Water quality criteria for selected pharmaceuticals and personal care products for the protection of marine ecosystems

Highlights: • Data to support marine water quality criteria (WQC) for PPCPs was reviewed. • Probabilistic WQC ranged 1–10 μg/L while assessment factor-derived WQC were consistently lower. • Log-logistic model fits better than log-normal model to Species Sensitivity Distributions. • Sulfamethoxazole showed the highest risk among the PPCP studied. Pharmaceuticals and personal care products (PPCP) are increasingly detected in estuarine and coastal waters and organisms but they are absent from lists of priority substances and no environmental quality criteria are available for these chemicals. Ten PPCP were selected on the basis of their occurrence and reported biological effects, oxytetracycline (OTC), sulfamethoxazole (SUL), ibuprofen (IBU), diclofenac (DCF), sertraline (SER), fluoxetine (FLU), carbamazepine (CAR), propranolol (PRO), benzophenone-3 (BP3), and triclosan (TCS). For these chemicals, probabilistic and deterministic water quality criteria (WQC) were derived from their species sensitive distribution (SSD) curves and the critical values respectively, prioritizing ecotoxicological information obtained with sensitive early-life stages of marine organisms. For most PPCP, the probability that the log-logistic model provided better fit than the log-normal model to the SSD was >95%. Using the lower end of the 5th percentile 95% confidence interval, resulting WQC ranged from 1 to 6 μg·L⁻¹ for OTC, IBU, SER, FLU, PRO and TCS. In contrast WQC derived applying the assessment factor method were consistently more conservative, ranging from 0.01 to 0.54 for the same chemicals.