- Email: [email protected]
Composition, species richness, and density of the germinable seed bank over 4 years in young and mature forests in Brazilian semiarid regions
Journal of Arid Environments 129 (2016) 93e101
Contents lists available at ScienceDirect
Journal of Arid Environments journal homepage: www.elsevier.com/locate/jaridenv
Composition, species richness, and density of the germinable seed bank over 4 years in young and mature forests in Brazilian semiarid regions ~o Fraga dos Santos a, Danielle Melo dos Santos a, *, Josiene Maria Falca b Kleber Andrade da Silva , Vanessa Kelly Rodrigues de Araújo a, Elcida de Lima Araújo a ^nica, Dois Irma ~os, 52171-900, Recife-PE, Brazil Universidade Federal Rural de Pernambuco, Departamento de biologia, Area Bota ~o-PE, ria, Rua do Alto do Reservato rio s/n, Bela Vista CEP: 55608-680, Vito ria de Santo Anta Universidade Federal de Pernambuco, Centro Acad^ emico de Vito Brazil a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 3 October 2014 Received in revised form 15 February 2016 Accepted 19 February 2016 Available online xxx
The soil seed bank is an important ecological component in forest regeneration. In semiarid regions, forest regeneration is highly affected by seasonal and interannual variations in precipitation, because these variations may affect the composition, species richness, and density of seeds in the soil. This study aimed to characterize and compare these parameters of the germinable seed bank in fragments of young and mature Caatinga forest between various seasons and over consecutive years. A total of 105 soil samples were collected in 20 20 5 cm plots in each forest (Young and Mature) at the end of the rainy and dry seasons over 4 years (2009e2012), totaling 840 samples. The composition, species richness, and density of seeds were determined by the method of seedling emergence. Over the 4 years, 121 species emerged from the soil seed bank, 86 in the young forest, and 109 in the mature forest. Significant differences in the composition, richness, and density were recorded between forests, seasons, and years, with a significant interaction between them. Relationship between rainfall and age of forests affects the dynamics of the soil seed bank in semiarid environments, which are important distinctions for the maintenance of these areas. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Caatinga Seedling emergence Natural regeneration
1. Introduction Severe anthropogenic disturbances, including complete destruction of bushland for the establishment of pasture lands and agriculture, have modified dry and humid environments in the world and endangered climatic characteristics as well as maintenance of ecosystem function and biodiversity conservation, necessitating the comprehension of their influence on natural regeneration, which involves resilience of forests (Santos et al., 2013a; Esaete et al., 2014; Randriamalala et al., 2015; Valenta et al., 2015). Among the processes involved in the regeneration of forests, the dynamic of seed bank plays an important role in carrying the germplasm, which enables the formation of new secondary forests and re-acquirement of areas previously occupied by humans
* Corresponding author. E-mail address: [email protected] (D.M. Santos). http://dx.doi.org/10.1016/j.jaridenv.2016.02.012 0140-1963/© 2016 Elsevier Ltd. All rights reserved.
(Kassahun et al., 2009; Golos and Dixon, 2014). However, type of and historical land use, associated with local climate changes, generates complexity on the dynamic of soil seed bank, considering the variation of the time taken for the recovery of species composition and structure of forests (Heydari et al., 2014; Mendes et al., 2015). Despite the complexity, three trends are reported: (1) With increased level of degradation of such area, a decrease in seed density and species richness of the soil seed bank occurs (Karlík and Pochlod, 2014; Mendes et al., 2015). (2) The new secondary forests consisting of abandoned areas of agriculture and pasture show changes in species composition and population density and low similarity with species composition of soil seed bank (Pereira et al., 2003; Kassahun et al., 2009; Heydari et al., 2014). (3) The recruitment and regeneration of degraded areas highly depend on the dormancy capacity of the seeds stored in the soil bank (Liu et al., 2009; Golos and Dixon, 2014). However, interannual variations in precipitation can increase the complexity and modify the trends reported in the literature,
94
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101
particularly in semiarid regions with well-defined rainy and dry seasons, because they induce differences in the time of production of fruits and seeds (Selwyn and Parthasarathy, 2006; Valdezndez et al., 2010; Souza et al., 2014), thereby affecting Herna dispersal and seed rain. For instance, interannual variations and irregularities in quantitative and rainfall distribution have a significant effect on the reproductive behavior of plants (Albuquerque et al., 2012; Santos et al., 2014) and the intensity of seed rain (Souza et al., 2014) in the vegetation of Brazilian semiarid Caatinga. This explains part of the wealth of species and seed density found in the soil bank of mature forests of this type of vegetation (Silva et al., 2013). Thus, this study characterizes and compares the composition, species richness, and density of the germinable seed bank in young and mature forests of Caatinga between various seasons and over consecutive years. In particular, we investigate whether (1) the species richness and seed density of the soil bank are smaller in the young forest, after 15 years of abandonment of agricultural activity; (2) anthropogenic disturbance and seasonal and interannual variations in rainfall affect the floristic composition of the remaining soil seed bank; and (3) the species richness and seed density of the soil seed bank are lower in dry seasons and driest years than in the rainy seasons and wettest years, in both young and mature forests.
2. Material and methods 2.1. Characterization of the study area The study was conducted at the Agronomic Institute of Pernambuco e IPA (8 140 S and 35 550 W, 537-m altitude), located in the municipality of Caruaru, Pernambuco, Brazil. The site is located in the rural zone at a distance of 9 km from the nearest city. This is a semiarid region, with minimum and maximum temperatures of 11 and 38 C, respectively. The average annual rainfall is 694 mm and the rainy season spans from March to August, with few months having rainfall >100 mm. The dry season spans from September to February, with average monthly rainfall of <30 mm. However, occasional or erratic rainfall may occur in the dry season and dry spells may occur in the rainy season (Araújo et al., 2005a). The range of total rainfall recorded in the study years 2009, 2010, 2011, and 2012 is 350.8e1031.2 mm (Fig. 1). The local seasonality determines the deciduousness of the woody flora in the dry season, and therophytic herbs can only be
observed in the rainy season (Santos et al., 2013a). Moreover, seasonal variations in rainfall influence the rhythm of seed rain in the local vegetation, existing in three species groups: (1) species that disperse seeds only in the rainy season; (2) species that disperse seeds only in the dry season; and (3) species that disperse seeds throughout the year (Souza et al., 2014). The experimental site occupied an area of 190 ha and was created with the primary aim of developing research for agriculture and livestock activities. Before the experiment, the area was occupied by a single patch of mature Caatinga vegetation; however, the current native vegetation is reduced to a small fragment of approximately 30 ha of mature forest. In this area, woody species of the families Mimosaceae, Caesalpiniaceae, Euphorbiaceae, and Cactaceae (Alcoforado-Filho et al., 2003; Araújo et al., 2007) form a closed canopy, which provides a longer period of soil moisture (Santos et al., 2013b). Species of the families Poaceae, Euphorbiaceae, Convolvulaceae, Malvaceae, Asteraceae, and Fabaceae (Araújo et al., 2005a; Reis et al., 2006) predominate in the herbaceous component. This fragment has been preserved for approximately 50 years by not allowing the entry of animals and removal of vegetation (Lopes et al., 2012; Santos et al., 2013a). This fragment is the first area of study, and is named the mature forest. For the last 19 years, a stretch of 3 ha near the mature forest suffered clear-cutting for the cultivation of Opuntia ficus-indica Mill., Cactaceae (commonly called “sweet palm”) (Santos et al., 2013a). No fire, fertilizer, or manure was used during the period of cultivation (Lopes et al., 2012; Souza et al., 2014). After 6 months, this fragment was abandoned and the young forest regenerated naturally. Currently, the herbaceous component of the young forest has rich species of Poaceae and Cyperaceae (Santos et al., 2013a) and the woody component is represented by some young species such as Poincianella pyramidalis (Tul.) L.P. Queiroz (catingueira), Acacia paniculata Willd. (unha de gato), and Anadenanthera macrocarpa (Benth.) Brenan (angico) (Lopes et al., 2012), which do not form a closed canopy, thereby promoting a higher incidence of light in this area (Andrade, unpublished data). This fragment is named the young forest.
2.2. Samples of the soil seed bank For both mature and young forest fragments, there exists a fragment of 1 ha, where studies on the woody and herbaceous
Fig. 1. Monthly precipitation and total precipitation during the rainy and dry seasons over 4 years. Solid arrows indicate the samples collected at the end of the rainy seasons and dashed arrows indicate the samples collected at the end of the dry seasons. Data were provided by the meteorological station of the Empresa Pernambucana de Pesquisa Agropecu aria (IPA) in Caruaru, Pernambuco, Brazil.
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101
components and seed rain are being undertaken (Araújo et al., 2005b; Souza et al., 2014). Therefore, in this fragment, 105 plots of dimension 1 1 m were randomly allocated for the study of herbaceous vegetation (Reis et al., 2006; Santos et al., 2013a), totaling 210 plots. In addition, in the area surrounding these plots (both forests), 105 soil samples were collected at the end of both rainy and dry seasons for 4 consecutive years (to enable comparison of the temporal effect between the areas), totaling 840 samples in each forest (Fig. 2). Soil from a depth of 5 cm, including the leaf litter layer, was collected in plots made of a stainless steel sheet of dimension 20 20 cm. This procedure was used in most studies on soil seed banks (Ma et al., 2006; Hegazy et al., 2009; Ne'eman and Izhaki, 2009; Quevedo-Robledo et al., 2010). All samples were packed in plastic bags and labeled by the respective plot and forest, and, in the greenhouse, each sample was placed in a polystyrene tray (20 38 3 cm) and irrigated daily for the seasonal duration of 6 months, enabling the comparison of the effect of seasonality between collections in young and mature forests. The trays were arranged in two rows with a control tray between them, containing sterilized soil, to detect possible contamination caused by seeds dispersed by wind. Contamination was not detected during the study period. Seedling emergence was the method used to determine the seed density in the soil seed bank (not considering the density of dormant seeds in the sample), following the methodology adopted by Silva et al. (2013). In this method, each seedling that emerged from soil represents one seed from the soil seed bank. Data were expressed in seeds per square meter to allow comparison with other studies (Ne'eman and Izhaki, 2009; Hegazy et al., 2009). Emerging seedlings of each soil sample were counted and labeled every day, noting the date of germination, plot number, and the forest in which the sample was collected. Once the size of seedlings reached approximately 5 cm, they were transplanted into polythene bags, irrigated daily, kept at an average temperature of 25 C, and monitored for 6 months in
Fig. 2. Schematic view of the collection of the soil seed bank (20 20 5 cm plots) surrounding the 1 1 m fixed plots in a semiarid region in the northeast of Brazil.
