STINK BUG POPULATION FLUCTUATION AND CONTROL IN SOUR PASSION FRUIT ORCHARDS IN SOUTHERN BRAZIL

ABSTRACT Sour passion fruit (Passiflora edulis Sims, 1818) crops in Brazil are drawing attention due to the national production, and the increasing consumption of this fruit as fresh and processed products. However, fruit losses due to damages caused by stink bugs (Hemiptera: Coreidae) are concerning to fruit growers in southern Brazil. The objective of this study was to evaluate the population fluctuation and control of stink bugs in P. edulis crops grown in the 2013-2014 and 2014-2015 crop seasons in Araquari, SC, Brazil, using natural and synthetic insecticides. The stink bug population was evaluated using the 0.25 m2 frame technique. The evaluations were carried out weekly during the two crop seasons. The species and distribution of stink bugs in the different parts of the plant were evaluated. The insecticides and rates used were: Neenmax (azadirachtin 1.0%) at 10 mL L-1; Orobor N1 (N-citric acid 1.0% + 0.20% B) at 2 mL L-1; and Decis 25 EC (deltamethrin 2.5%) at 1 mL L-1, diluted in aqueous solution. Specimens of three stink bug species where found: 86% Diactor bilineatus (Hemiptera: Coreidae); 13.5% Holhymenia histrio (Hemiptera: Coreidae), and 0.5% Nezara viridula (Hemiptera: Pentatomidae). The D. bilineatus species presented higher frequency, constancy, and dominance during the evaluated crop seasons; 64% of the stink bugs were found on fruits, 18.5% on leaves, 7.3% on floral buds, 5.8% on flowers, and 4.4% on branches. The use of azadirachtin or deltamethrin was efficient to reduce the stink bug infestation and the number of P. edulis withered fruits.


INTRODUCTION
Passion fruit (Passiflora sp.) is one of the main fruit species grown in Brazil (ASSUNÇÃO et al., 2015), where the sour passion fruit (Passiflora edulis Sims, 1818) is the most important commercial species of this fruit. P. edulis plants produce large fruits with acid-orange and aromatic pulp (BEZERRA et al., 2016). The harvested area in the country in 2016 was 49,889 hectares, with mean yield of 14.1 Mg ha -1 (ABF, 2018). The Northeast is the main producing region of passion fruit, with 64.9% of the Brazilian production; and the South region accounts for 6.6% of the national production (IBGE, 2016). Passion fruit orchards are mainly implemented by family farmers because the crop provides fast economic return and source of income (MELETTI, 2011).
Adult stink bugs feed on leaves, twigs, and fruits; and nymphs prefer flower buds and new fruits. Damaged fruits become smaller, withered, and deformed, and have dark punctuations at feeding sites (FADINI;SANTA-CECÍLIA, 2000). Deformations in fruits decrease their quality for fresh marketing (FANCELLI; ALMEIDA, 2002).
Despite few synthetic insecticides are approved for chemical control, it is the most used method for insect pest control in passion fruit ALMEIDA, 2006). Cultural control is an alternative, which consists in the collection and elimination of stink bug eggs, nymphs, adults, and their alternative hosts (MACHADO et al., 2017). The use of stink bug-resistant passion fruit genotype is also recommended as a control strategy (BALDIN; BOIÇA JUNIOR, 1999;CAETANO et al., 2000). Another option for controlling these pests is the use of botanical insecticides, such as citrus and neem (Azadirachta indica) oils, which present low toxicity to mammalian species and high persistence in the environment (ISMAN, 2008).
Considering the damages caused by stink bugs in sour passion fruit (P. edulis) orchards and the little information on the efficiency of control methods, the objective of this study was to evaluate the population fluctuation and control of stink bugs, and the control efficiency of natural and synthetic insecticides in the control of this insect pest in passion fruit orchards.

