Evaluating the occurrence of veterinary medicines at farm scale and the efficacy of current groundwater modelling.

The use of contaminated animal manures as an organic fertilizer offers a sustainable approach to waste recycling; however, it also presents a potential threat towards the environment, and human health. This study examines the presence/absence of veterinary medicines in pig slurries, soils, and groundwater at a commercial pig farm for one year. Pig slurry concentrations ranged from 0.004 μg/kg to 1.2 mg/kg (d/w) for azaperone, oxytetracycline, lincomycin, sulphadiazine and trimethoprim respectively. Following the application of contaminated slurry to land, antibiotics were detected (≤10 cm depth), although antibiotic concentrations were identified to dissipate below detectable limits following immediate incorporation to ≤20 cm. The veterinary antibiotics lincomycin, sulphadiazine, and trimethoprim were quantified at depths of 12.5 to 47 m, in addition, veterinary antibiotics were identified to persist throughout the sampling period (13/12/20 - 07/07/21, 0.011 to 1.5 ng/L). Following the application of manure to land, antibiotic concentrations in groundwater increased by a factors of 3 to 8.2 for sulphadiazine and trimethoprim respectively (12.5-22 m), respectively. As expected the groundwater leaching predictions were over predicted when following the worst-case risk assessment procedure. However, FOCUS_PEARL performed well at predicting the tendency for a chemical to leach - 66.67 % of the assessed were in agreement with presence/absence. In contrast, when corrected site-specific administration data were applied, PECGROUNDWATER predictions were underestimated. While overprediction within the risk assessment context reflects the suitability of the Tiered approach for conservative evaluations, underprediction with site-specific data highlights limitations in the modelling approach to accurately simulate the environmental fate of veterinary medicines.