Scientia Forestalis, volume 42, n. 103
p.371-382, setembro de 2014

Do alien species dominate plant communities undergoing restoration? A case study in the Brazilian savanna

Espécies exóticas dominam comunidades vegetais em restauração? Um estudo de caso no Cerrado

Caio Santilli1
Giselda Durigan2

1MSc., Alpes Environmental Consultancy and Engineering Agency. Pedro Niceto Filho Street, 20 - 13.033-253, Campinas, SP, Brazil. Corresponding author: Caio Santilli, E-mail:
2PhD, Forestry Institute of the State of São Paulo, Assis State Forest. P.O. Box 104, 19802-970, Assis, SP, Brazil

Recebido em 25/10/2013 - Aceito para publicação em 08/05/2014


Entre os princípios estabelecidos pela Sociedade Internacional para Restauração Ecológica - Society for Ecological Restoration International – SER, consta que o uso de espécies exóticas para fins de restauração deve ser evitado, e isso é delineado nos dois primeiros atributos esperados de um ecossistema restaurado pelo SER Primer. Essa recomendação é provavelmente baseada na hipótese de que as espécies exóticas dominarão comunidades em restauração e prejudicarão a biodiversidade local. Nós avaliamos a comunidade atual de plantas em uma área de Cerrado em restauração, onde 42 espécies – 6 nativas e 36 não nativas – foram plantadas, e comparamos com comunidade nativa adjacente. Nós visamos verificar se a comunidade nativa adjacente foi invadida pelas exóticas, e se essas espécies tendem a dominar a comunidade em restauração, que seria então distinta da flora nativa. Oito anos após o plantio, não detectamos a presença das espécies exóticas no ecossistema natural adjacente. Na comunidade em restauração, mesmo sendo exóticas 94% das árvores plantadas (86% das espécies), apenas 3% das plantas regenerantes (14% das espécies) pertencem a espécies não nativas, indicando que a similaridade florística tende a aumentar ao longo do tempo. Consideramos que as espécies não nativas utilizadas neste projeto não estão oferecendo perigo aos ecossistemas naturais situados nas proximidades e que, em longo prazo, espécies nativas dominarão o ecossistema em restauração. No entanto, recomendamos que políticas públicas priorizem e viabilizem o uso de espécies nativas do Cerrado, melhor adaptadas para a restauração deste ecossistema.
Palavras-chave: invasão biológica, Cerrado, restauração ecológica, espécies invasoras, efeito de prioridade.


Among the principles established by the Society for Ecological Restoration International – SER is that the use of exotic species for restoration purposes should be avoided, and this is outlined on the two first attributes expected of a restored ecosystem by the SER Primer. This recommendation is possibly based on the hypothesis that exotics will dominate the restored communities and jeopardize the local biodiversity. We assessed the current plant community in an area of the Brazilian Cerrado undergoing restoration, where 42 species – 6 natives and 36 non-natives – were planted, and compared it with the contiguous native community. We aimed at verifying if the surrounding native community has been invaded by alien species, and if the community being restored has been dominated by the latter, not resembling the native flora. Eight years after planting, the alien species were not recorded in the surrounding native ecosystem. In the community undergoing restoration, despite 94% of the planted trees being exotics (86% of the species) they corresponded to only 3% of plants regenerating (14% of the species), indicating that floristic similarity with the native vegetation is increasing over time. We consider that the non-native species planted do not offer threat to the native ecosystems in the vicinity, and tend to be defeated by the natives in the long term. Even though, public policies should prioritize and make feasible the use of native species, better adapted to the harsh environmental conditions of the Cerrado.
Keywords: biological invasion, Cerrado, ecological restoration, invasive species, priority effect.


The importance of invasive plant species has been highlighted because of their economic costs as weeds and because they may cause native biodiversity losses (WILCOVE et al., 1998; MACK et al., 2000; LEVINE et al., 2003) and alter ecosystem functions (VITOUSEK 1996; RAIZADA et al. 2008; EHRENFELD 2010). Ecological restoration aims at promoting natural succession, enabling high biodiversity and establishing a vegetation community similar to the pristine ecosystems. Among the principles established by the Society for Ecological Restoration International – SER is that ‘a restored ecosystem consists of indigenous species to the greatest practicable extent’, and ‘contains a characteristic assemblage of the species that occur in the reference ecosystem’ (SER 2004). Under these principles, we conclude that the use of exotic species for restoration purposes should be avoided. This recommendation is certainly based on the hypothesis that exotics can behave like Greek soldiers that were hidden on the huge wooden horse and brought as a gift into the city walls of Troy, defeating the native soldiers and conquering the new territory. Bringing this metaphor to the context of ecological restoration and biological invasions, we named this as “the Trojan horses’ hypothesis”, which we tested in the Brazilian savanna. This hypothesis is related to the priority effects, which occur when the arrival or earlier growth of one or more species leads to a different community structure than it would be if all species began growth simultaneously (SHULMAN et al. 1983). It has been tested in the context of native x alien species, and the advantage of exotics has been confirmed for herbs, grasses and shrubs (MCEWAN et al., 2009; DICKSON et al., 2012; WAINWRIGHT et al., 2012), but was never tested for tropical trees.

The Cerrado – the Brazilian savanna, with its squat trees of thick bark, twisted trunks and thick twigs, spread over a grass layer, is the second largest biome in Brazil the most extensive savanna in South America, and the richest savanna in the world. It covered more than 2 million km², but has lost about half of this area in the last four decades and is possibly the most threatened tropical savanna in the world (SILVA; BATES, 2002; KLINK; MACHADO, 2005), rapidly replaced by agriculture. As for other degraded biomes in the world, ecological restoration has becoming a priority action for biodiversity conservation of the Cerrado. Restoring the Cerrado vegetation, however, has been by far more difficult than restoring forest biomes in Brazil (DURIGAN; MELO, 2011; PILON; DURIGAN, 2013). The low availability of water and nutrients in the soil, and the unlikely propagation of its native plant species by seeds (LABORIAU et al., 1963; HOFFMANN, 1998; HOFFMANN et al., 2004; MELO et al., 2005) are the natural factors which, besides the invasion by African grasses, are constraining restoration success in the Cerrado (HOFFMANN; HARIDASAN, 2008; DURIGAN; MELO, 2011). Pilon and Durigan (2013), when indicating framework species to recover the Cerrado vegetation, adopted as criteria the ability of the species to overcome the obstacle posed by the invasive grasses.

