Afforestation effects on vegetation structure and diversity of grasslands in southern Brazil: The first years

Afforestation effects on vegetation structure and diversity of grasslands in southern Brazil: The first years

G Model JNC-25260; No. of Pages 7 ARTICLE IN PRESS Journal for Nature Conservation xxx (2012) xxx–xxx Contents lists available at SciVerse ScienceDi...

590KB Sizes 0 Downloads 4 Views

G Model JNC-25260; No. of Pages 7

ARTICLE IN PRESS Journal for Nature Conservation xxx (2012) xxx–xxx

Contents lists available at SciVerse ScienceDirect

Journal for Nature Conservation journal homepage: www.elsevier.de/jnc

Afforestation effects on vegetation structure and diversity of grasslands in southern Brazil: The first years Alexandre F. Souza ∗ , Nadiane P. Ramos, Marco Aurélio Pizo 1 , Ingo Hübel, Luciane O. Crossetti Programa de Pós-Graduac¸ão em Biologia: Diversidade e Manejo da Vida Silvestre, Universidade do Vale do Rio dos Sinos, Av. UNISINOS 950, C.P. 275, São Leopoldo 93022-000, RS, Brazil

a r t i c l e

i n f o

Article history: Received 25 July 2012 Received in revised form 17 October 2012 Accepted 17 October 2012 Keywords: Eucalyptus Grazing Livestock Management Pampa

a b s t r a c t We evaluated the effects of afforestation on the composition and structure of southern Brazilian subtropical grasslands directly beneath tree plantations, in pastures, and in permanent protection areas near watercourses. Cover of plants was registered in 120 plots of 1 m2 in grasslands located at two distances from small watercourses in pasture and eucalypt plantation areas within a livestock ranching landscape. Livestock density is reduced in plantation areas. Species composition beneath tree plantations was significantly different from both pasture and nearby permanent protection areas. Permanent protection areas near plantations were also compositionally different from pastures. Eucalypt plantation was the environment type where ruderals were most abundant. Growth form cover distribution in pasture areas showed a consistent pattern, with graminous herbs being the most common growth form followed by prostrate herbs, prostrate subshrubs, and standing subshrubs. In plantation areas, the dominance of the graminoid herb growth form increased markedly, as well as standing herbs. These increases occurred mainly at the expense of prostrate herbs and subshrubs, which were significantly reduced. Plot-level species density was significantly affected by distance from watercourse and, more importantly, was significantly influenced by the interaction between management regime and distance to watercourse. It was reduced in eucalypt plantations relative to permanent protected areas. Current management of eucalypt plantations in southern Brazil produce unsuitable conditions for grassland species to thrive within plantations. Livestock densities <1.0 animal unit ha−1 in eucalypt plantations undermines the ability of surrounding permanent protection areas to act as ecological networks for biodiversity conservation. © 2012 Elsevier GmbH. All rights reserved.

Introduction In recent years plantations of exotic tree species, mainly eucalypts (Eucalyptus spp.) and Pinus spp., have increased in Latin American landscapes and partially replaced cattle ranching as the main land use in natural and semi-natural grasslands (Ceccon & Martínez-Ramos 1999; Farley 2007). In Brazil ca. 5.4 million ha (55% eucalypts) are covered with short-rotation plantation forests (Brockerhoff et al. 2008). Initially, these plantations occupied mainly areas previously used for cattle ranching or agriculture due to the reduced cost of land clearing (Farley 2007). More recently, however, they have expanded to native grasslands. Southern Brazilian natural and semi-natural grasslands are biodiversity-rich and

∗ Corresponding author. Current address: Departamento de Botânica, Ecologia e Zoologia, CB, Universidade Federal do Rio Grande do Norte, Campus Universitário, Lagoa Nova, Natal 59072-970, RN, Brazil. E-mail address: [email protected] (A.F. Souza). 1 Current address: Universidade Estadual Paulista Júlio de Mesquita Filho, Av. 24A 1515, Bela Vista, Rio Claro 13506-900, SP, Brazil.

poorly studied (Overbeck et al. 2007; Saraiva & Souza 2012). These grassland areas coevolved with large herbivore mammals now extinct (MacFadden 1997), and have been maintained since their disappearance by human activities like livestock grazing and fire (Behling & Pillar 2007) and have seen large-scale and rapid expansion of forestry that has radically changed the landscape in many parts of the region, with significant implications for biodiversity ˜ 2004; Overbeck et al. 2007; Saraiva & Souza (Bilenca & Minarro 2012). Most of these grasslands lack formal nature conservation designations that preclude afforestation (Overbeck et al. 2007) and therefore sites of high biodiversity value are vulnerable to planting. Eucalypt plantations may be a useful conservation tool in the restoration process of cultural or previously forested landscapes through the promotion of secondary succession (Brockerhoff et al. 2008; Fonseca et al. 2009; Geldenhuys 1997; Parrotta et al. 1997). However, besides potentially undesirable socio-economic consequences (Teixeira Filho 2008) and direct habitat destruction, eucalypt plantations established on native grasslands have a number of indirect environmental effects. These include changes in land management such as the alteration of grazing regimes (Buscardo et al. 2008; Gibson 2009), fragmentation (White et al. 2000),

1617-1381/$ – see front matter © 2012 Elsevier GmbH. All rights reserved. http://dx.doi.org/10.1016/j.jnc.2012.10.002

Please cite this article in press as: Souza, A. F., et al. Afforestation effects on vegetation structure and diversity of grasslands in southern Brazil: The first years. Journal for Nature Conservation (2012), http://dx.doi.org/10.1016/j.jnc.2012.10.002

G Model JNC-25260; No. of Pages 7 2

ARTICLE IN PRESS A.F. Souza et al. / Journal for Nature Conservation xxx (2012) xxx–xxx

Fig. 1. (A) Geographical location of the study area () in the Pampa biome (grey area), and (B) schematic representation of the sampling design within sampling stations.  = Eucalypt plantation edge. The hatched area in the lower panel represents eucalypt plantations. Filled and empty circles and squares stand for sample plots.  = pasture grasslands in permanent protection areas,  = pasture grasslands in unprotected areas, 䊉 = grasslands in permanent protection areas in the vicinity of eucalypt plantations, and  = eucalypt plantations.

