Lepidoptera defoliators can be very damaging to eucalyptus plantations in Brazil. The objective of this study was to evaluate how plant age, the number of rotations, the tree growth rate (m3 of wood per ha per yr), the distance of native vegetation strips from the eucalyptus plantations, and the width of these strips affect the population dynamics of Lepidoptera defoliators in eucalyptus crops. The survey of the lepidopteran species was conducted fortnightly from Sep 1992 to Aug 1994 using light traps in Eucalyptus urophylla S. T. Blake (Myrtaceae) plantations in 4 areas of the Brazilian Amazon region. In total, 1,049, 1,096, 1,020, and 853 Lepidoptera species with 4,413, 3,457, 3,226, and 2,222 individuals and 11, 11, 11, and 10 species of primary pests were recorded. The primary pest species were represented by 272, 772, 963, and 411 individuals, corresponding to 1.1, 1.0, 1.1, and 1.2% of the species and of 6.2, 22.3, 29.8, and 18.5% of the individuals collected in the 4 areas, respectively. Eupseudosoma aberrans Schaus (Arctiidae), Eupseudosoma involuta Sepp (Arctiidae), Nystalea nyseus Cramer (Notodontidae), Oxydia vesulia Cramer (Geometridae), Stenalcidia grosica Schaus (Geometridae), and Thyrinteina arnobia Stoll (Geometridae) were the most abundant and represent 83.2% of primary pests species. The number of individuals of the primary pest species were not correlated with plant age, the number of rotations, the distance of native vegetation strips from the eucalyptus plantations, and the width these strips, but the total number of individuals of defoliating Lepidoptera had an inverse correlation with the growth rate (m3 of wood per ha per yr) of eucalyptus plants.
The study of factors affecting insect pest outbreaks is important in developing integrated management programs with reduced costs and adverse impacts. Environmental factors such as plant age and the diversity of the agro-ecosystem can affect populations of insect pests (Crocomo 1990).
Lepidoptera are among the most frequently used bio-indicators for monitoring ecosystems (Hilty & Merenlender 2000) and understanding environmental changes (Sparrow et al. 1994). They also are among the most destructive insects in affecting plant establishment, and can they be sampled easily (Axmacher & Fiedler 2004; Brehm & Axmacher 2006; Hawes et al. 2009).
Light traps are commonly used for monitoring, collecting, and defining methods of controlling insects. These traps have been used to evaluate Lepidoptera populations in eucalyptus crops (Pereira et al. 2001; Zanuncio et al. 1993, 2001a,b, 2003, 2014a), native forests (Ignatov et al. 2011; Santos et al. 2015; Vieira et al. 2015), and grazing areas (Delfina & Teston et al. 2013). They also are effective in the assessment of biodiversity. For example, in Altamira, Pará, Brazil, in a primary forest area, 78 species of Arctiinae were collected with 12 of them being new records (Teston et al. 2012). In Tijucas do Sul, Paraná, Brazil, 124 species of Bombycoidea were collected in a mixed rain forest (montane), with 10 new species records for this state (Santos et al. 2015).
The objective of this study was to observe the effects of plant age, number of rotations, tree growth (m3 of wood per ha per yr), distance of native vegetation strips from the eucalyptus plantations, and width of these strips on the population dynamics of Lepidoptera defoliators in Eucalyptus urophylla S. T. Blake (Myrtaceae) plantations in 4 areas of the Amazonian region of Brazil.
Materials and Methods
STUDY SITE
The population fluctuation of Lepidoptera defoliators was studied from Sep 1992 to Aug 1994 in E. urophylla plantations established in 4 areas of the Amazonian region of Brazil (Ponte Maria in Mar 1991, Pacanari in Mar 1992, and Caracurú and Felipe in Mar 1990; Table 1). The first 3 were in the municipality of Almeirim, Pará State, and the last in the municipality of Laranjal do Jari, Amapá State, in a region of tropical humidity climate, with an average distance of 50 km between them at latitudes from 1.00° to 2.00°S and longitudes of 52.00° to 53.00°W.
