Introduction
Restingas are coastal plain ecosystems located along Eastern Brazil, from latitude 4º N to 34º S, corresponding to about 5 000 km of the Atlantic coast. The largest restinga systems occur in the State of Rio Grande do Sul and deltas of the major rivers of the Brazilian Southeast and Northeast regions. The soil is composed by sediments classified as marine quartz sands deposited in the Quaternary period, so its origin is related to sedimentary geological phenomena and tidal regime of each location (Lacerda, Araujo, & Maciel, 1993).
The restinga vegetation is associated with the Atlantic rainforest biome and comprises four distinct main formation zones: coastal grasslands, shrublands, open-forests and marsh zones (Silva, Izeckson, & Silva, 2000). Since these shrublands patches are, generally, formed by dense bush islets of different sizes, widely spaced by open areas, we prefer to call them “nuclei open-scrub formation”, according to the nucleation theory byYarranton and Morrison (1974).
Especially due to coastal urbanization, this is a threatened ecosystem that, through its different shrub formations, exhibits a unique mosaic, as a result of the distribution of vegetation in nuclei of different covering, physiognomy and floristic composition (Menezes, Souza, & Castro, 2007; Monteiro, Giaretta, Pereira, & Menezes, 2014). As an aggravating factor, the knowledge about the regeneration dynamics of this ecosystem is limited, and there is a lack of information about the restinga seed banks.
Seed banks, dormant, viable seeds that are present on the litter or into the soil of a given area, are related to four levels of the regeneration process: population settlement and colonization, species diversity maintenance, ecologic groups and the species richness (Uhl, Clark, Dezzeo, & Maquirino, 1988; Baker, 1989; Garwood, 1989; Grombone-Guaratini & Rodrigues, 2002). This paper aimed to characterize the environmental resilience of a non-flooded open-scrub bush restinga formation, through vegetation, leaf litter and seed bank characterization, to better understand its dynamics.
Material and methods
Study area: The study site is in the municipality of Linhares, Espírito Santo, (19°39’26.125” S, 39°51’19.739” W), Brazil (Fig. 1). In this region, there are restinga strips of more than 30 km wide interspersed by extensive open areas (Colodete & Pereira, 2007). The climate is Aw type according to Köeppen: tropical hot, with moist summers (Oct-Jan; 166 mm on average) and dry winters (water deficit from Feb-Sep), reaching 1 200 mm per year. The average temperature is 20.7 to 26.2 °C, with the coldest temperatures (20.7 oC) in June and the warmest (26.2 oC) in January.
Vegetation survey: In March 2007, a perpendicular transect 150 m away from the shoreline towards the interior of the open-scrub was established. After this first transect, five transects were positioned, parallel to the shoreline and separated by 50 m from each other. The first six bush nuclei of each transect were sampled, totaling 30 nuclei, and they were mapped as well (Fig. 1). A bush nucleus was defined as one with a diameter equal or greater than 3 m and separated from others by a distance of at least 0.5 m of bare soil. The height and the largest and smallest diameter extensions of all nuclei sampled were registered in order to obtain their volume. Nuclei diameters were measured at 1.0 m of height. The vegetation survey was conducted using the line intercept method (Müller-Dombois & Ellenberg, 1974), where all the intercepted plants were identified. The species were identified in the field or by comparison with the herbaria VIES and MBML.
Seed bank: The seed bank was sampled in the same 30 shrub nuclei defined in the vegetation survey (Fig. 1). The soil samples were obtained using a 15 x 15 x 10 cm depth frame. In each nucleus, four random units were sampled, two collected from the litter layer and two from the soil right below the litter. Litter samples (after filtering in a mesh size 6) and topsoil samples were placed into germination trays (0.61 x 0.43 x 0.10 m) with inert nursery substrate, and irrigation in a nursery (shade net covering) near the site. Additionally, control trays were assembled next to each sample tray to certify no contamination by surrounding anemochoric plant species. The composition of the seed bank was estimated using the seedling emergence method in incubated soil, which detects only the fraction of viable seeds (Brown, 1992). The observations were made every 15 days, identifying and quantifying all emergent seedlings. After emergence of the cotyledon leaves, the seedlings were removed from the trays and cultivated in plastic bags with soil, to identify them. Every 15 days, the substrate or topsoil remnant within the trays were mixed to ensure that as many viable seeds as possible would come to the surface and germinate. This procedure lasted for 12 months. Seedlings were identified and counted.