shading (Ceccon & Martínez-Ramos 1999), soil carbon loss and nutrient depletion (Gibson 2009), reduced soil water retention capacity, and growth inhibition of other plant species in the vicinity by water competition (Jobbágy & Jackson 2004; Nosetto et al. 2005) or allelopathic effects (Florence 1986; Farley 2007, see Bremer & Farley 2010; Brockerhoff et al. 2008; Gibson 2009 for recent reviews). Tree plantations are particularly detrimental to biodiversity when established on natural grasslands and when exotic tree species are used (Bremer & Farley 2010; Brockerhoff et al. 2008), with general species loss and replacement of endemic and specialist species by ruderal and exotic species (Bremer & Farley 2010). White et al. (2000) analysed 90 grassland regions in both North and Latin America, and found that 37% occurred as small linear patches due to fragmentation. In southern Brazil, grasslands occurring within 30 m from the margins of watercourses <10 m constitute permanent protection areas according to Brazilian law (Law 4.771 by September 15th, 1965). They form a network of semi-natural grasslands embedded within extensive eucalypt plantations. Because of this configuration, permanent protection areas are likely to play an important, albeit unstudied, role as refugia and corridors for native plant and animal populations to persist in plantations and individuals to move between grassland fragments (Herrera & Laterra 2011; Samways et al. 2009). Given the pressure for the afforestation of landscapes to combat climate change, grasslands worldwide are likely to face increasing threats from tree plantations. The paucity of previous studies and the potential for significant biodiversity loss point to the need for studies that determine the effects of land use change from grassland to tree plantation. Here we evaluated patterns of variation in plant diversity in recently afforested and non-afforested grasslands in the biodiversity-rich and poorly-studied northern part of the Pampa biome. We tested the hypotheses that (1) eucalypt plantations represent unsuitable habitats for native species, indicated by reduced herbaceous cover and altered species composition and growth form distributions relative to baseline surrounding grazed native grasslands; (2) permanent protection grassland areas near watercourses are apt to function as biodiversity networks because they are not affected by newly-established tree plantations. They thus resemble baseline grasslands in composition and growth form distribution more than the herbaceous understories of nearby eucalypt stands.

Methods Study area The study area is located in the Vacaraí district of the São Gabriel Municipality, in the Brazilian state of Rio Grande do Sul (Fig. 1). The area is located in the Northern Campos biogeographical region ˜ (sensu Bilenca & Minarro 2004), in the Pampa biome. Vegetation consists in subtropical grassland that covers all of Uruguay, some areas of northeastern Argentina and southernmost Brazil. It is characterised by medium-tall grasslands, with areas of palm savanna, and riparian forests along the main rivers. The climate is temperate (there is no subtropical category in the Köppen–Geiger classification scheme) without a dry season and with a hot summer (temperature of the hottest month ≥22 ◦ C, Köppen–Geiger classification, Peel et al. 2007). Annual precipitation varies between 1312 and 1648 mm (Brazil 1973). Soils are deep and well-drained dystrophic Gleysols or flat and insufficiently-drained eutrophic Planosols (Streck et al. 2008). The regional relief is relatively flat, varying from 100 to 200 m a.s.l. Small natural watercourses (ca. 1–3 m wide) accompanied by wet grasslands occur between small hills, which are covered by dry grasslands. Sampling design and data collection Field data were collected in six sampling stations located at two sites. Of the six sampling stations, three were located in a 5000 ha ranch (site 1, 30◦ 34 40 S, 54◦ 33 42 W) supporting extensive cattle herds (hereafter referred to as pasture stations), and three were embedded in a 276.59 ha eucalypts plantation (site 2, 30◦ 33 44 S, 54◦ 30 29 W, hereafter referred to as plantation stations). Each sampling station was located adjacent to a different watercourse. Plantation areas were developed for pulp production, were planted for the first time and were composed of Eucalyptus saligna and Eucalyptus urograndis × globulus hybrids. Planted areas were previously used for cattle ranching with no known history of agricultural cultivation. Plantation rotation time is currently seven years, and planting spacing is 3.50 m × 2.15 m. Management practices include tree fall orientation away from native areas and machinery traffic over plantation debris as often as possible to avoid soil compaction.

Please cite this article in press as: Souza, A. F., et al. Afforestation effects on vegetation structure and diversity of grasslands in southern Brazil: The first years. Journal for Nature Conservation (2012), http://dx.doi.org/10.1016/j.jnc.2012.10.002

G Model JNC-25260; No. of Pages 7

ARTICLE IN PRESS A.F. Souza et al. / Journal for Nature Conservation xxx (2012) xxx–xxx

