Biological Control 60 (2012) 123–131
Contents lists available at SciVerse ScienceDirect
Biological Control journal homepage: www.elsevier.com/locate/ybcon
Hyperparasitism and seasonal patterns of parasitism as potential causes of low top-down control in Euproctis chrysorrhoea L. (Lymantriidae) Enric Frago a,b,⇑, Juli Pujade-Villar c, Miguel Guara d, Jesús Selfa a a
Universitat de València, Facultat de Biologia, Departament de Zoologia, València, Spain University of Oxford, Department of Zoology, Oxford, United Kingdom c Universitat de Barcelona, Facultat de Biologia, Departament de Biologia Animal, Barcelona, Spain d Universitat de València, Facultat de Biologia, Departament de Botànica, València, Spain b
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
" Browntail moth hyperparasitoids attack three out of the 17 primary parasitoids of this moth. " The attacked primary parasitoids are among the most important ones. " Seasonal patterns of parasitoid attack and flight help to understand browntail moth parasitism. " Hyperparasitism may explain browntail moth densities in certain habitats. " This study has implications for pest control in diverse forests.
a r t i c l e
i n f o
Article history: Received 21 April 2011 Accepted 22 November 2011 Available online 30 November 2011 Keywords: Behavioural interference Browntail moth Diversity Forest pest control Hyperparasitism
a b s t r a c t Pest suppression is an important ecosystem service provided by biodiversity, though antagonistic interactions may jeopardize its impact on pest suppression. Hyperparasitoids may release herbivore populations from natural enemy pressure and lead to outbreaks directly due to parasitism as well as indirect through behavioural interference. In a previous study we reported that in native populations of Euproctis chrysorrhoea L. (Lymantriidae) primary parasitism was very low and outbreaks were more likely in coastal habitats than inland. Here we hypothesise that hyperparasitoids are the underlying cause of such patterns by reporting data on direct hyperparasitism rates as well as seasonal patterns of parasitoid attack. Of the 17 primary parasitoids attacking E. chrysorrhoea, three were found to be hyperparasitized. Hyperparasitoids attack the most important E. chrysorrhoea primary parasitoids which may explain the pattern of moth density in some habitats. Seasonal patterns of parasitoids attack and flight also help to understand antagonistic interactions among E. chrysorrhoea parasitoids. We discuss the implications of our work in the context of pest control in diverse ecosystems. Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction Pest suppression is an important ecosystem service provided by biodiversity (Kremen, 2005; Tscharntke et al., 2005; Wilby and ⇑ Corresponding author at: University of Oxford, Department of Zoology, South Parks Road, Oxford OX1 3PS, United Kingdom. E-mail addresses:
[email protected] (E. Frago),
[email protected] (J. PujadeVillar),
[email protected] (M. Guara),
[email protected] (J. Selfa). 1049-9644/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.biocontrol.2011.11.013
Thomas, 2002). The combined action of different natural enemies on a specific herbivore is usually equal (additive) or greater (synergistic) than the summed mortality caused by each natural enemy separately. Natural enemy diversity, however, may also result in reduced herbivore suppression if they interact antagonistically (Letourneau et al., 2009) with higher trophic levels jeopardizing natural control of herbivorous pests. Understanding such interactions in diverse natural ecosystems is essential to resolving the
124
E. Frago et al. / Biological Control 60 (2012) 123–131
long-standing debate among biocontrol experts over whether parasitoid diversity improves pest suppression or not (Denoth et al., 2002; Kakehashi et al., 1984). Antagonistic interactions among natural enemies (especially among insect predators and parasitoids) may appear as a result of intraguild predation, behavioural interference or hyperparasitism. These interactions may release herbivore populations from natural enemy pressure and therefore lead to outbreaks (Rosenheim, 2007; Schmitz, 2007). It is usually assumed that hyperparasitism is a constraint to the natural control of pests, either through direct parasitism on the primary parasitoids or through indirect effects on the primary parasitoid’s foraging behaviour (Rosenheim, 1998). As Rosenheim pointed out, however, theoretical models predict that obligate hyperparasitoids may improve top–down control by stabilizing fluctuations in the food web. Identifying the mechanisms promoting synergistic or antagonistic interactions among hyperparasitoids and primary parasitoids of pest species is of interest to improve biocontrol strategies. The browntail moth, Euproctis chrysorrhoea L. (Lepidoptera: Lymantriidae), is a highly polyphagous pests of economic importance given its eruptive population dynamics in forest as well as in agricultural ecosystems (Fernald and Kirkland, 1903; Torossian et al., 1988). In