Data analysis: Phytosociology parameters -linear density, frequency and the Importance Value index of each species (Mueller-Dombois & Ellenberg, 1974)-and the Shannon-Weaver index of diversity for aboveground vegetation were calculated. The Sørensen Similarity index was used to compare the floristic composition between litter and topsoil seed bank, as well as between both and the aboveground vegetation survey. In addition, the regression between the nuclei volume and the number of species was calculated (α = 0.05) as well its determination coefficient (R²) value.
Results
Vegetation survey: The aboveground vegetation was composed by 54 species from 32 families. The families with higher species richness were Bromeliaceae (6 spp.), Orchidaceae (5), Myrtaceae (4), and Erythroxylaceae (3). We sampled a total of 1 098 individuals in 430 linear meters. Most of the species found in the aboveground were herbs (40.7 %), followed by shrubs and trees, representing 27.8 % each, and lianas (3.7 %). Davilla flexuosa, Smilax rufescens, Guapira pernambucensis, Allagoptera arenaria and Paullinia weinmanniaefolia presented the highest Importance Values (Table 1).
Family | Species | N | COV | AD | RD | AF | RF | IV | LF |
Dilleniaceae | Davilla flexuosa | 177 | 183 | 0.412 | 69.191 | 0.426 | 0.119 | 69.310 | S |
Smilacaceae | Smilax rufescens | 153 | 156 | 0.356 | 59.809 | 0.363 | 0.102 | 59.911 | L |
Nyctaginaceae | Guapira pernambucensis | 63 | 79 | 0.147 | 24.627 | 0.184 | 0.052 | 24.679 | T |
Arecaceae | Allagoptera arenaria | 62 | 143 | 0.144 | 24.236 | 0.333 | 0.093 | 24.330 | H |
Sapindaceae | Paullinia weinmanniaefolia | 60 | 60 | 0.14 | 23.455 | 0.140 | 0.039 | 23.494 | S |
Araceae | Anthurium raimundii | 59 | 61 | 0.137 | 23.064 | 0.142 | 0.040 | 23.103 | H |
Polygonaceae | Coccoloba alnifolia | 50 | 129 | 0.116 | 19.545 | 0.300 | 0.084 | 19.630 | S |
Peraceae | Pera glabrata | 50 | 127 | 0.116 | 19.545 | 0.295 | 0.083 | 19.628 | S |
Bromeliaceae | Aechmea nudicaulis | 48 | 55 | 0.112 | 18.764 | 0.128 | 0.036 | 18.800 | H |
Poaceae | Axonopus pressus | 47 | 47 | 0.109 | 18.373 | 0.109 | 0.031 | 18.403 | H |
Burseraceae | Protium heptaphyllum | 41 | 92 | 0.095 | 16.027 | 0.214 | 0.06 | 16.087 | T |
Malpighiaceae | Byrsonima sericea | 41 | 24 | 0.095 | 16.027 | 0.056 | 0.016 | 16.043 | T |
Bromeliaceae | Vriesea procera | 25 | 26 | 0.058 | 9.773 | 0.060 | 0.017 | 9.790 | H |
Clusiaceae | Clusia hilariana | 23 | 66 | 0.053 | 8.991 | 0.153 | 0.043 | 9.034 | S |
Chrysobalanaceae | Chrysobalanus icaco | 22 | 42 | 0.051 | 8.600 | 0.098 | 0.027 | 8.627 | S |
Sapindaceae | Cupania emarginata | 19 | 28 | 0.044 | 7.427 | 0.065 | 0.018 | 7.446 | T |
Primulaceae | Myrsine umbellata | 14 | 22 | 0.033 | 5.473 | 0.051 | 0.014 | 5.487 | T |
Rubiaceae | Salzmannia nitida | 14 | 18 | 0.033 | 5.473 | 0.042 | 0.012 | 5.484 | S |
