Survival analyses of Diaphorina citri immatures on young citrus orchard in São Paulo, Brazil Análises de sobrevivência de imaturos de Diaphorina citri em pomar jovem de citros em São Paulo, Brasil

Arq. Inst. Biol., v.86, 1-7, e1302018, 2019 RESUMO: A sobrevivência de ovos e ninfas de Diaphorina citri Kuwayama (Hemiptera: Liviidae) foi estimada em plantas jovens (< 1 ano) de Citrus sinensis (cultivar Valencia) enxertadas em Citrus limonia em 3 estações (inverno e primavera de 2013, e outono de 2014) em Ribeirão Preto, estado de São Paulo, Brasil. Ramos novos com ovos foram selecionados a partir de infestações naturais ou artificiais. Ovos, ninfas pequenas (ínstar 1 ao 3) e ninfas grandes (ínstar 4 e 5) foram contados a cada 3 dias. Ramos com ninfas grandes foram colocados em gaiolas até a emergência dos adultos. As menores sobrevivências de ovos e ninfas foram observadas no inverno (0,0%) e no outono (0,4%), e a maior sobrevivência ocorreu na primavera (12,2%). O principal fator biótico de mortalidade identificado foi o parasitismo de ninfas grandes por Tamarixia radiata Waterston (Hymenoptera: Eulophidae), observado em todas as estações. As temperaturas (mínima e máxima) e umidade relativa (mínima) do ar não explicaram as diferenças nas sobrevivências de imaturos entre a primavera (12,2%) e o outono (0,4%). As chuvas na primavera (165 mm) foram maiores que no outono (48 mm) e podem ter favorecido a sobrevivência de D. citri na primavera. Fatores abióticos adversos parecem ser mais importantes na sobrevivência de imaturos de D. citri que o fator biótico (parasitismo).


KEYWORDS:
D. citri adults may occur in any period or season; however, eggs and nymphs are observed when there are new flushes and leaves in citrus (HALL et al., 2008). In São Paulo State, before it became the target of management programs, the largest adult populations of D. citri occurred during Spring and Summer (YAMAMOTO et al., 2001). The frequent presence of new flushes in young plants may favor the survival of D. citri, and an early HLB infection (HALBERT; MANJUNATH, 2004). Therefore, newly planted citrus trees have been protected with long acting systemic insecticides , and techniques that avoid the access of insects to citrus plants have been adopted (CROXTON; MIRANDA et al., 2015).
The populations of D. citri are regulated by ecological factors of biotic nature: competition, natural enemies and host plant, and abiotic nature: environmental conditions (YANG et al., 2006). Among the natural enemies of D. citri, the parasitoid Tamarixia radiata Waterston (Hymenoptera: Eulophidae) was reported as the most important biological control agent (PLUKE et al., 2008), and it was the only natural enemy observed in citrus areas of São Paulo (PAIVA; PARRA, 2012a). Although predators are important in the control of immature D. citri in other citrus areas (MICHAUD, 2004;STANSLY, 2009), there is no evidence that they are important in Brazil (São Paulo state).
In contrast to what occurs under artificial conditions, where immature individuals of D. citri have high survival rates (> 70%) (NAVA et al., 2007), these immatures have presented low survivals rates in citrus under natural conditions; less than 10% in Florida, USA, (MICHAUD, 2004), and less than 20% in São Paulo, Brazil (PAIVA; PARRA, 2012a). A high mortality of D. citri was observed under adverse artificial conditions by the combination of high temperatures and low air relative humidity (MCFARLAND; HOY, 2001), or in unsuitable host plants (BORGONI et al., 2014;HALL et al., 2015). Considering new planted citrus trees are highly attractive to D. citri adults due to the constant emission of new flushes, and are also susceptible to HLB infection, the survival of immature D. citri was estimated and compared in different seasons.

