Incorporation of Intraguild Predation Into a Pest Management Decision-Making Tool: The Case of Thrips and Two Pollen-Feeding Predators in Strawberry

BIOLOGICAL AND MICROBIAL CONTROL Incorporation of Intraguild Predation Into a Pest Management Decision-Making Tool: The Case of Thrips and Two Pollen-Feeding Predators in Strawberry SULOCHANA SHAKYA, 1,2 MOSHE COLL, 1 AND PHYLLIS G. WEINTRAUB 3 J. Econ. Entomol. 103(4): 1086Ð1093 (2010); DOI: 10.1603/EC09373 ABSTRACT Action thresholds are traditionally based on the density of pests and the economic damage they cause to crops. Pest damage assessments are usually made in a sterile environment, devoid of extenuating factors such as predators, parasitoids, and alternative food sources. Recently, the effects of a predator or parasitoid species have been considered. However, interactions between natural enemy species (intraguild predation and interference), which are common in agricultural Þelds, have not been incorporated yet into decision-making tools. We conducted a series of leaf disc and potted plant trials to evaluate the effects of two predator species, the anthocorid Orius laevigatus (Fieber) and the phytoseiid Neoseiulus cucumeris (Oudemans) on the density of and fruit damage inßicted by western ßower thrips, Frankliniella occidentalis (Pergande). We then used the obtained results to develop a pest management decision-making tool for the control of western ßower thrips. Because strawberries (Fragaria ananassa Duchesne) ßower in cycles, pollen, a food source for both predators and the pest, is periodically available in the system and has also been incorporated in our decision-making tool. The developed new management tool would allow the relaxation of the economic threshold (ET) for western ßower thrips in strawberry ßowers. The presence of an average of a single O. laevigatus per ßower for example, may allow that relaxation of the ET by 40% (from 10 to 14 western ßower thrips per ßower) when pollen is available during the winter. Because Þeld monitoring shows that O. laevigatus populations in Israeli strawberry often reach mean densities of three to four per ßower, the new approach promises to drastically reduce the employment of toxic insecticides. KEY WORDS intraguild predation, pest management tools, predators, pollen, western ßower thrips 1 Department of Entomology, The Hebrew University of Jerusalem, P.O. Box 12, Rehovot 76100, Israel. 2 Current address: Himalayan College of Agricultural Sciences and Technology (HICAST), Purbanchal University, P.O. Box 13233, Kathmandu, Nepal. 3 Corresponding author: Department of Entomology, Agricultural Research Organization, Gilat Research Center, D.N. Negev 85280, Israel (e-mail: phyllisw@volcani.agri.gov.il). Traditionally, pest control recommendations have been based solely on estimates of pest density. In such pest management programs, results of pest monitoring in the Þeld are compared with established economic thresholds (ETs), and control measures are recommended to prevent pest populations from reaching economically damaging levels. More recently, a few studies have also considered the density of a pestõs natural enemy in the Þeld, thus allowing the relaxation of the ET (Brown 1997, Conway et al. 2006, Musser et al. 2006, Zhang and Swinton 2009). Yet, natural enemies rarely operate alone; they are often part of a suite or guild of predators and parasitoids that act simultaneously to reduce pest infestation. This guild may act additively or even synergistically, when one enemy facilitates the activity of another (e.g., Losey and Denno 1998). However, antagonistic interactions between natural enemies and intraguild predation (IGP) in particular, are common and may actually hamper herbivore suppression (Stiling 1993; Rosenheim et al. 1993, 1995). Intraguild (IG) predators and parasitoids have been shown to have a signiþcant effect on each other as well as on pest populations in many laboratory and Þeld experiments (see Colfer et al. 2003, and references therein; Hougardy and Mills, 2009). Yet, to date, the simultaneous activity of several natural enemy species have not been incorporated in the calculation of action thresholds and pest management decision-making tools. The overall objective of this study was to develop a decision making tool for the control of western ßower thrips, Frankliniella occidentalis (Pergande), in strawberry (Fragaria ananassa Duchesne), which incorporates the activities of two predatory species, the anthocorid Orius laevigatus (Fieber) and the phytoseiid Neoseiulus cucumeris (Oudemans), and the IG interactions between them. The three arthropod species in the system also are known to feed on pollen (Coll 1998, van Rijn and Sabelis 1990, Skirvin et al. 2007, Wackers et al. 2007). Because strawberry plants have three to four ßowering cycles during the growing season, we also incorporated pollen availability in our 0022-0493/10/1086Ð1093$04.00/0 2010 Entomological Society of America