Sapotaceae | Manilkara subsericea | 12 | 18 | 0.028 | 4.691 | 0.042 | 0.012 | 4.703 | S |
Primulaceae | Myrsine parvifolia | 11 | 12 | 0.026 | 4.300 | 0.028 | 0.008 | 4.308 | T |
Fabaceae | Swartzia apetala | 10 | 11 | 0.023 | 3.909 | 0.026 | 0.007 | 3.916 | T |
Cactaceae | Pilosocereus arrabidae | 9 | 12 | 0.021 | 3.518 | 0.028 | 0.008 | 3.526 | H |
Orchidaceae | Vanilla bahiana | 9 | 10 | 0.021 | 3.518 | 0.023 | 0.007 | 3.525 | H |
Fabaceae | Abarema jupunba | 7 | 15 | 0.016 | 2.736 | 0.035 | 0.010 | 2.746 | T |
Erythroxylaceae | Erythroxylum sp. (1) | 7 | 9 | 0.016 | 2.736 | 0.021 | 0.006 | 2.742 | S |
Anacardiaceae | Schinus terebinthifolius | 5 | 7 | 0.012 | 1.955 | 0.016 | 0.005 | 1.959 | T |
Bromeliaceae | Bromeliaceae sp. | 5 | 5 | 0.012 | 1.955 | 0.012 | 0.003 | 1.958 | H |
Asteraceae | Mikania glomerata | 5 | 5 | 0.012 | 1.955 | 0.012 | 0.003 | 1.958 | H |
Bromeliaceae | Vriesea neoglutinosa | 5 | 5 | 0.012 | 1.955 | 0.012 | 0.003 | 1.958 | H |
Myrtaceae | Eugenia rotundifolia | 4 | 7 | 0.009 | 1.564 | 0.016 | 0.005 | 1.568 | T |
Lauraceae | Ocotea notata | 4 | 5 | 0.009 | 1.564 | 0.012 | 0.003 | 1.567 | T |
Cactaceae | Cereus fernambucensis | 4 | 4 | 0.009 | 1.564 | 0.009 | 0.003 | 1.566 | H |
Apocynaceae | Oxypetalum banksii | 4 | 4 | 0.009 | 1.564 | 0.009 | 0.003 | 1.566 | H |
Calophyllaceae | Kielmeyera albopunctata | 3 | 4 | 0.007 | 1.173 | 0.009 | 0.003 | 1.175 | S |
Orchidaceae | Catasetum discolor | 2 | 3 | 0.005 | 0.782 | 0.007 | 0.002 | 0.784 | H |
Myrtaceae | Eugenia sp. | 2 | 4 | 0.005 | 0.782 | 0.009 | 0.003 | 0.784 | T |
Euphorbiaceae | Sebastiania glandulosa | 2 | 3 | 0.005 | 0.782 | 0.007 | 0.002 | 0.784 | S |
Orchidaceae | Tillandsia stricta | 2 | 3 | 0.005 | 0.782 | 0.007 | 0.002 | 0.784 | H |
Erythroxylaceae | Erythroxylum sp. (3) | 2 | 2 | 0.005 | 0.782 | 0.005 | 0.001 | 0.783 | S |
Bromeliaceae | Quesnelia quesneliana | 2 | 2 | 0.005 | 0.782 | 0.005 | 0.001 | 0.783 | H |
Bromeliaceae | Aechmea blanchetiana | 1 | 4 | 0.002 | 0.391 | 0.009 | 0.003 | 0.394 | H |
Polygonaceae | Coccoloba arborescens | 1 | 3 | 0.002 | 0.391 | 0.007 | 0.002 | 0.393 | S |
Bignoniaceae | Arrabidaea conjugata | 1 | 1 | 0.002 | 0.391 | 0.002 | 0.001 | 0.392 | H |
Rubiaceae | Borreria verticillata | 1 | 1 | 0.002 | 0.391 | 0.002 | 0.001 | 0.392 | H |
Capparaceae | Capparis flexuosa | 1 | 1 | 0.002 | 0.391 | 0.002 | 0.001 | 0.392 | H |
Fabaceae | Chamaecrista flexuosa | 1 | 1 | 0.002 | 0.391 | 0.002 | 0.001 | 0.392 | S |
Orchidaceae | Cyrtopodium polyphyllum | 1 | 2 | 0.002 | 0.391 | 0.005 | 0.001 | 0.392 | H |
Erythroxylaceae | Erytthroxylum sp. (2) | 1 | 1 | 0.002 | 0.391 | 0.002 | 0.001 | 0.392 | S |
Orchidaceae | Koellensteinia altissima | 1 | 2 | 0.002 | 0.391 | 0.005 | 0.001 | 0.392 | H |
