昆虫学报 ›› 2026, Vol. 69 ›› Issue (2): 288-301.doi: 10.16380/j.kcxb.2026.02.013

• 综 述 • 上一篇    下一篇

农药对蜜蜂的生态风险评估:暴露评估、效应评估和风险表征

徐超龙1, 荀杨1, 周蕾1, 袁善奎2,*, 尹晓辉1,*   

  1. (1. 浙江农林大学现代农学院, 杭州 311300; 2. 农业农村部农药检定所环境处, 北京 100125)
  • 出版日期:2026-02-20 发布日期:2026-03-19

Ecological risk assessment of pesticides on bees: Exposure assessment, effect assessment and risk characterization

XU Chao-Long1, XUN Yang1, ZHOU Lei1, YUAN Shan-Kui2,*, YIN Xiao-Hui1,*   

  1. (1. College of Advanced Agriculture, Zhejiang Agriculture and Forestry University, Hangzhou 311300, China; 2. Institute for Control of Agrochemicals, Ministry of Agriculture and Rural Affairs, Beijing 100125, China)
  • Online:2026-02-20 Published:2026-03-19

摘要:  蜜蜂作为农业生态系统的关键传粉者,其种群衰退与农药暴露的关联性已成为全球研究热点,不同类型的农药会对蜜蜂产生不同的毒性效应,开展农药对蜜蜂的生态风险评估对保护蜜蜂种群尤为重要。本文从暴露评估、效应评估和风险表征这几个方面综述了欧盟、美国及中国关于农药对蜜蜂的生态风险评估研究进展。暴露评估中,欧盟、美国及中国采用了不同暴露模型,研究表明,欧盟通过参数化暴露模型与三级评估体系(筛查级至田间试验)实现了定量化风险预测;美国的暴露模型采用多因素模拟,基于全球现实数据评估等特点;而中国虽已建立本土化评估标准,但在混合毒性评估、蜂群动态模型及本土蜂种的数据积累方面仍存短板。效应评估中,欧盟、美国及中国都采用了从低级到高级的评估方式,包括个体毒性效应、群体毒性效应和分子毒性效应。风险表征中,美国采用风险商值(risk quotient, RQ)进行评估[RQ=估计环境浓度(estimated environmental concentration, EEC)/致死中剂量(median lethal dose, LD50)],若RQ>关注程度(level of concern, LOC),后续需进一步评估。欧盟采用危害商值(hazard quotient, HQ)[HQ=施药量(application rate, AR)/LD50]进行评估,当HQ>50时,就必须开展高级风险评估。中国采用双轨评估框架: 在喷施场景中, RQ=AR/LD50×50;在土壤或种子处理场景下, RQ=预测暴露剂量(predicted exposure dose, PED)/预测无效应剂量(predicted no-effect dose, PNED);若RQ≤1,判定风险可接受; RQ>1,风险不可接受。总体而言,欧盟评估体系相对完善,将大黄蜂和独居蜂纳入评估体系;涵盖农药代谢物与混合物对蜜蜂的风险评估的方法与流程等。中国应尽快建立相关评估体系,完善评估标准,为农药的有效管理和保护蜜蜂种群提供切实可行的方案。

关键词: 农药残留, 蜜蜂, 暴露模型, 毒性效应, 生态风险评估

Abstract:  Bees are key pollinators in agricultural ecosystems, and the association between their population decline and pesticide exposure has become a global research hotspot. Different types of pesticides have different toxic effects on bees. Conducting ecological risk assessment of pesticides on bees is particularly important for protecting bee populations. In this article, we reviewed the research progress of ecological risk assessment of pesticides on bees in the European Union, the United States and China from the aspects of exposure assessment, effect assessment and risk characterization. In the exposure assessment, the European Union, the United States and China have adopted different exposure models. Studies have shown that the European Union has achieved quantitative risk prediction through the parameterized exposure model and the three-level assessment system (from screening level to field test). The exposure model in the United States features multi-factor simulation, and assessment based on real-world data. China although has established its own assessment standards, it still has shortcomings in the accumulation of data on mixed toxicity assessment, bee colony dynamic models and native bee species. In effect assessment, the European Union, the United States and China all adopt an assessment approach from low to high levels, including individual toxicity effects, population toxicity effects and molecular toxicity effects. In risk characterization, the United States uses the risk quotient (RQ) for assessment [RQ=Estimated environmental concentration (EEC)/median lethal dose (LD50)], and if RQ>level of concern (LOC), further assessment is required. The European Union uses the hazard quotient (HQ) [HQ=Application rate (AR)/LD50] for assessment, and when HQ>50, advanced risk assessment must be conducted. In China, double-line assessment frame is adopted: in the spraying scenario, RQ=AR/LD50×50, and in the soil or seed treatment scenario, RQ=Predicted exposure dose (PED)/predicted no-effect dose (PNED). If RQ≤1, the risk is considered acceptable, and if RQ>1, the risk is unacceptable. Overall, the European Union’s assessment system is relatively complete, including bumblebees and solitary bees in the assessment system, and covering methods and processes for risk assessment of pesticide metabolites and mixtures on bees. China should establish a relevant assessment system as soon as possible, improve assessment standards, and provide practical and feasible solutions for the effective management of pesticides and the protection of bee populations.

Key words:  Pesticide residues, bees, exposure model, toxic effects, ecological risk assessment