August 2010 SHAKYA ET AL.: IGP AND PEST MANAGEMENT DECISIONS 1087 model. The majority of the crop in Israel ( 400 ha total) is produced under low tunnels on the coastal plain between September and May. The tunnel cover is opened during most of the days, to reduce humidity buildup and the associated fungi infestation. Israeli growers usually monitor western ßower thrips in the ßowers (Coll et al. 2007) because (i) there is a strong correlation between thrips count in ßowers and fruit damage, and (ii) thrips occurrence in ßowers provides early warning for damage to young fruit (Steiner and Goodwin 2005; Coll et al. 2007). This is the Þrst time the incorporation of IG predatory interactions into a decision-making tool has been proposed for the management of an agricultural pest. The new approach promises to more accurately predict enemy impact on pest populations and thus increase reliability of the decision-making tool. This in turn would act to increase the adoption of these tools by growers, further reducing the employment of pesticides that hamper biological control and harm human health and the environment. Materials and Methods Plants. Strawberry plants (Ô328Õ) were transplanted in a mixture of vermiculite and potting soil, provided with slow release fertilizer (Osmocote Plus, 15N-8P- 11K tablets; Scotts Horticulture Products, Marysville, OH) and drip irrigation lines in the greenhouse. The plants were grown under natural light at 20 3 C. The ßowers were hand pollinated using a camelõs hair paint brush. To prevent spontaneous infestation of powdery mildew, plants were sprayed with Stroby 50 WP (active ingredient [AI] kresoxin-methyl, 0.03%; BASF, Ludwigshafen, Germany), Heliogofrit (AI sulfur, 0.02%; Action PIN, Dax, France), OÞr 2000 EW (AI penconazole, 0.05%; Syngenta, Wilmington, DE), and Triforine (AI piperazine with emulsiþers; American Cyanamid Company, Wayne, NJ) on an alternate basis according to label instructions. Mature, senescing leaves were removed occasionally. Western Flower Thrips. Field-collected western ßower thrips were reared in 1-liter glass jars and provided with Phaseolus vulgaris L. seedlings and pods (Ô4095Õ, Ben Shahar Co., Israel). The jars were kept in a growth chamber at 25 1 C and a photoperiod of 16:8 (L:D) h. Pods and seedlings were replaced every 3Ð4 d. Upon removal, old pods and seedlings were transferred to new rearing jars containing a fresh bean pod and seedling. Predators. O. laevigatus was cultured in the laboratory from females obtained from Bio-Bee Biological Systems (Kibbutz Sde Elyahu, Israel). Cultures were held in growth cabinets under 25 1 C, 70 10% RH, and a photoperiod of 16:8 (L:D) h and fed eggs of Ephestia kuehniella Zeller (Lepidoptera: Pyralidae). Bean pods were provided for moisture, to supplement predatorsõ diet, and to serve as an oviposition substrate. Ephestia eggs were obtained from a laboratory culture, and bean pods were collected from potted plants grown in the greenhouse. Table 1. Treatment combinations used to test the effect of pollen and two predators a on western flower thrips suppression and subsequent reduction in fruit damage Independent Treatment b variable 1 2 3 4 5 6 7 8 Pollen N. cucumeris (60 adults) O. laevigatus (3 adult females) a Predators were released on all plant organs. b indicates with and Ð indicates without the independent variables listed. N. cucumeris was obtained from Bio-Bee and reared on mold mites, Tyrophagus putrescentiae (Schrank), infested wheat bran. Before trials commenced, N. cucumeris were acclimated for several generations on strawberry pollen. Preliminary experiments indicated that the survival, development, and reproduction of these mites on strawberry pollen are approximately the same as that on thrips prey. Influence of Pollen Availability and Predator Activity on Thrips Population and Thrips-Inflicted Fruit Damage. The trial was conducted in a walk-in rearing room at 25 1 C, 60Ð70% RH, and a photoperiod of 13:11 (L:D) h. Potted strawberry plants (three leaves, three ßowers, and three whitish-pinkish fruit) were held individually in screened cages (0.4 by 0.4 by 0.4 m) and watered manually daily. Eight treatment combinations (the presence of pollen, N. cucumeris and O. laevigatus) were established and replicated four times in this nonchoice experiment (Table 1). The nonchoice setting, either with or without pollen, mimics the situation in the Þeld; each Þeld is either ßowering or not ßowering. In treatments without pollen, anthers were removed from all the ßowers before pollen maturation. On day 1, cohorts of 15 adult female western ßower thrips (3Ð4 d after eclosion) were introduced onto each caged plant. Thrips were allowed to oviposit for 6 d before the predators were released. Because these female thrips were expected to oviposit daily at 25 C (Whittaker and Kirk 2004), and taking into account a 3-d incubation period of the eggs, a continuous supply of immature thrips is most likely to have been produced over the duration of the trial. On day 6, three adult O. laevigatus and 60 adult N. cucumeris ( 0.3 and 6.6 bugs and mites per plant part, respectively) were released in each cage. All the releases were made 1Ð2 h before scotophase, to allow predators to acclimate. On day 10, the level of western ßower thrips-inßicted damage to the matured fruit was scored as follows: 0, no damage; 1, light spotting, slight browning of calyx; 2, 25% surface damage (bronzing, punctures); 3, 50% surface damage (bronzing, punctures, surface russeting); 4, 75% surface damage (bronzing, punctures, surface russeting, sucking parts around the achenes); 5, severe damage. During scotophase of that day (day 10), all plant parts were examined for O. laevigatus, then detached and separately washed in 70% ethyl

1088 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 103, no. 4 alcohol to remove all arthropods that were subsequently counted under a stereomicroscope. Data were analyzed by a three-way factorial analysis of variance (ANOVA) to test for interactive effects of predators and pollen. To homogenize error variance, data were square root transformed before analysis. Means were compared by TukeyÐKramer ( 0.05), if ANOVA results were signiþcant. Adjustment of ET. Thresholds for western ßower thrips have been developed for Israeli strawberries produced for the local and export markets (Coll et al. 2007). These thresholds were 24 and 10 western ßower thrips per ßower, respectively. To relax these ET, we tested predation rate of western ßower thrips by O. laevigatus and N. cucumeris when acting alone, in the presence and absence of pollen. The combined effect of the two predators on western ßower thrips was tested in a companion study (Shakya et al. 2009). Results indicated that in 24 h O. laevigatus preyed upon 3.41 0.16 N. cucumeris in the presence of pollen and 4.06 0.18 mites in its absence. These values together with results included in the present report, were used to adjust the ET used for western ßower thrips management. To quantify predation of western ßower thrips by the two predators, we conducted leaf disc experiments in agar-þlled ventilated experimental arenas (5 cm in diameter by 3 cm in height) in which a 4-cm-diameter strawberry leaf disc was placed bottom side up on 1% agar just before solidiþcation. All leaf disc experiments were conducted in a growth chamber at 25 1 C, 60 5% RH, and a photoperiod of 16:8 (L:D) h. Live and dead predators and prey were counted after 24 h, and all trials were repeated six times unless otherwise stated. We used Typha sp. pollen in the experiment because we wanted to be conservative in our assessment of the effect of pollen availability on pest suppression by the two predators; Typha pollen should have the strongest negative impact on pest suppression because it is known to be superior pollen type for many omnivorous predators (Shakya et al., 2009; Weintraub et al. 2007), Predation of Western Flower Thrips Adults by O. laevigatus Alone. We used 2Ð3-d-old O. laevigatus adult females that were starved (i.e., offered only moist cotton balls) in isolation for 24 h. Each individual was offered 5, 10, 15, or, 20 western ßower thrips adult, and the number of western ßower thrips consumed was recorded 24 h later. These functional response data were subjected to one-way ANOVA, with block over time effect and prey density as a source of variation in the model (JMP, version 6; Sall et al., 2001). When ANOVA results were signiþcant, means were compared using StudentÕs t-test ( 0.05). Effect of Pollen and Prey Stage on Larval Western Flower Thrips Predation by O. laevigatus. Each starved adult female O. laevigatus was offered 25 Þrst-instar or 25 second-instar western ßower thrips, with or without 1mgofTypha sp. pollen. Live and dead predators and prey were counted after 24 h. The experiment was replicated six times and the number of consumed western ßower thrips larvae was subjected to two-way ANOVA, with pollen at a whole plot level and prey stage at the subplot level (JMP, version 6; Sall et al., 2001). When ANOVA indicated signiþcant main effects, means were compared using StudentÕs t-test ( 0.05). Effect of Pollen on Predation of Immature Western Flower Thrips by N. cucumeris. A single starved predatory mite was offered six Þrst instar western ßower thrips with or without 1 mgoftypha sp. pollen. Predatory mites were held singly without food in the