Assessing the impact of arthropod natural ... - Wiley Online Library

Insect Science (2015) 22, 20–34, DOI 10.1111/1744-7917.12174

REVIEW

Assessing the impact of arthropod natural enemies on crop pests at the field scale Sarina Macfadyen1 , Andrew P. Davies2 and Myron P. Zalucki3 1 CSIRO Agriculture Flagship, Clunies Ross St, Acton, ACT 2601, Australia; 2 CSIRO, Australian Cotton Research Institute (ACRI), Narrabri,

NSW 2390, Australia and

3 School

of Biological Sciences, University of Queensland, St Lucia, Brisbane, QLD 4072, Australia

Abstract There are many reasons why it is important that we find ways to conserve, and better utilize natural enemies of invertebrate crop pests. Currently, measures of natural enemy impact are rarely incorporated into studies that purport to examine pest control. Most studies examine pest and natural enemy presence and/or abundance and then qualitatively infer impact. While this provides useful data to address a range of ecological questions, a measure of impact is critical for guiding pest management decision-making. Often some very simple techniques can be used to obtain an estimate of natural enemy impact. We present examples of field-based studies that have used cages, barriers to restrict natural enemy or prey movement, direct observation of natural enemy attack, and sentinel prey items to estimate mortality. The measure of natural enemy impact used in each study needs to be tailored to the needs of farmers and the specific pest problems they face. For example, the magnitude of mortality attributed to natural enemies may be less important than the timing and consistency of that mortality between seasons. Tailoring impact assessments will lead to research outcomes that do not simply provide general information about how to conserve natural enemies, but how to use these natural enemies as an integral part of decision-making. Key words mortality; parasitism; pest control; predation; semifield

Introduction The potential benefits of natural enemies, such as predators and parasitoids, for the control of a range of arthropod crop pests have been known for some time. Conservation biocontrol (Cullen et al., 2008; Jonsson et al., 2008) and augmentative biocontrol (Collier & van Steenwyk, 2004; Crowder, 2007) have been the focus of much research interest, and more recently pest suppression by

Correspondence: Sarina Macfadyen, CSIRO Ecosystem Sciences, Clunies Ross St, Acton, ACT 2601, Australia. Tel: +61 2 6246 4432; email: [email protected] † Current address: Saunders Havill Group, Bowen Hills, Brisbane, QLD 4006, Australia.

natural enemies has been appreciated as an important regulating ecosystem service (Cardinale et al., 2012; Holland et al., 2012). Since the advent of integrated pest management (IPM) principles (Kogan, 1998) and their use in agroecosystems, there has been an increase in interest in the use of natural enemies for pest control. IPM encourages insecticides only be applied when necessary with minimal nontarget impact via the use of spray thresholds and selective or “soft” insecticides. In addition, biological control can be achieved by manipulating the landscape through the provision of flowering resources for natural enemies (Landis et al., 2000; Van Driesche et al., 2008; Sigsgaard et al., 2013), establishing source habitats for natural enemies (Schellhorn et al., 2000a, 2008), and augmentative releases (above). The aim is to create an environment where natural enemies thrive to

20 C 2014 Institute of

Zoology, Chinese Academy of Sciences

Impact of arthropod natural enemies

maximize their effects on pest control. Despite this interest the reality is that even today those agricultural industries purporting to have adopted IPM rarely utilize natural enemies (Thomas, 1999; Ehler, 2006; Horne et al., 2008; Zalucki et al., 2009). Furthermore, relatively few research studies examining the role of predators and parasitoids in cropping systems actually measure their impact (see introductory paper in this series). Furlong and Zalucki (2010) found less than half of the studies on lepidopteran pests and their natural enemies they examined adopted methodologies that would allow the impact of natural enemies to be measured and objectively assessed. A meta-analysis by Chaplin-Kramer et al. (2011) on the response of pests and natural enemies to landscape complexity found only 13 of 46 studies included a measure of natural enemy impact. Most studies examine species presence and/or abundance and then qualitatively infer impact. This means it is difficult to make informed pest control decisions incorporating natural enemy activity in the absence of “real” evidence of their impact. For natural enemy impact to be incorporated into IPM it needs to be measured directly. Direct estimation, or “evidence,” of natural enemy impact can provide a tangible quantity applicable to modeling systems aimed at determining the point at which said impact no longer maintains populations of pests below economic damage thresholds. In such situations alternative or remedial methods, such as augmentative control or insecticides might need to be applied. This is the crux of natural enemy importance. Here we first dispel the myth that assessing natural enemy impact is often difficult, time consuming and cost prohibitive and therefore cannot be done. Given careful planning, this need not be the case. We examine the different measures of natural enemy impact currently reported in the scientific literature for predators and parastioids focusing mainly on broad-acre crop studies. Our second aim is to critically assess if and how such measures could be incorporated into decision-making by farmers. In short, we ask, are we measuring the right thing? We conclude by providing suggestions for how research in this field could be improved in the future to ensure that natural enemies are used to their full potential in agricultural landscapes. We acknowledge that much of this content is not new and there are previous articles that have covered some of this ground (e.g., Grant & Shepard, 1985; Luck et al., 1988; van Driesche & Bellows, 1996, see chapter 13; Mills, 1997; Furlong & Zalucki, 2010). However, our message that focusing solely on abundance patterns of pests and natural enemies, limits the ability of our research to solve real-world problems, is one worth repeating (Shackelford et al., 2013).

C 2014

Institute of Zoology, Chinese Academy of Sciences, 22, 20–34

21

Why do so few studies attempt to estimate impact? Figure 1 outlines a series of questions that need to be addressed in order to assess the potential of natural enemy species to control a pest species. Many of these questions require some measure of natural enemy impact in the field to answer. As we have already stated few studies have sought to quantify natural enemy impact in the field. Instead they monitor patterns in the diversity and abundance of natural enemy species across time and relate these to agronomic, climate, or landscape parameters. We are not suggesting this is a not a useful exercise. The first, but not defining step, in considering the potential of a natural enemy species to control a pest population is to determine it can potentially feed on the pest (see Furlong et al., 2015; Fig. 1, Q1) and is present in the crop at the same time as the target pest (Stanley, 1997; Bishop, 1978; Pearce et al., 2004; Fig. 1, Q6). However, we argue that other questions that need to be answered in order to incorporate natural enemies into pest management decision-making require that their impact on pest population growth be quantified (Fig. 1, Q2–Q5). There is a perception that measuring impact is more labor intensive than measuring abundance and species richness. However, there are benefits and challenges associated with each of these approaches and sometimes very simple techniques can be used to get an estimate of impact (see Table 1 and sections below). Widespread insecticide-use makes it challenging (but not impossible) to accurately assess predator impact in the field because results are confounded by insecticide drift into unsprayed control plots and there can be a general reduction in natural enemy abundance across a landscape (due to a combination of direct impacts from the insecticides and indirectly via a reduction in prey). This has been most apparent by the recent widespread of adoption of GM cotton in Australia with a concomitant reduction in pesticide application. As a consequence natural enemies are now generally more abundant (Whitehouse et al., 2005). Trichogramma was (re)-discovered in Northern NSW dispelling the myth that it was absent for bioclimatic reasons (Davies et al., 2011). This example illustrates that it is critical that we use techniques for assessing natural enemy impact in a variety of landscapes, despite these types of challenges. Natural enemies Whilst predators and parasitoids can be lumped together as natural enemies, this ignores the nuances that separates

