Size and shape matter: A preliminary analysis of microplastic sampling technique in seawater studies with implications for ecological risk assessment

Microplastic particles (MPs) are widely distributed in seawater. Fibrous MPs (microfibres) are often reported as the most commonly encountered shape of particle. To estimate MP concentrations in seawater, samples are often collected using towed nets (generally 300-350-μm mesh) and may underestimate the amount of microfibres present, which may pass through the mesh due to their narrow width. We compared the potential microplastic particle (PMP) concentration estimates provided by two different seawater sampling methods conducted at three commercial shellfish farms and three unfarmed sites in Baynes Sound, British Columbia, Canada. The methods were: 10-L bucket samples sieved through 63-μm mesh in situ and subsequently filtered through an 8-μm polycarbonate membrane; and 1-L bulk samples collected in jars and subsequently filtered to 8 μm. The jar samples yielded PMP concentrations averaging approximately 8.5 times higher than the bucket samples per L of water (at the site level), largely driven by differences in the number of microfibres. There was no significant difference in PNP concentration between shellfish farms and unfarmed sites. An analysis of MP concentrations and mesh sizes reported in the literature suggests that using a 300-350-μm mesh may underestimate total MP concentrations by one to four orders of magnitude compared with samples that are filtered through much smaller mesh sizes (e.g. <100 μm), despite the effect of sample volume. Particles <300 μm in diameter make up a large component of MPs commonly found in fish and invertebrates. As such, common sampling practices fail to adequately measure a biologically relevant class of MPs, thereby undermining the ability to quantify ecological risk. We suggest that seawater sampling methods be designed to filter to <10 μm (the approximate width of many microfibres), either using pressurized pumps for large-volume samples, or by using sufficient replication of small-volume discrete samples.