Gasta?o Viegas de Pinho Ju′nior?Andre′R.Terra Nascimento?Barbara Tahara Valverde?Lucas H.Clemente

Brazilian savanna re-establishment in a monoculture forest: diversity and environmental relations of native regenerating understory in Pinus caribaea Morelet.stands

Gasta?o Viegas de Pinho Ju′nior1?Andre′R.Terra Nascimento2?Barbara Tahara Valverde3?Lucas H.Clemente2

In this paper we analyze and compare natural regeneration in shrub-tree community in three areas below Pinus caribaea stands and a natural area(cerrado sensu stricto).We also analyze the influence of biophysical and environmentalvariables on the distribution of regenerating shrub-tree species.The areas were analyzed and compared in relation to dispersalsyndromes as well,being zoochory and anemochory the prevailing syndroms in the four areas. The sites below P.caribaea show a heterogeneous regeneration with the number of species ranging from 18 to 42. We analyzed the influence of the biophysical and environmental variables performing a canonical correspondence analyses,being found significantvalues forvariables distance from remnant and basal area.The species Siparuna guianensis,Miconia albicans,Xylopia aromatica, Sclerolobium paniculatum,Casearia sylvestris and P. caribaea were the most importante ones in the four areas.

Native regeneration·Cerrado·Commercial forests·Multivariate analysis·Plant ecology

Introduction

The differentphysiognomies thatcompose Cerrado biome, bordering a variety of another Brazilian biomes,help to explain the high biodiversity rates found in Cerrado,which holds more than 10,000 vascular plant species already catalogued(Machado et al.2004;Felfili and Silva-Ju′nior 2005;Mendonc?a et al.2008).Cerrado occurs frequently in savanna-like forms,but also as forest(closed arboreal canoppy),woodland(open arboreal canoppy),scrub and grassland foms(Gottsberger and Silberbauer-Gottsberger 2006),being cerrado sensu stricto,the savanna-like vegetation,which covers more than 60%oforiginalbiome area(Eiten 2001).

Nowadays,there are approximately 600,000 km2of cultivated land with African grasses in Cerrado.The corresponding cultivated area with annual crops equals 40% of the whole nationalproduction of soybean,22%of corn and 33%of cotton,being Cerrado therefore,regarded as a paramount area for Brazilian agribusiness(Ratter et al. 1997;Klink and Moreira 2002;Klink etal.2008;Lima and Silva 2008).

Despite the positive effects arising from the increase of economic activities in Cerrado biome,the fastexpansion of a highly intensive,mechanized and commercialagriculture has taken its toll:The employment of agrochemicals plus intensive burning process and deforestation has lead to biodiversity loss,rivers and groundwater pollution,soil erosion and imbalance in carbon stocks and flows,also causing climate alterations in a regional scale(Li and Zhang 2000;Ratter et al.2006;Klink et al.2008;Silva et al.2008).

Commercialforestry plays an importanteconomicalrole amongst agribusiness activities in Brazilian Cerrado,particularly in Minas Gerais state,which holds the largestareaof planted forests in Brazil(ABRAF 2011;MMA 2010). The commercial forest plantations have an array of particular features when compared to other monocultures such as,long cycle,reduced using of pesticides,high canopy cover and soil fertility changes that favour the removal of invasive dominant grasses.Hence,this monoculture has shown a higher permeability to flora and fauna than other agricultural environments,providing habitat for seed dispersers which in turns lead to recolonization of its understory by native plantspecies,which has led severalauthors to suggest that planted forest species would play the same role that pioneer species do under natural conditions,facilitating the stablishment of a regeneration layer composed by native species in their understory(Oliveira and Moreira 1992;Parrotta et al.1992;Lugo 1997;Parrotta etal.1997;Vianietal.2010).Therefore,studies on natural regeneration in the understory of forest plantations can provide data and basic information for development of mixed forestry exploitation models,using the understory vegetation found in these plantations as a genetic resource, providing saplings and seedlings for ex situ conservations projects.These studies can also provide vital information on how to take advantage of properties of fast growing exotic trees as nurse species forrestoration purposes,using the potential of some exotic species in catalyzing a secondary successional process(Aubert and Oliveira-Filho 1994;Lugo 1997;Vianietal.2010,Durigan et al.2011).