Insecticides (sulfur amide), herbicides (glyphosate and isoxaflutol) and fertilizers (natural reactive phosphate and NPK) were used in the first year of planting in each stand (Mauren K. Alves, personal communication). At the time of data collection, plantations were three years old and eucalypt heights were ca. 5 m. Pasture stations were located in a 120 ha pasture subdivision supporting a stocking density of 1.0 animal unit ha−1 . Neither chemical defensives nor prescribed burnings have been applied for at least 20 years and pasture management has been restricted to mechanical shrub control (Alcides M. P. Silva Neto, personal communication). Note that ungrazed grasslands are hardly ever found in the Northern Campos, and therefore the grazed system is the baseline in effect in this study. Plantation managers do not raise livestock in planted areas, but deteriorated fence conditions and intentional fence cutting by neighbours permit a reduced animal stocking density relative to pasture areas to graze grasslands in planted areas and in permanent protection areas embedded in plantations (Mauren K. Alves, personal communication). Stocking density estimates in plantation areas have not been carried out so far. Distances between plantation stations (average elevation = 169.3 m) varied from 410 to 1169 m, while distances between pasture stations (average elevation = 151.3 m) varied from 137 to 248 m. Site 1 and site 2 (i.e., plantation and pasture stations) were ca. 5 km apart. In each sampling station vegetation was sampled in 20 1 m × 1 m plots on four 100 m long transects established parallel to the nearest watercourse (Fig. 1). Five plots were randomly located on each transect while observing a minimum distance of 4 m between adjacent plots (n = 120 plots total). Transects were located at distances of 1, 26, 36, and 61 m from the nearest watercourse. Brazilian environmental regulation (Law 4.771 by September 15th, 1965) delimits permanent protection areas up to 30 m from watercourses, and plantations were established accordingly. Cattle ranging is allowed in permanent protection areas. Transects located at 1 and 26 m from watercourses were located in permanent protection areas (hereafter pasture or eucalypt permanent protection areas), while those located at 36 and 61 m lay between planting lines in eucalypt plantations (eucalypt areas) or grasslands outside permanent protection areas (pasture unprotected areas). Besides representing increasing distances from watercourses, transects also represented increasing distances from eucalypt plantation edges, which were located at ca. 30 m from watercourses (Fig. 1). Due to the early age of the studied plantations, plots in permanent protection areas were outside the shade cast by eucalypts. Vegetation was sampled between January and March 2010. To facilitate cover estimation, each 1 m × 1 m plot was subdivided into 100 subplots. Percent cover abundance of all vascular plants and bare ground was estimated in each plot. Cover was taken as the percentage of ground surface covered by the shadow of the foliage, estimated in 10% class intervals. The height of herbaceous vegetation was measured at five points per plot. Taxonomic identifications followed the APG II system (2003), and the keys by Burkart (1948, 1974, 1979, 1987), Barros (1960), Rosengurtt (1970), Irgang (1974), Porto et al. (1977), Boechat and Valls (1986, 1991), Longhi-Wagner (1987), Flores and Valls (1992), Guglieri and Longhi-Wagner (2000), Canto-Dorow (2001), Izaguirre and Beyhaut (2003), Luz (2004), Burkart and Bacigalupo (2005), Zuloaga and Morrone (2005), and O’Leary et al. (2007). Data analysis We used generalised linear modeling (GLM) to evaluate the effect of tree plantations on the variation in both log-transformed vegetation height and plot-scale species density. In both cases the analysis followed a split-plot design. In this analysis management regime (pasture or plantation) was the main factor, with site nested

3

within management regime. The four distances to watercourses classes formed the sub-plot factor, and the 1 × 1 plots were the observation units or random factor (Sokal & Rohlf 1995). Sites were explicitly included in the model in order to control for local history, soil type, and other site-specific differences (Sokal & Rohlf 1995). We used non-parametric MANCOVA (PERMANCOVA) with Bray–Curtis dissimilarities on arc-sin transformed species percent cover values to evaluate the effects of management regime on plot-level community composition, with data spatial structure (as PCNM eigenvalues, see explanation below) and distance to watercourses as covariables (Anderson 2001). Singletons – species only found in a single plot – were excluded from the analyses (Legendre & Legendre 1998). Significance of the pseudo-F ratio was evaluated using 4999 Monte Carlo permutations. Pair-wise a posteriori tests among treatment levels were performed with the same test (Anderson 2001) on Benjamini-Hochberg-corrected ␣ thresholds for multiple testing. PERMANCOVA was performed by the function ‘adonis’ of the package ‘VEGAN’ 1.6–22, in the R statistical software (version 2.9.1, R Core Team Development, Vienna, AT). Management regimes were spatially aggregated in the landscape due to segregation of management types in the ranch and in the plantation. Spatial autocorrelation between sampling units was thus a potentially confounding factor in the evaluation of the effect of management type on community composition. To control for the spatial autocorrelation effects, we used Principal Coordinates of Neighbour Matrices (PCNM, Borcard & Legendre 2002) to model the spatial structures present in the species data matrix. Eigenvectors corresponding to positive eigenvalues were then used as spatial descriptors in our PERMANCOVA model. PCNM analyses were run on Hellinger-transformed species abundances in R using function ‘quickPCNM’ function in package ‘PCNM’ version 1.9 after removing linear trends in the data (Borcard & Legendre 2002). Indicator Species Analysis (ISA, Dufrêne & Legendre 1997) was used as a complementary method to PERMANCOVA in order to identify the species that tend to be found in one of the management regimes versus another (McCune & Grace 2002). Indicator values (IV) range from zero (no indication) to 100 (perfect indication). Statistical significance was tested by 1000 Monte Carlo permutations using the ‘duleg’ function in the ‘LABDSV’ package in R. To illustrate pairwise floristic similarities among plots and management histories, we performed a non-metric multidimensional scaling (NMDS) using the function metaMDS of the package ‘VEGAN’ 1.17-0 in R and Bray–Curtis dissimilarities. This function standardises the scaling in the result by a principal components rotation. Dimensionality was assessed by examining the change in stress as a function of dimension while stepping down from a six- to one-dimensional solution. We chose the number of dimensions equal to four to minimise the stress (maximise the rank correlation between the calculated similarity distances and the plotted distances). We used the ‘envfit’ function of the package VEGAN to correlate distance to the watercourses with the NMDS axes. Correlation significance was assessed using 1000 permutations. Species were grouped as ruderals (sensu Grime 1977) or not according to the botanical literature (Boldrini 1997; Carneiro 1998; Damé et al. 1999; Gomar et al. 2004; Heringer & Jacques 2001; Lorenzi 2000). Species were also grouped into growth forms according to Ferri et al. (1981) and Gonc¸alves and Lorenzi (2007). Species were categorised as tussock grasses, erect herb, prostrate herb, stoloniferous herb, rosette herb, liana (woody stem), vine (herbaceous stem), shrub (woody stem, height >1.5 m), erect subshrub (woody stem, height ≤1.5 m), and prostrate subshrub. We used G-tests (Sokal & Rohlf 1995) to compare the proportions of ruderals and growth forms between management regimes using cover values. Individual species cover values were summed up to the corresponding growth form category before analyses, and only growth forms with at least 5% of total cover were included in the

Please cite this article in press as: Souza, A. F., et al. Afforestation effects on vegetation structure and diversity of grasslands in southern Brazil: The first years. Journal for Nature Conservation (2012), http://dx.doi.org/10.1016/j.jnc.2012.10.002