22

S. Macfadyen et al.

Fig. 1 Schematic illustrating a range of research questions that need to be addressed for any single crop–pest–natural enemy system. Those on the left require a measure of natural enemy impact in order to answer (Q1–Q5), those on the right are related to natural enemy impact (Q6–Q8) or crop management decisions (Q9 and Q10, out of scope of this article). Adapted from questions posed by Luck et al. (1988).

the two life-history strategies and influences the control potential of each. Predators are generally free-living organisms, capable of moving from prey to prey and source to source through each developmental stage, and require multiple prey individuals to complete their life-cycle. On the other hand, parasitoids are entirely dependent on a single host for development and survival during immature stages, and usually only disperse as adults. The impact of parasitoids must therefore be considered in a generational sense, whereas predator impact may be both within and between generations. For these reasons we will address each life strategy separately here.

prey mortality and subsequent abundance, and alternative prey preferences, are largely unknown. Predators are often described as “generalist” feeders that consume a wide range of prey types despite the lack of evidence to support or justify this conclusion. In fact, as new techniques for determining what a predator has recently eaten become more widely available, some species are clearly not as “generalist” as previously thought (Chapman et al., 2013; Furlong et al., 2015). Estimating the potential impact an individual predator species may have on pest populations involves not only understanding how it interacts with the pest species (Fig. 1, Q1–Q5), but also how this interaction changes in the presence of other natural enemy species in a complex community (Fig. 1, Q7). There has been much research done on the potential of predators to disrupt pest suppression through processes such as intraguild predation (Rosenheim et al., 1995; Hemptinne et al., 2012), and interspecific predation on other natural enemies such as parasitoids (Snyder & Ives, 2001). Interestingly, Traugott et al. (2012) found that more than half of the parasitoid DNA detected in the guts of generalist predators stems from direct predation of adult parasitoids. We will not explore these issues here,

Predators Free-living predatory arthropods are thought to be important for controlling pest species across a range of crops. They can reduce pest damage to crops by directly contributing to pest mortality and indirectly by disrupting the feeding activities of pest species. Johnson et al. (2000) record some 123 species of predators in Australian farming systems, however, while all these species are considered to be predators their impact on

C

2014 Institute of Zoology, Chinese Academy of Sciences, 22, 20–34


23

Table 1 The benefits and challenges associated with measuring different components of the ecology and impact of natural enemy species. These methods may be used to address some of the questions highlighted in Fig. 1. Method

Benefits

Challenges

Measuring species richness (Q6)

Identification of a species is often the 1st step towards managing the species May give some indication of the stability of pest control services

Measuring species abundance (Q6)

May give some idea of the magnitude of pest control services Ignores species that have low abundance (but may still have high impact)

Measuring impact– exclusion cage studies (Q2, Q3, Q4) Measuring impact– laboratory feeding studies (Q1, Q2, Q5) Measuring impact– direct mortality (e.g., gut content analysis, parasitism rate) (Q1, Q2, Q8, Q7)

Illustrates the combined impact of natural enemy community

Each sampling technique will not sample each species efficiently Rare species require much sampling to record Some species may not be involved in pest control services Species that are in high abundance may not always be important for providing pest control services More individuals greater levels of pest control Each sampling technique will not give an equivalent estimate of each species abundance Difficult to separate the impact of individual species

Gives an indication of maximum prey consumption ability and preferences Quantifies direct mortality to a pest species

but rather focus on the range of techniques that have been used to measure predator impact on pest numbers in the field via their contribution to vital mortality rates.

Direct impacts Prey mortality as a direct result of feeding by predators can be determined using a variety of methods including cage studies in small arenas such as a Petri dishes (perhaps the least informative) through to large field cages, the use of barriers to restrict predator or prey movement, direct observation of predation in the field (Costamagna & Landis, 2007), sentinel prey items exposed in the field, and by serological or molecular techniques that detect prey items in the guts of predators (Pompanon et al., 2012; Furlong et al., 2015). These methods, and their limitations are described in detail in Luck et al. (1988) (also see Macfadyen et al., 2014) and here we will highlight how they have been used in more recent studies. Cage studies, usually involving introducing both predator and prey into a known area (arena) and determin C 2014


Highly artificial environment. The natural enemy does not have to find prey in a complex environment Initial development costs high for molecular techniques Parasitism rate easy to measure but difficult to interpret

ing mortality after a set period of time are useful for determining the maximum number of pest species consumed under ideal but limited circumstances. These studies provide information on the preferred prey of the predator, potential consumption rates (Chenaux et al., 2011), the effect of search area on predation rates, and enable a quantitative comparison between predator species (Lingren et al., 1968; Lopez et al., 1976; Isenhour & Yeargan, 1981; Propp, 1982; Declercq & Degheele, 1994). The cage environment can range from simple to relatively complex. For example, studies on predation of Helicoverpa spp. using cages range from Petri dishes (Grundy & Maelzer, 2000; Pearce et al., 2004), small 400 mL containers (Horne et al., 2000), individually caged plants (Johnson, 1999), small 1-m squared cages (Dillon et al., 1994; Stanley, 1997), to 2-m square field cages (Titmarsh, 1992). Effective cage studies require an understanding of prey and predator preferences towards oviposition and feeding sites, prey dispersion and predator searching behavior. Cages change the microclimate on the plant surface, which may alter predator behavior (Hand & Keaster, 1967). Furthermore, confining prey and predators is likely to increase

24

S. Macfadyen et al.

predation rates (especially at low densities) as predators do not emigrate and therefore tend to research areas (Luck et al., 1988). Bahar et al. (2012) examined predation rate of H. armigera larvae by green lacewing larvae (Mallada signatus) in a variety of test arenas. Predation rate decreased from plastic cups to open trays to finally whole plants. Interestingly, a lower predation rate was recorded in the smallest containers (Petri dishes) as the lacewing larvae included the sides of the dish in their search, and the dish provided a hiding place for H. armigera larvae. More recently studies have focused on trying to link landscape factors with natural enemy diversity, abundance, and pest control services (reviewed by Chaplin-Kramer et al., 2011). Such studies often include a measure of natural enemy impact. For example, Thies et al. (2011) examined the impact of agricultural intensification on biological control services across Europe. In cereal fields experimental plots consisting of plastic barriers (circle of diameter 1 m, reaching 30 cm above the soil surface and 10 cm below the soil surface) were used to exclude ground-dwelling predators. Vegetationdwelling predators (and parasitoids) were excluded using a wire cage over the experimental plot. A similar sized area in the adjacent open field was used as the control. The impact on aphid densities of predator exclusion was assessed at both flowering and milk-ripening crop growth stages. Aphid populations were 28% higher in the ground predator exclusion treatment and 97% higher in the vegetation-dwelling predator (and parasitoid) exclusion treatment. The use of exclusion cages offers an opportunity for researchers to understand the biological control potential of a complex of predators in the field (Gardiner et al., 2009; Thies et al., 2011; Bennett & Gratton, 2012; Holland et al., 2012; Martin et al., 2013). The design options for these types of exclusion cages are endless and usually custom-made for the study site, and species involved (Fig. 2). Most importantly they do not have to be complex or expensive to develop. Recent cages designs (Fig. 2) bear a remarkable resemblance to those shown in an article published in 1985 by Grant and Shepard (1985) (see photos in Fig. 3). The limitation of these types of designs is that the actions of predator species (and in some cases parasitoids) cannot be disentangled unless a cage can be designed that excludes only one group. Sentinel prey items placed in the field and exposed to predators is a useful method for quantifying the direct impacts of predators. Schellhorn et al. (2000b) used Helicoverpa spp. egg cards or potted plants infested with eggs to determine parasitism and predation. Predation was determined by classifying the egg damage according