In this context,this study aimed to:compare the regeneration found in the understory of Pinus caribaea stands to the regeneration found in a cerrado sensu stricto remnant;analyze the relations between shrub-tree species distribution and some environmental and biophysical variables measured in the four sites.

Materials and methods

Study area

The study area consists of four sites located at Floresta do Lobo Farm,situated in the westpartofMinas Geraisstate,in Uberla?ndia municipality at a region known as Tria?ngulo Mineiro(Figs.1,2).The selected sites consist of three P. caribaea stands—Site1(48°8′13.85′′W;19°2′42.05′′S),Site 2(48°8′13.85′′W;19°3′51.68′′S),Site 3(48°10′41.19′′W; 19°5′13.91′′S)—and a cerrado sensu stricto remnant—Site 4 (48°05′26.13′′W;19°02′29.01′′S).The farm area includes large commercialstandsof P.caribaea and Eucalyptus sp.as wellas agriculturalcrops,pastures and cerrado sensu stricto remants atdistinctstages ofnaturalregeneration.The main physiognomies in the area are:cerrado sensu stricto,open shrubland,riparian forestand isolated fragments oftropical semi-deciduous forest.

Vegetation sampling

We analyzed the shrub-tree regeneration(Fig.2)in the understory of three selected P.caribaea stands(Sites 1,2 and 3)and one cerrado remnant(Site 4).For this purpose, we set in each site,four parallel lines 30 m distant from each other,in which we allocated twenty 10×10 m plots (100 m2)—five 10 m equidistantplots perline.These plots were assigned as Class IIplots.Inside each Class IIplotwe allocated a sub-plot 2×2 m(4 m2)—being these subplots assigned as Class I plots.We followed the same methodology when allocating the plots in cerrado sensu stricto remmant(Site 4).Therefore,60 plots were allocated in P.caribaea stands and 20 plots were allocated in cerrado sensu stricto remnant,a total of 80 plots in the four sites.

Biophysical and environmental variables

To evaluate the environmentalrelations,we measured five environmental and biophysical variables:average litter thickness,canopy coverage,basal area,declivity and distance from a propagule source(nearest natural remnant from each site).These variables were measured in each Class II plot in the four areas exceptthe variable distance from propagule source,which was measured only in the three stands—Site 1,2 and 3.

We measured the litter thickness using a 30 cm ruler. The measurements were taken at three points inside the plots(center and the two ends,following the lines direction).The average litter thickness per plotwas obtained by adding up these three measurements.The canopy coverage percentage was measured by using a convex spherical densiometer(Lemmon 1957).The basalarea was measured using a Bitterlich Relascope(Bitterlich 1984).For this purpose we count all individuals fitting in line 2 of the mirror representing in this case the basal area of each plot (Nascimento etal.2004).The declivity of Class Iplots was measured using a Sunto Clinometer.

Dispersal syndromes description

By analyzing morphological features,Van der Pijl (1982),classified fruits into three large groups according to dispersal characteristics:zoochory when diaspores are adapted to dispersal by animals,anemochory when diaspores are adapted to dispersal by wind and authochory when diaspores are dispersed by gravity or explosive dehiscence.We applied this same sort of classifications in this study,using the studies of Batalha(1997),Silva and Tabarelli(2000),Weiser and Godoy(2001),Ribeiro and Tabarelli(2002),Toppa(2004)and Lenza and Klink(2006)as reference.

Fig.1 Location map of study area in Brazil,at Minas Gerais state and in Uberla?ndia municipality,Tria?ngulo Mineiro region(Source:Henriques 2010)

Vegetation structure and comparison between areas

The corresponding data for density,basalarea and species richness were tested for normality using Lilliefors test.As the data did not fit the normal distribution we opted for using nonparametric statistics,employing Kruskal–Wallis (for comparison between vegetation structure parameters) and Chi squared test—for comparison between dispersal syndromes(Zar 1999).Both procedures were performed using the statisticalsoftware Systat10(SPSS 2000).

