A Case Study on Best Practices for Benchmarking Pesticide Degradation with Regulatory DT50 Values

Pesticide soil degradation half-lives (DT50) derived from laboratory studies are commonly summarized as representative reference values that support environmental exposure and risk assessment. Interpretation of degradation studies using more complex conditions compared to the single-compound standard studies may be difficult in terms of considering the variability of the underlying laboratory datasets. This work demonstrates a benchmarking and interpretation framework that uses regulatory reference datasets to evaluate whether experimentally determined DT50 values fall within experimental variability, illustrated through a multi-pesticide co-spike degradation case study. Degradation of a mixture of twelve pesticides (acetamiprid, azoxystrobin, bentazon, boscalid, chloridazon, dimethenamid-P, fenhexamid, metazachlor, metobromuron, mesosulfuron-methyl, propamocarb, and terbuthylazine) was investigated in seven agricultural soils. Experimental DT50 values were benchmarked against European Food Safety Authority (EFSA) reference datasets using reference value, reference interval, and expected variability derived from laboratory degradation studies. The interpretation was subsequently validated against a larger literature-derived reference dataset. The majority of experimentally determined DT50 values were consistent with the EFSA reference dataset. However, benchmarking against the expected variability identified several exceedances that could not be explained by differences in soil physicochemical properties alone. Validation against the literature-derived reference dataset refined the interpretation of several benchmark exceedances by accounting for a broader range of laboratory degradation behavior, while persistent exceedances identified across reference datasets were highlighted as candidates for further mechanistic investigation. Overall, the proposed benchmarking framework complements existing regulatory degradation endpoints by incorporating expected laboratory variability into the interpretation of newly generated degradation data. The framework provides an efficient first screening step for complex degradation scenarios, enabling researchers to distinguish expected experimental variation from observations requiring targeted mechanistic investigation.