experimental arena for 24 h before adding prey/pollen. The number of prey killed by the predator was recorded 24 h later and the experiment was replicated 10 times. Immature western ßower thrips fed upon by N. cucumeris are distinctive and can easily be differentiated from larval thrips that died from other causes. The number of consumed prey were subjected to one-way ANOVA (JMP, version 6; Sall et al., 2001). Results Influence of Pollen Availability and Predator Activity on Thrips Population and Thrips-Inflicted Fruit Damage. Pollen availability and the two predators affected the number of immature thrips on strawberry plants (pollen: F 1,24 8.81, P 0.007; Orius mite: F 1,24 12.45, P 0.002). No adult thrips were found, probably because they were consumed by O. laevigatus or perished naturally. The three factors did not interact signiþcantly in their effects on western ßower thrips (three-way interactions: F 1,24 0.81, P 0.38). Likewise, no signiþcant differences were found in the combined effects of pollen with each of the predators (pollen Orius: F 1,24 0.11, P 0.74; pollen mites: F 1,24 0.87, P 0.36). Yet, western ßower thrips populations were 2.5-fold higher on plants with pollen and no predators than on plants without pollen but in the presence of predatory mites (Fig. 1). Also, western ßower thrips distribution on the plants was affected signiþcantly by the presence of pollen and predators (Fig. 2). In the absence of predators, 31% of western ßower thrips larvae were found on the fruit in the pollenavailable treatment compared with 66% when pollen was absent (Fig. 2). Relatively few western ßower thrips were found on the fruit in the presence of either or both predators, irrespectively of pollen availability (0.1Ð 0.7 thrips per fruit; Fig. 2). The degree of western ßower thrips-inßicted fruit damage differed signiþcantly among treatments but no signiþcant three-way interactions were detected in the effects of pollen, O. laevigatus, and N. cucumeris (F 1,24 0.30; P 0.59). There were interactive effects of O. laevigatus and N. cucumeris on fruit damage (F 1,24 21.44; P 0.001). Likewise, there were signiþcant interactions in the effects of pollen and O. laevigatus on fruit damage (F 1,24 8.99; P 0.006). Yet, pollen and N. cucumeris did not interact in their effect on damage (F 1, 24 1.86; P 0.18). Overall, fruit damage correlated signiþcantly with western ßower thrips larval density per plant at end of the experiment (Y 4.29X; R 2 0.73, P 0.001), with less damage

August 2010 SHAKYA ET AL.: IGP AND PEST MANAGEMENT DECISIONS 1089 Fig. 1. Mean number of western ßower thrips (WFT) larvae (bars SE) recorded on strawberry plants, and thripsinßicted damage to strawberry fruit (dashed line) in different treatment combinations: 0, no damage; 1, light spotting, slight browning of calyx; 2, 25% surface damage (bronzing, punctures); 3, 50% surface damage (bronzing, punctures, surface russeting); 4, 75% surface damage (bronzing, punctures, surface russeting, sucking parts around the achenes); and 5, severe damage. recorded on the fruit of plants with predators, particularly on plants without pollen (Fig. 1). Adjustment of ET. Predation of Western Flower Thrips Adults by O. laevigatus. Predation on western ßower thrips by female O. laevigatus differed signiþcantly among prey densities (F 3,20 37.87; P 0.0001): this functional response increased linearly up to a density of 10 thrips, after which it remained stable at 12 western ßower thrips per predator in 24 h (Fig. 3). Effect of Pollen and Larval Stage on Western Flower Thrips Predation by O. laevigatus. Prey stage and pollen availability interacted in their effects on the number of prey consumed by the predator (F 1,10 48.26; P 0.0001) (Fig. 4). Predation on Þrst instar western ßower thrips was signiþcantly higher in the absence of pollen than in its presence. A similar effect was not detected for the second instar prey. Effect of Pollen on Predation of Immature Western Flower Thrips by N. cucumeris. In the presence of pollen, signiþcantly fewer Þrst-instar western ßower thrips were preyed upon in 24 h by N. cucumeris adults than when pollen was absent (F 1,36 25.77; P 0.0001; 2.0 0.4 and 3.5 0.7 western ßower thrips per 24 h, respectively). Adjustment of ETs. These results were used to modify the ET developed for western ßower thrips in strawberry (Coll et al. 2007). We took a conservative approach and used only 30Ð50% of the predation levels recorded in laboratory experiments (Table 2). For example, results indicate that an O. laevigatus female consumes 15Ð19 and 12Ð24 west- Fig. 2. Mean number of western ßower thrips (WFT) larvae per plant part ( SE) recorded on strawberry fruit, leaves and ßowers in all plants, in the presence and absence of pollen, O. laevigatus, and N. cucumeris.