Canonical correspondence analysis(CCA) between vegetation and environmental and biophysical variables

From the shrub-tree species density data for site,we developed density matrices per plotas wellas a matrix with environmental and biophysical variables(declivity,basal area,canopy cover,litter thickness and distance from natural remnant).The density matrices of the species in the four areas were correlated with environmental and biophysicaldata using CCA(Kentand Coker1992;ter Braack 1995;Silva Ju′nior et al.1996;ter Braack and Smilauer 2003).This kind of multivariate analysis allows to investigate the relation between the aforementioned environmental and biophysical variables and the distribution of shrub-tree species in a plant community(Kent and Coker 1992;Oliveira-Filho etal.1997;Oliveira-Filho and Fontes 2000;ter Braak and Smilauer 2003;Jonasˇova′etal.2006).

Results

Comparison of natural regeneration patterns between the four areas

The density values for sampled shrub-tree individuals in Class II ranged from 261 to 2066,the number of species from 36 to 90,the number of families from 23 to 33.The estimated Shannon diversity(H’)ranged from 1.3 to 3.0 at the four sites(Table 1).The density of individuals (Fig.3B)showed significant differences between regeneration in P.caribaea stands and regeneration in cerrado remnant(H=65.79;p＜0.0001).The basal area estimates and species richness also showed significant differences between areas(H=15.51;p＜0.0014)and (H=74.60;p＜0.0001)respectively,which indicates a considerable heterogeneity between naturalregeneration in the studied areas(Fig.3A).

When analysing the percentage of dispersal syndromes for shrub-tree species in the four areas,we found a similar pattern,being zoochory the predominant dispersal syndrome,represented by values over 70%in the four sites (Fig.4).Anemochory was the second most frequent syndrome in this study,showing percentage values higher than 20%in the areas.The less frequent syndrome was authocory,which showed percentage values lower than 8% and null values in two stand sites.We found a significant difference for dispersal syndromes between areas χ2=19.23(p＜0.05).

Fig.2 Some aspects of natural regeneration in Pinus caribaea understory:distance from propagules source on the left (A),new individuals emerging from Pinus caribaea litter(B), shrub tree community at secondary succession stage(B–C)and regeneration in natural remnant(D),Floresta do Lobo Farm,Uberla?ndia,MG

Table 1 Comparative data on natural regeneration between three sites below Pinus caribaea stands and a natural remnant. Floresta do Lobo farm, Uberla?ndia,MG

Relations between regeneration and environmental and byopshicalvariables

The relation species/environment explained from 57 to 83%of the variation in data gathered in the four areas (Tables 2,3,4).This relation also can be seen through Monte Carlo Permutation Test,which was significant for Site 1(p＜0.05;F=22.30)and Site 3(p＜0.05; F=1.724).The most important variables in the canonical ordering were distance from propagules source (natural remnant)and basal area(Figs.5,6).

These variables showed canonical correlation values higherthan 0.7,indicating a strong correlation with species distribution and significantprobability values according to Monte Carlo test(p＜0.05).

The high proportion of animal dispersed species corroborates the importance of the variable distance from propagules source(Figs.5,6).The percentage of zoochoricspecies in this study varies from 53 to 71%,which is the proportion usually found for cerrado sensu stricto and tropicalareas(Puig 2008;Pizo 2012).

Fig.3 Box plots of shrub-tree species per plot(A)and individuals per plot(B)in Sites 1,2,3(Pinus caribaea understory)and the natural area(cerrado sensu stricto remnant),Floresta do Lobo Farm, Uberla?ndia,MG

Table 2 Results of Canonical Correspondence Analysis(CCA)for regeneration at site 1(Pinus caribaea understory),Floresta do Lobo farm,Uberla?ndia,MG

Discussion

The native shrub-tree species showing the highest regeneration potential in the three P.caribaea stands share a common feature:environmental plasticity;Siparuna guianensis and Miconia albicans occur in cerrado sensu stricto and in the understory offorestformations like cerrada?o and mesophytic forests(Ratter et al.2003;Mendonc?a et al. 2008).M.albicans is commonly found in savanna formations in Central Brazil and also occurs in forest formationssuch as riparian forests and cerrada?o,being commonly found in disturbed areas as well.Xylopia aromatica occurs in cerrada?o,cerrado sensu stricto,dry forest,and riparian forest.In addition,the latter species is commonly found in large populations at cerrada?o and cerrado stricto sensu areas under human interference,as well as Maprounea guianensis which also occurs in riparian forests and cerrado formations(Arau′jo et al.1997;Ribeiro and Walter 2008; Mendonc?a etal.2008;Arau′jo etal.2011).

