Literature review of biological control enemy-risk effect studies
We carried out a systematic review of empirical studies on enemy-risk effects in biocontrol systems using combinations of the search terms “biological control”, “biocontrol”, and “pest” with the terms “non-consumptive”, “nonconsumptive”, “non-lethal”, “nonlethal”, “sub-lethal”, “sublethal”, “risk effect*”, “anti-predator”, or “anti-predator.” Studies were included if they were on arthropod pests, investigated some stage of the enemy-risk effect pathway depicted in Fig. 1, and demonstrated some relevance to pest control. Our review of the literature yields several takeaway messages: 1) enemy-risk effects are prevalent in arthropod pest systems, 2) enemy-induced trait shifts can interact with other aspects of agroecosystems, such as plant defenses, trap crops, and plant pathogen transmission, 3) risk effects produced by predators have been studied more extensively than those produced by parasitoids, 4) the importance of enemy-risk effects on non-target species has received little attention, and 5) few studies have examined the consequences of enemy-risk effects for plant damage in the field.
We organized papers in Table 1 according to the “level” of study, ranging from documentation of enemy-induced trait responses to explicit measure of NCEs on pest control and trait-mediated indirect effects on crops (see Supporting Information for expanded table format). This categorization is not meant to rank the quality or usefulness of studies, but rather to demonstrate where research has been focused and where room for growth remains. 54% of studies (32 of 59) aimed to assess the strength of pest responses, which is a critical step in the inclusion of enemy-risk effects in the design and implementation of biocontrol programs. Many of these studies incorporated other aspects relevant to pest management, such as variation in spatial scale (Lee et al. 2014), ability to transmit plant pathogens (Tholt et al. 2018), interactions with trap cropping (Lee et al. 2011), and plant defense (Thaler et al. 2014). Of the 27 remaining studies, about half documented demographic consequences for pests, and half documented levels of pest damage. Four studies measured changes in plant damage in the field (Griffin & Thaler 2006; Thaler & Griffin 2008; Steffan & Snyder 2010; Hermann & Thaler 2018). Only two studies measured the risk effects of enemies on non-target species (Walzer & Schausberger 2009; Fill et al. 2012); these effects are likely overlooked in many evaluations of host range, as we discuss in the following section.f
It can be difficult to scale up enemy-risk effect studies from measuring pest responses to measures of biocontrol efficacy, including effects on pest population dynamics or crop yield, as these typically require longer timescales and broader spatial scales (Hermann & Landis 2017). However, when moving from pest-agent interactions to the harvest and sale of a crop, there are many steps where the enemy-risk effects may attenuate (Hamburg & Hassell 1984; Godfray & Waage 1991; Collier & Van Steenwyk 2004; Kaplan et al. 2014). Additionally, there may be many interacting effects on pests and crop yield, ranging from environmental factors to pesticide applications. Due to these complications, enemy-risk effect studies that do not measure outcomes beyond pest responses may not fully capture the relevance of enemy-risk effects in pest management.
Some of the most fruitful areas for further research include 1) separating NCEs and CEs to improve predictions of pest population dynamics (see Box 4), 2) considering enemy-risk effects that include qualitative shifts, such as spatiotemporal location, and how they interact with agricultural practices in ways that differentiate them from CEs, 3) including enemy-risk effects in assessment of agent efficacy and non-target impacts, 4) expanding taxonomic breadth to include more parasitoids, and 5) expanding scales of study to better understand impacts on crop production. We believe ongoing empirical work would be well served by incorporating theory from the broader study of enemy-risk effects, which would facilitate predictions about when and where risk effects may play an important role in the efficacy of pest management programs.