EVALUATION AND INSTALLATION OF LIGHT TRAPS AND IDENTIFICATION OF INSECTS
Defoliating Lepidoptera were collected in traps with 55 amp blacklights powered by 12 V batteries. Four traps were fixed in wooden supports (1 per area) at the midpoint between 4 eucalyptus trees at 2 m height (Lara et al. 1977). The traps were installed fortnightly at 6:00 PM and withdrawn early in the next morning when they were turned off. A plastic bag (45 × 75 cm) was fixed at the base of each trap; it contained pieces of paper and a jar with ethyl acetate to kill the captured insects and reduce morphological damage (Ferreira & Martins 1982).
Table 1.
Locale and characteristics of Ponte Maria, Pacanari and Caracurú in Almerim Municipality, Pará State, and of Felipe, in Laranjal do Jari Municipality, Amapá State, Brazil, cultivated with Eucalyptus urophylla (Myrtaceae).
The insects collected were removed from the plastic bags, separated by size, packed in entomological envelopes (15 × 15 cm) lined with cotton, and identified with the date and place of collection. These insects were transported to the Universidade Federal de Viçosa (UFV) in Viçosa, Minas Gerais State, Brazil, where they were quantified, assembled, identified, and characterized as pests of eucalyptus based on the collection of the Regional Museum of Entomology (UFVB) (Zanuncio et al. 1993, 1994, 2003).
DATA ANALYSES
Pearson correlation and regression analyses were used to relate the numbers of individuals and Lepidoptera defoliator species with the plant age (yr), the number of rotations (2nd, 3rd, and 5th) (number of times the trees had been harvested since the plantation was initiated), the growth rate (m3 of wood per ha per yr) of eucalyptus plants, the distance (km) of native vegetation strips from the eucalyptus plantations, and the width (m) of these strips. The significance of the linear correlation between variables (r) was determined at the 5% level of probability (SAS Institute Inc. 2002–2003).
Results
In total, 1,049, 1,096, 1,020, and 853 Lepidoptera species with 4,413, 3,457, 3,226, and 2,222 individuals were collected in the areas of Ponte Maria, Pacanari, Caracurú, and Felipe, respectively, including 11, 11, 11, and 10 primary pest species with 272, 772, 963, and 411 individuals. These values corresponded to 1.1, 1.0, 1.1, and 1.2% of the total number of species, and 6.2, 22.3, 29.8, and 18.5% of those individuals collected in the 4 areas (Table 2).
The percentage of primary pest species ranged from 1.0 to 1.2%, but the individuals of the group represented 6.2 to 29.8% of the total collected in the areas of Ponte Maria, Pacanari, Caracurú, and Felipe. Eupseudosoma aberrans Schaus (Arctiidae), Eupseudosoma involuta Sepp (Arctiidae), Nystalea nyseus Cramer (Notodontidae), Oxydia vesulia Cramer (Geometridae), Stenalcidia grosica Schaus (Geometridae), and Thyrinteina arnobia Stoll (Geometridae) were the most abundant, comprising 83.2% of primary pest species (Table 2). Eupseudosoma aberrans and E. involuta, species with similar biological features, were most abundant in Caracurú, with 48.4 and 58.3% of the total number of primary pest individuals, as compared with 29.1 and 26.5%, 9.9 and 7.1%, and 12.6 and 8.1% of the individuals collected in Pacanari, Ponte Maria, and Felipe, respectively (Table 2). Misogada bleura Schaus (Notodontidae) and N. nyseus were most abundant in Caracurú with 54.7 and 65.9%, respectively, of the primary pest species individuals collected in the area. Misogada bleura comprised 11.2 and 17.6% of individuals of Ponte Maria and Felipe, respectively (Table 2). Oxydia vesulia and Sarsina violascens Herrich-Schäffer (Lymantriidae) comprised the highest percentages of individuals in Felipe (51.4 and 52.5%, respectively), whereas Dirphia rosacordis Walker (Saturniidae) was most abundant in Ponte Maria (24.0%), S. grosica in Caracurú (36.6%), and T. arnobia in Pacanari (79.2%) (Table 2).
