Out of site, out of mind? Considering pesticide drift and plant mutualisms

Abstract

Understanding the in situ impacts of pesticides on nontarget taxa entails investigating patterns of use, patterns of exposure, and the effects of exposure on survival, growth, or reproduction. Pesticide exposure is the overlap of pesticide contamination with an organism's spatiotemporal distribution. Pesticide use is a spatially discrete event, but the movement of pesticides (i.e. drift) beyond application areas during or postapplication, either as airborne particles or runoff, expands the spatial kernel in which exposure can occur. Despite practices aiming to mitigate drift through attention to applicator technology, product formulation, and spray conditions (e.g. temperature, relative humidity, or wind speed), off-site contamination is a persistent problem in agricultural landscapes. For example, herbicide drift potentially decreases neighboring crop yields (Egan et al., 2014) and insecticide drift can impact beneficial arthropods (Otto et al., 2013). Understanding the frequency and extent of pesticide drift is critical, particularly given efforts to enhance edge-of-field habitats for beneficial insects. The work by Baucom et al. is a leap forward because it demonstrates that drift-level herbicide exposure indirectly influences pollinator activity by altering floral traits.

The effects of exposure can be direct if a pesticide has an instantiated, unmediated impact on the survival, growth, or reproduction of an organism or indirect if the impact of a pesticide is mediated by another organism (Fig. 1). Considering these distinct pathways of influence disentangles how pesticide exposure propagates through ecosystems. How might multiple plant mutualisms be affected by both direct and indirect effects of pesticide exposure? The work by Baucom et al. contributes to a body of evidence demonstrating important, but often overlooked, indirect effects of herbicides, fungicides, and insecticides on plant–pollinator interactions.

Other studies have shown that herbicides alter floral displays and change pollinator visitation patterns. Bohnenblust et al. (2016) found that dicamba doses simulating particle drift delayed flowering and reduced flower numbers in Medicago sativa and Eupatorium perfoliatum; subsequently, these plants were visited less often by pollinators. That herbicide exposure decreases floral density is perhaps unsurprising. However, demonstrating that the indirect effects of pesticide use propagate to influence pollinator behavior, pollination, and plant reproduction (Fig. 1a) is a pathbreaking direction for applied ecological research. Moreover, these plastic changes in floral display may be accompanied by genetic responses, and herbicide-mediated selection may set eco-evo dynamics in motion (Iriart et al., 2021), which drive phenotypic change in traits associated with pollinator attraction (Kuester et al., 2017). The effects of herbicides on pollination are not all indirect; herbicides can have direct impacts on nontarget pollinator taxa (Cullen et al., 2019).

After pollinators, mycorrhizal fungi are one of the most well-known examples of plant mutualists. These symbionts form associations with nearly 90% of land plants (van der Heijden et al., 2015) and confer a wide array of benefits to their host plants, including improved floral display and reproduction (Hyjazie & Sargent, 2024). Fungicide application may indirectly affect plant–pollinator interactions by limiting mycorrhizal nutrient exchange, which in turn influences a plant's resource allocation to reproductive tissues (Fig. 1b). For example, plant mutualisms with arbuscular mycorrhizal fungi (AMF) influence flower number (Wolfe et al., 2005), flower size (Gange & Smith, 2005), pollen or nectar availability (Gange & Smith, 2005; Guzman et al., 2025), and flower scent (Bennett & Meek, 2020). These changes in floral traits often increase visits by pollinators, resulting in an improved seed set (Gange & Smith, 2005; Wolfe et al., 2005). Above- and belowground mutualisms may be disrupted by the nontarget effects of fungicide application. Plants grown in soils with a legacy of fungicide use experience reduced hyphal transfer of phosphate from AMF (Edlinger et al., 2022), which can alter floral traits (Cahill Jr. et al., 2008). As Baucom et al. suggest, understanding the multi-trophic impacts of pesticide use will not only yield novel insights but also be necessary evidence to support sustainable pest control, such as integrated pest–pollinator management (IPPM; Lundin et al., 2021).

An extensive literature shows largely negative direct effects of insecticides on pollinators, such as bees (Raine & Rundlöf, 2024). More complicated, however, are the indirect effects of insecticides on plant–pollinator interactions when considering that herbivory and the control of herbivores by natural enemies are also influenced by insecticide exposure (Fig. 1c). Ecological research in natural systems shows that herbivory of vegetative and floral tissues reduces floral displays, which in turn alters pollinator visitation (Strauss & Irwin, 2004). Therefore, control of herbivores, whether by insecticides or natural enemies, may improve pollination. This raises an important agronomic question: What is the net effect of biological pest control and pollination on crop yields? Studies have tested whether the joint effect of these ecosystem services is independent, compensatory, or synergistic. For example, Lundin et al. (2021) found positive synergistic effects: seed sets obtained when simultaneously increasing pollination and biocontrol outweighed the sum of seed set gains obtained when increasing each service separately. Insecticide use may also reduce the abundance of natural enemies that regulate herbivore populations (Bommarco et al., 2011), and if this insecticide exposure also reduces pollinator populations, then negative synergistic effects could occur. Multiple trade-offs occur between insecticide use, pest control, and pollinators (Knapp et al., 2022). Thus, holistic pest management plans (e.g. IPPM) are key to balancing agronomic and conservation concerns.

Baucom et al. found that the herbicide dicamba, which has no currently known direct effect on pollinating insects, altered pollinator behavior by reducing the abundance and complexity of floral resources provided by a community of plants that typically occur in natural habitats near crops (i.e. weeds). Indirect effects of pesticides are overlooked and likely widespread, and we urgently require tools to predict the cascading consequences of exposure to multiple pesticides in agriculture. These tools materialize by integrating ecotoxicology with community ecology, which has a long history and theoretical focus on interaction networks, spatiotemporal scales, and community structure (Rohr et al., 2006).

Off site. Nontarget. Indirect. Like most perturbations to complex systems, pesticide use creates externalities and unintended consequences. To forget this is to lose sight of the beneficial complexities essential to agroecosystems.

The New Phytologist Foundation remains neutral with regard to jurisdictional claims in maps and in any institutional affiliations.