Fig.4 Dispersal syndromes found in shrub-tree community in three regenerating areas—Pinus caribaea understory(A–C)and a natural area—cerrado sensu stricto remnant(D), Floresta do Lobo Farm, Uberla?ndia,MG.Zoo-zoochory, Ane-Anemochory and Aut-Authocory

Table 3 Results of Canonical Correspondence Analysis(CCA)for regeneration at site 2(Pinus caribaea understory),Floresta do Lobo farm,Uberla?ndia,MG

Table 4 Results of Canonical Correspondence Analysis(CCA)for regeneration at Site 3(Pinus caribaea understory),Floresta do Lobo farm,Uberla?ndia,MG

As we expected,the natural area showed the highest values for density and species richness when compared to the three P.caribaea stands(Fig.3).Studies on natural regeneration emphasize the strong regeneration potentialof naturalareas(Carvalho etal.2008;Costa and Arau′jo 2001; Ribeiro and Felfili2009;Luz etal.2008).The regeneration found in the understory of P.caribaea stands,however, resembles degraded or disturbed areas lacking a dominant tree species(Pinus spp.or Eucalyptus spp.).Therefore,the equability and species richness found in the P.caribaea stands are similar to those found by Vale et al.(2009)for plant community in a secondary cerrada?o remnant,showing signs ofstrong human disturbance in and to those found by Santos and Vieira(2005),for three areas of cerrado sensu stricto showing distinct levels of human disturbance.

Fig.5 Canonical Correspondence Analysis between species and environmental and biophysical variables(CCA)for Class II plots in Site 1(A)and Site 2(B)Floresta do Lobo Farm,Uberla?ndia,MG

Fig.6 Canonical Correspondence Analysis between species and environmental and biophysical variables(CCA)for Class II plots in Site 3(A)and Site 4—Cerrado remnant(B)Floresta do Lobo Farm, Uberla?ndia,MG

When analyzing the influence of adjacent vegetation on native understory regeneration in Eucalyptus sp.and Pinus sp.stands,Aubertand Oliveira-Filho(1994)found a strong relation between adjacent vegetation and the density gradients in the understory of stands.We found a similar relation in our study,and the low percentage of authocory followed by a high proportion of zoochory and anemochory help to explain the strong influence of naturalremnants on the understory regeneration bellow Pinus sp.stands,since these remnants can actas a refuge for native fauna which–in some cases and at different levels–tend to explore the nearby Pinus sp.stands,carrying with them the propagules of native species.Apart from being a fauna refuge,the natural remnants also work as a propagule source for anemochoric species,which in turns reach the stands and take advantage of the facilitating effects Pinus sp.might have on some native species,either anemochoryc or zoochoric ones(Aubertand Oliveira-Filho 1994;Gonzales and Nakashizuka 2010;Pizo 2012).

In summary,the species M.albicans,Matayba guianensis,X.aromatica,Dalbergia miscolobium,S.guianensis and M.guianensis were the common species between the four studied sites,suggesting a strong regeneration potential and tolerance for competition with already established P.caribaea individuals.The ecological features of these species such as their ability to thrive in acid soils found in P.caribaea stands might be useful when it comes to develop restoration protocols for degraded areas as well as enrichment programs for cerrado sensu strico vegetation in order to increase its biodiversity.

Our data indicates thatunderstory of P.caribaea stands can shelter native species catalyzing a secondary successional process with values of density and species richness resembling degraded or disturbed areas of cerrado sensu stricto.The regeneration found in the study sites is strongly related to proximity ofa propagule source(naturalremnant), presence of litter and basalarea of surrounding vegetation.

AcknowledgmentsWe would like to thank Capes for the scholarship provided for the first author;to managers and employees of Floresta do Lobo farm for supporting our work;to professors Jean Carlos Santos(UFU)and Solon Jonas Longhi(UFSM)for the importantsuggestions given;to professors Rosana Romero(UFU),Ivan Schiavini(UFU)and Glein M.Arau′jo(UFU)for helping to identify the collected botanical species.We would also like to thank Luzencort Junior,Mariana Crosara and Monica Zuffifor their hard work and assistance in field.