95
greenhouse to obtain reproductive material of therophytic herbs for correct taxonomic identification of the species. Species identification was performed through consultations in the literature and comparisons with herbarium specimens deposited in the herbaria Prof. Vasconcelos Sobrinho (PEUFR) and rdano de Andrade Lima (IPA), adopting the APG3 classification Da system (Souza and Lorenzi, 2012). Unidentified seedlings were listed as morphospecies. 2.3. Analyses of the seed bank The floristic compositions of both forests (young and mature) were compared between years by a nonmetric multidimensional scaling (NMDS) analysis using the BrayeCurtis dissimilarity matrix, based on the relative density of the species in the 105 sample units in each study area. The analysis of similarities (ANOSIM) was used to test the significance of the group formed in the NMDS. In order to verify the contribution of each species between seasons and between forests, a similarity percentage analysis (SIMPER) was performed. Software Primer version 6.1.6 (Clarke and Gorley, 2006) was used for ANOSIM, NMDS and SIMPER analyses. In order to verify the effect of predictor variables (age of the forest e young or mature; seasonal or interannual variation in precipitation) on species richness and seed density, a generalized linear model (GLM) analysis was performed. Differences in species richness and seed density between young and mature forests, rainy and dry seasons, and years were verified by the Tukey a posteriori test. All these analyses were carried out using the software Statistic 7.0 (StatSoft Inc., 2003). 3. Results 3.1. Floristic composition The floristic composition differs between the young and mature forests (R-global ¼ 0.898, p ¼ 0.001, Fig. 3). In the former, a seasonal variation in floristic composition was present (R-global ¼ 0.274, p ¼ 0.001, Fig. 4a), which was found to be similar between years (Rglobal ¼ 0.002, p ¼ 0.555). Seasonal (R-global ¼ 0.126, p ¼ 0.001, Fig. 4b) and interannual (R-global ¼ 0.018, p ¼ 0.031, Fig. 5) variations in floristic composition were present in the mature forest, where the assemblage of species recorded in the fourth year was
Fig. 3. Ordination formed after multidimensional scaling (MDS) analysis on the germinated seeds of the germinable soil seed bank in young and mature forests during 4 years of study, based on species richness per area. The symbols on the graph represent the samples of soil and their respective germinated species.
96
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101
seasons and between the mature and young forests. The five species that contributed most to the differences between rainy and dry seasons, in descending order, were Delilia biflora, Pilea hyalina, Panicum trichoides, Heliotropium angiospermum, and Panicum venezuelae (Table 1). The cumulative percentage contribution of these species in the differences between seasons was 40.05%. The analysis of dissimilarity between the young and mature forests showed the aforementioned five species as the main species, which contribute to the differences between the areas, accumulating 43.15% of the contribution (Table 1).
3.2. Species richness Over the 4 years, a total of 121 species were recorded in the soil seed bank, 86 species from the young forest, and 109 from the mature forest (Appendix 1). Of the total number of species, 16 were identified to family level, 17 to genus level, 84 to species level, and only five as morphospecies. A total of 13 and 36 species occurred exclusively in the young and mature forests, respectively (Appendix 1).
Table 1 SIMPER analysis between the rainy and dry seasons, between the mature and young forests, and between the 4 years of study with the contribution of each species dissimilar between sampling sites (Av.Diss.).
Fig. 4. Ordination formed after multidimensional scaling (MDS) analysis of the germinated seeds of the germinable soil seed bank between seasons (rainy and dry) in areas of young (A) and mature (B) forests over 4 consecutive years. This graph was produced based on species richness per area. The symbols on the graph represent the soil samples and their respective germinated species only considering the climatic seasons.
Fig. 5. Ordination formed after multidimensional scaling (MDS) analysis of the germinated seeds of the germinable soil seed bank over 4 consecutive years in mature forests. This graph was produced based on species richness per area. The symbols on the graph represent the soil samples and their respective germinated species, only considering the 4 years of study.
different from that recorded in the first 3 years (Fig. 5). The SIMPER analysis showed a high dissimilarity between
Dry and rainy seasons
Young and mature forests
Av. Diss. ¼ 86.29
Av. Diss. ¼ 92.18
Species
Av. Diss
Species
Av. Diss
Delília biflora Pilea Hyalina Panicum trichoides Heliotropium angiospermum Panicum venezuelae Poaceae 1 Phaseolus pedunculares Ruellia sp1 Begonia reniformis Tallinum triangulare Poaceae 2 Desmodium glabrum Gomphrena vaga Malvaceae 1 Bidens bipinnata Callisia repens Gnaphalium spicatum Ruellia asperulla Dioscorea coronata Ruellia bahiensis Pappophorum papipherum Hippeastrum sp1 Croton blanchetianus Cyperus uncinulatus Mimosa arenosa Croton heliotropiifolius Urochloa maxima Mimosaceae 1 Cactaceae 1 cie 2 Morfoespe cie 1 Morfoespe Chamaesyce hyssopifolia Blainvillea acmella Solanum americanum Myracrodruon urundeuva Vigna sp1 Conocliniopsis prasiifolia Dalechampia scandens Mollugo verticillata Phyllantus niruri Commelina obliqua Lippia americana
10.24 9.77 6.97 4.08 3.49 2.63 2.63 2.50 2.43 2.12 2.06 1.95 1.94 1.77 1.53 1.39 1.27 1.24 1.22 1.18 1.14 1.09 1.07 1.01 1.01 0.98 0.96 0.89 0.87 0.70 0.57 0.52 0.47 0.47 0.47 0.45 0.45 0.43 0.40 0.37 0.36 0.35
Pilea Hyalina Delília biflora Panicum trichoides Heliotropium angiospermum Panicum venezuelae Poaceae 1 Phaseolus pedunculares Begonia reniformis Ruellia sp1 Poaceae 2 Tallinum triangulare Desmodium glabrum Gomphrena vaga Malvaceae 1 Bidens bipinnata Callisia repens Gnaphalium spicatum Ruellia bahiensis Ruellia asperulla Pappophorum papipherum Dioscorea coronata Hippeastrum sp1 Croton blanchetianus Cyperus uncinulatus Mimosa arenosa Croton heliotropiifolius Urochloa maxima Mimosaceae 1 Cactaceae 1 cie 2 Morfoespe Chamaesyce hyssopifolia Blainvillea acmella cie 1 Morfoespe Solanum americanum Vigna sp1 Myracrodruon urundeuva Dalechampia scandens Conocliniopsis prasiifolia Mollugo verticillata Commelina obliqua Selaginella sulcata e
12.47 11.81 7.58 4.43 3.49 2.69 2.56 2.56 2.48 2.18 2.07 2.05 1.94 1.87 1.59 1.37 1.30 1.28 1.24 1.24 1.22 1.13 1.08 1.04 1.02 1.00 0.98 0.88 0.84 0.73 0.56 0.52 0.52 0.48 0.46 0.44 0.44 0.41 0.41 0.36 0.35 e
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101
97
Of the exclusive species in each forest, only four species occurred in the two seasons (rainy and dry) over the 4 years (Lippia americana L. in the young forest; Begonia reniformis Vell.; Cactaceae sp1; and Schinopsis brasiliensis Engl. in the mature forest). Of the total species found in both forests, 100 were found to be herbaceous, 11 were shrubs, and 10 were trees (Appendix 1). In both forests, families with higher species richness over the 4 years of study were Euphorbiaceae (14), Fabaceae (14), Poaceae (10), Asteraceae (11), Portulacaceae (five), and Malvaceae (six) (Appendix 1). Furthermore, we found that 28 species occurred in the two seasons (rainy and dry) over the 4 years of study (Appendix 1). Of the 28 species, only D. biflora (L.) Kuntze, P. trichoides Sw, and P. hyalina Fenzl. occurred in both forests. The GLM analysis showed that the age of the forest, seasonal and interannual variations in precipitation, and most of their interactions explain the variations in species richness (Table 2). Over the 4 years, in both the young and mature forests, species richness was higher in the dry season (Fig. 6). Comparing the study areas, species richness was higher in the mature forest than the young forest, only in the rainy season (Fig. 6). In the mature forest, species richness was higher in the second and third years (Fig. 7). In the young forest, species richness was highest in the first year and decreased significantly over time (Fig. 7). Considering each year separately, in the mature forest, the species richness was higher in the dry seasons of the second and fourth years (Fig. 8). In the young forest, species richness was higher in the dry seasons of the first and fourth years (Fig. 8). 3.3. Seed density During the 4 years of study, a total of 4097 seeds m2 were recorded in the soil seed bank, with 1594 and 2503 seeds m2 from the young and mature forests, respectively. Separate analysis of each season showed that, in the rainy and dry seasons, the respective total numbers of seeds recorded per square meter were 460 and 1133 for the young forest and 854 and 1649 for the mature forest. The GLM analysis showed that the age of the forest, seasonal and interannual variations in precipitation, and most of their interactions explained the variations in mean seed density (Table 2). Considering the 4 years, the mean density was higher in the mature forest. In both forests, the mean density was higher in the dry season (Fig. 6). In the mature forest, the mean density was higher in the second and third years. In the young forest, the mean density was higher in the first year and decreased significantly in the last year (Fig. 7). Considering each year separately, the mean density was higher in the dry seasons of the second and fourth years in the mature
Fig. 6. Spatiotemporal variation in mean richness (number of species/20 20 cm plot) and mean seed density (seeds/20 20 cm plot) in areas of young and mature forests over 4 years in a semiarid region in the northeast of Brazil. Different letters between seasons (rainy and dry) in each area (mature and young) and between areas in each season show a significant difference in the Tukey test. Vertical bars denote a confidence interval of 0.95.
forest, and first and fourth years in the young forest (Fig. 8). In the second and third years, the mean density was higher in the mature forest (Fig. 8).
