Stereological measurement of the specific surface area of seasonal snow types: Comparison to other methods, and implications for mm-scale vertical profiling

Density and specific surface area (SSA) are fundamental parameters determining the physical and chemical properties of porous materials like snow. The SSA of snow can be measured directly, using stereology, gas adsorption techniques, X-ray micro-tomography, or indirectly by optical methods deducing the SSA from the snow reflectance. Traditional stereological methods are unbiased only for convex shapes and isotropic snow. We applied new design- and model-based stereological principles, combined with digital image processing, to measure SSA without bias. The new approach only requires vertical sections and a simple, specific sampling design to estimate the SSA of snow. The SSA of thirty snow samples covering all typical snow types and a wide range of density was analyzed using micro-tomography and unbiased stereology. The results show that the SSA of all snow types, even of very complex shape, can be determined with very high precision. The SSA values determined using gas adsorption, X-ray micro-tomography, and stereology were directly compared. They were highly correlated and no systematic deviation was found between these methods. We show that the density and SSA of vertically inhomogeneous and finely layered snow can be analyzed precisely using this new technique. The technical requirements are modest compared to other methods, with similar or better precision. We suggest that this stereology-based method can be the laboratory method of choice to analyze density and SSA of snow in layered samples.