addition to its defoliating activity, this moth is also a public health risk because the larvae have a type of urticating hairs that can cause severe skin rashes in people (Blair, 1979) and even death (Schaefer, 1974). Few studies have addressed the factors influencing the population dynamics of the browntail moth. In native habitats, only Sterling and Speight (1989) and Frago et al. (2011) have thoroughly investigated browntail moth dynamics reporting an overall parasitoid impact of 5% and 12%, respectively. Studies carried out in the browntail moth populations introduced in America revealed that overall parasitoid impact was close to 30% (Schaefer, 1974). Browntail moth populations in America are currently restricted to coastal habitats because primary parasitism is lower than inland (Elkinton et al., 2006). Our previous work carried out in eastern Spain, partially confirmed these results: outbreak densities were found in coastal habitats (Thermo-mediterranean), but latent densities were found inland (Meso-mediterranean). Nevertheless, we found that overall primary parasitism was very low and that the underlying cause of such a density pattern was lower females’ fecundity inland (Frago et al., 2011). These results suggested that primary parasitism is not a key factor governing the density of the browntail moth in native habitats. Previous studies which reported an abundant and diverse complex of hyperparasitoids attacking browntail moth primary parasitoids (Arevalo-Durup, 1991; Proper, 1934; Schaefer, 1974) also suggest that higher order predators are one of the likely explanations for this. In this work, we hypothesize that the low impact of primary parasitoids on browntail moth populations, as well as the higher moth density in coastal habitats (Frago et al., 2011) is due to interactions with hyperparasitoids. Two general approaches are carried out, first we compare hyperparasitism rates between browntail moth natural populations found in both coastal and inland habitats. We hypothesise that higher hyperparasitism rates in coastal habitats may lead to the ecological release of these populations. Second, we report data on the phenology of both primary parasitoids and hyperparasitoids of the browntail moth. Hyperparasitoids may impact primary parasitism directly or indirectly via behavioral interference (i.e. non-consumptive effects, Werner and Peacor (2003)). Hence, seasonal patterns of parasitoid flight may help to understand whether the outcome of their interaction has a negative or positive effect on pest suppression.
2. Methods 2.1. Study sites The parasitoid complex attacking E. chrysorrhoea was studied in 4 sites, located in eastern Spain where this pest commonly infests Arbutus unedo trees (Generalitat Valenciana, 2008). Two sites were located in coastal Thermo-mediterranean thermotype habitats (Rivas-Martínez et al., 2002): El Garbí (latitude: 39°400 3100 N, longitude: 0°220 3600 E, altitude: 600 m a.s.l.) and El Rodeno (39°490 5100 N, 1°160 5900 E, 450 m a.s.l.); and two in inland Meso-mediterranean thermotype habitats (Rivas-Martínez et al., 2002): Ganacienda (39°230 5800 N, 1°170 1300 W, 650 m a.s.l.) and Zacaé (39°070 3700 N, 1°080 5600 W, 850 m a.s.l.). Coastal and inland habitats are approximately 80 km apart. These habitats are commonly inhabited by A. unedo, and in the studied sites this tree was the dominant species due to the degradation of the original forest. At each site, a study plot of 4–6 ha which included a minimum of 160 A. unedo trees taller than 1 m was established. Twenty attacked A. unedo trees per site were selected including individuals with different levels of browntail infestation. The main part of the study was carried out for two generations of the moth, from May 2003 to June 2005 (hereafter referred to as 2003–4 and 2004–5 generations). To study the trophic relationships among parasitoids, sampling were also conducted on 2002–3 and 2005–6 generations. 2.2. Collected insect material At each study site and browntail moth generation, the parasitoids of the different browntail moth stages were randomly sampled following a sequential sampling scheme as in Frago et al. (2011). The shifted phenology of this moth when feeding on A. unedo was specially taken into consideration: while larvae remain on diapause at least from September to March on deciduous trees (Fernald and Kirkland, 1903), on A. unedo larvae feed throughout winter (Frago et al., 2010). The number of collected samples depended on moth density; when browntail moth density was not enough to collect a significant number of winter nests, they were collected outside the study plot established at each study site. Samples were incubated at room temperature and in order to retrieve emerging parasitoids they were examined regularly. Samples were conserved in the laboratory until the next moth generation, and then examined carefully in order to assess whether parasitoids where primary or hyperparasitoids. If imagos did not emerge from their pupae after this period, pupae were considered aborted. Due to