Table 2 GLM analysis showing the influence of the age of the forest (young and mature), seasonal and annual variation in precipitation, and their interactions on the density and species richness of the germinable seed bank in a semiarid region in the northeast of Brazil. p values denote statistical significance (GL ¼ degrees of freedom; SQ ¼ sum of squares; QM ¼ mean square, F ¼ Fisher test). Variables
Interceptor Age Year Season Age*Year Age*Season Year*Season Age*Year*Season Error Total
Density
Richness
GL
SQ
QM
F
P
GL
SQ
QM
F
P
1 1 3 1 3 1 3 3 1664 1679
176607.5 8745.2 8525.4 22755.6 14297.0 162.8 3006.5 8421.8 591339.2 657253.5
176607.5 8745.2 2841.8 22755.6 4765.7 162.8 1002.2 2807.3 355.4
496.9650 24.6085 7.9967 64.0333 13.4104 0.4582 2.8200 7.8995
0.000000 0.000001 0.000027 0.000000 0.000000 0.498582 0.037721 0.000031
1 3 1 1 3 3 1 3 1664 1679
16783.39 621.34 1160.01 85.95 499.21 679.35 71.26 427.75 7415.73 10960.61
16783.39 207.11 1160.01 85.95 499.21 679.35 71.26 142.58 4.46
3.765.988 46.473 260.292 19.287 37.339 50.812 15.990 31.994
0.000000 0.000000 0.000000 0.000012 0.000000 0.000000 0.000066 0.000000
98
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101
Fig. 7. Annual change in mean richness (number of species/20 20 cm plot) and mean seed density (seeds/20 20 cm plot) in areas of young and mature forests over 4 years in a semiarid region in the northeast of Brazil. Uppercase letters between different years in each area and lowercase letters between different areas each year denote significant differences in the Tukey test. Vertical bars denote a confidence interval of 0.95.
4. Discussion 4.1. Effect of age of the forest on the soil seed bank In this study, we show that, on semiarid regions, species richness is not always lower in new young forests of the anthropogenic areas. In the dry seasons, wealth of young forests was found to be similar to that of the mature ones. Therefore, a part of first hypothesis was refused, although the literature reported a higher speed of resilience process in the dry forests than the wet forests (Lopes et al., 2012). This finding shows that studies evaluating the effect of anthropic actions on the species richness of the seed bank need to consider larger time intervals (4 years) for including temporal stochasticity of species richness of the soil seed bank of young and mature forests. Anthropogenic disturbances affect the number of native seeds reaching the soil bank and consequently the regeneration of degraded areas (Ferreira et al., 2015). Hence, some studies report that the seed density of the soil bank of disturbed areas is lower than that of the original forest (Tessema et al., 2012; Heydari et al., 2014; Karlík and Pochlod, 2014). This trend was also found in this study, endorsing the other part of the first hypothesis. In addition, we showed that anthropogenic disturbances cause a drastic change
in floristic composition, confirming part of our second hypothesis. Our results show that 90% of the difference in floristic composition was explained by forest age. Consequently, 15 years are not enough for the seed bank to recover the floristic composition and seed density of the original forest. Some studies indicate that a high number of seeds, weeds, or invasive plants occurs in areas that suffered disturbance. This is because (1) degradation promotes disappearance of native species with low colonization ability (Augusto et al., 2001; Ferreira et al., 2015) or (2) woody species of communities existing before the anthropogenic practices are not present on the seed bank left in the soil (Wang et al., 2009). Invasive or weed species can compete with native species, affecting the natural regeneration (Ferreira et al., 2015). However, despite the occurrence of drastic change in the set of colonizing species after disturbance, we believe that no risk occurred in the natural regeneration of the young forest in our study, because its seed bank was not composed of weed species. Most species found in the soil seed bank were annual herbaceous plants, and all of them are native species of the two forests (Santos et al., 2013b). Herbaceous species play an important role in the functioning of the ecosystem and correspond to most of the phytodiversity of dry forests with a significant influence on the dynamics of woody plants (Wassiea and Teketay, 2006; Wang et al., 2009; Santos et al., 2010). They also protect the seedlings of woody species against the impact of rain and assist in the retention of their seeds on the soil surface (Santos et al., 2013b). Furthermore, a specific assemblage of herbaceous species contributed to the differences in floristic composition between seasons and forests. Thus, this group of species has probably the highest potential for regeneration with a diverse soil seed bank exhibiting a higher longevity (Wassiea and Teketay, 2006; Wang et al., 2009; Golos and Dixon, 2014). Therefore, this study shows that perhaps this group of native and exclusive species of the disturbed forests can help recuperate the areas, as that occurred in our young forest, which suffered clear-cutting for the cultivation of sweet palm (O. ficus-indica Mill.). There was only one adult woody specie in the young forest that has enabled a higher light incidence and increased drying of the soil (Lopes et al., 2012; Santos et al., 2013a). On the contrary, the mature forest had woody plants reaching an average height of 5e7 m, with relatively overlapping crowns, providing more shading on the ground, which remained humid for more number of months (Santos et al., 2013b). Thus, these characteristics might have aided in the establishment of species, which requires more shade, promoting a higher seed density in the mature forest and a different floristic composition compared with the young forest. It is important to note that the floristic composition, richness, and density of seed bank can significantly increase during regeneration time. The longer the duration of regeneration of new forest, the higher the similarity between the floristic composition, species richness, and amount of seeds found in the bank of this soil in relation to the original forest (Chaideftou et al., 2009; Valenta et al., 2015). Thus, further studies on the seed bank of the Caatinga vegetation, with different ages and usage history, are necessary to find the time required to complete resilience of its forests. 4.2. Effect of temporal variation in precipitation on the soil seed bank In this study, seasonal variation in precipitation changed the floristic composition, species richness, and density of the soil seed bank, regardless of the age of the forest. This fact follows the pattern of dry forests, because the composition, richness, and seed density change faster with seasonal variation in these
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101
99
Fig. 8. Seasonal variation in mean richness (number of species/20 20 cm plot) and mean seed density (seeds/20 20 cm plot) in areas of young and mature forests over 4 years in a semiarid region in the northeast of Brazil. Different letters between seasons (rainy and dry) in each area show a significant difference in the Tukey test. Vertical bars denote a confidence interval of 0.95.
, 2000; Santos et al., 2010; environments (Pugnaire and Lazaro Mendes et al., 2015). Most studies on seed banks emphasize that the number of species and seeds is higher during the rainy season and the floristic pez, 2003; Mayor et al., composition is a function of this season (Lo 2003; Santos et al., 2010; Mendes et al., 2015). However, it is important to note that the trend for higher species richness and density in the rainy season was not recorded in this study. Thus, part of our third hypothesis was not accepted. The assemblage of species recorded only in the dry season was responsible for the changes in floristic composition. In addition, we also believe that the highest species richness and seed density recorded in this study in the dry season may be justified for the following reasons: (1) the sum of the amount of seeds dispersed during the rainy and dry seasons and (2) some
seeds dispersed in the rainy season need to spend the entire dry season in the soil to complete their maturation process. These reasons can also explain the difference in floristic composition between seasons. Therefore, further studies are necessary to test this hypothesis. With regard to interannual variations, the literature indicates that richness and composition of species, as well as the amount of seeds is directly proportional to the intensity of the annual rainfall pez, 2003; Facelli et al., in dry forests (Costa and Araújo, 2003; Lo 2005; Santos et al., 2010; Silva et al., 2013; Souza et al., 2014). In this study, this trend was recorded for species richness and seed density of the two forests, confirming another part of the third hypothesis. In the mature forest, species richness and seed density were higher in the second and third years, which have had the highest average annual rainfall (865 and 1031.2 mm, respectively).
