Seed coating transfers neonicotinoid risk to pollinators more efficiently than conventional sprays: A field study in...

Neonicotinoid residues are detected in 75% of honey samples worldwide, yet seed treatments are widely promoted as a precision application approach in rice systems. Whether seed treatments and foliar sprays construct fundamentally different pollinator exposure pathways remains mechanistically unresolved. Here, we conducted a five-month field study in Zhejiang Province, China, with weekly synchronized sampling yielding temporally paired soil-pollen-honey sample sets. Integrating toxicity-weighted risk indices with structural equation modeling (cross-validated R² = 0.87), we characterized neonicotinoid transfer pathways through the soil-pollen-honey continuum under contrasting application modes. Seed treatments operated through a high-fidelity plant-mediated pathway: the pollen-to-honey route dominated colony risk (path coefficient = 0.931), with honey contributing 93% of total hazard quotient, no significant direct environmental exposure, and negligible climate regulation (all P > 0.1). The inherently high toxicity of dominant compounds further elevated colony-level risk burdens. Foliar sprays reorganized this architecture: pollen-to-honey transfer weakened by 27%, a direct soil-to-honey route emerged (contributing 20% of colony exposure), and the entire process was strongly climate-regulated (P < 0.05). These findings reveal that seed treatments counterintuitively impose greater persistent pollinator risk through climate-insensitive, high-efficiency plant-mediated transfer, underscoring the need for application mode-specific protection strategies.