health reasons, samples were handled under a laminar flow cabinet. The following moth stages were sampled: (i) A maximum of 100 egg-batches per study site and generation were collected between May and June from different A. unedo trees. Eggs were preserved in 6.5 ml clear plastic vials. (ii) During the first year of study, a maximum of 140 winter nests were collected regularly from September to April at each site. In the following generation, nests were collected in dates determined after the local phenology of parasitoids for each site. Nests were placed individually in 650 cl containers. In order to assess trophic relationships inside winter nests, at the end of the study nests were dissected and carefully examined; parasitoid pupae were collected and inspected under a stereomicroscope. Hyperparasitism of Pediobius pyrgo (walker) on Trichomalopsis peregrina (Graham) was estimated by counting imagos emerged from winter nests; then nests were inspected and parasitized T. peregrina pupae counted to assess whether P. pyrgo was exclusively feeding on this host. (iii) Pupae of postdiapausing gregarious larval parasitoids were collected from the surface of browntail winter nest as well as from A. unedo trees. To do this, the selected A. unedo trees were examined every 14 days
E. Frago et al. / Biological Control 60 (2012) 123–131
from October to April. (iv) Pupae of dispersive larval parasitoids were collected from specific patches of small shrubby plants where dispersive larvae usually feed on (Frago et al., 2010). These patches were examined every 14 days from March to May. When Meteorus versicolor (Wesmael) pupae were found emerged in the field, they were classified as either unparasitized or parasitized by an unknown hyperparasitoid depending on the eclosion hole: M. versicolor imagos bite a round eclosion hole on one of the extremes of the pupae, while hyperparasitoids emerge through a small lateral hole with irregular margins. (v) Given that in 2002– 3 we were unable to find any browntail moth pupae in the field, prepupal and pupal parasitoids were obtained in 2003–4 generation by rearing in the laboratory a group of dispersive larvae. Once larvae spun the cocoon (i.e. they entered the prepupal stage) they were transplanted into their population of origin. They were fixed to a plastic mesh with a small drop of silicone based glue and placed in small groups at the base of those small shrubby plants where dispersive larvae usually feed on. Pupae were collected three to four weeks later. In 2004–5, parasitoids were obtained by restricting larval dispersion with PVC plastic barriers greased with Vaseline as described in Frago et al. (2011). Following Elkinton et al. (1992), when more than one hyperparasitoid attacked the same primary parasitoid species in the same site and generation, marginal attack rates were calculated. This value estimates the level of mortality associated to a single hyperparasitoid as if that was acting alone.
mi ¼ 1 ð1 qÞqi =q were mi is marginal attack rate from the ith hyperparasitoid, qi is the parasitism rate from the ith hyperparasitoid and q is combined parasitism rate from all hyperparasitoids. 2.3. Parasitoid exclusion experiment To understand the period that both T. peregrina and P. pyrgo, attack browntail larvae, a parasitoid exclusion experiment was carried out. A total of 48 browntail moth winter nests were collected in El Rodeno (where this parasitoid species was never found) and 12 in Ganacienda. They were deployed to a single A. unedo tree located in Ganacienda from September the 7th to December the 28th (i.e. the beginning and end of browntail moth larval diapause; Frago et al. (2010)). Nests were deployed at 1.5 m height homogeneously so that 15 nests were placed at each main cardinal point. The selected tree had an abundant density of browntail moth winter nests, and in the preceding year, winter nests were commonly attacked by both T. peregrina and P. pyrgo. Likewise, this tree was surrounded by other A. unedo trees also attacked by the browntail moth. At each cardinal point two of the nests collected in El Rodeno were assigned to the different treatments and controls represented in Fig. 1. The nests collected in Ganacienda were assigned to both controls (i.e. no exclusion and exclusion). Parasitoid exclusion was achieved by wrapping each nest with a fine cotton mesh. Although
125
parasitoids may be able to oviposit in browntail larvae through the mesh, this was not possible on winter nests since larvae were on diapause inside them. On December the 28th, nests were collected, taken to the laboratory, and reared as explained above. 2.4. Statistical analyses Differences in hyperparasitism between coastal and inland habitats were evaluated using generalized linear models for binary data (i.e. logistic regression; Agresti (2002)) corrected for overdispersion using the quasibinomial function in the statistical package R version 2.10.1 (R Development Core Team, 2008). Models were built considering primary parasitoids as either alive or dead due to hyperparasitoids.
