A novel binary pipette splitting sediment subsampling method for improving reproducibility in laser-diffraction particle-size analysis.

Laser-diffraction analysis has been established as one of the standard methods for particle-size distribution (PSD) measurement. However, the uncertainty when analyzing naturally heterogeneous sediment is poorly constrained for the lack of control on one of its largest error sources simply originating from subsampling. Here, we introduce a novel subsampling method, binary pipette splitting (BPS), and verify its precision by using sediment samples from ten flood-layer deposits that have formed in the wake of Hurricane Florence (2018). The BPS approach avoids extracting from only a fixed part of the suspended fluid but considers all the suspended sediment, resulting in the generation of twin subsamples. The median coefficient of variation (COV) for D10, D50, and D90 of subsamples obtained using BPS is 4%, 3%, and 2%, respectively. These values are significantly smaller than the corresponding values of 18%, 15%, and 13% obtained using the conventional pipette subsampling method. Therefore, the new BPS method represents a significant improvement in producing statistically identical subsamples for laser-diffraction particle-size analysis. •The binary pipette splitting (BPS) subsampling method dramatically improves the reproducibility of subsampling wet sediment.