Bayesian species sensitivity distribution modeling for microplastic particles: Integrating particle characteristics...

Area of Science:

• Environmental Science
• Ecotoxicology
• Risk Assessment

Background:

• Global concern over microplastic (MP) pollution necessitates robust ecological risk assessments.
• Assessing MP risks is challenging due to diverse particle characteristics (size, shape).
• Existing methods often do not fully account for MP diversity or censored ecotoxicity data.

Purpose of the Study:

• To develop species sensitivity distributions (SSDs) for microplastics using hierarchical Bayesian modeling.
• To incorporate microplastic characteristics (size, shape) and censored effect concentrations (HONECs) into SSDs.
• To improve the scientific defensibility of ecological risk assessments for microplastics.

Main Methods:

• Developed species sensitivity distributions (SSDs) using hierarchical Bayesian modeling.
• Accounted for intra-species variation and the influence of MP characteristics on chronic no observed effect concentrations (NOECs).
• Incorporated highest observed no-effect concentrations (HONECs) as right-censored data in SSD estimation.

Main Results:

• Models indicated smaller particle sizes (<83 μm) and fiber shapes were associated with lower chronic NOECs.
• Shorter particle lengths and non-spherical shapes (fragments, fibers) were linked to lower chronic NOECs.
• Hazardous concentration for 5% of species (HC5) estimates varied widely based on MP characteristics, ranging from 0.003 to 167 μg/L.

Conclusions:

• The developed SSD modeling framework offers a generalizable approach for microplastic ecological risk assessment.
• Incorporating MP characteristics and censored data improves the accuracy of risk assessments.
• This method can be applied to assess risks from microplastics and other environmental contaminants.