100
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101
For that reason, in the young forest, species richness and seed density decreased significantly in the fourth year, which showed the lowest annual rainfall (350.8 mm), compared with other years. Interannual variations in seed bank attributes can also be explained by the previous year's precipitation. In the same area where this study was conducted, the previous year's precipitation explained 21% of the variation in species richness and 16% of the variation in seed density in the mature forest (Silva et al., 2013). Therefore, we also believe that the interannual variation in species richness and density of the seed bank recorded in this study in young and mature forests can be explained by the previous year's precipitation. Nevertheless, interannual variations in floristic composition differed between the forests. No interannual variation in the floristic composition of the young forest was present. Therefore, the assemblage of species that make up this area is possibly not highly influenced by interannual variation in precipitation, not confirming part of the second hypothesis. Moreover, in the mature forest, differences in the floristic composition occurred over the years. It is noteworthy that this difference was very small, because the Rglobal (0.018) was very low. Interannual changes in the floristic composition of the seed bank were also observed in other studies (Janicka, 2006; Chaideftou et al., 2009; Santos et al., 2010, 2013b; Silva et al., 2013). In the mature forest, the assemblage of species found in the fourth year differed from that in other years. This may be due to the sharp reduction in the annual average precipitation from the third (1031.2 mm) to the fourth year (350.8 mm). Consequently, drastic reductions in annual rainfall may reduce seed production of the local community and, consequently, cause the disappearance of some species from the soil seed bank. This fact can be confirmed by comparing the amount of unique species in the third year (11) and the fourth year (five). In conclusion, we found that (1) Human activities cause changes in floristic composition, species richness, and density of the soil seed bank; (2) Temporal variation in precipitation caused changes in species richness and density of the soil seed bank, regardless of the age of the forest; and (3) The effect of variation of interannual precipitation on the floristic composition of young forest is not clear in semiarid environments, after disturbance. Acknowledgments The authors thank the Agronomic Institute of Agricultural Research (IPA) for logistics and permission to work on their property; researchers from the Laboratory of Plant Ecology and Northeastern Ecosystems (LEVEN) for their support, suggestions, and assistance in the execution of the project; and CAPES and CNPq (302645/2014-4) for the award of the student scholarship and productivity scholarship for researchers, as well as financial support of the project. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jaridenv.2016.02.012. References Albuquerque, U.P., Araújo, E.L., El-Deir, A.C.A., Lima, A.L.A., Souto, A., Bezerra, B.M., Ferraz, E.M.N., Freire, E.M.X., Sampaio, E.V.S.B., Las-Casas, F.M.G., Moura, G.J.B., Pereira, G.A., Melo, J.G., Ramos, M.A., Rodal, M.J.N., Schiel, N., Lyra-Neves, R.M., Alves, R.R.N., Azevedo-Junior, S.M., Telino-Junior, W.R., Severi, W., 2012. Caatinga revisited: ecology and conservation of an important seasonal dry forest. Sci. World J. 1e18. Alcoforado-Filho, F.G., Sampaio, E.V.S.B., Rodal, M.J.N., 2003. Florística
lia espinhosa arbo rea fitossociologia de um remanescente de vegetaç~ ao caducifo em Caruaru. Acta Bot. Bras. 17, 287e303. Araújo, E.L., Silva, K.A., Ferraz, E.M.N., Sampaio, E.V.S.B., Silva, S.I., 2005a. Diversirea de dade de herb aceas em microhabitats rochoso, plano e ciliar em uma a Caatinga, Caruaru-PE. Acta Bot. Bras. 19, 285e294. Araújo, E.L., Martins, F.R., Santos, A.M., 2005b. Establishment and death of two dry tropical forest woody species in dry and rainy seasons in northeastern Brazil. In: Nogueira, R.J.M.C., Araújo, E.L., Willadino, L.G., Cavalcante, U.M.T. (Eds.), fica e Editora, Estresses ambientais: danos e benefícios em plantas. MXM Gra Recife, pp. 76e91. Araújo, E.L., Castro, C.C., Albuquerque, U.P., 2007. Dynamics of brazilian Caatinga e a review concerning the plants, environment and people. Funct. Ecol. Comm. 1, 15e28. Augusto, L., Dupouey, J.L., Picard, J.F., Ranger, J., 2001. Potential contribution of the seed bank in coniferous plantations to the restoration of native deciduous forest vegetation. Acta Oecol. 22, 87e98. Chaideftou, E., Thanos, C.A., Bergneier, E., Kallimanis, A., Dimopoulos, P., 2009. Seed bank composition and above-ground vegetation in response to grazing in submediterranean oak forests (NW Greece). Plant Ecol. 201, 255e265. Clarke, K.R., Gorley, R.N., 2006. Primer v6: User Manual/tutorial. Plymouth Marine Laboratory, Plymouth. ~o e flora do banco de sementes Costa, R.C., Araújo, F.S., 2003. Densidade, germinaça do solo no final da estaç~ ao seca, em uma area de Caatinga, Quixad a, CE. Acta Bot. Bras. 17, 259e264. Esaete, J., Eycott, E.A., Reini, J., Telford, J.R., Vandvik, V., 2014. The seed and fern spore bank of a recovering african tropical forest. Biotropica 46 (6), 677e686. Facelli, J.M., Chesson, P., Barnes, N., 2005. Differences in seed biology of annual plants in arid lands: a key ingredient of the storage effect. Ecology 86, 2998e3006. Ferreira, C.M., Walter, M.T.B., Vieira, L.M.D., 2015. Topsoil translocation for Brazilian Savanna restoration: propagation of herbs, shrubs and trees. Restor. Ecol. 23 (6), 723e728. Golos, J.P., Dixon, W.K., 2014. Waterproofing topsoil stockpiles minimizes viability decline in the soil seed bank in an arid environment. Restor. Ecol. 22 (4), 495e501. Hegazy, A.K., Hammouda, O., Lovett-Doust, J., Gomaa, N.H., 2009. Variations of the germinable soil seed bank along the altitudinal gradient in the northwestern Red Sea region. Acta Ecol. Sin. 29, 20e29. Heydari, M., Pothier, D., Faramarzi1, M., Merzaei1, J., 2014. Short-term abandonment of human disturbances in Zagros Oak forest ecosystems: effects on secondary succession of soil seed bank and aboveground vegetation. Biodiversitas 15 (2), 147e161. Janicka, M., 2006. Species composition of the soil seed bank in comparison with the floristic composition of meadow sward. The Warsaw Agricultural University (SGGW). Grass Science 1, 200e202. Karlík, P., Poschlod, P., 2014. Soil seed-bank composition reveals the land-use history of calcareous grasslands. Acta Oecol. 58, 22e34. Kassahun, A., Snyman, H.A., Smit, G.N., 2009. Soil seed bank evaluation along a degradation gradient in arid rangelands of the Somali region, eastern Etiopia. Agric. Ecosyst. Environ. 129, 428e436. Liu, M., Jiang, G., Yu, S., Li, Y., Li, G., 2009. The role of soil seed banks in natural restoration of the degraded Hunshandak sandlands, northern China. Restor. Ecol. 17, 127e136. Lopes, C.G.R., Ferraz, E.M.N., Castro, C.C., Lima, E.N., Santos, J.M.F.F., Santos, D.M., Araújo, E.L., 2012. Forest succession and distance from preserved patches in the Brazilian semiarid region. For. Ecol. Manag. 271, 115e123. pez, R.P., 2003. Soil seed bank in the semi-arid Prepuna of Bolivia. Plant Ecol. 168, Lo 85e92. Ma, J.Y., Ren, J., Wang, G., Chen, F.H., 2006. Influence of different microhabitats and stand age on viable soil seed banks of sand-stabilising species. S. Afr. J. Bot. 72, 46e50. o, M.R., Pela ez, V.D., Elía, R.O., 2003. Seasonal variation of the soil Mayor, M.D., Bo seed bank of grasses in central Argentina as related to grazing and shrub cover. J. Arid Environ. 53, 467e477. Mendes, B.L., Silva, A.K., Santos, M.D., Santos, F.F.M.J., Albuquerque, P.U., Araújo, L.E., 2015. What happens to the soil seed bank 17 years after clear cutting of vegetations? J. Trop. Biol. 63 (2), 321e332. Ne'eman, G., Izhaki, I., 2009. The effect of stand age and microhabitat on soil seed banks in Mediterranean Aleppo pine forests after fire. Plant Ecol. 144, 115e125. Pereira, I.M., Andrade, L.A., Sampaio, E.V.S.B., Barbosa, M.R.V., 2003. Use-history effects on structure and Flora of Caatinga. Biotropica 35, 154e165. , R., 2000. Seed bank and understorey species composition in a Pugnaire, F.I., Lazaro semi-arid environment: the effect of shrub age and rainfall. Ann. Bot. e Lond. 86, 807e813. z, Y., 2010. Influences of interyear Quevedo-Robledo, L., Pucheta, E., Ribas-Fernande rainfall variability and microhabitat on the germinable seed bank of annual plants in a Sandy Monte Desert. J. Arid. Environ. 74, 167e172. , D., Letourmy, P., Carrie re, M.S., 2015. Effects of slash-andRandriamalala, R.J., Herve burn practices on soil seed banks in secondary forest successions in Madagascar. Agric. Ecosyst. Environ. 199, 312e319. ~es interanuais na Reis, A.M.S., Araújo, E.L., Ferraz, E.M.N., Moura, A.N., 2006. Variaço florística e fitossociologia do componente herb aceo de uma area de Caatinga, Pernambuco, Brasil. Acta Bot. Bras. 29, 497e508. Santos, M.G., Oliveira, M.T., Figueiredo, K.V., Falc~ ao, H.M., Arruda, E.C.P., AlmeidaCortez, J., Sampaio, E.V.S.B., Ometto, J.P.H.B., Menezes, R.S.C., Oliveira, A.F.M.,
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101 Pompelli, M.F., Antonino, A.C.D., 2014. Caatinga, the Brazilian dry tropical forest: can it tolerate climate changes? Theor. Exp. Plant Phys. 26, 83e99. Santos, J.M.F.F., Santos, D.M., Lopes, C.G.R.L., Silva, K.A., Sampaio, E.V.S.B., Araújo, E.L., 2013a. Natural regeneration of the herbaceous community in a semiarid region in Northeastern Brazil. Environ. Monit. Assess. 185 (10), 8287e8302. Santos, D.M., Silva, K.A., Albuquerque, U.P., Santos, J.M.F.F., Lopes, R.G.C., Araújo, E.L., 2013b. Can spatial variation and inter-annual variation in precipitation explain the seed density and species richness of the germinable soil seed bank in a tropical dry forest in north-eastern Brazil? Flora 208, 445e452. Santos, D.M., Silva, K.A., Santos, J.M.F.F., Lopes, C.G.R., Pimentel, R.M.M., Araújo, E.L., rea de floresta 2010. Variaç~ ao espaço-temporal do banco de sementes em uma a tropical seca (Caatinga) e Pernambuco. R. Geogr. 27, 234e253. Selwyn, M.A., Parthasarathy, N., 2006. Reproductive traits and phenology of plants in tropical dry evergreen forest on the Coramandel Coast of India. Biodivers. Conserv. 15, 3207e3234. Silva, K.A., Santos, D.M., Santos, J.M.F.F., Albuquerque, U.P., Ferraz, E.M.N., Araújo, E.L., 2013. Spatio-temporal variation in a seed bank of a semi-arid region in northeastern Brazil. Acta Oecol. 46, 25e32. Souza, J.T., Ferraz, E.M.N., Albuquerque, U.P., Araújo, E.L., 2014. Does proximity to a mature forest contribute to the seed rain and recovery of an abandoned
101
agriculture area in a semiarid climate? Plant Biol. 16, 748e756. tica: guia ilustrado para identificaça ~o Souza, V.C., Lorenzi, H., 2012. Bot^ anica Sistema gamas nativas e exo ticas no Brasil, baseado em APGIII, das famílias de Fanero third ed. Instituto Plantarum, Nova Odessa/SP, p. 768. StatSoft, Inc, 2003. Statistica for Windows (Data Analysis Software System), Version 7.0. StatSoft, Inc, USA. Tessema, Z.K., Boer, W.F., Baars, R.M., Prins, H.H., 2012. Influence of grazing on soil seed banks determines the restoration potential of aboveground vegetation in a semi-arid Savanna of Ethiopia. Biotropica 44, 211e219. Valdez-Hern andez, M., Andrade, J.L., Jackson, P.C., Rebolledo-Vieyra, M., 2010. xico: efPhenology of five trees species of a tropical dry forest in Yucatan, Me fects of environmental and physiological factors. Plant Soil 339, 155e171. Valenta, K., Steffens, S.T., Rafaliarison, R.R., Chapman, A.C., Lehman, M.S., 2015. Seed Banks in Savanna, Forest fragments, and continuous forest edges differ in a tropical dry forest in Madagascar. Biotropica 47 (4), 435e440. Wang, N., Jiao, J., Jia, Y., Zhang, X., 2009. Soil seed bank composition and distribution on eroded slopes in the hill-gully Loess Plateau region (China): influence on natural vegetation colonization. Earth Surf. Process. Landforms 36, 1825e1835. Wassiea, A., Teketay, D., 2006. Soil seed banks in church forests of northern Ethiopia: Implications for the conservation of woody plants. Flora 201, 32e43.