MATERIAL AND METHODS
Survival analyzes were carried out on orange trees younger than 1 year old (Citrus sinensis), cultivar Valencia, grafted on Citrus limonia. These trees were planted in June 2013, among orange trees (4 years old), cultivar Pera, in Ribeirão Preto, São Paulo state, Brazil (21°12'17"S, and 47°52'17"W). The citrus area was located at 630 m above sea level, in an eutrophic latosol with clay texture; and the climate, according to the classification by Köppen, was type Aw -tropical, with warm and humid Summers, and mild Winters with severe droughts.
Observations occurred in three periods: from July 1 st to July 25 th (Winter season); from October 18 th to November 22 nd (Spring season), 2013; and from March 23 rd to April 18 th (Autumn season), 2014. The trees were irrigated with 3 mm of water (drip system) in a daily basis, in dry periods. At the beginning of the experiments, the plants were pruned and fertilized (50 g of calcium nitrate per plant), and pesticides were not applied during the study.
In July 2013 (Winter), D. citri eggs were not observed, and new flushes were artificially infested. In each young flush, 5 adults of D. citri were caged for 2 days. The insects were reared in Murraya paniculata, at the facility of Escola Superior de Agricultura "Luiz de Queiroz" da Universidade de São Paulo (ESALQ-USP), in Piracicaba City, São Paulo. Fifty-seven flushes of 24 orange trees were infested during Winter. In October (Spring), in 13 orange trees, 52 flushes were naturally infested, and 30, artificially infested. In March of 2014 (Autumn), in 8 orange trees, 42 flushes with eggs were selected after a natural infestation.
Infested flushes were marked with a white line and a numbered label. Live eggs and nymphs were counted with a magnifying glass aid (10 ×) every 3 days. Small nymphs were considered from the 1 st to 3 rd instars, and large nymphs were those from the 4 th and 5 th instars. Branches with large nymphs received a cage, a voile tissue bag of 30 × 50 cm, until adult emergence. Emerged adults were withdrawn and counted. Nymphs parasitized by T. radiata were identified by characteristic mummification.
Daily data of maximum and minimum temperatures, air relative humidity (ARH) at 3 p.m., and rainfall were obtained in an automatic meteorological station located at 200 m from the citrus area. For all seasons, the number of days with: • minimum temperature below 13.5°C, a lower base temperature, estimated by NAVA et al. (2007); • maximum temperature above 32.0°C, temperature unsuitable to development of D. citri immatures (NAVA et al., 2007); and • ARH less than 30%, a condition in which the survival of nymphs is very low  were calculated.
The egg-adult duration was calculated by the weighted mean, considering the number of emerged adults and the period (days) between egg stage and adult emergence. To build the life table, the model proposed by SOUTHWOOD (1978) was adopted. The number of dead insects (dx) at each stage was obtained by the difference between the number of live insects (lx) of the stage and the number of live insects in the next stage (lx1). Mortality was obtained through the relation between dead and living insects at each stage (dx / lx), and survival, through their difference (Sx = 1 -dx / lx). The parasitism of D. citri nymphs by T. radiata was calculated by the ratio (%) between mummified nymphs and large nymphs.

RESULTS
The survival of D. citri eggs was low in the 2013 Winter. After artificial infestations, 567 eggs were obtained, of which only 15 nymphs (2.6%) hatched (Table 1). Inviable eggs, initially of a light-yellow color, became dark in 2 or 3 days. The survival of small nymphs was 20%, and no large nymphs were observed in this season. Thus, with a mean minimum temperature of 12.7°C and maximum of 26.0°C, and 42.8% ARH at 3 p.m., no D. citri adults emerged during Winter (Table 2 and Fig. 1).
In the Spring of 2013, with minimum and maximum temperatures of 17.8 and 30.5°C, and minimum ARH of 41.6% (Table 2 and Fig. 1), the survival of D. citri immatures was higher than in the Winter of 2013, and in the Autumn of 2014 (Tables 1 and 2). From 705 eggs obtained in 82 branches (30 artificially and 52 naturally infested), 268 nymphs hatched (38.0%) ( Table 1). From 705 eggs, 86 D. citri adults were obtained (12.2%) ( Table 1). From oviposition, that occurred  for one week, there was an emergence of adults in two weeks. The estimated egg-adult duration was 18.8 days in the Spring of 2013. In this season, egg-adult survival rates varied among the 13 citrus trees, and in six of them there was no emergence of adults (Table 3). The presence of large nymphs did not guarantee T. radiata parasitism, observed in 4 out of 9 citrus trees.
In the Autumn of 2014, with temperatures similar to those of the Spring of 2013, and slightly higher ARH (45.9%) ( Table 2), the survival of D. citri immatures was low (Table 1)    From 487 eggs, only 2 adults (0.4%) emerged. Despite similar temperatures and ARHs in Spring and Autumn, rainfall was higher and more frequent in the Spring of 2013 (165 mm) than in the Autumnof 2014 (48 mm) ( Table 2 and Fig. 1). D. citri eggs survival was higher during Spring than Winter and Autumn (Tables 1 and 2). The survival of small nymphs was similar during Winter (20%) and Autumn (21%) ( Table 1), and the survival of large nymphs was higher during Spring (47.3%) than Winter (0.0%) and Autumn (11.1%) ( Table 1). Only during 2 days of Winter, the minimum temperature was below the base temperature (13.5°C) estimated for this specie. In Spring and Autumn, during most of the study period, maximum temperatures were above 32.0°C (Table 2), a thermal condition that would limit immature development.