1090 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 103, no. 4 Table 2. Number of western flower thrips killed by O. laevigatus and N. cucumeris in 24 h, which were used to adjust the economic threshold in a pest management decision-making tool a Western ßower thrips predator No. western ßower thrips killed in the presence of pollen in the absence of pollen 1 O. laevigatus 4 6 1 N. cucumeris 1 2 1 O. laevigatus 1 N. cucumeris 4 6 a Data are from this study and from Shakya et al. (2009). Fig. 3. Functional response of O. laevigatus females to density of western ßower thrips prey, showing the mean number of prey ( SE) consumed by a predator in 24 h (n 6). ern ßower thrips per day in the presence and absence of pollen, respectively (Figs. 3 and 4). Accordingly, we modiþed the ET by using predation rates of four and six western ßower thrips per predator in the presence and absence of pollen, respectively (Table 2). Likewise, results indicate that N. cucumeris consumes 2 and 3.5 western ßower thrips per day in the presence and absence of pollen, respectively (see data in previous section). We therefore corrected the ET using values of one and two western ßower thrips per day (Table 2). Finally, results presented here and in Shakya et al. (2009) indicate that, in the presence of O. laevigatus, N. cucumeris does not inßict additional mortality on western ßower thrips. We therefore corrected the ET by using the values used for O. laevigatus alone, also for cases where both predators act together (Table 2). We developed a thrips management tool in the shape of a decision ßow-chart (Fig. 5). Parameterization also took into account fruit damage under different conditions (predator association and pollen availability; Fig. 1) and predator distribution pattern on different plant parts (i.e., proportion of predator population in ßowers; Shakya et al. 2009). The inputs needed for the adjustment of the ET are therefore the number of O. laevitatus and N. cucumeris predators per ßower, and the availability of pollen (ßowering or nonßowering plants). These values are to be collected Fig. 4. Mean number of immature western ßower thrips (WFT SE) consumed in 24 h by an adult O. laevigatus in the presence and absence of Typha pollen. weekly, during western ßower thrips monitoring (see Coll et al. 2007). Recommendations for western ßower thrips control measures are then made by comparing thrips density, as determined by ßower inspections, and the adjusted ET (i.e., the output of ßow chart in Fig. 5). Discussion The key pest in strawberries in many parts of the world is western ßower thrips. In Israel, naturally occurring Orius spp. invade pesticide-free Þelds and usually keep western ßower thrips below damaging levels (Coll et al. 2005, 2007). O. laevigatus and N. cucumeris are known to be effective predators of western ßower thrips on various crops when released separately. Two recent studies explored the ability of these two generalist predators to reduce western ßower thrips populations when released together (Skirvin et al. 2007, Shakya et al. 2009). Yet, the importance of these predators in reducing western ßower thrips-inßicted yield losses is being reported here for the Þrst time. In addition, we propose how the activities of these two predators could be incorporated to relax (increase) the ET of western ßower thrips in strawberry. A similar approach could be used to manage other pests in strawberry and other cropping systems. A few studies have explored the western ßower thrips predation capacity of adult O. laevigatus or N. cucumeris. Cocuzza et al. (1997) reported that O. laevigatus consumes approximately seven adult western ßower thrips in 24 h, and Montserrat et al. (2000) reported a satiation level of 9 s instar western ßower thrips in 6 h. Our Þndings are in general agreement with these reports (satiation at 11Ð12 western ßower thrips per d). N. cucumeris has been reported to prey on Þrst-instar F. occidentalis (Cloutier and Johnson 1993) and consume from 3.5 (current study) to 8Ð9 thrips per d (Wright and Williams 1999). Finally, O. laevigatus was reported to prey on N. cucumeris (Wittmann and Leather 1997, Shakya et al. 2009). Some studies suggested that pollen availability may act to retain predators in crop Þelds (van Rijn and Sabelis 1990, Coll and Bottrell 1991). During the winter and spring, strawberries go through several cycles of ßowering and fruit production; so, naturally pollen is alternately available then unavailable.