Ecological restoration or rehabilitation of the Cerrado vegetation is legally required in some situations (DURIGAN; MELO, 2011), but under the inexistence of seedlings available of native species, which are mostly endemic from this biome, there is not much option left but using species from other biomes or other countries. Many of the exotic species which have been planted in Cerrado restoration are already included in lists of invasive species in Brazil or elsewhere (Instituto Horus, Global Invasive Species Database, ZENNI; ZILLER, 2011). The impact of these non-native species on the Cerrado community, however, has not yet been assessed.

The presence of exotic species in plant communities undergoing restoration must not be considered a problem per se (ZAVALETA et al., 2001; D’ANTONIO; MEYERSON, 2002; EWEL; PUTZ, 2004; MARTÍNEZ, 2010; BRUDVIG, 2011). In fact, there are reports on exotics facilitating natural regeneration of native species and helping ecosystem reestablishment (LANTA; LEPS, 2008; SANTIAGO-GARCIA et al., 2008, MARTÍNEZ, 2010; MODNA et al., 2010) or improving ecosystem functioning (PARROTA; KNOWLES, 1999; VANDERHOEVEN, et al., 2005). The performance of exotic and natives often depends on growing conditions and no species can equally invade all ecosystems (DAEHLER, 2003; COLAUTTI; MCISAAC, 2004). Biological invasion proceeds when invasivity of the species matches with invasibility of the ecosystem (REJMÁNEK, 1999; ALPERT, et al., 2000; PYSEK; RICHARDSON, 2007; RICHARDSON; REJMÁNEK, 2011), and a species can be an effective invader only inside a particular ecological region (COLAUTTI; MCISAAC, 2004).

We assessed the current plant community in a stand of Cerrado vegetation undergoing restoration where a number of non-native species were inadvertently planted, and compared it with the contiguous native community. First, we investigated if the non-native species planted were colonizing the native vegetation around, already performing as invasive. In the community undergoing restoration, we verified if there is a tendency of the non native species planted to dominate the community over time, on the basis of their proportion among the trees planted and individuals spontaneously regenerating in the understory.


Study site

The Assis State Forest (Assis, state of Sao Paulo, Brazil, coordinates 22°37’41”S, 50°21’27”W) is a protected area in the southern borders of the Cerrado biome, in altitudes ranging from 500 to 588 m.a.s.l. The regional climate is between Cwa and Cfa subtropical climates (Köppen’s classification), with a rainy summer and a dry winter. Mean annual precipitation is around 1.400 mm, and mean temperature is 21.8°C (PINHEIRO; DURIGAN, 2009). The soil at the study site is classified as Red Dystrophic Oxysoil, which is deep, sandy, acid and poor as are most of Cerrado soils (REATTO et al., 2008). The local native vegetation is classified as Cerrado sensu lato, being the cerradão (woodland savanna) the dominant physiognomy. The woodland savanna has a forest physiognomy with a continous arboreal stratum, shading about 90% of the ground (PINHEIRO; DURIGAN, 2009). This study encompasses an area of 20 ha undergoing restoration and a remnant of woodland savanna of 8 ha in the vicinity, both situated more than 200 m far from the riparian zone.

Restoration background

In order to duplicate the width of an existing road near the Assis State Forest, a number of isolated trees and small fragments of native vegetation were removed. The Brazilian legislation requires compensation of the environmental impacts caused by an enterprise and that includes ecological restoration in the same region, and recovering the same vegetation type which was destroyed (see ARONSON et al., 2011; DURIGAN; MELO, 2011). For the Cerrado, in the state of São Paulo, an area four times greater than the impacted area must be restored (SÃO PAULO, 2009). To comply with the law by compensating the removal of native vegetation along the road, a restoration planting was carried out at Assis State Forest, aiming to recover an area previously occupied by pasture, and thus increasing the areas of habitat for the native species of the Cerrado biome.

Restoration planting took place in January 2003, after more than three decades of use as pasture of Urochloa decumbens (Poaceae), an invasive African grass that has been one of the most severe threats to the Cerrado biome (PIVELLO et al., 1999; KLINK; MACHADO, 2005; DURIGAN et al., 2007). Cattle had been excluded for some years and a natural regeneration process was taking place (data published by DURIGAN et al., 1998). Two years before the restoration planting, the entire area was accidentally burned and the high biomass of invasive grasses resulted in total destruction of the native biomass. From the set of tree species planted, only a small portion were native from the local Cerrado, and the majority came from other biomes in Brazil or from other countries. Some of the species planted are included amongst the most aggressive invasive species in the world according to the Global Invasive Species Database (2005), such as Leucaena leucocephala (Fabaceae Mimosoideae), Schinus terebinthifolius (Anacardiaceae) and Spathodea campanulata (Bignoniaceae). Others are amongst the species considered invasive in Brazil (ZENNI; ZILLER, 2011), including Clitorea fairchildiana (Fabaceae Faboideae), Hovenia dulcis (Rhamnaceae) and Tecoma stans (Bignoniaceae). The explanation from the company in charge of restoration for including this high number of non-native species was, as usual, the lack of seeds and seedlings from Cerrado species promptly available. Seedlings were planted in an average density of 1000 ind.ha-1, about 3 x 3 m spacing. The area being restored is bordered in its full extent by a fragment of native vegetation – the woodland savanna. We used this fragment as a reference ecosystem for the native flora and it was also surveyed in search of invasive individuals from the non-native species introduced in the planted stand.