Table 2.
Total numbers and percentages of individuals of the principal Lepidoptera defoliator species in 4 areas with Eucalyptus urophylla (Myrtaceae) plantations in Pará and Amapá states, Brazil.
Table 3.
Pearson correlation values between the total numbers of individuals and of the principal species of Lepidoptera defoliators in 4 areas cultivated with Eucalyptus urophylla (Myrtaceae) in Pará and Amapá states, Brazil, with plant age (A), the number of rotations (B), the tree growth (m3 of wood per ha per yr) (C), the distance of native vegetation strips from the eucalyptus plantations (D), and the width these strips (E).
Correlation and regression analyses of O. vesulia and S. violascens individuals versus the number of eucalyptus rotations revealed positive (r = 0.63; P < 0.001) and negative (r = −0.64; P < 0.001) relationships, respectively (Table 3; Fig. 1). The total number of individual Lepidoptera specimens collected was not correlated with plant age, number of rotations, distance of native vegetation strips from the eucalyptus plantations, and width these strips (Table 3). The most negative relationship was found between the average annual growth of eucalyptus trees with the total number of individuals of defoliating Lepidoptera (r = −0.72; P < 0.001) (Table 3; Fig. 2). Total numbers of individuals of primary pest species were not correlated with plant age, number of rotations, growth of eucalyptus trees, and width of native vegetation strips (Table 3), but they were positively correlated (r = 0.69; P < 0.001) with the distance of these strips from eucalyptus plantations. Correlation and regression analyses showed that the species M. bleura and S. violascens were negatively correlated (r = −0.63; P < 0.001 and r = −0.68; P < 0.001, respectively) with this parameter (Table 3; Fig. 3).
Discussion
The number of species of lepidopteran defoliators of eucalyptus in the Amazon region was similar to that reported for other areas of Brazil (Pereira et al. 2001; Zanuncio et al. 2003). This finding indicates that eucalyptus insect pests exist throughout the country, although their population densities depend on the presence of this crop. The 4 main species of lepidopteran pests of eucalyptus (E. aberrans, E. involuta, S. grosica, and T. arnobia) found in this region are generally those reported as the most harmful to eucalyptus in other regions of Brazil (Zanuncio et al. 1994).
The low diversity of species and the large number of the individuals of primary pests may be related to increased supply of food in eucalyptus cultivation (Zanuncio et al. 1998), or reduced or insufficient natural biological control of their populations. Modified ecosystems are simpler than natural ones, with a stable food supply available and lower impact by natural enemies, factors that favor the development and multiplication of pest insects such as defoliating caterpillars (Price 1984; Bragança et al. 1998; Zanuncio et al. 1998). On the other hand, the implementation and maintenance of complex ecosystems with high numbers of species favor the biological control of these pests (Botham et al. 2015).
The lower percentages of primary pest individuals in Ponte Maria (6.2 %) and Felipe (18.5 %) than in Pacanari (22.3 %) and Caracurú (29.8 %) suggest that the benefits of native vegetation in these areas may enhance in the biological control of Lepidoptera defoliators of eucalyptus. Light traps 1 and 4, the nearest to these native areas, had the lowest percentages of individuals of the primary pests. Native vegetation areas can provide ecological corridors that facilitate the dispersal of natural enemies to eucalyptus plantations (Altieri 1999). This effect was demonstrated in the Cerrado and Amazonian regions, where fewer Lepidoptera pests were captured in eucalyptus plantations with adjacent strips of native vegetation (Bragança et al. 1998; Zanuncio et al. 2014b). The numbers of primary pest individuals were lower, and those of natural enemies higher, near native areas (remnants of Atlantic Forest) compared with the interior of eucalyptus plantations (Bragança et al. 1998; Zanuncio et al. 1998). These findings show the importance of native vegetation as an environmental manipulation mechanism to reduce populations of eucalyptus-defoliating species.
