Quantified Effects of Particle Refractive Index Assumptions on Laser Diffraction Analyses of Selected Soils

The use of laser diffraction (LD) to conduct soil mechanical analysis is rapidly growing across a variety of disciplines. Almost always these techniques require a particle refractive index (RI) assumption. Here, for the first time, we quantify the effects of this assumption across the entire fine earth fraction (FEF) in soils with diverse pedogenic, textural, and mineralogical properties. While most discussions of RI emphasize clay-sized particle sensitivity, we demonstrate that the effects of particle RI assumption are often bimodal across the FEF, with pronounced modes in the 0.5- to 1.5- and 100- to 600-mu m diameter regions. The uncertainty derived from particle RI assumptions averages 5.7 +/- 2.9 (mean +/- 95% CI), 1.9 +/- 1.6, and 4.0 +/- 1.6% (v/v) for estimates of clay, silt, and sand concentration respectively, and 14.5 +/- 6.3, 0.20 +/- 0.04, and 0.10 +/- 0.05 mu m for estimates of the mean diameter of all soil particles, clay-sized particles, and sand-sized particles respectively across our samples. We expose a close relationship between particle RI assumption derived error and the textural properties of samples (i.e., coarse vs. fine-textured), and utilize this relationship to provide first-order constraints that quantitatively inform forthcoming LD analyses of soils about the effects of particle RI assumptions. Generally, uncertainty derived from particle RI assumption is similar in magnitude to that of a full process replicate and is minute compared to inter-soil variability. We suggest that quantitative inquiries evaluating particle RI assumptions across the entire FEF, and in edaphically diverse samples, are too rare. Such work fundamentally informs broader discussions of methodological standardization of LD mechanical analyses.