Data collection and analyses

We sampled 40 plots of 200 m² each (40 m x 5 m) randomly distributed in the area undergoing restoration, at eight years after planting, and 10 plots (the same size) in the reference ecosystem. In each plot, all woody plants from 50 cm in height were identified and recorded. In addition to the origin of the species (native or non-native), we categorized each individual sampled as planted (in regularly spaced lines) or not planted (naturally established). We also described the species by dispersal syndrome and shade tolerance, on the basis of previous studies (DURIGAN et al., 2004; PILON; DURIGAN, 2013), and literature on exotic species (LORENZI et al., 2003).  We separated the individuals in three size classes: a) stem diameter taken at 30 cm above ground – D30 < 1 cm; b) 1 cm ≤ D30 < 5 cm; and c) D30 ≥ 5 cm. Reproductive individuals were recorded, in order to indicate if the species could, potentially, leave descendants.

The stem diameter was measured for all individuals with D30 ≥ 5 cm, in order to estimate the basal area of the community. In addition, canopies and grass cover were estimated by the line interception method (CANFIELD, 1941), over a line 40 m long in the middle of each plot.

Absolute and relative densities of the species in the community being restored and in the native vegetation were calculated separately by size class, species origin (native or non-native) and for planted or naturally established individuals. Relative density was calculated as the proportion of individuals of a particular group among all individuals in the community. We applied Chi-square test to verify if the proportions differ between groups.

The Jaccard’s similarity index - ISj (MÜLLER-DOMBOIS; ELLENBERG, 1974) between the floristic composition of the reference ecosystem and that being restored was calculated for each size class.  We considered the smaller plants as the last to arrive at the community (the youngest). The Jaccard´s similarity index between two communities corresponds to the proportion of the total number of species sampled which is common to both communities. Two communities are considered floristically similar if at least 25% of the species (ISj = 0.25) are in common (MÜLLER-DOMBOIS; ELLENBERG, 1974).

For each non-native species planted, we compared the relative density among the planted individuals and among those naturally regenerating. We considered as potentially invasive a species if its frequency (relative density) was higher among plants in natural regeneration than among the planted individuals.


Community structure: The reference ecosystem presented a basal area of 19.76 m2.ha-1, canopy cover of 90% and grasses were absent. The density was 1225 individuals ha-1 in the arboreal layer and 6040 individuals ha-1 in the understory (stem diameter below 5 cm), from 53 species. In the community undergoing restoration, basal area was 5.12 m2.ha-1, canopy cover was 66% and invasive grasses occupied 67% of the area; density was 489 individuals ha-1 of planted trees and 2418 individuals ha-1 of plants in natural regeneration, summing up 78 species (Table 1).

Origin of the species: In the Cerrado fragment (reference ecosystem), all individuals surveyed in the arboreal layer as well as in the understory were from native species (53 species recorded). In the community being restored, among the 42 species planted, 14% were native (6 species) and 86% were non-natives (36 species). From the regenerating community, 86% of the species (38 species) were native and 14% were non-natives (6 species) (Table 1).

