The limits of biological control datasets to investigate ecological processes at fine scale

Get Complete Project Material File(s) Now! »

Landscape ecology

Biological control is usually deployed over multiple locally restrained areas, granting the opportunity to investigate the role of landscape features on population interactions and dynamics. Furthermore, agroecosystems, which is the target ecosystem for biological practitioners, are a particular kind of landscape. In fact, anthropogenic actions simplify the landscape by, among other things, increasing fragmentation of natural habitats (Tscharntke et al 2005, Baessler and Klotz 2006). Furthermore, agroecosystems are composed of farmlands of which management can be actively modified (Carcamo 1995, Bengtsson et al 2005) and need sustainable regulation of pest populations (Lawton and Brown 1993, Swift et al 1996, Koss et al 2005). Overall, the lower complexity of ecological interactions within agroecosystems and the ability to replicate experimentation, and control ecological variables motivates the use of biological control for investigating landscape ecology.
Landscape structure has been shown to affect community structure, species richness and abundance, population dynamics and interactions within and between trophic levels and naturally the efficiency of biological control (Kareiva 1987, Marino and Landis 1996, Zabel and Tscharntke 1998, Tscharntke and Brandl 2004, Bianchi et al 2006, Finke and Denno 2006, Woodcock et al 2007). In fact, communities, are made up of species with different spatial strategies (Ettema and Wardle 2002, Kareiva 1990, Steffan-Dewenter et al 2002) and the spatial scale of population processes is contingent on the species’ trophic level (Holt et al 1999, Lawton 1995, Pimm 1991). Hence, decreasing size and connectivity of habitats as well as changes of the landscape type between habitats may not only decrease population densities and species richness, but also disrupt plant-herbivore, herbivore-enemy, and plant-pollinator interactions (Didham et al 1996, Matthies et al 1995, Steffan-Dewenter and Tscharntke 1999, Tscharntke and Kruess 1999). In the context of pest biological control, landscape structure also mediates the interactions between native and introduced biological control agents (Didham et al 2007). It is known to influence natural enemy abundance and pest control in other agricultural systems (Bianchi et al. 2006, Chaplin-Kramer et al. 2011, Rusch et al. 2016). For example, biological control has been shown to be hindered by landscape simplification correlating with increased pest numbers and significantly lower yield (Grab et al 2018). Semi-natural habitats have been considered has key landscape features involved in biological control success. For instance, in mango orchards, parasitism rate of the mango gall fly (Proncontarinia matteiana) by its parasitoid Chrysonotomyia pulcherrima decreased as distance from natural vegetation increased (Morgan et al 2017). Tomasetto et al (2017) have even suggested that the changes in landscape (mainly the loss of natural habitat acting as refuge sites for natural enemies), mediated by the intensification of agricultural practices, causes the agroecosystem to evolve a resistance towards biological control.