Contents lists available at ScienceDirect
Journal of Arid Environments journal homepage: www.elsevier.com/locate/jaridenv
Composition, species richness, and density of the germinable seed bank over 4 years in young and mature forests in Brazilian semiarid regions ~o Fraga dos Santos a, Danielle Melo dos Santos a, *, Josiene Maria Falca b Kleber Andrade da Silva , Vanessa Kelly Rodrigues de Araújo a, Elcida de Lima Araújo a ^nica, Dois Irma ~os, 52171-900, Recife-PE, Brazil Universidade Federal Rural de Pernambuco, Departamento de biologia, Area Bota ~o-PE, ria, Rua do Alto do Reservato rio s/n, Bela Vista CEP: 55608-680, Vito ria de Santo Anta Universidade Federal de Pernambuco, Centro Acad^ emico de Vito Brazil a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 3 October 2014 Received in revised form 15 February 2016 Accepted 19 February 2016 Available online xxx
The soil seed bank is an important ecological component in forest regeneration. In semiarid regions, forest regeneration is highly affected by seasonal and interannual variations in precipitation, because these variations may affect the composition, species richness, and density of seeds in the soil. This study aimed to characterize and compare these parameters of the germinable seed bank in fragments of young and mature Caatinga forest between various seasons and over consecutive years. A total of 105 soil samples were collected in 20 20 5 cm plots in each forest (Young and Mature) at the end of the rainy and dry seasons over 4 years (2009e2012), totaling 840 samples. The composition, species richness, and density of seeds were determined by the method of seedling emergence. Over the 4 years, 121 species emerged from the soil seed bank, 86 in the young forest, and 109 in the mature forest. Significant differences in the composition, richness, and density were recorded between forests, seasons, and years, with a significant interaction between them. Relationship between rainfall and age of forests affects the dynamics of the soil seed bank in semiarid environments, which are important distinctions for the maintenance of these areas. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Caatinga Seedling emergence Natural regeneration
1. Introduction Severe anthropogenic disturbances, including complete destruction of bushland for the establishment of pasture lands and agriculture, have modified dry and humid environments in the world and endangered climatic characteristics as well as maintenance of ecosystem function and biodiversity conservation, necessitating the comprehension of their influence on natural regeneration, which involves resilience of forests (Santos et al., 2013a; Esaete et al., 2014; Randriamalala et al., 2015; Valenta et al., 2015). Among the processes involved in the regeneration of forests, the dynamic of seed bank plays an important role in carrying the germplasm, which enables the formation of new secondary forests and re-acquirement of areas previously occupied by humans
* Corresponding author. E-mail address: [email protected] (D.M. Santos). http://dx.doi.org/10.1016/j.jaridenv.2016.02.012 0140-1963/© 2016 Elsevier Ltd. All rights reserved.
(Kassahun et al., 2009; Golos and Dixon, 2014). However, type of and historical land use, associated with local climate changes, generates complexity on the dynamic of soil seed bank, considering the variation of the time taken for the recovery of species composition and structure of forests (Heydari et al., 2014; Mendes et al., 2015). Despite the complexity, three trends are reported: (1) With increased level of degradation of such area, a decrease in seed density and species richness of the soil seed bank occurs (Karlík and Pochlod, 2014; Mendes et al., 2015). (2) The new secondary forests consisting of abandoned areas of agriculture and pasture show changes in species composition and population density and low similarity with species composition of soil seed bank (Pereira et al., 2003; Kassahun et al., 2009; Heydari et al., 2014). (3) The recruitment and regeneration of degraded areas highly depend on the dormancy capacity of the seeds stored in the soil bank (Liu et al., 2009; Golos and Dixon, 2014). However, interannual variations in precipitation can increase the complexity and modify the trends reported in the literature,
94
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101
particularly in semiarid regions with well-defined rainy and dry seasons, because they induce differences in the time of production of fruits and seeds (Selwyn and Parthasarathy, 2006; Valdezndez et al., 2010; Souza et al., 2014), thereby affecting Herna dispersal and seed rain. For instance, interannual variations and irregularities in quantitative and rainfall distribution have a significant effect on the reproductive behavior of plants (Albuquerque et al., 2012; Santos et al., 2014) and the intensity of seed rain (Souza et al., 2014) in the vegetation of Brazilian semiarid Caatinga. This explains part of the wealth of species and seed density found in the soil bank of mature forests of this type of vegetation (Silva et al., 2013). Thus, this study characterizes and compares the composition, species richness, and density of the germinable seed bank in young and mature forests of Caatinga between various seasons and over consecutive years. In particular, we investigate whether (1) the species richness and seed density of the soil bank are smaller in the young forest, after 15 years of abandonment of agricultural activity; (2) anthropogenic disturbance and seasonal and interannual variations in rainfall affect the floristic composition of the remaining soil seed bank; and (3) the species richness and seed density of the soil seed bank are lower in dry seasons and driest years than in the rainy seasons and wettest years, in both young and mature forests.
2. Material and methods 2.1. Characterization of the study area The study was conducted at the Agronomic Institute of Pernambuco e IPA (8 140 S and 35 550 W, 537-m altitude), located in the municipality of Caruaru, Pernambuco, Brazil. The site is located in the rural zone at a distance of 9 km from the nearest city. This is a semiarid region, with minimum and maximum temperatures of 11 and 38 C, respectively. The average annual rainfall is 694 mm and the rainy season spans from March to August, with few months having rainfall >100 mm. The dry season spans from September to February, with average monthly rainfall of <30 mm. However, occasional or erratic rainfall may occur in the dry season and dry spells may occur in the rainy season (Araújo et al., 2005a). The range of total rainfall recorded in the study years 2009, 2010, 2011, and 2012 is 350.8e1031.2 mm (Fig. 1). The local seasonality determines the deciduousness of the woody flora in the dry season, and therophytic herbs can only be
observed in the rainy season (Santos et al., 2013a). Moreover, seasonal variations in rainfall influence the rhythm of seed rain in the local vegetation, existing in three species groups: (1) species that disperse seeds only in the rainy season; (2) species that disperse seeds only in the dry season; and (3) species that disperse seeds throughout the year (Souza et al., 2014). The experimental site occupied an area of 190 ha and was created with the primary aim of developing research for agriculture and livestock activities. Before the experiment, the area was occupied by a single patch of mature Caatinga vegetation; however, the current native vegetation is reduced to a small fragment of approximately 30 ha of mature forest. In this area, woody species of the families Mimosaceae, Caesalpiniaceae, Euphorbiaceae, and Cactaceae (Alcoforado-Filho et al., 2003; Araújo et al., 2007) form a closed canopy, which provides a longer period of soil moisture (Santos et al., 2013b). Species of the families Poaceae, Euphorbiaceae, Convolvulaceae, Malvaceae, Asteraceae, and Fabaceae (Araújo et al., 2005a; Reis et al., 2006) predominate in the herbaceous component. This fragment has been preserved for approximately 50 years by not allowing the entry of animals and removal of vegetation (Lopes et al., 2012; Santos et al., 2013a). This fragment is the first area of study, and is named the mature forest. For the last 19 years, a stretch of 3 ha near the mature forest suffered clear-cutting for the cultivation of Opuntia ficus-indica Mill., Cactaceae (commonly called “sweet palm”) (Santos et al., 2013a). No fire, fertilizer, or manure was used during the period of cultivation (Lopes et al., 2012; Souza et al., 2014). After 6 months, this fragment was abandoned and the young forest regenerated naturally. Currently, the herbaceous component of the young forest has rich species of Poaceae and Cyperaceae (Santos et al., 2013a) and the woody component is represented by some young species such as Poincianella pyramidalis (Tul.) L.P. Queiroz (catingueira), Acacia paniculata Willd. (unha de gato), and Anadenanthera macrocarpa (Benth.) Brenan (angico) (Lopes et al., 2012), which do not form a closed canopy, thereby promoting a higher incidence of light in this area (Andrade, unpublished data). This fragment is named the young forest.
2.2. Samples of the soil seed bank For both mature and young forest fragments, there exists a fragment of 1 ha, where studies on the woody and herbaceous
Fig. 1. Monthly precipitation and total precipitation during the rainy and dry seasons over 4 years. Solid arrows indicate the samples collected at the end of the rainy seasons and dashed arrows indicate the samples collected at the end of the dry seasons. Data were provided by the meteorological station of the Empresa Pernambucana de Pesquisa Agropecu aria (IPA) in Caruaru, Pernambuco, Brazil.
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101
components and seed rain are being undertaken (Araújo et al., 2005b; Souza et al., 2014). Therefore, in this fragment, 105 plots of dimension 1 1 m were randomly allocated for the study of herbaceous vegetation (Reis et al., 2006; Santos et al., 2013a), totaling 210 plots. In addition, in the area surrounding these plots (both forests), 105 soil samples were collected at the end of both rainy and dry seasons for 4 consecutive years (to enable comparison of the temporal effect between the areas), totaling 840 samples in each forest (Fig. 2). Soil from a depth of 5 cm, including the leaf litter layer, was collected in plots made of a stainless steel sheet of dimension 20 20 cm. This procedure was used in most studies on soil seed banks (Ma et al., 2006; Hegazy et al., 2009; Ne'eman and Izhaki, 2009; Quevedo-Robledo et al., 2010). All samples were packed in plastic bags and labeled by the respective plot and forest, and, in the greenhouse, each sample was placed in a polystyrene tray (20 38 3 cm) and irrigated daily for the seasonal duration of 6 months, enabling the comparison of the effect of seasonality between collections in young and mature forests. The trays were arranged in two rows with a control tray between them, containing sterilized soil, to detect possible contamination caused by seeds dispersed by wind. Contamination was not detected during the study period. Seedling emergence was the method used to determine the seed density in the soil seed bank (not considering the density of dormant seeds in the sample), following the methodology adopted by Silva et al. (2013). In this method, each seedling that emerged from soil represents one seed from the soil seed bank. Data were expressed in seeds per square meter to allow comparison with other studies (Ne'eman and Izhaki, 2009; Hegazy et al., 2009). Emerging seedlings of each soil sample were counted and labeled every day, noting the date of germination, plot number, and the forest in which the sample was collected. Once the size of seedlings reached approximately 5 cm, they were transplanted into polythene bags, irrigated daily, kept at an average temperature of 25 C, and monitored for 6 months in
Fig. 2. Schematic view of the collection of the soil seed bank (20 20 5 cm plots) surrounding the 1 1 m fixed plots in a semiarid region in the northeast of Brazil.
