Skip to main content
SLU publication database (SLUpub)

Abstract

Soil particle density (Dp) is an important soil property for calculating soil porosity expressions. However, many studies assume a constant value, typically 2.65 Mg m(-3) for arable, mineral soils. Few models exist for the prediction of Dp from soil organic matter (SOM) content. We hypothesized that better predictions may be obtained by including the soil clay content in least squares prediction equations. A calibration data set with 79 soil samples from 16 locations in Denmark, comprising both topsoil and subsoil horizons, was selected from the literature. Simple linear regression indicated that Dp of clay particles was approximately 2.86 Mg m(-3), while that of sand + silt particles could be estimated to similar to 2.65 Mg m(-3). Multiple linear regression showed that a combination of clay and SOM contents could explain nearly 92% of the variation in measured Dp. The clay and SOM prediction equation was validated against a combined data set with 227 soil samples representing A, B, and C horizons from temperate North America and Europe. The new prediction equation performed better than two SOM-based models from the literature. Validation of the new clay and SOM model using the 227 soil samples gave a root mean square error and mean error of 0.041 and +0.013 Mg m(-3), respectively. Predictions were accurate for all levels of SOM content in the validation data set. The model gave very precise predictions for soils with clay contents lower than 03 kg kg 1, while a moderate over-prediction was observed for soils very high in clay. Finally, we developed a texture-enhanced curvilinear model that will be useful for predicting Dp of soils with high contents of clay and in particular SOM. (C) 2016 Elsevier B.V. All rights reserved.

Keywords

Soil particle density; Soil organic matter; Clay; Pedotransfer function; Regression

Published in

Geoderma
2017, volume: 286, pages: 83-87
Publisher: ELSEVIER SCIENCE BV

SLU Authors

UKÄ Subject classification

Soil Science

Publication identifier

  • DOI: https://doi.org/10.1016/j.geoderma.2016.10.020

Permanent link to this page (URI)

https://res.slu.se/id/publ/78413