1991Soil ScienceRequires access

ON THE CALCULATION OF MEAN PARTICLE DIAMETER AND STANDARD DEVIATION FROM SAND, SILT, AND CLAY FRACTIONS

Seiji Shiozawa, Gaylon S. Campbell

Open publisher page 71 citations

Abstract

Shirazi and Boersma (SB) proposed that soil texture be described using two parameters, a geometric mean particle diameter and a geometric standard deviation, and suggested transformations for obtaining these parameters from sand, silt, and clay fractions. The SB description of texture has been more useful for predicting hydraulic properties than have conventional sand, silt, and clay percentages, but the proposal and its underlying assumption of a log-normal size distribution have not been tested on soils data. This note describes tests of the proposal and its assumption on samples from six Washington soils. A unimodal log-normal model gave a satisfactory fit for only one of the soils, while a bimodal model gave a good fit to all of the data. Even though the unimodal log-normal distribution failed to fit the particle size data in five of the six cases, the geometric means and standard deviations of all samples were accurately predicted from measurements of sand, silt, and clay fractions. A description of soil texture in terms of geometric mean and geometric standard deviation therefore appears valid, at least for the range of textures in this study. Geometric mean particle diameter, dg, and geometric standard deviation, σ g (both in μM) can be calculated directly from silt (mt) and clay (my) fractions using:

About this research paper

What this paper is about

Shirazi and Boersma (SB) proposed that soil texture be described using two parameters, a geometric mean particle diameter and a geometric standard deviation, and suggested transformations for obtaining these parameters from sand, silt, and clay fractions. The SB description of texture has been more useful for predicting hydraulic properties than have conventional sand, silt, and clay percentages, but the proposal and its underlying assumption of a log-normal size distribution have not been tested on soils data. This note describes tests of the proposal and its assumption on samples from six Washington soils. A unimodal log-normal model gave a satisfactory fit for only one of the soils, while a bimodal model gave a good fit to all of the data. Even though the unimodal log-normal distribution failed to fit the particle size data in five of the six cases, the geometric means and standard deviations of all samples were accurately predicted from measurements of sand, silt, and clay fractions. A description of soil texture in terms of geometric mean and geometric standard deviation therefore appears valid, at least for the range of textures in this study. Geometric mean particle diameter, dg, and geometric standard deviation, σ g (both in μM) can be calculated directly from silt (mt) and clay (my) fractions using:

Why it matters

OpenAlex reports 71 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Shirazi and Boersma (SB) proposed that soil texture be described using two parameters, a geometric mean particle diameter and a geometric standard deviation, and suggested transformations for obtaining these parameters from sand, silt, and clay fractions. The SB description of texture has been more useful for predicting hydraulic properties than have conventional sand, silt, and clay percentages, but the proposal and its underlying assumption of a log-normal size distribution have not been tested on soils data. This note describes tests of the proposal and its assumption on samples from six Washington soils. A unimodal log-normal model gave a satisfactory fit for only one of the soils, while a bimodal model gave a good fit to all of the data. Even though the unimodal log-normal distribution failed to fit the particle size data in five of the six cases, the geometric means and standard deviations of all samples were accurately predicted from measurements of sand, silt, and clay fractions. A description of soil texture in terms of geometric mean and geometric standard deviation therefore appears valid, at least for the range of textures in this study. Geometric mean particle diameter, dg, and geometric standard deviation, σ g (both in μM) can be calculated directly from silt (mt) and clay (my) fractions using:

Key concepts: Silt, Geometric mean, Geometric standard deviation, Standard deviation, Soil texture, Mathematics, Soil water, Texture (cosmology)

Related papers

Back to paper searchBrowse research topicsOriginal source
ON THE CALCULATION OF MEAN PARTICLE DIAMETER AND STANDARD DEVIATION FROM SAND, SILT, AND CLAY FRACTIONS — Research Paper | ScholarLens