Table 1. Arboreal species sampled in the Cerrado area undergoing restoration, their functional attributes (shade tolerance and dispersal syndrome), if planted or not, reproducing or not, native or non native, status as invasive in Brazil, and relative density among planted trees, among individuals in natural regeneration and in the reference ecosystem (Assis State Forest, Assis, SP, Brazil).
Tabela 1. Espécies arbóreas amostradas na área de restauração de Cerrado, atributos funcionais (tolerância à sombra e síndrome de dispersão), se plantadas ou não, reprodutivas ou não, nativas ou não, status como invasoras no Brasil, e densidade relativa entre árvores plantadas, entre indivíduos em regeneração natural e no ecossistema de referência (Floresta Estadual de Assis, Assis, SP, Brasil).
Species Shade tolerance Dispersal syndrome Planted species Reproducing Geographical origin Status as invasive Relative density (%)
Among planted trees Among regenerating individuals in the restored area In the Reference ecosystem
Acacia mangium Intolerant Barochory Planted Yes non native Invasive 0.26 0.00 0.00
Acacia paniculata Intolerant Barochory Planted native 0.51 0.10 0.00
Actinostemon conceptionis Tolerant Barochory Yes native 0.00 0.00 0,08
Aegiphila lhotskiana Intolerant Zoochory Yes native 0.00 0.78 0,04
Albizia lebbeck Intolerant Barochory Planted non native 1.28 0.00 0.00
Amaioua intermedia Tolerant Zoochory Yes native 0.00 0.00 0,04
Anadenanthera falcata Intolerant Barochory Planted Yes native 1.54 0.00 0.00
Anadenanthera macrocarpa Intolerant Barochory Planted Yes non native 11.03 0.16 0.00
Annona dioica Intolerant Zoochory Yes native 0.00 0.62 0.00
Annona muricata Intolerant Zoochory Planted non native 0.51 0.00 0.00
Astronium graveolens Tolerant Anemochory native 0.00 0.00 0,04
Baccharis dracunculifolia Intolerant Anemochory Yes native 0.00 21.93 0.00
Bauhinia longifolia Tolerant Barochory Planted Yes non native 1.79 0.26 0.00
Bauhinia variegata Intolerant Barochory Planted Yes non native 3.85 0.00 0.00
Bixa orellana Intolerant Zoochory Planted Yes non native 1.03 0.00 0.00
Bredemeyera floribunda Intolerant Anemochory Yes native 0.00 1.98 0,12
Byrsonima intermedia Intolerant Zoochory native 0.00 0.78 0,29
Byrsonima laxiflora Tolerant Zoochory native 0.00 0.10 0,08
Caesalpinia ferrea Tolerant Barochory Planted non native 0.10 0.05 0.00
Caesalpinia peltophoroides Intolerant Barochory Planted non native 2.05 0.00 0.00
Callicarpa reevesii Intolerant Zoochory Planted non native 0.26 0.00 0.00
Caryocar brasiliense Intolerant Zoochory Yes native 0.00 0.00 0,62
Casearia silvestris Tolerant Zoochory Yes native 0.00 0.10 0.00
Cecropia pachystachya Intolerant Zoochory Planted Yes native 2.05 0.00 0.00
Cedrela fissilis Tolerant Anemochory Planted non native 1.28 0.00 0.00
Ceiba speciosa Intolerant Anemochory Planted non native 0.26 0.00 0.00
Clitoria fairchildiana Intolerant Barochory Planted Yes non native Invasive 4.62 0.00 0.00
Cojoba sophorocarpa Intolerant Barochory Planted non native 0.77 0.00 0.00
Copaifera langsdorffii Tolerant Zoochory native 0.00 0.05 4,44
Cordia sellowiana Tolerant Zoochory Yes native 0.00 0.00 0,29
Couepia grandiflora Intolerant Zoochory Yes native 0.00 0.00 0,21
Croton floribundus Intolerant Barochory Planted Yes non native 0.26 0.31 0,25
Croton urucurana Intolerant Barochory Planted Yes non native 1.79 0.00 0.00
Cupania tenuivalvis Tolerant Zoochory Yes native 0.00 0.00 0,99
Dimorphandra mollis Intolerant Barochory Yes native 0.00 0.16 0.00
Diospyros hispida Intolerant Zoochory native 0.00 0.05 0.00
Diospyros inconstans Tolerant Zoochory native 0.00 0.16 0.00
Duguetia furfuracea Intolerant Zoochory Yes native 0.00 0.26 0,08
Enterolobium contortisiliquum Intolerant Barochory Planted non native 0.26 0.00 0.00
Erythoxylum pelleterianum Tolerant Zoochory Yes native 0.00 0.00 0,21
Eugenia lambertiana Tolerant Zoochory Yes native 0.00 0.00 0,25
Eugenia punicifolia Intolerant Zoochory native 0.00 0.21 0,41
Genipa americana Intolerant Zoochory Planted non native 0.77 0.00 0.00
Gochnatia barrosii Intolerant Anemochory native 0.00 5.82 0,33
Gochnatia polymorpha Intolerant Anemochory Yes native 0.00 0.16 0,21
Helicteres lhotzkyana Intolerant Anemochory Planted Yes native 0.26 0.00 0.00
Hovenia dulcis Tolerant Zoochory Planted non native Invasive 3.85 0.00 0.00
Hymenaea courbaril Tolerant Zoochory Planted non native 1,03 0.00 0.00
Hymenaea stigonocarpa Tolerant Zoochory Yes native 0.00 0.00 0,04
Inga laurina Intolerant Zoochory Planted non native 1.28 0.00 0.00
Jacaranda caroba Intolerant Anemochory Yes native 0.00 0.52 0,99
Jacaranda cuspidifolia Intolerant Anemochory Planted non native 4.10 0.00 0.00
Lacistema hasslerianum Tolerant Zoochory Yes native 0.00 0.00 0,04
Lafoensia pacari Tolerant Anemochory Planted Yes native 0.26 0.05 0.00
Luehea candicans Tolerant Anemochory native 0.26 0.05 0.00
Mabea fistulifera Tolerant Barochory Yes native 0.00 4.99 5,80
Machaerium aculeatum Intolerant Anemochory Yes native 0.00 0.00 0,21
Machaerium acutifolium Intolerant Anemochory Yes native 0.00 0.00 0,99
Machaerium brasiliense Intolerant Anemochory Yes native 0.00 0.68 0.00
Maprounea guianensis Tolerant Zoochory Yes native 0.00 0.00 4,03
Melia azedarach Intolerant Zoochory Planted non native Invasive 0.77 0.00 0.00
Memora axillaris Intolerant Anemochory Yes native 0.00 0.10 0.00
Miconia albicans Tolerant Zoochory native 0.00 0.16 4,57
Miconia ligustroides Tolerant Zoochory Yes native 0.00 0.00 0,08
Miconia stenostachya Tolerant Zoochory Yes native 0.00 0.00 0,95
Mimosa caesalpiniifolia Intolerant Barochory Planted Yes non native Invasive 11.54 0.00 0.00
Mimosa setosa Intolerant Barochory Planted Yes non native 3.08 2.23 0.00
Myrcia fallax Tolerant Zoochory native 0.00 0.10 3,13
Myrcia guianensis Tolerant Zoochory Yes native 0.00 0.57 12,59
Myrcia lingua Intolerant Zoochory Yes native 0.00 0.00 0,04
Myrcia multiflora Tolerant Zoochory Yes native 0.00 0.00 2,18
Myrciaria floribunda Tolerant Zoochory Yes native 0.00 0.00 0,62
Nectandra cuspidata Tolerant Zoochory Yes native 0.00 0.00 1,73
Nerium oleander Intolerant Anemochory Planted non native 0.26 0.00 0.00
Ocotea corymbosa Tolerant Zoochory native 0.00 0.99 7,32
Peltophorum dubium Intolerant Anemochory Planted Yes non native 13.59 0.10 0.00
Pera obovata Tolerant Barochory Yes native 0.00 0.10 0.00
Persea wildenovi Tolerant Zoochory Yes native 0.00 0.00 0,25
Pouteria ramiflora Tolerant Zoochory native 0.00 0.05 2,71
Protium heptaphyllum Tolerant Zoochory Yes native 0.00 0.00 0,04
Pseudolmedia laevigata Tolerant Zoochory Yes native 0.00 0.00 0,25
Psidium guajava Intolerant Zoochory Planted Yes non native Invasive 0.26 0.00 0.00
Qualea cordata Intolerant Anemochory Yes native 0.00 0.00 0,21
Qualea grandiflora Intolerant Anemochory Yes native 0.00 0.00 0,25
Roupala montana Tolerant Anemochory Yes native 0.00 0.57 0,45
Schefflera vinosa Intolerant Zoochory Yes native 0.00 0.00 1,60
Schinus molle Intolerant Zoochory Planted non native 1.03 0.00 0.00
Schinus terebinthifolius Intolerant Zoochory Planted Yes non native 10.51 0.16 0.00
Senna alata Intolerant Barochory Planted Yes non native 0.26 0.00 0.00
Senna rugosa Intolerant Zoochory Yes native 0.00 0.78 0,04
Senna siamea Intolerant Barochory Planted non native 0.26 0.00 0.00
Siparuna guianensis Tolerant Zoochory Yes native 0.00 0.00 15,14
Solanum paniculatum Intolerant Zoochory Yes native 0.00 37.32 0.00
Spathodea nilotica Intolerant Anemochory Planted non native 0.77 0.00 0.00
Strychnos brasiliensis Intolerant Zoochory native 0.00 0.26 1,85
Stryphnodendron obovatum Intolerant Barochory Yes native 0.00 14.40 0,99
Tabebuia impetiginosa Intolerant Anemochory Planted non native 4.87 0.00 0.00
Tabebuia ochracea Intolerant Anemochory Yes native 0.00 0.00 0,21
Tabernaemontana catharinensis Tolerant Zoochory Yes native 0.00 0.83 0.00
Tapirira guianensis Tolerant Zoochory native 0.00 0.05 0,78
Tecoma stans Intolerant Anemochory Planted Yes non native Invasive 2.82 0.00 0.00
Terminalia glabrescens Intolerant Anemochory Planted native 1.03 0.00 1,07
Tibouchina granulosa Intolerant Anemochory Planted Yes non native 1.79 0.00 0.00
Vernonia polyanthes Intolerant Anemochory Yes native 0.00 0.42 0.00
Vochysia tucanorum Intolerant Anemochory Yes native 0.00 0.10 4,48
Xylopia aromatica Intolerant Zoochory Yes native 0.00 0.42 15,43