95
greenhouse to obtain reproductive material of therophytic herbs for correct taxonomic identification of the species. Species identification was performed through consultations in the literature and comparisons with herbarium specimens deposited in the herbaria Prof. Vasconcelos Sobrinho (PEUFR) and rdano de Andrade Lima (IPA), adopting the APG3 classification Da system (Souza and Lorenzi, 2012). Unidentified seedlings were listed as morphospecies. 2.3. Analyses of the seed bank The floristic compositions of both forests (young and mature) were compared between years by a nonmetric multidimensional scaling (NMDS) analysis using the BrayeCurtis dissimilarity matrix, based on the relative density of the species in the 105 sample units in each study area. The analysis of similarities (ANOSIM) was used to test the significance of the group formed in the NMDS. In order to verify the contribution of each species between seasons and between forests, a similarity percentage analysis (SIMPER) was performed. Software Primer version 6.1.6 (Clarke and Gorley, 2006) was used for ANOSIM, NMDS and SIMPER analyses. In order to verify the effect of predictor variables (age of the forest e young or mature; seasonal or interannual variation in precipitation) on species richness and seed density, a generalized linear model (GLM) analysis was performed. Differences in species richness and seed density between young and mature forests, rainy and dry seasons, and years were verified by the Tukey a posteriori test. All these analyses were carried out using the software Statistic 7.0 (StatSoft Inc., 2003). 3. Results 3.1. Floristic composition The floristic composition differs between the young and mature forests (R-global ¼ 0.898, p ¼ 0.001, Fig. 3). In the former, a seasonal variation in floristic composition was present (R-global ¼ 0.274, p ¼ 0.001, Fig. 4a), which was found to be similar between years (Rglobal ¼ 0.002, p ¼ 0.555). Seasonal (R-global ¼ 0.126, p ¼ 0.001, Fig. 4b) and interannual (R-global ¼ 0.018, p ¼ 0.031, Fig. 5) variations in floristic composition were present in the mature forest, where the assemblage of species recorded in the fourth year was
Fig. 3. Ordination formed after multidimensional scaling (MDS) analysis on the germinated seeds of the germinable soil seed bank in young and mature forests during 4 years of study, based on species richness per area. The symbols on the graph represent the samples of soil and their respective germinated species.
96
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101
seasons and between the mature and young forests. The five species that contributed most to the differences between rainy and dry seasons, in descending order, were Delilia biflora, Pilea hyalina, Panicum trichoides, Heliotropium angiospermum, and Panicum venezuelae (Table 1). The cumulative percentage contribution of these species in the differences between seasons was 40.05%. The analysis of dissimilarity between the young and mature forests showed the aforementioned five species as the main species, which contribute to the differences between the areas, accumulating 43.15% of the contribution (Table 1).
3.2. Species richness Over the 4 years, a total of 121 species were recorded in the soil seed bank, 86 species from the young forest, and 109 from the mature forest (Appendix 1). Of the total number of species, 16 were identified to family level, 17 to genus level, 84 to species level, and only five as morphospecies. A total of 13 and 36 species occurred exclusively in the young and mature forests, respectively (Appendix 1).
Table 1 SIMPER analysis between the rainy and dry seasons, between the mature and young forests, and between the 4 years of study with the contribution of each species dissimilar between sampling sites (Av.Diss.).
Fig. 4. Ordination formed after multidimensional scaling (MDS) analysis of the germinated seeds of the germinable soil seed bank between seasons (rainy and dry) in areas of young (A) and mature (B) forests over 4 consecutive years. This graph was produced based on species richness per area. The symbols on the graph represent the soil samples and their respective germinated species only considering the climatic seasons.
Fig. 5. Ordination formed after multidimensional scaling (MDS) analysis of the germinated seeds of the germinable soil seed bank over 4 consecutive years in mature forests. This graph was produced based on species richness per area. The symbols on the graph represent the soil samples and their respective germinated species, only considering the 4 years of study.
different from that recorded in the first 3 years (Fig. 5). The SIMPER analysis showed a high dissimilarity between
Dry and rainy seasons
Young and mature forests
Av. Diss. ¼ 86.29
Av. Diss. ¼ 92.18
Species
Av. Diss
Species
Av. Diss
Delília biflora Pilea Hyalina Panicum trichoides Heliotropium angiospermum Panicum venezuelae Poaceae 1 Phaseolus pedunculares Ruellia sp1 Begonia reniformis Tallinum triangulare Poaceae 2 Desmodium glabrum Gomphrena vaga Malvaceae 1 Bidens bipinnata Callisia repens Gnaphalium spicatum Ruellia asperulla Dioscorea coronata Ruellia bahiensis Pappophorum papipherum Hippeastrum sp1 Croton blanchetianus Cyperus uncinulatus Mimosa arenosa Croton heliotropiifolius Urochloa maxima Mimosaceae 1 Cactaceae 1 cie 2 Morfoespe cie 1 Morfoespe Chamaesyce hyssopifolia Blainvillea acmella Solanum americanum Myracrodruon urundeuva Vigna sp1 Conocliniopsis prasiifolia Dalechampia scandens Mollugo verticillata Phyllantus niruri Commelina obliqua Lippia americana
10.24 9.77 6.97 4.08 3.49 2.63 2.63 2.50 2.43 2.12 2.06 1.95 1.94 1.77 1.53 1.39 1.27 1.24 1.22 1.18 1.14 1.09 1.07 1.01 1.01 0.98 0.96 0.89 0.87 0.70 0.57 0.52 0.47 0.47 0.47 0.45 0.45 0.43 0.40 0.37 0.36 0.35
Pilea Hyalina Delília biflora Panicum trichoides Heliotropium angiospermum Panicum venezuelae Poaceae 1 Phaseolus pedunculares Begonia reniformis Ruellia sp1 Poaceae 2 Tallinum triangulare Desmodium glabrum Gomphrena vaga Malvaceae 1 Bidens bipinnata Callisia repens Gnaphalium spicatum Ruellia bahiensis Ruellia asperulla Pappophorum papipherum Dioscorea coronata Hippeastrum sp1 Croton blanchetianus Cyperus uncinulatus Mimosa arenosa Croton heliotropiifolius Urochloa maxima Mimosaceae 1 Cactaceae 1 cie 2 Morfoespe Chamaesyce hyssopifolia Blainvillea acmella cie 1 Morfoespe Solanum americanum Vigna sp1 Myracrodruon urundeuva Dalechampia scandens Conocliniopsis prasiifolia Mollugo verticillata Commelina obliqua Selaginella sulcata e
12.47 11.81 7.58 4.43 3.49 2.69 2.56 2.56 2.48 2.18 2.07 2.05 1.94 1.87 1.59 1.37 1.30 1.28 1.24 1.24 1.22 1.13 1.08 1.04 1.02 1.00 0.98 0.88 0.84 0.73 0.56 0.52 0.52 0.48 0.46 0.44 0.44 0.41 0.41 0.36 0.35 e
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101
97
Of the exclusive species in each forest, only four species occurred in the two seasons (rainy and dry) over the 4 years (Lippia americana L. in the young forest; Begonia reniformis Vell.; Cactaceae sp1; and Schinopsis brasiliensis Engl. in the mature forest). Of the total species found in both forests, 100 were found to be herbaceous, 11 were shrubs, and 10 were trees (Appendix 1). In both forests, families with higher species richness over the 4 years of study were Euphorbiaceae (14), Fabaceae (14), Poaceae (10), Asteraceae (11), Portulacaceae (five), and Malvaceae (six) (Appendix 1). Furthermore, we found that 28 species occurred in the two seasons (rainy and dry) over the 4 years of study (Appendix 1). Of the 28 species, only D. biflora (L.) Kuntze, P. trichoides Sw, and P. hyalina Fenzl. occurred in both forests. The GLM analysis showed that the age of the forest, seasonal and interannual variations in precipitation, and most of their interactions explain the variations in species richness (Table 2). Over the 4 years, in both the young and mature forests, species richness was higher in the dry season (Fig. 6). Comparing the study areas, species richness was higher in the mature forest than the young forest, only in the rainy season (Fig. 6). In the mature forest, species richness was higher in the second and third years (Fig. 7). In the young forest, species richness was highest in the first year and decreased significantly over time (Fig. 7). Considering each year separately, in the mature forest, the species richness was higher in the dry seasons of the second and fourth years (Fig. 8). In the young forest, species richness was higher in the dry seasons of the first and fourth years (Fig. 8). 3.3. Seed density During the 4 years of study, a total of 4097 seeds m2 were recorded in the soil seed bank, with 1594 and 2503 seeds m2 from the young and mature forests, respectively. Separate analysis of each season showed that, in the rainy and dry seasons, the respective total numbers of seeds recorded per square meter were 460 and 1133 for the young forest and 854 and 1649 for the mature forest. The GLM analysis showed that the age of the forest, seasonal and interannual variations in precipitation, and most of their interactions explained the variations in mean seed density (Table 2). Considering the 4 years, the mean density was higher in the mature forest. In both forests, the mean density was higher in the dry season (Fig. 6). In the mature forest, the mean density was higher in the second and third years. In the young forest, the mean density was higher in the first year and decreased significantly in the last year (Fig. 7). Considering each year separately, the mean density was higher in the dry seasons of the second and fourth years in the mature
Fig. 6. Spatiotemporal variation in mean richness (number of species/20 20 cm plot) and mean seed density (seeds/20 20 cm plot) in areas of young and mature forests over 4 years in a semiarid region in the northeast of Brazil. Different letters between seasons (rainy and dry) in each area (mature and young) and between areas in each season show a significant difference in the Tukey test. Vertical bars denote a confidence interval of 0.95.
forest, and first and fourth years in the young forest (Fig. 8). In the second and third years, the mean density was higher in the mature forest (Fig. 8).
