Does what we find depend on how we sample? Measured streambed microplastic concentrations can be affected by the choice of sampling method

Microplastics (MPs) are prevalent in rivers worldwide and can adversely impact riverine ecosystems. To sample for MPs in streambeds, a variety of different sampling techniques is applied, including (i) scooping, (ii) coring, (iii) freeze coring, (iv) resuspension method, and (v) piezometer sampling. These common sampling techniques capture different parts of the streambed and different sampling volumes. However, the resulting MP concentrations are usually reported without discussing how the sampling method may affect MP recovery, thus limiting rigorous inter-study comparisons. To better understand the impact of these techniques on MP recovery in different streambed environments, we conducted mesocosm experiments. Two mesocosms were filled with sediment of different composition and spiked with a known concentration of polyamide (PA) fragments prior to employing the various sampling techniques. MP recovery rates from scooping, coring and freeze coring varied by a factor of two compared to the input concentration, while resuspension and piezometer techniques overestimated PA recovery by three to nine times. Furthermore, variations in recovered PA particle size distributions emphasize the importance of selecting methods based on research objectives and streambed characteristics. Our results underline the need for a critical evaluation of reported streambed microplastic concentrations with a focus on the chosen sampling technique.