The positive correlation between the number of rotations on eucalyptus plantations and the number of O. vesulia individuals shows that populations of this species may increase with the time of cultivation of plants in the same area. In addition, the increase in the number of rotations of the culture may reduce the soil fertility, leaving the plants more susceptible to pests and diseases. This correlation indicates the importance of monitoring for pest insects and of silvicultural treatments such as the removal of lower branches to cultivate forests with high productivity and low pest susceptibility. On the other hand, the negative correlation between the number of S. violascens individuals with the number of rotations suggests differences in the behavior of eucalyptus-defoliating species that may become pests throughout the crop cycle.
The presence of dense undergrowth in eucalyptus stands at Pacanari and Caracurú, where traps 2 and 3 were placed, respectively, was not enough to foster adequate natural biological control of Lepidoptera pests, although this lack of natural control may also be due to the great distance from areas of native vegetation (Thomas et al. 2001; Teja & Roland 2004). The negative correlation (r = −0.72; P < 0.001) between the total number of individual Lepidoptera defoliators and the annual growth of eucalyptus plants indicates that low growth rate may favor the multiplication of these insects. However, the abundance of these defoliating insects could also be related to this reduction in growth. This correlation confirms reports that most of the outbreaks of Lepidoptera defoliators were observed in eucalyptus plantations with low productivity, such as those in sandy soils or with non-adapted clones and species (Zanuncio et al. 2003, 2014a). Furthermore, these reports reinforce the importance of selecting plants with good quality genotypes, especially those adapted to growth in various soil types and those resistant or tolerant to damage by pests and diseases (Silva et al. 2013).
The positive correlation (r = 0.69; P < 0.001) between the number of individuals of primary pests and the distance from the native vegetation to eucalyptus plantations may indicate that more diverse vegetation may result in more numerous natural enemies, which can suppress pest species in the eucalyptus plantations (Bragança et al. 1998; Zanuncio et al. 1998). Previous research has shown that Hymenoptera natural enemies were more abundant at the edge of eucalyptus plantations adjacent to native forests and inside native forests than inside eucalyptus plantations (Freitas et al. 2002). The abundance of Hymenoptera parasitoids was also higher at the edge of native vegetation adjacent to eucalyptus plantations compared with their abundance inside these plantations (Dall'oglio et al. 2003). Strips of native vegetation can increase the diversity and abundance of natural enemies and their effectiveness in biological control (Zanuncio et al. 1998; Freitas et al. 2002). The region of eucalyptus plantations sampled has more than 95% of its areas covered with native forests, but the distribution of these forests should be improved to facilitate the dispersion of natural enemies. The presence and dispersion of natural enemies might help to increase eucalyptus productivity, making it possible to reduce the number of hectares planted without reducing total wood production. In addition, planting on slopes is difficult, and therefore plantations in steep areas are being converted back to natural vegetation that can increase populations of natural enemies and reduce problems with insect pests. The negative correlation between the numbers of M. bleura and S. violascens individuals with the distance of native vegetation may indicate that these pests are dispersing to eucalyptus stands. However, their natural enemies can also migrate to these areas and contribute to keeping pest populations at low levels.
Numbers of adult Lepidoptera pests of eucalyptus showed positive and negative correlations with environmental and cultural factors such as average annual growth, rotation number, and distance to the native vegetation. This result shows the need for considering these factors in integrated pest management programs to increase the efficiency of biological control in cultivated forests.
Acknowledgments
We express our thanks to “Conselho Nacional de Desenvolvimento Cientifico e Tecnológico (CNPq)”, “Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)”, “Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG)” and the “Programa Cooperativo sobre Proteção Florestal/PROTEF do Instituto de Pesquisas e Estudos Florestais/IPEF” for the financial support. Dr. Phillip John Villani (University of Melbourne, Australia) revised and corrected the English language used in this manuscript.