Table 2 GLM analysis showing the influence of the age of the forest (young and mature), seasonal and annual variation in precipitation, and their interactions on the density and species richness of the germinable seed bank in a semiarid region in the northeast of Brazil. p values denote statistical significance (GL ¼ degrees of freedom; SQ ¼ sum of squares; QM ¼ mean square, F ¼ Fisher test). Variables
Interceptor Age Year Season Age*Year Age*Season Year*Season Age*Year*Season Error Total
Density
Richness
GL
SQ
QM
F
P
GL
SQ
QM
F
P
1 1 3 1 3 1 3 3 1664 1679
176607.5 8745.2 8525.4 22755.6 14297.0 162.8 3006.5 8421.8 591339.2 657253.5
176607.5 8745.2 2841.8 22755.6 4765.7 162.8 1002.2 2807.3 355.4
496.9650 24.6085 7.9967 64.0333 13.4104 0.4582 2.8200 7.8995
0.000000 0.000001 0.000027 0.000000 0.000000 0.498582 0.037721 0.000031
1 3 1 1 3 3 1 3 1664 1679
16783.39 621.34 1160.01 85.95 499.21 679.35 71.26 427.75 7415.73 10960.61
16783.39 207.11 1160.01 85.95 499.21 679.35 71.26 142.58 4.46
3.765.988 46.473 260.292 19.287 37.339 50.812 15.990 31.994
0.000000 0.000000 0.000000 0.000012 0.000000 0.000000 0.000066 0.000000
98
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101
Fig. 7. Annual change in mean richness (number of species/20 20 cm plot) and mean seed density (seeds/20 20 cm plot) in areas of young and mature forests over 4 years in a semiarid region in the northeast of Brazil. Uppercase letters between different years in each area and lowercase letters between different areas each year denote significant differences in the Tukey test. Vertical bars denote a confidence interval of 0.95.
4. Discussion 4.1. Effect of age of the forest on the soil seed bank In this study, we show that, on semiarid regions, species richness is not always lower in new young forests of the anthropogenic areas. In the dry seasons, wealth of young forests was found to be similar to that of the mature ones. Therefore, a part of first hypothesis was refused, although the literature reported a higher speed of resilience process in the dry forests than the wet forests (Lopes et al., 2012). This finding shows that studies evaluating the effect of anthropic actions on the species richness of the seed bank need to consider larger time intervals (4 years) for including temporal stochasticity of species richness of the soil seed bank of young and mature forests. Anthropogenic disturbances affect the number of native seeds reaching the soil bank and consequently the regeneration of degraded areas (Ferreira et al., 2015). Hence, some studies report that the seed density of the soil bank of disturbed areas is lower than that of the original forest (Tessema et al., 2012; Heydari et al., 2014; Karlík and Pochlod, 2014). This trend was also found in this study, endorsing the other part of the first hypothesis. In addition, we showed that anthropogenic disturbances cause a drastic change
in floristic composition, confirming part of our second hypothesis. Our results show that 90% of the difference in floristic composition was explained by forest age. Consequently, 15 years are not enough for the seed bank to recover the floristic composition and seed density of the original forest. Some studies indicate that a high number of seeds, weeds, or invasive plants occurs in areas that suffered disturbance. This is because (1) degradation promotes disappearance of native species with low colonization ability (Augusto et al., 2001; Ferreira et al., 2015) or (2) woody species of communities existing before the anthropogenic practices are not present on the seed bank left in the soil (Wang et al., 2009). Invasive or weed species can compete with native species, affecting the natural regeneration (Ferreira et al., 2015). However, despite the occurrence of drastic change in the set of colonizing species after disturbance, we believe that no risk occurred in the natural regeneration of the young forest in our study, because its seed bank was not composed of weed species. Most species found in the soil seed bank were annual herbaceous plants, and all of them are native species of the two forests (Santos et al., 2013b). Herbaceous species play an important role in the functioning of the ecosystem and correspond to most of the phytodiversity of dry forests with a significant influence on the dynamics of woody plants (Wassiea and Teketay, 2006; Wang et al., 2009; Santos et al., 2010). They also protect the seedlings of woody species against the impact of rain and assist in the retention of their seeds on the soil surface (Santos et al., 2013b). Furthermore, a specific assemblage of herbaceous species contributed to the differences in floristic composition between seasons and forests. Thus, this group of species has probably the highest potential for regeneration with a diverse soil seed bank exhibiting a higher longevity (Wassiea and Teketay, 2006; Wang et al., 2009; Golos and Dixon, 2014). Therefore, this study shows that perhaps this group of native and exclusive species of the disturbed forests can help recuperate the areas, as that occurred in our young forest, which suffered clear-cutting for the cultivation of sweet palm (O. ficus-indica Mill.). There was only one adult woody specie in the young forest that has enabled a higher light incidence and increased drying of the soil (Lopes et al., 2012; Santos et al., 2013a). On the contrary, the mature forest had woody plants reaching an average height of 5e7 m, with relatively overlapping crowns, providing more shading on the ground, which remained humid for more number of months (Santos et al., 2013b). Thus, these characteristics might have aided in the establishment of species, which requires more shade, promoting a higher seed density in the mature forest and a different floristic composition compared with the young forest. It is important to note that the floristic composition, richness, and density of seed bank can significantly increase during regeneration time. The longer the duration of regeneration of new forest, the higher the similarity between the floristic composition, species richness, and amount of seeds found in the bank of this soil in relation to the original forest (Chaideftou et al., 2009; Valenta et al., 2015). Thus, further studies on the seed bank of the Caatinga vegetation, with different ages and usage history, are necessary to find the time required to complete resilience of its forests. 4.2. Effect of temporal variation in precipitation on the soil seed bank In this study, seasonal variation in precipitation changed the floristic composition, species richness, and density of the soil seed bank, regardless of the age of the forest. This fact follows the pattern of dry forests, because the composition, richness, and seed density change faster with seasonal variation in these
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101
99
Fig. 8. Seasonal variation in mean richness (number of species/20 20 cm plot) and mean seed density (seeds/20 20 cm plot) in areas of young and mature forests over 4 years in a semiarid region in the northeast of Brazil. Different letters between seasons (rainy and dry) in each area show a significant difference in the Tukey test. Vertical bars denote a confidence interval of 0.95.
, 2000; Santos et al., 2010; environments (Pugnaire and Lazaro Mendes et al., 2015). Most studies on seed banks emphasize that the number of species and seeds is higher during the rainy season and the floristic pez, 2003; Mayor et al., composition is a function of this season (Lo 2003; Santos et al., 2010; Mendes et al., 2015). However, it is important to note that the trend for higher species richness and density in the rainy season was not recorded in this study. Thus, part of our third hypothesis was not accepted. The assemblage of species recorded only in the dry season was responsible for the changes in floristic composition. In addition, we also believe that the highest species richness and seed density recorded in this study in the dry season may be justified for the following reasons: (1) the sum of the amount of seeds dispersed during the rainy and dry seasons and (2) some
seeds dispersed in the rainy season need to spend the entire dry season in the soil to complete their maturation process. These reasons can also explain the difference in floristic composition between seasons. Therefore, further studies are necessary to test this hypothesis. With regard to interannual variations, the literature indicates that richness and composition of species, as well as the amount of seeds is directly proportional to the intensity of the annual rainfall pez, 2003; Facelli et al., in dry forests (Costa and Araújo, 2003; Lo 2005; Santos et al., 2010; Silva et al., 2013; Souza et al., 2014). In this study, this trend was recorded for species richness and seed density of the two forests, confirming another part of the third hypothesis. In the mature forest, species richness and seed density were higher in the second and third years, which have had the highest average annual rainfall (865 and 1031.2 mm, respectively).
100
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101
