THE LONG & THE SHORT OF IT: CAN WE DISTINGUISH MICROPLASTIC SOURCES IN SENECA LAKE USING FIBER LENGTH?

Microplastics—synthetic polymer fragments, pellets, films, and fibers smaller than 5 mm in diameter—are ubiquitous in the environment and humans consume microplastic that contaminates surface drinking water sources. Fibers dominate microplastic shape in freshwater systems. We explore whether fiber length can be used to distinguish the relative contribution of sources feeding Seneca Lake, a major drinking water resource in the Finger Lakes of New York. We sampled various routes fibers can take into the lake: precipitation, direct airfall from the atmosphere, surface flow (Castle Creek), and wastewater effluent. Samples were passed through a benchtop vacuum filtration system with a .45 µm Millipore filter. Filters were illuminated using a NightSea UV/royal blue light to induce fluorescence in many polymers, and photographed; fibers were measured in ImageJ. We found considerable variability in fiber length, which ranged from 0.0078 mm to 117 mm. Most fibers are small (< 0.8 mm) resulting in distributions with a strong right skew. We found small but statistically significant differences in length between fibers deposited from the atmosphere (air fall median = 0.055 mm, precipitation median = 0.0557 mm) compared to those carried in water (Castle Creek median = 0.0633 mm, effluent median = 0.075 mm). On July 13, 2023 a line of thunderstorms interrupted an airfall sampling period, which allowed us to compare direct airfall and precipitation from the same airmass. During the 20-minute storm, we retrieved the airfall sample trays and collected rainwater. Fiber lengths from airfall (median = 0.0483 mm) differed significantly from those in precipitation (median = 0.23 mm). This contradicts results from our larger data set where the median lengths for air (0.055 mm) and precipitation (0.0577) are not significantly different. This contrast highlights strong variability in this contaminant system. Environmental monitoring and modeling must account for this variability. Additionally, the abundance of small fibers and the unimodal distribution of fiber lengths in the Seneca Lake demonstrate that differential length cannot be used to track fiber source.