G Model JNC-25260; No. of Pages 7

ARTICLE IN PRESS A.F. Souza et al. / Journal for Nature Conservation xxx (2012) xxx–xxx

4

Fig. 2. Southern Brazilian grasslands vegetation height and the effect of eucalypt plantations.  = Edge between eucalypt plantations and permanent protected areas.

analysis. Species that had not been identified to the species level and could not be safely assigned to any life form group were considered as part of total vegetation cover in calculation of relative cover values of each growth form group, but otherwise left out from the analyses. Results Vegetation height (log-transformed) and ground cover (arcsintransformed) were strongly and negatively related (Pearson correlation, r = −0.73, P < 0.0001). After controlling for the effect of site, there was a significant interaction between management regime (pasture or plantation) and distance to watercourses (F = 61.78, df = 3, P < 0.0001) on vegetation height. This means that the effect of distance from watercourses on vegetation height differed depending on the management regime. In the pasture areas, vegetation height did not vary in a consistent manner at increasing distances from watercourses (average ± SD = 19.6 ± 7.5 cm, Fig. 2). In plantation areas, vegetation height outside planted stands was higher than in pasture areas (80.0 ± 27.4 cm), while inside the eucalypt plantations it was much lower (5.8 ± 6.9 cm). Within eucalypt stands, there was a plantation edge effect on vegetation height, with vegetation height in the transects near plantation edges (9.3 ± 8.2 cm) higher than those further away from the edge (2.4 ± 2.1 cm, Fig. 2). A total 135 species distributed in 79 genera and 30 botanical families were censused (Supplementary material related to this article (Appendix S1) is contained in the online version at http://dx.doi.org/10.1016/j.jnc.2012.10.002). Of these, 118 were identified to the species level. Poaceae, Asteraceae, Cyperaceae, Fabaceae, and Apiaceae were the most speciose families. Eragrostis plana (Poaceae), a common invader of degraded pastures,

Fig. 3. Non-metric multidimensional scaling (NMDS) ordination of herbaceous communities in grasslands with different management histories in southern Brazil.  = Pasture grasslands in permanent protection areas,  = pasture grasslands in unprotected areas, 䊉 = grasslands in permanent protection areas in the vicinity of eucalypt plantations, and  = eucalypt plantations.

was the only exotic species. After controlling for spatial effects, both eucalypt plantations and distance to watercourses had significant effects on herbaceous species composition, as well as the interaction of these two factors (Table 1). Pairwise comparisons controlling for spatial effects (not shown) revealed that the species composition did not differ between plots in permanent protection and unprotected pasture grasslands (PERMANCOVA, F = 1.63, R2 = 0.03, P = 0.0652). In the eucalypt plantation sites, species composition differed significantly between permanent protection areas and planted areas (F = 11.93, R2 = 0.16, P = 0.0002). A significant compositional difference was also found between plots in pasture and plantation permanent protection areas (F = 29.42, R2 = 0.31, P = 0.0084), as well as between planted eucalypt and unprotected pasture grasslands (F = 34.30, R2 = 0.34, P = 0.027). The first ordination axis in the NMDS (final solution with four dimensions, stress = 11.67, Fig. 3) clearly reflected compositional differences in herbaceous communities between pasture and plantation areas, with pasture areas to the right, and eucalypt areas to the left. The second ordination axis was significantly correlated with distance to the watercourses (R2 = 0.99, P = 0.000999, Fig. 3), with pasture plots clumped in the lower part of the ordination space and eucalypt plots vertically scattered. The exotic invader species Eragrostis plana was not found in plantations. Its abundance was much higher in pasture (46.5 ± 33.6% cover) than in permanent protection areas in the vicinity of eucalypt plantations (0.1 ± 0.7% cover, Kruskal–Wallis ANOVA on arcsin-transformed data, H = 90.63, df = 2, P < 2.2 × 10−16 ). Indicator Species Analyses revealed 16 diagnostic indicator species for pasture grasslands, 20 for grasslands in permanent protection areas in the vicinity of eucalypt plantations, and six for planted stands (Table 2).

Table 1 PERMANCOVA of management regime (pasture or eucalypt plantation) and distance to watercourses influencing herbaceous species abundances. Source of variation

df

SS

F

R2

P

Management regime Distance to watercourses Spatial structure Management × distance to watercourses Residuals Total

1 1 9 1 107 119

9.82 1.60 3.49 1.54 20.57 37.03

51.12 8.33 2.02 8.00

0.27 0.04 0.09 0.04 0.56 1.00

0.0002 0.0002 0.0002 0.0002

Please cite this article in press as: Souza, A. F., et al. Afforestation effects on vegetation structure and diversity of grasslands in southern Brazil: The first years. Journal for Nature Conservation (2012), http://dx.doi.org/10.1016/j.jnc.2012.10.002

ARTICLE IN PRESS

G Model JNC-25260; No. of Pages 7

A.F. Souza et al. / Journal for Nature Conservation xxx (2012) xxx–xxx

5

Table 2 Indicator species for grasslands subjected to different degrees of influence of eucalypt plantations based on indicator species analysis. Figures show indicator values (P-values in parentheses). PPA plantation = permanent protection areas in the vicinity of plantations. Species

Pasture

Fimbristylis dichotoma Sporobolus indicusa Axonopus affinis Dichondra sericea Vernonia squarrosa Andropogon selloanus Paspalum notatum Elephantopus mollisa Agalinis communis Desmodium incanum Coelorhachis selloana Eryngium horridum Cuphea glutinosa Pterocaulon angustifolium Baccharis trimera Setaria parvifloraa Aristida laevis Eupatorium ivifolium Eriocaulon gomphrenoides Paspalum guenoarum Eupatorium candolleanum Sacciolepis campestris Mikania micrantha Tibouchina gracilis Andropogon lateralis Andropogon virgatus Rhynchospora barrosiana Fuirena incompleta Sorghastrum agrostoides Rhynchospora tenuis Briza poaemorpha Axonopus argentinus Eryngium ebracteatum Schizachyrium microstachyuma Eragrostis airoides Digitaria ciliarisa Ipomoea trilobaa Hydrocotyle exigua Aspilia montevidensis Piptochaetium montevidense Scutellaria racemosaa Richardia brasiliensis

