Calculating Soil Particle Density From Soil Survey Data.

Soil particle density (D_p) represents the composite average of the density of all the particles (both mineral and organic) that make up the <2 mm fraction of soil. In soil survey an estimation of D_p is needed for horizons of all map unit components for use in models and interpretations. The objective was to develop a model that estimates particle density for a wide range of soils. Particle density was estimated by evaluating the different mineral components and organic content by percentage and applying known particle densities weighted by that same percentage. The particle density fraction for organic matter, free iron, salts, gypsum, mica, calcium carbonate, gibbsite, volcanic glass, and total clay contents were calculated. Taxonomic mineralogy class and/or CEC were used to identify the dominant clay mineralogy or presence of gibbsite and secondary iron minerals. Then the density fractions were summed and the remaining unaccounted for fraction, assumed the siliceous fraction was assigned 2.65 Mg m^-3. The general form of calculating D_p was as follows: D_p = organic matter_D_p_fraction + Fe_D_p_fraction + salt_D_p_fraction + gypsum_D_p_fraction + mica_D_p_fraction + calcium carbonate_D_p_fraction + gibbsite_D_p_fraction + volcanic glass_D_p_fraction + clay_D_p_fraction + siliceous_D_p_fraction. For pedons that contain organic matter in the near surface, D_p generally decreased with depth. As the organic matter content decreased, the particle density increased. Seventy-nine soil layers with measured D_p values from the national soil survey characterization database were available for validating the model. Analysis of measured versus predicted densities produced an R² of 0.692 and RMSE of 0.074 Mg m^-3. It is cautioned that if a significant amount of primary heavy minerals is present, the model will underestimate the D_p of the soil.