to the number of missing eggs (ants), collapsed eggs (true bugs and lacewings, or spiders if a brown stain was present), or partially chewed eggs (beetles) after 24–72 h field exposure. During this time field observations provide direct qualitative evidence of predation. Head et al. (2005) not only compared the composition of arthropod populations on Bt (Bollgard) cotton and non-Bt (conventional) cotton in the United States and examined predation rate as well. Sentinel prey items (H. zea eggs, and 1st instar beet armyworm larvae, Spodoptera exigua) were placed in paired fields and exposed to predators and parasitoids for 24 h. After this time fewer eggs and larvae were recovered from Bt cotton fields than the non-Bt fields. The insecticide sprays applied to the non-Bt cotton fields were used to explain the lower predation rates. Indirect impacts Herbivores utilize a range of techniques for detecting and avoiding predators while feeding on host plants. Often predator avoidance comes at a physiological cost. While you are watching out for predators you are usually not able to feed or reproduce. Predator presence alone will have a number of indirect impacts on herbivores (Johnson et al., 2007) and some of these may lead to reduced herbivory rates (Lima, 1988; Chapman et al., 2013). While many studies examine the impact of antipredator behavior on the herbivores’ ecology, few extend this to examining the impact on plant damage. Lemos et al. (2010) found that the red spider mite (Tetranychus evansi) laid more eggs suspended on their silken web, away from the leaf surface, in response to cues produced by a predatory mite (Phytoseiulus longipes). Laying eggs away from the leaf surface has energetic costs for the mother and potentially may lead to greater mortality of the young as they hatch and have to move to the leaf surface, however no quantification of this flow-on effect was made by Lemos et al. (2010). The response of the pea aphid (Acyrthosiphon pisum) to predator induced alarm pheromone includes dropping from the feeding site onto the ground and walking from the feeding site (Dill et al., 1990). Both strategies incur a cost due to lost feeding opportunity and desiccationinduced mortality is likely if the aphid cannot find a new plant quickly (see Perovic et al., 2008 for similar work on H. armigera). Antipredator activity has been shown to alter the transmission of viruses vectored by aphids. Belliure et al. (2011) found that predators (Syrphidae and a Coccinellidae) increased dispersal of the aphid Myzus persicae, but did not increase virus transmission on bean plants. C



25

Fig. 2 Examples of different exclusion cages used in field studies that assess natural enemy impact. (A) and (B) shows cages surrounding cabbage seedlings used by Martin et al. (2013) in South Korea to exclude flying insects and birds (shown open in B for illustrative purposes). (C) Caged and exposed wheat plant used to assess predation of aphids by Bennett and Gratton (2012). (D) Egg card made of Helicoverpa armigera eggs used to assess predation and parasitism in a soybean crop (Macfadyen unpublished). Photos were kindly supplied by A.B. Bennett and E.A. Martin.

Direct and indirect impacts Field studies that artificially enhance the numbers of predators in a crop, or reduce them in some dramatic way (e.g., use of a selective insecticide) both allow us to quantify both the direct and indirect impact of predators. As this is a complex field situation it is often impossible to partition out the direct mortality effects from the indirect effects (antipredator behaviors, movement away from site). Releases of 10–30 predatory bug nymphs (Anthocoris nemoralis and A. nemorum) per tree in a pear orchard resulted in reductions of 31%–40% in pear psyllid populations (Cacopsylla pyri) (Sigsgaard et al., 2006). In contrast, augmentative release of green lacewing eggs (Chrysoperla rufilabris) in cotton leads to no differences in cotton aphid (Aphis gossyppi) density (Knutson & Tedders, 2002). This was attributed to high mortality of the immature lacewing stages. The use of predators in augmentative biological control programmes in broad-acre crop situations has generally been unsuccessful (Collier

C 2014


& van Steenwyk, 2004) due to the cost of releasing the huge numbers required. However, these “experiments” serve as a good method for quantifying impact, and if a reduction in pest damage results from a flux of natural enemies into a field, then this clearly demonstrates their potential. Other techniques for enhancing populations of predators long-term, such as increased area of noncrop vegetation (Gardiner et al., 2009), or more flowering plant resources, can then be explored. Parasitoids Parasitoids (Hymenoptera and Diptera) are considered important biocontrol agents for a range of pest species around the world. In comparison to predators, their impact is easier to quantify as a researcher can rear a host in the laboratory and record if a parasitoid species emerges. As such, direct estimates of field mortality rates are not difficult to obtain (usually expressed as parasitism rate).

26

S. Macfadyen et al.

There have been attempts to suggest a universal threshold of parasitism rate beyond which control of a pest species in the field is likely. Hawkins and Cornell (1994) found that successful biological control was likely at parasitism rates above 32%, and Plećaˇs et al. (2014) found that parasitism rates of 22%–24% were associated with declines in aphid populations. However, much higher parasitism rates have also been observed in the field. Davies et al. (2009) directly estimated mortality of Helicoverpa eggs in winter-grown Bt-transgenic cotton attributed to Trichogramma wasp activity in tropical northern Australia. Once established via immigration and in the absence of pesticides, these parasitoids were responsible for 98% of Helicoverpa egg mortality during high host density early season when the crop was relatively young and most vulnerable to pest damage. We continue to use Trichogramma examples below as they are the most widely applied biological control agent and a species that illustrates many of the issues associated with determining impact for parasitoid species in general. There are 3 main techniques for estimating the impact of parasitoids in field situations: collecting and rearing naturally occurring hosts (e.g., Furlong et al., 2008; Macfadyen et al., 2009; Lohaus et al., 2013), molecular identification of parasitism in field-collected hosts (Derocles et al., 2012; also see Furlong et al., 2015), or the placement of sentinel hosts in the field for a set exposure period (e.g., Letourneau et al., 2012; Thomson & Hoffmann, 2013). Presence of hosts is of course critical to host–parasitoid interaction experiments. Limited host numbers in the field that make assessment impractical can be avoided by the placement of laboratory-raised hosts, most commonly as sentinel egg cards (Fig. 2) for egg parasitoid studies (e.g., Keller & Lewis, 1985; Glenn & Hoffmann, 1997). Parasitoid activity can then be monitored and experimental procedure followed year-round regardless of host densities. This allows comparisons of standardized parasitism rates across different crop types, at different times throughout the season, and across multiple habitats in agricultural landscapes (e.g., Thomson & Hoffmann, 2013) Despite successful examples of integration of parasitoids into IPM programs difficulties arise when attempting to determine the actual impact of parasitoids on pest population growth. The most common approach is to estimate parasitoid abundance using parasitism rate, assuming high apparent parasitism rate equals high parasitoid abundance and equates to high levels of possible control (e.g., Yu & Byers, 1994; Mailafiya et al., 2010). However, additional factors such as host and parasitoid immigration and emigration, rates of reproduction and other sources of mortality will impact parasitoid abun-

