|
In 1949 the stability of dry surface soil to water was usually measured either by wet sieving or the weight of particles <50 micro-meter present after shaking a suspension of aggregates end-over-end in water. Neitehr method distinguished between the immediate microscopic break-up of dry aggregates on wetting (slaking) and the subsequent break-up and release of clay (dispersion) caused by the mechanical and osmotic stresses applied to the wet aggregates. My aim was to design visual tests which separated these effects. As a result a simple classification of soil has been developed mainly based on the ease of dispersion of soil in water. Physical properties of soil in the field have been related to class number and reasons for soil being a member of a given class investigated.
Dry aggregates of Class 1 soil disperse severely when dropped into water. If subsoil in this class uis used to buildl a dam to store run-off in a semi-arid area, then the wall can fail by piping. Dispersion is caused b sufficient exchange sites being occupied by Na and Mg rather than Ca ions and the presence of clay minerals besides kaolin.
Soils in class 3 has to be worked wet before a portion will disperse when dropped into water. If the minimum water content required is less than field capacity, tillage wet followed by rain can result in surface crusting. The minimum can be reduced by organic matter. For example, crusting of a non-sodic kaolinitic soil was affecting the emergence of cotton. After peroxidation, the soil only disperse slightly however wet it was worked. Further when a water extract of eucalypt leaves was added to the peroxidised soil, the soil dispersed at less than field capactiy as before.
With sufficient organic matter present, clay particles can be strongly held together. Result of periodate oxidation and reported ultra-thin sections indicate that the additional bonds in aggregates from grassland are due to the clay tactoids being enveloped in the acid carbohydrate gel exuded by bacteria.
Under grass, available water on a gravimetric basis increases approximately linearly with % C. I suggested this could be wholly due to water held by carbohydrate gel. However, this did not explain satifactorily why the reported plastic limits of silty soils increase linearly with % C. The possible role of plant residues coated mainly with silt particles was overlooked. Such residues could not only retain additional water, but also have sufficient coherence to remain intact when determining the plastic limit. Residues could also explain why the reported increase in the available water per g of C due to rotavating FYM annually into a sandy loam was more than twice that produced under grass in silt loam soils. Rotavation could also produce the fine granular seedbed essential for the uniform emergence of onions!
My favourite experiment aimed to simulate the wetting up of a drained clay subsoil after a summer crop. A portion of a core of Rothamsted subsoil, dried to the wilting point, was suspended from one arm of a balance. 10 mM CaCl2 was then dripped on slowly using a capillary siphon. After a rapid increase in weight initially, there was a further slow increase over the next 100 days! The delayed water uptake was consistent with reported data for the Rothamsted drain gauges and the increase in field mole drains in the Gault clay. I assumed the delayed uptake was due to the slow release of air entrapped in the initial wetting.
The two papers which best cover the application of my work for dams and agriculture are respectively:
1) Emerson (1978). Aggregate classification and hydraulic conductivity of compacted soil. Ch. 30 in "Modification of soil structure", pp. 239-262. John Wiley, London.
2) Emerson (1991). Structural decline of soils, assessment and prevention. Aust. J. Soil Res. 29, 905-921.
| 