ABRAF(Associac?a?o Brasileira dos Produtores de Florestas Plantadas)(2011)Anua′rio Estat?′stico ABRAF.Available at:http:// www.abraflor.org.br/estatisticas.asp.Accessed 29 Aug 2013

Arau′jo GM,Nunes JJ,Rosa AG,Resende EJ(1997)Estrutura comunita′ria de vinte a′reas de cerrado residuais no munic?′pio de Uberla?ndia,MG.Daphne 7:7–14

Arau′jo GM,Nascimento ART,Lopes SF,Rodrigues RF,Ratter JA (2011)Structure and floristics of the arboreal component of a dystrophic Cerrada?o and comparions with other cerrado?es in central Brazil.Edinb J Bot 68:401–418

Aubert E,Oliveira-Filho AT(1994)Ana′lise multivariada da estrutura fitossociolo′gica do sub-bosque de plantios experimentais de Eucalyptus spp.e Pinus spp.em Lavras(MG).Rev A′rvore 18:192–214

Batalha MA(1997)Ana′lise da vegetac?a?o da ARIE Cerrado Pe′-degigante(Santa Rita do Passa Quatro,SP).Dissertac?a?o(Mestrado em ecologia),Universidade de Sa?o Paulo,Sa?o Paulo

Bitterlich W(1984)The relascope idea.Commonwealth Agricultural Bureau,Slough

Carvalho AF,Rodrigues PHV,Kilca VR,Siqueira SA,Arau′jo GM, Schiavini I(2008)Composic?a?o flor?′stica,riqueza e diversidade de um cerrado sensu stricto no sudeste do estado de Goia′s. Biosci J 24:64–72

Costa AA,Arau′jo GM(2001)Comparac?a?o da vegetac?a?o arbo′rea de cerrada?o e de cerrado na Reserva do Panga,Uberla?ndia,Minas Gerais.Acta Bot Bras 15:63–72

Durigan G,Melo GCA,Max JCM,Villas Boas O,Contieri AW, Ramos SV(2011)Manual para a recuperac?a?o da vegetac?a?o de cerrado.SMA,Sa?o Paulo,p 24

Eiten G(2001)Vegetac?a?o natural do Distrito Federal.UnB, SEBRAE,Bras?′lia,p 162

Felfili JM,Silva-Junior MC(2005)Diversidade alfa e beta no cerrado sensu stricto,Distrito Federal,Goia′s,Minas Gerais e Bahia.In: Scariot A,Sousa-Silva JC,Felfili JM(eds)Cerrado:Ecologia, biodiversidade e conservac?a?o.Embrapa,Sa?o Paulo,pp 141–154

Gonzales RS,Nakashizuka T(2010)Broad-leaf species composition in Cryptomeria japonica plantations with respect to distance from natural forest.For Ecol Manag 259:2133–2140

Gottsberger G,Silberbauer-Gottsberger I(2006)Life in the Cerrado: a South American Tropical Seasonal Vegetation,vol.II. Pollination and Seed Dispersal.Reta Verlag,Ulm,p 384

Henriques TR(2010)Geoprocessamento aplicado nas mudanc?as de uso da terra e cobertura vegetal na Fazenda Floresta do Lobo, Uberla?ndia-MG.Monografia.(Bacharelado em Cie?ncias Biolo′gicas)-Universidade Federal de Uberla?ndia,Uberla?ndia,p 33

Jonasˇova′MJ,van Hees A,Prach K(2006)Rehabilitation of monotonous exotic coniferous plantations:A case study of spontaneous establishment of different tree species.Ecol Eng 28:141–148

Kent M,Coker P(1992)Vegetation description and analysis:a practical approach.CRC Press,Boca Raton,p 363

Klink AC,Moreira GA(2002)Past and current human occupation, and land use.In:Oliveira SP,Marquis JR(eds)The cerrados of Brazil:ecology and natural history of a neotropical savanna. Columbia University Press,New York,pp 69–83

Klink AC,Santos GH,Campari SJ,Matsumoto HM,Freitas KG, Baumgarten L(2008)Conservac?a?o dos recursos naturais em terras privadas:O papel das reservas legais no arranjo funcional das paisagens produtivas do bioma Cerrado.In:Sano SM, Almeida SP,Ribeiro JF(eds)Cerrado:ecologia e flora. Embrapa,Planaltina,pp 401–406