The evidence from this study indicates that the woodland savanna has been resistant to invasion by the exotic species, since not even a single individual from the alien species already reproducing was recorded in the native ecosystem. However, considering that a lag phase is common between establishment and spread, when the invasive species adapt to the new community (SAKAI et al., 2001; BARNEY, 2006), the time period since the introduction of the non-natives was too short for a conclusion, requiring monitoring of the native ecosystem in the long term. Studies have shown that the community assembly in the woodland savanna is mediated by competition for light and soil water (ABREU et al., 2011; ASSIS et al., 2011), and these are, probably, the ecological filters locally constraining the establishment of the alien species in the native ecosystem.

The alien species planted are not colonizing the area being restored either. It was hypothesized that alien species are not adapted to the harsh environmental conditions of the Cerrado and, therefore, do not reproduce or regenerate in the study site. Soil water deficiency in addition to acid and poor soils, can have either delayed or inhibited those species to reach their reproductive stage or to establish. In fact, Figure 1 reveals that, among the species planted, the natives have been relatively better succeeded than exotics in reproduction.

Reproductive stage: 42 of all 78 species sampled in the community undergoing restoration (planted or regenerating) were already in reproductive age in the restored area or in the native vegetation around. From these species, 26 species are native from Cerrado, while the others (16) do not occur in the study region. Considering only the planted trees, 20 species have reached reproductive stage, being four natives and 16 non-natives. The proportion of species reproducing among those planted (χ²= 10.021; p<0.05; d.f=1) is higher among natives than non-natives (Figure 1).

Figura 1. Proportions of the species planted (natives and non-natives) which 1) did not reach the reproductive stage (non-reproductive), 2) are in reproductive stage, but do not leave descendants (reproductive not established) or 3) are naturalized (reproductive and leaving descendants).
Figure 1. Proporção de espécies plantadas (nativas e não nativas) que 1) não atingiram estado reprodutivo (não reprodutivas), 2) estão em estado reprodutivo, mas não deixam descendentes (reprodutivas mas não estabelecidas) e 3) estão naturalizadas (reprodutivas e deixando descendentes).

The relative density (proportion of individuals) by the origin of the species (natives x non-natives) (Figure 2) was distinct between the regenerating stratum and the set of planted trees (Chi-square test: χ²=165; p<0.001, d.f=1, Figure 2).The restoration planting was composed mainly by exotic species, which summed up to 86% of all species. This unbalance, however, has not been sufficient to favor the alien species in colonizing the area being restored, since the regenerating stratum is strongly dominated by the natives (Figure 2 a). The remarkably higher relative abundance of exotics among the planted trees (Figure 2 b) could potentially increase the propagule pressure, which is one of the predictors of invasion success cited in many studies (WILLIAMSON, 1996). The priority effects being stronger for alien than native species (MCEWAN et al. 2009; DICKSON et al. 2012; WAINWRIGHT et al. 2012) were not observed in this study.

Figura 2. Proportion of individuals from native and non-native species in an area of Cerrado undergoing restoration, southeastern Brazil. (a) among the naturally regenerated individuals; (b) among the planted individuals.
Figure 2. Proporção de indivíduos de espécies nativas e não nativas em área de Cerrado que se encontra em processo de restauração, sudeste do Brasil. (a) entre indivíduos naturalmente regenerantes, (b) entre indivíduos plantados.

When every single species is analyzed, the relative abundance in the community regenerating was never higher for the alien species than the proportion of the species among the individuals planted (Table 1). Mimosa setosa (Fabaceae Mimosoideae) presented the highest relative density among individuals naturally regenerating (2.23%), but it is still lower than among planted trees of this species (3.08%). This is a short lived treelet native from the dry regions of northeastern Brazil, which has adapted to the severity of the local environmental conditions, and requires attention. Since this species is not shade tolerant, however, its persistence in the area undergoing restoration will probably not be possible in the future, when the forest structure of the native vegetation is reached. Four native species not planted – Solanum paniculatum (zoochorous and shade intolerant), Stryphnodendron obovatum (barochorous and shade intolerant), Gochnatia barrosii (anemochorous and shade intolerant) and Mabea fistulifera (barochorous and shade tolerant), have higher relative densities than M. setosa. These species sum up 62.51% of individuals in the understory (Table 1), clearly dominating the community, and not showing a functional pattern among colonizers. All other non-native introduced species may reproduce occasionally in the community undergoing restoration but have not been able to replace adult trees and rely on repeated introductions to sustain their populations. Although further studies are required to draw definite conclusion, we observe that these species have, actually, performed as nurse trees (CALLAWAY, 1995) for the natives and are expected to go extinct when the adult trees die out. Consequently, they would not result in invasion if used for restoration purposes in the Cerrado region, even if they are invasive elsewhere. The same function, however, should certainly be performed by some native species, like those indicated as framework species for Cerrado restoration by Pilon and Durigan (2013), if there were seedlings available. As observed by Daehler (2003) and Colautti and McIsaac (2004), the performance of exotic and natives within a particular ecological region often depends on growing conditions and no species can equally invade all ecosystems. Richardson et al. (2000) states that “only a small fraction of all introduced taxa reproduce and spread over large areas; most taxa fail at some stage before reaching such levels of success”.