dance. These must be taken into consideration; otherwise parasitism rate alone only gives you a picture of impact relating to certain life stages of both the host and parasitoid, and provides an indication of parasitoid activity only. When combined with information about host abundance, development and survival, adult parasitoid survival, reproductive potential and searching ability, parasitism rate can be used to estimate the control potential of the parasitoid species (van Driesche, 1983) via the measurement of recruitment, stage–frequency analyses and death rate analyses (van Driesche et al., 1991). The interaction between host and parasitoid is thought to be strongly influenced by density dependence (Pareja et al., 2008). However, density dependence, or increasing levels of parasitism with increasing host density, is not universally detectable in the field, which suggests its importance has been overrated (Walter & Zalucki, 1999; Bastos et al., 2010; Rand, 2013). Asynchronous development alone does not explain, in terms of density dependence, cases of Trichogramma inoculation that achieve high parasitism levels concurrent with high pest abundance that ultimately result in limited control and unacceptable crop damage (see Twine & Lloyd, 1982). If unlimited hosts are available (such as occurs at high pest densities), why then do not Trichogramma multiply to similar densities and achieve respectable control potential? Trichogramma activity (adult survival, reproductive potential and searching ability, and the development of immatures) can be limited by environmental constraints such as chemical applications (Varma & Singh, 1987; Campbell et al., 1991), and unsuitable climatic conditions (Orr et al., 1997) or plant traits (Romeis et al., 1999a,b), despite the presence of large numbers of hosts. Host abundance itself may not be the driving force behind the dynamics of a parasitoid population, but rather a reflection of the interaction between the attributes of individuals within that population and the characteristics of the environment (Hengeveld & Walter, 1999). Furthermore, the relative reproductive potential of the host may be greater than that of the parasitoid species and therefore not show any top-down control. Host preference may influence parasitoid activity in crops. Many Trichogramma are regarded as polyphagous (Pinto & Stouthamer, 1994) and so are assumed to attack a number of hosts indiscriminately provided the correct conditions for development are present. Why then at sustained high host abundance do we usually see poor parasitism levels and occasionally the opposite? If a preferred host is abundant and available elsewhere, parasitoids may ignore abundant nonpreferred pest species that are potential hosts within the crop. Similarly, if host abundance in the crop is low, but the species is a preferred host, localized Trichogramma may parasitize available eggs resulting in C



high parasitism rates. Host preference in Trichogramma is difficult to assess in the field but has been measured under laboratory conditions (Pluke & Leibee, 2006; Siqueira et al., 2012). It is possible that host discrimination influences levels of parasitism observed in a target pest species, however we can not yet describe the circumstances in which this will increase or limit pest control services. Certainly for a range of parasitoid species host preference in relation to the plant species on which hosts are present has been shown (Ortiz-Martinez et al., 2013; Sankara et al., 2014).

Novel and underutilized techniques for determining impact We commonly measure the abundance of natural enemies and pests in the field using a variety of techniques but translating those abundance estimates into some measure of impact is rarely attempted. Impact is simply assumed (i.e., low abundance equates to negligible impact, high abundance means more impact). However, there are novel ways for elucidating food webs and potential trophic interactions from simple count data. Bohan et al. (2011) used an automated machine learning approach to generate plausible and testable food webs from sample data collected in arable fields across Great Britain. Forty five invertebrate species were hypothesized to be linked due to their cooccurrence patterns and it is thought that those links reflect trophic connections (also see Tamaddoni-Nezhad et al., 2013). Bell et al. (2010) used within-field spatial co-occurrence networks to predict feeding behavior of predators. They found a negative relationship between decreasing predator–prey co-occurrence in a wheat field and the number of prey positives in the guts of those predators. If a consistent and predictable link can be made between abundance estimates and prey consumption by predators this provides more rigor to the use of abundance estimates in decision-making by farmers. If we can predict with confidence that a certain predator community composition is capable of suppressing a certain pest species then measurements of community composition may be used as a proxy for natural enemy impact (Fig. 1, Q6). However this will not be possible until we have a more comprehensive understanding of this relationship, and abiotic factors (such as temperature and rainfall, Fig. 1, Q8) that alter this relationship. The interactions between natural enemies and prey can be modeled at a variety of scales and can provide useful insights into the conditions necessary for pest suppression and the characteristics of natural enemies required for high impact (Bianchi et al., 2009). To date, very few C 2014


27

models have been implemented that specifically explore the impact of natural enemies. Bianchi et al. (2010) developed a spatially explicit simulation model with pest populations that grew in suitable patches and predators that dispersed across the hypothetical landscape. They found that the spatial arrangement of source habitats for predators had profound effects on their potential to colonize crop fields and suppress pest populations. More detailed models that include a comparison of natural enemy impact and pesticide-use (Bianchi et al., 2013), and the behavioral responses of natural enemies to pesticide exposure (Banks & Stark, 2011), offer a new method for exploring how natural enemies can be most effectively integrated into IPM programmes. The next step is to validate these models by examining how well they reflect what occurs in commercial fields.

Are we measuring the right thing? Many researchers over the years have suggested that until we know more about the impact of natural enemies they cannot be effectively incorporated into management programs. Fitt (1989) noted that “until the efficacy of beneficials (natural enemies) is quantified, their potential is unlikely to be utilized efficiently.” Strickland et al. (1996) suggests that one of the key focuses of IPM in the future should be maximizing the use of natural enemies. These are just a few examples of many references that argue for the need to better conserve and understand our natural enemy communities. We do not disagree with these sentiments, but would suggest that a clearer understanding of what natural enemy impact means to a farmer may better guide research in the future. There are a variety of ways we can measure natural enemy impact (Table 2). These include direct pest mortality due to natural enemies, the reliability or stability of pest mortality due to natural enemies across time, reduction in pest population growth across time, and the relative timing of pest and natural enemy arrival into fields. Depending on the agricultural landscape and pest problems involved, some of these measures may provide more useful information to farmers for decision-making than others. As an example of our line of reasoning we use an exemplary study by Gardiner et al. (2009) that measured the amount of biological control service supplied by naturally occurring arthropod predators in soybean across north– central United States. Impact was measured by experimentally excluding or allowing access to soybean aphid (Aphis glycines) infested plants in the field using simple cages. The change in aphid numbers (after 14 d) in

Reduction in pest population growth across time (Q3)

Reliability of pest mortality across time/space (Q3)

Direct pest mortality (in field) (Q2, Q3)

Potential pest mortality (in laboratory) (Q1)

Type of impact measure Functional response of natural enemy species to pest density Prey diet breadth in no-choice and choice tests Estimate of the rate at which prey are consumed in the field by directly measuring prey mortality due to natural enemies

Description

Sentinel prey caged and exposed to natural enemies, metric of biological control suppression

Predation or parasitism rate measured across time/space

Predation rate in the field Parasitism rate in the field (might require the use of sentinel prey)

Number and type of prey eaten by natural enemy individuals under controlled conditions

Measurement required

C

(to be continued)

Bennett and Gratton (2012), Furlong et al. (2008), Gardiner et al. (2009), Holland et al. (2012), Thies et al. (2011)

Thomson and Hoffmann (2013), Macfadyen et al. (2011)

Latham and Mills (2010), Costamagna and Landis (2007), Furlong et al. (2015)

Chenaux et al. (2011), Propp (1982), Pearce et al. (2004)

Example references

Table 2 The different ways we can measure impact of natural enemies (NE) on pest species. These methods may be used to address some of the questions highlighted in Fig. 1.