Evolutionary Ecology

As detailed above, biological control subjects both biological control agents and native species to modifications of a variety of ecological interactions. In response to this change in environment, some protagonists may be subject to evolutionary forces.
Firstly, as the result of specific antagonistic coevolution, which is vastly illustrated with host-parasitoid models, species may evolve to exploit resource or avoid predation more efficiently. This is a process referred to as « arms race ». For example, the parsnip Pastinaca sativa is known to increase in toxic furanocoumarins as they coevolve with their major specialist herbivore, the parsnip webworm, Depressaria pastinacella (Zangerl and Berenbaum 2005). Here the increased toxicity of the weed was systematically observed after reassociation with its coevolved herbivore in non-native areas. In fact, after invading new areas in the world the weed reallocated its resources from chemical defense into growth and reproduction. However, after its herbivore resumed the interaction with the weed in the invaded areas decades later, weeds developed unexpectedly high levels of toxic furanocoumarins (ibid.). In classical biological control, pest may experience evolution driven by the introduction of a biological control agent. In fact, sometimes pests outperform their natural enemies in this arms-race and lead to significant decrease of biological control success. For instance, Tomasetto et al (2017) showed that parasitism rates of an introduced biological control agent may decrease over time due to growing resistance to parasitism among pest populations. In their case, this is supported by the fact that the parasitoid undergoes parthenogenic (thelytokous) reproduction, whereas the pest reproduces sexually. Similarly, Stastny and Sargent (2017) reported that the chrysomelid beetle Neogalerucella calmariensis, introduced into Canada for control of invasive Lythrum salicaria can rapidly select for increased resistance (increased antiherbivore defenses) and tolerance (faster regrowth). Sap feeders such as whiteflies, aphids or mealybugs are known to have coevolved (each separately) with eubacterial endosymbionts (Clark et al 1992) providing essential nutrients to the host (Srivastava 1987). This background, in relation to pest biological control, resulted in the first report of parasitism resistance in aphids induced by a coevolved endosymbiont (Oliver et al 2003).
Secondly, evolutionary forces could stem from a response to an environmental perturbation. Invading species (e.g., ClBC agents) experience novel abiotic and biotic conditions in their introduced environments that can include climates that differ from what they are adapted to, altered availability, distribution, genetic composition, defense, or phenology of their hosts or novel predators, parasitoids, and competitors. These novel ecological conditions may impose strong natural selection, which can lead to evolutionary change (Reznick & Ghalambor, 2001, McEvoy et al 2012). Natural selection is the differential fitness of individuals due to variations in phenotype that lead to the spread of advantageous traits through heritability in a population (Endler et al 1986). For example, biological control agents have been shown to experience, changes in critical daylength for diapause inductions (Bean et al 2012) or increasing development speed and survival when exposed to shorter growing seasons (McEvoy et al 2012, Szucs et al 2012).
Finally, over certain circumstances, the ecological interactions induced by biological control may lead some populations to persist at small sizes, having several potential evolutionary implications. Firstly, the number of founders in a newly introduced population as well as their allele composition may have strong impact on their fitness, population dynamics, dispersal and their ability to coevolve with an antagonistic organism (Briskie and Mackintosh 2004, Hufbauer et al 2013, Szucs et al 2014). In fact, the gene pools of a few individuals (which may not reflect the gene pool of the source population) will restrain the allele composition of the invading population. This evolutionary mechanism is referred to as the “founder effect”. This process may take place into any population that experience a bottleneck (i.e., a drastic reduction in size). ClBC is particularly prone to creating bottlenecks in natural enemy populations before introducing them in the target area (e.g., during sampling, rearing or releasing). This founder effect is at the root of two major evolutionary forces that impacts small populations: genetic drift and inbreeding.
Genetic drift refers to random change in the frequencies of alleles from generation to generation due to stochastic fluctuations (Masel 2011). Genetic drift may cause gene variants to disappear completely and thereby reduce genetic variation (Star and Spencer 2013). It can also cause initially rare or deleterious alleles to become much more frequent and even fixed. This may happen after the population experience a bottleneck and its size is greatly decreased. When populations are small, the rate of inbreeding increases (mating amongst siblings), increasing the damage done by recessive deleterious mutations, in a process known as inbreeding depression (Wright 1977, Shields 1987). Concern about inbreeding is particularly great when population sizes remain small for long periods, as it is often the case for small introduced population that may experience a demographic lag phase (Coutts et al 2018). For instance, some literature (e.g., Baker et al., 2003; Hufbauer et al., 2004; Lloyd et al., 2005) suggest that biological control agents do indeed experience bottlenecks in population size that reduce variation in neutral loci as predicted on theoretical grounds (Hopper et al., 1993). Although the consequences of lower neutral loci variation have not been studied directly in classical biological control, it has been shown to reduce fitness of a parasitoid used in augmentative biological control (Hufbauer 2002, Hufbauer et al 2004).