0.98 (0.001) 0.96 (0.001) 0.95 (0.001) 0.94 (0.001) 0.91 (0.001) 0.91 (0.001) 0.90 (0.001) 0.88 (0.002) 0.88 (0.001) 0.88 (0.001) 0.81 (0.025) 0.79 (0.001) 0.75 (0.045) 0.66 (0.005) 0.56 (0.030) 0.56 (0.001)

a

PPA Plantation

Plantation

0.96 (0.006) 0.96 (0.001) 0.95 (0.031) 0.95 (0.014) 0.95 (0.006) 0.93 (0.001) 0.92 (0.036) 0.92 (0.001) 0.91 (0.008) 0.91 (0.008) 0.89 (0.032) 0.89 (0.006) 0.89 (0.001) 0.86 (0.015) 0.84 (0.026) 0.83 (0.029) 0.81 (0.026) 0.71 (0.016) 0.48 (0.001) 0.15 (0.020)

Fig. 4. Distribution of growth forms between pasture and eucalypt grasslands in southern Brazil. PPA = permanent protection area; TG = tussock grasses; PH = prostrate herb; PS = prostrate subshrub; EH = erect herb; ES = erect subshrub.

0.95 (0.005) 0.93 (0.001) 0.89 (0.001) 0.88 (0.001) 0.86 (0.013) 0.85 (0.034)

Ruderals.

Overall, 17 ruderal species were present in the studied grasslands (12.9% of the total species). Ruderal abundance differed between management regimes (G = 180.8, df = 1, P < 0.0001). Except for eucalypt permanent protection areas and unprotected grasslands (G = 0.28, df = 1, P = 0.60), where ruderal proportion was lowest (5% of the flora), the proportion of ruderals varied between all management categories (P < 0.001 in all comparisons). Eucalypt plantation was the environment type where ruderals were most abundant (12.5% of the flora). Overall, tussock grasses were by far the most common growth form in the studied grasslands (52.7% of total cover), followed by prostrate herb (14.5%), and prostrate shrub (13.5%). Rosette herbs, vines, and shrubs represented <5% of total cover and were excluded from the statistical analysis. The relative abundances of different growth forms differed significantly between management regimes (G = 4636.7, df = 12, P < 0.0001, Fig. 4). Growth form cover distribution in pasture areas showed a consistent pattern, with graminous herbs being the most common growth form followed by prostrate herbs, prostrate subshrubs, and standing subshrubs (Fig. 4). This pattern differed in eucalypt areas, where the dominance of the graminoid herb growth form increased markedly, as well as standing herbs, which increased from ca. 2% in pasture areas to ca. 11% in eucalypt

areas. These increases occurred mainly at the expense of prostrate herbs and subshrubs, which were significantly reduced. Standing subshrubs decreased in plantation areas but not in permanently protected grasslands close to them. Species density at the plot (1 m2 ) scale varied from 1 to 17 and varied significantly between sampling stations (GLM, F = 8.8, df = 4, P ≤ 0.0001). It was marginally affected by management regime alone (F = 3.2, df = 1, P = 0.076), with plots in plantations or permanent protection areas neighbouring plantations presenting 5.9 ± 2.7 species m−2 , and plots in pasture areas presenting 10.9 ± 2.5 species m−2 . It was significantly affected by distance from watercourses (F = 5.7, df = 1, P = 0.018) and, more importantly, was significantly influenced by the interaction between management regime and distance to watercourses (F = 25.0, df = 1, P ≤ 0.0001). This resulted from the fact that species density was reduced in eucalypt plantations (4.8 ± 2.1 species m−2 ) relative to permanent protected areas (7.0 ± 2.9 species m−2 ). Particularly marked reductions were observed in plots further inside plantations, at 30 m from the plantation edge or 60 m from the watercourses (3.4 ± 1.18 species m−2 ). In contrast, plots at increasing distances from watercourses in pasture areas did not suffer noticeable reductions in species density (11.5 ± 2.3 species m−2 at 60 m from the watercourses). Discussion Despite the large-scale Pampa grasslands conversion to eucalypt plantations in recent years (Brockerhoff et al. 2008; Ceccon & Martínez-Ramos 1999; Jobbágy & Jackson 2004; Nosetto et al. 2005; Teixeira Filho 2008), information on the effects of habitat conversion on native grasslands is scant (but see Pillar et al. 2002; Saraiva & Souza 2012). To our knowledge, our results represent the first attempt to bridge this gap. They point to the unsuitability of eucalypt plantations for native grassland species and to marked community changes in grassland communities around plantations. Furthermore, the overall number and listing of native species is an important register for a poorly surveyed landscape that is increasingly dominated by sylviculture and annual cropfields.

Please cite this article in press as: Souza, A. F., et al. Afforestation effects on vegetation structure and diversity of grasslands in southern Brazil: The first years. Journal for Nature Conservation (2012), http://dx.doi.org/10.1016/j.jnc.2012.10.002

G Model JNC-25260; No. of Pages 7

ARTICLE IN PRESS A.F. Souza et al. / Journal for Nature Conservation xxx (2012) xxx–xxx