3. Results 3.1. Parasitoid phenology and life history In this study 17 primary parasitoids attacking the browntail moth and 9 hyperparasitoids in eastern Spain were reported (Fig. 2). Egg parasitoids always emerged from browntail moth egg-batches during the few weeks following browntail moth females’ oviposition. Trichogramma spp. species were found in all studied sites, and imagos emerged within 10 days after browntail moth egg laying. Telenomus pinnatus Kozlov & Kononova was only found in coastal sites, and the incubation period was longer. This species was especially abundant in El Garbí 2004–5 where imagos emerged within 2 and 5 weeks after browntail moth egg laying (Fig. 3). A total of 542 browntail moth winter nests were collected between 2003 and 2005. The gregarious idiobiont ectoparasitoid T. peregrina and its associated hyperparasitoid, the solitary idiobiont endoparasitoid P. pyrgo, were abundant in inland habitats but scarcely found in the coast (Table 1). Only 2 winter nests were attacked by the primary parasitoid in El Garbí, and it was never found in El Rodeno. The scrutiny of browntail moth winter nests as well as parasitoid pupae revealed that P. pyrgo never fed on browntail larvae. Thus, this parasitoid was always obtained from browntail moth winter nests where T. peregrina was present. T. peregrina pupae attacked by P. pyrgo were easily identified by the emergence hole of the later species on the pupae of the former. Imagos of T. peregrina as well as P. pyrgo emerged from winter nests from April to June following a unimodal emergence pattern (Fig. 4). In the parasitism exclusion experiment, none of the winter nests collected in El Rodeno and deployed in Ganacienda were attacked by either T. peregrina or P. pyrgo. We found both parasitoid species in two winter nests collected in the same area where the experiment was carried out, specifically in the exclusion control treatment. Apart from P. pyrgo, less than 1% of T. peregrina pupae were hyperparasitized by Aprostocetus sp. in inland habitats.
Fig. 1. Euproctis chrysorrhoea winter nest parasitoid exclusion experiment in 2004. Grey bars indicate the weeks during which nests were exposed to natural enemies.
126
E. Frago et al. / Biological Control 60 (2012) 123–131
Fig. 2. The parasitoid complex of Euproctis chrysorrhoea feeding on Arbutus unedo indicate trophic links pointing from host to parasitoid. Primary parasitoids were previously reported in Frago et al. (2011). ⁄Gregarious larval parasitoids include both diapausing and post-diapausing larval parasitoids.
Fig. 3. Emergence of the imagos of Telenomus pinnatus from Euproctis chrysorrhoea egg-batches in El Garbí 2004.
127
E. Frago et al. / Biological Control 60 (2012) 123–131 Table 1 Primary parasitism (Trichomalopsis peregrina and Apanteles lacteicolor) and hyperparasitism (Pediobius pyrgo and Elasmus nudus) on Euproctis chrysorrhoea winter nests. E. chrysorrhoea winter nests
Coastal habitats
Inland habitats
El Garbi
Collected Attacked by Attacked by Attacked by Attacked by
T. peregrina P. pyrgo A. lacteicolor E. nudus
El Rodeno
Ganacienda
Zacaé
2003–4
2004–5
2003–4
2004–5
2003–4
2004–5
2003–4
2004–5
93 2 1 21 0
103 0 0 16 0
24 0 0 5 0
136 0 0 2 0
31 4 4 2 0
94 5 3 10 3
25 9 2 2 0
36 13 5 3 2
Fig. 4. Emergence of both Trichomalopsis peregrina and Pediobius pyrgo imagos from Euproctis chrysorrhoea winter nests in Ganacienda and Zacaé in 2003–4 and 2004–5 generations.
The solitary koinobiont endoparasitoid Apanteles lacteicolor (Viereck) was found in all the studied sites. Its pupae were mainly obtained from nests collected from February to March, imagos appearing from December until April. Unfortunately, the overall number of pupae was not high enough to well define the flight period for this species. A. lacteicolor was found to be parasitized by the gregarious idiobiont endoparasitoid Elasmus nudus (Nees) in inland sites (Table 1). Although the main part of the imagos of this elasmid appeared from April to June, 3 out of the 33 obtained emerged form A. lacteicolor pupae in September. In April 2006,
two A. lacteicolor pupae collected in Zacaé were superparasitized by both E. nudus and P. pyrgo. The dissection of these pupae revealed that P. pyrgo was living at expense of E. nudus as tertiary parasitoid (Fig. 2). In El Rodeno, M. versicolor pupae were collected from browntail moth winter nests as well as from the small shrubby plants where browntail dispersive larvae usually feed. In El Garbí M. versicolor pupae were found exclusively on winter nests, whereas in inland habitats they were obtained from dispersive larvae. The presence of M. versicolor pupae in El Rodeno 2004–5, followed a bimodal
128
E. Frago et al. / Biological Control 60 (2012) 123–131