28 S. Macfadyen et al.


C 2014

(related to impact) Abundance and diversity of natural enemies (Q6)

Relative abundance or richness of natural enemy species in different crop-types or habitats. Ratio of natural enemy abundance to pest abundance that provides adequate pest control

Sampling for pests and natural enemies Mansfield et al. (2006), Shackelford et at different sites then species al. (2013) identification

Bianchi et al. (2009) The time at which natural enemies and pests arrive in the field and resulting change in pest population growth Relative timing of pest and natural enemy arrival in field (Q3, Q6)

Type of impact measure

Table 2 Continue.

Description

Measurement required

Example references



29

open and caged plants was used to create a Biocontrol Services Index (BSI) (also see the “natural pest suppression index” of Bennett & Gratton, 2012). Gardiner et al. (2009) found that only fairly low numbers of predators, principally Coccinellidae, were required to achieve aphid suppression. The authors noted the times at which the aphid populations in the cage and in the crop surrounding the cages reached an economic threshold level. They concluded that “without predation aphid populations exceeded threshold earlier in the season and more often than in the presence of predators” and estimated that the presence of predators reduced the need for insecticide treatment by 25%–43%. The authors not only quantified the impact of natural enemies but translated these results into information that would be useful for farmers making pest management decisions. They concluded that small but consistent predation early in the season prevented aphids from reaching damaging levels in many fields. It was not just the magnitude of natural enemy impact but the timing and consistency of impact that were important in this production system. In some cases the impact of natural enemies may not only be related to pest mortality but their activities may be valuable for other issues being managed in a production system. Davies et al. (2009) showed that Trichogramma were responsible for 98% of Helicoverpa egg mortality during periods of high host density early season. More importantly, parasitized Helicoverpa eggs will not hatch so at high parasitism levels relatively few larvae emerge and are exposed to Bt-transgenic cotton tissue. It is for this reason that Trichogramma are integral to the insect resistance management strategy and cotton production in northern Australia (also see Heimpel et al., 2005). We need to develop ways that we can measure and incorporate these “auxiliary impact” factors into pest management decision-making.

Conclusions As we have previously mentioned there is a wide range of research questions that need to be addressed in single crop–pest–natural enemy systems, many of which require some estimate of natural enemy impact (Fig. 1, Q1–Q5). However, the measure of impact used needs to be tailored to the needs of farmers and the pest problems they face. This will lead to research outcomes that go beyond simply providing general information about how to conserve natural enemies, but how to use these natural enemies as an integral part of the decision-making process. It may be that in certain production systems the timing of natural enemy arrival into a field is a more useful measure of potential

30

S. Macfadyen et al.

References

impact than say direct mortality estimates (e.g., Meihls et al., 2010; Table 2). In another system, the diversity of natural enemy species (and their activity patterns across space and time) that attack a pest species may be the critical piece of information that farmers require. We have illustrated through the studies cited here that techniques for measuring natural enemy impact do not have to be costly or complex and can in theory be incorporated into a range of studies (Fig. 2). From simple field cages to sentinel prey items, a range of techniques can be used to obtain a measure of impact (Table 2). We would encourage all field studies in production landscapes to incorporate natural enemy impact assessment into their project design and determine how best to communicate the results to farmers in way that can inform their decision-making. There are challenges to incorporating measures of natural enemy impact (Table 1) in any production system; however, the benefits that may accrue from using these techniques more routinely are significant. Eventually, we would like to be able to predict (and even manipulate) where populations of natural enemies occur in space and time, so we can improve pest suppression. To do this we need to develop new and more specific techniques (e.g., landscape engineering, automated trapping networks, precision pest control techniques) which explicitly incorporate natural enemy impacts into pest management systems, rather than just attempting to conserve species. Needless to say, with our current state of knowledge around the impact of natural enemies, this will be a challenge.

Bahar, M.H., Stanley, J.N., Gregg, P.C., Del Socorro, A.P. and Kristiansen, P. (2012) Comparing the predatory performance of green lacewing on cotton bollworm on conventional and Bt cotton. Journal of Applied Entomology, 136, 263– 270. Banks, J.E. and Stark, J.D. (2011) Effects of a nicotinic insecticide, imidacloprid, and vegetation diversity on movement of a common predator, Coccinella septempunctata. Biopesticides International, 7, 113–122. Bastos, C.S., Torres, J.B. and Suinaga, F.A. (2010) Parasitism of cotton leafworm Alabama argillacea eggs by Trichogramma pretiosum in commercial cotton fields. Journal of Applied Entomology, 134, 572–580. Bell, J.R., Andrew King, R., Bohan, D.A. and Symondson, W.O.C. (2010) Spatial co-occurrence networks predict the feeding histories of polyphagous arthropod predators at field scales. Ecography, 33, 64–72. Belliure, B., Amoros-Jimenez, R., Fereres, A. and MarcosGarcia, M.A. (2011) Antipredator behaviour of Myzus persicae affects transmission efficiency of broad bean wilt virus 1. Virus Research, 159, 206–214. Bennett, A.B. and Gratton, C. (2012) Measuring natural pest suppression at different spatial scales affects the importance of local variables. Environmental Entomology, 41, 1077–1085. Bianchi, F., Schellhorn, N.A., Buckley, Y.M. and Possingham, H.P. (2010) Spatial variability in ecosystem services: simple rules for predator-mediated pest suppression. Ecological Applications, 20, 2322–2333. Bianchi, F.J.J.A., Ives, A.R. and Schellhorn, N.A. (2013) Interactions between conventional and organic farming for biocontrol services across the landscape. Ecological Applications, 23, 1531–1543. Bianchi, F.J.J.A., Schellhorn, N.A. and Werf, W.V.D. (2009) Predicting the time to colonization of the parasitoid Diadegma semiclausum: the importance of the shape of spatial dispersal kernels for biological control. Biological Control, 50, 267– 274. Bishop, A.L. (1978) The Role of Spiders as Predators in A Cotton Ecosystem. Department of Entomology, The University of Queensland, Brisbane. Bohan, D.A., Caron-Lormier, G., Muggleton, S., Raybould, A. and Tamaddoni-Nezhad, A. (2011) Automated discovery of food webs from ecological data using logic-based machine learning. PLoS ONE, 6, e29028. Campbell, C.D., Walgenbach, J.F. and Kennedy, G.G. (1991) Effect of parasitoids on lepidopterous pests in insecticdetreated and untreated tomatoes in Western North Carolina. Journal of Economic Entomology, 84, 1662–1667.

Acknowledgments S.M.’s research is funded by the Grains Research and Development Corporation and the National Invertebrate Pest Initiative. Thanks go to John Roberts and an anonymous reviewer for helpful comments on a draft of the manuscript, and A.B. Bennett and E.A. Martin who kindly supplied photos taken during their field work activities for Figure 2.

Disclosure S. Macfadyen, A.P. Davies, and M.P. Zalucki are not aware of any affiliations, memberships, funding, or financial holdings that might be perceived as affecting the objectivity of this work. A.P. Davies’s past research work whilst employed by CSIRO investigated nontarget effects of Bt cotton and was partly funded by Monsanto.