6

Results showed that permanent protection areas embedded in eucalypt plantations are suffering ecological degradation due to increased biomass and soil cover, altered species composition, reduced species density, and increased dominance of tussock grasses over other growth forms relative to pasture grasslands. These results hold despite spatial effects. A number of environmental impacts are related to tree plantation establishment and could influence nearby grasslands. These include shade (Crawford 1989), reduced grazing intensity and nutrient management (Buscardo et al. 2008), and competition for water with planted trees during drier months (Florence 1986; Jobbágy & Jackson 2004; Nosetto et al. 2005), among others (Bremer & Farley 2010; Brockerhoff et al. 2008; Ceccon & Martínez-Ramos 1999; Farley 2007; Gibson 2009; White et al. 2000). We agree with Buscardo et al. (2008) that reduced grazing intensity is the most likely process to cause the important alterations in afforested grassland structure. Eucalypt plantations and permanent protection areas around them experience a reduced animal stocking density relative to pasture areas. Reduced grazing is known to allow for biomass accumulation and higher canopy height (Karki et al. 2000; McNaughton 1984), and lead to compositional changes due to reduced herbivore pressure on palatable species (Karki et al. 2000; O’Connor 1994). Intense vegetative reproduction is known to give competitive advantages to species found to indicate pasture areas, like Eryngium horridum (Fidelis et al. 2007). Grazing response has also been found to be associated with plant growth form (Noy-Meir et al. 1989), with tall tussock grasses dominating ungrazed sites, followed by tall erect plants (Jacquemyn et al. 2011). Accordingly, grasslands in the lightly-grazed permanent protection areas we studied presented increased abundance of tall grasses and standing herbs, while prostrate herbs and subshrubs were absent. The reduced diversity we found in these areas relative to pastures agrees with the finding that grasslands subjected to less intensive grazing may have lower diversity due to competitive exclusion of shade-sensitive species by dominant tall grasses (Herrera & Laterra 2011; Jacquemyn et al. 2011; Karki et al. 2000). Experimental work is needed in order to establish the relative contributions of grazing intensity and plantations on floristic composition of structure. Conservation implications As stated in the Methods section, planted and pasture areas shared similar management histories previous to the establishment of eucalypt plantations, and thus the effects we found cannot be attributed to historical land use differences. Two years after plantation establishment, herbaceous vegetation under eucalypt stands was shorter, scarcer, had fewer species, increased proportion of ruderals relative to pasture areas, and fewer life-form types. Pasture plots showed high floristic similarity regardless of distance to watercourses, but altered conditions in plantation areas, both eucalypt stands and nearby permanent protected areas, produced large floristic differences between plots. Floristic variation in Northern Campos follows topographic-moisture gradients, and distinct community types are found on slopes and on wet lowlands (Focht & Pillar 2003). The planted sites we studied were located in lowlands, and their understory floristic distinctiveness is not, thus, attributable to topographic differences. Although this picture may reflect an ongoing colonisation process of planted sites by nearby grasslands, eucalypt understories are unlikely to represent viable habitats for most native grassland species (Bremer & Farley 2010; Brockerhoff et al. 2008). This is mainly due to stressful conditions that may include reduced irradiance intensity (Pillar et al. 2002), altered red:far red ratio (Crawford 1989), allelopathy, the physical barrier of litter to germination (Bremer & Farley 2010), intense competition for water with eucalypt trees (Jobbágy & Jackson 2004; Nosetto et al. 2005), and reduced wind velocity that impairs

pollination and seed dispersion of grasses (Gibson 2009). These effects act in synergy and lead to decreased species richness with plantation age after grassland afforestation (Bremer & Farley 2010). Light to moderate livestock grazing is probably needed for permanent protection areas to act as ecological networks for biodiversity conservation (Carvalho & Batello 2009; Overbeck et al. 2007; Questad et al. 2011). Possible limitations of such practice during plantation establishment due to browsing damage of younger eucalypt trees need to be investigated. Acknowledgments Financial support was provided by Celulose Riograndense/CMPC (formerly ARACRUZ Celulose), in which plantations part of the data were collected. Funding sources had no part in the study design, in the collection, analysis and interpretation of data, in the writing of the report, nor in the decision to submit the paper for publication. We also thank Jose Antonio Carchedi for permission to work in his property, in which pasture areas were established, and Alcides M. P. Silva Neto for logistical support during work in those areas. Maurem K. L. Alves and Elias from CMPC kindly provided information about the studied areas as well as logistic support in initial field surveys. Ronaldo Patias and Thais Monero provided valuable help during fieldwork. We are indebted to Dr. Ilsi Iob Boldrini for instrumental help in botanical identification of specimens. We thank Tiago C. de Marchi for valuable assistance in the field and help in the botanical identification of specimens. We wish to thank Andrew Rayburn, Charlotte Reemts, Gerard Overbeck, Jean Carlos Budke, Juliano Morales de Oliveira and Laura Six for reading and commenting an earlier version of the manuscript. References Anderson, M. J. (2001). A new method for non-parametric multivariate analysis of variance. Austral Ecology, 26, 32–46. Behling, H., & Pillar, V. D. (2007). Late quaternary vegetation, biodiversity and fire dynamics on the southern Brazilian highland and their implication for conservation and management of modern Araucaria forest and grassland ecosystems. Philosophical Transactions of the Royal Society B: Biological Sciences, 362, 243–251. APG II. (2003). An update of the Angiosperm Phylogeny Group classification for orders and families of flowering plants, APG II. Botanical Journal of the Linnean Society, 141, 399–436. Barros, M. (1960). Las Ciperaceas de Santa Catalina. Sellowia, 12, 181–450. ˜ F. (2004). Identificación de áreas valiosas de pastizal en las Bilenca, D., & Minarro, pampas y campos de Argentina, Uruguay y sur de Brasil (AVPs). Buenos Aires: Fundación Vida Silvestre Argentina. Boechat, S. C., & Valls, J. F. M. (1986). O gênero Eragrostis von Wolf (Gramineae, Chloridoideae). no Rio Grande do Sul, Brasil. Iheringia série Botânica, 34, 51–130. Boechat, S. C., & Valls, J. F. M. (1991). As espécies do gênero Sporobolus R. Br. (Gramineae-Chloridoideae) no Rio Grande do Sul. Iheringia série Botânica, 41, 9–45. Boldrini, I. I. (1997). Rio Grande do Sul Grasslands, physiognomy and occupational problems. Boletim do Instituto de Biociências da UFGRS, 56, 1–39. Borcard, D., & Legendre, P. (2002). All-scale spatial analysis of ecological data by means of principal coordinates of neighbour matrices. Ecological Modeling, 153, 51–68. Brazil. (1973). Levantamento de reconhecimento dos solos do Estado do Rio Grande do Sul. Boletim Técnico n◦ 30, Ministério da Agricultura. Bremer, L. L., & Farley, K. A. (2010). Does plantation forestry restore biodiversity or create green deserts? A synthesis of the effects of land-use transitions on plant species richness. Biodiversity and Conservation, 19, 3893–3915. Brockerhoff, E. G., Jactel, H., Parrotta, J. A., Quine, C. P., & Sayer, J. (2008). Plantation forests and biodiversity, oxymoron or opportunity? Biodiversity and Conservation, 17, 925–951. Burkart, A. (1948). Las espécies de Mimosa de la Flora Argentina. Darwiniana, 8, 9–257. Burkart, A. (1974). Flora Ilustrada de Entre Rios (Argentina). Parte VI, Dicotiledoneas Metaclamídeas (Gamopétalas), B, Rubiales, Cucurbitales, Campanulales (incluso Compuestas). Tomo 6. Buenos Aires: Colección Cientifica del INTA. Burkart, A. (1979). Flora Ilustrada de Entre Rios (Argentina). Parte V, Dicotiledoneas Metaclamídeas (Gamopétalas)., A, Primulales a Plantaginales. Tomo 6. Buenos Aires: Colección Cientifica del INTA. Burkart, A. (1987). Flora Ilustrada de Entre Rios (Argentina). Parte III, Dicotiledoneas Arquiclamídeas (Dialipétalas)., A, Salicales a Rosales (Leguminosas). Tomo 6. Buenos Aires: Colección Cientifica del INTA.