curve: the first generation of pupae appeared in January and February, and the second one from March to June. Imagos appeared within 1 and 2 weeks later, following a bimodal curve as well. The flight period of the second generation of M. versicolor was suppressed by the action of hyperparasitoids which killed a large number of pupae (Fig. 5). In this regard, a total of nine different species of hyperparasitoid were obtained from M. versicolor pupae though in some cases only one individual per species was found (Table 2). Among them, Brachymeria secundaria (Habu), a solitary idiobiont endoparasitoid, was widely distributed and the most abundant. The rest of hyperparasitoid species were obtained from M. versicolor pupae collected in the second generation of this parasitoid and the imagos emerged during April to June (Fig. 6). Four different species of tachinid and 5 hymenopterans were obtained from late instar larvae, prepupae and pupae (Fig. 2). Of them, Compsilura concinnata was the only species found in all study sites. All parasitoid pupae were obtained from late February to May. Imagos usually emerged some weeks after except for Carcelia laxifrons and Eremotylus divisor (Aubert) which imagos emerged during the next browntail moth generation (Fig. 7). 3.2. Impact of hyperparasitoids Among the 17 primary parasitoids found attacking the browntail moth, only the larval parasitoids T. peregrina, A. lacteicolor and M. versicolor suffered hyperparasitism. No hyperparasitoids were found attacking the primary parasitoids of late instar larvae, prepupae or pupae. The hyperparasitoid P. pyrgo, was present in 45% of the winter nests attacked by T. peregrina (Table 1). Out of the 2449 T. peregrina pupae studied 35% were hyperparasitized in El Garbí, 5% in Ganacienda and 8% in Zacaé. The logistic regression analysis of data from 2 winter nests parasitized by T. peregrina in coastal sites, and 31 inland, revealed that the odds of mortality of T. peregrina due to P. pyrgo parasitism was 92% lower in inland habitats (F1,2448 = 6.17; p < 0.001). A total of 203 A. lacteicolor pupae were obtained; E. nudus was found attacking them exclusively in inland habitats: 15% in Ganacienda and 50% in Zacaé. A. lacteicolor pupal mortality was 97% lower in the coast than inland (F1,2448 = 4.63; p < 0.001).
To study the hyperparasitoids of M. versicolor, a total of 563 pupae were obtained. Overall, 18% of the samples were hyperparasitised. The highest marginal rates of mortality were found in El Rodeno (Table 2). The hyperparasitoid with the greatest impact was B. secundaria, which reached rates of parasitism up to 20% in some cases. M. versicolor pupal mortality (including all hyperparasitoid species as well as aborted pupae) was not significantly different between coastal and inland sites (F1,562 = 1.31; p = 0.189). 4. Discussion In our previous work 17 primary parasitoids which accounted for 12% of the browntail moth mortality were reported (Frago et al., 2011). The present study adds 10 hyperparasitoids to the complex and at least one facultative hyperparasitoid: Dibrachys lignicola Graham that was found attacking both the browntail moth and M. versicolor pupae (Fig. 2). Unexpectedly, after an intensive sampling effort several browntail moth stages which are usually reported to be hyperparasitized were found to be free from hyperparasitoids (Fig. 2). For instance, Proper (1934) and Schaefer (1974) obtained several hyperparasitoids from the tachinids that attack late instar larvae and prepupae. Although hyperparasitoids were only found attacking three out of the 17 species of primary parasitoids (i.e. T. peregrina, A. lacteicolor and M. versicolor), the attacked species accounted for half the parasitism and 6% overall mortality in Frago et al. (2011). Hyperparasitism suffered by these primary parasitoids merits further investigation first because they kill the larvae before they dramatically defoliate the host tree and therefore they are important in terms of pest control. Second, because they are usually reported as the most important ones parasitizing natural browntail moth populations (Arevalo-Durup, 1991; Burgess and Crossman, 1929; Schaefer, 1974; Sterling and Speight, 1989). Interactions among higher-order predators may govern the population dynamics of herbivores (Rosenheim, 1998). For example, while spider mite populations may be effectively suppressed by a single natural enemy, when several generalist predators are found in the same habitat spider mite outbreaks are common (Rosenheim, 2005). In the browntail moth, outbreak densities were
Fig. 5. Phenology of Meteorus versicolor attacking Euproctis chrysorrhoea larvae in El Rodeno in 2004–5 generation. Also shown is the proportion of hyperparasitized M. versicolor pupae and imagos’ emergence of the hyperparasitoid Brachymeria secundaria.
129
E. Frago et al. / Biological Control 60 (2012) 123–131
Fig. 6. Seasonal pattern of hyperparasitoid attack (black) and imago emergence (grey) on Meteorus versicolor pupae, a primary parasitoid of Euproctis chrysorrhoea. Data from 2003–4 and 2004–5 generations are presented together. Phenology of the hyperparasitoid Brachymeria secundaria is already shown in Fig. 5.