The relative density of native species regenerating differs among size classes (χ²= 7.6; p < 0.05; d.f = 2), with natives increasing as the plant size decreases (Figure 3). Considering that the smaller plants were the last to arrive to the community, the proportion of natives has increased with time after restoration planting.

Figura 3. Relative density of native and non-native individuals in the community undergoing restoration, in each size class.
Figure 3. Densidade relativa de indivíduos nativos e não nativos na comunidade em processo de restauração, em cada classe de tamanho.

The floristic similarity with the reference ecosystem (Figure 4) tends to increase with time, considering the youngest plants as the last to arrive to the community. When the younger plants are analyzed (diameter below 1 cm); the community undergoing restoration can already be considered floristically similar to the reference ecosystem (ISj > 0.25).

Figura 4. Jaccard´s Similarity Index (ISj) between the floristic composition of the reference ecosystem and the community undergoing restoration, calculated for three size classes according to the stem diameter.
Figure 4. Indície de Similaridade de Jaccard (ISj) entre a composição florística do ecossistema de referência e a comunidade em restauração, calculado para três classes de tamanho, conforme diâmetro do caule.

Our results revealed that non-native species introduced aiming at restoration made the community floristically distinct from the native vegetation. The young trees regenerating, however, which are mostly native (Figure 3), have already reached floristic similarity with the reference vegetation (Figure 4), with Jaccard’s Similarity Index higher than 25% (MÜLLER-DOMBOIS; ELLENBERG, 1974) if the smaller size class is included in the analysis. That is a consequence of the continuous immigration of native species from the surrounding native vegetation and from native adult trees in the area undergoing restoration (Figure 2). Therefore, the plant community is evolving towards native ecosystem despite of the inadvertently massive introduction of alien species. Apparently, the community assembly has followed the ‘self-design’ capacity of nature (MITSCH; WILSON, 1996) with the native species assembling by themselves given a long time period (ca 15–20 years).


Our findings did not support the Trojan’s horse hypothesis, since the alien species, even if intentionally introduced, have not proven to be able to spread over the ecosystem undergoing restoration and to dominate the community or invading the native ecosystems around. Alternatively, the natives are winning the battle, and the community is becoming more similar to the native vegetation through time. The presence of exotic species in plant communities being restored has not been considered a problem per se. If native and non-native species are both planted in a restoration project or if there are seed sources in the vicinity, the natives tend to succeed relatively better in regenerating and colonizing areas of woodland savanna (cerradão) undergoing restoration. The non-natives tend to be naturally eliminated from the community.

Monitoring the non-native species in the long term is recommended, to identify possible threats of invasion (species to be necessarily banned from restoration) as well as exotic species which can act as nurse trees (“friendly” alien species), fostering the natural regeneration of Cerrado species. The use of native species for restoration is always preferable, particularly if seedlings of a set of framework species can be obtained. However, as far as production of seedlings of native Cerrado species in large scale persists as a technical obstacle to be surpassed, a pool of friendly non-natives can, potentially, help recovering at least some ecosystem services and catalyzing the latter arrival of the natives in severely degraded areas.


We thank Edson Damasceno and Edson Adriano Berto for helping with the field work, Lourens Poorter for his help with data analysis, the Forestry Institute of São Paulo state for supporting the project (research permit # 260108 – 013.650/2009, SMA-SP/IF), and the Brazilian Council for Scientific Research – CNPq for the productivity grant to G.D. (Process 303402/2012-1).


ABREU, R. C. R., G. B.; ASSIS, S.; FRISON, S.; AGUIRRE, A.; DURIGAN, G. Can native vegetation recover after slash pine cultivation in the Brazilian Savanna? Forest Ecology and Management, Amsterdam, v. 262, n. 8, p.1452–1459, 2011.

ALPERT, P.; BONE, E.; HOLZAPFEL, C. Invasiveness, invasibility and the role of environmental stress in the spread of non-native plants. Perspectives in Plant Ecology Evolution and Systematics, Amsterdam, v. 3, n. 1, p. 52–66. 2000.

ARONSON, J.; BRANCALION, P. H. S.; DURIGAN, G.; RODRIGUES, R. R.; ENGEL, V. L.; TABARELLI, M.; TOREZAN, J. M. D.; GANDOLFI, S.; MELO, A. C. G.; KAGEYAMA, P. Y.; MARQUES, M. C. M.; NAVE, A.; MARTINS, S. V.; GANDARA, F. B.; REIS, A.; BARBOSA, L. M.; SCARANO, F. R. What role should government regulation play in ecological restoration? Ongoing debate in São Paulo State, Brazil. Restoration Ecology, Tucson, v. 19, n. 6, p. 690–695, 2011.

ASSIS, A. C. C., COELHO, R. M.; PINHEIRO, E. S.; DURIGAN, G. Water availability determines physiognomic gradient in an area of low-fertility soils under Cerrado vegetation. Plant Ecology, Oxford, v. 212, n. 7, p. 1135–1147, 2011.

BARNEY, J. N. North American history of two invasive plant species: phytogeographical distribution, dispersal vectors, and multiple introductions. Biological Invasions, v. 8, n. 4, p. 703–713. 2006.

BRUDVIG, L. A. The Restoration of Biodiversity: where has research been and where does it need to go? American Journal of Botany, Columbus, v. 98, n. 3, p. 549–558. 2011.

CALLAWAY, R. M. Positive interactions among plants. The Botanical Review, Lancaster, v. 61, n. 4, p. 306–349. 1995.

CANFIELD, R. H. Application of the line interception method in sampling range vegetation. Journal of Forestry, Bethesdam, v. 39, n. 3, p. 388–394. 1941.COLAUTTI, R. I.; MACISAAC, H. J. A neutral terminology to define ‘invasive’ species. Diversity and Distributions, Matieland, v. 10, p. 135–141, 2004.

DAEHLER, C. C. Performance Comparisons of Co-Occurring Native and Alien Invasive Plants: Implications for Conservation and Restoration. Annual Review of Ecology, Evolution and Systematics, Palo Alto, v. 34, p. 183–211, 2003.