C


Impact of arthropod natural enemies Cardinale, B.J., Duffy, J.E., Gonzalez, A., Hooper, D.U., Perrings, C., Venail, P., Narwani, A., Mace, G.M., Tilman, D., Wardle, D.A., Kinzig, A.P., Daily, G.C., Loreau, M., Grace, J.B., Larigauderie, A., Srivastava, D.S. and Naeem, S. (2012) Biodiversity loss and its impact on humanity. Nature, 486, 59–67. Chaplin-Kramer, R., O’rourke, M.E., Blitzer, E.J. and Kremen, C. (2011) A meta-analysis of crop pest and natural enemy response to landscape complexity. Ecology Letters, 14, 922– 932. Chapman, E.G., Schmidt, J.M., Welch, K.D. and Harwood, J.D. (2013) Molecular evidence for dietary selectivity and pest suppression potential in an epigeal spider community in winter wheat. Biological Control, 65, 72–86. Chenaux, B., Costamagna, A.C., Bianchi, F.J.J.A. and Schellhorn, N.A. (2011) Functional response of two common Australian predators, Dicranolaius bellulus (GuerinMeneville) (Coleoptera: Melyridae) and Micraspis frenata (Erichson) (Coleoptera: Coccinellidae), attacking Aphis gossypii Glover (Hemiptera: Aphididae). Australian Journal of Entomology, 50, 453–459. Collier, T. and van Steenwyk, R. (2004) A critical evaluation of augmentative biological control. Biological Control, 31, 245–256. Costamagna, A.C. and Landis, D.A. (2007) Quantifying predation on soybean aphid through direct field observations. Biological Control, 42, 16–24. Crowder, D.W. (2007) Impact of release rates on the effectiveness of augmentative biological control agents. Journal of Insect Science, 7, 11. Cullen, R., Warner, K.D., Jonsson, M. and Wratten, S.D. (2008) Economics and adoption of conservation biological control. Biological Control, 45, 272–280. Davies, A.P., Carr, C.M., Scholz, B.C.G. and Zalucki, M.P. (2011) Using Trichogramma Westwood (Hymenoptera: Trichogrammatidae) for insect pest biological control in cotton crops: an Australian perspective. Australian Journal of Entomology, 50, 424–440. Davies, A.P., Pufke, U.S. and Zalucki, M.P. (2009) Trichogramma (Hymenoptera: Trichogrammatidae) ecology in a tropical Bt transgenic cotton cropping system: Sampling to improve seasonal pest impact estimates in the Ord River irrigation area, Australia. Journal of Economic Entomology, 102, 1018–1031. Declercq, P. and Degheele, D. (1994) Laboratory measurements of pedation by Podisus maculiventris and P. sagitta (Hemiptera, Pentatomidae) on beet armyworm (Lepidopetra, Noctuidae). Journal of Economic Entomology, 87, 76–83. Derocles, S.a.P., Plantegenest, M., Simon, J.-C., Taberlet, P. and Le Ralec, A. (2012) A universal method for the detection and identification of Aphidiinae parasitoids within their aphid hosts. Molecular Ecology Resources, 12, 634–645.

C 2014


31

Dill, L.M., Fraser, A.H.G. and Roitberg, B.D. (1990) The economics of escape behaviour in the pea aphid, Acyrthosiphon pisum. Oecologia, 83, 473–478. Ehler, E.L. (2006) Integrated pest management (IPM): definition, historical development and implementation, and the other IPM. Pest Management Science, 62, 787–789. Fitt, G.P. (1989) The ecology of Heliothis species in relation to agroecosystems. Annual Review of Entomology, 34, 17–52. Furlong, M.J., Ju, K.H., Su, P.W., Chol, J.K., Il, R.C. and Zalucki, M.P. (2008) Integration of endemic natural enemies and Bacillus thuringiensis to manage insect pests of Brassica crops in North Korea. Agriculture Ecosystems & Environment, 125, 223–238. Furlong, M.J. and Zalucki, M.P. (2010) Exploiting predators for pest management: the need for sound ecological assessment. Entomologia Experimentalis et Applicata, 135, 225–236. Gardiner, M.M., Landis, D.A., Gratton, C., Difonzo, C.D., O’neal, M., Chacon, J.M., Wayo, M.T., Schmidt, N.P., Mueller, E.E. and Heimpel, G.E. (2009) Landscape diversity enhances biological control of an introduced crop pest in the north-central USA. Ecological Applications, 19, 143–154. Glenn, D.C. and Hoffmann, A.A. (1997) Developing a commercially viable system for biological control of light brown apple moth (Lepidoptera: Tortricidae) in grapes using endemic Trichogramma (Hymenoptera: Trichogrammatidae). Journal of Economic Entomology, 90, 370–382. Grant, J.F. and Shepard, M. (1985) Techniques for evaluating predators for control of insect pests. Journal of Agronomic Entomology, 2, 99–116. Grundy, P. and Maelzer, D. (2000) Predation by the assassin bug Pristhesancus plagipennis (Walker) (Hemiptera : Reduviidae) of Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae) and Nezara viridula (L.) (Hemiptera: Pentatomidae) in the laboratory. Australian Journal of Entomology, 39, 280– 282. Hawkins, B.A. and Cornell, H.V. (1994) Maximum parasitism rates and successful biologcial control. Science, 266, 1886– 1886. Head, G., Moar, W., Eubanks, M., Freeman, B., Ruberson, J., Hagerty, A. and Turnipseed, S. (2005) A multiyear, largescale comparison of arthropod populations on commercially managed Bt and non-Bt cotton fields. Environmental Entomology, 34, 1257–1266. Heimpel, G.E., Neuhauser, C. and Andow, D.A. (2005) Natural enemies and the evolution of resistance to transgenic insecticidal crops by pest insects: the role of egg mortality. Environmental Entomology, 34, 512–526. Hemptinne, J.L., Magro, A., Saladin, C. and Dixon, A.F.G. (2012) Role of intraguild predation in aphidophagous guilds. Journal of Applied Entomology, 136, 161–170. Hengeveld, R. and Walter, G.H. (1999) The two coexisting ecological paradigms. Acta Biotheoretica, 47, 141–170.