Poaceae | Melinis minutiflora | 1 | 1 | 0.002 | 0.391 | 0.002 | 0.001 | 0.392 | H |
Passifloraceae | Passiflora sp. | 1 | 1 | 0.002 | 0.391 | 0.002 | 0.001 | 0.392 | L |
Schoepfiaceae | Schoepfia brasiliensis | 1 | 2 | 0.002 | 0.391 | 0.005 | 0.001 | 0.392 | T |
Myrtaceae | Myrciaria floribunda | 1 | 1 | 0.002 | 0.391 | 0.002 | 0.001 | 0.392 | T |
Myrtaceae | Psidium macahense | 1 | 1 | 0.002 | 0.391 | 0.002 | 0.001 | 0.392 | T |
Total | 1 098 | 1.53 | 2.55 | 429.22 | 3.56 | 1.00 | 430.22 | ||
Mean | 20.33 | 28.33 | 0.09 | 7 949 | 0.06 | 0.019 | 7 967 |
N = number of individuals; COV = coverage (m); AD = absolute linear density; RD = relative linear density (%); AF = absolute frequency (%); RF = relative frequency (%); IV = importance value; LF = life form: H = herb, L = liana, S = shrub, T = tree.
Three of the identified species are listed in the red book of species of Espírito Santo flora threatened by extinction (Simonelli & Fraga, 2007): Axonopus pressus, restricted to the North of the state, is listed as critically endangered, as well as Aechmea blanchetiana and Vriesea neoglutinosa, classified as vulnerable. The most frequent species was Allagoptera arenaria (Arecaceae), which occurred in all lines. Other species, such as Axonopus pressus, G. pernambucensis, S. rufescens, P. weinmanniaefolia, Pera glabrata and D. flexuosa, were also highly frequent in more than 20 nuclei. The diversity index of Shannon-Weaver (H’) was 3.08 nats for the whole experimentation site. The volumes of the studied nuclei varied from 29.3 m3 to 22 562.5 m3, with an average diameter of 11.5 m, mean area of 526.4 m2 and mean height of 2.9 m. Most nuclei had a maximum volume of 1 000 m³, with richness of 5-17 species each. The largest number of species was observed in the most voluminous nucleus (22 562.54 m³), with 31 species sampled. The second richest nucleus, despite having a volume almost eight times smaller than the first, sheltered 30 different species. Therefore, a weak tendency of increasing richness with the increasing of the nucleus size was noted, with a coefficient of termination (R²) of 0.487. The maximum nucleus height was 6 m, particularly represented by Clusia hilariana and Coccoloba alnifolia.
Seed bank: The highest proportion of germinated seeds occurred in the first four months of observation, showing a decrease in the following seven months followed by an increase in the last three months. A total of 1 839 seedlings were recorded in the seed bank (litter plus topsoil) distributed in 32 species and 19 families (Table 2). Enydra sessilis presented the highest density (544), followed by Poaceae sp. 2 (371). The most abundant family was Cyperaceae, represented by four species. Most species were herbs (56.25 %), but species of trees (12.5 %), shrubs (6.25 %) and lianas (6.25 %) were also observed.
