Introduction
Biological control (or biocontrol) is the use of an organism to reduce or prevent the unwanted impact of another organism, typically through an exploitative interaction (Eilenberg et al. 2001). While competitive relationships are sometimes utilized (Tyndale-Biscoe & Vogt 1996), most biological control agents, including predators, parasitoids, pathogens, and herbivores, are consumers of pest organisms (Heimpel & Mills 2017). Perhaps the best-known form of biological control is “classical” or importation biological control, where a natural enemy is imported from a region other than the target area, often from the native home range of the pest. Today, this involves a rigorous process of enemy selection, efficacy testing, and non-target testing (Bigler et al. 2006), since history is filled with examples of exotic enemies wreaking havoc on naïve, native communities (Simberloff & Stiling 1996). Inundative and inoculative releases of natural enemies, collectively referred to as “augmentative control,” involve the release of large numbers of enemies, either to bolster existing populations or to provide a short pulse of control without long-term establishment. In contrast, conservation biological control is the attempt to increase the effectiveness of already-present enemies. Methods include the provision of alternative resources for biocontrol agents (e.g., extrafloral or floral nectar, pollen), changes in landscape complexity, and the preservation of natural areas beneficial to enemies (Bianchi et al. 2006; Tscharntke et al. 2007, 2016). Altogether, these various methods of biological control provide significant ecosystem services in both natural and agricultural ecosystems (Losey & Vaughan 2006; Zhang & Swinton 2012; Naranjo et al. 2015).
Biological control and predator-prey/parasitoid-host (“natural enemy”) ecology have a long relationship (Hassell & Varley 1969; McMurtry et al. 1970; Murdoch et al. 1985). Early theory in natural enemy ecology was heavily influenced by examples of classical biological control, and broader natural enemy ecology has served to inform biocontrol practice. However, biological control has lagged behind natural enemy ecology by not recognizing the impact and importance of enemy-risk effects, often referred to as non-consumptive effects (NCEs), fear effects, risk effects, non-lethal effects, or trait-mediated effects. Biocontrol typically focuses on direct lethal effects of enemies on pests, whether through consumption or parasitism (which we refer to as consumptive effects or CEs) or infection. However, natural enemy ecology has long recognized the importance of enemy-risk effects (Abrams et al. 1996; Werner & Anholt 1996; Schmitz 1998; Werner & Peacor 2003). Enemies induce behavioral, physiological, morphological or life history changes in their prey that can lead to significant changes in individual fitness, population dynamics, and community dynamics. Meta-analyses and reviews have noted that even when natural enemies kill relatively few prey or hosts, they can have major impacts via enemy-risk effects (Preisser et al. 2005; Peckarsky et al. 2008; Preisser & Bolnick 2008; Schmitz et al. 2008; Sih et al. 2010; Buchanan et al. 2017). While numerous studies have demonstrated major enemy-risk effects in many biological control systems, this knowledge has not been implemented in standard thinking about biocontrol. Several ways that enemy-risk effects connect to biocontrol include understanding: 1) the dynamics of trophic cascades where natural enemies have positive impacts on plants not only by killing pests (CEs), but also by altering pest traits; 2) the role of risk effects in governing interactions in biocontrol systems with multiple enemies, intraguild predation, and bottom-up effects; 3) the impacts of enemy-induced pest dispersal on the spatiotemporal ecology of biocontrol; and 4) how effects of natural enemies differ on coevolved versus naïve prey, as is common for target versus non-target prey, respectively. Insights about enemy-risk effects can thus help to better guide agent selection, non-target testing, integrated pest management (IPM) programs, and other biocontrol practices. Conversely, biocontrol systems are ideal for the general study of enemy-risk effects, offering opportunities to study risk at multiple scales, across multiple trophic levels, with varying levels of co-evolution, and in systems amenable to experimental manipulation.
We provide a systematic overview of insights gained from integrating enemy-risk effects into the ecology of biocontrol, focusing on management of arthropod pests. We begin with a conceptual overview of current literature on enemy-risk effects, including work outside of biocontrol systems, then review studies of enemy-risk effects in biocontrol, and finish by demonstrating and discussing in some detail how a conceptual knowledge of risk effects can inform and improve pest management and biocontrol programs (see Box 1 for a well-studied example).