August 2010 SHAKYA ET AL.: IGP AND PEST MANAGEMENT DECISIONS 1091 Fig. 5. Adjustment of ET for western ßower thrips (WFT) in strawberry according to the presence of pollen and thrips predators. Number of thrips and predators (Orius bugs [X], N. cucumeris mites [Y]) per ßower are determined through Þeld sampling. In the absence of predators, the ET for export and local markets are set at 10 and 25 western ßower thrips/ßower (Coll et al. 2007). Because O. laevigatus, N. cucumeris, and western ßower thrips feed and develop on pollen, we hypothesized that the presence of pollen may reduce the level of IGP and predation on thrips by both predators possibly leading to an increased level of thrips-inßicted damage. Shakya et al. (2009) showed that more thrips were found in strawberry ßowers bearing pollen than in those without pollen, regardless of the presence of predators. In the absence of pollen, thrips were recorded primarily on fruit. O. laevigatus primarily inhabits the ßowers when pollen is present, but the leaves and fruit in its absence. When pollen is present, N. cucumeris is found in the ßowers only when O. laevigatus is absent; the mite is an intraguild prey for O. laevigatus. The mites mostly inhabit the fruit or leaves in the presence of O. laevigatus or when pollen is absent. Similarly in greenhouse, Weintraub et al. (2004) reported that in the absence of O. laevigatus, N. cucumeris is found mainly in pepper ßowers. Fitzgerald et al. (2008) reported that 4 wk after being released in the greenhouse and Þeld, signiþcantly more N. cucumeris eggs, immatures and adults were found in strawberry ßowers and fruit clusters than on any other part of the plant. These results and the current study suggest therefore that western ßower thrips and both predators primarily occupy the ßowers when pollen is available. In the absence of predators, western ßower thrips caused signiþcantly higher fruit damage when pollen was absent than in its presence. The reverse was true when both predators were present. These results strongly suggest that the two predators feed on pollen when present and are thus less efþcient in reducing thrips numbers. Results indicate that N. cucumeris is able to reduce fruit damage irrespective of the presence of pollen; however, when O. laevigatus was the only predator, there was less fruit damage in the absence of pollen. It can be concluded that in spite of O. laevigatus feeding on N. cucumeris, thrips and pollen, both predators act to substantially suppress F. occidentalis populations and reduce western ßower thripsinßicted fruit damage. Most agricultural systems have a diverse assemblage of pests, predators, and parasitoids with a multitude of major and minor interactions occurring between consumers at several trophic levels. These complex interactions make it difþcult to establish realistic economic injury levels (EILs). As a result, pest densities alone were traditionally used for recommending control measures. Peterson and Hunt (2003) attempted to develop a probabilistic EIL to account for variability in a system. To date, however, most decision-making tools do not take into account the role natural enemies have in suppressing pest populations (but see Giles et al. 2003; Naranjo and Ellsworth, 2009, and references therein), and antagonistic interactions among natural enemies have been ignored all together in determination of action levels. Results indicate that predator activity may allow us to dramatically relax the ET for western ßower thrips in strawberry ßowers. The presence of an average of a single O. laevigatus per ßower for example, may allow that relaxation of the ET by 40% (from 10 to 14 western ßower thrips per ßower) when pollen is available during the winter (export market). Field monitoring shows that O. laevigatus populations in Israeli strawberry Þelds often reach mean densities of three to four per ßower (Shouster 2003) and that western ßower thrips populations rarely exceeded the calculated ET (Coll et al. 2007). This study shows, for the Þrst time, how consideration of pollen availability and the simultaneous activity of two predators could modify pest man-