Family | Species | N | Ts | Li | LF |
Asclepiadaceae | Oxypetalum banksii | 2 | X | liana | |
Asteraceae | Enydra sessilis | 544 | X | X | herb |
Pluchea sagittalis | 139 | X | X | herb | |
Vernonia scorpioides | 1 | X | shrub | ||
Blechnaceae | Telmatoblechnum serrulatum | 5 | X | X | herb |
Bromeliaceae | Quesnelia quesneliana | 1 | X | herb | |
Cactaceae | Pilosocereus arrabidae | 8 | X | X | herb |
Curcubitaceae | sp. 1 | 1 | X | indet. | |
Cyperaceae | Bulbostylis capillaris | 8 | X | X | herb |
Cyperus haspan | 1 | X | herb | ||
Cyperus ligularis | 138 | X | X | herb | |
Fimbristylis aspera | 3 | X | herb | ||
Erythroxylaceae | Erythroxylum sp. | 10 | X | X | tree |
Euphorbiaceae | Euphorbia heterophylla | 1 | X | herb | |
Sebastiania glandulosa | 34 | X | X | herb | |
Fabaceae | Chamaecrista ramosa | 3 | X | herb | |
Melastomataceae | Miconia albicans | 3 | X | tree | |
Molluginaceae | Mollugo verticillata | 358 | X | X | herb |
Poaceae | sp. 2 | 371 | X | X | herb |
Portulacaceae | sp. 3 | 28 | X | X | herb |
Primulaceae | Myrsine umbellata | 4 | X | tree | |
Pteridaceae | Pityrogramma calomelanos | 76 | X | X | herb |
Rubiaceae | Borreria verticillata | 2 | X | herb | |
Diodia sp. | 3 | X | herb | ||
Smilacaceae | Smilax rufescens | 1 | X | liana | |
Solanaceae | Solanum americanum | 4 | X | X | shrub |
sp. 4 | 5 | X | X | indet. | |
Indet. | sp. 1 | 54 | X | X | indet. |
sp. 2 | 3 | X | indet. | ||
sp. 3 | 16 | X | indet. | ||
sp. 4 | 5 | X | indet. | ||
Urticaceae | Cecropia hololeuca | 7 | X | tree | |
Total | 1 839 |
N = total of seedlings; Ts = topsoil layer; Li = leaf litter; LF = life form; indet. = indeterminate species.
Among the individuals listed, 64.6 % emerged from topsoil samples and 35.3 % emerged from litter. A number of 15 species were recorded exclusively in the topsoil samples, 2 exclusively in the litter and 15 species were common to both samples (Table 2). The Sørensen Similarity index (0.10) between the species sampled in the aboveground vegetation survey and those emerged from the seed bank (both litter and topsoil combined), showed low similarity (10 %), indicating that most of the species are not shared between both. Only five species were common to both aboveground and seed bank: the herbs Borreria verticillata, Pilosocereus arrabidae, Quesnelia quesneliana, Sebastiania glandulosa and the liana S. rufescens.
Discussion
In the aboveground vegetation, D. flexuosa was classified as the most abundant species in our study, and it seems to be a common species in this type of ecosystem. The palm tree A. arenaria occurred in all the transects of the sampled area and was included among the three species with the highest Importance Value. Allagoptera arenaria has a wide occurrence in both open-scrub and “closed” restinga formations with the ability to be disseminated by seed or rhizomatous growth and sprouting after burning (Pereira, Cordeiro, & Araujo, 2004). Protium heptaphyllum was among the ten species with the highest IV, like studies conducted in other restingas (Pereira& Araujo, 2000). G. pernambucensis is widely distributed throughout tropical South America, according to Araujo, Oliveira, Vieira, Barros and Lima (2001). Both B. sericea and S. terebinthifolius are more common in disturbed areas. In restinga formations, Bromeliaceae is known to be common (as the ones found in our study Aechmea blanchetiana, Aechmea nudicaulis, Q. quesneliana and Vriesea procera) and are fundamental elements in this physiognomy composition, since they play an important ecological role, including the capacity to store water in their tanks and act as sites for germination and development of other plant species, which makes them “focal plants”, important in maintaining the diversity of restinga habitats (Scarano, 2002). Melinis minutiflora, an invasive grass of African origin, reproduces both by seeds and vegetatively. It is sensitive to fire, but adapted to the condition of low soil fertility, being present in open and sunny environments. It has invaded large areas of tropical ecosystems, displacing native species due to its aggressiveness and superior competitive capacity. This species occurred in low abundance in our study, but still deserves special attention due to its biological invasive potential. A. pressus occurs in the open formation of Ericaceae restinga and was listed in the Espírito Santo red book as critically endangered, as well as A. blanchetiana and V. neoglutinosa, which are listed as vulnerable species (Simonelli & Fraga, 2007). About rare species, in our study they were represented only by 1.45 % of the total sampled, analogous to the rare species percentage of a restinga in Rio de Janeiro (2 %) (Pereira et al., 2004), but lower than the rate in another restinga in Espírito Santo (7.1 %) (Monteiro et al., 2014). Therefore, the Shannon-Weaner diversity index (H’) obtained in our study (H = 3.08 nats) was higher than the H’ = 2.84 nats found by Pereira& Araujo (2000) and H’ = 1.89 nats registered in Pereira et al. (2004), both in open restingas, probably because it is a more conserved site than the other studies cited above.