For that reason, in the young forest, species richness and seed density decreased significantly in the fourth year, which showed the lowest annual rainfall (350.8 mm), compared with other years. Interannual variations in seed bank attributes can also be explained by the previous year's precipitation. In the same area where this study was conducted, the previous year's precipitation explained 21% of the variation in species richness and 16% of the variation in seed density in the mature forest (Silva et al., 2013). Therefore, we also believe that the interannual variation in species richness and density of the seed bank recorded in this study in young and mature forests can be explained by the previous year's precipitation. Nevertheless, interannual variations in floristic composition differed between the forests. No interannual variation in the floristic composition of the young forest was present. Therefore, the assemblage of species that make up this area is possibly not highly influenced by interannual variation in precipitation, not confirming part of the second hypothesis. Moreover, in the mature forest, differences in the floristic composition occurred over the years. It is noteworthy that this difference was very small, because the Rglobal (0.018) was very low. Interannual changes in the floristic composition of the seed bank were also observed in other studies (Janicka, 2006; Chaideftou et al., 2009; Santos et al., 2010, 2013b; Silva et al., 2013). In the mature forest, the assemblage of species found in the fourth year differed from that in other years. This may be due to the sharp reduction in the annual average precipitation from the third (1031.2 mm) to the fourth year (350.8 mm). Consequently, drastic reductions in annual rainfall may reduce seed production of the local community and, consequently, cause the disappearance of some species from the soil seed bank. This fact can be confirmed by comparing the amount of unique species in the third year (11) and the fourth year (five). In conclusion, we found that (1) Human activities cause changes in floristic composition, species richness, and density of the soil seed bank; (2) Temporal variation in precipitation caused changes in species richness and density of the soil seed bank, regardless of the age of the forest; and (3) The effect of variation of interannual precipitation on the floristic composition of young forest is not clear in semiarid environments, after disturbance. Acknowledgments The authors thank the Agronomic Institute of Agricultural Research (IPA) for logistics and permission to work on their property; researchers from the Laboratory of Plant Ecology and Northeastern Ecosystems (LEVEN) for their support, suggestions, and assistance in the execution of the project; and CAPES and CNPq (302645/2014-4) for the award of the student scholarship and productivity scholarship for researchers, as well as financial support of the project. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jaridenv.2016.02.012. References Albuquerque, U.P., Araújo, E.L., El-Deir, A.C.A., Lima, A.L.A., Souto, A., Bezerra, B.M., Ferraz, E.M.N., Freire, E.M.X., Sampaio, E.V.S.B., Las-Casas, F.M.G., Moura, G.J.B., Pereira, G.A., Melo, J.G., Ramos, M.A., Rodal, M.J.N., Schiel, N., Lyra-Neves, R.M., Alves, R.R.N., Azevedo-Junior, S.M., Telino-Junior, W.R., Severi, W., 2012. Caatinga revisited: ecology and conservation of an important seasonal dry forest. Sci. World J. 1e18. Alcoforado-Filho, F.G., Sampaio, E.V.S.B., Rodal, M.J.N., 2003. Florística
lia espinhosa arbo rea fitossociologia de um remanescente de vegetaç~ ao caducifo em Caruaru. Acta Bot. Bras. 17, 287e303. Araújo, E.L., Silva, K.A., Ferraz, E.M.N., Sampaio, E.V.S.B., Silva, S.I., 2005a. Diversirea de dade de herb aceas em microhabitats rochoso, plano e ciliar em uma a Caatinga, Caruaru-PE. Acta Bot. Bras. 19, 285e294. Araújo, E.L., Martins, F.R., Santos, A.M., 2005b. Establishment and death of two dry tropical forest woody species in dry and rainy seasons in northeastern Brazil. In: Nogueira, R.J.M.C., Araújo, E.L., Willadino, L.G., Cavalcante, U.M.T. (Eds.), fica e Editora, Estresses ambientais: danos e benefícios em plantas. MXM Gra Recife, pp. 76e91. Araújo, E.L., Castro, C.C., Albuquerque, U.P., 2007. Dynamics of brazilian Caatinga e a review concerning the plants, environment and people. Funct. Ecol. Comm. 1, 15e28. Augusto, L., Dupouey, J.L., Picard, J.F., Ranger, J., 2001. Potential contribution of the seed bank in coniferous plantations to the restoration of native deciduous forest vegetation. Acta Oecol. 22, 87e98. Chaideftou, E., Thanos, C.A., Bergneier, E., Kallimanis, A., Dimopoulos, P., 2009. Seed bank composition and above-ground vegetation in response to grazing in submediterranean oak forests (NW Greece). Plant Ecol. 201, 255e265. Clarke, K.R., Gorley, R.N., 2006. Primer v6: User Manual/tutorial. Plymouth Marine Laboratory, Plymouth. ~o e flora do banco de sementes Costa, R.C., Araújo, F.S., 2003. Densidade, germinaça do solo no final da estaç~ ao seca, em uma area de Caatinga, Quixad a, CE. Acta Bot. Bras. 17, 259e264. Esaete, J., Eycott, E.A., Reini, J., Telford, J.R., Vandvik, V., 2014. The seed and fern spore bank of a recovering african tropical forest. Biotropica 46 (6), 677e686. Facelli, J.M., Chesson, P., Barnes, N., 2005. Differences in seed biology of annual plants in arid lands: a key ingredient of the storage effect. Ecology 86, 2998e3006. Ferreira, C.M., Walter, M.T.B., Vieira, L.M.D., 2015. Topsoil translocation for Brazilian Savanna restoration: propagation of herbs, shrubs and trees. Restor. Ecol. 23 (6), 723e728. Golos, J.P., Dixon, W.K., 2014. Waterproofing topsoil stockpiles minimizes viability decline in the soil seed bank in an arid environment. Restor. Ecol. 22 (4), 495e501. Hegazy, A.K., Hammouda, O., Lovett-Doust, J., Gomaa, N.H., 2009. Variations of the germinable soil seed bank along the altitudinal gradient in the northwestern Red Sea region. Acta Ecol. Sin. 29, 20e29. Heydari, M., Pothier, D., Faramarzi1, M., Merzaei1, J., 2014. Short-term abandonment of human disturbances in Zagros Oak forest ecosystems: effects on secondary succession of soil seed bank and aboveground vegetation. Biodiversitas 15 (2), 147e161. Janicka, M., 2006. Species composition of the soil seed bank in comparison with the floristic composition of meadow sward. The Warsaw Agricultural University (SGGW). Grass Science 1, 200e202. Karlík, P., Poschlod, P., 2014. Soil seed-bank composition reveals the land-use history of calcareous grasslands. Acta Oecol. 58, 22e34. Kassahun, A., Snyman, H.A., Smit, G.N., 2009. Soil seed bank evaluation along a degradation gradient in arid rangelands of the Somali region, eastern Etiopia. Agric. Ecosyst. Environ. 129, 428e436. Liu, M., Jiang, G., Yu, S., Li, Y., Li, G., 2009. The role of soil seed banks in natural restoration of the degraded Hunshandak sandlands, northern China. Restor. Ecol. 17, 127e136. Lopes, C.G.R., Ferraz, E.M.N., Castro, C.C., Lima, E.N., Santos, J.M.F.F., Santos, D.M., Araújo, E.L., 2012. Forest succession and distance from preserved patches in the Brazilian semiarid region. For. Ecol. Manag. 271, 115e123. pez, R.P., 2003. Soil seed bank in the semi-arid Prepuna of Bolivia. Plant Ecol. 168, Lo 85e92. Ma, J.Y., Ren, J., Wang, G., Chen, F.H., 2006. Influence of different microhabitats and stand age on viable soil seed banks of sand-stabilising species. S. Afr. J. Bot. 72, 46e50. o, M.R., Pela ez, V.D., Elía, R.O., 2003. Seasonal variation of the soil Mayor, M.D., Bo seed bank of grasses in central Argentina as related to grazing and shrub cover. J. Arid Environ. 53, 467e477. Mendes, B.L., Silva, A.K., Santos, M.D., Santos, F.F.M.J., Albuquerque, P.U., Araújo, L.E., 2015. What happens to the soil seed bank 17 years after clear cutting of vegetations? J. Trop. Biol. 63 (2), 321e332. Ne'eman, G., Izhaki, I., 2009. The effect of stand age and microhabitat on soil seed banks in Mediterranean Aleppo pine forests after fire. Plant Ecol. 144, 115e125. Pereira, I.M., Andrade, L.A., Sampaio, E.V.S.B., Barbosa, M.R.V., 2003. Use-history effects on structure and Flora of Caatinga. Biotropica 35, 154e165. , R., 2000. Seed bank and understorey species composition in a Pugnaire, F.I., Lazaro semi-arid environment: the effect of shrub age and rainfall. Ann. Bot. e Lond. 86, 807e813. z, Y., 2010. Influences of interyear Quevedo-Robledo, L., Pucheta, E., Ribas-Fernande rainfall variability and microhabitat on the germinable seed bank of annual plants in a Sandy Monte Desert. J. Arid. Environ. 74, 167e172. , D., Letourmy, P., Carrie re, M.S., 2015. Effects of slash-andRandriamalala, R.J., Herve burn practices on soil seed banks in secondary forest successions in Madagascar. Agric. Ecosyst. Environ. 199, 312e319. ~es interanuais na Reis, A.M.S., Araújo, E.L., Ferraz, E.M.N., Moura, A.N., 2006. Variaço florística e fitossociologia do componente herb aceo de uma area de Caatinga, Pernambuco, Brasil. Acta Bot. Bras. 29, 497e508. Santos, M.G., Oliveira, M.T., Figueiredo, K.V., Falc~ ao, H.M., Arruda, E.C.P., AlmeidaCortez, J., Sampaio, E.V.S.B., Ometto, J.P.H.B., Menezes, R.S.C., Oliveira, A.F.M.,
D.M. Santos et al. / Journal of Arid Environments 129 (2016) 93e101 Pompelli, M.F., Antonino, A.C.D., 2014. Caatinga, the Brazilian dry tropical forest: can it tolerate climate changes? Theor. Exp. Plant Phys. 26, 83e99. Santos, J.M.F.F., Santos, D.M., Lopes, C.G.R.L., Silva, K.A., Sampaio, E.V.S.B., Araújo, E.L., 2013a. Natural regeneration of the herbaceous community in a semiarid region in Northeastern Brazil. Environ. Monit. Assess. 185 (10), 8287e8302. Santos, D.M., Silva, K.A., Albuquerque, U.P., Santos, J.M.F.F., Lopes, R.G.C., Araújo, E.L., 2013b. Can spatial variation and inter-annual variation in precipitation explain the seed density and species richness of the germinable soil seed bank in a tropical dry forest in north-eastern Brazil? Flora 208, 445e452. Santos, D.M., Silva, K.A., Santos, J.M.F.F., Lopes, C.G.R., Pimentel, R.M.M., Araújo, E.L., rea de floresta 2010. Variaç~ ao espaço-temporal do banco de sementes em uma a tropical seca (Caatinga) e Pernambuco. R. Geogr. 27, 234e253. Selwyn, M.A., Parthasarathy, N., 2006. Reproductive traits and phenology of plants in tropical dry evergreen forest on the Coramandel Coast of India. Biodivers. Conserv. 15, 3207e3234. Silva, K.A., Santos, D.M., Santos, J.M.F.F., Albuquerque, U.P., Ferraz, E.M.N., Araújo, E.L., 2013. Spatio-temporal variation in a seed bank of a semi-arid region in northeastern Brazil. Acta Oecol. 46, 25e32. Souza, J.T., Ferraz, E.M.N., Albuquerque, U.P., Araújo, E.L., 2014. Does proximity to a mature forest contribute to the seed rain and recovery of an abandoned
101
agriculture area in a semiarid climate? Plant Biol. 16, 748e756. tica: guia ilustrado para identificaça ~o Souza, V.C., Lorenzi, H., 2012. Bot^ anica Sistema gamas nativas e exo ticas no Brasil, baseado em APGIII, das famílias de Fanero third ed. Instituto Plantarum, Nova Odessa/SP, p. 768. StatSoft, Inc, 2003. Statistica for Windows (Data Analysis Software System), Version 7.0. StatSoft, Inc, USA. Tessema, Z.K., Boer, W.F., Baars, R.M., Prins, H.H., 2012. Influence of grazing on soil seed banks determines the restoration potential of aboveground vegetation in a semi-arid Savanna of Ethiopia. Biotropica 44, 211e219. Valdez-Hern andez, M., Andrade, J.L., Jackson, P.C., Rebolledo-Vieyra, M., 2010. xico: efPhenology of five trees species of a tropical dry forest in Yucatan, Me fects of environmental and physiological factors. Plant Soil 339, 155e171. Valenta, K., Steffens, S.T., Rafaliarison, R.R., Chapman, A.C., Lehman, M.S., 2015. Seed Banks in Savanna, Forest fragments, and continuous forest edges differ in a tropical dry forest in Madagascar. Biotropica 47 (4), 435e440. Wang, N., Jiao, J., Jia, Y., Zhang, X., 2009. Soil seed bank composition and distribution on eroded slopes in the hill-gully Loess Plateau region (China): influence on natural vegetation colonization. Earth Surf. Process. Landforms 36, 1825e1835. Wassiea, A., Teketay, D., 2006. Soil seed banks in church forests of northern Ethiopia: Implications for the conservation of woody plants. Flora 201, 32e43.