Our aim was to assess the seasonal abundance and diversity of arthropods on Acacia mangium Willd. (Fabales: Fabaceae) trees. Phytophagous arthropods, natural enemies and pollinators were quantified biweekly in 20 trees during 3 years. The Shannon biodiversity index (H') of arthropods was greater in the summer and smaller in the winter and spring on A. magium. The diversities of species and abundances of individuals of all arthropod taxa were greater in the autumn and smaller in the winter and in the spring. The H' index of arthropods was greater in 2005 and smaller in 2007. The diversity of species and abundance of individuals of phytophagous taxa were greater in 2005 and 2006, respectively, whereas those of natural enemies and pollinators was greater in 2006. Aethalion reticulatum L. (Hemiptera: Aetalionidae) was the most abundant phytophagous species found in the various seasons, while Camponotus sp.2 (Hymenoptera: Formicidae), Trigona spinipes Fabricius (Hymenoptera: Apidae, Meliponini), Tetragonisca angustula Latreille (Hymenoptera: Meliponinae, Meliponini) and Polistes sp. (Hymenoptera: Vespidae) were the most abundant natural enemies and pollinators. The great diversity of predators on this tree species can explain the little damage to its leaves. These results may be applied to support programs of pest control and maintenance of natural enemies and pollinators in future plantations of A. mangium.
Distributions of arthropod species may differ in different years in response to changes in climate and food availability, which may directly impact their abundances and diversity (Wolda 1988; Chilima & Leather 2001; Leite et al. 2005). In tropical areas with well-defined dry and wet seasons, the population dynamics of many insect species is influenced by climatic changes, which probably cause changes in the physiology of the host plants (Hamer et al. 2005; Leite et al. 2006a).
The abundance and diversity of arthropods may be greater in some developmental stage of the host plant. The nutritional quality and chemical defenses of plants, which are linked to the age of leaves, are important for phytophagous insects (Bowers & Stamp 1993; Coley & Barone 1996; Leite et al. 2006b). Acacia mangium Willd. (Fabales: Fabaceae) is a pioneer species that can grow on acid and infertile soils due to its capacity of N-fixing (Galiana et al. 1998), with this nutrient being one of the most important for the tree growth (Lea & Azevedo 2006).
Interactions between plants of Acacia spp. and insects have been studied (Fleming et al. 2007; McLeish et al. 2007; Palmer et al. 2007; Silva et al. 2014), and insects are the main organisms responsible for the decline of Arabic gum production from these trees in Sudan (Jamal 1994). The main herbivores found damaging A. mangium were leafcutter ants (Atta spp. and Acromyrmex spp.), termites and Costalimaita ferruginea Fabr. (Coleoptera: Chrysomelidae) (Arco-Verde 2002) in the northern Brazil, and Trigona spinipes Fabr. (Hymenoptera: Apidae), Aethalion reticulatum (Hemiptera: Aetalionidae) and Pentatomidae sp. 1 (Hemiptera) in southeastern Brazil (Silva et al. 2014).
Acacia spp. are broadly used in the initial restoration process of degraded lands (Tsai 1988; Yu & Li 2007), in places where initially a number of barriers prevent plant development (Chada et al. 2004). Moreover, they are also used as windbreaks to prevent wind and water erosion (Michels et al. 1998; Bird et al. 2002) and as barriers to movements of arthropods (Rao et al. 2000).
The aim of this study was to assess the seasonal abundance and diversity of arthropods during 3 years of sampling on A. mangium trees.
Materials and Methods
STUDY SITES
This study was carried out in a pasture area of the Institute of Agricultural Sciences at the Universidade Federal de Minas Gerais (ICA/ UFMG), Brazil. Samplings occurred from Jan 2005 to Mar 2007 in an area with an Aw climate, i.e., tropical savanna according to the classification of Köppen with a dry winter and a rainy summer and a dystrophic red-yellow latosol.
STUDY DESIGN
Windbreaks, 100 m long with 2 rows of A. mangium spaced 3 × 3 m were used. Saplings were prepared in a nursery and planted in Sep 2003 in 30 × 30 × 30 cm holes with 360 grams of natural reactive phosphate mixed into the subsoil of a Brachiaria decumbens Stapf. (Poales: Poaceae) pasture. The phytophagous, natural enemies and pollinators arthropods in twenty 16-month old A. mangium trees were visually counted biweekly every yr. The arthropods were counted on the adaxial and abaxial leaf surfaces in the upper, median and smaller apical canopy on branches facing north, south, east and west; and with a total of 12 leaves per canopy and 9 per tree branch position in each sampling. Arthropod collection also occurred on the trunks of 20 trees per sampling. All material collected was stored in flasks with 70% ethanol, separated by morphospecies and sent for identification. The climatic data (rainfall, temperature, relative humidity, solar irradiation and wind speed) were obtained from the Main Climatic Station of Montes Claros of the 5th DISME-INMET.