1092 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 103, no. 4 agement decisions that may drastically reduce the use of costly and harmful insecticides. Acknowledgments The study was supported in part by CRAFT Project QLK5-CT-2001-70484 (BIOPROTECT) from the European Community (to M.C.). References Cited Brown, G. C. 1997. Simple models of natural enemy action and economic thresholds. Am. Entomol. 43: 117Ð124. Cloutier, C., and G. S. Johnson. 1993. Interaction between life stages in a phytoseiid predator: western ßower thrips prey killed by adults as food for protonymphs of Amblyseius cucumeris. Exp. Appl. Acarol. 17: 441Ð449. Cocuzza, G. E., P. de Clercq, S. Lizzio, M. van de Veire, L. Tirry, D. Degheele, and V. Vacante. 1997. Life tables and predation activity of Orius laevigatus and O. albidipennis at three constant temperatures. Entomol. Exp. Appl. 85: 189Ð198. Colfer, R. G., J. A. Rosenheim, L. D. Godfrey, and C. L. Hsu. 2003. Interactions between the augmentatively released predaceous mite Galendromus occidentalis (Acari: Phytoseiidae) and naturally occurring generalist predators. Environ. Entomol. 32: 840Ð852. Coll, M. 1998. Living and feeding on plants in predatory Heteroptera, pp. 89Ð129. In M. Coll and J. R. Ruberson [eds.], Predatory Heteroptera: their ecology and use in biological control. Entomological Society of America, Lanham, MD. Coll, M., and D. G. Bottrell. 1991. Microhabitat and resource selection of the European corn borer (Lepidoptera: Pyralidae) and its natural enemies in Þeld corn. Environ. Entomol. 20: 526Ð533. Coll, M., I. Shouster, and S. Steinberg. 2005. Removal of a predatory bug from a biological control package facilitated an augmentative program in Israeli strawberry. In Proceedings of the 2nd International Congress of Biological Control of Arthropods, 11Ð16 September 2005, Davos, Switzerland. Publ. FHTET-2005-08, USDA Forest Service, Morgantown, WV. Coll, M., S. Shakaya, I. Shouster, Y. Nenner, and S. Steinberg. 2007. Decision-making tools for Frankliniella occidentalis management in strawberry: consideration of target markets. Entomol. Exp. Appl. 122: 59Ð67. Conway, H. E., D. C. Steinkraus, J. R. Ruberson, and T. J. Kring. 2006. Experimental treatment threshold for the cotton aphid (Homoptera: Aphididae) using natural enemies in Arkansas cotton. J. Entomol. Sci. 41: 361Ð373. Fitzgerald, J., X. Xiangming, N. Pepper, M. Easterbrook, and M. Solomon. 2008. The spatial and temporal distribution of predatory and phytophagous mites in Þeld-grown strawberry in the UK. Exp. Appl. Acarol. 44: 293Ð306. Giles, K. L., D. B. Jones, T. A. Royer, N. C. Elliot, and S. D. Kindler. 2003. Development of a sampling plan in winter wheat that estimates cereal aphid parasitism levels and predicts population suppression. J. Econ. Entomol. 96: 975Ð982. Hougardy, E., and N. J. Mills. 2009. Factors inßuencing the abundance of Trioxys pallidus, a successful introduced biological control agent of walnut aphid in California. Biol. Control 48: 22Ð29. Losey, J. E., and R. F. Denno. 1998. Positive predator-predator interactions: enhanced predation rates and synergistic suppression of aphid populations. Ecology 79: 2143Ð 2152. Montserrat, M., R. Albajes, and C. Castañé. 2000. Functional response of four heteropteran predators preying on greenhouse whiteßy (Homoptera: Aleyrodidae) and western ßower thrips (Thysanoptera: Thripidae). Environ. Entomol. 29: 1075Ð1082. Musser, F. R., J. P. Nyrop, and A. M. Shelton. 