Trichogramma species: stars of augmentation biological control

READ  POTENTIAL APPLICATIONS OF A NEW METHOD FOR PYRITIC SULPHUR AND ORGANIC SULPHUR QUANTIFICATION USING ROCK-EVAL 7S

Their pre-imaginal development occurs inside the host eggs (Fig 7), the host embryo being usually quickly killed. According to current taxonomy, this genus contains about 210 described species worldwide, 40 of which occurring in Europe. At the genus level, the host range of Trichogramma covers ten insect orders, mainly Lepidoptera (Consoli Trichogramma Egg- Laying Nymphs Larvae Drawing by Alexia Crézé. et al 2010).
Trichogramma species are studied for two main reasons. They are conveniently easy to rear and manipulate in laboratory conditions and they are used as biological control agents worldwide. For instance, in the early 2000 it was estimated that Trichogramma were used for biological control in more than twenty million hectares (Smith 1996). Several Trichogramma species are commercialized for the control of crop pests (T. brassicae against Ostrinia nubilalis), greenhouse productions (T. evanescens against Noctuidae), fruit orchards (e.g., T. cacoeciae against Cydia pomonella) (see Websites of biocontrol agents’ manufacturers Smith et al 2008, van Lenteren 2012). This demonstrates the potential of Trichogramma to provide efficient and economically competitive pest control. However, the current situation is still unsatisfactory as inter and intra-specific biodiversity is poorly documented. Many Trichogramma species are described as highly polyphagous and habitat-generalists, which has been presented as potential drawback for their use in biocontrol (Babendreier et al 2003, Yong and Hoffman 2006, Paraiso et al 2013).

Smarter biological control: Learning more about Trichogramma’s ecology

Efficient biological control stems in major part from the understanding of the biology and ecology of the biological control agent. In the case of Trichogramma, only a few species are used and their ecology (e.g., host range, ecological distribution, etc.) or taxonomy (e.g., species delimitation) are usually poorly known. Moreover, information obtained on the local biodiversity of Trichogramma, non-intentional impacts and their geographical/ecological distributions is essential for regulation agencies to more objectively evaluate requests for the introduction of new species. That is why the INRAE developed a nation-wide initiative to survey the species of Trichogramma in various habitats and host plants. In this endeavor, the survey was carried out exclusively by using surrogate eggs of Ephestia kuehniella through two different methods. Indeed, the eggs were either sprinkled under the leaves or introduced as a “manufactured” patch. In complement to this wide-scale standardized survey, we attempted to document more closely the natural diversity of egg parasitoids in a single location, but using naturally occurring eggs. We focused our efforts on the eggs of Iphiclides podalirius, a common Rhopalocera that is endangered in some parts of Europe (e.g., Belgium, Fichefet et al 2008).

The hidden side of the moon: Sampling wild populations

The scarce swallowtail butterfly (Fig 8), Iphiclides podalirius (Lepidoptera: Papillionidae) is found in a larger geographical area, from south Europe to Western China (Mazel, 2014). It feeds on plants from the Rosaceae family, with a preference for the Prunus genus in Europe (Tolman and Lewington 1997). Eggs are laid singly, mostly under the leaves of the plants mentioned above. They hatch after one to four weeks, depending on temperature. Caterpillars are highly sedentary, especially in the first instars. They spend most of their time on a silk cushion spun on the surface of the leaf selected as a resting site normally the one on which the egg was laid and move only to feed upon nearby leaves. The closely related species Iphiclides feisthamelii is parasitized by several species of Trichogramma (Stefanescu et al 2010).
This project had three main objectives. Firstly, we wanted to describe the egg-parasitoid complex associated with a wild species of butterfly. Secondly, we hoped to get a better understanding of how the different species of egg-parasitoids are distributed as well as the relationships between host eggs, and parasitoids in a natural situation. Thirdly, we aimed at comparing the diversity of Trichogramma obtained from wild eggs with the results collected using sentinel eggs of E. kuehniella.
In addition to the work described in Chapter 4 on this topic, the data acquired allowed the redaction (in preparation) of another manuscript dedicated to the first recording of Trichogramma gicai on Iphiclides podalirius and other hosts. In this article, one-of-a-kind data on the wasp’s behavior are provided, as long as new molecular data, wild footage and information about the holes T. gicai leaves behind after emergence with a comparison between host species.