STATISTICAL ANALYSIS
The natural enemies and pollinators were grouped as evaluating them separately would not meet the requirements of the tests. The ecological indices (number of individuals, richness, diversity and abundance of species) were calculated for the arthropod species identified. All ecological indices were measured by calculating the dataset of taxa by samples in BioDiversity Pro Version 2 software. Diversity was calculated by the Shannon-Weaver formula: H' = pi ln (pi). Abundance and species richness (S) were calculated by the Simpson formula: D = (ni /N) * 100, where: pi = ni /N; ni = number of individuals per species; N = total number of individuals; S = richness (number of species). k-Dominance were calculated by plotting the percentage cumulative abundance against log species rank (Lambshead et al. 1983). The kdominance values indicate the dominance and evenness distribution of individuals among species (Gee et al. 1985). The Spearman correlation was applied for the data (P < 0.05).
Results and Discussion
A total of 418 individuals of phytophagous arthropods were sampled, with Hemiptera having the greatest diversity (4 species, 4 genera, 7 families and 4 unidentified species), followed by Coleoptera (1 species, 2 genera, 2 families and 1 unidentified species), Orthoptera (1 species, 1 genus, 2 families and 1 unidentified species), Lepidoptera (1 genus, 1 family and 1 unidentified species) and Diptera (1 genus and 1 family). Aethalion reticulatum L. (Hemiptera: Aetalionidae) was the most abundant phytophagous species on A. mangium trees during the various seasons of the year (Figs. 1 and 2). Aethalion reticulatum feed on plant sap, which can affect the development of fruits and sprouting, and at high infestations, kill the plant (Silva et al. 2007; Vanin et al. 2008).
A total of 1,148 specimens of natural enemies and pollinators arthropods were sampled. The greatest diversity was found in the Hymenoptera (4 species, 8 genera, 3 families and 3 morphospecies), followed by Araneae (2 species, 2 genera, 3 families and 1 unidentified species), Coleoptera (1 species, 1 genus, 2 families and 1 unidentified species), Neuroptera (1 species, 1 genus, 1 family and 1 unidentified species), Mantodea (1 species, 1 genus and 1 family) and Hemiptera (1 unidentified species). The most abundant natural enemies and pollinators were Camponotus sp. 2 (Hymenoptera: Formicidae), Trigona spinipes Fabricius (Hymenoptera: Apidae, Meliponini), Tetragonisca angustula Latreille (Hymenoptera: Meliponinae, Meliponini) and Polistes sp. (Hymenoptera: Vespidae) (Figs. 1 and 2). Despite species of the genus Camponotus being known as predators (or natural enemies) (Cortez et al. 2012), they are also found tending sucking insects, such as A. reticulatum (Brown 1976); protecting them against predators and parasitoids (Renault et al. 2005). Camponotus sp. 1 was correlated positively with A. reticulatum (r = 0.37), so this could explain the greatest abundance of this phytophage on A. mangium. Therefore, Camponotus sp. 1 can indirectly affect host plants by hindering the impact of other natural enemies. However, T. angustula and T. spinipes are important for pollination and for increasing the genetic variability of plants (Proni & Macieira 2004; Costa et al. 2008). Polistes spp. are important predators of different species, mainly lepidopterans (Prezoto et al. 2006).
Arthropods had the greatest H' index in 2005, and the smallest H' in 2007 (Table 1). The greatest number of species and individuals of phytophages was observed in 2005 and 2006, respectively; while natural enemies and pollinators had the greatest number of species and individuals in 2006 (Table 1). The phytophages had the greatest H' index in 2005 owing the presence of rare species, leading thus to greater equitability between number of species and abundance of individuals. This might have occurred owing to the initial colonization stage of different phytophagous arthropod species in the first year of the study. Linzmeier & Ribeiro-Costa (2008) found a similar colonization pattern for species of Galerucinae (Coleoptera: Chrysomelidae) in the first year of study in an Araucaria (Pinales: Araucariaceae) forest.