Table 2 Hyperparasitism on Meteorus versicolor pupae, a primary larval parasitoid of Euproctis chrysorrhoea. Also shown is the number of M. versicolor pupae collected as well as the proportion of viable and aborted pupae. See species names in full in Fig. 1. Site and year
Secondary parasitism (marginal rate of attack)a
M. versicolor pupae
Collected Viable Aborted Unknown Brachymeria Gelis Gelis Dibrachys Dibrachys Pteromalus Pediobius Mesochorus Melittobia secundaria carbonarius liparae cavus lignicola chrysos pyrgo tenuis acasta El Garbi 02–03 7 El Garbi 03–04 23 El Garbi 04–05 9 El Rodeno 02–03 69 El Rodeno 03–04 146 El Rodeno 04–05 254 Ganacienda 02–03 27 Ganacienda 03–04 5 Ganacienda 04–05 14 Zacaé 02–03 7 Zacaé 03–04 1 Zacaé 04–05 1 a
0.57 0.74 0.89 0.55 0.67 0.37 0.41 0.40 0.36 0.71 0.00 0.00
0.14 0.26 0.11 0.20 0.32 0.35 0.41 0.60 0.50 0.14 1.00 0.00
0.14 0.00 0.00 0.00 0.01 0.05 0.00 0.00 0.07 0.00 0.00 0.00
0.14 0.00 0.00 0.14 0.00 0.21 0.08 0.00 0.07 0.14 0.00 0.00
0.00 0.00 0.00 0.06 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00 0.00 0.08 0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.05 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.02 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00
0.00 0.00 0.00 0.00 0.00 0.00 0.04 0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Marginal attack rates were calculated when more than one parasitoid species attacked M. versicolor in the same site and generation.
Fig. 7. Emergence of late instar larval, prepupal and pupal parasitoids of Euproctis chrysorrhoea, from their pupae or puparium. Data from 2003–4 and 2004–5 generations are presented together.
found in coastal habitats but latent densities inland (Frago et al., 2011). Both M. versicolor and A. lacteicolor were widely distributed in the studied sites but their pupae did not suffer a higher level of mortality in coastal habitats (Tables 1 and 2). Hyperparasitism on these two species is therefore unlikely to explain outbreak densities in coastal habitats. However, hyperparasitism of P. pyrgo on T. peregrina may impact browntail moth density. As reported in Frago et al. (2011) T. peregrina, together with E. divisor, were the most important parasitoids in inland sites. If T. peregrina is actually a key species in the natural control of browntail moth populations, it is possible that the low impact of P. pyrgo inland has an indirect effect that releases T. peregrina and in turn may maintain browntail moth populations at latent densities. This interpretation would be in agreement with Arevalo-Durup (1991) who found as well low rates of P. pyrgo hyperparasitism studying browntail moth outbreaking populations in France. Evidence is emerging that changes in natural enemy diversity and density influence the functioning of
ecosystems (e.g. Cardinale et al. (2003)). Further research may elucidate whether browntail moth hyperparasitoids are actually attacking keystone species in the trophic web leading to an overall reduction of primary parasitism. Seasonal patterns of parasitoid attack may explain the coexistence of several species attacking the same host in the same habitat. Competition among browntail moth larval parasitoids may be relaxed by attacking gregarious and non-gregarious stages so that their niches only partially overlap (Godfray, 1994). Competition among parasitoids of the same browntail stage, however, may be strong as in the case of the larval parasitoids T. peregrina, M. versicolor and A. lacteicolor. The former is likely to have a competitive advantage over the other two because it is an ectoparasitoid, this condition typically being advantageous over endoparasitoids because host growth is arrested after ectoparasitoid attack (Godfray, 1994). In addition, T. peregrina kills browntail larvae earlier: it kills diapausing larvae whereas M. versicolor and A. lacteicolor kill them
130
E. Frago et al. / Biological Control 60 (2012) 123–131
once they resume post-diapausal feeding activity (Burgess and Crossman, 1929). The early presence of T. peregrina inside winter nests was demonstrated by the exclusion experiment as this species was exclusively obtained from nests in the exclusion control treatment. Based on the competitive advantage of T. peregrina, low hyperparasitism rates of P. pyrgo in inland sites may explain why in Frago et al. (2011) M. versicolor and A. lacteicolor were scarcely found attacking diapausing larvae in these habitats. M. versicolor parasitism on post-diapausing dispersive larvae, however, was higher in inland sites (Frago et al., 2011) supporting the hypothesis that T. peregrina may over-compete this species especially during the gregarious phase of the browntail moth. The competitive disadvantage of M. versicolor with respect to T. peregrina may have a synergistic negative effect due to the fact that the hyperparasitoid P. pyrgo, which is commonly associated with T. peregrina, also attacks