EXPLORING THE REPRODUCIBILITY OF LASER PARTICLE SIZE ANALYSIS

Grain size analysis in sedimentary systems provides insight to large-scale processes including sediment transport mechanisms as well as climatic interpretations in both modern sediment and ancient rocks. However, grain size results from laser particle size analysis (LPSA) can vary between operators depending on techniques and methodology, ultimately affecting precision and reproducibility of these data. The goal of this study is to determine the effects of operator-dependent variables on siliciclastic sediment of various size classes - very-fine (<20 µm), fine (20 – 63 µm), coarse (>63 µm), and mixed (unsieved) - to identify parameters that impact reproducibility. After treatments for organic and carbonate removal, multiple commonly utilized techniques were performed when extracting and dispersing an aliquot into the LPSA (Malvern Mastersizer 3000). Specifically, for each grain-size fraction, we test the impacts of the following variables: obscuration range, sample mass, sediment dispersal technique, time spent in dispersant, and techniques of sample introduction to the LPSA. The degree of reproducibility (highly, moderately, or rarely) is qualitatively determined by visually comparing resultant particle size distributions (PSD), and quantitatively determined by application of Kolmogorov-Smirnov (K-S) tests. Preliminary results indicate that PSDs of coarser fractions (>63 µm) are moderately-to-highly reproducible, whereas very-fine grained (<20 µm) and mixed samples (unsieved) are variably reproducible when operator approach is not consistent. Introduction of the sample into the LPSA introduces the greatest variability in results, perhaps owning to particle settling in the time it takes to transfer the aliquot to the LPSA. Traditionally, grain size data from a single sample are reported as an average of three measurements from one aliquot measured by the LPSA, but our results suggest the need to report data from multiple aliquots, and to provide detailed reporting of operator techniques to bolster reproducibility among research labs and operators.