Spatial variability, source identification and risks assessment of antibiotics in multimedia of North China's largest freshwater lake using positive matrix factorization and Monte Carlo simulation.

Antibiotics are widely found in aquatic ecosystems and pose a serious threat to human and the ecological system. Samples of surface water (SW), overlying water (OW), pore water (PW) and sediments (Sedi) were collected to investigate the spatial variability, potential sources, ecological risk (RQs) and health risks (HQs) of nine common antibiotics in Baiyangdian Lake using positive matrix factorization (PMF), and Monte Carlo simulation. Significant spatial autocorrelation of most antibiotics were observed in PW and Sedi samples rather than in SW and OW samples, and higher antibiotic levels were found in the northwest of waters and the southwest of sediments. Livestock (26.74-35.57%) and aquaculture (21.62-37.70%) were identified as primary sources of antibiotics in the water and sediments. Norfloxacin and roxithromycin showed high levels of RQ and HQ in more than 50% of samples, respectively. The combined RQ (ΣRQ) in the PW can be used as a sign of across multimedia risk. Notably, appreciable health risks were observed for the combined HQ (ΣHQ) in about 80% of samples, indicating the importance of taking health risk of antibiotics into consideration. The findings of this work provides a reference for antibiotics pollution control and risk management in shallow lake.