Lei-Hong Li,Wan-li Zhang(2000)Native vegetation and its ecosystem current situation in Brazil.J For Res 11:140–144

Lemmon PE(1957)A new instrumentformeasuring forestoverstorey density.J For 55:67–668

Lenza E,Klink AC(2006)Comportamento fenolo′gico de espe′cies lenhosas em um cerrado sentido restrito de Bras?′lia,DF.Rev Bras Bot 29:627–638

Lima JFEW,Silva EM(2008)Recursos h?′dricos do bioma cerrado: importa?ncia e situac?a?o.In:Sano MS,Almeida SP,Ribeiro JF (eds)Cerrado:ecologia e flora.Embrapa,Planaltina,pp 91–106

Lugo AE(1997)The apparent paradox of restablishing species richness on degraded lands with tree monocultures.For Ecol Manag 99:9–19

Luz,GR,Menino GOG,Mota SG,Nunes FRY(2008)S?′ndromes de dispersa?o de espe′cies arbustivo-arbo′reas em diferentes fitofisionomias no norte de Minas Gerais.In:IX Simpo′sio Nacional Cerrado e IISimpo′sio Internacional Savanas Tropicais,Embrapa,Planaltina,p 1132

Machado BR,Ramos Neto MB,Pereira PGP,Caldas EF,Gonc?alves DA,Santos NS,Tabor K,Steininger M(2004)Estimativas de perda da a′rea do cerrado Brasileiro.Conservac?a?o Internacional, Sa?o Paulo,p 23

Mendonc?a CR,Felfili JM,Walter BMT,Silva Ju′nior CM,Rezende AV,Filgueiras TS,Nogueira PE(2008)Flora vascular do Cerrado.In:Sano SM,Almeida SP,Ribeiro JF(eds)Cerrado: ambiente e flora.Embrapa,Planaltina,pp 289–556

Ministe′rio do Meio Ambiente(2010)Florestas do Brasilem resumo. Servic?o Florestal Brasileiro.Available at:http://www.ciflorestas. com.br/arquivos/doc_florestas_resumo_22648.pdf.Accessed 29 Aug 2013

Nascimento ART,Scariot AO,Silva JA,Sevilha AC(2004) Estimativas de a′rea basal e uso de relasco′pio de Bitterlich em amostragem de Floresta Estacional Decidual.Cie?nc Florest 14:169–176

Oliveira PE,Moreira AG(1992)Anemocoria em espe′cies de cerrado e de mata de galeria.Rev Bras Bot 15:163–174

Oliveira-Filho AT,Fontes MAL(2000)Patterns of floristic differentiation among Atlantic forests in southeastern Brazil,and the influence of climate.Biotropica 32:793–810

Oliveira-Filho AT,Mello JM,Scolforo JRS(1997)Effects of past disturbance and edges on tree community structure and dynamics within a fragment of tropical semideciduous forest in southeastern Brazil over a five-year period(1987–1992).Plant Ecol 131:45–66

Parrottta JA(1992)The role of plantation forests in rehabilitating degraded tropicalecosystems.Agric EcosystEnviron 41:115–133

Parrotta JA,Turnbull JW,Jones N(1997)Catalyzin native Forest regeneration on degraded tropical lands.Forest Ecol Manag 99:1–7

Pizo MAO(2012)O movimento dos animais frug?′voros e das sementes em paisagens fragmentadas.In:Del-Claro K,Torezan-Silingardi HM(eds)Ecologia das interac?o?es plantas-animais uma abordagem ecolo′gico-evolutiva.Technical Books Editora, Rio de Janeiro,pp 143–154

Puig H(2008)A floresta tropical u′mida.Imprensa Oficial do Estado de Sa?o Paulo/Editora da UNESP,Sa?o Paulo,p 496

Ratter JA,Ribeiro JF,Bridgewater S(1997)The Brazilian Cerrado vegetation and threats to its biodiversity.Ann Bot 80:223–230

Ratter JA,Bridgewater S,Ribeiro JF(2003)Analysis of the floristic composition of the Brazilian Cerrado vegetation III:comparison of the woody vegetation of 376 areas.Edinb J Bot 60:57–109