32

S. Macfadyen et al.

Holland, J.M., Oaten, H., Moreby, S., Birkett, T., Simper, J., Southway, S. and Smith, B.M. (2012) Agri-environment scheme enhancing ecosystem services: a demonstration of improved biological control in cereal crops. Agriculture Ecosystems & Environment, 155, 147–152. Horne, P.A., Edward, C.L. and Kourmouzis, T. (2000) Dicranolaius bellulus (Guerin-Meneville) (Coleoptera : Melyridae: Malachiinae), a possible biological control agent of lepidopterous pests in inland Australia. Australian Journal of Entomology, 39, 47–48. Horne, P.A., Page, J. and Nicholson, C. (2008) When will integrated pest management strategies be adopted? Example of the development and implementation of integrated pest management strategies in cropping systems in Victoria. Australian Journal of Experimental Agriculture, 48, 1601– 1607. Isenhour, D.J. and Yeargan, K.V. (1981) Interactive behavior of Orius insidious Hem, Anthocoridae and Sericothrips variabilis Thys, Thripidae–predator searching strategies and prey escape tatics. Entomophaga, 26, 213–220. Johnson, M.L., Armitage, S., Scholz, B.C.G., Merritt, D., Cribb, B. and Zalucki, M.P. (2007) Predator presence moves Helicoverpa armigera larvae to distraction. Journal of Insect Behaviour, 20, 1–18. Johnson, M.L., Pearce, S., Wade, M., Davies, A., Silberbauer, L., Gregg, P. and Zalucki, M.P. (2000) Review of Beneficials in Cotton Farming Systems. Cotton Research and Development Corporation, Narrabri, Australia. 82 pp. Jonsson, M., Wratten, S.D., Landis, D.A. and Gurr, G.M. (2008) Recent advances in conservation biological control of arthropods by arthropods. Biological Control, 45, 172–175. Keller, M.A. and Lewis, W.J. (1985) Movements by Trichogramma pretiosum (Hymenoptera, Trichogrammatidae) released into cotton. Southwestern Entomologist, 99–109. Knutson, A.E. and Tedders, L. (2002) Augmentation of green lacewing, Chrysoperla rufilabris, in cotton in Texas. Southwestern Entomologist, 27, 231–239. Kogan, M. (1998) Integrated pest management: Historical perspectives and contemporary developments. Annual Review of Entomology, 43, 243–270. Landis, D.A., Wratten, S.D. and Gurr, G.M. (2000) Habitat management to conserve natural enemies of arthropod pests in agriculture. Annual Review of Entomology, 45, 175–201. Latham, D.R. and Mills, N.J. (2010) Quantifying aphid predation: the mealy plum aphid Hyalopterus pruni in California as a case study. Journal of Applied Ecology, 47, 200–208. Lemos, F., Sarmento, R.A., Pallini, A., Dias, C.R., Sabelis, M.W. and Janssen, A. (2010) Spider mite web mediates antipredator behaviour. Experimental and Applied Acarology, 52, 1–10. Letourneau, D.K., Allen, S.G.B. and Stireman, J.O. (2012) Perennial habitat fragments, parasitoid diversity and para-

sitism in ephemeral crops. Journal of Applied Ecology, 49, 1405–1416. Lima, S.L. (1988) Nonlethal effects in the ecology of predator– prey interactions. BioScience, 48, 25–34. Lingren, P.D., Ridgway, R.L., Cowan, C.B., Davis, J.W. and Watkins, W.C. (1968) Biological control of bollworm and tobacco budworm by arthropod predators affected by insecticides. Journal of Economic Entomology, 61, 1521– 1525. Lohaus, K., Vidal, S. and Thies, C. (2013) Farming practices change food web structures in cereal aphid–parasitoid– hyperparasitoid communities. Oecologia, 171, 249–259. Lopez, J.D., Jr., Ridgway, R.L. and Pinnell, R.E. (1976) Comparative efficacy of four insect predators of the bollworm and tobacco budworm. Environmental Entomology, 5, 1160– 1164. Luck, R.F., Shepard, M.B. and Kenmore, P.E. (1988) Experimental methods for evaluating arthropod natural enemies. Annual Review of Entomology, 33, 367–391. Macfadyen, S., Banks, J.E., Stark, J.D. and Davies, A.P. (2014) Using semifield studies to examine the effects of pesticides on mobile terrestrial invertebrates. Annual Review of Entomology, 59, 383–404. Macfadyen, S., Craze, P.G., Polaszek, A., van Achterberg, K. and Memmott, J. (2011) Parasitoid diversity reduces the variability in pest control services across time on farms. Proceedings of the Royal Society B-Biological Sciences, 278, 3387–3394. Macfadyen, S., Gibson, R., Polaszek, A., Morris, R.J., Craze, P.G., Planque, R., Symondson, W.O.C. and Memmott, J. (2009) Do differences in food web structure between organic and conventional farms affect the ecosystem service of pest control? Ecology Letters, 12, 229–238. Mailafiya, D.M., Le Ru, B.P., Kairu, E.W., Calatayud, P.A. and Dupas, S. (2010) Factors affecting stem borer parasitoid species diversity and parasitism in cultivated and natural habitats. Environmental Entomology, 39, 57–67. Mansfield, S., Dillon, M.L. and Whitehouse, M.E.A. (2006) Are arthropod communities in cotton really disrupted? An assessment of insecticide regimes and evaluation of the beneficial disruption index. Agriculture Ecosystems & Environment, 113, 326–335. Martin, E.A., Reineking, B., Seo, B. and Steffan-Dewenter, I. (2013) Natural enemy interactions constrain pest control in complex agricultural landscapes. Proceedings of the National Academy of Sciences of the United States of America, 110, 5534–5539. Meihls, L.N., Clark, T.L., Bailey, W.C. and Ellersieck, M.R. (2010) Population growth of soybean aphid, Aphis glycines, under varying levels of predator exclusion. Journal of Insect Science, 10, 1–18. Mills, N. (1997) Techniques to evaluate the efficacy of natural enemies. Methods in Ecological and Agricultural

C


Impact of arthropod natural enemies Entomology (eds. D.R. Dent & M.P. Waltons), pp. 271–291. CAB International, New York. Orr, D.B., Landis, D.A., Mutch, D.R., Manley, G.V., Stuby, S.A. and King, R.L. (1997) Ground cover influence on microclimate and Trichogramma (Hymenoptera: Trichogrammatidae) augmentation in seed corn production. Environmental Entomology, 26, 433–438. Ortiz-Martinez, S.A., Ramirez, C.C. and Lavandero, B. (2013) Host acceptance behavior of the parasitoid Aphelinus mali and its aphid-host Eriosoma lanigerum on two Rosaceae plant species. Journal of Pest Science, 86, 659– 667. Pareja, M., Brown, V.K. and Powell, W. (2008) Aggregation of parasitism risk in an aphid–parasitoid system: Effects of plant patch size and aphid density. Basic and Applied Ecology, 9, 701–708. Pearce, S., Hebron, W.M., Raven, R.J., Zalucki, M.P. and Hassan, E. (2004) Spider fauna of soybean crops in south-east Queensland and their potential as predators of Helicoverpa spp. (Lepidoptera: Noctuidae). Australian Journal of Entomology, 43, 57–65. Perovic, D., Johnson, M.L., Scholz, B.C.G. and Zalucki, M.P. (2008) The mortality of Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae) neonate larvae in relation to drop-off and soil surface temperature: the dangers of bungy jumping. Australian Journal of Entomology, 47, 289–296. Pinto, J.D. and Stouthamer, R. (1994) Systematics of the Trichogrammatidae with emphasis on Trichogramma. Biological Control with Egg Parasitoids (eds. E. Wajnberg & S.A. Hassan), pp. 1–36. CAB International, Wallingford. Plećaˇs, M., Gagić, V., Janković, M., Petrović-Obradović, O., ˇ Thies, C., Tscharntke, Kavallieratos, N.G., Tomanović, Z., ´ T. and Cetković, A. (2014) Landscape composition and configuration influence cereal aphid–parasitoid–hyperparasitoid interactions and biological control differentially across years. Agriculture, Ecosystems & Environment, 183, 1–10. Pluke, R.W.H. and Leibee, G.L. (2006) Host preferences of Trichogramma pretiosum and the influence of prior ovipositional experience on the parasitism of Plutella xylostella and Pseudoplusia includens eggs. BioControl, 51, 569–583. Pompanon, F., Deagle, B.E., Symondson, W.O., Brown, D.S., Jarman, S.N. and Taberlet, P. (2012) Who is eating what: diet assessment using next generation sequencing. Molecular Ecology, 21, 1931–1950. Propp, G.D. (1982) Functional response of Nabis americoferus to two of its prey, Spodoptera exigua and Lygus hesperus. Environmental Entomology, 11, 670–674. Rand, T.A. (2013) Host density drives spatial variation in parasitism of the alfalfa weevil, Hypera postica, across dryland and irrigated alfalfa cropping systems. Environmental Entomology, 42, 116–122.