MATERIAL AND METHODS
The experiment was conducted in a 0.51 ha sour passion fruit orchard, with 860 hybrid plants from crosses of the ovulado, amarelo cerrado, and gigante cerrado lines grown in trellising system, in Araquari, Santa Catarina (SC), Brazil (26º21'56''S, 48º42'26''W, and mean altitude of 4 m). The climate of the region is Cfa, humid subtropical, according to the Köppen classification, with annual average temperature of 20 ºC, relative air humidity of 85%, and annual rainfall of 1,700 to 1,900 mm (PEEL; FINLAYSON; MCMAHON, 2007).  Caetano et al. (2000). A 50×50 cm (0.25 m 2 ) wooden frame was fixed on one side of the trellis, with the aid of a metal hook in the center of the plants and at 1.5 m from the ground. The stink bug species were quantified through this frame, considering the distribution of these insects in the different organs of the plants and the injuries caused by their infestation.
The stink bugs were collected, placed in bottles with 70% alcohol, and sent to the Entomology Laboratory of the Federal University of Pará (UFPA) to determine their species. The frequency, constancy, and dominance of species of stink bugs were determined by using the formulas described by Silveira Neto et al. (1976). The frequency (F) was determined by F = (ni / N) × 100, where F is the frequency of the species i in percentage; ni is the number of individuals of the species i; and N is the total number of individuals collected in the sampled area, classifying the species i as infrequent, frequent, or very frequent. The constancy was determined by the formula C = (p × 100) / N, where C is the percentage of constancy; p is the number of collections containing the species i; and N is the total number of collections.
The species were classified according to the classification proposed by Silveira Neto et al., (1976), which considers constant the species that are found in more than 50% of the collections, accessory when present in 25% to 50% of the collections, and accidental when present in less than 25% of the collections. Dominance was determined by D = Nmax / NT, where D is the dominance, Nmax is the number of individuals of the most abundant species, and NT is the total number of individuals in the sample.
The use of natural and synthetic insecticides was evaluated in the same area and time as the monitoring of the stink bug populations. The treatments consisted of Neenmax (1.0% azadirachtin) at a rate of 10 mL of the commercial product per liter of water; Orobor N1 (Citric acid N 1.0% + B 0.20%) at 2.0 mL of the commercial product per liter of water; Decis 25 EC (deltamethrin 2.5%) at 1.0 mL of the commercial product per liter of water; and a control treatment containing only water. A randomized block experimental design was used, with plots consisting of ten plants.
The treatments were applied fortnightly in the mornings. The products were applied using a 16liters backpack sprayer Model SP16l, Guarany ® , maximum pressure of 6.9 kPa (75 psi), and flow rate of 600 mL min -1 with average speed of 3 km h -1 , and a spray volume rate of 700 L ha -1 . The percentage of withered fruits was recorded during the harvest period in each treatment to determine the effectiveness of the insecticides in the control of stink bugs.
The obtained data were subjected to analysis of variance, and the means were compared by the Tukey's test at 5% significance level. The insect mortality values found were corrected by Abbott's formula (1925) the values expressed as percentage and the direct count were transformed to arcsen √ (x / 100) and √ (x + 0.5), respectively. The statistical analysis was performed with the aid of the DSAASTAT 1.101 program (ONOFRI, 2010).
Regarding the plant parts where the stink bugs species were found, 64% was on fruits, 18.5% on leaves, 7.3% on flower buds, 5.8% on flowers, and 4.4% on fruit branches. The occurrence of stink bugs in five passion fruit species found by Caetano et al. (2000) in Jaboticabal, SP, was predominantly (65.6%) on fruits. In the present study, the stink bugs were mainly on flowers and fruits in development, which directly affects the quantity and quality of fruits.