D’ANTONIO, C. M.; MEYERSON, L. A. Exotic plant species as problems and solutions in ecological restoration: a synthesis. Restoration Ecology, Tucson, v. 10, n. 4, p. 703–13, 2002.

DICKSON, T. L.; HOPWOOD, J. L.; WILSEY, B. J. Do priority effects benefit invasive plants more than native plants? An experiment with six grassland species. Biological Invasion, v. 14, n. 12, p. 2617–2624, 2012

DURIGAN G.; MELO, A. C. G. An overview of public policies and research on ecological restoration in the state of São Paulo, Brazil. In: FIGUEROA, E. (Ed.). Biodiversity Conservation in the Americas: Lessons and Policy Recommendations. Santiago: Editorial FEN - Universidad de Chile, 2011. p. 325-355.

DURIGAN, G.; SIQUEIRA, M. F.; FRANCO, G. A. D. C. Threats to the Cerrado Remnants of the state of São Paulo, Brazil. Scientia Agricola, Piracicaba. v. 64, n. 4, p. 355–363, 2007.

DURIGAN, G.; SIQUEIRA, M. F.; FRANCO, G. A. D. C.; CONTIERI, W. A. A Flora Arbustivo-Arbórea do Médio Paranapanema: Base para a Restauração dos Ecossistemas Naturais, In: VILAS BÔAS, O.; DURIGAN, G. (Org.). Pesquisas em conservação e recuperação ambiental no Oeste Paulista: resultados da cooperação Brasil/ Japão. 1.ed. São Paulo: Páginas & Letras, 2004, p. 199–239.

DURIGAN, G.; CONTIERI, W. A.; FRANCO, G. A. D. C.; GARRIDO, M. A. O. Indução do processo de regeneração da vegetação de cerrado em área de pastagem, Assis, SP. Acta Botanica Brasilica, São Paulo, v. 12, n. 3, p. 421–429, 1998.

EHRENFELD, J. Ecosystem Consequences of Biological Invasions. Annual Review of Ecology, Evolution, and Systematics, Palo Alto, v. 41, p. 59–80, 2010.

EWEL, J. J.; PUTZ. F. E. A place for alien species in ecosystem restoration. Frontiers in Ecology and the Environment, v. 2, n. 7, p. 354–360, 2004

GLOBAL INVASIVE SPECIES DATABASE, 100 of the World's Worst Invasive Alien Species, 2005.  Disponível em: <>. Acesso em: 01 maio 2012.

HOFFMANN, W. A. Post-burn reproduction of woody plants in a Neotropical savanna: the relative importance of sexual and vegetative reproduction. Journal of Applied Ecology, Oxford, v. 35, n.3, p. 422-433, 1998

HOFFMANN, W. A.; HARIDASAN. M. The invasive grass, Melinis minutiflora, inhibits tree regeneration in a Neotropical savanna. Austral Ecology, v. 33, n. 1, p. 29–36, 2008.

HOFFMANN, W. A.; ORTHEN, B.; FRANCO. A. C. Constraints to seedling success of savanna and forest trees across the savanna-forest boundary. Oecologia, Berlin, v. 140, n. 2, p. 252–260, 2004


KLINK, C. A.; MACHADO, R. B. Conservation of the Brazilian Cerrado. Conservation Biology, Barking, v. 19, n. 3, p. 707–713, 2005.

LABOURIAU, L. G. Problemas da fisiologia ecológica dos cerrados. In M. G. Ferri (Coord.). I Simpósio sobre Cerrado. São Paulo: Edgard Blücher e EDUSP, 1963. p.233–276.

LANTA, V.; LEPS, E. J. Effects of plant species richness on invasibility of experimental plant communities. Plant Ecology, Oxford, v. 198, n. 2, p. 253–263. 2008.

LEVINE, J. M.; VILÀ, M.; D'ANTONIO, C. M.; DUKES, J. S.; GRIGULIS, K.; LAVOREL, S. Mechanisms underlying the impacts of invasive species. Proceedings of the Royal Society, v. 270, n. 1517, p. 775–781, 2003.

LORENZI, H.; SOUZA, H. M.; TORRES, M. A. V.; BACKER, L. B. Árvores exóticas no Brasil. Nova Odessa: Instituto Plantarum. 2003, 368 p.

MACK, R. N., SIMBERLOFF, D.; LONSDALE, W. M.; EVANS, H.; CLOUT, M.; BAZZAZ, F. A. Biotic invasions: causes, epidemiology, global consequences and control. Ecological Applications, Washington, v. 10, n. 3, p. 689–710, 2000.

MARTÍNEZ, O. J. A. Invasion by native tree species prevents biotic homogenization in novel forests of Puerto Rico. Plant Ecology, Oxford, v. 211, n. 1, p. 49–64, 2010.

MCEWAN, R. W.; BIRCHFIELD, M. K.; SCHOERGENDORFER, A.; ARTHUR, M. A.  Leaf phenology and freeze tolerance of the invasive shrub Amur honeysuckle and potential native competitors. Journal of the Torrey Botanical Society, v. 136, n. 2, p. 212–220, 2009.

MELO, J. T.; TORRES, R. A. A.; SILVEIRA, C.; CALDAS, L. Coleta, propagação e desenvolvimento inicial de plantas do Cerrado. In: SCARIOT, A.; SOUSA-SILVA, J. C.; FELFILI, J. M. (Org.) Cerrado: ecologia, biodiversidade e conservação. Brasília: Ministério do Meio Ambiente. 2005, p. 321-350.

MITSCH, W. J.; WILSON, R. F. Improving the success of Wetland creation and restoration with know-how, time, and self-design. Ecological Applications, Washington, v. 6, n. 1, p. 77–83, 1996.

MODNA, D.; DURIGAN, G.; VITAL, M. V. C. Pinus elliottii Engelm como facilitadora da regeneração natural da mata ciliar em região de Cerrado, Assis, SP, Brasil. Scientia Forestalis, Piracicaba, v. 38, n. 85, p. 73–83, 2010.