Ratter JA,Bridgewater S,Ribeiro JF(2006)Biodiversity pattherns of woody vegetation of the Brazilian Cerrado.In:Pennington T, Lewis GP,Ratter JA(eds)Neotropical savannas and seasonally dry forests:Plantdiversity,biogeography and conservation.CRC Press,Boca Raton,pp 31–66

Ribeiro MPHG,Felfili JM(2009)Regenerac?a?o naturalem diferentes ambientes da mata de galeria do Capetinga,na fazenda A′gua Limpa-DF.Cerne 15:1–9

Ribeiro LC,Tabarelli M(2002)A structural gradient in cerrado vegetation of Brazil:changes in woody plant density,species richness,life history and plant composition.J Trop Ecol 18:775–794

Ribeiro JF,Walter BMT(2008)As principais fitofisionomias do bioma Cerrado.In:Sano SM,Almeida SM,Ribeiro JF(eds) Cerrado:ecologia e flora.Embrapa,Planaltina,pp 153–212

Santos MR,Vieira AF(2005)Ana′lise estrutural do componente arbo′reo de tre?s a′reas de cerrado em diferentes esta′dios de conservac?a?o no munic?′pio de Tre?s Marias,Minas Gerais,Brasil. Cerne 11:399–408

Silva JMC,Tabarelli M(2000)Tree species impoverishmentand the future flora of the Atlantic forest of northeast Brazil.Nature 404:72–74

Silva MAF,Assad DE,Evangelista AB(2008)Caracterizac?a?o clima′tica do bioma Cerrado.In:Sano SM,Almeida SP,Ribeiro JF(eds)Cerrado:ecologia e flora.Embrapa,Planaltina, pp 69–87

Silva-Ju′nior MC,Furley PA,Ratter JA(1996)Variations in tree communities and soils with slope in gallery forest,Federal District,Brazil.In:Anderson MG,Brooks SM(eds)Advances in hill slope processes.Wiley,London,pp 451–469

ter Braak CJF(1995)Ordination.In:Jongman RHG,ter Braak CFJ, van Tongeren OFR(eds)Data analysis in community and landscape ecology.Cambridge University Press,Cambridge, pp 91–173

ter Braak CJF,Smilauer P(2003)Canoco?for windows version 4.5. Centre for Biometry,Wageningen

Toppa HR(2003)Estrutura e diversidade flor?′stica das diferentes fisionomias de Cerrado e suas correlac?o?es com o solo na Estac?a?o Ecolo′gica de Jata?′,Luiz Anto?nio,SP.Tese(Doutorado em Ecologia e Recursos Naturais)-Universidade Federal de Sa?o Carlos,Sa?o Carlos,p 127

Vale VS,Crespilho RF,Schiavini I(2009)Ana′lise da regenerac?a?o natural em uma comunidade vegetal de cerrado no parque Victo′rio Siquierolli,Uberla?ndia-MG.Biosc J 25:131–145

van der PijlL(1982)Principles ofdispersalin higherplants.Springer, Berlin

Viani GAR,Durigan G,Melo GCA(2010)A regenerac?a?o naturalsob plantios florestais:desertos verdes ou redutos da biodiversidade? Cie?nc Florest 20:533–552

Weiser LV,Godoy PAS(2001)Flor?′stica em um hectare de cerrado stricto sensu na ARIE Cerrado Pe′de Gigante,Santa Rita do Passa Quatro,SP.Acta Bot Bras 15:201–212

Zar JH(1999)Biostatisticalanalysis.Prentice-Hall,New York,p 944

21 August 2013/Accepted:9 June 2014/Published online:28 April 2015

?Northeast Forestry University and Springer-Verlag Berlin Heidelberg 2015

The online version is available at http://www.springerlink.com

Corresponding editor:Hu Yanbo

?Andre′R.Terra Nascimento arnterra@gmail.com

1Elementary/High Schools in Uberla?ndia,Uberla?ndia, Minas Gerais 38412-436,Brazil

2Universidade Federal de Uberla?ndia,Uberla?ndia, Minas Gerais 38412-436,Brazil

3Avenida Ortizio Borges,2836,Bairro Santa Mo?nica, Uberla?ndia,Minas Gerais 38412-436,Brazil