Three-dimensional distribution and key drivers of neonicotinoid residues in hilly agricultural areas.

Neonicotinoids (NNIs) raise global concern due to their substantial soil residues and potential health risks to animal and human health. High water solubility and low soil adsorption enhanced vertical and horizontal migration of NNIs. However, understanding of NNIs' three-dimensional distribution in soils and influencing factors remains limited, limiting accurate risk assessment and remediation strategies for agriculture ecosystems. This study selected typical mountainous farmland soil to investigate the three-dimensional distribution of NNIs contents and composition. The findings indicated that the average detection rate of imidacloprid (IMI) in the 0-20 cm layer was 33% higher than that in the 30-40 cm layer, whereas clothianidin (CLO) detection rates remained consistent across 0-40 cm layer. The contents of eight NNIs (∑NNIs) in the study area ranged from 0.09 to 37.08 ng/g, with the 6.58±8.65 ng/g in the 0-10 cm and 2.60±7.78 ng/g in the 30-40 cm layer. The contents of ∑NNIs, IMI, and CLO decreased by 60%, 62%, and 75%, respectively, with increasing depth. The proportion of IMI and CLO to ∑NNIs decreased and increased by 35% and 12%, respectively, in the 0-40 cm soil, leading to IMI predominance in the topsoil (60%) and CLO in the deeper soil (29%). Correlation analysis revealed that soil particle size, slope, and elevation were significantly associated with both the ∑NNIs and the proportions of IMI and CLO. These results highlighted the substantial influence of topography and soil structure on the vertical distribution of NNIs. Additionally, the ∑NNIs content in stem mustard soil was higher than sweet potato, rice, corn, and forest. Overall, the study found very low health risks to humans (hazard index, HI<1) and no overall potential ecological risk in the study area, though localized sublethal risks to non-target organisms were identified. Furthermore, the spatial correlation between IMI and CLO health risk regions identified overlapping high-risk areas.