MÜLLER-DOMBOIS, D.; ELLENBERG, H. Aims and methods of vegetation ecology. New York: John Wiley and Sons, 1974. 547 p.

PARROTA, J. A.; KNOWLES, O. H. Restoration of tropical moist forests on bauxite-mined lands in the Brazilian Amazon. Restoration Ecology, Tucson, v. 7, n. 2, p. 103–116, 1999.

PILON, N. A. L.; DURIGAN, G. Critérios para indicação de espécies prioritárias para a restauração da vegetação de Cerrado. Scientia Forestalis, Piracicaba, v.41, n. 99, p. 389-399, 2013.

PINHEIRO, E. S.; DURIGAN, G. Dinâmica espaço-temporal (1962-2006) das fitofisionomias em unidade de conservação do Cerrado no sudeste do Brasil. Revista Brasileira de Botânica, São Paulo, v. 32, n. 3, p. 441–454, 2009.

PIVELLO, V. R.; CARVALHO, M. C. V.; PECCININI, A. A.; ROSSO, S. Abundance and distribution of native and alien grasses in a “Cerrado” (Brazilian Savanna) biological reserve. Biotropica, Lawrence, v. 31, n. 1, p. 71–82, 1999.

PYSEK, P.; RICHARDSON, D. M. Traits associated with invasiveness in alien plants: where do we stand? Ecological Studies, v. 193, p. 97–126, 2007.

RAIZADA P; RAGHUBANSHI, A. S.; SINGH, J. S. Impact of invasive alien plant species on soil processes: a review. Proceedings of the National Academy of Sciences of India, Section B, v. 78, pt. IV, p. 288–98, 2008.

REATTO, A.; CORREIA, J. R.; SPERA, S. T.; MARTINS, E. S. Solos do Bioma Cerrado: aspectos pedológicos. In: SANO, S. M.; ALMEIDA, S. P.; RIBEIRO, J. F. (Ed.). Cerrado: Ecologia e Flora. Planaltina: Embrapa. 2008, p. 109-149.

REJMÁNEK, M. Invasive plant species and invasible ecosystems. In: SANDLUND, O. T.; SCHEI, P. J.; VIKEN, A. (Ed.). Invasive species and biodiversity management. Dordrecht: Kluwer, 1999, p.79-102.

RICHARDSON, D. M.; PYSEK, P.; REJMÁNEK, M.; BARBOUR, M. G.; PANETTA, F. D.; WEST, C. J. Naturalization and invasion of alien plants: concepts and definitions. Diversity and Distributions, Matieland, v. 6, n. 2, p. 93-107, 2000.

RICHARDSON, D. M.; REJMÁNEK. D. M. Trees and shrubs as invasive alien species - a global review. Diversity and Distributions, Matieland, v. 17, n. 5, p. 788–809, 2011.

SAKAI, A. K.; ALLENDORF, F. W.; HOLT, J. S.; LODGE, D. M.; MOLOFSKY, J.; WITH, K. A.; BAUGHMAN, S.; CABIN, R. J.; COHEN, J. E.; ELLSTRAND, N. C.; MCCAULEY, D. E.;  O’NEIL, P.; PARKER, I. M.; THOMPSON, J. N.; WELLER, S. G. The population biology of invasive species. Annual Review of Ecology, Evolution, and Systematics, v. 32, p. 305–332, 2001.

SANTIAGO-GARCIA, R. J.; COLON, S. M.; SOLLINS, P.; BLOEM. S. J. V. The role of nurse trees in mitigating fire effects on tropical dry forest restoration: a case study. Ambio, Stockholm, v. 37, n. 7-8, p. 604-608, 2008.

SÃO PAULO. Lei nº 13.550, de 2 de Junho de 2009, Dispõe sobre a utilização e proteção da vegetação nativa do Bioma Cerrado no Estado, e dá providências correlatas. 2009. Disponível em: <>. Acesso em: 01 maio de 2012.

SER - SOCIETY FOR ECOLOGICAL RESTORATION Society for Ecological  Restoration International’s primer of ecological restoration. 2004. Disponível em: <> Acesso em: 13 out. 2013.

SHULMAN, M. J.; OGDEN, J. C.; EBERSOLE, J. P.; MCFARLAND, W. N.; MILLER, S. L.; WOLF, N. G. Priority effects in the recruitment of juvenile coral reef fishes. Ecology, Washington, v. 64, n. 6, p. 1508–1513, 1983.

SILVA, J. M. C.; BATES, J. M. Biogeographic patterns and conservation in the South American Cerrado: a Tropical Savanna Hotspot. BioScience, Washington, v. 52, n. 3, p. 225-233, 2002.

VANDERHOEVEN, S.; DASSONVILLE, N.; MEERTS, P. Increased topsoil mineral nutrient concentrations under exotic invasive plants in Belgium. Plant and Soil, Dordrecht, v. 275, n. 1-2, p. 169-179, 2005.

VITOUSEK, P. M.; D’ANTONIO, C. M.; LOOPE, L. L.; WESTBROOKS, R. Biological invasions as global environmental change. American Scientist, v. 84, p. 468–478, 1996.

WAINWRIGHT C. E.; WOLKOVICH, E. M.; CLELAND, E. E. Seasonal priority effects: implications for invasion and restoration in a semi-arid system. Journal of Applied Ecology, Oxford, v. 49, n. 1, p. 234–241, 2012.

WILCOVE, D. S.; ROTHSTEIN, D.; DUBOW, J.; PHILIPS, A.; LOSOS, E. Quantifying threats to imperiled species in the United States. BioScience, Washington, v. 48, n. 8, p. 607-615, 1998.

WILLIAMSON, M. Biological Invasions. London: Chapman & Hall, 1996. 244 p.

ZAVALETA, E. S.; HOBBS, R. J.; MOONEY, H. A. Viewing invasive species removal in a whole-ecosystem context. Trends in Ecology & Evolution, Amsterdam, v. 16, n. 8, p. 454-459. 2001.

ZENNI, R. D.; ZILLER, S. R. An overview of invasive plants in Brazil. Revista Brasileira de Botânica, São Paulo, v. 34, n. 3, p. 431-446, 2011.