, M.A., Grillo, A.S., Tabarelli, M., 2004. Forest edge in the Brazilian Atlantic forest: drastic changes in tree species assemblages. Oryx 38, 389–394. 1017/S0030605304000754. Oliveira, M.A., Santos, A.M.M., Tabarelli, M., 2008. Profound impoverishment of the largetree stand in a hyper-fragmented landscape of the Atlantic Forest. For. Ecol. Manag. 256, 1910–1917. Paciencia, M.L.B., Prado, J., 2005. Effects of forest fragmentation on pteridophyte diversity in a tropical rain forest in Brazil. Plant Ecol. 180, 87–104. s11258-005-3025-x. Pande, P.K., 2004. Nutrient cycling in disturbed tropical dry deciduous teak forest of Saptura plateau, Madhya Pradesh, India. 2004. J. Trop. For. Sci. 16, 94–105. Petit, S., Griffiths, L., Smart, S.S., Smith, G.M., Stuart, R.C., Wright, S.M., Petit, S., Griffiths, L., Smart, S.S., Smith, G.M., Stuart, R.C., Wright, S.M., 2004. Effects of area and isolation of woodland patches on herbaceous plant species richness across Great Britain. Landsc. Ecol. 19, 463–472. Poulsen, A.D., 1996. Species richness and diversity of ground herbs within a plot of lowland rainforest in northwesht Borneo. J. Trop. Ecol. 12, 177–190. Poulsen, A.D., Balslev, H., 1991. Abundance and cover of ground herbs in an Amazonian rain forest. J. Veg. Sci. 2, 315–322. Prado, J., 2005. Flora da Reserva Ducke, Amazônia, Brasil. Pteridophyta — Schizaeaceae, Rodriguésia 56 pp. 93–97. Rasingam, L., Parthasarathy, N., 2009. Diversity of understory plants in undisturbed and disturbed tropical lowland forests of Little Andaman Island, India. Biodivers. Conserv. 18, 1045–1065. Ribeiro, M.C., Metzger, J.P., Martensen, A.C., Ponzoni, F.J., Hirota, M.M., 2009. The Brazilian Atlantic Forest: how much is left, and how is the remaining forest distributed? Implications for conservation. Biol. Conserv. 142, 1141–1153. biocon.2009.02.021. Ribeiro, M.B., Bruna, E.M., Mantovani, W., 2010. Influence of post-clearing treatment on the recovery of herbaceous plant communities in Amazonian secondary forests. Restor. Ecol. 18, 50–58. Richards, P.W., 1996. The Tropical Rain Forest: an Ecological Study. Cambridge University Press, Cambridge. Santos, B.A., Benítez-Malvido, J., 2012. Insect herbivory and leaf disease in natural and human disturbed habitats: lessons from early-successional Heliconia herbs. Biotropica 44, 53–62. Santos, A.M.M., Cavalcanti, D.R., Silva, J.M.C., Tabarelli, M., 2007. Biogeographical relationships among tropical forests in north-eastern Brazil. J. Biogeogr. 34, 437–446. http:// Santos, B.A., Peres, C.A., Oliveira, M.A., Grillo, A., Alves-Costa, C.P., Tabarelli, M., 2008. Drastic erosion in functional attributes of tree assemblages in Atlantic forest fragments of


northeastern Brazil. Biol. Conserv. 141, 249–260. 2007.09.018. Santos, G.G.A., Santos, B.A., Nascimento, H.E.M., Tabarelli, M., 2012. Contrasting demographic structure of short- and long-lived pioneer tree species on Amazonian forest edges. Biotropica 44, 771–778. Silva, J.M.C., Tabarelli, M., 2000. Tree species impoverishment and the future flora of the Atlantic forest of northeast Brazil. Nature 404, 72–74. 35003563. Silva, K.A., Araújo, E.L., Ferraz, E.M.N., 2009. Estudo florístico do componente herbáceo e relação com solos em áreas de caatinga do embasamento cristalino e bacia sedimentar, Petrolândia, PE, Brasil. Acta Bot. Bras. 23, 100–110. 