Fig. 2. Half-lives (DT50, d, panels A–E) of the pesticides most frequently detected in soils under potato cultivation, i.e., ametryn (herbicide (H), n = 25), azoxystrobin (fungicide (F), n = 27), chlorothalonil (F, n = 20), cyproconazole (F, n = 10), and S-metolachlor (H, n = 38) observed at different sites. Only individual true-positive cases (applied-detected) were used for the boxplots. Data points are categorised according to three sampling times: before planting potato (s1, circle), during peak pesticide application (s2, triangle), around harvest (s3, cross) and cultivation period: 2018–2019 (CP1, black), 2019–2020 (CP2, red), and 2020–2021 (CP3, blue). If a sampling time was not depicted, the residue fell below the limit of quantification and/or was not a true-positive case. The boxes represent the 25th to 75th percentile, and the bold white line dividing the box is the median of single data points. Geometric means of data from single sites served to indicate the median over all sites (dashed line). Dotted line shows DT50,field of the Pesticide Properties Database (PPDB44; for S-metolachlor, the line of geometric mean and the PPDB44 coincide).
