Hardsetting in the surface horizons of sandy soils and its implications for soil classification and management

R.J. Harper, Robert Gilkes

Research output: Contribution to journalArticlepeer-review

17 Citations (Scopus)

Abstract

Marked variations in hardsetting occur in the sandy surface horizons of duplex deep sandy soils in a semi-arid area of Western Australia. Hardsetting by definition only occurs in dry soils and increases with field texture. Soil strength measured on remoulded samples in the laboratory conformed with field assessments of strength (consistence). Most (79%) of the variation in strength between Ap horizon samples was explained by clay content, with small differences in clay content resulting in large differences in strength. Half of the maximum measured strength in the Ap horizons was achieved at a clay content of only 8%. The A2 horizons were markedly stronger than corresponding Ap horizons, despite similar clay contents, and this difference in strength is related to the larger organic matter content of the Ap horizons. Hardsetting of these sandy soils may be explained in terms of the cementing action provided by clay which forms bridges between particles. Organic material weakens these bridges.Hardsetting may affect the wind erodibility of sandy soils, through differences in surface conditions (i.e. loose v. compact) and by increasing the resistance to abrasion by saltating sand. It is not clear what effect it will have on plant performance. Hardsetting is a continuous, rather than discrete soil attribute, and if it is to be described in the field, and used in soil classification schemes, objectives class limits should be defined, perhaps using dry consistence ratings.
Original languageEnglish
Pages (from-to)603-619
JournalAustralian Journal of Soil Research
Volume32
DOIs
Publication statusPublished - 1994

Fingerprint

Dive into the research topics of 'Hardsetting in the surface horizons of sandy soils and its implications for soil classification and management'. Together they form a unique fingerprint.

Cite this