Community-level risk assessments on organophosphate esters in the sediments from the Bohai Sea of China based on multimodal species sensitivity distributions coupled with the equilibrium partitioning method

Abstract

Organophosphate esters (OPEs), increasingly used as alternatives to brominated flame retardants, are ubiquitous in the global aquatic environment. Despite their potential toxicological impact on ecosystems, community-level risk assessments for OPEs in sediments remain scarce. This study investigated OPE occurrences and composition characteristics in the Bohai Sea's sediments and appraised both individual and joint ecological risks posed by characteristic OPE homologs using ten commonly used species sensitivity distribution (SSD) models, integrating acute-to-chronic conversion and phase equilibrium partitioning. OPEs were detected across all sediment samples, with total concentrations ranging from 0.213 ng/g dry weight (dw) to 91.1 ng/g dw. The predominant congeners included tri-n-butyl phosphate (TnBP), triisobutyl phosphate (TiBP), tri(2-ethylhexyl) phosphate, tris(2-chloroethyl) phosphate (TCEP), tris(1-chloro-2-propyl) phosphate (TCPP), tris(1, 3-dichloro-2-propyl) phosphate (TDCIPP), and triphenylphosphine oxide. Best-fit SSD models varied among TnBP, TiBP, TCEP, TCPP, and TDCIPP, demonstrating Sigmoid, Burr III, Sigmoid, Burr III, and Burr III, respectively. The same parametric model demonstrated variability in the fitting process for different OPE congeners, which also happened to the fitting results of ten parametric models for the same specific characteristic congener, underscoring the necessity of employing multiple models for precise community-level risk assessments. Hazard concentrations for a 5% cumulative probability were 0.116 mg/L, 2.88 mg/L, 1.30 mg/L, 1.44 mg/L, and 1.85 mg/L for each respective congener. The resulting risk quotients (RQ) and overall hazard index (HI) were selected as criteria to assess the individual and joint ecological risks of OPEs in sediments from the Bohai Sea, respectively. RQ and HI were both below 0.1, indicating a low risk to the local ecosystems. Multi-model SSD analysis could provide refined data for community-level risk evaluation, offering valuable insights for the development of evidence-based environmental standards and pollution control strategies.