The smallest H' index may be explained by the reduced number of rare species in the second year when other likely more adapted species to A. mangium were predominant. The natural enemies had the greatest H' index in 2005 and the greatest number of species and individuals in 2006, likely owing to a greater abundance of phytophages in the second year, as observed in the positive correlation between phytophages and their natural enemies (r = 0.45), in agreement with other studies (Donaldson et al. 2007; Öberg et al. 2008; Philpott et al. 2008; Venturino et al. 2008). In general, natural enemies (such as spiders in this study) have greater population densities in more complex habitats (Bragança et al. 1998; Landis et al. 2000) in response to more favorable microclimates and reductions of cannibalism and competition (Ramalho et al. 2007).
Table 1.
Shannon (H') biodiversity indices of phytophagous and natural enemies + pollinators arthropods, number of leaves per branch and number of branches per tree of Acacia mangium. Climatic data during 3 years of sampling.
The increase in number of phytophagous arthropods in the second year is likely due to the greatest number of branches and leaves present on A. mangium trees, which implies an enhanced food resource (Table 1). Other studies have also found this same positive correlation between phytophagous arthropods and complexity of plant architecture (Lara et al. 2008; Obermaier et al. 2008; Sinclair & Hughes 2008). Another possibility is that temperature, rainfall and wind speed had smaller values in 2006 than in 2005 (Table 1), which could have favored some phytophagous, natural enemies and pollinators. This is shown by the negative correlation between temperature (r = -0.44) and rainfall (r = -0.32) with A. reticulatum; temperature with phytophagous arthropods (Hemiptera) (r = -0.42); T. spinipes with temperature (r = - 0.56) and wind speed (r = -0.44); and Camponotus sp. 1 with temperature (r = -0.44). The wind speed found here (over 2.0 m/s) might negatively affect visits of bees (such as of T. spinipes) to flowers (Dutra & Machado 2001).
In the first year, the greatest number of some natural enemies, such as Camponotus sp. 2, Camponotus sp. 5, Polistes spp. and spiders on A. mangium trees (Table 1 and Fig. 1) might have negatively affected the number of phytophages, as indicated by the negative correlation between Camponotus sp. 2 and A. reticulatum (r = -0.41). Some studies have reported Camponotus sp. as important natural enemy in the system (Prezoto et al. 2006; Philpott et al. 2008). This negative correlation between Camponotus sp. 2 and A. reticulatum may also be a result of competition with Camponotus sp. 1, which is reported as frequently tending this hemipteran trophobiont (see the positive correlation above). Thus, Camponotus sp. 2 and 5 could be expelled from plants hosting Camponotus sp. 1 and A. reticulatum.
All arthropod species had greater Shannon (H') biodiversity indices in the summer than in the winter and spring (Table 1). We found a greater number of species and greater abundance of individual arthropod species in the autumn than in the winter and spring, respectively (Table 1). The greatest diversity of phytophagous arthropods may be associated with the availability of resources in the wetter seasons (Wolda 1978). In biomes with well-defined seasons (such as dry and wet in the Cerrado), because population densities and herbivory rates decrease in the dry season (winter) and increase gradually during the wet season (summer) (Tanaka & Tanaka 1982; Coley & Barone 1996). This implies that phytophagous arthropods thrive when present during seasons when trees have new leaves (Wolda 1978), and as observed in this work, when A. mangium had more grown leaves in the summer and fewer in the winter (Table 1).
The greatest abundance of phytophagous arthropods in the autumn is likely due to the effect of herbivore concentration during seasons in which A. mangium has few leaves (Table 1). This was also found for other tree species such as Copaifera langsdorffii Desf. (Fabales: Fabaceae) in the same biome (Almeida et al. 2006). Whereas abundance of natural enemies and pollinators seems to vary according to the phytophagous population density, because their abundance is maximal in the autumn during the peak of population density of phytophages (r = 0.45) (Table 1), and minimal in the spring after the decrease of prey availability (Donaldson et al. 2007; Öberg et al. 2008; Philpott et al. 2008; Venturino et al. 2008).
In conclusion, the diversity of natural enemies and pollinators on A. mangium trees varies according to the population density of phytophagous arthropods. The diversity of natural enemies may explain the small damage on leaves and flowers of this tree species. Finally, control strategies of phytophagous arthropods should be implemented whenever needed at the beginning of tree development and/or in the wet season.
Acknowledgments
We wish to thank Dr Antônio Domingos Brescovit (Instituto Butantã) (Aracnidae), Dr Ayr de Moura Bello (Coleoptera), Dr Ivan Cardoso Nascimento (CEPLAC) (Formicidae), Dr Paulo Sérgio Fiuza Ferreira (UFV) (Hemiptera) and Dr Luci Boa Nova Coelho (UFRJ) (Cicadellidae) for species identification. We also wish to thank “Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)” and “Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG)”.