Although the regression´s coefficient of determination was low, there was an increase in the number of species in relation to the nuclei volume, and the highest number of species was observed in the most voluminous nuclei. This might be related to the fact that some species might be more abundant under shrubs and others would prefer areas without coverage to develop (Shmida & Whittaker, 1981). Even though restingas are highly stressful ecosystems in terms of salinity, low fertilization, high radiation, low water availability on soil, wind exposure and sand burial, dense groups of plants facilitate the establishment of new species in these ecosystems due to their improvement in microclimate factors (Shumway, 2000). The nuclei found in this study may act in a nucleation process (Bechara et. al, 2016). These plants may play a key role in plants assemblages, especially in xerophytic environments, since they contribute to the improvement of conditions for germination, establishment and growth of other plant species (Zaluar & Scarano, 2000). For example, Clusia hilariana, which is often described as a nurse-plant, since it presents positive association between adults and juvenile density of other woody species and works as dispersers attractive in restingas (Dias, Zaluar, Ganade, & Scarano, 2005; Correia, Dias, & Scarano 2010).
The seedlings diversity from seed bank in our study is like other studies in restingas. Cyperaceae, the most abundant family, and Cyperus spp., are frequently found on restingas seed banks. Asteraceae, the second most abundant family, was also one of the most significant families in the floristic composition of Santa Catarina’s restingas (Klein, Citadini-Zanette & Santos, 2007; Korte, de Gasper, Kruger, & Sevegnani, 2013). The most abundant species, E. sessilis, occurs in aquatic and terrestrial sites in restingas, so it can be considered an amphibian species. Poaceae, as the second family with more individuals, also occur in other restingas sites. The third most abundant species, M. verticillata, is one of the ten most abundant species as well in a restinga in Rio de Janeiro (Pereira et al., 2004). The predominance of herbs in the seed bank and in the aboveground vegetation is typical in restingas, and they are important especially in terms of soil coverage, interaction with other life form species and early ecological succession.
Regarding the germination over time, it is commonly reported that seed banks have rapid germination response in the first months of studies. Costa and Araújo (2003) reported the highest proportion of germinated seeds in the first month of observation. Caldato, Floss, Croce and Longhi (1996) also found that most of the seedlings emerged in the first four months, as in our study. Nonetheless, it is important to highlight that the soil seed bank of open areas such as restingas is mainly composed of species that germinate under constant sunlight. Most of the taxa emerged from seed bank in our study follow a pioneer species behavior, which is common in restingas, where the successional groups form mosaics defined by the diversity of different levels of soil-dependent climax. The richness of this seed bank (33 spp.) is like the results obtained by Bechara and Reis (2009), where the authors found a diversity of 35 species in a restinga site in Santa Catarina.
The low similarity between the aboveground vegetation floristic composition and the sampled seed bank indicates that early communities, after disturbances, could be significantly different in these delicate ecosystems. The higher diversity in the aboveground vegetation indicates a more mature community. The high abundance of seedlings found in seed banks indicates that they may be used as an important low-cost tool for open-scrub restinga restoration in nuclei (see Bechara et al., 2016).
Ethical statement: authors declare that they all agree with this publication and made significant contributions; that there is no conflict of interest of any kind; and that we followed all pertinent ethical and legal procedures and requirements. All financial sources are fully and clearly stated in the acknowledgements section. A signed document has been filed in the journal archives.