Please cite this article in press as: Souza, A. F., et al. Afforestation effects on vegetation structure and diversity of grasslands in southern Brazil: The first years. Journal for Nature Conservation (2012), http://dx.doi.org/10.1016/j.jnc.2012.10.002

G Model JNC-25260; No. of Pages 7

ARTICLE IN PRESS A.F. Souza et al. / Journal for Nature Conservation xxx (2012) xxx–xxx

Burkart, A., & Bacigalupo, N. M. (2005). Flora Ilustrada de Entre Rios (Argentina). Parte IV, Dicotiledóneas Arquiclamídeas, B, Geraniales a Umbelliflorales. Tomo 6. Buenos Aires: Colección Cientifica del INTA. Buscardo, E., Smith, G. F., Kelly, D. L., Freitas, H., Iremonger, S., Mitchell, F. J. G., et al. (2008). The early effects of afforestation on biodiversity of grasslands in Ireland. Biodiversity and Conservation, 17, 1057–1072. Canto-Dorow, T. S. (2001). O gênero Digitaria Haller (Poaceae-Panicoideae-Paniceae). no Brasil (PhD Thesis, Universidade Federal do Rio Grande do Sul). Carneiro, A. M. (1998). Vegetac¸ão ruderal da Vila de Santo Amaro, Mun. Gal. Câmara, RS, BR. Ruas, muros, terrenos baldios e passeios públicos (MSc Thesis, Universidade Federal do Rio Grande do Sul). Carvalho, P., & Batello, C. (2009). Access to land, livestock production and ecosystem conservation in the Brazilian Campos biome, the natural grasslands dilemma. Livestock Science, 120, 158–162. Ceccon, E., & Martínez-Ramos, M. (1999). Aspectos ambientales referentes al estabelecimiento de plantaciones de eucalipto de gran escala en áreas tropicales, aplicación al caso de México. Interciencia, 24, 352–359. Crawford, R. M. M. (1989). Studies in plant survival. Ecological case histories of plant adaptation to adversity. Oxford: Blackwell Scientific. Damé, P. R. V., Rocha, M. G., Quadros, F. L. F., & Pereira, C. F. S. (1999). Estudo florístico de pastagem natural sob pastejo. Empresa Brasileira de Pesquisa Agropecuária, 5, 45–49. Dufrêne, M., & Legendre, P. (1997). Species assemblages and indicators species, the need for a flexible asymmetrical approach. Ecological Monographs, 67, 345–366. Farley, K. A. (2007). Grasslands to Tree Plantations, forest transition in the Andes of Ecuador. Annals of the Association of American Geographers, 97, 755–771. Ferri, M. G., Menezes, N. L., & Monteiro, W. L. (1981). Glossário ilustrado de botânica. São Paulo: Nobel. Fidelis, A., Overbeck, G., Pillar, V. D., & Pfadenhauer, J. (2007). Effects of disturbance on population biology of the rosette species Eryngium horridum Malme in grasslands in southern Brazil. Plant Ecology, 195, 55–67. Florence, R. G. (1986). Cultural problems of eucalyptus as exotics. Commonwealth Forestry Research, 65, 141–163. Flores, A. I. P., & Valls, J. F. M. (1992). O gênero Sorghastrum Nash (Gramineae, Andropogoneae). no Rio Grande do Sul. Iheringia série Botânica, 42, 25–54. Focht, T., & Pillar, V. D. (2003). Spatial patterns and relations with site factors in a campos grassland under grazing. Brazilian Journal of Biology, 63, 423–436. Fonseca, C. R., Ganade, G., Baldissera, R., Becker, C., Boelter, C. R., Brescovit, A. D., et al. (2009). Towards an ecologically-sustainable forestry in the Atlantic Forest. Biological Conservation, 142, 1209–1219. Geldenhuys, C. J. (1997). Native forest regeneration in pine and eucalypt plantations in Northern Province, South Africa. Forest Ecology and Management, 99, 101–115. Gibson, D. J. (2009). Grasses and grassland ecology. Oxford: Oxford University Press. Gomar, E. P., Reichert, J. M., Reinert, D. J., Prechac, F. G., Berreta, E., & Marchesi, C. (2004). Semeadura direta de forrageiras de estac¸ão fria em campo natural submetido a aplicac¸ões de herbicidas. II Composic¸ão botânica. Ciência Rural, 34, 769–777. Gonc¸alves, E. G., & Lorenzi, E. G. (2007). Morfologia vegetal. São Paulo: Instituto Plantarum. Grime, J. P. (1977). Evidence for the existence of three primary strategies in plants and its relevance for ecological and evolutionary theory. American Naturalist, 111, 1169–1194. Guglieri, A., & Longhi-Wagner, H. M. (2000). Gramineae – Paniceae Gênero Panicum L. B. Boletim do Instituto de Biociências da UFRGS, 59, 1–163. Heringer, I., & Jacques, A. V. A. (2001). Adaptac¸ão das plantas ao fogo, enfoque na transic¸ão floresta-campo. Ciência Rural, 31, 1085–1090. Herrera, L. P., & Laterra, P. (2011). Relative influence of size, connectivity and disturbance history on plant species richness and assemblages in fragmented grasslands. Applied Vegetation Science, 14, 181–188. Irgang, B. E. (1974). Umbelliferae II – Gênero Eryngium L. Flora Ilustrada do Rio Grande do Sul. Boletim do Instituto Central de Biociências, 32, 1–86. Izaguirre, P., & Beyhaut, R. (2003). Las leguminosas en Uruguay y regiones vecinas. Montevideo: Hemisferio Sur.

