Visual, Near-Infrared, Short Wave Infrared Reflectance Spectroscopy for Microplastic Identification and Quantification in soils and the Environment

This study investigates the viability of using visible, near-infrared (VNIR), and short wave infrared spectroscopy, specifically the PSR+ spectroradiometer, for the identification and quantification of microplastic pollution across various environmental matrices. Partial Least Squares Regression (PLSR) models were developed using Tennessee red clay soil combined with known concentrations of four plastic polymers: low-density polyethylene (LDPE), polyvinyl chloride (PVC), polystyrene (PS), and polypropylene (PP). The models yielded strong predictive performance with R² values of 0.83 (LDPE), 0.91 (PVC), 0.946 (PS), and 0.98 (PP). Environmental samples were collected from a semi-urban agricultural site in Blount County, Tennessee, and included dry atmospheric deposition, precipitation, groundwater, surface water, and soil. A total of 1,799 microplastic particles were confirmed through microscopic analysis, with fibers being the most abundant type, followed by fragments, films, pellets, microbeads, and foams. Spectral analysis identified LDPE, PVC, nylon, polystyrene, and polypropylene in the environmental samples, with the highest detection rates occurring in soil and surface water. This research demonstrates that the PSR+ spectroradiometer, when paired with properly constructed PLSR models and standardized sample processing, can serve as an effective, non-destructive tool for detecting and characterizing microplastics in terrestrial and aquatic environments.