A Probabilistic Risk Framework for Microplastics Integrating Uncertainty Across Toxicological and Environmental Variability : Development and Application to Marine and Freshwater Ecosystems

Title data

Coffin, Scott ; Bertrand, Lidwina ; Ahmed, Kazi Towsif ; de Souza Leite, Luan ; Cowger, Win ; Siña, SinaMariella ; Barrick, Andrew ; Kukkola, Anna ; Almroth, Bethanie Carney ; Miller, Ezra ; Yeh, Andrew ; Kennedy, Stephanie ; Mair, Magdalena:
A Probabilistic Risk Framework for Microplastics Integrating Uncertainty Across Toxicological and Environmental Variability : Development and Application to Marine and Freshwater Ecosystems.
In: Journal of Hazardous Materials. (1 January 2026) . - 141021.
ISSN 0304-3894
DOI: https://doi.org/10.1016/j.jhazmat.2025.141021

Project information

Abstract in another language

Quantitative risk assessment for microplastics (MPs) is complicated by misalignments between environmentally relevant particles and those used in toxicity studies. Previous approaches addressed this using ecologically relevant metrics (ERMs) and species sensitivity distributions (SSDs), but did not propagate uncertainty from particle-trait alignments or intraspecies variability. Here, we present a novel probabilistic framework that propagates uncertainty through ERM alignments using Monte Carlo (MC) simulation, paired with a modified probabilistic SSD model (PSSD++). Using high-quality data from the updated Toxicity of Microplastics Explorer (ToMEx 2.0), we compared hazard thresholds derived by three approaches: traditional SSD, MC+SSD, and PSSD++. PSSD++ consistently produced the most health-protective median thresholds and lowest 5th-percentile values, which generally exhibited the widest relative confidence intervals. MC+SSDs produced the narrowest uncertainty ranges. Uncertainty was greater for food dilution than for tissue translocation, and greater for freshwater environments than marine. Sensitivity analysis identified ERM-alignment parameters as the dominant drivers of threshold variability, contributing up to two orders of magnitude difference. This framework emphasizes the importance of propagating alignments uncertainty in MP risk assessments and highlights key research needs, including improved models for tissue translocation and more representative environmental particle characterizations.