M. versicolor (Fig. 2). Though little is known about the biology of some of the reported late instar larvae, prepupal and pupal parasitoids (Noyes, 2001;Yu et al., 2004), competition among them is likely because they attack the same host species. Coexistence among these parasitoids may be possible because several of them are generalist so that the possibility of feeding on many host species lessens the risk of extinction (Godfray, 1994). For instance, M. versicolor is a highly polyphagous species (Yu et al., 2004), and considering that adults can be found in the field from January to June (Fig. 5) they may attack several other species. For instance, this species was reported in El Rodeno attacking the pine processionary moth and at least one tussock moth species (E. Frago, pers obs.). In the same way, the tachinid C. concinnata, is known to parasitize more than 100 species in the Palaeartic region (Herting, 1960). On the contrary, the risk of competitive exclusion is higher in the case of monophagous parasitoids such as T. nidicola and C. laxifrons (Herting, 1976). Theoretical studies predict that spatial and temporal variability may promote species coexistence, with the poorer competitor exploiting patches where the superior competitor is not present (Godfray, 1994). In this regard, different flight periods among late instar larval, prepupal and pupal parasitoids of E. chrysorrhoea (Fig. 7) may reflect different host exploitation strategies. Given that hyperparasitoids may interfere with primary parasitism due to behavioural interference (Rosenheim, 2007), is tempting to speculate that the low impact of primary parasitoids attacking late instar larvae, prepupae and pupae (Frago et al., 2011) is due to the abundant diversity of polyphgous hyperparasitoids flying around from March to May (Figs. 4 and 5). Nevertheless, further studies are necessary to confirm this. In conclusion, we have found that hyperparasitoids attack some of the most important E. chrysorrhoea primary parasitoids. As reported in Frago et al. (2011) E. chrysorrhoea mortality is positively correlated with moth density, and outbreaks are more likely in coastal than in inland habitats. This suggests that in coastal habitats population declines after an outbreak situation are likely to be the result of intra-specific competition. In inland habitats, latent densities were mainly due to lower moth’s realized fecundity. The current work also suggests that the indirect effect of hyperparasitoids might cascade down to the trophic web and maintain inland populations at low density. Although high hyperparasitism rates usually lead to poor biological control (Rosenheim, 1998) other factors may reduce the effectiveness of primary parasitism. In some cases primary parasitoids may be able to control their host in spite of high levels (80–100%) of hyperparasitism (e.g. Hougardy and Mills, 2009). Given that an important topic in community ecology is whether diversity of higher order predators increases herbivore suppression (Letourneau et al., 2009), our results stress that such complex interactions may be context-dependent and tightly related to local conditions. This work also opens the door to future experiments
aiming to elucidate direct as well as indirect effects among the naturally associated parasitoids of this pest. Acknowledgments We especially thank Lee Henry and Mario X. Ruiz-González for their valuable comments on an earlier version of this manuscript. We also thank Rafael Currás and José Martín-Cano for logistic support and two anonymous referees for their constructive comments. Information concerning Euproctis chrysorrhoea in València was kindly provided by the regional Ministry of ‘‘Medi Ambient, Aigua, Urbanisme i Habitatge’’ in València. Parasitoids were identified by the following specialists: Dick Askew, Hannes Baur, Peter Neerup Buhl, Pilar Navarro, María Teresa Oltra, Bernard Pintureau, Heinz Schnee, Martin Schwarz, Hans-Peter Tschorsnig and María Jesús Verdú. This research was funded by the ‘‘Ministerio de Educación y Ciencia de España’’ (Project I+D BOS2002-03820 with an associate 4-year Grant FPI). References Agresti, A., 2002. Categorical Data Analysis. John Wiley & Sons, New York. Arevalo-Durup, P., 1991. Le nid d’hiver d’ Euproctis chrysorrhoea L. (Lepidoptera: Lymantriidae) comme estimateur de population en milieu forestier. Dissertation. Université Paul Sabatier, Toulouse, France. Blair, C.P., 1979. Browntail moth, its caterpillar and their rash. Clinical and Experimental Dermatology 4, 215–222. Burgess, A.F., Crossman, S.S., 1929. Imported insect enemies of the Gypsy moth and the Brown tail moth. U.S.D.A. Technical Bulletin 86, 1–147. Cardinale, B.J., Harvey, C.T., Gross, K., Ives, A.R., 2003. Biodiversity and biocontrol: emergent impacts of a multi-enemy assemblage on pest suppression and crop yield in an agroecosystem. Ecology Letters 6, 857–865. Denoth, M., Frid, L., Myers, J.H., 