DISCUSSION
Under natural conditions with thermal and hydric variations, which directly influence the insect and the host plant, the survival of D. citri eggs and nymphs was low in this study. In a 4 years old Valencia sweet orange trees, PAIVA; PARRA (2012a) reported survival variation of D. citri immatures from 10% in Spring (2006) and Autumn (2007) to 20% in Summer and late Autumn (2007). The mortality of eggs and small nymphs was the key phase for the population growth of this insect. The results obtained under field conditions differ from those obtained under ideal conditions for insect development (25°C, and 60 -80% ARH, constants). Under controlled conditions, the host plant did not affect the survival of D. citri eggs and nymphs; 75% of the immature D. citri survived in M. paniculata plants, which originated from seeds (LIU; TSAI, 2000). As to C. limonia, the immatures survival was of 72% (NAVA et al., 2007). Similarly, in Valencia sweet orange grafted on C. limonia, the survival rate was 66% (ALVES et al., 2014), and greater than 60% in Pera and Natal oranges BORGONI et al., 2014). In our study, in the field, for C. sinensis trees younger than 1 year old , grafted on C. limonia, there was no immature survival of D. citri in the Winter of 2013, and only 0.4% of them reached their adult life in the Autumn of 2014.
The higher survival of eggs and nymphs may explain the higher population density and population peaks of D. citri during Spring and Summer (YANG et al., 2006;YAMAMOTO et al., 2001). During Autumn and Winter, the population was reduced, probably due to ecological factors, such as competition, natural enemies, host plants, and abiotic nature (YANG et al., 2006). Although D. citri adults occur at any period and season, eggs and nymphs were only observed when there were citrus flushes. Moreover, population outbreaks can occur if environmental conditions are favorable and new shoots are available (HALL et al., 2008).
The frequent presence of flushes in young plants may favor the survival of D. citri and an early HLB infection (HALBERT; MANJUNATH, 2004). Therefore, in new plantations, young trees are rigorously protected with systemic insecticides of long residual action , or foliar insecticide sprays, based on a fixed schedule (BELASQUE JUNIOR et al., 2010). Additionally, methods that prevent insect access to plants may be adopted (CROXTON;MIRANDA et al., 2015).
Adults of D. citri exhibit a high tolerance to extreme temperatures, and can survive in low-temperature conditions (0 or 5°C for 1 day) (EL-SHESHENY et al., 2016), as well as oviposit in warm environments (35°C), depositing about 30 eggs within 48 hours (HALL et al., 2011). Similarly, D. citri adults can survive in very dry environments (MCFARLAND; HOY, 2001). This fact may explain the peak population of D. citri during late Winter and early Spring, as observed by YAMAMOTO et al. (2001) for the conditions of São Paulo.
D. citri parasitism by T. radiata occurred in all seasons of the study, even in very low densities from the third to the fifth instar nymphs (2013 Winter, and 2014 Autumn). In previous research, the highest rates of parasitism were observed during Summer (25%) and Autumn (15%), and the lowest, during Spring (11%) and Winter (6%) (PAIVA; PARRA, 2012b). The data obtained in this study for Spring were similar to those obtained by these authors, indicating that this season is the most favorable to the development of D. citri, and that the lower parasitism by T. radiata may be one key factor for the greater population observed during Spring.
In addition to the parasitoids, the role of predators in the regulation of the D. citri population should be considered, mainly of nymphs, a stage with low mobility and restricted to new shoots. MICHAUD (2002) considered the contribution of predators to the biological control of D. citri greater than that of T. radiata in Florida (USA), and suggested that parasitism could be overestimated, because mummified (parasitized) nymphs are not predated and remain longer in citrus leaves. The T. radiata parasitoid, as far as it is concerned, is the only natural enemy associated with D. citri in São Paulo (PAIVA; PARRA, 2012a).
The low survival of eggs and small nymphs, and additional parasitism of the fourth and fifth instars by T. radiata may limit the natural growth of D. citri population in citrus. Although abiotic factors are not enough to prevent the spreading of HLB pathogens, they play an important role in the natural mortality of eggs and small nymphs, preventing the insect from reaching high populations in São Paulo citrus orchards.

ACKNOWLEDGMENTS
We thank APTA Regional Leste for the assignment of the area for the experiment, the Integrated Pest Management Laboratory of ESALQ-USP for supplying the insects, and Instituto Agronômico de Campinas (IAC) for climatological data.