C 2014


33

Romeis, J., Shanower, T.G. and Zebitz, C.P.W. (1999a) Trichogramma egg parasitism of Helicoverpa armigera on pigeonpea and sorghum in southern India. Entomologia Experimentalis et Applicata, 90, 69–81. Romeis, J., Shanower, T.G. and Zebitz, C.P.W. (1999b) Why Trichogramma (Hymenoptera: Trichogrammatidae) egg parasitoids of Helicoverpa armigera (Lepidoptera: Noctuidae) fail on chickpea. Bulletin of Entomological Research, 89, 89– 95. Rosenheim, J.A., Kaya, H.K., Ehler, L.E., Marois, J.J. and Jaffee, B.A. (1995) Intraguild predation among biological-control agents–theory and evidence. Biological Control, 5, 303–335. Sankara, F., Dabire, L.C.B., Ilboudo, Z., Dugravot, S., Cortesero, A.M. and Sanon, A. (2014) Influence of host origin on host choice of the parasitoid Dinarmus basalis: Does upbringing influence choices later in life? Journal of Insect Science, 14, 1–11. Schellhorn, N.A., Bellati, J., Paull, C.A. and Maratos, L. (2008) Parasitoid and moth movement from refuge to crop. Basic and Applied Ecology, 9, 691–700. Schellhorn, N.A., Harmon, J.P. and Andow, D.A. (2000a) Using cultural practices to enhance insect pest control by natural enemies. Insect Pest Management, 147–170. Schellhorn, N.S., Manners, A. and Fitt, G.P. (2000b) Augmentation and conservation of parasitoids of Helicoverpa spp.: findings from the first field season. 10th Australian Cotton Conference. Brisbane. Shackelford, G., Steward, P.R., Benton, T.G., Kunin, W.E., Potts, S.G., Biesmeijer, J.C. and Sait, S.M. (2013) Comparison of pollinators and natural enemies: a meta-analysis of landscape and local effects on abundance and richness in crops. Biological Reviews of the Cambridge Philosophical Society, 88, 1002–1021. Sigsgaard, L., Betzer, C., Naulin, C., Eilenberg, J., Enkegaard, A. and Kristensen, K. (2013) The effect of floral resources on parasitoid and host longevity: Prospects for conservation biological control in strawberries. Journal of Insect Science, 13, 1–7. Sigsgaard, L., Esbjerg, P. and Philipsen, H. (2006) Experimental releases of Anthocoris nemoralis F. and Anthocoris nemorum (L.) (Heteroptera: Anthocoridae) against the pear psyllid Cacopsylla pyri L. (Homoptera: Psyllidae) in pear. Biological Control, 39, 87–95. Siqueira, J.R., Oliveira De Freitas Bueno, R.C., Bueno, A.D.F. and Vieira, S.S. (2012) Host preference of the egg parasitoid Trichogramma pretiosum. Ciencia Rural, 42, 1–5. Snyder, W.E. and Ives, A.R. (2001) Generalist predators disrupt biological control by a specialist parasitoid. Ecology, 82, 705– 716. Stanley, J. (1997) The seasonal abundance and impact of predatory arthropods on Helicoverpa spp. in Australian

34

S. Macfadyen et al.

cotton fields. School of Rural Science and Natural Resources. University of New England, Armidale. Strickland, G., Lacey, I., Heading, L. and Yeates, S. (1996) Preliminary pest management studies in winter grown cotton in the Ord River Irrigation Area (ORIA). Proceedings of the 8th Australian Cotton Conference, pp. 189–198. Australian Cotton growers research association, Gold Coast. Tamaddoni-Nezhad, A., Milani, G.A., Raybould, A., Muggleton, S. and Bohan, D.A. (2013) Construction and validation of food webs using logic-based machine learning and text mining. Advances in Ecological Research (eds. W. Guy & A.B. Davids), Volume 49 pp. 225–289. Academic Press, Amsterdam, The Netherlands Thies, C., Haenke, S., Scherber, C., Bengtsson, J., Bommarco, R., Clement, L.W., Ceryngier, P., Dennis, C., Emmerson, M., Gagic, V., Hawro, V., Liira, J., Weisser, W.W., Winqvist, C. and Tscharntke, T. (2011) The relationship between agricultural intensification and biological control: experimental tests across Europe. Ecologcial Applications, 21, 2187– 2196. Thomas, M.B. (1999) Ecological approaches and the development of “truly integrated” pest management. Proceedings of the National Academy of Sciences of the United States of America, 96, 5944–5951. Thomson, L.J. and Hoffmann, A.A. (2013) Spatial scale of benefits from adjacent woody vegetation on natural enemies within vineyards. Biological Control, 64, 57–65. Traugott, M., Bell, J.R., Raso, L., Sint, D. and Symondson, W.O. (2012) Generalist predators disrupt parasitoid aphid control by direct and coincidental intraguild predation. Bulletin of Entomology Research, 102, 239–247. Twine, P.H. and Lloyd, R.J. (1982) Observations on the effect of regular releases of Trichogramma spp. in controlling Heliothis spp. and other insects in cotton. Queensland Journal of Agricultural and Animal Sciences, 39, 159–167.

van Driesche, R.G. (1983) Meaning of percent parasitism in studies of insect parasitoids. Environmental Entomology, 12, 1611–1622. van Driesche, R.G. and Bellows, T.S. (1996) Biological Control. Chapman & Hall, New York. van Driesche, R.G., Bellows, T.S., Elkinton, J.S., Gould, J.R. and Ferro, D.N. (1991) The meaning of percent parasitism revisitied–solutions to the problem of accurately estimating total losses from parasitism. Environmental Entomology, 20, 1–7. van Driesche, R.G., Lyon, S., Sanderson, J.P., Bennett, K.C., Stanek, E.J., III and Zhang, R. (2008) Greenhouse trials of Aphidius colemani (Hymenoptera: Braconidae) banker plants for control of aphids (Hemiptera: Aphididae) in greenhouse spring floral crops. The Florida Entomologist, 91, 583–591. Varma, G.C. and Singh, P.P. (1987) Effect of insecticides on the emergence of Trichogramma brasiliensis Hymenoptera, Trichogrammatidae from parasitized host eggs. Entomophaga, 32, 443–448. Walter, G.H. and Zalucki, M.P. (1999) Rare butterflies and theories of evolution and ecology. Monographs on Australian Lepidoptera, 6, 349–368. Whitehouse, M.E.A., Wilson, L.J. and Fitt, G.P. (2005) A comparison of arthropod communities in transgenic Bt and conventional cotton in Australia. Environmental Entomology, 34, 1224–1241. Yu, D.S. and Byers, J.R. (1994) Inundative release of Trichogramma brassicae Bezdenko (Hymenoptera, Trichogrammatidae) for control of European corn borer in sweet corn. Canadian Entomologist, 126, 291–301. Zalucki, M.P., Adamson, D. and Furlong, M.J. (2009) The future of IPM: whither or wither? Australian Journal of Entomology, 48, 85–96. Accepted September 2, 2014

C