Inferring dates of colonization

Overall, the base growth model predicted years of colonization that are mostly anterior to the release years of close sites. First, in the Alpes Maritimes region (Figure 10) there is only one release site (2011) and several colonization sites predicted to be invaded from 2008-2009 until 2012-2013. In this region, most of colonization sites are predicted to be invaded by T. sinensis prior to its release in the region. Conversely in Corsica (Figure 11), there is a great number of release sites for only one colonization site. Here the colonization site is predicted to be invaded between 2011 and 2012 for a first release within the island in 2011.
The region of Ardèche (Figure 12) is the region with the most balanced numbered of colonization versus release sites. In Ardèche, the first release happened in 2011 and 13 colonization sites are predicted to have been invaded prior to 2011 with the first invasion predicted in 2009.

The Torymus sinensis case study: some more background

In this chapter, I will again use data from the classical biological control program involving Torymus sinensis against the Asian chestnut gall wasp Dryocosmus kuriphilus in France (See Chapter 1 Section I).
The long post release monitoring of the biological control agent, its target pest and the native community of parasitoids recently associated with D. kuriphilus represents an ideal situation in which to study non-intentional impacts of the biological control program. In fact, the population of native parasitoids has associated itself with D. kuriphilus only a few years prior to the release of T. sinensis when the pest invaded France. Therefore, the stability of such system may be deeply impacted by the perturbation that would represent the introduction of an exotic super-efficient biological control agent.

Table of contents :

Introduction
Experimental Ecology
Observation, theory and experimentation: the triplets of Natural Sciences
Biological control: an opportunity for experimental ecology
Biological control as a playground for experimental ecology
Population dynamics
Community ecology
Landscape ecology
Evolutionary Ecology
Research question
Chapter 1: My case-studies
Torymus sinensis and the Asian chestnut gall wasp Dryocosmus kuriphilus
Biology and Ecology
Classical biological control against the ACGW
Mastrus ridens and the codling moth Cydia pomonella
Biology and Ecology
Classical biological control against C. pomonella
Trichogramma species: stars of augmentation biological control
Smarter biological control: Learning more about Trichogramma’s ecology
The hidden side of the moon: Sampling wild populations
Chapter 2: Population Ecology
The Torymus sinensis case study: some background
Manuscript 1: When did you get there? Inferring time since colonization in naturally colonized locations. (in prep.)
Abstract
Introduction
Methods
Results
Discussion
Assets and Limits of the Dataset
Chapter 3: Community Ecology
The Torymus sinensis case study: some more background
Manuscript 2: The open bar is closed: restructuration of a native parasitoid community following successful control of an invasive pest. (Recommended by Peer Community In Zoology)
Abstract
Introduction
Methodology
Discussion
Assets and Limits of the Dataset
Chapter 4: The limits of biological control datasets to investigate ecological processes at fine scale
Influence of fine-scale landscape structure on the establishment and early expansion of a biological control agent
Release protocol
Mastrus ridens: a partial failure?
Documenting the diversity of native Trichogramma: from the description of species to the
ecological processes structuring communities
Manuscript 3: Survey of Trichogramma species in France and neighboring countries: what drives their local presence and diversity? (First Draft)
Introduction
Methodology
Results
Discussion (bullet points only)
Manuscript 4: Behind the scenes of Trichogramma: a close-scale monitoring study of egg parasitism within the natural population of a native butterfly (in prep.)
Introduction
Methodology
Results
Discussion
Discussion
Biological control: ecological observations and hypotheses
Rolling out the red carpet: The importance of experimental design.
The calm before the storm: The initial state of the system
Is establishment the key to successful experimental ecology?
Post-release monitoring: the issue of scale
References

GET THE COMPLETE PROJECT

Related Posts