1590/S0102-33062009000100013. Siqueira, L.P., Matos, D.M.S., Portela, R.C.Q., Braz, M.I.G., Silva-Lima, L., Matos, M.B., 2004. Using the variances of microclimate variables to determine edge effects in small forest Atlantic Rain Forest fragments, South-Eastern Brazil. Ecotropica 10, 59–64. Smith, A.R., Pryer, K.M., Schuettpelz, E., Korall, P., Schneider, H., Wolf, P.G., 2006. A classification for extant ferns. Taxon 55, 705–731. Smith, A.R., Pryer, K.M., Schuettpelz, E., Korall, P., Schneider, H., Wolf, P.G., 2008. Fern classification. In: Ranker, T.A., Haufler, C.H. (Eds.), Biology and Evolution Ferns and Lycophytes. Cambridge University Press, New York, pp. 417–467. Tabarelli, M., Lopes, A.V., Peres, C.A., 2008. Edge-effects drive tropical forest fragments towards an early-successional system. Biotropica 40, 657–661. 1111/j.1744-7429.2008.00454.x. Tabarelli, M., Aguiar, A.V., Girão, L.C., Peres, C.A., Lopes, A.V., 2010a. Effects of pioneer tree species hyperabundance on forest fragments in northeastern Brazil. Conserv. Biol. 24, 1654–1663. Tabarelli, M., Aguiar, A.V., Milton, R., Metzger, J.P., Peres, C., 2010b. Prospects for biodiversity conservation in the Atlantic forest: lessons from aging human-modified landscapes. Biol. Conserv. 143, 2328–2340. 02.005. Tabarelli, M., Peres, C.A., Melo, F.P.L., 2012. The few winners and many losers paradigm revisited: emerging prospects for tropical forest biodiversity. Biol. Conserv. 155, 136–140. Ter Braak, C.J.F., 1988. Canoco — a Fortran Program for Canonical Community Ordination by (Partial) (Detrended) (Canonical) Correspondence Analysis and Redundancy Analysis, version 2.1. Intitute of Applied Computer Science, Wageningen. Tomimatsu, H., Yamagishi, H., Tanaka, I., Sato, M., Kondo, R., Konno, Y., 2011. Consequences of forest fragmentation in an understory plant community: extensive range expansion of native dwarf bamboo. Plant Species Biol. 26, 3–12. http://dx.doi. org/10.1111/j.1442-1984.2010.00310.x. Tuomisto, H., Ruokolainen, K., Poulsen, A.D., Moran, R.C., Quintana, C., Cañas, G., Cell, J., 2002. Distribution and diversity of pteridophytes and Melastomataceae along edaphic gradients in Yasuní National Park, Ecuadorian Amazonia. Biotropica 34, 516–533. Tuomisto, H., Ruokolainen, K., Yli-Halla, M., 2003. Dispersal, environment, and floristic variation of western Amazonian forests. Science 299, 241–244. 1126/science.1078037. Tuomisto, H., Ruokolainen, L., Ruokolainen, K., 2012. Modelling niche and neutral dynamics: on the ecological interpretation of variation partitioning results. Ecography 35, 961–971. Turner, I.M., 1996. Species loss in fragments of tropical rain forest: a review of the evidence. J. Appl. Ecol. 33, 200–209. Uriarte, M., Bruna, E.M., Rubim, P., Anciães, M., Jonckheere, I., 2010. Effects of forest fragmentation on the seedling recruitment of a tropical herb: assessing seed vs. safe-site limitation. Ecology 91, 1317–1328. Vasconcelos, H.L., Luizão, F.J., 2004. Litter production and litter nutrient concentrations in a fragmented Amazonian landscape. Ecol. Appl. 14, 884–892. 1890/03-5093. Verheyen, K., Honnay, O., Motzkin, G., Hermy, M., Foster, D.R., 2003. Response of forest plant species to land-use change: a life-history trait-based approach. J. Ecol. 91, 563–577. Vieira, C.M., Pessoa, S.V.A., 2001. Estrutura e composição florística do estrato herbáceosubarbustivo de um pasto abandonado na Reserva Biológica de Poço das Antas, município de Silva Jardim, Rio de Janeiro. Rodriguésia 52, 17–30. Whigham, D.F., 2004. Ecology of woodland herbs in temperate deciduous forests. Annu. Rev. Ecol. Evol. Syst. 35, 583–621. 021103.105708. Zar, J., 1996. Biostatistical Analysis. two ed. Prentice Hall, New Jersey.