In general, insect population distribution is connected to host plant availability, which provide them with shelter, food, and mating site. There were guava trees around the evaluated orchard, whose fruits are hosts of H. clavigera (LUNZ et al., 2006). However, this phytophagous stink bug species was not found in the samplings of the present study. There is a hypothesis that the absence of certain stink bug species may be related to the antibiosis factors found in some passion fruit species, such as P. edulis for the stink bugs of the species H. histrio (BALDIN; BOIÇA JÚNIOR, 1999) and L. gonagra (CAETANO et al., 2000). The peak population of stink bugs occurred in March in both crop seasons (Figure 1). Stink bugs were present in the passion fruit orchard during the whole evaluation period. However, in June (end of the harvest period) to October, the number of specimens found in the evaluated areas decreased (Figure 1). Decreases in temperature (CAETANO et al., 2000) and presence of natural enemies (FANCELLI; ALMEIDA, 2002) are factors that can decrease the stink bug population in passion fruit orchards. However, the absence of reproductive structures such as flowers, flower buds, and fruits (main food sources), especially from June to October, was the main factor for the decrease in the stink bug population (Figure 1). The stink bug population fluctuation was influenced by the application of natural or synthetic insecticides and differed from the control treatment. The plants in the treatment without insecticides presented variation of 8 to 28 stink bugs per month during the passion fruit production period, October to June (Figure 1). In this period, the number of stink bugs on plants treated with insecticides was less than one per month, presenting significant difference when compared to the control treatment (Table 2).  Regarding the control efficiency of the insecticides, the stink bugs population on plants in the treatments were 93% lower for the citric acid, 95% lower for the azadirachtin, and 98% lower for the deltamethrin, when compared to the control treatment. Thus, no significant differences in control efficiency were found when comparing the treatments containing natural or synthetic insecticides ( Table 2).
The lower incidence of stink bugs on passion fruit plants treated with azadirachtin and deltamethrin resulted in a 23% and 29% higher fruit yield, respectively, when compared to the control treatment ( and deltamethrin resulted in significantly lower percentage of withered fruits due to stink bug damages in both crop seasons; and the one with citric acid differed from the control treatment only in the 2014-2015 crop season (Table 2). These results show that both synthetic and natural insecticides can decrease the percentage of fruit damaged by stink bugs, ensuring the fruit quantity and quality of passion fruit orchards for both fresh and industry markets. However, plants treated with citric acid showed higher percentage of withered fruits when compared to those treated with azadirachtin or deltamethrin, in both crop seasons (Table 2). According to Lebedenco et al. (2007), the use of chemical control of insect pests by applying synthetic insecticides may result in significant higher production cost, especially when used without other control strategies. Another disadvantage of this method is that these insecticides usually have low selectivity to natural enemies, as occurs for deltamethrin (HOHMANN et al., 2010;STECCA et al., 2017).
The natural citric acid-based insecticides used in integrated pest management (IPM) programs present selectivity for the parasitoid Trichogramma pretiosum (LUCKMANN et al., 2014). In the present study, despite the use of citric acid did not result in a satisfactory control of stink bug damages, it contributed to the reduction of the insect population.
Botanical insecticides have been shown to have low toxicity to mammals and low persistence in agroecosystems (ISMAN, 2008). Neem-based compounds applied in vineyards have been found to be less toxic to certain natural enemies, such as T. pretiosum (MORANDI FILHO et al., 2006). However, Ndakidemi, Mtei andNdakidemi (2016) showed that these compounds present different levels of toxicity to natural enemies and pollinators, and reported that neem-based biopesticides may, in some situations, reduce parasitism rates of natural enemies and cause lethal and sublethal effects to predators and pollinators. Contrastingly, Badshah et al. (2017) showed that pure oil and seed extract of neem are proven to be less toxic to natural enemies than synthetic insecticides. Moreover, Ratnakar et al. (2017) reported that organophosphates and pyrethroids were significantly more toxic to pollinators than neem oil, according to the toxicity classes stablished by the International Organization for Biological and Integrated Control of Noxious Animals and Plants (IOBC).
Therefore, chemical compounds for controlling insect pests in IPM systems need to be used with caution, regardless their synthetic or natural origin. Moreover, constant studies are needed to optimize the concentrations to be used, the application times, and the residual effects of these compounds on natural enemies and pollinators. The optimization of these factors makes possible to create and improve more sustainable ways of using these chemical compounds, especially those of natural origin.

CONCLUSIONS
The stink bug species Diactor bilineatus is constant, frequent, and dominant in the sour passion fruit (Passiflora edulis) crops in Araquari, SC, Brazil. D. bilineatus specimens are found from October (beginning of flowering) to June (end of harvesting), with population peak in March. Application of azadirachtin or deltamethrin is effective to control stink bug infestation and to reduce the occurrence of P. edulis withered fruits.