2006. Integrating biological and chemical controls in decision making: European corn borer (Lepidoptera: Crambidae) control in sweet corn as an example. J. Econ. Entomol. 99: 1538Ð1349. Naranjo, S. E., and P. C. Ellsworth. 2009. Fifty years of the integrated control concept: moving the model and implementation forward in Arizona. Pest Manag. Sci. 65: 1267Ð1286. Peterson, R.K.D., and T. E. Hunt. 2003. The probabilistic economic injury level: incorporating uncertainty into pest management decision-making. J. Econ. Entomol. 96: 536Ð542. Rosenheim, J. A., L. R. Wilhoit, and C. A. Armer. 1993. Inßuence of intraguild predation among generalist insect predators on the suppression of an herbivore population. Oecologia 96: 439Ð449. Rosenheim, J. A., H. K. Kaya, L. E. Ehler, J. J. Marois, and B. A. Jaffee. 1995. Intraguild predation among biological-control agents: theory and evidence. Biol. Control 5: 303Ð335. Sall, J., A. Lehman, and L. Crighton. 2001. JMP start statistics: a guide to statistics and data analysis using JMP & JMP IN software, 2nd ed. Duxbury, Thomson Learning, PaciÞc Grove, CA. Shakya, S., P. G. Weintraub, and M. Coll. 2009. Effect of pollen supplement on intraguild predatory interactions between two omnivores: the importance of spatial dynamics. Biol. Control 50: 281Ð287. Shouster, I. 2003. Ecological and agricultural implications of tritrophic level interactions between strawberry plants, the western ßower thrips Frankliniella occidentalis, and the predatory bug Orius laevigatus, M.S. thesis, The Hebrew University of Jerusalem, Jerusalem, Israel. Skirvin, D. J., L. Kravar-Garde, K. Reynolds, J. Jones, A. Mead, and J. Fenlon. 2007. Supplemental food affects thrips predation and movement of Orius laevigatus (Hemiptera: Anthocoridae) and Neoseiulus cucumeris (Acari: Phytoseiidae). Bull. Entomol. Res. 97: 309Ð315. Steiner, M. Y., and S. Goodwin. 2005. Management of thrips (Thysanoptera: Thripidae) in Australian strawberry crops: within-plant distribution characteristics and action thresholds. Aust. J. Entomol. 44: 175Ð185. Stiling, P. 1993. Why do natural enemies fail in classical biological control programs? Am. Entomol. 39: 31Ð37. van Rijn, P.C.J., and M. W. Sabelis. 1990. Pollen as an alternative food source for predatory mites and its effect on the biological control of thrips in greenhouses. Proc. Exp. Appl. Entomol. 1: 44Ð48. Wackers, F. L., J. Romeis, and P. van Rijn. 2007. Nectar and pollen feeding by insect herbivores and implications for multitrophic interactions. Annu. Rev. Entomol. 52: 301Ð 323. Weintraub, P. G., V. Alchanatis, and E. Palevsky. 2004. Distribution of the predatory mite, Neoseiulus cucumeris, in greenhouse pepper. Acta Hortic. 659: 281Ð285. Weintraub, P. G., S. Kleitman, V. Alchanatis, and E. Palevsly. 2007. Factors affecting the distribution of a predatory mite on greenhouse sweet pepper. Exp. Appl. Acarol. 42: 23Ð35.

August 2010 SHAKYA ET AL.: IGP AND PEST MANAGEMENT DECISIONS 1093 Whittaker, M. S., and W.D.J. Kirk. 2004. The effect of photoperiod on walking, feeding, and oviposition in the western ßower thrips. Entomol. Exp. Appl. 111: 209Ð214. Wittmann, E. J., and S. R. Leather. 1997. Compatibility of Orius laevigatus (Fieber) (Hemiptera: Anthocoridae) with Neoseiulus (Amblyseius) cucumeris (Oudemans) (Acari: Phytoseiidae) and Iphiseius (Amblyseius) degenerans (Berlese) (Acari: Phytoseiidae) in the biocontrol of Frankliniella occidentalis Pergande (Thysanoptera: Thripidae). Exp. Appl. Acarol. 21: 523Ð538. Wright, E., and M.D.C. Williams. 1999. A bioassay technique to determine the functional response of different predators of Frankliniella occidentalis in ornamentals. Bull. IOBC/wprs 22: 287Ð290. Zhang, W., and S. M. Swinton. 2009. Incorporating natural enemies in an economic threshold for dynamically optimal pest management. Ecol. Model. 220: 1315-1324. Received 4 November 2009; accepted 16 April 2010.