2002. Multiple agents in biological control: improving the odds? Biological control 24, 20–30. Elkinton, J.S., Buonaccorsi, J.P., Bellows, T.S., Vandriesche, R.G., 1992. Marginal attack rate, K-values and density dependence in the analysis of contemporaneous mortality factors. Researches on Population Ecology 34, 29–44. Elkinton, J.S., Parry, D., Boettner, G.H., 2006. Implicating an introduced generalist parasitoid in the invasive browntail moth’s enigmatic demise. Ecology 87, 2664–2672. Fernald, C., Kirkland, A., 1903. The Brown-tail Moth, Euproctis chrysorrhoea: A Report on the Life History and Habits of the Imported Brown-tail Moth. Wright and Potter Printing, Boston. Frago, E., Guara, M., Pujade-Villar, J., Selfa, J., 2010. Winter feeding leads to a shifted phenology in the browntail moth Euproctis chrysorrhoea on the evergreen strawberry tree Arbutus unedo. Agricultural and Forest Entomology 12, 381– 388. Frago, E., Pujade-Villar, J., Guara, M., Selfa, J., 2011. Providing insights into browntail moth local outbreaks by combining life table data and semi-parametric statistics. Ecological Entomology 36, 188–199. Godfray, H.C.J., 1994. Parasitoids: Behavioral and Evolutionary Ecology. Princeton University Press, Princeton, New Jersey. Herting, B., 1960. Biologie der westpaläarktischen Raupenfliegen Dipt., Tachinidae. Monographien zur angewandte Entomologie 16, 1–188. Herting, B., 1976. Lepidoptera, Part 2 (Macrolepidoptera) A Catalogue of Parasites and Predators of Terrestrial Arthropods. Section A. Host or Prey/Enemy, vol. 7. Commonwealth Institute of Biological Control, Commonwealth Agricultural Bureaux. Hougardy, E., Mills, N.J., 2009. Factors influencing the abundance of Trioxys pallidus, a successful introduced biological control agent of walnut aphid in California. Biological control 48, 22–29. Kakehashi, N., Suzuki, Y., Iwasa, Y., 1984. Niche overlap of parasitoids in host parasitoid systems – its consequence to single versus multiple introduction controversy in biological-control. Journal of Applied Ecology 21, 115–131. Kremen, C., 2005. Managing ecosystem services: what do we need to know about their ecology? Ecology Letters 8, 468–479. Letourneau, D.K., Jedlicka, J.A., Bothwell, S.G., Moreno, C.R., 2009. Effects of natural enemy biodiversity on the suppression of arthropod herbivores in terrestrial ecosystems. Annual Review of Ecology Evolution and Systematics 40, 573–592. Noyes, J.S., 2001. Interactive Catalogue of World Chalcidoidea 2001. Taxapad. CDROM, Canada. Proper, A.B., 1934. Hyperparasitism in the case of some introduced Lepidopterous tree defoliators. Journal of Agricultural Research 48, 359–376. R Development Core Team, 2008. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna. Rivas-Martínez, S., Díaz, T.E., Fernánez-González, F., Izco, J., Loidi, J., Lousã, M., 2002. Vascular plant communities of Spain and Portugal. Itinera Geobotanica 15, 5– 432. Rosenheim, J.A., 1998. Higher-order predators and the regulation of insect herbivore populations. Annual Review of Entomology 43, 421–447.
E. Frago et al. / Biological Control 60 (2012) 123–131 Rosenheim, J.A., 2005. Intraguild predation of Orius tristicolor by Geocoris spp. and the paradox of irruptive spider mite dynamics in California cotton. Biological Control 32, 172–179. Rosenheim, J.A., 2007. Intraguild predation: New theoretical and empirical perspectives. Ecology 88, 2679–2680. Schaefer, P.W., 1974. Population ecology of the browntail moth, Euproctis chrysorrhoea (Lepidoptera: Lymantriidae). University of Maine, Orono, Maine, USA. Schmitz, O.J., 2007. Predator diversity and trophic interactions. Ecology 88, 2415– 2426. Sterling, P.H., Speight, M.R., 1989. Comparative Mortalities of the Brown-Tail Moth, Euproctis chrysorrhoea (L) (Lepidoptera, Lymantriidae), in Southeast England. Botanical Journal of the Linnean Society 101, 69–78. Torossian, C., Torossian, F., Roques, L., 1988. Le bombyx cul brun: Euproctis chrysorrhoea, (1) Cycle biologique-ecologie-nuisibilite. Bulletin de la Societe d’Histoire Naturelle de Toulouse 124, 127–174.
131
Tscharntke, T., Klein, A.M., Kruess, A., Steffan-Dewenter, I., Thies, C., 2005. Landscape perspectives on agricultural intensification and biodiversity – ecosystem service management. Ecology Letters 8, 857–874. Generalitat Valenciana (2008). La salud de nuestros bosques, resultados de la prospección de la Conselleria de Medi Ambient, Aigua, Urbanisme i Habitatge. Dirección General de Gestión del Medio Natural. Werner, E.E., Peacor, S.D., 2003. A review of trait-mediated indirect interactions in ecological communities. Ecology 84, 1083–1100. Wilby, A., Thomas, M.B., 2002. Natural enemy diversity and pest control: patterns of pest emergence with agricultural intensification. Ecology Letters 5, 353–360. Yu, D.S., van Achterberg, K., Horstmann, K., 2004. World Ichneumonoidea 2004: Taxonomy, Biology, Morphology and distribution. Taxapad. CD-ROM.