7

Jacquemyn, H., Mechelen, C. V., Brys, R., & Honnay, O. (2011). Management effects on the vegetation and soil seed bank of calcareous grasslands, an 11-year experiment. Biological Conservation, 144, 416–422. Jobbágy, E. G., & Jackson, R. B. (2004). Groundwater use and salinization with grassland afforestation. Global Change Biology, 10, 1299–1312. Karki, J. B., Jhala, Y. V., & Khanna, P. P. (2000). Grazing Lawns in Terai Grasslands, Royal Bardia Nationalna, Park, Nepal. Biotropica, 32, 423–429. Legendre, P., & Legendre, L. (1998). Numerical ecology. Amsterdam: Elsevier. Longhi-Wagner, H. M. (1987). Gramineae Tribo Poeae. Boletim do Instituto de Biociências da UFRGS, 41, 1–191. Lorenzi, H. (2000). Plantas daninhas do Brasil, terrestres, aquáticas parasitas e tóxicas (3rd ed.). Nova Odessa: Plantarum. Luz, C. L. (2004). Taxonomia da família Juncaceae Juss. no Rio Grande do Sul, Brasil (MSc Thesis, Universidade Federal do Rio Grande do Sul). MacFadden, B. J. (1997). Origin and evolution of the grazing guild in New World terrestrial mammals. Trends in Ecology & Evolution, 12, 182–187. McCune, B., & Grace, J. B. (2002). Analysis of ecological communities. Gleneden Beach: MJM Software. McNaughton, S. J. (1984). Grazing lawns, animals in herds, plant form, and coevolution. American Naturalist, 124, 863–886. Nosetto, M. D., Jobbagy, E. G., & Paruelo, J. M. (2005). Land-use change and water losses, the case of grassland afforestation across a soil textural gradient in central Argentina. Global Change Biology, 11, 1101–1117. Noy-Meir, I., Gutman, M., & Kaplan, Y. (1989). Responses of Mediterranean plants to grazing and protection. Journal of Ecology, 77, 290–310. O’Connor, T. G. (1994). Composition and population responses of an African grassland to rainfall and grazing. Journal of Applied Ecology, 31, 155–171. O’Leary, N., Múlgura, M. E., & Morrone, O. (2007). Revisión taxonómica de las especies del género Verbena (Verbenaceae), série Pachystachyae. Annals of the Missouri Botanical Garden, 94, 571–621. Overbeck, G. E., Müller, S. C., Fidelis, A., Pfadenhauer, J., Pillar, V. D., Blanco, C. C., et al. (2007). Brazil’s neglected biome, the south Brazilian campos. Perspectives in Plant Ecology and Evolution, 9, 101–116. Parrotta, J. A., Turnbull, J. W., & Jones, N. (1997). Catalyzing native forest regeneration on degraded tropical lands. Forest Ecology and Management, 99, 1–7. Peel, M. C., Finlayson, B. L., & McMahon, T. A. (2007). Updated world map of the Köppen–Geiger climate classification. Hydrology and Earth System Sciences, 11, 1633–1644. Pillar, V. P., Boldrini, I. I., & Lange, O. (2002). Spatial patterns of grassland communities under eucalyptus plantation. Pesquisa agropecuária brasileira, 37, 753–761. Porto, M. L., Callegari, S., Miotto, S. T. S., Waechter, J. L., & Detoni, M. L. (1977). Rubiaceae – Tribo Spermacoceae. Flora Ilustrada do Rio Grande do Sul. Boletim do Instituto de Biociências da UFRGS, 35, 1–114. Questad, E. J., Foster, B. L., Jog, S., Kindscher, K., & Loring, H. (2011). Evaluating patterns of biodiversity in managed grasslands using spatial turnover metrics. Biological Conservation, 144, 1050–1058. Rosengurtt, B. (1970). Gramineas uruguayas. Montevideo: Universidad de la Republica. Samways, M. J., Bazelet, C. S., & Pryke, J. S. (2009). Provision of ecosystem services by large scale corridors and ecological networks. Biodiversity and Conservation, 19, 2949–2962. Saraiva, D. D., & Souza, A. F. (2012). Effects of environmental factors and plantation forests on endangered cactus diversity and composition in subtropical South American grasslands. Perspectives in Plant Ecology and Evolution, http://dx.doi.org/10.1016/j.ppees.2012.02.002 Sokal, R. R., & Rohlf, F. J. (1995). Biometry the principles and practice of statistics in biological research. Nova York: W.H. Freeman. Streck, E. V., Kämpf, N., Dalmolin, R. S. D., Klant, E., Nascimento, P. C., Schneider, P., et al. (2008). Solos do rio grande do sul. Porto Alegre: EMATER/RS. Teixeira Filho, A. (Ed.). (2008). Eucaliptais, que Rio Grande do Sul desejamos?. Pelotas: Authors. White, R., Murray, S., & Rohweder, M. (2000). Pilot analysis of global ecosystems, grassland ecosystems technical report. Washington: World Resources Institute. Zuloaga, F. O., & Morrone, O. (2005). Revisión de las especies de Paspalum para América del Sur Austral. Saint Louis: Missouri Botanical Garden Press.

Please cite this article in press as: Souza, A. F., et al. Afforestation effects on vegetation structure and diversity of grasslands in southern Brazil: The first years. Journal for Nature Conservation (2012), http://dx.doi.org/10.1016/j.jnc.2012.10.002